Sox3-A (xSox3, xSox-B1; Xenbase sox3.S) is a SoxB1-group HMG-box transcription factor of Xenopus laevis. Its HMG box binds A/T-rich Sox motifs (AACAAT, AACAAAG), and in the early embryo it occupies thousands of enhancers, often with Pou5f3 factors at Oct-Sox composite sites. It is a maternal factor: the RNA is most abundant in early oocytes and the protein is localised to the animal hemisphere of the egg, cytoplasmic at first and increasingly nuclear as development proceeds. Acting as a sequence-specific repressor, animal Sox3 binds the promoter of the VegT/beta-catenin target Xnr5 and, with induction of Nodal antagonists (Ectodermin, Xema, Coco), restricts nodal-related gene expression to the vegetal hemisphere, setting the normal pattern of germ-layer formation. Acting as an activator, Sox3 together with Pou5f3 drives zygotic genome activation, with differential recruitment to the L and S subgenomes. With Sox2 it is required for the competence of pluripotent blastula animal-pole cells to respond to mesoderm- and endoderm-inducing signals. Sox3 is the only SoxB1 factor expressed throughout naive ectoderm before neural induction; it is induced and maintained by BMP antagonism and FGF, directly activates the early neural genes sox2 and geminin, and is required for BMP-inhibition-mediated neural induction. By the end of gastrulation Sox3 is confined to the neural plate and preplacodal region and is excluded from the neural crest, where SoxE factors take over; forced SoxB1 activity at the neural plate border lowers crest specifier expression, and SoxB1 cannot replace SoxE in crest induction. Later it marks brain, otic vesicle, branchial arches and lens, and in adults is mainly ovarian. SoxB1 blastula and neural functions are ancestral in chordates, and the downregulation of SoxB1 in premigratory crest with a switch to SoxE is conserved in lamprey.
| GO Term | Evidence | Action | Reason |
|---|---|---|---|
| GO:0000122 negative regulation of transcription by RNA polymerase II | IBA GO_REF:0000033 | ACCEPT | Summary: Phylogenetic inference that SoxB1 proteins repress RNA polymerase II transcription; P55863 is itself among the experimental donors (Xnr5 repression). Reason: Sox3 binds the Xnr5 promoter and represses it in embryos and animal caps; loss of function raises Xnr5. The SoxB1 node placement is sound and the target's own IDA grounds it. Maternal repression of nodal-related genes is a core function of Sox3. Supporting Evidence: PMID:14522872 XSOX3 acts as a transcriptional repressor of Xnr5 in both the intact embryo and animal caps injected with VegT RNA. |
| GO:0000122 negative regulation of transcription by RNA polymerase II | IDA PMID:14522872 The beta-catenin/VegT-regulated early zygotic gene Xnr5 is a... | ACCEPT | Summary: Sox3 binds sites in the Xnr5 promoter and represses Xnr5 transcription; Sox3-VP16, morpholino or a DNA-binding-blocking antibody all raise Xnr5. Reason: Direct, DNA-binding-dependent repression of an RNA polymerase II gene, shown by gain and loss of function. The abstract explicitly rules out beta-catenin binding as the mechanism. Core maternal function. Supporting Evidence: PMID:14522872 rather to SOX3 binding to sites within the promoter of the early VegT- and beta-catenin-regulated dorsal-mesoderm-inducing gene Xnr5 PMID:14522872 Expression of a chimeric polypeptide composed of XSOX3 and a VP16 transcriptional activation domain or morpholino-induced decrease in endogenous XSOX3 polypeptide levels lead to an increase in Xnr5 expression |
| GO:0000122 negative regulation of transcription by RNA polymerase II | IEA GO_REF:0000118 | ACCEPT | Summary: TreeGrafter transfer of the SoxB1 repression annotation. Reason: Consistent with the IBA and with direct experimental evidence of Xnr5 repression by Sox3. Supporting Evidence: PMID:14522872 maternal XSOX3 acts in a novel manner to restrict Xnr5 expression to the vegetal hemisphere |
