APOOL, also called MIC27, is a subunit of the MICOS complex (mitochondrial contact site and cristae organizing system) of the mitochondrial inner membrane. It is synthesised with an N-terminal mitochondrial targeting sequence, inserts into the inner membrane through two short transmembrane helices connected by a strongly basic loop, and presents both its N-terminal and its long, intrinsically disordered C-terminal region to the intermembrane space. Which face of the membrane the basic loop itself contacts is not settled: the bulk of the protein was shown by protease protection to face the intermembrane space, while the short loop between the two helices is only predicted, as matrix-facing. The gene is named for a superficial resemblance to the plasma apolipoproteins, but the protein is not secreted and is not part of any lipoprotein particle: endogenous APOOL is absent from the cytosol, resists alkaline carbonate extraction, and has been shown in four human cell lines to exist only as the ~30 kDa mitochondrial form, with no glycosylated or secreted species. Within MICOS, APOOL belongs to the membrane-shaping MIC10 module together with MIC10, MIC26 and MIC13, rather than to the MIC60-MIC19-MIC25 module; its incorporation depends on MIC13, whose loss causes APOOL to be degraded along with MIC10 and MIC26. Because MICOS associates across the intermembrane space with the outer-membrane SAMM50-metaxin machinery, APOOL is also recovered in the larger MIB (mitochondrial intermembrane space bridging) assembly, and MICOS itself is concentrated at crista junctions, the narrow tubular necks that connect cristae to the inner boundary membrane. The one molecular activity demonstrated for the protein is lipid binding: recombinant APOOL binds cardiolipin specifically and does not bind its biosynthetic precursor phosphatidylglycerol or any other common membrane phospholipid. The basic inter-transmembrane loop that is the proposed lipid contact is conserved across the whole MIC26/MIC27 family. Functionally, loss of APOOL reduces the number of crista junctions and cristae per mitochondrial section and impairs respiration, and simultaneous loss of APOOL and its paralogue APOO/MIC26 produces the full MICOS phenotype of concentric, onion-like cristae together with reduced cardiolipin and destabilised respiratory chain supercomplexes and ATP synthase. APOOL and APOO regulate each other's protein levels reciprocally and post-transcriptionally, so each partly masks the other's loss. The pair arose late in evolution and is restricted to opisthokonts; Drosophila, for example, carries a single combined MIC26/MIC27 protein, CG5903.
| GO Term | Evidence | Action | Reason |
|---|---|---|---|
| GO:0001401 SAM complex | HDA PMID:26477565 Evolution and structural organization of the mitochondrial c... | REMOVE | Summary: APOOL is not a subunit of the outer-membrane SAM complex; the co-migration behind this high-throughput row is the MICOS-SAM bridge, which the same reference already records correctly as MIB complex membership. Reason: GO:0001401 is defined as a complex of the mitochondrial OUTER membrane that sorts and assembles imported outer-membrane proteins. APOOL is an inner-membrane protein. It resists alkaline carbonate extraction, and its protease-accessibility profile matches the intermembrane-space marker Mitofilin, not the outer-membrane marker TOM20. A protein integrated in the inner membrane cannot be part_of an outer-membrane complex, so this is a compartment and complex mismatch rather than a matter of degree. The cited reference does not support the row either. Its own resolution of the complexome profile places MIC27 in form (A), the MICOS complex, with SAMM50 and MTX2 appearing only when the membrane-bridging subcomplex is formed and DNAJC11 and MTX1 only in the complete MIB complex. The real relationship the profile shows, that APOOL co-migrates with SAMM50-containing assemblies because MICOS is bridged to the outer membrane, is already captured from this same reference by GO:0140275 MIB complex and GO:0061617 MICOS complex, both of which are accepted here. Removing this row therefore loses no information and removes a false compartment assignment. Supporting Evidence: PMID:26477565 We find three main forms of the complex: A) The MICOS complex, containing all the MICOS proteins, B) a membrane bridging subcomplex, containing in addition SAMM50, MTX2 and the previously uncharacterized MTX3, and C) the complete MIB complex containing in addition DNAJC11 and MTX1. PMID:23704930 APOOL was found to remain fully in the membrane fraction whereas the chaperone mtHsp60 was released efficiently, yet not completely, to the soluble fraction PMID:23704930 Taken together, we conclude that APOOL is located in the intermembrane space consistent with the proposed location of the Mitofilin/MINOS complex. |
| GO:0005515 protein binding | IPI PMID:25764979 The non-glycosylated isoform of MIC26 is a constituent of th... | REMOVE | Summary: A real interaction with MIC60/Mitofilin, but recorded as bare protein binding, which conveys no molecular function and is already implied by the accepted MICOS complex membership. Reason: The partner resolves against UniProt to Q16891, IMMT/MIC60 (MIC60_HUMAN), the central MICOS subunit, and the interaction is independently supported by reciprocal co-immunoprecipitation in PMID:23704930. Removing this row does not mean the reported interaction is false; it means the annotation carries no functional information. GO:0005515 states only that two proteins touch, and the biologically meaningful content, that both are subunits of one complex, is already asserted by GO:0061617 part_of MICOS complex, which this review accepts on six rows drawn from five distinct references. MODIFY was considered and rejected because the cited paper supports no more informative molecular function. The obvious candidate, an oligomer-stabilising or scaffolding activity, comes from yeast (PMID:26968360, PMID:29733859) and is contradicted for human MIC27 by the CRISPR knockouts in PMID:32788226, which show MIC26 and MIC27 are dispensable for the stability and integration of the remaining MICOS subunits; inventing a specific function from interaction evidence alone is not an option here. Supporting Evidence: PMID:25764979 The latter isoform spans the mitochondrial inner membrane and physically interacts with several MICOS complex subunits such as MIC60, MIC27, and MIC10. PMID:32788226 MIC26 and MIC27 are dispensable for the stability and integration of the remaining MICOS subunits into the complex suggesting that they assemble late into the MICOS complex. |
