Angiotensinogen is the liver-secreted plasma glycoprotein that is the sole source of the angiotensin peptides, and through them the origin of the renin-angiotensin system that sets blood pressure, blood volume and electrolyte balance. It is synthesised with a cleaved 24-residue signal peptide, carries N-glycans at four sites, and circulates in plasma at high, physiologically rate-influencing concentration. Structurally it belongs to the serpin superfamily and retains the serpin fold despite only about 22% identity to its closest serpin relatives, but it is a non-inhibitory serpin: it has no functional reactive-centre bond, its reactive centre loop does not undergo the stressed-to-relaxed transition by which inhibitory serpins trap proteases, and it inhibits no protease. The fold instead serves as a carrier and delivery device for the hormone held in its N-terminal tail. The rate-limiting step of the whole cascade is cleavage of that tail by renin, an aspartyl protease. Angiotensinogen is not a passive substrate in this reaction. The scissile Leu-Val bond is buried in an ordered 63-residue N-terminal superstructure and becomes accessible only through a large conformational change: renin binds through a tail-into-mouth allosteric mechanism that threads the N terminus into a pocket equivalent to the hormone-binding site of the carrier serpins, unwinds helix H, and forms an extensive body-to-body interface with the protease outside its catalytic cleft. Cleavage specificity and rate are set by angiotensinogen residues and glycans in that interface. A labile disulfide bridge between Cys42 and Cys162 (Cys18-Cys138 in mature numbering), the only two cysteines conserved across species, links the tail to the body of the molecule and gives plasma angiotensinogen a mixture of reduced and oxidised forms; the oxidised form is more efficiently cleaved by receptor-bound renin, although removing the bridge in mice does not measurably change angiotensin II production or blood pressure. Renin cleavage releases angiotensin I, which ACE converts to the vasoconstrictor octapeptide angiotensin II, and further processing by ACE2, aminopeptidases, neprilysin and other peptidases generates angiotensin III, angiotensin IV, angiotensin-(1-9) and angiotensin-(1-7). Angiotensin II acts on the AGTR1 and AGTR2 G-protein-coupled receptors: through AGTR1 it constricts arterioles, drives aldosterone secretion from the adrenal zona glomerulosa, promotes renal sodium and water retention, and stimulates cardiac, vascular and fibroblast growth and matrix deposition, while through AGTR2 it engages SHP-1, inhibits ERK and can promote apoptosis. Angiotensin-(1-7), acting on MAS1, opposes several of these effects. Loss of angiotensinogen is profound and informative: knockout mice have no circulating angiotensin I and are markedly hypotensive, and biallelic human loss-of-function mutations cause autosomal recessive renal tubular dysgenesis, with absent proximal tubules, fetal anuria and perinatal death, attributed to chronically low fetal renal perfusion pressure. Common coding variants, notably M235T, associate with essential hypertension and pre-eclampsia and appear to act by raising circulating angiotensinogen concentration rather than by altering the protein's function.
| GO Term | Evidence | Action | Reason |
|---|---|---|---|
| GO:0005576 extracellular region | IBA GO_REF:0000033 | ACCEPT | Summary: Correct family inference. PANTHER node PTN000156123 is the secreted-serpin node of PTHR11461, and its 51 seeds span the whole functional range of the family - inhibitory members (SERPINA1, SERPINC1, SERPINE1, SERPINF2) alongside non-inhibitory ones (SERPINF1/PEDF, SERPINA6/corticosteroid-binding globulin, SERPINA7/thyroxine-binding globulin, chicken ovalbumin-related protein Y). What those proteins actually share is secretion, not inhibition, so the node is both correctly placed and correctly broad. AGT is a liver-secreted plasma protein with a cleaved signal peptide (residues 1-24) and no transmembrane segment, so it sits squarely inside the clade that inherited this property. Reason: Extracellular localisation is a core, directly evidenced property of AGT and the phylogenetic inference is sound. That the same family node carries both inhibitory and non-inhibitory serpins is exactly why this term transfers where GO:0004867 does not. Propagation Review Root cause: NO FAILURE CORE Sources checked: AGI_LocusCode:AT1G47710 · Serpin-ZX (Arabidopsis thaliana) SUPPORTS TRANSFER Serpin-ZX (Arabidopsis thaliana), a secreted serpin; extracellular localisation is correct for it and transfers correctly to AGT, which is likewise secreted. FB:FBgn0002930 · nec (Drosophila melanogaster) SUPPORTS TRANSFER nec (Drosophila melanogaster), a secreted serpin; extracellular localisation is correct for it and transfers correctly to AGT, which is likewise secreted. FB:FBgn0015586 · Acp76A (Drosophila melanogaster) SUPPORTS TRANSFER Acp76A (Drosophila melanogaster), a secreted serpin; extracellular localisation is correct for it and transfers correctly to AGT, which is likewise secreted. FB:FBgn0028984 · Spn88Ea (Drosophila melanogaster) SUPPORTS TRANSFER Spn88Ea (Drosophila melanogaster), a secreted serpin; extracellular localisation is correct for it and transfers correctly to AGT, which is likewise secreted. FB:FBgn0028986 · Spn38F (Drosophila melanogaster) SUPPORTS TRANSFER Spn38F (Drosophila melanogaster), a secreted serpin; extracellular localisation is correct for it and transfers correctly to AGT, which is likewise secreted. FB:FBgn0028990 · Spn27A (Drosophila melanogaster) SUPPORTS TRANSFER Spn27A (Drosophila melanogaster), a secreted serpin; extracellular localisation is correct for it and transfers correctly to AGT, which is likewise secreted. FB:FBgn0031973 · Spn28Dc (Drosophila melanogaster) SUPPORTS TRANSFER Spn28Dc (Drosophila melanogaster), a secreted serpin; extracellular localisation is correct for it and transfers correctly to AGT, which is likewise secreted. FB:FBgn0262057 · Spn77Ba (Drosophila melanogaster) SUPPORTS TRANSFER Spn77Ba (Drosophila melanogaster), a secreted serpin; extracellular localisation is correct for it and transfers correctly to AGT, which is likewise secreted. FB:FBgn0265137 · Spn42Da (Drosophila melanogaster) SUPPORTS TRANSFER Spn42Da (Drosophila melanogaster), a secreted serpin; extracellular localisation is correct for it and transfers correctly to AGT, which is likewise secreted. MGI:MGI:101780 · Serpine2 (Mus musculus) SUPPORTS TRANSFER Serpine2 (Mus musculus), a secreted serpin; extracellular localisation is correct for it and transfers correctly to AGT, which is likewise secreted. MGI:MGI:105045 · Serpina3n (Mus musculus) SUPPORTS TRANSFER Serpina3n (Mus musculus), a secreted serpin; extracellular localisation is correct for it and transfers correctly to AGT, which is likewise secreted. MGI:MGI:108080 · Serpinf1 (Mus musculus) SUPPORTS TRANSFER Serpinf1 (Mus musculus), a secreted serpin; extracellular localisation is correct for it and transfers correctly to AGT, which is likewise secreted. MGI:MGI:1913472 · Serpinb1a (Mus musculus) SUPPORTS TRANSFER Serpinb1a (Mus musculus), a secreted serpin; extracellular localisation is correct for it and transfers correctly to AGT, which is likewise secreted. MGI:MGI:1915304 · Serpina12 (Mus musculus) SUPPORTS TRANSFER Serpina12 (Mus musculus), a secreted serpin; extracellular localisation is correct for it and transfers correctly to AGT, which is likewise secreted. MGI:MGI:87963 · Agt (Mus musculus) SUPPORTS TRANSFER Agt (Mus musculus), a secreted serpin; extracellular localisation is correct for it and transfers correctly to AGT, which is likewise secreted. MGI:MGI:88278 · Serpina6 (Mus musculus) SUPPORTS TRANSFER Serpina6 (Mus musculus), a secreted serpin; extracellular localisation is correct for it and transfers correctly to AGT, which is likewise secreted. MGI:MGI:891968 · Serpina1d (Mus musculus) SUPPORTS TRANSFER Serpina1d (Mus musculus), a secreted serpin; extracellular localisation is correct for it and transfers correctly to AGT, which is likewise secreted. MGI:MGI:891969 · Serpina1c (Mus musculus) SUPPORTS TRANSFER Serpina1c (Mus musculus), a secreted serpin; extracellular localisation is correct for it and transfers correctly to AGT, which is likewise secreted. MGI:MGI:891970 · Serpina1b (Mus musculus) SUPPORTS TRANSFER Serpina1b (Mus musculus), a secreted serpin; extracellular localisation is correct for it and transfers correctly to AGT, which is likewise secreted. MGI:MGI:891971 · Serpina1a (Mus musculus) SUPPORTS TRANSFER Serpina1a (Mus musculus), a secreted serpin; extracellular localisation is correct for it and transfers correctly to AGT, which is likewise secreted. MGI:MGI:894696 · Serping1 (Mus musculus) SUPPORTS TRANSFER Serping1 (Mus musculus), a secreted serpin; extracellular localisation is correct for it and transfers correctly to AGT, which is likewise secreted. PANTHER:PTN000156123 · PAINT ancestral node SUPPORTS TRANSFER Secreted-serpin IBD node of PTHR11461, seeded by 51 extant serpins across plants, insects and vertebrates. Its seeds include non-inhibitory serpins, so the node captures secretion rather than inhibition and transfers correctly to AGT. RGD:1306692 · Serpinf2 (Rattus norvegicus) SUPPORTS TRANSFER Serpinf2 (Rattus norvegicus), a secreted serpin; extracellular localisation is correct for it and transfers correctly to AGT, which is likewise secreted. RGD:1307404 · Serpinc1 (Rattus norvegicus) SUPPORTS TRANSFER Serpinc1 (Rattus norvegicus), a secreted serpin; extracellular localisation is correct for it and transfers correctly to AGT, which is likewise secreted. RGD:2069 · Agt (Rattus norvegicus) SUPPORTS TRANSFER Agt (Rattus norvegicus), a secreted serpin; extracellular localisation is correct for it and transfers correctly to AGT, which is likewise secreted. RGD:3249 · Serpine1 (Rattus norvegicus) SUPPORTS TRANSFER Serpine1 (Rattus norvegicus), a secreted serpin; extracellular localisation is correct for it and transfers correctly to AGT, which is likewise secreted. RGD:3326 · Serpina1 (Rattus norvegicus) SUPPORTS TRANSFER Serpina1 (Rattus norvegicus), a secreted serpin; extracellular localisation is correct for it and transfers correctly to AGT, which is likewise secreted. RGD:3747 · Serpina3n (Rattus norvegicus) SUPPORTS TRANSFER Serpina3n (Rattus norvegicus), a secreted serpin; extracellular localisation is correct for it and transfers correctly to AGT, which is likewise secreted. RGD:3748 · Serpine2 (Rattus norvegicus) SUPPORTS TRANSFER Serpine2 (Rattus norvegicus), a secreted serpin; extracellular localisation is correct for it and transfers correctly to AGT, which is likewise secreted. RGD:619833 · Serpina7 (Rattus norvegicus) SUPPORTS TRANSFER Serpina7 (Rattus norvegicus), a secreted serpin; extracellular localisation is correct for it and transfers correctly to AGT, which is likewise secreted. RGD:619896 · Serpini1 (Rattus norvegicus) SUPPORTS TRANSFER Serpini1 (Rattus norvegicus), a secreted serpin; extracellular localisation is correct for it and transfers correctly to AGT, which is likewise secreted. RGD:631369 · Serpinf1 (Rattus norvegicus) SUPPORTS TRANSFER Serpinf1 (Rattus norvegicus), a secreted serpin; extracellular localisation is correct for it and transfers correctly to AGT, which is likewise secreted. RGD:735225 · Serping1 (Rattus norvegicus) SUPPORTS TRANSFER Serping1 (Rattus norvegicus), a secreted serpin; extracellular localisation is correct for it and transfers correctly to AGT, which is likewise secreted. UniProtKB:P01008 · SERPINC1 (Homo sapiens) SUPPORTS TRANSFER SERPINC1 (Homo sapiens), a secreted serpin; extracellular localisation is correct for it and transfers correctly to AGT, which is likewise secreted. UniProtKB:P01009 · SERPINA1 (Homo sapiens) SUPPORTS TRANSFER SERPINA1 (Homo sapiens), a secreted serpin; extracellular localisation is correct for it and transfers correctly to AGT, which is likewise secreted. UniProtKB:P01011 · SERPINA3 (Homo sapiens) SUPPORTS TRANSFER SERPINA3 (Homo sapiens), a secreted serpin; extracellular localisation is correct for it and transfers correctly to AGT, which is likewise secreted. UniProtKB:P01014 · SERPINB14B (Gallus gallus) SUPPORTS TRANSFER SERPINB14B (Gallus gallus), a secreted serpin; extracellular localisation is correct for it and transfers correctly to AGT, which is likewise secreted. UniProtKB:P01019 · AGT (Homo sapiens) SUPPORTS TRANSFER AGT itself, the target. Its own IDA (PMID:4300938) and HDA plasma-proteomics evidence are among the descendant evidences the PAINT curator used to place this IBD, which is expected and is not circular. UniProtKB:P05121 · SERPINE1 (Homo sapiens) SUPPORTS TRANSFER SERPINE1 (Homo sapiens), a secreted serpin; extracellular localisation is correct for it and transfers correctly to AGT, which is likewise secreted. UniProtKB:P05154 · SERPINA5 (Homo sapiens) SUPPORTS TRANSFER SERPINA5 (Homo sapiens), a secreted serpin; extracellular localisation is correct for it and transfers correctly to AGT, which is likewise secreted. UniProtKB:P05155 · SERPING1 (Homo sapiens) SUPPORTS TRANSFER SERPING1 (Homo sapiens), a secreted serpin; extracellular localisation is correct for it and transfers correctly to AGT, which is likewise secreted. UniProtKB:P07093 · SERPINE2 (Homo sapiens) SUPPORTS TRANSFER SERPINE2 (Homo sapiens), a secreted serpin; extracellular localisation is correct for it and transfers correctly to AGT, which is likewise secreted. UniProtKB:P08697 · SERPINF2 (Homo sapiens) SUPPORTS TRANSFER SERPINF2 (Homo sapiens), a secreted serpin; extracellular localisation is correct for it and transfers correctly to AGT, which is likewise secreted. UniProtKB:P29508 · SERPINB3 (Homo sapiens) SUPPORTS TRANSFER SERPINB3 (Homo sapiens), a secreted serpin; extracellular localisation is correct for it and transfers correctly to AGT, which is likewise secreted. UniProtKB:P30740 · SERPINB1 (Homo sapiens) SUPPORTS TRANSFER SERPINB1 (Homo sapiens), a secreted serpin; extracellular localisation is correct for it and transfers correctly to AGT, which is likewise secreted. UniProtKB:P36955 · SERPINF1 (Homo sapiens) SUPPORTS TRANSFER SERPINF1 (Homo sapiens), a secreted serpin; extracellular localisation is correct for it and transfers correctly to AGT, which is likewise secreted. UniProtKB:P50453 · SERPINB9 (Homo sapiens) SUPPORTS TRANSFER SERPINB9 (Homo sapiens), a secreted serpin; extracellular localisation is correct for it and transfers correctly to AGT, which is likewise secreted. UniProtKB:Q7QIJ8 · SRPN2 (Anopheles gambiae) SUPPORTS TRANSFER SRPN2 (Anopheles gambiae), a secreted serpin; extracellular localisation is correct for it and transfers correctly to AGT, which is likewise secreted. UniProtKB:Q90935 · SERPINI1 (Gallus gallus) SUPPORTS TRANSFER SERPINI1 (Gallus gallus), a secreted serpin; extracellular localisation is correct for it and transfers correctly to AGT, which is likewise secreted. UniProtKB:Q95121 · SERPINF1 (Bos taurus) SUPPORTS TRANSFER SERPINF1 (Bos taurus), a secreted serpin; extracellular localisation is correct for it and transfers correctly to AGT, which is likewise secreted. UniProtKB:Q99574 · SERPINI1 (Homo sapiens) SUPPORTS TRANSFER SERPINI1 (Homo sapiens), a secreted serpin; extracellular localisation is correct for it and transfers correctly to AGT, which is likewise secreted. UniProtKB:Q9TTE1 · SERPINA3-1 (Bos taurus) SUPPORTS TRANSFER SERPINA3-1 (Bos taurus), a secreted serpin; extracellular localisation is correct for it and transfers correctly to AGT, which is likewise secreted. Supporting Evidence: file:human/AGT/AGT-uniprot.txt SUBCELLULAR LOCATION: Secreted file:human/AGT/AGT-uniprot.txt TISSUE SPECIFICITY: Expressed by the liver and secreted in plasma. |
| GO:0038166 angiotensin-activated signaling pathway | IBA GO_REF:0000033 | ACCEPT | Summary: Well-placed, angiotensinogen-specific inference. Node PTN008518321 is not a family-wide node: it is seeded by mouse Agt (MGI:MGI:87963 = P11859) and human AGT itself, at taxon 117571 (Euteleostomi). The term describes the pathway initiated when angiotensin II binds AGTR1/AGTR2, and angiotensin II exists only as a cleavage product of angiotensinogen, so no other gene could seed it. Reason: This is the core biological process of the gene and the node is tightly and correctly drawn around the angiotensinogens. The three-token WITH/FROM is short because the clade is small, not because the evidence is thin. Propagation Review Root cause: NO FAILURE CORE Sources checked: MGI:MGI:87963 · Agt (Mus musculus) SUPPORTS TRANSFER Mouse Agt (P11859), the 1:1 orthologue. Its own knockout abolishes plasma angiotensin I (PMID:7989296), which is the strongest possible descendant evidence that this pathway depends on angiotensinogen. PANTHER:PTN008518321 · PAINT ancestral node SUPPORTS TRANSFER Angiotensinogen-specific IBD node (taxon:117571), seeded by mouse and human angiotensinogen. Confirmed in the committed PAINT slice for PTHR11461, where it also carries GO:0042981. UniProtKB:P01019 · AGT (Homo sapiens) SUPPORTS TRANSFER Human AGT, the target's own accession. Its five direct GO:0038166 annotations are among the descendant evidences behind the IBD; the target appearing in its own WITH/FROM is expected and marks experimental grounding on the target, not circularity. Supporting Evidence: PMID:1567413 Ligand binding studies of the cloned receptor expressed in COS cells suggested that it is pharmacologically a type 1 angiotensin II receptor subtype. PMID:7989296 These mice do not produce angiotensinogen in the liver, resulting in the complete loss of plasma immunoreactive angiotensin I. |
| GO:0004867 serine-type endopeptidase inhibitor activity | IBA GO_REF:0000033 | REMOVE | Summary: Over-propagated family inference, and the clearest defect in this gene's record. GO:0004867 is defined as 'Binds to and stops, prevents or reduces the activity of a serine-type endopeptidase'. Angiotensinogen does none of that. It is the SUBSTRATE of renin, an ASPARTYL protease - wrong catalytic class and wrong role. The serpin inhibitory mechanism requires cleavage at the reactive centre loop followed by a stressed-to-relaxed transition that traps the protease, and AGT has been shown crystallographically to have lost that transition. Sequence analysis agrees: AGT carries no UniProt 'Reactive bond' site (10 of 12 inhibitors in the panel do), its P17-P9 hinge is ADEREPTES against SERPINA1's EKGTEAAGA with a helix-breaking Pro430 at the P12 position that must be small for beta-sheet A insertion, and MEROPS classes it I04.953, in the range reserved for non-inhibitor homologues. Every one of the 7 resolvable seed proteins of node PTN008970140 is a MEROPS inhibitor. Reason: A family-level inference contradicted by the target's own structure and sequence. The IBD node PTN008970140 is the SERPINA-clade node and every seed that donated the term is a genuine inhibitory serpin; AGT is the one non-inhibitory member of that clade and inherits the activity mechanically. No seed annotation is wrong, so this is a propagation failure, not a source failure, and removing it overrules no experimental annotation. Note that PAINT already uses IRD in this same family (node PTN002606963 blocks GO:0005576 for the intracellular subclade); an IRD at the angiotensinogen node would fix every inherited copy at once. Propagation Review Root cause: PROPAGATION BAD Failure modes: FUNCTIONAL DIVERGENCE PSEUDO OR SUBACTIVITY LOSS Sources checked: MGI:MGI:891970 · Serpina1b (Mus musculus) SUPPORTS SOURCE BUT NOT TARGET Serpina1b (Mus musculus): MEROPS I04.001. A genuine inhibitory serpin with an intact reactive centre, so the term is right for this seed. It does not transfer to AGT, which is MEROPS I04.953 - the range reserved for non-inhibitor homologues - and which has no annotated reactive bond and a proline at the P12 hinge position. PANTHER:PTN008970140 · PAINT ancestral node SOURCE BAD SERPINA-clade IBD node (taxon:7711) seeded exclusively by inhibitory serpins. The node assertion is correct for those seeds but its placement covers angiotensinogen, the one clade member that lost the inhibitory mechanism, so as an assertion about the clade as a whole it is wrong. RGD:1307404 · Serpinc1 (Rattus norvegicus) SUPPORTS SOURCE BUT NOT TARGET Serpinc1 (Rattus norvegicus): no MEROPS cross-reference on its own UniProt entry, though its human orthologue is MEROPS-classified as an inhibitor. A genuine inhibitory serpin with an intact reactive centre, so the term is right for this seed. It does not transfer to AGT, which is MEROPS I04.953 - the range reserved for non-inhibitor homologues - and which has no annotated reactive bond and a proline at the P12 hinge position. RGD:3326 · Serpina1 (Rattus norvegicus) SUPPORTS SOURCE BUT NOT TARGET Serpina1 (Rattus norvegicus): MEROPS I04.001. A genuine inhibitory serpin with an intact reactive centre, so the term is right for this seed. It does not transfer to AGT, which is MEROPS I04.953 - the range reserved for non-inhibitor homologues - and which has no annotated reactive bond and a proline at the P12 hinge position. RGD:619817 · Serpina5 (Rattus norvegicus) SUPPORTS SOURCE BUT NOT TARGET Serpina5 (Rattus norvegicus): no MEROPS cross-reference on its own UniProt entry, though its human orthologue is MEROPS-classified as an inhibitor. A genuine inhibitory serpin with an intact reactive centre, so the term is right for this seed. It does not transfer to AGT, which is MEROPS I04.953 - the range reserved for non-inhibitor homologues - and which has no annotated reactive bond and a proline at the P12 hinge position. UniProtKB:P01009 · SERPINA1 (Homo sapiens) SUPPORTS SOURCE BUT NOT TARGET SERPINA1 (Homo sapiens): MEROPS I04.001. A genuine inhibitory serpin with an intact reactive centre, so the term is right for this seed. It does not transfer to AGT, which is MEROPS I04.953 - the range reserved for non-inhibitor homologues - and which has no annotated reactive bond and a proline at the P12 hinge position. UniProtKB:P05154 · SERPINA5 (Homo sapiens) SUPPORTS SOURCE BUT NOT TARGET SERPINA5 (Homo sapiens): MEROPS I04.004. A genuine inhibitory serpin with an intact reactive centre, so the term is right for this seed. It does not transfer to AGT, which is MEROPS I04.953 - the range reserved for non-inhibitor homologues - and which has no annotated reactive bond and a proline at the P12 hinge position. UniProtKB:Q3ZEJ6 · SERPINA3-3 (Bos taurus) SUPPORTS SOURCE BUT NOT TARGET SERPINA3-3 (Bos taurus): MEROPS I04.027. A genuine inhibitory serpin with an intact reactive centre, so the term is right for this seed. It does not transfer to AGT, which is MEROPS I04.953 - the range reserved for non-inhibitor homologues - and which has no annotated reactive bond and a proline at the P12 hinge position. UniProtKB:Q86WD7 · SERPINA9 (Homo sapiens) SUPPORTS SOURCE BUT NOT TARGET SERPINA9 (Homo sapiens): MEROPS I04.082. A genuine inhibitory serpin with an intact reactive centre, so the term is right for this seed. It does not transfer to AGT, which is MEROPS I04.953 - the range reserved for non-inhibitor homologues - and which has no annotated reactive bond and a proline at the P12 hinge position. UniProtKB:Q9TTE1 · SERPINA3-1 (Bos taurus) SUPPORTS SOURCE BUT NOT TARGET SERPINA3-1 (Bos taurus): MEROPS I04.027. A genuine inhibitory serpin with an intact reactive centre, so the term is right for this seed. It does not transfer to AGT, which is MEROPS I04.953 - the range reserved for non-inhibitor homologues - and which has no annotated reactive bond and a proline at the P12 hinge position. Supporting Evidence: PMID:20927107 angiotensinogen-a non-inhibitory member of the serpin family of protease inhibitors PMID:30563843 To confirm conclusions from biochemical experiments that cleavage of the reactive center loop of AGT does not trigger the stressed-to-relaxed (S-to-R) transition characteristic of inhibitory serpins, we solved the structure of AGT cleaved by thermolysin treatment and compared it with that of intact AGT. PMID:30563843 In contrast, it has been shown that AGT has lost the ability to undergo this typical serpin S-to-R transition (29), confirmed here by our structure of loop-cleaved AGT, so it was very puzzling why the serpin framework was selected in the course of evolution as an angiotensin carrier. file:human/AGT/AGT-bioinformatics/RESULTS.md All 7 resolvable seed proteins of IBD node `PTN008970140` are MEROPS inhibitors (7/7). |
| GO:0042981 regulation of apoptotic process | IBA GO_REF:0000033 | KEEP AS NON CORE | Summary: Defensible but peripheral and deliberately generic. The node is the same angiotensinogen-specific PTN008518321 seeded by mouse Agt, rat Agt and human AGT, so the inference is not over-propagated. The bare parent term is however the right level rather than a granularity failure: angiotensin II is pro-apoptotic through AGTR2 (PMID:10406457, and AGT already carries GO:2001238 positive regulation of extrinsic apoptotic signaling pathway) and anti-apoptotic/proliferative through AGTR1, so the effect has no single sign to inherit - not because the donor angiotensinogens disagree with one another, but because the biology itself is bidirectional and receptor-dependent. The neutral parent is therefore the correct least common ancestor. GO's own comment on the term endorses this use. Reason: Real but secondary: apoptotic regulation is one downstream consequence of angiotensin receptor engagement, not what angiotensinogen is for. Kept at the parent level deliberately - this is not a GRANULARITY_MISMATCH, because a more specific term would have to pick a sign the biology does not support family-wide. Propagation Review Root cause: NO FAILURE NON CORE Sources checked: MGI:MGI:87963 · Agt (Mus musculus) SUPPORTS TRANSFER Mouse Agt (P11859); angiotensin II regulates apoptosis in mouse tissues in both directions depending on receptor subtype. PANTHER:PTN008518321 · PAINT ancestral node SUPPORTS TRANSFER Angiotensinogen-specific IBD node, the same node that correctly seeds GO:0038166. Correctly placed; the reservation is about the term's breadth, not the node. RGD:2069 · Agt (Rattus norvegicus) SUPPORTS TRANSFER Rat Agt (P01015), a 1:1 orthologue whose GO record carries experimentally grounded angiotensin phenotypes (verified live in QuickGO). UniProtKB:P01019 · AGT (Homo sapiens) SUPPORTS TRANSFER Human AGT, the target itself; its GO:2001238 IDA is one of the descendant evidences behind the node. Expected, not circular. Supporting Evidence: PMID:10406457 Deletion of residues 240-244 within the intermediate portion of the i3 loop resulted in a complete loss of AT2-mediated apoptosis, inhibition of extracellular signal-regulated kinases (ERK), and SHP-1 activation. PMID:10406457 Our data demonstrate that the intermediate portion of the i3 loop is important for AT2 function and that SHP-1 is a proximal effector of the AT2 receptor that is implicated in the inhibition of ERKs and in the apoptotic effect of this receptor. |
| GO:0003081 regulation of systemic arterial blood pressure by renin-angiotensin | IEA GO_REF:0000002 | ACCEPT | Summary: Correctly scoped InterPro2GO inference, and the control that shows what went wrong with the GO:0004867 IEA. The two signatures in the WITH/FROM are IPR000227 (Angiotensinogen) and IPR033834 (Angiotensinogen_serpin_dom) - angiotensinogen-specific, not family-wide - and they map to the angiotensinogen-specific process term. This is exactly what AGT does. Reason: Gene-level InterPro signatures mapped to the gene's core physiological process. The same protein matches the generic serpin signature IPR000215 as well, and that route produces the wrong molecular function - the contrast is diagnostic of where InterPro2GO fails and where it works. Propagation Review Root cause: NO FAILURE CORE Sources checked: InterPro:IPR000227 SUPPORTS TRANSFER InterPro 'Angiotensinogen' - a signature specific to this gene, not to the serpin family, so its GO mapping cannot leak across the family. InterPro:IPR033834 SUPPORTS TRANSFER InterPro 'Angiotensinogen_serpin_dom' - the angiotensinogen-specific serpin domain, again gene-level rather than family-level. Supporting Evidence: PMID:7989296 The systolic blood pressure of the homozygous mutant mice was 66.9 +/- 4.1 mm Hg, significantly lower than that of wild-type mice (100.4 +/- 4.4 mm Hg). PMID:25691624 AAV-driven expression of both forms of AGT led to equivalent increases of systolic blood pressure and augmentation of atherosclerotic lesion size in hepAGT-/- mice. |
| GO:0004867 serine-type endopeptidase inhibitor activity | IEA GO_REF:0000002 | REMOVE | Summary: The InterPro route to the same false claim as the GO:0004867 IBA. The single signature is IPR000215 'Serpin_fam', which matches the whole serpin family including its many non-inhibitory members (corticosteroid-binding globulin, thyroxine-binding globulin, PEDF, ovalbumin, HSP47, and angiotensinogen). The interpro2go mapping IPR000215 -> GO:0004867 is therefore only as good as the assumption that every serpin inhibits, which is false, and AGT is one of the members where it fails. Reason: A demonstrably over-broad InterPro2GO mapping applied to a member that lacks the activity. Removing it is safe on exactly the grounds the guidelines allow for IEA rows: AGT has no reactive bond, a proline-blocked hinge, MEROPS non-inhibitor classification, and a published crystal structure showing it cannot undergo the stressed-to-relaxed transition. Note that AGT's own gene-level signatures (IPR000227, IPR033834) give a correct annotation via the same pipeline. Propagation Review Root cause: PROPAGATION BAD Failure modes: FUNCTIONAL DIVERGENCE Sources checked: InterPro:IPR000215 SUPPORTS SOURCE BUT NOT TARGET InterPro 'Serpin_fam' - a whole-family signature. It correctly identifies AGT as a serpin (UniProt agrees: 'Belongs to the serpin family'), but the activity its interpro2go mapping carries is not a property of every family member and is not a property of this one. Supporting Evidence: PMID:20927107 angiotensinogen-a non-inhibitory member of the serpin family of protease inhibitors PMID:30563843 In contrast, it has been shown that AGT has lost the ability to undergo this typical serpin S-to-R transition (29), confirmed here by our structure of loop-cleaved AGT, so it was very puzzling why the serpin framework was selected in the course of evolution as an angiotensin carrier. file:human/AGT/AGT-uniprot.txt SIMILARITY: Belongs to the serpin family. |
| GO:0005576 extracellular region | IEA GO_REF:0000120 | ACCEPT | Summary: Correct. Two independent automatic routes agree: the serpin family signature IPR000215 (secretion is genuinely typical of the family) and, more directly, UniProtKB-SubCell SL-0243, which is the curated 'Secreted' subcellular location on this very entry. The second is essentially a re-expression of AGT's own curated annotation rather than an inference from anything else. Reason: Correct location, and the SubCell route is grounded in AGT's own curated record. Not independent of the IBA, ISS, IDA and HDA rows for the same term, but not wrong. Propagation Review Root cause: NO FAILURE CORE Sources checked: InterPro:IPR000215 SUPPORTS TRANSFER Serpin family signature. Over-broad for molecular function, but secretion is a real family-wide tendency and is correct for AGT. UniProtKB-SubCell:SL-0243 SUPPORTS TRANSFER The 'Secreted' SubCell term drawn from AGT's own curated SUBCELLULAR LOCATION line, which cites PubMed:4300938, 7259779 and 7539791. Not an inference from another gene. Supporting Evidence: file:human/AGT/AGT-uniprot.txt SUBCELLULAR LOCATION: Secreted |
| GO:0005515 protein binding | IPI PMID:18202720 Conformational switch of angiotensin II type 1 receptor unde... | MODIFY | Summary: Real interaction, uninformative term. The partner is rat Agtr1 (P25095), the type-1 angiotensin II receptor - UniProt records the pair as a cross-species IntAct entry with 10 experiments. The molecule on the AGT side is the angiotensin II octapeptide, and what it does to AGTR1 is act as its agonist. 'Protein binding' says none of that, and GO curation guidance is to replace bare GO:0005515 with an informative molecular function wherever the partner is known. Reason: The specific term GO:0031702 type 1 angiotensin receptor binding is available, is already used elsewhere in this gene's record for the same interaction from a different reference, and states what the bare parent hides. Proposed replacements: type 1 angiotensin receptor binding Supporting Evidence: PMID:18202720 Here, we show that cell stretch leads to activation of the AT(1) receptor, which undergoes an anticlockwise rotation and a shift of transmembrane (TM) 7 into the ligand-binding pocket. file:human/AGT/AGT-uniprot.txt P01019; P25095: Agtr1; Xeno; NbExp=10 |
| GO:0005515 protein binding | IPI PMID:20927107 A redox switch in angiotensinogen modulates angiotensin rele... | MODIFY | Summary: The single most important interaction in this gene's record, recorded as bare 'protein binding'. The partner is renin (P00797), and the association is not a transient enzyme-substrate encounter: the crystal structure of the human angiotensinogen-renin complex shows a 670 A^2 body-to-body interface outside the active-site cleft, with helix A of AGT lying across renin's active site and helix C and the CD-loop contacting renin's N-terminal lobe. The follow-up structures show a tail-into-mouth allosteric mechanism and demonstrate by mutagenesis and kinetics that specificity is set by AGT residues and glycans outside that cleft. Reason: GO:0002020 protease binding is the informative molecular function this interaction supports and is the only term in AGT's whole record that describes what the full-length precursor itself does. The interface is extensive, structurally characterised, and functionally load-bearing rather than incidental to catalysis. Proposed replacements: protease binding Supporting Evidence: PMID:20927107 there is a substantial contact surface of 670Å2 between the bodies of the two proteins, primarily hydrophobic in nature PMID:30563843 These structures revealed that AGT undergoes profound conformational changes and binds renin through a tail-into-mouth allosteric mechanism that inserts the N terminus into a pocket equivalent to a hormone-binding site on other serpins. PMID:30563843 Mutagenesis and kinetic analyses confirmed that renin-mediated production of angiotensin I is controlled by interactions of amino acid residues and glycan components outside renin's active-site cleft. |
| GO:0005515 protein binding | IPI PMID:23082758 Structural basis of peptide recognition by the angiotensin-1... | MARK AS OVER ANNOTATED | Summary: A cross-phylum enzyme-substrate co-crystal reported as a protein-protein interaction. The partner Q10714 is AnCE, the Drosophila melanogaster angiotensin-converting enzyme homologue, and UniProt flags the pair 'Xeno'. What was actually crystallised is the insect enzyme with the mammalian angiotensin I decapeptide (and bradykinin, and a snake-venom peptide) sitting in its active site. That is a peptide in a catalytic cleft, not an interaction of human angiotensinogen with a human partner. Reason: Structurally real but of no interpretive value for human AGT: a 10-residue cleavage product of AGT bound in the active site of a Drosophila enzyme, used as a structural model for human ACE. Human ACE processing of angiotensin I is already represented properly in this record by the Reactome reactions. Kept rather than removed because the structure is genuine. Supporting Evidence: PMID:23082758 Here, we report the structures of an ACE homologue from Drosophila melanogaster (AnCE; a proven structural model for the more complex human ACE) co-crystallized with mammalian peptide substrates (bradykinin, Thr(6) -bradykinin, angiotensin I and a snake venom peptide inhibitor, bradykinin-potentiating peptide-b). file:human/AGT/AGT-uniprot.txt P01019; Q10714: Ance; Xeno; NbExp=2 |
| GO:0005515 protein binding | IPI PMID:32814053 Interactome Mapping Provides a Network of Neurodegenerative ... | MARK AS OVER ANNOTATED | Summary: One of ten GO:0005515 rows from a single large-scale yeast two-hybrid interactome of neurodegenerative disease proteins. Angiotensinogen is not a neurodegeneration protein and appears here as library prey. The partner in this row is UniProtKB:C9J082 (NPHP1, Nephrocystin 1), for which UniProt annotates no subcellular location at all, so the pairing cannot even be assessed for compartment plausibility. No functional consequence of this pairing has been reported for either protein. Reason: Bare 'protein binding' from one high-throughput screen, with no functional follow-up, no orthogonal validation, and a systematic compartment mismatch: AGT is a secreted, signal-peptide-cleaved, disulfide-bonded, four-site N-glycosylated plasma protein, while yeast two-hybrid reconstitutes a transcription factor in the yeast nucleus. Across all ten partners, 0 of 10 share a secreted or extracellular compartment with AGT. The paper itself calls these candidate interactions, and UniProt's NbExp=3 is replicates within this one study rather than three independent studies. Marked over-annotated rather than removed: the screen is reproducible, it is simply uncharacterised for this gene. Supporting Evidence: PMID:32814053 candidate interactions and is generated by systematic yeast two-hybrid interaction screening of file:human/AGT/AGT-bioinformatics/RESULTS.md None of the ten PMID:32814053 partners is annotated as secreted or extracellular. file:human/AGT/AGT-uniprot.txt SUBCELLULAR LOCATION: Secreted |
| GO:0005515 protein binding | IPI PMID:32814053 Interactome Mapping Provides a Network of Neurodegenerative ... | MARK AS OVER ANNOTATED | Summary: One of ten GO:0005515 rows from a single large-scale yeast two-hybrid interactome of neurodegenerative disease proteins. Angiotensinogen is not a neurodegeneration protein and appears here as library prey. The partner in this row is UniProtKB:O00746 (NME4, Nucleoside diphosphate kinase D, mitochondrial), which UniProt places in Mitochondrion intermembrane space; Mitochondrion matrix; Mitochondrion outer membrane - an intracellular compartment, not the secreted one AGT occupies. No functional consequence of this pairing has been reported for either protein. Reason: Bare 'protein binding' from one high-throughput screen, with no functional follow-up, no orthogonal validation, and a systematic compartment mismatch: AGT is a secreted, signal-peptide-cleaved, disulfide-bonded, four-site N-glycosylated plasma protein, while yeast two-hybrid reconstitutes a transcription factor in the yeast nucleus. Across all ten partners, 0 of 10 share a secreted or extracellular compartment with AGT. The paper itself calls these candidate interactions, and UniProt's NbExp=3 is replicates within this one study rather than three independent studies. Marked over-annotated rather than removed: the screen is reproducible, it is simply uncharacterised for this gene. Supporting Evidence: PMID:32814053 candidate interactions and is generated by systematic yeast two-hybrid interaction screening of file:human/AGT/AGT-bioinformatics/RESULTS.md None of the ten PMID:32814053 partners is annotated as secreted or extracellular. file:human/AGT/AGT-uniprot.txt SUBCELLULAR LOCATION: Secreted |
| GO:0005515 protein binding | IPI PMID:32814053 Interactome Mapping Provides a Network of Neurodegenerative ... | MARK AS OVER ANNOTATED | Summary: One of ten GO:0005515 rows from a single large-scale yeast two-hybrid interactome of neurodegenerative disease proteins. Angiotensinogen is not a neurodegeneration protein and appears here as library prey. The partner in this row is UniProtKB:O14744 (PRMT5, Protein arginine N-methyltransferase 5), which UniProt places in Cytoplasm; Nucleus; Chromosome; Golgi apparatus - an intracellular compartment, not the secreted one AGT occupies. No functional consequence of this pairing has been reported for either protein. Reason: Bare 'protein binding' from one high-throughput screen, with no functional follow-up, no orthogonal validation, and a systematic compartment mismatch: AGT is a secreted, signal-peptide-cleaved, disulfide-bonded, four-site N-glycosylated plasma protein, while yeast two-hybrid reconstitutes a transcription factor in the yeast nucleus. Across all ten partners, 0 of 10 share a secreted or extracellular compartment with AGT. The paper itself calls these candidate interactions, and UniProt's NbExp=3 is replicates within this one study rather than three independent studies. Marked over-annotated rather than removed: the screen is reproducible, it is simply uncharacterised for this gene. Supporting Evidence: PMID:32814053 candidate interactions and is generated by systematic yeast two-hybrid interaction screening of file:human/AGT/AGT-bioinformatics/RESULTS.md None of the ten PMID:32814053 partners is annotated as secreted or extracellular. file:human/AGT/AGT-uniprot.txt SUBCELLULAR LOCATION: Secreted |
| GO:0005515 protein binding | IPI PMID:32814053 Interactome Mapping Provides a Network of Neurodegenerative ... | MARK AS OVER ANNOTATED | Summary: One of ten GO:0005515 rows from a single large-scale yeast two-hybrid interactome of neurodegenerative disease proteins. Angiotensinogen is not a neurodegeneration protein and appears here as library prey. The partner in this row is UniProtKB:O43493-5 (TGOLN2, Trans-Golgi network integral membrane protein 2), which UniProt places in Cell membrane; Golgi apparatus, trans-Golgi network membrane - an intracellular compartment, not the secreted one AGT occupies. No functional consequence of this pairing has been reported for either protein. Reason: Bare 'protein binding' from one high-throughput screen, with no functional follow-up, no orthogonal validation, and a systematic compartment mismatch: AGT is a secreted, signal-peptide-cleaved, disulfide-bonded, four-site N-glycosylated plasma protein, while yeast two-hybrid reconstitutes a transcription factor in the yeast nucleus. Across all ten partners, 0 of 10 share a secreted or extracellular compartment with AGT. The paper itself calls these candidate interactions, and UniProt's NbExp=3 is replicates within this one study rather than three independent studies. Marked over-annotated rather than removed: the screen is reproducible, it is simply uncharacterised for this gene. Supporting Evidence: PMID:32814053 candidate interactions and is generated by systematic yeast two-hybrid interaction screening of file:human/AGT/AGT-bioinformatics/RESULTS.md None of the ten PMID:32814053 partners is annotated as secreted or extracellular. file:human/AGT/AGT-uniprot.txt SUBCELLULAR LOCATION: Secreted |
