ACTL7A (Arp7A, T-actin 2) is a testis-enriched actin-related protein that provides structural scaffolding for the sperm head rather than acting as a conventional actin. It couples a 435-residue actin fold to a unique disordered 64-residue N-terminal extension that is not present in actin and that carries its best-defined biochemical activity: the extension binds the tandem LIM2-3 domains of testin (TES), forming an ACTL7A-TES-ENAH heterotrimer whose structure has been solved. ACTL7A concentrates in the subacrosomal layer of the perinuclear theca, the F-actin- and keratin-containing plate called the acroplaxome that lies between the developing acrosome and the nuclear envelope, and is also found transiently on the Golgi apparatus during acrosome biogenesis, in the germ-cell nucleus, and later in postacrosomal regions of the sperm head. Loss of ACTL7A abolishes subacrosomal filamentous actin specifically, without affecting F-actin elsewhere in the spermatid, and the acrosome then mislocalises its granule and peels away from the nucleus; ACTL7A therefore organises a spatially restricted actin network and anchors the acrosome to the nuclear envelope. It binds several other perinuclear-theca proteins, including ACTL9 and cylicin-1. Its actin fold is intermediate in character: the nucleotide cleft is largely conserved, but the histidine that triggers ATP hydrolysis on filament incorporation and the bulk of the inter-protomer filament interface are not, and the protein does not assemble into filaments when expressed in somatic cells. In humans, bi-allelic ACTL7A variants cause autosomal-recessive spermatogenic failure 86, in which morphologically unremarkable sperm carry acrosomal and perinuclear-theca ultrastructural defects and fail to activate the oocyte because phospholipase C zeta is depleted or mislocalised, producing total fertilization failure or early embryonic arrest that can be bypassed by artificial oocyte activation.
Definition: A cytoskeletal plate that forms the subacrosomal region of the perinuclear theca of developing spermatids and spermatozoa, lying between the inner acrosomal membrane and the nuclear envelope and containing filamentous actin and keratin 5, and which anchors the developing acrosome to the nuclear envelope. It is bounded caudally by a marginal ring apposed to the leading edge of the acrosome.
Justification: GO's perinuclear theca definition already names its two parts, a subacrosomal layer and, continuing caudally beyond the acrosomic system, the postacrosomal sheath, and GO has a term for the caudal part, GO:0033150 cytoskeletal calyx, but none for the subacrosomal layer. The gap has a concrete curation cost that this review encountered directly: ACTL7A's characterised location is the acroplaxome, and with no term available curators have had to choose between GO:0001669 acrosomal vesicle, which is the wrong compartment, its definition being the hydrolase-containing lysosome-derived structure with exact synonyms acrosome and acrosomal granule, and GO:0033011, which is correct but merges the subacrosomal and postacrosomal pools that ACTL7A demonstrably moves between during spermiogenesis. The same term is needed for a growing set of genes assigned to this structure, including ACTL9, CYLC1, ACTRT2, SPACA1, MYO6 and IZUMO3. OLS and QuickGO searches for acroplaxome and subacrosomal return no GO term.
Parent term: perinuclear theca
Supporting Evidence:
Definition: The process by which the developing acrosome becomes and remains physically anchored to the nuclear envelope of a spermatid, mediated by the intervening acroplaxome and involving linkage of the inner acrosomal membrane through subacrosomal cytoskeletal and theca proteins to nuclear envelope proteins.
Justification: GO:0001675 acrosome assembly is defined narrowly as the formation of the acrosome from the spermatid Golgi and so does not cover the distinct, separately mutable step of holding the formed acrosome onto the nucleus. That step is what fails in a coherent and expanding set of human infertility genes: ACTL7A loss causes acrosomes to peel off and bubble-shaped acrosomes, cylicin-1 loss causes acrosome detachment from the nucleus through an inner-acrosomal-membrane to nuclear-envelope sandwich with SPACA1 and FAM209, and ACTL9 loss disorganises the perinuclear theca. Without a term, all of these are annotated to acrosome assembly, which conflates biogenesis with attachment and loses the shared mechanism.
Parent term: acrosome assembly
Supporting Evidence:
| GO Term | Evidence | Action | Reason |
|---|---|---|---|
| GO:0005634 nucleus | IBA GO_REF:0000033 | KEEP AS NON CORE | Summary: Nuclear localisation of ACTL7A is genuine, but no nuclear molecular activity has been shown, so the is_active_in qualifier overstates what is known. Kept as non-core. Reason: Resolving the five WITH/FROM tokens shows this is not a fold artefact even though the PANTHER family mixes cytoplasmic and nuclear actin-related proteins. Two sources are yeast and Candida ARP9, bona fide nuclear SWI/SNF-complex subunits with their own IDA nucleus evidence, and one is an unreviewed Trypanosoma protein; but the mouse ortholog Actl7a also carries its own experimental nucleus evidence, and mouse ACTL7A is seen inside spermatid nuclei by 3D confocal imaging. The term is therefore right. What is absent is any demonstrated nuclear function: the reported nuclear roles, namely HDAC association and docking on INO80 and SWI/SNF HSA domains, rest on a preprint and in-silico modelling. is_active_in asserts activity at this location, which the evidence does not reach, and ACTL7A's characterised function is at the acroplaxome, so this is non-core; located_in would be the safer qualifier. No specificity upgrade is proposed on this particular row, because the donor set is genuinely heterogeneous - a yeast SWI/SNF subunit, a Candida homologue, a trypanosome protein and two mammalian ACTL7 proteins - and GO:0005634 is the correct least common ancestor of what those donors support. The descendant GO:0001673 male germ cell nucleus would fit the mammalian evidence, but it would not fit the fungal and protozoan donors, so it belongs on the ortholog-specific ISS row if anywhere; see suggested_questions. Propagation Review Root cause: NO FAILURE NON CORE Sources checked: MGI:MGI:1343051 · mouse Actl7a (true ortholog) SUPPORTS TRANSFER Carries its own nucleus EXP/IDA (PMID:12672658, PMID:21278383, PMID:36734600) and the more specific GO:0001673 male germ cell nucleus by IDA. SGD:S000004636 · S. cerevisiae ARP9 SUPPORTS SOURCE BUT NOT TARGET Nuclear SWI/SNF and RSC complex subunit with nucleus IDA (PMID:11011149); a nuclear ARP, not an ACTL7 ortholog, so it supports the source not the transfer. CGD:CAL0000196900 · Candida albicans Arp9 SUPPORTS SOURCE BUT NOT TARGET Resolves to unreviewed TrEMBL entry Q5A9X7; RSC-type complex IDA (PMID:33151931), nucleus only by IBA/IEA. UniProtKB:Q57ZL0 · Trypanosoma brucei, unreviewed SOURCE WEAK OR INFERRED No reviewed entry and no assigned protein name; nucleus IDA PMID:28727848. Reported rather than hidden, since an unreviewed source is weaker support. PANTHER:PTN008986520 · PANTHER tree node NOT RELEVANT An internal tree node, not a protein; carries no evidence of its own. Supporting Evidence: PMID:36734600 Our intracellular localization data show ACTL7A to be dynamically present within the nucleus and subacrosomal space and later associated with postacrosomal regions of developing spermatids. PMID:12672658 T-ACTIN-1 was found in the cytoplasm and T-ACTIN-2 was located in the nucleus |