| GO:0005515 protein binding | IPI PMID:25764979 The non-glycosylated isoform of MIC26 is a constituent of th... | REMOVE | Summary: A real interaction with MIC10, the membrane-shaping core of the MICOS module APOOL belongs to, but again recorded only as bare protein binding. Reason: The partner resolves against UniProt to Q5TGZ0, MICOS10/MIC10 (MIC10_HUMAN), and the interaction is corroborated by the reciprocal co-immunoprecipitation of APOOL with MINOS1 in PMID:23704930 and by the co-degradation of MIC10, MIC26 and MIC27 when MIC13 is lost (PMID:25997101, PMID:27479602). Removal is not a claim that the interaction is false - it is the best corroborated of the three - but that the generic term is uninformative. This is also the most functionally interesting of the three partners, because the yeast literature assigns Mic27 a role in stabilising Mic10 oligomers, which would be the MODIFY target. That role has not been demonstrated for the human protein, and the human knockout data run the other way, so promoting this row to an oligomer-stabilisation molecular function would be stating a yeast mechanism as human fact. With no informative human molecular function supported by the cited paper, the generic row goes; the real content, shared membership of the MICOS complex, is already recorded by GO:0061617. Supporting Evidence: PMID:25764979 The latter isoform spans the mitochondrial inner membrane and physically interacts with several MICOS complex subunits such as MIC60, MIC27, and MIC10. PMID:25997101 Using quantitative proteomics, we show that loss of QIL1 resulted in MICOS disassembly with the accumulation of a MIC60-MIC19-MIC25 sub-complex and degradation of MIC10, MIC26, and MIC27. |
| GO:0005515 protein binding | IPI PMID:25764979 The non-glycosylated isoform of MIC26 is a constituent of th... | REMOVE | Summary: A real interaction with the paralogue MIC26/APOO, recorded as bare protein binding; the biologically important relationship between these two proteins is reciprocal regulation of abundance, which GO:0005515 does not express. Reason: The partner resolves against UniProt to Q9BUR5, APOO/MIC26 (MIC26_HUMAN), the other human member of PTHR14564. The interaction is real and both proteins sit in the same MICOS module, and removing the row disputes neither. But the substantive finding of this very paper about the MIC26-MIC27 relationship is not binding. It is that the two regulate each other's levels antagonistically, which is why every single-gene depletion before the double knockout was hard to interpret. GO:0005515 captures none of that, and GO has no molecular function term for reciprocal control of a paralogue's abundance, so there is nothing to MODIFY to. Shared complex membership is already recorded by GO:0061617, so the generic row is redundant as well as uninformative. Supporting Evidence: PMID:25764979 We further demonstrate that MIC26 and MIC27, a homologous protein formerly termed APOOL, regulate their levels in an antagonistic manner. PMID:32788226 MIC26 and MIC27 are reciprocally regulated at the posttranscriptional level |
| GO:0005576 extracellular region | TAS Reactome:R-HSA-481007 | REMOVE | Summary: A name-era artefact. APOOL is not secreted; the row is a projection of a Reactome platelet pathway membership that itself rests on a 2004 bulk releasate proteomics screen. Reason: Three independent lines of positive evidence place this protein inside mitochondria and nowhere else. Endogenous APOOL is entirely absent from the cytosolic fraction and enriched in the mitochondrial fraction, and resists alkaline carbonate extraction (PMID:23704930). UniProt records only a mitochondrial TRANSIT peptide, not a signal peptide, and gives SUBCELLULAR LOCATION Mitochondrion inner membrane. Most directly, PMID:37279200 tested the secreted-isoform hypothesis for this family head-on in four human cell lines with four antibodies and tagged constructs and concluded that MIC26 and MIC27 are exclusively mitochondrial, explicitly excluding a high-molecular-weight glycosylated MIC27. The provenance of the row is equally clear. Reactome lists Q6UXV4 among the 27 participants of Platelet degranulation (R-HSA-114608), whose only PubMed-backed literature reference is a thrombin-releasate shotgun screen returning more than 300 proteins, published in 2004, nine years before the protein was shown to be mitochondrial. The reference projects two terms onto 67 gene products indiscriminately, and Reactome itself now separately places Q6UXV4 in the mitochondrial inner membrane under Cristae formation. Supporting Evidence: PMID:37279200 Taken together, we conclude that both MIC26 and MIC27 are exclusively localized in mitochondria and that the observed phenotypes reported previously are exclusively due to their mitochondrial function. PMID:23704930 Endogenous APOOL was found to be entirely absent in the cytosolic fraction and instead was detected in the mitochondrial fraction. PMID:14630798 Using a proteomics approach, we have identified more than 300 proteins released by human platelets following thrombin activation. |
| GO:0005739 mitochondrion | HTP PMID:34800366 Quantitative high-confidence human mitochondrial proteome an... | ACCEPT | Summary: Correct. APOOL is in the MitoCoP high-confidence human mitochondrial proteome; general relative to the inner-membrane term the gene also carries, but appropriate to the method. Reason: A high-throughput proteomic assignment, which is exactly the level of resolution a whole-cell mitochondrial proteome supports. The call is independently confirmed at higher resolution by antibody immunofluorescence and subcellular fractionation (PMID:23704930) and by the inner-membrane EXP rows on this gene, so there is no risk of a bulk-preparation artefact here. The reference is broad, returning 1235 annotations over 1234 distinct gene products in a paginated QuickGO query, all to GO:0005739, but breadth is the point of a proteome resource and does not make the individual assignment weak when the protein is independently established as mitochondrial. Kept rather than modified to the inner-membrane term, because GO:0005743 is already asserted four times on this gene from evidence that actually resolves the submitochondrial compartment. Supporting Evidence: PMID:34800366 We classified >8,000 proteins in mitochondrial preparations of human cells and defined a mitochondrial high-confidence proteome of >1,100 proteins (MitoCoP). PMID:23704930 Endogenous APOOL was found to be entirely absent in the cytosolic fraction and instead was detected in the mitochondrial fraction. |
| GO:0005739 mitochondrion | IDA GO_REF:0000052 | ACCEPT | Summary: Human Protein Atlas immunofluorescence; agrees with the antibody immunofluorescence and fractionation in the primary literature. Reason: Immunofluorescence curation is well suited to a compartment call at this granularity, and the independent antibody immunofluorescence in PMID:23704930 showed clear colocalisation of endogenous APOOL with a mitochondrial marker in HeLa cells. Nothing in the record conflicts. Supporting Evidence: PMID:23704930 We observed a clear colocalization of the APOOL corresponding signal with the mitochondrial marker mitoDsRed indicating that endogenous APOOL is a mitochondrial protein |
