Six-transmembrane protein of the endoplasmic reticulum membrane that forms a symmetric homodimer and binds a molecule of coenzyme A in a deep, cytosol-facing cavity within each protomer. Its fold resembles the very-long-chain fatty acid elongase ELOVL7, but the pocket cannot accommodate a long acyl chain and no elongase activity has been detected; the protein is a CoA-binding accessory factor of glycerolipid synthesis rather than an enzyme. TMEM120A associates with the ER acyl-CoA synthetases ACSL1 and ACSL3 and promotes their retention at the ER, supporting local long-chain acyl-CoA production and the re-esterification of fatty acids into triacylglycerol during lipolysis; it also binds the glycerol-3-phosphate acyltransferase GPAT4 and, together with myristoylated CHP1, synergistically stimulates GPAT4 activity, driving acyl-CoA into glycerolipid. Both activities require an intact CoA-binding site. The protein is enriched in adipose tissue, where its loss causes defective fatty acid re-esterification, ER stress and a lipodystrophy-like metabolic phenotype, and the role in glycerolipid synthesis is conserved in Caenorhabditis elegans. TMEM120A was originally described as a nuclear envelope protein (NET29) and later, under the name TACAN, as a plasma membrane mechanosensitive ion channel of nociceptors; the channel assignment is contradicted by four independent structural and electrophysiological studies that could not detect mechanically evoked currents, and the protein instead modulates mechanical sensitivity indirectly by raising cellular phosphatidic acid and lysophosphatidic acid, which inhibit PIEZO2. Additional reported roles include promotion of STING trafficking from the ER during the antiviral response and inhibition of the PKD2 channel.
Definition: Binding to and increasing the activity of a glycerol-3-phosphate O-acyltransferase, an enzyme that catalyses the transfer of an acyl group from acyl-CoA to sn-glycerol 3-phosphate to form lysophosphatidic acid.
Justification: TMEM120A acts on GPAT4 exactly as an enzyme activator: it binds the enzyme at the endoplasmic reticulum membrane and, synergistically with myristoylated CHP1, increases its acyltransferase activity in a reconstituted in vitro assay, without itself catalysing any reaction. GO has no term at this specificity - a QuickGO search returns no glycerol-3-phosphate O-acyltransferase activator activity - so the only available option is the generic GO:0008047 enzyme activator activity, which loses the identity of the regulated enzyme and therefore the biology. There is direct precedent for acyltransferase-specific activator children, notably GO:0060228 phosphatidylcholine-sterol O-acyltransferase activator activity. The term would also fit CHP1, the only previously known GPAT4 activator, so it is not a single-gene term.
Parent term: enzyme activator activity
Supporting Evidence:
| GO Term | Evidence | Action | Reason |
|---|---|---|---|
| GO:0005637 nuclear inner membrane | IBA GO_REF:0000033 | KEEP AS NON CORE | Summary: Phylogenetic propagation of the original NET29 nuclear-envelope assignment across the TMEM120 family. Reason: The inner nuclear membrane assignment rests on real experiments - nuclear envelope proteomics and imaging in adipocytes (PMID:26024229), plus a genome-organisation phenotype in adipocyte-specific knockout mice (PMID:35027552) - and the INM is continuous with the ER, so this is not simply wrong. But the compartment is explicitly contested: the two most recent mechanistic studies place the protein in the bulk ER, by fractionation and imaging in adipocytes (PMID:41423633) and by endogenous tagging in C. elegans plus TurboID proximity labelling in human cells (PMID:42098142). All functional partners now identified (ACSL1, ACSL3, GPAT4, CHP1, STING1) are ER proteins. Retained as a secondary/disputed location rather than a core one. Supporting Evidence: PMID:26024229 Here we present two nuclear envelope trans-membrane proteins TMEM120A and TMEM120B that are paralogs encoded by the Tmem120A and Tmem120B genes. PMID:42098142 Intriguingly, a consensus has yet to emerge regarding the subcellular localization of TMEM120A. |