| GO:0005737 cytoplasm | IBA GO_REF:0000033 | KEEP AS NON CORE | Summary: True but uninformative root-level compartment; the specific location, perinuclear theca, is separately annotated. Reason: Every WITH/FROM token resolves inside the ACTL7 clade: mouse Actl7a and rat Actl7a carry experimental cytoplasm evidence, mouse Actl7b carries an IDA that needs the caveat below, and one token is ACTL7A itself, a self-referential IBA recording a PAINT curator's judgement rather than a circular inference. The term is not in doubt, because the true ortholog supports it independently. GO:0005737 is simply a parent of the informative terms already present, since perinuclear theca is a cytoplasm descendant, so it adds no resolution and is not core. One caution on the evidence trail: PMID:12672658 names T-ACTIN-1 as mACTL7B and T-ACTIN-2 as mACTL7A, and assigns T-ACTIN-2 to the nucleus and T-ACTIN-1 to the cytoplasm, so that paper supports the mouse cytoplasm annotation for the paralogue Actl7b, not for Actl7a. The cytoplasmic evidence that actually applies to ACTL7A comes from PMID:21278383 and PMID:36734600. Propagation Review Root cause: NO FAILURE NON CORE Failure modes: GRANULARITY MISMATCH Sources checked: MGI:MGI:1343051 · mouse Actl7a SUPPORTS TRANSFER Cytoplasm IDA PMID:21278383 and IDA PMID:36734600. Note that the EXP PMID:12672658 evidence on this term traces to a paper that places T-ACTIN-2 (mACTL7A) in the nucleus and T-ACTIN-1 (mACTL7B) in the cytoplasm. MGI:MGI:1343053 · mouse Actl7b (paralogue) SUPPORTS SOURCE BUT NOT TARGET Cytoplasm IDA PMID:12672658 - and that IDA is precisely the paralogue-specific half of the experiment, since the same paper places T-ACTIN-1 (mACTL7B) in the cytoplasm and T-ACTIN-2 (mACTL7A) in the nucleus. A paralogue donor is legitimate for IBA, but this one supports cytoplasm for ACTL7B rather than for ACTL7A. The transfer survives only because the true ortholog has independent cytoplasm IDAs. RGD:1304697 · rat Actl7a SUPPORTS TRANSFER Cytoplasm EXP PMID:21278383. UniProtKB:Q9Y615 · ACTL7A itself SUPPORTS TRANSFER Self-referential IBA; valid by construction, recording the PAINT curator judging the compartment. PANTHER:PTN001377938 · PANTHER tree node NOT RELEVANT Internal tree node, not a protein. Supporting Evidence: PMID:21278383 In contrast, when expressed in HeLa cells, GFP-Arp7A is cytoplasmic and does not assemble into filaments or co-localize with and/or affect the localization of Tes and Mena at focal adhesions (data not shown). PMID:36734600 Our intracellular localization data show ACTL7A to be dynamically present within the nucleus and subacrosomal space and later associated with postacrosomal regions of developing spermatids. |
| GO:0005198 structural molecule activity | IBA GO_REF:0000033 | ACCEPT | Summary: ACTL7A's core molecular function, and not a vague fallback: this term is the level GO's own phylogenetic pipeline deliberately chose for this clade, having explicitly rejected the specific child GO:0005200 on the same node and the same date. Reason: The cached PAINT record for this family settles the level at which the structural claim should be made. At the ancestral actin node PTN000940351, GO:0005200 structural constituent of cytoskeleton is IBD-supported from experimentally annotated conventional actins and Arp2/Arp3 (seeds include UniProtKB:P60709 beta-actin, P61160 ARP2 and P61158 ARP3). On descent into the divergent ARP branches it is then negated: PTN008986528, the node that carries ACTL7A, has GO:0005200 as an IRD with negated=true, and carries GO:0005198 by IBA instead - both rows dated 2025-08-05, and GO:0005198 appears nowhere else in the family table. So this row is not a parent left standing for want of a better term; it is the product of a single, deliberate, current curatorial act that replaced the child with the parent for exactly this clade. That judgement is independently what this review's own structure audit supports: ACTL7A's inter-protomer filament interface is 42.3% identical to actin's against 93-100% for conventional actins, and the ATP-hydrolysis trigger His161 is replaced by tyrosine, so the route by which real actins earn GO:0005200 is closed for ACTL7A. Meanwhile the claim GO:0005198 does make - contributing to the structural integrity of a complex - is exactly what the phenotypes show: knock-out spermatids lose subacrosomal F-actin entirely while retaining it elsewhere, their acrosomes peel off the nucleus, and patient sperm show irregular perinuclear theca. Accepted as the core molecular function on that basis. An earlier round of this review promoted GO:0005200 here instead, arguing that ACTL7A's residence in GO:0033011 (a GO:0005856 descendant) made the specific term apply; that argument is withdrawn, because residing in a cytoskeletal structure is not the same as being a structural constituent of it, and because the IRD is dated after all of the knock-out and patient literature, so it cannot be dismissed as a judgement made without that evidence. Propagation Review Root cause: NO FAILURE CORE Sources checked: PANTHER:PTN000940351 · ancestral actin node of PTHR11937 SUPPORTS TRANSFER An internal tree node rather than a protein, but not therefore uninformative. It is the node at which GO:0005200 is IBD-supported from experimentally annotated conventional actins and Arp2/Arp3, and it is the seed for both the GO:0005198 IBA and the GO:0005200 IRD on the descendant ACTL7 node. PANTHER:PTN008986528 · PANTHER node carrying ACTL7A SUPPORTS TRANSFER The decisive source. This node carries GO:0005198 by IBA and, on the same date of 2025-08-05, GO:0005200 as an IRD with negated set to true, which is an explicit rejection of the specific child for the clade containing ACTL7A rather than a silent omission. Recording this node as merely not-a-protein would drop the one fact about it that bears on the decision. Supporting Evidence: PMID:36734600 Significantly, we found a complete loss of subacrosomal filamentous actin (F-actin) structures in knock-out spermatids suggesting a regulatory role for subacrosomal F-actin. PMID:36734600 we propose that ACTL7A forms a unique testis-specific protein complex to anchor and stabilize acrosomal adherence to the acroplaxome at least in part by facilitating the presence of F-actin in the subacrosomal space file:human/ACTL7A/ACTL7A-bioinformatics/RESULTS.md * **The filament interface is not retained.** ACTL7A matches actin at 42.3% of the 79 inter-protomer contact positions, versus 93.2% for conventional actins and 55.7% for Arp1, which does form a filament (in dynactin). file:interpro/panther/PTHR11937/PTHR11937-paint.tsv PTHR11937 PTN008986528 GO:0005198 F IBA false PANTHER:PTN000940351 |
