DNAAF2, also called KTU, is a cytoplasmic dynein assembly factor that helps build outer and inner axonemal dynein arms before their delivery to motile cilia and flagella. It associates with other assembly factors and chaperone machinery, supporting heavy-chain stability and incorporation into dynein complexes. A cytosolic pool exchanges with dynamic dynein axonemal particles in vertebrate multiciliated cells. Loss of DNAAF2 disrupts respiratory-ciliary and sperm motility and causes primary ciliary dyskinesia.
| GO Term | Evidence | Action | Reason |
|---|---|---|---|
| GO:0003351 epithelial cilium movement involved in extracellular fluid movement | IBA GO_REF:0000033 | KEEP AS NON CORE | Summary: Conserved assembly supports epithelial ciliary movement. Reason: The literal PTN002744107 IBD lies on the human DNAAF2 lineage and is grounded in mouse Ktu evidence. The mouse knockout causes tracheal and ependymal dynein defects and loss of motility, consistent with human respiratory disease and conserved preassembly. Retain this genuine physiological consequence as non-core: DNAAF2 performs dynein-arm assembly rather than generating the ciliary power stroke. |
| GO:0005515 protein binding | IPI PMID:23872636 DYX1C1 is required for axonemal dynein assembly and ciliary ... | KEEP AS NON CORE | Summary: Direct DYX1C1 association is supported. Reason: Full-length DYX1C1 and DNAAF2 gave a positive yeast two-hybrid result, corroborated by tagged co-immunoprecipitation in human cells. This is a targeted positive interaction, not an inferred association from shared phenotype. The evidence does not establish an intrinsic folding activity or a more specific binding-function class for DNAAF2. Retain the supported generic interaction as non-core under the project binding policy. Supporting Evidence: PMID:23872636 DYX1C1 and DNAAF2 interact directly by yeast two-hybrid assay |
| GO:0005515 protein binding | IPI PMID:27173435 An organelle-specific protein landscape identifies novel dis... | KEEP AS NON CORE | Summary: UBE3D association is supported by affinity purification. Reason: The imported Q7Z6J8 partner is UBE3D, as confirmed by its official sequence and the literal human interaction record. PMID:27173435 reports a HEK293T tandem-affinity-purification spoke with DNAAF2 bait and UBE3D prey. This supports association in the captured assembly, not purified binary recognition, a ubiquitination substrate relationship or a specific ubiquitin-ligase-binding mechanism. Retain the positive generic association as non-core. Supporting Evidence: |
| GO:0005515 protein binding | IPI PMID:28041644 Mutations in PIH1D3 Cause X-Linked Primary Ciliary Dyskinesi... | KEEP AS NON CORE | Summary: PIH1D3 association is experimentally supported. Reason: The source tests sequence-verified human constructs by yeast two-hybrid and reciprocal FLAG-PIH1D3/HA-DNAAF2 co-immunoprecipitation in HEK293T cells. These experiments support interaction between assembly factors. They do not establish a specific catalytic or folding molecular function of DNAAF2, so retain the positive generic association as non-core. Supporting Evidence: |
| GO:0005515 protein binding | IPI PMID:29727692 C11orf70 Mutations Disrupting the Intraflagellar Transport-D... | KEEP AS NON CORE | Summary: CFAP300 binding is independently supported, with a citation-attribution caveat. Reason: The imported PMID:29727692 and CFAP300 partner are preserved. Its inspected original Methods and target-text search did not recover the DNAAF2 binding experiment. The companion PMID:29727693 explicitly reports C11ORF70/CFAP300βDNAAF2 yeast two-hybrid and tagged co-immunoprecipitation with controls. This independent positive evidence supports the gene-level interaction, while the exact attribution of the imported source remains unresolved; no wrong-gene or nonbinding conclusion is inferred. The official UniProt Q9NVR5 subunit annotation independently cites PMID:29727693 for this same CFAP300 pair. This corroborates the positive interaction, but does not establish that the imported PMID:29727692 was an identifier error. Supporting Evidence: PMID:29727693 C11orf70 shows an interaction with cytoplasmic ODA/IDA assembly factor DNAAF2 UniProtKB:Q9NVR5 Interacts with CFAP300 (PubMed:29727693). |
| GO:0005515 protein binding | IPI PMID:33961781 Dual proteome-scale networks reveal cell-specific remodeling... | KEEP AS NON CORE | Summary: BioPlex supports UBE3D-containing associations. Reason: Three literal human Q7Z6J8/UBE3DβDNAAF2 records are recovered: two HEK293T spoke-expanded associations and one HCT116 physical association. UBE3D is the C-terminally tagged bait and DNAAF2 the recovered partner; both participant sequences and taxon checks match the official products. These AP-MS results support association but not a purified binary interface, ubiquitination of DNAAF2 or a product-specific function. Retain non-core generic binding. Supporting Evidence: |
