ARGLU1

Existing Annotations Review

GO Term	Evidence	Action	Reason
GO:0000381 regulation of alternative mRNA splicing, via spliceosome	IEA GO_REF:0000107	ACCEPT	Summary: Ensembl Compara transfer from mouse Arglu1 (Q3UL36) of the alternative-splicing regulation term. The mouse annotation is itself experimental and the human protein independently carries the same term by IMP from the same study, so this row is a correct ortholog transfer that happens to be redundant with direct human evidence. Reason: The donor is the genuine one-to-one mouse ortholog, not a paralog, and the transferred term is the one both species' experiments support. Depletion of ARGLU1 in neural cells changes alternative splicing genome-wide, and the same result is reproduced in the mouse conditional knockout cortex. Keeping the Compara row costs nothing and preserves the cross-species support. Supporting Evidence: PMID:30698747 RNA-seq of neural cells depleted of ARGLU1 revealed significant changes in the expression and alternative splicing of distinct genes involved in neurogenesis. PMID:37612280 RNA seq analysis reveals widespread changes in alternative splicing in the mutant mouse embryonic cortex, preferentially affecting genes involved in neuronal functions.
GO:0000381 regulation of alternative mRNA splicing, via spliceosome	IMP PMID:30698747 ARGLU1 is a transcriptional coactivator and splicing regulat...	ACCEPT	Summary: Core biological process. Knockdown of ARGLU1 in neural cells changes alternative splicing genome-wide, and the N-terminal region that binds pre-mRNA is the region required for the splicing outcomes, tying the phenotype to a direct activity rather than to an indirect consequence of transcriptional change. Reason: This is one of the gene's two best-supported functions and the term is at the right level of specificity. The paper separates the transcriptional and splicing activities onto different regions of the protein, and the splicing activity maps to the RNA-binding N-terminus, which argues the effect is direct for at least a subset of targets. The specificity control matters: a target that is ARGLU1-dependent for splicing but lacks G-rich sequence near its splice sites is not enriched in the RNA pulldown, so the authors explicitly allow that some of the splicing changes are indirect. Supporting Evidence: PMID:30698747 Biochemical studies determined that ARGLU1 interacts with GR through its C-terminal domain; whereas, the N-terminal domain mediates interactions with splicing factors and contributes to both basal and GC-dependent splicing outcomes. PMID:30698747 In contrast, Brwd3 pre-mRNA (undergoes ARGLU1-dependent AS but does not contain G-rich sequences near splice sites) is not enriched (Supplementary Figure S9E).
GO:0003713 transcription coactivator activity	IEA GO_REF:0000107	ACCEPT	Summary: Ensembl Compara transfer of coactivator activity from mouse Arglu1. Redundant with the human IMP row from the same study but correct. Reason: Correct ortholog, correct term. The coactivation assays in the source study were performed with human ARGLU1 constructs in human HEK293 cells, so the human annotation does not depend on this transfer; the row is retained as corroborating rather than load-bearing. Supporting Evidence: PMID:30698747 These studies revealed that the construct containing only the C-terminus (ARGLU1C-term) retained the ability to coactivate GR, comparable to that of the full-length protein; whereas, the construct containing only the N-terminus (ARGLU1N-term) did not (Figure 2B).
GO:0003713 transcription coactivator activity	IMP PMID:30698747 ARGLU1 is a transcriptional coactivator and splicing regulat...	ACCEPT	Summary: Core molecular function. ARGLU1 was isolated in a screen for modulators of glucocorticoid signalling and coactivates multiple nuclear receptors in GAL4-LBD reporter assays; domain-swap constructs localise the activity to the glutamate-rich C-terminal region. Reason: GO:0003713 is the correct and maximal term. The substrate-specific alternative a reader would reach for, GO:0030374 nuclear receptor coactivator activity, is obsolete in the current ontology, so no more specific child exists and none should be proposed. The activity is co-operative rather than autonomous, which is consistent with a coregulator term rather than a DNA-binding transcription factor term. Worth recording that the obvious mechanistic explanation was tested and failed: mutating both degenerate LXXLL motifs did not disrupt coactivation, so the motif content of the C-terminus does not account for the activity. Supporting Evidence: PMID:30698747 These studies revealed that the construct containing only the C-terminus (ARGLU1C-term) retained the ability to coactivate GR, comparable to that of the full-length protein; whereas, the construct containing only the N-terminus (ARGLU1N-term) did not (Figure 2B). PMID:30698747 we observed a synergistic increase in ligand-induced luciferase activity (Figure 2A), implying that ARGLU1 forms a complex with other coactivators and GR to regulate gene expression. PMID:30698747 These data suggest that while the C-terminal domain is important for activation, these motifs alone cannot account for the ARGLU1 mediated transactivation of GR, thus, other non-canonical interactions are likely involved.
