ufsp2 encodes a UFM1-specific cysteine protease (peptidase C78 family) that functions as the primary deUFMylase in the UFM1 conjugation pathway. The enzyme cleaves UFM1 from conjugated substrates (deconjugation) and processes UFM1 precursors to expose the C-terminal glycine required for conjugation (maturation). Its principal substrate is UFMylated RPL26 on ER-bound 60S ribosomal subunits, where deUFMylation by ufsp2 is required for ribosome recycling during ER-associated ribosome quality control. All annotations for zebrafish ufsp2 are inferred from mammalian orthologs (human UFSP2 and mouse Ufsp2) by sequence similarity.
Definition: The process in which the ubiquitin-fold modifier UFM1 is cleaved from its target protein, removing one or more covalently attached UFM1 groups from a substrate. This is the deconjugation arm of the UFMylation cycle, catalyzed by UFM1-specific proteases (e.g. UFSP2).
Justification: GO provides a molecular function term for deUFMylase activity (GO:0071567) and a biological process term for the forward reaction, protein ufmylation (GO:0071569), but lacks a corresponding biological process term for the reverse (deconjugation) reaction. Analogous deconjugation processes exist for other ubiquitin-like modifiers as children of GO:0070646 protein modification by small protein removal, namely protein deubiquitination (GO:0016579), protein desumoylation (GO:0016926), and protein deneddylation (GO:0000338). A protein deUFMylation term would complete this set and allow UFSP2 deconjugation activity to be annotated to an appropriately specific BP.
Parent term: protein modification by small protein removal
Supporting Evidence:
| GO Term | Evidence | Action | Reason |
|---|---|---|---|
|
GO:0005783
endoplasmic reticulum
|
IBA
GO_REF:0000033 |
ACCEPT |
Summary: Correct. UFSP2 is recruited to the ER via its N-terminal domain interaction with DDRGK1, and acts on UFMylated RPL26 on ER-bound ribosomes. This is a functionally important localization site.
Reason: The ER localization is well supported by structural studies showing the N-terminal domain recruits UFSP2 to the ER via DDRGK1, and functional studies demonstrating its role in deUFMylation of ER-associated ribosomal subunits.
Supporting Evidence:
PMID:21228277
plays a role in the recognition of its cellular substrate C20orf116 and thus in the recruitment of UfSP2 to the endoplasmic reticulum
PMID:30626644
RPL26 UFMylation and de-UFMylation is catalyzed by enzyme complexes tethered to the cytoplasmic surface of the ER
file:DANRE/ufsp2/ufsp2-deep-research-falcon.md
In the current pathway-centric view, this ER tethering spatially matches UFSP2 to major UFMylation events on **ER-associated ribosomes**, and thus to ER proteostasis/ribosome quality control processes.
|
|
GO:0071567
deUFMylase activity
|
IBA
GO_REF:0000033 |
ACCEPT |
Summary: Correct. deUFMylase activity is the core molecular function of UFSP2. The enzyme cleaves UFM1 from conjugated substrates with strict specificity, using a catalytic cysteine-dependent mechanism.
Reason: This is the defining molecular function of UFSP2, supported by direct biochemical evidence from the mammalian orthologs showing specific cleavage of UFM1 conjugates, loss of activity upon Cys-to-Ser mutation, and accumulation of UFMylated RPL26 when UFSP2 is disrupted.
Supporting Evidence:
PMID:17182609
cleaved the C-terminal extension of Ufm1 but not that of ubiquitin
PMID:17182609
replacement of the conserved Cys residue by Ser resulted in a complete loss of the UfSP1 and UfSP2 activities
PMID:30626644
RPL26, is the principal target of UFM1 conjugation...UFMylated RPL26 is highly enriched on ER membrane-bound ribosomes and polysomes
file:DANRE/ufsp2/ufsp2-deep-research-bioreason-sft.md
A cytoplasmic cysteine protease that specializes in processing the ubiquitin-like modifier UFM1
file:DANRE/ufsp2/ufsp2-deep-research-falcon.md
UFSP2 is described as a UFM1-specific **cysteine protease/isopeptidase**, with activity dependent on a catalytic cysteine and a conserved **Cys–Asp–His** catalytic triad (often discussed alongside an additional conserved Tyr typical of cysteine-based Ub/UBL proteases). UFSP activity can be inhibited by thiol-reactive reagents such as **N-ethylmaleimide (NEM)** or by mutation of the catalytic cysteine.
file:DANRE/ufsp2/ufsp2-deep-research-falcon.md
- **UFSP2**: a major **de-UFMylase** for ER-associated substrates, particularly efficiently removing UFM1 from the ribosomal protein **RPL26 (uL24)**; UFSP2 knockout causes a **drastic accumulation of UFMylated proteins** in cell systems summarized in reviews.
|
|
GO:0005634
nucleus
|
IBA
GO_REF:0000033 |
KEEP AS NON CORE |
Summary: Likely correct but non-core. Nuclear localization of UFSP2 has been reported for the mammalian ortholog, and UFSP2 has been observed to be recruited to DNA double-strand breaks. However, the primary functional site is the ER.
Reason: Nuclear localization is reported for the mammalian ortholog and there is evidence for UFSP2 function at DNA damage sites, but the core function of UFSP2 is at the ER where it deUFMylates RPL26 on ribosomal subunits. Falcon deep research notes that reviews list nucleus/cytoplasm/ER localization, but ER tethering is the strongest mechanistic conclusion, consistent with retaining nucleus as non-core.
Supporting Evidence:
file:DANRE/ufsp2/ufsp2-deep-research-falcon.md
Some reviews also list nucleus/cytoplasm/ER localization, but ER tethering is the strongest mechanistic conclusion.
|
|
GO:0006508
proteolysis
|
IBA
GO_REF:0000033 |
MODIFY |
Summary: Correct but too general. UFSP2 performs a highly specific form of proteolysis -- it cleaves the isopeptide bond between UFM1 and its conjugated substrates. The term proteolysis does not capture this specificity.
Reason: While proteolysis is technically accurate, it is overly broad for this specialized isopeptidase. The specific process is removal of the ubiquitin-like modifier UFM1 from substrate proteins (deUFMylation). No dedicated GO term for protein deUFMylation exists, so the closest correct existing BP term is GO:0070646 protein modification by small protein removal, whose definition explicitly covers removal of ubiquitin-like proteins (a sibling of protein deneddylation and protein desumoylation). A dedicated protein deUFMylation term is requested in proposed_new_terms.
Proposed replacements:
protein modification by small protein removal
Supporting Evidence:
file:DANRE/ufsp2/ufsp2-deep-research-falcon.md
Cross-species evidence supports that UFSP2 catalyzes:
- **Proteolytic maturation of pro-UFM1** to mature UFM1 required for UFMylation, and
- **Hydrolysis of isopeptide bonds** linking UFM1 to substrate proteins (de-UFMylation).
|
|
GO:0005634
nucleus
|
IEA
GO_REF:0000044 |
KEEP AS NON CORE |
Summary: Redundant with the IBA annotation for nucleus. Consistent with subcellular localization data from the mammalian ortholog.
Reason: Redundant with the IBA-supported nucleus annotation. Retained as non-core since the primary functional site is the ER.
|
|
GO:0005737
cytoplasm
|
IEA
GO_REF:0000044 |
KEEP AS NON CORE |
Summary: Correct. UFSP2 is a soluble enzyme that resides in the cytoplasm and is recruited to the ER membrane surface where it acts on UFMylated substrates.
Reason: Cytoplasm is listed in UniProt subcellular location data and is consistent with UFSP2 being a soluble protease. However, the mechanistic picture points to constitutive ER tethering via the N-terminal domain interaction with DDRGK1/ODR4 rather than dynamic recruitment, so cytoplasm is retained as a non-core localization.
Supporting Evidence:
file:DANRE/ufsp2/ufsp2-deep-research-falcon.md
Some reviews also list nucleus/cytoplasm/ER localization, but ER tethering is the strongest mechanistic conclusion.
|
|
GO:0005783
endoplasmic reticulum
|
IEA
GO_REF:0000044 |
ACCEPT |
Summary: Redundant with the IBA annotation for ER. Correct and consistent with the role of UFSP2 at the ER membrane surface.
Reason: Redundant with the IBA-supported ER annotation, but independently correct based on UniProt subcellular location data from the mammalian ortholog.
Supporting Evidence:
PMID:21228277
plays a role in the recognition of its cellular substrate C20orf116 and thus in the recruitment of UfSP2 to the endoplasmic reticulum
file:DANRE/ufsp2/ufsp2-deep-research-falcon.md
The UniProt accession **Q7T347** corresponds to **Danio rerio ufsp2**, annotated as **Ufm1-specific protease 2 (UfSP2)**, a cysteine protease in the UFMylation (UFM1 conjugation) system. In the retrieved literature corpus, “UFSP2/Ufsp2” consistently refers to the **UFM1-specific protease 2** (not an unrelated gene), functioning as a UFM1-directed deconjugating enzyme and pro-UFM1 maturase, typically positioned at the **endoplasmic reticulum (ER)** through interaction with **ODR4**.
|
|
GO:0071567
deUFMylase activity
|
IEA
GO_REF:0000117 |
ACCEPT |
Summary: Redundant with the IBA annotation for deUFMylase activity. Correct annotation from ARBA machine learning models.
Reason: Redundant with the IBA-supported annotation but independently correct. deUFMylase activity is the core molecular function.
Supporting Evidence:
PMID:17182609
cleaved the C-terminal extension of Ufm1 but not that of ubiquitin
file:DANRE/ufsp2/ufsp2-deep-research-falcon.md
- **UFSP2**: a major **de-UFMylase** for ER-associated substrates, particularly efficiently removing UFM1 from the ribosomal protein **RPL26 (uL24)**; UFSP2 knockout causes a **drastic accumulation of UFMylated proteins** in cell systems summarized in reviews.
|
|
GO:0032790
ribosome disassembly
|
ISS
GO_REF:0000024 |
ACCEPT |
Summary: Correct. UFSP2-mediated deUFMylation of RPL26 is a required step in the recycling of 60S ribosomal subunits from the ER. UFMylation marks stalled ribosomes for quality control, and UFSP2 removes UFM1 to enable ribosome recycling.
Reason: Supported by studies showing that RPL26 UFMylation at the ER triggers ribosome quality control and that UFSP2-mediated deUFMylation is needed for 60S subunit recycling.