| GO:0001669 acrosomal vesicle | IEA GO_REF:0000120 | MODIFY | Summary: ACTL7A lies in the subacrosomal layer between the acrosome and the nuclear envelope, not inside the acrosomal vesicle. Replace with perinuclear theca, which GO defines as including the subacrosomal layer. Reason: This is the compartment-resolution error that the 2011 structural paper explicitly corrected: ACTL7A was reassigned from the nucleus of spermatids and the acrosome to the subacrosomal layer, also called the acroplaxome. GO:0001669 is defined as the hydrolase-containing structure derived from the lysosome, with exact synonyms acrosome and acrosomal granule, so it denotes a membrane-bounded vesicle lumen; ACTL7A is a cytosolic and cytoskeletal protein on the outside of the inner acrosomal membrane and cannot be located_in it. The GO:0033011 perinuclear theca definition explicitly comprises a subacrosomal layer and the postacrosomal sheath, which is where every high-resolution observation of ACTL7A places it, and that term is already annotated to this gene by IDA. Modifying rather than removing because the biology is adjacent and the intent of the annotation is clear. Propagation Review Root cause: TERM SCOPING PROBLEM Failure modes: COMPARTMENT OR COMPLEX MISMATCH Sources checked: UniProtKB:Q9QY84 · mouse Actl7a SUPPORTS SOURCE BUT NOT TARGET The mouse acrosome annotation reflects the same light-microscopy resolution limit; the better-resolved mouse call is perinuclear theca (IDA PMID:41169243). UniProtKB-SubCell:SL-0007 · UniProt SubCell acrosome SUPPORTS SOURCE BUT NOT TARGET Keyword-level mapping; carries no compartment resolution of its own. Proposed replacements: perinuclear theca Supporting Evidence: PMID:21278383 Previous observations have shown that Arp7A is expressed in testis and is associated with the nucleus of spermatids (18, 19). We have now extended these earlier studies and shown that Arp7A is actually localized in the subacrosomal layer, which is also known as the acroplaxome (40). |
| GO:0005634 nucleus | IEA GO_REF:0000120 | KEEP AS NON CORE | Summary: Real localisation with no known nuclear function; retained as a secondary compartment. Reason: Same judgement as the IBA and HDA nucleus rows. Mouse ortholog immunofluorescence and human sperm-nucleus proteomics both place ACTL7A in the germ-cell nucleus, so the term is not wrong, but no nuclear activity is established and the characterised function is at the acroplaxome. The more precise term GO:0001673 male germ cell nucleus is available and is what the mouse ortholog carries by IDA; refining all four nucleus rows to it would be an improvement, and is raised in suggested_questions rather than applied here so that the rows stay mutually consistent. Propagation Review Root cause: NO FAILURE NON CORE Sources checked: UniProtKB:Q9QY84 · mouse Actl7a SUPPORTS TRANSFER Nucleus IDA PMID:36734600 and male germ cell nucleus IDA PMID:12672658. Supporting Evidence: PMID:36734600 Our intracellular localization data show ACTL7A to be dynamically present within the nucleus and subacrosomal space and later associated with postacrosomal regions of developing spermatids. |
| GO:0005737 cytoplasm | IEA GO_REF:0000120 | KEEP AS NON CORE | Summary: Correct but root-level; the specific compartment is separately annotated. Reason: Transferred from the mouse ortholog, which has experimental cytoplasm evidence. Kept because it is true, marked non-core because perinuclear theca and Golgi apparatus give the resolution that matters. Propagation Review Root cause: NO FAILURE NON CORE Failure modes: GRANULARITY MISMATCH Sources checked: UniProtKB:Q9QY84 · mouse Actl7a SUPPORTS TRANSFER Cytoplasm IDA PMID:21278383 and PMID:36734600. |
| GO:0005794 Golgi apparatus | IEA GO_REF:0000120 | KEEP AS NON CORE | Summary: Genuine but transient association during acrosome biogenesis rather than a steady-state residence. Reason: Well supported by immunofluorescence in rodent testis sections, where ACTL7A co-localises with GM130 before and throughout acrosome formation. Kept as non-core because it reflects a stage-specific stop on the route to the acroplaxome, where the protein does its structural work, rather than a compartment in which its function is exercised. Propagation Review Root cause: NO FAILURE NON CORE Sources checked: UniProtKB:Q9QY84 · mouse Actl7a SUPPORTS TRANSFER Golgi apparatus IDA PMID:21278383. Supporting Evidence: PMID:21278383 Arp7A is associated with the Golgi apparatus prior to and throughout the biogenesis of the acrosome |
| GO:0005856 cytoskeleton | IEA GO_REF:0000044 | KEEP AS NON CORE | Summary: True by subsumption, since perinuclear theca is a cytoskeleton descendant, but adds no information. Reason: Derived from the UniProt SubCell cytoskeleton keyword. GO:0033011 perinuclear theca is itself a descendant of GO:0005856, so this row is redundant rather than wrong. Marked non-core; the informative statement is the perinuclear theca annotation. Propagation Review Root cause: NO FAILURE NON CORE Failure modes: GRANULARITY MISMATCH Sources checked: UniProtKB-SubCell:SL-0090 · UniProt SubCell cytoskeleton SUPPORTS TRANSFER Keyword mapping; the target already carries the specific descendant term. |
| GO:0007010 cytoskeleton organization | IEA GO_REF:0000108 | MODIFY | Summary: The process is right but reached by the wrong route and far too general. The evidenced process is actin cytoskeleton organization, from the mouse knock-out's complete and spatially restricted loss of subacrosomal F-actin. Reason: The WITH/FROM field on this row is not a gene product but the GO term GO:0005200, so this is an inter-ontology molecular-function to biological-process inference from the gene's own 1999 TAS row, which is itself a name-derived annotation from a positional-cloning paper with no functional experiment. The process term therefore inherits weak evidence at second hand. Independent evidence supports a more specific term: Actl7a knock-out spermatids lose subacrosomal F-actin completely while retaining F-actin everywhere else, which is actin cytoskeleton organization, a child of GO:0007010. Neither human ACTL7A nor mouse Actl7a currently carries any annotation in that branch, so moving this row there fills a real gap rather than duplicating one. Whether ACTL7A is itself the filament or nucleates and anchors filaments made by conventional actins is explicitly unresolved, which is why the non-committal involved_in GO:0030036 is preferred over a polymerisation or nucleation term. Propagation Review Root cause: TERM SCOPING PROBLEM Failure modes: GRANULARITY MISMATCH SOURCE EVIDENCE WEAK Sources checked: GO:0005200 · structural constituent of cytoskeleton (inter-ontology link, not a gene product) SOURCE WEAK OR INFERRED The source is the gene's own TAS row citing PMID:10373328, a cloning and mapping paper with no functional data, so the inferred process rests on a name-based assertion. Proposed replacements: actin cytoskeleton organization Supporting Evidence: PMID:36734600 The KO spermatids did not lose F-actin in other cellular regions, indicating that ACTL7A is not globally required for F-actin formation within developing spermatids PMID:36734600 it is not yet clear whether ACTL7A may be the filament-forming actin in the acroplaxome, or whether its role is to nucleate, stabilize formation, and/or anchor F-actin formed by conventional ARPs to the nuclear surface |
