{"gene":"UPF3A","run_date":"2026-06-10T10:51:56","timeline":{"discoveries":[{"year":2001,"finding":"Human UPF3A (hUpf3p) was identified as a human orthologue of S. cerevisiae Upf3p; co-immunoprecipitation of epitope-tagged proteins in HeLa cells demonstrated that hUpf2p interacts with both hUpf3p-X (UPF3B) and hUpf3p (UPF3A), and the domains required for these interactions were defined. UPF3A localizes primarily to nuclei and is a shuttling protein, indicating NMD has both nuclear and cytoplasmic components.","method":"Co-immunoprecipitation of epitope-tagged proteins in HeLa cells; indirect immunofluorescence for subcellular localization","journal":"Molecular and cellular biology","confidence":"Medium","confidence_rationale":"Tier 2 / Moderate — reciprocal Co-IP with domain mapping and localization by immunofluorescence, single lab","pmids":["11113196"],"is_preprint":false},{"year":2004,"finding":"Crystal structure at 1.95 Å of the complex between the interacting domains of human UPF2 (MIF4G domain) and UPF3B (RNP domain) revealed that the protein-protein interface is mediated by highly conserved charged residues and involves the beta-sheet surface of the UPF3B RNP domain. The UPF3b RNP domain does not bind RNA, whereas the UPF2 construct and the complex do. The same RNP domain architecture is shared by UPF3A.","method":"X-ray crystallography (1.95 Å resolution); RNA-binding assays","journal":"Nature structural & molecular biology","confidence":"High","confidence_rationale":"Tier 1 / Strong — crystal structure with functional validation of RNA-binding; foundational structural result replicated by later studies","pmids":["15004547"],"is_preprint":false},{"year":2006,"finding":"Using a tethering system, UPF3A was shown to be much less active than UPF3B in inducing NMD and stimulating translation. The C-terminal domain region that discriminates UPF3A from UPF3B in NMD function mediates interaction with EJC components Y14, Magoh, BTZ, and eIF4AIII; this interaction is required for NMD induction. Translation stimulation is independent of EJC interaction and is determined by other regions of the UPF3 proteins.","method":"lambdaN/boxB tethering reporter assay; co-immunoprecipitation for EJC interaction mapping","journal":"RNA (New York, N.Y.)","confidence":"Medium","confidence_rationale":"Tier 2 / Moderate — functional tethering assay with domain dissection, single lab, two orthogonal approaches","pmids":["16601204"],"is_preprint":false},{"year":2007,"finding":"Recombinant EJC core was sufficient to reconstitute a stable heptameric complex on RNA with UPF1, UPF2, and UPF3b. EJC proteins MAGOH, Y14, and eIF4AIII provide a composite binding site for UPF3b that bridges to UPF2 and UPF1. In the trimeric UPF complex, UPF2 and UPF3b cooperatively stimulate both ATPase and RNA helicase activities of UPF1.","method":"In vitro reconstitution; ATPase assay; RNA helicase assay with recombinant proteins","journal":"Nature structural & molecular biology","confidence":"High","confidence_rationale":"Tier 1 / Strong — in vitro reconstitution of heptameric complex with enzymatic assays, multiple orthogonal methods","pmids":["18066079"],"is_preprint":false},{"year":2009,"finding":"UPF3A protein levels are regulated post-transcriptionally: the presence of UPF3B promotes destabilization/reduction of UPF3A levels via a conserved UPF3B-dependent mechanism. When UPF3B is absent (e.g., due to mutations), UPF3A levels rise and partially compensate for NMD, but UPF3A also impairs NMD by competing with UPF3B for binding to UPF2. This UPF3B-dependent destabilization of UPF3A constitutes a post-transcriptional regulatory switch maintaining appropriate NMD levels.","method":"Western blotting of UPF3A/UPF3B levels in cells with UPF3B mutations; competitive binding assays; NMD reporter assays","journal":"Nature structural & molecular biology","confidence":"High","confidence_rationale":"Tier 2 / Strong — reciprocal protein-level regulation demonstrated in multiple patient-derived cell lines and knockdown experiments, multiple orthogonal readouts","pmids":["19503078"],"is_preprint":false},{"year":2016,"finding":"UPF3A acts primarily as a potent NMD inhibitor, stabilizing hundreds of transcripts, rather than an NMD activator. It acquired repressor activity through impairment of a critical domain (C-terminal EJC-binding domain is weakened relative to UPF3B). Conditional knockout of UPF3A in mice causes 'hyper' NMD and leads to defects in embryogenesis and gametogenesis. UPF3A competes with UPF3B for binding to UPF2, and its NMD-inhibitory function is explained by its weaker NMD activation capacity displacing the stronger activator UPF3B.","method":"Loss-of-function in vitro (siRNA/shRNA) and in vivo (conditional knockout mouse); RNA-seq; NMD reporter assays; co-immunoprecipitation","journal":"Cell","confidence":"High","confidence_rationale":"Tier 2 / Strong — multiple orthogonal methods in vitro and in vivo, conditional KO mouse with defined phenotypes, replicated by later studies","pmids":["27040500"],"is_preprint":false},{"year":2019,"finding":"In zebrafish, Upf3a (a NMD pathway member) and components of the COMPASS complex including Wdr5 are required for the genetic compensation response (GCR) triggered by mRNAs bearing a premature termination codon (PTC). The GCR is accompanied by enhancement of H3K4me3 at the transcription start site regions of compensatory genes, linking Upf3a to transcriptional upregulation of homologous genes in response to PTC-containing mRNA.","method":"Zebrafish knockdown vs. knockout models; transgene analysis; ChIP for H3K4me3; genetic epistasis using upf3a mutants","journal":"Nature","confidence":"High","confidence_rationale":"Tier 2 / Strong — genetic epistasis with multiple designed transgenes, ChIP-seq, replicated across two gene models (capn3a, nid1a)","pmids":["30944473"],"is_preprint":false},{"year":2020,"finding":"Reconstituted UPF3A expression in KM12 CRC cells (which have a frameshift mutation in UPF3A) caused down-regulation of several enzymes involved in cholesterol biosynthesis and altered phosphorylation of 85 phosphosites in 52 phosphoproteins, predominantly nuclear proteins involved in gene expression regulation and RNA splicing, suggesting UPF3A influences cellular signaling pathways beyond NMD.","method":"SILAC-based quantitative proteomics; phosphoproteomics in CRC cell lines with reconstituted UPF3A expression","journal":"International journal of molecular sciences","confidence":"Medium","confidence_rationale":"Tier 2 / Weak — quantitative proteomics with reconstitution in a single CRC cell line, single lab","pmids":["32718059"],"is_preprint":false},{"year":2022,"finding":"UPF3A and UPF3B share structural homology comprising an RRM-like domain (RRM-L), a NONA/paraspeckle-like domain (NOPS-L), and an extended α-helical domain; these domains are essential for RNA/ribosome-binding, RNA-induced oligomerization, and UPF2 interaction. Crystal structures of UPF2's MIF4GIII domain in complex with