GATM encodes L-arginine:glycine amidinotransferase (AGAT), the enzyme catalysing the first, rate-limiting step of creatine biosynthesis. It transfers an amidino group from arginine to glycine, forming guanidinoacetate and ornithine; GAMT then methylates guanidinoacetate to creatine. The principal mitochondrial homodimer acts in the intermembrane compartment through a Cys407βHis303βAsp254 catalytic system and an amidino-cysteine intermediate. Expression is broad and highest in kidney. Recombinant human GATM also accepts other amino compounds in vitro, without established physiological flux for each reaction. Biallelic loss causes AGAT deficiency (cerebral creatine deficiency syndrome 3), with cognitive and muscle impairment. Distinct dominant aggregation-prone variants cause Fanconi renotubular syndrome 1 with proximal tubular dysfunction and kidney failure. Adipocyte GATM supports thermogenic capacity in mouse models.
| GO Term | Evidence | Action | Reason |
|---|---|---|---|
| GO:0005515 protein binding | IPI PMID:32814053 Interactome Mapping Provides a Network of Neurodegenerative ... | REMOVE | Summary: Reported binding to SCGB1A1. Reason: Positive GATMβSCGB1A1 interaction evidence is retained. Generic protein binding does not identify a biological activity, and this pair does not establish a supported specific molecular-function replacement. Supporting Evidence: |
| GO:0005515 protein binding | IPI PMID:32814053 Interactome Mapping Provides a Network of Neurodegenerative ... | REMOVE | Summary: Reported binding to GRB10. Reason: Positive GATMβGRB10 interaction evidence is retained. Generic protein binding does not identify a biological activity, and this pair does not establish a supported specific molecular-function replacement. The literal partner isoform is preserved; the gene-level supplement does not independently resolve its splice sequence. Supporting Evidence: |
| GO:0005515 protein binding | IPI PMID:32814053 Interactome Mapping Provides a Network of Neurodegenerative ... | MODIFY | Summary: Supported binding to an informative partner class. Reason: The positive GATMβITGB3BP pair can be represented as coregulator binding because ITGB3BP has an established coregulator role. This does not assign transcriptional regulation to GATM. Its isoform-specific nuclear-receptor coactivation is not generalized to all products. The interaction source and independent partner-class record are distinct; native co-localization or a physiological consequence is not established. Proposed replacements: transcription coregulator binding |
| GO:0005515 protein binding | IPI PMID:32814053 Interactome Mapping Provides a Network of Neurodegenerative ... | REMOVE | Summary: Reported binding to TNFAIP1. Reason: Positive GATMβTNFAIP1 interaction evidence is retained. Generic protein binding does not identify a biological activity, and this pair does not establish a supported specific molecular-function replacement. Supporting Evidence: |
| GO:0005515 protein binding | IPI PMID:32814053 Interactome Mapping Provides a Network of Neurodegenerative ... | REMOVE | Summary: Reported binding to BECN1. Reason: Positive GATMβBECN1 interaction evidence is retained. Generic protein binding does not identify a biological activity, and this pair does not establish a supported specific molecular-function replacement. Supporting Evidence: |
| GO:0005515 protein binding | IPI PMID:32814053 Interactome Mapping Provides a Network of Neurodegenerative ... | MODIFY | Summary: Supported binding to an informative partner class. Reason: The positive GATMβLPIN1 pair supports phosphatase binding because LPIN1 is a phosphatidate phosphatase. This is not protein-phosphatase binding, and it does not assign phosphatase activity, lipid synthesis or regulation of LPIN1 to GATM. The interaction source and independent partner-class record are distinct; native co-localization or a physiological consequence is not established. Proposed replacements: phosphatase binding |
| GO:0005515 protein binding | IPI PMID:32814053 Interactome Mapping Provides a Network of Neurodegenerative ... | REMOVE | Summary: Reported binding to ADAMTSL4. Reason: Positive GATMβADAMTSL4 interaction evidence is retained. Generic protein binding does not identify a biological activity, and this pair does not establish a supported specific molecular-function replacement. The literal partner isoform is preserved; the gene-level supplement does not independently resolve its splice sequence. Supporting Evidence: |
