ADO (2-aminoethanethiol dioxygenase, also called cysteamine dioxygenase; EC 1.13.11.19) is a cytosolic non-heme mononuclear Fe(II) thiol dioxygenase of the cupin superfamily and one of only two mammalian thiol dioxygenases (the other being cysteine dioxygenase, CDO1). It has two distinct, well-established catalytic activities that act on chemically different substrate classes. First, as a cysteamine dioxygenase it oxidizes the sulfur of the small molecule cysteamine (2-aminoethanethiol) to hypotaurine using molecular oxygen, providing a second route of hypotaurine and taurine biosynthesis that complements the cysteine-sulfinate route through CDO1 and CSAD. Second, ADO is the animal N-terminal cysteine dioxygenase of the Cys/N-degron pathway: it oxidizes the thiol of N-terminal cysteine residues of substrate proteins (e.g. RGS4, RGS5 and interleukin-32) to cysteine-sulfinic acid, priming them for arginylation by ATE1 and subsequent ubiquitin-mediated degradation. Because its oxygen affinity is low (a high apparent KmO2, above the physiological range), this protein-modifying activity couples substrate stability to oxygen availability, making ADO an enzymatic cellular oxygen sensor analogous to the plant cysteine oxidases. The active site is a 3-His-ligated ferrous iron center in a jelly-roll beta-barrel; ADO can autocatalytically form a Cys220-Tyr222 thioether cross-linked cofactor that amplifies catalysis.
Definition: Catalysis of the reaction: N-terminal L-cysteinyl-[protein] + O2 = N-terminal S-hydroxy-S-oxy-L-cysteinyl-[protein] (N-terminal cysteine-sulfinic acid) + H(+). This oxygen-dependent dioxygenation of a protein N-terminal cysteine thiol primes the substrate for the Cys/N-degron pathway. Distinct from cysteamine dioxygenase activity (GO:0047800), which acts on the free small molecule cysteamine, and from cysteine dioxygenase activity (GO:0017172), which acts on free L-cysteine.
Justification: ADO (and plant cysteine oxidases) catalyze dioxygenation of the N-terminal cysteine of protein substrates (RGS4, RGS5, IL-32) as the committed enzymatic step of the Cys/N-degron oxygen-sensing pathway. This is a distinct molecular function from the small-molecule cysteamine dioxygenase activity, corresponds to a defined Rhea reaction (RHEA:70895) already used by UniProt, but has no dedicated GO MF term. A specific term is needed so that ADO's second, physiologically central catalytic activity can be annotated precisely rather than approximated by GO:0047800 plus process terms.
Supporting Evidence:
| GO Term | Evidence | Action | Reason |
|---|---|---|---|
| GO:0005739 mitochondrion | IBA GO_REF:0000033 | KEEP AS NON CORE | Summary: Phylogenetic (IBA) inference of mitochondrial localization, seeded by mouse Ado. ADO's best-characterized activity is cytosolic (N-degron oxidation of RGS4/5 and IL-32; Reactome cytosol), but mouse fractionation and human and mouse mitochondrial proteomes also place a pool in mitochondria. Not a core localization. Reason: Revised from MARK_AS_OVER_ANNOTATED. ORGANELLES check (2026-10-24): the earlier verdict (cytosolic only; mitochondrial calls are co-purification) does not hold up. The three GO:0005739 rows rest on independent experimental sources: the IBA's PAINT node PTN000539917 (PTHR22966) is seeded by mouse Ado (MGI:2685083); the Ensembl projection is from mouse Q6PDY2, whose own mitochondrion rows are HDA (MitoCarta, PMID:18614015) and IDA (fractionation of primary inguinal and brown adipocytes, PMID:37630723), so it is not circular; and the human HTP row is from MitoCoP (PMID:34800366), which was built with subtractive proteomics designed to exclude co-purifying contaminants. UniProt gives ADO no curated subcellular location at all, so nothing curated contradicts the organelle claim; the cytosolic pool (Reactome TAS, N-degron substrates RGS4/5 and IL-32) remains the core location. A dual cytosolic and mitochondrial distribution is the reading the data support, and all three rows are KEEP_AS_NON_CORE. No sub-compartment row: ADO has no TM segment and no curated or cleaved presequence; the mitochondrion-only lower band of the Ado doublet in PMID:37630723 hints at processing but is a single line, so the row stays at the root. Supporting Evidence: PMID:37630723 Intriguingly, subcellular fractionation experiments revealed that Ado is localized to both the mitochondrial and cytoplasmic compartments PMID:34800366 To overcome the limitations of individual approaches, we combined subtractive proteomics, spatial proteomics, mitochondria-specific importomics, and literature/database curation by stringent criteria. |