UPF3B or UPF3A revealed intimate binding interfaces. UPF3A binds UPF2 with ~10-fold higher affinity than UPF3B. The disease-causing UPF3B mutation Y160D in the NOPS-L domain displaces Y160 from a hydrophobic cleft in UPF2, reducing binding affinity ~40-fold. UPF3A and UPF3B compete for the same UPF2 binding site.","method":"X-ray crystallography; binding affinity measurements; mutagenesis; RNA/ribosome binding assays","journal":"Nucleic acids research","confidence":"High","confidence_rationale":"Tier 1 / Strong — crystal structures of both paralogs with UPF2, mutagenesis validation, quantitative binding measurements, multiple orthogonal methods","pmids":["35640974"],"is_preprint":false},{"year":2022,"finding":"In HCT116 cells deleted for UPF3B, UPF3A strongly activates NMD; in cells lacking both UPF3A and UPF3B, NMD is only partially active. Complementation studies show the EJC-binding domain of UPF3 paralogs is dispensable for NMD; instead, the conserved 'mid' domain is consequential for NMD activity. UPF3A can activate NMD independently of EJC binding.","method":"CRISPR knockout cell lines; NMD reporter assays; RNA-seq; complementation with domain mutants","journal":"The EMBO journal","confidence":"High","confidence_rationale":"Tier 2 / Strong — CRISPR KO with complementation and domain mutants, RNA-seq, corroborated by independent parallel study","pmids":["35451102"],"is_preprint":false},{"year":2022,"finding":"Co-depletion of UPF3A and UPF3B (but not single depletion of either) results in marked NMD inhibition and transcriptome-wide upregulation of NMD substrates, demonstrating functional redundancy between UPF3A and UPF3B. Rescue experiments show UPF2-binding or EJC-binding-deficient UPF3B largely retains NMD activity, but deletion of the middle domain combined with other mutations synergistically impairs NMD.","method":"siRNA knockdown and CRISPR knockout; RNA-seq; rescue experiments with domain mutants","journal":"The EMBO journal","confidence":"High","confidence_rationale":"Tier 2 / Strong — combinatorial KO with domain dissection rescue, RNA-seq, corroborated by independent parallel study in same journal issue","pmids":["35451084"],"is_preprint":false},{"year":2023,"finding":"In zebrafish leg1 deleterious mutants, Upf3a (but not Upf1) is essential for the homology-dependent genetic compensation response (HDGCR) induced by nonsense mutations; this occurs in an H3K4me3-independent manner. Upf3a is also responsible for correcting the expression of hundreds of genes dysregulated in leg1 mutants.","method":"Zebrafish single and double knockout mutants; RNA-seq (71 samples); ULI-NChIP-seq for H3K4me3; genetic epistasis","journal":"Cell discovery","confidence":"High","confidence_rationale":"Tier 2 / Strong — multiple single and double KO lines with RNA-seq and ChIP-seq, large sample size, genetic epistasis","pmids":["37369707"],"is_preprint":false},{"year":2023,"finding":"In mouse embryonic stem cells, somatic cells, and major organs (liver, spleen, thymus), UPF3A is dispensable for NMD when UPF3B is present; UPF3A may weakly and selectively promote NMD in certain murine organs. UPF3A does not repress NMD in these contexts.","method":"Conditional knockout mouse (Upf3a); qRT-PCR and RNA analysis of 33 NMD targets in multiple cell lines and organs","journal":"Life science alliance","confidence":"Medium","confidence_rationale":"Tier 2 / Moderate — conditional KO mouse with analysis of multiple NMD targets across cell types and organs, single lab","pmids":["36997282"],"is_preprint":false}],"current_model":"UPF3A is a paralog of UPF3B that functions as a context-dependent modulator of nonsense-mediated mRNA decay (NMD): it competes with UPF3B for binding to UPF2 (with ~10-fold higher affinity) via shared RRM-L and NOPS-L structural domains, acts as a potent NMD inhibitor when UPF3B is present (by displacing the stronger activator), yet can activate NMD independently of EJC binding—through its conserved mid domain—when UPF3B is absent; additionally, UPF3A plays a distinct NMD-independent role in the genetic compensation response (GCR), where PTC-bearing mRNA recruits Upf3a together with COMPASS components to transcriptionally upregulate homologous compensatory genes, and UPF3A protein levels are themselves regulated post-transcriptionally by a UPF3B-dependent destabilization mechanism."},"narrative":{"mechanistic_narrative":"UPF3A is a paralog of UPF3B that acts as a context-dependent modulator of nonsense-mediated mRNA decay (NMD), a nuclear-and-cytoplasmic surveillance pathway in which it bridges the exon-junction complex (EJC) to the core UPF machinery [PMID:11113196, PMID:16601204]. UPF3A engages UPF2 through shared structural elements—an RRM-like domain, a NONA/paraspeckle-like (NOPS-L) domain, and an extended α-helical region that also support RNA/ribosome binding and RNA-induced oligomerization—and binds UPF2 with ~10-fold higher affinity than UPF3B, competing for the same interface [PMID:35640974]. Because UPF3A is intrinsically a weaker NMD activator than UPF3B, this competition allows UPF3A to behave as a potent NMD inhibitor that stabilizes hundreds of transcripts when UPF3B is present; conditional loss of UPF3A in mice causes hyperactive NMD and defects in embryogenesis and gametogenesis [PMID:27040500]. When UPF3B is absent, however, UPF3A switches to an activator, driving NMD through its conserved 'mid' domain independently of EJC binding, and the two paralogs are functionally redundant such that only co-depletion strongly blocks NMD [PMID:35451102, PMID:35451084]. This activity is held in balance by a post-transcriptional switch in which UPF3B promotes destabilization of UPF3A protein, so that UPF3A levels rise to partially compensate only when UPF3B is lost [PMID:19503078]. Beyond canonical NMD, UPF3A is required—together with COMPASS components such as Wdr5—for the genetic compensation response triggered by premature-termination-codon-bearing mRNAs, transcriptionally upregulating homologous genes via enhanced H3K4me3, while a distinct homology-dependent compensation role in zebrafish operates in an H3K4me3-independent manner [PMID:30944473, PMID:37369707].","teleology":[{"year":2001,"claim":"Established UPF3A as a human NMD-pathway component by showing it, like UPF3B, binds the central adaptor UPF2 and shuttles between nucleus and cytoplasm, framing NMD as a process with both compartmental phases.","evidence":"Reciprocal Co-IP of epitope-tagged proteins with domain mapping and immunofluorescence in HeLa cells","pmids":["11113196"],"confidence":"Medium","gaps":["Did not resolve why UPF3A and UPF3B differ functionally","No quantitative comparison of UPF2-binding affinity"]},{"year":2004,"claim":"Defined the atomic basis of the UPF2-UPF3 interaction, showing the UPF3 RNP/RRM-like domain docks onto UPF2 via conserved charged residues and that this domain does not itself bind RNA, clarifying the architecture shared by UPF3A.","evidence":"1.95 Å crystal structure of UPF2 MIF4G–UPF3B RNP