| GO:0005515 protein binding | IPI PMID:32814053 Interactome Mapping Provides a Network of Neurodegenerative ... | REMOVE | Summary: Reported binding to WDR88. Reason: Positive GATMβWDR88 interaction evidence is retained. Generic protein binding does not identify a biological activity, and this pair does not establish a supported specific molecular-function replacement. The literal partner isoform is preserved; the gene-level supplement does not independently resolve its splice sequence. Supporting Evidence: |
| GO:0005515 protein binding | IPI PMID:32814053 Interactome Mapping Provides a Network of Neurodegenerative ... | REMOVE | Summary: Reported binding to TMEM185A. Reason: Positive GATMβTMEM185A interaction evidence is retained. Generic protein binding does not identify a biological activity, and this pair does not establish a supported specific molecular-function replacement. Supporting Evidence: |
| GO:0005515 protein binding | IPI PMID:32814053 Interactome Mapping Provides a Network of Neurodegenerative ... | MODIFY | Summary: Supported binding to an informative partner class. Reason: The positive GATMβRPUSD4 pair supports enzyme binding: RPUSD4 catalyses RNA pseudouridylation. The interaction does not assign RNA modification or translation activity to GATM, and no more specific generic isomerase-binding term was found. The interaction source and independent partner-class record are distinct; native co-localization or a physiological consequence is not established. Proposed replacements: enzyme binding |
| GO:0005515 protein binding | IPI PMID:32814053 Interactome Mapping Provides a Network of Neurodegenerative ... | REMOVE | Summary: Reported binding to SIRPG. Reason: Positive GATMβSIRPG interaction evidence is retained. SIRPG is described as a probable receptor with an adhesion role; its signaling-receptor classification is not resolved in the inspected source. Generic binding is uninformative, and a more specific binding term is not established here. Supporting Evidence: |
| GO:0005515 protein binding | IPI PMID:32814053 Interactome Mapping Provides a Network of Neurodegenerative ... | MODIFY | Summary: Supported binding to an informative partner class. Reason: The positive GATMβEIF2B4 pair supports initiation-factor binding. EIF2B4 is a regulatory polypeptide of eIF2B, fitting the term definition. Binding does not establish GATM membership in eIF2B, nucleotide exchange activity or regulation of translation. The interaction source and independent partner-class record are distinct; native co-localization or a physiological consequence is not established. Proposed replacements: translation initiation factor binding |
| GO:0005515 protein binding | IPI PMID:32814053 Interactome Mapping Provides a Network of Neurodegenerative ... | REMOVE | Summary: Reported binding to COMMD10. Reason: Positive GATMβCOMMD10 interaction evidence is retained. Generic protein binding does not identify a biological activity, and this pair does not establish a supported specific molecular-function replacement. Supporting Evidence: |
| GO:0005737 cytoplasm | IEA GO_REF:0000044 | KEEP AS NON CORE | Summary: Retain the broad compartment with its alternative-product source context. Reason: UniProt explicitly assigns product 2 to cytoplasm, although the inspected sequence submission and original paper do not establish a localization assay. Retain this source context instead of replacing it with the principal mitochondrial compartment. The incoming GOA assertion is unqualified, so no isoform field is added. Broad cytoplasm is compatible with the independently established mitochondrial location; product-2 experimental localization remains unresolved. |
| GO:0005739 mitochondrion | HTP PMID:34800366 Quantitative high-confidence human mitochondrial proteome an... | ACCEPT | Summary: Retain the mitochondrial claim with precise atlas provenance. Reason: Mitochondrial localization is supported independently of this atlas. The original MitoCoP supplementary table includes GATM through its literature pool rather than as a newly identified or validated proteomics hit. Preserve the HTP source fields without claiming a GATM-specific new mass-spectrometry measurement; inclusion supports the established compartment. |
| GO:0005739 mitochondrion | IDA GO_REF:0000052 | ACCEPT | Summary: Mitochondrial localization supports the principal enzymatic role. Reason: Retain this informative organelle annotation on the combined mitochondrial evidence. The incoming HPA experimental citation is preserved; its target images were not independently inspected. The accepted gene-level compartment does not assign every natural product to mitochondria. |
