SEPHS1

Existing Annotations Review

GO Term	Evidence	Action	Reason
GO:0005737 cytoplasm	IBA GO_REF:0000033	KEEP AS NON CORE	Summary: Phylogenetic (IBA) annotation placing SEPHS1 activity in the cytoplasm. This is consistent with experimental data: several splice variants and the Sec-biosynthesis-associated oligomer localize to the cytoplasm, and the protein is also nuclear. Correct but a broad cellular-component term. Reason: Correct compartment but a general location; the nuclear/chromosomal localization tied to the Zincore coregulator role is the more informative site. Supporting Evidence: PMID:28414460 SEPHS1 and SEPHS2 form oligomers in eukaryotic cells
GO:0016260 L-selenocysteine biosynthetic process	IBA GO_REF:0000033	MARK AS OVER ANNOTATED	Summary: Phylogenetic (IBA) annotation to L-selenocysteine biosynthetic process, transferred across the SelD/selenophosphate-synthetase family. For human SEPHS1 this ancestral role is uncertain: in vitro the human protein does not synthesize selenophosphate (SPS2 does), and its knockdown does not affect selenoprotein biosynthesis. Any contribution is at most a selenium salvage/recycling role. Reason: Family-level Sec-biosynthesis role propagated by phylogeny to the human paralog, which has diverged (Thr in place of the catalytic Sec) and lacks de novo selenophosphate synthase function in vivo. Propagation Review Root cause: PROPAGATION BAD Failure modes: FUNCTIONAL DIVERGENCE Supporting Evidence: PMID:20471958 In in vitro experiments, SPS2 synthesized SeP from selenide and ATP, but SPS1 did not have this activity file:human/SEPHS1/SEPHS1-uniprot.txt unclear and several studies suggest that it does not act as a
GO:0004756 selenide, water dikinase activity	IBA GO_REF:0000033	MARK AS OVER ANNOTATED	Summary: Phylogenetic (IBA) assignment of the family catalytic activity (selenide, water dikinase / selenophosphate synthetase). Human SEPHS1 retains the fold and ATP binding but has a threonine where SPS2 has the catalytic selenocysteine, and does not display this activity in vitro. This is an ancestral/family activity that is likely not performed by human SEPHS1. Reason: Family-level catalytic activity propagated by homology; the human paralog has substituted the catalytic nucleophile and does not show the activity in vitro, so this is not its core molecular function. Human SEPHS1 carries THREONINE at the position corresponding to SEPHS2's selenocysteine (SEPHS2 reads KGUGCKVPQ at U60, SEPHS1 reads KGTGCKVPQ at T29). The cited review states the residue is arginine, which is true of Drosophila Sps1 (R49) but not of the human protein; the paper generalises across SPS1 orthologs. Note also that SEPHS1 retains a conserved cysteine two residues downstream (C31) which UniProt annotates as its own active site -- that Cys is present in both paralogs and is not the position that distinguishes them. Propagation Review Root cause: PROPAGATION BAD Failure modes: PSEUDO OR SUBACTIVITY LOSS Supporting Evidence: PMID:20471958 In in vitro experiments, SPS2 synthesized SeP from selenide and ATP, but SPS1 did not have this activity
GO:0004756 selenide, water dikinase activity	IEA GO_REF:0000120	MARK AS OVER ANNOTATED	Summary: Electronic (RHEA:18737 / EC:2.7.9.3) mapping to selenide, water dikinase activity. In UniProt this catalytic activity is asserted only by similarity (ECO:0000250\|UniProtKB:P97364, mouse), not demonstrated for human SEPHS1, which lacks the activity in vitro. The EC/RHEA mapping is therefore an over-annotation of the human protein. Reason: Automated EC-to-GO mapping of a family activity not experimentally supported for human SEPHS1. Supporting Evidence: file:human/SEPHS1/SEPHS1-uniprot.txt selenophosphate synthase in vivo or plays an non-essential role