complex plus RNA-binding assays","pmids":["15004547"],"confidence":"High","gaps":["Structure was of UPF3B, not UPF3A directly","Did not explain functional divergence of the paralogs"]},{"year":2006,"claim":"Showed UPF3A is intrinsically a weaker NMD inducer than UPF3B and localized the discriminating function to a C-terminal EJC-binding region, separating EJC-dependent NMD induction from EJC-independent translation stimulation.","evidence":"lambdaN/boxB tethering reporter assays with EJC-interaction Co-IP mapping","pmids":["16601204"],"confidence":"Medium","gaps":["Tethering bypasses normal mRNP recruitment","Did not establish physiological consequence of weak UPF3A activity"]},{"year":2007,"claim":"Reconstituted the heptameric EJC–UPF1–UPF2–UPF3 complex in vitro and showed UPF2/UPF3 cooperatively stimulate UPF1 ATPase and helicase activity, placing UPF3 mechanistically as an activator of the central NMD enzyme.","evidence":"In vitro reconstitution with recombinant proteins plus ATPase and RNA helicase assays","pmids":["18066079"],"confidence":"High","gaps":["Assays used UPF3B; UPF3A-specific enzymatic stimulation not measured","In vitro stoichiometry may not reflect cellular complexes"]},{"year":2009,"claim":"Revealed a post-transcriptional regulatory switch in which UPF3B destabilizes UPF3A protein, so UPF3A rises and partially compensates only when UPF3B is lost—while simultaneously impairing NMD through UPF2 competition.","evidence":"Western blotting in UPF3B-mutant patient cells, competitive binding and NMD reporter assays","pmids":["19503078"],"confidence":"High","gaps":["Molecular mechanism of UPF3A destabilization not identified","Did not quantify net NMD outcome in normal cells"]},{"year":2016,"claim":"Reframed UPF3A as primarily a potent NMD inhibitor whose loss causes hyper-NMD and developmental defects, attributing its repressor role to a weakened EJC-binding domain that lets it displace the stronger activator UPF3B.","evidence":"siRNA/shRNA, conditional knockout mouse, RNA-seq, NMD reporters and Co-IP","pmids":["27040500"],"confidence":"High","gaps":["Did not test UPF3A function in the complete absence of UPF3B","Mechanism reconciling inhibition with later activator findings unresolved"]},{"year":2019,"claim":"Identified an NMD-independent role for Upf3a in the genetic compensation response, linking PTC-bearing mRNA to COMPASS/Wdr5-mediated H3K4me3 deposition and transcriptional upregulation of homologous genes.","evidence":"Zebrafish knockdown-vs-knockout models, transgenes, H3K4me3 ChIP and genetic epistasis","pmids":["30944473"],"confidence":"High","gaps":["How Upf3a couples PTC mRNA to chromatin modifiers is unknown","Conservation of GCR mechanism in mammals not established"]},{"year":2020,"claim":"Indicated UPF3A influences pathways beyond NMD, with reconstitution altering cholesterol-biosynthesis enzyme levels and the phosphorylation of nuclear gene-expression/splicing regulators.","evidence":"SILAC quantitative proteomics and phosphoproteomics in UPF3A-reconstituted CRC cells","pmids":["32718059"],"confidence":"Medium","gaps":["Single CRC cell line, single lab","Direct versus indirect effects of UPF3A not distinguished"]},{"year":2022,"claim":"Provided the quantitative and structural basis for paralog competition: crystal structures of UPF2 bound to either UPF3A or UPF3B showed shared interfaces, UPF3A binds UPF2 ~10-fold tighter, and a disease UPF3B mutation in NOPS-L weakens UPF2 binding ~40-fold.","evidence":"X-ray crystallography, affinity measurements, mutagenesis and RNA/ribosome-binding assays","pmids":["35640974"],"confidence":"High","gaps":["Did not resolve how tighter UPF2 binding maps onto inhibitor-versus-activator behavior","In-cell stoichiometry of competition not measured"]},{"year":2022,"claim":"Resolved the activator/inhibitor paradox by showing UPF3A strongly activates NMD when UPF3B is deleted and that the EJC-binding domain is dispensable while the conserved 'mid' domain drives activity, with the two paralogs functionally redundant.","evidence":"CRISPR knockout and combinatorial co-depletion, NMD reporters, RNA-seq and domain-mutant complementation across two parallel studies","pmids":["35451102","35451084"],"confidence":"High","gaps":["Biochemical function of the 'mid' domain undefined","Why outcomes differ between cell systems not fully explained"]},{"year":2023,"claim":"Refined the in vivo NMD picture and the compensation mechanism: UPF3A is largely dispensable for NMD when UPF3B is present in mouse, and in zebrafish leg1 mutants Upf3a drives homology-dependent compensation in an H3K4me3-independent manner.","evidence":"Conditional knockout mice with NMD-target qRT-PCR; zebrafish single/double knockouts with RNA-seq and ULI-NChIP-seq","pmids":["36997282","37369707"],"confidence":"Medium","gaps":["Reconciliation of H3K4me3-dependent and -independent compensation modes unresolved","Molecular effectors of homology-dependent correction unknown"]},{"year":null,"claim":"The molecular mechanism by which the conserved 'mid' domain activates NMD, and how UPF3A links PTC-containing mRNA to transcriptional compensation, remain unresolved.","evidence":"","pmids":[],"confidence":"High","gaps":["No biochemical activity assigned to the 'mid' domain","No identified molecular bridge from UPF3A to COMPASS/chromatin in the GCR","Mechanism of UPF3B-dependent UPF3A destabilization unknown"]}],"mechanism_profile":{"molecular_activity":[{"term_id":"GO:0003723","term_label":"RNA binding","supporting_discovery_ids":[8]},{"term_id":"GO:0098772","term_label":"molecular function regulator activity","supporting_discovery_ids":[4,5,9]},{"term_id":"GO:0060090","term_label":"molecular adaptor activity","supporting_discovery_ids":[0,2,3]}],"localization":[{"term_id":"GO:0005634","term_label":"nucleus","supporting_discovery_ids":[0]},{"term_id":"GO:0005829","term_label":"cytosol","supporting_discovery_ids":[0]}],"pathway":[{"term_id":"R-HSA-8953854","term_label":"Metabolism of RNA","supporting_discovery_ids":[3,5,9,10]},{"term_id":"R-HSA-74160","term_label":"Gene expression (Transcription)","supporting_discovery_ids":[6,11]}],"complexes":["EJC-UPF surveillance complex (UPF1-UPF2-UPF3-EJC)"],"partners":["UPF2","UPF3B","UPF1","EIF4A3","MAGOH","RBM8A","WDR5"],"other_free_text":[]}},"prefetch_data":{"uniprot":{"accession":"Q9H1J1","full_name":"Regulator of nonsense transcripts 3A","aliases":["Nonsense mRNA reducing factor 3A","Up-frameshift suppressor 3 homolog A","hUpf3"],"length_aa":476,"mass_kda":54.7,"function":"Involved in nonsense-mediated decay (NMD) of mRNAs containing premature stop codons by associating with the nuclear exon junction complex (EJC) and serving as link between the EJC core and NMD machinery. Recruits UPF2 at the cytoplasmic side of the nuclear envelope and the subsequent formation of an UPF1-UPF2-UPF3 surveillance complex (including UPF1 bound to release factors at the stalled ribosome) is believed to activate NMD. However, UPF3A is shown to be only marginally active