| GO:0045444 fat cell differentiation | IBA GO_REF:0000033 | KEEP AS NON CORE | Summary: Phylogenetic propagation of the adipocyte-differentiation role documented for TMEM120A/B in mouse and human. Reason: Well supported by knockdown in 3T3-L1 preadipocytes (PMID:26024229) and by the lipodystrophy phenotype of adipocyte-specific knockout mice (PMID:35027552), but this is a tissue-level consequence of the protein's lipid-handling activity at the ER rather than the activity itself. The conserved role, present down to C. elegans which has no adipocytes, is glycerolipid synthesis (PMID:42098142). Kept as a genuine but non-core process. Supporting Evidence: PMID:26024229 Accordingly, TMEM120A and B knockdown individually and together impacted on adipocyte differentiation/metabolism as measured by lipid accumulation through binding of Oil Red O and coherent anti-Stokes Raman scattering microscopy (CARS). |
| GO:0005637 nuclear inner membrane | IEA GO_REF:0000120 | KEEP AS NON CORE | Summary: Electronic assignment of the inner nuclear membrane location, projected from mouse Q8C1E7 and the UniProt subcellular-location vocabulary. Reason: Same judgment as the IBA row for this term - a real but disputed secondary location, superseded as the primary compartment by the ER evidence in PMID:41423633 and PMID:42098142. |
| GO:0005783 endoplasmic reticulum | IEA GO_REF:0000044 | ACCEPT | Summary: Electronic mapping of the UniProt endoplasmic reticulum subcellular-location annotation. Reason: Correct, and on current evidence this is the primary compartment. Confirmed by subcellular fractionation and imaging in adipocytes, by endogenous tagging in C. elegans, and by the ER localisation of every functional partner identified to date. Supporting Evidence: PMID:41423633 Here, we identify TMEM120A as an ER-resident CoA-binding protein that regulates intracellular FA metabolism. PMID:42098142 Taken together, our results firmly suggest that TMEM-120 is an ER-resident protein, with its N- and C-termini both oriented towards the cytoplasm. |
| GO:0005886 plasma membrane | IEA GO_REF:0000120 | KEEP AS NON CORE | Summary: Electronic projection of the plasma membrane location from mouse Q8C1E7. Reason: The mouse source annotations for plasma membrane derive from the TACAN channel study (PMID:32084332) and from a PIEZO2-regulation study, both of which used heterologous overexpression. Surface localisation of an overexpressed multi-pass ER protein is a weak basis for a physiological site, and PMID:42098142 notes that the plasma membrane assignment was tied to the channel model. Retained, not removed, because the human IDA row for this term stands on its own, but demoted to non-core. Supporting Evidence: PMID:42098142 In contrast, the assignment of TMEM120A as a mechanosensory channel placed it at the plasma membrane when overexpressed17. |
| GO:0016020 membrane | IEA GO_REF:0000002 | MODIFY | Summary: InterPro signature IPR012926 maps to the generic membrane term. Reason: Correct but uninformative. TMEM120A is a polytopic protein of a specific membrane; the endoplasmic reticulum membrane is the supported and far more useful assignment. Proposed replacements: endoplasmic reticulum membrane |
| GO:0005515 protein binding | IPI PMID:32296183 A reference map of the human binary protein interactome. | REMOVE | Summary: Bare protein binding from a high-throughput binary interactome map (24 partners listed in GOA). Reason: Per project guidance, bare GO:0005515 carries no functional information. The interactions that matter mechanistically (ACSL1/ACSL3, GPAT4, CHP1) are not in this set, and the informative molecular function is captured by the coenzyme A binding and the proposed enzyme activator terms. No more informative molecular function is supportable from this evidence, so the row is removed rather than rewritten. Removal does not imply the reported interaction is false. |
| GO:0034220 monoatomic ion transmembrane transport | IEA GO_REF:0000107 | REMOVE | Summary: Ensembl Compara projection from mouse Q8C1E7 of an ion-transport process that follows only from the refuted channel model. Reason: The mouse source annotation for this term is from PMID:32084332, the original TACAN paper. Ion transmembrane transport is a direct downstream consequence of channel activity, and GO already carries three human experimental NOT|enables annotations against GO:0005216 monoatomic ion channel activity for this protein (IMP PMID:34374645, IDA PMID:34409941, IDA PMID:34465718). Four independent groups failed to detect mechanically evoked or constitutive currents in cells and in reconstituted membranes. A process annotation that presupposes the negated molecular function cannot stand. Supporting Evidence: PMID:34409941 Thus, under our experimental conditions with the use of Piezo channels and HsTMEM63a as proper positive controls, we conclude that TMEM120A is not sufficient to mediate poking- or stretch-induced currents in P1-KO-HEK cells. PMID:34374644 Whilst its physiological function remains unclear, we anticipate that TACAN is not a mechanosensitive ion channel. |
