{"gene":"GNPAT","run_date":"2026-06-10T01:55:21","timeline":{"discoveries":[{"year":1991,"finding":"GNPAT (DHAP-AT) is a peroxisomal enzyme with ~80% latency in selectively permeabilized fibroblasts. Its latency is critically dependent on reduced glutathione (GSH) in the medium, and ATP stimulates latent DHAP-AT activity through a mechanism involving cations but not a proton gradient. The integrity of peroxisomal membrane thiol-groups is essential for maintaining DHAP-AT latency.","method":"Selective permeabilization of human skin fibroblasts, enzyme activity assays with defined medium components (GSH/GSSG manipulation, ATP, cation chelation)","journal":"Biochimica et biophysica acta","confidence":"Medium","confidence_rationale":"Tier 2 / Moderate — direct biochemical assay in permeabilized cells with multiple medium component manipulations, single lab","pmids":["1657193"],"is_preprint":false},{"year":1993,"finding":"Isolated deficiency of the peroxisomal enzyme DHAP-AT (GNPAT) causes rhizomelic chondrodysplasia punctata with deficient de novo plasmalogen biosynthesis, establishing GNPAT as essential for the plasmalogen synthesis pathway in humans, with all other peroxisomal functions normal.","method":"Enzyme activity measurement in patient fibroblasts, clinical phenotype characterization, peroxisomal function panel","journal":"Archives of disease in childhood","confidence":"Medium","confidence_rationale":"Tier 2 / Moderate — direct enzyme activity assays in patient cells with comprehensive peroxisomal function panel ruling out other deficiencies, single lab","pmids":["8466247"],"is_preprint":false},{"year":1995,"finding":"GNPAT (DHAPAT) enzymatic activity can be reliably measured in chorionic villous samples, blood cells, cultured skin fibroblasts, cultured chorionic villus fibroblasts, and cultured amniocytes, establishing the enzyme's presence and activity across multiple human cell types and its utility for prenatal diagnosis.","method":"Enzyme activity assay in multiple human cell types and tissue samples","journal":"Journal of inherited metabolic disease","confidence":"Medium","confidence_rationale":"Tier 2 / Moderate — direct enzyme activity measurement across multiple cell types, single lab, single method","pmids":["9053559"],"is_preprint":false},{"year":2000,"finding":"A missense mutation in GNPAT (R211H) causes isolated DHAPAT deficiency, resulting in deficient de novo plasmalogen synthesis in patient fibroblasts while alkyl-DHAP synthase activity, VLCFA profile, phytanic acid concentration, and peroxisomal thiolase remain normal. The deficiency leads to abnormal myelin formation, linking GNPAT-dependent plasmalogen biosynthesis to myelin integrity.","method":"Enzyme activity assays in patient fibroblasts, mutation identification by cDNA sequencing, MRI of white matter","journal":"Developmental medicine and child neurology","confidence":"Medium","confidence_rationale":"Tier 2 / Moderate — direct mutation identification with functional enzyme activity assay in patient cells, orthogonal MRI evidence, single lab","pmids":["10972423"],"is_preprint":false},{"year":2001,"finding":"The human GNPAT gene spans ~28 kb on chromosome 1q42.12-43, consists of 16 exons and 15 introns, and nine distinct mutations (missense, deletions, insertion, splice site mutations) in GNPAT cause DHAPAT enzyme deficiency and impaired etherphospholipid biosynthesis. This confirmed GNPAT as the sole gene responsible for RCDP type 2.","method":"Genomic sequencing, exon mapping, patient mutation analysis across 12 GNPAT-deficient patients, splice site and transcript analysis","journal":"Biochemical and biophysical research communications","confidence":"Medium","confidence_rationale":"Tier 2 / Moderate — comprehensive genomic characterization with mutation-phenotype correlation in 12 patients, single lab, multiple mutation types","pmids":["11237722"],"is_preprint":false},{"year":2011,"finding":"GNPAT and AGPS physically partner on the luminal peroxisomal membrane surface; the presence of AGPS is required for full GNPAT activity, and full GNPAT activity also depends on intact substrate channeling from GNPAT to AGPS. Milder RCDP phenotypes correlate with residual GNPAT or AGPS protein function.","method":"Cell line protein quantification (GNPAT and AGPS levels compared across patient and control lines), protein modeling for structural consequences, transcript analysis, functional enzyme assays in patient fibroblasts","journal":"Human mutation","confidence":"Medium","confidence_rationale":"Tier 2 / Moderate — multiple patient cell lines, protein quantification, transcript analysis, and functional assays; single lab but orthogonal methods","pmids":["21990100"],"is_preprint":false},{"year":2015,"finding":"siRNA-mediated knockdown of GNPAT in HepG2/C3A human liver-derived cells resulted in a >17-fold decrease in hepcidin (HAMP) mRNA expression, indicating that GNPAT participates in hepcidin regulation in hepatocytes.","method":"siRNA knockdown of GNPAT in HepG2/C3A cells, quantitative RT-PCR for hepcidin mRNA","journal":"Hepatology (Baltimore, Md.)","confidence":"Low","confidence_rationale":"Tier 3 / Weak — single knockdown experiment with one readout, single lab, no mechanistic follow-up","pmids":["25605615"],"is_preprint":false},{"year":2020,"finding":"Gnpat knockout mice (Gnpat-/-, Gnpat-/-Hfe-/-, hepatocyte-specific Gnpat-/-) showed no significant difference in serum or tissue iron levels or hepcidin expression compared to controls under normal or high-iron diet conditions. However, Gnpat knockdown in primary hepatocytes repressed BMP6-induced hepcidin expression, suggesting a specific role only under BMP6-stimulated conditions. This NEGATIVE result contradicts the earlier HepG2 siRNA study and the hypothesis that GNPAT directly mediates systemic iron homeostasis.","method":"Gnpat knockout mouse models (germline and hepatocyte-specific), high-iron diet challenge, serum iron measurement, tissue iron quantification, hepcidin mRNA expression, primary hepatocyte knockdown with BMP6 stimulation","journal":"Journal of cellular and molecular medicine","confidence":"High","confidence_rationale":"Tier 2 / Strong — multiple orthogonal mouse models (germline KO, double KO, tissue-specific KO), dietary challenge, and primary cell studies, comprehensive panel of iron parameters","pmids":["32108988"],"is_preprint":false},{"year":2020,"finding":"Gnpat heterozygous mice fed a high-iron diet had lower