Testis-enriched member of the NOL1/NOP2/Sun (NSUN) protein family that, unlike its paralogues NSUN1-NSUN6, is not an active RNA m5C methyltransferase. NSUN7 retains the canonical NSUN fold and motifs IV and VI, but the aspartate of motif IV that contacts S-adenosyl-L-methionine in the active members is replaced by leucine; purified mouse and human NSUN domains fail to bind SAM, loss of Nsun7 does not alter RNA m5C levels measured by mass spectrometry or bisulfite sequencing, and mice carrying a substitution of the putative catalytic cysteine are phenotypically normal. The protein instead behaves as an RNA-binding protein: CLIP-seq from testis recovers a largely mRNA set of interactors specific to elongated spermatids, and NSUN7 protein stability is itself RNA-dependent. NSUN7 is expressed in spermatids and localises to the cytoplasm and to the sperm flagellum, where it is required for normal axoneme structure, for correct positioning of the fibrous-sheath longitudinal columns, and hence for progressive sperm motility; loss-of-function causes male infertility in mice and variants are associated with asthenozoospermia in men. Whether NSUN7 stabilises or destabilises its bound transcripts is unresolved, with knockout studies reporting effects in opposite directions. Reports that NSUN7 deposits m5C on specific transcripts in kidney, ovarian granulosa cells and glioblastoma rest on knockdown-plus-mapping correlations rather than on direct tests of catalysis.
| GO Term | Evidence | Action | Reason |
|---|---|---|---|
| GO:0005737 cytoplasm | IEA GO_REF:0000044 | ACCEPT | Summary: Mapping of the UniProt subcellular-location annotation; NSUN7 is a cytoplasmic protein of spermatids. Reason: Correct, although the sperm flagellum annotation (GO:0036126) carried by this gene is the informative location. |
| GO:0008168 methyltransferase activity | IEA GO_REF:0000002 | REMOVE | Summary: Electronic assignment of methyltransferase activity from the NOL1/NOP2/Sun family signature InterPro:IPR001678, i.e. from the fold rather than from any measurement on NSUN7 itself. Reason: Directly contradicted by biochemistry on the protein. Isothermal titration calorimetry shows that neither the mouse nor the human NSUN domain binds the methyl donor SAM, because the motif IV aspartate that coordinates SAM in NSUN1-NSUN6 is a leucine in NSUN7; knockout testis shows no change in RNA m5C by mass spectrometry or ultrafast bisulfite sequencing (PMID:41381527); and a mouse carrying a substitution of the putative catalytic cysteine of motif IV is fertile with normal sperm motility and normal longitudinal-column positioning, so the knockout phenotype is not caused by loss of methyl transfer (PMID:40545153). UniProt has itself added a CAUTION stating that NSUN7 lacks RNA cytosine C5-methyltransferase activity. Three disease papers do assert NSUN7-catalysed m5C (PMID:41871257 kidney injury, PMID:41062800 polycystic ovary syndrome, PMID:41275291 glioblastoma), but each infers catalysis from a knockdown or knockout followed by MeRIP or bisulfite mapping; none measures SAM binding, in vitro methyl transfer by purified NSUN7, or rescue by a catalytically dead allele, and all place a testis-restricted protein as a dominant m5C writer in somatic tissues. Removing an electronic, signature-derived inference that direct assays on the protein refute is the case the project guidelines explicitly allow; the informative replacement function is mRNA binding (GO:0003729), proposed in core_functions. Supporting Evidence: PMID:41381527 Intriguingly, we found that the NSUN domain of mouse NSUN7 does not bind SAM at all PMID:41381527 Consistent with the result of mouse NSUN7, the NSUN domain of human NSUN7 also failed to bind SAM PMID:41381527 Through transcriptome-wide m5C mass spectrometry and ultrafast bisulfite sequencing, we found that Nsun7 knockout has no impact on RNA m5C modification. PMID:40545153 Contrary to predictions based on the typical reaction mechanism of NOP2/Sun family methyltransferases, Nsun7C382A mice did not exhibit any phenotypical characteristics of knockouts and had normal fertility, sperm motility, and longitudinal column positioning, similar to wild-type mice. |