in NMD as compared to UPF3B. Binds spliced mRNA upstream of exon-exon junctions. In vitro, weakly stimulates translation","subcellular_location":"Nucleus; Cytoplasm","url":"https://www.uniprot.org/uniprotkb/Q9H1J1/entry"},"depmap":{"release":"DepMap","has_data":true,"is_common_essential":false,"resolved_as":"","url":"https://depmap.org/portal/gene/UPF3A","classification":"Not Classified","n_dependent_lines":51,"n_total_lines":1208,"dependency_fraction":0.042218543046357616},"opencell":{"profiled":false,"resolved_as":"","ensg_id":"","cell_line_id":"","localizations":[],"interactors":[{"gene":"SSRP1","stoichiometry":0.2}],"url":"https://opencell.sf.czbiohub.org/search/UPF3A","total_profiled":1310},"omim":[{"mim_id":"609012","title":"WD REPEAT-CONTAINING PROTEIN 5; WDR5","url":"https://www.omim.org/entry/609012"},{"mim_id":"607032","title":"SMG1 NONSENSE-MEDIATED mRNA DECAY-ASSOCIATED PI3K-RELATED KINASE; SMG1","url":"https://www.omim.org/entry/607032"},{"mim_id":"606447","title":"RNA-BINDING PROTEIN S1; RNPS1","url":"https://www.omim.org/entry/606447"},{"mim_id":"605530","title":"UPF3A REGULATOR OF NONSENSE-MEDIATED mRNA DECAY; UPF3A","url":"https://www.omim.org/entry/605530"},{"mim_id":"605529","title":"UPF2 REGULATOR OF NONSENSE-MEDIATED mRNA DECAY; UPF2","url":"https://www.omim.org/entry/605529"}],"hpa":{"profiled":true,"resolved_as":"","reliability":"Supported","locations":[{"location":"Nucleoplasm","reliability":"Supported"},{"location":"Vesicles","reliability":"Supported"},{"location":"Plasma membrane","reliability":"Supported"},{"location":"Primary cilium","reliability":"Additional"},{"location":"Primary cilium tip","reliability":"Additional"},{"location":"Basal body","reliability":"Additional"}],"tissue_specificity":"Low tissue specificity","tissue_distribution":"Detected in all","driving_tissues":[],"url":"https://www.proteinatlas.org/search/UPF3A"},"hgnc":{"alias_symbol":["RENT3A","UPF3","HUPF3A"],"prev_symbol":[]},"alphafold":{"accession":"Q9H1J1","domains":[{"cath_id":"3.30.70.330","chopping":"70-151","consensus_level":"high","plddt":90.4766,"start":70,"end":151},{"cath_id":"-","chopping":"172-214","consensus_level":"high","plddt":78.134,"start":172,"end":214},{"cath_id":"1.20.5","chopping":"216-284","consensus_level":"medium","plddt":74.8246,"start":216,"end":284}],"viewer_url":"https://alphafold.ebi.ac.uk/entry/Q9H1J1","model_url":"https://alphafold.ebi.ac.uk/files/AF-Q9H1J1-F1-model_v6.cif","pae_url":"https://alphafold.ebi.ac.uk/files/AF-Q9H1J1-F1-predicted_aligned_error_v6.png","plddt_mean":62.16},"mouse_models":{"mgi_url":"https://www.informatics.jax.org/marker/summary?nomen=UPF3A","jax_strain_url":"https://www.jax.org/strain/search?query=UPF3A"},"sequence":{"accession":"Q9H1J1","fasta_url":"https://rest.uniprot.org/uniprotkb/Q9H1J1.fasta","uniprot_url":"https://www.uniprot.org/uniprotkb/Q9H1J1/entry","alphafold_viewer_url":"https://alphafold.ebi.ac.uk/entry/Q9H1J1"}},"corpus_meta":[{"pmid":"30944473","id":"PMC_30944473","title":"PTC-bearing mRNA elicits a genetic compensation response via Upf3a and COMPASS components.","date":"2019","source":"Nature","url":"https://pubmed.ncbi.nlm.nih.gov/30944473","citation_count":375,"is_preprint":false},{"pmid":"18066079","id":"PMC_18066079","title":"NMD factors UPF2 and UPF3 bridge UPF1 to the exon junction complex and stimulate its RNA helicase activity.","date":"2007","source":"Nature structural & molecular biology","url":"https://pubmed.ncbi.nlm.nih.gov/18066079","citation_count":274,"is_preprint":false},{"pmid":"11113196","id":"PMC_11113196","title":"Identification and characterization of human orthologues to Saccharomyces cerevisiae Upf2 protein and Upf3 protein (Caenorhabditis elegans SMG-4).","date":"2001","source":"Molecular and cellular biology","url":"https://pubmed.ncbi.nlm.nih.gov/11113196","citation_count":208,"is_preprint":false},{"pmid":"15004547","id":"PMC_15004547","title":"The structural basis for the interaction between nonsense-mediated mRNA decay factors UPF2 and UPF3.","date":"2004","source":"Nature structural & molecular biology","url":"https://pubmed.ncbi.nlm.nih.gov/15004547","citation_count":166,"is_preprint":false},{"pmid":"27040500","id":"PMC_27040500","title":"The Antagonistic Gene Paralogs Upf3a and Upf3b Govern Nonsense-Mediated RNA Decay.","date":"2016","source":"Cell","url":"https://pubmed.ncbi.nlm.nih.gov/27040500","citation_count":133,"is_preprint":false},{"pmid":"16601204","id":"PMC_16601204","title":"Functions of hUpf3a and hUpf3b in nonsense-mediated mRNA decay and translation.","date":"2006","source":"RNA (New York, N.Y.)","url":"https://pubmed.ncbi.nlm.nih.gov/16601204","citation_count":110,"is_preprint":false},{"pmid":"19503078","id":"PMC_19503078","title":"A UPF3-mediated regulatory switch that maintains RNA surveillance.","date":"2009","source":"Nature structural & molecular biology","url":"https://pubmed.ncbi.nlm.nih.gov/19503078","citation_count":101,"is_preprint":false},{"pmid":"22025558","id":"PMC_22025558","title":"Nonsense-mediated mRNA decay factors, UPF1 and UPF3, contribute to plant defense.","date":"2011","source":"Plant & cell physiology","url":"https://pubmed.ncbi.nlm.nih.gov/22025558","citation_count":61,"is_preprint":false},{"pmid":"21317294","id":"PMC_21317294","title":"Drosophila Upf1 and Upf2 loss of function inhibits cell growth and causes animal death in a Upf3-independent manner.","date":"2011","source":"RNA (New York, N.Y.)","url":"https://pubmed.ncbi.nlm.nih.gov/21317294","citation_count":54,"is_preprint":false},{"pmid":"18676617","id":"PMC_18676617","title":"Upf1 potentially serves as a RING-related E3 ubiquitin ligase via its association with Upf3 in yeast.","date":"2008","source":"RNA (New York, N.Y.)","url":"https://pubmed.ncbi.nlm.nih.gov/18676617","citation_count":46,"is_preprint":false},{"pmid":"35451102","id":"PMC_35451102","title":"Mammalian UPF3A and UPF3B can activate nonsense-mediated mRNA decay independently of their exon junction complex binding.","date":"2022","source":"The EMBO journal","url":"https://pubmed.ncbi.nlm.nih.gov/35451102","citation_count":43,"is_preprint":false},{"pmid":"35451084","id":"PMC_35451084","title":"Human UPF3A and UPF3B enable fault-tolerant activation of nonsense-mediated mRNA decay.","date":"2022","source":"The EMBO journal","url":"https://pubmed.ncbi.nlm.nih.gov/35451084","citation_count":36,"is_preprint":false},{"pmid":"9671745","id":"PMC_9671745","title":"The upf3 protein is a component of the surveillance complex that monitors both translation and mRNA turnover and affects viral propagation.","date":"1998","source":"Proceedings of the National Academy of Sciences of the United States of America","url":"https://pubmed.ncbi.nlm.nih.gov/9671745","citation_count":35,"is_preprint":false},{"pmid":"25820429","id":"PMC_25820429","title":"The feedback control of UPF3 is crucial for RNA surveillance in plants.","date":"2015","source":"Nucleic acids research","url":"https://pubmed.ncbi.nlm.nih.gov/25820429","citation_count":22,"is_preprint":false},{"pmid":"37369707","id":"PMC_37369707","title":"Upf3a