| GO:0005739 mitochondrion | IDA PMID:25781180 Detailed analysis of the human mitochondrial contact site co... | ACCEPT | Summary: Direct microscopy of FLAG-tagged Mic27/ApoOL colocalised with MitoTracker; correct. Reason: This study imaged FLAG-tagged MICOS and MIB components, including Mic27/ApoOL, against MitoTracker in HeLa cells. Tagged expression is a caveat for a fine-grained submitochondrial call but not for organelle assignment, and the same conclusion follows from endogenous-protein fractionation and immunofluorescence in PMID:23704930. Correct at the level asserted. Supporting Evidence: PMID:25781180 HeLa cells were grown on cover slips and transfected using Lipofectamine 2000 with pCDNA3 plasmids carrying information for FLAG-tagged Sam50 (A), Mic60/Mitofilin (B), Mic19/CHCHD3 (C), Mic25/CHCHD6 (D), Mic27/ApoOL (E) or Mic23/ApoO (F). |
| GO:0005739 mitochondrion | IDA PMID:25997101 QIL1 is a novel mitochondrial protein required for MICOS com... | ACCEPT | Summary: MICOS interaction proteomics from isolated human mitochondria; correct. Reason: MIC27 was recovered from digitonin-solubilised mitochondria of 293T and HCT116 cells in immunoprecipitations of MICOS components, alongside the other MICOS subunits and their outer-membrane interactors. Recovery from purified mitochondria is a direct mitochondrial assignment, consistent with every other line of evidence on this gene. Supporting Evidence: PMID:25997101 The MIC10-MIC19-MIC25-MIC26-MIC27-MIC60 complex is thought to reside in the IM at the site of CJs. PMID:25997101 While C-terminally tagged QIL1 efficiently co-purified with endogenous MIC60, MIC19, MIC27, and MIC10 when mitochondria were solubilized with 1% digitonin, no specific binding was detected when mitochondria were solubilized with Triton X-100, consistent with an association of QIL1 with the mature MICOS complex |
| GO:0005739 mitochondrion | IEA GO_REF:0000120 | ACCEPT | Summary: Automatic assignment from a family rule and from the Swiss-Prot subcellular location; both resolve to conditions APOOL genuinely satisfies, and both give the right answer. Reason: Both sources were traced. ARBA00026962 confers GO:0005739 and has 380 condition sets; the one APOOL satisfies is set 26, IPR019166 together with taxon Eukaryota. That is a sound condition, since IPR019166 is the MIC26/MIC27 family and every characterised member of it is mitochondrial. UniProtKB-SubCell SL-0173 is Mitochondrion and maps to GO:0005739 in the UniProt locations API; the Swiss-Prot SUBCELLULAR LOCATION line it derives from is itself ECO:0000269-backed by PubMed:23704930 and PubMed:25764979. An automatic route that lands on a term four experimental rows already support is not an over-reach. Propagation Review Root cause: NO FAILURE CORE Sources checked: ARBA:ARBA00026962 · ARBA rule ARBA00026962 (confers GO:0005739) SUPPORTS TRANSFER Fetched from the UniProt ARBA API. 380 condition sets, one annotation (GO:0005739). APOOL matches condition set 26, IPR019166 AND taxon Eukaryota, a family-plus-lineage condition that is correct for the MIC26/MIC27 family, every characterised member of which is mitochondrial. UniProtKB-SubCell:SL-0173 · UniProt subcellular location SL-0173 Mitochondrion SUPPORTS TRANSFER Resolved through the UniProt locations API. SL-0173 Mitochondrion maps to GO:0005739. The Q6UXV4 SUBCELLULAR LOCATION line that triggers it carries ECO:0000269 evidence from PubMed:23704930 and PubMed:25764979, so this is an experimentally grounded curated statement rather than a prediction. Supporting Evidence: file:human/APOOL/APOOL-uniprot.txt CC -!- SUBCELLULAR LOCATION: Mitochondrion inner membrane PMID:23704930 Endogenous APOOL was found to be entirely absent in the cytosolic fraction and instead was detected in the mitochondrial fraction. |
| GO:0005743 mitochondrial inner membrane | EXP PMID:23704930 APOOL is a cardiolipin-binding constituent of the Mitofilin/... | ACCEPT | Summary: The primary localisation result for this protein and a core annotation. Membrane-integrated, in the inner membrane, with the bulk of the protein facing the intermembrane space. Reason: Three complementary experiments in the cited paper, all on endogenous protein. Subcellular fractionation of 143B cells puts APOOL in the mitochondrial fraction and not the cytosol. Alkaline carbonate extraction leaves it entirely in the membrane pellet while the matrix chaperone mtHSP60 is released, showing it is integral or tightly membrane-associated rather than soluble. Graded-digitonin protease protection gives it the accessibility profile of the intermembrane-space marker Mitofilin, distinct from the outer-membrane marker TOM20 and from the matrix marker mtHSP60. UniProt annotates two transmembrane helices at 111-129 and 138-155 with intermembrane-space-facing termini, matching. This is the location at which every described function of the protein occurs. Supporting Evidence: PMID:23704930 APOOL was found to remain fully in the membrane fraction whereas the chaperone mtHsp60 was released efficiently, yet not completely, to the soluble fraction PMID:23704930 Taken together, we conclude that APOOL is located in the intermembrane space consistent with the proposed location of the Mitofilin/MINOS complex. PMID:23704930 We show that APOOL is a mitochondrial membrane protein facing the intermembrane space. |
| GO:0005743 mitochondrial inner membrane | EXP PMID:25764979 The non-glycosylated isoform of MIC26 is a constituent of th... | ACCEPT | Summary: Independent experimental confirmation of inner-membrane residence from a second study of this protein family. Reason: The cached record for this paper is abstract-only, so the specific experiment behind the row could not be read; the curator who made it will have had the full text. The abstract does state that the mitochondrial form of MIC26 spans the inner membrane and physically interacts with MIC60, MIC27 and MIC10, which places MIC27 in an inner-membrane complex, and the conclusion agrees with the direct carbonate-extraction and protease-protection data on MIC27 itself in PMID:23704930. Nothing in the record argues against it, so this is accepted as a duplicate of a well-established location rather than second-guessed. Supporting Evidence: PMID:25764979 The latter isoform spans the mitochondrial inner membrane and physically interacts with several MICOS complex subunits such as MIC60, MIC27, and MIC10. |
| GO:0005743 mitochondrial inner membrane | IEA GO_REF:0000044 | ACCEPT | Summary: Automatic mapping of the Swiss-Prot subcellular location to GO; the location statement it maps is itself experimentally backed. Reason: SL-0168 resolves through the UniProt locations API to Mitochondrion inner membrane and maps to exactly GO:0005743, so the transformation is faithful. The statement it transforms is not a prediction. The Q6UXV4 SUBCELLULAR LOCATION line carries ECO:0000269 evidence from PubMed:23704930 and PubMed:25764979 and adds Multi-pass membrane protein from PubMed:23704930. The automatic row therefore restates, at the correct granularity, what the two EXP rows on this gene assert directly. Propagation Review Root cause: NO FAILURE CORE Sources checked: UniProtKB-SubCell:SL-0168 · UniProt subcellular location SL-0168 Mitochondrion inner membrane SUPPORTS TRANSFER Resolved through the UniProt locations API. SL-0168 maps to GO:0005743 and to nothing else, so the mapping cannot over- or under-shoot. The Q6UXV4 location line it is derived from is ECO:0000269-backed by PubMed:23704930 and PubMed:25764979, which are also the two EXP rows on this gene. Supporting Evidence: file:human/APOOL/APOOL-uniprot.txt CC -!- SUBCELLULAR LOCATION: Mitochondrion inner membrane PMID:23704930 We show that APOOL is a mitochondrial membrane protein facing the intermembrane space. |
