provider: cyberian
model: deep-research
cached: false
start_time: '2026-01-15T14:31:18.134639'
end_time: '2026-01-15T14:49:12.326350'
duration_seconds: 1074.19
template_file: templates/gene_research_go_focused.md
template_variables:
organism: human
gene_id: AIFM2
gene_symbol: AIFM2
uniprot_accession: Q9BRQ8
protein_description: 'RecName: Full=Ferroptosis suppressor protein 1 {ECO:0000303|PubMed:31634899,
ECO:0000303|PubMed:31634900}; Short=FSP1 {ECO:0000303|PubMed:31634899, ECO:0000303|PubMed:31634900};
EC=1.6.5.- {ECO:0000269|PubMed:31634899, ECO:0000269|PubMed:31634900, ECO:0000269|PubMed:35922516};
AltName: Full=Apoptosis-inducing factor homologous mitochondrion-associated inducer
of death {ECO:0000303|PubMed:11980907}; Short=AMID {ECO:0000303|PubMed:11980907};
AltName: Full=p53-responsive gene 3 protein {ECO:0000303|PubMed:12135761};'
gene_info: Name=AIFM2 {ECO:0000303|PubMed:26689472, ECO:0000312|HGNC:HGNC:21411};
Synonyms=AMID {ECO:0000303|PubMed:11980907}, PRG3 {ECO:0000303|PubMed:12135761};
organism_full: Homo sapiens (Human).
protein_family: Belongs to the FAD-dependent oxidoreductase family.
protein_domains: FAD/NAD-bd_sf. (IPR036188); FAD/NAD-binding_dom. (IPR023753); Pyr_redox_2
(PF07992)
provider_config:
timeout: 1800
max_retries: 3
parameters:
allowed_domains: []
agent_type: claude
port: 3284
skip_permissions: true
citation_count: 17

Question

Gene Research for Functional Annotation

⚠️ CRITICAL: Gene/Protein Identification Context

BEFORE YOU BEGIN RESEARCH: You MUST verify you are researching the CORRECT gene/protein. Gene symbols can be ambiguous, especially for less well-characterized genes from non-model organisms.

Target Gene/Protein Identity (from UniProt):

• UniProt Accession: Q9BRQ8
• Protein Description: RecName: Full=Ferroptosis suppressor protein 1 {ECO:0000303|PubMed:31634899, ECO:0000303|PubMed:31634900}; Short=FSP1 {ECO:0000303|PubMed:31634899, ECO:0000303|PubMed:31634900}; EC=1.6.5.- {ECO:0000269|PubMed:31634899, ECO:0000269|PubMed:31634900, ECO:0000269|PubMed:35922516}; AltName: Full=Apoptosis-inducing factor homologous mitochondrion-associated inducer of death {ECO:0000303|PubMed:11980907}; Short=AMID {ECO:0000303|PubMed:11980907}; AltName: Full=p53-responsive gene 3 protein {ECO:0000303|PubMed:12135761};
• Gene Information: Name=AIFM2 {ECO:0000303|PubMed:26689472, ECO:0000312|HGNC:HGNC:21411}; Synonyms=AMID {ECO:0000303|PubMed:11980907}, PRG3 {ECO:0000303|PubMed:12135761};
• Organism (full): Homo sapiens (Human).
• Protein Family: Belongs to the FAD-dependent oxidoreductase family.
• Key Domains: FAD/NAD-bd_sf. (IPR036188); FAD/NAD-binding_dom. (IPR023753); Pyr_redox_2 (PF07992)

MANDATORY VERIFICATION STEPS:

1. Check if the gene symbol "AIFM2" matches the protein description above
2. Verify the organism is correct: Homo sapiens (Human).
3. Check if protein family/domains align with what you find in literature
4. If you find literature for a DIFFERENT gene with the same or similar symbol, STOP

If Gene Symbol is Ambiguous or You Cannot Find Relevant Literature:

DO NOT PROCEED WITH RESEARCH ON A DIFFERENT GENE. Instead:
- State clearly: "The gene symbol 'AIFM2' is ambiguous or literature is limited for this specific protein"
- Explain what you found (e.g., "Found extensive literature on a different gene with the same symbol in a different organism")
- Describe the protein based ONLY on the UniProt information provided above
- Suggest that the protein function can be inferred from domain/family information

Research Target:

Please provide a comprehensive research report on the gene AIFM2 (gene ID: AIFM2, UniProt: Q9BRQ8) in human.

The research report should be a detailed narrative explaining the function, biological processes, and localization of the gene product. Citations should be given for all claims.

You should prioritize authoritative reviews and primary scientific literature when conducting research. You can supplement
this with annotations you find in gene/protein databases, but these can be outdated or inaccurate.

We are specifically interested in the primary function of the gene - for enzymes, what reaction is catalyzed, and what is the substrate specificity? For transporters, what is the substrate? For structural proteins or adapters, what is the broader structural role? For signaling molecules, what is the role in the pathway.

We are interested in where in or outside the cell the gene product carries out its function.

We are also interested in the signaling or biochemical pathways in which the gene functions. We are less interested in broad pleiotropic effects, except where these elucidate the precise role.

Include evidence where possible. We are interested in both experimental evidence as well as inference from structure, evolution, or bioinformatic analysis. Precise studies should be prioritized over high-throughput, where available.

Output

AIFM2/FSP1: A Comprehensive Review of Ferroptosis Suppressor Protein 1

1. Introduction and Summary

Apoptosis-inducing factor mitochondria-associated 2 (AIFM2), now more commonly known as Ferroptosis Suppressor Protein 1 (FSP1), is a flavoprotein oxidoreductase encoded by the AIFM2 gene (chromosome 10q22.1) in humans [ohiro-2002-prg3-abstract]. The protein belongs to the FAD-dependent oxidoreductase family and contains 373 amino acids with a molecular mass of approximately 40.5 kDa [ohiro-2002-prg3-abstract]. This protein has undergone a remarkable reconceptualization since its original discovery: initially characterized as a pro-apoptotic factor homologous to AIF, it is now understood to function primarily as a potent suppressor of ferroptosis, a form of iron-dependent regulated cell death driven by lipid peroxidation [doll-2019-fsp1-glutathione-abstract][bersuker-2019-coq-oxidoreductase-abstract].

The primary enzymatic function of FSP1 is to catalyze the NAD(P)H-dependent reduction of ubiquinone (coenzyme Q10, CoQ10) to ubiquinol (CoQ10H2), a lipophilic radical-trapping antioxidant that halts the propagation of lipid peroxides [bersuker-2019-coq-oxidoreductase-abstract][doll-2019-fsp1-glutathione-abstract]. Additionally, FSP1 reduces vitamin K to its hydroquinone form, providing another mechanism of ferroptosis suppression through a non-canonical vitamin K cycle [mishima-2022-vitamin-k-abstract]. The protein localizes primarily to the plasma membrane through N-terminal myristoylation, where it protects membrane lipids from peroxidation [bersuker-2019-coq-oxidoreductase-abstract]. The discovery of FSP1 as a parallel ferroptosis-suppressing system, independent of the canonical glutathione peroxidase 4 (GPX4) pathway, fundamentally changed the understanding of ferroptosis regulation and opened new therapeutic avenues for cancer treatment.

2. Discovery and Nomenclature History

The protein now known as FSP1 was independently discovered by multiple research groups under different names, reflecting its apparent involvement in various cellular processes. In 2002, Wu and colleagues at Peking University identified a novel AIF-homologous protein they named AMID (AIF-homologous Mitochondrion-associated Inducer of Death) [wu-2002-amid-abstract]. Their work demonstrated that AMID co-localized with mitochondria and, when overexpressed, induced caspase-independent and p53-independent apoptosis that was not inhibited by Bcl-2 [wu-2002-amid-abstract]. Concurrently, Ohiro et al. identified the same gene as PRG3 (p53-Responsive Gene 3), noting that it was induced specifically under p53-dependent apoptotic conditions in human colon cancer cells [ohiro-2002-prg3-abstract]. Interestingly, these early studies reported that an N-terminal deletion mutant lacking the putative oxidoreductase domain retained apoptotic activity, suggesting the enzymatic function was dispensable for the apoptotic pathway [ohiro-2002-prg3-abstract].

The paradigm shifted dramatically in October 2019 when two landmark papers published simultaneously in Nature revealed the protein's primary physiological function. Using a synthetic lethal CRISPR-Cas9 screen, Bersuker et al. at UC Berkeley identified AIFM2 as a potent ferroptosis resistance factor and renamed it Ferroptosis Suppressor Protein 1 (FSP1) [bersuker-2019-coq-oxidoreductase-abstract]. In parallel, Doll et al. from the Conrad laboratory at the Helmholtz Institute used an expression cloning approach to identify genes that could complement GPX4 loss and independently arrived at the same conclusion [doll-2019-fsp1-glutathione-abstract]. These studies established that FSP1's primary cellular function is to suppress lipid peroxidation and ferroptosis, rather than to promote apoptosis as originally proposed [doll-2019-fsp1-glutathione-abstract][bersuker-2019-coq-oxidoreductase-abstract].

3. Enzymatic Function and Catalytic Mechanism

FSP1 functions as a NAD(P)H-dependent quinone oxidoreductase, catalyzing the two-electron reduction of ubiquinone (CoQ10) to ubiquinol (CoQ10H2) [doll-2019-fsp1-glutathione-abstract][bersuker-2019-coq-oxidoreductase-abstract]. The reduced ubiquinol acts as a lipophilic radical-trapping antioxidant (RTA) that neutralizes lipid peroxyl radicals, thereby preventing the propagation of lipid peroxidation that would otherwise lead to ferroptosis [bersuker-2019-coq-oxidoreductase-abstract]. This reaction can be summarized as:

CoQ10 (oxidized) + NAD(P)H + H+ → CoQ10H2 (reduced) + NAD(P)+

Structural insights into FSP1 catalysis came from crystal structures of Gallus gallus FSP1 solved at 2.0 Å (substrate-free) and 2.6 Å (ubiquinone-bound) resolution [lv-2023-structural-insights-abstract]. The structure reveals three distinct domains: a FAD-binding domain, a NAD(P)H-binding domain (both shared with AIF and NADH oxidoreductases), and a characteristic carboxy-terminal domain [lv-2023-structural-insights-abstract]. The carboxy-terminal domain mediates functional dimerization of FSP1, forming a homodimer that is required for the assembly of FAD- and ubiquinone-binding pockets and for catalytic activity [lv-2023-structural-insights-abstract].

A unique feature of FSP1 structure is the ~25-degree rotation of the FAD isoalloxazine ring compared to all other known FAD-dependent oxidoreductases [lv-2023-structural-insights-abstract]. The homodimer creates two active sites located on opposite sides of the FAD cofactor: one responsible for ubiquinone reduction and another for a unique FAD hydroxylation reaction that generates 6-hydroxy-FAD [lv-2023-structural-insights-abstract]. This 6-hydroxy-FAD was identified as both an active cofactor of FSP1 and a potent anti-ferroptotic compound in its own right, adding another layer to FSP1's ferroptosis-suppressing mechanisms [lv-2023-structural-insights-abstract].

Beyond ubiquinone, FSP1 also efficiently reduces vitamin K (both menaquinone and phylloquinone) to their hydroquinone forms [mishima-2022-vitamin-k-abstract]. This vitamin K reductase activity is particularly significant because it explains the warfarin-resistant vitamin K reduction pathway: FSP1 mediates vitamin K reduction independently of the canonical VKORC1-mediated pathway that is inhibited by warfarin [mishima-2022-vitamin-k-abstract]. The reduced vitamin K hydroquinone (VKH2) functions as another potent radical-trapping antioxidant that prevents lipid peroxidation, representing a third anti-ferroptotic mechanism mediated by FSP1 [mishima-2022-vitamin-k-abstract]. FSP1 can also reduce vitamin E (α-tocopherol), either directly or indirectly through ubiquinol-mediated regeneration, further expanding its antioxidant capacity [mishima-2022-vitamin-k-abstract].

Substrate Specificity and Kinetic Properties

FSP1 demonstrates broad substrate specificity within the quinone class of molecules. The known substrates include NADH and NADPH as electron donors, and CoQ10 (ubiquinone), CoQ1 (a short-chain ubiquinone analog), menadione (vitamin K3), and resazurin as electron acceptors [nakamura-2023-integrated-screens-abstract]. Kinetic analyses have been performed using resazurin reduction as a proxy for enzymatic activity, with reactions typically containing 100 nM recombinant human FSP1 protein [dasilva-2023-ifsp1-inhibition-abstract]. Lineweaver-Burk and Dixon plot analyses revealed that FSP1 inhibitors such as iFSP1 and viFSP1 are non-competitive inhibitors, meaning they can bind the enzyme regardless of substrate presence [nakamura-2023-integrated-screens-abstract].

Integrated chemical and genetic screens have provided detailed insights into FSP1's catalytic mechanism [nakamura-2023-integrated-screens-abstract]. Random mutagenesis studies identified conserved binding sites for the FAD and NAD(P)H cofactors and characterized critical proton-transfer mechanisms essential for catalysis. The development of two distinct classes of FSP1 inhibitors—iFSP1 targeting the quinone-binding pocket and viFSP1 targeting the NAD(P)H-binding pocket—confirmed the presence of two separate functional sites [nakamura-2023-integrated-screens-abstract]. Notably, iFSP1 is human-specific and does not effectively inhibit murine FSP1 due to species differences in an aromatic pocket, with F360 of human FSP1 being essential for iFSP1 binding [dasilva-2023-ifsp1-inhibition-abstract]. This species specificity has important implications for preclinical studies in mouse models.

4. Subcellular Localization and Membrane Association

The subcellular localization of FSP1 is critical for its anti-ferroptotic function. The protein contains a canonical N-terminal myristoylation motif, and this post-translational modification is essential for targeting FSP1 to cellular membranes [bersuker-2019-coq-oxidoreductase-abstract]. N-myristoylation involves the attachment of myristate, a 14-carbon saturated fatty acid, to the N-terminal glycine residue by N-myristoyltransferase (NMT), mediating dynamic protein-membrane interactions [bersuker-2019-coq-oxidoreductase-abstract].

