FoxN2/3 (Strongylocentrotus purpuratus, UniProt Q2V887) - curation notes

Identity

• UniProt Q2V887 (Q2V887_STRPU, unreviewed/TrEMBL, 197 aa, flagged as a
  Fragment with a non-terminal C-terminus) is the S. purpuratus FoxN2/3 cDNA
  deposited by the Davidson lab (EMBL DQ286744 / ABB89482.1, GN Name=FoxN2/3).
  Its single RX line is PMID:17081512, the genome-wide Forkhead survey of Tu,
  Brown, Davidson & Oliveri (2006) that defined the 22 S. purpuratus fox genes.
  NCBI GeneID 590815 (LOC590815, "forkhead transcription factor N2/3", alias
  FoxN2/3; checked via NCBI gene esummary, 2026-10-04) is the same locus.
• The automated RecName "Forkhead box protein N3" and the ARBA function note
  ("Acts as a transcriptional repressor. May be involved in DNA damage-inducible
  cell cycle arrests") are family-level transfers from vertebrate FOXN3/CHES1 and
  are not sea urchin data. The sea urchin gene is equally related to vertebrate
  FOXN2 and FOXN3, hence the community symbol FoxN2/3 (also written foxN2/3,
  foxn2/3, Sp-FoxN2/3). The FoxN subfamily has only two sea urchin members:
  PMID:21303847. Because the
  accession is the Davidson-lab FoxN2/3 clone itself, the identity between the
  accession and the GRN gene is not in doubt (confidence: high).
• Domain content: a single Fork-head DNA-binding domain (PROSITE PS50039, residues
  86-173; CDD cd20059 FH_FOXN3; InterPro IPR047119 "FOXN2/3-like"). PANTHER
  PTHR13962:SF22, whose official label (from interpro/panther/panther.obo) is
  "FORKHEAD BOX PROTEIN N3-LIKE PROTEIN" (family PTHR13962 "FORKHEAD BOX PROTEIN
  N3-LIKE PROTEIN-RELATED").
• Tu et al. 2006 is abstract-only in the cache; the abstract describes the
  survey, not foxN2/3 specifically: PMID:17081512.

Organism caveat for the functional literature

• The only functional (perturbation) studies of foxN2/3 are from the McClay lab
  and were done in Lytechinus variegatus, not S. purpuratus:
  PMID:21303847 and
  PMID:24598159. The authors explicitly matched the Lytechinus
  expression pattern to the published S. purpuratus pattern: PMID:21303847.
  The two species are both camarodont euechinoids with essentially the same PMC
  GRN architecture, so the data are used here as the primary functional evidence
  for the S. purpuratus orthologue, but every functional annotation derived from
  them is flagged as inferred across species.
• S. purpuratus-specific data are expression-level only: the Davidson-lab
  nCounter time course measured foxn2/3 transcript prevalence through
  mid-gastrula PMID:20398801, and Materna & Davidson note that in S. purpuratus foxN2/3
  is first skeletogenic and then switches to NSM: PMID:23261933.
• Note: the brief listed Oliveri, Tu & Davidson 2008 (PMID:18413610) and Rafiq
  et al. 2014 (PMID:24496631) as placing foxN2/3 in the PMC GRN. The cached
  text of PMID:18413610 does not name foxN2/3 anywhere (it covers foxb and
  foxo), and the cached abstract of PMID:24496631 does not name it either, so
  neither is cited in the review YAML. Shashikant & Ettensohn (2018) do
  attribute the beta-catenin/Pmar1 dependence of foxN2/3 to Oliveri 2008 (see
  below), so the data are probably in that paper's supplementary material.

Expression (where and when)

• Transient expression in the large-micromere / primary mesenchyme cell (PMC)
  lineage, then NSM, then endoderm: PMID:21303847. Double FISH
  with tbr (PMC) and gcm (NSM) confirmed the lineage assignments: PMID:21303847 and
  PMID:21303847.
• Listed among the regulatory genes expressed selectively in the PMC lineage in
  the 2018 synthesis of the S. purpuratus skeletogenic GRN: PMID:30264451.
• Later gut expression (Lytechinus; reported as not previously described in
  S. purpuratus): "foxN2/3 mRNA was expressed in the endoderm from the early
  gastrula stage and remained in the mid- and hindgut area through the late
  gastrula stage".
• The brief's mention of oral-ectoderm expression is not supported by anything
  in the cached literature; the documented later domains are NSM and endoderm.