| GO:0005743 mitochondrial inner membrane | NAS PMID:25997101 QIL1 is a novel mitochondrial protein required for MICOS com... | ACCEPT | Summary: ComplexPortal's statement of where the MICOS complex sits; correct for APOOL as a subunit of that complex. Reason: Non-traceable author statement drawn from the MICOS ComplexPortal entry CPX-6141, which APOOL is a member of. The cited paper states plainly that the MIC10-MIC19-MIC25-MIC26-MIC27-MIC60 complex resides in the inner membrane at crista junctions. The claim is weakly evidenced as an annotation type but is the same claim the two EXP rows on this gene make from direct experiment, so there is nothing to correct. Supporting Evidence: PMID:25997101 The MIC10-MIC19-MIC25-MIC26-MIC27-MIC60 complex is thought to reside in the IM at the site of CJs. |
| GO:0007007 inner mitochondrial membrane organization | IC PMID:26477565 Evolution and structural organization of the mitochondrial c... | MODIFY | Summary: True but uninformatively general. The specific process APOOL participates in is cristae formation, a descendant of this term that the gene already carries. Reason: The inference is legitimate. A curator reasoned from MIB complex membership (GO:0140275, the supporting entity on this row) to participation in inner mitochondrial membrane organisation, and that is a correct entailment for a MICOS subunit. The problem is granularity. QuickGO's ancestor list for GO:0042407 cristae formation contains GO:0007007, so the term asserted here is a strict parent of the process for which this gene has direct human evidence. Single CRISPR knockout of MIC27 significantly reduces crista junctions and cristae per mitochondrial section (PMID:32788226), and miRNA knockdown significantly alters cristae ultrastructure (PMID:23704930). Nothing in the record suggests APOOL has any inner-membrane role outside cristae and crista junctions. The replacement term is already asserted on this gene by IBA, InterPro IEA and ComplexPortal NAS, so the practical effect of the modification is to consolidate rather than to add. It removes a parent-level row that conveys less than the child already present. Propagation Review Root cause: TERM SCOPING PROBLEM Failure modes: GRANULARITY MISMATCH Sources checked: GO:0140275 · MIB complex (the GO term this IC was inferred from) SUPPORTS TRANSFER The source is not a gene product but the gene's own MIB complex annotation, which this review accepts. The inference from it is valid; only the resulting term is coarser than the evidence supports, since cristae formation is a descendant of GO:0007007 and is already on the gene. Proposed replacements: cristae formation Supporting Evidence: PMID:26477565 In mitochondrial protein complexome profiles, the MIB complex occurs as a defined complex and as separate subcomplexes, potentially reflecting various assembly stages. PMID:32788226 We observed that both SKOs and DKO cells show significantly reduced CJs per mitochondrial section compared with the controls |
| GO:0031093 platelet alpha granule lumen | TAS Reactome:R-HSA-481007 | REMOVE | Summary: The second half of the same Reactome platelet projection, and the more specific of the two false locations. APOOL is not a platelet granule cargo protein. Reason: GO:0031093 is the volume enclosed by the platelet alpha granule membrane, so this row asserts that APOOL is a soluble granule cargo protein. It is not. It is an integral or tightly membrane-associated protein that resists alkaline carbonate extraction and carries two predicted transmembrane helices and a mitochondrial transit peptide rather than a signal peptide (PMID:23704930, UniProt Q6UXV4). The secreted-glycoprotein reading of this family was tested directly and rejected. PMID:37279200 excluded a high-molecular-weight glycosylated MIC27 and concluded that MIC26 and MIC27 are exclusively mitochondrial. The row's origin is a Reactome pathway membership traceable to a thrombin-releasate shotgun proteomics screen from 2004, a class of experiment in which abundant mitochondrial membrane proteins are a recognised background, and the reference projects GO:0005576 and GO:0031093 onto 67 gene products indiscriminately. This row and the extracellular region row are one error, not two independent observations. Supporting Evidence: PMID:37279200 We further excluded the presence of a glycosylated, high-molecular weight MIC27 protein. PMID:23704930 APOOL was found to remain fully in the membrane fraction whereas the chaperone mtHsp60 was released efficiently, yet not completely, to the soluble fraction PMID:14630798 Many of the proteins identified were not previously attributed to platelets |
| GO:0042407 cristae formation | IBA GO_REF:0000033 | ACCEPT | Summary: A correct phylogenetic call that has since been confirmed directly in human cells. Single CRISPR knockout of MIC27 significantly reduces crista junctions and cristae per mitochondrial section. Reason: The node behind this row, PTN001803267 in PTHR14564, carries an IBD for GO:0042407 whose sole gene-level seed is UniProtKB:Q9BUR5, human APOO/MIC26. That seed is experimentally grounded. QuickGO shows MIC26 carries an IMP for GO:0042407 from PMID:25764979. So this IBA is, in phylogenetic terms, MIC26's loss-of-function experiment inherited by its paralogue. APOOL is plainly inside the inheriting clade, being one of only two human members of PTHR14564 and experimentally a MICOS subunit, and there is no target-specific evidence of divergence. The family's conserved basic inter-transmembrane loop is not merely retained in APOOL but is the most basic version of it in the family alignment, with 10 of 10 members net positive and human APOOL at RKGSKFKK (residues 130-137, net +5). What makes this an unusually strong IBA to accept is that MIC27-specific human evidence has arrived independently since. Quantified electron microscopy of HAP1 MIC27 single knockouts shows significantly fewer crista junctions and fewer cristae per section, and miRNA knockdown earlier showed significantly altered cristae ultrastructure. The one negative report (PMID:25781180) scored cristae morphology qualitatively in shRNA knockdowns without controlling for compensatory upregulation of the paralogue, and the double knockout shows why that matters. Propagation Review Root cause: NO FAILURE CORE Sources checked: PANTHER:PTN001803267 · PTHR14564 MICOS complex subunit MIC26/MIC27 IBD node SUPPORTS TRANSFER The ancestral node that actually carries the assertion. The local PAINT slice (interpro/panther/PTHR14564/PTHR14564-paint.tsv) shows exactly two IBD rows in this family, both at this node, and no IRD or IKR anywhere, so no curator has recorded