SHLD2 (FAM35A, RINN2) is the central, non-enzymatic subunit of shieldin, a vertebrate DNA double-strand-break (DSB) end-protection complex made of SHLD1, SHLD2, SHLD3 and REV7 (MAD2L2). Its short N-terminal segment binds the SHLD3-REV7 recruitment module, which links shieldin to the 53BP1-RIF1 pathway at damaged chromatin, and its C-terminal region carries three RPA1-like OB folds that associate with SHLD1 and bind single-stranded DNA with high affinity. By occupying resected 3' single-stranded overhangs at DSBs, shieldin limits further end resection and RAD51 loading, so it favours non-homologous end joining over homologous recombination. This activity supports immunoglobulin class-switch recombination and end joining of deprotected telomeres. In BRCA1-deficient cells, loss of SHLD2 restores resection and confers resistance to PARP inhibitors. Alternative splicing removes a 69-residue exon from the second OB fold. The UniProt displayed (835-residue) sequence is this short form. The 904-residue long form is the one used in the biochemical and rescue studies.
| GO Term | Evidence | Action | Reason |
|---|---|---|---|
| GO:0000785 chromatin | NAS PMID:29656893 DNA Repair Network Analysis Reveals Shieldin as a Key Regula... | ACCEPT | Summary: Shieldin acts on damaged chromatin at DSBs. Reason: Shieldin is recruited to DSB chromatin through the ATM-RNF8-RNF168-53BP1-RIF1 axis, and SHLD2 co-localises with damage foci and LacR-tethered RIF1 on chromatin. Chromatin is an appropriate location for this damage-induced pool; it does not imply a constitutive chromosome-wide distribution. Supporting Evidence: PMID:29656893 Recruitment of shieldin to DSBs, via the ATM-RNF8-RNF168-53BP1-RIF1 axis, promotes NHEJ-dependent repair of intrachromosomal breaks, immunoglobulin class-switch recombination (CSR), and fusion of unprotected telomeres. |
| GO:0002208 somatic diversification of immunoglobulins involved in immune response | NAS PMID:29656893 DNA Repair Network Analysis Reveals Shieldin as a Key Regula... | MODIFY | Summary: Shieldin's role in antibody diversification is specifically in class-switch recombination. Reason: The evidence concerns NHEJ-dependent joining of AID-induced switch-region breaks during class-switch recombination, not somatic hypermutation or V(D)J recombination. GO:0045190 isotype switching (an is_a descendant of GO:0002208 via somatic recombination of immunoglobulin genes) is the precise term, and its term note says it is intended for gene products in the DNA recombination mechanism of switching, which is where shieldin-mediated end protection acts. The gene also carries positive regulation of isotype switching (GO:0045830) from the same paper. Kept as a non-core, immune-specific use of the general DSB end-protection function. Proposed replacements: isotype switching Supporting Evidence: PMID:29656893 Immunoglobulin CSR involves generation of activation-induced cytidine deaminase (AID)-induced DSBs in the S regions and subsequent repair by NHEJ. PMID:29656893 Recruitment of shieldin to DSBs, via the ATM-RNF8-RNF168-53BP1-RIF1 axis, promotes NHEJ-dependent repair of intrachromosomal breaks, immunoglobulin class-switch recombination (CSR), and fusion of unprotected telomeres. |
| GO:0005515 protein binding | IPI PMID:29656893 DNA Repair Network Analysis Reveals Shieldin as a Key Regula... | MODIFY | Summary: SHLD2 binds SHLD3 through its N-terminus as part of its shieldin scaffold role. Reason: Bare protein binding is uninformative. Interaction mapping places the SHLD3-REV7 module on SHLD2 residues 1-50/60 and SHLD1 on the C-terminal OB-fold region, so SHLD2 bridges the recruitment and DNA-binding modules of shieldin. The informative molecular function is protein-macromolecule adaptor activity (GO:0030674), already the scaffold core function in this review. Proposed replacements: protein-macromolecule adaptor activity Supporting Evidence: PMID:30022168 Further mapping studies suggest that Shieldin consists of a DSB-recruitment module composed of SHLD3-REV7 that binds to the N-terminus of SHLD2 (residues 1-50; ED Fig 6, ED Fig 7a-c), and a presumptive DNA-binding module (SHLD2-SHLD1) featuring the OB-fold domains at the SHLD2 C-terminus (SHLD2C residues 421-904; ED Fig 7a). file:human/SHLD2/SHLD2-uniprot.txt SHLD2 forms a scaffold which interacts with a SHLD3-MAD2L2 subcomplex via its N-terminus, and with SHLD1 via its C-terminus |