| GO:0031514 motile cilium | IEA GO_REF:0000044 | ACCEPT | Summary: The sperm flagellum is a motile cilium, and NSUN7 localises to the flagella of elongated spermatids. Reason: Correct; the more specific sperm flagellum term (GO:0036126) is also carried by this gene. Supporting Evidence: PMID:41381527 Thus, our results showed that NSUN7 is predominantly enriched in adult mouse testis, with specific localization to the flagella of elongated spermatids |
| GO:0036126 sperm flagellum | ISS GO_REF:0000024 | ACCEPT | Summary: Transfer from mouse Nsun7 (UniProtKB:Q14AW5), which is imaged in the flagella of elongated spermatids. Reason: Core location, well supported in the mouse orthologue and consistent with the testis-restricted expression of the human gene. Supporting Evidence: PMID:41381527 Thus, our results showed that NSUN7 is predominantly enriched in adult mouse testis, with specific localization to the flagella of elongated spermatids |
| GO:0005737 cytoplasm | ISS GO_REF:0000024 | ACCEPT | Summary: Transfer of the cytoplasmic localisation from mouse Nsun7. Reason: Correct, although the sperm flagellum annotation (GO:0036126) carried by this gene is the informative location. |
| GO:0030317 flagellated sperm motility | ISS GO_REF:0000024 | ACCEPT | Summary: Nsun7 loss in mice severely reduces progressive sperm motility, and human NSUN7 variants are associated with asthenozoospermia. Reason: Core biological process. Two independent knockout lines and an earlier ENU allele all give a progressive-motility defect, and this is the phenotype through which the gene was originally found. Supporting Evidence: PMID:41381527 Nsun7 deficiency impairs sperm progressive motility, accompanied by defects in axonemes and mispositioning of longitudinal columns. PMID:17442852 Mutant sperm exhibited depressed progressive motility associated with a rigid flagellar midpiece (but not principal piece) segment |
| GO:0120316 sperm flagellum assembly | ISS GO_REF:0000024 | ACCEPT | Summary: Nsun7 knockout spermatozoa have axonemal microtubule defects and mispositioned fibrous-sheath longitudinal columns, i.e. a flagellar assembly defect. Reason: Core biological process, and the one the gene is now named for. Importantly this role is independent of methyl transfer, since a catalytic-cysteine knock-in mouse assembles normal flagella. Supporting Evidence: PMID:41381527 Nsun7 deficiency impairs sperm progressive motility, accompanied by defects in axonemes and mispositioning of longitudinal columns. PMID:40032361 A physiologic consequence of Nsun7 gene knockout is male infertility, which is mechanistically explained by the observed mispositioning of longitudinal columns relative to the axonemal microtubular doublets leading to a motility defect. |
| GO:0003729 mRNA binding | ISS PMID:40032361 Positioning of sperm tail longitudinal columns depends on NS... | NEW | Summary: Proposed NEW annotation, and the informative molecular function that should stand in place of the removed GO:0008168 methyltransferase activity. UV CLIP-seq from mouse testis recovers a set of NSUN7-crosslinked RNAs that is predominantly mRNA and is enriched for transcripts of elongated spermatids; NSUN7 protein stability is itself RNase-sensitive. The human protein is a one-to-one orthologue and already carries the UniProt RNA-binding keyword, so the evidence code is given as ISS by transfer from mouse Nsun7 (UniProtKB:Q14AW5). Reason: A catalytically dead member of an enzyme family still has a molecular function, and for NSUN7 the measured one is RNA binding. Asserting it makes the removal of the family-signature methyltransferase term informative rather than merely subtractive. Supporting Evidence: PMID:40032361 The CLIP-seq experiment performed on testes extract allowed us to identify multiple NSUN7 binding RNAs. The majority of those belong to mRNAs, although several noncoding RNAs have also been identified. PMID:41381527 Our results show that RNase treatment has an impact on NSUN7 protein stability, further indicating that the association of NSUN7 with RNA contributes to maintaining its protein stability. |
Loading supporting content…
Download this section (compressed HTML)Q: Does NSUN7 stabilise or destabilise the transcripts it binds? CLIP-seq plus RNA-seq of the Nsun7 knockout reports upregulation of NSUN7-bound RNAs, implying destabilisation, whereas single-cell RNA-seq of an independent knockout reports downregulation of cilium-organisation mRNAs, implying stabilisation. Are these opposite signs explained by different transcript sets, different knockout alleles, or by secondary consequences of the structural defect?