GO:0005515 protein binding	IPI PMID:22365833 Dynamic protein-protein interaction wiring of the human spli...	MODIFY	Summary: Yeast two-hybrid interaction with SRPK2 from a targeted spliceosome protein interaction map. SRPK2 is a protein kinase, so an informative replacement term is available and the bare protein binding row should not stand. Reason: Per the project's policy, a bare GO:0005515 row is replaced by a more informative term when the evidence supports one. SRPK2 (P78362) is a serine/arginine protein kinase, so GO:0019901 protein kinase binding is directly supported and strictly more informative. The interaction is not a one-screen artefact: the committed IntAct audit finds three rows for this pair from three independent studies using three orthogonal methods (yeast two-hybrid here, an in vitro protein kinase assay in PMID:23602568, and AP-MS in PMID:33961781), and all four of ARGLU1's curated partners resolve to reviewed Swiss-Prot entries at canonical length with nuclear localisation. SRPK2 is itself annotated to nuclear speckles, the compartment ARGLU1 occupies. Worth stating explicitly, because this review admits GO:1990935 splicing factor binding on the U2AF2 and PUF60 rows using a measured criterion: SRPK2 passes that criterion too, carrying two IDA annotations under GO:0008380 RNA splicing, so both terms are defensible for this partner. GO:0019901 is preferred because the relationship that was actually measured here is catalytic rather than merely associative, and the kinase term carries that mechanistic content where the splicing-factor term would discard it. Proposed replacements: protein kinase binding Supporting Evidence: PMID:22365833 screen generated a protein interaction map comprising 632 interactions between 196 proteins.
GO:0005515 protein binding	IPI PMID:23602568 The protein interaction landscape of the human CMGC kinase g...	MODIFY	Summary: SRPK2 interaction from the human CMGC kinase-group interaction proteome. IntAct records this particular entry not as a physical association but as a phosphorylation, detected by protein kinase assay in vitro, so it reports ARGLU1 as an SRPK2 substrate rather than merely a binding partner. Reason: GO:0019901 protein kinase binding is supported and more informative than bare protein binding. This is the strongest of the three SRPK2 rows because the assay is functional rather than a binding screen. It is also structurally coherent rather than incidental: the committed composition analysis shows ARGLU1's N-terminal region is a genuine RS-repeat region (33.8 percent Arg, 27.0 percent Ser, 12 RS plus 12 SR dipeptides with an unbroken 8-dipeptide alternating run), which is the substrate class SRPK-family kinases act on, and three of UniProt's six annotated phosphosites (Ser-58, Ser-60, Thr-61) fall inside it. Note the restraint: nothing here licenses a biological-process term, because no study has tested what SRPK2 phosphorylation does to ARGLU1. That question is recorded under suggested_experiments instead. Proposed replacements: protein kinase binding Supporting Evidence: PMID:23602568 Furthermore, the analysis revealed a kinase-kinase subnetwork and candidate substrates for CMGC kinases.
GO:0005515 protein binding	IPI PMID:30698747 ARGLU1 is a transcriptional coactivator and splicing regulat...	MODIFY	Summary: Reciprocal co-immunoprecipitation with U2AF2 in the ARGLU1 primary paper, with confocal colocalisation in the same study. The interaction is real and mechanistically apt, but GO:0005515 conveys nothing about what ARGLU1 does. Reason: An informative replacement term exists, so the bare protein binding row is modified rather than removed. GO:1990935 splicing factor binding is defined as binding to a protein involved in removing sections of the primary transcript, and U2AF2 qualifies on GOA's own record: it carries an IDA to GO:0000398 mRNA splicing, via spliceosome. It is the U2 snRNP auxiliary factor that recognises polypyrimidine tracts at 3' splice sites, so contact with it is the expected shape for a regulator of cassette-exon choice, and the evidence here is among the best on the gene: reciprocal co-immunoprecipitation with confocal colocalisation in the primary paper. Correction worth recording: an earlier draft of this review removed this row on the ground that no such term existed, having searched for spliceosomal complex binding. GO's search is token-based, so that query could never have returned splicing factor binding. Proposed replacements: splicing factor binding Supporting Evidence: PMID:30698747 The MS results were confirmed with co-IP western blotting experiments in which ARGLU1 was found to interact with PUF60, U2AF2 and JMJD6 (Figure 3B). PMID:30698747 Reverse IPs with the FLAG antibody led to the HA-ARGLU1 being pulled down with FLAG-PUF60 and FLAG-U2AF2.