| GO:0045444 fat cell differentiation | IEA GO_REF:0000107 | KEEP AS NON CORE | Summary: Ensembl Compara projection of the adipocyte-differentiation role from mouse. Reason: Consistent with the human IMP and IBA rows for this term; a real but downstream, tissue-restricted consequence of the protein's ER lipid-handling activity. |
| GO:0050966 detection of mechanical stimulus involved in sensory perception of pain | IEA GO_REF:0000107 | MODIFY | Summary: Ensembl Compara projection from mouse of the nociceptor mechanical-pain phenotype, framed as stimulus detection. Reason: The organismal phenotype behind the mouse annotation is real - nociceptor-specific inducible knockout reduced behavioural responses to painful mechanical stimuli (PMID:32084332) - and independent rodent work supports a contribution to mechanical hyperalgesia. What has been refuted is the mechanism: this term asserts that the gene product receives and transduces the mechanical stimulus, which is exactly the channel claim negated three times in GOA. The current model is indirect modulation: TMEM120A inhibits PIEZO2 (PMID:35819364), and the mediator is a glycerolipid, since TMEM120A expression raises phosphatidic acid and lysophosphatidic acid, which inhibit PIEZO2 (PMID:39147733) - and lysophosphatidic acid is the product of the GPAT4 reaction that TMEM120A activates (PMID:42098142). A regulation term is the defensible altitude. Flagged honestly: the sign of the effect is unresolved, since knockout reduces mechanical pain behaviour whereas knockdown increases PIEZO2 current amplitudes. Proposed replacements: regulation of sensory perception of pain Supporting Evidence: PMID:35819364 Our data identify TMEM120A as a negative modulator of PIEZO2 channel activity, and do not support TMEM120A being a mechanically activated ion channel. PMID:39147733 Here we find that TMEM120A expression elevates cellular levels of phosphatidic acid and lysophosphatidic acid (LPA), aligning with its structural resemblance to lipid-modifying enzymes. |
| GO:0051260 protein homooligomerization | IEA GO_REF:0000107 | ACCEPT | Summary: Ensembl Compara projection of the homodimerisation property from mouse. Reason: Uncontested. Every independent cryo-EM structure of TMEM120A resolves a symmetric homodimer, so this is a well-established architectural property. |
| GO:0051291 protein heterooligomerization | IEA GO_REF:0000107 | KEEP AS NON CORE | Summary: Ensembl Compara projection of the reported TMEM120A-TMEM120B heterooligomer from mouse. Reason: The TMEM120A/TMEM120B heterooligomer rests on co-immunoprecipitation in PMID:26024229; all subsequent structural work has resolved only homodimers of each paralog, and the TurboID comparison in PMID:42098142 treats them as parallel rather than obligately partnered proteins. Retained but not treated as a core property. |
| GO:0120225 coenzyme A binding | IEA GO_REF:0000107 | ACCEPT | Summary: Ensembl Compara projection of coenzyme A binding from mouse Q8C1E7. Reason: Agrees with the human IDA row and with four independent structural studies. This is the best-established molecular function of the protein. |
| GO:0140374 antiviral innate immune response | IEA GO_REF:0000107 | KEEP AS NON CORE | Summary: Ensembl Compara projection of the STING-dependent antiviral role from mouse. Reason: Consistent with the human IMP row for this term; a plausible but single-laboratory role that is peripheral to the conserved lipid-metabolic function. |
| GO:0005637 nuclear inner membrane | ISS GO_REF:0000024 | KEEP AS NON CORE | Summary: Curator sequence-similarity transfer of the inner nuclear membrane location from mouse Q8C1E7. Reason: Same judgment as the other rows for this term - a real but disputed secondary location. Unlike the channel-derived rows in the same GO_REF:0000024 block, the nuclear envelope assignment has independent experimental support in both mouse and human and is not simply a projection of the TACAN model. |