| GO:0005515 protein binding | IPI PMID:32814053 Interactome Mapping Provides a Network of Neurodegenerative ... | MARK AS OVER ANNOTATED | Summary: One of ten GO:0005515 rows from a single large-scale yeast two-hybrid interactome of neurodegenerative disease proteins. Angiotensinogen is not a neurodegeneration protein and appears here as library prey. The partner in this row is UniProtKB:Q8IYM2 (SLFN12, Ribonuclease SLFN12), which UniProt places in Nucleus; Cytoplasm, cytosol - an intracellular compartment, not the secreted one AGT occupies. No functional consequence of this pairing has been reported for either protein. Reason: Bare 'protein binding' from one high-throughput screen, with no functional follow-up, no orthogonal validation, and a systematic compartment mismatch: AGT is a secreted, signal-peptide-cleaved, disulfide-bonded, four-site N-glycosylated plasma protein, while yeast two-hybrid reconstitutes a transcription factor in the yeast nucleus. Across all ten partners, 0 of 10 share a secreted or extracellular compartment with AGT. The paper itself calls these candidate interactions, and UniProt's NbExp=3 is replicates within this one study rather than three independent studies. Marked over-annotated rather than removed: the screen is reproducible, it is simply uncharacterised for this gene. Supporting Evidence: PMID:32814053 candidate interactions and is generated by systematic yeast two-hybrid interaction screening of file:human/AGT/AGT-bioinformatics/RESULTS.md None of the ten PMID:32814053 partners is annotated as secreted or extracellular. file:human/AGT/AGT-uniprot.txt SUBCELLULAR LOCATION: Secreted |
| GO:0005515 protein binding | IPI PMID:32814053 Interactome Mapping Provides a Network of Neurodegenerative ... | MARK AS OVER ANNOTATED | Summary: One of ten GO:0005515 rows from a single large-scale yeast two-hybrid interactome of neurodegenerative disease proteins. Angiotensinogen is not a neurodegeneration protein and appears here as library prey. The partner in this row is UniProtKB:Q8N0U8 (VKORC1L1, Vitamin K epoxide reductase complex subunit 1-like protein 1), which UniProt places in Endoplasmic reticulum membrane - an intracellular compartment, not the secreted one AGT occupies. No functional consequence of this pairing has been reported for either protein. Reason: Bare 'protein binding' from one high-throughput screen, with no functional follow-up, no orthogonal validation, and a systematic compartment mismatch: AGT is a secreted, signal-peptide-cleaved, disulfide-bonded, four-site N-glycosylated plasma protein, while yeast two-hybrid reconstitutes a transcription factor in the yeast nucleus. Across all ten partners, 0 of 10 share a secreted or extracellular compartment with AGT. The paper itself calls these candidate interactions, and UniProt's NbExp=3 is replicates within this one study rather than three independent studies. Marked over-annotated rather than removed: the screen is reproducible, it is simply uncharacterised for this gene. Supporting Evidence: PMID:32814053 candidate interactions and is generated by systematic yeast two-hybrid interaction screening of file:human/AGT/AGT-bioinformatics/RESULTS.md None of the ten PMID:32814053 partners is annotated as secreted or extracellular. file:human/AGT/AGT-uniprot.txt SUBCELLULAR LOCATION: Secreted |
| GO:0005515 protein binding | IPI PMID:32814053 Interactome Mapping Provides a Network of Neurodegenerative ... | MARK AS OVER ANNOTATED | Summary: One of ten GO:0005515 rows from a single large-scale yeast two-hybrid interactome of neurodegenerative disease proteins. Angiotensinogen is not a neurodegeneration protein and appears here as library prey. The partner in this row is UniProtKB:Q8NFB2 (TMEM185A, Transmembrane protein 185A), which UniProt places in Cell projection, dendrite; Membrane - an intracellular compartment, not the secreted one AGT occupies. No functional consequence of this pairing has been reported for either protein. Reason: Bare 'protein binding' from one high-throughput screen, with no functional follow-up, no orthogonal validation, and a systematic compartment mismatch: AGT is a secreted, signal-peptide-cleaved, disulfide-bonded, four-site N-glycosylated plasma protein, while yeast two-hybrid reconstitutes a transcription factor in the yeast nucleus. Across all ten partners, 0 of 10 share a secreted or extracellular compartment with AGT. The paper itself calls these candidate interactions, and UniProt's NbExp=3 is replicates within this one study rather than three independent studies. Marked over-annotated rather than removed: the screen is reproducible, it is simply uncharacterised for this gene. Supporting Evidence: PMID:32814053 candidate interactions and is generated by systematic yeast two-hybrid interaction screening of file:human/AGT/AGT-bioinformatics/RESULTS.md None of the ten PMID:32814053 partners is annotated as secreted or extracellular. file:human/AGT/AGT-uniprot.txt SUBCELLULAR LOCATION: Secreted |
| GO:0005515 protein binding | IPI PMID:32814053 Interactome Mapping Provides a Network of Neurodegenerative ... | MARK AS OVER ANNOTATED | Summary: One of ten GO:0005515 rows from a single large-scale yeast two-hybrid interactome of neurodegenerative disease proteins. Angiotensinogen is not a neurodegeneration protein and appears here as library prey. The partner in this row is UniProtKB:Q8TAS3 (PRRG2, PRRG2 protein), for which UniProt annotates no subcellular location at all, so the pairing cannot even be assessed for compartment plausibility. No functional consequence of this pairing has been reported for either protein. Reason: Bare 'protein binding' from one high-throughput screen, with no functional follow-up, no orthogonal validation, and a systematic compartment mismatch: AGT is a secreted, signal-peptide-cleaved, disulfide-bonded, four-site N-glycosylated plasma protein, while yeast two-hybrid reconstitutes a transcription factor in the yeast nucleus. Across all ten partners, 0 of 10 share a secreted or extracellular compartment with AGT. The paper itself calls these candidate interactions, and UniProt's NbExp=3 is replicates within this one study rather than three independent studies. Marked over-annotated rather than removed: the screen is reproducible, it is simply uncharacterised for this gene. Supporting Evidence: PMID:32814053 candidate interactions and is generated by systematic yeast two-hybrid interaction screening of file:human/AGT/AGT-bioinformatics/RESULTS.md None of the ten PMID:32814053 partners is annotated as secreted or extracellular. file:human/AGT/AGT-uniprot.txt SUBCELLULAR LOCATION: Secreted |
| GO:0005515 protein binding | IPI PMID:32814053 Interactome Mapping Provides a Network of Neurodegenerative ... | MARK AS OVER ANNOTATED | Summary: One of ten GO:0005515 rows from a single large-scale yeast two-hybrid interactome of neurodegenerative disease proteins. Angiotensinogen is not a neurodegeneration protein and appears here as library prey. The partner in this row is UniProtKB:Q9UI10 (EIF2B4, Translation initiation factor eIF2B subunit delta), which UniProt places in Cytoplasm, cytosol - an intracellular compartment, not the secreted one AGT occupies. No functional consequence of this pairing has been reported for either protein. Reason: Bare 'protein binding' from one high-throughput screen, with no functional follow-up, no orthogonal validation, and a systematic compartment mismatch: AGT is a secreted, signal-peptide-cleaved, disulfide-bonded, four-site N-glycosylated plasma protein, while yeast two-hybrid reconstitutes a transcription factor in the yeast nucleus. Across all ten partners, 0 of 10 share a secreted or extracellular compartment with AGT. The paper itself calls these candidate interactions, and UniProt's NbExp=3 is replicates within this one study rather than three independent studies. Marked over-annotated rather than removed: the screen is reproducible, it is simply uncharacterised for this gene. Supporting Evidence: PMID:32814053 candidate interactions and is generated by systematic yeast two-hybrid interaction screening of file:human/AGT/AGT-bioinformatics/RESULTS.md None of the ten PMID:32814053 partners is annotated as secreted or extracellular. file:human/AGT/AGT-uniprot.txt SUBCELLULAR LOCATION: Secreted |
| GO:0005515 protein binding | IPI PMID:32814053 Interactome Mapping Provides a Network of Neurodegenerative ... | MARK AS OVER ANNOTATED | Summary: One of ten GO:0005515 rows from a single large-scale yeast two-hybrid interactome of neurodegenerative disease proteins. Angiotensinogen is not a neurodegeneration protein and appears here as library prey. The partner in this row is UniProtKB:Q9UMY4 (SNX12, Sorting nexin-12), which UniProt places in Membrane - an intracellular compartment, not the secreted one AGT occupies. No functional consequence of this pairing has been reported for either protein. Reason: Bare 'protein binding' from one high-throughput screen, with no functional follow-up, no orthogonal validation, and a systematic compartment mismatch: AGT is a secreted, signal-peptide-cleaved, disulfide-bonded, four-site N-glycosylated plasma protein, while yeast two-hybrid reconstitutes a transcription factor in the yeast nucleus. Across all ten partners, 0 of 10 share a secreted or extracellular compartment with AGT. The paper itself calls these candidate interactions, and UniProt's NbExp=3 is replicates within this one study rather than three independent studies. Marked over-annotated rather than removed: the screen is reproducible, it is simply uncharacterised for this gene. Supporting Evidence: PMID:32814053 candidate interactions and is generated by systematic yeast two-hybrid interaction screening of file:human/AGT/AGT-bioinformatics/RESULTS.md None of the ten PMID:32814053 partners is annotated as secreted or extracellular. file:human/AGT/AGT-uniprot.txt SUBCELLULAR LOCATION: Secreted |
| GO:0005576 extracellular region | TAS Reactome:R-HSA-1989774 | ACCEPT | Summary: One of 27 Reactome TAS rows on this gene, all asserting the same term - located_in extracellular region - and differing only in which reaction is cited. Together they are 24% of AGT's entire GOA record expressing a single correct claim once per reaction the protein or one of its peptides participates in. This row cites Reactome:R-HSA-1989774, "Expression of AGT". Reason: The term is correct: angiotensinogen and all of its peptide products act in the extracellular compartment, and this reaction takes place there. Accepted on that basis. The caveat is about reading the record rather than about this row: 27 identical assertions from one source database are one projection, not 27 independent findings, and the count should not be mistaken for weight of evidence. Supporting Evidence: file:human/AGT/AGT-uniprot.txt SUBCELLULAR LOCATION: Secreted file:human/AGT/AGT-uniprot.txt TISSUE SPECIFICITY: Expressed by the liver and secreted in plasma. |
| GO:0005576 extracellular region | TAS Reactome:R-HSA-2022403 | ACCEPT | Summary: One of 27 Reactome TAS rows on this gene, all asserting the same term - located_in extracellular region - and differing only in which reaction is cited. Together they are 24% of AGT's entire GOA record expressing a single correct claim once per reaction the protein or one of its peptides participates in. This row cites Reactome:R-HSA-2022403, "Renin:Prorenin Receptor hydrolyzes Angiotensinogen to AGT(25-34) (Angiotensin-(1-10))". Reason: The term is correct: angiotensinogen and all of its peptide products act in the extracellular compartment, and this reaction takes place there. Accepted on that basis. The caveat is about reading the record rather than about this row: 27 identical assertions from one source database are one projection, not 27 independent findings, and the count should not be mistaken for weight of evidence. Supporting Evidence: file:human/AGT/AGT-uniprot.txt SUBCELLULAR LOCATION: Secreted file:human/AGT/AGT-uniprot.txt TISSUE SPECIFICITY: Expressed by the liver and secreted in plasma. |
| GO:0005576 extracellular region | TAS Reactome:R-HSA-2022412 | ACCEPT | Summary: One of 27 Reactome TAS rows on this gene, all asserting the same term - located_in extracellular region - and differing only in which reaction is cited. Together they are 24% of AGT's entire GOA record expressing a single correct claim once per reaction the protein or one of its peptides participates in. This row cites Reactome:R-HSA-2022412, "Renin hydrolyzes Angiotensinogen to AGT(25-34) (Angiotensin-(1-10))". Reason: The term is correct: angiotensinogen and all of its peptide products act in the extracellular compartment, and this reaction takes place there. Accepted on that basis. The caveat is about reading the record rather than about this row: 27 identical assertions from one source database are one projection, not 27 independent findings, and the count should not be mistaken for weight of evidence. Supporting Evidence: file:human/AGT/AGT-uniprot.txt SUBCELLULAR LOCATION: Secreted file:human/AGT/AGT-uniprot.txt TISSUE SPECIFICITY: Expressed by the liver and secreted in plasma. |
| GO:0005576 extracellular region | TAS Reactome:R-HSA-2065357 | ACCEPT | Summary: One of 27 Reactome TAS rows on this gene, all asserting the same term - located_in extracellular region - and differing only in which reaction is cited. Together they are 24% of AGT's entire GOA record expressing a single correct claim once per reaction the protein or one of its peptides participates in. This row cites Reactome:R-HSA-2065357, "Prorenin:Prorenin Receptor hydrolyzes Angiotensinogen to AGT(25-35) (Angiotensin-(1-10))". Reason: The term is correct: angiotensinogen and all of its peptide products act in the extracellular compartment, and this reaction takes place there. Accepted on that basis. The caveat is about reading the record rather than about this row: 27 identical assertions from one source database are one projection, not 27 independent findings, and the count should not be mistaken for weight of evidence. Supporting Evidence: file:human/AGT/AGT-uniprot.txt SUBCELLULAR LOCATION: Secreted file:human/AGT/AGT-uniprot.txt TISSUE SPECIFICITY: Expressed by the liver and secreted in plasma. |
| GO:0005576 extracellular region | TAS Reactome:R-HSA-2022368 | ACCEPT | Summary: One of 27 Reactome TAS rows on this gene, all asserting the same term - located_in extracellular region - and differing only in which reaction is cited. Together they are 24% of AGT's entire GOA record expressing a single correct claim once per reaction the protein or one of its peptides participates in. This row cites Reactome:R-HSA-2022368, "MME:Zn2+ (Neprilysin) hydrolyses AGT(25-33) (Angiotensin-(1-9)) to AGT(25-31) (Angiotensin-(1-7)". Reason: The term is correct: angiotensinogen and all of its peptide products act in the extracellular compartment, and this reaction takes place there. Accepted on that basis. The caveat is about reading the record rather than about this row: 27 identical assertions from one source database are one projection, not 27 independent findings, and the count should not be mistaken for weight of evidence. Supporting Evidence: file:human/AGT/AGT-uniprot.txt SUBCELLULAR LOCATION: Secreted file:human/AGT/AGT-uniprot.txt TISSUE SPECIFICITY: Expressed by the liver and secreted in plasma. |
| GO:0005576 extracellular region | TAS Reactome:R-HSA-2022378 | ACCEPT | Summary: One of 27 Reactome TAS rows on this gene, all asserting the same term - located_in extracellular region - and differing only in which reaction is cited. Together they are 24% of AGT's entire GOA record expressing a single correct claim once per reaction the protein or one of its peptides participates in. This row cites Reactome:R-HSA-2022378, "ACE2(18-805):Zn2+ hydrolyzes AGT(25-34) (Angiotensin-(1-10)) to AGT(25-33) (Angiotensin-(1-9))". Reason: The term is correct: angiotensinogen and all of its peptide products act in the extracellular compartment, and this reaction takes place there. Accepted on that basis. The caveat is about reading the record rather than about this row: 27 identical assertions from one source database are one projection, not 27 independent findings, and the count should not be mistaken for weight of evidence. Supporting Evidence: file:human/AGT/AGT-uniprot.txt SUBCELLULAR LOCATION: Secreted file:human/AGT/AGT-uniprot.txt TISSUE SPECIFICITY: Expressed by the liver and secreted in plasma. |
| GO:0005576 extracellular region | TAS Reactome:R-HSA-2022379 | ACCEPT | Summary: One of 27 Reactome TAS rows on this gene, all asserting the same term - located_in extracellular region - and differing only in which reaction is cited. Together they are 24% of AGT's entire GOA record expressing a single correct claim once per reaction the protein or one of its peptides participates in. This row cites Reactome:R-HSA-2022379, "ACE2(18-805):Zn2+ hydrolyzes AGT(25-32) (Angiotensin-(1-8)) to AGT(25-31) (Angiotensin-(1-7))". Reason: The term is correct: angiotensinogen and all of its peptide products act in the extracellular compartment, and this reaction takes place there. Accepted on that basis. The caveat is about reading the record rather than about this row: 27 identical assertions from one source database are one projection, not 27 independent findings, and the count should not be mistaken for weight of evidence. Supporting Evidence: file:human/AGT/AGT-uniprot.txt SUBCELLULAR LOCATION: Secreted file:human/AGT/AGT-uniprot.txt TISSUE SPECIFICITY: Expressed by the liver and secreted in plasma. |
| GO:0005576 extracellular region | TAS Reactome:R-HSA-2022381 | ACCEPT | Summary: One of 27 Reactome TAS rows on this gene, all asserting the same term - located_in extracellular region - and differing only in which reaction is cited. Together they are 24% of AGT's entire GOA record expressing a single correct claim once per reaction the protein or one of its peptides participates in. This row cites Reactome:R-HSA-2022381, "Cathepsin Z (Cathepsin X) hydrolyzes AGT(25-34) (Angiotensin-(1-10)) to AGT(25-32) (Angiotensin-(1-8))". Reason: The term is correct: angiotensinogen and all of its peptide products act in the extracellular compartment, and this reaction takes place there. Accepted on that basis. The caveat is about reading the record rather than about this row: 27 identical assertions from one source database are one projection, not 27 independent findings, and the count should not be mistaken for weight of evidence. Supporting Evidence: file:human/AGT/AGT-uniprot.txt SUBCELLULAR LOCATION: Secreted file:human/AGT/AGT-uniprot.txt TISSUE SPECIFICITY: Expressed by the liver and secreted in plasma. |
| GO:0005576 extracellular region | TAS Reactome:R-HSA-2022383 | ACCEPT | Summary: One of 27 Reactome TAS rows on this gene, all asserting the same term - located_in extracellular region - and differing only in which reaction is cited. Together they are 24% of AGT's entire GOA record expressing a single correct claim once per reaction the protein or one of its peptides participates in. This row cites Reactome:R-HSA-2022383, "Chymase hydrolyzes AGT(25-34) (Angiotensin-(1-10)) to AGT(25-32) (Angiotensin-(1-8))". Reason: The term is correct: angiotensinogen and all of its peptide products act in the extracellular compartment, and this reaction takes place there. Accepted on that basis. The caveat is about reading the record rather than about this row: 27 identical assertions from one source database are one projection, not 27 independent findings, and the count should not be mistaken for weight of evidence. Supporting Evidence: file:human/AGT/AGT-uniprot.txt SUBCELLULAR LOCATION: Secreted file:human/AGT/AGT-uniprot.txt TISSUE SPECIFICITY: Expressed by the liver and secreted in plasma. |
| GO:0005576 extracellular region | TAS Reactome:R-HSA-2022396 | ACCEPT | Summary: One of 27 Reactome TAS rows on this gene, all asserting the same term - located_in extracellular region - and differing only in which reaction is cited. Together they are 24% of AGT's entire GOA record expressing a single correct claim once per reaction the protein or one of its peptides participates in. This row cites Reactome:R-HSA-2022396, "MME:Zn2+ (Neprilysin) hydrolyses AGT(25-34) (Angiotensin-(1-10)) to AGT(25-31) (Angiotensin-(1-7))". Reason: The term is correct: angiotensinogen and all of its peptide products act in the extracellular compartment, and this reaction takes place there. Accepted on that basis. The caveat is about reading the record rather than about this row: 27 identical assertions from one source database are one projection, not 27 independent findings, and the count should not be mistaken for weight of evidence. Supporting Evidence: file:human/AGT/AGT-uniprot.txt SUBCELLULAR LOCATION: Secreted file:human/AGT/AGT-uniprot.txt TISSUE SPECIFICITY: Expressed by the liver and secreted in plasma. |