Supporting Evidence:
PMID:30626644
RPL26, is the principal target of UFM1 conjugation...UFMylated RPL26 is highly enriched on ER membrane-bound ribosomes and polysomes
PMID:37036982
UFMylation of translocon-bound 60S subunits modulates the RTJ to promote access of proteasomes and RQC machinery to ER-APs
PMID:38383785
UFM1 conjugation (UFMylation) facilitates the rescue of 60S ribosomal subunits (60S) that are released after ribosome-associated quality-control-mediated splitting of ribosomes that stall during co-translational translocation of secretory proteins into the ER.
file:DANRE/ufsp2/ufsp2-deep-research-falcon.md
A 2024 Nature study (“UFM1 E3 ligase promotes recycling of 60S ribosomal subunits from the ER”) provides primary, high-authority support for a model in which UFMylation machinery promotes **recycling of 60S ribosomal subunits** from ER-associated states, and it explicitly frames UFSP2 as an **ER-tethered UFM1-specific hydrolase** in this functional context.
|
|
GO:0006508
proteolysis
|
ISS
GO_REF:0000024 |
MODIFY |
Summary: Correct but too general. Same issue as the IBA proteolysis annotation. UFSP2 performs specific isopeptide bond cleavage of UFM1 conjugates, not general proteolysis.
Reason: Overly broad for this specialized isopeptidase. Should be replaced with a more specific term capturing UFM1 removal. As for the IBA proteolysis annotation, no dedicated protein deUFMylation BP term exists, so the closest correct existing term is GO:0070646 protein modification by small protein removal (a sibling of protein desumoylation and protein deneddylation). A dedicated protein deUFMylation term is requested in proposed_new_terms.
Proposed replacements:
protein modification by small protein removal
Supporting Evidence:
file:DANRE/ufsp2/ufsp2-deep-research-falcon.md
- **Hydrolysis of isopeptide bonds** linking UFM1 to substrate proteins (de-UFMylation).
|
|
GO:0071567
deUFMylase activity
|
ISS
GO_REF:0000024 |
ACCEPT |
Summary: Redundant with IBA and IEA annotations for deUFMylase activity. Correct and well-supported by direct experimental data from the human ortholog.
Reason: Core molecular function, well supported by ISS from human UFSP2 (Q9NUQ7). Consistent with all other evidence lines.
Supporting Evidence:
PMID:17182609
replacement of the conserved Cys residue by Ser resulted in a complete loss of the UfSP1 and UfSP2 activities
file:DANRE/ufsp2/ufsp2-deep-research-falcon.md
UFSP2 can also reverse UFMylation on model substrates such as **DDRGK1/UFBP1** and **ASC1** in vitro/cell studies.
|
|
GO:0005634
nucleus
|
ISS
GO_REF:0000024 |
KEEP AS NON CORE |
Summary: Redundant with IBA and IEA nucleus annotations. Non-core localization.
Reason: Nucleus is a secondary localization site. The primary functional site is the ER.
|
|
GO:0005737
cytoplasm
|
ISS
GO_REF:0000024 |
KEEP AS NON CORE |
Summary: Redundant with IEA cytoplasm annotation. Correct default localization for this soluble protease.
Reason: Cytoplasm is listed in UniProt subcellular location data for this soluble enzyme, but the mechanistic picture points to constitutive ER tethering via DDRGK1/ODR4 rather than dynamic recruitment. Non-core localization.
Supporting Evidence:
file:DANRE/ufsp2/ufsp2-deep-research-falcon.md
Some reviews also list nucleus/cytoplasm/ER localization, but ER tethering is the strongest mechanistic conclusion.
|
|
GO:0005783
endoplasmic reticulum
|
ISS
GO_REF:0000024 |
ACCEPT |
Summary: Redundant with IBA and IEA ER annotations. Correct and core localization for UFSP2 function.
Reason: ER is the primary functional site where UFSP2 deUFMylates RPL26 on ribosomal subunits. Well supported by structural and functional studies.
Supporting Evidence:
PMID:21228277
plays a role in the recognition of its cellular substrate C20orf116 and thus in the recruitment of UfSP2 to the endoplasmic reticulum
file:DANRE/ufsp2/ufsp2-deep-research-falcon.md
In the current pathway-centric view, this ER tethering spatially matches UFSP2 to major UFMylation events on **ER-associated ribosomes**, and thus to ER proteostasis/ribosome quality control processes.
|
Q: Has ufsp2 function been directly tested in zebrafish using morpholino knockdown or CRISPR knockout? Are there any zebrafish-specific phenotypes?
Q: Does zebrafish ufsp2 play a role in erythropoiesis or hematopoietic development, as suggested by the UFMylation pathway's role in erythroid differentiation in mammals?
Experiment: Generate ufsp2 CRISPR knockout zebrafish and characterize phenotypes including skeletal development, neurological function, and ER stress markers to determine if the mammalian disease phenotypes are conserved.
Hypothesis: Loss of ufsp2 in zebrafish will cause ER stress-related phenotypes and potentially skeletal or neurological defects analogous to human UFSP2 mutations.
Experiment: Perform immunofluorescence microscopy in zebrafish embryos to confirm subcellular localization of ufsp2 at the ER, cytoplasm, and nucleus.
Hypothesis: Zebrafish ufsp2 will show predominantly ER and cytoplasmic localization, consistent with mammalian orthologs.
I begin with the domain architecture. The N-terminus carries IPR049387 (UFSP2, second domain, residues 3–188), followed by a C-terminal catalytic module IPR012462 (UFSP1/2/DUB catalytic domain, residues 210–393). This arrangement places a UFSP-specific noncatalytic region upstream of a papain-like cysteine protease core. The UFSP1/2/DUB catalytic domain dictates a thiol protease chemistry tailored to ubiquitin-like modifiers, while the UFSP2-specific N-terminal domain provides substrate-recognition and positioning surfaces that guide the modifier’s C-terminal tail into the active site. The linear order—recognition domain leading into a cysteine protease core—causally enables selective hydrolysis of isopeptide bonds formed by a ubiquitin-like modifier rather than general proteolysis.
This architecture enforces a molecular function of ubiquitin-like protein peptidase activity (GO:0019783). The catalytic domain supplies the nucleophilic cysteine and general base histidine (with an acidic residue completing the triad), while the UFSP2-specific domain aligns the UFM1 C-terminus for precise scissile bond cleavage. The enzyme therefore processes UFM1 in two ways: it removes UFM1 from conjugated substrates (deconjugation) and trims the UFM1 precursor to its mature form (maturation). The same recognition logic explains specificity for UFM1 and the lack of activity toward SUMO1/2/3 or ISG15—surface features in the UFSP2-specific domain favor the hydrophobic patch and tail geometry unique to UFM1.
From this molecular activity, the biological process follows. By maturing UFM1 and editing UFM1 conjugations, the enzyme tunes the UFM1 conjugation cycle that supports ER-proximal protein quality control and proteostasis. In a developing vertebrate, such control impacts tissues with high secretory and proteostatic demand, notably the inner ear and visual system. Perturbations in UFM1 processing alter the function of UFM1-ubiquitin-like conjugation on ER-associated substrates, which in sensory neurons and hair cells influences protein folding, trafficking, and signaling pathways that underlie orientation in gravity fields and photoreceptor performance. This mechanistic link rationalizes involvement in gravitaxis (GO:0042332) and response to light stimulus (GO:0009416): precise UFM1 processing maintains cellular fitness in sensory circuits that execute these behaviors.
The cellular component is inferred from the soluble, two-domain protease lacking transmembrane segments, pointing to a cytosolic enzyme. Its role in UFM1 maturation and deconjugation requires access to UFM1 precursors and conjugates in the cytoplasm, consistent with cytoplasm (GO:0005737). The enzyme likely shuttles near the ER interface through transient recruitment by UFM1-pathway scaffolds, but its core residence and action remain cytoplasmic.
A coherent mechanistic model emerges: the UFSP2-specific N-terminal domain captures UFM1—either as a precursor or as the C-terminally attached modifier on target proteins—and funnels its tail into the C-terminal catalytic cleft for isopeptidase cleavage. This activity is coordinated with the UFM1 conjugation machinery. The enzyme likely associates with Ubiquitin-like modifier-activating enzyme 5 (UBA5), Ubiquitin-fold modifier-conjugating enzyme 1 (UBC13-like E2), and E3 UFM1-protein ligase 1 (UFL1) to balance conjugation with deconjugation. Scaffolds such as DDRGK domain-containing protein 1 and adaptors like CDK5 regulatory subunit-associated protein 3 and Thyroid hormone receptor interactor 4 can recruit the protease to ER-proximal sites or specific substrates. Direct binding to Ubiquitin-fold modifier 1 itself is expected during both maturation and de-UFMylase reactions. Through this cycle, the enzyme maintains UFM1 homeostasis that supports sensory physiology linked to gravitaxis and light responsiveness.
## Functional Summary
A cytoplasmic cysteine protease that specializes in processing the ubiquitin-like modifier UFM1. It both matures UFM1 precursors to the mature form and removes UFM1 from conjugated substrates, thereby balancing the UFM1 conjugation cycle that supports ER-proximal protein quality control. By editing UFM1 attachments in sensory cells, it helps sustain proteostasis required for proper orientation in gravity fields and photoreceptor responsiveness. Mechanistically, an N-terminal recognition module positions the UFM1 tail for cleavage by a C-terminal papain-like catalytic core, and the enzyme likely operates in concert with the UFM1 E1, E2, and E3 enzymes and ER-associated scaffolds.
## UniProt Summary
DUB-like protease which recognizes and hydrolyzes the peptide bond at the C-terminal Gly of UFM1, a ubiquitin-like modifier protein bound to a number of target proteins. Involved in the maturation of ufmylation by processing the precursor of UFM1 to mature UFM1. Does not cleave SUMO1, SUMO2, SUMO3 or ISG15 (By similarity).
## InterPro Domains
- IPR049387: UFSP2, second domain (domain) [3-188]
- IPR012462: UFSP1/2/DUB, catalytic domain (domain) [210-393]
## GO Term Predictions
### Molecular Function
### Biological Process
### Cellular Component
The research report should be a detailed narrative explaining the function, biological processes, and localization of the gene product. Citations should be given for all claims.
You should prioritize authoritative reviews and primary scientific literature when conducting research. You can supplement
this with annotations you find in gene/protein databases, but these can be outdated or inaccurate.