| GO:0005515 protein binding | IPI PMID:29656893 DNA Repair Network Analysis Reveals Shieldin as a Key Regula... | MODIFY | Summary: SHLD2 binds SHLD1 through its C-terminal OB-fold region as part of its shieldin scaffold role. Reason: Bare protein binding is uninformative. SHLD1 binds the SHLD2 C-terminus (UniProt region 652-822) while SHLD3-REV7 bind the N-terminus, so this interaction is one arm of SHLD2's bridging function. The informative molecular function is protein-macromolecule adaptor activity (GO:0030674). Proposed replacements: protein-macromolecule adaptor activity Supporting Evidence: PMID:30022168 Further mapping studies suggest that Shieldin consists of a DSB-recruitment module composed of SHLD3-REV7 that binds to the N-terminus of SHLD2 (residues 1-50; ED Fig 6, ED Fig 7a-c), and a presumptive DNA-binding module (SHLD2-SHLD1) featuring the OB-fold domains at the SHLD2 C-terminus (SHLD2C residues 421-904; ED Fig 7a). file:human/SHLD2/SHLD2-uniprot.txt SHLD2 forms a scaffold which interacts with a SHLD3-MAD2L2 subcomplex via its N-terminus, and with SHLD1 via its C-terminus |
| GO:0005515 protein binding | IPI PMID:29656893 DNA Repair Network Analysis Reveals Shieldin as a Key Regula... | MODIFY | Summary: SHLD2 binds REV7 (with SHLD3) through its N-terminus as part of its shieldin scaffold role. Reason: Bare protein binding is uninformative. The SHLD2 N-terminal segment is sufficient for association with the SHLD3-REV7 module, linking it to the C-terminal SHLD1/ssDNA-binding module. The informative molecular function is protein-macromolecule adaptor activity (GO:0030674). Proposed replacements: protein-macromolecule adaptor activity Supporting Evidence: PMID:30022168 Further mapping studies suggest that Shieldin consists of a DSB-recruitment module composed of SHLD3-REV7 that binds to the N-terminus of SHLD2 (residues 1-50; ED Fig 6, ED Fig 7a-c), and a presumptive DNA-binding module (SHLD2-SHLD1) featuring the OB-fold domains at the SHLD2 C-terminus (SHLD2C residues 421-904; ED Fig 7a). file:human/SHLD2/SHLD2-uniprot.txt SHLD2 forms a scaffold which interacts with a SHLD3-MAD2L2 subcomplex via its N-terminus, and with SHLD1 via its C-terminus |
| GO:0005515 protein binding | IPI PMID:29789392 FAM35A associates with REV7 and modulates DNA damage respons... | REMOVE | Summary: Co-immunoprecipitation of FAM35A with 53BP1; bare protein binding is uninformative. Reason: The association with 53BP1 is real and places SHLD2 in the 53BP1 pathway, but it was recovered by immunoprecipitation and is probably indirect (bridged by RIF1 and SHLD3-REV7). Bare GO:0005515 carries no functional information, and no more informative MF is supported for this pair. Removal does not dispute the interaction. Supporting Evidence: PMID:29789392 DNA repair proteins associating with FAM35A included REV7, RIF1, BLM, and TOP3A |
| GO:0005515 protein binding | IPI PMID:29789392 FAM35A associates with REV7 and modulates DNA damage respons... | REMOVE | Summary: Co-immunoprecipitation of FAM35A with RIF1; bare protein binding is uninformative. Reason: RIF1 co-purifies with FAM35A and acts upstream of shieldin recruitment, but the association does not establish a direct SHLD2-RIF1 contact or a specific molecular activity. Bare GO:0005515 is removed as uninformative; the interaction itself is not disputed. Supporting Evidence: PMID:29789392 DNA repair proteins associating with FAM35A included REV7, RIF1, BLM, and TOP3A |
