pou5f1.1 (Xenopus laevis, Q7T103) review notes

Project: NEURAL_CREST_ORIGINS, Tier 3 (blastula pluripotency programme retained in the crest).
Reviewed 2026-10-07.

Identity and paralog bookkeeping

• UniProt Q7T103, "POU domain, class 5, transcription factor 1.1", AltName Oct-25 / XOct-25.
  RefSeq NP_001079832.1; Xenbase pou5f3.2.L (XB-GENE-919851). UniProt calls it pou5f1.1,
  Xenbase calls it pou5f3.2; same protein (RefSeq cross-reference in the UniProt entry; York 2024
  lists "Xenopus laevis pou5f3.2 NP_001079832.1" in its methods).
• The three X. laevis POU-V (Pou5f3) paralogs, with names used in the literature:
  | Literature name | Xenbase | UniProt (reviewed) | Expression |
  |---|---|---|---|
  | Oct-25 / XOct-25 | pou5f3.2 | Q7T103 (pou5f1.1) | maternal low + zygotic, peaks gastrula |
  | Oct-91 | pou5f3.1 | B7ZQA9 (pou5f1.2) | zygotic after MBT |
  | Oct-60 | pou5f3.3 | Q91989 (pou5f1.3) | maternal only |
  PMID:25931449
  PMID:1732736
• Homeologs: Q7T103 is the .L homeolog; GOA rows sit on Q7T103 only. York 2024 used L and S
  morpholinos for both pou5f3.1 and pou5f3.2.

Which paralog each experiment addressed

Molecular activity

• Octamer-binding POU-V transcription factor (POU-specific + POU homeodomain)
  PMID:1732736.
• Direct promoter binding (EMSA + ChIP in embryos) at Xnr1 and Siamois
  PMID:17541407
  and at Gsc PMID:18922797.
• Activates when bound directly to octamer sites (Xvent-2B; artificial reporters)
  PMID:15292233
  PMID:18922797.
• But its dominant embryonic mode is repression via protein-protein recruitment onto
  promoters bound by other TFs/signal transducers — a co-repressor mode:
• VegT and Tcf3 (Wnt): PMID:17541407
  PMID:17541407
• FoxH1/FAST1, WBSCR11 (Gtf2ird1), Smad2 (Nodal/activin): PMID:18922797
• Smad1/4 and Vent2 at Xvent-2B (BMP): PMID:15292233
• MF conclusion: GO:0000981 (pol II DNA-binding TF) as core; the protein binding IPIs
  (Vent2, VegT, Tcf3) map to GO:0140297 DNA-binding transcription factor binding (all three
  partners are DNA-binding TFs); NEW GO:0003714 transcription corepressor activity
  (repression independent of its own site, through Tcf3/VegT).

Biological role 1: germ-layer timing / anti-differentiation (best supported, core)

• Antagonises the primary mesendoderm inducers in the nucleus: VegT, beta-catenin/Tcf3,
  Nodal/activin (and FGF) — and BMP in the ectoderm:
  PMID:16860542
  PMID:16860542
  PMID:18922797
• Ectoderm: limits competence to respond to BMP, promotes neural vs epidermal fate
  PMID:17950579
  PMID:17950579
  PMID:16860542.
• It must then be switched off for neural differentiation: Suv4-20h/H4K20me3 silences it
  PMID:23382689.
  Expression: PMID:23382689.
• Mammalian Oct4 behaves as a functional homolog in these frog assays (PMID:16860542,
  PMID:17541407), so the anti-differentiation role is a conserved POU-V function.

Biological role 2: neural crest formation (competence factor) — York 2024

• Expression: PMID:39060477
  (Extended Data; note the tension with Nicetto 2013, which describes notoplate restriction by
  midneurula — open issue.)
• GOF: PMID:39060477
• LOF (pou5f3.1 + pou5f3.2 MOs, both homeologs): PMID:39060477
  pax3 reduced, zic1 expanded (neural expansion). Rescue with pou5f3.2 (= this protein) alone:
  PMID:39060477.
• Authors' interpretation is competence/progenitor-state maintenance, not specification:
  PMID:39060477.
• Amniote corroboration (orthologs; not frog): mouse Oct4 reactivated in CNCC precursors and
  needed for ectomesenchyme PMID:33542111;
  chick/human OCT4-SOX2 bind NC enhancers with TFAP2A
  PMID:36182685.
  The amniote data show the POU5 protein acting at crest enhancers — participation, not only necessity.

Layer placement and term choice

• Oct25 is both: a general blastula/gastrula anti-differentiation (germ-layer timing) factor
  — the best-evidenced, core role — and a competence factor whose activity at the neural
  plate border is required for crest formation. It is not a crest specifier: it is expressed
  broadly in ectoderm before the border is defined, alone it promotes neural as well as border
  fates (sox3 expansion, epidermis loss in York 2024), and its LOF reduces border (pax3) as
  well as crest markers. Under the project convention it gets GO:0014029 (not GO:0014036).
• Participation test: Oct25 is a DNA-binding TF expressed in the border/crest cells it
  affects; ortholog OCT4 binds crest enhancers (Hovland 2022). It does part of the work
  (transcriptional regulation in the forming crest), not substrate/consumed. Passes, with the
  caveat that direct frog Oct25 occupancy of crest enhancers is not shown.
• GO:0019827 stem cell population maintenance: not added. In frog there is no
  self-renewing stem cell population maintained by Oct25: the blastula animal cap is transient,
  and Oct25 phenotypes are changes in germ-layer specification and ectodermal fate choice
  (ectopic endoderm, epidermis vs neural), plus delayed neural differentiation when it is not
  silenced. GO:0019827 on mouse Pou5f1 (MGI IMP, ICM/ES self-renewal) does not transfer to this
  biology. Raised as a question for the module's competence knowledge gap.