| GO:0000122 negative regulation of transcription by RNA polymerase II | IMP PMID:17608734 The Sox axis, Nodal signaling, and germ layer specification. | ACCEPT | Summary: Review plus new data showing animal Sox3 inhibits Xnr5 and Xnr6 expression. Reason: The abstract states that Sox3 inhibits Nodal-related gene (Xnr5, Xnr6) expression; abstract-only in cache, deferring to the curator on the loss-of-function data. Fits the established repressor role. Supporting Evidence: PMID:17608734 Animally localized Sox3 acts to inhibit Nodal (Xnr5 and Xnr6) expression, and induces the expression of genes (Ectodermin, Xema, and Coco) whose products repress Nodal signaling. |
| GO:0000978 RNA polymerase II cis-regulatory region sequence-specific DNA binding | IBA GO_REF:0000033 | ACCEPT | Summary: Sox-family HMG-box binding to RNA polymerase II cis-regulatory regions. Reason: Sox3 binds the Xnr5 promoter, and V5-tagged Sox3.S (the P55863 sequence) occupies thousands of embryonic enhancers by CUT&RUN, with the SOX3 motif as a top de novo hit. Core MF. Supporting Evidence: PMID:37787392 Homer de novo motif analysis recovered the OCT4 and SOX3 binding sequences as top hits PMID:37787392 To interrogate Pou5f3 and Sox3 chromatin binding across the subgenomes, we performed CUT&RUN on stage 8 embryos injected with mRNA encoding V5 epitope-tagged pou5f3.3.L and sox3.S. |
| GO:0000978 RNA polymerase II cis-regulatory region sequence-specific DNA binding | IEA GO_REF:0000118 | ACCEPT | Summary: TreeGrafter transfer of SoxB1 cis-regulatory DNA binding. Reason: Consistent with the IBA and direct promoter and enhancer binding data. Supporting Evidence: PMID:14522872 rather to SOX3 binding to sites within the promoter of the early VegT- and beta-catenin-regulated dorsal-mesoderm-inducing gene Xnr5 |
| GO:0001228 DNA-binding transcription activator activity, RNA polymerase II-specific | IBA GO_REF:0000033 | ACCEPT | Summary: Sox-family transcription activator activity. Reason: Frog Sox3 directly activates sox2 and geminin without new protein synthesis, and Sox3 with Pou5f3.3 activates zygotic genome activation from bound enhancers. This is the principal MF; the Xnr5 repression is context-specific and is captured at the process level. Supporting Evidence: PMID:18992330 Sox3 functions as an activator to induce expression of the early neural genes, sox2 and geminin in the absence of protein synthesis and to indirectly inhibit the Bmp target Xvent2. PMID:37787392 Together, our findings establish the pluripotency factors Pou5f3.3 and Sox3 as maternal activators of embryonic genome activation, which are differentially recruited to the two homeologous subgenomes of X. |
| GO:0001228 DNA-binding transcription activator activity, RNA polymerase II-specific | IEA GO_REF:0000118 | ACCEPT | Summary: TreeGrafter transfer of activator activity. Reason: Consistent with IBA and frog experimental data. Supporting Evidence: PMID:18992330 Sox3 functions as an activator to induce expression of the early neural genes, sox2 and geminin in the absence of protein synthesis and to indirectly inhibit the Bmp target Xvent2. |
| GO:0001704 formation of primary germ layer | IMP PMID:15302595 Repression of nodal expression by maternal B1-type SOXs regu... | ACCEPT | Summary: Blocking maternal Sox3 DNA binding (antiSOX3c antibody) shifts mesoderm and endoderm markers animally, raises endoderm markers and Xnr5, and causes gastrulation failure, rescued by a Sox3-EnR repressor or by Cerberus-short. Reason: Sox3 does part of the work of germ-layer formation by repressing nodal-related genes in the animal hemisphere (a transcriptional step, not mere necessity). Rescue by a Sox3-EnR repressor and by a Nodal inhibitor places the effect on Sox3 repression of nodal. The antibody may also block Sox3-B; the term is shared by both homeologs (sox3-b by ISS). Core maternal function. Supporting Evidence: PMID:15302595 maternal B1-type SOX functions together with the VegT/beta-catenin system to regulate nodal expression and to establish the normal pattern of germ layer formation in Xenopus. PMID:15302595 In Xenopus, SOX3 acts as a negative regulator of Xnr5, which encodes a nodal-related TGFbeta-family protein. |
