Ferroptosis Project

Bottom line: ferroptosis is iron-dependent regulated cell death driven by
peroxidation of polyunsaturated phospholipids, and GO treats it as an evolved
programmed process (GO:0097707).
The project reviewed all 22 human genes that supply the oxidizable
PUFA-phospholipid substrate, tune the labile iron pool, suppress the execution
step through four parallel defense systems, or reset those defenses
transcriptionally. Those gene-level reviews are now summarized in a
decomposable ferroptosis module with production
GO-CAM associations for AIFM2, GPX4, NFE2L2, and SLC7A11 switch models.

The 22 reviewed gene files, 21 COMPLETE plus ACSL4 still DRAFT, cover 2,362
seeded GOA annotations plus 10 proposed new ones as of 2026-10-04. Across those
rows, 1,458 annotations were accepted as core, 297 were kept as non-core, 113
were marked over-annotated, 97 were modified, 377 were removed, and 20 were left
undecided. TP53 alone contributed 872 seeded rows, 344 of which were removed.
The core suppressor term GO:0110076 negative regulation of ferroptosis is
accepted on GPX4, SLC7A11, AIFM2, FTH1, and NFE2L2 and proposed as a new
annotation for DHODH.

Mechanistic Scope

Ferroptosis is best modelled as one execution process, not as a linear cascade.
PUFA-containing phospholipids and redox-active iron set up the chemistry; GPX4,
CoQ10, DHODH, and BH4 arms quench the radicals or lipid hydroperoxides in
parallel; NFE2L2, KEAP1, ATF4, and TP53 tune the balance by changing
defense-gene expression.

This seed deliberately scoped out adjacent or newer partners such as FTL,
GCLM, FADS2, MBOAT1/2, and the GPX4 and SLC7A11 abundance-switch regulators
captured in production GO-CAMs.

Curation Findings

• Parallel defenses are first-class biology. AIFM2 and DHODH were key
  test cases because their 2019-2021 discovery split the field from a
  GPX4-only model into multiple independent suppressor arms. The DHODH review
  proposes GO:0110076 for the mitochondrial CoQ10 arm, and the module now
  places that process directly on the DHODH quinone oxidoreductase annoton.
• p53 is not a blanket ferroptosis label. The TP53 review retained
  well-supported SLC7A11 repression and lipid-peroxidation contexts but removed
  hundreds of indirect DNA-damage, apoptosis, and broad regulation rows from the
  ferroptosis-focused interpretation.
• The curated module is more precise than the old pathway summary. The
  maintained artifact is modules/ferroptosis.yaml: one GO:0097707 execution
  node fed by PUFA-phospholipid and iron drivers, negatively regulated by
  GPX4-glutathione, AIFM2/FSP1-CoQ10, DHODH-CoQ10, and GCH1-BH4, and optionally
  tuned by NFE2L2/NRF2, KEAP1, ATF4, and p53.
• GO-CAM switch models extend beyond the 22-gene seed. The module already
  folds in GPX4 chaperone-mediated autophagy and SLC7A11 CRL3/USP18 abundance
  switches from reviewed production GO-CAMs, but LAMP2, EGLN3, USP18, and
  related ligase/adaptor components remain second-batch review scope.

Key Discoveries

1. AIFM2/FSP1-CoQ10 pathway (2019) - GPX4-independent ferroptosis
   suppression by ubiquinol regeneration.
2. GCH1-BH4 pathway (2020) - a tetrahydrobiopterin radical-trapping axis
   parallel to GPX4 and CoQ.
3. DHODH in mitochondria (2021) - mitochondrial-inner-membrane CoQ10
   regeneration by the pyrimidine-biosynthesis enzyme DHODH.
4. MBOAT1/2 resistance (2023) - sex-hormone-linked lipid remodeling that
   should be considered with other second-batch suppressor and switch genes.

Module

The ferroptosis mechanism is captured as a recursively decomposable module
grounded to UniProt and GO:

• Ferroptosis module - one execution node
  (GO:0097707) fed by PUFA-phospholipid and labile-iron driver arms,
  redundantly suppressed by four independent defense axes, and tuned by a
  transcriptional regulatory layer. Source: modules/ferroptosis.yaml.

Key References

• Stockwell BR et al. (2017) Cell - foundational review
• Doll S et al. (2019) Nature - FSP1 discovery
• Mao C et al. (2021) Nature - DHODH
• Kraft VAN et al. (2020) ACS Cent Sci - GCH1
• Jiang X et al. (2021) Nat Rev Mol Cell Biol - comprehensive review
• Chen X et al. (2021) Signal Transduct Target Ther - mechanisms update

Project Status

• [x] Review the initial 22 human ferroptosis genes
• [x] Validate the 22 seeded gene reviews
• [x] Capture the integrated mechanism as MODULE:ferroptosis
• [x] Attach production GO-CAM associations for reviewed GPX4, SLC7A11, AIFM2,
  and NFE2L2 ferroptosis models
• [ ] Complete optional switch and adjacent-gene follow-ups
  (#4084)

STATUS

Initial 22-gene human ferroptosis seed reviewed and valid; optional switch
scope tracked in #4084
(2026-10-04).

NOTES

2026-10-04

• Re-audited the 22 human gene reviews: all validate, ACSL4 remains DRAFT,
  and the aggregate seeded/NEW/action counts still match the project summary.
• Refreshed the project page and slide deck to lead with the mechanism and the
  concrete curation results instead of the original priority-order plan.
• Fixed the main module so FADS2 is treated as adjacent follow-up scope, DHODH's
  annoton carries GO:0110076, and the USP18 SLC7A11-stabilization role points
  at the current GO-CAM reference-swap problem.
• Opened #4084 for
  the remaining optional switch genes, MBOAT1/2 and adjacent-gene scope,
  GO-CAM evidence-reference follow-up, and the stale standalone pathway summary.

2026-01-19

• Manual second passes completed for GPX4, SLC7A11, ACSL4, and AIFM2/FSP1; each
  review validated after its pass.
• GPX4 HTP/HDA evidence summaries were updated for mitochondrial, nuclear, and
  exosome entries; nuclear matrix and spermatogenesis primary-evidence follow-up
  remains in #4084.
• SLC7A11 deep-research support was added for the core antiporter activity;
  optional 2025-2026 disease/regulation papers remain in #4084.
• ACSL4 had missing HDA supporting text filled and 2025 context papers folded in.
• AIFM2/FSP1 had missing reference support resolved and newer RNF126 and
  temsirolimus papers integrated.

2025-12-28

• Completed the initial 22-gene human ferroptosis review batch across the core
  machinery, regulatory network, and supporting genes.
• Validated all 22 seeded gene reviews and drafted the standalone human pathway
  summary at genes/human/FERROPTOSIS.md.