GO:0005515 protein binding	IPI PMID:30698747 ARGLU1 is a transcriptional coactivator and splicing regulat...	MODIFY	Summary: The same U2AF2 interaction from the same study, curated a second time by UniProt alongside the IntAct row. GOA carries both; this entry restores the row the fetch-gene stub collapsed. Reason: Same judgement as the IntAct-assigned row for this partner: MODIFY to GO:1990935 splicing factor binding, which U2AF2 satisfies by its own IDA to GO:0000398. Recorded separately rather than merged because GOA holds two rows differing only in assigning database, and the row count in this review is reconciled against the GOA tsv rather than against the seeded stub. Proposed replacements: splicing factor binding Supporting Evidence: PMID:30698747 Reverse IPs with the FLAG antibody led to the HA-ARGLU1 being pulled down with FLAG-PUF60 and FLAG-U2AF2.
GO:0005515 protein binding	IPI PMID:30698747 ARGLU1 is a transcriptional coactivator and splicing regulat...	REMOVE	Summary: Co-immunoprecipitation with JMJD6, the oxygenase that hydroxylates U2AF2, in the ARGLU1 primary paper. Real interaction, uninformative term. Reason: The generic term is removed as uninformative, not as incorrect. GO:1990935 splicing factor binding, used on the U2AF2 and PUF60 rows, was considered here and withheld, on a narrower ground than an earlier draft of this review claimed. That draft said JMJD6 carries zero annotations anywhere under GO:0008380 RNA splicing. The query behind it used is_a and part_of descendants only, which structurally cannot reach regulation terms, because GO relates regulation to its target by regulates. Widening it shows JMJD6 does hold GO:0048024 regulation of mRNA splicing, via spliceosome, by IMP. What remains true is the finer distinction: JMJD6 has no annotation to splicing itself, only to its regulation, and it is an oxygenase that hydroxylates a splicing factor rather than a component of the splicing machinery. Whether a regulator of splicing satisfies the term's requirement that the partner be involved in removing sections of the primary transcript is a question the definition does not settle, so the term is withheld here rather than asserted, and the ambiguity is recorded rather than hidden. The JMJD6 contact is arguably the most mechanistically interesting of the four, because a later study proposes that ARGLU1 acts by competing JMJD6 away from BRD4 to enhance promoter-proximal Pol II pausing; but that mechanism is hedged by its own authors and does not yield a molecular function term for ARGLU1. The committed IntAct audit notes that JMJD6 is the weakest of the four partners by method diversity: all five of its rows use anti-tag co-IP, three of them spoke-expanded rows from two releases of the same AP-MS pipeline, so the independent support is the primary paper's directed co-IP plus one pipeline. Supporting Evidence: PMID:30698747 The MS results were confirmed with co-IP western blotting experiments in which ARGLU1 was found to interact with PUF60, U2AF2 and JMJD6 (Figure 3B). PMID:30698747 whereas JMJD6 is an enzyme which has been found to hydroxylate U2AF2 and alter its activity
GO:0005515 protein binding	IPI PMID:30698747 ARGLU1 is a transcriptional coactivator and splicing regulat...	MODIFY	Summary: Reciprocal co-immunoprecipitation with PUF60 in the ARGLU1 primary paper, plus confocal colocalisation. Real interaction, uninformative term. Reason: MODIFY to GO:1990935 splicing factor binding, as for U2AF2. PUF60 qualifies under the term's definition and is annotated under GO:0008380 RNA splicing in GOA, though by IBA rather than experimentally, which is weaker than U2AF2's IDA and is recorded here rather than glossed. PUF60, like U2AF2, is a polypyrimidine-tract binding factor that promotes U2 snRNP recruitment, and the interaction is supported by reciprocal co-immunoprecipitation and confocal colocalisation in the primary paper. Proposed replacements: splicing factor binding Supporting Evidence: PMID:30698747 Reverse IPs with the FLAG antibody led to the HA-ARGLU1 being pulled down with FLAG-PUF60 and FLAG-U2AF2. PMID:30698747 PUF60 and U2AF2 (also known as U2AF65) bind polypyrimidine tract sequences adjacent to 3' splice sites and facilitate the recruitment of U2 snRNP during spliceosome assembly
GO:0005515 protein binding	IPI PMID:33961781 Dual proteome-scale networks reveal cell-specific remodeling...	MODIFY	Summary: SRPK2 interaction recovered independently by BioPlex 3.0 AP-MS, the third of three orthogonal methods supporting this pair. Reason: Replaced with GO:0019901 protein kinase binding for the same reason as the other two SRPK2 rows. BioPlex is a very large screen, and the committed reference-scope audit shows this reference alone produced 9509 GO:0005515 annotations, too many to enumerate, so on its own it would carry little weight. It is retained as supporting because it independently reproduces a pair that a directed yeast two-hybrid map and an in vitro kinase assay had already found, in different laboratories using different methods. The convergence, not this row, is what justifies the replacement term. Proposed replacements: protein kinase binding Supporting Evidence: PMID:23602568 Furthermore, the analysis revealed a kinase-kinase subnetwork and candidate substrates for CMGC kinases.