| GO:0005515 protein binding | IPI PMID:29789392 FAM35A associates with REV7 and modulates DNA damage respons... | MODIFY | Summary: FAM35A associates with REV7, an interaction mapped to the SHLD2 N-terminal scaffold segment. Reason: Bare protein binding is uninformative. The REV7 association reported here is the same one mapped by PMID:30022168 to the SHLD2 N-terminus, as the bridge between the SHLD3-REV7 recruitment module and the SHLD1/ssDNA-binding module. MODIFY to protein-macromolecule adaptor activity for consistency with the other REV7 row. Proposed replacements: protein-macromolecule adaptor activity Supporting Evidence: PMID:29789392 DNA repair proteins associating with FAM35A included REV7, RIF1, BLM, and TOP3A PMID:30022168 Further mapping studies suggest that Shieldin consists of a DSB-recruitment module composed of SHLD3-REV7 that binds to the N-terminus of SHLD2 (residues 1-50; ED Fig 6, ED Fig 7a-c), and a presumptive DNA-binding module (SHLD2-SHLD1) featuring the OB-fold domains at the SHLD2 C-terminus (SHLD2C residues 421-904; ED Fig 7a). |
| GO:0005515 protein binding | IPI PMID:34354233 ASTE1 promotes shieldin-complex-mediated DNA repair by atten... | REMOVE | Summary: Interaction with the shieldin-dependent nuclease ASTE1; bare protein binding is uninformative. Reason: ASTE1 localises to damage sites in a shieldin-dependent manner, and UniProt records an interaction region on SHLD2 (residues 1-568). Only the abstract is cached, so the assay is not inspected here. Either way, bare GO:0005515 carries no functional information, and no more specific SHLD2 MF follows from it. Removal does not dispute the interaction. Supporting Evidence: PMID:34354233 ASTE1 localizes to DNA damage sites in a shieldin-dependent manner. |
| GO:0005634 nucleus | EXP PMID:29656893 DNA Repair Network Analysis Reveals Shieldin as a Key Regula... | ACCEPT | Summary: SHLD2 is a nuclear protein that forms foci at DNA damage. Reason: Nuclear localisation and damage-induced nuclear foci are shown for FAM35A/SHLD2 and are the compartment in which shieldin protects DSB ends. The EXP, IDA, IBA and SubCell-mapping rows all assert the same, correct location. Supporting Evidence: PMID:29789392 FAM35A re-localized in damaged cell nuclei |
| GO:0005634 nucleus | IBA GO_REF:0000033 | ACCEPT | Summary: Nuclear activity inherited across the SHLD2 family node. Reason: The PAINT node PTN001025652 is seeded by human SHLD2's own experimental nuclear annotation, which is expected. All PTHR14495 members fall in subfamily SF2, so inheritance of a nuclear site of action is sound. Supporting Evidence: PMID:29789392 FAM35A re-localized in damaged cell nuclei |
| GO:0005634 nucleus | IDA PMID:29789392 FAM35A associates with REV7 and modulates DNA damage respons... | ACCEPT | Summary: FAM35A relocalises within damaged nuclei. Reason: Direct imaging of FAM35A in damaged cell nuclei. Supporting Evidence: PMID:29789392 FAM35A re-localized in damaged cell nuclei |
| GO:0005634 nucleus | IEA GO_REF:0000044 | ACCEPT | Summary: UniProt subcellular location (Nucleus) mapping, backed by experiment. Reason: The UniProt Nucleus location is itself experimentally supported (PubMed:29656893, PubMed:29789392). Supporting Evidence: PMID:29789392 FAM35A re-localized in damaged cell nuclei |
| GO:0005654 nucleoplasm | IDA GO_REF:0000052 | ACCEPT | Summary: HPA immunofluorescence places SHLD2 mainly in the nucleoplasm. Reason: HPA reports the nucleoplasm as the main location, which fits the nuclear DNA-repair role of SHLD2. The secondary actin-filament signal in the same HPA record is not annotated in GOA and is not considered here. Supporting Evidence: file:human/SHLD2/SHLD2-hpa.xml Mainly localized to the nucleoplasm. In addition localized to the actin filaments. |