| GO:0005576 extracellular region | TAS Reactome:R-HSA-2022398 | ACCEPT | Summary: One of 27 Reactome TAS rows on this gene, all asserting the same term - located_in extracellular region - and differing only in which reaction is cited. Together they are 24% of AGT's entire GOA record expressing a single correct claim once per reaction the protein or one of its peptides participates in. This row cites Reactome:R-HSA-2022398, "ACE:Zn2+ hydrolyzes AGT(25-33) (Angiotensin-(1-9)) to AGT(25-31) (Angiotensin-(1-7))". Reason: The term is correct: angiotensinogen and all of its peptide products act in the extracellular compartment, and this reaction takes place there. Accepted on that basis. The caveat is about reading the record rather than about this row: 27 identical assertions from one source database are one projection, not 27 independent findings, and the count should not be mistaken for weight of evidence. Supporting Evidence: file:human/AGT/AGT-uniprot.txt SUBCELLULAR LOCATION: Secreted file:human/AGT/AGT-uniprot.txt TISSUE SPECIFICITY: Expressed by the liver and secreted in plasma. |
| GO:0005576 extracellular region | TAS Reactome:R-HSA-2022399 | ACCEPT | Summary: One of 27 Reactome TAS rows on this gene, all asserting the same term - located_in extracellular region - and differing only in which reaction is cited. Together they are 24% of AGT's entire GOA record expressing a single correct claim once per reaction the protein or one of its peptides participates in. This row cites Reactome:R-HSA-2022399, "ENPEP:Zn2+ hydrolyzes AGT(25-32) (Angiotensin-(1-8)) to AGT(26-32) (Angiotensin-(2-8))". Reason: The term is correct: angiotensinogen and all of its peptide products act in the extracellular compartment, and this reaction takes place there. Accepted on that basis. The caveat is about reading the record rather than about this row: 27 identical assertions from one source database are one projection, not 27 independent findings, and the count should not be mistaken for weight of evidence. Supporting Evidence: file:human/AGT/AGT-uniprot.txt SUBCELLULAR LOCATION: Secreted file:human/AGT/AGT-uniprot.txt TISSUE SPECIFICITY: Expressed by the liver and secreted in plasma. |
| GO:0005576 extracellular region | TAS Reactome:R-HSA-2022405 | ACCEPT | Summary: One of 27 Reactome TAS rows on this gene, all asserting the same term - located_in extracellular region - and differing only in which reaction is cited. Together they are 24% of AGT's entire GOA record expressing a single correct claim once per reaction the protein or one of its peptides participates in. This row cites Reactome:R-HSA-2022405, "ACE:Zn2+ hydrolyzes AGT(25-34) (Angiotensin-(1-10)) to AGT(25-32) (Angiotensin-(1-8))". Reason: The term is correct: angiotensinogen and all of its peptide products act in the extracellular compartment, and this reaction takes place there. Accepted on that basis. The caveat is about reading the record rather than about this row: 27 identical assertions from one source database are one projection, not 27 independent findings, and the count should not be mistaken for weight of evidence. Supporting Evidence: file:human/AGT/AGT-uniprot.txt SUBCELLULAR LOCATION: Secreted file:human/AGT/AGT-uniprot.txt TISSUE SPECIFICITY: Expressed by the liver and secreted in plasma. |
| GO:0005576 extracellular region | TAS Reactome:R-HSA-2022411 | ACCEPT | Summary: One of 27 Reactome TAS rows on this gene, all asserting the same term - located_in extracellular region - and differing only in which reaction is cited. Together they are 24% of AGT's entire GOA record expressing a single correct claim once per reaction the protein or one of its peptides participates in. This row cites Reactome:R-HSA-2022411, "Cathepsin G hydrolyzes AGT(25-34) (Angiotensin-(1-10)) to AGT(25-32) (Angiotensin-(1-8))". Reason: The term is correct: angiotensinogen and all of its peptide products act in the extracellular compartment, and this reaction takes place there. Accepted on that basis. The caveat is about reading the record rather than about this row: 27 identical assertions from one source database are one projection, not 27 independent findings, and the count should not be mistaken for weight of evidence. Supporting Evidence: file:human/AGT/AGT-uniprot.txt SUBCELLULAR LOCATION: Secreted file:human/AGT/AGT-uniprot.txt TISSUE SPECIFICITY: Expressed by the liver and secreted in plasma. |
| GO:0005576 extracellular region | TAS Reactome:R-HSA-2028294 | ACCEPT | Summary: One of 27 Reactome TAS rows on this gene, all asserting the same term - located_in extracellular region - and differing only in which reaction is cited. Together they are 24% of AGT's entire GOA record expressing a single correct claim once per reaction the protein or one of its peptides participates in. This row cites Reactome:R-HSA-2028294, "Mast Cell Carboxypeptidase hydrolyzes AGT(25-34) (Angiotensin-(1-10)) to AGT(25-33) (Angiotensin-(1-9))". Reason: The term is correct: angiotensinogen and all of its peptide products act in the extracellular compartment, and this reaction takes place there. Accepted on that basis. The caveat is about reading the record rather than about this row: 27 identical assertions from one source database are one projection, not 27 independent findings, and the count should not be mistaken for weight of evidence. Supporting Evidence: file:human/AGT/AGT-uniprot.txt SUBCELLULAR LOCATION: Secreted file:human/AGT/AGT-uniprot.txt TISSUE SPECIFICITY: Expressed by the liver and secreted in plasma. |
| GO:0005576 extracellular region | TAS Reactome:R-HSA-2065355 | ACCEPT | Summary: One of 27 Reactome TAS rows on this gene, all asserting the same term - located_in extracellular region - and differing only in which reaction is cited. Together they are 24% of AGT's entire GOA record expressing a single correct claim once per reaction the protein or one of its peptides participates in. This row cites Reactome:R-HSA-2065355, "Secreted ACE:Zn2+ hydrolyzes AGT(25-34) (Angiotensin-(1-10)) to AGT(25-32) (Angiotensin-(1-8))". Reason: The term is correct: angiotensinogen and all of its peptide products act in the extracellular compartment, and this reaction takes place there. Accepted on that basis. The caveat is about reading the record rather than about this row: 27 identical assertions from one source database are one projection, not 27 independent findings, and the count should not be mistaken for weight of evidence. Supporting Evidence: file:human/AGT/AGT-uniprot.txt SUBCELLULAR LOCATION: Secreted file:human/AGT/AGT-uniprot.txt TISSUE SPECIFICITY: Expressed by the liver and secreted in plasma. |
| GO:0005576 extracellular region | TAS Reactome:R-HSA-374173 | ACCEPT | Summary: One of 27 Reactome TAS rows on this gene, all asserting the same term - located_in extracellular region - and differing only in which reaction is cited. Together they are 24% of AGT's entire GOA record expressing a single correct claim once per reaction the protein or one of its peptides participates in. This row cites Reactome:R-HSA-374173, "AGT(25-32) binds to AGTR1". Reason: The term is correct: angiotensinogen and all of its peptide products act in the extracellular compartment, and this reaction takes place there. Accepted on that basis. The caveat is about reading the record rather than about this row: 27 identical assertions from one source database are one projection, not 27 independent findings, and the count should not be mistaken for weight of evidence. Supporting Evidence: file:human/AGT/AGT-uniprot.txt SUBCELLULAR LOCATION: Secreted file:human/AGT/AGT-uniprot.txt TISSUE SPECIFICITY: Expressed by the liver and secreted in plasma. |
| GO:0005576 extracellular region | TAS Reactome:R-HSA-379048 | ACCEPT | Summary: One of 27 Reactome TAS rows on this gene, all asserting the same term - located_in extracellular region - and differing only in which reaction is cited. Together they are 24% of AGT's entire GOA record expressing a single correct claim once per reaction the protein or one of its peptides participates in. This row cites Reactome:R-HSA-379048, "Liganded Gq/11-activating GPCRs act as GEFs for Gq/11". Reason: The term is correct: angiotensinogen and all of its peptide products act in the extracellular compartment, and this reaction takes place there. Accepted on that basis. The caveat is about reading the record rather than about this row: 27 identical assertions from one source database are one projection, not 27 independent findings, and the count should not be mistaken for weight of evidence. Supporting Evidence: file:human/AGT/AGT-uniprot.txt SUBCELLULAR LOCATION: Secreted file:human/AGT/AGT-uniprot.txt TISSUE SPECIFICITY: Expressed by the liver and secreted in plasma. |
| GO:0005576 extracellular region | TAS Reactome:R-HSA-380073 | ACCEPT | Summary: One of 27 Reactome TAS rows on this gene, all asserting the same term - located_in extracellular region - and differing only in which reaction is cited. Together they are 24% of AGT's entire GOA record expressing a single correct claim once per reaction the protein or one of its peptides participates in. This row cites Reactome:R-HSA-380073, "Liganded Gi-activating GPCR acts as a GEF for Gi". Reason: The term is correct: angiotensinogen and all of its peptide products act in the extracellular compartment, and this reaction takes place there. Accepted on that basis. The caveat is about reading the record rather than about this row: 27 identical assertions from one source database are one projection, not 27 independent findings, and the count should not be mistaken for weight of evidence. Supporting Evidence: file:human/AGT/AGT-uniprot.txt SUBCELLULAR LOCATION: Secreted file:human/AGT/AGT-uniprot.txt TISSUE SPECIFICITY: Expressed by the liver and secreted in plasma. |
| GO:0005576 extracellular region | TAS Reactome:R-HSA-749448 | ACCEPT | Summary: One of 27 Reactome TAS rows on this gene, all asserting the same term - located_in extracellular region - and differing only in which reaction is cited. Together they are 24% of AGT's entire GOA record expressing a single correct claim once per reaction the protein or one of its peptides participates in. This row cites Reactome:R-HSA-749448, "Liganded Gq-activating GPCRs bind inactive heterotrimeric Gq". Reason: The term is correct: angiotensinogen and all of its peptide products act in the extracellular compartment, and this reaction takes place there. Accepted on that basis. The caveat is about reading the record rather than about this row: 27 identical assertions from one source database are one projection, not 27 independent findings, and the count should not be mistaken for weight of evidence. Supporting Evidence: file:human/AGT/AGT-uniprot.txt SUBCELLULAR LOCATION: Secreted file:human/AGT/AGT-uniprot.txt TISSUE SPECIFICITY: Expressed by the liver and secreted in plasma. |
| GO:0005576 extracellular region | TAS Reactome:R-HSA-749452 | ACCEPT | Summary: One of 27 Reactome TAS rows on this gene, all asserting the same term - located_in extracellular region - and differing only in which reaction is cited. Together they are 24% of AGT's entire GOA record expressing a single correct claim once per reaction the protein or one of its peptides participates in. This row cites Reactome:R-HSA-749452, "The Ligand:GPCR:Gq complex dissociates". Reason: The term is correct: angiotensinogen and all of its peptide products act in the extracellular compartment, and this reaction takes place there. Accepted on that basis. The caveat is about reading the record rather than about this row: 27 identical assertions from one source database are one projection, not 27 independent findings, and the count should not be mistaken for weight of evidence. Supporting Evidence: file:human/AGT/AGT-uniprot.txt SUBCELLULAR LOCATION: Secreted file:human/AGT/AGT-uniprot.txt TISSUE SPECIFICITY: Expressed by the liver and secreted in plasma. |
| GO:0005576 extracellular region | TAS Reactome:R-HSA-749454 | ACCEPT | Summary: One of 27 Reactome TAS rows on this gene, all asserting the same term - located_in extracellular region - and differing only in which reaction is cited. Together they are 24% of AGT's entire GOA record expressing a single correct claim once per reaction the protein or one of its peptides participates in. This row cites Reactome:R-HSA-749454, "The Ligand:GPCR:Gi complex dissociates". Reason: The term is correct: angiotensinogen and all of its peptide products act in the extracellular compartment, and this reaction takes place there. Accepted on that basis. The caveat is about reading the record rather than about this row: 27 identical assertions from one source database are one projection, not 27 independent findings, and the count should not be mistaken for weight of evidence. Supporting Evidence: file:human/AGT/AGT-uniprot.txt SUBCELLULAR LOCATION: Secreted file:human/AGT/AGT-uniprot.txt TISSUE SPECIFICITY: Expressed by the liver and secreted in plasma. |
| GO:0005576 extracellular region | TAS Reactome:R-HSA-749456 | ACCEPT | Summary: One of 27 Reactome TAS rows on this gene, all asserting the same term - located_in extracellular region - and differing only in which reaction is cited. Together they are 24% of AGT's entire GOA record expressing a single correct claim once per reaction the protein or one of its peptides participates in. This row cites Reactome:R-HSA-749456, "Liganded Gi-activating GPCRs bind inactive heterotrimeric G-protein Gi". Reason: The term is correct: angiotensinogen and all of its peptide products act in the extracellular compartment, and this reaction takes place there. Accepted on that basis. The caveat is about reading the record rather than about this row: 27 identical assertions from one source database are one projection, not 27 independent findings, and the count should not be mistaken for weight of evidence. Supporting Evidence: file:human/AGT/AGT-uniprot.txt SUBCELLULAR LOCATION: Secreted file:human/AGT/AGT-uniprot.txt TISSUE SPECIFICITY: Expressed by the liver and secreted in plasma. |
| GO:0005576 extracellular region | TAS Reactome:R-HSA-9615348 | ACCEPT | Summary: One of 27 Reactome TAS rows on this gene, all asserting the same term - located_in extracellular region - and differing only in which reaction is cited. Together they are 24% of AGT's entire GOA record expressing a single correct claim once per reaction the protein or one of its peptides participates in. This row cites Reactome:R-HSA-9615348, "AGT(25-32) binds to AGTR2". Reason: The term is correct: angiotensinogen and all of its peptide products act in the extracellular compartment, and this reaction takes place there. Accepted on that basis. The caveat is about reading the record rather than about this row: 27 identical assertions from one source database are one projection, not 27 independent findings, and the count should not be mistaken for weight of evidence. Supporting Evidence: file:human/AGT/AGT-uniprot.txt SUBCELLULAR LOCATION: Secreted file:human/AGT/AGT-uniprot.txt TISSUE SPECIFICITY: Expressed by the liver and secreted in plasma. |
| GO:0005576 extracellular region | TAS Reactome:R-HSA-9944473 | ACCEPT | Summary: One of 27 Reactome TAS rows on this gene, all asserting the same term - located_in extracellular region - and differing only in which reaction is cited. Together they are 24% of AGT's entire GOA record expressing a single correct claim once per reaction the protein or one of its peptides participates in. This row cites Reactome:R-HSA-9944473, "AGT(25-31) is decarboxylated to yield Alamandine". Reason: The term is correct: angiotensinogen and all of its peptide products act in the extracellular compartment, and this reaction takes place there. Accepted on that basis. The caveat is about reading the record rather than about this row: 27 identical assertions from one source database are one projection, not 27 independent findings, and the count should not be mistaken for weight of evidence. Supporting Evidence: file:human/AGT/AGT-uniprot.txt SUBCELLULAR LOCATION: Secreted file:human/AGT/AGT-uniprot.txt TISSUE SPECIFICITY: Expressed by the liver and secreted in plasma. |
| GO:0005576 extracellular region | TAS Reactome:R-HSA-9944540 | ACCEPT | Summary: One of 27 Reactome TAS rows on this gene, all asserting the same term - located_in extracellular region - and differing only in which reaction is cited. Together they are 24% of AGT's entire GOA record expressing a single correct claim once per reaction the protein or one of its peptides participates in. This row cites Reactome:R-HSA-9944540, "AGT(25-32) is decarboxylated to yield 1-Ala-AGT(25-32) (des(Asp1)-(Ala1)-Ang II, Angiotensin A)". Reason: The term is correct: angiotensinogen and all of its peptide products act in the extracellular compartment, and this reaction takes place there. Accepted on that basis. The caveat is about reading the record rather than about this row: 27 identical assertions from one source database are one projection, not 27 independent findings, and the count should not be mistaken for weight of evidence. Supporting Evidence: file:human/AGT/AGT-uniprot.txt SUBCELLULAR LOCATION: Secreted file:human/AGT/AGT-uniprot.txt TISSUE SPECIFICITY: Expressed by the liver and secreted in plasma. |
| GO:0005576 extracellular region | TAS Reactome:R-NUL-2022369 | ACCEPT | Summary: One of 27 Reactome TAS rows on this gene, all asserting the same term - located_in extracellular region - and differing only in which reaction is cited. Together they are 24% of AGT's entire GOA record expressing a single correct claim once per reaction the protein or one of its peptides participates in. This row cites Reactome:R-NUL-2022369, "ANPEP hydrolyzes Angiotensin III to Angiotensin IV". Reason: The term is correct: angiotensinogen and all of its peptide products act in the extracellular compartment, and this reaction takes place there. Accepted on that basis. The caveat is about reading the record rather than about this row: 27 identical assertions from one source database are one projection, not 27 independent findings, and the count should not be mistaken for weight of evidence. Supporting Evidence: file:human/AGT/AGT-uniprot.txt SUBCELLULAR LOCATION: Secreted file:human/AGT/AGT-uniprot.txt TISSUE SPECIFICITY: Expressed by the liver and secreted in plasma. |
| GO:0038166 angiotensin-activated signaling pathway | IDA PMID:18971559 Angiotensin II upregulates acyl-CoA:cholesterol acyltransfer... | ACCEPT | Summary: Angiotensin-activated signalling is the core biological process of this gene: angiotensin II is produced from angiotensinogen and from nothing else, and this term names the pathway it initiates. Here it is evidenced in primary human monocyte-macrophages, where added angiotensin II raises ACAT1 through AGTR1 and the effect is abolished by AT1 antagonists but not by an AT2 antagonist - a clean receptor-specific demonstration of the pathway. Reason: Core process, correctly evidenced with receptor-subtype controls. Supporting Evidence: PMID:18971559 Two-fold increases in ACAT1 protein expression and ACAT activity with Ang II treatment were completely inhibited by AT(1) receptor antagonists PMID:18971559 Application of an Ang II type 1 (AT(1)) receptor agonist (L162313), but not an Ang II type 2 (AT(2)) receptor agonist (CGP42112A), mimicked the effects of Ang II treatment in inducing ACAT1 protein expression. |
| GO:0090205 positive regulation of cholesterol metabolic process | IDA PMID:18971559 Angiotensin II upregulates acyl-CoA:cholesterol acyltransfer... | KEEP AS NON CORE | Summary: A downstream cellular response to added angiotensin II, correctly qualified. Angiotensin II raises ACAT1, which esterifies free cholesterol for storage in lipid droplets. AGT is not itself a participant in cholesterol metabolism; it is the source of a hormone whose receptor engagement changes the expression of a cholesterol enzyme. The 'acts_upstream_of' qualifier is what makes this defensible, and it is the qualifier several sibling rows from the same class of experiment should have used but did not. Reason: Real and reproducible, but two steps removed from anything angiotensinogen does: hormone -> receptor -> transcriptional change in an unrelated enzyme. Peripheral to the gene's function and correctly marked as upstream rather than participatory. Supporting Evidence: PMID:18971559 Ang II significantly increased ACAT1 protein expression in a time- or concentration-dependent manner. |