| GO:0120225 coenzyme A binding | IDA PMID:34374645 TMEM120A is a coenzyme A-binding membrane protein with struc... | ACCEPT | Summary: Direct demonstration that purified TMEM120A carries a bound coenzyme A molecule in a deep transmembrane pocket. Reason: Core molecular function, established independently four times over: cryo-EM density modelled as CoA plus a fluorometric CoA assay on purified protein (PMID:34374645), a defined CoASH site whose central tryptophan is required for binding (PMID:34409941), a further cryo-EM structure (PMID:34465718), and native mass spectrometry (PMID:34374644). The site is functionally load-bearing rather than incidental: the W193A mutation abolishes the interaction with ACSL1 and ACSL3 (PMID:41423633), and mutating the equivalent linker residue in C. elegans attenuates function (PMID:42098142). Kept at this altitude - the protein binds CoA but, lacking a hydrophobic tunnel for an acyl chain, does not catalyse elongation, so no catalytic term is warranted. Supporting Evidence: PMID:34374645 Instead, the six TMs form an Ξ±-barrel with a deep pocket where a coenzyme A (CoA) molecule is bound. PMID:34409941 Mutation of a central tryptophan residue involved in binding CoASH dramatically reduced the binding affinity of HsTMEM120A with CoASH. PMID:41423633 As previously noted20, although TMEM120A contains a CoA-binding site, it lacks a defined hydrophobic pocket capable of accommodating long-chain fatty acyl chain. |
| GO:0005216 monoatomic ion channel activity | IMP NOT PMID:34374645 TMEM120A is a coenzyme A-binding membrane protein with struc... | ACCEPT | Summary: Negative assertion - TMEM120A does NOT enable ion channel activity, from failure to reproduce mechanosensitive currents in cells and in reconstituted liposomes. Reason: The negated annotation is correct and should stand. This row is one of three independent experimental negatives GO carries for this term on this protein, and GO has already sided against the channel model on their basis. Supporting Evidence: PMID:34374645 However, we were unable to reproduce the mechanosensitive activity of TMEM120A expressed in HEK293 or CHO cells, nor did we observe any mechanosensitive channel activity in giant liposome patching using TMEM120A protein reconstituted into lipid vesicles. |
| GO:0005216 monoatomic ion channel activity | IDA NOT PMID:34409941 TMEM120A contains a specific coenzyme A-binding site and mig... | ACCEPT | Summary: Negative assertion - TMEM120A is not sufficient to mediate poking- or stretch-induced currents, assayed with Piezo1 and TMEM63A positive controls in Piezo1-knockout HEK cells. Reason: The negated annotation is correct and should stand. The controls make this a strong negative rather than a failure of assay sensitivity, and the same study identified the CoA site that replaces the channel interpretation. Supporting Evidence: PMID:34409941 Here we report that expression of TMEM120A is not sufficient in mediating poking- or stretch-induced currents in cells and have solved cryo-electron microscopy (cryo-EM) structures of human TMEM120A (HsTMEM120A) in complex with an endogenous metabolic cofactor (coenzyme A, CoASH) and in the apo form. |
| GO:0005216 monoatomic ion channel activity | IDA NOT PMID:34465718 Cryo-EM structures of human TMEM120A and TMEM120B. | ACCEPT | Summary: Negative assertion - cryo-EM structures reveal no ion-conduction pathway and the authors decline to call TMEM120A a channel. Reason: The negated annotation is correct and should stand. This study is notable for making the negative carefully - it observed some conductance in a bilayer system yet still concluded that this does not establish channel identity, since non-channel proteins also conduct in that assay. Supporting Evidence: PMID:34465718 Although TMEM120A shows the ability to permeate ions as measured in bilayer system, our results indicate TMEM120A does not respond to poking or stretch mechanical stimuli in heterologous expression system. PMID:34465718 We even cannot conclude TMEM120A is an ion channel as conducting currents were also measured for several non-channel proteins in the bilayer system8. |
