SPEN is the only member of the real-gap floor whose epigenetic readout is direct and published: its SPOC domain recruits NCoR/SMRT, NuRD and the m6A machinery to active enhancers and promoters, and it disengages from chromatin the moment silencing succeeds
also called SPEN SHARP MINT SPOC · q181 SPEN
SPEN is a repressor that latches onto genes while they are still switched on, pulls in enzymes that strip acetyl groups off histones and remodel the packaging, and then lets go once the gene is quiet. It is the one protein in this beat's list of true gaps that has a clear, published chromatin job.
The same thing, for a scientist
gene-spen is the outlier in this beat's eight-entry sampling-free real-gap floor: the other seven are the P-TEFb reserve/release axis, and SPEN is a corepressor with a direct, well-characterised chromatin mechanism that simply was never written into its entry. Dossin and colleagues showed that SPEN is recruited to the X chromosome immediately on Xist upregulation, is targeted to enhancers and promoters of ACTIVE genes, and rapidly disengages from chromatin on silencing - implying that active transcription tethers it. They defined the SPOC domain as the major effector of silencing, showed that tethering SPOC to Xist RNA suffices, and identified SPOC's partners as NCoR/SMRT, the m6A RNA methylation machinery, the NuRD complex, RNA Pol II, and initiation and elongation regulators. So SPEN's epigenetic readout is histone deacetylation and nucleosome remodelling recruited to active regulatory elements - a fully RECORDED state, not a bounded absence, and the annotation gap here is purely notational.
The mechanism, and the unusual part of it
1:
- SPEN is essential for initiating gene silencing on the X in preimplantation mouse embryos and ES cells, and dispensable for maintenance in neural progenitors, though it still reduces expression of escapee genes.
- It is recruited to the X immediately on Xist upregulation, and is targeted to enhancers and promoters of ACTIVE genes.
- It rapidly disengages from chromatin on silencing, suggesting active transcription tethers it.
- The SPOC domain is the major silencing effector, and tethering SPOC to Xist RNA is sufficient to silence.
- SPOC's partners: NCoR/SMRT, the m6A RNA methylation machinery, the NuRD complex, RNA Pol II, and transcription initiation and elongation regulators.
The disengagement result is the interesting one for this corpus. A repressor that binds active chromatin and leaves once the gene is off is a transient occupant, which means a steady-state ChIP in a quiescent locus will find nothing -- an assay design trap that generalises well beyond SPEN and beyond X-inactivation.
Why it was in the gap list, and what that says about the gap list
SPEN's data{} in this corpus carries localization: "Nucleus" and a subnuclear note about X-inactivation, and nothing on the epigenetics dimension: no field, no mark token, no chromatin concept term. That is a pure notation gap on a protein whose mechanism is histone deacetylation.
It is also a useful check on this beat's own instrument. The eight-member floor was defined as 'names nothing in any vocabulary tested', and the counter-evidence on ana-q181-the-only-chromatin-machinery-with-genuinely-nothing-is-the-ptefb-brake warned that reading the entries end to end could shrink it. SPEN is the case where that warning cashes: its entry really does say nothing on this dimension, so the floor holds at eight, but the reason it says nothing is that nobody wrote it down -- not that the knowledge is missing. Even inside the sampling-free floor, the schema-artefact explanation is present.
Relevance and its limits
SPEN's connection to this project's index disease is not established and this entry does not manufacture one. The evidence is mouse X-inactivation. What makes it worth writing here is:
- It is a corepressor bridging an RNA to NCoR/HDAC3 and NuRD -- the same architecture as several fusion-adjacent corepressor arms this corpus already holds.
- Its partner list includes the m6A machinery, which is the dimension-boundary question
ana-q181-the-only-chromatin-machinery-with-genuinely-nothing-is-the-ptefb-brakeleaves open: several m6A entries (FTO, WTAP, YTHDC1, ZC3H13) sit in this corpus's empty stratum and were excluded from the machinery list by fiat. - It is an X-linked-silencing factor in a disease with a documented sex skew in infant presentation -- a connection that is speculative and is flagged as such, not asserted.