GO:0005515 protein binding	IPI PMID:33961781 Dual proteome-scale networks reveal cell-specific remodeling...	REMOVE	Summary: JMJD6 interaction recovered by BioPlex 3.0 AP-MS, corroborating the primary paper's directed co-immunoprecipitation. Reason: Uninformative term, removed on the same grounds as the other JMJD6 row: JMJD6's only splicing-related annotation is to the regulation of splicing rather than to splicing itself, so GO:1990935 splicing factor binding is withheld here even though it applies to U2AF2 and PUF60. The interaction itself is corroborated. This row and the BioPlex 2.0 row that IntAct also holds for this pair come from two releases of one AP-MS pipeline, so they are not two independent confirmations. Supporting Evidence: PMID:30698747 The MS results were confirmed with co-IP western blotting experiments in which ARGLU1 was found to interact with PUF60, U2AF2 and JMJD6 (Figure 3B).
GO:0005515 protein binding	IPI PMID:33961781 Dual proteome-scale networks reveal cell-specific remodeling...	MODIFY	Summary: PUF60 interaction recovered by BioPlex 3.0 AP-MS, corroborating the primary paper's reciprocal co-immunoprecipitation. Reason: MODIFY to GO:1990935 splicing factor binding, on the same grounds as the other PUF60 row. This AP-MS row does not establish the interaction on its own, but it independently reproduces the primary paper's reciprocal co-immunoprecipitation. Proposed replacements: splicing factor binding Supporting Evidence: PMID:30698747 The MS results were confirmed with co-IP western blotting experiments in which ARGLU1 was found to interact with PUF60, U2AF2 and JMJD6 (Figure 3B).
GO:0005515 protein binding	IPI PMID:39251607 Systematic identification of post-transcriptional regulatory...	MODIFY	Summary: U2AF2 interaction recovered by proximity-dependent biotin identification in a systematic survey of post-transcriptional regulatory modules. Reason: MODIFY to GO:1990935 splicing factor binding, as for the other U2AF2 rows. The supporting evidence on this particular row is weaker in one respect worth recording: the method is proximity labelling, which reports neighbourhood rather than binding, and the committed reference-scope audit shows this reference produced 201 GO:0005515 annotations over 122 distinct gene products. That does not make it wrong here, because the pair is independently established by directed reciprocal co-immunoprecipitation in the primary paper and both proteins are nuclear, but this row is not independent confirmation of binding. This reference also carries an erratum (PMID:39468017); nothing in this review depends on its scope. Proposed replacements: splicing factor binding Supporting Evidence: PMID:30698747 Reverse IPs with the FLAG antibody led to the HA-ARGLU1 being pulled down with FLAG-PUF60 and FLAG-U2AF2.
GO:0005634 nucleus	EXP PMID:21454576 Arginine and glutamate-rich 1 (ARGLU1) interacts with mediat...	ACCEPT	Summary: Nuclear localisation established alongside colocalisation with MED1 and ligand-dependent recruitment to estrogen receptor target gene promoters. Reason: Uncontroversial and supported by several independent studies. Nuclear localisation is a precondition for both of ARGLU1's characterised activities, and the same paper places the protein on chromatin by ChIP and ChIP-reChIP, which is a stronger claim than nuclear residence alone. Supporting Evidence: PMID:21454576 We found that ARGLU1 (arginine and glutamate rich 1) not only colocalizes with MED1 in the nucleus, but also directly interacts with a far C-terminal region of MED1.
GO:0005634 nucleus	EXP PMID:30698747 ARGLU1 is a transcriptional coactivator and splicing regulat...	ACCEPT	Summary: Nuclear localisation independently confirmed in the ARGLU1 primary paper by confocal microscopy of tagged protein in HEK293 and Neuro-2a cells. Reason: Correct and independently replicated. The same imaging resolves a punctate subnuclear pattern that the study identifies as nuclear speckles, so the more specific GO:0016607 row from this paper is the informative one and this row is its parent; both are retained because the protein is present in the nucleoplasm as well as in speckles. Supporting Evidence: PMID:30698747 Studies performed with GFP-tagged nuclear speckle marker SRSF2 demonstrated very high co-localization with ARGLU1 with a Pearson correlation coefficient (PCC) of 0.83
GO:0005634 nucleus	IEA GO_REF:0000044	ACCEPT	Summary: Automatic transfer of the UniProt subcellular location keyword Nucleus to the corresponding GO term. Reason: The keyword this maps from is itself supported by two experimental references on the human protein, so the electronic row agrees with the direct evidence. No action needed. Supporting Evidence: PMID:30698747 Studies performed with GFP-tagged nuclear speckle marker SRSF2 demonstrated very high co-localization with ARGLU1 with a Pearson correlation coefficient (PCC) of 0.83