a loss or divergence event that would exclude APOOL. For GO:0042407 the node's seed list is a single gene product, Q9BUR5; that is not weak support, it is one well-characterised human descendant with a direct loss-of-function experiment. APOOL is one of the two human members descending from this node. UniProtKB:Q9BUR5 · APOO / MIC26 (human), the sole gene-level seed of this IBD SUPPORTS TRANSFER Resolved against UniProt as MIC26_HUMAN, Swiss-Prot, the paralogue of the target. QuickGO confirms the grounding for this exact term. MIC26 carries GO:0042407 with IMP evidence from PMID:25764979, not a further inference. Note that the target does not appear in this row's WITH/FROM, unlike the MICOS-complex IBA, because APOOL had no experimental cristae-formation annotation of its own when the node was placed in 2018; PMID:32788226 has since supplied human MIC27-specific evidence for the same process. Supporting Evidence: PMID:32788226 We observed that both SKOs and DKO cells show significantly reduced CJs per mitochondrial section compared with the controls PMID:32788226 In addition, the number of cristae per mitochondrial section was significantly reduced in all the KO cells compared with control, with a slightly more pronounced reduction in DKO cells compared with SKOs PMID:23704930 The frequency of mitochondrial sections showing cristae with small concentric structures that appeared branched and interconnected was significantly increased upon downregulation of APOOL file:human/APOOL/APOOL-bioinformatics/RESULTS.md 10/10 members carry at least two basic residues and a net positive charge in this block, and none carries an acidic residue there. |
| GO:0042407 cristae formation | IEA GO_REF:0000002 | ACCEPT | Summary: InterPro2GO from the MIC26/MIC27 family signatures; the mapping is appropriate and the conclusion is independently established for this gene. Reason: Both signatures were fetched from the InterPro API. IPR019166 (MICOS complex subunit MIC26/MIC27) and IPR033182 (the animal-restricted subset) each carry exactly two GO mappings, GO:0042407 and GO:0061617, which is what this and the companion IEA row assert. The mapping is well scoped. This is a small, functionally homogeneous family in which every characterised member is a MICOS subunit implicated in cristae architecture, so a family-level process inference does not over-generalise. APOOL matches both signatures, and the conclusion is supported directly by the human knockout and knockdown data. Propagation Review Root cause: NO FAILURE CORE Sources checked: InterPro:IPR019166 · MICOS complex subunit MIC26/MIC27 (InterPro family) SUPPORTS TRANSFER Fetched from the InterPro API. A family entry covering fungal and animal MIC26 plus mammalian MIC27, carrying exactly GO:0042407 and GO:0061617. APOOL matches it (UniProt DR line). The entry explicitly excludes budding yeast Mic27, which is the correct scoping, so the family is not being stretched across a divergent member. InterPro:IPR033182 · MICOS complex subunit MIC26/MIC27, animal (InterPro family) SUPPORTS TRANSFER The animal-restricted child entry, same two GO mappings. Narrower than IPR019166 and equally correct for a human protein; the two signatures agree rather than pulling in different directions. Supporting Evidence: PMID:32788226 We observed that both SKOs and DKO cells show significantly reduced CJs per mitochondrial section compared with the controls file:human/APOOL/APOOL-uniprot.txt DR InterPro; IPR019166; MIC26/MIC27. |
| GO:0042407 cristae formation | NAS PMID:25997101 QIL1 is a novel mitochondrial protein required for MICOS com... | ACCEPT | Summary: ComplexPortal's process statement for the MICOS complex, applied to APOOL as a subunit; correct, and now backed by direct human evidence. Reason: A non-traceable author statement derived from the MICOS ComplexPortal entry CPX-6141. As an evidence type it is weak, but the claim itself is not. The cited paper shows that disassembling MICOS by removing QIL1/MIC13, which specifically degrades MIC10, MIC26 and MIC27, produces aberrant cristae, and MIC27-specific loss-of-function evidence for the same process has since been published. Accepted as a duplicate of a well-supported process assignment. Supporting Evidence: PMID:25997101 Using quantitative proteomics, we show that loss of QIL1 resulted in MICOS disassembly with the accumulation of a MIC60-MIC19-MIC25 sub-complex and degradation of MIC10, MIC26, and MIC27. PMID:32788226 We obtained single and double knockout (DKO) human cells of MIC26 and MIC27 and found that DKO show more concentric onion-like cristae with loss of CJs than any single deletion indicating overlapping roles in formation of CJs. |
| GO:0044284 mitochondrial crista junction | NAS PMID:25997101 QIL1 is a novel mitochondrial protein required for MICOS com... | ACCEPT | Summary: The MICOS complex is concentrated at crista junctions, and APOOL is in the module whose loss costs the cell crista junctions; the location is right. Reason: GO:0044284 is the tubular neck connecting a crista to the inner boundary membrane. The cited paper states that the MIC10-MIC19-MIC25-MIC26-MIC27-MIC60 complex resides in the inner membrane at the site of crista junctions, and APOOL is a subunit of it. The functional data agree with the localisation rather than merely being compatible with it. Removing MIC27 significantly reduces the number of crista junctions per mitochondrial section, which is what one expects of a protein that acts at them. Non-traceable as an evidence type, so accepted on the strength of the underlying claim rather than of the annotation route. Supporting Evidence: PMID:25997101 The MIC10-MIC19-MIC25-MIC26-MIC27-MIC60 complex is thought to reside in the IM at the site of CJs. PMID:32788226 We observed that both SKOs and DKO cells show significantly reduced CJs per mitochondrial section compared with the controls |
| GO:0061617 MICOS complex | HDA PMID:26477565 Evolution and structural organization of the mitochondrial c... | ACCEPT | Summary: Complexome profiling assigns MIC27 to the MICOS form of the complex; this is the annotation the cited reference actually supports. Reason: High-throughput, but exactly on target. The study resolves the profile into three nested forms and places all MICOS proteins, MIC27 included, in form (A), the MICOS complex proper. The same assignment is made independently by direct co-immunoprecipitation (PMID:23704930) and by interaction proteomics in two human cell lines (PMID:25997101), and it is what UniProt records in its SUBUNIT block. Worth contrasting with the SAM complex row derived from this same reference, which the reference does not support and which is removed. Supporting Evidence: PMID:26477565 We find three main forms of the complex: A) The MICOS complex, containing all the MICOS proteins, B) a membrane bridging subcomplex, containing in addition SAMM50, MTX2 and the previously uncharacterized MTX3, and C) the complete MIB complex containing in addition DNAJC11 and MTX1. file:human/APOOL/APOOL-uniprot.txt CHCHD3/MIC19, CHCHD6/MIC25, APOOL/MIC27, IMMT/MIC60, APOO/MIC23/MIC26 |
