AGK

Existing Annotations Review

GO Term	Evidence	Action	Reason
GO:0001729 ceramide kinase activity	IBA GO_REF:0000033	REMOVE	Summary: Human AGK substrate assays detected phosphorylation of monoacylglycerols and diacylglycerols but not ceramide. Reason: Human AGK expressed in mammalian cells produced MAG- and DAG-phosphorylation products but no significant ceramide product in PMID:15939762, which explicitly discusses the positive bacterial mouse-MuLK result. PMID:16269826 also reports a negative MuLK assay, while cautioning that its substrate conclusions were not definitive. These data favor excluding ceramide kinase from the human assignment under the tested conditions; they do not prove a universal absence under every membrane or assay condition. PAINT places the function on PTN008994514 using mouse Agk and fly Mulk experimental descendants, so the disagreement is with inheritance of this substrate activity on the human branch, not with donor count or a presumed CERK-only clade. The positive fly evidence in PMID:22069480 includes increased embryonic C1P after Dmulk overexpression. This is relevant ortholog evidence, although it does not isolate purified Dmulk or human AGK ceramide phosphorylation. PMID:18004883 supports broad MuLK glycerolipid stereoselectivity; its abstract does not establish a human ceramide assay. Neither source removes the need to reconcile the positive ortholog findings with the human substrate tests. Propagation Review Root cause: PROPAGATION BAD Failure modes: FUNCTIONAL DIVERGENCE Sources checked: PANTHER:PTN008994514 · PTN008994514 SUPPORTS SOURCE BUT NOT TARGET Cached PTHR12358 IBD has mouse Agk and fly Mulk descendants. PMID:22069480 supplies positive fly evidence, with Dmulk overexpression increasing embryonic C1P about 3.3-fold. The human-branch FUNCTIONAL_DIVERGENCE argument is made against this genuine ortholog evidence using the negative human substrate assays; assay context versus evolutionary divergence remains to be resolved. Supporting Evidence: PMID:15939762 Significant phosphorylated products were only detected with monoacylglycerols and diacylglycerols as substrates, but not with any other lipid tested, including ceramide and sphingosine PMID:16269826 No evidence for phosphorylation of ceramide by the recently described multiple lipid kinase was found PMID:22069480 a 3.3-fold increase in C1P levels upon Dmulk overexpression.
GO:0004143 ATP-dependent diacylglycerol kinase activity	IBA GO_REF:0000033	ACCEPT	Summary: Phylogenetic propagation of ATP-dependent diacylglycerol kinase activity is supported by direct human enzymology. Reason: AGK directly phosphorylates diacylglycerol to phosphatidic acid in the human study and UniProt records EC 2.7.1.107 for this activity. Propagation Review Root cause: NO FAILURE CORE Sources checked: PANTHER:PTN008994514 · PTN008994514 SUPPORTS TRANSFER The seeded ancestral assertion is independently supported by the human localization or glycerolipid-kinase experiments in PMID:15939762. Human AGK among descendant evidence is legitimate experimental grounding. Supporting Evidence: PMID:15939762 that phosphorylates monoacylglycerol and diacylglycerol to form LPA and PA, respectively file:human/AGK/AGK-uniprot.txt Lipid kinase that can phosphorylate both monoacylglycerol and diacylglycerol to form lysophosphatidic acid (LPA) and phosphatidic acid (PA), respectively
GO:0005739 mitochondrion	IBA GO_REF:0000033	ACCEPT	Summary: AGK is a mitochondrial protein with lipid-kinase and TIM22-complex functions. Reason: The core AGK functions occur in mitochondria: the original lipid-kinase paper localized AGK to mitochondria and later TIM22 studies place AGK at the mitochondrial inner membrane/intermembrane-space face. Propagation Review Root cause: NO FAILURE CORE Sources checked: PANTHER:PTN008994514 · PTN008994514 SUPPORTS TRANSFER The seeded ancestral assertion is independently supported by the human localization or glycerolipid-kinase experiments in PMID:15939762. Human AGK among descendant evidence is legitimate experimental grounding. Supporting Evidence: PMID:15939762 Confocal microscopy and subcellular fractionation suggest that AGK is localized to the mitochondria file:human/AGK/AGK-uniprot.txt Mitochondrion inner membrane
GO:0005886 plasma membrane	IBA GO_REF:0000033	UNDECIDED	Summary: The plasma-membrane assignment remains unresolved from the recovered phylogenetic and fractionation evidence. Reason: The seeded PAINT source is PTN000270644, but the current cached slice does not reconstruct its plasma-membrane assertion. PMID:15939762 detects less AGK in the P4 plasma-membrane fraction than in P2 mitochondria. Figure 2 also assayed MOG kinase activity in each fraction and reports the highest AGK-specific activity in P2; the experiment was not limited to protein abundance. The P4-specific quantitative result and marker purity were not independently resolved here, so a functional plasma-membrane pool cannot be confidently assigned from the recovered evidence. Retain uncertainty rather than infer either genuine residence or contamination from the small fraction signal. Propagation Review Root cause: UNRESOLVED Sources checked: PANTHER:PTN000270644 · PTN000270644 UNRESOLVED The seeded ancestral node is preserved. Its plasma-membrane IBD was not reconstructed from the current PAINT slice. Human fractionation detects a smaller P4 signal but does not resolve a functional resident pool in this audit. Supporting Evidence: PMID:15939762 Much less AGK was present in the P3 fraction containing intracellular membranes of the ER and Golgi or in the P4 plasma membrane fraction. PMID:15939762 AGK activity was also determined in each subcellular fraction with MOG as substrate.