| GO:0005515 protein binding | IPI PMID:21278383 Molecular recognition of the Tes LIM2-3 domains by the actin... | MODIFY | Summary: The TES interaction is one of the best-characterised facts about this gene and deserves an informative term: ACTL7A's N-terminal extension is a tandem-LIM-domain ligand. Reason: The WITH/FROM accession resolves to UniProtKB:Q9UGI8, human testin (TES), a genuine partner rather than a screen artefact: the interaction was mapped biochemically to ACTL7A residues 28-49 and the LIM2 and LIM3 domains of TES, crystallised as a ternary complex with the ENAH EVH1 domain (PDB 2XQN), and abolished by single alanine-to-tyrosine substitutions at ACTL7A Ala31 or Ala41. GO:0030274 LIM domain binding is a GO:0005515 descendant that states exactly this and replaces an uninformative parent. Note that the binding surface is the disordered N-terminal extension unique to ACTL7A, not the actin fold. Proposed replacements: LIM domain binding Supporting Evidence: PMID:21278383 The crystal structure of the 1-65(Arp7A)·LIM2-3(Tes)·EVH1(Mena) complex reveals that residues 28-49 of Arp7A contact the LIM2-3 domains of Tes. PMID:21278383 The N-terminal 65-residue extension to the actin-like fold of Arp7A interacts directly with Tes. |
| GO:0001675 acrosome assembly | IEA GO_REF:0000107 | ACCEPT | Summary: Core process. Ortholog transfer from mouse Actl7a, which carries two independent IMP annotations for this term, and matched by the human disease phenotype. Reason: The single WITH/FROM source, mouse Actl7a, carries acrosome assembly by IMP from both the knock-in and the knock-out study, so this is an ortholog transfer resting on experimental evidence rather than on further inference. In humans, ACTL7A variants produce acrosomal ultrastructural defects directly, so the process is independently supported in the target species. Propagation Review Root cause: NO FAILURE CORE Sources checked: UniProtKB:Q9QY84 · mouse Actl7a SUPPORTS TRANSFER Acrosome assembly IMP PMID:32923619 and IMP PMID:36734600. Supporting Evidence: PMID:36734600 Through the generation of an Actl7a knock-out mouse model, we consistently observed disruption of acrosomal biogenesis with abnormal migration of the acrosomal granule and peeling acrosomes during spermatid elongation. PMID:32923619 This mutation is deleterious and causes sperm acrosomal ultrastructural defects. |
| GO:0007286 spermatid development | IEA GO_REF:0000107 | ACCEPT | Summary: Core process. Mouse knock-out disrupts the acrosome-acroplaxome-manchette complex and sperm head shaping. Reason: The mouse ortholog carries spermatid development by IMP. The phenotype is specific to spermiogenesis: loss of ACTL7A damages formation of the acrosome-acroplaxome-manchette complex and produces small, misshapen sperm heads, while flagellar integrity, motility and capacitation are unaffected. Propagation Review Root cause: NO FAILURE CORE Sources checked: UniProtKB:Q9QY84 · mouse Actl7a SUPPORTS TRANSFER Spermatid development IMP PMID:37667331. Supporting Evidence: PMID:37667331 Immunofluorescence and TEM analyses revealed that the deletion of ACTL7A damaged the formation of acrosome-acroplaxome-manchette complex, leading to abnormalities in the shaping of sperm heads. |
| GO:0009566 fertilization | IEA GO_REF:0000107 | ACCEPT | Summary: Core process, and the strongest human evidence in the record: bi-allelic ACTL7A variants cause total fertilization failure. Reason: Mouse Actl7a carries fertilization by IMP from two independent knock-out and knock-in studies, and the human genetics is direct: patients with bi-allelic ACTL7A variants show total fertilization failure or early embryonic arrest after IVF and ICSI. The term is correct but does not capture the mechanism, which is failure of sperm-borne oocyte activation; the descendant term GO:0007343 egg activation is proposed as an additional annotation rather than as a replacement, since fertilization remains true. Propagation Review Root cause: NO FAILURE CORE Sources checked: UniProtKB:Q9QY84 · mouse Actl7a SUPPORTS TRANSFER Fertilization IMP PMID:32923619 and IMP PMID:35921706. Supporting Evidence: PMID:35921706 Actl7a deficiency led to malformed formation of sperm acrosomes, male infertility, fertilization failure during in vitro fertilization (IVF) and intracytoplasmic sperm injection (ICSI), and reduced sperm-zona pellucida (ZP) binding ability. |
| GO:0033011 perinuclear theca | IEA GO_REF:0000107 | ACCEPT | Summary: The correct and core location. GO defines the perinuclear theca as comprising a subacrosomal layer, which is where ACTL7A sits. Reason: Ortholog transfer from mouse Actl7a, which carries perinuclear theca by IDA. This is the best-resolved compartment assignment available for ACTL7A: the 2011 structural study localised it specifically to the subacrosomal layer, and patient sperm show irregular perinuclear theca ultrastructure. GO lacks a term for the acroplaxome itself, so GO:0033011 is the most specific available term, and an acroplaxome term is proposed below. Propagation Review Root cause: NO FAILURE CORE Sources checked: UniProtKB:Q9QY84 · mouse Actl7a SUPPORTS TRANSFER Perinuclear theca IDA PMID:41169243. Supporting Evidence: PMID:21278383 We have now extended these earlier studies and shown that Arp7A is actually localized in the subacrosomal layer, which is also known as the acroplaxome PMID:34727571 TEM showed that the ACTL7A variants caused ultrastructural defects in the acrosome and perinuclear theca. |
| GO:0033011 perinuclear theca | IDA GO_REF:0000052 | ACCEPT | Summary: Direct human immunofluorescence evidence for the core location, independent of the ortholog transfer above. Reason: Curated from immunofluorescence imaging of human tissue. This is the only human direct-assay annotation that places ACTL7A in the compartment where its characterised function is exercised, and it agrees with the rodent testis-section work and with the perinuclear-theca ultrastructural defects seen in patient sperm. Supporting Evidence: PMID:41169243 ACTL7A, another essential PT protein, is required for acrosome biogenesis and mediates the tethering of the acrosomal outer membrane to the nucleus. |
| GO:0005634 nucleus | ISS GO_REF:0000024 | MODIFY | Summary: This transfer landed one level above the term its own donor holds: mouse Actl7a carries GO:0001673 male germ cell nucleus by IDA, so the specific term should have come across. Refine downward. Reason: A curator-judged sequence-similarity transfer whose single donor is mouse Actl7a, and that donor carries both GO:0005634 and the descendant GO:0001673 male germ cell nucleus, the latter by IDA from the study that first reported nuclear T-ACTIN-2 in haploid germ cells. A propagated annotation that is less precise than the evidence its own donor holds loses information for no reason, and the fix is to refine downward rather than to leave the parent standing. This applies to this row specifically and not to the other three nucleus rows, which is why their actions differ: the phylogenetic row's donor set also contains a yeast SWI/SNF subunit, a Candida homologue and a trypanosome protein, for which a germ-cell term is meaningless, so GO:0005634 is the correct least common ancestor there; the GO_REF:0000120 row's WITH/FROM mixes the ortholog with a generic subcellular-location keyword; and the HDA row is a human experimental observation best left as the curator recorded it. Every observation of ACTL7A protein in a nucleus, in either species, has been in a spermatocyte, spermatid or sperm, so the specific term fits the target. Still non-core in substance, since no nuclear molecular function has been demonstrated. GOA carries two identical rows for this term and evidence code; both should be refined the same way. Propagation Review Root cause: TERM SCOPING PROBLEM Failure modes: GRANULARITY MISMATCH Sources checked: UniProtKB:Q9QY84 · mouse Actl7a SUPPORTS TRANSFER Carries GO:0001673 male germ cell nucleus by IDA (PMID:12672658) as well as the general nucleus term; the specific term is the one that should transfer. Proposed replacements: male germ cell nucleus Supporting Evidence: PMID:12672658 Although the cellular locations of these two proteins are quite different (T-ACTIN-1 was found in the cytoplasm and T-ACTIN-2 was located in the nucleus), the expression of their proteins and mRNAs is controlled during development and limited during spermiogenesis. |