GO:0005654 nucleoplasm	IBA GO_REF:0000033	ACCEPT	Summary: Phylogenetic propagation of nucleoplasm from PANTHER node PTN001271790. The WITH/FROM field names Drosophila Arglu1, mouse Arglu1 and human ARGLU1 itself; the self-reference records that the PAINT curator used this gene's own experimental annotation as one of the descendant evidences, which is how IBA is meant to work. Reason: Sound propagation. The committed PANTHER node audit resolved every WITH/FROM token and queried each donor's own evidence: three gene products in the propagated set hold nucleoplasm by their own experimental annotation, and all three are reviewed Swiss-Prot ARGLU1 orthologs (human Q9NWB6 by EXP and IDA, mouse Q3UL36 by IDA, Drosophila Q9VL63 by EXP). The donors agree with each other and with the target, so the node placement is supported by ortholog-strength evidence in three lineages. This row is also the control that makes the mitochondrion row on the same node diagnosable, since the two terms have identical reach and differ only in donor quality. Supporting Evidence: PMID:30698747 Studies performed with GFP-tagged nuclear speckle marker SRSF2 demonstrated very high co-localization with ARGLU1 with a Pearson correlation coefficient (PCC) of 0.83
GO:0005654 nucleoplasm	IDA GO_REF:0000052	ACCEPT	Summary: Human Protein Atlas immunofluorescence localising ARGLU1 to the nucleoplasm. Reason: Consistent with every other localisation result on this protein, and useful because it is antibody-based detection of the endogenous protein rather than imaging of an overexpressed tagged construct, which is how most of the other localisation evidence was obtained. Supporting Evidence: PMID:30698747 Studies performed with GFP-tagged nuclear speckle marker SRSF2 demonstrated very high co-localization with ARGLU1 with a Pearson correlation coefficient (PCC) of 0.83
GO:0005694 chromosome	EXP PMID:21454576 Arginine and glutamate-rich 1 (ARGLU1) interacts with mediat...	ACCEPT	Summary: ARGLU1 is recruited to endogenous estrogen receptor target gene promoters in a ligand-dependent manner, and ChIP-reChIP places it and MED1 on the same promoter. Reason: This is chromatin association demonstrated by ChIP at named endogenous loci, with a ligand dependence that argues it is regulated occupancy rather than background. It is also the physical counterpart of the coactivator molecular function, so it is a core rather than incidental localisation. Supporting Evidence: PMID:21454576 Importantly, ARGLU1 is recruited, in a ligand-dependent manner, to endogenous estrogen receptor target gene promoters and is required for their expression. PMID:21454576 Furthermore, by ChIP-reChIP assay, we confirm that ARGLU1 and MED1 colocalize on the same estrogen receptor target gene promoter upon estrogen induction.
GO:0005694 chromosome	IEA GO_REF:0000044	ACCEPT	Summary: Automatic transfer of the UniProt Chromosome subcellular location keyword. Reason: The keyword rests on the same ChIP evidence as the experimental row above, so the electronic annotation is in agreement with direct evidence on the human protein. Supporting Evidence: PMID:21454576 Furthermore, by ChIP-reChIP assay, we confirm that ARGLU1 and MED1 colocalize on the same estrogen receptor target gene promoter upon estrogen induction.
GO:0005739 mitochondrion	IBA GO_REF:0000033	REMOVE	Summary: Phylogenetic propagation of mitochondrion from PANTHER node PTN001271790. Resolving the single non-tree token in the WITH/FROM field shows the whole propagation rests on one Candida albicans protein, TLO16, which is a telomere-associated Mediator-tail paralog rather than an ARGLU1 ortholog, and whose mitochondrial localisation its own source paper restricts to one clade of a lineage-specific gene-family expansion. Reason: The committed PANTHER node audit makes this measurable rather than argumentative. Node PTN001271790 propagates both nucleoplasm and mitochondrion to exactly the same 111 gene products across 64 NCBI taxa, so the two terms differ only in their donor pools. For nucleoplasm, three gene products in the set hold the term by their own experimental evidence and all three are reviewed ARGLU1 orthologs. For mitochondrion, exactly one does: CGD:CAL0000179812, which resolves to the unreviewed TrEMBL entry A0A1D8PRM3, Tlo16p of Candida albicans, holding GO:0005739 by a single IDA from PMID:22923044. Every one of the 111 mitochondrion rows names that same accession, so the annotation of the entire eukaryotic ARGLU1 family, including plants, green algae, Selaginella, Marchantia, Trichoplax, Daphnia and all vertebrates, traces to that one protein. Two facts finish the argument. First, TLO16 is architecturally not an ARGLU1 ortholog: InterPro assigns it IPR021017 Mediator complex subunit Med2 (Pfam PF11214) and the AH/BAR superfamily in addition to the shared IPR033371, whereas human ARGLU1 carries IPR033371 and nothing else. Second, its own source paper reports the mitochondrial localisation as a property of one clade of a Candida-specific expansion from one or two ancestral copies to fourteen, with the other two clades nuclear. A localisation that is not even uniform across one lineage-specific family within one species cannot support a pan-eukaryotic node. Human ARGLU1 has no mitochondrial evidence of any kind, no targeting sequence, and four experimentally supported cellular component terms that are all nuclear. Noted deliberately: 213 merged reviews in this repository carry a GO:0005739 IBA row and their modal verdict is ACCEPT, so