| GO:0016702 oxidoreductase activity, acting on single donors with incorporation of molecular oxygen, incorporation of two atoms of oxygen | IEA GO_REF:0000002 | MODIFY | Summary: Correct but generic parent dioxygenase MF from InterPro2GO (IPR012864, PCO/ADO). ADO is a true dioxygenase (incorporates both oxygen atoms), so the term is not wrong, but a more specific catalytic term is available and experimentally supported. Reason: The specific experimentally supported activity GO:0047800 (cysteamine dioxygenase activity) should be used instead of this generic parent. Proposed replacements: cysteamine dioxygenase activity |
| GO:0047800 cysteamine dioxygenase activity | IEA GO_REF:0000120 | ACCEPT | Summary: Electronic (ARBA/rule-based) call of the core cysteamine dioxygenase activity, concordant with multiple experimental annotations. This is ADO's founding and correct core molecular function; the electronic call reinforces it. |
| GO:0005515 protein binding | IPI PMID:28514442 Architecture of the human interactome defines protein commun... | MARK AS OVER ANNOTATED | Summary: Bare protein-binding IPI from a high-throughput affinity-purification interactome (BioPlex), with partner FAM76B (Q5HYJ3). Uninformative about ADO's molecular function and not one of its established interactors of functional significance. Reason: Bare "protein binding" (GO:0005515) from a large-scale screen conveys no specific function; per policy an IPI is marked over-annotated rather than removed. |
| GO:0005515 protein binding | IPI PMID:32296183 A reference map of the human binary protein interactome. | MARK AS OVER ANNOTATED | Summary: Bare protein-binding IPI from the HuRI binary (yeast two-hybrid) interactome, with partners CAP2 (P40123) and DISC1 (Q9NRI5-2). These interactions appear in the UniProt INTERACTION block but do not define a specific molecular function for ADO. Reason: Bare "protein binding" from a high-throughput binary screen is uninformative; marked over-annotated per policy for IPI annotations. |
| GO:0005515 protein binding | IPI PMID:33961781 Dual proteome-scale networks reveal cell-specific remodeling... | MARK AS OVER ANNOTATED | Summary: Bare protein-binding IPI from a proteome-scale interactome (BioPlex 3.0), partner FAM76B (Q5HYJ3). Uninformative about ADO's molecular function. Reason: Bare "protein binding" from a large-scale screen conveys no specific function; marked over-annotated per policy for IPI annotations. |
| GO:0005506 iron ion binding | IEA GO_REF:0000120 | ACCEPT | Summary: Electronic call of iron ion binding, concordant with the experimental structural evidence that ADO has a catalytic non-heme iron center. This is a core cofactor function of ADO. |
| GO:0005739 mitochondrion | IEA GO_REF:0000107 | KEEP AS NON CORE | Summary: Ensembl-Compara IEA transfer of mitochondrial localization from the mouse ortholog, whose own annotation rests on mouse proteomic (HDA) and fractionation (IDA) data. A secondary pool alongside the cytosolic core location. Reason: Revised from REMOVE. ORGANELLES check (2026-10-24): the earlier verdict (cytosolic only; mitochondrial calls are co-purification) does not hold up. The three GO:0005739 rows rest on independent experimental sources: the IBA's PAINT node PTN000539917 (PTHR22966) is seeded by mouse Ado (MGI:2685083); the Ensembl projection is from mouse Q6PDY2, whose own mitochondrion rows are HDA (MitoCarta, PMID:18614015) and IDA (fractionation of primary inguinal and brown adipocytes, PMID:37630723), so it is not circular; and the human HTP row is from MitoCoP (PMID:34800366), which was built with subtractive proteomics designed to exclude co-purifying contaminants. UniProt gives ADO no curated subcellular