| GO:0005737 cytoplasm | ISS GO_REF:0000024 | KEEP AS NON CORE | Summary: Correct root-level compartment, superseded by the specific annotations. Reason: Sequence-similarity transfer from mouse Actl7a, which has cytoplasm by IDA. True, but GO:0005737 is a parent of the perinuclear theca and Golgi annotations already present. GOA carries two identical rows for this term and evidence code; both are non-core. Propagation Review Root cause: NO FAILURE NON CORE Failure modes: GRANULARITY MISMATCH Sources checked: UniProtKB:Q9QY84 · mouse Actl7a SUPPORTS TRANSFER Cytoplasm IDA PMID:21278383 and PMID:36734600. |
| GO:0005794 Golgi apparatus | ISS GO_REF:0000024 | KEEP AS NON CORE | Summary: Stage-specific association during acrosome biogenesis; real but not where the structural function is exercised. Reason: Transferred from mouse Actl7a, whose Golgi annotation is an IDA based on GM130 co-localisation in round spermatids. The association is confined to the period of acrosome biogenesis, when pro-acrosomal vesicles are being delivered to the acroplaxome, so it is a trafficking waypoint rather than the site of ACTL7A's structural role. Note that the same source also supports a colocalizes_with row, which is the better-hedged qualifier for this observation. Propagation Review Root cause: NO FAILURE NON CORE Sources checked: UniProtKB:Q9QY84 · mouse Actl7a SUPPORTS TRANSFER Golgi apparatus IDA PMID:21278383. Supporting Evidence: PMID:21278383 immunofluorescence analysis of testis sections reveals that Arp7A co-localizes with GM130 on the Golgi apparatus of round spermatids |
| GO:0007286 spermatid development | ISS GO_REF:0000024 | ACCEPT | Summary: Core process; curator-judged transfer from the mouse ortholog, duplicating the Ensembl-Compara row above. Reason: Same evidence base as the GO_REF:0000107 row: mouse Actl7a knock-out shows defective acrosome-acroplaxome-manchette complex formation and abnormal sperm head shaping, with normal flagella, motility and capacitation, so the deficit is specific to spermatid morphogenesis. Propagation Review Root cause: NO FAILURE CORE Sources checked: UniProtKB:Q9QY84 · mouse Actl7a SUPPORTS TRANSFER Spermatid development IMP PMID:37667331. Supporting Evidence: PMID:37667331 The new phenotype of small head sperm associated with loss of ACTL7A in patients was discovered, and further confirmed in Actl7a-KO mice. |
| GO:0009566 fertilization | ISS GO_REF:0000024 | ACCEPT | Summary: Core process; duplicates the Ensembl-Compara row and is independently supported by human patient genetics. Reason: Curator-judged transfer from mouse Actl7a, which carries fertilization by IMP. The human evidence is at least as strong: bi-allelic ACTL7A variants cause total fertilization failure, and the failure is bypassed by artificial oocyte activation, which localises the defect to the oocyte-activation step of fertilization. Propagation Review Root cause: NO FAILURE CORE Sources checked: UniProtKB:Q9QY84 · mouse Actl7a SUPPORTS TRANSFER Fertilization IMP PMID:32923619 and PMID:35921706. Supporting Evidence: PMID:34727571 AOA by calcium ionophore (A23187) after ICSI successfully rescued the TFF and achieved a live birth for the patient with ACTL7A variants. |
| GO:0005515 protein binding | IPI PMID:38573307 Disruption in CYLC1 leads to acrosome detachment, sperm head... | KEEP AS NON CORE | Summary: Genuine interaction with the perinuclear-theca protein cylicin-1, but no informative GO molecular-function term exists for it, so the row is kept and marked non-core. Reason: The WITH/FROM accession resolves to UniProtKB:P35663, human cylicin-1, a major constituent of the perinuclear theca whose own loss causes acrosome detachment from the nucleus in mice and men. The interaction is therefore biologically coherent rather than screen noise, and UniProt records that it may be relevant for proper acrosome attachment to the nuclear envelope. Bare protein binding carries no functional information, but unlike the TES row there is no specific descendant term available, since cylicins are not a recognised GO binding class, so MODIFY has no target. The informative content is that ACTL7A participates in a perinuclear-theca protein assembly, which is recorded in core_functions and by the protein-containing complex annotation. Supporting Evidence: PMID:38573307 Loss of cylicin-1 in mice leads to a high incidence of malformed sperm heads with acrosome detachment from the nucleus. |
| GO:0001669 acrosomal vesicle | IMP PMID:34727571 Novel bi-allelic variants in ACTL7A are associated with male... | MODIFY | Summary: The observation is an acrosomal-region signal in patient sperm, which at light-microscope resolution includes the acroplaxome immediately beneath the acrosome. Proposed for refinement to perinuclear theca rather than removal. Reason: The cited study used immunofluorescence and western blotting on patient spermatozoa and transmission electron microscopy, and reported defects in both the acrosome and the perinuclear theca; UniProt cites it for acrosomal localisation. The cached record is abstract-only, so the figures cannot be inspected, and this is not a proposal to delete a curator's experimental call. It is a term-scoping correction on definitional grounds: GO:0001669 denotes the lysosome-derived, hydrolase-containing vesicle, and a cytosolic cytoskeletal protein sitting on the nuclear side of the inner acrosomal membrane is not located_in it. Every high-resolution study of ACTL7A places it in the subacrosomal layer or, in mature sperm, the postacrosomal sheath, both of which are parts of the perinuclear theca, and GO:0033011 is already annotated to this gene by IDA. UNDECIDED was the conservative alternative and was considered; MODIFY is preferred because the argument does not depend on anything in the figures - it turns on what GO:0001669 denotes and on where a cytosolic protein can be - so deferring would withhold an actionable, checkable proposal rather than avoid an unsupported one. A curator with access to the figures should confirm; the underlying limitation is that GO has no acroplaxome term with which to state the distinction. Proposed replacements: perinuclear theca Supporting Evidence: PMID:34727571 TEM showed that the ACTL7A variants caused ultrastructural defects in the acrosome and perinuclear theca. PMID:21278383 We have now extended these earlier studies and shown that Arp7A is actually localized in the subacrosomal layer, which is also known as the acroplaxome |