| GO:0005743 mitochondrial inner membrane | IEA GO_REF:0000044 | KEEP AS NON CORE | Summary: Peripheral association with the mitochondrial inner membrane. Reason: UniProt assigns product 1 to the intermembrane side of the inner membrane and qualifies its attachment as probable. Retain that peripheral association as a secondary location context; it does not establish an integral membrane topology or a membrane-spanning catalytic mechanism. Supporting Evidence: UniProt:P50440 |
| GO:0005758 mitochondrial intermembrane space | IBA GO_REF:0000033 | ACCEPT | Summary: Catalysis in the mitochondrial intermembrane space. Reason: The actual PAINT ancestor places the activity in this compartment, with the human experimental annotation among its descendants. The target follows the verified ancestral path without a retained loss assertion. Independent curated localization and the GATM reaction model support this core location; the target appearing among the donors is valid experimental grounding. |
| GO:0005758 mitochondrial intermembrane space | IDA PMID:9218780 Crystal structure and mechanism of human L-arginine:glycine ... | ACCEPT | Summary: Intermembrane-space localization accompanies the core activity. Reason: The human enzyme is established as mitochondrial, and the curatorβs intermembrane-space annotation agrees with UniProt and the compartmentalized GATM reaction model. The inspected structural abstract does not expose the localization experiment, so no independent imaging or fractionation reading is claimed. Retain the correct location without treating that access limit as contrary evidence. |
| GO:0005758 mitochondrial intermembrane space | TAS Reactome:R-HSA-71275 | ACCEPT | Summary: The curated reaction places GATM activity in the intermembrane space. Reason: The original human Reactome event explicitly names the GATM dimer catalyst in the mitochondrial intermembrane space, with arginine and glycine inputs and guanidinoacetate and ornithine outputs. This supports the core compartment as model evidence. The inspected event is not a new localization assay, and the downstream physical-entity record was unavailable in this private access attempt. Supporting Evidence: Reactome:R-HSA-71275 |
| GO:0006601 creatine biosynthetic process | IBA GO_REF:0000033 | ACCEPT | Summary: Inherited participation in creatine biosynthesis. Reason: The valid PAINT ancestral assertion is supported by direct human evidence for the pathwayβs guanidinoacetate-forming step. GATM catalyses that step itself; downstream creatine utilization and tissue phenotypes are distinct from this core biosynthetic role. |
| GO:0006601 creatine biosynthetic process | IDA PMID:9218780 Crystal structure and mechanism of human L-arginine:glycine ... | ACCEPT | Summary: Catalytic entry step of creatine synthesis. Reason: GATM forms guanidinoacetate from arginine and glycine, performing the first chemical step of creatine biosynthesis before GAMT methylation. The human structural study and independent purified-enzyme evidence support this direct pathway role. The inspected structural abstract does not establish every experimental construct detail. |
| GO:0006601 creatine biosynthetic process | IEA GO_REF:0000041 | ACCEPT | Summary: Electronic pathway assignment agrees with the measured reaction. Reason: The creatine-biosynthesis assignment agrees with the experimentally characterized guanidinoacetate-producing reaction. GATM performs pathway chemistry, rather than merely responding to creatine availability. This judgment does not claim an independent reading of the UniPathway record. |
| GO:0007611 learning or memory | IMP PMID:26490222 Arginine:glycine amidinotransferase (AGAT) deficiency: Clini... | MARK AS OVER ANNOTATED | Summary: Cognitive impairment is an established consequence of GATM deficiency. Reason: Formal cognitive testing and treatment observations establish genuine impairment in AGAT deficiency. They do not identify GATM performing information acquisition, processing, storage or retrieval. Its demonstrated role is creatine-precursor synthesis; metabolic necessity and clinical consequences alone do not establish participation in learning or memory. Retain the positive cognitive findings while narrowing this process interpretation, as for the atrophy assertion. Supporting Evidence: PMID:26490222 Late treated patients had limited improvement of cognitive functions. |