but not Upf1 mediates the genetic compensation response induced by leg1 deleterious mutations in an H3K4me3-independent manner.","date":"2023","source":"Cell discovery","url":"https://pubmed.ncbi.nlm.nih.gov/37369707","citation_count":20,"is_preprint":false},{"pmid":"35640974","id":"PMC_35640974","title":"Structures of nonsense-mediated mRNA decay factors UPF3B and UPF3A in complex with UPF2 reveal molecular basis for competitive binding and for neurodevelopmental disorder-causing mutation.","date":"2022","source":"Nucleic acids research","url":"https://pubmed.ncbi.nlm.nih.gov/35640974","citation_count":17,"is_preprint":false},{"pmid":"37794510","id":"PMC_37794510","title":"Serum methylation of GALNT9, UPF3A, WARS, and LDB2 as noninvasive biomarkers for the early detection of colorectal cancer and advanced adenomas.","date":"2023","source":"Clinical epigenetics","url":"https://pubmed.ncbi.nlm.nih.gov/37794510","citation_count":14,"is_preprint":false},{"pmid":"12172963","id":"PMC_12172963","title":"Role of RNA surveillance proteins Upf1/CpaR, Upf2 and Upf3 in the translational regulation of yeast CPA1 gene.","date":"2002","source":"Current genetics","url":"https://pubmed.ncbi.nlm.nih.gov/12172963","citation_count":14,"is_preprint":false},{"pmid":"33132159","id":"PMC_33132159","title":"Functional roles of human Up-frameshift suppressor 3 (UPF3) proteins: From nonsense-mediated mRNA decay to neurodevelopmental disorders.","date":"2020","source":"Biochimie","url":"https://pubmed.ncbi.nlm.nih.gov/33132159","citation_count":11,"is_preprint":false},{"pmid":"36997282","id":"PMC_36997282","title":"UPF3A is dispensable for nonsense-mediated mRNA decay in mouse pluripotent and somatic cells.","date":"2023","source":"Life science alliance","url":"https://pubmed.ncbi.nlm.nih.gov/36997282","citation_count":10,"is_preprint":false},{"pmid":"32606924","id":"PMC_32606924","title":"Functions and Clinical Significance of UPF3a Expression in Human Colorectal Cancer.","date":"2020","source":"Cancer management and research","url":"https://pubmed.ncbi.nlm.nih.gov/32606924","citation_count":10,"is_preprint":false},{"pmid":"32718059","id":"PMC_32718059","title":"(Phospho)proteomic Profiling of Microsatellite Unstable CRC Cells Reveals Alterations in Nuclear Signaling and Cholesterol Metabolism Caused by Frameshift Mutation of NMD Regulator UPF3A.","date":"2020","source":"International journal of molecular sciences","url":"https://pubmed.ncbi.nlm.nih.gov/32718059","citation_count":5,"is_preprint":false},{"pmid":"36807382","id":"PMC_36807382","title":"Genetic compensation response could exist in colorectal cancer: UPF3A upregulates the oncogenic homologue gene SRSF3 expression corresponding to SRSF6 to promote colorectal cancer metastasis.","date":"2023","source":"Journal of gastroenterology and hepatology","url":"https://pubmed.ncbi.nlm.nih.gov/36807382","citation_count":4,"is_preprint":false},{"pmid":"25463387","id":"PMC_25463387","title":"An UPF3-based nonsense-mediated decay in Paramecium.","date":"2014","source":"Research in microbiology","url":"https://pubmed.ncbi.nlm.nih.gov/25463387","citation_count":4,"is_preprint":false},{"pmid":"37961809","id":"PMC_37961809","title":"Spatial expression of the nonsense-mediated mRNA decay factors UPF3A and UPF3B among mouse tissues.","date":"2023","source":"Journal of Zhejiang University. Science. B","url":"https://pubmed.ncbi.nlm.nih.gov/37961809","citation_count":2,"is_preprint":false},{"pmid":"41668711","id":"PMC_41668711","title":"The nonsense-mediated mRNA decay factor Upf3 negatively regulates bulk autophagy progression in Saccharomyces cerevisiae.","date":"2026","source":"Autophagy reports","url":"https://pubmed.ncbi.nlm.nih.gov/41668711","citation_count":0,"is_preprint":false},{"pmid":null,"id":"bio_10.1101_2025.02.21.639497","title":"Most Upf1-associated mRNAs have short poly(A) tails, lack a premature termination codon and are targeted by NMD","date":"2025-02-27","source":"bioRxiv","url":"https://doi.org/10.1101/2025.02.21.639497","citation_count":0,"is_preprint":true},{"pmid":null,"id":"bio_10.1101_2024.12.18.629111","title":"Mapping interactions within the NMD decapping complex reveals how Upf1 recruits Dcp2 onto mRNA targets","date":"2025-01-04","source":"bioRxiv","url":"https://doi.org/10.1101/2024.12.18.629111","citation_count":0,"is_preprint":true},{"pmid":null,"id":"bio_10.1101_2024.10.26.620407","title":"Dynamic RNA binding and unfolding by nonsense-mediated mRNA decay factor UPF2","date":"2024-10-28","source":"bioRxiv","url":"https://doi.org/10.1101/2024.10.26.620407","citation_count":0,"is_preprint":true}],"cost":{"stage1":{"model":"claude-sonnet-4-6","input_tokens":15841,"output_tokens":3704,"usd":0.051541,"stage1_stop_reason":"end_turn"},"stage2":{"model":"claude-opus-4-8","input_tokens":11251,"output_tokens":4161,"usd":0.08014,"stage2_stop_reason":"end_turn"},"total_usd":0.131681,"stage1_batch_id":"msgbatch_017qqVz33Ga63SCDwzb35XPb","stage2_batch_id":"msgbatch_01Scr2Ei9x4HJMiRGxwrmvLK","note":"batch pricing = 50% of standard"},"stage1_raw":"```json\n{\n  \"discoveries\": [\n    {\n      \"year\": 2001,\n      \"finding\": \"Human UPF3A (hUpf3p) was identified as a human orthologue of S. cerevisiae Upf3p; co-immunoprecipitation of epitope-tagged proteins in HeLa cells demonstrated that hUpf2p interacts with both hUpf3p-X (UPF3B) and hUpf3p (UPF3A), and the domains required for these interactions were defined. UPF3A localizes primarily to nuclei and is a shuttling protein, indicating NMD has both nuclear and cytoplasmic components.\",\n      \"method\": \"Co-immunoprecipitation of epitope-tagged proteins in HeLa cells; indirect immunofluorescence for subcellular localization\",\n      \"journal\": \"Molecular and cellular biology\",\n      \"confidence\": \"Medium\",\n      \"confidence_rationale\": \"Tier 2 / Moderate — reciprocal Co-IP with domain mapping and localization by immunofluorescence, single lab\",\n      \"pmids\": [\"11113196\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2004,\n      \"finding\": \"Crystal structure at 1.95 Å of the complex between the interacting domains of human UPF2 (MIF4G domain) and UPF3B (RNP domain) revealed that the protein-protein interface is mediated by highly conserved charged residues and involves the beta-sheet surface of the UPF3B RNP domain. The UPF3b RNP domain does not bind RNA, whereas the UPF2 construct and the complex do. The same RNP domain architecture is shared by UPF3A.\",\n      \"method\": \"X-ray crystallography (1.95 Å resolution); RNA-binding assays\",\n      \"journal\": \"Nature structural & molecular biology\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 1 / Strong — crystal structure with functional validation of RNA-binding; foundational structural result replicated by later studies\",\n      \"pmids\": [\"15004547\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2006,\n      \"finding\": \"Using a tethering system, UPF3A was shown to be much less active than UPF3B in inducing NMD and stimulating translation. The C-terminal domain region that discriminates UPF3A from UPF3B in NMD function mediates interaction with EJC components Y14, Magoh, BTZ, and eIF4AIII; this interaction is required for NMD induction. Translation stimulation is independent of EJC interaction and is determined by other regions of the UPF3 proteins.