PSIP1-SDC1 - DECLINED as an interaction, kept for the splicing thread: nuclear syndecan-1 co-IP in a mesothelioma line recovers LEDGF among an RNA-splicing-enriched network
also called LEDGF-SDC1 · PSIP1-syndecan-1 · nuclear syndecan-1
Syndecan-1 is normally a cell-surface molecule, but in mesothelioma cells some of it travels to the nucleus. When researchers pulled it down there, LEDGF came with it, inside a network full of RNA-splicing proteins.
The same thing, for a scientist
**Verdict: DECLINED as an interaction; RECORDED for the splicing thread.** IntAct holds one PSIP1-SDC1 record: **anti-BAIT co-IP** - so the pull-down was through SDC1 with an antibody against SDC1 itself rather than a tag - *physical association*, host recorded as **human mesothelioma cell line STAV-AB**, PSIP1 as prey, intact-miscore 0.40, from **PMID 32664515** [E1]. Kumar-Singh et al., *Biomolecules* 2020, co-immunoprecipitated SDC1 in a mesothelioma line, state that *the precipitates contained a large number of proteins, indicating the recovery of protein networks*, found the nuclear-protein fraction enriched for **cell proliferation and RNA synthesis, splicing and transport**, verified the top RNA-splicing candidates by co-IP and Western blot, and showed SDC1 loss- and gain-of-function changes RNA levels [E2]. **The authors' own description - a large number of proteins, protein networks - is the decline.** What survives is the observation that LEDGF sits in a splicing-enriched nuclear network, which is a recurring and under-examined shape in this census.
The decline, and the one thing this row has that most do not
| criterion | verdict |
|---|---|
| replication | no - one record, one publication |
| the source is about this pair | no - PSIP1 is not named in the abstract |
| endogenous bait, untagged | yes - anti bait coip against native SDC1 |
| non-standard, disease-relevant cell line | yes - STAV-AB mesothelioma, not HEK293 or U2OS |
| the authors themselves call it a network recovery | yes, and that is the decline |
Two criteria pass that almost nothing else in this census passes. The bait is endogenous and untagged, and the cell is neither embryonic kidney nor osteosarcoma. This corpus declines most prey rows on tagged over-expression in a non-expressing line, and neither objection is available here.
It is declined anyway, on the authors' own framing: a co-IP that recovers a large number of proteins and is analysed as a network is a network experiment, and a single member of such a network is a candidate. LEDGF is additionally one of the most frequently recovered proteins in nuclear pull-downs generally - this corpus already records it as prey to nine of fifty RBP baits in an unrelated map.
The splicing thread, which recurs across this census and is not being followed
Count the splicing machinery among KMT2A's and PSIP1's uncovered prey-only partners: EFTUD2, SF3B4, U2AF2, SRSF4, SRSF5, SRSF6, HNRNPA1, RAVER1 - eight spliceosomal or SR/hnRNP proteins, mostly from the Fasci crosslink-MS study, plus this SDC1 row whose entire finding is a splicing-enriched network around a nuclear protein.
That is a pattern and this beat is not treating it as one, deliberately. Rule 84's lesson is that a category with a large base rate produces apparent enrichment for free, and spliceosomal and hnRNP proteins are among the most abundant nuclear proteins and the commonest recoveries in any nuclear AP-MS or crosslink experiment. Eight of 135 is not obviously above what any nuclear bait would give.
The null is computable and is not computed here. What fraction of a comparable nuclear protein's IntAct prey list is spliceosomal? Until that is known, KMT2A associates with the spliceosome is an observation about nuclear abundance. It is named as an unrun null rather than reported as a finding, and it is the single most obvious follow-up this census leaves behind.
| partner a | gene-psip1 |
|---|---|
| partner b | SDC1 |
| partner b uniprot | P18827 |
| partner b length AA | 310 |
| partner a uniprot | O75475 |
| partner a length AA | 530 |
| verdict | DECLINED AS AN INTERACTION, RECORDED FOR THE SPLICING THREAD |
| adjudicated by | q160, cycle 146, 2026-09-24 |
| druggability | Not assessable. |
| n methods | 1 |
| n platforms | 1 |
| n publications | 1 |
| endogenous side | yes - anti-bait antibody against native SDC1, not a tag |
| intact miscore | 0.4 |
| cell state | unstimulated; the study's manipulations are SDC1 loss- and gain-of-function |
| epigenetics | None directly. The study's functional readout is RNA level and splicing, not chromatin. |
| named unrun null | What fraction of a comparable abundant nuclear protein's IntAct prey list is spliceosomal? Without it, the eight spliceosomal partners among 135 are an observation about nuclear abundance. |
| absence class | never-measured |
| absence class note | No LEDGF co-IP in a leukaemia cell has been analysed for spliceosomal content against a matched abundance null. |
interface residues a
[]
structures
[]
ligands
[]
records
[ "PSIP1-SDC1, anti bait coip (ENDOGENOUS bait), physical association, STAV-AB mesothelioma, PSIP1 prey, miscore 0.40, PMID 32664515" ]
cell systems
[ "STAV-AB human mesothelioma" ]
localization
{
"compartment": "nucleus (the study's subject is nuclear-translocated SDC1)",
"subnuclear": [],
"shown_for_fusion": "NO - nobody has looked in any fusion."