| GO:0014889 muscle atrophy | IMP PMID:26490222 Arginine:glycine amidinotransferase (AGAT) deficiency: Clini... | MARK AS OVER ANNOTATED | Summary: Muscle atrophy is a consequence of GATM deficiency. Reason: The original study documents biopsy atrophy, so the phenotype is real. However, loss of precursor synthesis produces muscle dysfunction; it does not establish that GATM performs the muscle-shrinking process described by this term. Retain the positive clinical evidence while distinguishing downstream deficiency pathology from participation in atrophy. Supporting Evidence: PMID:26490222 8 patients also had myopathy/proximal muscle weakness. |
| GO:0015067 amidinotransferase activity | IDA PMID:36543883 Guanidino acid hydrolysis by the human enzyme annotated as a... | KEEP AS NON CORE | Summary: Retain the measured non-glycine acceptor breadth as a secondary catalytic activity. Reason: The human GATM acceptor panel demonstrated activity with GABA, beta-alanine and taurine as well as glycine. Retain the broad amidinotransferase annotation for this measured secondary chemistry. Glycine remains the preferred acceptor and the sole physiological core represented here; this in-vitro breadth does not establish additional human pathways or a distinct natural-product function. Supporting Evidence: PMID:36543883 GABA, Ξ²-alanine and taurine were accepted as substrates |
| GO:0015067 amidinotransferase activity | IEA GO_REF:0000002 | KEEP AS NON CORE | Summary: Retain broad amidinotransferase activity with independent experimental corroboration. Reason: The InterPro family mapping supports amidino-group transfer. Independent human GATM assays demonstrate non-glycine acceptor activity, so this broad term preserves measured chemistry beyond the glycine-specific child. Retain it as non-core alongside the experimental broad-activity annotation. The family mapping alone does not establish the complete acceptor range; the glycine reaction remains the single physiological core, without assigning additional human pathways. |
| GO:0015068 glycine amidinotransferase activity | EXP PMID:36543883 Guanidino acid hydrolysis by the human enzyme annotated as a... | ACCEPT | Summary: Direct human activity with the principal acceptor glycine. Reason: Glycine is the preferred acceptor in the inspected human GATM experiment. This supports the specific core reaction. The reference assessment distinguishes the ornithine readout, engineered protein and unresolved physiological significance of other acceptors. Supporting Evidence: PMID:36543883 highest activity with glycine |
| GO:0015068 glycine amidinotransferase activity | EXP PMID:3800397 The purification and characterization of human kidney L-argi... | ACCEPT | Summary: Purified human kidney enzyme catalyses the glycine reaction. Reason: Purified human kidney amidinotransferase was enzymatically characterized with arginine and glycine. This direct human evidence supports the core catalytic activity. The accessible abstract establishes the substrate assay and dimeric preparation, but does not resolve a modern splice product or all purification details. Supporting Evidence: PMID:3800397 The Km for arginine and glycine were both 2.5 mM |
| GO:0015068 glycine amidinotransferase activity | EXP PMID:9266688 Substrate binding and catalysis by L-arginine:glycine amidin... | ACCEPT | Summary: Catalytic mutagenesis supports the amidino-transfer mechanism. Reason: Human recombinant mutagenesis and kinetic characterization support the catalytic role of Cys407, His303 and Asp254. The abstract distinguishes nearby Cys410 from catalytic residues. These engineered-mutant results support the enzyme activity without establishing the function of the natural C-terminally altered product. Supporting Evidence: PMID:9266688 key residues for activity and are part of a disjoint triad |
| GO:0015068 glycine amidinotransferase activity | IBA GO_REF:0000033 | ACCEPT | Summary: Inherited glycine amidinotransferase chemistry. Reason: The actual PAINT ancestral placement includes human GATM, matches the canonical sequence and has no retained loss annotation on its path. Human experimental evidence independently grounds the catalytic assertion. The target in the descendant evidence is legitimate support, not circularity. |
| GO:0015068 glycine amidinotransferase activity | IDA PMID:27233232 Arginine-Glycine Amidinotransferase Deficiency and Functiona... | ACCEPT | Summary: Wild-type GATM activity provides the variant-assay comparator. Reason: The authentic abstract reports functional testing of GATM variants against wild-type activity. This supports the established catalytic annotation together with independent human enzymology. Full assay Methods and the novel transcript accession remain uninspected; the result is not assigned to every natural product. Supporting Evidence: PMID:27233232 seven missense variants retaining 0% of wild-type GATM activity |