| GO:0005694 chromosome | IDA PMID:29656893 DNA Repair Network Analysis Reveals Shieldin as a Key Regula... | MODIFY | Summary: The chromosomal pool of SHLD2 is at DSB sites. Reason: The chromosome term is correct but general. The evidence is damage-induced foci at DSBs, and GO:0035861 site of double-strand break is part_of GO:0005694 chromosome. The gene already carries GO:0035861 by IDA from the same paper. Proposed replacements: site of double-strand break Supporting Evidence: PMID:29656893 Recruitment of shieldin to DSBs, via the ATM-RNF8-RNF168-53BP1-RIF1 axis, promotes NHEJ-dependent repair of intrachromosomal breaks, immunoglobulin class-switch recombination (CSR), and fusion of unprotected telomeres. |
| GO:0005694 chromosome | IEA GO_REF:0000044 | MODIFY | Summary: UniProt Chromosome location; the relevant chromosomal site is the DSB. Reason: The UniProt location note says SHLD2 localises to nuclear foci in response to DNA damage. GO:0035861 site of double-strand break (part_of chromosome) is the more precise location and is already on the gene. Proposed replacements: site of double-strand break Supporting Evidence: file:human/SHLD2/SHLD2-uniprot.txt Note=Localizes to nuclear foci in response to DNA damage. |
| GO:0010569 regulation of double-strand break repair via homologous recombination | IBA GO_REF:0000033 | MODIFY | Summary: SHLD2 negatively regulates HR by blocking end resection. Reason: The family-level IBA (node PTN001025652, seeded by human SHLD2) is sound, but the sign is known for SHLD2. Loss of SHLD2 restores resection and RAD51 foci in BRCA1-deficient cells, and wild-type long-form SHLD2 suppresses RAD51 focus formation. The precise term is GO:2000042, which the gene carries by IDA. The PAINT curator placed the general regulation term and not the NHEJ or isotype-switching terms at this node. That choice does not affect the human rows, which rest on human experiments. Propagation Review Root cause: TERM SCOPING PROBLEM Failure modes: GRANULARITY MISMATCH Sources checked: PANTHER:PTN001025652 · PTHR14495 ancestral node (Eumetazoa), IBD seeded by human SHLD2 SUPPORTS TRANSFER The node carries the unsigned regulation term. For the target, the sign is experimentally known to be negative, so the transfer is correct but less precise than the target's own IDA. UniProtKB:Q86V20 · human SHLD2, the node's seed and the target itself SUPPORTS TRANSFER The target's own experimental annotation grounds the IBD. This is expected and not circular. Proposed replacements: negative regulation of double-strand break repair via homologous recombination Supporting Evidence: PMID:29789392 However, the BRCA1‐mutant cells formed damage‐dependent nuclear RAD51 foci following FAM35A depletion, suggesting that 5′ end resection was more active in the absence of FAM35A (Fig 5B). PMID:30022168 Expression of full-length SHLD2-m1 and SHLD2S in BRCA1-KO SHLD2-KO cells also failed to suppress RAD51 IR-induced focus formation, unlike wild-type SHLD2 (ED Fig 10ab). |
| GO:0035861 site of double-strand break | IBA GO_REF:0000033 | ACCEPT | Summary: SHLD2 acts at DSB sites, inherited across the SHLD2 family node. Reason: SHLD2 is recruited to DSBs downstream of 53BP1-RIF1 and protects the ends there. The IBA node is seeded by human SHLD2's experimental annotation, which is expected and not circular. Supporting Evidence: PMID:29656893 Recruitment of shieldin to DSBs, via the ATM-RNF8-RNF168-53BP1-RIF1 axis, promotes NHEJ-dependent repair of intrachromosomal breaks, immunoglobulin class-switch recombination (CSR), and fusion of unprotected telomeres. |
| GO:0035861 site of double-strand break | IDA PMID:29656893 DNA Repair Network Analysis Reveals Shieldin as a Key Regula... | ACCEPT | Summary: SHLD2 is recruited to DSBs. Reason: Direct localisation of shieldin/SHLD2 to DSB sites, consistent with independent FokI-break recruitment in PMID:30022168, where SHLD2 OB-fold mutants still localise to DSBs. Supporting Evidence: PMID:29656893 Recruitment of shieldin to DSBs, via the ATM-RNF8-RNF168-53BP1-RIF1 axis, promotes NHEJ-dependent repair of intrachromosomal breaks, immunoglobulin class-switch recombination (CSR), and fusion of unprotected telomeres. PMID:30022168 Importantly, both mutants localized to DSB sites (ED Fig 10cd) and interacted with the other members of the Shieldin complex (ED Fig 10e). |