hepatic hepcidin mRNA expression but significantly higher serum iron levels and transferrin saturation compared to wild-type littermates, suggesting that reduced GNPAT expression impairs the hepcidin response to dietary iron challenge.","method":"Gnpat+/- mouse model, high-iron diet challenge, hepatic hepcidin (HAMP) mRNA quantification, serum iron and transferrin saturation measurements","journal":"Bioscience reports","confidence":"Medium","confidence_rationale":"Tier 2 / Moderate — in vivo mouse model with dietary challenge and multiple iron parameter readouts, single lab","pmids":["32766721"],"is_preprint":false},{"year":2023,"finding":"CSE (cigarette smoke extract) downregulates SIRT4 expression, which leads to increased acetylation of GNPAT protein; SIRT4 directly deacetylates GNPAT, and this deacetylation reduces GNPAT protein levels and activity. Knockdown of GNPAT mitigated CSE-induced ferroptosis in A549 cells, while GNPAT overexpression reversed SIRT4-mediated inhibition of ferroptosis, establishing a SIRT4→GNPAT acetylation axis in CSE-induced ferroptosis.","method":"Immunoprecipitation for GNPAT acetylation level, siRNA knockdown and overexpression in A549 cells, cell viability, LDH, ROS, lipid ROS, GSH, GPX4, and MDA measurement, western blot, qRT-PCR, immunofluorescence","journal":"Respiratory research","confidence":"Medium","confidence_rationale":"Tier 3 / Moderate — immunoprecipitation for acetylation with functional knockdown/overexpression experiments, multiple readouts, single lab","pmids":["38041059"],"is_preprint":false},{"year":2023,"finding":"Xenopus GNPAT is present in both soluble and membrane fractions, but only the membrane-bound enzyme displays acyltransferase activity. The amino-terminal domain of GNPAT, conserved in humans, shows lipid-binding capacity that is enhanced by phosphatidic acid, identifying a regulatory lipid-binding function of the N-terminus.","method":"Heterologous expression of Xenopus Gnpat in yeast, subcellular fractionation, enzyme activity assay in soluble vs. membrane fractions, lipid-binding assay with phosphatidic acid","journal":"Investigative ophthalmology & visual science","confidence":"Medium","confidence_rationale":"Tier 1-2 / Moderate — reconstituted activity in heterologous yeast system with fractionation and lipid-binding assays, single lab","pmids":["37204785"],"is_preprint":false},{"year":2025,"finding":"GNPAT recruits USP30 (ubiquitin-specific protease 30), which stabilizes DRP1 protein (dynamin-related protein 1) by deubiquitination, promoting mitochondrial fission, mitochondrial dysfunction, and cell apoptosis in CSE-exposed A549 cells. Co-immunoprecipitation confirmed direct binding among GNPAT, USP30, and DRP1.","method":"Co-immunoprecipitation assay in A549 cells, plasmid overexpression and knockdown of GNPAT/USP30/DRP1, transmission electron microscopy for mitochondrial morphology, cell viability, apoptosis, LDH, ATP production, ROS measurement, western blot, qRT-PCR, mouse COPD model","journal":"The Kaohsiung journal of medical sciences","confidence":"Medium","confidence_rationale":"Tier 3 / Moderate — co-immunoprecipitation with functional overexpression/knockdown rescue experiments, multiple readouts, single lab","pmids":["40709564"],"is_preprint":false},{"year":2026,"finding":"GNPAT promotes M2 macrophage polarization in hepatocellular carcinoma via a plasmalogen-PPARγ pathway: GNPAT overexpression in HCC cells enhances plasmalogen synthesis, activates PPAR signaling in an autocrine manner, and polarizes macrophages to an M2-like immunosuppressive phenotype.","method":"Functional overexpression experiments in HCC cell lines, macrophage co-culture polarization assay, plasmalogen quantification, PPARγ pathway readouts, migration, proliferation, and apoptosis assays","journal":"Frontiers in immunology","confidence":"Low","confidence_rationale":"Tier 3 / Weak — functional overexpression with co-culture assay and pathway readouts, single lab, single study with limited mechanistic depth at abstract level","pmids":["41816340"],"is_preprint":false}],"current_model":"GNPAT (DHAPAT) is a peroxisomal membrane-bound acyltransferase that catalyzes the first step in plasmalogen (ether-phospholipid) biosynthesis; only the membrane-associated form is catalytically active, its activity requires phosphatidic acid-enhanced N-terminal lipid binding, depends on reduced glutathione for peroxisomal integrity, and is potentiated by its partner enzyme AGPS through substrate channeling. GNPAT is post-translationally regulated by SIRT4-mediated deacetylation, and beyond plasmalogen synthesis it promotes ferroptosis susceptibility, interacts with USP30 to stabilize DRP1 and drive mitochondrial fission, and activates a plasmalogen-PPARγ axis that drives M2 macrophage polarization; its role in iron homeostasis/hepcidin regulation remains contested, with mouse knockout studies failing to confirm a direct systemic effect."},"narrative":{"mechanistic_narrative":"GNPAT (DHAPAT) is a peroxisomal acyltransferase that catalyzes the committed step of de novo plasmalogen (ether-phospholipid) biosynthesis; its isolated deficiency causes rhizomelic chondrodysplasia punctata type 2 while leaving all other peroxisomal functions intact, and GNPAT is the sole gene responsible for this disorder [PMID:8466247, PMID:11237722]. Catalytic activity is restricted to the membrane-bound form of the enzyme, with an N-terminal lipid-binding domain whose binding is enhanced by phosphatidic acid [PMID:37204785], and full activity depends on its luminal-membrane partner AGPS through substrate channeling from GNPAT to AGPS [PMID:21990100]. The enzyme exhibits high latency that requires reduced glutathione and intact peroxisomal membrane thiol groups, with ATP stimulating the latent activity through a cation-dependent mechanism [PMID:1657193]. Loss of GNPAT-dependent plasmalogen synthesis disrupts myelin integrity [PMID:10972423]. Beyond plasmalogen production, GNPAT is regulated by SIRT4-mediated deacetylation that lowers its protein level and activity and thereby modulates cigarette-smoke-induced ferroptosis [PMID:38041059], and it has been linked to mitochondrial fission via recruitment of USP30 to deubiquitinate and stabilize DRP1 [PMID:40709564]. A proposed role in hepatocyte hepcidin/iron regulation is contested: a HepG2 knockdown reduced hepcidin expression [PMID:25605615], but comprehensive Gnpat knockout mouse models showed no systemic iron