this REMOVE diverges from the corpus; it does so because in this case the entire donor pool is a single architecturally divergent fungal paralog. Propagation Review Root cause: SOURCE BAD Failure modes: WRONG ORTHOLOG OR PARALOG LINEAGE OR TAXON MISMATCH COMPARTMENT OR COMPLEX MISMATCH Sources checked: CGD:CAL0000179812 · TLO16 SOURCE BAD Resolves via UniProt xref to A0A1D8PRM3, an unreviewed TrEMBL entry, Tlo16p of Candida albicans SC5314. It holds GO:0005739 by IDA from PMID:22923044 and is the only gene product in the 111-member propagated set with its own experimental evidence for the term. It is a member of the TLO family, a Candida-specific expansion of Mediator subunit Med2 paralogs, and InterPro gives it IPR021017 and the AH/BAR superfamily in addition to IPR033371, so its domain architecture differs from human ARGLU1's. Its source paper confines the mitochondrial localisation to the gamma clade of that family. CGD itself assigns the protein ND for both molecular function and biological process. PANTHER:PTN001271790 · ARGLU1 family ancestral node SUPPORTS SOURCE BUT NOT TARGET The ancestral node, not a protein. It is a legitimate node that correctly carries GO:0005654 nucleoplasm to the same 111 gene products on the strength of three reviewed ARGLU1 orthologs. The defect is not the node's existence but the attachment of a mitochondrion term to it from a single divergent occupant. Supporting Evidence: PMID:22923044 clade α and clade β have no introns and encode proteins that localize primarily to the nucleus; clade γ sometimes undergoes splicing, and the gene products localize within the mitochondria as well as the nuclei. PMID:22923044 The telomere-associated TLO gene family underwent a recent expansion from one or two copies in other CUG clade members to 14 expressed copies in C. albicans. PMID:22923044 The 14 expressed TLO gene family members have a conserved Med2 domain at the N terminus, suggesting a role in general transcription.
GO:0016607 nuclear speck	EXP PMID:36533631 Distinct biogenesis pathways may have led to functional dive...	ACCEPT	Summary: Nuclear speckle localisation shown in the sisRNA study, where binding of the ARGLU1-derived stable intronic sequence RNA drives the protein into speckles and away from its own gene locus. Reason: Correct, and unusually informative for a localisation annotation because the study identifies a regulated mechanism that controls it rather than reporting a static steady-state distribution. The speckle pool is the functionally sequestered pool in this model, which makes the localisation part of the regulatory circuit rather than an incidental property. Supporting Evidence: PMID:36533631 Mechanistically, Arglu1 sisRNA represses the splicing-inhibitory activity of ARGLU1 protein by binding to ARGLU1 protein and promoting its localization to nuclear speckles, away from the Arglu1 gene locus.
GO:0016607 nuclear speck	IDA PMID:30698747 ARGLU1 is a transcriptional coactivator and splicing regulat...	ACCEPT	Summary: Confocal colocalisation of ARGLU1 with the nuclear speckle marker SRSF2 at a Pearson coefficient of 0.83, independently establishing the speckle localisation. Reason: A quantified colocalisation against a canonical speckle marker, replicated by a second study using a different route. One caveat for the record rather than for the action: the imaging used transfected mCherry-tagged ARGLU1 and GFP-tagged SRSF2 rather than endogenous protein, so the annotation would more conventionally be IMP or at least flagged as ectopic expression. The conclusion is not in doubt, since the Human Protein Atlas detects endogenous ARGLU1 in the nucleoplasm and an independent study reaches the speckle result by a different method. Supporting Evidence: PMID:30698747 Studies performed with GFP-tagged nuclear speckle marker SRSF2 demonstrated very high co-localization with ARGLU1 with a Pearson correlation coefficient (PCC) of 0.83
GO:0016607 nuclear speck	IEA GO_REF:0000120	ACCEPT	Summary: Combined automated annotation reaching nuclear speck from the mouse ortholog and the UniProt subcellular location vocabulary. Reason: The electronic call agrees with two independent experimental results on the human protein, so it adds no risk. The donor is the one-to-one mouse ortholog rather than a paralog. Supporting Evidence: PMID:30698747 Studies performed with GFP-tagged nuclear speckle marker SRSF2 demonstrated very high co-localization with ARGLU1 with a Pearson correlation coefficient (PCC) of 0.83