| GO:0001819 positive regulation of cytokine production | IDA PMID:17906677 Enhanced angiotensin converting enzyme 2 regulates the insul... | KEEP AS NON CORE | Summary: Real but indirect, and from a paper about a different gene. The study transfects ACE2 into an endothelial cell line and asks what that does to insulin signalling; the AGT-side observation is that angiotensin II and angiotensin IV induce macrophage migration inhibitory factor, which ACE2 transfer then suppresses. So the evidence for AGT is 'added peptide raises a cytokine in a cell line', which is genuine but sits at the far end of the hormone's action. Reason: Angiotensin II is genuinely pro-inflammatory and cytokine-inducing, so the biology is sound, but this is a cellular response to the hormone rather than a function of angiotensinogen, and the citation's subject is ACE2. Supporting Evidence: PMID:17906677 Gene transfer of ACE2 suppressed the expression of p22phox and MIF induced by angiotensin (Ang) II and Ang IV, accompanied by a decreased level of malondialdehyde in cells. |
| GO:0050729 positive regulation of inflammatory response | IDA PMID:17906677 Enhanced angiotensin converting enzyme 2 regulates the insul... | KEEP AS NON CORE | Summary: The inflammatory arm of the same experiment. Angiotensin II induces MIF, a pro-inflammatory cytokine, in the EAhy926 endothelial line; ACE2 over-expression, which degrades angiotensin II, reverses it. Duplicated by a TAS row for the same term from the PMID:17159080 editorial. Reason: A well-established consequence of angiotensin II signalling, but a response of the target cell rather than an activity of angiotensinogen, and evidenced here only in an immortalised endothelial line. Supporting Evidence: PMID:17906677 Gene transfer of ACE2 suppressed the expression of p22phox and MIF induced by angiotensin (Ang) II and Ang IV, accompanied by a decreased level of malondialdehyde in cells. |
| GO:2000379 positive regulation of reactive oxygen species metabolic process | IDA PMID:17906677 Enhanced angiotensin converting enzyme 2 regulates the insul... | KEEP AS NON CORE | Summary: The oxidative arm of the same experiment: angiotensin II induces p22phox, an NADPH oxidase subunit, raising malondialdehyde, and ACE2 transfer reverses it. Angiotensin II-driven NADPH oxidase activation is one of the best-replicated effects in vascular biology, so the direction is not in doubt; the reservation is the same as for the sibling rows. Reason: Real downstream redox signalling initiated by the hormone, not a molecular or cellular role of the precursor protein. Supporting Evidence: PMID:17906677 In addition, Ang II diminished insulin-stimulated phosphorylation of Akt (at Ser(473)) and eNOS (at Ser(1177)) and NO generation, effects which were reversed by ACE2 gene transfer and anti-MIF treatment in endothelial cells. |
| GO:0008083 growth factor activity | IDA PMID:10406457 Analysis of functional domains of angiotensin II type 2 rece... | MARK AS OVER ANNOTATED | Summary: Contradicted by the paper it cites. GO:0008083 is defined as 'The function that stimulates a cell to grow or proliferate'. PMID:10406457 is an AT2-receptor intracellular-loop mutagenesis study in PC12 cells, and everything it measures runs the other way: AT2-mediated apoptosis, inhibition of ERK, and SHP-1 activation. It contains no growth or proliferation assay. A second paper cited elsewhere in this same record independently reports that angiotensin II does not increase proliferation in its system. Reason: Angiotensin II is genuinely mitogenic for vascular smooth muscle through AGTR1, so the term is not absurd for the gene and I am not removing it. But the cited evidence is an apoptosis-and-ERK-inhibition study, which cannot support a growth factor activity, and no growth-factor assay on AGT or its peptides is cited anywhere in the record. GO's own comment on this term also suggests considering GO:0048018 receptor agonist activity, which would fit angiotensin II far better. Supporting Evidence: PMID:10406457 Deletion of residues 240-244 within the intermediate portion of the i3 loop resulted in a complete loss of AT2-mediated apoptosis, inhibition of extracellular signal-regulated kinases (ERK), and SHP-1 activation. PMID:10406457 Our data demonstrate that the intermediate portion of the i3 loop is important for AT2 function and that SHP-1 is a proximal effector of the AT2 receptor that is implicated in the inhibition of ERKs and in the apoptotic effect of this receptor. PMID:15652490 Moreover, Ang II induces a time- and dose-dependent augmentation in cell migration, but does not affect HUVEC proliferation. |
| GO:0038166 angiotensin-activated signaling pathway | IMP PMID:15153556 Angiotensin II increases Pax-2 expression in fetal kidney ce... | ACCEPT | Summary: Angiotensin-activated signalling is the core biological process of this gene: angiotensin II is produced from angiotensinogen and from nothing else, and this term names the pathway it initiates. Evidenced here in mouse embryonic renal cell lines, where added angiotensin II upregulates Pax-2 through AGTR2 and the effect is blocked by the AT2 antagonist PD123319 but not by the AT1 antagonist losartan. The receptor-subtype pharmacology is clean. The IMP code is loose - the perturbations are receptor antagonists and kinase inhibitors, not mutations in AGT - but the pathway claim holds. Reason: Core process, correctly identified, with receptor-specific antagonist controls. The evidence-code choice is questionable but the annotation is not. Supporting Evidence: PMID:15153556 Angiotensin II (AngII) upregulated Pax-2 protein and Pax-2 mRNA expression via the AngII type 2 (AT(2)) receptor in MK4 but not in MK3 cells. |
| GO:0038060 nitric oxide-cGMP-mediated signaling | TAS PMID:17159080 Cross-talk between angiotensin II receptor types 1 and 2: po... | KEEP AS NON CORE | Summary: Sourced from PMID:17159080, which PubMed types as Comment / Editorial / Review. It is a two-page commentary on another paper and the cached record contains no abstract text at all - only the citation line. It is the sole source for eleven GOA rows on this gene, three of them coded TAS and eight NAS. The commentary's subject is AGTR1/AGTR2 cross-talk in vascular remodelling, and the AT2 arm is conventionally described as signalling through bradykinin, nitric oxide and cGMP. That is a receptor-level statement about AGTR2, applied here to the ligand's precursor. Reason: The AT2/NO-cGMP arm is real and is the counter-regulatory limb of the pathway, so the term is not wrong. But it is a statement about receptor signalling taken from an editorial with no primary data, and it is peripheral to the gene's core role. Supporting Evidence: PMID:17906677 In addition, Ang II diminished insulin-stimulated phosphorylation of Akt (at Ser(473)) and eNOS (at Ser(1177)) and NO generation, effects which were reversed by ACE2 gene transfer and anti-MIF treatment in endothelial cells. |
| GO:1901203 positive regulation of extracellular matrix assembly | IDA PMID:30729664 Novel role of the clustered miR-23b-3p and miR-27b-3p in enh... | KEEP AS NON CORE | Summary: Three steps removed from angiotensinogen. In this atrial fibrosis model angiotensin II raises miR-23b-3p and miR-27b-3p in human atrial fibroblasts; those miRNAs target TGFBR3; loss of TGFBR3 activates Smad3; and it is over-expression of the miRNAs, not angiotensin II, that raises COL1A1, COL3A1 and ACTA2. The 'acts_upstream_of' qualifier is doing necessary work here. Reason: Real fibrotic biology downstream of angiotensin II and correctly qualified as upstream rather than participatory, but too indirect to be a function of the gene. Supporting Evidence: PMID:30729664 A cell model of atrial fibrosis was achieved in Ang-II-induced HAFs. PMID:30729664 Moreover, Smad3 was activated in HAFs in response to Ang-II treatment and inactivation of Smad3 attenuated up-regulation of miR-23b-3p and miR-27b-3p in Ang-II-treated HAFs. |
| GO:1902895 positive regulation of miRNA transcription | IMP PMID:30729664 Novel role of the clustered miR-23b-3p and miR-27b-3p in enh... | KEEP AS NON CORE | Summary: Directly measured: miR-23b-3p and miR-27b-3p rise in angiotensin II-treated human atrial fibroblasts, and Smad3 inactivation blocks that rise. This is the most immediate of the three rows from this paper. Reason: A real transcriptional response to the hormone in a primary human cell type, correctly qualified as upstream. Peripheral to the gene's function. The IMP code sits oddly on an experiment whose perturbation is peptide addition rather than a mutation in AGT. Supporting Evidence: PMID:30729664 We found that miR-23b-3p and miR-27b-3p were markedly increased in atrial appendage tissues of AF patients and in Ang-II-treated HAFs. |
| GO:1990776 response to angiotensin | IDA PMID:30729664 Novel role of the clustered miR-23b-3p and miR-27b-3p in enh... | KEEP AS NON CORE | Summary: Nearly tautological as written: angiotensinogen 'acts upstream of' the response to angiotensin. It is true - angiotensinogen is the sole source of the ligand, so it necessarily precedes any response to it - but it conveys no information beyond what GO:0038166 already states, and it would be equally true of every angiotensin experiment ever done. Reason: Not wrong, and the qualifier is the correct one, but uninformative: it restates the definitional relationship between the precursor and its own peptide. Kept because it is harmless and correctly qualified. Supporting Evidence: PMID:30729664 A cell model of atrial fibrosis was achieved in Ang-II-induced HAFs. |
| GO:0038166 angiotensin-activated signaling pathway | IDA PMID:15652490 Angiotensin II induces focal adhesion kinase/paxillin phosph... | ACCEPT | Summary: Angiotensin-activated signalling is the core biological process of this gene: angiotensin II is produced from angiotensinogen and from nothing else, and this term names the pathway it initiates. Here in primary human umbilical vein endothelial cells, where angiotensin II drives FAK and paxillin phosphorylation and migration through PI3K, Src and EGFR transactivation. Reason: Core process, in a primary human cell type, with pathway inhibitor dissection. Supporting Evidence: PMID:15652490 In the present study, we demonstrated that Ang II provokes a transitory enhancement of focal adhesion kinase (FAK) and paxillin phosphorylation in human umbilical endothelial cells (HUVEC). |
| GO:0051897 positive regulation of phosphatidylinositol 3-kinase/protein kinase B signal transduction | IDA PMID:15652490 Angiotensin II induces focal adhesion kinase/paxillin phosph... | KEEP AS NON CORE | Summary: A downstream signalling response to added angiotensin II. Wortmannin and LY294002 attenuate the FAK/paxillin phosphorylation, placing PI3K in the pathway. This is a property of the AGTR1-bearing endothelial cell, not of angiotensinogen. Reason: Real and well-controlled, but a generic intracellular signalling module engaged downstream of the receptor. Every GPCR agonist that transactivates a receptor tyrosine kinase would earn this term. Supporting Evidence: PMID:15652490 The effect of Ang II on FAK and paxillin phosphorylation was markedly attenuated in cells pretreated with wortmannin and LY294002, indicating that phosphoinositide 3-kinase (PI3K) plays an important role in regulating FAK activation. |
| GO:0005179 hormone activity | IDA PMID:1567413 Cloning and characterization of a human angiotensin II type ... | ACCEPT | Summary: Correct, with a caveat about what the annotation unit means. GO:0005179 requires a substance formed in one tissue and carried to another where it has a specific regulatory action through a receptor. That describes angiotensin II exactly, and does not describe the 476-residue precursor, which is inert until renin cleaves it. But GOA's annotation unit is the UniProt accession, angiotensin II has no gene of its own, and UniProt itself carries the peptide functions on P01019 as 'FUNCTION: [Angiotensin-2]' and as PRO_ chain-level interactions. So the hormone activity belongs here. This GOA row is exactly duplicated (it appears twice with identical term, evidence, reference and qualifier), which is why 114 GOA rows reconcile to 113 review entries. Reason: Core molecular function of the gene product, understood as including its cleavage products - the convention GOA and UniProt both follow. The evidence is thin for an IDA (the cited paper is a receptor-cloning study whose angiotensin II is synthetic peptide), but the claim is correct and is independently supported by the ISS from rat Agt and by the IC from receptor binding. Supporting Evidence: PMID:1567413 Ligand binding studies of the cloned receptor expressed in COS cells suggested that it is pharmacologically a type 1 angiotensin II receptor subtype. PMID:1567413 Electrophysiological studies of the receptor expressed in Xenopus laevis oocytes revealed that it could functionally couple to a second messenger system leading to the mobilization of intracellular stores of calcium. |
| GO:0038166 angiotensin-activated signaling pathway | IDA PMID:1567413 Cloning and characterization of a human angiotensin II type ... | ACCEPT | Summary: Angiotensin-activated signalling is the core biological process of this gene: angiotensin II is produced from angiotensinogen and from nothing else, and this term names the pathway it initiates. The cited receptor-cloning paper shows the human AT1 receptor coupling to calcium mobilisation on angiotensin II binding, which is the pathway this term names. Reason: Core process with direct receptor-coupling evidence. Supporting Evidence: PMID:1567413 Electrophysiological studies of the receptor expressed in Xenopus laevis oocytes revealed that it could functionally couple to a second messenger system leading to the mobilization of intracellular stores of calcium. PMID:1567413 Ligand binding studies of the cloned receptor expressed in COS cells suggested that it is pharmacologically a type 1 angiotensin II receptor subtype. |
| GO:0010718 positive regulation of epithelial to mesenchymal transition | ISS GO_REF:0000024 | KEEP AS NON CORE | Summary: Sequence-similarity transfer from mouse Agt (P11859), a true 1:1 orthologue, not a paralogue. I confirmed against QuickGO that the donor still carries this exact term with experimental evidence (IMP, PMID:34615811), so the transfer is not stale and the source is not itself an inference. Reason: Defensible orthologue transfer, but epithelial-mesenchymal transition is a context-specific fibrotic/oncogenic response to angiotensin signalling rather than a core function of the gene. Propagation Review Root cause: NO FAILURE NON CORE Sources checked: UniProtKB:P11859 · Agt (Mus musculus) SUPPORTS TRANSFER Mouse Agt, the 1:1 orthologue. Verified live in QuickGO to still hold GO:0010718 by IMP from PMID:34615811, so this is an experimentally grounded donor rather than a chain of inferences. Supporting Evidence: PMID:7989296 These mice do not produce angiotensinogen in the liver, resulting in the complete loss of plasma immunoreactive angiotensin I. |
| GO:0002034 maintenance of blood vessel diameter homeostasis by renin-angiotensin | IDA PMID:1567413 Cloning and characterization of a human angiotensin II type ... | ACCEPT | Summary: The gene's defining physiological role, stated at the right level of specificity. This is a renin-angiotensin-specific term rather than a generic vascular one, and the human genetics is unambiguous: loss of angiotensinogen in mice abolishes plasma angiotensin I and drops systolic pressure by a third, and re-expression restores it. Reason: Core biological process. Among the strongest claims in the whole record and one of the few where perturbation evidence exists in both directions. Supporting Evidence: PMID:7989296 The systolic blood pressure of the homozygous mutant mice was 66.9 +/- 4.1 mm Hg, significantly lower than that of wild-type mice (100.4 +/- 4.4 mm Hg). PMID:25691624 AAV-driven expression of both forms of AGT led to equivalent increases of systolic blood pressure and augmentation of atherosclerotic lesion size in hepAGT-/- mice. PMID:1567413 Ligand binding studies of the cloned receptor expressed in COS cells suggested that it is pharmacologically a type 1 angiotensin II receptor subtype. |
| GO:0005576 extracellular region | IDA PMID:4300938 Enzymatic degradation and electrophoresis of human angiotens... | ACCEPT | Summary: The oldest and, for the 'is_active_in' qualifier, the most apt localisation row: a 1968 study of the enzymatic degradation and electrophoresis of human angiotensin I, i.e. the peptide chemistry done on circulating material. UniProt cites this paper for both the signal peptide (residues 1-24) and the 'Secreted' subcellular location. PubMed holds no abstract for it, so the underlying experiment cannot be inspected here, but the conclusion is not in question. Reason: AGT is a liver-secreted plasma protein and the extracellular compartment is where it is cleaved by renin and where its peptides meet their receptors, so 'is_active_in' is the right qualifier. Corroborated by five HDA proteomics datasets, the curated UniProt location, the IBA and the ISS. Supporting Evidence: file:human/AGT/AGT-uniprot.txt SUBCELLULAR LOCATION: Secreted file:human/AGT/AGT-uniprot.txt TISSUE SPECIFICITY: Expressed by the liver and secreted in plasma. |
| GO:0010744 positive regulation of macrophage derived foam cell differentiation | IDA PMID:18971559 Angiotensin II upregulates acyl-CoA:cholesterol acyltransfer... | KEEP AS NON CORE | Summary: Same experiment, one step further downstream: raised ACAT1 activity drives cholesteryl-ester accumulation and foam-cell formation in cultured monocyte-macrophages loaded with acetylated LDL. This is atherosclerosis cell biology driven by an added peptide hormone, not a function of the precursor. Reason: Genuine downstream physiology of angiotensin II in macrophages and worth keeping, but well outside the gene's core role. Note the qualifier here is 'involved_in' while the cholesterol row from the same paper and the same experiment uses 'acts_upstream_of' - the two should agree. Supporting Evidence: PMID:18971559 Ang II significantly increased ACAT1 protein expression in a time- or concentration-dependent manner. |
| GO:0031012 extracellular matrix | HDA PMID:25037231 Extracellular matrix signatures of human primary metastatic ... | MARK AS OVER ANNOTATED | Summary: High-throughput mass spectrometry of extracellular matrix preparations from human primary and metastatic colon cancers. Angiotensinogen is among the most abundant proteins in plasma, and plasma proteins are the standard background of any tissue-derived ECM proteome. Nothing in the study distinguishes matrix-incorporated angiotensinogen from blood carried in the tissue, and no matrix function has ever been reported for it. Reason: Detection is not localisation of function. A specific structural compartment term is being assigned to an abundant soluble plasma protein on the basis of its presence in a tissue proteome. The general term GO:0005576, which the record already carries many times over, is what this observation supports. Supporting Evidence: file:human/AGT/AGT-uniprot.txt SUBCELLULAR LOCATION: Secreted file:human/AGT/AGT-uniprot.txt TISSUE SPECIFICITY: Expressed by the liver and secreted in plasma. |
| GO:0005576 extracellular region | HDA PMID:27068509 Extracellular matrix remodelling in response to venous hyper... | ACCEPT | Summary: High-throughput proteomics detection of angiotensinogen in a study of extracellular matrix remodelling in venous hypertension. The compartment assigned is the correct general one for a plasma protein. Reason: Correct term. Detection of an abundant plasma protein in a tissue proteome is weak evidence in itself, but 'extracellular region' is broad enough to be right whatever the source of the material. Supporting Evidence: file:human/AGT/AGT-uniprot.txt SUBCELLULAR LOCATION: Secreted |
| GO:0043407 negative regulation of MAP kinase activity | IMP PMID:28283184 Angiotensin-(1-7) regulates angiotensin II-induced matrix me... | KEEP AS NON CORE | Summary: Correct, and about a genuinely different peptide. The negative regulation comes from angiotensin-(1-7), an AGT-derived peptide acting through MAS1, which suppresses the p38 MAPK activation that angiotensin II induces in vascular smooth muscle cells. So angiotensinogen is the source of peptides that both activate and inhibit MAP kinase signalling, and this row captures the counter-regulatory limb. Reason: Real peptide pharmacology of the Ang-(1-7)/MAS1 axis, worth retaining because it is the only row in the record representing that limb. Non-core because it is one downstream signalling consequence among many. The IMP code is loose - the perturbations are peptide agonists and the MAS1 antagonist A779, not AGT mutations. Supporting Evidence: PMID:28283184 Furthermore, Ang II induced p38 MAPK activation, and this was inhibited by the treatment of Ang-(1-7). |