or hepcidin phenotype [PMID:32108988].","teleology":[{"year":1991,"claim":"Established the biochemical behavior of the enzyme by showing its peroxisomal latency depends on reduced glutathione and membrane thiol integrity, defining the conditions under which it is catalytically active.","evidence":"selective permeabilization of human fibroblasts with defined-medium enzyme assays manipulating GSH/GSSG, ATP, and cations","pmids":["1657193"],"confidence":"Medium","gaps":["molecular basis of thiol-dependent latency not defined","no structural mechanism for ATP/cation stimulation"]},{"year":1993,"claim":"Defined the disease relevance of GNPAT by showing that its isolated deficiency causes RCDP with selectively deficient plasmalogen synthesis, establishing the enzyme as essential and non-redundant in this pathway.","evidence":"enzyme activity assay in patient fibroblasts with a comprehensive peroxisomal function panel","pmids":["8466247"],"confidence":"Medium","gaps":["causative gene mutation not yet identified at this stage","tissue-level consequences beyond fibroblasts not addressed"]},{"year":2000,"claim":"Linked a specific GNPAT missense mutation to deficient plasmalogen synthesis and abnormal myelin, connecting the enzyme's biochemical role to a downstream physiological consequence.","evidence":"patient fibroblast enzyme assays, cDNA mutation identification (R211H), and white-matter MRI","pmids":["10972423"],"confidence":"Medium","gaps":["mechanism connecting plasmalogen loss to myelin defect not established","single mutation"]},{"year":2001,"claim":"Confirmed GNPAT as the sole gene underlying RCDP type 2 by cataloguing diverse loss-of-function mutations, cementing the genotype-phenotype relationship.","evidence":"genomic sequencing, exon mapping, and mutation analysis across 12 patients","pmids":["11237722"],"confidence":"Medium","gaps":["functional impact of individual mutation classes on protein not resolved"]},{"year":2011,"claim":"Identified AGPS as a physical partner required for full GNPAT activity via substrate channeling, revealing that the enzyme functions within a coupled complex rather than alone.","evidence":"protein quantification, structural modeling, transcript analysis, and functional assays across patient cell lines","pmids":["21990100"],"confidence":"Medium","gaps":["direct structural interface of GNPAT-AGPS not defined","channeling mechanism inferred rather than reconstituted"]},{"year":2023,"claim":"Resolved which form of the enzyme is active and identified an N-terminal regulatory lipid-binding domain, explaining the soluble-versus-membrane activity difference.","evidence":"heterologous expression of Xenopus Gnpat in yeast, subcellular fractionation, activity assays, and phosphatidic-acid lipid-binding assays","pmids":["37204785"],"confidence":"Medium","gaps":["functional role of N-terminal lipid binding in human enzyme regulation not demonstrated in vivo","ortholog system"]},{"year":2023,"claim":"Established post-translational control of GNPAT by SIRT4 deacetylation and connected the enzyme to ferroptosis susceptibility, extending its role beyond constitutive plasmalogen synthesis.","evidence":"immunoprecipitation for acetylation plus knockdown/overexpression rescue with ferroptosis readouts in A549 cells","pmids":["38041059"],"confidence":"Medium","gaps":["acetylated residues not mapped","link between GNPAT activity and ferroptosis lipid species not defined"]},{"year":2025,"claim":"Proposed a non-enzymatic scaffolding role in which GNPAT recruits USP30 to stabilize DRP1 and drive mitochondrial fission, broadening its function into mitochondrial dynamics.","evidence":"co-immunoprecipitation, overexpression/knockdown rescue, electron microscopy, and a mouse COPD model in A549 cells","pmids":["40709564"],"confidence":"Medium","gaps":["reciprocal validation and endogenous-level interaction not shown","how a peroxisomal enzyme engages mitochondrial machinery unexplained"]},{"year":2020,"claim":"Tested whether GNPAT directly controls systemic iron homeostasis and found no requirement, with multiple knockout models showing normal iron and hepcidin, refining the contested hepcidin hypothesis.","evidence":"germline, double, and hepatocyte-specific Gnpat knockout mice with dietary iron challenge and primary hepatocyte BMP6 stimulation","pmids":["32108988"],"confidence":"High","gaps":["conflicting in vitro knockdown data unresolved","any conditional BMP6-stimulated role mechanistically undefined"]},{"year":2026,"claim":"Extended GNPAT function to the tumor microenvironment by linking its plasmalogen output to PPARγ-driven M2 macrophage polarization in hepatocellular carcinoma.","evidence":"overexpression in HCC cell lines, macrophage co-culture polarization, plasmalogen quantification, and PPAR pathway readouts","pmids":["41816340"],"confidence":"Low","gaps":["single low-confidence study with limited mechanistic depth","autocrine PPARγ activation not validated in vivo","specific plasmalogen ligand not identified"]},{"year":null,"claim":"How GNPAT's enzymatic plasmalogen-synthesis function mechanistically connects to its reported roles in ferroptosis, mitochondrial fission, and macrophage polarization remains unresolved.","evidence":"","pmids":[],"confidence":"Low","gaps":["no unified mechanism linking peroxisomal lipid synthesis to mitochondrial and immune phenotypes","structural basis of catalysis and partner binding undetermined","in vivo relevance of non-enzymatic functions untested"]}],"mechanism_profile":{"molecular_activity":[{"term_id":"GO:0016740","term_label":"transferase activity","supporting_discovery_ids":[0,1,5,10]},{"term_id":"GO:0008289","term_label":"lipid binding","supporting_discovery_ids":[10]}],"localization":[{"term_id":"GO:0005777","term_label":"peroxisome","supporting_discovery_ids":[0,1,5]}],"pathway":[{"term_id":"R-HSA-1430728","term_label":"Metabolism","supporting_discovery_ids":[1,3,4]}],"complexes":[],"partners":["AGPS","SIRT4","USP30","DRP1"],"other_free_text":[]}},"prefetch_data":{"uniprot":{"accession":"O15228","full_name":"Dihydroxyacetone phosphate acyltransferase","aliases":["Acyl-CoA:dihydroxyacetonephosphateacyltransferase","Glycerone-phosphate O-acyltransferase"],"length_aa":680,"mass_kda":77.2,"function":"Dihydroxyacetonephosphate acyltransferase catalyzing the first step in the biosynthesis of plasmalogens, a subset of phospholipids that differ from other glycerolipids by having an alkyl chain attached through a