location at all, so nothing curated contradicts the organelle claim; the cytosolic pool (Reactome TAS, N-degron substrates RGS4/5 and IL-32) remains the core location. A dual cytosolic and mitochondrial distribution is the reading the data support, and all three rows are KEEP_AS_NON_CORE. No sub-compartment row: ADO has no TM segment and no curated or cleaved presequence; the mitochondrion-only lower band of the Ado doublet in PMID:37630723 hints at processing but is a single line, so the row stays at the root. Supporting Evidence: PMID:37630723 Intriguingly, subcellular fractionation experiments revealed that Ado is localized to both the mitochondrial and cytoplasmic compartments |
| GO:0005829 cytosol | IEA GO_REF:0000107 | ACCEPT | Summary: Cytosolic localization transferred electronically from the mouse ortholog. This is the correct core location of ADO, consistent with the Reactome TAS annotation and the enzyme's cytosolic substrates. |
| GO:0042412 taurine biosynthetic process | IEA GO_REF:0000107 | ACCEPT | Summary: ADO oxidizes cysteamine to hypotaurine, the immediate precursor of taurine, and is the sole gene product for this cytosolic route to hypotaurine/taurine. This is a correct core biological process for ADO, supported experimentally (PMID:17581819) and by Reactome. |
| GO:0005739 mitochondrion | HTP PMID:34800366 Quantitative high-confidence human mitochondrial proteome an... | KEEP AS NON CORE | Summary: Detection of ADO in the MitoCoP high-confidence human mitochondrial proteome. ADO is not discussed in the cached full text (listed in supplementary data only). The assay reports the organelle and cannot resolve a sub-compartment. Reason: Revised from MARK_AS_OVER_ANNOTATED. Under ORGANELLES rule 2 a direct whole-organelle assay is kept for what it showed, and the GOA curator saw the supplementary table we do not have. ORGANELLES check (2026-10-24): the earlier verdict (cytosolic only; mitochondrial calls are co-purification) does not hold up. The three GO:0005739 rows rest on independent experimental sources: the IBA's PAINT node PTN000539917 (PTHR22966) is seeded by mouse Ado (MGI:2685083); the Ensembl projection is from mouse Q6PDY2, whose own mitochondrion rows are HDA (MitoCarta, PMID:18614015) and IDA (fractionation of primary inguinal and brown adipocytes, PMID:37630723), so it is not circular; and the human HTP row is from MitoCoP (PMID:34800366), which was built with subtractive proteomics designed to exclude co-purifying contaminants. UniProt gives ADO no curated subcellular location at all, so nothing curated contradicts the organelle claim; the cytosolic pool (Reactome TAS, N-degron substrates RGS4/5 and IL-32) remains the core location. A dual cytosolic and mitochondrial distribution is the reading the data support, and all three rows are KEEP_AS_NON_CORE. No sub-compartment row: ADO has no TM segment and no curated or cleaved presequence; the mitochondrion-only lower band of the Ado doublet in PMID:37630723 hints at processing but is a single line, so the row stays at the root. Supporting Evidence: PMID:34800366 To overcome the limitations of individual approaches, we combined subtractive proteomics, spatial proteomics, mitochondria-specific importomics, and literature/database curation by stringent criteria. PMID:37630723 Intriguingly, subcellular fractionation experiments revealed that Ado is localized to both the mitochondrial and cytoplasmic compartments |
| GO:0005506 iron ion binding | IDA PMID:34508780 Crystal structure of human cysteamine dioxygenase provides a... | ACCEPT | Summary: Direct structural evidence (1.78 A crystal structure, PDB 7REI) that ADO binds a catalytic metal ion via a 3-His facial triad (His112/His114/His193); the native metal of thiol dioxygenases is ferrous iron. Core cofactor function. |
| GO:0047800 cysteamine dioxygenase activity | IMP PMID:29752763 Cofactor Biogenesis in Cysteamine Dioxygenase: C-F Bond Clea... | ACCEPT | Summary: Mutagenesis (Tyr222) demonstrating the cysteamine dioxygenase activity of human ADO and its Cys220-Tyr222 cross-linked catalytic amplifier. Core molecular function. |