| epigenetic readout | Corepressor recruitment to active enhancers and promoters: the SPOC domain binds NCoR/SMRT (histone deacetylase corepressor), the NuRD complex (nucleosome remodelling and deacetylation) and the m6A RNA methylation machinery, and is necessary and sufficient for Xist-directed silencing (PMID 32025035). SPEN occupies chromatin only while the target is actively transcribed and disengages on silencing. |
|---|---|
| epigenetic mark named | Histone acetylation, as the substrate REMOVED - via NCoR/SMRT-HDAC3 and NuRD recruitment. No specific residue is identified in the source. |
| eraser of that mark | Indirect: SPEN itself has no catalytic activity; it recruits HDAC-containing corepressor complexes. |
| epigenetic bounded absence | No specific histone residue is named in the cited work, and no SPEN ChIP, CUT&RUN or histone-mark measurement exists in any leukaemia system or in any KMT2A-rearranged cell. |
| epigenetics not applicable | false |
| uniprot | Q96T58 |
| shown for fusion | NO - nobody has looked in any KMT2A fusion. The entire evidence base is mouse X-inactivation. |
| assay design trap | SPEN occupies chromatin only at ACTIVE targets and disengages on silencing, so a steady-state ChIP at an already-silenced locus returns nothing. An absence of SPEN signal is not an absence of SPEN function. |
measurements
[
{
"quantity": "real-gap-floor members whose epigenetic readout is direct and already published",
"value": 1,
"units": "of 8",
"subject": "SPEN, against the seven P-TEFb-axis members whose readout is complex-level or indirect",
"universe": "the eight-entry sampling-free floor in ana-q181-the-only-chromatin-machinery-with-genuinely-nothing-is-the-ptefb-brake",
"null": null,
"note": "so even inside the floor, at least one entry is a pure notation gap rather than a knowledge gap"
}
]- SPEN's mechanism runs through HDAC-containing corepressors, and HDAC inhibitors are in active use in this disease - this corpus holds panobinostat and chidamide entries. An HDAC inhibitor would be expected to block SPEN-directed silencing along with everything else, so any SPEN-dependent tumour-suppressive silencing in a blast is collateral damage from a drug given for another reason. Nobody has measured whether that matters here, and the absence of a SPEN record in this corpus is why the question has not been asked.
- The assay-design trap is a bite-back on evidence rather than on therapy, and it generalises: a factor that leaves chromatin when its job succeeds will be scored absent by exactly the steady-state ChIP a coverage-filling exercise would reach for. Filling the epigenetics field for SPEN from a negative ChIP would record a false negative as a measurement.
- This entry links SPEN to a disease with a documented sex skew in infant presentation only as a flagged speculation. Treating that as a lead would be building on a coincidence of X-linkage, and it is named here so that a later reader does not find the association in the graph and assume it was argued.
- Is SPEN expressed in KMT2A-rearranged blasts, and does it occupy any fusion target? Nothing has been measured; the entire evidence base is mouse X-inactivation.
- Is the deacetylase mechanism required, or only present? A SPOC mutant retaining NCoR binding but losing silencing, or silencing under HDAC3 inhibition, would separate them, and neither experiment was searched for.
- SPOC binds the m6A machinery. Several m6A entries in this corpus (gene-fto, gene-wtap, gene-ythdc1, gene-zc3h13) sit in the epigenetics-empty stratum and were excluded from this beat's chromatin-machinery list by fiat. Is RNA modification part of the epigenetics dimension? SPEN is the entry where the two meet and the boundary has to be decided.
- Run the bare-operand controls and check data/symbol-collisions.json for SPEN, SHARP and MINT before any further literature claim. SHARP is an obviously collided token and the registry was not consulted for this entry.
- E1strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
SPEN is essential for initiating X-linked gene silencing in preimplantation mouse embryos and ES cells but dispensable for maintenance in neural progenitors; it is recruited to the X chromosome immediately upon Xist upregulation and is targeted to enhancers and promoters of ACTIVE genes, disengaging rapidly from chromatin upon gene silencing; the SPOC domain is the major effector of its silencing function and tethering SPOC to Xist RNA is sufficient to mediate silencing; and SPOC's protein partners include NCoR/SMRT, the m6A RNA methylation machinery, the NuRD complex, RNA polymerase II and regulators of transcription initiation and elongation.
SPEN integrates transcriptional and epigenetic control of X-inactivation · Dossin F, Pinheiro I, Zylicz JJ, Roensch J, Collombet S, Le Saux A, Chelmicki T, Attia M, Kapoor V, Zhan Y, Dingli F, Loew D, Mercher T, Dekker J, Heard E · Nature 578(7795):455-460 2020What this evidence cannot show
Abstract-level read; the full text was not retrieved. The system is mouse X-inactivation - preimplantation embryos, embryonic stem cells and neural progenitors - and expected_absent_terms records that this is NOT a leukaemia or KMT2A paper, which is part of what the entry asserts. The SPOC partner list comes from a proteomic pull-down and the abstract does not state which partners are required for silencing; NCoR/SMRT and NuRD are named as partners, and the deacetylase mechanism is the authors' proposal ('we propose that SPEN acts as a molecular integrator'), not a demonstrated requirement.
Evidence that argues against this (2)
- The deacetylase readout this entry records as RECORDED is weaker than the label suggests: the paper demonstrates that SPOC is necessary and sufficient for silencing and that it BINDS NCoR/SMRT and NuRD, but a proteomic partner list plus a proposal is not a demonstration that histone deacetylation is the silencing mechanism. A SPOC point mutant that retains NCoR binding but loses silencing, or silencing that survives HDAC3 inhibition, would leave the partner list intact and remove the mechanism. Neither experiment was searched for.
- No paper disagreeing with the SPEN/SPOC silencing model was located, but the search was one PubMed query on the paper's own title terms; the bare-operand controls on SPEN, SHARP and MINT were not run, and SHARP in particular is a badly collided token. Per this project's Rule 88 the symbol-collision registry was not consulted for SPEN/SHARP/MINT before these searches, so the absence of contradiction here is unchecked rather than clean.
supplies the epigenetics record gene-spen lacks: SPOC recruits NCoR/SMRT and NuRD to active enhancers
the one floor member whose readout is direct and published, so even inside the floor the gap is partly notational