| GO:0005634 nucleus | IBA GO_REF:0000033 | ACCEPT | Summary: Nuclear site of action for Sox transcription factors; P55863 is among the donors. Reason: Sox3 acts on chromatin; nuclear localisation increases as development proceeds. Supporting Evidence: file:XENLA/sox3-a/sox3-a-uniprot.txt Nuclear localization |
| GO:0005634 nucleus | IDA PMID:14522872 The beta-catenin/VegT-regulated early zygotic gene Xnr5 is a... | ACCEPT | Summary: Immunolocalisation of Sox3 in the nucleus of embryonic cells (UniProt cites this paper for nuclear and cytoplasmic localisation). Reason: Abstract-only in cache; deferring to the curator, and consistent with Sox3 promoter binding in vivo. Supporting Evidence: file:XENLA/sox3-a/sox3-a-uniprot.txt Nuclear localization |
| GO:0005634 nucleus | IEA GO_REF:0000120 | ACCEPT | Summary: Automated nuclear localisation. Reason: Correct for a DNA-binding transcription factor. Supporting Evidence: file:XENLA/sox3-a/sox3-a-uniprot.txt Nuclear localization |
| GO:0005737 cytoplasm | IDA PMID:14522872 The beta-catenin/VegT-regulated early zygotic gene Xnr5 is a... | KEEP AS NON CORE | Summary: Maternal Sox3 protein is mostly cytoplasmic in early embryos. Reason: Real observation (UniProt cites this paper) reflecting storage or retention of maternal protein before it enters nuclei; not the site of its function. Abstract-only in cache. Supporting Evidence: file:XENLA/sox3-a/sox3-a-uniprot.txt Primarily cytoplasmic in early embryos |
| GO:0005737 cytoplasm | IEA GO_REF:0000044 | KEEP AS NON CORE | Summary: Mapping from the UniProt subcellular location (cytoplasm). Reason: Reflects the early cytoplasmic pool of maternal protein; non-core. Supporting Evidence: file:XENLA/sox3-a/sox3-a-uniprot.txt Primarily cytoplasmic in early embryos |
| GO:0006355 regulation of DNA-templated transcription | IEA GO_REF:0000002 | ACCEPT | Summary: InterPro SOX-family mapping to transcription regulation. Reason: Correct though general; more specific activation and repression terms are also present. Supporting Evidence: file:XENLA/sox3-a/sox3-a-deep-research-falcon.md Depending on regulatory context, Xenopus Sox3 can promote an early ectodermal/neural-progenitor transcriptional program or repress a mesendoderm-promoting target. PMID:14522872 XSOX3 acts as a transcriptional repressor of Xnr5 in both the intact embryo and animal caps injected with VegT RNA. |
| GO:0007420 brain development | IBA GO_REF:0000033 | KEEP AS NON CORE | Summary: SoxB1 phylogenetic inference for brain development. Reason: Plausible (frog sox3 is expressed in the developing brain; mouse Sox3 has forebrain phenotypes), but no frog Sox3 brain-specific experiment exists; its documented neural role is earlier, in neural induction. Kept as non-core, matching the sox2 review. Supporting Evidence: file:XENLA/sox3-a/sox3-a-uniprot.txt Throughout tail bud |
| GO:0007420 brain development | IEA GO_REF:0000118 | KEEP AS NON CORE | Summary: TreeGrafter transfer of brain development. Reason: As for the IBA row. Supporting Evidence: file:XENLA/sox3-a/sox3-a-uniprot.txt Throughout tail bud |
| GO:0008013 beta-catenin binding | IDA PMID:10549281 Regulation of Wnt signaling by Sox proteins: XSox17 alpha/be... | KEEP AS NON CORE | Summary: XSox3 binds beta-catenin and inhibits TCF-mediated signalling. Reason: Informative binding activity, stated directly in the abstract. But Zhang et al. 2003 showed that Sox3's suppression of dorsal axis formation in vivo does not depend on beta-catenin binding; it depends on Sox3 binding the Xnr5 promoter. So the interaction is real but not shown to be the physiological mechanism. Supporting Evidence: PMID:10549281 Two additional Sox proteins, XSox17 alpha and XSox3, likewise bind to beta-catenin and inhibit its TCF-mediated signaling activity. PMID:14522872 this effect is due not to the binding of XSOX3 to beta-catenin nor to its competition with beta-catenin-regulated TCF-type transcription factors for specific DNA binding sites |