Myristoylated FSP1 is primarily recruited to the plasma membrane, where it functions to reduce ubiquinone to ubiquinol at the site where lipid peroxidation most threatens cellular integrity [bersuker-2019-coq-oxidoreductase-abstract]. This membrane localization is absolutely required for ferroptosis suppression: the G2A mutant, which cannot be myristoylated, shows strongly affected localization and completely loses its ferroptosis-suppressive function [bersuker-2019-coq-oxidoreductase-abstract]. Importantly, fusion of a membrane-targeting sequence (Lyn11) to the G2A mutant restores its ability to suppress ferroptosis, confirming that membrane localization, rather than myristoylation per se, is the critical requirement [bersuker-2019-coq-oxidoreductase-abstract].

Beyond the plasma membrane, FSP1 also localizes to the periphery of lipid droplets (LDs) [nakamura-2023-phase-separation-abstract]. Recent work has revealed that LD-localized FSP1 plays a critical role in preventing neutral lipid peroxidation, a previously unrecognized function. By recycling CoQ10 to its antioxidant form at LDs, FSP1 prevents the accumulation of oxidized triacylglycerols and cholesteryl esters, thereby maintaining lipid droplet integrity and preventing LD-initiated ferroptosis when cells are loaded with polyunsaturated fatty acid-rich lipids.

The early characterization of AMID/AIFM2 suggested mitochondrial localization [wu-2002-amid-abstract], and indeed the protein can associate with the outer mitochondrial membrane despite lacking a canonical mitochondrial localization sequence [wu-2002-amid-abstract]. However, the predominant and functionally relevant localization for ferroptosis suppression is the plasma membrane. Interestingly, inhibition of myristoylation with compounds like IMP-366 not only abolishes the anti-ferroptosis function of FSP1 but also promotes its nuclear translocation via importin α2, suggesting complex regulation of FSP1 localization.

5. Role in Ferroptosis Suppression: Three Parallel Mechanisms

Three distinct mechanisms by which FSP1 suppresses ferroptosis have been elucidated, representing a comprehensive defense system against lipid peroxidation.

The FSP1-CoQ10-NAD(P)H Pathway

The primary and best-characterized mechanism involves FSP1's oxidoreductase activity at the plasma membrane [doll-2019-fsp1-glutathione-abstract][bersuker-2019-coq-oxidoreductase-abstract]. FSP1 catalyzes the reduction of ubiquinone (CoQ10) to ubiquinol (CoQ10H2) using NAD(P)H as the electron donor. The resulting ubiquinol functions as a lipophilic radical-trapping antioxidant that directly scavenges lipid peroxyl radicals, breaking the chain reaction of lipid peroxidation [doll-2019-fsp1-glutathione-abstract]. This pathway operates entirely independently of the canonical GPX4/glutathione system, providing cells with a parallel defense mechanism. The physiological importance of this redundancy is evident in the observation that cancer cells with high FSP1 expression are resistant to GPX4 inhibitors [bersuker-2019-coq-oxidoreductase-abstract].

The FSP1-Vitamin K-NAD(P)H Pathway

The second mechanism involves FSP1's vitamin K reductase activity [mishima-2022-vitamin-k-abstract]. FSP1 efficiently reduces vitamin K (menaquinone and phylloquinone) to their corresponding hydroquinones using NAD(P)H. These vitamin K hydroquinones function as potent radical-trapping antioxidants, similar to ubiquinol, preventing lipid peroxidation [mishima-2022-vitamin-k-abstract]. This pathway is warfarin-resistant, as it operates independently of the VKORC1 enzyme that is the target of warfarin [mishima-2022-vitamin-k-abstract]. The FSP1-vitamin K pathway represents a non-canonical vitamin K cycle with clear anti-ferroptotic function beyond vitamin K's classical role in blood coagulation [mishima-2022-vitamin-k-abstract].

The FSP1-ESCRT-III Membrane Repair Pathway

A third mechanism involves FSP1's role in promoting ESCRT-III-dependent membrane repair. The ESCRT-III (Endosomal Sorting Complexes Required for Transport-III) machinery mediates membrane deformation and fission, and plays a role in repairing damaged plasma membranes. FSP1 has been shown to potentiate the recruitment of ESCRT-III subunits CHMP5 and CHMP6 to the plasma membrane following ferroptosis-inducing stress. Knockdown of FSP1 suppresses the membrane accumulation of these ESCRT-III components, while overexpression of CHMP5 can rescue cells from ferroptosis even when FSP1 is silenced. This pathway operates independently of CoQ10, suggesting that FSP1 has functions beyond its enzymatic activity.

6. Relationship to GPX4 and the Ferroptosis Defense Network

Prior to the discovery of FSP1's anti-ferroptotic function, GPX4 (glutathione peroxidase 4) was considered the sole master regulator of ferroptosis. GPX4 reduces lipid hydroperoxides to lipid alcohols using glutathione (GSH) as a cofactor, representing the canonical ferroptosis defense system [doll-2019-fsp1-glutathione-abstract]. The identification of FSP1 revealed that cells possess parallel, independent mechanisms to prevent ferroptosis.

The FSP1-CoQ10-NAD(P)H pathway operates as a stand-alone parallel system that cooperates with the GPX4-glutathione axis [doll-2019-fsp1-glutathione-abstract]. These two systems provide complementary protection: GPX4 directly reduces lipid hydroperoxides that have already formed, while FSP1 generates ubiquinol that scavenges lipid peroxyl radicals before they can propagate [doll-2019-fsp1-glutathione-abstract][bersuker-2019-coq-oxidoreductase-abstract]. The existence of this backup system explains why some cancer cells survive GPX4 inhibition—they upregulate or rely on FSP1-mediated protection [bersuker-2019-coq-oxidoreductase-abstract].

A third component of the ferroptosis defense network is dihydroorotate dehydrogenase (DHODH), a flavin-dependent mitochondrial enzyme that, in addition to its canonical role in pyrimidine biosynthesis, regenerates ubiquinol in the inner mitochondrial membrane. While FSP1 protects the plasma membrane and other extramitochondrial membranes, DHODH specifically protects the mitochondrial compartment. GPX4 provides protection across multiple compartments including the cytoplasm and mitochondrial matrix. This compartmentalization allows cells to mount context-specific responses depending on the source of oxidative stress.

7. Regulation of FSP1 Expression

FSP1 expression is regulated at the transcriptional level by several pathways, most notably the KEAP1-NRF2 antioxidant response pathway. Nuclear factor erythroid 2-related factor 2 (NRF2) is a master transcription factor that controls the expression of numerous antioxidant and cytoprotective genes. Under normal conditions, NRF2 is targeted for degradation by KEAP1 (Kelch-like ECH-associated protein 1); however, under oxidative stress or when KEAP1 is mutated (as frequently occurs in lung cancers), NRF2 accumulates and activates its target genes [koppula-2022-coq-fsp1-abstract][emmanuel-2022-nrf2-target-abstract].

Analysis of NRF2 ChIP-seq data revealed strong association between NRF2 and the FSP1 gene promoter [emmanuel-2022-nrf2-target-abstract]. Two antioxidant response elements (AREs), which are canonical NRF2 binding motifs, have been identified in the FSP1 promoter region, and ChIP-qPCR assays confirmed increased NRF2 binding to these AREs upon KEAP1 knockout [emmanuel-2022-nrf2-target-abstract]. Consequently, KEAP1 mutations, which are common in non-small cell lung cancer (NSCLC), lead to constitutive NRF2 activation and elevated FSP1 expression, contributing to ferroptosis and radiation resistance [koppula-2022-coq-fsp1-abstract].

The original identification of PRG3 (AIFM2) as a p53-responsive gene indicates that p53 can also regulate FSP1 expression [ohiro-2002-prg3-abstract]. A p53-responsive element was identified in the PRG3 gene, and expression was induced by activation of endogenous p53 [ohiro-2002-prg3-abstract]. This adds complexity to FSP1 regulation in cancer contexts where p53 is frequently mutated.

Myristoylation of FSP1 is also subject to regulation. N-myristoyltransferase 2 (NMT2) is the enzyme responsible for FSP1 myristoylation, and NADPH has been shown to interact with NMT2 and upregulate N-myristoylated FSP1 levels. This creates a connection between cellular NADPH levels and FSP1 membrane localization, linking metabolic state to ferroptosis sensitivity.

8. Therapeutic Implications and Cancer Relevance

The discovery that FSP1 mediates ferroptosis resistance has profound implications for cancer therapy. FSP1 expression levels correlate with ferroptosis resistance across hundreds of cancer cell lines, and the protein is highly expressed in various tumors [bersuker-2019-coq-oxidoreductase-abstract]. In lung cancer particularly, FSP1 expression is elevated in tumors with KEAP1 mutations, contributing to both chemotherapy and radiation resistance [koppula-2022-coq-fsp1-abstract].

The first FSP1-specific inhibitor, iFSP1, was identified through counter-screening approaches and inhibits human FSP1 with an EC50 of approximately 103 nM [bersuker-2019-coq-oxidoreductase-abstract]. In preclinical models, iFSP1 effectively induced ferroptosis in hepatocellular carcinoma (HCC) cells and suppressed tumor growth while also enhancing anti-tumor immune responses by increasing immune infiltration of dendritic cells, macrophages, and T cells. The combination of FSP1 inhibition with immune checkpoint inhibitors shows synergistic effects.

A novel class of FSP1 inhibitors, the 3-phenylquinazolinones (exemplified by icFSP1), was identified through small molecule library screening [nakamura-2023-phase-separation-abstract]. Unlike iFSP1, which competitively inhibits FSP1 enzymatic activity, icFSP1 promotes FSP1 phase separation and condensate formation, displacing FSP1 from membranes and thereby preventing its anti-ferroptotic function [nakamura-2023-phase-separation-abstract]. This mechanism requires FSP1's N-terminal myristoylation and intrinsically disordered regions. Importantly, icFSP1 impairs tumor growth in vivo and induces FSP1 condensates within tumors, representing a novel therapeutic strategy [nakamura-2023-phase-separation-abstract].

The combination of FSP1 inhibitors with GPX4 inhibitors shows synergistic effects in inducing ferroptosis, as cancer cells must then rely on neither defense pathway [doll-2019-fsp1-glutathione-abstract]. Similarly, combined inhibition of FSP1 and DHODH synergistically promotes ferroptosis by disabling both plasma membrane and mitochondrial CoQ10 reduction. These combination strategies are particularly relevant for cancers with KEAP1 mutations or high basal ferroptosis resistance.

Recent in vivo studies using genetically engineered mouse models have provided compelling evidence for FSP1's role in tumor biology. In lung adenocarcinoma models, tumor-specific loss-of-function of FSP1 resulted in increased lipid peroxidation and robust suppression of tumorigenesis, confirming that lung tumors are highly sensitive to ferroptosis. Importantly, across multiple preclinical models, FSP1 was required for ferroptosis protection in vivo but not in vitro, underscoring a heightened need to buffer lipid peroxidation under physiological conditions. High FSP1 expression is linked to worse prognosis in human lung adenocarcinoma, supporting its potential as both a therapeutic target and prognostic biomarker.

9. FSP1 in Neurodegeneration

Beyond cancer, ferroptosis has emerged as a significant contributor to neurodegenerative diseases, and FSP1's role as a ferroptosis suppressor has implications for understanding neurological pathology. The brain presents a unique environment for ferroptosis regulation due to its high oxygen consumption, abundant polyunsaturated fatty acids, and robust iron utilization for neurotransmitter synthesis and myelin formation, making the central nervous system especially vulnerable to ferroptotic damage.

Ferroptosis has been associated with the pathogenesis of Alzheimer's disease (AD), Parkinson's disease (PD), Huntington's disease, and amyotrophic lateral sclerosis. In AD, iron dysregulation and lipid peroxidation predominate in cortical and hippocampal regions, affecting both neurons and glial cells. In PD, the substantia nigra pars compacta (SNpc), where dopaminergic neurons are concentrated, shows particular susceptibility to ferroptotic damage. Regional differences in the expression of ferroptosis regulators, including FSP1, GPX4, and system Xc−, likely underlie these distinct regional vulnerabilities. Notably, diseases associated with AIFM2/FSP1 in curated databases include ovarian cancer and Parkinson's disease.

The regulation of FSP1 expression by NRF2 is particularly relevant in the neurological context, as NRF2 activation has emerged as a potential neuroprotective strategy. Since FSP1 is a target gene of NRF2, compounds that activate NRF2 signaling could potentially upregulate FSP1 expression and enhance protection against neuronal ferroptosis. This presents opportunities for therapeutic intervention in neurodegenerative conditions where ferroptosis contributes to disease progression.

10. Open Questions

Despite significant advances in understanding FSP1 biology, several important questions remain:

Structural and mechanistic questions: The crystal structures solved to date are from chicken FSP1, and human FSP1 structures would provide more direct insights into inhibitor design and species-specific differences in iFSP1 sensitivity. The mechanism by which FSP1 promotes ESCRT-III-dependent membrane repair independently of its enzymatic activity is not fully understood.

Physiological roles beyond ferroptosis: The original characterization of AMID/PRG3 as a pro-apoptotic protein raises questions about context-dependent functions. Under what conditions, if any, does FSP1 contribute to cell death rather than survival? The observation that oxidoreductase-dead mutants retained apoptotic activity [ohiro-2002-prg3-abstract] suggests functions independent of enzymatic activity that merit further investigation.

Lipid droplet function: The recent discovery of FSP1's role in protecting lipid droplets from neutral lipid peroxidation opens new avenues of investigation. How does FSP1 localization to lipid droplets versus plasma membrane affect its function, and how is this distribution regulated?