Suggested experts: Petr V Sergiev, Ru-Juan Liu
Q: Given that NSUN7 does not bind SAM and that a catalytic-cysteine knock-in mouse is phenotypically normal, what accounts for the reports that Nsun7 deletion lowers bulk m5C in kidney and that NSUN7 knockdown lowers m5C on specific transcripts in granulosa cells and glioblastoma? Is NSUN7 acting indirectly on a genuine NSUN enzyme, or are these off-target or mapping artefacts?
Suggested experts: Xiao-Ming Meng
Q: Does NSUN7 retain SAM binding or methyl-transfer activity in any vertebrate lineage? The motif IV aspartate is replaced by leucine in most vertebrate NSUN7 orthologues but by valine or threonine in some, and it is unknown whether any of these variants restores the cofactor site.
Suggested experts: Ru-Juan Liu
Q: What is the RNA sequence or structure element that NSUN7 recognises, and does the PUA-like or NSUN-domain surface that normally positions the substrate cytosine now serve purely as an RNA-binding platform?
Suggested experts: Petr V Sergiev
Experiment: Determine a crystal or cryo-EM structure of the NSUN7 NSUN domain with and without a CLIP-derived RNA ligand, and test SAM and SAH binding directly by ITC and by differential scanning fluorimetry alongside NSUN2 and NSUN6 as positive controls. A structure showing an occluded or re-arranged cofactor pocket together with an occupied RNA groove would settle the pseudoenzyme assignment.
Hypothesis: NSUN7 acts as a sequence-specific mRNA-binding protein in elongated spermatids, not as a methyltransferase.
Type: Structural biology and biophysics
Experiment: Generate knock-in mice carrying point substitutions that abolish RNA binding (guided by the structure above) while leaving the fold intact, and compare them with the existing null and with the catalytic-cysteine Nsun7C382A line for fertility, progressive motility, axonemal ultrastructure and longitudinal-column position.
Hypothesis: The flagellar phenotype of Nsun7 knockout mice is caused by loss of RNA binding, not by loss of catalysis.
Type: Mouse genetics
Experiment: Measure transcript half-lives directly in sorted wild-type and Nsun7 knockout elongated spermatids using metabolic labelling (SLAM-seq or TimeLapse-seq) rather than steady-state RNA-seq, and restrict the analysis to CLIP-defined direct targets, so that secondary effects of the structural defect are separated from the primary effect on decay.
Hypothesis: The direction of the NSUN7 effect on transcript abundance can be resolved by measuring decay rather than steady-state levels.
Type: RNA metabolic labelling
Experiment: In a kidney injury model, compare Nsun7 knockout with knock-in of the catalytically dead-equivalent allele; if bulk m5C falls in the null but not in the point mutant, the m5C change is not produced by NSUN7 catalysis. In parallel, quantify NSUN7 protein in kidney by targeted mass spectrometry and test in vitro methylation of the proposed SPARC and NLRP3 transcripts by purified NSUN7.
Hypothesis: The reported somatic m5C-writer activity of NSUN7 is indirect.
Type: Mouse genetics and in vitro biochemistry
Loading supporting content…
Download this section (compressed HTML)Loading supporting content…
Download this section (compressed HTML)