SNURF (SNRPN upstream reading frame protein) is a highly basic 71-amino-acid nuclear protein encoded by the upstream open reading frame of the bicistronic SNURF-SNRPN transcript on chromosome 15q11-q13; the downstream ORF of the same mRNA encodes the spliceosomal SmN protein (SNRPN). The locus lies in the Prader-Willi syndrome imprinting centre and is expressed only from the paternal allele, and both proteins are made in normal but not Prader-Willi cells. The ORF is conserved across eutherian mammals with substitutions concentrated at wobble positions, which indicates selection for coding function, and the protein carries a conserved basic nuclear localization motif. A GFP-tagged protein localizes to the nucleus and antibody staining places it in nuclear speckles. Its molecular function and biological role are unknown. It is unrelated to the RING finger protein RNF4, which was originally also named SNURF.
| GO Term | Evidence | Action | Reason |
|---|---|---|---|
| GO:0005634 nucleus | IEA GO_REF:0000120 | ACCEPT | Summary: Combined automated nuclear annotation, consistent with the GFP-fusion nuclear localization of the human protein and its conserved basic NLS-like motif. Reason: Supported by direct localization of the human protein (PMID:10318933) and by HPA nuclear-speckle staining. Supporting Evidence: PMID:10318933 C-terminally tagged SNURF–GFP is indeed targeted to the nucleus when ectopically expressed |
| GO:0005634 nucleus | NAS PMID:10318933 An imprinted, mammalian bicistronic transcript encodes two i... | ACCEPT | Summary: Gray et al. 1999 state that SNURF is nuclear-localized, based on their GFP-fusion experiment in the same paper. Reason: The statement is backed by the paper's own direct observation of a GFP fusion, so the location is sound (it could equally have been coded IDA). Supporting Evidence: PMID:10318933 SNURF encodes a highly basic 71-aa protein that is nuclear-localized (as is SmN). PMID:10318933 C-terminally tagged SNURF–GFP is indeed targeted to the nucleus when ectopically expressed |
| GO:0016607 nuclear speck | IBA GO_REF:0000033 | ACCEPT | Summary: PAINT inference at a SNURF-family node, grounded in the human HPA nuclear-speck observation (Q9Y675 is itself the experimental source). Reason: The IBA says the speckle location is inherited across the eutherian SNURF clade, which is reasonable for an ORF conserved at 93-100% amino-acid identity. The human protein is its own grounding evidence, which is expected and not circular. No contrary localization data exist, and the term is a refinement of the established nuclear location. Supporting Evidence: PMID:10318933 Amino acid identity with human SNURF ranges from 93% to 100%, with most substitutions being quite conservative. |
| GO:0016607 nuclear speck | IDA GO_REF:0000052 | ACCEPT | Summary: Human Protein Atlas immunofluorescence places SNURF in nuclear speckles (main location, reliability "Supported"). Reason: Antibody-based localization consistent with the GFP-fusion nuclear localization in PMID:10318933. Antibody specificity against the SmN product of the same transcript has not been reported independently, which is worth confirming, but there is no reason to reject it. |
| GO:0051117 ATPase binding | IEA GO_REF:0000107 | REMOVE | Summary: Ensembl Compara transfer of an MGI IPI on mouse Snurf (Q9WU12, WITH the SNF2-family ATPase ARIP4/Rad54l2, PMID:12058073). The SNURF in that paper is the RING finger protein RNF4, not the SNRPN upstream reading frame protein. Reason: Symbol collision at the source. RNF4 was first described as "SNURF", small nuclear RING finger protein, a 194-residue C3HC4 RING protein from the same steroid-receptor coregulator group (PMID:9710597) that published the ARIP4 paper. The ARIP4 paper's full text (PMC117628, read online; the local cache is abstract-only) defines its SNURF as "small nuclear RING finger protein (SNURF)" and uses a "SNURF residues 20-177" VP16 fusion as an androgen-receptor-binding control. A 20-177 fragment cannot come from the 71-residue SNRPN-uORF protein. The MGI IPI (and its reciprocal protein-binding IPI on ARIP4) was therefore attached to the wrong mouse gene product (Snurf instead of Rnf4), and the human IEA inherits that error. Nothing links Q9Y675 to ATPase binding. Supporting Evidence: PMID:9710597 194 amino acid residues and comprising a consensus C3HC4 zinc finger (RING) structure in the C-terminal region and a bipartite nuclear localization signal near the N terminus PMID:10318933 SNURF encodes a highly basic 71-aa protein that is nuclear-localized (as is SmN). |
Loading supporting content…
Download this section (compressed HTML)Q: Does the SNURF protein itself (as opposed to the SNURF-SNRPN transcript and the imprinting centre) contribute to any Prader-Willi syndrome phenotype?
Q: Does the arginine-rich 71-aa protein bind RNA or speckle components (SR proteins, snRNPs), consistent with its nuclear-speckle localization?
Experiment: Introduce a start-codon or frameshift mutation into the mouse Snurf ORF that leaves the imprinting centre and Snrpn translation intact, and phenotype the mice. Combine with proximity labelling or RNA crosslinking of tagged endogenous SNURF in neurons.
Hypothesis: The SNURF protein has a function separate from the imprinting-centre DNA and the SNRPN product.
Loading supporting content…
Download this section (compressed HTML)Loading supporting content…
Download this section (compressed HTML)