| GO:0005515 protein binding | IPI PMID:35013224 Gain-of-function genetic screening identifies the antiviral ... | REMOVE | Summary: Bare protein binding capturing the TMEM120A-STING1 (Q86WV6) interaction. Reason: Per project guidance, bare GO:0005515 is uninformative. The underlying interaction is real and its functional consequence is already captured by the antiviral innate immune response annotation from the same paper. No more informative molecular function is supportable from this evidence, so the row is removed rather than rewritten. |
| GO:0005783 endoplasmic reticulum | IDA PMID:35013224 Gain-of-function genetic screening identifies the antiviral ... | ACCEPT | Summary: Direct observation of TMEM120A in the endoplasmic reticulum. Reason: Correct and now the best-supported compartment, independently confirmed by adipocyte fractionation and imaging (PMID:41423633) and by endogenous tagging and proximity labelling (PMID:42098142). Supporting Evidence: PMID:35013224 Mechanistically, the antiviral activity of TMEM120A is dependent on STING, as TMEM120A interacts with STING, promotes the translocation of STING from the endoplasmic reticulum (ER) to ER-Golgi intermediate compartment (ERGIC) and enhances the phosphorylation of downstream TBK1 and IRF3, resulting in the expression of multiple antiviral cytokines and interferon-stimulated genes. |
| GO:0005886 plasma membrane | IDA PMID:35013224 Gain-of-function genetic screening identifies the antiviral ... | KEEP AS NON CORE | Summary: Direct localisation assay in which a fraction of TMEM120A signal was assigned to the plasma membrane. Reason: Not removed - this is an experimental annotation made by a curator with access to the full paper, and CLAUDE.md is explicit that such rows are not to be overturned from a cached abstract. But no function has been localised to the plasma membrane now that the channel model is negated, and all identified partners are ER proteins, so the site is recorded as secondary rather than core. |
| GO:0140374 antiviral innate immune response | IMP PMID:35013224 Gain-of-function genetic screening identifies the antiviral ... | KEEP AS NON CORE | Summary: Gain- and loss-of-function evidence that TMEM120A restricts Zika virus replication through STING. Reason: Genuine experimental support, including a mouse knockout, and mechanistically coherent for an ER membrane protein: the readout is ER-to-ERGIC export of STING1, a process that happens in the compartment where TMEM120A resides. Kept as non-core because it rests on a single laboratory, is not conserved outside vertebrates, and is unrelated to the lipid-metabolic activity shared from worm to human. Supporting Evidence: PMID:35013224 TMEM120A overexpression significantly inhibits ZIKV replication, while TMEM120A knockdown increases ZIKV infection in cell lines. |
| GO:0005515 protein binding | IPI PMID:36420836 Regulation of PKD2 channel function by TACAN. | REMOVE | Summary: Bare protein binding capturing the TMEM120A-PKD2 (Q13563) interaction. Reason: Per project guidance, bare GO:0005515 is uninformative. The cached record for this paper is abstract-only, so the interaction is not further adjudicated here; UniProt records that TMEM120A inhibits PKD2 channel activity through this physical association, which would be better expressed as a channel-regulator term if the evidence is confirmed. No more informative molecular function is supportable from this evidence, so the row is removed rather than rewritten. Removal does not imply the reported interaction is false. |
| GO:0005216 monoatomic ion channel activity | ISS GO_REF:0000024 | REMOVE | Summary: Positive orthology transfer of ion channel activity from mouse Q8C1E7, standing directly against three experimental negatives on the same term in the same protein. Reason: This is the single annotation in GOA still asserting that TMEM120A is an ion channel, and it should go. Provenance: the donor is mouse Q8C1E7, whose enables GO:0005216 annotation derives from PMID:32084332, the original TACAN paper; the transfer is dated 2020-03-11, three weeks after that paper and more than a year before the refutations. Since then four independent groups have failed to detect mechanically evoked or constitutive currents in cells, in reconstituted liposomes and in bilayers (PMID:34374645, PMID:34409941, PMID:34465718, PMID:34374644), and GO itself has recorded three human NOT|enables annotations against this exact term. Removing it is not overruling an experimentalist - it is retiring a curator-judgment sequence-similarity projection whose donor evidence has been superseded on the human protein itself. The negated rows for this term are accepted rather than removed, because a negated and a non-negated annotation of one term are opposite assertions and cannot take the same action. Supporting Evidence: PMID:34374644 Using cellular patch-recording methods, we failed to identify mechanosensitive ion channel activity. PMID:41423633 Our findings clarify this ambiguity by demonstrating that in adipocytes, TMEM120A is localized to the ER rather than the nuclear envelope, where it functions not as an ion channel or fatty acid elongase, but as a CoA-binding protein that facilitates lipid metabolic flux. |
| GO:0005886 plasma membrane | ISS GO_REF:0000024 | KEEP AS NON CORE | Summary: Curator sequence-similarity transfer of the plasma membrane location from mouse Q8C1E7, part of the same 2020 block that projected the channel model. Reason: Same judgment as the other plasma membrane rows. The donor annotation comes from heterologous overexpression in the TACAN study, so this is weak evidence for a physiological site, but the human IDA row independently supports some surface pool and the location is therefore demoted rather than deleted. |
| GO:0034220 monoatomic ion transmembrane transport | ISS GO_REF:0000024 | REMOVE | Summary: Curator sequence-similarity transfer of ion transmembrane transport from mouse Q8C1E7, derived from the refuted channel model. Reason: Removed for the same reason as the GO_REF:0000107 row for this term. The process presupposes the ion channel activity that GO negates three times over on this protein, and its donor evidence (PMID:32084332) is the claim that four independent groups could not reproduce. Supporting Evidence: PMID:34374645 Thus, we were unable to detect any mechanosensitive channel activity of TMEM120A in our electrophysiological assays. |
| GO:0050966 detection of mechanical stimulus involved in sensory perception of pain | ISS GO_REF:0000024 | MODIFY | Summary: Curator sequence-similarity transfer of the mouse nociceptor mechanical-pain phenotype, framed as stimulus detection. Reason: Same judgment as the GO_REF:0000107 row for this term. The rodent behavioural phenotype is not disputed, but the detection framing encodes the negated channel mechanism; the evidence now supports indirect modulation of mechanical sensitivity through lipid signalling onto PIEZO2, so a regulation term is the defensible altitude. Proposed replacements: regulation of sensory perception of pain |
| GO:0045444 fat cell differentiation | IMP PMID:26024229 TMEM120A and B: Nuclear Envelope Transmembrane Proteins Impo... | KEEP AS NON CORE | Summary: Knockdown of TMEM120A and TMEM120B in 3T3-L1 cells altered adipogenic gene expression and lipid accumulation. Reason: Solid experimental evidence, reinforced by the lipodystrophy phenotype of adipocyte-specific Tmem120a knockout mice (PMID:35027552) and by the adipose phenotype in PMID:41423633. Treated as non-core because it is a tissue-level outcome of the protein's conserved ER glycerolipid-synthesis activity rather than the activity itself. Supporting Evidence: PMID:26024229 Knockdown of one or the other protein altered expression of several genes required for adipocyte differentiation, Gata3, Fasn, Glut4, while knockdown of both together additionally affected Pparg and Adipoq. PMID:35027552 Here we report that adipocyte-specific knockout of the gene encoding nuclear envelope transmembrane protein Tmem120a disrupts fat genome organisation, thus causing a lipodystrophy syndrome. |
| GO:0051260 protein homooligomerization | IDA PMID:26024229 TMEM120A and B: Nuclear Envelope Transmembrane Proteins Impo... | ACCEPT | Summary: Direct evidence that TMEM120A self-associates. Reason: Confirmed at atomic resolution by every subsequent structure - TMEM120A is an obligate symmetric homodimer, with each protomer contributing a CoA site. This is a core architectural property rather than an incidental interaction. Supporting Evidence: PMID:34374644 TACAN is an Ξ±-helical TM protein that forms a symmetric dimer (Figure 3A). PMID:34374645 A bound CoA ligand was later identified in the pocket (Figure 2f and g).TMEM120A forms a tightly packed dimer with extensive dimerization interactions involving multiple parts of the protein (Figure 3a). |