| GO:0001669 acrosomal vesicle | IDA PMID:32923619 Disruption in ACTL7A causes acrosomal ultrastructural defect... | MODIFY | Summary: Same acrosome-versus-acroplaxome refinement as the other two acrosomal-vesicle rows; the compartment ACTL7A actually occupies is the perinuclear theca. Reason: This study characterised human and mouse ACTL7A-mutant sperm by transmission electron microscopy and immunofluorescence, and UniProt cites it for acrosomal localisation. The cached record is abstract-only. The refinement is proposed on the same definitional grounds as the IMP row: ACTL7A is a cytosolic protein of the subacrosomal cytoskeletal plate, not a constituent of the hydrolase-containing acrosomal vesicle, and the same paper's mouse knock-in work shows ACTL7A failing to attach to the acroplaxome when mutated, which presupposes acroplaxome rather than acrosomal residence. Modified rather than removed, since the observation is real and the intended compartment is adjacent. Proposed replacements: perinuclear theca Supporting Evidence: PMID:32923619 This mutation is deleterious and causes sperm acrosomal ultrastructural defects. PMID:35863052 Furthermore, mutant ACTL7A failed to attach to the acroplaxome and was discharged by cytoplasmic droplets, which led to the absence of ACTL7A in epididymal spermatozoa in mice. |
| GO:0001675 acrosome assembly | ISS GO_REF:0000024 | ACCEPT | Summary: Core process; curator-judged transfer duplicating the Ensembl-Compara row, and directly supported in human by the disease phenotype. Reason: Mouse Actl7a carries acrosome assembly by two independent IMP annotations. In the knock-out, the acrosomal granule mislocalises and acrosomes peel away during spermatid elongation, and in human patients ACTL7A variants produce acrosomal ultrastructural defects, so the process holds in the target species too. Propagation Review Root cause: NO FAILURE CORE Sources checked: UniProtKB:Q9QY84 · mouse Actl7a SUPPORTS TRANSFER Acrosome assembly IMP PMID:32923619 and IMP PMID:36734600. Supporting Evidence: PMID:36734600 Through the generation of an Actl7a knock-out mouse model, we consistently observed disruption of acrosomal biogenesis with abnormal migration of the acrosomal granule and peeling acrosomes during spermatid elongation. |
| GO:0005515 protein binding | IPI PMID:33626338 Homozygous pathogenic variants in ACTL9 cause fertilization ... | KEEP AS NON CORE | Summary: Genuine interaction with the perinuclear-theca actin-like protein ACTL9; no informative molecular-function term is available, so kept and marked non-core. Reason: The WITH/FROM accession resolves to UniProtKB:Q8TC94, human ACTL9, another testis-specific actin-like protein whose own bi-allelic loss causes abnormal perinuclear-theca ultrastructure, mislocalised phospholipase C zeta and total fertilization failure, the same phenotype as ACTL7A loss. The interaction is therefore strongly corroborated by convergent genetics rather than being an isolated pull-down. As for the cylicin-1 row, no GO molecular-function descendant describes binding to an actin-like protein of this kind, so there is no MODIFY target; the informative statement is shared complex membership. Supporting Evidence: PMID:33626338 ACTL9 variations led to abnormal ultrastructure of the perinuclear theca (PT), and PLCζ was absent in the head and present in the neck of the mutant sperm, which contributed to failed normal calcium oscillations in oocytes and subsequent TFF. |
| GO:0005634 nucleus | HDA PMID:21630459 Proteomic characterization of the human sperm nucleus. | KEEP AS NON CORE | Summary: Human sperm-nucleus proteomics detects ACTL7A; retained as a secondary compartment because perinuclear-theca carry-over cannot be excluded from a nuclear preparation alone and no nuclear activity is known. Reason: The study prepared sperm nuclei to high purity and reports the absence of acrosome, tails and mitochondria, so this is a careful dataset. The perinuclear theca, however, is a shell tightly apposed to the nuclear envelope and is not what such a preparation is designed to remove, so a proteomic hit alone cannot separate a truly intranuclear pool from theca carry-over. What makes the nuclear pool credible is independent immunofluorescence in mouse germ cells, including 3D confocal imaging in the knock-out study. Kept, but non-core: no nuclear molecular function has been demonstrated, and the mouse knock-out has normal chromatin compaction, with only a modest increase in TUNEL positivity as a hint of a nuclear role. Supporting Evidence: PMID:21630459 sperm nuclei were obtained through CTAB treatment and isolated to over 99.9% purity without any tail fragments, acrosome or mitochondria as assessed by optical microscopy and transmission electron microscopy PMID:36734600 Three-dimensional (3D) rendered confocal z-stacks of spermatids at different developmental steps illustrating a clear intranuclear localization pattern of the protein in spermatocytes, round spermatids, elongating spermatids, and condensing spermatids. |
| GO:0005794 Golgi apparatus | ISS GO_REF:0000024 | KEEP AS NON CORE | Summary: The best-hedged of the three Golgi rows: co-localisation during acrosome biogenesis is exactly what was observed. Reason: The underlying observation is co-localisation of ACTL7A with the Golgi marker GM130 in round spermatids, so colocalizes_with reports the evidence accurately rather than asserting residence. Non-core because the Golgi association is confined to the period of pro-acrosomal vesicle delivery and is not the site of ACTL7A's structural function. Propagation Review Root cause: NO FAILURE NON CORE Sources checked: UniProtKB:Q9QY84 · mouse Actl7a SUPPORTS TRANSFER Golgi apparatus IDA PMID:21278383. Supporting Evidence: PMID:21278383 immunofluorescence analysis of testis sections reveals that Arp7A co-localizes with GM130 on the Golgi apparatus of round spermatids |
| GO:0032991 protein-containing complex | IDA PMID:21278383 Molecular recognition of the Tes LIM2-3 domains by the actin... | ACCEPT | Summary: Accepted: the complex is real and structurally defined, an ACTL7A-TES-ENAH heterotrimer, even though GO offers only the root complex term to say so. Reason: This is not a vague complex claim. ACTL7A, TES and ENAH co-immunoprecipitate from testis extract, all three are present in the acroplaxome, and the ternary complex of ACTL7A residues 1-65, TES LIM2-3 and the ENAH EVH1 domain was reconstituted from purified components and crystallised as PDB 2XQN. Retained at the root level because GO has no term for this complex; the informative statement is captured in core_functions, and a specific complex term would be a reasonable ontology request. Supporting Evidence: PMID:21278383 Here we show that Arp7A, which has 43% sequence identity with β-actin, forms a complex with the cytoskeletal proteins Tes and Mena in the subacrosomal layer of round spermatids. PMID:21278383 immunoblot analysis of Tes immunoprecipitations (IP) with the indicated antibodies reveals that Arp7A, Tes, and Mena are complexed to each other in rat testes extracts |