We are specifically interested in the primary function of the gene - for enzymes, what reaction is catalyzed, and what is the substrate specificity? For transporters, what is the substrate? For structural proteins or adapters, what is the broader structural role? For signaling molecules, what is the role in the pathway.
We are interested in where in or outside the cell the gene product carries out its function.
We are also interested in the signaling or biochemical pathways in which the gene functions. We are less interested in broad pleiotropic effects, except where these elucidate the precise role.
Include evidence where possible. We are interested in both experimental evidence as well as inference from structure, evolution, or bioinformatic analysis. Precise studies should be prioritized over high-throughput, where available.
The UniProt accession Q7T347 corresponds to Danio rerio ufsp2, annotated as Ufm1-specific protease 2 (UfSP2), a cysteine protease in the UFMylation (UFM1 conjugation) system. In the retrieved literature corpus, “UFSP2/Ufsp2” consistently refers to the UFM1-specific protease 2 (not an unrelated gene), functioning as a UFM1-directed deconjugating enzyme and pro-UFM1 maturase, typically positioned at the endoplasmic reticulum (ER) through interaction with ODR4. (millrine2023aguideto pages 2-4, zhou2024ufmylationaubiquitinlike pages 19-22, millrine2023aguideto pages 4-5)
Important evidence limitation: despite targeted searches, direct zebrafish (Danio rerio) primary studies specifically on ufsp2 (Q7T347; zgc:64113) were not recovered in this tool run. Accordingly, zebrafish ufsp2 function below is annotated primarily via conserved orthology and cross-species mechanistic studies (mainly human/mouse cells) and is clearly labeled as such. (millrine2023aguideto pages 2-4, zhou2024ufmylationaubiquitinlike pages 19-22, millrine2023aguideto pages 4-5)
UFMylation is a ubiquitin-like post-translational modification in which the ubiquitin-fold modifier UFM1 is covalently attached to lysine residues of protein substrates through an E1–E2–E3 cascade. This pathway has been strongly linked to ER-associated functions, particularly ribosome quality control and ER proteostasis. (millrine2023aguideto pages 2-4, zhou2024ufmylationaubiquitinlike pages 19-22)
The UFMylation pathway is regulated by UFM1-specific proteases (UFSP1 and UFSP2) that perform two central reactions:
1. pro-UFM1 maturation (processing the UFM1 precursor to expose the C-terminal glycine required for conjugation), and
2. de-UFMylation (isopeptidase activity removing UFM1 from substrates). (millrine2023aguideto pages 2-4, zhou2024ufmylationaubiquitinlike pages 19-22)
A key mechanistic detail is that pro-UFM1 activation requires removal of a Ser84–Cys85 dipeptide to expose the conjugatable UFM1 C-terminus. (millrine2023aguideto pages 2-4)
UFSP2 is described as a UFM1-specific cysteine protease/isopeptidase, with activity dependent on a catalytic cysteine and a conserved Cys–Asp–His catalytic triad (often discussed alongside an additional conserved Tyr typical of cysteine-based Ub/UBL proteases). UFSP activity can be inhibited by thiol-reactive reagents such as N-ethylmaleimide (NEM) or by mutation of the catalytic cysteine. (zhou2024ufmylationaubiquitinlike pages 19-22)
Cross-species evidence supports that UFSP2 catalyzes:
- Proteolytic maturation of pro-UFM1 to mature UFM1 required for UFMylation, and
- Hydrolysis of isopeptide bonds linking UFM1 to substrate proteins (de-UFMylation). (millrine2023aguideto pages 2-4, zhou2024ufmylationaubiquitinlike pages 19-22)
Recent synthesis of mechanistic work supports a functional partitioning:
- UFSP2: a major de-UFMylase for ER-associated substrates, particularly efficiently removing UFM1 from the ribosomal protein RPL26 (uL24); UFSP2 knockout causes a drastic accumulation of UFMylated proteins in cell systems summarized in reviews. (zhou2024ufmylationaubiquitinlike pages 19-22)
- UFSP1: more efficient for pro-UFM1 maturation and for removing UFM1 from the E2 enzyme UFC1 at Lys122 (a site near UFC1 catalytic Cys116), consistent with a role in maintaining UFMylation flux. (zhou2024ufmylationaubiquitinlike pages 19-22, millrine2023aguideto pages 2-4)
This division-of-labor model is central to interpreting zebrafish ufsp2: if conserved, zebrafish Ufsp2 would be expected to primarily tune ER/ribosome-associated de-UFMylation, especially on ribosomal targets like Rpl26/uL24. (millrine2023aguideto pages 2-4, zhou2024ufmylationaubiquitinlike pages 19-22, millrine2023aguideto pages 4-5)
A consistent mechanistic conclusion is that UFSP2 is positioned at the endoplasmic reticulum via an interaction with ODR4, an ER membrane protein. UFSP2 is commonly described as ER-associated/tethered rather than intrinsically membrane-spanning, and loss of one partner can destabilize the other in the ER-tethering model summarized in reviews. (millrine2023aguideto pages 4-5, komatsu2025ufm1atthe pages 1-2)
In the current pathway-centric view, this ER tethering spatially matches UFSP2 to major UFMylation events on ER-associated ribosomes, and thus to ER proteostasis/ribosome quality control processes. (millrine2023aguideto pages 2-4, millrine2023aguideto pages 4-5)
Multiple 2023–2024 sources emphasize that UFMylation is strongly engaged at the ER and that RPL26/uL24 is a principal UFMylation target in cells, especially in contexts of translational stalling and ER translocon stress; UFSP2 is positioned as the counteracting de-UFMylase. (millrine2023aguideto pages 2-4, zhou2024ufmylationaubiquitinlike pages 19-22, millrine2023aguideto pages 4-5)
A 2024 Nature study (“UFM1 E3 ligase promotes recycling of 60S ribosomal subunits from the ER”) provides primary, high-authority support for a model in which UFMylation machinery promotes recycling of 60S ribosomal subunits from ER-associated states, and it explicitly frames UFSP2 as an ER-tethered UFM1-specific hydrolase in this functional context. (darosa2024ufm1e3ligase pages 1-6)
No directly retrieved primary study in this run measured zebrafish ufsp2 (Q7T347) enzymatic activity, localization, developmental expression, or knockout phenotype. Therefore, any zebrafish-specific statements about ufsp2 beyond orthology-based inference cannot be made from the present evidence set. (millrine2023aguideto pages 2-4, zhou2024ufmylationaubiquitinlike pages 19-22)
A human genetics study on UBA5 (the UFM1 E1 enzyme) reported zebrafish experiments in which uba5 silencing decreased motility and induced abnormal movements suggestive of seizures, supporting that perturbing the UFMylation pathway has measurable neurobehavioral consequences in zebrafish. However, this is not ufsp2-specific; it supports pathway importance in zebrafish rather than attributing a phenotype to ufsp2. (colin2016biallelicvariantsin pages 1-2)
Millrine et al. (FEBS Journal, publication date: Feb 2023) emphasized that RPL26/uL24 is one of the main targets of UFMylation in cells and described the UFMylation pathway as operating at the ER in coordination with ribosome quality control. This review also summarized then-recent evidence that human UFSP1 is catalytically active (via non-canonical translation initiation) and that ER-associated UFSP2 (via ODR4) is positioned to regulate ribosome-linked UFMylation. (millrine2023aguideto pages 2-4, millrine2023aguideto pages 4-5)
Zhou et al. (Trends in Biochemical Sciences, publication date: Jan 2024) provided an authoritative synthesis (“Box 1”) that UFSP2 knockout causes drastic accumulation of UFMylated proteins, and that UFSP2 efficiently removes UFMylation from RPL26, whereas UFSP1 is more efficient for pro-UFM1 maturation and UFC1 Lys122 de-UFMylation. This is a current expert consensus-like framing of UFSP2’s core biochemical niche. (zhou2024ufmylationaubiquitinlike pages 19-22)
DaRosa et al. (Nature, publication date: Feb 2024) provides high-impact primary evidence that the UFM1 system is mechanistically involved in 60S ribosome recycling from the ER, and frames UFSP2 as an ER-tethered UFM1 hydrolase in this process context—strengthening the interpretation that UFSP2’s key physiological substrates are ER-proximal ribosomal UFMylation targets. (darosa2024ufm1e3ligase pages 1-6)
Although zebrafish-specific ufsp2 disease models were not identified here, UFSP2’s pathway is medically relevant because defects in UFMylation components (including UFSP2 in broader literature) are associated with human developmental and neurological phenotypes and have become an emerging target space for understanding ER proteostasis defects. The strongest mechanistic-to-application bridge in the retrieved set is that the UFMylation system is now understood as a regulator of ER ribosome homeostasis, a process broadly implicated in proteostasis-associated disorders. (zhou2024ufmylationaubiquitinlike pages 19-22, millrine2023aguideto pages 2-4, darosa2024ufm1e3ligase pages 1-6)
The UFSP enzymes’ UFM1 specificity has enabled use of UFM1 activity-based probes (e.g., UFM1-VME) in biochemical studies to capture UFSP activity and define protease specificity, supporting pathway-mapping and mechanistic dissection that can be adapted to non-mammalian orthologs such as zebrafish ufsp2. (millrine2023aguideto pages 2-4)
The following table consolidates the most supportable functional annotation elements for zebrafish ufsp2 (Q7T347), explicitly separating conserved mechanistic evidence from the limited direct zebrafish evidence.