GO:0046513 ceramide biosynthetic process	IBA GO_REF:0000033	UNDECIDED	Summary: The ancestral ceramide-biosynthesis inference is unresolved; negative phosphorylation assays alone do not test this biosynthetic process. Reason: The cached PTHR12358 IBD places ceramide biosynthesis on PTN008994514 using mouse Agk evidence. The mouse-source trail points to PMID:15252046, whose accessible abstract describes lipid phosphorylation; its complete process evidence was not recovered. Formation of ceramide must be distinguished from phosphorylation of ceramide to ceramide phosphate. Human negative substrate assays alone therefore do not resolve this process transfer; retain UNDECIDED pending the donor pathway evidence. Propagation Review Root cause: UNRESOLVED Sources checked: PANTHER:PTN008994514 · PTN008994514 UNRESOLVED The ceramide-biosynthesis assertion is placed on the AGK/Mulk ancestor using mouse Agk evidence. Kinase-negative target assays do not independently test formation of ceramide; pathway participation remains unresolved.
GO:0047620 acylglycerol kinase activity	IBA GO_REF:0000033	ACCEPT	Summary: Acylglycerol kinase activity is a core AGK molecular function. Reason: This is the defining lipid-kinase activity of AGK and is directly supported by the human JCB study. GO:0047620 represents monoacylglycerol phosphorylation (EC 2.7.1.94); the distinct DAG reaction is represented by GO:0004143. Propagation Review Root cause: NO FAILURE CORE Sources checked: PANTHER:PTN008994514 · PTN008994514 SUPPORTS TRANSFER The seeded ancestral assertion is independently supported by the human localization or glycerolipid-kinase experiments in PMID:15939762. Human AGK among descendant evidence is legitimate experimental grounding. Supporting Evidence: PMID:15939762 that phosphorylates monoacylglycerol and diacylglycerol to form LPA and PA, respectively file:human/AGK/AGK-uniprot.txt Lipid kinase that can phosphorylate both monoacylglycerol and diacylglycerol to form lysophosphatidic acid (LPA) and phosphatidic acid (PA), respectively
GO:0046512 sphingosine biosynthetic process	IBA GO_REF:0000033	UNDECIDED	Summary: The ancestral sphingosine-biosynthesis inference is unresolved; negative phosphorylation assays alone do not test this biosynthetic process. Reason: The cached PTHR12358 IBD places sphingosine biosynthesis on the broad eukaryotic PTN000270644 node using SPHK1/SPHK2 evidence. The exact experimental process assertions and their inheritance into the AGK branch were not recovered. Sphingosine formation is distinct from its phosphorylation, and a negative AGK sphingosine-kinase assay cannot alone decide this biosynthetic process. Retain UNDECIDED. Propagation Review Root cause: UNRESOLVED Sources checked: PANTHER:PTN000270644 · PTN000270644 UNRESOLVED The sphingosine-biosynthesis assertion is at a broad eukaryotic ancestor containing sphingosine-kinase descendants. The exact biosynthetic role inherited by AGK is unresolved; lack of sphingosine phosphorylation alone does not resolve lipid formation.
GO:0001729 ceramide kinase activity	IEA GO_REF:0000120	REMOVE	Summary: Human AGK substrate assays detected phosphorylation of monoacylglycerols and diacylglycerols but not ceramide. Reason: Human AGK expressed in mammalian cells produced MAG- and DAG-phosphorylation products but no significant ceramide product in PMID:15939762, which explicitly discusses the positive bacterial mouse-MuLK result. PMID:16269826 also reports a negative MuLK assay, while cautioning that its substrate conclusions were not definitive. These data favor excluding ceramide kinase from the human assignment under the tested conditions; they do not prove a universal absence under every membrane or assay condition. This combined IEA has ortholog, EC and Rhea inputs. In particular, cached UniProt attaches RHEA:43312 (C6-ceramide phosphorylation) to PMID:15939762 even though Figure 1 tests C6-ceramide with a negative result; that source attribution does not support the positive reaction. Propagation Review Root cause: SOURCE BAD Failure modes: SOURCE MISCITATION FUNCTIONAL DIVERGENCE Sources checked: UniProtKB:Q9ESW4 · Agk SUPPORTS SOURCE BUT NOT TARGET Mouse Agk is the ortholog donor; positive recombinant MuLK enzymology conflicts with the tested human substrate profile. ensembl:ENSMUSP00000031977 UNRESOLVED GOA lists this mouse protein alongside Q9ESW4; the current Ensembl model was not independently retrieved. RHEA:17929 SUPPORTS SOURCE BUT NOT TARGET Cached UniProt assigns the general ceramide reaction by similarity to mouse Q9ESW4, rather than a positive human assay. RHEA:43312 SOURCE BAD Cached UniProt cites human PMID:15939762 for C6-ceramide phosphorylation, but its Figure 1 reports no significant product with that substrate. EC:2.7.1.138 SUPPORTS SOURCE BUT NOT TARGET The reaction classification is ceramide kinase; the human assignment inherits the disputed substrate claim. Supporting Evidence: PMID:15939762 Significant phosphorylated products were only detected with monoacylglycerols and diacylglycerols as substrates, but not with any other lipid tested, including ceramide and sphingosine PMID:16269826 No evidence for phosphorylation of ceramide by the recently described multiple lipid kinase was found