| GO:0005200 structural constituent of cytoskeleton | TAS PMID:10373328 Cloning, mapping, and expression of two novel actin genes, a... | REMOVE | Summary: A gene-name-derived annotation that GO's own phylogenetic pipeline has since explicitly negated for this clade, and that this review's structure audit independently argues against. The defensible level, GO:0005198, is already on the gene. Reason: Three independent arguments converge on removal. First, the cited evidence does not exist: PMID:10373328 identified ACTL7A and ACTL7B by cDNA selection and genomic sequencing, mapped them, and surveyed expression by Northern blot. It contains no biochemistry, no localisation and no phenotype; the only structural statement in it is the homology-based label actin-like. TAS is not an experimental code, so removing this row overrules no curator who read data - it retracts an inference drawn from a gene name. Second, GO's own phylogenetic pipeline has ruled against the term for this clade. The cached PAINT record for PTHR11937 shows GO:0005200 IBD-supported at the ancestral actin node PTN000940351 from conventional actins and Arp2/Arp3, then negated on descent: PTN008986528, the node carrying ACTL7A, has GO:0005200 as an IRD with negated set to true, and carries the parent GO:0005198 by IBA instead, both rows dated 2025-08-05. That is a deliberate, current, gene-covering rejection, and it postdates the whole knock-out and patient-variant literature, so it cannot be set aside as a judgement made without that evidence. Third, this review's own structure audit argues the same way: the route by which real actins earn this term is filament formation, and ACTL7A's inter-protomer interface is 42.3% identical to actin's against 93 to 100% for conventional actins, with the ATP-hydrolysis trigger His161 replaced by tyrosine. Nothing is lost by removal, because the defensible level of the claim, GO:0005198 structural molecule activity, is already annotated to the gene and is accepted there as the core molecular function. The independently reviewed paralogue ACTL7B removed the same row on the same reasoning, so the pair ends up consistent. Recorded for completeness: an earlier round of this review accepted this term and promoted it to the core molecular function, arguing that ACTL7A's residence in GO:0033011, a GO:0005856 descendant, made the specific term apply. That argument is withdrawn - residing in a cytoskeletal structure is not the same as being a structural constituent of it. One consequence of removal is worth flagging: GO_REF:0000108 uses this molecular-function term as the literal WITH/FROM for the GO:0007010 cytoskeleton organization row, so that row loses its basis too, which is part of why it is modified to the phenotype-supported GO:0030036 rather than left standing. Supporting Evidence: PMID:10373328 Two novel human actin-like genes, ACTL7A and ACTL7B, were identified by cDNA selection and direct genomic sequencing from the familial dysautonomia candidate region on 9q31. PMID:12672658 The two deduced amino acid sequences had 60% homology, and they had approximately 40% homology with other actins. file:human/ACTL7A/ACTL7A-bioinformatics/RESULTS.md * **The ATP-hydrolysis trigger is not retained.** Of the 5 literature-defined catalytic positions ACTL7A keeps 3 (DQEVY); changed: D154->E, H161->Y. Every conventional actin and Arp1/Arp2/Arp3 in the panel keeps all of them. file:interpro/panther/PTHR11937/PTHR11937-paint.tsv PTHR11937 PTN008986528 GO:0005200 F IRD true PANTHER:PTN000940351 file:interpro/panther/PTHR11937/PTHR11937-paint.tsv PTHR11937 PTN000940351 GO:0005200 F IBD false |
| GO:0005856 cytoskeleton | TAS PMID:10373328 Cloning, mapping, and expression of two novel actin genes, a... | KEEP AS NON CORE | Summary: Same name-derived source as the molecular-function row, but the compartment claim survives on later evidence because perinuclear theca is a cytoskeleton descendant. Kept as non-core. Reason: Like the paired GO:0005200 row, this comes from a cloning paper with no localisation data, so the cited evidence does not support it. Unlike that row, however, the statement is now redundant rather than merely unsupported: ACTL7A is annotated to GO:0033011 perinuclear theca, which is a descendant of GO:0005856, so the compartment claim is entailed by better-evidenced annotations. Kept for that reason and marked non-core, since the parent adds no resolution. Supporting Evidence: PMID:10373328 ACTL7A is expressed in a wide variety of adult tissues, while the ACTL7B message was detected only in the testis and, to a lesser extent, in the prostate. |
| GO:0007343 egg activation | IMP PMID:34727571 Novel bi-allelic variants in ACTL7A are associated with male... | NEW | Summary: Proposed new annotation. ACTL7A is an oocyte-activation-deficiency gene: patient variants deplete and mislocalise phospholipase C zeta so the oocyte never activates, and the defect is rescued by a calcium ionophore. GOA has nothing more specific than fertilization. Reason: GO:0007343 egg activation is a descendant of GO:0009566 fertilization and is exactly the step that fails when ACTL7A is lost. The human evidence is patient-variant based and mechanistically explicit: ACTL7A and phospholipase C zeta protein are both markedly reduced in the affected sperm, the phospholipase is mislocalised, calcium oscillations fail, and artificial oocyte activation by calcium ionophore rescues fertilization and yields a live birth. The same axis is reproduced in mouse knock-in and knock-out models and in three further independent patient reports. There is direct GO precedent for annotating a sperm-borne factor to this term on patient-variant evidence: human PLCZ1 carries GO:0007343 with an IMP. This is additive; the existing fertilization annotations remain correct. Supporting Evidence: PMID:34727571 Protein expression of ACTL7A and phospholipase C zeta, a key sperm-borne oocyte activation factor, was significantly reduced in the affected sperm compared to healthy controls, suggesting that the ACLT7A variants lead to an oocyte activation deficiency and TFF. PMID:34727571 AOA by calcium ionophore (A23187) after ICSI successfully rescued the TFF and achieved a live birth for the patient with ACTL7A variants. PMID:32923619 Furthermore, the sperm from ACTL7A/Actl7a-mutated men and mice showed reduced expression and abnormal localization of PLCζ as a potential cause of embryonic arrest and failure of fertilization. PMID:35863052 The mutant sperm failed to activate the oocyte, and sperm-borne oocyte activation factor phospholipase C zeta (PLCζ) discharge accompanied by ACTL7A was observed, leading to total fertilization failure (TFF). |
Loading supporting content…
Download this section (compressed HTML)Q: Does ACTL7A bind ATP or ADP? The nucleotide cleft is 63 percent identical to actin's, well above the divergent SWI/SNF Arps at 37 percent, and two of five variants reported in SPGF86 patients map into it while none of four population polymorphisms do, yet no binding measurement exists. A negative result would justify dropping the actin-derived framing altogether; a positive one would make ATP binding a legitimate molecular function.