GO:0005524 ATP binding	IEA GO_REF:0000002	ACCEPT	Summary: InterPro-based electronic annotation to ATP binding. This is strongly supported: crystal structures of human SEPHS1 were solved with ADP, Mg2+ and phosphate, and mutations in the nucleotide-binding region alter ATP binding. A genuine, structurally established molecular function. Supporting Evidence: PMID:7665581 reduced selenium labeling was due to altered ATP binding
GO:0005634 nucleus	IEA GO_REF:0000044	ACCEPT	Summary: Electronic (UniProt subcellular-location keyword) annotation to nucleus. Corroborated by direct experimental evidence: as a component of the Zincore transcription-coregulator complex, SEPHS1 localizes to the nucleus and chromatin. Supporting Evidence: file:human/SEPHS1/SEPHS1-uniprot.txt SUBCELLULAR LOCATION: Nucleus
GO:0005694 chromosome	IEA GO_REF:0000044	ACCEPT	Summary: Electronic (UniProt subcellular-location keyword) annotation to chromosome. Reflects the Zincore complex binding zinc-finger transcription factors on chromatin at a conserved promoter motif; supported by the same experimental study as the EXP chromosome annotation. Supporting Evidence: PMID:40608935 Zincore, a protein complex consisting of QRICH1 and SEPHS1, as a ZNF-specific
GO:0005737 cytoplasm	IEA GO_REF:0000044	KEEP AS NON CORE	Summary: Electronic (UniProt subcellular-location keyword) annotation to cytoplasm, consistent with the experimentally observed cytoplasmic localization of SEPHS1 and its splice variants. Correct but a broad compartment relative to the nuclear/chromatin site of the coregulator function. Reason: Correct but general location; non-core relative to the nuclear Zincore role. Supporting Evidence: PMID:28414460 SEPHS1 and SEPHS2 form oligomers in eukaryotic cells
GO:0005886 plasma membrane	IEA GO_REF:0000044	KEEP AS NON CORE	Summary: Electronic (UniProt subcellular-location keyword) annotation to plasma membrane. This derives from an isoform-1-specific observation (the major type localizes to plasma and nuclear membranes, whereas other variants are cytoplasmic); no membrane-targeting signal is known and the mechanism is unclear. It is an isoform-restricted, peripheral association, not a core site of function. Reason: Isoform-1-specific peripheral membrane localization with no identified targeting signal; not the core functional compartment. Supporting Evidence: PMID:20471958 MT was localized on both plasma and the nuclear membrane
GO:0031965 nuclear membrane	IEA GO_REF:0000044	KEEP AS NON CORE	Summary: Electronic (UniProt subcellular-location keyword) annotation to nuclear membrane, derived from the isoform-1-specific membrane localization reported by Kim et al. Isoform-restricted and mechanistically unexplained; not the core functional compartment. Reason: Isoform-1-specific peripheral nuclear-membrane localization; retained as non-core. Supporting Evidence: PMID:20471958 MT was localized on both plasma and the nuclear membrane
GO:0005515 protein binding	IPI PMID:21516116 Next-generation sequencing to generate interactome datasets.	MARK AS OVER ANNOTATED	Summary: IntAct binary interaction (IPI) mapped to the uninformative term protein binding. The underlying interactions are real, but the bare term conveys no functional information. Some partners (e.g. QRICH1, zinc-finger proteins) are biologically meaningful (Zincore complex), which is better captured by the coregulator core function. Reason: Bare "protein binding" is uninformative per curation guidelines; retained (not removed) as it is an experimental IPI.
GO:0005515 protein binding	IPI PMID:25416956 A proteome-scale map of the human interactome network.	MARK AS OVER ANNOTATED	Summary: IntAct binary interaction (IPI) mapped to the uninformative term protein binding, from a large-scale human interactome map. Real but uninformative at this level of granularity. Reason: Bare "protein binding"; uninformative, kept as an experimental IPI.