| Aspect | Key points | Evidence type (review/primary) | Best citation IDs (pqac-...) | Publication (first author year journal) | URL/DOI |
|---|---|---|---|---|---|
| Identity/orthology | Target is zebrafish ufsp2 / Ufm1-specific protease 2 (UniProt Q7T347), a peptidase C78 family member. Direct zebrafish-specific primary literature is sparse, so functional annotation relies substantially on conserved cross-species UFSP2/UFMylation evidence plus UniProt/domain context. | Review + database-informed inference | (millrine2023aguideto pages 2-4, zhou2024ufmylationaubiquitinlike pages 19-22) | Millrine 2023 FEBS J; Zhou 2024 Trends Biochem Sci | https://doi.org/10.1111/febs.16730 ; https://doi.org/10.1016/j.tibs.2023.10.004 |
| Enzymatic activity | UFSP2 is a UFM1-specific cysteine protease/deUFMylase. It catalyzes both pro-UFM1 maturation and removal of UFM1 from modified proteins, although current understanding emphasizes stronger de-UFMylase/substrate-editing roles for UFSP2 compared with UFSP1 in many mammalian systems. | Review | (wang2023theposttranslationalrole pages 2-4, millrine2023aguideto pages 2-4, zhou2024ufmylationaubiquitinlike pages 19-22) | Wang 2023 Cells; Millrine 2023 FEBS J; Zhou 2024 Trends Biochem Sci | https://doi.org/10.3390/cells12212543 ; https://doi.org/10.1111/febs.16730 ; https://doi.org/10.1016/j.tibs.2023.10.004 |
| Reaction catalyzed | Proteolytic cleavage of the C-terminal extension of pro-UFM1 exposes the terminal glycine required for conjugation; UFSP2 also hydrolyzes isopeptide bonds linking UFM1 to substrate lysines (de-UFMylation). UFSP catalytic activity depends on a conserved Cys-Asp-His catalytic core/triad. | Review | (kuang2026theufm1conjugation pages 4-5, zhou2024ufmylationaubiquitinlike pages 19-22, millrine2023aguideto pages 2-4) | Kuang 2026 Biology; Zhou 2024 Trends Biochem Sci; Zhou 2024 accepted manuscript; Millrine 2023 FEBS J | https://doi.org/10.3390/biology15050382 ; https://doi.org/10.1016/j.tibs.2023.10.004 ; https://doi.org/10.1111/febs.16730 |
| Substrate specificity | UFSP2 is reported to efficiently de-UFMylate RPL26/uL24 on ER-associated ribosomes; UFSP1 instead shows stronger roles in pro-UFM1 maturation and removing UFM1 from UFC1 Lys122. UFSP2 can also reverse UFMylation on model substrates such as DDRGK1/UFBP1 and ASC1 in vitro/cell studies. | Review summarizing primary studies | (kuang2026theufm1conjugation pages 5-7, millrine2023aguideto pages 2-4, zhou2024ufmylationaubiquitinlike pages 19-22, kuang2026theufm1conjugation pages 4-5) | Kuang 2026 Biology; Millrine 2023 FEBS J; Zhou 2024 Trends Biochem Sci | https://doi.org/10.3390/biology15050382 ; https://doi.org/10.1111/febs.16730 ; https://doi.org/10.1016/j.tibs.2023.10.004 |
| Localization | UFSP2 is primarily positioned at the endoplasmic reticulum (ER) through association with the tail-anchored factor ODR4; its extended N-terminus contributes to ER localization and substrate selectivity. Some reviews also list nucleus/cytoplasm/ER localization, but ER tethering is the strongest mechanistic conclusion. | Review | (kuang2026theufm1conjugation pages 5-7, kuang2026theufm1conjugation pages 4-5, millrine2023aguideto pages 4-5, komatsu2025ufm1atthe pages 1-2) | Kuang 2026 Biology; Millrine 2023 FEBS J; Komatsu 2025 Essays Biochem | https://doi.org/10.3390/biology15050382 ; https://doi.org/10.1111/febs.16730 ; https://doi.org/10.1042/ebc20253054 |
| Pathway role | UFSP2 functions in the UFMylation/de-UFMylation pathway, especially at the ER where it helps regulate ribosome quality control, ER homeostasis, and likely ER-phagy-associated processes by editing ribosome-linked UFM1 signals. | Review | (kuang2026theufm1conjugation pages 5-7, millrine2023aguideto pages 2-4, komatsu2025ufm1atthe pages 1-2) | Kuang 2026 Biology; Millrine 2023 FEBS J; Komatsu 2025 Essays Biochem | https://doi.org/10.3390/biology15050382 ; https://doi.org/10.1111/febs.16730 ; https://doi.org/10.1042/ebc20253054 |
| Zebrafish evidence / phenotypes | Zebrafish-specific evidence for ufsp2 itself is limited in the retrieved sources. A 2016 human-disease paper used zebrafish to model uba5 deficiency, not ufsp2, showing reduced motility and abnormal seizure-like movements after uba5 silencing; this supports pathway importance in vertebrate neurobiology but is not direct ufsp2 functional evidence. | Primary (pathway-level, not ufsp2-specific) | (colin2016biallelicvariantsin pages 1-2) | Colin 2016 Am J Hum Genet | https://doi.org/10.1016/j.ajhg.2016.06.030 |
| Recent 2023-2024 developments | 2023-2024 literature clarified a division of labor between UFSP1 and UFSP2, resolved the long-standing issue that human UFSP1 can be active, and strengthened the model that ER-tethered UFSP2 edits ribosome UFMylation, especially on RPL26/uL24. A 2024 Nature study further linked UFM1 machinery to recycling of 60S ribosomal subunits from the ER. | Review + primary | (millrine2023aguideto pages 2-4, zhou2024ufmylationaubiquitinlike pages 19-22, darosa2024ufm1e3ligase pages 1-6) | Millrine 2023 FEBS J; Zhou 2024 Trends Biochem Sci; DaRosa 2024 Nature | https://doi.org/10.1111/febs.16730 ; https://doi.org/10.1016/j.tibs.2023.10.004 ; https://doi.org/10.1038/s41586-024-07073-0 |
Table: This table summarizes the best-supported functional annotation for zebrafish ufsp2 (UniProt Q7T347), distinguishing conserved cross-species mechanistic evidence from the limited direct zebrafish literature. It is useful for quickly identifying UFSP2's enzymatic role, substrates, localization, pathway context, and where evidence gaps remain.
Given the evidence base available here, the most defensible functional annotation for Danio rerio ufsp2 (Q7T347) is:
- Molecular function (inferred by orthology): UFM1-specific cysteine protease (peptidase C78 family) acting as a de-UFMylase and pro-UFM1 maturase, with the dominant cellular role likely being de-UFMylation of ER-proximal substrates including UFMylated ribosomal proteins (RPL26/uL24). (millrine2023aguideto pages 4-5, zhou2024ufmylationaubiquitinlike pages 19-22, millrine2023aguideto pages 2-4)
- Cellular component (inferred by orthology): ER-associated via ODR4 tethering, aligning UFSP2 activity to ER ribosome quality control and ER proteostasis. (millrine2023aguideto pages 4-5, komatsu2025ufm1atthe pages 1-2)
- Biological process (inferred by pathway conservation): regulation of ER homeostasis and ribosome-associated quality control through dynamic editing of UFMylation on ribosomes and ER-localized pathway components. (millrine2023aguideto pages 2-4, darosa2024ufm1e3ligase pages 1-6)
References
(millrine2023aguideto pages 2-4): David Millrine, Joshua J. Peter, and Yogesh Kulathu. A guide to ufmylation, an emerging posttranslational modification. The FEBS Journal, 290:5040-5056, Feb 2023. URL: https://doi.org/10.1111/febs.16730, doi:10.1111/febs.16730. This article has 71 citations.
(zhou2024ufmylationaubiquitinlike pages 19-22): Xingchen Zhou, Sayyed J. Mahdizadeh, Matthieu Le Gallo, Leif A. Eriksson, Eric Chevet, and Elodie Lafont. Ufmylation: a ubiquitin-like modification. Trends in Biochemical Sciences, 49:52-67, Jan 2024. URL: https://doi.org/10.1016/j.tibs.2023.10.004, doi:10.1016/j.tibs.2023.10.004. This article has 82 citations and is from a domain leading peer-reviewed journal.
(millrine2023aguideto pages 4-5): David Millrine, Joshua J. Peter, and Yogesh Kulathu. A guide to ufmylation, an emerging posttranslational modification. The FEBS Journal, 290:5040-5056, Feb 2023. URL: https://doi.org/10.1111/febs.16730, doi:10.1111/febs.16730. This article has 71 citations.
(komatsu2025ufm1atthe pages 1-2): Masaaki Komatsu and Gaoxin Mao. Ufm1 at the endoplasmic reticulum: linking er stress, ribosome quality control, and er-phagy. Essays in Biochemistry, 69:281-290, Oct 2025. URL: https://doi.org/10.1042/ebc20253054, doi:10.1042/ebc20253054. This article has 3 citations and is from a peer-reviewed journal.
(darosa2024ufm1e3ligase pages 1-6): Paul A. DaRosa, Ivan Penchev, Samantha C. Gumbin, Francesco Scavone, Magda Wąchalska, Joao A. Paulo, Alban Ordureau, Joshua J. Peter, Yogesh Kulathu, J. Wade Harper, Thomas Becker, Roland Beckmann, and Ron R. Kopito. Ufm1 e3 ligase promotes recycling of 60s ribosomal subunits from the er. Nature, 627:445-452, Feb 2024. URL: https://doi.org/10.1038/s41586-024-07073-0, doi:10.1038/s41586-024-07073-0. This article has 62 citations and is from a highest quality peer-reviewed journal.
(colin2016biallelicvariantsin pages 1-2): Estelle Colin, Jens Daniel, Alban Ziegler, Jamal Wakim, Aurora Scrivo, Tobias B. Haack, Salim Khiati, Anne-Sophie Denommé, Patrizia Amati-Bonneau, Majida Charif, Vincent Procaccio, Pascal Reynier, Kyrieckos A. Aleck, Lorenzo D. Botto, Claudia Lena Herper, Charlotte Sophia Kaiser, Rima Nabbout, Sylvie N’Guyen, José Antonio Mora-Lorca, Birgit Assmann, Stine Christ, Thomas Meitinger, Tim M. Strom, Holger Prokisch, Antonio Miranda-Vizuete, Georg F. Hoffmann, Guy Lenaers, Pascale Bomont, Eva Liebau, Dominique Bonneau, Emmanuelle Génin, Dominique Campion, Jean-François Dartigues, Jean-François Deleuze, Jean-Charles Lambert, Richard Redon, Thomas Ludwig, Benjamin Grenier-Boley, Sébastien Letort, Pierre Lindenbaum, Vincent Meyer, Olivier Quenez, Christian Dina, Céline Bellenguez, Camille Charbonnier -Le Clézio, Joanna Giemza, Stéphanie Chatel, Claude Férec, Hervé Le Marec, Luc Letenneur, Gaël Nicolas, Karen Rouault, Delphine Bacq, Anne Boland, and Doris Lechner. Biallelic variants in uba5 reveal that disruption of the ufm1 cascade can result in early-onset encephalopathy. American journal of human genetics, 99 3:695-703, Sep 2016. URL: https://doi.org/10.1016/j.ajhg.2016.06.030, doi:10.1016/j.ajhg.2016.06.030. This article has 131 citations and is from a highest quality peer-reviewed journal.