vinyl ether linkage at the sn-1 position of the glycerol backbone, and which unique physical properties have an impact on various aspects of cell signaling and membrane biology","subcellular_location":"Peroxisome membrane","url":"https://www.uniprot.org/uniprotkb/O15228/entry"},"depmap":{"release":"DepMap","has_data":true,"is_common_essential":false,"resolved_as":"","url":"https://depmap.org/portal/gene/GNPAT","classification":"Not Classified","n_dependent_lines":5,"n_total_lines":1208,"dependency_fraction":0.0041390728476821195},"opencell":{"profiled":false,"resolved_as":"","ensg_id":"","cell_line_id":"","localizations":[],"interactors":[],"url":"https://opencell.sf.czbiohub.org/search/GNPAT","total_profiled":1310},"omim":[{"mim_id":"603051","title":"ALKYLGLYCERONE-PHOSPHATE SYNTHASE; AGPS","url":"https://www.omim.org/entry/603051"},{"mim_id":"602744","title":"GLYCERONEPHOSPHATE O-ACYLTRANSFERASE; GNPAT","url":"https://www.omim.org/entry/602744"},{"mim_id":"600121","title":"RHIZOMELIC CHONDRODYSPLASIA PUNCTATA, TYPE 3; RCDP3","url":"https://www.omim.org/entry/600121"},{"mim_id":"222765","title":"RHIZOMELIC CHONDRODYSPLASIA PUNCTATA, TYPE 2; RCDP2","url":"https://www.omim.org/entry/222765"},{"mim_id":"215100","title":"RHIZOMELIC CHONDRODYSPLASIA PUNCTATA, TYPE 1; RCDP1","url":"https://www.omim.org/entry/215100"}],"hpa":{"profiled":true,"resolved_as":"","reliability":"Approved","locations":[{"location":"Vesicles","reliability":"Approved"},{"location":"Cell Junctions","reliability":"Approved"}],"tissue_specificity":"Low tissue specificity","tissue_distribution":"Detected in all","driving_tissues":[],"url":"https://www.proteinatlas.org/search/GNPAT"},"hgnc":{"alias_symbol":["DHAPAT","DAPAT","DAP-AT"],"prev_symbol":[]},"alphafold":{"accession":"O15228","domains":[{"cath_id":"3.40.1130","chopping":"36-53_63-381","consensus_level":"high","plddt":91.9638,"start":36,"end":381},{"cath_id":"1.10.10","chopping":"384-457","consensus_level":"medium","plddt":93.8836,"start":384,"end":457},{"cath_id":"1.10.10.10","chopping":"577-678","consensus_level":"medium","plddt":88.0216,"start":577,"end":678}],"viewer_url":"https://alphafold.ebi.ac.uk/entry/O15228","model_url":"https://alphafold.ebi.ac.uk/files/AF-O15228-F1-model_v6.cif","pae_url":"https://alphafold.ebi.ac.uk/files/AF-O15228-F1-predicted_aligned_error_v6.png","plddt_mean":88.5},"mouse_models":{"mgi_url":"https://www.informatics.jax.org/marker/summary?nomen=GNPAT","jax_strain_url":"https://www.jax.org/strain/search?query=GNPAT"},"sequence":{"accession":"O15228","fasta_url":"https://rest.uniprot.org/uniprotkb/O15228.fasta","uniprot_url":"https://www.uniprot.org/uniprotkb/O15228/entry","alphafold_viewer_url":"https://alphafold.ebi.ac.uk/entry/O15228"}},"corpus_meta":[{"pmid":"25605615","id":"PMC_25605615","title":"Exome sequencing in HFE C282Y homozygous men with extreme phenotypes identifies a GNPAT variant associated with severe iron overload.","date":"2015","source":"Hepatology (Baltimore, Md.)","url":"https://pubmed.ncbi.nlm.nih.gov/25605615","citation_count":64,"is_preprint":false},{"pmid":"21990100","id":"PMC_21990100","title":"Functional characterization of novel mutations in GNPAT and AGPS, causing rhizomelic chondrodysplasia punctata (RCDP) types 2 and 3.","date":"2011","source":"Human mutation","url":"https://pubmed.ncbi.nlm.nih.gov/21990100","citation_count":61,"is_preprint":false},{"pmid":"16997000","id":"PMC_16997000","title":"A single nucleotide polymorphism fine mapping study of chromosome 1q42.1 reveals the vulnerability genes for schizophrenia, GNPAT and DISC1: Association with impairment of sustained attention.","date":"2006","source":"Biological psychiatry","url":"https://pubmed.ncbi.nlm.nih.gov/16997000","citation_count":55,"is_preprint":false},{"pmid":"8466247","id":"PMC_8466247","title":"Rhizomelic chondrodysplasia punctata with isolated DHAP-AT deficiency.","date":"1993","source":"Archives of disease in childhood","url":"https://pubmed.ncbi.nlm.nih.gov/8466247","citation_count":29,"is_preprint":false},{"pmid":"10972423","id":"PMC_10972423","title":"Abnormal myelin formation in rhizomelic chondrodysplasia punctata type 2 (DHAPAT-deficiency).","date":"2000","source":"Developmental medicine and child neurology","url":"https://pubmed.ncbi.nlm.nih.gov/10972423","citation_count":23,"is_preprint":false},{"pmid":"38041059","id":"PMC_38041059","title":"CSE triggers ferroptosis via SIRT4-mediated GNPAT deacetylation in the pathogenesis of COPD.","date":"2023","source":"Respiratory research","url":"https://pubmed.ncbi.nlm.nih.gov/38041059","citation_count":21,"is_preprint":false},{"pmid":"9843043","id":"PMC_9843043","title":"Developmental delay and growth failure caused by a peroxisomal disorder, dihydroxyacetonephosphate acyltransferase (DHAP-AT) deficiency.","date":"1998","source":"American journal of medical genetics","url":"https://pubmed.ncbi.nlm.nih.gov/9843043","citation_count":17,"is_preprint":false},{"pmid":"9053559","id":"PMC_9053559","title":"Measurement of dihydroxyacetone-phosphate acyltransferase (DHAPAT) in chorionic villous samples, blood cells and cultured cells.","date":"1995","source":"Journal of inherited metabolic disease","url":"https://pubmed.ncbi.nlm.nih.gov/9053559","citation_count":16,"is_preprint":false},{"pmid":"1657193","id":"PMC_1657193","title":"Factors influencing the latency of the peroxisomal enzyme dihydroxyacetone-phosphate acyltransferase (DHAP-AT) in permeabilized human skin fibroblasts.","date":"1991","source":"Biochimica et biophysica acta","url":"https://pubmed.ncbi.nlm.nih.gov/1657193","citation_count":14,"is_preprint":false},{"pmid":"11237722","id":"PMC_11237722","title":"Etherphospholipid biosynthesis and dihydroxyactetone-phosphate acyltransferase: resolution of the genomic organization of the human gnpat gene and its use in the identification of novel mutations.","date":"2001","source":"Biochemical and biophysical research communications","url":"https://pubmed.ncbi.nlm.nih.gov/11237722","citation_count":11,"is_preprint":false},{"pmid":"27740525","id":"PMC_27740525","title":"GNPAT variant (D519G) is not associated with an elevated serum ferritin or iron removed by phlebotomy in patients referred for C282Y-linked hemochromatosis.","date":"2016","source":"Annals of hepatology","url":"https://pubmed.ncbi.nlm.nih.gov/27740525","citation_count":6,"is_preprint":false},{"pmid":"22371097","id":"PMC_22371097","title":"Synthesis and structure-activity relationships of