GO:0005515 protein binding	IPI PMID:32296183 A reference map of the human binary protein interactome.	MARK AS OVER ANNOTATED	Summary: IntAct binary interactions (IPI) from the human reference (binary) interactome, mapped to protein binding. Partners include zinc-finger proteins and QRICH1, consistent with the Zincore coregulator role, but the bare term itself is uninformative. Reason: Bare "protein binding"; uninformative, kept as an experimental IPI.
GO:0005515 protein binding	IPI PMID:33961781 Dual proteome-scale networks reveal cell-specific remodeling...	MARK AS OVER ANNOTATED	Summary: IntAct binary interactions (IPI) from a proteome-scale interactome study, mapped to protein binding. Includes SEPHS2 and QRICH1 partners, relevant to the Sec-biosynthesis oligomer and Zincore complex respectively, but the generic term is uninformative. Reason: Bare "protein binding"; uninformative, kept as an experimental IPI.
GO:0042802 identical protein binding	IPI PMID:16189514 Towards a proteome-scale map of the human protein-protein in...	ACCEPT	Summary: IPI self-interaction annotation (SEPHS1 with SEPHS1, P49903:P49903), reflecting homodimerization. SEPHS1 is a well-established homodimer (crystal structures; co-IP of splice variants), so this specific term is correct and experimentally supported. Supporting Evidence: file:human/SEPHS1/SEPHS1-uniprot.txt SUBUNIT: Homodimer
GO:0042802 identical protein binding	IPI PMID:21516116 Next-generation sequencing to generate interactome datasets.	ACCEPT	Summary: IPI self-interaction annotation supporting SEPHS1 homodimerization, consistent with the crystallographic homodimer and co-immunoprecipitation of splice variants. Supporting Evidence: file:human/SEPHS1/SEPHS1-uniprot.txt SUBUNIT: Homodimer
GO:0042802 identical protein binding	IPI PMID:25416956 A proteome-scale map of the human interactome network.	ACCEPT	Summary: IPI self-interaction annotation supporting SEPHS1 homodimerization. Consistent with the established homodimeric structure. Supporting Evidence: file:human/SEPHS1/SEPHS1-uniprot.txt SUBUNIT: Homodimer
GO:0042802 identical protein binding	IPI PMID:32296183 A reference map of the human binary protein interactome.	ACCEPT	Summary: IPI self-interaction annotation supporting SEPHS1 homodimerization from the binary interactome map. Consistent with the homodimeric crystal structure. Supporting Evidence: file:human/SEPHS1/SEPHS1-uniprot.txt SUBUNIT: Homodimer
GO:0045454 cell redox homeostasis	IEA GO_REF:0000120	ACCEPT	Summary: Electronic annotation (ARBA / ortholog transfer from mouse Q8BH69) to cell redox homeostasis. This is corroborated by direct experimental IMP evidence in human cells (PMID:31607477) and by SPS1 loss increasing ROS in Drosophila and human cells. A genuine biological role. Supporting Evidence: PMID:31607477 hSEPHS1 is a regulator of selenium-mediated redox-signaling
GO:0004756 selenide, water dikinase activity	ISS GO_REF:0000024	MARK AS OVER ANNOTATED	Summary: Sequence-similarity (ISS) transfer of selenide, water dikinase activity from mouse SEPHS1 (P97364). As with the IBA/IEA versions, this reflects the conserved fold rather than a demonstrated human activity; the human protein lacks this activity in vitro and its selenocysteine-biosynthesis role is considered unclear. Reason: Homology-based catalytic assignment; the human paralog appears catalytically deficient for selenophosphate synthesis. Propagation Review Root cause: PROPAGATION BAD Failure modes: PSEUDO OR SUBACTIVITY LOSS Supporting Evidence: file:human/SEPHS1/SEPHS1-uniprot.txt selenophosphate synthase in vivo or plays an non-essential role
GO:0005634 nucleus	EXP PMID:40608935 Zincore, an atypical coregulator, binds zinc finger transcri...	ACCEPT	Summary: Direct experimental (EXP) annotation to nucleus from the Zincore study. SEPHS1, as part of the Zincore complex, acts on nuclear chromatin to stabilize zinc-finger transcription factors on DNA. This is a core functional localization. Supporting Evidence: PMID:40608935 Zincore, a protein complex consisting of QRICH1 and SEPHS1, as a ZNF-specific