(wang2023theposttranslationalrole pages 2-4): Xingde Wang, Xingzhi Xu, and Zhifeng Wang. The post-translational role of ufmylation in physiology and disease. Cells, 12:2543, Oct 2023. URL: https://doi.org/10.3390/cells12212543, doi:10.3390/cells12212543. This article has 27 citations.
(kuang2026theufm1conjugation pages 4-5): Meiqian Kuang, Haigang Xu, Hongjun Huang, Caifang Ren, Pan Huang, and Aihua Gong. The ufm1 conjugation system: a master regulator of cellular stress surveillance in human disease. Biology, 15(5):382, Feb 2026. URL: https://doi.org/10.3390/biology15050382, doi:10.3390/biology15050382. This article has 0 citations.
(kuang2026theufm1conjugation pages 5-7): Meiqian Kuang, Haigang Xu, Hongjun Huang, Caifang Ren, Pan Huang, and Aihua Gong. The ufm1 conjugation system: a master regulator of cellular stress surveillance in human disease. Biology, 15(5):382, Feb 2026. URL: https://doi.org/10.3390/biology15050382, doi:10.3390/biology15050382. This article has 0 citations.
ufsp2 encodes UFM1-specific protease 2, a thiol-dependent cysteine protease belonging to the peptidase C78 family. The enzyme specifically cleaves UFM1 (ubiquitin-fold modifier 1) from conjugated proteins and also processes UFM1 precursors to their mature form by removing C-terminal residues. This gene is the zebrafish ortholog of human UFSP2 (UniProt Q9NUQ7).
The UFM1 conjugation system was first described by Komatsu et al. PMID:15071506. UFM1 is a 9.1 kDa ubiquitin-fold modifier that "lacks obvious sequence identity" to ubiquitin but shares a similar tertiary structure. The pathway is conserved in metazoa and plants but absent from yeast PMID:15071506.
Two UFM1-specific proteases, UfSP1 and UfSP2, were identified by Kang et al. PMID:17182609. These enzymes "cleaved the C-terminal extension of Ufm1 but not that of ubiquitin" or related modifiers like SUMO-1 and ISG15, demonstrating strict substrate specificity PMID:17182609. The catalytic mechanism is cysteine-dependent: "replacement of the conserved Cys residue by Ser resulted in a complete loss of the UfSP1 and UfSP2 activities" PMID:17182609.
The crystal structure of mouse UfSP2 was determined at 2.6 angstrom resolution PMID:21228277. UfSP2 comprises two domains: an N-terminal domain (approximately 240 residues) with a novel fold, and a C-terminal catalytic domain (approximately 200 residues) resembling papain-like proteases. The active site employs an unconventional catalytic triad: "Cys294, Asp418, and His420 form the catalytic machinery" PMID:21228277. The N-terminal domain plays a critical role in substrate recognition: "the N-terminal domain of UfSP2 plays a key role in the recognition of C20orf116 and thus in the recruitment of UfSP2 to the ER" PMID:21228277.
The ribosomal protein RPL26 was identified as the principal substrate of UFMylation PMID:30626644. UFMylation occurs at lysines K132 and K134 specifically on 60S ribosomal subunits docked at the ER membrane. The UFMylation machinery (UFL1/CDK5RAP3/DDRGK1 E3 complex) is localized to the ER, ensuring "RPL26 UFMylation occurs exclusively at the ER surface" PMID:30626644. UFSP2 reverses this modification: "UFMylated RPL26 increased in abundance upon disruption of UFSP2" PMID:30626644.
RPL26 UFMylation is essential for ribosome-associated quality control (RQC) at the ER PMID:37036982. When ribosomes stall during cotranslational translocation, "UFMylation of translocon-bound 60S subunits modulates the RTJ to promote access of proteasomes and RQC machinery to ER-APs" PMID:37036982. UFSP2-mediated deUFMylation is required to recycle the modified ribosomal subunits after quality control is completed.
UFSP2 participates in ribosome disassembly at the ER. The UFM1 E3 ligase complex promotes recycling of 60S ribosomal subunits from the ER by UFMylating RPL26, and UFSP2-mediated removal of UFM1 from RPL26 is a required step in the recycling cycle. This connects UFSP2 directly to the GO:0032790 (ribosome disassembly) annotation.
Heterozygous pathogenic variants in the C-terminal catalytic domain of human UFSP2 cause skeletal dysplasias. A Y290H (equivalent to mouse Y282H) variant was identified in Beukes hip dysplasia (BHD), an autosomal dominant disorder causing "bilateral dysmorphism of the proximal femur, which results in severe degenerative osteoarthropathy" PMID:26428751. This mutation "abolishes UFSP2-mediated C-terminal cleavage of its substrate, Ufm1" PMID:26428751.
A separate autosomal recessive variant (V115E, in the N-terminal domain) causes severe neurodevelopmental disease with epilepsy PMID:33473208. Patient fibroblasts showed that "the variant more prominently affects de-UFMylation rather than UFMylation," with reduced UFSP2 protein stability PMID:33473208.
Based on studies of mammalian orthologs, UFSP2 localizes to the ER, cytoplasm, and nucleus. The ER localization is functionally significant given the role in ER-associated ribosome quality control and the recruitment via the N-terminal domain interacting with DDRGK1 (C20orf116) at the ER membrane PMID:21228277. Cytoplasmic and nuclear localizations are also reported, consistent with additional roles including at DNA double-strand breaks.
All annotations for zebrafish ufsp2 are transferred from mammalian orthologs (human UFSP2 Q9NUQ7 and mouse Ufsp2 Q99K23) by sequence similarity (ISS, IBA, IEA). No direct experimental evidence exists for zebrafish ufsp2 itself. The protein is highly conserved across vertebrates, so the transferred annotations are expected to be reliable.
Source: ufsp2-deep-research-bioreason-sft.md
The BioReason SFT functional summary states:
A cytoplasmic cysteine protease that specializes in processing the ubiquitin-like modifier UFM1. It both matures UFM1 precursors to the mature form and removes UFM1 from conjugated substrates, thereby balancing the UFM1 conjugation cycle that supports ER-proximal protein quality control. By editing UFM1 attachments in sensory cells, it helps sustain proteostasis required for proper orientation in gravity fields and photoreceptor responsiveness. Mechanistically, an N-terminal recognition module positions the UFM1 tail for cleavage by a C-terminal papain-like catalytic core, and the enzyme likely operates in concert with the UFM1 E1, E2, and E3 enzymes and ER-associated scaffolds.
Strengths:
The core molecular function is correctly described. The summary accurately identifies UFSP2 as a cysteine protease with dual activity: maturation of UFM1 precursors and deconjugation of UFM1 from substrates. This matches the biochemical evidence that UFSP2 "cleaved the C-terminal extension of Ufm1" PMID:17182609 and functions as a deUFMylase.
The two-domain architecture is correctly described: "an N-terminal recognition module positions the UFM1 tail for cleavage by a C-terminal papain-like catalytic core." This aligns with the crystal structure showing the N-terminal domain mediates substrate recognition and the C-terminal domain contains the catalytic triad PMID:21228277.
The connection to ER-proximal protein quality control is correct. UFSP2 balances the UFM1 conjugation cycle at the ER, where UFMylation of RPL26 on 60S ribosomal subunits supports ribosome-associated quality control [PMID:30626644, PMID:37036982].
The mention of partnership with "UFM1 E1, E2, and E3 enzymes and ER-associated scaffolds" is accurate, though generic.
Weaknesses:
The claim about "gravitaxis (GO:0042332) and response to light stimulus (GO:0009416)" is unsupported. The thinking trace speculates that "perturbations in UFM1 processing alter the function of UFM1-ubiquitin-like conjugation on ER-associated substrates, which in sensory neurons and hair cells influences...orientation in gravity fields and photoreceptor performance." No published evidence links UFSP2 to gravitaxis or light response. This appears to be an over-interpretation, possibly influenced by zebrafish-specific GO terms that are enriched in the training data but not specific to UFSP2.
The summary describes the localization as "cytoplasmic" and the thinking trace explicitly states "cytoplasm (GO:0005737)" as the core residence, downplaying the ER. While UFSP2 is a soluble cytoplasmic protein, its primary functional site is the ER, where it is recruited via DDRGK1 to deUFMylate RPL26 on ribosomal subunits. The summary mentions "ER-proximal" quality control but fails to list the ER as a primary localization.
No mention of RPL26 as the principal UFMylation target. The curated review identifies RPL26 UFMylation as the central biological context for UFSP2 function PMID:30626644, including the role in ribosome disassembly (GO:0032790). The BioReason summary only vaguely refers to "ER-proximal protein quality control" without identifying the specific substrate or process.
No mention of disease relevance. While not strictly a GO annotation concern, the role of UFSP2 mutations in skeletal dysplasia PMID:26428751 and neurodevelopmental disease PMID:33473208 provides important functional validation.
Scoring rationale:
Correctness (4/5): The core biochemistry (cysteine protease, UFM1 specificity, dual maturation/deconjugation activity, two-domain architecture, ER quality control link) is accurate. The gravitaxis/light stimulus claims are unsupported but do not dominate the summary. No fundamentally wrong claims about the enzymatic mechanism.
Completeness (3/5): The summary captures the basic enzymatic function and domain architecture but misses the specific biological context (RPL26 UFMylation, ribosome disassembly, specific ER localization as a functional site). The gravitaxis/light stimulus speculation takes up space that could have been used for the actual characterized biology.
The ai-review.yaml does not contain GO_REF:0000002 (InterPro2GO) annotations for ufsp2. The IEA annotations present are from GO_REF:0000044 (UniProt subcellular location mapping) and GO_REF:0000117 (ARBA). The IBA annotations are from phylogenetic inference (GO_REF:0000033).
The InterPro domains for ufsp2 are IPR049387 (UFSP2, second domain) and IPR012462 (UFSP1/2/DUB catalytic domain). These domains would map via InterPro2GO to deUFMylase activity, which is correct and matches the curated annotation.
BioReason's SFT output goes beyond what InterPro2GO alone would predict in several ways:
- It correctly infers the two-domain architecture and the division of labor (recognition vs. catalysis), which is biological insight beyond simple domain-to-function mapping.
- It correctly connects the enzyme to the broader UFM1 conjugation cycle and ER quality control, which InterPro2GO would not capture.
- However, it also adds unsupported claims (gravitaxis, light response) that InterPro2GO would not make, demonstrating a tendency toward over-interpretation when extending beyond domain-based prediction.
Overall, the BioReason SFT output provides genuine biological insight beyond InterPro2GO for the core function, but introduces confabulated phenotype-level claims that a conservative domain-based approach would avoid.