o-sulfonamido-arylhydrazides as inhibitors of LL-diaminopimelate aminotransferase (LL-DAP-AT).","date":"2012","source":"Organic & biomolecular chemistry","url":"https://pubmed.ncbi.nlm.nih.gov/22371097","citation_count":6,"is_preprint":false},{"pmid":"32108988","id":"PMC_32108988","title":"Gnpat does not play an essential role in systemic iron homeostasis in murine model.","date":"2020","source":"Journal of cellular and molecular medicine","url":"https://pubmed.ncbi.nlm.nih.gov/32108988","citation_count":3,"is_preprint":false},{"pmid":"37555240","id":"PMC_37555240","title":"RNA N6-methyladenosine of DHAPAT and PAP involves in regulation of diapause of Bombyx mori via the lipid metabolism pathway.","date":"2023","source":"Bulletin of entomological research","url":"https://pubmed.ncbi.nlm.nih.gov/37555240","citation_count":3,"is_preprint":false},{"pmid":"37323250","id":"PMC_37323250","title":"Neonatal rhizomelic chondrodysplasia punctata type 2 caused by a novel homozygous variant in the GNPAT gene.","date":"2023","source":"Clinical case reports","url":"https://pubmed.ncbi.nlm.nih.gov/37323250","citation_count":3,"is_preprint":false},{"pmid":"40394457","id":"PMC_40394457","title":"A bovine model of rhizomelic chondrodysplasia punctata caused by a deep intronic splicing variant in the GNPAT gene.","date":"2025","source":"Genetics, selection, evolution : GSE","url":"https://pubmed.ncbi.nlm.nih.gov/40394457","citation_count":3,"is_preprint":false},{"pmid":"40709564","id":"PMC_40709564","title":"GNPAT/USP30 Stabilizes DRP1 Protein to Promote Mitochondrial Fission and Functional Damage in COPD Progression.","date":"2025","source":"The Kaohsiung journal of medical sciences","url":"https://pubmed.ncbi.nlm.nih.gov/40709564","citation_count":2,"is_preprint":false},{"pmid":"37204785","id":"PMC_37204785","title":"Differential Eye Expression of Xenopus Acyltransferase Gnpat and Its Biochemical Characterization Shed Light on Lipid-Associated Ocular Pathologies.","date":"2023","source":"Investigative ophthalmology & visual science","url":"https://pubmed.ncbi.nlm.nih.gov/37204785","citation_count":1,"is_preprint":false},{"pmid":"34110102","id":"PMC_34110102","title":"A new GNPAT variant of foetal rhizomelic chondrodysplasia punctata.","date":"2021","source":"Molecular genetics & genomic medicine","url":"https://pubmed.ncbi.nlm.nih.gov/34110102","citation_count":1,"is_preprint":false},{"pmid":"32766721","id":"PMC_32766721","title":"Dysregulated hepcidin response to dietary iron in male mice with reduced Gnpat expression.","date":"2020","source":"Bioscience reports","url":"https://pubmed.ncbi.nlm.nih.gov/32766721","citation_count":0,"is_preprint":false},{"pmid":"41816340","id":"PMC_41816340","title":"GNPAT promotes immunosuppression in hepatocellular carcinoma by activating the plasmalogen-PPARγ pathway to drive M2 macrophage polarization.","date":"2026","source":"Frontiers in immunology","url":"https://pubmed.ncbi.nlm.nih.gov/41816340","citation_count":0,"is_preprint":false},{"pmid":null,"id":"bio_10.1101_2025.09.10.675385","title":"Impaired lysine biosynthesis drives constitutive energy stress priming and darkness stress responses","date":"2025-09-12","source":"bioRxiv","url":"https://doi.org/10.1101/2025.09.10.675385","citation_count":0,"is_preprint":true}],"cost":{"stage1":{"model":"claude-sonnet-4-6","input_tokens":12687,"output_tokens":3722,"usd":0.046946,"stage1_stop_reason":"end_turn"},"stage2":{"model":"claude-opus-4-8","input_tokens":11245,"output_tokens":3198,"usd":0.068087,"stage2_stop_reason":"end_turn"},"total_usd":0.115033,"stage1_batch_id":"msgbatch_01DUPACZi7AAWwuZiUjseQF4","stage2_batch_id":"msgbatch_018yt7chE3f48PEFtDwtj6AE","note":"batch pricing = 50% of standard"},"stage1_raw":"Looking at the papers, I need to identify which ones are about the canonical human GNPAT (glyceronephosphate O-acyltransferase, also known as DHAPAT/DHAP-AT/DAP-AT) gene involved in plasmalogen biosynthesis.\n\nPaper [12] is about LL-diaminopimelate aminotransferase (LL-DAP-AT) in plants/bacteria - symbol collision, exclude.\nPaper [14] is about Bombyx mori DHAPAT - this is an insect ortholog, but the study is about silkworm diapause regulation, not canonical mammalian function. The context is fundamentally about insect biology with m6A methylation. I'll treat cautiously.\nPaper [22] is about a plant lysine biosynthesis mutant \"dapat\" - symbol collision with plant gene, exclude.\nPaper [3] is a genetic association study for schizophrenia - no mechanistic experiments on GNPAT protein, exclude.\n\nNow extracting mechanistic discoveries from relevant papers:\n\n```json\n{\n  \"discoveries\": [\n    {\n      \"year\": 1991,\n      \"finding\": \"GNPAT (DHAP-AT) is a peroxisomal enzyme with ~80% latency in selectively permeabilized fibroblasts. Its latency is critically dependent on reduced glutathione (GSH) in the medium, and ATP stimulates latent DHAP-AT activity through a mechanism involving cations but not a proton gradient. The integrity of peroxisomal membrane thiol-groups is essential for maintaining DHAP-AT latency.\",\n      \"method\": \"Selective permeabilization of human skin fibroblasts, enzyme activity assays with defined medium components (GSH/GSSG manipulation, ATP, cation chelation)\",\n      \"journal\": \"Biochimica et biophysica acta\",\n      \"confidence\": \"Medium\",\n      \"confidence_rationale\": \"Tier 2 / Moderate — direct biochemical assay in permeabilized cells with multiple medium component manipulations, single lab\",\n      \"pmids\": [\"1657193\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 1993,\n      \"finding\": \"Isolated deficiency of the peroxisomal enzyme DHAP-AT (GNPAT) causes rhizomelic chondrodysplasia punctata with deficient de novo plasmalogen biosynthesis, establishing GNPAT as essential for the plasmalogen synthesis pathway in humans, with all other peroxisomal functions normal.\",\n      \"method\": \"Enzyme activity measurement in patient fibroblasts, clinical phenotype characterization, peroxisomal function panel\",\n      \"journal\": \"Archives of disease in childhood\",\n      \"confidence\": \"Medium\",\n      \"confidence_rationale\": \"Tier 2 / Moderate — direct enzyme activity assays in patient cells with comprehensive peroxisomal function panel ruling out other deficiencies, single lab\",\n      \"pmids\": [\"8466247\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 1995,\n      \"finding\": \"GNPAT (DHAPAT) enzymatic activity can be reliably measured in chorionic villous samples, blood cells, cultured skin fibroblasts, cultured chorionic villus fibroblasts, and cultured amniocytes, establishing the enzyme's presence and activity across multiple human cell types and its utility for prenatal diagnosis.