\",\n      \"method\": \"lambdaN/boxB tethering reporter assay; co-immunoprecipitation for EJC interaction mapping\",\n      \"journal\": \"RNA (New York, N.Y.)\",\n      \"confidence\": \"Medium\",\n      \"confidence_rationale\": \"Tier 2 / Moderate — functional tethering assay with domain dissection, single lab, two orthogonal approaches\",\n      \"pmids\": [\"16601204\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2007,\n      \"finding\": \"Recombinant EJC core was sufficient to reconstitute a stable heptameric complex on RNA with UPF1, UPF2, and UPF3b. EJC proteins MAGOH, Y14, and eIF4AIII provide a composite binding site for UPF3b that bridges to UPF2 and UPF1. In the trimeric UPF complex, UPF2 and UPF3b cooperatively stimulate both ATPase and RNA helicase activities of UPF1.\",\n      \"method\": \"In vitro reconstitution; ATPase assay; RNA helicase assay with recombinant proteins\",\n      \"journal\": \"Nature structural & molecular biology\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 1 / Strong — in vitro reconstitution of heptameric complex with enzymatic assays, multiple orthogonal methods\",\n      \"pmids\": [\"18066079\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2009,\n      \"finding\": \"UPF3A protein levels are regulated post-transcriptionally: the presence of UPF3B promotes destabilization/reduction of UPF3A levels via a conserved UPF3B-dependent mechanism. When UPF3B is absent (e.g., due to mutations), UPF3A levels rise and partially compensate for NMD, but UPF3A also impairs NMD by competing with UPF3B for binding to UPF2. This UPF3B-dependent destabilization of UPF3A constitutes a post-transcriptional regulatory switch maintaining appropriate NMD levels.\",\n      \"method\": \"Western blotting of UPF3A/UPF3B levels in cells with UPF3B mutations; competitive binding assays; NMD reporter assays\",\n      \"journal\": \"Nature structural & molecular biology\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 2 / Strong — reciprocal protein-level regulation demonstrated in multiple patient-derived cell lines and knockdown experiments, multiple orthogonal readouts\",\n      \"pmids\": [\"19503078\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2016,\n      \"finding\": \"UPF3A acts primarily as a potent NMD inhibitor, stabilizing hundreds of transcripts, rather than an NMD activator. It acquired repressor activity through impairment of a critical domain (C-terminal EJC-binding domain is weakened relative to UPF3B). Conditional knockout of UPF3A in mice causes 'hyper' NMD and leads to defects in embryogenesis and gametogenesis. UPF3A competes with UPF3B for binding to UPF2, and its NMD-inhibitory function is explained by its weaker NMD activation capacity displacing the stronger activator UPF3B.\",\n      \"method\": \"Loss-of-function in vitro (siRNA/shRNA) and in vivo (conditional knockout mouse); RNA-seq; NMD reporter assays; co-immunoprecipitation\",\n      \"journal\": \"Cell\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 2 / Strong — multiple orthogonal methods in vitro and in vivo, conditional KO mouse with defined phenotypes, replicated by later studies\",\n      \"pmids\": [\"27040500\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2019,\n      \"finding\": \"In zebrafish, Upf3a (a NMD pathway member) and components of the COMPASS complex including Wdr5 are required for the genetic compensation response (GCR) triggered by mRNAs bearing a premature termination codon (PTC). The GCR is accompanied by enhancement of H3K4me3 at the transcription start site regions of compensatory genes, linking Upf3a to transcriptional upregulation of homologous genes in response to PTC-containing mRNA.\",\n      \"method\": \"Zebrafish knockdown vs. knockout models; transgene analysis; ChIP for H3K4me3; genetic epistasis using upf3a mutants\",\n      \"journal\": \"Nature\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 2 / Strong — genetic epistasis with multiple designed transgenes, ChIP-seq, replicated across two gene models (capn3a, nid1a)\",\n      \"pmids\": [\"30944473\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2020,\n      \"finding\": \"Reconstituted UPF3A expression in KM12 CRC cells (which have a frameshift mutation in UPF3A) caused down-regulation of several enzymes involved in cholesterol biosynthesis and altered phosphorylation of 85 phosphosites in 52 phosphoproteins, predominantly nuclear proteins involved in gene expression regulation and RNA splicing, suggesting UPF3A influences cellular signaling pathways beyond NMD.\",\n      \"method\": \"SILAC-based quantitative proteomics; phosphoproteomics in CRC cell lines with reconstituted UPF3A expression\",\n      \"journal\": \"International journal of molecular sciences\",\n      \"confidence\": \"Medium\",\n      \"confidence_rationale\": \"Tier 2 / Weak — quantitative proteomics with reconstitution in a single CRC cell line, single lab\",\n      \"pmids\": [\"32718059\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2022,\n      \"finding\": \"UPF3A and UPF3B share structural homology comprising an RRM-like domain (RRM-L), a NONA/paraspeckle-like domain (NOPS-L), and an extended α-helical domain; these domains are essential for RNA/ribosome-binding, RNA-induced oligomerization, and UPF2 interaction. Crystal structures of UPF2's MIF4GIII domain in complex with UPF3B or UPF3A revealed intimate binding interfaces. UPF3A binds UPF2 with ~10-fold higher affinity than UPF3B. The disease-causing UPF3B mutation Y160D in the NOPS-L domain displaces Y160 from a hydrophobic cleft in UPF2, reducing binding affinity ~40-fold. UPF3A and UPF3B compete for the same UPF2 binding site.\",\n      \"method\": \"X-ray crystallography; binding affinity measurements; mutagenesis; RNA/ribosome binding assays\",\n      \"journal\": \"Nucleic acids research\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 1 / Strong — crystal structures of both paralogs with UPF2, mutagenesis validation, quantitative binding measurements, multiple orthogonal methods\",\n      \"pmids\": [\"35640974\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2022,\n      \"finding\": \"In HCT116 cells deleted for UPF3B, UPF3A strongly activates NMD; in cells lacking both UPF3A and UPF3B, NMD is only partially active. Complementation studies show the EJC-binding domain of UPF3 paralogs is dispensable for NMD; instead, the conserved 'mid' domain is consequential for NMD activity. UPF3A can activate NMD independently of EJC binding.