| GO:0047800 cysteamine dioxygenase activity | IDA PMID:31273118 Conserved N-terminal cysteine dioxygenases transduce respons... | ACCEPT | Summary: Recombinant human ADO oxidizes its thiol substrates using molecular oxygen; this paper directly assays ADO catalytic activity (alongside establishing the N-terminal cysteine dioxygenase reaction). Core molecular function. |
| GO:0047800 cysteamine dioxygenase activity | IDA PMID:32601061 Characterization of the nonheme iron center of cysteamine di... | ACCEPT | Summary: Direct biochemical/spectroscopic characterization of ADO acting on cysteamine at its ferrous iron center. Core molecular function. |
| GO:0071456 cellular response to hypoxia | IDA PMID:31273118 Conserved N-terminal cysteine dioxygenases transduce respons... | ACCEPT | Summary: ADO is an enzymatic oxygen sensor: its N-terminal cysteine dioxygenase activity has a high apparent KmO2, so under hypoxia its substrates (RGS4/5, IL-32) are stabilized, transducing responses to low oxygen. Core biological process (oxygen sensing). |
| GO:0005506 iron ion binding | ISS GO_REF:0000024 | ACCEPT | Summary: Iron ion binding transferred by sequence similarity from the mouse ortholog (Q6PDY2), concordant with the human structural IDA evidence. Core cofactor function. |
| GO:0047800 cysteamine dioxygenase activity | IMP PMID:17581819 Discovery and characterization of a second mammalian thiol d... | ACCEPT | Summary: Original identification of ADO as the mammalian cysteamine dioxygenase: recombinant ADO oxidizes cysteamine to hypotaurine and RNAi of human ADO reduces hypotaurine production. Core molecular function. |
| GO:0005829 cytosol | TAS Reactome:R-HSA-6814153 | ACCEPT | Summary: Curated (Reactome) cytosolic localization of ADO for the cysteamine-to-hypotaurine reaction. This is the authoritative core location of ADO. |
| GO:0018171 peptidyl-cysteine oxidation | IDA PMID:31273118 Conserved N-terminal cysteine dioxygenases transduce respons... | NEW | Summary: Proposed new annotation capturing ADO's second core activity: it oxidizes the N-terminal cysteine of substrate proteins (RGS4, RGS5, IL-32) to cysteine-sulfinic acid, the committed step of the Cys/N-degron pathway. GO lacks a dedicated MF term for the protein N-terminal cysteine dioxygenase reaction (see proposed_new_terms), so this peptidyl-cysteine oxidation process term records the modification chemistry. Reason: Directly demonstrated by mass spectrometry showing +32 Da (two oxygen atoms) addition to the N-terminal cysteine of RGS4/RGS5 peptides by recombinant ADO. |
| GO:0031648 protein destabilization | IMP PMID:31273118 Conserved N-terminal cysteine dioxygenases transduce respons... | NEW | Summary: Proposed new annotation: by dioxygenating N-terminal cysteines, ADO destabilizes its substrate proteins (RGS4, RGS5, IL-32), priming them for arginylation and degradation. Loss of ADO constitutively stabilizes these substrates. Reason: CRISPR inactivation of ADO causes constitutive upregulation of RGS4/RGS5, demonstrating that ADO activity destabilizes these proteins. |
Loading supporting content…
Download this section (compressed HTML)Q: Beyond RGS4, RGS5 and IL-32, what is the full physiological substrate repertoire of human ADO in the Cys/N-degron pathway, and how is substrate selectivity encoded?
Q: How is the balance between ADO's small-molecule (cysteamine) and protein (N-terminal cysteine) activities regulated in vivo, and does the Cys220-Tyr222 cross-linked cofactor tune this balance?
Experiment: N-terminomics (e.g. TAILS) comparing wild-type and ADO-knockout human cells under normoxia and hypoxia to identify the endogenous set of N-terminal-cysteine proteins whose stability is oxygen- and ADO-dependent.
Experiment: Quantitative subcellular fractionation plus immunofluorescence of endogenous ADO to resolve the reported mitochondrial signal versus its curated cytosolic localization, with protease protection of isolated mitochondria to test whether the mitochondrial pool is imported (and whether the lower band of the doublet is a processed form).
Loading supporting content…
Download this section (compressed HTML)Loading supporting content…
Download this section (compressed HTML)