| GO:0061617 MICOS complex | IBA GO_REF:0000033 | ACCEPT | Summary: A well-seeded phylogenetic assertion of MICOS membership, with the target's own experimental evidence among the descendant evidences that placed the node. Reason: The IBD sits at PTN001803267 in PTHR14564 and is seeded by three gene-level donors spanning the family, namely both human paralogues, MIC27 (the target) and MIC26, and the single-copy Drosophila member behind FB:FBgn0038400. The target appearing in its own WITH/FROM is correct and expected, not circular. APOOL has three independent IDA and HDA rows for this very term, and those are among the descendant evidences the PAINT curator used to decide where MICOS membership arose. What the IBA then adds is the further claim that membership is ancestral to the family rather than lineage-specific, and the fly data support exactly that, since knockdown of CG5903/Mic26-27 phenocopies Mitofilin/MIC60 and QIL1/MIC13 knockdown with loss of crista junctions. There is no evidence of divergence in the target. APOOL is experimentally in human MICOS, requires MIC13 for its incorporation like the rest of the MIC10 module, and retains the family's conserved basic inter-transmembrane loop. Note the evolutionary scope the transfer correctly does not exceed. MIC26 and MIC27 are opisthokont-restricted, and the node is placed within the family rather than deeper. Propagation Review Root cause: NO FAILURE CORE Sources checked: FB:FBgn0038400 · Mic26-27 / CG5903 (Drosophila melanogaster) SUPPORTS TRANSFER A FlyBase identifier, so not a key in any accession-keyed PANTHER entries index; resolved instead through the UniProt cross-reference to Q9VEY5 (TrEMBL, MICOS complex subunit, gene names including Mic26-27, dApoO and CG5903). This is the protein characterised in PMID:33268479 as a Drosophila MICOS component, so the seed has real experimental grounding and supplies the invertebrate, single-copy branch of the family. PANTHER:PTN001803267 · PTHR14564 MICOS complex subunit MIC26/MIC27 IBD node SUPPORTS TRANSFER The node carrying the assertion. The local PAINT slice shows this is one of only two IBD rows in the family, both at this node, with no IRD or IKR recorded anywhere, so no curator has flagged a loss of MICOS membership in any branch. The target descends from it. UniProtKB:Q6UXV4 · APOOL / MIC27 (human), the target itself SUPPORTS TRANSFER Self-reference, and it is the expected and correct pattern rather than circularity. APOOL's own IDA rows for GO:0061617 from PMID:25781180 and PMID:25997101, plus the HDA row from PMID:26477565, are among the descendant evidences that let the curator place MICOS membership at this node. Its presence marks experimental grounding on the target, and the IBA adds the separate claim that the membership is inherited. UniProtKB:Q9BUR5 · APOO / MIC26 (human), the paralogue SUPPORTS TRANSFER Resolved against UniProt as MIC26_HUMAN, Swiss-Prot. QuickGO confirms three independent IDA annotations to GO:0061617 on this donor, from PMID:25764979, PMID:25781180 and PMID:25997101, so it contributes experimental grounding rather than another inference. Supporting Evidence: PMID:25997101 we identified 5 core MICOS subunits (MIC60, MIC19, MIC25, MIC26 and MIC27), 3 OM known interactors (SAMM50, MTX1 and MTX2) as well as the chaperone DNAJC11 and TMEM11 in association with QIL1 PMID:33268479 We found that CG5903/MIC26-MIC27 colocalizes and functions with Mitofilin/MIC60 and QIL1/MIC13 as a Drosophila MICOS component; knocking down expression of any of these three genes predictably altered mitochondrial morphology, causing loss of cristae junctions, and disruption of cristae packing. PMID:26477565 In contrast, Mic23 and Mic27 appear to be the youngest MICOS proteins, as they only occur in opisthokonts. |
| GO:0061617 MICOS complex | IDA PMID:25781180 Detailed analysis of the human mitochondrial contact site co... | ACCEPT | Summary: Direct demonstration that Mic27/ApoOL behaves as a MICOS and MIB component in human cells; core. Reason: This study analysed all then-known human MICOS and MIB subunits together, using inducible knockdown cell lines, native gel analysis and in vitro import. Mic27/ApoOL is treated throughout as a complex member, and the interdependence data show it, since depletion of Mic60/Mitofilin or of Mic19/CHCHD3 dramatically reduces Mic27/ApoOL levels. The same paper concludes that MIC27 is a peripheral rather than a core subunit, which is a statement about importance within the complex, not about membership, and does not conflict with this row. Supporting Evidence: PMID:25781180 it also dramatically affected MICOS components Mic19/CHCHD3, Mic25/CHCHD6, Mic27/ApoOL, Mic10/MINOS1, as well as Mic23/ApoO PMID:25781180 In human mitochondria, the inner membrane MICOS complex interacts with the outer membrane sorting and assembly machinery (SAM) complex, to form the mitochondrial intermembrane space bridging complex (MIB). |
| GO:0061617 MICOS complex | IDA PMID:25997101 QIL1 is a novel mitochondrial protein required for MICOS com... | ACCEPT | Summary: MIC27 recovered as a MICOS subunit by interaction proteomics in two human cell lines, and specifically assigned to the MIC13-dependent module; core. Reason: MIC27 was identified as one of the MICOS subunits in FLAG and HA immunoprecipitations from 293T and HCT116 cells, in a complex of about 700 kDa, and co-purified with tagged QIL1 under digitonin but not Triton X-100, the detergent sensitivity expected of a genuine native complex rather than a co-precipitation artefact. The paper goes further and places MIC27 within MICOS, since loss of QIL1/MIC13 degrades MIC10, MIC26 and MIC27 while the MIC60-MIC19-MIC25 subcomplex accumulates, which is the basis for the MIC10-module assignment used throughout this review and is independently reproduced in PMID:27479602. Supporting Evidence: PMID:25997101 Using quantitative proteomics coupled with native gel analysis, we show that QIL1 is a component of a ∼700 KDa MICOS complex and its depletion from cells results in loss of MIC10, MIC26, and MIC27 from the complex and a reduction in the abundance of these proteins in mitochondria. PMID:27479602 MIC13 is required for the assembly of MIC10, MIC26, and MIC27 into the MICOS complex. |
| GO:0061617 MICOS complex | IEA GO_REF:0000002 | ACCEPT | Summary: InterPro2GO from the same two family signatures; the complex-membership half of the mapping, and correct. Reason: IPR019166 and IPR033182 each carry GO:0061617 among their two GO mappings, and APOOL matches both. Family-level complex membership is a safe inference here because the family is defined by MICOS participation, every characterised member in fungi, animals and nematodes being a MICOS subunit. The automatic row reaches the same conclusion as three independent experimental rows on this gene, so there is no over-reach to correct. Propagation Review Root cause: NO FAILURE CORE Sources checked: InterPro:IPR019166 · MICOS complex subunit MIC26/MIC27 (InterPro family) SUPPORTS TRANSFER Fetched from the InterPro API; carries GO:0042407 and GO:0061617 and nothing else. The family is defined by MICOS participation, so transferring complex membership from the signature is inference within, not beyond, the entry's own definition. InterPro:IPR033182 · MICOS complex subunit MIC26/MIC27, animal (InterPro family) SUPPORTS TRANSFER The animal-restricted child entry with the same two mappings. Both signatures match APOOL and agree, and neither extends the claim past the clade in which MIC26/MIC27 exist. Supporting Evidence: file:human/APOOL/APOOL-uniprot.txt DR InterPro; IPR033182; MIC26/MIC27_animal. PMID:25997101 we identified 5 core MICOS subunits (MIC60, MIC19, MIC25, MIC26 and MIC27), 3 OM known interactors (SAMM50, MTX1 and MTX2) as well as the chaperone DNAJC11 and TMEM11 in association with QIL1 |