| GO:0005515 protein binding | IPI PMID:40205054 Multimodal cell maps as a foundation for structural and func... | KEEP AS NON CORE | Summary: The primary U2OS network directly records UBE3DβDNAAF2 association. Reason: The paper-linked primary U2OS AP-MS graph contains UBE3D bait node 608476, DNAAF2 prey node 608478 and their edge 608480. Official Q7Z6J8 identifies UBE3D. This source-specific positive edge supports the imported association even though the separate current IntAct/MINT query did not return it. The assay does not establish direct molecular recognition, a ubiquitination substrate relationship or physiological regulation by UBE3D, so retain generic binding as non-core. Supporting Evidence: |
| GO:0005576 extracellular region | IEA GO_REF:0000108 | REMOVE | Summary: The extracellular inference confuses moved fluid with protein location. Reason: The automatic source is epithelial cilium movement involved in extracellular fluid movement. That process names cilia moving extracellular fluid; it does not put the dynein assembly factor outside the cell. Human respiratory fractionation, cytosolic HPA staining and ortholog DynAP imaging instead support intracellular preassembly. This particular inter-ontology localization inference is biologically incorrect, without implying that every intracellular pool is identical. |
| GO:0005737 cytoplasm | IBA GO_REF:0000033 | MODIFY | Summary: The inherited cytoplasmic claim can be made more precise. Reason: The cytoplasm IBD at PTN000543181 is ancestral to the sequence-matched human target; the target itself is legitimate experimental descendant grounding. Human HPA ICC/IF supports cytosol, while Xenopus Ktu imaging supports a DynAP pool. These nonexclusive refinements retain the inherited cytoplasmic claim without implying that PAINT independently resolved both compartments. Basal-body localization is assessed in its separate HPA-sourced annotation and is not a replacement for this source assertion. Propagation Review Root cause: TERM SCOPING PROBLEM Failure modes: GRANULARITY MISMATCH Sources checked: PANTHER:PTN000543181 SUPPORTS TRANSFER The literal source supports cytoplasm. Independent localization evidence refines this to cytosol and DynAP without asserting failed orthology, circular evidence or a basal-body assignment from this source. Proposed replacements: cytosol dynein axonemal particle |
| GO:0005737 cytoplasm | IDA PMID:19052621 Ktu/PF13 is required for cytoplasmic pre-assembly of axonema... | MODIFY | Summary: The broad respiratory cytoplasmic observation is refined using independent human compartment evidence. Reason: Human respiratory fractionation in Figure 2d resolves cytoplasm from cilia; unsuccessful human respiratory immunohistochemistry does not negate the successful medaka renal-tubule study. In medaka, Ktu concentrates in apical cytoplasm near the gamma-tubulin-positive region but mostly does not colocalize with it, supporting proximity rather than pericentriolar-material membership. Independent human HPA evidence supports cytosol as a refinement of the broad fractionation result. The separately observed HPA basal-body pool is not attributed to this paper or used to replace its cytoplasmic assertion. Proposed replacements: cytosol Supporting Evidence: PMID:19052621 In most cases, Ktu and Ξ³-tubulin signals did not co-localize. GO_REF:0000052 |
| GO:0005737 cytoplasm | IEA GO_REF:0000120 | MODIFY | Summary: Matching family and ortholog evidence supports the more specific cytoplasmic pools. Reason: The target matches the non-fragment eukaryotic HAMAP MF_03069 rule and Kintoun InterPro entry, and the mouse Q8BPI1 identity and broad mapped cytoplasm assertions are sound. Human HPA ICC/IF supports cytosol; ortholog imaging separately supports DynAPs. These nonexclusive gene-level refinements preserve the cytoplasmic scope without alleging failed orthology or direct human DynAP imaging. The separate HPA basal-body annotation is not a claim made by this rule or a replacement for its cytoplasmic assertion. Propagation Review Root cause: TERM SCOPING PROBLEM Failure modes: GRANULARITY MISMATCH Sources checked: UniRule:UR000369588 SUPPORTS TRANSFER The literal source supports cytoplasm. Independent localization evidence refines this to cytosol and DynAP without asserting failed orthology, circular evidence or a basal-body assignment from this source. Proposed replacements: cytosol dynein axonemal particle |
| GO:0005829 cytosol | IDA GO_REF:0000052 | ACCEPT | Summary: Human cytosolic localization supports the assembly role. Reason: The current HPA target record matches ENSG00000165506 and Q9NVR5 and reports approved cytosolic ICC/IF localization. This is coherent with human respiratory cytoplasmic fractionation and the preassembly function. The HPA record also lists other pools; no exclusive cytosolic distribution or isoform resolution is asserted. |