GO:0004143 ATP-dependent diacylglycerol kinase activity	IEA GO_REF:0000120	ACCEPT	Summary: Automated diacylglycerol kinase activity is supported by direct human evidence. Reason: The IEA call is consistent with the human biochemical characterization and UniProt catalytic-activity record. Propagation Review Root cause: NO FAILURE CORE Sources checked: ARBA:ARBA00028046 UNRESOLVED The rule identifier is present in GOA; its current predicates were not retrieved. The activity judgment rests on independent human enzymology. UniProtKB:Q9ESW4 · Agk SUPPORTS TRANSFER Mouse Agk donor chemistry agrees with the independently demonstrated human MAG/DAG reaction. ensembl:ENSMUSP00000031977 UNRESOLVED Seeded Ensembl donor identifier; current model was not separately inspected. RHEA:10272 SUPPORTS TRANSFER The exact reaction is present in the cached human UniProt catalytic-activity section with experimental PMID:15939762 support. RHEA:40327 SUPPORTS TRANSFER The exact reaction is present in the cached human UniProt catalytic-activity section with experimental PMID:15939762 support. EC:2.7.1.107 SUPPORTS TRANSFER The EC reaction class matches the substrate-specific human enzyme assay and cached UniProt reaction. Supporting Evidence: PMID:15939762 that phosphorylates monoacylglycerol and diacylglycerol to form LPA and PA, respectively file:human/AGK/AGK-uniprot.txt Lipid kinase that can phosphorylate both monoacylglycerol and diacylglycerol to form lysophosphatidic acid (LPA) and phosphatidic acid (PA), respectively
GO:0005743 mitochondrial inner membrane	IEA GO_REF:0000044	ACCEPT	Summary: Mitochondrial inner-membrane localization is well supported. Reason: Multiple direct AGK/TIM22 studies and UniProt place AGK at the mitochondrial inner membrane. Propagation Review Root cause: NO FAILURE CORE Sources checked: UniProtKB-SubCell:SL-0168 SUPPORTS TRANSFER Cached UniProt explicitly records mitochondrial inner membrane with the human TIM22 publications; the vocabulary mapping agrees with this localization. Supporting Evidence: PMID:15939762 Confocal microscopy and subcellular fractionation suggest that AGK is localized to the mitochondria file:human/AGK/AGK-uniprot.txt Mitochondrion inner membrane PMID:28712724 TIM22 complex in the mitochondrial inner membrane
GO:0005758 mitochondrial intermembrane space	IEA GO_REF:0000044	ACCEPT	Summary: Mitochondrial intermembrane-space localization is well supported. Reason: UniProt and the TIM22 literature place AGK in the intermembrane-space side of the inner membrane, matching this localization annotation. Propagation Review Root cause: NO FAILURE CORE Sources checked: UniProtKB-SubCell:SL-0169 SUPPORTS TRANSFER Cached UniProt explicitly records mitochondrial intermembrane space with the human TIM22 publications; the vocabulary mapping agrees with this localization. Supporting Evidence: file:human/AGK/AGK-uniprot.txt Mitochondrion intermembrane space file:human/AGK/AGK-uniprot.txt Localizes in the mitochondrion intermembrane space, where it associates with the inner membrane
GO:0016301 kinase activity	IEA GO_REF:0000002	MODIFY	Summary: Direct human substrate assays permit refinement of generic kinase activity to the two glycerolipid-kinase reactions. Reason: The InterPro catalytic-domain mapping correctly identifies a kinase, but it does not specify the phosphorylated substrate. Human substrate assays in PMID:15939762 directly resolve MAG-to-LPA and DAG-to-PA chemistry, supporting GO:0047620 and GO:0004143 as informative replacements. This is a molecular-chemistry refinement based on measured substrates, not an inference from family name alone. It differs from assigning a specific compartment to every broad membrane source: the localization evidence establishes membrane association without resolving the attachment site of every pool. The replacements are already seeded and do not create new annotation coverage. Propagation Review Root cause: TERM SCOPING PROBLEM Failure modes: GRANULARITY MISMATCH Sources checked: InterPro:IPR001206 SUPPORTS TRANSFER IPR001206 is present in the cached human entry and its broad kinase mapping is valid. Human MAG/DAG substrate assays, rather than this domain alone, justify the specified chemistry refinement. Proposed replacements: acylglycerol kinase activity ATP-dependent diacylglycerol kinase activity Supporting Evidence: PMID:15939762 that phosphorylates monoacylglycerol and diacylglycerol to form LPA and PA, respectively file:human/AGK/AGK-uniprot.txt Lipid kinase that can phosphorylate both monoacylglycerol and diacylglycerol to form lysophosphatidic acid (LPA) and phosphatidic acid (PA), respectively
GO:0047620 acylglycerol kinase activity	IEA GO_REF:0000120	ACCEPT	Summary: Automated acylglycerol kinase activity is supported by direct human evidence. Reason: This IEA agrees with the experimentally characterized AGK activity. GO:0047620 represents monoacylglycerol phosphorylation (EC 2.7.1.94); the distinct DAG reaction is represented by GO:0004143. Propagation Review Root cause: NO FAILURE CORE Sources checked: ARBA:ARBA00092986 UNRESOLVED The rule identifier is present in GOA; its current predicates were not retrieved. The activity judgment rests on independent human enzymology. UniProtKB:Q9ESW4 · Agk SUPPORTS TRANSFER Mouse Agk donor chemistry agrees with the independently demonstrated human MAG/DAG reaction. ensembl:ENSMUSP00000031977 UNRESOLVED Seeded Ensembl donor identifier; current model was not separately inspected. RHEA:19293 SUPPORTS TRANSFER The exact reaction is