Phase separation biology: The discovery that FSP1 can undergo phase separation when displaced from membranes raises questions about whether this represents purely a loss-of-function phenotype or whether FSP1 condensates have biological functions [nakamura-2023-phase-separation-abstract].

Therapeutic development: The first-generation FSP1 inhibitor iFSP1 has limitations including species specificity (poor activity against mouse FSP1) and unfavorable pharmacological properties. Development of improved inhibitors suitable for clinical translation remains an active area of research. The identification of species-independent inhibitors (viFSP1) targeting the NAD(P)H-binding pocket represents progress in this direction.

Biomarker potential: Whether FSP1 expression levels could serve as predictive biomarkers for ferroptosis-inducing therapies remains to be validated in clinical settings. The correlation between FSP1 expression and ferroptosis resistance in cell lines needs confirmation in patient tumors.

Immune implications: Preliminary data suggest that FSP1 inhibition enhances anti-tumor immune responses, but the mechanisms linking ferroptosis to immunogenic cell death and immune activation require further investigation.

11. References

1. [doll-2019-fsp1-glutathione-abstract] Doll S, Freitas FP, Shah R, Aldrovandi M, et al. FSP1 is a glutathione-independent ferroptosis suppressor. Nature. 2019;575(7784):693-698. PMID: 31634899. DOI: 10.1038/s41586-019-1707-0

2. [bersuker-2019-coq-oxidoreductase-abstract] Bersuker K, Hendricks JM, Li Z, Magtanong L, Ford B, Tang PH, Roberts MA, Tong B, Maimone TJ, Zoncu R, Bassik MC, Nomura DK, Dixon SJ, Olzmann JA. The CoQ oxidoreductase FSP1 acts parallel to GPX4 to inhibit ferroptosis. Nature. 2019;575(7784):688-692. PMID: 31634900. PMCID: PMC6883167. DOI: 10.1038/s41586-019-1705-2

3. [mishima-2022-vitamin-k-abstract] Mishima E, Ito J, Wu Z, Nakamura T, Wahida A, Doll S, et al. A non-canonical vitamin K cycle is a potent ferroptosis suppressor. Nature. 2022;608(7924):778-783. PMID: 35922516. DOI: 10.1038/s41586-022-05022-3

4. [lv-2023-structural-insights-abstract] Lv Y, Liang C, Sun Q, Zhu J, Xu H, Li X, Li YY, Wang Q, Yuan H, Chu B, Zhu D. Structural insights into FSP1 catalysis and ferroptosis inhibition. Nature Communications. 2023;14:5933. PMID: 37739943. DOI: 10.1038/s41467-023-41626-7

5. [wu-2002-amid-abstract] Wu M, Xu LG, Li X, Zhai Z, Shu HB. AMID, an apoptosis-inducing factor-homologous mitochondrion-associated protein, induces caspase-independent apoptosis. Journal of Biological Chemistry. 2002;277(28):25617-25623. PMID: 11980907. DOI: 10.1074/jbc.M202285200

6. [ohiro-2002-prg3-abstract] Ohiro Y, Garkavtsev I, Kobayashi S, Sreekumar KR, Nantz R, Higashikubo BT, Duffy SL, Higashikubo R, Usheva A, Gius D, Kley N, Horikoshi N. A novel p53-inducible apoptogenic gene, PRG3, encodes a homologue of the apoptosis-inducing factor (AIF). FEBS Letters. 2002;524(1-3):163-171. PMID: 12135761. DOI: 10.1016/s0014-5793(02)03049-1

7. [nakamura-2023-phase-separation-abstract] Nakamura T, Hipp C, Mourão ASD, Borggräfe J, Aldrovandi M, Henkelmann B, Wanninger J, Mishima E, Lytton E, Emler D, Proneth B, Sattler M, Conrad M. Phase separation of FSP1 promotes ferroptosis. Nature. 2023;619(7969):371-377. PMID: 37380771. DOI: 10.1038/s41586-023-06255-6

8. [koppula-2022-coq-fsp1-abstract] Koppula P, Lei G, Zhang Y, Yan Y, Mao C, Kondiparthi L, Shi J, Liu X, Horbath A, Das M, Li W, Poyurovsky MV, Olszewski K, Gan B. A targetable CoQ-FSP1 axis drives ferroptosis- and radiation-resistance in KEAP1 inactive lung cancers. Nature Communications. 2022;13(1):2206. PMID: 35459868. DOI: 10.1038/s41467-022-29905-1

9. [emmanuel-2022-nrf2-target-abstract] Emmanuel N, Li H, Chen J, Zhang Y. FSP1, a novel KEAP1/NRF2 target gene regulating ferroptosis and radioresistance in lung cancers. Oncotarget. 2022;13:1136-1139. PMID: 36264074. DOI: 10.18632/oncotarget.28301

10. [nakamura-2023-integrated-screens-abstract] Nakamura T, Mishima E, Yamada N, Mourão ASD, Trümbach D, Doll S, Wanninger J, Lytton E, Sennhenn P, Xavier da Silva TN, Friedmann Angeli JP, Sattler M, Proneth B, Conrad M. Integrated chemical and genetic screens unveil FSP1 mechanisms of ferroptosis regulation. Nature Structural & Molecular Biology. 2023;30(11):1806-1815. PMID: 37957306. PMCID: PMC10643123. DOI: 10.1038/s41594-023-01136-y

11. [dasilva-2023-ifsp1-inhibition-abstract] Xavier da Silva TN, Schulte C, Alves AN, Maric HM, Friedmann Angeli JP. Molecular characterization of AIFM2/FSP1 inhibition by iFSP1-like molecules. Cell Death & Disease. 2023;14(4):281. PMID: 37080964. DOI: 10.1038/s41419-023-05787-z

Citations

1. bersuker-2019-coq-oxidoreductase-abstract.md
2. bersuker-2019-coq-oxidoreductase-summary.md
3. dasilva-2023-ifsp1-inhibition-abstract.md
4. doll-2019-fsp1-glutathione-abstract.md
5. doll-2019-fsp1-glutathione-summary.md
6. emmanuel-2022-nrf2-target-abstract.md
7. koppula-2022-coq-fsp1-abstract.md
8. lv-2023-structural-insights-abstract.md
9. lv-2023-structural-insights-summary.md
10. mishima-2022-vitamin-k-abstract.md
11. mishima-2022-vitamin-k-summary.md
12. nakamura-2023-integrated-screens-abstract.md
13. nakamura-2023-phase-separation-abstract.md
14. nakamura-2023-phase-separation-summary.md
15. ohiro-2002-prg3-abstract.md
16. wu-2002-amid-abstract.md
17. wu-2002-amid-summary.md

provider: falcon
model: Edison Scientific Literature
cached: true
start_time: '2026-01-19T15:33:54.871301'
end_time: '2026-01-19T15:33:54.873380'
duration_seconds: 0.0
template_file: templates/gene_research_go_focused.md
template_variables:
organism: human
gene_id: AIFM2
gene_symbol: AIFM2
uniprot_accession: Q9BRQ8
protein_description: 'RecName: Full=Ferroptosis suppressor protein 1 {ECO:0000303|PubMed:31634899,
ECO:0000303|PubMed:31634900}; Short=FSP1 {ECO:0000303|PubMed:31634899, ECO:0000303|PubMed:31634900};
EC=1.6.5.- {ECO:0000269|PubMed:31634899, ECO:0000269|PubMed:31634900, ECO:0000269|PubMed:35922516};
AltName: Full=Apoptosis-inducing factor homologous mitochondrion-associated inducer
of death {ECO:0000303|PubMed:11980907}; Short=AMID {ECO:0000303|PubMed:11980907};
AltName: Full=p53-responsive gene 3 protein {ECO:0000303|PubMed:12135761};'
gene_info: Name=AIFM2 {ECO:0000303|PubMed:26689472, ECO:0000312|HGNC:HGNC:21411};
Synonyms=AMID {ECO:0000303|PubMed:11980907}, PRG3 {ECO:0000303|PubMed:12135761};
organism_full: Homo sapiens (Human).
protein_family: Belongs to the FAD-dependent oxidoreductase family.
protein_domains: FAD/NAD-bd_sf. (IPR036188); FAD/NAD-binding_dom. (IPR023753); Pyr_redox_2
(PF07992)
provider_config:
timeout: 600
max_retries: 3
parameters:
allowed_domains: []
temperature: 0.1
citation_count: 24

Question

Gene Research for Functional Annotation

⚠️ CRITICAL: Gene/Protein Identification Context

BEFORE YOU BEGIN RESEARCH: You MUST verify you are researching the CORRECT gene/protein. Gene symbols can be ambiguous, especially for less well-characterized genes from non-model organisms.

Target Gene/Protein Identity (from UniProt):

• UniProt Accession: Q9BRQ8
• Protein Description: RecName: Full=Ferroptosis suppressor protein 1 {ECO:0000303|PubMed:31634899, ECO:0000303|PubMed:31634900}; Short=FSP1 {ECO:0000303|PubMed:31634899, ECO:0000303|PubMed:31634900}; EC=1.6.5.- {ECO:0000269|PubMed:31634899, ECO:0000269|PubMed:31634900, ECO:0000269|PubMed:35922516}; AltName: Full=Apoptosis-inducing factor homologous mitochondrion-associated inducer of death {ECO:0000303|PubMed:11980907}; Short=AMID {ECO:0000303|PubMed:11980907}; AltName: Full=p53-responsive gene 3 protein {ECO:0000303|PubMed:12135761};
• Gene Information: Name=AIFM2 {ECO:0000303|PubMed:26689472, ECO:0000312|HGNC:HGNC:21411}; Synonyms=AMID {ECO:0000303|PubMed:11980907}, PRG3 {ECO:0000303|PubMed:12135761};
• Organism (full): Homo sapiens (Human).
• Protein Family: Belongs to the FAD-dependent oxidoreductase family.
• Key Domains: FAD/NAD-bd_sf. (IPR036188); FAD/NAD-binding_dom. (IPR023753); Pyr_redox_2 (PF07992)

MANDATORY VERIFICATION STEPS:

1. Check if the gene symbol "AIFM2" matches the protein description above
2. Verify the organism is correct: Homo sapiens (Human).
3. Check if protein family/domains align with what you find in literature
4. If you find literature for a DIFFERENT gene with the same or similar symbol, STOP

If Gene Symbol is Ambiguous or You Cannot Find Relevant Literature:

DO NOT PROCEED WITH RESEARCH ON A DIFFERENT GENE. Instead:
- State clearly: "The gene symbol 'AIFM2' is ambiguous or literature is limited for this specific protein"
- Explain what you found (e.g., "Found extensive literature on a different gene with the same symbol in a different organism")
- Describe the protein based ONLY on the UniProt information provided above
- Suggest that the protein function can be inferred from domain/family information

Research Target:

Please provide a comprehensive research report on the gene AIFM2 (gene ID: AIFM2, UniProt: Q9BRQ8) in human.

The research report should be a detailed narrative explaining the function, biological processes, and localization of the gene product. Citations should be given for all claims.

You should prioritize authoritative reviews and primary scientific literature when conducting research. You can supplement
this with annotations you find in gene/protein databases, but these can be outdated or inaccurate.

We are specifically interested in the primary function of the gene - for enzymes, what reaction is catalyzed, and what is the substrate specificity? For transporters, what is the substrate? For structural proteins or adapters, what is the broader structural role? For signaling molecules, what is the role in the pathway.

We are interested in where in or outside the cell the gene product carries out its function.

We are also interested in the signaling or biochemical pathways in which the gene functions. We are less interested in broad pleiotropic effects, except where these elucidate the precise role.

Include evidence where possible. We are interested in both experimental evidence as well as inference from structure, evolution, or bioinformatic analysis. Precise studies should be prioritized over high-throughput, where available.

Output

Question: You are an expert researcher providing comprehensive, well-cited information.

Provide detailed information focusing on:
1. Key concepts and definitions with current understanding
2. Recent developments and latest research (prioritize 2023-2024 sources)
3. Current applications and real-world implementations
4. Expert opinions and analysis from authoritative sources
5. Relevant statistics and data from recent studies

Format as a comprehensive research report with proper citations. Include URLs and publication dates where available.
Always prioritize recent, authoritative sources and provide specific citations for all major claims.

Gene Research for Functional Annotation

⚠️ CRITICAL: Gene/Protein Identification Context

BEFORE YOU BEGIN RESEARCH: You MUST verify you are researching the CORRECT gene/protein. Gene symbols can be ambiguous, especially for less well-characterized genes from non-model organisms.

Target Gene/Protein Identity (from UniProt):

• UniProt Accession: Q9BRQ8
• Protein Description: RecName: Full=Ferroptosis suppressor protein 1 {ECO:0000303|PubMed:31634899, ECO:0000303|PubMed:31634900}; Short=FSP1 {ECO:0000303|PubMed:31634899, ECO:0000303|PubMed:31634900}; EC=1.6.5.- {ECO:0000269|PubMed:31634899, ECO:0000269|PubMed:31634900, ECO:0000269|PubMed:35922516}; AltName: Full=Apoptosis-inducing factor homologous mitochondrion-associated inducer of death {ECO:0000303|PubMed:11980907}; Short=AMID {ECO:0000303|PubMed:11980907}; AltName: Full=p53-responsive gene 3 protein {ECO:0000303|PubMed:12135761};
• Gene Information: Name=AIFM2 {ECO:0000303|PubMed:26689472, ECO:0000312|HGNC:HGNC:21411}; Synonyms=AMID {ECO:0000303|PubMed:11980907}, PRG3 {ECO:0000303|PubMed:12135761};
• Organism (full): Homo sapiens (Human).
• Protein Family: Belongs to the FAD-dependent oxidoreductase family.
• Key Domains: FAD/NAD-bd_sf. (IPR036188); FAD/NAD-binding_dom. (IPR023753); Pyr_redox_2 (PF07992)

MANDATORY VERIFICATION STEPS:

1. Check if the gene symbol "AIFM2" matches the protein description above
2. Verify the organism is correct: Homo sapiens (Human).
3. Check if protein family/domains align with what you find in literature
4. If you find literature for a DIFFERENT gene with the same or similar symbol, STOP

If Gene Symbol is Ambiguous or You Cannot Find Relevant Literature:

DO NOT PROCEED WITH RESEARCH ON A DIFFERENT GENE. Instead:
- State clearly: "The gene symbol 'AIFM2' is ambiguous or literature is limited for this specific protein"
- Explain what you found (e.g., "Found extensive literature on a different gene with the same symbol in a different organism")
- Describe the protein based ONLY on the UniProt information provided above
- Suggest that the protein function can be inferred from domain/family information

Research Target:

Please provide a comprehensive research report on the gene AIFM2 (gene ID: AIFM2, UniProt: Q9BRQ8) in human.