The thinking trace shows a methodical domain-by-domain reasoning approach. It correctly identifies the UFSP2-specific N-terminal domain and the papain-like catalytic core, and derives the molecular function from domain architecture. The mechanistic chain from domain architecture to molecular function to biological process is logical.
The trace goes off track when it extrapolates from "ER-proximal protein quality control" to "tissues with high secretory and proteostatic demand, notably the inner ear and visual system," and then to gravitaxis and light response. This is a multi-step inference chain where each step introduces uncertainty, and the final claims are not supported by experimental evidence for UFSP2. This pattern of plausible but unsupported phenotypic extrapolation is a characteristic failure mode of language model-based functional prediction.
The trace also correctly notes the absence of transmembrane segments and predicts a soluble, cytoplasmic enzyme, though it could have given more weight to the documented ER recruitment mechanism.
id: Q7T347
gene_symbol: ufsp2
product_type: PROTEIN
status: DRAFT
taxon:
id: NCBITaxon:7955
label: Danio rerio
description: ufsp2 encodes a UFM1-specific cysteine protease (peptidase C78 family)
that functions as the primary deUFMylase in the UFM1 conjugation pathway. The enzyme
cleaves UFM1 from conjugated substrates (deconjugation) and processes UFM1 precursors
to expose the C-terminal glycine required for conjugation (maturation). Its principal
substrate is UFMylated RPL26 on ER-bound 60S ribosomal subunits, where deUFMylation
by ufsp2 is required for ribosome recycling during ER-associated ribosome quality
control. All annotations for zebrafish ufsp2 are inferred from mammalian orthologs
(human UFSP2 and mouse Ufsp2) by sequence similarity.
references:
- id: GO_REF:0000024
title: Manual transfer of experimentally-verified manual GO annotation data to orthologs
by curator judgment of sequence similarity
findings: []
- id: GO_REF:0000033
title: Annotation inferences using phylogenetic trees
findings: []
- id: GO_REF:0000044
title: Gene Ontology annotation based on UniProtKB/Swiss-Prot Subcellular Location
vocabulary mapping, accompanied by conservative changes to GO terms applied by
UniProt
findings: []
- id: GO_REF:0000117
title: Electronic Gene Ontology annotations created by ARBA machine learning models
findings: []
- id: PMID:15071506
title: A novel protein-conjugating system for Ufm1, a ubiquitin-fold modifier
findings:
- statement: UFM1 was identified as a novel ubiquitin-fold modifier with a dedicated
conjugation system involving E1 (UBA5) and E2 (UFC1) enzymes, conserved in metazoa
and plants but not yeast.
supporting_text: Ufm1, Uba5, and Ufc1 are all conserved in metazoa and plants
but not in yeast
- id: PMID:17182609
title: Two novel ubiquitin-fold modifier 1 (Ufm1)-specific proteases, UfSP1 and
UfSP2
findings:
- statement: UfSP1 and UfSP2 were identified as cysteine proteases specific for
UFM1, with UfSP2 conserved across multicellular organisms.
supporting_text: UfSP2 is present in most, if not all, of multicellular organisms
- statement: UfSP1/2 are strictly specific for UFM1 and do not cleave ubiquitin,
SUMO-1, or ISG15.
supporting_text: cleaved the C-terminal extension of Ufm1 but not that of ubiquitin
- statement: The catalytic cysteine is essential for protease activity.
supporting_text: replacement of the conserved Cys residue by Ser resulted in a
complete loss of the UfSP1 and UfSP2 activities
- id: PMID:21228277
title: Structure of ubiquitin-fold modifier 1-specific protease UfSP2
findings:
- statement: Crystal structure of mouse UfSP2 reveals a two-domain architecture
with an unconventional catalytic triad (Cys294, Asp418, His420).
supporting_text: The C-terminal catalytic domain is similar to UfSP1 with Cys(294),
Asp(418), His(420), Tyr(282), and a regulatory loop participating in catalysis
- statement: The N-terminal domain of UfSP2 mediates substrate recognition and ER
recruitment via interaction with DDRGK1 (C20orf116).
supporting_text: The novel N-terminal domain shows a unique structure and plays a
role in the recognition of its cellular substrate C20orf116 and thus in the
recruitment of UfSP2 to the endoplasmic reticulum
- id: PMID:30626644
title: Ribosomal protein RPL26 is the principal target of UFMylation
findings:
- statement: RPL26 is the primary cellular UFMylation target, modified at K132/K134
on ER-bound 60S subunits. UFSP2 reverses this modification.
supporting_text: a largely uncharacterized ribosomal protein, RPL26, is the principal
target of UFM1 conjugation
- statement: RPL26 UFMylation and deUFMylation occur at the ER surface via
tethered enzyme complexes.
supporting_text: RPL26 UFMylation and de-UFMylation is catalyzed by enzyme complexes
tethered to the cytoplasmic surface of the ER
- statement: UFMylated RPL26 is enriched on ER-associated ribosomes.
supporting_text: UFMylated RPL26 is highly enriched on ER membrane-bound ribosomes
and polysomes
- id: PMID:37036982
title: RPL26/uL24 UFMylation is essential for ribosome-associated quality control
at the endoplasmic reticulum
findings:
- statement: RPL26 UFMylation is required for degradation of stalled nascent chains
at the ER via RQC and the ubiquitin-proteasome system.
supporting_text: UFMylation of translocon-bound 60S subunits modulates the RTJ
to promote access of proteasomes and RQC machinery to ER-APs
- id: PMID:26428751
title: Identification of a mutation in the ubiquitin-fold modifier 1-specific peptidase
2 gene, UFSP2, in an extended South African family with Beukes hip dysplasia
findings:
- statement: A Y290H mutation in UFSP2 abolishes deUFMylase activity and causes
autosomal dominant Beukes hip dysplasia.
supporting_text: the BHD mutation abolishes UFSP2-mediated C-terminal cleavage
of its substrate, Ufm1
- id: PMID:33473208
title: A pathogenic UFSP2 variant in an autosomal recessive form of pediatric neurodevelopmental
anomalies and epilepsy
findings:
- statement: A homozygous V115E variant in the N-terminal domain of UFSP2 causes
neurodevelopmental delay and epilepsy by impairing deUFMylation.
supporting_text: the variant more prominently affects de-UFMylation rather than
UFMylation
- id: PMID:38383785
title: UFM1 E3 ligase promotes recycling of 60S ribosomal subunits from the ER
findings:
- statement: UFMylation of uL24/RPL26 on ER-associated 60S ribosomal subunits
facilitates rescue and recycling of the large subunit after ribosome-associated
quality control splits stalled ribosomes during co-translational translocation
at the ER.
supporting_text: UFM1 conjugation (UFMylation) facilitates the rescue of 60S
ribosomal subunits (60S) that are released after ribosome-associated
quality-control-mediated splitting of ribosomes that stall during
co-translational translocation of secretory proteins into the ER.
- id: file:DANRE/ufsp2/ufsp2-deep-research-bioreason-sft.md
title: BioReason SFT functional summary for ufsp2 (Danio rerio)
findings:
- statement: BioReason SFT correctly identifies ufsp2 as a cytoplasmic cysteine
protease specializing in UFM1 processing and deconjugation, with a two-domain
architecture supporting ER-proximal protein quality control.
- id: file:DANRE/ufsp2/ufsp2-deep-research-falcon.md
title: Falcon deep research report on ufsp2 (Danio rerio)
findings:
- statement: |
Zebrafish ufsp2 (UniProt Q7T347) is the Ufm1-specific protease 2, a peptidase
C78 family cysteine protease in the UFMylation system that acts as a UFM1-directed
deconjugating enzyme and pro-UFM1 maturase, typically positioned at the
endoplasmic reticulum through interaction with ODR4. No direct zebrafish primary
studies were recovered, so function is inferred from conserved cross-species
(human/mouse) mechanistic studies.
reference_section_type: OTHER
supporting_text: |-
The UniProt accession **Q7T347** corresponds to **Danio rerio ufsp2**, annotated as **Ufm1-specific protease 2 (UfSP2)**, a cysteine protease in the UFMylation (UFM1 conjugation) system. In the retrieved literature corpus, “UFSP2/Ufsp2” consistently refers to the **UFM1-specific protease 2** (not an unrelated gene), functioning as a UFM1-directed deconjugating enzyme and pro-UFM1 maturase, typically positioned at the **endoplasmic reticulum (ER)** through interaction with **ODR4**.
- statement: |
UFSP2 is a UFM1-specific cysteine protease/isopeptidase whose activity depends
on a catalytic cysteine and a conserved Cys-Asp-His catalytic triad; activity is
abolished by thiol-reactive reagents such as N-ethylmaleimide or by mutation of
the catalytic cysteine.
reference_section_type: OTHER
supporting_text: |-
UFSP2 is described as a UFM1-specific **cysteine protease/isopeptidase**, with activity dependent on a catalytic cysteine and a conserved **Cys–Asp–His** catalytic triad (often discussed alongside an additional conserved Tyr typical of cysteine-based Ub/UBL proteases). UFSP activity can be inhibited by thiol-reactive reagents such as **N-ethylmaleimide (NEM)** or by mutation of the catalytic cysteine.
- statement: |
UFSP2 catalyzes two reactions: proteolytic maturation of pro-UFM1 to mature UFM1
required for UFMylation, and hydrolysis of isopeptide bonds linking UFM1 to
substrate proteins (de-UFMylation). pro-UFM1 activation requires removal of a
Ser84-Cys85 dipeptide to expose the conjugatable UFM1 C-terminus.
reference_section_type: OTHER
supporting_text: |-
Cross-species evidence supports that UFSP2 catalyzes:
- **Proteolytic maturation of pro-UFM1** to mature UFM1 required for UFMylation, and
- **Hydrolysis of isopeptide bonds** linking UFM1 to substrate proteins (de-UFMylation).
- statement: |
A functional division of labor is described: UFSP2 is the major de-UFMylase for
ER-associated substrates, efficiently removing UFM1 from ribosomal protein RPL26
(uL24); UFSP2 knockout causes drastic accumulation of UFMylated proteins. UFSP1
is more efficient for pro-UFM1 maturation and de-UFMylation of UFC1 at Lys122.
reference_section_type: OTHER
supporting_text: |-
- **UFSP2**: a major **de-UFMylase** for ER-associated substrates, particularly efficiently removing UFM1 from the ribosomal protein **RPL26 (uL24)**; UFSP2 knockout causes a **drastic accumulation of UFMylated proteins** in cell systems summarized in reviews.