Place in the GRN: inputs

• Downstream of beta-catenin and the Pmar1/HesC double-negative gate:
  PMID:21303847; PMID:30264451.
• Immediate upstream regulators in micromeres are Ets1 and Tbr, not Alx1:
  PMID:21303847; PMID:21303847;
  PMID:21303847; PMID:21303847. Tbr is necessary but not
  sufficient (AND logic): PMID:21303847. The brief's statement that foxN2/3 lies downstream of alx1 is
  therefore not correct for the micromere phase; it lies downstream of
  ets1/tbr.
• The later NSM/endoderm phase is independent of the micromere phase, of
  Delta/Notch and of the Blimp1 torus subcircuit: PMID:21303847; the authors infer
  PMID:21303847. No cis-regulatory module has actually been isolated.
• No autoregulation: foxN2/3 transcription is unchanged in foxN2/3 morphants
  ("QPCR results also showed that the expression level of foxN2/3 was not
  significantly changed by foxN2/3 MASOs").

Place in the GRN: outputs and loss-of-function phenotype

• Two non-overlapping translation-blocking morpholinos with 5'UTR-RFP
  specificity controls and mRNA rescue: PMID:21303847.
• Skeletogenic effector genes require FoxN2/3 early: PMID:21303847;
  PMID:21303847; msp130 protein absent at 12 h. VEGFR is also a target:
  PMID:21303847. All of these recover later, so FoxN2/3 is an early
  input, not a maintenance factor: PMID:21303847.
• Ingression (EMT): PMID:21303847. Saunders & McClay 2014 confirmed foxn2/3
  among the 13 PMC transcription factors required for EMT and found it has the
  strongest apical-constriction / de-adhesion defect: PMID:24598159; PMID:24598159.
• Cell-autonomous skeletal defect shown by micromere transplantation:
  PMID:21303847;
  syncytium: PMID:21303847.
• Non-autonomous: morphant PMCs fail to send the signal that normally
  suppresses NSM transfating: PMID:21303847. This is an indirect, signalling-mediated effect and is not
  annotated.
• Summary: PMID:21303847; PMID:21303847.

Curation decisions

• Electronic rows: GO:0003700 (DNA-binding TF activity), GO:0006355, GO:0005634
  are consistent with a Forkhead TF whose knockdown abolishes early expression
  of a battery of PMC effector genes; accepted (nucleus and TF activity) or
  kept as non-core generic parent (GO:0006355). GO:0003677 is over-general
  (MODIFY to GO:0043565 which already exists as a separate row, which is
  accepted). The UniProt ARBA "transcriptional repressor" note is a vertebrate
  FOXN3 transfer; in the sea urchin the only direction shown is activation
  (loss of target expression), and direct binding to any target CRM has not
  been shown, so the sign-specific MF child is NOT proposed.
• NEW rows kept minimal because the functional data are Lytechinus-only:
  GO:0045944 positive regulation of transcription by RNA polymerase II (IMP;
  skeletal-matrix genes, msp130, VEGFR lost in morphants), GO:0010718 positive
  regulation of epithelial to mesenchymal transition (IMP; two independent
  studies), GO:0070169 positive regulation of biomineral tissue development
  (IMP; cell-autonomous failure of transplanted morphant PMCs to contribute to
  the skeleton, loss of sm30/sm50/msp130). Participation test: FoxN2/3 is the
  TF that drives these effector genes, so it does the regulatory work.
  Comparator: Alx1 and Ets1 reviews in this directory carry GO:0010718 and
  GO:0070169 for the same role.
• Not proposed: mesodermal cell fate specification (foxN2/3 is activated after
  the lineage is specified, and alx1/tbr are expressed normally in morphants,
  so it is a differentiation input, not a specification factor); NSM or
  endoderm roles (expression only, no perturbation of those phases); syncytium
  formation / cell fusion (no suitable term established for echinoderm PMC
  fusion, and the mechanism is unknown); regulation of VEGF signalling (VEGFR
  transcript loss is one of many downstream effector losses).