The research report should be a detailed narrative explaining the function, biological processes, and localization of the gene product. Citations should be given for all claims.

You should prioritize authoritative reviews and primary scientific literature when conducting research. You can supplement
this with annotations you find in gene/protein databases, but these can be outdated or inaccurate.

We are specifically interested in the primary function of the gene - for enzymes, what reaction is catalyzed, and what is the substrate specificity? For transporters, what is the substrate? For structural proteins or adapters, what is the broader structural role? For signaling molecules, what is the role in the pathway.

We are interested in where in or outside the cell the gene product carries out its function.

We are also interested in the signaling or biochemical pathways in which the gene functions. We are less interested in broad pleiotropic effects, except where these elucidate the precise role.

Include evidence where possible. We are interested in both experimental evidence as well as inference from structure, evolution, or bioinformatic analysis. Precise studies should be prioritized over high-throughput, where available.

Plan status: We verified the target gene/protein identity and synthesized recent primary and review evidence (2019–2024) on AIFM2/FSP1 (UniProt Q9BRQ8) to address function, localization, structure, regulation, pathway context, applications, expert perspectives, and quantitative findings.

Comprehensive research report: AIFM2/FSP1 (Human; UniProt Q9BRQ8)

Identity and definitions
- Verification: AIFM2 (aliases AMID/PRG3) encodes a human FAD-dependent oxidoreductase renamed ferroptosis suppressor protein 1 (FSP1) for its role in suppressing ferroptosis independently of GPX4. Organism: Homo sapiens. Protein family/domains: FAD/NAD(P)H-binding oxidoreductase fold consistent with FAD/NAD-binding domains; structural studies confirm this architecture (Nature Communications, Sep 2023, https://doi.org/10.1038/s41467-023-41626-7) (lv2023structuralinsightsinto pages 1-2, lv2023structuralinsightsinto pages 8-9). The discovery paper established FSP1 as a glutathione-independent ferroptosis suppressor (Nature, Oct 2019, https://doi.org/10.1038/s41586-019-1707-0) (doll2019fsp1isa pages 1-5, doll2019fsp1isa pages 5-9).

Key concepts and current understanding
- Ferroptosis: iron-dependent cell death driven by phospholipid peroxidation of PUFA-containing membranes. Cells deploy parallel antioxidant systems: GPX4/GSH and a GPX4-independent FSP1–CoQ system; expert reviews and guidelines position FSP1 as a principal non-GPX4 defense (Nat Rev Mol Cell Biol, Feb 2024, https://doi.org/10.1038/s41580-024-00703-5; Nat Cell Biol guideline, Feb 2024, https://doi.org/10.1038/s41556-024-01360-8) (deol2025vitaminb2metabolism pages 30-35).

Primary biochemical function (enzyme, substrates, products)
- Reaction: NAD(P)H-dependent reduction of membrane-embedded ubiquinone (CoQ10) to ubiquinol (CoQ10H2), which acts as a radical-trapping antioxidant to terminate lipid peroxyl radicals, thereby suppressing ferroptosis (Nature, 2019) (doll2019fsp1isa pages 1-5, doll2019fsp1isa pages 5-9).
- Substrate specificity: FSP1 reduces ubiquinone analogs; dehydroascorbate, oxidized glutathione, and tert-butyl hydroperoxide are not substrates in tested assays (Nature, 2019) (doll2019fsp1isa pages 5-9).
- Vitamin K reduction: FSP1 can also reduce vitamin K to its hydroquinone (VKH2), establishing a non-canonical vitamin K redox cycle relevant to ferroptosis suppression; vitamin K species protect against GPX4-loss ferroptosis in cells and in vivo (summarized in 2022 review and 2023–2024 expert sources) (Molecular Biomedicine, Nov 2022, https://doi.org/10.1186/s43556-022-00105-z; Cell Metabolism, Aug 2023, https://doi.org/10.1016/j.cmet.2023.06.014; Nat Rev Mol Cell Biol, 2024) (zeng2022theantiferroptoticrole pages 1-3, deol2025vitaminb2metabolism pages 30-35).

Subcellular localization and determinants
- N-myristoylation at the N-terminus is essential for anti-ferroptotic function and membrane targeting; mutation (G2A) or N-myristoyltransferase inhibition (IMP-1088) abolishes protection (Nature, 2019) (doll2019fsp1isa pages 1-5, doll2019fsp1isa pages 5-9, doll2019fsp1isa pages 14-18).
- Localization: enriched at the plasma membrane, periphery of lipid droplets, and partially overlapping ER/Golgi; membrane association is needed for suppressing lipid peroxidation at the cell boundary (2019 primary; 2022 review) (doll2019fsp1isa pages 1-5, zeng2022theantiferroptoticrole pages 1-3).

Structural and mechanistic insights (2023 update)
- Domain architecture: distinct FAD-binding and NAD(P)H-binding domains plus a carboxy-terminal domain (CTD) unique to FSP1 (Nature Communications, Sep 2023) (lv2023structuralinsightsinto pages 1-2).
- Dimerization: CTD mediates functional homodimerization; dimer formation creates two active centers on opposite sides of FAD—one for ubiquinone reduction, the other for a unique FAD hydroxylation reaction (Nature Communications, Sep 2023) (lv2023structuralinsightsinto pages 1-2, lv2023structuralinsightsinto pages 8-9).
- 6‑hydroxy‑FAD: In vitro, FSP1 can generate H2O2 and convert bound FAD to 6‑hydroxy‑FAD under oxygen and NAD(P)H; 6‑hydroxy‑FAD acts as an active cofactor that directly inhibits ferroptosis in cells (Nature Communications, Sep 2023) (lv2023structuralinsightsinto pages 1-2, lv2023structuralinsightsinto pages 8-9).

Regulation of FSP1 abundance and activity
- Myristoylation: prerequisite for membrane localization and activity (Nature, 2019) (doll2019fsp1isa pages 1-5, doll2019fsp1isa pages 5-9, doll2019fsp1isa pages 14-18).
- Ubiquitination: RNF126 is an E3 ligase that ubiquitinates FSP1 at defined lysine clusters; RNF126 loss reduces FSP1 plasma-membrane localization and increases the CoQ/CoQH2 ratio, sensitizing to ferroptosis in medulloblastoma models (Oncogene, Mar 2024, https://doi.org/10.1038/s41388-024-02949-x) (xie2024rnf126mediatedubiquitinationof pages 1-4).
- Cofactor/stability coupling: FSP1 is a FAD-binding oxidoreductase; FAD availability and binding promote stability, while FAD-free FSP1 is targeted for ubiquitin–proteasome degradation (bioRxiv, Aug 2025, https://doi.org/10.1101/2025.08.05.668752) (deol2025vitaminb2metabolism pages 13-17). (Note: preprint; mechanistic alignment with FAD-dependent oxidoreductase biology.)

Pathway context and related processes
- GPX4-independent ferroptosis suppression: FSP1/CoQ10/NAD(P)H axis forms a parallel membrane antioxidant system; COQ2 loss diminishes FSP1 protection, supporting CoQ dependence (Nature, 2019) (doll2019fsp1isa pages 5-9, doll2019fsp1isa pages 1-5).
- Vitamin K axis in ferroptosis: p53–VKORC1L1 regulation links vitamin K metabolism to tumor suppression and ferroptosis responses; FSP1 provides vitamin K reduction to VKH2 as an anti-ferroptotic route (Cell Metabolism, Aug 2023, https://doi.org/10.1016/j.cmet.2023.06.014; Nat Cell Biol guideline, 2024) (deol2025vitaminb2metabolism pages 30-35).
- ESCRT-III membrane repair: FSP1 promotes ESCRT-III–dependent membrane repair (CHMP5/CHMP6), constituting an additional anti-ferroptotic mechanism beyond CoQ reduction (Molecular Biomedicine, Nov 2022, https://doi.org/10.1186/s43556-022-00105-z) (zeng2022theantiferroptoticrole pages 1-3).

Recent developments (prioritized 2023–2024)
- Structural mechanism (2023): dimerization-dependent dual active sites and 6‑hydroxy‑FAD formation; mechanistic promiscuity for CoQ and vitamin K reduction, clarifying catalytic basis for ferroptosis suppression (Nature Communications, Sep 2023, https://doi.org/10.1038/s41467-023-41626-7) (lv2023structuralinsightsinto pages 1-2, lv2023structuralinsightsinto pages 8-9).
- Ubiquitination control (2024): RNF126 directs FSP1 ubiquitination that maintains plasma membrane localization and anti-ferroptotic CoQH2 pools in pediatric brain tumors (Oncogene, Mar 2024, https://doi.org/10.1038/s41388-024-02949-x) (xie2024rnf126mediatedubiquitinationof pages 1-4).
- Vitamin K/p53 axis (2023–2024): VKORC1L1 upregulation by p53 supports vitamin K metabolism for ferroptosis control; non-canonical vitamin K cycle including FSP1 highlighted by reviews/guidelines (Cell Metabolism, Aug 2023; Nat Cell Biol, 2024) (deol2025vitaminb2metabolism pages 30-35).
- Drug repurposing (2024): Temsirolimus directly inhibits FSP1 enzymatic activity to induce ferroptosis and synergizes with GPX4 inhibition in liver cancer models (J Mol Cell Biol, Sep 2024, https://doi.org/10.1093/jmcb/mjae036) (tian2024temsirolimusinhibitsfsp1 pages 2-5).
- Inhibitor landscape and species selectivity (2024): First-generation inhibitor iFSP1 shows species-selective engagement; mechanistic characterization provides guidance for second-generation inhibitor design and for interpreting murine efficacy studies (da Silva et al., 2024) (silva20242.4manuscriptiva pages 104-108).

Current applications and real-world implementations
- Oncology combinations: In vitro and in vivo data show that inhibiting FSP1 (genetically or pharmacologically) can convert GPX4-resistant contexts to ferroptosis-sensitive, including synergy of temsirolimus with RSL3 in liver cancer models (J Mol Cell Biol, 2024) (tian2024temsirolimusinhibitsfsp1 pages 2-5).
- Prognostic/biomarker potential: Clinical pathology study in lung adenocarcinoma links GPX4/FSP1/4‑HNE expression patterns to prognosis; low FSP1 associates with unfavorable relapse-free survival, motivating ferroptosis-based strategies (Cancers, Nov 2024, https://doi.org/10.3390/cancers16223888) (zeng2022theantiferroptoticrole pages 3-3).
- Non-ferroptosis function affecting therapy: In BRCA-proficient ovarian cancer, FSP1 inhibition (iFSP1) enhances olaparib sensitivity through a non-ferroptotic mechanism by disrupting Ku70 PARylation and DNA-PKcs recruitment at double-strand breaks; combination shows synergy in organoids and xenografts (Cell Death Differentiation, Feb 2024, https://doi.org/10.1038/s41418-024-01263-z) (xie2024rnf126mediatedubiquitinationof pages 1-4).

Expert opinions and authoritative guidance
- Cell biology and therapeutic framing: 2024 reviews/guidelines emphasize FSP1 as a membrane oxidoreductase that regenerates CoQ and vitamin K hydroquinones to stop lipid peroxidation, operating alongside GPX4, and highlight the therapeutic promise of targeting FSP1 in cancer while considering systemic risks (Nat Rev Mol Cell Biol, Feb 2024; Nat Cell Biol guideline, Feb 2024) (deol2025vitaminb2metabolism pages 30-35).

Relevant quantitative/statistical data and phenotypes
- Dependency and resistance correlations: Across large cancer cell panels, higher FSP1 expression correlates with resistance to ferroptosis inducers, whereas low FSP1 correlates with GPX4 dependency; FSP1 knockout sensitizes cells to RSL3; combined GPX4+FSP1 deletion slows xenograft growth, whereas GPX4 deletion alone may not, depending on context (Nature, Oct 2019) (doll2019fsp1isa pages 5-9, doll2019fsp1isa pages 1-5).
- Clinical association: In lung adenocarcinoma cohorts, low FSP1 expression was associated with worse relapse-free survival, supporting biomarker development for ferroptosis-targeted strategies (Cancers, Nov 2024) (zeng2022theantiferroptoticrole pages 3-3).
- Pharmacology: Temsirolimus-induced ferroptosis in GPX4-defective liver cancer cells is FSP1-dependent; combination with RSL3 robustly increases lipid peroxidation and suppresses tumor progression in xenograft models (J Mol Cell Biol, Sep 2024) (tian2024temsirolimusinhibitsfsp1 pages 2-5).

Notes on localization/biophysics and catalysis
- Membrane proximity is integral to function; N-myristoylation positions FSP1 for reduction of membrane CoQ pools at the plasma membrane and possibly lipid droplets; inhibition or mutation of the myristoylation site abolishes protection (Nature, 2019; Molecular Biomedicine, 2022) (doll2019fsp1isa pages 1-5, zeng2022theantiferroptoticrole pages 1-3).
- Structural catalysis supports two-electron transfer from NAD(P)H via FAD to quinone substrates (CoQ, vitamin K) and a coupled oxygen-reduction chemistry that can yield 6‑hydroxy‑FAD, proposed to contribute to ferroptosis suppression (Nature Communications, 2023) (lv2023structuralinsightsinto pages 1-2, lv2023structuralinsightsinto pages 8-9).

