zic1 (Xenopus laevis, O73689) — curation notes
Project: projects/NEURAL_CREST_ORIGINS.md, Tier 2 (neural plate border specifiers).
Reviewer: claude-code, 2026-10-05.
Identity
- O73689 = Zic1 / Opl / Zic-r1, 443 aa, GLI-like C2H2 zinc-finger protein (five C2H2 fingers 221-380; ZOC/ZF-N
region; C-terminal regulatory region). Xenbase cross-reference in the UniProt record is zic1.S
(XB-GENE-864815); the L homeolog zic1.L is unreviewed TrEMBL A0A1L8GAZ5 (443 aa) and carries no experimental
GO rows. All experimental GO rows sit on O73689, while the reagents (mRNAs, the Sato 2005 morpholino) were
generic "Zic1" reagents; not "fixed" here by assertion.
- Family: Zic1-5, sister to GLI/Ci within PANTHER PTHR45718. Drosophila ortholog = odd-paired (opa).
Evidence synthesis (layer placement)
1. Expression
2. Molecular activity
- Transcriptional activator with a C-terminal regulatory (inhibitory) domain.
PMID:9655809
- Sequence-specific DNA binding to a conserved snail1 upstream element (EMSA, mutant oligo control).
PMID:24360906
Mouse Zic1 binds a GLI-like site PMID:11053430.
- => MF:
GO:0001228 DNA-binding transcription activator activity, RNA polymerase II-specific (NEW; existing MF rows
stop at GO:0000981). No frog evidence for repressor activity, so no repressor term.
3. Direct targets
4. Gain of function
5. Loss of function
- Zic1 MO: loss of foxd3/slug, rescued by zic1 (Sato 2005; numbers in deep research); co-presence required:
PMID:15843410
- Zic1 MO: loss of PE (Six1, preplacodal Sox2) and crest Snail2 induction:
PMID:17409353
PMID:17409353
- Zic1+Zic3 jointly required for definitive neural fate:
PMID:19805078
6. Conservation / outgroups
- Amphioxus: Zic and Pax3/7 mark the neural plate border, but crest specifiers are not expressed there.
PMID:18562679
PMID:18562679
- Interpretation: Zic belongs to the ancestral chordate border (and neural plate) layer. The vertebrate innovation is
the downstream wiring: a Zic1 site in the snail1 element and Pax3+Zic1 co-activation of snail/foxd3/sox8/twist1.
Whether that is cis-regulatory (target elements) or protein change is untested (suggested experiment).
- Drosophila opa regulates wingless/engrailed; frog Zic1 -> wnt1 -> en-2 parallels this PMID:16892174.
- Mouse: Zic1 mutants have cerebellar defects; mammalian crest induction does not rely on Zic1 alone (paralog
redundancy) — deep research note; not used as evidence for annotations.
Network-layer decision
Zic1 is a neural plate border specifier (Tier 2), not a crest specifier and not a competence factor:
- It is upstream of and directly activates the first-wave crest specifiers (snail1, sox8 via Pax3+Zic1).
- Alone it does not make crest; its output at the border depends on partner level (with Pax3 -> crest; without
Pax3 -> preplacodal ectoderm). That is the definition of a border specifier.
- It also has an earlier, separate role as a BMP-inhibition-responsive neural plate gene (with Zic3) and later roles
in dorsal neural tube and midbrain/hindbrain patterning (via Wnt). These are real but non-core relative to the
project question; neural plate development is kept as core (the neural plate and its border are where all Zic1
outputs are executed).
NC-branch term decision (GO:0014029 vs GO:0014034 vs GO:0014036)
GO:0014029 neural crest formation = formation of the ectodermal region between neural plate and non-neural
ectoderm. For a border specifier that is exactly the right level: ACCEPT all four IMP rows + IEA.GO:0014034 neural crest cell fate commitment: Pax3+Zic1 are sufficient for full determination and directly
activate specifiers; Zic1 does the work (DNA-binding activator of snail1). ACCEPT the three rows — matches
pax3-a (parallel review) and foxd3-a.- Not adding
GO:0014036 fate specification: it is part_of GO:0014034 (redundant), and the project convention
reserves it for the downstream specifiers (sox8/9/10, snai2, twist1) that execute specification.