GO:0005694 chromosome	EXP PMID:40608935 Zincore, an atypical coregulator, binds zinc finger transcri...	ACCEPT	Summary: Direct experimental (EXP) annotation to chromosome. The Zincore complex binds a conserved promoter motif (CTTTAAR) and locks DNA-bound zinc-finger transcription factors onto chromatin, placing SEPHS1 on chromosomes. Core functional localization tied to the coregulator role. Supporting Evidence: PMID:40608935 recognition of different ZNFs and stabilizes ZFP91 onto its cognate DNA motif
GO:0005737 cytoplasm	EXP PMID:20471958 Human selenophosphate synthetase 1 has five splice variants ...	KEEP AS NON CORE	Summary: Experimental annotation to cytoplasm. Most SEPHS1 splice variants localize homogeneously to the cytoplasm (only the major type shows membrane association). Correct but a general compartment relative to the nuclear/chromatin site of the coregulator function. Reason: Correct but broad location; retained as non-core relative to the nuclear Zincore role. Supporting Evidence: PMID:20471958 the other three types were localized in the cytoplasm homogeneously
GO:0005886 plasma membrane	EXP PMID:20471958 Human selenophosphate synthetase 1 has five splice variants ...	KEEP AS NON CORE	Summary: Experimental annotation to plasma membrane, based on the isoform-1 (major-type) immunocytochemistry showing localization to plasma and nuclear membranes. Isoform-restricted, with no identified membrane- targeting signal; not a core functional site. Reason: Isoform-1-specific membrane localization of unclear mechanism; non-core. Supporting Evidence: PMID:20471958 MT was localized on both plasma and the nuclear membrane
GO:0031965 nuclear membrane	EXP PMID:20471958 Human selenophosphate synthetase 1 has five splice variants ...	KEEP AS NON CORE	Summary: Experimental annotation to nuclear membrane, from the isoform-1 (major type) localization. Isoform-restricted; distinct from the nucleoplasmic/ chromatin localization of the Zincore complex. Non-core. Reason: Isoform-1-specific nuclear-membrane localization; retained as non-core. Supporting Evidence: PMID:20471958 MT was localized on both plasma and the nuclear membrane
GO:0045454 cell redox homeostasis	IMP PMID:31607477 The role of selenium-mediated redox signaling by selenophosp...	ACCEPT	Summary: Direct experimental (IMP) annotation to cell redox homeostasis. SEPHS1 knockdown in human ESCs altered ROS-pathway and apoptosis gene sets and impaired reprogramming/survival, and selenium treatment improved survival, establishing SEPHS1 as a regulator of selenium-mediated redox signaling. A well-supported biological process. Supporting Evidence: PMID:31607477 hSEPHS1 is a regulator of selenium-mediated redox-signaling PMID:31607477 the ROS pathway and apoptosis in SEPHS1-knockdown cells
GO:0005737 cytoplasm	IDA PMID:28414460 Analysis of Novel Interactions between Components of the Sel...	KEEP AS NON CORE	Summary: Direct (IDA) annotation to cytoplasm from the study of Sec-biosynthesis pathway interactions, where SEPHS1 oligomerizes with SEPHS2, SEPSECS and SECp43 in eukaryotic cells. Correct but a general compartment. Reason: Correct but broad location; non-core relative to the nuclear coregulator role. Supporting Evidence: PMID:28414460 SEPHS1 and SEPHS2 form oligomers in eukaryotic cells
GO:0016260 L-selenocysteine biosynthetic process	IMP NOT PMID:15534230 Selenophosphate synthetase genes from lung adenocarcinoma ce...	ACCEPT	Summary: NOT annotation: SEPHS1 does not function in de novo L-selenocysteine biosynthesis. Human Sps1 only weakly complemented an E. coli selD mutant and only when L-selenocysteine was supplied, indicating dependence on a selenium salvage/recycling system rather than selenite assimilation (which SPS2 supports). This correctly negates the biosynthetic role and is consistent with the in vitro absence of selenophosphate synthase activity. Supporting Evidence: PMID:15534230 recycles l-selenocysteine, whereas the Sps2 enzyme can function with a