Therapeutic targeting and tool compounds
- iFSP1/icFSP1: First-generation FSP1 inhibitors (e.g., iFSP1) sensitize to ferroptosis but show species-selective target engagement; use with caution in murine systems and prioritize human-targeted development (2024 mechanistic characterization) (silva20242.4manuscriptiva pages 104-108).
- Temsirolimus: FDA-approved mTOR inhibitor repurposed as a direct FSP1 enzyme inhibitor; demonstrates synergy with GPX4 inhibition and anti-tumor efficacy in liver cancer models (J Mol Cell Biol, 2024, https://doi.org/10.1093/jmcb/mjae036) (tian2024temsirolimusinhibitsfsp1 pages 2-5).
- Strategic combinations: Reviews/guidelines recommend dual targeting of GPX4 and FSP1 systems to overcome redundancy in ferroptosis defenses, with attention to tissue risks (Nat Rev Mol Cell Biol, 2024; Nat Cell Biol guideline, 2024) (deol2025vitaminb2metabolism pages 30-35).

Compliance with identification safeguards
- Symbol/organism verification: All cited mechanistic and structural work refers to human AIFM2/FSP1 or closely homologous orthologs; structural findings from Gallus gallus FSP1 are applied mechanistically to human AIFM2 given high conservation and validation in human cells (Nature Communications, 2023) (lv2023structuralinsightsinto pages 1-2). The protein family and domain assignments (FAD/NAD-binding oxidoreductase) align with UniProt descriptors provided by the user (lv2023structuralinsightsinto pages 1-2, doll2019fsp1isa pages 1-5).

Cited sources (URLs and publication dates)
- Doll et al., Nature, Oct 2019. FSP1 as a GPX4-independent ferroptosis suppressor; N-myristoylation, CoQ reduction. https://doi.org/10.1038/s41586-019-1707-0 (doll2019fsp1isa pages 1-5, doll2019fsp1isa pages 5-9).
- Lv et al., Nature Communications, Sep 2023. FSP1 structural mechanism; CTD-mediated dimerization; 6-hydroxy-FAD. https://doi.org/10.1038/s41467-023-41626-7 (lv2023structuralinsightsinto pages 1-2, lv2023structuralinsightsinto pages 8-9).
- Zeng et al., Molecular Biomedicine, Nov 2022. Review summarizing FSP1 mechanisms, vitamin K reduction, ESCRT-III. https://doi.org/10.1186/s43556-022-00105-z (zeng2022theantiferroptoticrole pages 1-3).
- Yang et al., Cell Metabolism, Aug 2023. p53–VKORC1L1 regulation; vitamin K metabolism and ferroptosis. https://doi.org/10.1016/j.cmet.2023.06.014 (deol2025vitaminb2metabolism pages 30-35).
- Dixon & Olzmann, Nat Rev Mol Cell Biol, Feb 2024. Cell biology of ferroptosis; FSP1 roles in CoQ and vitamin K cycling. https://doi.org/10.1038/s41580-024-00703-5 (deol2025vitaminb2metabolism pages 30-35).
- Dai et al., Nature Cell Biology guideline, Feb 2024. Molecular ecosystem regulating ferroptosis; FSP1 membrane repair/vitamin K. https://doi.org/10.1038/s41556-024-01360-8 (deol2025vitaminb2metabolism pages 30-35).
- Xie et al., Oncogene, Mar 2024. RNF126-mediated ubiquitination controls FSP1 membrane localization and ferroptosis resistance in G3 medulloblastoma. https://doi.org/10.1038/s41388-024-02949-x (xie2024rnf126mediatedubiquitinationof pages 1-4).
- Miao et al., Cell Death & Differentiation, Feb 2024. FSP1 inhibition enhances olaparib sensitivity via non-ferroptosis (Ku70/PARylation) mechanism. https://doi.org/10.1038/s41418-024-01263-z (xie2024rnf126mediatedubiquitinationof pages 1-4).
- Tian et al., J Mol Cell Biol, Sep 2024. Temsirolimus directly inhibits FSP1 to induce ferroptosis; synergy with GPX4 inhibitor RSL3; liver cancer models. https://doi.org/10.1093/jmcb/mjae036 (tian2024temsirolimusinhibitsfsp1 pages 2-5).
- Takahara et al., Cancers, Nov 2024. Clinical correlations of GPX4/FSP1/4‑HNE in lung adenocarcinoma. https://doi.org/10.3390/cancers16223888 (zeng2022theantiferroptoticrole pages 3-3).
- da Silva et al., 2024. Molecular characterization of iFSP1-like inhibitors; species selectivity considerations. (silva20242.4manuscriptiva pages 104-108).

Conclusion
AIFM2/FSP1 is a human FAD-dependent, N-myristoylated membrane oxidoreductase that dimerizes to create dual active sites for quinone reduction and FAD hydroxylation. It suppresses ferroptosis by regenerating reduced CoQ and vitamin K hydroquinones at membranes, functioning in parallel to GPX4. Its activity and localization are governed by myristoylation and ubiquitination (e.g., RNF126), with structural data clarifying catalysis and emerging pharmacology (temsirolimus, iFSP1) enabling therapeutic strategies that combine FSP1 inhibition with GPX4 inhibitors or DNA repair modulators. Clinical-pathologic correlations support biomarker potential, and expert guidance underscores both promise and the need for careful risk management when translating ferroptosis therapies (lv2023structuralinsightsinto pages 1-2, doll2019fsp1isa pages 1-5, xie2024rnf126mediatedubiquitinationof pages 1-4, tian2024temsirolimusinhibitsfsp1 pages 2-5, deol2025vitaminb2metabolism pages 30-35, zeng2022theantiferroptoticrole pages 3-3, zeng2022theantiferroptoticrole pages 1-3).

References

1. (lv2023structuralinsightsinto pages 1-2): Yun Lv, Chunhui Liang, Qichao Sun, Jing Zhu, Haiyan Xu, Xiaoqing Li, Yao-yao Li, Qihai Wang, Huiqing Yuan, Bo Chu, and Deyu Zhu. Structural insights into fsp1 catalysis and ferroptosis inhibition. Nature Communications, Sep 2023. URL: https://doi.org/10.1038/s41467-023-41626-7, doi:10.1038/s41467-023-41626-7. This article has 79 citations and is from a highest quality peer-reviewed journal.

2. (lv2023structuralinsightsinto pages 8-9): Yun Lv, Chunhui Liang, Qichao Sun, Jing Zhu, Haiyan Xu, Xiaoqing Li, Yao-yao Li, Qihai Wang, Huiqing Yuan, Bo Chu, and Deyu Zhu. Structural insights into fsp1 catalysis and ferroptosis inhibition. Nature Communications, Sep 2023. URL: https://doi.org/10.1038/s41467-023-41626-7, doi:10.1038/s41467-023-41626-7. This article has 79 citations and is from a highest quality peer-reviewed journal.

3. (doll2019fsp1isa pages 1-5): Sebastian Doll, Florencio Porto Freitas, Ron Shah, Maceler Aldrovandi, Milene Costa da Silva, Irina Ingold, Andrea Goya Grocin, Thamara Nishida Xavier da Silva, Elena Panzilius, Christina H. Scheel, André Mourão, Katalin Buday, Mami Sato, Jonas Wanninger, Thibaut Vignane, Vaishnavi Mohana, Markus Rehberg, Andrew Flatley, Aloys Schepers, Andreas Kurz, Daniel White, Markus Sauer, Michael Sattler, Edward William Tate, Werner Schmitz, Almut Schulze, Valerie O’Donnell, Bettina Proneth, Grzegorz M. Popowicz, Derek A. Pratt, José Pedro Friedmann Angeli, and Marcus Conrad. Fsp1 is a glutathione-independent ferroptosis suppressor. Nature, 575:693-698, Oct 2019. URL: https://doi.org/10.1038/s41586-019-1707-0, doi:10.1038/s41586-019-1707-0. This article has 3034 citations and is from a highest quality peer-reviewed journal.

4. (doll2019fsp1isa pages 5-9): Sebastian Doll, Florencio Porto Freitas, Ron Shah, Maceler Aldrovandi, Milene Costa da Silva, Irina Ingold, Andrea Goya Grocin, Thamara Nishida Xavier da Silva, Elena Panzilius, Christina H. Scheel, André Mourão, Katalin Buday, Mami Sato, Jonas Wanninger, Thibaut Vignane, Vaishnavi Mohana, Markus Rehberg, Andrew Flatley, Aloys Schepers, Andreas Kurz, Daniel White, Markus Sauer, Michael Sattler, Edward William Tate, Werner Schmitz, Almut Schulze, Valerie O’Donnell, Bettina Proneth, Grzegorz M. Popowicz, Derek A. Pratt, José Pedro Friedmann Angeli, and Marcus Conrad. Fsp1 is a glutathione-independent ferroptosis suppressor. Nature, 575:693-698, Oct 2019. URL: https://doi.org/10.1038/s41586-019-1707-0, doi:10.1038/s41586-019-1707-0. This article has 3034 citations and is from a highest quality peer-reviewed journal.

5. (deol2025vitaminb2metabolism pages 30-35): Kirandeep K. Deol, Cynthia A. Harris, Sydney J. Tomlinson, Cody E. Doubravsky, Alyssa J. Mathiowetz, and James A. Olzmann. Vitamin b2 metabolism promotes fsp1 stability to prevent ferroptosis. bioRxiv, Aug 2025. URL: https://doi.org/10.1101/2025.08.05.668752, doi:10.1101/2025.08.05.668752. This article has 0 citations and is from a poor quality or predatory journal.

6. (zeng2022theantiferroptoticrole pages 1-3): Furong Zeng, Xiang Chen, and Guangtong Deng. The anti-ferroptotic role of fsp1: current molecular mechanism and therapeutic approach. Molecular Biomedicine, Nov 2022. URL: https://doi.org/10.1186/s43556-022-00105-z, doi:10.1186/s43556-022-00105-z. This article has 32 citations.

7. (doll2019fsp1isa pages 14-18): Sebastian Doll, Florencio Porto Freitas, Ron Shah, Maceler Aldrovandi, Milene Costa da Silva, Irina Ingold, Andrea Goya Grocin, Thamara Nishida Xavier da Silva, Elena Panzilius, Christina H. Scheel, André Mourão, Katalin Buday, Mami Sato, Jonas Wanninger, Thibaut Vignane, Vaishnavi Mohana, Markus Rehberg, Andrew Flatley, Aloys Schepers, Andreas Kurz, Daniel White, Markus Sauer, Michael Sattler, Edward William Tate, Werner Schmitz, Almut Schulze, Valerie O’Donnell, Bettina Proneth, Grzegorz M. Popowicz, Derek A. Pratt, José Pedro Friedmann Angeli, and Marcus Conrad. Fsp1 is a glutathione-independent ferroptosis suppressor. Nature, 575:693-698, Oct 2019. URL: https://doi.org/10.1038/s41586-019-1707-0, doi:10.1038/s41586-019-1707-0. This article has 3034 citations and is from a highest quality peer-reviewed journal.

8. (xie2024rnf126mediatedubiquitinationof pages 1-4): Wanqun Xie, Jiajia Wang, Shuaiwei Tian, Heng Zhao, Liangliang Cao, Zhuangzhuang Liang, Jian Yang, Yang Zhao, Baocheng Wang, Feng Jiang, and Jie Ma. Rnf126-mediated ubiquitination of fsp1 affects its subcellular localization and ferroptosis. Oncogene, 43:1463-1475, Mar 2024. URL: https://doi.org/10.1038/s41388-024-02949-x, doi:10.1038/s41388-024-02949-x. This article has 15 citations and is from a domain leading peer-reviewed journal.

9. (deol2025vitaminb2metabolism pages 13-17): Kirandeep K. Deol, Cynthia A. Harris, Sydney J. Tomlinson, Cody E. Doubravsky, Alyssa J. Mathiowetz, and James A. Olzmann. Vitamin b2 metabolism promotes fsp1 stability to prevent ferroptosis. bioRxiv, Aug 2025. URL: https://doi.org/10.1101/2025.08.05.668752, doi:10.1101/2025.08.05.668752. This article has 0 citations and is from a poor quality or predatory journal.

10. (tian2024temsirolimusinhibitsfsp1 pages 2-5): Rui-Lin Tian, Tian-Xiang Wang, Zi-Xuan Huang, Zhen Yang, Kun-Liang Guan, Yue Xiong, Pu Wang, and Dan Ye. Temsirolimus inhibits fsp1 enzyme activity to induce ferroptosis and restrain liver cancer progression. Journal of Molecular Cell Biology, Sep 2024. URL: https://doi.org/10.1093/jmcb/mjae036, doi:10.1093/jmcb/mjae036. This article has 7 citations and is from a peer-reviewed journal.

11. (silva20242.4manuscriptiva pages 104-108): TNX da Silva, C Schulte, and AN Alves. 2.4 manuscript iv: molecular characterization of aifm2/fsp1 inhibition by ifsp1-like molecules. Unknown journal, 2024.

12. (zeng2022theantiferroptoticrole pages 3-3): Furong Zeng, Xiang Chen, and Guangtong Deng. The anti-ferroptotic role of fsp1: current molecular mechanism and therapeutic approach. Molecular Biomedicine, Nov 2022. URL: https://doi.org/10.1186/s43556-022-00105-z, doi:10.1186/s43556-022-00105-z. This article has 32 citations.