\",\n      \"method\": \"CRISPR knockout cell lines; NMD reporter assays; RNA-seq; complementation with domain mutants\",\n      \"journal\": \"The EMBO journal\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 2 / Strong — CRISPR KO with complementation and domain mutants, RNA-seq, corroborated by independent parallel study\",\n      \"pmids\": [\"35451102\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2022,\n      \"finding\": \"Co-depletion of UPF3A and UPF3B (but not single depletion of either) results in marked NMD inhibition and transcriptome-wide upregulation of NMD substrates, demonstrating functional redundancy between UPF3A and UPF3B. Rescue experiments show UPF2-binding or EJC-binding-deficient UPF3B largely retains NMD activity, but deletion of the middle domain combined with other mutations synergistically impairs NMD.\",\n      \"method\": \"siRNA knockdown and CRISPR knockout; RNA-seq; rescue experiments with domain mutants\",\n      \"journal\": \"The EMBO journal\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 2 / Strong — combinatorial KO with domain dissection rescue, RNA-seq, corroborated by independent parallel study in same journal issue\",\n      \"pmids\": [\"35451084\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2023,\n      \"finding\": \"In zebrafish leg1 deleterious mutants, Upf3a (but not Upf1) is essential for the homology-dependent genetic compensation response (HDGCR) induced by nonsense mutations; this occurs in an H3K4me3-independent manner. Upf3a is also responsible for correcting the expression of hundreds of genes dysregulated in leg1 mutants.\",\n      \"method\": \"Zebrafish single and double knockout mutants; RNA-seq (71 samples); ULI-NChIP-seq for H3K4me3; genetic epistasis\",\n      \"journal\": \"Cell discovery\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 2 / Strong — multiple single and double KO lines with RNA-seq and ChIP-seq, large sample size, genetic epistasis\",\n      \"pmids\": [\"37369707\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2023,\n      \"finding\": \"In mouse embryonic stem cells, somatic cells, and major organs (liver, spleen, thymus), UPF3A is dispensable for NMD when UPF3B is present; UPF3A may weakly and selectively promote NMD in certain murine organs. UPF3A does not repress NMD in these contexts.\",\n      \"method\": \"Conditional knockout mouse (Upf3a); qRT-PCR and RNA analysis of 33 NMD targets in multiple cell lines and organs\",\n      \"journal\": \"Life science alliance\",\n      \"confidence\": \"Medium\",\n      \"confidence_rationale\": \"Tier 2 / Moderate — conditional KO mouse with analysis of multiple NMD targets across cell types and organs, single lab\",\n      \"pmids\": [\"36997282\"],\n      \"is_preprint\": false\n    }\n  ],\n  \"current_model\": \"UPF3A is a paralog of UPF3B that functions as a context-dependent modulator of nonsense-mediated mRNA decay (NMD): it competes with UPF3B for binding to UPF2 (with ~10-fold higher affinity) via shared RRM-L and NOPS-L structural domains, acts as a potent NMD inhibitor when UPF3B is present (by displacing the stronger activator), yet can activate NMD independently of EJC binding—through its conserved mid domain—when UPF3B is absent; additionally, UPF3A plays a distinct NMD-independent role in the genetic compensation response (GCR), where PTC-bearing mRNA recruits Upf3a together with COMPASS components to transcriptionally upregulate homologous compensatory genes, and UPF3A protein levels are themselves regulated post-transcriptionally by a UPF3B-dependent destabilization mechanism.\"\n}\n```","stage2_raw":"{\n  \"mechanistic_narrative\": \"UPF3A is a paralog of UPF3B that acts as a context-dependent modulator of nonsense-mediated mRNA decay (NMD), a nuclear-and-cytoplasmic surveillance pathway in which it bridges the exon-junction complex (EJC) to the core UPF machinery [#0, #2]. UPF3A engages UPF2 through shared structural elements\\u2014an RRM-like domain, a NONA/paraspeckle-like (NOPS-L) domain, and an extended \\u03b1-helical region that also support RNA/ribosome binding and RNA-induced oligomerization\\u2014and binds UPF2 with ~10-fold higher affinity than UPF3B, competing for the same interface [#8]. Because UPF3A is intrinsically a weaker NMD activator than UPF3B, this competition allows UPF3A to behave as a potent NMD inhibitor that stabilizes hundreds of transcripts when UPF3B is present; conditional loss of UPF3A in mice causes hyperactive NMD and defects in embryogenesis and gametogenesis [#5]. When UPF3B is absent, however, UPF3A switches to an activator, driving NMD through its conserved 'mid' domain independently of EJC binding, and the two paralogs are functionally redundant such that only co-depletion strongly blocks NMD [#9, #10]. This activity is held in balance by a post-transcriptional switch in which UPF3B promotes destabilization of UPF3A protein, so that UPF3A levels rise to partially compensate only when UPF3B is lost [#4]. Beyond canonical NMD, UPF3A is required\\u2014together with COMPASS components such as Wdr5\\u2014for the genetic compensation response triggered by premature-termination-codon-bearing mRNAs, transcriptionally upregulating homologous genes via enhanced H3K4me3, while a distinct homology-dependent compensation role in zebrafish operates in an H3K4me3-independent manner [#6, #11].\",\n  \"teleology\": [\n    {\n      \"year\": 2001,\n      \"claim\": \"Established UPF3A as a human NMD-pathway component by showing it, like UPF3B, binds the central adaptor UPF2 and shuttles between nucleus and cytoplasm, framing NMD as a process with both compartmental phases.\",\n      \"evidence\": \"Reciprocal Co-IP of epitope-tagged proteins with domain mapping and immunofluorescence in HeLa cells\",\n      \"pmids\": [\"11113196\"],\n      \"confidence\": \"Medium\",\n      \"confidence_rationale\": \"\",\n      \"gaps\": [\"Did not resolve why UPF3A and UPF3B differ functionally\", \"No quantitative comparison of UPF2-binding affinity\"]\n    },\n    {\n      \"year\": 2004,\n      \"claim\": \"Defined the atomic basis of the UPF2-UPF3 interaction, showing the UPF3 RNP/RRM-like domain docks onto UPF2 via conserved charged residues and that this domain does not itself bind RNA, clarifying the architecture shared by UPF3A.\",\n      \"evidence\": \"1.95 \\u00c5 crystal structure of UPF2 MIF4G\\u2013UPF3B RNP complex plus RNA-binding assays\",\n      \"pmids\": [\"15004547\"],\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"\",\n      \"gaps\": [\"Structure was of UPF3B, not UPF3A directly\", \"Did not explain functional divergence of the paralogs\"]\n    },\n    {\n      \"year\": 2006,\n      \"claim\": \"Showed UPF3A is intrinsically a weaker NMD inducer than UPF3B and localized the discriminating function to a C-terminal EJC-binding region, separating EJC-dependent NMD induction from EJC-independent translation stimulation.