| GO:0051291 protein heterooligomerization | IDA PMID:26024229 TMEM120A and B: Nuclear Envelope Transmembrane Proteins Impo... | KEEP AS NON CORE | Summary: Direct evidence that TMEM120A and its paralog TMEM120B co-associate. Reason: The heterooligomer is reported only in this study; all later structural work resolves homodimers of TMEM120A and of TMEM120B separately, and the proximity-labelling comparison in PMID:42098142 treats the paralogs as parallel factors with overlapping interactomes rather than obligate partners. Retained as a plausible but non-core property. |
| GO:0008047 enzyme activator activity | IDA PMID:42098142 A conserved ER protein prevents lipotoxicity by stimulating ... | NEW | Summary: Proposed new annotation - TMEM120A binds GPAT4 and, with myristoylated CHP1, synergistically increases its acyltransferase activity in a reconstituted in vitro assay. Reason: GOA currently has no molecular function for TMEM120A beyond coenzyme A binding and bare protein binding, so nothing records what the protein does. The 2026 work supplies it: an in vitro acyltransferase assay with purified components shows that TMEM120A raises GPAT4 turnover, and TMEM120A itself catalyses nothing. GO:0008047 is the closest existing term; a glycerol-3-phosphate O-acyltransferase-specific child is requested in proposed_new_terms. Supporting Evidence: PMID:42098142 Here, we have used genetic, biochemical, and imaging techniques to identify TMEM120A as GPAT4-activating protein. PMID:42098142 We reproduced the observation that CHP1 enhanced GPAT4 activity23, and so did TMEM120A (Fig. |
| GO:0045017 glycerolipid biosynthetic process | IMP PMID:42098142 A conserved ER protein prevents lipotoxicity by stimulating ... | NEW | Summary: Proposed new annotation - TMEM120A promotes the incorporation of acyl-CoA into glycerolipid, a role conserved from Caenorhabditis elegans to mammalian cells. Reason: This is the conserved biological process the protein serves and it is currently absent from GOA, which carries only the downstream tissue-level fat cell differentiation term. Supported by loss-of-function in worm and by gain- and loss-of-function in mammalian cells. Supporting Evidence: PMID:42098142 We show that ER-localized TMEM120A and CHP1 synergistically activate GPAT4 and promote the incorporation of medium and long chain acyl-CoA into glycerolipid. PMID:42098142 Our results indicate that TMEM-120 deficiency causes a broad impairment in the incorporation of excess fatty acids into glycerolipid. |
| GO:0019432 triglyceride biosynthetic process | IMP PMID:41423633 TMEM120A maintains adipose tissue lipid homeostasis through ... | NEW | Summary: Proposed new annotation - TMEM120A promotes re-esterification of fatty acids into triacylglycerol during lipolysis, traced with labelled palmitate. Reason: Absent from GOA. Overexpression increases and knockdown decreases 13C-palmitate incorporation into triacylglycerol species in adipocytes, and the adipocyte-specific knockout mouse shows the corresponding metabolic phenotype, so this is a directly measured process rather than an inference from the interaction data. Supporting Evidence: PMID:41423633 At the basal level, TMEM120A overexpression enhanced the 13C-palmitate incorporation into TGs including 48:0, 48:1, 50:0, and 52:0, while Tmem120a KD reduced 13C-palmitate re-esterification into TGs (Fig. PMID:41423633 TMEM120A interacts with the ER-localized acyl-CoA synthetase ACSL1 and ACSL3 to promote long-chain acyl-CoA synthesis and channeling into the ER, thereby facilitating FA re-esterification and lipid cycling during lipolysis. |
Loading supporting contentβ¦
Download this section (compressed HTML)Q: Where does endogenous, untagged TMEM120A actually reside in a human cell - the bulk endoplasmic reticulum, the inner nuclear membrane, or both in a cell-type-dependent way? The two accounts are supported by different laboratories using different systems, and the answer determines whether the genome-organisation phenotype of the knockout mouse is a direct nuclear function or a downstream consequence of disturbed lipid flux.