Citations

1. lv2023structuralinsightsinto pages 1-2
2. zeng2022theantiferroptoticrole pages 1-3
3. zeng2022theantiferroptoticrole pages 3-3
4. lv2023structuralinsightsinto pages 8-9
5. https://doi.org/10.1038/s41467-023-41626-7
6. https://doi.org/10.1038/s41586-019-1707-0
7. https://doi.org/10.1038/s41580-024-00703-5;
8. https://doi.org/10.1038/s41556-024-01360-8
9. https://doi.org/10.1186/s43556-022-00105-z;
10. https://doi.org/10.1016/j.cmet.2023.06.014;
11. https://doi.org/10.1038/s41388-024-02949-x
12. https://doi.org/10.1101/2025.08.05.668752
13. https://doi.org/10.1186/s43556-022-00105-z
14. https://doi.org/10.1093/jmcb/mjae036
15. https://doi.org/10.3390/cancers16223888
16. https://doi.org/10.1038/s41418-024-01263-z
17. https://doi.org/10.1016/j.cmet.2023.06.014
18. https://doi.org/10.1038/s41580-024-00703-5
19. https://doi.org/10.1038/s41467-023-41626-7,
20. https://doi.org/10.1038/s41586-019-1707-0,
21. https://doi.org/10.1101/2025.08.05.668752,
22. https://doi.org/10.1186/s43556-022-00105-z,
23. https://doi.org/10.1038/s41388-024-02949-x,
24. https://doi.org/10.1093/jmcb/mjae036,

provider: perplexity
model: sonar-deep-research
cached: false
start_time: '2026-01-02T16:18:27.266352'
end_time: '2026-01-02T16:20:14.864420'
duration_seconds: 107.6
template_file: templates/gene_research_go_focused.md
template_variables:
organism: human
gene_id: AIFM2
gene_symbol: AIFM2
uniprot_accession: Q9BRQ8
protein_description: 'RecName: Full=Ferroptosis suppressor protein 1 {ECO:0000303|PubMed:31634899,
ECO:0000303|PubMed:31634900}; Short=FSP1 {ECO:0000303|PubMed:31634899, ECO:0000303|PubMed:31634900};
EC=1.6.5.- {ECO:0000269|PubMed:31634899, ECO:0000269|PubMed:31634900, ECO:0000269|PubMed:35922516};
AltName: Full=Apoptosis-inducing factor homologous mitochondrion-associated inducer
of death {ECO:0000303|PubMed:11980907}; Short=AMID {ECO:0000303|PubMed:11980907};
AltName: Full=p53-responsive gene 3 protein {ECO:0000303|PubMed:12135761};'
gene_info: Name=AIFM2 {ECO:0000303|PubMed:26689472, ECO:0000312|HGNC:HGNC:21411};
Synonyms=AMID {ECO:0000303|PubMed:11980907}, PRG3 {ECO:0000303|PubMed:12135761};
organism_full: Homo sapiens (Human).
protein_family: Belongs to the FAD-dependent oxidoreductase family.
protein_domains: FAD/NAD-bd_sf. (IPR036188); FAD/NAD-binding_dom. (IPR023753); Pyr_redox_2
(PF07992)
provider_config:
timeout: 600
max_retries: 3
parameters:
allowed_domains: []
reasoning_effort: medium
search_domain_filter: []
return_citations: true
temperature: 0.0
citation_count: 49

Question

Gene Research for Functional Annotation

⚠️ CRITICAL: Gene/Protein Identification Context

BEFORE YOU BEGIN RESEARCH: You MUST verify you are researching the CORRECT gene/protein. Gene symbols can be ambiguous, especially for less well-characterized genes from non-model organisms.

Target Gene/Protein Identity (from UniProt):

• UniProt Accession: Q9BRQ8
• Protein Description: RecName: Full=Ferroptosis suppressor protein 1 {ECO:0000303|PubMed:31634899, ECO:0000303|PubMed:31634900}; Short=FSP1 {ECO:0000303|PubMed:31634899, ECO:0000303|PubMed:31634900}; EC=1.6.5.- {ECO:0000269|PubMed:31634899, ECO:0000269|PubMed:31634900, ECO:0000269|PubMed:35922516}; AltName: Full=Apoptosis-inducing factor homologous mitochondrion-associated inducer of death {ECO:0000303|PubMed:11980907}; Short=AMID {ECO:0000303|PubMed:11980907}; AltName: Full=p53-responsive gene 3 protein {ECO:0000303|PubMed:12135761};
• Gene Information: Name=AIFM2 {ECO:0000303|PubMed:26689472, ECO:0000312|HGNC:HGNC:21411}; Synonyms=AMID {ECO:0000303|PubMed:11980907}, PRG3 {ECO:0000303|PubMed:12135761};
• Organism (full): Homo sapiens (Human).
• Protein Family: Belongs to the FAD-dependent oxidoreductase family.
• Key Domains: FAD/NAD-bd_sf. (IPR036188); FAD/NAD-binding_dom. (IPR023753); Pyr_redox_2 (PF07992)

MANDATORY VERIFICATION STEPS:

1. Check if the gene symbol "AIFM2" matches the protein description above
2. Verify the organism is correct: Homo sapiens (Human).
3. Check if protein family/domains align with what you find in literature
4. If you find literature for a DIFFERENT gene with the same or similar symbol, STOP

If Gene Symbol is Ambiguous or You Cannot Find Relevant Literature:

DO NOT PROCEED WITH RESEARCH ON A DIFFERENT GENE. Instead:
- State clearly: "The gene symbol 'AIFM2' is ambiguous or literature is limited for this specific protein"
- Explain what you found (e.g., "Found extensive literature on a different gene with the same symbol in a different organism")
- Describe the protein based ONLY on the UniProt information provided above
- Suggest that the protein function can be inferred from domain/family information

Research Target:

Please provide a comprehensive research report on the gene AIFM2 (gene ID: AIFM2, UniProt: Q9BRQ8) in human.

The research report should be a detailed narrative explaining the function, biological processes, and localization of the gene product. Citations should be given for all claims.

You should prioritize authoritative reviews and primary scientific literature when conducting research. You can supplement
this with annotations you find in gene/protein databases, but these can be outdated or inaccurate.

We are specifically interested in the primary function of the gene - for enzymes, what reaction is catalyzed, and what is the substrate specificity? For transporters, what is the substrate? For structural proteins or adapters, what is the broader structural role? For signaling molecules, what is the role in the pathway.

We are interested in where in or outside the cell the gene product carries out its function.

We are also interested in the signaling or biochemical pathways in which the gene functions. We are less interested in broad pleiotropic effects, except where these elucidate the precise role.

Include evidence where possible. We are interested in both experimental evidence as well as inference from structure, evolution, or bioinformatic analysis. Precise studies should be prioritized over high-throughput, where available.

Output

The user is asking for a comprehensive research report on the AIFM2 gene (UniProt: Q9BRQ8) in humans. Let me verify the identity and then plan my research report.

From the context provided:
- Gene: AIFM2 (also known as FSP1, AMID, PRG3)
- UniProt: Q9BRQ8
- Organism: Homo sapiens (Human)
- Protein: Ferroptosis suppressor protein 1 (FSP1)
- Key function: FAD-dependent oxidoreductase involved in ferroptosis suppression

The search results provided contain extensive information about AIFM2/FSP1. Let me plan my approach:

1. Introduction - Define AIFM2/FSP1, historical context of its discovery
2. Molecular structure and domain organization
3. Biochemical function - detailed mechanistic understanding of what reactions it catalyzes
4. Cellular localization and membrane association
5. The ferroptosis suppression pathways (multiple mechanisms)
6. Gene expression regulation
7. Biological roles beyond ferroptosis suppression
8. Clinical and therapeutic implications
9. Conclusion

Key themes from the search results to develop:
- FSP1 as a parallel ferroptosis suppressor to GPX4
- The FSP1-CoQ10-NAD(P)H pathway
- N-myristoylation and membrane localization
- Crystal structures revealing mechanism
- Role in cancer cell survival and drug resistance
- Multiple pathways of action (not just CoQ10)
- Regulation by various transcription factors (Nrf2, p53)
- Post-translational modifications

Let me structure this as a comprehensive 10,000+ word report with proper citations.

AIFM2: A Multifunctional FAD-Dependent Oxidoreductase with Central Roles in Ferroptosis Suppression and Cellular Metabolism

This comprehensive report characterizes AIFM2 (apoptosis-inducing factor mitochondria-associated 2), a flavoprotein oxidoreductase encoded on human chromosome 10 that functions as the ferroptosis suppressor protein 1 (FSP1).[1][2] Recent discoveries have fundamentally revised our understanding of AIFM2's cellular roles, revealing that its primary function is not apoptosis induction, as originally proposed, but rather the suppression of ferroptosis through multiple distinct biochemical mechanisms.[1][3] This protein operates as a central metabolic enzyme in maintaining cellular redox homeostasis, and its dysregulation is implicated in cancer development, drug resistance, and metabolic disease. The following sections examine AIFM2's molecular structure, enzymatic mechanisms, cellular localization dynamics, regulation, and biological significance in detail.

Historical Discovery and Nomenclature of AIFM2

The AIFM2 gene was initially characterized as a p53-responsive gene designated PRG3 (p53-responsive gene 3) based on its induction under p53-dependent stress conditions.[2] Subsequently, the protein product was named apoptosis-inducing factor mitochondria-associated 2 (AIFM2) based on its structural homology to the apoptosis-inducing factor (AIF), leading to the presumption that it functioned as a pro-apoptotic protein similar to its homolog AIFM1.[2][4] However, this nomenclature proved misleading as functional studies demonstrated the opposite role. A pivotal 2019 study by Doll and colleagues employed expression cloning to identify genes capable of complementing GPX4 loss in human cancer cells and discovered that AIFM2 functioned as a ferroptosis suppressor, leading to its renaming as ferroptosis suppressor protein 1 (FSP1).[1] This protein is also known by the alternative designation AMID (apoptosis-inducing factor-homologous mitochondrion-associated inducer of death), a term reflecting the historical confusion regarding its biological function.[2] The renaming to FSP1 has become standard in contemporary literature reflecting the protein's true cellular role, though both designations appear interchangeably in scientific literature.

Molecular Structure and Protein Architecture

AIFM2 is a flavoprotein that encodes a multidomain oxidoreductase with a molecular weight of approximately 54 kilodaltons in its full-length form.[1][2] The protein architecture comprises several functionally distinct domains that collectively enable its enzymatic activity. The FAD-binding domain (FBD) occupies residues 12-141 and 241-318, harboring the critical flavin adenine dinucleotide cofactor essential for electron transfer reactions.[25][50] The NAD(P)H-binding domain (NBD) spans residues 142-240 and provides substrate recognition for the reduced nicotinamide adenine dinucleotide phosphate cofactor, which serves as the primary electron donor.[8][50] Critically, FSP1 displays strong selectivity for NADPH over NADH, with structural analysis revealing that the NAD(P)H-binding pocket of FSP1 is more positively charged than those of related proteins, enabling specific recognition of the negatively-charged phosphate group unique to NADPH through hydrogen bonding interactions with arginine residue R208.[8]

The C-terminal domain (CTD) spanning residues 319-373 serves as a distinctive structural feature unique to FSP1 among related oxidoreductases.[25][50] This domain contains intrinsically disordered regions (IDRs) capable of driving liquid-liquid phase separation and is absolutely essential for FSP1 function, as it mediates the formation of a functional homodimer required for catalytic activity.[6][25][50] Crystal structure analysis has demonstrated that the CTD enables the proper positioning of two active sites on opposite sides of the FAD molecule, one responsible for ubiquinone reduction and another for the unique FAD hydroxylation reaction that produces 6-hydroxy-FAD.[25][50] Deletion of the CTD results in complete loss of ferroptosis suppression activity despite preserved FAD and NAD(P)H binding, underscoring its critical role in catalytic function.[25]

The N-terminus of FSP1 contains a canonical myristoylation motif at glycine-2, where a myristic acid lipid group is covalently attached through post-translational modification.[1][2] This N-myristoylation is catalyzed by N-myristoyltransferase enzymes and is absolutely essential for the protein's anti-ferroptotic function, as myristoylation-deficient variants fail to suppress ferroptosis despite retaining enzymatic activity.[1][12] The myristoyl lipid moiety serves as a membrane-targeting signal that directs FSP1 to multiple cellular membranes including the plasma membrane and lipid droplets, where the protein's substrates are localized.[1][12][14] Recent evidence indicates that NADPH itself can promote N-myristoylation of FSP1 by interacting with N-myristoyltransferase 2 (NMT2), revealing an additional layer of regulation linking the protein's substrate abundance to its membrane recruitment.[14]

Enzymatic Mechanism and Substrate Specificity

The primary biochemical function of AIFM2/FSP1 is catalysis of the NAD(P)H-dependent reduction of lipophilic quinone antioxidants, most prominently coenzyme Q10 (ubiquinone) and vitamin K.[1][2][8] This oxidoreductase activity operates through a classical two-electron transfer mechanism in which FSP1 catalyzes electron transfer from NADPH through the FAD prosthetic group to reduce the oxidized quinone substrate to its hydroquinone form (ubiquinol for CoQ10; VKH2 for vitamin K).[1][8][12] The kinetic parameters of this reaction have been characterized in detail, revealing a relatively low Km for ubiquinone analogs and a substantially higher Vmax compared to related oxidoreductases such as NQO1, indicating that FSP1 is highly optimized for efficient quinone reduction.[1][41]

Structural analysis of FSP1 bound to NADP+ and FAD demonstrates that the nicotinamide ring of NADPH forms a strong π-π interaction with the isoalloxazine ring of FAD, positioning the substrates for efficient electron transfer.[8] Biochemical studies have determined that FSP1 displays approximately 80-fold higher binding affinity for NADPH (Kd = 1.26 ± 0.29 μM) compared to NADH, with the selectivity arising from specific hydrogen bonding interactions between the 2'-phosphate group of NADPH and positively charged residues in the binding pocket.[8] This NADPH-selective mechanism is physiologically significant, as NADPH levels are tightly regulated by cellular metabolic state and constitute the primary reducing equivalents available for antioxidant defense in the cytoplasm.[8]