\",\n      \"method\": \"Enzyme activity assay in multiple human cell types and tissue samples\",\n      \"journal\": \"Journal of inherited metabolic disease\",\n      \"confidence\": \"Medium\",\n      \"confidence_rationale\": \"Tier 2 / Moderate — direct enzyme activity measurement across multiple cell types, single lab, single method\",\n      \"pmids\": [\"9053559\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2000,\n      \"finding\": \"A missense mutation in GNPAT (R211H) causes isolated DHAPAT deficiency, resulting in deficient de novo plasmalogen synthesis in patient fibroblasts while alkyl-DHAP synthase activity, VLCFA profile, phytanic acid concentration, and peroxisomal thiolase remain normal. The deficiency leads to abnormal myelin formation, linking GNPAT-dependent plasmalogen biosynthesis to myelin integrity.\",\n      \"method\": \"Enzyme activity assays in patient fibroblasts, mutation identification by cDNA sequencing, MRI of white matter\",\n      \"journal\": \"Developmental medicine and child neurology\",\n      \"confidence\": \"Medium\",\n      \"confidence_rationale\": \"Tier 2 / Moderate — direct mutation identification with functional enzyme activity assay in patient cells, orthogonal MRI evidence, single lab\",\n      \"pmids\": [\"10972423\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2001,\n      \"finding\": \"The human GNPAT gene spans ~28 kb on chromosome 1q42.12-43, consists of 16 exons and 15 introns, and nine distinct mutations (missense, deletions, insertion, splice site mutations) in GNPAT cause DHAPAT enzyme deficiency and impaired etherphospholipid biosynthesis. This confirmed GNPAT as the sole gene responsible for RCDP type 2.\",\n      \"method\": \"Genomic sequencing, exon mapping, patient mutation analysis across 12 GNPAT-deficient patients, splice site and transcript analysis\",\n      \"journal\": \"Biochemical and biophysical research communications\",\n      \"confidence\": \"Medium\",\n      \"confidence_rationale\": \"Tier 2 / Moderate — comprehensive genomic characterization with mutation-phenotype correlation in 12 patients, single lab, multiple mutation types\",\n      \"pmids\": [\"11237722\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2011,\n      \"finding\": \"GNPAT and AGPS physically partner on the luminal peroxisomal membrane surface; the presence of AGPS is required for full GNPAT activity, and full GNPAT activity also depends on intact substrate channeling from GNPAT to AGPS. Milder RCDP phenotypes correlate with residual GNPAT or AGPS protein function.\",\n      \"method\": \"Cell line protein quantification (GNPAT and AGPS levels compared across patient and control lines), protein modeling for structural consequences, transcript analysis, functional enzyme assays in patient fibroblasts\",\n      \"journal\": \"Human mutation\",\n      \"confidence\": \"Medium\",\n      \"confidence_rationale\": \"Tier 2 / Moderate — multiple patient cell lines, protein quantification, transcript analysis, and functional assays; single lab but orthogonal methods\",\n      \"pmids\": [\"21990100\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2015,\n      \"finding\": \"siRNA-mediated knockdown of GNPAT in HepG2/C3A human liver-derived cells resulted in a >17-fold decrease in hepcidin (HAMP) mRNA expression, indicating that GNPAT participates in hepcidin regulation in hepatocytes.\",\n      \"method\": \"siRNA knockdown of GNPAT in HepG2/C3A cells, quantitative RT-PCR for hepcidin mRNA\",\n      \"journal\": \"Hepatology (Baltimore, Md.)\",\n      \"confidence\": \"Low\",\n      \"confidence_rationale\": \"Tier 3 / Weak — single knockdown experiment with one readout, single lab, no mechanistic follow-up\",\n      \"pmids\": [\"25605615\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2020,\n      \"finding\": \"Gnpat knockout mice (Gnpat-/-, Gnpat-/-Hfe-/-, hepatocyte-specific Gnpat-/-) showed no significant difference in serum or tissue iron levels or hepcidin expression compared to controls under normal or high-iron diet conditions. However, Gnpat knockdown in primary hepatocytes repressed BMP6-induced hepcidin expression, suggesting a specific role only under BMP6-stimulated conditions. This NEGATIVE result contradicts the earlier HepG2 siRNA study and the hypothesis that GNPAT directly mediates systemic iron homeostasis.\",\n      \"method\": \"Gnpat knockout mouse models (germline and hepatocyte-specific), high-iron diet challenge, serum iron measurement, tissue iron quantification, hepcidin mRNA expression, primary hepatocyte knockdown with BMP6 stimulation\",\n      \"journal\": \"Journal of cellular and molecular medicine\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 2 / Strong — multiple orthogonal mouse models (germline KO, double KO, tissue-specific KO), dietary challenge, and primary cell studies, comprehensive panel of iron parameters\",\n      \"pmids\": [\"32108988\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2020,\n      \"finding\": \"Gnpat heterozygous mice fed a high-iron diet had lower hepatic hepcidin mRNA expression but significantly higher serum iron levels and transferrin saturation compared to wild-type littermates, suggesting that reduced GNPAT expression impairs the hepcidin response to dietary iron challenge.\",\n      \"method\": \"Gnpat+/- mouse model, high-iron diet challenge, hepatic hepcidin (HAMP) mRNA quantification, serum iron and transferrin saturation measurements\",\n      \"journal\": \"Bioscience reports\",\n      \"confidence\": \"Medium\",\n      \"confidence_rationale\": \"Tier 2 / Moderate — in vivo mouse model with dietary challenge and multiple iron parameter readouts, single lab\",\n      \"pmids\": [\"32766721\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2023,\n      \"finding\": \"CSE (cigarette smoke extract) downregulates SIRT4 expression, which leads to increased acetylation of GNPAT protein; SIRT4 directly deacetylates GNPAT, and this deacetylation reduces GNPAT protein levels and activity. Knockdown of GNPAT mitigated CSE-induced ferroptosis in A549 cells, while GNPAT overexpression reversed SIRT4-mediated inhibition of ferroptosis, establishing a SIRT4→GNPAT acetylation axis in CSE-induced ferroptosis.