Q: Is the substitution of actin's His161 by tyrosine in ACTL7A functionally meaningful? In actin, flipping of this histidine on filament incorporation triggers ATP hydrolysis. ACTL7B retains the histidine while ACTL7A and ACTL9 do not, which suggests a within-family functional split worth testing rather than neutral drift.
Q: How much of ACTL7A's annotation set should transfer to its head-to-head paralogue ACTL7B, and how much should not? The two are the obvious candidates for cross-transfer, and this review found they are separable on three independent axes. Residues: ACTL7B keeps the His161 hydrolysis trigger that ACTL7A has lost, and has a better-conserved nucleotide cleft (68.4 versus 63.2 percent) but a worse-conserved filament interface (36.4 versus 42.3 percent). Localisation: the study that first described both proteins put T-ACTIN-2 (ACTL7A) in the nucleus and in sperm heads and tails, and T-ACTIN-1 (ACTL7B) in the cytoplasm and not in sperm, although both were later found enriched in the acroplaxome. Expression: the original human survey detected ACTL7A message broadly but ACTL7B only in testis and prostate. Most consequentially for molecular function, ACTL7A's LIM-domain binding is carried by its N-terminal extension, and the equivalent extension in ACTL7B is reported to be distinct, so GO:0030274 should not be assumed to transfer between them. Mouse Actl7b is already a WITH/FROM donor on ACTL7A's cytoplasm IBA row, which is where the pair most needs watching.
Q: Should the acrosomal-vesicle annotations for ACTL7A be reassigned to the acroplaxome, and should GO gain a term for it? The only high-resolution study places ACTL7A in the subacrosomal layer and explicitly corrects earlier acrosome and nucleus assignments, but light-microscopy studies of patient sperm cannot distinguish the two compartments and GO has no term with which to state the difference.
Q: Should the ortholog-derived nucleus rows be refined to GO:0001673 male germ cell nucleus? Every observation of nuclear ACTL7A protein has been in a spermatocyte, spermatid or sperm, and the mouse ortholog carries that specific term by IDA from the same experiment the human ISS row transfers from. Two counter-considerations, neither decisive: the phylogenetic row's donor set includes fungal and protozoan Arps for which the germ-cell term is meaningless, so GO:0005634 is the right least common ancestor there; and no study has looked for ACTL7A protein in a somatic nucleus, so the germ-cell restriction is established by where the protein has been observed rather than by exclusion elsewhere. ACTL7A mRNA has been reported across many adult tissues, but a Northern survey does not speak to protein localisation and is not the reason for hesitating.
Q: What, if anything, does the intranuclear pool of ACTL7A do? It is reproducible across mouse germ-cell immunofluorescence, human sperm-nucleus proteomics and heterologous expression, where ACTL7A alone reorganises the nucleus, yet chromatin compaction is normal in the knock-out. Reported chromatin-remodeller links rest on a preprint and in-silico docking.
Q: Is the ACTL7A-TES-ENAH heterotrimer required in vivo? The complex is defined crystallographically and the interaction is abolished by single point substitutions, but no mouse carrying an interaction-null ACTL7A allele has been reported, so the complex's contribution to acroplaxome function is untested.
Q: Why is ACTL7A protein absent rather than merely dysfunctional in most patients? Truncating, missense and knock-in alleles alike lead to loss of ACTL7A from sperm, by degradation or by shedding in cytoplasmic droplets. Whether a quality-control pathway removes misfolded ACTL7A, and whether that is the proximate cause of phospholipase C zeta loss, is unknown.
Experiment: Express and purify recombinant full-length human ACTL7A and the actin-fold-only construct lacking residues 1-64. Measure nucleotide binding by isothermal titration calorimetry and with a fluorescent nucleotide analogue such as etheno-ATP, and measure steady-state and single-turnover phosphate release with a malachite-green or EnzChek assay, in parallel with beta-actin as a positive control and a cleft mutant of ACTL7A as a negative control. Include the patient variants G246A, G362R and D75A, which the structure-guided audit places in or adjacent to the cleft, and test whether substituting histidine for tyrosine at the position equivalent to actin His161 restores hydrolysis.
Hypothesis: ACTL7A binds a nucleotide but cannot hydrolyse it, because the cleft is conserved while the His161 hydrolysis trigger is not.
Type: biochemistry
Experiment: Perform pyrene-actin polymerisation assays with recombinant ACTL7A alone and titrated into actin, measuring nucleation lag, elongation rate and critical concentration; follow with total internal reflection fluorescence microscopy of labelled filaments to see whether ACTL7A incorporates, caps or bundles. Examine any polymer by negative-stain electron microscopy. Interpret against the finding that ACTL7A matches actin at only 42 percent of inter-protomer contact positions.
Hypothesis: ACTL7A does not homopolymerise but co-assembles with or caps conventional actin, explaining why subacrosomal F-actin is lost when it is absent.
Type: in vitro reconstitution
Experiment: Apply immunogold electron microscopy and expansion microscopy or cryo-electron tomography to wild-type round spermatids, using antibodies against ACTL7A, beta-actin and keratin 5, to determine the molecular composition of the subacrosomal filaments and the spatial relationship of ACTL7A to them. This directly addresses the 2011 observation that actin could not be detected in the subacrosomal layer of round spermatids while filaments were morphologically present.
Hypothesis: The acroplaxome filaments that disappear in Actl7a knock-out spermatids contain ACTL7A itself.
Type: imaging
Experiment: Generate a knock-in mouse carrying the alanine-to-tyrosine substitution at the ACTL7A residue equivalent to human Ala31 or Ala41, which abolishes TES binding in vitro without deleting the protein, and score fertility, acrosome attachment, subacrosomal F-actin by phalloidin, phospholipase C zeta localisation, and ACTL7A retention in epididymal sperm. This separates the LIM-binding activity from the rest of the protein.
Hypothesis: The ACTL7A-TES interaction is required for acroplaxome integrity and male fertility.
Type: mouse genetics
Experiment: Test for a direct interaction by reciprocal co-immunoprecipitation and proximity labelling of ACTL7A and PLCZ1 in spermatids and in a heterologous co-expression system, then compare phospholipase C zeta localisation and abundance across ACTL7A, ACTL9 and CYLC1 mutant sperm. A shared phospholipase phenotype across structurally distinct theca mutants would establish the indirect model and support annotating ACTL7A to egg activation through an architectural role rather than as a carrier.
Hypothesis: Phospholipase C zeta loss in ACTL7A mutants is an indirect consequence of perinuclear-theca failure rather than of a direct ACTL7A-phospholipase interaction.
Type: interaction proteomics
Loading supporting content…
Download this section (compressed HTML)Loading supporting content…
Download this section (compressed HTML)Loading supporting content…
Download this section (compressed HTML)Loading supporting content…
Download this section (compressed HTML)