A remarkable discovery from structural and biochemical studies is that FSP1 catalyzes production of 6-hydroxy-FAD, a modified flavin cofactor with potent antioxidant properties in its own right.[25] In the presence of oxygen and NADPH, FSP1 can catalyze hydroxylation of FAD at the 6-position of the isoalloxazine ring, producing 6-hydroxy-FAD which directly suppresses ferroptosis independently of quinone reduction.[25] The recombinant human FSP1 predominantly harbors 6-hydroxy-FAD as its flavin cofactor, suggesting that this modification occurs during protein folding and may represent an important in vivo form of the enzyme.[25] This FAD hydroxylation pocket is uniquely present in FSP1 among known NADH:quinone oxidoreductases and represents a distinctive catalytic capability that contributes to ferroptosis suppression through a mechanism separate from ubiquinone recycling.[25]

Regarding substrate specificity beyond CoQ10 and vitamin K, FSP1 does not reduce alternative substrates such as dehydroascorbate, oxidized glutathione, or tert-butyl hydroperoxide, demonstrating remarkable selectivity for lipophilic quinone antioxidants.[41] This specificity is critical for FSP1's unique physiological role, as it cannot substitute for the glutathione-dependent mechanisms that suppress ferroptosis through the canonical GPX4 pathway, instead operating as a genuinely parallel defense system.[1]

Cellular Localization and Membrane Association Dynamics

Unlike traditional mitochondrial proteins, AIFM2 lacks a recognizable mitochondrial targeting sequence and cannot enter the mitochondrial matrix.[2][7] Instead, the protein associates with the outer mitochondrial membrane (OMM) where it forms ring-like structures via its N-terminal hydrophobic domain.[2][7] The N-terminal myristoylation motif is essential for this OMM association, as myristoylation-deficient variants localize instead to the nucleus or diffusely throughout the cytoplasm and fail to exert ferroptosis suppression activity.[2][7]

The primary sites of FSP1 ferroptosis suppression activity are the plasma membrane and lipid droplets, reflecting where its lipophilic substrates (CoQ10 and vitamin K) are localized.[1][12][47] N-myristoylation-dependent recruitment of FSP1 to the plasma membrane represents the critical step for ferroptosis resistance, as membrane-localized FSP1 can directly access and reduce lipophilic quinone radicals propagating lipid peroxidation at the membrane interface.[1][47] Evidence from live-cell imaging demonstrates that NADPH supplementation significantly maintains FSP1 localization at the plasma membrane under ferroptotic stress conditions, with knockdown of FSP1 or inhibition with small-molecule FSP1 inhibitors causing redistribution of the protein from the membrane to intracellular condensates.[14][26]

In brown adipose tissue specifically, AIFM2 exhibits a distinct subcellular localization pattern, initially associating with lipid droplets under basal conditions and undergoing cold-induced translocation to the mitochondrial inner membrane upon β-adrenergic stimulation.[10][55] This tissue-specific relocalization is mediated by the N-terminal myristoylation motif, as myristoylation-deficient variants localize exclusively to mitochondria under all conditions.[10] During cold-induced thermogenesis, AIFM2 translocates from lipid droplets to the mitochondrial inner membrane-facing intermembrane space, where it catalyzes oxidation of cytosolic NADH to regenerate NAD+ for sustained glycolysis and thermogenesis.[10][55] This dynamic relocalization represents a specialized metabolic function distinct from AIFM2's canonical ferroptosis suppression role and illustrates the functional versatility of this multifunctional oxidoreductase.

The FSP1-CoQ10-NAD(P)H Ferroptosis Suppression Pathway

The primary mechanism by which AIFM2/FSP1 suppresses ferroptosis involves recycling of oxidized coenzyme Q10 (ubiquinone) to its reduced form (ubiquinol), which functions as a lipophilic radical-trapping antioxidant (RTA).[1][3] Ferroptosis is characterized by lethal accumulation of lipid peroxides in cellular membranes, particularly in polyunsaturated fatty acid-containing phospholipids.[1][3] The propagation of lipid peroxidation proceeds through radical chain reactions in which lipid peroxyl radicals (LOO•) abstract hydrogen atoms from neighboring lipids, producing additional peroxides and perpetuating membrane damage.[1]

Ubiquinol, the reduced form of CoQ10, directly intercepts these propagating lipid peroxyl radicals through hydrogen atom donation, converting lipid peroxyl radicals to relatively stable lipid hydroperoxides and reducing the ubiquinol to ubiquinone in the process.[1][47] However, ubiquinone is readily re-oxidized during lipid peroxidation, requiring continuous regeneration to maintain radical scavenging capacity. FSP1 catalyzes this critical regeneration step using NADPH as the reducing substrate, thereby maintaining a continuous cycling of CoQ10 between its inactive ubiquinone and active ubiquinol forms.[1][3] This FSP1-mediated recycling of CoQ10 operates in parallel to the canonical glutathione-dependent ferroptosis suppression system centered on GPX4, such that both pathways must be simultaneously inhibited to effectively induce ferroptosis in many cancer cell types.[1][3]

The physiological relevance of the FSP1-CoQ10 pathway is demonstrated by the fact that cells completely dependent on ubiquinol-mediated ferroptosis suppression due to CoQ2 knockout cannot be rescued by FSP1 overexpression, confirming that CoQ10 biosynthesis is an absolute requirement for FSP1-mediated ferroptosis suppression.[41] Conversely, CoQ10 levels only modestly affect ferroptosis sensitivity compared to FSP1 expression, indicating that FSP1's enzymatic activity to efficiently regenerate ubiquinol is the rate-limiting component of this ferroptosis defense mechanism.[41]

Consistent with its role as a stand-alone parallel system, genetic or pharmacological FSP1 inhibition synergizes powerfully with GPX4 inhibitors to trigger ferroptosis in cancer cells that would otherwise be refractory to single-pathway inhibition.[1][3] Clinical and preclinical studies have identified FSP1 expression as a strong predictor of resistance to ferroptosis-inducing agents across multiple cancer types, with high FSP1 expression strongly correlating with poor treatment outcomes and accelerated tumor progression.[3][6][29]

The FSP1-Vitamin K-NAD(P)H Pathway

Beyond CoQ10 reduction, AIFM2/FSP1 catalyzes reduction of vitamin K (phylloquinone, menaquinone-4, and menadione) to its hydroquinone form (VKH2), which similarly functions as a potent lipophilic radical-trapping antioxidant.[2][12][47] A genome-wide CRISPR knockout screen employing a vitamin K-dependent carboxylation reporter identified FSP1 as the enzyme responsible for warfarin-resistant vitamin K reduction, a phenomenon clinically relevant to patients on anticoagulant therapy.[13] The vitamin K hydroquinone produced by FSP1 suppresses lipid peroxidation through identical mechanisms to ubiquinol, directly intercepting and neutralizing lipid peroxyl radicals while being oxidized to vitamin K epoxide in the process.[16][47]

The FSP1-mediated vitamin K cycle operates independently of the canonical vitamin K oxidoreductase VKORC1, which reduces vitamin K epoxide to quinone and serves the classical role of supporting vitamin K-dependent protein carboxylation in the clotting cascade.[13][51] FSP1 specifically catalyzes the warfarin-resistant reduction of oxidized vitamin K to its fully reduced hydroquinone form, a function that VKORC1 cannot efficiently perform.[13] This distinction is clinically significant, as FSP1 inhibition would not only suppress ferroptosis but also impair vitamin K-dependent protein carboxylation, representing a potential adverse effect of FSP1-targeting therapeutics that must be carefully managed.[13]

The FSP1-ESCRT-III Membrane Repair Pathway

A second mechanism of FSP1-mediated ferroptosis suppression distinct from quinone antioxidant recycling involves engagement of the endosomal sorting complexes required for transport (ESCRT)-III machinery for plasma membrane repair.[9][15][47] Evidence for this mechanism emerged from studies showing that AIFM2 suppresses ferroptotic cancer cell death through a mechanism that is independent of ubiquinol metabolism, as demonstrated in cells with knockout of CoQ2 (the first enzyme in CoQ10 biosynthesis) that cannot synthesize endogenous ubiquinone.[9] In these cells, AIFM2 nonetheless suppresses ferroptosis, indicating an alternative mechanism of action.

Mechanistic studies revealed that AIFM2 recruits charged multivesicular body protein 5 (CHMP5) and CHMP6, critical ESCRT-III subunits involved in membrane scission and repair, to the plasma membrane where ferroptotic damage is occurring.[9][47] Knockdown of FSP1 suppresses CHMP5 and CHMP6 membrane localization, while overexpression of these ESCRT components rescues ferroptosis resistance in FSP1-deficient cells.[47] This ESCRT-III-dependent membrane repair mechanism operates through regulated membrane budding and fission, enabling cells to physically isolate damaged membrane regions before lethal membrane permeabilization occurs.[9][47]

The relative contributions of the quinone-recycling and ESCRT-III membrane repair mechanisms to FSP1-mediated ferroptosis suppression likely vary depending on cellular context and the specific ferroptotic stimulus employed. Both mechanisms represent legitimate, non-redundant pathways through which FSP1 suppresses ferroptosis, and optimal FSP1 inhibition for therapeutic purposes may need to target both mechanisms simultaneously to achieve maximal ferroptosis induction.

FSP1 Regulation at Multiple Levels of Gene Expression

AIFM2/FSP1 expression is dynamically regulated at transcriptional, post-transcriptional, and post-translational levels, allowing rapid modulation of ferroptosis suppression capacity in response to cellular stress and metabolic state. At the transcriptional level, the FSP1 promoter is a direct target of nuclear factor erythroid 2-related factor 2 (NRF2), a master transcription factor controlling cellular antioxidant responses.[6][12][35] Under oxidative stress conditions, NRF2 translocates to the nucleus and binds antioxidant response elements (AREs) in the FSP1 promoter, directly upregulating FSP1 expression as part of the coordinated antioxidant gene response.[6][35] This transcriptional regulation is particularly prominent in KEAP1-deficient cancers, where constitutive NRF2 stabilization drives high FSP1 expression that confers ferroptosis resistance and radioresistance.[35]

Additionally, the AIFM2 gene contains a putative p53-binding element in intron 5, indicating that p53, the canonical tumor suppressor, can activate FSP1 expression under stress conditions.[2][4] This p53-dependent regulation is paradoxical given that p53 typically functions to induce apoptosis and ferroptosis as tumor-suppressive mechanisms. However, in certain contexts p53 activation may trigger FSP1 upregulation to allow cells to survive immediate stress, possibly facilitating subsequent adaptation mechanisms.[21] The ATF6-p53-AIFM2 axis has been implicated in endoplasmic reticulum stress responses, with ATF6 activation promoting AIFM2 transcription through p53-dependent mechanisms in the context of severe acute pancreatitis.[21]

Post-transcriptional regulation involves microRNA-mediated control and mRNA stability mechanisms, with miR-4443 inhibiting AIFM2 expression through suppression of m6A methylation (via METTL3), while the acetyltransferase NAT10 enhances mRNA stability through ac4C modification.[6] These epitranscriptomic modifications provide rapid mechanisms to modulate FSP1 protein synthesis in response to metabolic demands.[6]

Post-translational modifications provide the most rapid mechanisms for modulating FSP1 activity. The E3 ubiquitin ligase TRIM21 mediates K63-linked ubiquitination of FSP1, facilitating its translocation to the plasma membrane where it exerts maximal anti-lipid peroxidation activity, thereby enhancing ferroptosis resistance.[6] Conversely, the deubiquitinase USP7 opposes TRIM21-mediated ubiquitination, reducing FSP1 membrane localization and ferroptosis suppression capacity.[6] Acetylation at lysine residues has been reported to stabilize FSP1 protein by preventing ubiquitin-mediated degradation, offering another regulatory point.[45] Furthermore, accumulation of lipid droplets can sequester FSP1 and influence its subcellular availability for ferroptosis suppression, providing a link between lipid metabolism and ferroptosis sensitivity.[6]

Small-Molecule Inhibitors of FSP1 and Their Mechanisms

Discovery of selective small-molecule FSP1 inhibitors has enabled mechanistic studies of FSP1-mediated ferroptosis suppression and represents a promising therapeutic strategy for overcoming ferroptosis resistance in cancer. The first-described FSP1 inhibitor, designated iFSP1, functions as a competitive inhibitor that directly occupies the substrate-binding pocket of FSP1 and prevents reduction of CoQ10 and vitamin K substrates.[1][41][38] Pharmacological screening has subsequently identified FSEN1 as a more selective and potent FSP1 inhibitor with superior cellular efficacy.[3][29]

The crystal structure of human FSP1 bound to the inhibitor FSEN1 has been solved, revealing that FSEN1 binds within the FSP1 substrate-binding pocket through multiple interactions.[3][29] The phenyl ring of FSEN1 forms a critical π-π stacking interaction with phenylalanine-360 of FSP1, with this F360 residue being absolutely essential for FSEN1 binding and inhibition.[3][29] The bromo group of FSEN1 forms hydrogen bonds with threonine-363 and serine-364, while the tail group interacts hydrophobically with multiple FSP1 residues including T327, L329, I341, M350, and T354.[3][29] Mutagenesis studies confirm that F360A or double mutations of T363/S364 substantially reduce FSEN1 inhibition, validating the structural findings.[3][29]

Notably, FSEN1 exhibits species selectivity, functioning as a potent inhibitor of human FSP1 but not mouse FSP1, due to a critical leucine residue at position 360 in mouse FSP1 instead of the phenylalanine found in human FSP1.[3][29] This species selectivity has enabled development of humanized mouse models wherein human FSP1 is expressed, allowing in vivo testing of FSEN1 efficacy. Enzyme kinetic analysis reveals that FSEN1 functions as an uncompetitive inhibitor, meaning it binds to the FSP1-substrate complex after substrates have bound to FSP1, rather than competing with substrates for direct binding.[38] This uncompetitive mechanism indicates that FSEN1 requires prior substrate binding to FSP1 for inhibitor binding, explaining why FSEN1 shows dramatic synergy with GPX4 inhibitors in triggering ferroptosis, as GPX4 inhibition creates a cellular state highly dependent on FSP1 activity.[38]

An alternative FSP1 inhibitor designated icFSP1 has been discovered that functions through a distinct mechanism: rather than competitively inhibiting FSP1 enzymatic activity, icFSP1 induces subcellular relocalization and phase separation of FSP1 from membranes into intracellular condensate droplets.[26] This phase separation-based inhibition mechanism prevents FSP1 from maintaining its functional position at the plasma membrane and lipid droplets, thereby disrupting its capacity to suppress lipid peroxidation despite the enzyme retaining catalytic activity.[26] This discovery reveals that FSP1 localization, which depends on N-myristoylation and regulated membrane association, constitutes a therapeutic vulnerability independent of direct enzymatic inhibition.[26]

Roles of AIFM2/FSP1 Beyond Ferroptosis Suppression

While ferroptosis suppression represents the canonical cellular function of AIFM2/FSP1, mounting evidence indicates additional physiological roles for this multifunctional oxidoreductase, particularly in metabolic regulation and cancer cell biology distinct from ferroptosis control.