GO:0016260 L-selenocysteine biosynthetic process	IDA PMID:7665581 Cloning and functional characterization of human selenophosp...	MARK AS OVER ANNOTATED	Summary: Historical direct (IDA) annotation from the 1995 cloning paper, where human selD transfection increased selenium labeling of type-1 iodothyronine deiodinase and weakly complemented a bacterial selD mutant. This indirect readout established the family assignment, but later work shows the human protein lacks selenophosphate synthase activity in vitro and is non-essential for selenoprotein synthesis; its role, if any, is in selenium/Sec salvage. Retained (experimental IDA) but the biosynthetic role is contradicted by the later NOT annotation and is at most non-core. Reason: Experimental (IDA) evidence retained per policy, but the de novo biosynthetic interpretation is superseded by evidence that human SEPHS1 lacks selenophosphate synthase activity; treat as over-annotation of a salvage-context role. Supporting Evidence: PMID:7665581 increased selenium labeling of a mammalian selenoprotein PMID:20471958 inhibition of SPS1 expression did not affect the biosynthesis of selenoprotein
GO:0005524 ATP binding	TAS PMID:7665581 Cloning and functional characterization of human selenophosp...	ACCEPT	Summary: Traceable-author (TAS) annotation to ATP binding from the original cloning paper, which identified an ATP/GTP-binding domain and showed that point mutations altered ATP-binding properties. Strongly corroborated by later crystal structures (ADP/Mg2+/phosphate). A core molecular function. Supporting Evidence: PMID:7665581 reduced selenium labeling was due to altered ATP binding
GO:0005525 GTP binding	TAS PMID:7665581 Cloning and functional characterization of human selenophosp...	MARK AS OVER ANNOTATED	Summary: TAS annotation to GTP binding derived from the 1995 description of a "consensus ATP/GTP binding domain". GTP binding per se was not demonstrated; the functionally relevant nucleotide is ATP, and all structural and mutagenesis data concern ATP. This appears to be an over-annotation stemming from a generic sequence-motif description. Reason: GTP binding not experimentally shown; inferred from a generic ATP/GTP motif description. ATP binding is the supported nucleotide function. Supporting Evidence: PMID:7665581 reduced selenium labeling was due to altered ATP binding
GO:0036211 protein modification process	TAS PMID:7665581 Cloning and functional characterization of human selenophosp...	MARK AS OVER ANNOTATED	Summary: TAS annotation to the very general term protein modification process. This derives from SEPHS1's presumed role in supplying selenophosphate for selenoprotein (selenocysteine) synthesis, i.e. co-translational incorporation of a modified amino acid. It is both extremely broad and resting on the disputed catalytic role; uninformative and over-annotated. Reason: Overly general BP term, tied to the disputed selenophosphate-synthesis role; not informative of SEPHS1 function. Supporting Evidence: PMID:7665581 essential component of selenoprotein synthesis

Gene Description

Existing Annotations Review

Core Functions

Cellular redox homeostasis / selenium-mediated redox signaling: SEPHS1 regulates reactive-oxygen-species balance and supports cell proliferation and survival; it is highly expressed in undifferentiated stem cells, and its loss increases ROS and impairs survival/reprogramming.

References

Suggested Questions for Experts

Suggested Experiments

📚 Additional Documentation

Notes

📄 View Raw YAML