| GO:0030154 cell differentiation | IEA GO_REF:0000118 | KEEP AS NON CORE | Summary: Very general TreeGrafter term. Reason: True at a high level (Sox3 influences ectodermal and neural fate) but uninformative; more specific terms capture the biology. Matches the sox2 review. Supporting Evidence: PMID:18992330 Sox3 increases cell proliferation, delays neurogenesis and inhibits epidermal and neural crest formation to expand the neural plate. |
| GO:0030182 neuron differentiation | IBA GO_REF:0000033 | KEEP AS NON CORE | Summary: Pan-bilaterian SoxB1 node for neuron differentiation (fly SoxNeuro/Dichaete, worm, mouse, zebrafish donors). Reason: The node placement is sound and Sox3 is in the clade. In frog, Sox3 acts upstream, keeping neural progenitors proliferative and delaying neurogenesis (gain-of-function data), not driving neuron differentiation. Kept as non-core. Frog sox2 ACCEPTs this term because it has retina-specific differentiation data; Sox3 has none. Supporting Evidence: PMID:18992330 Sox3 increases cell proliferation, delays neurogenesis and inhibits epidermal and neural crest formation to expand the neural plate. |
| GO:0030182 neuron differentiation | IEA GO_REF:0000118 | KEEP AS NON CORE | Summary: TreeGrafter transfer of neuron differentiation. Reason: As for the IBA row. Supporting Evidence: PMID:18992330 Sox3 increases cell proliferation, delays neurogenesis and inhibits epidermal and neural crest formation to expand the neural plate. |
| GO:0030514 negative regulation of BMP signaling pathway | IMP PMID:17950579 The Xenopus POU class V transcription factor XOct-25 inhibit... | KEEP AS NON CORE | Summary: Cited paper is about XOct-25 limiting ectodermal BMP competence; Sox3 is not mentioned in the cached abstract. Reason: Abstract-only; the full text may include Sox3 knockdown data the curator saw, so this is not removed. Independently, Rogers et al. 2009 show Sox3 represses bmp4 and the BMP target Xvent2 and lowers BMP-responsive reporter activity, but indirectly, through Sox3 targets. Kept as a non-core downstream consequence of Sox3's neural-competence role. Supporting Evidence: PMID:18992330 These experiments demonstrate that Sox3 functions as an activator to inhibit expression of bmp4 and Xvent2 and epidermis formation |
| GO:0043565 sequence-specific DNA binding | IDA PMID:14522872 The beta-catenin/VegT-regulated early zygotic gene Xnr5 is a... | ACCEPT | Summary: Sox3 binds Sox sites in the Xnr5 promoter. Reason: Direct DNA-binding evidence; the more specific GO:0000978 is also present. Supporting Evidence: PMID:14522872 rather to SOX3 binding to sites within the promoter of the early VegT- and beta-catenin-regulated dorsal-mesoderm-inducing gene Xnr5 |
| GO:0043565 sequence-specific DNA binding | IDA PMID:9099866 The Xenopus Sox3 gene expressed in oocytes of early stages. | ACCEPT | Summary: Recombinant XSox3 binds AACAAT and AACAAAG motifs. Reason: Direct in vitro binding with recombinant protein. Supporting Evidence: PMID:9099866 Recombinant XSox3 protein produced in Escherichia coli bound specifically to sequences containing the binding motif for the HMG box of SRY or SOX proteins, AACAAT or AACAAAG, demonstrating its sequence-specific DNA binding property. |
| GO:0043565 sequence-specific DNA binding | IDA PMID:9328277 Isolation and characterization of Xenopus laevis xSox-B1 cDN... | ACCEPT | Summary: Recombinant xSox-B1 HMG domain preferentially binds AACAAT. Reason: Direct in vitro binding; xSox-B1 is the same 309-aa protein. Supporting Evidence: PMID:9328277 Recombinant polypeptide of the xSox-B1 HMG domain preferentially binds to the AACAAT sequence. |
| GO:0045892 negative regulation of DNA-templated transcription | IDA PMID:14522872 The beta-catenin/VegT-regulated early zygotic gene Xnr5 is a... | ACCEPT | Summary: Sox3 represses Xnr5 transcription. Reason: Correct; the parent of GO:0000122, which is also annotated from the same paper. Supporting Evidence: PMID:14522872 XSOX3 acts as a transcriptional repressor of Xnr5 in both the intact embryo and animal caps injected with VegT RNA. |