}measurements
[
{
"quantity": "count",
"subject": "spliceosomal, SR or hnRNP proteins among the uncovered prey-only partners",
"value": 8,
"units": "partners",
"n": 135,
"universe": "the 135 uncovered human prey-only partners, 2026-09-24; class assigned by hand, not by an ontology join",
"null": null,
"note": "NULL NOT COMPUTED. Most come from one crosslink-MS study (PMID 30021884), so they are not independent. Do not read this as enrichment."
}
]- Recording an eight-member 'splicing thread' without its null is exactly the failure this project has committed four times over the DepMap anti-selective half. The count is written into the entry with the null field explicitly empty so that nobody can quote it as enrichment without noticing.
- What fraction of an abundance-matched nuclear protein's IntAct prey list is spliceosomal? This is the null for the eight-partner splicing observation and it is computable from data already pulled.
- Where did PSIP1 rank among the SDC1 co-IP recoveries? The paper's tables would say whether it was a top hit or near the threshold, and they were not opened.
- Does nuclear translocation of a normally extracellular proteoglycan happen in leukaemia? SDC1 (CD138) is the canonical plasma-cell marker, so the question is live in myeloma and has never been asked in a KMT2A-rearranged blast.
- E1established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.database
IntAct REST returns one PSIP1-SDC1 (O75475-P18827) record: detection method 'anti bait coip', interaction type 'physical association', host recorded as 'human mesothelioma cell line STAV-AB', PSIP1 as PREY, intact-miscore 0.40, from PMID 32664515. UniProtKB P18827 (SDC1_HUMAN) is 310 aa.
IntAct molecular interaction database, REST findInteractions on O75475, retrieved 2026-09-24 by q160 (partner SDC1) · IntAct (EMBL-EBI) 2026What this evidence cannot show
One record. 'anti bait coip' indicates an antibody against the bait protein itself rather than an epitope tag, which means the bait is endogenous - unusual in this census and recorded as a strength. The host-organism string is free text naming a cell line that is not a standard reference line, so its identity and authentication were not checked here.
Evidence that argues against this (1)
- An endogenous antibody pull-down trades one artefact class for another: antibody cross-reactivity replaces tag effects, and a co-IP of a heavily glycosylated proteoglycan is prone to non-specific capture. The paper validated its top splicing candidates by Western blot; PSIP1 was not among the validated ones. Kumar-Singh A et al. 2020
- E2established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper
PMID 32664515 (Kumar-Singh et al., Biomolecules 2020) co-immunoprecipitated SDC1 from a mesothelioma cell line to identify interacting proteins, stating that the precipitates contained a large number of proteins indicating the recovery of protein networks; proteomic analysis focused on nuclear proteins revealed association with pathways related to cell proliferation and RNA synthesis, splicing and transport; the top RNA-splicing candidates were verified by co-IP and Western blot; and loss- and gain-of-function experiments showed SDC1 influences RNA levels. The paper had previously shown SDC1 translocates to the nucleus by a tubulin-dependent mechanism. PSIP1 and LEDGF are not named in the abstract.
Mapping the Interactome of the Nuclear Heparan Sulfate Proteoglycan Syndecan-1 in Mesothelioma Cells. · Kumar-Singh A, Shrinet J, Parniewska MM, Fuxe J, Dobra K, Hjerpe A · Biomolecules 2020What this evidence cannot show
Abstract read in full; the proteomic tables were not opened, so it is not known where PSIP1 ranked among the recovered proteins or whether it was near a threshold. The authors' own phrase 'a large number of proteins' is quoted because it is the decline, and it is their characterisation rather than this beat's.
Evidence that argues against this (1)
- The study is a genuine, focused investigation of a specific biological question - what nuclear syndecan-1 does - rather than an atlas, and its RNA-level functional follow-up is real. Declining a row from it on the grounds that the co-IP was broad is a criterion that would also decline most successful interaction discoveries at the discovery stage. Kumar-Singh A et al. 2020
- E3moderate confidence 3 of 5: One strong study, or several smaller consistent ones.dataset
Among the 135 uncovered human prey-only partners in this census, eight are spliceosomal, SR or hnRNP proteins: EFTUD2, SF3B4, U2AF2, SRSF4, SRSF5, SRSF6, HNRNPA1 and RAVER1. No null has been computed for what fraction of any comparable nuclear protein's IntAct prey list is spliceosomal, so this count is not evidence of enrichment.
analyses/q160-the-prey-only-interactome/census.json and rest-roles.json, generated 2026-09-24 · this project 2026What this evidence cannot show
The eight were identified by hand from the census symbol list, not by an ontology join, so the count is approximate and the class boundary ('spliceosomal, SR or hnRNP') was drawn by this beat. Most come from a single crosslink-MS study (PMID 30021884, Fasci et al.), which this beat's atlas heuristic DID flag at 16 partners, so they are not independent of each other.
Evidence that argues against this (1)
- Spliceosomal and hnRNP proteins are among the most abundant nuclear proteins and the commonest recoveries in nuclear AP-MS and crosslink experiments. Eight of 135 may be at or below the base rate for any nuclear bait, in which case there is no pattern here at all - which is why this is recorded as an unrun null rather than as a finding. this project 2026
endogenous untagged SDC1 co-IP in mesothelioma recovers LEDGF in a splicing-enriched network; DECLINED
one of the 135 uncovered human prey-only partners; eight of them are spliceosomal and the null is unrun