| GO:0050821 protein stabilization | NAS PMID:29844425 The RPAP3-Cterminal domain identifies R2TP-like quaternary c... | KEEP AS NON CORE | Summary: Dynein heavy-chain stability is supported, but the cited client assay has a different target. Reason: DNAAF2/PF13 contributes to dynein heavy-chain stability during preassembly: the 2008 mutant has depleted cytoplasmic heavy chains and defective heavy-chain incorporation while an intermediate/light-chain subcomplex still forms. This supports the broad stabilization role. However, the 2018 Figure 7 stabilization and liprin-alpha2 assembly assays test PIH1D2/R2SP, not DNAAF2. Preserve that source distinction and retain stabilization as a non-core aspect of the integrated assembly mechanism, without transferring those clients or an autonomous folding activity. |
| GO:0060285 cilium-dependent cell motility | IBA GO_REF:0000033 | KEEP AS NON CORE | Summary: The motility phenotype is a conserved consequence of dynein preassembly. Reason: The literal PTN002744107 IBD is on the target lineage with human and medaka experimental grounding. Human KTU variants and medaka loss of Ktu disrupt dynein arms and ciliary or sperm motility. Retain this biologically sound physiological consequence as non-core; the defining performed work is dynein-arm preassembly, not motor force generation. Tissue-specific compensation in zebrafish does not erase the positive human and medaka evidence. |
| GO:0060285 cilium-dependent cell motility | IEA GO_REF:0000104 | KEEP AS NON CORE | Summary: The matching UniRule correctly captures a downstream motility consequence. Reason: UR000369588 matches the target HAMAP family, eukaryotic taxon and non-fragment sequence. Its motility annotation agrees with human and ortholog experimental evidence. Retain the true motility role as non-core alongside the more mechanistically precise dynein-arm assembly core; do not infer ATPase or motor activity from this process annotation. Supporting Evidence: |
| GO:0060285 cilium-dependent cell motility | IMP PMID:19052621 Ktu/PF13 is required for cytoplasmic pre-assembly of axonema... | KEEP AS NON CORE | Summary: Human and model-organism motility defects support a non-core phenotype. Reason: The source reports combined dynein-arm abnormalities and impaired respiratory-ciliary or sperm motility in human KTU deficiency, with complementary medaka and Chlamydomonas experiments. The factor works during cytoplasmic dynein preassembly. Retain the experimentally established movement consequence as non-core rather than treating DNAAF2 as the axonemal motor. Supporting Evidence: |
| GO:0070286 axonemal dynein complex assembly | IBA GO_REF:0000033 | MODIFY | Summary: Inherited arm assembly is supported and can be split into its two established branches. Reason: The actual PTN002744107 assembly IBD is on the human target lineage and has human, medaka, mouse and zebrafish experimental grounding. Human and Chlamydomonas results establish outer- and inner-arm assembly roles, although the affected subtypes and compensation differ between organisms and tissues. Replace the broad assembly parent with both supported children; do not assert universal loss of every inner arm or reject an IBD because the target contributes its own experimental evidence. Propagation Review Root cause: TERM SCOPING PROBLEM Failure modes: GRANULARITY MISMATCH Sources checked: PANTHER:PTN002744107 SUPPORTS TRANSFER The actual source supports the inherited or mapped broad claim. Supported specific descendants refine its scope; no erroneous donor, lineage loss or circular target evidence is asserted. Proposed replacements: outer dynein arm assembly inner dynein arm assembly |
| GO:0070286 axonemal dynein complex assembly | IEA GO_REF:0000120 | MODIFY | Summary: The matching electronic assembly annotation is refined to both arm types. Reason: The Kintoun InterPro mapping and matching UR000369588 rule assert axonemal dynein complex assembly. Human disease material and ortholog biochemical experiments independently support both outer- and inner-arm assembly. Use both supported specific descendants instead of the generic parent; the term refinement does not challenge the family assignment. Propagation Review Root cause: TERM SCOPING PROBLEM Failure modes: GRANULARITY MISMATCH Sources checked: InterPro:IPR034727 SUPPORTS TRANSFER The actual source supports the inherited or mapped broad claim. Supported specific descendants refine its scope; no erroneous donor, lineage loss or circular target evidence is asserted. Proposed replacements: outer dynein arm assembly inner dynein arm assembly |