| GO:0005515 protein binding | IPI PMID:16237761 Screening of hepatocyte proteins binding to F protein of hep... | MARK AS OVER ANNOTATED | Summary: Could not be verified against the cited paper. The row asserts that AGT binds the hepatitis C virus F/ARFP protein (P0C045), from a yeast two-hybrid screen of a liver cDNA library, and was assigned by AgBase. The cached full text of that paper is complete - abstract, introduction, results and the whole discussion, in which every hit is discussed in turn - and the word 'angiotensinogen' does not appear anywhere in it. That is weaker evidence than it first appears, and the limit is worth stating plainly: the Results identify only 25 of the 36 positive colonies, and the Discussion names a handful more, so roughly a third of the colonies are never identified anywhere in the text. Angiotensinogen therefore cannot be positively excluded from the screen; it simply cannot be found in it. The serpin the paper does report is C1 inhibitor. Independently, the assay is a poor fit: AGT is a secreted, signal-peptide-cleaved, disulfide-bonded, four-site N-glycosylated plasma protein and yeast two-hybrid reconstitutes a transcription factor in the yeast nucleus, while UniProt places the F protein in host cytoplasm. Reason: Bare 'protein binding' with no functional follow-up, a compartment mismatch, and a citation in which I could not locate the claim despite having the complete text. Not removed, because a table that did not survive text extraction cannot be excluded and I will not overrule an experimental annotation on a cache I cannot fully guarantee. Raised as a question for the assigning group instead. Supporting Evidence: PMID:16237761 Thirty-six colonies were selected and sequenced. Among them, 11 colonies were zymogen granule protein, 5 colonies were zinc finger protein, 4 colonies were zinc-alpha-2-glycoprotein, 1 colony was sialyltransferase, 1 colony was complement control protein factor I, 1 colony was vitronectin, and 2 colonies were new genes with unknown function. file:human/AGT/AGT-uniprot.txt SUBCELLULAR LOCATION: Secreted |
| GO:1903779 regulation of cardiac conduction | IGI PMID:17416596 Angiotensin II-induced sudden arrhythmic death and electrica... | KEEP AS NON CORE | Summary: From the dTGR model: rats carrying BOTH the human renin and the human angiotensinogen transgenes. Human renin cleaves human angiotensinogen efficiently and rat angiotensinogen poorly, so the phenotype requires the two transgenes together - which is precisely what IGI with UniProtKB:P00797 is for, and an unusually well-justified use of the code. Ventricular tachycardia was inducible in 88% of untreated dTGR versus 33% of losartan-treated dTGR and none of the controls, with prolonged and inhomogeneous depolarisation and repolarisation on cardiac magnetic field mapping. Reason: A genuine in vivo consequence of sustained angiotensin II exposure, but arrhythmic remodelling is downstream target-organ damage rather than a core function of angiotensinogen. Propagation Review Root cause: NO FAILURE NON CORE Sources checked: UniProtKB:P00797 · REN (Homo sapiens) SUPPORTS TRANSFER Human renin; required alongside the human AGT transgene to generate the angiotensin II exposure that produces the electrical remodelling. Supporting Evidence: PMID:17416596 Rats harboring the human renin and angiotensinogen genes (dTGR) feature angiotensin (ANG) II/hypertension-induced cardiac damage and die suddenly between wk 7 and 8. PMID:17416596 Untreated dTGR show electrical remodeling and probably die from VT. |
| GO:0008217 regulation of blood pressure | IGI PMID:17416596 Angiotensin II-induced sudden arrhythmic death and electrica... | ACCEPT | Summary: From the dTGR model: rats carrying BOTH the human renin and the human angiotensinogen transgenes. Human renin cleaves human angiotensinogen efficiently and rat angiotensinogen poorly, so the phenotype requires the two transgenes together - which is precisely what IGI with UniProtKB:P00797 is for, and an unusually well-justified use of the code. Blood pressure rises progressively in dTGR alongside cardiac hypertrophy, and losartan reverses the whole syndrome, tying it to AGTR1. This is the gene's central physiological output measured in vivo with the human genes. Reason: Core biological process, demonstrated in vivo with the human gene products and with receptor-blockade rescue. One of the best-supported rows in the record. Propagation Review Root cause: NO FAILURE CORE Sources checked: UniProtKB:P00797 · REN (Homo sapiens) SUPPORTS TRANSFER Human renin, the obligate genetic partner in the dTGR model. Correct IGI partner: neither transgene produces the phenotype alone, because human renin is species-selective for human angiotensinogen. Supporting Evidence: PMID:17416596 Rats harboring the human renin and angiotensinogen genes (dTGR) feature angiotensin (ANG) II/hypertension-induced cardiac damage and die suddenly between wk 7 and 8. PMID:7989296 The systolic blood pressure of the homozygous mutant mice was 66.9 +/- 4.1 mm Hg, significantly lower than that of wild-type mice (100.4 +/- 4.4 mm Hg). |
| GO:1901201 regulation of extracellular matrix assembly | IGI PMID:17416596 Angiotensin II-induced sudden arrhythmic death and electrica... | MODIFY | Summary: From the dTGR model: rats carrying BOTH the human renin and the human angiotensinogen transgenes. Human renin cleaves human angiotensinogen efficiently and rat angiotensinogen poorly, so the phenotype requires the two transgenes together - which is precisely what IGI with UniProtKB:P00797 is for, and an unusually well-justified use of the code. dTGR show increased perivascular and interstitial fibrosis and raised connective tissue growth factor from week 5 onwards, progressing with time and ameliorated by losartan. The term used is the neutral parent, but the paper determines the direction. Reason: The sign is determined by the data - matrix accumulation increases - so the neutral parent understates what was shown. GO:1901203 positive regulation of extracellular matrix assembly is the accurate term and is already used elsewhere in this record for the same direction of effect. Propagation Review Root cause: TERM SCOPING PROBLEM Failure modes: GRANULARITY MISMATCH Sources checked: UniProtKB:P00797 · REN (Homo sapiens) SUPPORTS TRANSFER Human renin; correct IGI partner. The problem is the term's granularity, not the genetic interaction. Proposed replacements: positive regulation of extracellular matrix assembly Supporting Evidence: PMID:17416596 Already by wk 5, untreated dTGR showed increased perivascular and interstitial fibrosis, connective tissue growth factor expression, and monocyte infiltration compared with SD rats, differences that progressed through time. |
| GO:1903598 positive regulation of gap junction assembly | IGI PMID:17416596 Angiotensin II-induced sudden arrhythmic death and electrica... | MODIFY | Summary: From the dTGR model: rats carrying BOTH the human renin and the human angiotensinogen transgenes. Human renin cleaves human angiotensinogen efficiently and rat angiotensinogen poorly, so the phenotype requires the two transgenes together - which is precisely what IGI with UniProtKB:P00797 is for, and an unusually well-justified use of the code. This row has the sign backwards. GOA asserts POSITIVE regulation of gap junction assembly, but what the paper measured is that left-ventricular connexin 43 - the ventricular gap-junction protein - was significantly REDUCED in dTGR relative to both losartan-treated dTGR and normal controls. More angiotensinogen-derived angiotensin II means less connexin 43, and that gap-junction loss is part of the electrical remodelling that makes these animals arrhythmic. One caveat that does not change the verdict: what was measured is left-ventricular connexin 43 mRNA abundance, not gap-junction assembly itself, so the term is a step removed from the readout in both its original and its corrected form. The direction, which is what this row gets wrong, is unambiguous either way. Reason: Regulatory sign inversion, verifiable against the abstract of the cited paper. The correct term is GO:1903597 negative regulation of gap junction assembly. The genetic interaction and the reference are both right; only the direction is wrong. Propagation Review Root cause: TERM SCOPING PROBLEM Failure modes: REGULATORY SIGN INVERSION Sources checked: UniProtKB:P00797 · REN (Homo sapiens) SUPPORTS TRANSFER Human renin; the IGI partner is correct and the model is the right one. The defect is entirely in the direction of the term chosen from it. Proposed replacements: negative regulation of gap junction assembly Supporting Evidence: PMID:17416596 Left-ventricular mRNA expression of potassium channel subunit Kv4.3 and gap-junction protein connexin 43 were significantly reduced in dTGR compared with Los-treated dTGR and SD. PMID:17416596 Rats harboring the human renin and angiotensinogen genes (dTGR) feature angiotensin (ANG) II/hypertension-induced cardiac damage and die suddenly between wk 7 and 8. |
| GO:0070062 extracellular exosome | HDA PMID:23533145 In-depth proteomic analyses of exosomes isolated from expres... | MARK AS OVER ANNOTATED | Summary: High-throughput proteomics of exosomes isolated from expressed prostatic secretions. Abundant plasma and body-fluid proteins co-purify with extracellular vesicles by density and by adsorption, and this term is assigned to hundreds of such proteins on exactly this basis. No exosomal sorting signal, no vesicle-associated function and no follow-up exists for angiotensinogen. Reason: A specific vesicle compartment assigned from co-purification with an abundant soluble protein. Reproducible but uncharacterised, which is what this action is for. Supporting Evidence: file:human/AGT/AGT-uniprot.txt SUBCELLULAR LOCATION: Secreted |
| GO:0005576 extracellular region | HDA PMID:16502470 Human colostrum: identification of minor proteins in the aqu... | ACCEPT | Summary: High-throughput proteomics detection in the aqueous phase of human colostrum. Angiotensinogen is a normal constituent of extracellular fluids, so the general compartment term is correct. Reason: Correct term from a weak but unobjectionable observation; 'extracellular region' cannot be wrong for a secreted plasma protein found in a body fluid. Supporting Evidence: file:human/AGT/AGT-uniprot.txt SUBCELLULAR LOCATION: Secreted |
| GO:0072562 blood microparticle | HDA PMID:22516433 Proteomic analysis of microvesicles from plasma of healthy d... | MARK AS OVER ANNOTATED | Summary: High-throughput proteomics of microvesicles from the plasma of healthy donors. The same co-purification argument applies, and rather more forcefully here: the source material is plasma, where angiotensinogen is present at microgram-per-millilitre concentrations, so its appearance in any plasma-derived vesicle preparation is expected regardless of whether it is vesicle-associated. Reason: Assigns a specific particle compartment on the basis of abundance in the starting material. The soluble plasma pool, already captured by GO:0005576, is the parsimonious explanation. Supporting Evidence: file:human/AGT/AGT-uniprot.txt SUBCELLULAR LOCATION: Secreted file:human/AGT/AGT-uniprot.txt TISSUE SPECIFICITY: Expressed by the liver and secreted in plasma. |
| GO:0003014 renal system process | IDA PMID:21183621 Protein microarrays discover angiotensinogen and PRKRIP1 as ... | MARK AS OVER ANNOTATED | Summary: The evidence is an autoantibody biomarker screen, not a renal-process experiment. Protein microarrays found raised anti-angiotensinogen autoantibody titres in the serum of patients with chronic kidney disease. The authors are explicit that they cannot say what the antibodies bind, cannot establish pathogenicity, and suspect the titres reflect kidney damage rather than cause it. Nothing in the paper measures a renal system process attributable to angiotensinogen. Reason: An IDA whose 'direct assay' is an antibody titre in patient sera. The gene's genuine renal biology is already carried, and far better, by GO:0002019 regulation of renal output by angiotensin, GO:0035813 regulation of renal sodium excretion, and the renal tubular dysgenesis genetics. Marked rather than removed, in keeping with not overruling an experimental annotation outright, and raised as a question for the assigning group. Supporting Evidence: PMID:21183621 We cannot distinguish whether these auto-Ab are targeting angiotensinogen or angiotensin I, as cross-reactivity in antibodies against these two antigens has been described ( 33 ). PMID:21183621 Significant elevations in the titer of novel auto-Ab were noted against angiotensinogen and PRKRIP1 in renal insufficiency. |
| GO:2001238 positive regulation of extrinsic apoptotic signaling pathway | IDA PMID:10406457 Analysis of functional domains of angiotensin II type 2 rece... | KEEP AS NON CORE | Summary: Supported, and the specific term the growth-factor row should have been. The same PC12 study maps AT2 residues Lys240, Asn242 and Ser243 as required for AT2-induced apoptosis and shows SHP-1 is the proximal effector. Angiotensin II acting through AGTR2 is genuinely pro-apoptotic. Reason: Real peptide pharmacology, but demonstrated in a transfected PC12 neuroendocrine line with mutant receptors, and peripheral to the gene's cardiovascular core. Supporting Evidence: PMID:10406457 Deletion of residues 240-244 within the intermediate portion of the i3 loop resulted in a complete loss of AT2-mediated apoptosis, inhibition of extracellular signal-regulated kinases (ERK), and SHP-1 activation. PMID:10406457 Our data demonstrate that the intermediate portion of the i3 loop is important for AT2 function and that SHP-1 is a proximal effector of the AT2 receptor that is implicated in the inhibition of ERKs and in the apoptotic effect of this receptor. |
| GO:0035813 regulation of renal sodium excretion | NAS PMID:17159080 Cross-talk between angiotensin II receptor types 1 and 2: po... | ACCEPT | Summary: Sourced from PMID:17159080, which PubMed types as Comment / Editorial / Review. It is a two-page commentary on another paper and the cached record contains no abstract text at all - only the citation line. It is the sole source for eleven GOA rows on this gene, three of them coded TAS and eight NAS. Sodium handling is where the pathway's blood-pressure effect is finally executed, via direct proximal tubular action and via aldosterone. Reason: Core physiology of the pathway, accepted on the biology with the citation flagged. Supporting Evidence: file:human/AGT/AGT-uniprot.txt potent regulator of blood pressure, body fluid and electrolyte PMID:7989296 The systolic blood pressure of the homozygous mutant mice was 66.9 +/- 4.1 mm Hg, significantly lower than that of wild-type mice (100.4 +/- 4.4 mm Hg). |
| GO:0045893 positive regulation of DNA-templated transcription | IDA PMID:15153556 Angiotensin II increases Pax-2 expression in fetal kidney ce... | MARK AS OVER ANNOTATED | Summary: A generic term that casts a hormone as a transcriptional regulator. What the paper shows is that added angiotensin II raises Pax-2 mRNA and protein in mouse MK4 cells through AGTR2 and JAK2/STAT signalling. Angiotensinogen does not regulate transcription; it is the source of a ligand whose receptor engages a signalling cascade that ends in changed transcription. The record already carries GO:0038166, which states that chain correctly and is annotated from this very paper. Reason: 'Positive regulation of DNA-templated transcription' is the terminal readout of almost any signalling experiment and is uninformative here. It also mis-assigns the role: the substrate that supplies a receptor agonist is not the agent of the transcriptional change. GO:0038166 from the same experiment carries the real content. Supporting Evidence: PMID:15153556 Moreover, embryonic kidney explants in culture confirmed that AngII upregulates Pax-2 gene expression via the AT(2) receptor. |
| GO:0045893 positive regulation of DNA-templated transcription | IDA PMID:18607644 Deficiency of intrarenal angiotensin II type 2 receptor impa... | MARK AS OVER ANNOTATED | Summary: The same generic transcription term from the companion study, which uses AT2-receptor knockout mice and shows that angiotensin II raises Pax-2 and N-myc in wild-type embryonic kidney but not in AT2R-null kidney. Good genetics for the receptor; still the wrong role assignment for angiotensinogen. Reason: Same objection as the sibling row: an uninformative terminal-readout term that represents the hormone's precursor as a transcriptional regulator. The specific and defensible content of this experiment is the ureteric-bud branching row. Supporting Evidence: PMID:18607644 In ex vivo studies, Ang II stimulated Pax-2 and N-myc mRNA expression in embryonic kidneys of wild-type mice, but this stimulatory effect was absent in embryonic kidneys of AT(2)R KO mice. |
| GO:0090190 positive regulation of branching involved in ureteric bud morphogenesis | IDA PMID:18607644 Deficiency of intrarenal angiotensin II type 2 receptor impa... | KEEP AS NON CORE | Summary: Specific, developmentally meaningful, and properly controlled by receptor genetics: AT2R-null embryonic kidneys have smaller glomeruli and reduced Pax-2 and N-myc, and do not respond to angiotensin II. This is the defensible annotation from the paper, and it aligns with the human loss-of-function phenotype (renal tubular dysgenesis). Reason: A real developmental role of angiotensin signalling in nephrogenesis, but downstream developmental physiology rather than the gene's core function, and demonstrated in mouse explants. Supporting Evidence: PMID:18607644 Glomerular size was significantly smaller, and Pax-2 and N-myc expression down-regulated, in kidneys of AT(2)R KO mice compared with those of wild-type mice. |
| GO:0005576 extracellular region | IDA PMID:18971559 Angiotensin II upregulates acyl-CoA:cholesterol acyltransfer... | ACCEPT | Summary: The term is right but this reference does not establish it. PMID:18971559 is a study in which synthetic angiotensin II was added to cultured human monocyte-macrophages and ACAT1 expression measured; it contains no localisation experiment on angiotensinogen. AGT's extracellular location is abundantly evidenced elsewhere in this record - by direct plasma protein chemistry (PMID:4300938), by five independent HDA proteomics datasets (two of which assign GO:0005576 itself and three a more specific extracellular compartment), by the curated UniProt SUBCELLULAR LOCATION, and by the IBA and ISS rows. Reason: The term is correct and is core for this gene, so the action follows the biology rather than the citation: extracellular localisation is accepted here as it is on every other row for this term. The defect is in the evidence attribution alone - this reference contains no localisation experiment - and it is recorded in the reference_review for PMID:18971559 rather than by downgrading a correct term. Supporting Evidence: file:human/AGT/AGT-uniprot.txt SUBCELLULAR LOCATION: Secreted |
| GO:0034374 low-density lipoprotein particle remodeling | NAS PMID:17069818 Angiotensin II type 1-receptor antagonism prevents type IIA ... | MARK AS OVER ANNOTATED | Summary: Two steps removed and a non-traceable author statement. The cited work shows that angiotensin II acting through AGTR1 raises secretory phospholipase A2 type IIA in rat aortic smooth muscle cells, and that sPLA2-IIA is what modifies and peroxidises LDL - an effect abolished by the sPLA2 inhibitor LY311727. So the LDL remodelling is the enzyme's activity, with angiotensin II two steps upstream of it. Reason: Role conflation: the term names what sPLA2-IIA does to LDL particles, and assigns it to the hormone precursor that induces the enzyme. The paper's own inhibitor control shows the activity is the enzyme's. Supporting Evidence: PMID:17069818 Stimulation of rat aortic smooth muscle cells with ANG II (10(-7) mol/L) enhanced sPLA2-IIA protein expression, activity as well as LDL-peroxidation, determined by western blot, activity assay and malondialdehyde (MDA)-assay and diene formation, respectively, and were blunted by AT1-receptor blockade (Losartan, 10(-5) mol/L). |
| GO:0005179 hormone activity | ISS GO_REF:0000024 | ACCEPT | Summary: Sequence-similarity transfer from rat Agt (P01015), a 1:1 orthologue. I confirmed against QuickGO that the donor still holds this term with experimental evidence, so the source is neither stale nor itself inferred. Donor evidence: IDA from PMID:8348686. Corroborates the two direct IDA rows and the IC row for the same term. Reason: Core molecular function, independently supported on the orthologue by direct assay. Together with the human IDA rows this is the best-evidenced molecular function in the record. Propagation Review Root cause: NO FAILURE CORE Sources checked: UniProtKB:P01015 · Agt (Rattus norvegicus) SUPPORTS TRANSFER Rat Agt, 1:1 orthologue; holds GO:0005179 by IDA from PMID:8348686 in the current QuickGO record. Supporting Evidence: PMID:1567413 Ligand binding studies of the cloned receptor expressed in COS cells suggested that it is pharmacologically a type 1 angiotensin II receptor subtype. |
| GO:0005576 extracellular region | ISS GO_REF:0000024 | ACCEPT | Summary: Sequence-similarity transfer from rat Agt (P01015), a 1:1 orthologue. I confirmed against QuickGO that the donor still holds this term with experimental evidence, so the source is neither stale nor itself inferred. Donor evidence: two independent IDA rows (PMID:12242043, PMID:8252633) plus IBA and ISO. Rat angiotensinogen is, like the human protein, a liver-secreted plasma protein. Reason: Correct location, well grounded on the orthologue and directly evidenced on the human protein as well. Propagation Review Root cause: NO FAILURE CORE Sources checked: UniProtKB:P01015 · Agt (Rattus norvegicus) SUPPORTS TRANSFER Rat Agt; holds GO:0005576 by IDA from PMID:12242043 and PMID:8252633 in the current QuickGO record. Supporting Evidence: file:human/AGT/AGT-uniprot.txt SUBCELLULAR LOCATION: Secreted file:human/AGT/AGT-uniprot.txt TISSUE SPECIFICITY: Expressed by the liver and secreted in plasma. |