\",\n      \"method\": \"Immunoprecipitation for GNPAT acetylation level, siRNA knockdown and overexpression in A549 cells, cell viability, LDH, ROS, lipid ROS, GSH, GPX4, and MDA measurement, western blot, qRT-PCR, immunofluorescence\",\n      \"journal\": \"Respiratory research\",\n      \"confidence\": \"Medium\",\n      \"confidence_rationale\": \"Tier 3 / Moderate — immunoprecipitation for acetylation with functional knockdown/overexpression experiments, multiple readouts, single lab\",\n      \"pmids\": [\"38041059\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2023,\n      \"finding\": \"Xenopus GNPAT is present in both soluble and membrane fractions, but only the membrane-bound enzyme displays acyltransferase activity. The amino-terminal domain of GNPAT, conserved in humans, shows lipid-binding capacity that is enhanced by phosphatidic acid, identifying a regulatory lipid-binding function of the N-terminus.\",\n      \"method\": \"Heterologous expression of Xenopus Gnpat in yeast, subcellular fractionation, enzyme activity assay in soluble vs. membrane fractions, lipid-binding assay with phosphatidic acid\",\n      \"journal\": \"Investigative ophthalmology & visual science\",\n      \"confidence\": \"Medium\",\n      \"confidence_rationale\": \"Tier 1-2 / Moderate — reconstituted activity in heterologous yeast system with fractionation and lipid-binding assays, single lab\",\n      \"pmids\": [\"37204785\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2025,\n      \"finding\": \"GNPAT recruits USP30 (ubiquitin-specific protease 30), which stabilizes DRP1 protein (dynamin-related protein 1) by deubiquitination, promoting mitochondrial fission, mitochondrial dysfunction, and cell apoptosis in CSE-exposed A549 cells. Co-immunoprecipitation confirmed direct binding among GNPAT, USP30, and DRP1.\",\n      \"method\": \"Co-immunoprecipitation assay in A549 cells, plasmid overexpression and knockdown of GNPAT/USP30/DRP1, transmission electron microscopy for mitochondrial morphology, cell viability, apoptosis, LDH, ATP production, ROS measurement, western blot, qRT-PCR, mouse COPD model\",\n      \"journal\": \"The Kaohsiung journal of medical sciences\",\n      \"confidence\": \"Medium\",\n      \"confidence_rationale\": \"Tier 3 / Moderate — co-immunoprecipitation with functional overexpression/knockdown rescue experiments, multiple readouts, single lab\",\n      \"pmids\": [\"40709564\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2026,\n      \"finding\": \"GNPAT promotes M2 macrophage polarization in hepatocellular carcinoma via a plasmalogen-PPARγ pathway: GNPAT overexpression in HCC cells enhances plasmalogen synthesis, activates PPAR signaling in an autocrine manner, and polarizes macrophages to an M2-like immunosuppressive phenotype.\",\n      \"method\": \"Functional overexpression experiments in HCC cell lines, macrophage co-culture polarization assay, plasmalogen quantification, PPARγ pathway readouts, migration, proliferation, and apoptosis assays\",\n      \"journal\": \"Frontiers in immunology\",\n      \"confidence\": \"Low\",\n      \"confidence_rationale\": \"Tier 3 / Weak — functional overexpression with co-culture assay and pathway readouts, single lab, single study with limited mechanistic depth at abstract level\",\n      \"pmids\": [\"41816340\"],\n      \"is_preprint\": false\n    }\n  ],\n  \"current_model\": \"GNPAT (DHAPAT) is a peroxisomal membrane-bound acyltransferase that catalyzes the first step in plasmalogen (ether-phospholipid) biosynthesis; only the membrane-associated form is catalytically active, its activity requires phosphatidic acid-enhanced N-terminal lipid binding, depends on reduced glutathione for peroxisomal integrity, and is potentiated by its partner enzyme AGPS through substrate channeling. GNPAT is post-translationally regulated by SIRT4-mediated deacetylation, and beyond plasmalogen synthesis it promotes ferroptosis susceptibility, interacts with USP30 to stabilize DRP1 and drive mitochondrial fission, and activates a plasmalogen-PPARγ axis that drives M2 macrophage polarization; its role in iron homeostasis/hepcidin regulation remains contested, with mouse knockout studies failing to confirm a direct systemic effect.\"\n}\n```","stage2_raw":"{\n  \"mechanistic_narrative\": \"GNPAT (DHAPAT) is a peroxisomal acyltransferase that catalyzes the committed step of de novo plasmalogen (ether-phospholipid) biosynthesis; its isolated deficiency causes rhizomelic chondrodysplasia punctata type 2 while leaving all other peroxisomal functions intact, and GNPAT is the sole gene responsible for this disorder [#1, #4]. Catalytic activity is restricted to the membrane-bound form of the enzyme, with an N-terminal lipid-binding domain whose binding is enhanced by phosphatidic acid [#10], and full activity depends on its luminal-membrane partner AGPS through substrate channeling from GNPAT to AGPS [#5]. The enzyme exhibits high latency that requires reduced glutathione and intact peroxisomal membrane thiol groups, with ATP stimulating the latent activity through a cation-dependent mechanism [#0]. Loss of GNPAT-dependent plasmalogen synthesis disrupts myelin integrity [#3]. Beyond plasmalogen production, GNPAT is regulated by SIRT4-mediated deacetylation that lowers its protein level and activity and thereby modulates cigarette-smoke-induced ferroptosis [#9], and it has been linked to mitochondrial fission via recruitment of USP30 to deubiquitinate and stabilize DRP1 [#11]. A proposed role in hepatocyte hepcidin/iron regulation is contested: a HepG2 knockdown reduced hepcidin expression [#6], but comprehensive Gnpat knockout mouse models showed no systemic iron or hepcidin phenotype [#7].\"\n,\n  \"teleology\": [\n    {\n      \"year\": 1991,\n      \"claim\": \"Established the biochemical behavior of the enzyme by showing its peroxisomal latency depends on reduced glutathione and membrane thiol integrity, defining the conditions under which it is catalytically active.\",\n      \"evidence\": \"selective permeabilization of human fibroblasts with defined-medium enzyme assays manipulating GSH/GSSG, ATP, and cations\",\n      \"pmids\": [\"1657193\"],\n      \"confidence\": \"Medium\",\n      \"gaps\": [\"molecular basis of thiol-dependent latency not defined\", \"no structural mechanism for ATP/cation stimulation\"]\n    },\n    {\n      \"year\": 1993,\n      \"claim\": \"Defined the disease relevance of GNPAT by showing that its isolated deficiency causes RCDP with selectively deficient plasmalogen synthesis, establishing the enzyme as essential and non-redundant in this pathway.