| GO:0070286 axonemal dynein complex assembly | IMP PMID:19052621 Ktu/PF13 is required for cytoplasmic pre-assembly of axonema... | MODIFY | Summary: Direct experiments support outer- and inner-arm preassembly. Reason: Human KTU deficiency affects outer and inner dynein arms. In Chlamydomonas, residual outer-arm heavy chain fails to co-precipitate intermediate/light chains while the intermediate/light-chain subcomplex still forms, placing PF13/Ktu in performed preassembly work. Inner-arm heavy-chain defects provide the complementary scope. Both specific assembly terms are warranted, without claiming every inner-arm subtype is identically affected. Proposed replacements: outer dynein arm assembly inner dynein arm assembly Supporting Evidence: PMID:19052621 intermediate and light chains can still pre-assemble in pf13 cytoplasm |
| GO:0101031 protein folding chaperone complex | IPI PMID:29844425 The RPAP3-Cterminal domain identifies R2TP-like quaternary c... | KEEP AS NON CORE | Summary: Broad chaperone-assembly association is supported; stable R2SD composition is not. Reason: The current term includes complexes participating in macromolecular assembly or stabilization. Human DNAAF2 forms a controlled DYX1C1-isoform-a association in LUMIER and HeLa proteomics, with weaker RUVBL association. This supports a broad co-chaperone assembly context, reinforced by mouse KtuβHsp70 association. The proposed DNAAF2βSPAG1βRUVBL R2SD complex remains hypothetical because it was not recovered by proteomics; the demonstrated R2SP client assays concern PIH1D2. Retain non-core broad membership without assigning a fixed R2SD stoichiometry, purified DNAAF2 folding activity or a separate core. Supporting Evidence: PMID:29844425 this complex remains hypothetical since it was detected in LUMIER but not in proteomic experiments PMID:19052621 only Hsp70 co-immunoprecipitates with mKtu |
| GO:0120293 dynein axonemal particle | IEA GO_REF:0000044 | ACCEPT | Summary: The mapped DynAP localization is supported by the relevant ortholog experiments. Reason: The SL-0542 mapping agrees with the current DynAP definition. Tagged Xenopus Ktu/Dnaaf2 colocalizes with dynein subunits and assembly factors in dynamic cytoplasmic foci, distinct from the tested P-body and Golgi controls. This supports the conserved assembly compartment; it is not an endogenous human DNAAF2 imaging experiment or proof of a stable stoichiometric complex. |
| GO:0120293 dynein axonemal particle | ISS GO_REF:0000024 | ACCEPT | Summary: The literal Xenopus donor supports DynAP localization. Reason: B1H1W9 is the experimentally annotated Xenopus Ktu donor, with IDA DynAP records from PMID:30561330 and PMID:33263282. Their tagged-construct imaging and colocalization support localization in this assembly compartment. Preserve the curated ortholog transfer and its organism/construct limits; rapid Ktu exchange in DynAPs is distinct from stable dynein retention. |
| GO:0036064 ciliary basal body | IDA GO_REF:0000052 | NEW | Summary: Human Protein Atlas imaging supports a basal-body pool of DNAAF2. Reason: The current HPA ICC/IF record for ENSG00000165506 and Q9NVR5 assigns basal body as a main location with approved gene-level reliability. HPA002894 gives basal-body calls in ASC52telo, serum-starved hTERT-RPE1 and RPTEC/TERT1, using cilium and basal-body reference channels. This is one antibody across three cell models, not independent antibody validation or proof of a basal-body assembly function. This distinct location is recorded once from its actual HPA evidence; cytosol remains independently supported. Supporting Evidence: GO_REF:0000052 file:human/DNAAF2/DNAAF2-hpa.xml Mainly localized to the cytosol and basal body. |
Loading supporting contentβ¦
Download this section (compressed HTML)Q: How do the two natural DNAAF2 products differ in abundance, localization and dynein-arm assembly activity?
Q: Which native DNAAF2 assemblies perform heavy-chain stabilization, and when does the DYX1C1 heterodimer associate with Hsp70, SPAG1 or RUVBL proteins?
Q: Do endogenous human DNAAF2 basal-body-associated and nearby apical pools overlap DynAPs, or mark distinct stages of dynein-arm assembly and delivery?
Q: How closely does endogenous human DNAAF2 DynAP dynamics match the tagged Xenopus Ktu observations?
Experiment: Compare accession-verified products at matched expression in deficient human multiciliated cells, measuring dynein heavy-chain incorporation, partner association and ciliary motility with rescue and fraction-purity controls.
Hypothesis: Natural DNAAF2 products may differ in assembly-factor recruitment without losing the entire preassembly role.
Type: Product-resolved rescue and complex assembly
Experiment: Combine endogenous tagging, quantitative native-complex purification and defined-component reconstitution to separate direct contacts, transient co-chaperone associations and heavy-chain stabilization from downstream motility.
Hypothesis: DNAAF2 couples a dynamic chaperone assembly to heavy-chain incorporation.
Type: Native and reconstituted assembly analysis
Loading supporting contentβ¦
Download this section (compressed HTML)Loading supporting contentβ¦
Download this section (compressed HTML)