GO:0036002 pre-mRNA binding	IDA PMID:30698747 ARGLU1 is a transcriptional coactivator and splicing regulat...	ACCEPT	Summary: Core molecular function. RNAcompete with purified GST-ARGLU1 defines a G-rich CGG(A/G)GG preference, and cross-linking RNA immunoprecipitation shows the full-length protein and its N-terminal region, but not its C-terminal region, enrich specific endogenous pre-mRNAs. Reason: Direct binding, with both an in vitro specificity determination on purified protein and a cellular assay on endogenous transcripts, and with a region mapping that ties the activity to the arginine/serine-rich N-terminus. The specificity control is what makes pre-mRNA binding, rather than generic RNA binding, the right call: a transcript that is ARGLU1-dependent for splicing but lacks G-rich sequence near its splice sites is not enriched, so the enrichment reflects sequence-specific recognition and not bulk association with nascent RNA. Retained as core. Supporting Evidence: PMID:30698747 Computational analysis of the microarray data found full length ARGLU1 preferentially bound to CGG(A/G)GG type k-mers (Figure 3C, Supplementary Figure S9). PMID:30698747 Our results showed that full-length ARGLU1 and ARGLU1N-term (but not ARGLU1C-term) pulldowns enrich Lpin1 and Bin1 pre-mRNA PMID:30698747 In contrast, Brwd3 pre-mRNA (undergoes ARGLU1-dependent AS but does not contain G-rich sequences near splice sites) is not enriched (Supplementary Figure S9E).
GO:0036002 pre-mRNA binding	IEA GO_REF:0000107	ACCEPT	Summary: Ensembl Compara transfer of pre-mRNA binding from the mouse ortholog. Reason: Correct ortholog and correct term, redundant with the human IDA from the same study. Note that the cellular RNA immunoprecipitation in that study was performed in mouse Neuro-2a cells on mouse transcripts using tagged human constructs, so the mouse and human annotations in fact rest on overlapping experiments rather than on independent ones. That does not change the action but is worth recording. Supporting Evidence: PMID:30698747 Our results showed that full-length ARGLU1 and ARGLU1N-term (but not ARGLU1C-term) pulldowns enrich Lpin1 and Bin1 pre-mRNA
GO:0045296 cadherin binding	HDA PMID:25468996 E-cadherin interactome complexity and robustness resolved by...	MARK AS OVER ANNOTATED	Summary: Derived from a proximity-biotinylation survey of the E-cadherin cytoplasmic tail. The assay reports neighbourhood within the labelling radius, and the paper itself describes its hits as proteins in the vicinity of the tail, not as cadherin-binding proteins. ARGLU1 is nuclear in every curated location it holds. Reason: Three lines converge. First, the assay-to-term mapping overreaches: the study used proximity biotinylation to identify 561 proteins near the E-cadherin cytoplasmic tail, and proximity is not binding. Second, the scale is that of a screen rather than a finding: the committed reference-scope audit shows this single reference produced 272 GO:0045296 annotations over 272 distinct gene products, 271 of them HDA, all assigned by BHF-UCL, against a control in the same table where the ARGLU1 primary paper annotates four gene products. Third, ARGLU1 is topologically implausible as a junctional cadherin ligand: its UniProt subcellular location is Nucleus, Nucleus speckle and Chromosome, all four of its experimentally supported cellular component terms are nuclear, and no study in its literature places it at a cell junction. MARK_AS_OVER_ANNOTATED rather than REMOVE because the high-throughput observation itself is not being called false, only the inference from proximity to binding, and because this is the modal verdict on this exact row across the corpus: the committed sibling-verdict survey finds 38 merged reviews carrying this same term and reference, resolved as MARK_AS_OVER_ANNOTATED 21 times, KEEP_AS_NON_CORE 11, REMOVE 3, and one each UNDECIDED, ACCEPT and PENDING. Supporting Evidence: PMID:25468996 We used proximity biotinylation and quantitative proteomics to identify 561 proteins in the vicinity of the cytoplasmic tail of E-cadherin.
GO:0045893 positive regulation of DNA-templated transcription	IEA GO_REF:0000108	ACCEPT	Summary: Inferred automatically from the transcription coactivator activity annotation by inter-ontology logical inference. Reason: The inference is valid and the premise annotation is experimentally supported in human cells. One nuance that the term does not capture and that a curator should be aware of: ARGLU1 is not uniformly a positive regulator. A later study reports that it enhances promoter-proximal RNA polymerase II pausing, and a 2025 study describes it as a transcriptional repressor at adenoviral promoters by the same mechanism. Both directions are published, at different loci, and the positive regulation term remains correct for the nuclear receptor target genes on which it was established. Supporting Evidence: PMID:30698747 we observed a synergistic increase in ligand-induced luciferase activity (Figure 2A), implying that ARGLU1 forms a complex with other coactivators and GR to regulate gene expression. PMID:38520408 Transcriptional effects induced by ARGLU1 occur via enhancement of promoter-proximal RNA polymerase II pausing, likely by inhibiting the interaction between JMJD6 and BRD4.