| GO:0061617 MICOS complex | NAS PMID:25997101 QIL1 is a novel mitochondrial protein required for MICOS com... | ACCEPT | Summary: The ComplexPortal statement of MICOS composition; duplicates, at weaker evidence, what the IDA rows on this gene establish directly. Reason: Derived from ComplexPortal entry CPX-6141, which lists APOOL/MIC27 as a MICOS component. The underlying paper names MIC27 among the MICOS subunits it recovers, and UniProt's curated SUBUNIT block lists it in the same composition. Weak as an evidence type, but there is nothing to correct, since it asserts the same membership as three experimental rows on this gene. Supporting Evidence: file:human/APOOL/APOOL-uniprot.txt CHCHD3/MIC19, CHCHD6/MIC25, APOOL/MIC27, IMMT/MIC60, APOO/MIC23/MIC26 PMID:25997101 The MIC10-MIC19-MIC25-MIC26-MIC27-MIC60 complex is thought to reside in the IM at the site of CJs. |
| GO:0140275 MIB complex | HDA PMID:26477565 Evolution and structural organization of the mitochondrial c... | ACCEPT | Summary: Correct, and it is this row rather than the SAM complex row that records APOOL's relationship to the outer-membrane machinery. Reason: GO:0140275 is defined as a bridging complex consisting of MICOS components in the inner membrane, SAM components in the outer membrane, DNAJC11 and metaxin 1. APOOL qualifies through the MICOS half of that definition, which is precisely why the MIB row is right where the SAM row is wrong. MIB membership does not require the protein to be in the outer membrane, whereas SAM membership does. The cited complexome profiling resolves the MIB complex as a defined assembly containing all MICOS proteins plus SAMM50, MTX1, MTX2 and DNAJC11, and the physical link is independently visible in APOOL co-immunoprecipitations, which pull down SAMM50 (PMID:23704930), and in MICOS interaction proteomics that recover SAMM50, MTX1 and MTX2 alongside MIC27 (PMID:25997101). Supporting Evidence: PMID:26477565 We find three main forms of the complex: A) The MICOS complex, containing all the MICOS proteins, B) a membrane bridging subcomplex, containing in addition SAMM50, MTX2 and the previously uncharacterized MTX3, and C) the complete MIB complex containing in addition DNAJC11 and MTX1. PMID:23704930 These experiments revealed that APOOL is co-purified with Mitofilin, MINOS1, and SAMM50 PMID:25781180 In human mitochondria, the inner membrane MICOS complex interacts with the outer membrane sorting and assembly machinery (SAM) complex, to form the mitochondrial intermembrane space bridging complex (MIB). |
| GO:1901612 cardiolipin binding | IDA PMID:23704930 APOOL is a cardiolipin-binding constituent of the Mitofilin/... | NEW | Summary: The one molecular activity ever demonstrated for this protein, and it is absent from GOA entirely. Recombinant APOOL binds cardiolipin specifically and does not bind its precursor phosphatidylglycerol. Reason: APOOL currently has no molecular function annotation of any kind except three bare GO:0005515 protein binding rows, and neither GOA nor the UniProt DR GO block carries cardiolipin binding, even though it is the finding the defining paper is titled for and the activity UniProt states in its own curated FUNCTION block. GO:1901612 exists, is not obsolete, and is defined simply as binding to cardiolipin. The evidence is a direct in vitro binding assay on the purified gene product. GST-APOOL-HIS6 was expressed, affinity purified in two steps and applied to membrane lipid strips in triplicate, where it bound cardiolipin and none of diacylglycerol, phosphatidic acid, phosphatidylserine, phosphatidylethanolamine, phosphatidylcholine, phosphatidylglycerol or phosphatidylinositol, while the GST-only control bound nothing. Failure to bind phosphatidylglycerol, the immediate biosynthetic precursor which differs from cardiolipin essentially by the second phosphatidyl moiety, is what makes this a specific rather than a generic anionic-lipid interaction, so GO:1901612 is the right level and GO:0008289 lipid binding would be too broad. IDA is the correct code, since this is a direct assay of the gene product and not a perturbation of its level. The biological plausibility is strong, cardiolipin being concentrated in the inner membrane where APOOL sits and the family's basic inter-transmembrane loop that is the proposed contact site being conserved in all ten family members tested here, but the binding claim rests on the in vitro assay alone and its cellular consequence is recorded as an open knowledge gap rather than asserted. Supporting Evidence: PMID:23704930 GST-APOOL-HIS6 was found to bind specifically to cardiolipin (CL) but not to any of the other lipids tested, notably not even to phosphatidylglycerol the precursor of cardiolipin PMID:23704930 The control GST non-fusion protein did not bind to any lipid. file:human/APOOL/APOOL-uniprot.txt cardiolipin (in vitro) but not to the precursor lipid file:human/APOOL/APOOL-bioinformatics/RESULTS.md The inter-TM loop is therefore residues 130–137, `RKGSKFKK`: five of eight residues are lysine or arginine, no acidic residue, net charge +5. |
Loading supporting content…
Download this section (compressed HTML)Q: Does the basic inter-transmembrane loop of MIC27 (human residues 130-137, RKGSKFKK) mediate the cardiolipin binding seen on lipid strips, and is that binding required for crista junction formation, or does MIC27 support crista junctions through protein contacts within the MIC10 module regardless of lipid binding?
Suggested experts: Andreas S. Reichert, Ruchika Anand
Q: What is the stoichiometry and arrangement of MIC10, MIC26, MIC27 and MIC13 within the MIC10 module, and do MIC26 and MIC27 occupy the same position in the module, as their reciprocal abundance regulation and partial functional redundancy would suggest?
Suggested experts: Arun Kumar Kondadi, Siewert J. Marrink
Q: By what post-transcriptional mechanism does loss of one of MIC26 or MIC27 raise the level of the other, and is the compensating paralogue actually taking over the vacated function or merely accumulating?
Suggested experts: Ruchika Anand, Sebastian Koob
Q: Do the cardiolipin reduction and the OXPHOS supercomplex destabilisation in MIC26/MIC27 double knockout cells lie downstream of the loss of crista junctions, or does the pair influence cardiolipin metabolism more directly?