| GO:0045944 positive regulation of transcription by RNA polymerase II | IBA GO_REF:0000033 | ACCEPT | Summary: Sox-family positive regulation of RNA polymerase II transcription. Reason: Supported in frog by direct (cycloheximide-resistant) activation of sox2 and geminin and by Sox3-dependent zygotic genome activation. Supporting Evidence: PMID:18992330 Sox3 functions as an activator to induce expression of the early neural genes, sox2 and geminin in the absence of protein synthesis and to indirectly inhibit the Bmp target Xvent2. PMID:37787392 Down-regulated genes are highly significantly nearer to predicted regulatory elements with enriched Pou5f3 and Sox3 binding |
| GO:0045944 positive regulation of transcription by RNA polymerase II | IEA GO_REF:0000118 | ACCEPT | Summary: TreeGrafter transfer. Reason: Consistent with IBA and frog experimental data. Supporting Evidence: PMID:18992330 Sox3 functions as an activator to induce expression of the early neural genes, sox2 and geminin in the absence of protein synthesis and to indirectly inhibit the Bmp target Xvent2. |
| GO:0090090 negative regulation of canonical Wnt signaling pathway | IDA PMID:17875931 Sox17 and Sox4 differentially regulate beta-catenin/T-cell f... | KEEP AS NON CORE | Summary: Cited paper tests Sox proteins on beta-catenin/TCF reporter activity in colon carcinoma cells; Sox3 is not named in the abstract. Reason: Abstract-only and full text not retrievable via PMC; the abstract says some Sox proteins antagonise TCF activity, so Sox3 may be among those tested, and the row is not removed. Zorn et al. 1999 independently show XSox3 inhibits TCF-mediated signalling. However, in the embryo Sox3 suppresses beta-catenin-driven axis formation by repressing Xnr5, not via beta-catenin, so Wnt-pathway antagonism is a reporter-level, non-core activity. Supporting Evidence: PMID:17875931 We found that some Sox proteins antagonize while others enhance beta-catenin/T-cell factor (TCF) activity. PMID:10549281 Two additional Sox proteins, XSox17 alpha and XSox3, likewise bind to beta-catenin and inhibit its TCF-mediated signaling activity. |
| GO:0141064 zygotic genome activation | IMP PMID:37787392 Hybridization led to a rewired pluripotency network in the a... | NEW | Summary: Proposed: maternal Sox3, with Pou5f3, activates zygotic gene expression at the maternal-to-zygotic transition, binding enhancers of the genes it activates. Reason: Phelps et al. 2023 performed CUT&RUN with V5-tagged Sox3.S (NM_001090679 = P55863, so the binding data are specific to this homeolog) and morpholino knockdown with RNA-seq, including under cycloheximide to isolate primary activation. Genes that fail to activate lie near Sox3/Pou5f3-bound enhancers, whereas up-regulated genes do not. Participation: Sox3 is the sequence-specific activator bound at the regulatory elements of activated genes, so it performs a step of the process. Comparator (QuickGO 2026-10-07): zebrafish sox19b (Q9DDD7, a maternal SoxB1) and zebrafish pou5f3 both carry GO:0141064 by IMP (PMID:35145080); same role, same term. Caveat: the strongest loss-of-function effects need combined sox3 and pou5f3 morpholinos. Supporting Evidence: PMID:37787392 Together, our findings establish the pluripotency factors Pou5f3.3 and Sox3 as maternal activators of embryonic genome activation, which are differentially recruited to the two homeologous subgenomes of X. PMID:37787392 Down-regulated genes are highly significantly nearer to predicted regulatory elements with enriched Pou5f3 and Sox3 binding |