present in the cached human UniProt catalytic-activity section with experimental PMID:15939762 support. RHEA:33747 SUPPORTS TRANSFER The exact reaction is present in the cached human UniProt catalytic-activity section with experimental PMID:15939762 support. RHEA:39847 SUPPORTS TRANSFER The exact reaction is present in the cached human UniProt catalytic-activity section with experimental PMID:15939762 support. EC:2.7.1.94 SUPPORTS TRANSFER The EC reaction class matches the substrate-specific human enzyme assay and cached UniProt reaction. Supporting Evidence: PMID:15939762 that phosphorylates monoacylglycerol and diacylglycerol to form LPA and PA, respectively file:human/AGK/AGK-uniprot.txt Lipid kinase that can phosphorylate both monoacylglycerol and diacylglycerol to form lysophosphatidic acid (LPA) and phosphatidic acid (PA), respectively
GO:0005515 protein binding	IPI PMID:28514442 Architecture of the human interactome defines protein commun...	REMOVE	Summary: The HCCS interaction row is a generic protein-binding annotation from a proteome-scale interactome dataset. Reason: The reported interaction does not supply a separate molecular function through the generic protein binding label. Remove this uninformative term while retaining the interaction evidence; the lipid-kinase and TIM22 functions are described by informative annotations. The partner identity is traceable in GOA and UniProt, but the pair-level supplementary record was not independently recovered. No specific alternative AGK molecular function is established by the accessible screen description. Supporting Evidence: PMID:28514442 BioPlex 2.0 contains more than 29,000 previously unknown co-associations
GO:0005515 protein binding	IPI PMID:32814053 Interactome Mapping Provides a Network of Neurodegenerative ...	REMOVE	Summary: The HTT interaction row is a generic protein-binding annotation from a neurodegeneration interactome map. Reason: The reported interaction does not supply a separate molecular function through the generic protein binding label. Remove this uninformative term while retaining the interaction evidence; the lipid-kinase and TIM22 functions are described by informative annotations. The partner identity is traceable in GOA and UniProt, but the pair-level supplementary record was not independently recovered. No specific alternative AGK molecular function is established by the accessible screen description. Supporting Evidence: PMID:32814053 This network reveals interconnectivity across diseases and links many known ND-causing proteins
GO:0005515 protein binding	IPI PMID:33961781 Dual proteome-scale networks reveal cell-specific remodeling...	REMOVE	Summary: The HCCS interaction row is a generic protein-binding annotation from BioPlex-style AP-MS. Reason: The reported interaction does not supply a separate molecular function through the generic protein binding label. Remove this uninformative term while retaining the interaction evidence; the lipid-kinase and TIM22 functions are described by informative annotations. The partner identity is traceable in GOA and UniProt, but the pair-level supplementary record was not independently recovered. No specific alternative AGK molecular function is established by the accessible screen description. Supporting Evidence: PMID:33961781 Through affinity-purification mass spectrometry, we have created two proteome-scale, cell-line-specific interaction networks
GO:0005515 protein binding	IPI PMID:37009826 LINC00493-encoded microprotein SMIM26 exerts anti-metastatic...	REMOVE	Summary: The SMIM26 interaction is biologically interesting but generic protein binding is not an informative AGK molecular function. Reason: The source directly reports SMIM26–AGK association and SMIM26-dependent changes in AGK mitochondrial localization and AKT signaling. Indexed primary Figure 6 describes reciprocal co-immunoprecipitation and cellular localization experiments; these do not by themselves establish a new catalytic, adapter or transporter activity of AGK toward SMIM26. Remove the uninformative binding label without denying the interaction or transferring SMIM26 regulatory effects to AGK. Supporting Evidence: PMID:37009826 SMIM26, but not LINC00493, suppresses ccRCC growth and metastatic lung colonization by interacting with acylglycerol kinase (AGK)
GO:0001727 lipid kinase activity	IEA GO_REF:0000107	MODIFY	Summary: Human enzymology resolves the broad lipid-kinase transfer to specific MAG and DAG kinase reactions. Reason: The mouse-source transfer supports lipid kinase activity broadly. Direct human assays in PMID:15939762 resolve the substrates and products as MAG-to-LPA and DAG-to-PA, so refinement to GO:0047620 and GO:0004143 is evidence-based. The broad source does not itself identify these substrates; independent target experiments do. This chemistry is better resolved than the precise compartment of every membrane-associated pool, for which the broad membrane assertion is retained. The existing specific annotations already provide coverage, so no NEW row is proposed. Propagation Review Root cause: TERM SCOPING PROBLEM Failure modes: GRANULARITY MISMATCH Sources checked: UniProtKB:Q9ESW4 · Agk SUPPORTS TRANSFER Mouse Agk is a lipid kinase and the broad transfer is supported; direct human MAG/DAG assays justify the requested refinement. ensembl:ENSMUSP00000031977 UNRESOLVED The seeded Ensembl donor is recorded; its current model was not independently retrieved. Proposed replacements: acylglycerol kinase activity ATP-dependent diacylglycerol kinase activity Supporting Evidence: PMID:15939762 that phosphorylates monoacylglycerol and diacylglycerol to form LPA and PA, respectively file:human/AGK/AGK-uniprot.txt Lipid kinase that can phosphorylate both monoacylglycerol and diacylglycerol to form lysophosphatidic acid (LPA) and phosphatidic acid (PA), respectively