| GO:0035861 site of double-strand break | NAS PMID:29656893 DNA Repair Network Analysis Reveals Shieldin as a Key Regula... | ACCEPT | Summary: SHLD2 is recruited to DSBs. Reason: Same claim as the IDA row from the same paper; consistent. Supporting Evidence: PMID:29656893 Recruitment of shieldin to DSBs, via the ATM-RNF8-RNF168-53BP1-RIF1 axis, promotes NHEJ-dependent repair of intrachromosomal breaks, immunoglobulin class-switch recombination (CSR), and fusion of unprotected telomeres. |
| GO:0043247 telomere maintenance in response to DNA damage | NAS PMID:29656893 DNA Repair Network Analysis Reveals Shieldin as a Key Regula... | KEEP AS NON CORE | Summary: Shieldin acts on deprotected telomeres, a specialised DSB-end context. Reason: The cited evidence is NHEJ-mediated fusion of unprotected telomeres, i.e. shieldin treats deprotected telomere ends as DSB ends. This is a damage response at telomeres, not routine telomere-length maintenance. The term is kept as a non-core context of the end-protection function. It is not core, and the fusion readout itself is pathological. A more literal reading of GO:0043247 would need shieldin-CST-dependent fill-in at telomeres, which is not in the cached literature. Supporting Evidence: PMID:29656893 In the physiological context, the 53BP1/RIF1/REV7 pathway promotes NHEJ, telomere maintenance, and immunoglobulin class-switch recombination (CSR). |
| GO:0045830 positive regulation of isotype switching | IDA PMID:29656893 DNA Repair Network Analysis Reveals Shieldin as a Key Regula... | KEEP AS NON CORE | Summary: Shieldin promotes immunoglobulin class-switch recombination. Reason: Shieldin loss impairs CSR in B cells, a specialised physiological use of shieldin-dependent end joining. This experimental annotation is kept as non-core, alongside isotype switching (GO:0045190) proposed above for the mechanistic participation. Supporting Evidence: PMID:29656893 Recruitment of shieldin to DSBs, via the ATM-RNF8-RNF168-53BP1-RIF1 axis, promotes NHEJ-dependent repair of intrachromosomal breaks, immunoglobulin class-switch recombination (CSR), and fusion of unprotected telomeres. |
| GO:2000042 negative regulation of double-strand break repair via homologous recombination | IDA PMID:29656893 DNA Repair Network Analysis Reveals Shieldin as a Key Regula... | ACCEPT | Summary: SHLD2 suppresses resection-dependent homologous recombination. Reason: Loss of SHLD2 restores damage-induced RAD51 foci in BRCA1-deficient cells, and wild-type (long-form) SHLD2 suppresses them, while the ssDNA-binding-defective m1 mutant and the short splice form SHLD2S do not. This is a direct consequence of SHLD2's own ssDNA-binding end-protection activity, so it is core. Supporting Evidence: PMID:29789392 However, the BRCA1‐mutant cells formed damage‐dependent nuclear RAD51 foci following FAM35A depletion, suggesting that 5′ end resection was more active in the absence of FAM35A (Fig 5B). PMID:30022168 Importantly, both mutants localized to DSB sites (ED Fig 10cd) and interacted with the other members of the Shieldin complex (ED Fig 10e). Therefore, the SHLD2-m1 and SHLD2S mutants are defective in suppressing HR. |
| GO:2001034 positive regulation of double-strand break repair via nonhomologous end joining | IDA PMID:29656893 DNA Repair Network Analysis Reveals Shieldin as a Key Regula... | ACCEPT | Summary: SHLD2 promotes non-homologous end joining. Reason: Shieldin identification, together with FAM35A depletion and rescue, shows that SHLD2 promotes NHEJ. It does so as an end-protecting ssDNA-binding scaffold, not by catalysing end processing or ligation, so the regulation term (rather than direct NHEJ participation) is appropriate. Supporting Evidence: PMID:29656893 We propose that 53BP1-RIF1 act upstream to facilitate the recruitment of the shieldin complex, which we postulate shields DSBs against resection and thereby promotes NHEJ. PMID:29789392 Expression of FAM35A isoform 1 in 293T cells (Fig 4D) restored NHEJ to normal levels (Fig 4E). |