| GO:0021990 neural plate formation | IMP PMID:18992330 Xenopus Sox3 activates sox2 and geminin and indirectly repre... | NEW | Summary: Proposed: Sox3 is required in naive ectoderm for neural induction by BMP inhibition and directly activates early neural plate genes. Reason: Sox3 is the only SoxB1 expressed throughout the ectoderm before neural induction. A Sox3 morpholino blocks Noggin-induced neural markers in explants, and MO-resistant sox3 RNA rescues. Sox3 directly (cycloheximide-resistant) activates sox2 and geminin. Combined Sox2/Sox3 depletion blocks Chordin neural induction, rescued by Sox3 (Buitrago-Delgado 2018). GO has no neural induction term; GO:0021990 is defined as formation of the neural plate from ectoderm in response to signals from dorsal mesoderm, which is this process. Participation: Sox3 is the transcription factor that activates the neural plate programme in responding cells. Comparator (QuickGO 2026-10-07): no SoxB1 carries GO:0001840 or its descendants, but frog sox2 carries the broader GO:0007399 by IMP for equivalent neural-induction experiments, and zebrafish and mouse Sox3 carry nervous-system descendants. The gap is one of granularity, not a convention against SoxB1 in neural induction. Supporting Evidence: PMID:18992330 With morpholino-mediated knockdown of Sox3, we demonstrate that it is required for induction of neural tissue by BMP inhibition. PMID:18992330 We were able to rescue this phenotype by injecting sox3 mRNA in which the Sox3MO annealing sequence was mutated PMID:30144418 We first showed that morpholino-mediated depletion of Sox2 and Sox3 prevented chordin-mediated neural induction (Figure 6A–B). |
Loading supporting content…
Download this section (compressed HTML)Q: Combined Sox2 and Sox3 morpholinos abolish the competence of blastula animal-pole cells to form mesoderm and endoderm (Buitrago-Delgado 2018). Only mammalian SOX2 carries stem cell population maintenance terms; no SOX3 does in any species (QuickGO 2026-10-07). Should frog Sox3 carry such a term for blastula pluripotency? The same question is open for sox2, id3-a, myc-a, snai1 and lin28a.
Q: Forced SoxB1 activity at the neural plate border lowers foxd3 and snail2, and Sox3 overexpression delays crest specification. But no Sox3 loss of function expands the crest, and no SoxB1 protein in any species carries a neural crest process term, positive or negative (QuickGO 2026-10-07). Is there loss-of-function evidence that endogenous Sox3 limits crest formation (which would support GO:0090301), or is Sox3's exclusion from the crest a consequence of crest specifiers repressing it (cf. Snail2–Sox3 reciprocal repression in amniote gastrulation)?
Q: Sox3 represses Xnr5 through DNA binding but activates sox2, geminin and zygotic genes. Should it also carry GO:0001227 (repressor activity), or GO:1900108 negative regulation of nodal signaling pathway (no SoxB1 carries it yet), or is repression partner-dependent and best left at the process level?
Q: The module knowledge gap describes OCT4–SOX2 redeployed to TFAP2A-bound crest enhancers. In frog, the blastula Oct–Sox pair is Pou5f3–Sox3 at genome-activation enhancers, and Sox3 is excluded from definitive crest by neurula. Does any Oct–SoxB1 complex act at frog crest enhancers, or does Oct25 partner with SoxE there?
Q: All experimental rows are on sox3-a (sox3.S); sox3-b (Q5FWM3, presumably sox3.L) carries only ISS copies, and UniProt lists "sox11" as a synonym for sox3-b. Should the synonym be corrected, and do the two homeologs differ in maternal versus zygotic expression?
Experiment: F0 CRISPR knockout of sox3.L, sox3.S or both (maternal-effect via oocyte host transfer), scoring zygotic gene activation by RNA-seq and animal-cap competence to respond to activin.
Hypothesis: Sox3 alone, not only SoxB1 jointly, is required for blastula competence and zygotic genome activation, and Sox3.L and Sox3.S act redundantly.
Type: CRISPR knockout
Experiment: CUT&RUN for Snai2, Sox9/10 and Sox3 in dissected border tissue at stages 12-14, combined with border-restricted sox3 knockdown or knockout to test whether the crest domain expands.
Hypothesis: Sox3 must be actively repressed at the neural plate border, and its persistence there blocks crest specification.
Type: ChIP-seq
Loading supporting content…
Download this section (compressed HTML)Loading supporting content…
Download this section (compressed HTML)Loading supporting content…
Download this section (compressed HTML)