GO:0016020 membrane	IEA GO_REF:0000120	ACCEPT	Summary: Broad membrane association is supported at the source resolution. Reason: The combined electronic sources establish membrane association but do not uniquely assign the attachment site of every AGK pool. The inner-membrane/IMS TIM22 population is independently well supported and already annotated, while source-specific outer- and plasma-membrane assertions remain unresolved. Replacing this broad location with both an individual membrane and an aqueous subcompartment would exceed the resolution of the combined source. This differs from the two kinase refinements: direct human substrate assays identify exact MAG/DAG chemistry, whereas compartment identity is not uniquely resolved for every source. Retain the valid broad CC assertion and its independent membrane-association support. Propagation Review Root cause: NO FAILURE CORE Sources checked: ARBA:ARBA00028763 UNRESOLVED Rule predicates were not inspected; independent AGK localization supports the broad annotation. UniProtKB:Q9ESW4 · Agk SUPPORTS TRANSFER The mouse membrane-associated donor is compatible with direct human membrane fractionation. ensembl:ENSMUSP00000031977 UNRESOLVED Current Ensembl model was not separately inspected. Supporting Evidence: PMID:15939762 Confocal microscopy and subcellular fractionation suggest that AGK is localized to the mitochondria file:human/AGK/AGK-uniprot.txt Mitochondrion inner membrane file:human/AGK/AGK-uniprot.txt Localizes in the mitochondrion intermembrane space, where it associates with the inner membrane
GO:0046486 glycerolipid metabolic process	IEA GO_REF:0000041	ACCEPT	Summary: AGK directly catalyzes glycerolipid phosphorylation. Reason: MAG-to-LPA and DAG-to-PA conversion are steps of glycerolipid metabolism. This source-level pathway term is supported by AGK catalysis, independently of downstream growth or mitochondrial phenotypes. Propagation Review Root cause: NO FAILURE CORE Sources checked: UniPathway:UPA00230 SUPPORTS TRANSFER GOA names this pathway mapping and cached UniProt explicitly assigns glycerolipid metabolism using PMID:15939762. Supporting Evidence: PMID:15939762 that phosphorylates monoacylglycerol and diacylglycerol to form LPA and PA, respectively file:human/AGK/AGK-uniprot.txt Lipid kinase that can phosphorylate both monoacylglycerol and diacylglycerol to form lysophosphatidic acid (LPA) and phosphatidic acid (PA), respectively
GO:0005743 mitochondrial inner membrane	IDA PMID:32901109 Cryo-EM structure of the human mitochondrial translocase TIM...	ACCEPT	Summary: The TIM22 cryo-EM/ComplexPortal row supports mitochondrial inner-membrane localization. Reason: The primary structural article resolves AGK within the human TIM22 assembly and its membrane-anchoring region (Figure 1h and supplementary figures). This supports the inner-membrane complex location and agrees with independent TIM22 localization studies. The local publication cache is bibliographic only; primary indexed Results supplied the structural detail. Supporting Evidence: PMID:28712726 AGK functions in a kinase-independent manner to maintain the integrity of the TIM22 complex, where it facilitates the import and assembly of mitochondrial carrier proteins
GO:0042721 TIM22 mitochondrial import inner membrane insertion complex	IPI PMID:32901109 Cryo-EM structure of the human mitochondrial translocase TIM...	ACCEPT	Summary: AGK is a supported TIM22 complex subunit. Reason: The human TIM22 structure places AGK in contact with Tim29 and the small-Tim chaperone assembly. This physical membership corroborates the independent biochemical TIM22 studies; it does not imply that AGK alone forms the translocation pore. Supporting Evidence: PMID:28712726 AGK functions in a kinase-independent manner to maintain the integrity of the TIM22 complex, where it facilitates the import and assembly of mitochondrial carrier proteins
GO:0045039 protein insertion into mitochondrial inner membrane	IDA PMID:32901109 Cryo-EM structure of the human mitochondrial translocase TIM...	ACCEPT	Summary: AGK participates in protein insertion into the mitochondrial inner membrane through TIM22. Reason: AGK supplies a structural contribution to the TIM22 machinery that imports and inserts multipass inner-membrane proteins. Structural membership is complemented by the independent human knockout/rescue and carrier-import evidence in PMID:28712726; the role is not inferred simply from a generic protein-transport classification. Supporting Evidence: PMID:28712726 AGK functions in a kinase-independent manner to maintain the integrity of the TIM22 complex, where it facilitates the import and assembly of mitochondrial carrier proteins
GO:0005739 mitochondrion	IDA GO_REF:0000052	ACCEPT	Summary: HPA mitochondrial localization is consistent with AGK biology. Reason: Independent localization and TIM22 data support AGK as a mitochondrial protein. Supporting Evidence: PMID:15939762 Confocal microscopy and subcellular fractionation suggest that AGK is localized to the mitochondria file:human/AGK/AGK-uniprot.txt Mitochondrion inner membrane