GO:0036033 mediator complex binding	IPI PMID:21454576 Arginine and glutamate-rich 1 (ARGLU1) interacts with mediat...	NEW	Summary: Proposed new annotation. ARGLU1 binds directly to a far C-terminal region of MED1, the Mediator subunit that contacts nuclear receptors, and the two proteins occupy the same estrogen receptor target gene promoter by ChIP-reChIP. This is the title finding of the 2011 paper and it currently produces no molecular function annotation anywhere in GOA. Reason: A real coverage gap. PMID:21454576 appears in ARGLU1's GOA record only as support for two cellular component rows, so the direct protein interaction it reports, which is the paper's central result, is not represented by any molecular function term. GO:0036033 mediator complex binding is the available term and its definition, binding to a mediator complex, is satisfied by direct binding to a Mediator subunit within the complex; the caveat that the demonstrated interaction is with the MED1 subunit rather than with reconstituted holo-Mediator is recorded here rather than hidden. Evidence code IPI with MED1 (UniProtKB:Q15648) as the interactor; the experiments were performed on human proteins in human breast cancer and HEK cells, so an experimental code is appropriate. This annotation is also what makes ARGLU1's coactivator activity mechanistically legible, since Mediator is the bridge from a ligand-bound nuclear receptor to RNA polymerase II. Supporting Evidence: PMID:21454576 We found that ARGLU1 (arginine and glutamate rich 1) not only colocalizes with MED1 in the nucleus, but also directly interacts with a far C-terminal region of MED1. PMID:21454576 Furthermore, by ChIP-reChIP assay, we confirm that ARGLU1 and MED1 colocalize on the same estrogen receptor target gene promoter upon estrogen induction.
GO:0048025 negative regulation of mRNA splicing, via spliceosome	IMP PMID:27899669 An Ultraconserved Element (UCE) controls homeostatic splicin...	NEW	Summary: Proposed new annotation capturing the direction of ARGLU1's autoregulatory activity on its own transcript. Raising ARGLU1 protein shifts its own mRNA towards the retained-intron and nonsense-mediated-decay isoforms and away from the fully spliced isoform; removing the protein shifts splicing the other way. Reason: GOA carries only GO:0000381, which is unsigned and says nothing about direction. The homeostatic circuit that is arguably the best-understood thing about this gene is specifically inhibitory on its own host transcript, and a second study states the point directly by describing the sisRNA as repressing the splicing-inhibitory activity of ARGLU1 protein. GO:0048025 is the signed term for that claim and is complementary to, not a replacement for, GO:0000381, which remains correct for the broader set of cassette exons ARGLU1 regulates in both directions. Evidence code IMP and species human: the experiments were done in HeLa, HEK293 Flp-In T-REx and MCF7 cells, including a CRISPR frameshift line that abolishes ARGLU1 protein while retaining the mRNA. Supporting Evidence: PMID:27899669 Further, overexpression of the ARGLU1 protein shifted the splicing of endogenous ARGLU1 mRNA, resulting in an increase in the retained intron isoform and nonsense mediated decay susceptible isoform and a decrease in the fully spliced isoform. PMID:27899669 This was confirmed in HeLa, HEK293 Flp-In TRex and MCF7 cells by RT-PCR and Northern blot (Figure 2A and B, Supplementary Figure S1B). PMID:36533631 Mechanistically, Arglu1 sisRNA represses the splicing-inhibitory activity of ARGLU1 protein by binding to ARGLU1 protein and promoting its localization to nuclear speckles, away from the Arglu1 gene locus.
GO:0042921 nuclear receptor-mediated glucocorticoid signaling pathway	IMP PMID:30698747 ARGLU1 is a transcriptional coactivator and splicing regulat...	NEW	Summary: Proposed new annotation. ARGLU1 was identified in a screen for modulators of glucocorticoid signalling and is required for a full glucocorticoid receptor transcriptional response; dexamethasone-induced alternative splicing changes are largely lost when ARGLU1 is absent. Reason: GOA represents ARGLU1's coactivator activity and its general positive regulation of transcription, but not the specific signalling pathway in which that activity was characterised, even though the pathway is the subject of the paper's title. The term is well defined for this use, covering a pathway initiated by a glucocorticoid binding an intracellular nuclear receptor and ending in regulation of a downstream cellular process. The evidence is loss and gain of ARGLU1 function in reporter and RNA-seq assays rather than a direct biochemical measurement, hence IMP. The human arm of the study, the expression screen and the GAL4-NR-LBD coactivation assays, was performed in HEK293 cells with human ARGLU1 constructs; the neural depletion and dexamethasone RNA-seq work was in mouse Neuro-2a cells, and the mouse ortholog carries the corresponding annotations separately. Supporting Evidence: PMID:30698747 These studies revealed that the construct containing only the C-terminus (ARGLU1C-term) retained the ability to coactivate GR, comparable to that of the full-length protein; whereas, the construct containing only the N-terminus (ARGLU1N-term) did not (Figure 2B). PMID:30698747 Treatment with dexamethasone, a GR activator, also induces changes in the pattern of alternatively spliced genes, many of which were lost when ARGLU1 was absent.

Gene Description

Existing Annotations Review

Core Functions

References

Suggested Questions for Experts

Suggested Experiments

Knowledge Gaps

Deep Research

Affinage

📚 Additional Documentation

Notes

Bioinformatics Results

📄 View Raw YAML