- statement: |
UFSP2 can also reverse UFMylation on model substrates such as DDRGK1/UFBP1 and
ASC1 in vitro and in cell studies, in addition to RPL26/uL24.
reference_section_type: OTHER
supporting_text: |-
UFSP2 can also reverse UFMylation on model substrates such as **DDRGK1/UFBP1** and **ASC1** in vitro/cell studies.
- statement: |
UFSP2 is positioned at the endoplasmic reticulum via interaction with the
ER membrane protein ODR4, tethering it spatially to major UFMylation events on
ER-associated ribosomes and thus to ER proteostasis and ribosome quality control.
Reviews also list nucleus/cytoplasm/ER localization, but ER tethering is the
strongest mechanistic conclusion.
reference_section_type: OTHER
supporting_text: |-
In the current pathway-centric view, this ER tethering spatially matches UFSP2 to major UFMylation events on **ER-associated ribosomes**, and thus to ER proteostasis/ribosome quality control processes.
- statement: |
A 2024 Nature study frames UFSP2 as an ER-tethered UFM1-specific hydrolase in a
model where the UFMylation machinery promotes recycling of 60S ribosomal subunits
from the ER, consistent with UFSP2's role in ribosome disassembly/recycling
during ER-associated quality control.
reference_section_type: OTHER
supporting_text: |-
A 2024 Nature study (“UFM1 E3 ligase promotes recycling of 60S ribosomal subunits from the ER”) provides primary, high-authority support for a model in which UFMylation machinery promotes **recycling of 60S ribosomal subunits** from ER-associated states, and it explicitly frames UFSP2 as an **ER-tethered UFM1-specific hydrolase** in this functional context.
existing_annotations:
- term:
id: GO:0005783
label: endoplasmic reticulum
evidence_type: IBA
original_reference_id: GO_REF:0000033
review:
summary: Correct. UFSP2 is recruited to the ER via its N-terminal domain interaction
with DDRGK1, and acts on UFMylated RPL26 on ER-bound ribosomes. This is a functionally
important localization site.
action: ACCEPT
reason: The ER localization is well supported by structural studies showing the
N-terminal domain recruits UFSP2 to the ER via DDRGK1, and functional studies
demonstrating its role in deUFMylation of ER-associated ribosomal subunits.
supported_by:
- reference_id: PMID:21228277
supporting_text: plays a role in the recognition of its cellular substrate C20orf116
and thus in the recruitment of UfSP2 to the endoplasmic reticulum
- reference_id: PMID:30626644
supporting_text: RPL26 UFMylation and de-UFMylation is catalyzed by enzyme complexes
tethered to the cytoplasmic surface of the ER
- reference_id: file:DANRE/ufsp2/ufsp2-deep-research-falcon.md
supporting_text: |-
In the current pathway-centric view, this ER tethering spatially matches UFSP2 to major UFMylation events on **ER-associated ribosomes**, and thus to ER proteostasis/ribosome quality control processes.
- term:
id: GO:0071567
label: deUFMylase activity
evidence_type: IBA
original_reference_id: GO_REF:0000033
review:
summary: Correct. deUFMylase activity is the core molecular function of UFSP2.
The enzyme cleaves UFM1 from conjugated substrates with strict specificity,
using a catalytic cysteine-dependent mechanism.
action: ACCEPT
reason: This is the defining molecular function of UFSP2, supported by direct
biochemical evidence from the mammalian orthologs showing specific cleavage
of UFM1 conjugates, loss of activity upon Cys-to-Ser mutation, and accumulation
of UFMylated RPL26 when UFSP2 is disrupted.
supported_by:
- reference_id: PMID:17182609
supporting_text: cleaved the C-terminal extension of Ufm1 but not that of ubiquitin
- reference_id: PMID:17182609
supporting_text: replacement of the conserved Cys residue by Ser resulted in
a complete loss of the UfSP1 and UfSP2 activities
- reference_id: PMID:30626644
supporting_text: RPL26, is the principal target of UFM1 conjugation...UFMylated
RPL26 is highly enriched on ER membrane-bound ribosomes and polysomes
- reference_id: file:DANRE/ufsp2/ufsp2-deep-research-bioreason-sft.md
supporting_text: A cytoplasmic cysteine protease that specializes in processing
the ubiquitin-like modifier UFM1
- reference_id: file:DANRE/ufsp2/ufsp2-deep-research-falcon.md
supporting_text: |-
UFSP2 is described as a UFM1-specific **cysteine protease/isopeptidase**, with activity dependent on a catalytic cysteine and a conserved **Cys–Asp–His** catalytic triad (often discussed alongside an additional conserved Tyr typical of cysteine-based Ub/UBL proteases). UFSP activity can be inhibited by thiol-reactive reagents such as **N-ethylmaleimide (NEM)** or by mutation of the catalytic cysteine.
- reference_id: file:DANRE/ufsp2/ufsp2-deep-research-falcon.md
supporting_text: |-
- **UFSP2**: a major **de-UFMylase** for ER-associated substrates, particularly efficiently removing UFM1 from the ribosomal protein **RPL26 (uL24)**; UFSP2 knockout causes a **drastic accumulation of UFMylated proteins** in cell systems summarized in reviews.
- term:
id: GO:0005634
label: nucleus
evidence_type: IBA
original_reference_id: GO_REF:0000033
review:
summary: Likely correct but non-core. Nuclear localization of UFSP2 has been reported
for the mammalian ortholog, and UFSP2 has been observed to be recruited to DNA
double-strand breaks. However, the primary functional site is the ER.
action: KEEP_AS_NON_CORE
reason: Nuclear localization is reported for the mammalian ortholog and there
is evidence for UFSP2 function at DNA damage sites, but the core function of
UFSP2 is at the ER where it deUFMylates RPL26 on ribosomal subunits. Falcon
deep research notes that reviews list nucleus/cytoplasm/ER localization, but
ER tethering is the strongest mechanistic conclusion, consistent with retaining
nucleus as non-core.
supported_by:
- reference_id: file:DANRE/ufsp2/ufsp2-deep-research-falcon.md
supporting_text: |-
Some reviews also list nucleus/cytoplasm/ER localization, but ER tethering is the strongest mechanistic conclusion.
- term:
id: GO:0006508
label: proteolysis
evidence_type: IBA
original_reference_id: GO_REF:0000033
review:
summary: Correct but too general. UFSP2 performs a highly specific form of proteolysis
-- it cleaves the isopeptide bond between UFM1 and its conjugated substrates.
The term proteolysis does not capture this specificity.
action: MODIFY
reason: While proteolysis is technically accurate, it is overly broad for this
specialized isopeptidase. The specific process is removal of the ubiquitin-like
modifier UFM1 from substrate proteins (deUFMylation). No dedicated GO term for
protein deUFMylation exists, so the closest correct existing BP term is
GO:0070646 protein modification by small protein removal, whose definition
explicitly covers removal of ubiquitin-like proteins (a sibling of protein
deneddylation and protein desumoylation). A dedicated protein deUFMylation term
is requested in proposed_new_terms.
proposed_replacement_terms:
- id: GO:0070646
label: protein modification by small protein removal
description: Closest correct existing BP term for UFM1 deconjugation; its
definition explicitly includes removal of ubiquitin-like proteins. Unlike
GO:0016579 (protein deubiquitination), it is not ubiquitin-specific and is
the direct parent of the analogous protein desumoylation (GO:0016926) and
protein deneddylation (GO:0000338) terms.
supported_by:
- reference_id: file:DANRE/ufsp2/ufsp2-deep-research-falcon.md
supporting_text: |-
Cross-species evidence supports that UFSP2 catalyzes:
- **Proteolytic maturation of pro-UFM1** to mature UFM1 required for UFMylation, and
- **Hydrolysis of isopeptide bonds** linking UFM1 to substrate proteins (de-UFMylation).
- term:
id: GO:0005634
label: nucleus
evidence_type: IEA
original_reference_id: GO_REF:0000044
review:
summary: Redundant with the IBA annotation for nucleus. Consistent with subcellular
localization data from the mammalian ortholog.
action: KEEP_AS_NON_CORE
reason: Redundant with the IBA-supported nucleus annotation. Retained as non-core
since the primary functional site is the ER.
- term:
id: GO:0005737
label: cytoplasm
evidence_type: IEA
original_reference_id: GO_REF:0000044
review:
summary: Correct. UFSP2 is a soluble enzyme that resides in the cytoplasm and
is recruited to the ER membrane surface where it acts on UFMylated substrates.
action: KEEP_AS_NON_CORE
reason: Cytoplasm is listed in UniProt subcellular location data and is consistent
with UFSP2 being a soluble protease. However, the mechanistic picture points to
constitutive ER tethering via the N-terminal domain interaction with DDRGK1/ODR4
rather than dynamic recruitment, so cytoplasm is retained as a non-core
localization.
supported_by:
- reference_id: file:DANRE/ufsp2/ufsp2-deep-research-falcon.md
supporting_text: |-
Some reviews also list nucleus/cytoplasm/ER localization, but ER tethering is the strongest mechanistic conclusion.
- term:
id: GO:0005783
label: endoplasmic reticulum
evidence_type: IEA
original_reference_id: GO_REF:0000044
review:
summary: Redundant with the IBA annotation for ER. Correct and consistent with
the role of UFSP2 at the ER membrane surface.
action: ACCEPT
reason: Redundant with the IBA-supported ER annotation, but independently correct
based on UniProt subcellular location data from the mammalian ortholog.
supported_by:
- reference_id: PMID:21228277
supporting_text: plays a role in the recognition of its cellular substrate C20orf116
and thus in the recruitment of UfSP2 to the endoplasmic reticulum
- reference_id: file:DANRE/ufsp2/ufsp2-deep-research-falcon.md
supporting_text: |-
The UniProt accession **Q7T347** corresponds to **Danio rerio ufsp2**, annotated as **Ufm1-specific protease 2 (UfSP2)**, a cysteine protease in the UFMylation (UFM1 conjugation) system. In the retrieved literature corpus, “UFSP2/Ufsp2” consistently refers to the **UFM1-specific protease 2** (not an unrelated gene), functioning as a UFM1-directed deconjugating enzyme and pro-UFM1 maturase, typically positioned at the **endoplasmic reticulum (ER)** through interaction with **ODR4**.