GO:0005741 mitochondrial outer membrane	TAS Reactome:R-HSA-5696074	UNDECIDED	Summary: An IMS-proximity source does not independently establish outer-membrane attachment of AGK. Reason: The cached Reactome event attributes outer-membrane binding to Hung et al. (PMID:25002142). The cited APEX study maps IMS proximity, and PMID:28712726 interprets the AGK result as an IMS-exposed domain. Recovering the original AGK supplementary row would verify the identification but, by itself, would not distinguish an outer- from an inner-membrane anchor. Later protease-protection, membrane-association and TIM22 experiments independently support an inner-membrane pool; they do not exclude every other pool. The outer-membrane assertion therefore remains UNDECIDED pending topology-resolving evidence, rather than being supported by its own Reactome statement. Supporting Evidence: PMID:28712726 AGK functions in a kinase-independent manner to maintain the integrity of the TIM22 complex, where it facilitates the import and assembly of mitochondrial carrier proteins
GO:0001729 ceramide kinase activity	ISS GO_REF:0000024	REMOVE	Summary: Human AGK substrate assays detected phosphorylation of monoacylglycerols and diacylglycerols but not ceramide. Reason: Human AGK expressed in mammalian cells produced MAG- and DAG-phosphorylation products but no significant ceramide product in PMID:15939762, which explicitly discusses the positive bacterial mouse-MuLK result. PMID:16269826 also reports a negative MuLK assay, while cautioning that its substrate conclusions were not definitive. These data favor excluding ceramide kinase from the human assignment under the tested conditions; they do not prove a universal absence under every membrane or assay condition. The ISS donor is mouse Agk Q9ESW4; this is an ortholog transfer, distinct from the PAINT node and combined electronic mapping. Propagation Review Root cause: PROPAGATION BAD Failure modes: FUNCTIONAL DIVERGENCE Sources checked: UniProtKB:Q9ESW4 · Agk SUPPORTS SOURCE BUT NOT TARGET Positive mouse recombinant MuLK enzymology is genuine ortholog evidence. Negative human assays limit transfer of ceramide substrate specificity; they do not invalidate the mouse experiment. Supporting Evidence: PMID:15939762 Significant phosphorylated products were only detected with monoacylglycerols and diacylglycerols as substrates, but not with any other lipid tested, including ceramide and sphingosine PMID:16269826 No evidence for phosphorylation of ceramide by the recently described multiple lipid kinase was found
GO:0005739 mitochondrion	HTP PMID:34800366 Quantitative high-confidence human mitochondrial proteome an...	ACCEPT	Summary: High-throughput mitochondrial proteome evidence is consistent with AGK localization. Reason: This HTP localization agrees with direct AGK and TIM22 complex evidence. Supporting Evidence: PMID:34800366 We defined a human mitochondrial high-confidence proteome
GO:0005829 cytosol	TAS Reactome:R-HSA-6802927	REMOVE	Summary: The Reactome cytosolic entity in these events is an oncogenic AGK-BRAF fusion. Reason: The Reactome entity R-HSA-6802510 is explicitly AGK(1-33)InsW-p-BRAF(328-766) fusion [cytosol] and participates in these BRAF/RAF fusion events. Its link to the AGK reference sequence does not make the fusion localization evidence for intact AGK. This is a verified entity-context mismatch, not an inference from an event title or a claim that native AGK can never enter cytosol. See https://reactome.org/content/detail/R-HSA-6802510 for the modified entity and participant hierarchy. The source event concerns braf and raf fusion mutant dimers are phosphorylated. Supporting Evidence: Reactome:R-HSA-6802927 Fusion mutants of BRAF and RAF1 are believed to form constitutive dimers and activate downstream signaling independent of RAS and external stimuli
GO:0005829 cytosol	TAS Reactome:R-HSA-6802932	REMOVE	Summary: The Reactome cytosolic entity in these events is an oncogenic AGK-BRAF fusion. Reason: The Reactome entity R-HSA-6802510 is explicitly AGK(1-33)InsW-p-BRAF(328-766) fusion [cytosol] and participates in these BRAF/RAF fusion events. Its link to the AGK reference sequence does not make the fusion localization evidence for intact AGK. This is a verified entity-context mismatch, not an inference from an event title or a claim that native AGK can never enter cytosol. See https://reactome.org/content/detail/R-HSA-6802510 for the modified entity and participant hierarchy. The source event concerns dissociation of braf/raf fusion complex. Supporting Evidence: Reactome:R-HSA-6802932 After phosphorylation by MAP2Ks, the scaffolded kinase complex assembled by BRAF and RAF fusion dimers presumably dissociates, as is the case for WT complexes
GO:0005829 cytosol	TAS Reactome:R-HSA-6802933	REMOVE	Summary: The Reactome cytosolic entity in these events is an oncogenic AGK-BRAF fusion. Reason: The Reactome entity R-HSA-6802510 is explicitly AGK(1-33)InsW-p-BRAF(328-766) fusion [cytosol] and participates in these BRAF/RAF fusion events. Its link to the AGK reference sequence does not make the fusion localization evidence for intact AGK. This is a verified entity-context mismatch, not an inference from an event title or a claim that native AGK can never enter cytosol. See https://reactome.org/content/detail/R-HSA-6802510 for the modified entity and participant hierarchy. The source event concerns p-braf and raf fusion dimers phosphorylate map2ks. Supporting Evidence: Reactome:R-HSA-6802933 BRAF and RAF fusion dimers constitutively phosphorylate MAP2Ks