| GO:0010595 positive regulation of endothelial cell migration | IDA PMID:15652490 Angiotensin II induces focal adhesion kinase/paxillin phosph... | KEEP AS NON CORE | Summary: Directly measured in the cited paper: angiotensin II increases HUVEC migration in a time- and dose-dependent way, through PI3K, Src and EGFR. The same sentence records that proliferation was unaffected, which is a useful negative given that the record elsewhere claims growth factor activity. Reason: A real, directly measured cellular response to the hormone in primary human endothelial cells, but downstream physiology rather than a core role of the gene. Supporting Evidence: PMID:15652490 Moreover, Ang II induces a time- and dose-dependent augmentation in cell migration, but does not affect HUVEC proliferation. |
| GO:0010613 positive regulation of cardiac muscle hypertrophy | ISS GO_REF:0000024 | KEEP AS NON CORE | Summary: Sequence-similarity transfer from rat Agt (P01015), a 1:1 orthologue. I confirmed against QuickGO that the donor still holds this term with experimental evidence, so the source is neither stale nor itself inferred. Donor evidence: IDA from PMID:8348686. Angiotensin II-driven cardiac hypertrophy is one of the best-replicated effects of the pathway and the dTGR rows in this same record show the same biology in vivo. Reason: Real and well replicated, but a pathological/adaptive tissue response to sustained angiotensin II rather than a core function of angiotensinogen. Propagation Review Root cause: NO FAILURE NON CORE Sources checked: UniProtKB:P01015 · Agt (Rattus norvegicus) SUPPORTS TRANSFER Rat Agt; holds GO:0010613 by IDA from PMID:8348686 in the current QuickGO record. Experimentally grounded, not an inference chain. Supporting Evidence: PMID:17416596 Rats harboring the human renin and angiotensinogen genes (dTGR) feature angiotensin (ANG) II/hypertension-induced cardiac damage and die suddenly between wk 7 and 8. |
| GO:0014873 response to muscle activity involved in regulation of muscle adaptation | ISS GO_REF:0000024 | KEEP AS NON CORE | Summary: Sequence-similarity transfer from rat Agt (P01015), a 1:1 orthologue. I confirmed against QuickGO that the donor still holds this term with experimental evidence, so the source is neither stale nor itself inferred. Donor evidence: IMP from PMID:8252633. A narrow, physiology-specific term about muscle adaptation to activity. Reason: Defensible orthologue transfer of a specific rat physiological observation, but peripheral to the gene's function and not independently established in human. Propagation Review Root cause: NO FAILURE NON CORE Sources checked: UniProtKB:P01015 · Agt (Rattus norvegicus) SUPPORTS TRANSFER Rat Agt; holds GO:0014873 by IMP from PMID:8252633 in the current QuickGO record. Supporting Evidence: file:human/AGT/AGT-uniprot.txt TISSUE SPECIFICITY: Expressed by the liver and secreted in plasma. |
| GO:0045742 positive regulation of epidermal growth factor receptor signaling pathway | IDA PMID:15652490 Angiotensin II induces focal adhesion kinase/paxillin phosph... | KEEP AS NON CORE | Summary: EGFR transactivation downstream of AGTR1 - a genuine and much-replicated phenomenon, demonstrated here by AG1478 blocking angiotensin II-induced FAK and paxillin phosphorylation in HUVEC. Reason: Real receptor cross-talk, but a property of the responding cell's signalling architecture rather than a function of the hormone's precursor. Supporting Evidence: PMID:15652490 Furthermore, FAK and paxillin phosphorylation was markedly blocked after treatment of HUVEC with AG1478, a selective inhibitor of epidermal growth factor receptor (EGFR) phosphorylation. |
| GO:0048146 positive regulation of fibroblast proliferation | ISS GO_REF:0000024 | KEEP AS NON CORE | Summary: Sequence-similarity transfer from rat Agt (P01015), a 1:1 orthologue. I confirmed against QuickGO that the donor still holds this term with experimental evidence, so the source is neither stale nor itself inferred. Donor evidence: IDA from PMID:8348686, the same rat study behind the hormone-activity and cardiac-hypertrophy transfers. Angiotensin II is a well-documented fibroblast mitogen. Reason: Real proliferative effect of the hormone on fibroblasts, and consistent with the fibrosis seen in the dTGR model, but a downstream tissue response rather than a core function. Worth noting that three of the six ISS rows trace to this single donor paper, so they are not three independent findings. Propagation Review Root cause: NO FAILURE NON CORE Sources checked: UniProtKB:P01015 · Agt (Rattus norvegicus) SUPPORTS TRANSFER Rat Agt; holds GO:0048146 by IDA from PMID:8348686 in the current QuickGO record. Same donor study as the GO:0005179 and GO:0010613 transfers. Supporting Evidence: PMID:17416596 Rats harboring the human renin and angiotensinogen genes (dTGR) feature angiotensin (ANG) II/hypertension-induced cardiac damage and die suddenly between wk 7 and 8. |
| GO:0001558 regulation of cell growth | NAS PMID:17159080 Cross-talk between angiotensin II receptor types 1 and 2: po... | MARK AS OVER ANNOTATED | Summary: Sourced from PMID:17159080, which PubMed types as Comment / Editorial / Review. It is a two-page commentary on another paper and the cached record contains no abstract text at all - only the citation line. It is the sole source for eleven GOA rows on this gene, three of them coded TAS and eight NAS. 'Regulation of cell growth' is about as uninformative as a biological process term can be, and the source contains no data. The record's one attempt to say something specific in this area - GO:0008083 growth factor activity - is contradicted by its own citation, and a paper cited elsewhere here reports that angiotensin II does not affect proliferation in HUVEC. Reason: An uninformative parent term with a non-traceable author statement from an editorial as its only support. It conveys nothing that the specific growth-related rows do not, and those are themselves poorly supported. Supporting Evidence: PMID:15652490 Moreover, Ang II induces a time- and dose-dependent augmentation in cell migration, but does not affect HUVEC proliferation. |
| GO:0002016 regulation of blood volume by renin-angiotensin | NAS PMID:17159080 Cross-talk between angiotensin II receptor types 1 and 2: po... | ACCEPT | Summary: Sourced from PMID:17159080, which PubMed types as Comment / Editorial / Review. It is a two-page commentary on another paper and the cached record contains no abstract text at all - only the citation line. It is the sole source for eleven GOA rows on this gene, three of them coded TAS and eight NAS. The term is renin-angiotensin-specific and is straightforwardly correct: angiotensin II raises blood volume through aldosterone-driven sodium retention, vasopressin release and thirst. Reason: Core physiology of the pathway, accepted on the biology. The citation is weak and is flagged in reference_review, but the claim itself is not in doubt and is corroborated by the knockout and rescue data. Supporting Evidence: file:human/AGT/AGT-uniprot.txt potent regulator of blood pressure, body fluid and electrolyte PMID:7989296 The systolic blood pressure of the homozygous mutant mice was 66.9 +/- 4.1 mm Hg, significantly lower than that of wild-type mice (100.4 +/- 4.4 mm Hg). |
| GO:0002018 renin-angiotensin regulation of aldosterone production | NAS PMID:17159080 Cross-talk between angiotensin II receptor types 1 and 2: po... | ACCEPT | Summary: Sourced from PMID:17159080, which PubMed types as Comment / Editorial / Review. It is a two-page commentary on another paper and the cached record contains no abstract text at all - only the citation line. It is the sole source for eleven GOA rows on this gene, three of them coded TAS and eight NAS. Angiotensin II stimulating adrenal zona glomerulosa cells to make and release aldosterone is the canonical endocrine output of this pathway, and UniProt states it directly in the FUNCTION line for the angiotensin-2 chain. Reason: Core physiology, and the one term here that names the pathway's endocrine arm. Accepted on the biology despite the weak citation. Supporting Evidence: file:human/AGT/AGT-uniprot.txt potent regulator of blood pressure, body fluid and electrolyte PMID:7989296 The systolic blood pressure of the homozygous mutant mice was 66.9 +/- 4.1 mm Hg, significantly lower than that of wild-type mice (100.4 +/- 4.4 mm Hg). |
| GO:0002019 regulation of renal output by angiotensin | NAS PMID:17159080 Cross-talk between angiotensin II receptor types 1 and 2: po... | ACCEPT | Summary: Sourced from PMID:17159080, which PubMed types as Comment / Editorial / Review. It is a two-page commentary on another paper and the cached record contains no abstract text at all - only the citation line. It is the sole source for eleven GOA rows on this gene, three of them coded TAS and eight NAS. Another pathway-specific term. Angiotensin II constricts the efferent arteriole, alters glomerular filtration and promotes proximal sodium reabsorption; loss of the pathway in humans produces fetal anuria. Reason: Core renal physiology of the pathway, independently corroborated by the renal tubular dysgenesis genetics in this same record. Supporting Evidence: file:human/AGT/AGT-uniprot.txt potent regulator of blood pressure, body fluid and electrolyte PMID:7989296 The systolic blood pressure of the homozygous mutant mice was 66.9 +/- 4.1 mm Hg, significantly lower than that of wild-type mice (100.4 +/- 4.4 mm Hg). |
| GO:0005179 hormone activity | IC PMID:1378723 Cloning, expression, and characterization of a gene encoding... | ACCEPT | Summary: Curator inference from GO:0031702 type 1 angiotensin receptor binding, which AGT holds by IPI from this same reference. The inference is sound - a molecule that binds a cell-surface angiotensin receptor with high affinity and triggers inositol phosphate turnover is acting as a hormone. It is however entirely redundant: AGT already holds GO:0005179 twice by IDA and once by ISS from rat, all of which are stronger. Reason: Accepted: hormone activity is a core molecular function of this gene product and the inference behind this row is valid. Its redundancy is a property of the evidence route rather than of the term, so it is recorded in propagation_review as EVIDENCE_CIRCULAR_OR_REDUNDANT instead of by giving a core term a non-core action. Propagation Review Root cause: EVIDENCE CIRCULAR OR REDUNDANT Sources checked: GO:0031702 CIRCULAR OR REDUNDANT The IC is drawn from another GO term on the same gene from the same reference. The inference itself is sound; it simply restates, at lower evidential strength, a conclusion the target already holds by IDA and ISS. Supporting Evidence: PMID:1378723 The expressed gene exhibited high-affinity AII and Dup753 binding and was functionally coupled to inositol phosphate turnover. |
| GO:0005576 extracellular region | IC PMID:1378723 Cloning, expression, and characterization of a gene encoding... | ACCEPT | Summary: Curator inference from GO:0031702 type 1 angiotensin receptor binding, which AGT holds by IPI from this same reference. Equally sound and equally redundant: binding a cell-surface receptor implies being extracellular. The record already carries GO:0005576 thirty-six times, including by direct assay, two HDA proteomics datasets, IBA, ISS and the curated UniProt location. Reason: Accepted: the localisation is correct and core, and the inference is sound. As with the companion hormone-activity IC row, the redundancy is a property of the evidence route and is recorded in propagation_review, not in the action. Propagation Review Root cause: EVIDENCE CIRCULAR OR REDUNDANT Sources checked: GO:0031702 CIRCULAR OR REDUNDANT Same-gene, same-reference term-to-term inference. Valid reasoning, but the target already has direct experimental and curated evidence for this location. Supporting Evidence: file:human/AGT/AGT-uniprot.txt SUBCELLULAR LOCATION: Secreted |
| GO:0007199 G protein-coupled receptor signaling pathway coupled to cGMP nucleotide second messenger | TAS PMID:17159080 Cross-talk between angiotensin II receptor types 1 and 2: po... | KEEP AS NON CORE | Summary: Sourced from PMID:17159080, which PubMed types as Comment / Editorial / Review. It is a two-page commentary on another paper and the cached record contains no abstract text at all - only the citation line. It is the sole source for eleven GOA rows on this gene, three of them coded TAS and eight NAS. Asserts that angiotensin signalling proceeds through a cGMP-coupled GPCR pathway. AGTR1 is a Gq/PLC receptor and AGTR2 reaches cGMP indirectly, through bradykinin and nitric oxide rather than by direct cyclase coupling, so the term overstates the directness of the link. Reason: Not wrong in outcome - the AT2 limb does raise cGMP - but the coupling it implies is indirect, and the only source is an editorial I cannot read. Kept at low confidence rather than removed, since I cannot inspect the text that supports it. Supporting Evidence: PMID:10406457 Our data demonstrate that the intermediate portion of the i3 loop is important for AT2 function and that SHP-1 is a proximal effector of the AT2 receptor that is implicated in the inhibition of ERKs and in the apoptotic effect of this receptor. |
| GO:0007200 phospholipase C-activating G protein-coupled receptor signaling pathway | NAS PMID:17159080 Cross-talk between angiotensin II receptor types 1 and 2: po... | ACCEPT | Summary: Sourced from PMID:17159080, which PubMed types as Comment / Editorial / Review. It is a two-page commentary on another paper and the cached record contains no abstract text at all - only the citation line. It is the sole source for eleven GOA rows on this gene, three of them coded TAS and eight NAS. This one names the canonical mechanism correctly: AGTR1 is a Gq/11-coupled receptor and angiotensin II binding activates phospholipase C. The cited AT1 cloning papers in this same record demonstrate it directly, with inositol phosphate turnover and calcium mobilisation. Reason: Mechanistically informative and independently demonstrated by two receptor-cloning papers already cited on this gene. The evidence code and source are weak but the annotation is the textbook mechanism. Supporting Evidence: PMID:1378723 The expressed gene exhibited high-affinity AII and Dup753 binding and was functionally coupled to inositol phosphate turnover. PMID:1567413 Electrophysiological studies of the receptor expressed in Xenopus laevis oocytes revealed that it could functionally couple to a second messenger system leading to the mobilization of intracellular stores of calcium. |
| GO:0019229 regulation of vasoconstriction | NAS PMID:17159080 Cross-talk between angiotensin II receptor types 1 and 2: po... | ACCEPT | Summary: Sourced from PMID:17159080, which PubMed types as Comment / Editorial / Review. It is a two-page commentary on another paper and the cached record contains no abstract text at all - only the citation line. It is the sole source for eleven GOA rows on this gene, three of them coded TAS and eight NAS. Angiotensin II is the archetypal endogenous vasoconstrictor; UniProt keywords the entry 'Vasoactive' and 'Vasoconstrictor'. Reason: Core function. The weak citation does not put the claim in question. Supporting Evidence: file:human/AGT/AGT-uniprot.txt potent regulator of blood pressure, body fluid and electrolyte PMID:7989296 The systolic blood pressure of the homozygous mutant mice was 66.9 +/- 4.1 mm Hg, significantly lower than that of wild-type mice (100.4 +/- 4.4 mm Hg). |
| GO:0042127 regulation of cell population proliferation | NAS PMID:17159080 Cross-talk between angiotensin II receptor types 1 and 2: po... | MARK AS OVER ANNOTATED | Summary: Sourced from PMID:17159080, which PubMed types as Comment / Editorial / Review. It is a two-page commentary on another paper and the cached record contains no abstract text at all - only the citation line. It is the sole source for eleven GOA rows on this gene, three of them coded TAS and eight NAS. Same objection as 'regulation of cell growth': a top-level regulatory term with no direction, no cell type and no primary data behind it. The specific proliferation claims in this record are the rat-orthologue fibroblast transfer (which is kept) and the growth-factor row (which is contradicted). Reason: Generic, direction-free, and sourced from a commentary. Adds nothing over the specific rows and dilutes the record. Supporting Evidence: PMID:15652490 Moreover, Ang II induces a time- and dose-dependent augmentation in cell migration, but does not affect HUVEC proliferation. |
| GO:0050729 positive regulation of inflammatory response | TAS PMID:17159080 Cross-talk between angiotensin II receptor types 1 and 2: po... | KEEP AS NON CORE | Summary: Sourced from PMID:17159080, which PubMed types as Comment / Editorial / Review. It is a two-page commentary on another paper and the cached record contains no abstract text at all - only the citation line. It is the sole source for eleven GOA rows on this gene, three of them coded TAS and eight NAS. Duplicates the IDA row for the same term from PMID:17906677, where angiotensin II is shown to induce MIF in endothelial cells. Reason: Real biology, but redundant with the IDA row for the identical term and sourced from a commentary with no data of its own. Supporting Evidence: PMID:17906677 Gene transfer of ACE2 suppressed the expression of p22phox and MIF induced by angiotensin (Ang) II and Ang IV, accompanied by a decreased level of malondialdehyde in cells. |
| GO:0001974 blood vessel remodeling | TAS PMID:10406457 Analysis of functional domains of angiotensin II type 2 rece... | MARK AS OVER ANNOTATED | Summary: Term plausible for the gene, citation does not support it. PMID:10406457 is an AT2-receptor i3-loop deletion study in PC12 cells measuring apoptosis, ERK and SHP-1. It contains no vessel, no vascular smooth muscle and no remodelling measurement. Angiotensin II certainly does drive vascular remodelling, but not in this paper. Reason: A TAS whose traceable source contains nothing about blood vessels. Retained rather than removed because the biology is real and well documented elsewhere, but this row carries no usable evidence. Supporting Evidence: PMID:10406457 Deletion of residues 240-244 within the intermediate portion of the i3 loop resulted in a complete loss of AT2-mediated apoptosis, inhibition of extracellular signal-regulated kinases (ERK), and SHP-1 activation. |
| GO:0002034 maintenance of blood vessel diameter homeostasis by renin-angiotensin | TAS PMID:10406457 Analysis of functional domains of angiotensin II type 2 rece... | ACCEPT | Summary: Correct term, wrong citation. Maintenance of blood vessel diameter by the renin-angiotensin system is exactly what angiotensinogen exists for, and the same term is independently annotated by IDA from PMID:1567413. This particular TAS points at the PC12 AT2 apoptosis paper, which measures nothing of the kind. Reason: Accepted on the biology: this is a core function of angiotensinogen and it is independently evidenced on this gene by IDA. Core-versus-non-core is a judgement about the term, not about the citation, so a weak citation is not a reason to downgrade a correct core term. The citation problem - a PC12-cell AT2 apoptosis study standing behind a vascular-diameter term - is recorded in the reference_review for PMID:10406457. Supporting Evidence: PMID:1567413 Ligand binding studies of the cloned receptor expressed in COS cells suggested that it is pharmacologically a type 1 angiotensin II receptor subtype. |
| GO:0007186 G protein-coupled receptor signaling pathway | IDA PMID:10406457 Analysis of functional domains of angiotensin II type 2 rece... | MODIFY | Summary: True but uninformative parent. Angiotensin receptors are GPCRs, so any angiotensin signalling is GPCR signalling - but AGT already carries the specific child GO:0038166 angiotensin-activated signaling pathway five times over, which says the same thing and names the ligand. Reason: Too general. The specific term is available, is already in use on this gene, and is what the evidence actually shows; the bare parent adds nothing and dilutes the record. Proposed replacements: angiotensin-activated signaling pathway Supporting Evidence: PMID:10406457 Deletion of residues 240-244 within the intermediate portion of the i3 loop resulted in a complete loss of AT2-mediated apoptosis, inhibition of extracellular signal-regulated kinases (ERK), and SHP-1 activation. |