\",\n      \"evidence\": \"lambdaN/boxB tethering reporter assays with EJC-interaction Co-IP mapping\",\n      \"pmids\": [\"16601204\"],\n      \"confidence\": \"Medium\",\n      \"confidence_rationale\": \"\",\n      \"gaps\": [\"Tethering bypasses normal mRNP recruitment\", \"Did not establish physiological consequence of weak UPF3A activity\"]\n    },\n    {\n      \"year\": 2007,\n      \"claim\": \"Reconstituted the heptameric EJC\\u2013UPF1\\u2013UPF2\\u2013UPF3 complex in vitro and showed UPF2/UPF3 cooperatively stimulate UPF1 ATPase and helicase activity, placing UPF3 mechanistically as an activator of the central NMD enzyme.\",\n      \"evidence\": \"In vitro reconstitution with recombinant proteins plus ATPase and RNA helicase assays\",\n      \"pmids\": [\"18066079\"],\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"\",\n      \"gaps\": [\"Assays used UPF3B; UPF3A-specific enzymatic stimulation not measured\", \"In vitro stoichiometry may not reflect cellular complexes\"]\n    },\n    {\n      \"year\": 2009,\n      \"claim\": \"Revealed a post-transcriptional regulatory switch in which UPF3B destabilizes UPF3A protein, so UPF3A rises and partially compensates only when UPF3B is lost\\u2014while simultaneously impairing NMD through UPF2 competition.\",\n      \"evidence\": \"Western blotting in UPF3B-mutant patient cells, competitive binding and NMD reporter assays\",\n      \"pmids\": [\"19503078\"],\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"\",\n      \"gaps\": [\"Molecular mechanism of UPF3A destabilization not identified\", \"Did not quantify net NMD outcome in normal cells\"]\n    },\n    {\n      \"year\": 2016,\n      \"claim\": \"Reframed UPF3A as primarily a potent NMD inhibitor whose loss causes hyper-NMD and developmental defects, attributing its repressor role to a weakened EJC-binding domain that lets it displace the stronger activator UPF3B.\",\n      \"evidence\": \"siRNA/shRNA, conditional knockout mouse, RNA-seq, NMD reporters and Co-IP\",\n      \"pmids\": [\"27040500\"],\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"\",\n      \"gaps\": [\"Did not test UPF3A function in the complete absence of UPF3B\", \"Mechanism reconciling inhibition with later activator findings unresolved\"]\n    },\n    {\n      \"year\": 2019,\n      \"claim\": \"Identified an NMD-independent role for Upf3a in the genetic compensation response, linking PTC-bearing mRNA to COMPASS/Wdr5-mediated H3K4me3 deposition and transcriptional upregulation of homologous genes.\",\n      \"evidence\": \"Zebrafish knockdown-vs-knockout models, transgenes, H3K4me3 ChIP and genetic epistasis\",\n      \"pmids\": [\"30944473\"],\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"\",\n      \"gaps\": [\"How Upf3a couples PTC mRNA to chromatin modifiers is unknown\", \"Conservation of GCR mechanism in mammals not established\"]\n    },\n    {\n      \"year\": 2020,\n      \"claim\": \"Indicated UPF3A influences pathways beyond NMD, with reconstitution altering cholesterol-biosynthesis enzyme levels and the phosphorylation of nuclear gene-expression/splicing regulators.\",\n      \"evidence\": \"SILAC quantitative proteomics and phosphoproteomics in UPF3A-reconstituted CRC cells\",\n      \"pmids\": [\"32718059\"],\n      \"confidence\": \"Medium\",\n      \"confidence_rationale\": \"\",\n      \"gaps\": [\"Single CRC cell line, single lab\", \"Direct versus indirect effects of UPF3A not distinguished\"]\n    },\n    {\n      \"year\": 2022,\n      \"claim\": \"Provided the quantitative and structural basis for paralog competition: crystal structures of UPF2 bound to either UPF3A or UPF3B showed shared interfaces, UPF3A binds UPF2 ~10-fold tighter, and a disease UPF3B mutation in NOPS-L weakens UPF2 binding ~40-fold.\",\n      \"evidence\": \"X-ray crystallography, affinity measurements, mutagenesis and RNA/ribosome-binding assays\",\n      \"pmids\": [\"35640974\"],\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"\",\n      \"gaps\": [\"Did not resolve how tighter UPF2 binding maps onto inhibitor-versus-activator behavior\", \"In-cell stoichiometry of competition not measured\"]\n    },\n    {\n      \"year\": 2022,\n      \"claim\": \"Resolved the activator/inhibitor paradox by showing UPF3A strongly activates NMD when UPF3B is deleted and that the EJC-binding domain is dispensable while the conserved 'mid' domain drives activity, with the two paralogs functionally redundant.\",\n      \"evidence\": \"CRISPR knockout and combinatorial co-depletion, NMD reporters, RNA-seq and domain-mutant complementation across two parallel studies\",\n      \"pmids\": [\"35451102\", \"35451084\"],\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"\",\n      \"gaps\": [\"Biochemical function of the 'mid' domain undefined\", \"Why outcomes differ between cell systems not fully explained\"]\n    },\n    {\n      \"year\": 2023,\n      \"claim\": \"Refined the in vivo NMD picture and the compensation mechanism: UPF3A is largely dispensable for NMD when UPF3B is present in mouse, and in zebrafish leg1 mutants Upf3a drives homology-dependent compensation in an H3K4me3-independent manner.\",\n      \"evidence\": \"Conditional knockout mice with NMD-target qRT-PCR; zebrafish single/double knockouts with RNA-seq and ULI-NChIP-seq\",\n      \"pmids\": [\"36997282\", \"37369707\"],\n      \"confidence\": \"Medium\",\n      \"confidence_rationale\": \"\",\n      \"gaps\": [\"Reconciliation of H3K4me3-dependent and -independent compensation modes unresolved\", \"Molecular effectors of homology-dependent correction unknown\"]\n    },\n    {\n      \"year\": null,\n      \"claim\": \"The molecular mechanism by which the conserved 'mid' domain activates NMD, and how UPF3A links PTC-containing mRNA to transcriptional compensation, remain unresolved.\",\n      \"evidence\": \"\",\n      \"pmids\": [],\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"\",\n      \"gaps\": [\"No biochemical activity assigned to the 'mid' domain\", \"No identified molecular bridge from UPF3A to COMPASS/chromatin in the GCR\", \"Mechanism of UPF3B-dependent UPF3A destabilization unknown\"]\n    }\n  ],\n  \"mechanism_profile\": {\n    \"molecular_activity\": [\n      {\"term_id\": \"GO:0003723\", \"supporting_discovery_ids\": [8]},\n      {\"term_id\": \"GO:0098772\", \"supporting_discovery_ids\": [4, 5, 9]},\n      {\"term_id\": \"GO:0060090\", \"supporting_discovery_ids\": [0, 2, 3]}\n    ],\n    \"localization\": [\n      {\"term_id\": \"GO:0005634\", \"supporting_discovery_ids\": [0]},\n      {\"term_id\": \"GO:0005829\", \"supporting_discovery_ids\": [0]}\n    ],\n    \"pathway\": [\n      {\"term_id\": \"R-HSA-8953854\", \"supporting_discovery_ids\": [3, 5, 9, 10]},\n      {\"term_id\": \"R-HSA-74160\", \"supporting_discovery_ids\": [6, 11]}\n    ],\n    \"complexes\": [\"EJC-UPF surveillance complex (UPF1-UPF2-UPF3-EJC)\"],\n    \"partners\": [\"UPF2\", \"UPF3B\", \"UPF1\", \"EIF4A3\", \"MAGOH\", \"RBM8A\", \"WDR5\"],\n    \"other_free_text\": []\n  }\n}","audit_flag":null,"evaluation":{"pairwise":"win","faith_supported":6,"faith_total":6,"faith_pct":100.0}}