| GO:2001034 positive regulation of double-strand break repair via nonhomologous end joining | IDA PMID:29789392 FAM35A associates with REV7 and modulates DNA damage respons... | ACCEPT | Summary: FAM35A depletion reduces NHEJ, and long-isoform expression restores it. Reason: A direct NHEJ reporter assay with rescue by FAM35A isoform 1 (the 904-residue long form, UniProt Q86V20-2). Supporting Evidence: PMID:29789392 Expression of FAM35A isoform 1 in 293T cells (Fig 4D) restored NHEJ to normal levels (Fig 4E). |
| GO:2001034 positive regulation of double-strand break repair via nonhomologous end joining | NAS PMID:29656893 DNA Repair Network Analysis Reveals Shieldin as a Key Regula... | ACCEPT | Summary: SHLD2 promotes non-homologous end joining. Reason: Same claim as the IDA rows; consistent. Supporting Evidence: PMID:29656893 We propose that 53BP1-RIF1 act upstream to facilitate the recruitment of the shieldin complex, which we postulate shields DSBs against resection and thereby promotes NHEJ. |
| GO:0003697 single-stranded DNA binding | IDA PMID:30022168 The shieldin complex mediates 53BP1-dependent DNA repair. | NEW | Summary: The SHLD2 C-terminal OB-fold region binds ssDNA directly, with about 10 nM affinity. Reason: Purified SHLD2C (residues 421-904 of the long form) binds a radiolabelled ssDNA probe in EMSA with about 10 nM affinity and prefers ssDNA over dsDNA. Binding does not require SHLD1. An OB-fold mutant (m1) abolishes binding and fails to suppress HR, although it still localises to DSBs and assembles with shieldin. This direct biochemical activity is the effector mechanism of shieldin and is missing from GOA. The ASTE1 paper restates it independently ("binds to single-stranded DNA ends ... through OB-fold domains"). Supporting Evidence: PMID:30022168 We estimate the binding affinity of the SHLD2C-ssDNA interaction to be ~10 nM (Fig 4e and ED Fig 9c). We conclude that SHLD2 possesses ssDNA-binding activity. PMID:30022168 competition with unlabeled oligonucleotides revealed that SHLD2C preferentially binds to ssDNA over dsDNA (Fig 4c) PMID:34354233 The SHLD2 subunit binds to single-stranded DNA ends and blocks end resection through OB-fold domains. |
Loading supporting content…
Download this section (compressed HTML)Q: Is the 835-residue SHLD2 splice form, which lacks 69 residues of OB fold B and is UniProt's displayed sequence, expressed as protein at physiologically relevant levels? Does it act as a dominant-negative or regulatory variant of shieldin end protection?
Q: GO has no cellular-component term for the shieldin complex (ComplexPortal and HGNC group 1656 define it). Should one be created so that SHLD1, SHLD2, SHLD3 and REV7 can carry part_of shieldin complex?
Experiment: Compare the endogenous protein abundance of the 904- and 835-residue SHLD2 isoforms across cell types and B-cell activation states, and test whether the 835-residue form affects end protection by the long form when co-expressed.
Hypothesis: The ratio of the two splice forms tunes how strongly shieldin suppresses resection.
Experiment: Determine the structure of the SHLD2 C-terminal OB-fold region bound to SHLD1 and a 3' overhang ssDNA substrate, and map the polarity and minimum length of ssDNA recognised.
Hypothesis: SHLD2 OB folds bind 3' overhangs with an RPA-like polarity that positions SHLD1 to recruit CST-Pol alpha fill-in.
Loading supporting content…
Download this section (compressed HTML)Loading supporting content…
Download this section (compressed HTML)Loading supporting content…
Download this section (compressed HTML)