- term:
id: GO:0071567
label: deUFMylase activity
evidence_type: IEA
original_reference_id: GO_REF:0000117
review:
summary: Redundant with the IBA annotation for deUFMylase activity. Correct annotation
from ARBA machine learning models.
action: ACCEPT
reason: Redundant with the IBA-supported annotation but independently correct.
deUFMylase activity is the core molecular function.
supported_by:
- reference_id: PMID:17182609
supporting_text: cleaved the C-terminal extension of Ufm1 but not that of ubiquitin
- reference_id: file:DANRE/ufsp2/ufsp2-deep-research-falcon.md
supporting_text: |-
- **UFSP2**: a major **de-UFMylase** for ER-associated substrates, particularly efficiently removing UFM1 from the ribosomal protein **RPL26 (uL24)**; UFSP2 knockout causes a **drastic accumulation of UFMylated proteins** in cell systems summarized in reviews.
- term:
id: GO:0032790
label: ribosome disassembly
evidence_type: ISS
original_reference_id: GO_REF:0000024
review:
summary: Correct. UFSP2-mediated deUFMylation of RPL26 is a required step in the
recycling of 60S ribosomal subunits from the ER. UFMylation marks stalled ribosomes
for quality control, and UFSP2 removes UFM1 to enable ribosome recycling.
action: ACCEPT
reason: Supported by studies showing that RPL26 UFMylation at the ER triggers
ribosome quality control and that UFSP2-mediated deUFMylation is needed for
60S subunit recycling.
supported_by:
- reference_id: PMID:30626644
supporting_text: RPL26, is the principal target of UFM1 conjugation...UFMylated
RPL26 is highly enriched on ER membrane-bound ribosomes and polysomes
- reference_id: PMID:37036982
supporting_text: UFMylation of translocon-bound 60S subunits modulates the RTJ
to promote access of proteasomes and RQC machinery to ER-APs
- reference_id: PMID:38383785
supporting_text: UFM1 conjugation (UFMylation) facilitates the rescue of 60S
ribosomal subunits (60S) that are released after ribosome-associated
quality-control-mediated splitting of ribosomes that stall during
co-translational translocation of secretory proteins into the ER.
- reference_id: file:DANRE/ufsp2/ufsp2-deep-research-falcon.md
supporting_text: |-
A 2024 Nature study (“UFM1 E3 ligase promotes recycling of 60S ribosomal subunits from the ER”) provides primary, high-authority support for a model in which UFMylation machinery promotes **recycling of 60S ribosomal subunits** from ER-associated states, and it explicitly frames UFSP2 as an **ER-tethered UFM1-specific hydrolase** in this functional context.
- term:
id: GO:0006508
label: proteolysis
evidence_type: ISS
original_reference_id: GO_REF:0000024
review:
summary: Correct but too general. Same issue as the IBA proteolysis annotation.
UFSP2 performs specific isopeptide bond cleavage of UFM1 conjugates, not general
proteolysis.
action: MODIFY
reason: Overly broad for this specialized isopeptidase. Should be replaced with
a more specific term capturing UFM1 removal. As for the IBA proteolysis
annotation, no dedicated protein deUFMylation BP term exists, so the closest
correct existing term is GO:0070646 protein modification by small protein
removal (a sibling of protein desumoylation and protein deneddylation). A
dedicated protein deUFMylation term is requested in proposed_new_terms.
proposed_replacement_terms:
- id: GO:0070646
label: protein modification by small protein removal
description: Closest correct existing BP term for UFM1 deconjugation; not
ubiquitin-specific (unlike GO:0016579), with a definition that explicitly
covers removal of ubiquitin-like proteins.
supported_by:
- reference_id: file:DANRE/ufsp2/ufsp2-deep-research-falcon.md
supporting_text: |-
- **Hydrolysis of isopeptide bonds** linking UFM1 to substrate proteins (de-UFMylation).
- term:
id: GO:0071567
label: deUFMylase activity
evidence_type: ISS
original_reference_id: GO_REF:0000024
review:
summary: Redundant with IBA and IEA annotations for deUFMylase activity. Correct
and well-supported by direct experimental data from the human ortholog.
action: ACCEPT
reason: Core molecular function, well supported by ISS from human UFSP2 (Q9NUQ7).
Consistent with all other evidence lines.
supported_by:
- reference_id: PMID:17182609
supporting_text: replacement of the conserved Cys residue by Ser resulted in
a complete loss of the UfSP1 and UfSP2 activities
- reference_id: file:DANRE/ufsp2/ufsp2-deep-research-falcon.md
supporting_text: |-
UFSP2 can also reverse UFMylation on model substrates such as **DDRGK1/UFBP1** and **ASC1** in vitro/cell studies.
- term:
id: GO:0005634
label: nucleus
evidence_type: ISS
original_reference_id: GO_REF:0000024
review:
summary: Redundant with IBA and IEA nucleus annotations. Non-core localization.
action: KEEP_AS_NON_CORE
reason: Nucleus is a secondary localization site. The primary functional site
is the ER.
- term:
id: GO:0005737
label: cytoplasm
evidence_type: ISS
original_reference_id: GO_REF:0000024
review:
summary: Redundant with IEA cytoplasm annotation. Correct default localization
for this soluble protease.
action: KEEP_AS_NON_CORE
reason: Cytoplasm is listed in UniProt subcellular location data for this soluble
enzyme, but the mechanistic picture points to constitutive ER tethering via
DDRGK1/ODR4 rather than dynamic recruitment. Non-core localization.
supported_by:
- reference_id: file:DANRE/ufsp2/ufsp2-deep-research-falcon.md
supporting_text: |-
Some reviews also list nucleus/cytoplasm/ER localization, but ER tethering is the strongest mechanistic conclusion.
- term:
id: GO:0005783
label: endoplasmic reticulum
evidence_type: ISS
original_reference_id: GO_REF:0000024
review:
summary: Redundant with IBA and IEA ER annotations. Correct and core localization
for UFSP2 function.
action: ACCEPT
reason: ER is the primary functional site where UFSP2 deUFMylates RPL26 on ribosomal
subunits. Well supported by structural and functional studies.
supported_by:
- reference_id: PMID:21228277
supporting_text: plays a role in the recognition of its cellular substrate C20orf116
and thus in the recruitment of UfSP2 to the endoplasmic reticulum
- reference_id: file:DANRE/ufsp2/ufsp2-deep-research-falcon.md
supporting_text: |-
In the current pathway-centric view, this ER tethering spatially matches UFSP2 to major UFMylation events on **ER-associated ribosomes**, and thus to ER proteostasis/ribosome quality control processes.
core_functions:
- description: UFSP2 functions as a thiol-dependent isopeptidase that cleaves UFM1
from conjugated substrates (deUFMylation) and processes UFM1 precursors to their
mature form. Its principal substrate is UFMylated RPL26 on ER-bound 60S ribosomal
subunits, where deUFMylation is required for ribosome recycling during ER-associated
quality control.
molecular_function:
id: GO:0071567
label: deUFMylase activity
directly_involved_in:
- id: GO:0032790
label: ribosome disassembly
locations:
- id: GO:0005783
label: endoplasmic reticulum
substrates:
- id: UniProtKB:Q7SXA1
label: UFMylated Rpl26 (60S ribosomal protein, zebrafish ortholog)
supported_by:
- reference_id: PMID:30626644
supporting_text: a largely uncharacterized ribosomal protein, RPL26, is the principal
target of UFM1 conjugation
- reference_id: PMID:17182609
supporting_text: cleaved the C-terminal extension of Ufm1 but not that of ubiquitin
- reference_id: PMID:37036982
supporting_text: UFMylation of translocon-bound 60S subunits modulates the RTJ
to promote access of proteasomes and RQC machinery to ER-APs
- reference_id: file:DANRE/ufsp2/ufsp2-deep-research-falcon.md
supporting_text: |-
- **UFSP2**: a major **de-UFMylase** for ER-associated substrates, particularly efficiently removing UFM1 from the ribosomal protein **RPL26 (uL24)**; UFSP2 knockout causes a **drastic accumulation of UFMylated proteins** in cell systems summarized in reviews.
proposed_new_terms:
- proposed_name: protein deUFMylation
proposed_definition: The process in which the ubiquitin-fold modifier UFM1 is
cleaved from its target protein, removing one or more covalently attached UFM1
groups from a substrate. This is the deconjugation arm of the UFMylation cycle,
catalyzed by UFM1-specific proteases (e.g. UFSP2).
justification: GO provides a molecular function term for deUFMylase activity
(GO:0071567) and a biological process term for the forward reaction, protein
ufmylation (GO:0071569), but lacks a corresponding biological process term for
the reverse (deconjugation) reaction. Analogous deconjugation processes exist
for other ubiquitin-like modifiers as children of GO:0070646 protein
modification by small protein removal, namely protein deubiquitination
(GO:0016579), protein desumoylation (GO:0016926), and protein deneddylation
(GO:0000338). A protein deUFMylation term would complete this set and allow
UFSP2 deconjugation activity to be annotated to an appropriately specific BP.
proposed_parent:
id: GO:0070646
label: protein modification by small protein removal
supported_by:
- reference_id: file:DANRE/ufsp2/ufsp2-deep-research-falcon.md
supporting_text: |-
Cross-species evidence supports that UFSP2 catalyzes:
- **Proteolytic maturation of pro-UFM1** to mature UFM1 required for UFMylation, and
- **Hydrolysis of isopeptide bonds** linking UFM1 to substrate proteins (de-UFMylation).
suggested_questions:
- question: Has ufsp2 function been directly tested in zebrafish using morpholino
knockdown or CRISPR knockout? Are there any zebrafish-specific phenotypes?
experts: []
- question: Does zebrafish ufsp2 play a role in erythropoiesis or hematopoietic development,
as suggested by the UFMylation pathway's role in erythroid differentiation in
mammals?
experts: []
suggested_experiments:
- description: Generate ufsp2 CRISPR knockout zebrafish and characterize phenotypes
including skeletal development, neurological function, and ER stress markers to
determine if the mammalian disease phenotypes are conserved.
hypothesis: Loss of ufsp2 in zebrafish will cause ER stress-related phenotypes and
potentially skeletal or neurological defects analogous to human UFSP2 mutations.
- description: Perform immunofluorescence microscopy in zebrafish embryos to confirm
subcellular localization of ufsp2 at the ER, cytoplasm, and nucleus.
hypothesis: Zebrafish ufsp2 will show predominantly ER and cytoplasmic localization,
consistent with mammalian orthologs.