| GO:0015068 glycine amidinotransferase activity | IDA PMID:9218780 Crystal structure and mechanism of human L-arginine:glycine ... | ACCEPT | Summary: Human substrate/product-bound structure supports catalysis. Reason: The human recombinant enzyme structures and proposed CysβHisβAsp mechanism support glycine amidinotransferase activity. Independent measured human activity strengthens the assignment. The inspected abstract is not a full Methods reading and does not identify an isolated natural splice product. |
| GO:0015068 glycine amidinotransferase activity | IEA GO_REF:0000120 | ACCEPT | Summary: The mapped reaction is GATMβs established catalytic step. Reason: The asserted RHEA/EC reaction transfers an amidino group from arginine to glycine, producing ornithine and guanidinoacetate. This is the directly characterized human enzyme activity, not a pathway attribution based only on family membership. |
| GO:0070062 extracellular exosome | HDA PMID:23533145 In-depth proteomic analyses of exosomes isolated from expres... | KEEP AS NON CORE | Summary: Observed occurrence in a urinary extracellular-vesicle preparation. Reason: Original supplementary Table 2 directly identifies P50440/GATM in the urinary extracellular-vesicle preparation, with two unique peptides in each duplicate non-cancer pooled analysis. Retain this non-core experimental occurrence. A cellular-component annotation does not require a demonstrated extracellular enzymatic function. The pooled preparation and contamination caveat limit conclusions about donor replication, natural isoform and vesicle topology. Supporting Evidence: PMID:23533145 exosome preparations were characterized by a shotgun proteomics procedure |
| GO:0120162 positive regulation of cold-induced thermogenesis | IEA GO_REF:0000107 | KEEP AS NON CORE | Summary: Metabolic support for cold-induced thermogenic capacity. Reason: Mouse adipocyte-specific GATM loss impairs the cold-temperature response, supporting a contextual positive regulatory contribution through creatine precursor supply. GATM is not the heat-dissipating reaction. Dietary creatine rescue concerned sustained adrenergic oxygen consumption, not the cold-temperature endpoint; direct human perturbation was not reported. Supporting Evidence: PMID:28844881 impaired thermal homeostasis in response to environmental cold |
| GO:0120162 positive regulation of cold-induced thermogenesis | ISS PMID:28844881 Genetic Depletion of Adipocyte Creatine Metabolism Inhibits ... | KEEP AS NON CORE | Summary: Metabolic support for cold-induced thermogenic capacity. Reason: Mouse adipocyte-specific GATM loss impairs the cold-temperature response, supporting a contextual positive regulatory contribution through creatine precursor supply. GATM is not the heat-dissipating reaction. Dietary creatine rescue concerned sustained adrenergic oxygen consumption, not the cold-temperature endpoint; direct human perturbation was not reported. Supporting Evidence: PMID:28844881 impaired thermal homeostasis in response to environmental cold |
Loading supporting contentβ¦
Download this section (compressed HTML)Q: What are the endogenous locations, processing states and catalytic activities of the three sequence-defined human GATM products?
Q: Do the non-glycine acceptor reactions carry appreciable flux at physiological substrate concentrations?
Q: Do the specific interaction partners alter GATM activity or localization in the cells where both proteins are expressed?
Experiment: Express sequence-verified natural products at matched levels and compare endogenous localization, processing and arginine/glycine turnover; include engineered AT38/AT56 only as separate controls.
Hypothesis: The three natural products may differ in processing, compartment or catalytic activity; sequence variation alone does not establish the outcome.
Type: Natural-product localization and activity
Experiment: Use isotope tracing to compare guanidinoacetate synthesis with non-glycine products under physiological substrate conditions, with GATM loss and rescue controls.
Hypothesis: Non-glycine acceptor activity may contribute measurable cellular flux under some conditions, but in-vitro turnover alone does not establish that contribution.
Type: Substrate flux tracing
Loading supporting contentβ¦
Download this section (compressed HTML)Loading supporting contentβ¦
Download this section (compressed HTML)Loading supporting contentβ¦
Download this section (compressed HTML)