GO:0005829 cytosol	TAS Reactome:R-HSA-6802934	REMOVE	Summary: The Reactome cytosolic entity in these events is an oncogenic AGK-BRAF fusion. Reason: The Reactome entity R-HSA-6802510 is explicitly AGK(1-33)InsW-p-BRAF(328-766) fusion [cytosol] and participates in these BRAF/RAF fusion events. Its link to the AGK reference sequence does not make the fusion localization evidence for intact AGK. This is a verified entity-context mismatch, not an inference from an event title or a claim that native AGK can never enter cytosol. See https://reactome.org/content/detail/R-HSA-6802510 for the modified entity and participant hierarchy. The source event concerns p-braf and raf fusion dimers bind map2ks and mapks. Supporting Evidence: Reactome:R-HSA-6802934 BRAF and RAF fusion proteins expressed in cancer constitutively activate downstream signaling and promote cellular transformation
GO:0005829 cytosol	TAS Reactome:R-HSA-6802935	REMOVE	Summary: The Reactome cytosolic entity in these events is an oncogenic AGK-BRAF fusion. Reason: The Reactome entity R-HSA-6802510 is explicitly AGK(1-33)InsW-p-BRAF(328-766) fusion [cytosol] and participates in these BRAF/RAF fusion events. Its link to the AGK reference sequence does not make the fusion localization evidence for intact AGK. This is a verified entity-context mismatch, not an inference from an event title or a claim that native AGK can never enter cytosol. See https://reactome.org/content/detail/R-HSA-6802510 for the modified entity and participant hierarchy. The source event concerns mapks are phosphorylated downstream of braf and raf fusion dimers. Supporting Evidence: Reactome:R-HSA-6802935 MAPKs are phosphorylated downstream of constitutively active BRAF and RAF fusion proteins
GO:0004143 ATP-dependent diacylglycerol kinase activity	IDA PMID:15939762 A novel acylglycerol kinase that produces lysophosphatidic a...	ACCEPT	Summary: Direct human evidence supports ATP-dependent diacylglycerol kinase activity. Reason: PMID:15939762 directly characterized AGK phosphorylation of diacylglycerol to phosphatidic acid. Supporting Evidence: PMID:15939762 that phosphorylates monoacylglycerol and diacylglycerol to form LPA and PA, respectively file:human/AGK/AGK-uniprot.txt Lipid kinase that can phosphorylate both monoacylglycerol and diacylglycerol to form lysophosphatidic acid (LPA) and phosphatidic acid (PA), respectively
GO:0031966 mitochondrial membrane	IDA PMID:15939762 A novel acylglycerol kinase that produces lysophosphatidic a...	ACCEPT	Summary: Direct human evidence supports mitochondrial membrane localization. Reason: PMID:15939762 localized AGK to mitochondria and subcellular fractions enriched for mitochondrial AGK activity. Supporting Evidence: PMID:15939762 Confocal microscopy and subcellular fractionation suggest that AGK is localized to the mitochondria file:human/AGK/AGK-uniprot.txt Mitochondrion inner membrane
GO:0047620 acylglycerol kinase activity	IDA PMID:15939762 A novel acylglycerol kinase that produces lysophosphatidic a...	ACCEPT	Summary: Direct human evidence supports acylglycerol kinase activity. Reason: PMID:15939762 directly characterized AGK phosphorylation of monoacylglycerol to LPA. Supporting Evidence: PMID:15939762 that phosphorylates monoacylglycerol and diacylglycerol to form LPA and PA, respectively file:human/AGK/AGK-uniprot.txt Lipid kinase that can phosphorylate both monoacylglycerol and diacylglycerol to form lysophosphatidic acid (LPA) and phosphatidic acid (PA), respectively
GO:0031966 mitochondrial membrane	IDA PMID:16269826 Further characterization of mammalian ceramide kinase: subst...	ACCEPT	Summary: The original mitochondrial-membrane annotation is retained with explicit organelle-level limits on the accessible abstract. Reason: The PMID:16269826 abstract explicitly reports mitochondrial localization of MuLK, but that sentence alone resolves neither membrane versus organelle interior nor inner versus outer membrane. Indexed publisher discussion distinguishes the human MuLK work from mouse recombinant studies. Retain the experimentally curated membrane assignment with deference to the full-paper curator and independent human membrane/topology evidence in PMID:15939762 and PMID:28712726. The supporting abstract quotation is organelle-level corroboration, not the entire basis for membrane localization. Supporting Evidence: PMID:16269826 The latter kinase is localized in the mitochondria
GO:0005743 mitochondrial inner membrane	IDA PMID:28712724 Acylglycerol Kinase Mutated in Sengers Syndrome Is a Subunit...	ACCEPT	Summary: PMID:28712724 supports AGK mitochondrial inner-membrane localization. Reason: This paper identifies AGK as a TIM22 complex constituent at the mitochondrial inner membrane. Supporting Evidence: PMID:28712724 we have identified AGK as a constituent of the TIM22 complex in the mitochondrial inner membrane PMID:28712724 AGK assembles with TIMM22 and TIMM29 and supports the import of a subset of multi-spanning membrane proteins
GO:0005743 mitochondrial inner membrane	IDA PMID:28712726 Sengers Syndrome-Associated Mitochondrial Acylglycerol Kinas...	ACCEPT	Summary: PMID:28712726 supports AGK mitochondrial inner-membrane localization. Reason: The primary author manuscript reports mitochondrial subfractionation, protease protection and membrane extraction of tagged AGK, supporting an inner-membrane protein with its C terminus exposed to the IMS. The N-terminal anchor has moderate hydrophobicity; the finding is compatible with the later TIM22 structure. Supporting Evidence: PMID:28712726 we identified AGK as a subunit of the mitochondrial TIM22 protein import complex PMID:28712726 AGK functions in a kinase-independent manner to maintain the integrity of the TIM22 complex, where it facilitates the import and assembly of mitochondrial carrier proteins