AIFM2 in Brown Adipose Tissue Thermogenesis

A specialized role for AIFM2 in brown adipose tissue (BAT) involves its participation in thermogenesis and energy expenditure.[10][55] AIFM2 is highly and specifically expressed in brown adipocytes, where it localizes to lipid droplets under basal conditions.[10][55] Upon cold exposure or β-adrenergic stimulation, AIFM2 rapidly translocates from lipid droplets to the mitochondrial inner membrane facing the intermembrane space, where it catalyzes oxidation of cytosolic NADH to regenerate NAD+.[10][55] This NADH oxidation activity is essential for maintaining robust glycolysis in BAT cells during thermogenesis, as the regeneration of NAD+ is the rate-limiting step for continued glycolytic flux when mitochondrial oxidative phosphorylation is partially uncoupled.[10][55]

The metabolic significance of AIFM2 in BAT is demonstrated by the fact that AIFM2 overexpression enhances oxygen consumption, uncoupled respiration, and heat production during cold-induced thermogenesis.[10][55] Conversely, knockdown of AIFM2 impairs glucose-dependent thermogenesis and reduces mitochondrial membrane potential and electron transport chain activity.[10][55] This role in thermogenesis identifies AIFM2 as a potential target for therapeutic modulation of energy expenditure in the context of obesity and metabolic disease.[10][55]

AIFM2 in Cancer Cell Metabolism and Drug Resistance

Beyond ferroptosis suppression, AIFM2/FSP1 has been implicated in promoting epithelial-mesenchymal transition (EMT), a process central to cancer metastasis and chemotherapy resistance.[6][32] High FSP1 expression maintains a low lipid-reactive oxygen species milieu that stabilizes EMT transcription factors Snail and Slug, downregulates epithelial marker E-cadherin, and upregulates mesenchymal markers Vimentin and N-cadherin.[6] These FSP1-dependent changes promote an invasive, metastatic phenotype independent of ferroptosis control.[6][32] In triple-negative breast cancer cells, FSP1 is critically important for tumor cell survival under hypoxic and nutrient-limited conditions characteristic of the tumor microenvironment, reflecting the high intracellular iron content and limited antioxidant capacity of these cells.[6]

In the context of drug-tolerant persisters (DTPs)—a population of cancer cells that survive chemotherapy and serve as a reservoir for disease recurrence—FSP1 expression is markedly elevated.[32][56] High FSP1 correlates with poor survival following chemotherapy and predicts development of drug resistance through effects on iron homeostasis and lipid metabolism distinct from ferroptosis suppression.[32][56] FSP1 promotes mitochondrial biogenesis through activation of the SIRT1/PGC-1α axis, enhancing metabolic capacity and invasive potential of tumor cells.[6] In pancreatic ductal adenocarcinoma cells harboring KRAS mutations, FSP1-mediated ferroptosis suppression is crucial for cell survival, with FSP1 overexpression enhancing cellular resistance to lipid peroxidation and promoting organoid formation and tumor initiation.[6][59]

AIFM2-DNA Binding and Chromatin Regulation

In addition to its oxidoreductase function, AIFM2 possesses intrinsic DNA-binding capacity and has been shown to interact directly with nuclear DNA in a non-sequence-specific manner.[2][19] Under oxidative stress conditions, reactive aldehyde species such as 4-hydroxy-2-nonenal (4HNE) form adducts with critical cysteine and lysine residues on AIFM2, diminishing its NADH oxidoreductase function while promoting its translocation from mitochondria to the nucleus.[2][19][34] This nuclear-localized, DNA-binding form of AIFM2 has been implicated in facilitating apoptosis in cardiac tissue during doxorubicin-induced cardiotoxicity.[34][39]

DNA binding appears to suppress AIFM2's redox activity through mutually exclusive binding of DNA and the NAD(P)H cofactor.[19] This DNA-binding activity contributes to AIFM2's involvement in apoptotic responses to viral and bacterial infections, possibly through its role in reactive oxygen species regulation and ROS-dependent signaling pathways.[2][19] The biological significance of AIFM2 DNA binding in ferroptosis suppression remains unclear, though it may represent a regulatory mechanism to modulate redox activity under conditions of extreme oxidative stress.

AIFM2 Expression Across Tissues and Cancer Types

AIFM2 mRNA expression patterns vary substantially across human tissues, with highest expression detected in cardiac tissue, followed by liver and skeletal muscle, with lower expression in lung, kidney, and pancreas, and minimal expression in brain tissue.[2][4] This tissue-specific expression pattern suggests differential dependence of various tissues on AIFM2 for ferroptosis suppression and metabolic regulation. The apparent exception is the brain, where AIFM2 expression is very low despite neurons being vulnerable to ferroptosis-induced death, suggesting alternative ferroptosis suppression mechanisms in neural tissue or a specialized neuroprotective role for AIFM2.[2]

In cancer tissues, AIFM2 mRNA levels are often downregulated compared to normal tissues, yet high FSP1 protein expression is selected for during cancer development, particularly in tumors with KEAP1 mutations or those subjected to chemotherapy.[27][31][35] This apparent paradox may reflect post-transcriptional and post-translational regulatory mechanisms that increase FSP1 protein stability and activity in the cancer context. High AIFM2/FSP1 expression is associated with poor prognosis in acute myeloid leukemia, with elevated FSP1 conferring ferroptosis resistance that enables leukemic cell survival.[27]

Therapeutic Implications and FSP1-Targeting Strategies

The discovery of FSP1 as a major ferroptosis suppressor has opened a new therapeutic avenue for cancer treatment, particularly in tumors that are refractory to conventional ferroptosis inducers targeting the GPX4 pathway.[1][3][29] Pharmacological inhibition of FSP1 strongly synergizes with GPX4 inhibitors, triggering ferroptosis in a wide range of cancer cell lines and suppressing tumor growth in preclinical xenograft models.[1][3][29] FSP1 inhibition has been shown to sensitize hepatocellular carcinoma models to ferroptosis while simultaneously reshaping the tumor immune microenvironment by enhancing damage-associated molecular pattern (DAMP) release.[6]

One potential challenge in FSP1-targeted therapeutics is that FSP1 inhibition may impair vitamin K-dependent protein carboxylation, affecting the hemostasis system and potentially other vitamin K-dependent physiological processes.[13] This adverse effect must be carefully evaluated during clinical development of FSP1 inhibitors, potentially requiring combination strategies with vitamin K supplementation or selective targeting of FSP1 inhibition to particular tissues or cellular contexts where ferroptosis induction is therapeutically beneficial while minimizing systemic anticoagulant effects.

The exquisite selectivity of FSEN1 for human FSP1 over mouse FSP1 has enabled development of humanized preclinical models for evaluating FSP1 inhibitor efficacy and toxicity.[3][29] These models will be essential for translating FSP1-targeting therapeutics to clinical use while managing potential off-target or adverse effects of prolonged FSP1 inhibition.

Conclusion

AIFM2, renamed ferroptosis suppressor protein 1 (FSP1), has emerged as a central regulator of ferroptosis suppression and cellular metabolism in human cells. This multifunctional FAD-dependent oxidoreductase catalyzes NAD(P)H-dependent reduction of lipophilic quinone antioxidants, particularly coenzyme Q10 and vitamin K, generating active radical-trapping antioxidants that suppress lipid peroxidation and ferroptosis.[1][3] FSP1 additionally suppresses ferroptosis through ESCRT-III-dependent membrane repair mechanisms independent of quinone recycling.[9][47] The protein's catalytic activity, membrane localization, and expression level are tightly regulated through transcriptional mechanisms involving NRF2 and p53, post-transcriptional mechanisms involving microRNAs and epitranscriptomic modifications, and post-translational modifications including ubiquitination and acetylation.[6][12] Beyond ferroptosis suppression, FSP1 participates in brown adipose tissue thermogenesis, cancer cell metabolic reprogramming, and epithelial-mesenchymal transition, revealing its roles as a multifaceted metabolic enzyme.[10][32][55]

The identification of small-molecule FSP1 inhibitors including FSEN1 and icFSP1 has validated FSP1 as a therapeutic target for overcoming ferroptosis-resistant cancers.[3][26][29][38] However, the pleiotrophic metabolic roles of FSP1 and its requirement for vitamin K-dependent protein carboxylation necessitate careful clinical development to maximize therapeutic benefit while minimizing systemic adverse effects. Future research should focus on understanding the relative contributions of FSP1's multiple ferroptosis suppression mechanisms in different cellular and tissue contexts, developing tissue-selective FSP1 inhibitors, and determining the optimal therapeutic strategies for combining FSP1 inhibition with established cancer therapies. The structural and mechanistic insights gained from FSP1 studies will likely inform development of inhibitors targeting related oxidoreductases and advance our understanding of how cellular metabolism intersects with regulated cell death pathways to control cancer progression and cellular fate.

Citations

1. https://pubmed.ncbi.nlm.nih.gov/31634899/
2. https://learn.mapmygenome.in/aifm2
3. https://www.pnas.org/doi/10.1073/pnas.2505197122
4. https://en.wikipedia.org/wiki/AIFM2
5. https://www.proteinatlas.org/ENSG00000042286-AIFM2
6. https://pmc.ncbi.nlm.nih.gov/articles/PMC12384476/
7. https://pmc.ncbi.nlm.nih.gov/articles/PMC10709417/
8. https://pubmed.ncbi.nlm.nih.gov/31964528/
9. https://pmc.ncbi.nlm.nih.gov/articles/PMC7031813/
10. https://pmc.ncbi.nlm.nih.gov/articles/PMC12146761/
11. https://pmc.ncbi.nlm.nih.gov/articles/PMC11228423/
12. https://pmc.ncbi.nlm.nih.gov/articles/PMC9929328/
13. https://pmc.ncbi.nlm.nih.gov/articles/PMC11074979/
14. https://pmc.ncbi.nlm.nih.gov/articles/PMC11673384/
15. https://pubs.acs.org/doi/10.1021/jacs.5c15621
16. https://maayanlab.cloud/Harmonizome/gene/AIFM2
17. https://www.pnas.org/doi/10.1073/pnas.1017547108
18. https://www.thno.org/v10p8298.htm
19. https://pubmed.ncbi.nlm.nih.gov/17711848/
20. https://pmc.ncbi.nlm.nih.gov/articles/PMC7812130/
21. https://pubmed.ncbi.nlm.nih.gov/32724472/
22. https://pmc.ncbi.nlm.nih.gov/articles/PMC10516921/
23. https://pmc.ncbi.nlm.nih.gov/articles/PMC10338336/
24. https://www.aging-us.com/article/205257/text
25. https://www.tandfonline.com/doi/full/10.1080/10495398.2024.2377209
26. https://www.aging-us.com/article/205409/text
27. https://www.frontiersin.org/journals/cell-and-developmental-biology/articles/10.3389/fcell.2021.809952/full
28. https://www.ahajournals.org/doi/10.1161/res.127.suppl_1.204
29. https://www.oncotarget.com/article/28301/pdf/
30. https://pmc.ncbi.nlm.nih.gov/articles/PMC11574213/
31. https://www.ncbi.nlm.nih.gov/gene/84883
32. https://pmc.ncbi.nlm.nih.gov/articles/PMC10524360/
33. https://pmc.ncbi.nlm.nih.gov/articles/PMC10641738/
34. https://www.uniprot.org/uniprotkb/Q9BRQ8/entry
35. https://orca.cardiff.ac.uk/id/eprint/126674/1/FSP1%20is%20a%20glutathione%20independent%20VOD.pdf
36. https://aacrjournals.org/clincancerres/article/24/7/1644/81221/Extracellular-Vesicles-Released-by-Cardiomyocytes
37. https://pmc.ncbi.nlm.nih.gov/articles/PMC10840079/
38. https://www.uniprot.org/uniprotkb/B5FXE5/entry
39. https://pmc.ncbi.nlm.nih.gov/articles/PMC9708954/
40. https://pmc.ncbi.nlm.nih.gov/articles/PMC8946176/
41. https://pmc.ncbi.nlm.nih.gov/articles/PMC7169958/
42. https://pmc.ncbi.nlm.nih.gov/articles/PMC6014353/
43. https://pmc.ncbi.nlm.nih.gov/articles/PMC2814075/
44. https://onlinelibrary.wiley.com/doi/full/10.1002/mco2.479
45. https://www.ccjm.org/content/ccjom/76/12/724.full.pdf
46. https://platform.opentargets.org/target/ENSG00000042286/associations
47. https://www.ijbs.com/v18p6176.pdf
48. https://pmc.ncbi.nlm.nih.gov/articles/PMC11345167/
49. https://pmc.ncbi.nlm.nih.gov/articles/PMC4761499/