Comparator check (QuickGO, 2026-10-07)

Query 1 — NC branch on POU5 family members, all evidence:
https://www.ebi.ac.uk/QuickGO/services/annotation/downloadSearch?geneProductId=UniProtKB:Q7T103,UniProtKB:B7ZQA9,UniProtKB:Q91989,UniProtKB:B3DM25,UniProtKB:B3DM23,UniProtKB:Q01860,UniProtKB:P20263,UniProtKB:Q90270,UniProtKB:A7Y7W2&goId=GO:0014029,GO:0014032,GO:0014033,GO:0001755,GO:0014034,GO:0014036&goUsage=descendants&goUsageRelationships=is_a,part_of
(X. laevis Oct-25/Oct-91/Oct-60, X. tropicalis pou5f1.1/pou5f1.3, human POU5F1, mouse Pou5f1,
zebrafish pou5f3, chicken POU5F3) → 0 rows. Positive control with same query on mouse
Tfap2a P34056 and frog id3-a Q91399 returned their known NC rows (GO:0014032 IMP; GO:0014029 IMP x5).

Query 2 — per-taxon NC-branch download (taxonId=7955|10090|9606|8355|8364|9031|10116,
taxonUsage=exact, same goIds) and grep for symbols starting pou/oct → 0 POU-family rows in
any taxon. The same files show the same-layer blastula/crest potency peers do carry NC
terms: zebrafish mych GO:0014032 IMP, zebrafish foxd3 GO:0014032/GO:0001755, X. laevis id3-a
GO:0014029 IMP, snai1 GO:0014036 IMP, foxd3-a GO:0014034 IMP.

Query 3 — GO:0019827 descendants on the same POU5 set: mouse Pou5f1 GO:0019827 IMP (5 papers)
and GO:0030718 germ-line stem cell population maintenance; human POU5F1 GO:0035019 somatic stem
cell population maintenance (IMP/IDA). No frog or fish POU5 carries any.

Query 4 — are the crest papers curated? grep of all nine POU5 annotation sets for
PMID:33542111, 36182685, 39060477 → none.

Interpretation: the absence of NC terms on POU5 proteins is not a convention. The POU5 crest
literature is recent (2021-2024) and entirely uncurated; same-layer peers carry the term. So the
gap is ordinary curation lag, and NEW GO:0014029 on Q7T103 (IMP+rescue, PMID:39060477) is
justified. Analogous to the myc-a outcome.

Evolution

• pou5 is a vertebrate innovation that evolved from a pou3-like ancestor:
  PMID:39060477.
  pou3 inhibits crest and promotes neural plate; pou5 retains the ancestral neural-promoting
  activity (sox3 expansion) and added crest-promoting activity.
• Lamprey pou5 is expressed in animal pole cells but not in crest
  PMID:39060477,
  yet lamprey pou5 rescues frog crest — so the protein activity was present at the vertebrate
  base, and crest expression was either lost in cyclostomes or gained in gnathostomes
  (a cis-regulatory change). Pou5 and Xenopus Pou5f3 cannot engage the lamprey crest GRN.
  This is the mirror image of Id3/SoxE co-option: here the protein is new and its deployment
  in the crest differs between the two vertebrate clades.
• The blastula role (anti-differentiation, partner repression of inducers) is the conserved
  POU-V function shared with mammalian Oct4 (cross-species functional rescue in frog assays).

GOA row decisions (summary)

• MF: DNA binding/TF/cis-regulatory rows accepted. 3 protein binding → MODIFY GO:0140297.
  SMAD binding and DNA-binding TF binding accepted. NEW GO:0003714 corepressor.
• CC: nucleus, transcription regulator complex (repression complexes) accepted.
• BP: negative regulation of activin receptor signalling and BMP signalling, negative regulation
  of epidermal differentiation accepted (core); negative regulation of DNA-templated
  transcription → GO:0000122; gastrulation, neural tube, head, neurogenesis rows non-core;
  A/P axis specification and post-anal tail morphogenesis over-annotated (indirect phenotypes
  of combined Oct-25/Oct-60 knockdown; abstract only). NEW GO:0014029.

Open issues

• Nomenclature: UniProt pou5f1.1 vs Xenbase pou5f3.2 for the same protein.
• Expression conflict: York 2024 places zygotic pou5f3.2 at the NPB/crest; Nicetto 2013 says
  notoplate by midneurula. Needs stage-matched in situ / HCR with paralog-specific probes.
• York's LOF hit pou5f3.1 and pou5f3.2 together; single-paralog requirement untested, although
  pou5f3.2 alone rescues.
• No direct evidence that frog Oct25 binds crest enhancers (OCT4-SOX2-TFAP2A shown in chick/human).