Suggested experts: Eric C Schirmer, Yun-Hee Lee, Ho Yi Mak
Q: What is the sign of TMEM120A's effect on mechanical nociception, and can it be explained entirely by glycerolipid signalling? Nociceptor-specific knockout reduces behavioural responses to painful mechanical stimuli, whereas knockdown increases PIEZO2 current amplitudes and lowers their thresholds. Both cannot be complete accounts, and neither requires TMEM120A to be a channel.
Suggested experts: Tibor Rohacs, Reza Sharif-Naeini
Q: Does TMEM120A move coenzyme A across the endoplasmic reticulum membrane, or does it only present bound coenzyme A to partner enzymes? The phrase "CoA channeling" has been used for the adipocyte phenotype, but the authors hedge the mechanism, and no transport assay has been reported. A reconstituted flux measurement would decide whether a transporter molecular function term is ever warranted.
Suggested experts: Yun-Hee Lee
Q: Is the activation of GPAT4 by TMEM120A direct and stoichiometric, and does bound coenzyme A participate in it? The W193A and G195E mutations disrupt both CoA binding and partner interactions, leaving open whether CoA is a cofactor for the activation step or merely a structural requirement for the fold.
Suggested experts: Ho Yi Mak
Q: Do TMEM120A and TMEM120B form a genuine heterodimer in cells, and if so does it have different activity toward GPAT4 or the ACSLs than either homodimer? All structures to date resolve homodimers only.
Suggested experts: Eric C Schirmer
Experiment: Reconstitute purified human GPAT4, CHP1 and wild-type or W193A TMEM120A into nanodiscs of defined lipid composition and measure glycerol-3-phosphate acyltransferase kinetics (Km and kcat for acyl-CoA and for sn-glycerol 3-phosphate) with and without each partner. Deplete or add back free coenzyme A to test whether the bound cofactor is required for activation as opposed to for folding.
Hypothesis: TMEM120A increases GPAT4 catalytic turnover by direct binding, and the effect requires an intact coenzyme A site.
Type: reconstituted enzyme kinetics
Experiment: In nociceptor-specific Tmem120a knockout mice, quantify phosphatidic acid and lysophosphatidic acid in dorsal root ganglia by lipidomics, record PIEZO2-mediated and high-threshold mechanically activated currents, and test whether intracellular delivery of phosphatidic acid or the non-hydrolysable analogue restores wild-type mechanical thresholds. A GPAT4 loss-of-function allele should phenocopy the TMEM120A knockout if the lipid route is the operative one.
Hypothesis: The mechanical-pain phenotype of TMEM120A loss is mediated by reduced phosphatidic acid and lysophosphatidic acid rather than by loss of a channel.
Type: in vivo lipidomics plus patch-clamp rescue
Experiment: Knock a small epitope tag into the endogenous TMEM120A locus in a human adipocyte model and in a sensory-neuron model, then compare localisation by Airyscan or expansion microscopy against calnexin, lamin B1 and an inner nuclear membrane marker, with quantitative subcellular fractionation and split-GFP topology reporters as orthogonal readouts. Avoid overexpression throughout, since it is the principal confound in the existing literature.
Hypothesis: Endogenous TMEM120A is an endoplasmic reticulum protein in most cell types, with any inner nuclear membrane pool being a minor subfraction.
Type: endogenous tagging and quantitative imaging
Experiment: Perform a blinded, multi-site electrophysiological comparison of TMEM120A against a positive control (PIEZO1 or TMEM63A) and an empty vector, in Piezo1-knockout HEK cells, in native dorsal root ganglion neurons, and in proteoliposomes, using a common poking and stretch protocol agreed in advance. This directly addresses the residual disagreement rather than adding another single-laboratory report.
Hypothesis: No ion flux is attributable to TMEM120A under conditions where a genuine mechanosensitive channel would be detected.
Type: blinded multi-laboratory electrophysiology
Loading supporting contentβ¦
Download this section (compressed HTML)Loading supporting contentβ¦
Download this section (compressed HTML)