TP53 notes

This notes file was created for the premetazoan-origin pass below. The main
review (TP53-ai-review.yaml, status COMPLETE) predates it; its sources are the
deep-research files in this folder. Automated deep research was not run for the
section below; it is based on PubMed searches and the cached papers cited inline.

Premetazoan origin (ORIGINS_OF_MULTICELLULARITY, 2026-10-01)

What is in unicellular relatives of animals

The choanoflagellate genome has p53-family genes
PMID:18273011.
The authors suggest a role in cell-type differentiation, but this is speculation
PMID:18273011.
Monosiga has two family members, one with and one without a SAM domain
PMID:17924139.
Abstract only. The "p53-type" label means "no SAM domain". It does not show
orthology to vertebrate TP53, which arose much later (below).
Homologs are also found in Filasterea (Capsaspora), Ichthyosporea and Corallochytrea
PMID:31861340.
The DNA-binding domain is conserved, including the DNA-contact arginines R248 and R273 and two zinc ligands
PMID:31861340;
PMID:31861340.
But there is no transactivation domain
PMID:31861340,
and no experiments have been done
PMID:31861340.
The family is restricted to Holozoa; the reported Entamoeba homolog was rejected as a false positive (same paper, Discussion).
I found no experimental study of any choanoflagellate, Capsaspora or
ichthyosporean p53-family protein (PubMed searches: "p53 family evolution
(choanoflagellate OR Monosiga OR Capsaspora OR unicellular)", "Nematostella p53").

Early-branching animals: the p63-like germline role

Sea anemone nvp63 binds the p53 response element and mediates UV-induced death of
gametes, not somatic cells
PMID:17848985;
PMID:17848985.
The authors read this as an ancestral germ-cell-protective pathway
PMID:17848985.
Fly and nematode p53 proteins also kill mainly germ cells after DNA damage
PMID:20081368.
The same abstract places p53 and its inhibitor Mdm2 in placozoans
PMID:20081368. Abstract only;
I did not read the primary Drosophila or C. elegans papers.
Vertebrate p53 is a later duplication of a p63/p73-like ancestor
PMID:20516129.
Abstract only (review). Its claim that the family is "first detected" in sea
anemones is out of date given PMID:17924139 and PMID:31861340.

Classification of the review's core functions

Core function (existing review)	Classification	Basis
Sequence-specific DNA binding to p53 response elements (GO:0000976)	Ancestral as a domain property (inferred); experimentally shown only from cnidarians onward	Unicellular DBDs keep DNA-contact residues (PMID:31861340), but binding is untested; nvp63 binds the consensus RE (PMID:17848985)
Transcriptional activation / repression by RNA pol II (GO:0045944, GO:0000122)	Unresolved before animals; activation shown in cnidarians	Unicellular homologs lack a transactivation domain (PMID:31861340)
DNA damage response and p53-mediated intrinsic apoptosis (GO:0030330, GO:0072332)	Ancestral within animals, in the germ line, through a p63-like protein; unknown in unicellular holozoans	PMID:17848985, PMID:20081368
Negative regulation of mitotic cell cycle; regulation of senescence (GO:0045930, GO:2000772)	Animal-specific on current evidence; not studied in invertebrates in the papers I read	No source above tests it
Somatic tumour suppression (TP53 as a gene)	Vertebrate-specific	TP53 arises by duplication in cartilaginous fish (PMID:20516129)

So the ancestral role, as far as the cited papers go, is p63-like: DNA damage-induced
germ-cell death. The vertebrate TP53 role in somatic tumour suppression came later.
Nothing is known about what the unicellular homologs do.

Track C (propagation audit) check

TP53 has 8 IBA rows on two PANTHER nodes (PTHR11447):
- PTN000893833: chromatin, nucleus, regulation of apoptotic process (GO:0042981),
positive regulation of transcription by RNA pol II, promoter-specific chromatin
binding, DNA-binding TF activity (RNA pol II), and pol II cis-regulatory sequence-specific DNA binding.
- PTN000154790: intrinsic apoptotic signaling pathway in response to DNA damage by
p53 class mediator (GO:0042771).

QuickGO withFrom queries (2026-10-01; taxonUsage=descendants):
- PTN000893833 / Choanoflagellata (28009): 14 IBA rows. All 7 terms reach both
Monosiga brevicollis proteins, A9UZX3 (MONBRDRAFT_25618) and A9V4M3
(MONBRDRAFT_27210). None reach Salpingoeca. None reach Filasterea or Ichthyosporea.
- PTN000154790 / Filasterea (2687318): 3 TreeGrafter IEA rows (GO_REF:0000118) on
Capsaspora A0A0D2X0F0 (CAOG_000511): GO:0000978, GO:0000981, GO:0006357. The
p53-class apoptosis term GO:0042771 was not transferred.
Choanoflagellata and Ichthyosporea: 0 rows.
- No animal tissue, organ or development term reaches a unicellular organism.
- Borderline: GO:0042981 regulation of apoptotic process (IBA, PTN000893833)
reaches both Monosiga proteins. It is a cellular process, not a tissue term, but
nothing shows these proteins regulate cell death. For comparison, InterPro2GO maps
IPR002117 (p53 tumour suppressor family) to GO:0006915 apoptotic process on all
three unicellular proteins (A9UZX3, A9V4M3, A0A0D2X0F0). Both are untested
extrapolations from animals (PMID:31861340 "The biological role of the remote homologs found in Holozoa is unclear.").
- Changes to the human TP53 rows: none. These node placements do not contradict the
human annotations, and GO:0042771 sits on a node that does not reach choanoflagellates by IBA.

2026-11-06 ORGANELLES ER batch 4 (endoplasmic reticulum rows)

Row: GO:0005783 IEA (GO_REF:0000044, SL-0095), the only ER row. No sub-compartment or MAM rows.

UniProt: "Endoplasmic reticulum {ECO:0000269|PubMed:17170702}" (curated). PMID:17170702 (Synoviolin; cached text lacks the results figures): PMID:17170702.
Independent: PMID:25624484 (Giorgi et al. 2015 PNAS; fetched, abstract-only): PMID:25624484 and PMID:25624484.
The old ACCEPT cited PMID:14744935 ("Endoplasmic reticulum stress induces p53 cytoplasmic localization"), which is about cytoplasm, not ER; replaced.
Outcome: ACCEPT -> KEEP_AS_NON_CORE (recruited regulatory pool of a soluble nuclear TF). Root only; no NEW MAM (GO:0044233) row because it rests on one abstract-only study of a recruited pool. Mitochondrion rows untouched.