GO:0005758 mitochondrial intermembrane space	IDA PMID:28712724 Acylglycerol Kinase Mutated in Sengers Syndrome Is a Subunit...	ACCEPT	Summary: PMID:28712724 supports AGK intermembrane-space/inner-membrane association. Reason: The TIM22 study places AGK at the mitochondrial inner membrane/intermembrane-space side of the complex. Supporting Evidence: PMID:28712724 we have identified AGK as a constituent of the TIM22 complex in the mitochondrial inner membrane PMID:28712724 AGK assembles with TIMM22 and TIMM29 and supports the import of a subset of multi-spanning membrane proteins file:human/AGK/AGK-uniprot.txt Mitochondrion intermembrane space file:human/AGK/AGK-uniprot.txt Localizes in the mitochondrion intermembrane space, where it associates with the inner membrane
GO:0005758 mitochondrial intermembrane space	IDA PMID:28712726 Sengers Syndrome-Associated Mitochondrial Acylglycerol Kinas...	ACCEPT	Summary: PMID:28712726 supports AGK intermembrane-space/inner-membrane association. Reason: The primary author manuscript Figure 1B reports protease accessibility of the C-terminal tag after outer-membrane disruption, directly supporting IMS exposure. This is the IMS-facing portion of membrane-associated AGK, not an assertion that the entire protein is freely soluble. Supporting Evidence: PMID:28712726 we identified AGK as a subunit of the mitochondrial TIM22 protein import complex PMID:28712726 AGK functions in a kinase-independent manner to maintain the integrity of the TIM22 complex, where it facilitates the import and assembly of mitochondrial carrier proteins file:human/AGK/AGK-uniprot.txt Mitochondrion intermembrane space file:human/AGK/AGK-uniprot.txt Localizes in the mitochondrion intermembrane space, where it associates with the inner membrane
GO:0042721 TIM22 mitochondrial import inner membrane insertion complex	IDA PMID:28712724 Acylglycerol Kinase Mutated in Sengers Syndrome Is a Subunit...	ACCEPT	Summary: PMID:28712724 directly supports AGK as part of the TIM22 complex. Reason: AGK assembles with TIMM22 and TIMM29 and supports import of multi-spanning membrane proteins. Supporting Evidence: PMID:28712724 we have identified AGK as a constituent of the TIM22 complex in the mitochondrial inner membrane PMID:28712724 AGK assembles with TIMM22 and TIMM29 and supports the import of a subset of multi-spanning membrane proteins
GO:0042721 TIM22 mitochondrial import inner membrane insertion complex	IDA PMID:28712726 Sengers Syndrome-Associated Mitochondrial Acylglycerol Kinas...	ACCEPT	Summary: PMID:28712726 directly supports AGK as part of the human TIM22 complex. Reason: AGK maintains TIM22 complex integrity and facilitates carrier import and assembly. Supporting Evidence: PMID:28712726 we identified AGK as a subunit of the mitochondrial TIM22 protein import complex PMID:28712726 AGK functions in a kinase-independent manner to maintain the integrity of the TIM22 complex, where it facilitates the import and assembly of mitochondrial carrier proteins
GO:0045039 protein insertion into mitochondrial inner membrane	IDA PMID:28712724 Acylglycerol Kinase Mutated in Sengers Syndrome Is a Subunit...	ACCEPT	Summary: PMID:28712724 supports AGK involvement in protein insertion into the mitochondrial inner membrane. Reason: The paper ties AGK-containing TIM22 to import of multi-spanning inner-membrane proteins, matching this process annotation. Supporting Evidence: PMID:28712724 we have identified AGK as a constituent of the TIM22 complex in the mitochondrial inner membrane PMID:28712724 AGK assembles with TIMM22 and TIMM29 and supports the import of a subset of multi-spanning membrane proteins
GO:0045039 protein insertion into mitochondrial inner membrane	IDA PMID:28712726 Sengers Syndrome-Associated Mitochondrial Acylglycerol Kinas...	ACCEPT	Summary: PMID:28712726 supports AGK involvement in protein insertion into the mitochondrial inner membrane. Reason: The primary study combines AGK knockout, wild-type and kinase-dead rescue with carrier import/assembly assays. Carrier biogenesis is impaired and rescued independently of kinase activity, whereas effects vary among TIM22 substrates; the annotation does not imply AGK is required equally for every inner-membrane protein. Supporting Evidence: PMID:28712726 we identified AGK as a subunit of the mitochondrial TIM22 protein import complex PMID:28712726 AGK functions in a kinase-independent manner to maintain the integrity of the TIM22 complex, where it facilitates the import and assembly of mitochondrial carrier proteins

Gene Description

Existing Annotations Review

Core Functions

Phosphorylates monoacylglycerols with ATP to produce lysophosphatidic acid in mitochondrial glycerolipid metabolism.

Phosphorylates diacylglycerol with ATP to produce phosphatidic acid in mitochondrial glycerolipid metabolism.

Supports TIM22 complex integrity and the import and assembly of multipass mitochondrial carrier proteins, independently of AGK lipid-kinase activity.

References

Suggested Questions for Experts

Suggested Experiments

Tags

Deep Research

Falcon

OpenScientist

📚 Additional Documentation

Notes

Pn Notes

📄 View Raw YAML