GO:0014033 neural crest cell differentiation (IMP PMID:9655809, gain-of-function marker induction) —
MODIFY to GO:0014034: Zic1 initiates the program; differentiation is executed by downstream factors.
Comparator check (QuickGO, 2026-10-05)
Query (experimental + TAS only, exact terms):
https://www.ebi.ac.uk/QuickGO/services/annotation/downloadSearch?goId=GO:0014029,GO:0014034,GO:0014036,GO:0014033,GO:0014032,GO:0001755&goUsage=exact&evidenceCode=ECO:0000269,ECO:0000304&evidenceCodeUsage=descendants
(805 rows), filtered for symbols matching zic|pax3|pax7|msx1|msx2|tfap2a|gbx2|hes4|dlx5:
- X. laevis zic1: GO:0014029 x4, GO:0014034 x3, GO:0014033 x1 (this gene).
- X. laevis zic2-a: GO:0014029 x2, GO:0014034 x1; zic3: GO:0014034; zic4: GO:0014029; zic5: GO:0014029.
- Zebrafish zic2a, zic2b: GO:0014032 neural crest cell development (IMP).
- X. laevis pax3-a/-b: GO:0014029, GO:0014034; hes4-a/-b: GO:0014029.
- Zebrafish tfap2a: GO:0014036 (16) and GO:0014032 (32); mouse Tfap2a: GO:0014032; mouse Pax3, Gbx2: GO:0001755.
- Mouse/human Zic1: no experimental NC-branch annotation.
Conclusion: Xenopus border specifiers (Zic family, Pax3, Hes4) are consistently at GO:0014029/GO:0014034 — no
Zic in any species carries GO:0014036. Keeping zic1 at GO:0014029 + GO:0014034 is consistent with the curation
convention for this layer. Only zebrafish tfap2a (a border/competence gene) sits at GO:0014036 — an
inconsistency to note for the TFAP2A review, not a reason to change zic1.
Preplacodal comparator (GO:0060788 ectodermal placode formation, experimental, exact): zebrafish tfap2a, tfap2c,
gata3, foxi1, fgf8a; X. laevis tbx1-a, ripply3, lsm14a-a. Preplacodal competence/border factors do carry this term,
so a NEW GO:0060788 for Zic1 (necessary and sufficient for PE fate, PMID:17409353, full text) follows convention.
Participation: Zic1 is the transcription factor that activates the PE specifiers six1/eya1 (it does work in the
process). There is no "preplacodal ectoderm" term.
GO gaps / questions
- No neural plate border formation/specification term (project-level; raise as question). GO:0021998 neural plate
mediolateral regionalization exists but is only annotated to mouse/rat Bmpr1a; not proposed.
- PMID:9435279 (Mizuseki 1998) is the source of a GO:0014029 IMP row but its abstract only describes neural
induction; full text not available here — ACCEPT deferring to curator (consistent with GoF neural crest data in
sibling papers), flagged.
Other row decisions (summary)
- Cytoplasm (IEA SubCell / ISS from mouse Zic1 P46684): mouse evidence is I-mfa-mediated cytoplasmic retention in
cultured cells PMID:15207726;
KEEP_AS_NON_CORE.
- Positive regulation of protein import into nucleus (ISS from mouse Zic1, PMID:11238441 Gli co-translocation on
overexpression): MARK_AS_OVER_ANNOTATED — Zic acts as a binding partner that carries Gli, not a regulator of
import machinery; no frog evidence.
- Positive regulation of Wnt signaling (IGI PMID:16892174): Zic1 transcriptionally induces wnt1/4/8b; KEEP_AS_NON_CORE
(regionalisation output; mechanism is ligand gene expression).
- CNS development (IBA), D/V neural tube patterning, neurogenesis: KEEP_AS_NON_CORE.