| GO:0031703 type 2 angiotensin receptor binding | IPI PMID:10406457 Analysis of functional domains of angiotensin II type 2 rece... | ACCEPT | Summary: Correct and informative. Angiotensin II binds AGTR2 (P50052); UniProt encodes the interaction precisely, on the angiotensin chain rather than the whole precursor (PRO_0000032459 with AGTR2). This and the AGTR1 row are the two molecular functions in the record that name a real, specific partner. Reason: Core molecular function of the gene product's principal peptide, with a named partner and a specific term. Supporting Evidence: PMID:10406457 Deletion of residues 240-244 within the intermediate portion of the i3 loop resulted in a complete loss of AT2-mediated apoptosis, inhibition of extracellular signal-regulated kinases (ERK), and SHP-1 activation. file:human/AGT/AGT-uniprot.txt PRO_0000032459; P50052: AGTR2; NbExp=2 |
| GO:0051387 negative regulation of neurotrophin TRK receptor signaling pathway | IDA PMID:10406457 Analysis of functional domains of angiotensin II type 2 rece... | KEEP AS NON CORE | Summary: Plausible and consistent with the cited data, in a specialised context. PC12 cells are the classic NGF/TrkA model, and this study shows AT2 engagement activates SHP-1 and inhibits ERK - the pathway TrkA signals through. So angiotensin II acting on AT2 does dampen neurotrophin receptor signalling in these cells. Reason: A genuine but narrow cross-talk effect, shown in one transfected cell line, far from the gene's core cardiovascular role. Supporting Evidence: PMID:10406457 Our data demonstrate that the intermediate portion of the i3 loop is important for AT2 function and that SHP-1 is a proximal effector of the AT2 receptor that is implicated in the inhibition of ERKs and in the apoptotic effect of this receptor. |
| GO:0001822 kidney development | IMP PMID:16116425 Mutations in genes in the renin-angiotensin system are assoc... | KEEP AS NON CORE | Summary: The strongest human loss-of-function evidence in the record. Biallelic mutations in AGT - alongside REN, ACE and AGTR1 - cause autosomal recessive renal tubular dysgenesis, with absent proximal tubules, fetal anuria and perinatal death. The authors' own mechanism is haemodynamic rather than a direct developmental instruction: chronic low perfusion pressure of the fetal kidney as a consequence of an inactive renin-angiotensin system. Reason: Unambiguous human genetics, kept without hesitation. Marked non-core rather than core because the paper's own reading makes kidney development a downstream consequence of losing the pathway's haemodynamic output, not a separate developmental activity of angiotensinogen. The core role that produces it is already annotated as GO:0003081 and GO:0002034. Supporting Evidence: PMID:16116425 We studied 11 individuals with renal tubular dysgenesis, belonging to nine families, and found that they had homozygous or compound heterozygous mutations in the genes encoding renin, angiotensinogen, angiotensin converting enzyme or angiotensin II receptor type 1. PMID:16116425 We propose that renal lesions and early anuria result from chronic low perfusion pressure of the fetal kidney, a consequence of renin-angiotensin system inactivity. |
| GO:0031702 type 1 angiotensin receptor binding | IPI PMID:1378723 Cloning, expression, and characterization of a gene encoding... | ACCEPT | Summary: Correct and informative, the AGTR1 counterpart. The cited paper clones the human AT1a receptor and shows high-affinity angiotensin II binding coupled to inositol phosphate turnover. UniProt again attaches the interaction to the angiotensin-2 chain (PRO_0000032458 with AGTR1) rather than to full-length angiotensinogen, which is the correct reading: the binding entity is the octapeptide. Reason: Core molecular function. AGTR1 is the receptor through which essentially all of the pressor, aldosterone-releasing and proliferative actions of the pathway run. Supporting Evidence: PMID:1378723 The expressed gene exhibited high-affinity AII and Dup753 binding and was functionally coupled to inositol phosphate turnover. file:human/AGT/AGT-uniprot.txt PRO_0000032458; P30556: AGTR1; NbExp=2 |
| GO:0005576 extracellular region | NAS PMID:14718574 The human plasma proteome: a nonredundant list developed by ... | ACCEPT | Summary: A non-traceable author statement drawn from the 2004 consolidated human plasma proteome list, which merged four methodologies into 1175 non-redundant gene products. It is a catalogue entry rather than an experiment, but angiotensinogen is one of the least controversial members of that catalogue. Reason: Correct localisation from a weak evidence type. Adds breadth of corroboration rather than new evidence. Supporting Evidence: PMID:14718574 We have merged four different views of the human plasma proteome, based on different methodologies, into a single nonredundant list of 1175 distinct gene products. file:human/AGT/AGT-uniprot.txt SUBCELLULAR LOCATION: Secreted |
| GO:0004867 serine-type endopeptidase inhibitor activity | TAS PMID:3397061 Molecular cloning of the mouse angiotensinogen gene. | REMOVE | Summary: The origin of the serpin-inhibitor claim, and its own wording shows it was never an activity measurement. This is a 1988 paper reporting the cloning and sequencing of the MOUSE angiotensinogen gene. Its statement about inhibition is a sequence alignment: because angiotensinogen is homologous to serpins, the authors aligned the putative reactive centre across species. The same abstract then notes that the human site differs from the rodent one and that the significance of that divergence was unknown. No inhibition assay, no protease, no kinetics - a homology observation recorded as a traceable author statement and thereafter read as a molecular function. Reason: The claim is false and its source does not even assert it. Angiotensinogen does not inhibit serine endopeptidases: it has no reactive bond, a proline-blocked hinge at P12, MEROPS non-inhibitor classification I04.953, and a crystal structure of loop-cleaved protein showing it cannot undergo the stressed-to-relaxed transition that the inhibitory mechanism requires. Its protease partner renin is an aspartyl protease and AGT is its substrate. Removing this, the IBA and the IEA together eliminates the claim from the record; no experimental annotation is being overruled, because none exists. Supporting Evidence: PMID:3397061 Because angiotensinogen is homologous to other members of the serine protease inhibitor family, we aligned the putative reactive center of angiotensinogens from various species. PMID:3397061 This alignment shows that the inhibitor site in human angiotensinogen is different from its rodent counterpart, but the role of this sequence divergence in the pathogenesis of human disease remains to be established. PMID:20927107 angiotensinogen-a non-inhibitory member of the serpin family of protease inhibitors PMID:30563843 In contrast, it has been shown that AGT has lost the ability to undergo this typical serpin S-to-R transition (29), confirmed here by our structure of loop-cleaved AGT, so it was very puzzling why the serpin framework was selected in the course of evolution as an angiotensin carrier. |
| GO:0007267 cell-cell signaling | TAS PMID:8513325 A molecular variant of angiotensinogen associated with preec... | MARK AS OVER ANNOTATED | Summary: A generic term from a genetic association study. PMID:8513325 is the classic report linking the T235 angiotensinogen variant to preeclampsia in Caucasian and Japanese cohorts. It is a case-control association, containing no signalling experiment. The structural work also argues that this variant acts by raising angiotensinogen concentration rather than by changing what the protein does. Reason: 'Cell-cell signalling' is an uninformative high-level term, and the traceable source is an allele-frequency study. The specific and correct version of this claim, GO:0038166 angiotensin-activated signaling pathway, is already annotated five times on this gene from papers that actually measured signalling. Supporting Evidence: PMID:8513325 In a series of Caucasian women with pregnancy-induced hypertension, we have observed a significant association of preeclampsia with a molecular variant of angiotensinogen, T235, found previously to be associated with essential hypertension. PMID:20927107 This strengthens previous deductions5,6 that the predisposition to hypertension results from the small increase in concentration of the polymorphic angiotensinogen rather than a change in its function. |
Loading supporting content…
Download this section (compressed HTML)Q: For the GO editorial office: is there a way to make a prohormone's substrate role machine-readable without annotating the precursor to the processing term? GO:0002003 angiotensin maturation is_a GO:0016486 peptide hormone processing and so sits under proteolysis, which is why angiotensinogen is correctly absent from it - and equally why INS is absent from GO:0030070, POMC, NPPA, GCG and PENK from GO:0016486, and APP from GO:0034205. The GO-CAM for this pathway (6246724f00000549) does capture it, as UniProtKB:P01019 molecule instances feeding the REN and ACE activities, but nothing equivalent survives in the GAF. Would has_input extensions on the proteases' GO:0002003 rows be the intended answer, or is substrate-side representation deliberately out of scope for standard annotation? The concrete instance in this review is the GO:0005515 -> GO:0002020 protease binding MODIFY on the renin interaction: that is an MF on the substrate side of an enzyme-substrate pair, justified here by a 670 A2 body-to-body interface and specificity determinants outside renin's catalytic cleft rather than by the cleavage itself, and it is the annotation a decision on this question would settle.
Q: For the PAINT curators of PTHR11461: would an IRD at the angiotensinogen node block GO:0004867 the way the existing IRD at PTN002606963 blocks GO:0005576 for the intracellular serpin subclade? Angiotensinogen is the one non-inhibitory member of the SERPINA clade that node PTN008970140 covers, and the same leak will reach every angiotensinogen orthologue in the family, not just the human gene.
Q: For InterPro: should the interpro2go mapping IPR000215 (Serpin_fam) -> GO:0004867 exist at all? The family contains many non-inhibitory members - corticosteroid-binding globulin, thyroxine-binding globulin, PEDF, ovalbumin, angiotensinogen - and MEROPS already distinguishes them by reserving the I04.9xx range. The gene-level angiotensinogen signatures IPR000227 and IPR033834 produce a correct annotation through the same pipeline.
Q: For BHF-UCL: GO:1903598 positive regulation of gap junction assembly is annotated to AGT by IGI from PMID:17416596, but that paper reports connexin 43 significantly REDUCED in the double-transgenic rats. Should this be GO:1903597, negative regulation of gap junction assembly?
Q: For AgBase: can the AGT-HCV F protein interaction annotated from PMID:16237761 be traced? Angiotensinogen is not mentioned anywhere in the retrievable full text, and the serpin the paper does report is C1 inhibitor - but the Results identify only 25 of the 36 positive colonies, so the claim may rest on one of the unnamed ones. A pointer to the supporting colony would settle it either way.
Q: For UniProt and GOA jointly: the pregnancy-specific disulfide-linked 2:2 heterotetramer of angiotensinogen with pro-PRG2, and the probable 2:2:2 complex with pro-PRG2 and C3dg, are curated in UniProt from PMID:20927107 and PMID:7539791 but appear nowhere in the GO record. What is the barrier - a missing complex term, or simply that no one has curated it?
Q: Would GOA accept annotations against the PRO chain identifiers that already exist for this entry? Roughly forty rows on AGT describe the pharmacology of an octapeptide, and UniProt already keys the receptor interactions to PRO_0000032458 and PRO_0000032459 rather than to P01019.
Q: Is the eleven-row dependence of this gene's record on PMID:17159080, a two-page editorial with no primary data and no abstract, intentional? Three of those rows are coded TAS, which implies traceability to a primary source that the commentary itself would have to supply, and the other eight are NAS.
Experiment: Test directly whether angiotensinogen inhibits any serine endopeptidase. Incubate purified plasma-derived and recombinant glycosylated angiotensinogen with a panel of serine proteases spanning the targets of its SERPINA clade-mates - neutrophil elastase and proteinase 3 for SERPINA1, thrombin and activated protein C for SERPINA5, cathepsin G for SERPINA3 - and look for both residual activity loss and, critically, the SDS-stable covalent acyl-enzyme complex that is diagnostic of the serpin mechanism. The prediction from the structure and from MEROPS I04.953 is that no complex forms with any of them. This is the experiment that would convert the GO:0004867 removal from an argument into a measurement, and its absence from the literature is why the claim survived for nearly forty years.
Experiment: Discriminate the redox switch's physiology from its biochemistry. The in vitro effect requires cell-surface renin bound to the prorenin receptor, while the in vivo test used whole-animal plasma angiotensin II and blood pressure, both strongly buffered by renin feedback. Cross the Cys18Ser/Cys137Ser knock-in onto a background with tissue-specific prorenin receptor readouts and measure local angiotensin generation in kidney and vascular wall, rather than circulating levels; add a pregnancy challenge, since the human observation that motivated the model is pre-eclampsia.
Experiment: Ask what the serpin body is for. Use the renin-complex structure to design angiotensinogen variants that retain the N-terminal tail and the renin-binding surface but disrupt distinct patches elsewhere on the molecule, express them as correctly glycosylated secreted protein, and test each for renin cleavage kinetics (to confirm the presentation function is intact) and then for binding partners, cellular uptake and plasma half-life. Any patch whose disruption changes clearance or partner binding without changing cleavage would be the first evidence of a second function for the 400 residues C-terminal to the angiotensin sequence.
Experiment: Replace the ten yeast two-hybrid interactions with something interpretable. Pull down native angiotensinogen from human plasma with a monoclonal antibody raised against the body of the molecule rather than the angiotensin tail, and identify co-eluting proteins by mass spectrometry, in plasma from non-pregnant donors and from pregnant donors with and without pre-eclampsia. The positive control is built in: the pro-PRG2/C3dg complex should appear in the pregnancy samples and not the others, which both validates the method and would supply the curation that the GO record currently lacks.
What is not known — curated, literature-grounded statements of the open unknowns (the inverse of core functions).
Gap: Whether the Cys42-Cys162 redox switch has any physiological role is unresolved, and the two decisive experiments disagree. The bridge is Cys18-Cys138 in human mature numbering (UniProt DISULFID 42..162 on a chain beginning at residue 25) and Cys18-Cys137 in the mouse, which is the numbering used by the knock-in study below; the two names denote the same conserved bond.
OPEN BIOLOGY RESIDUAL_SUBGAP
What is known: Firmly established: the disulfide exists and is the only cysteine pair conserved across species; it is labile, with plasma holding a near 40:60 reduced:oxidised mixture; the oxidised form is preferentially cleaved by receptor-bound renin, giving a 4-fold increase in angiotensin release in vitro; and the oxidised form is elevated in the maternal circulation in pre-eclampsia. Equally firmly established: removing the bridge in mice changes nothing measurable.
Significance: This is the only proposed mechanism by which angiotensinogen itself, rather than renin abundance, could regulate the cascade's output, and it is the stated rationale for antioxidant approaches to pre-eclampsia. If real it would justify a molecular function on the precursor; if not, angiotensinogen is a concentration variable and nothing more. No GO term is proposed here for exactly this reason.
What would resolve it: The in vitro and in vivo experiments differ in more than species: the in vitro effect requires the prorenin receptor and cell-surface renin, while the mouse experiment measured whole-animal plasma angiotensin II, blood pressure and lesion size, which are strongly buffered by renin feedback. A tissue-level or prorenin-receptor-dependent readout in the Cys-to-Ser mice, or a pregnancy challenge, would discriminate.
Provenance (the field's own admissions):
Gap: Why the angiotensin peptides are carried on a serpin scaffold at all is unknown. The fold is retained but the mechanism that defines the superfamily has been lost, and no function has been assigned to the retained body of the molecule beyond providing the renin-binding surface.
OPEN BIOLOGY MF_DARK
What is known: Established: angiotensinogen retains the typical serpin fold at only about 22% identity to its closest serpin relatives; it has no functional reactive centre and loop cleavage does not trigger the stressed-to-relaxed transition; and the related non-inhibitory serpins SERPINA6 and SERPINA7 do use that transition to release their bound hormones, which angiotensinogen cannot. The 400-odd residues C-terminal to the angiotensin sequence have no assigned activity.
Significance: The authors of the structural work call this puzzling in print. If the serpin body has a function beyond presenting the tail to renin - a second ligand, a clearance receptor, a protein partner - it would be the first molecular activity of the precursor beyond protease binding, and would change what should be annotated on this gene.
What would resolve it: Structure-guided mutagenesis separating the renin-presentation surface from the rest of the serpin body, combined with an unbiased interaction screen using correctly folded, glycosylated, secreted angiotensinogen rather than a yeast two-hybrid system.
Provenance (the field's own admissions):
Gap: Angiotensinogen has no curated protein interaction partner outside the renin-angiotensin cascade that is supported by anything better than a large-scale yeast two-hybrid screen.
OPEN BIOLOGYCURATION MF_DARK
What is known: Established: the renin interaction is structurally solved, and UniProt records a disulfide-linked 2:2 heterotetramer with the proform of PRG2 and a probable 2:2:2 complex with pro-PRG2 and C3dg during pregnancy. Beyond those, the sixteen IPI rows - fourteen of them bare GO:0005515, two of them receptor-binding - resolve to the two angiotensin receptors, renin, a Drosophila enzyme co-crystallised with the peptide, a hepatitis C virus protein, and the ten partners of one interactome screen. Only renin and the Drosophila enzyme are themselves extracellular, and only renin has functional follow-up.
Significance: The pregnancy heterotetramer with pro-PRG2 and C3dg is a genuine, curated complex that appears nowhere in this gene's GO record - no complex term, no interaction row - while ten unvalidated yeast two-hybrid pairings do appear. The record's interaction content is therefore close to inverted with respect to evidence quality.
What would resolve it: Curation of the pro-PRG2/C3dg complex from PMID:20927107 and PMID:7539791, and an interaction screen using native or correctly processed secreted angiotensinogen from plasma.
Provenance (the field's own admissions):
Gap: GO cannot express the distinction between a precursor protein and the peptides cleaved from it, so every function of angiotensin I, II, III, IV and (1-7) is recorded indistinguishably on the 476-residue precursor.
OPEN ONTOLOGYCURATION MF_DARK
What is known: Established, and handled better elsewhere: UniProt separates the layers explicitly, with eight PEPTIDE features, per-chain FUNCTION blocks for angiotensin-2, angiotensin-3 and angiotensin-(1-7), and IntAct entries keyed to PRO_ chain identifiers rather than to P01019. Reactome likewise names its participants AGT(25-32), AGT(25-31) and so on. GOA has only the accession, so roughly forty rows on this gene describe the peptides while a handful describe the protein, with nothing to tell them apart.
Significance: This is why the record looks the way it does. A reader cannot tell from GOA that 'hormone activity' and 'type 1 angiotensin receptor binding' belong to an octapeptide while 'protease binding' belongs to the precursor, nor that the precursor has essentially no annotated activity of its own. The same problem affects every prohormone and polyprotein.
What would resolve it: Annotation against PRO identifiers, which already exist for these chains (PRO_0000032456 for the chain, PRO_0000032457-032459 and PRO_0000420659-420663 for the eight angiotensin peptides) and which UniProt already uses for the receptor interactions. No new ontology term is needed; what is needed is for GOA to accept the finer-grained entity.
Provenance (the field's own admissions):
Loading supporting content…
Download this section (compressed HTML)Loading supporting content…
Download this section (compressed HTML)Loading supporting content…
Download this section (compressed HTML)Loading supporting content…
Download this section (compressed HTML)