\",\n      \"evidence\": \"enzyme activity assay in patient fibroblasts with a comprehensive peroxisomal function panel\",\n      \"pmids\": [\"8466247\"],\n      \"confidence\": \"Medium\",\n      \"gaps\": [\"causative gene mutation not yet identified at this stage\", \"tissue-level consequences beyond fibroblasts not addressed\"]\n    },\n    {\n      \"year\": 2000,\n      \"claim\": \"Linked a specific GNPAT missense mutation to deficient plasmalogen synthesis and abnormal myelin, connecting the enzyme's biochemical role to a downstream physiological consequence.\",\n      \"evidence\": \"patient fibroblast enzyme assays, cDNA mutation identification (R211H), and white-matter MRI\",\n      \"pmids\": [\"10972423\"],\n      \"confidence\": \"Medium\",\n      \"gaps\": [\"mechanism connecting plasmalogen loss to myelin defect not established\", \"single mutation\"]\n    },\n    {\n      \"year\": 2001,\n      \"claim\": \"Confirmed GNPAT as the sole gene underlying RCDP type 2 by cataloguing diverse loss-of-function mutations, cementing the genotype-phenotype relationship.\",\n      \"evidence\": \"genomic sequencing, exon mapping, and mutation analysis across 12 patients\",\n      \"pmids\": [\"11237722\"],\n      \"confidence\": \"Medium\",\n      \"gaps\": [\"functional impact of individual mutation classes on protein not resolved\"]\n    },\n    {\n      \"year\": 2011,\n      \"claim\": \"Identified AGPS as a physical partner required for full GNPAT activity via substrate channeling, revealing that the enzyme functions within a coupled complex rather than alone.\",\n      \"evidence\": \"protein quantification, structural modeling, transcript analysis, and functional assays across patient cell lines\",\n      \"pmids\": [\"21990100\"],\n      \"confidence\": \"Medium\",\n      \"gaps\": [\"direct structural interface of GNPAT-AGPS not defined\", \"channeling mechanism inferred rather than reconstituted\"]\n    },\n    {\n      \"year\": 2023,\n      \"claim\": \"Resolved which form of the enzyme is active and identified an N-terminal regulatory lipid-binding domain, explaining the soluble-versus-membrane activity difference.\",\n      \"evidence\": \"heterologous expression of Xenopus Gnpat in yeast, subcellular fractionation, activity assays, and phosphatidic-acid lipid-binding assays\",\n      \"pmids\": [\"37204785\"],\n      \"confidence\": \"Medium\",\n      \"gaps\": [\"functional role of N-terminal lipid binding in human enzyme regulation not demonstrated in vivo\", \"ortholog system\"]\n    },\n    {\n      \"year\": 2023,\n      \"claim\": \"Established post-translational control of GNPAT by SIRT4 deacetylation and connected the enzyme to ferroptosis susceptibility, extending its role beyond constitutive plasmalogen synthesis.\",\n      \"evidence\": \"immunoprecipitation for acetylation plus knockdown/overexpression rescue with ferroptosis readouts in A549 cells\",\n      \"pmids\": [\"38041059\"],\n      \"confidence\": \"Medium\",\n      \"gaps\": [\"acetylated residues not mapped\", \"link between GNPAT activity and ferroptosis lipid species not defined\"]\n    },\n    {\n      \"year\": 2025,\n      \"claim\": \"Proposed a non-enzymatic scaffolding role in which GNPAT recruits USP30 to stabilize DRP1 and drive mitochondrial fission, broadening its function into mitochondrial dynamics.\",\n      \"evidence\": \"co-immunoprecipitation, overexpression/knockdown rescue, electron microscopy, and a mouse COPD model in A549 cells\",\n      \"pmids\": [\"40709564\"],\n      \"confidence\": \"Medium\",\n      \"gaps\": [\"reciprocal validation and endogenous-level interaction not shown\", \"how a peroxisomal enzyme engages mitochondrial machinery unexplained\"]\n    },\n    {\n      \"year\": 2020,\n      \"claim\": \"Tested whether GNPAT directly controls systemic iron homeostasis and found no requirement, with multiple knockout models showing normal iron and hepcidin, refining the contested hepcidin hypothesis.\",\n      \"evidence\": \"germline, double, and hepatocyte-specific Gnpat knockout mice with dietary iron challenge and primary hepatocyte BMP6 stimulation\",\n      \"pmids\": [\"32108988\"],\n      \"confidence\": \"High\",\n      \"gaps\": [\"conflicting in vitro knockdown data unresolved\", \"any conditional BMP6-stimulated role mechanistically undefined\"]\n    },\n    {\n      \"year\": 2026,\n      \"claim\": \"Extended GNPAT function to the tumor microenvironment by linking its plasmalogen output to PPARγ-driven M2 macrophage polarization in hepatocellular carcinoma.\",\n      \"evidence\": \"overexpression in HCC cell lines, macrophage co-culture polarization, plasmalogen quantification, and PPAR pathway readouts\",\n      \"pmids\": [\"41816340\"],\n      \"confidence\": \"Low\",\n      \"gaps\": [\"single low-confidence study with limited mechanistic depth\", \"autocrine PPARγ activation not validated in vivo\", \"specific plasmalogen ligand not identified\"]\n    },\n    {\n      \"year\": null,\n      \"claim\": \"How GNPAT's enzymatic plasmalogen-synthesis function mechanistically connects to its reported roles in ferroptosis, mitochondrial fission, and macrophage polarization remains unresolved.\",\n      \"evidence\": \"\",\n      \"pmids\": [],\n      \"confidence\": \"Low\",\n      \"gaps\": [\"no unified mechanism linking peroxisomal lipid synthesis to mitochondrial and immune phenotypes\", \"structural basis of catalysis and partner binding undetermined\", \"in vivo relevance of non-enzymatic functions untested\"]\n    }\n  ],\n  \"mechanism_profile\": {\n    \"molecular_activity\": [\n      {\"term_id\": \"GO:0016740\", \"supporting_discovery_ids\": [0, 1, 5, 10]},\n      {\"term_id\": \"GO:0008289\", \"supporting_discovery_ids\": [10]}\n    ],\n    \"localization\": [\n      {\"term_id\": \"GO:0005777\", \"supporting_discovery_ids\": [0, 1, 5]}\n    ],\n    \"pathway\": [\n      {\"term_id\": \"R-HSA-1430728\", \"supporting_discovery_ids\": [1, 3, 4]}\n    ],\n    \"complexes\": [],\n    \"partners\": [\"AGPS\", \"SIRT4\", \"USP30\", \"DRP1\"],\n    \"other_free_text\": []\n  }\n}","audit_flag":null,"evaluation":{"pairwise":"win","faith_supported":6,"faith_total":6,"faith_pct":100.0}}