Suggested experts: Ilka Wittig, Ruchika Anand
Q: Is the yeast finding that Mic27 stabilises Mic10 oligomers true of human MIC27, given that human MIC26 and MIC27 knockouts leave the assembly and stability of the other MICOS subunits intact?
Suggested experts: Andreas S. Reichert
Q: Is there any physiological condition, tissue or developmental stage in which MIC27 is not functionally redundant with MIC26, given that the two are co-expressed and mutually compensating in every cell line examined so far?
Suggested experts: Vera Kozjak-Pavlovic
Experiment: Re-express wild-type APOOL and a loop-neutralised variant, with the basic residues of 130-137 mutated to alanine or glutamine, in MIC27 knockout cells, and compare cardiolipin binding by liposome flotation, MICOS assembly by blue-native PAGE, and crista junction number by quantitative electron microscopy. This is the single experiment that would convert cardiolipin binding from a demonstrated in vitro activity into a demonstrated cellular function.
Hypothesis: Cardiolipin binding by the basic inter-transmembrane loop is required for MIC27 to support crista junction formation, and is separable from MICOS assembly.
Type: structure-function rescue in a knockout background
Experiment: Measure APOOL-cardiolipin binding quantitatively rather than on lipid strips, using surface plasmon resonance or isothermal titration calorimetry against liposomes of defined cardiolipin content with phosphatidylglycerol and phosphatidic acid as specificity controls, to establish an affinity and test whether the specificity survives in a membrane.
Hypothesis: The discrimination between cardiolipin and phosphatidylglycerol seen on lipid strips reflects a real affinity difference in a bilayer context.
Type: quantitative lipid binding
Experiment: Determine the structure of the human MIC10 module (MIC10, MIC26, MIC27, MIC13) in a lipid environment, by cryo-EM of a detergent- or nanodisc-reconstituted complex, to test the predicted MIC10-MIC27 dimer arrangement and locate the cardiolipin sites.
Hypothesis: MIC27 and MIC10 form a wedge-shaped heterodimer that positions cardiolipin at the inter-helical loops, as the simulations predict.
Type: structural biology
Experiment: Perform cycloheximide chase and selective protease inhibition in MIC26 and MIC27 single knockout cells to determine whether the compensation acts on protein turnover rather than synthesis, and test dose dependence by expressing MIC26 from a heterologous promoter in MIC27 knockout cells.
Hypothesis: The reciprocal abundance regulation of MIC26 and MIC27 is degradative, with the surviving paralogue stabilised by occupying the position vacated in the MIC10 module.
Type: protein turnover
Experiment: Restore cardiolipin genetically in MIC26/MIC27 double knockout cells, as already shown to rescue respiratory complex stability, and ask by quantitative electron microscopy whether crista junctions return. If they do not, the lipid and the membrane geometry are separable.
Hypothesis: The OXPHOS phenotype of the MIC26/MIC27 double knockout is the cardiolipin arm alone, separable from the loss of crista junctions.
Type: genetic rescue with ultrastructural readout
Experiment: Express Drosophila Mic26-27 (CG5903) in human MIC26/MIC27 double knockout cells and score crista junction rescue, which would show whether the mammalian pair genuinely partitioned an ancestral function or merely duplicated it.
Hypothesis: The two mammalian paralogues partitioned an ancestral function that the single-copy invertebrate protein still performs in full.
Type: cross-species complementation
What is not known — curated, literature-grounded statements of the open unknowns (the inverse of core functions).
Gap: No cardiolipin-binding-deficient APOOL mutant has ever been made, so the protein's only demonstrated molecular activity has never been connected to any cellular phenotype.
OPEN BIOLOGY MF_DARK
What is known: Firmly established: purified recombinant APOOL binds cardiolipin and not phosphatidylglycerol or any other common membrane phospholipid, with a GST-only negative control. Also firmly established: losing APOOL costs the cell crista junctions and cristae. What has never been shown is that the first causes the second. Every phenotype in the literature comes from removing the whole protein, never from removing only its lipid binding, and no dissociation constant, binding stoichiometry or structure of the complex has been reported.
Significance: This is the hinge of the entire model for the protein. If cardiolipin binding is what couples the MICOS scaffold to the inner membrane, a binding-dead APOOL should fail to support crista junctions while still assembling into MICOS, a clean separation that the existing null alleles cannot provide. If instead the loop is dispensable, APOOL's role must be reassigned to protein-protein contacts within the MIC10 module and cardiolipin binding becomes incidental.
What would resolve it: The 2026 simulations nominate a specific structural element and so make the experiment designable for the first time. Neutralise the basic inter-transmembrane loop (human residues 130-137, RKGSKFKK, retained across all ten family members tested here) and re-express it in MIC27 knockout cells, then score MICOS assembly by blue-native gel and crista junctions by quantitative electron microscopy. Binding should be measured directly rather than on lipid strips, for instance by liposome flotation or surface plasmon resonance with defined cardiolipin content.
Provenance (the field's own admissions):
Gap: Whether MIC26 and MIC27 act directly on respiratory chain supercomplexes and the F1Fo-ATP synthase, or only indirectly through reduced cardiolipin and lost crista junctions, is unresolved, and the authors who reported the phenotype say so themselves.
OPEN BIOLOGY BP_DARK
What is known: Established: double knockout of MIC26 and MIC27 in human cells lowers cardiolipin, destabilises respiratory chain complexes and supercomplexes and the monomeric F1Fo-ATP synthase, and overexpressing cardiolipin synthase in those cells restores respiratory complex stability. Also established: the effect needs both paralogues gone, and single knockouts show little or none of it. Not established: whether either protein contacts or chaperones any OXPHOS complex, or whether the whole effect is downstream of the lipid and the membrane geometry.
Significance: The answer decides whether MIC26 and MIC27 belong in the OXPHOS assembly literature at all, and whether any GO biological process annotation for respiratory chain complex assembly is warranted for this gene. No such annotation is proposed here precisely because the question is open, a double-knockout phenotype rescued by a lipid enzyme being exactly the pattern an indirect effect produces.
What would resolve it: Distinguishing experiments would be complexome profiling of cells in which cardiolipin is restored genetically but crista junctions are not, to separate the lipid from the geometry; proximity labelling from MIC27 to test for physical neighbourhood with OXPHOS subunits; and rescue of the double knockout with a binding-dead APOOL, which should restore complex assembly only if the contribution is structural rather than lipidic.
Provenance (the field's own admissions):
Gap: The mechanism by which MIC26 and MIC27 reciprocally control each other's protein abundance is unknown, and it is the confound that made a decade of single-gene depletion studies of this pair difficult to interpret.
OPEN BIOLOGYCURATION RESIDUAL_SUBGAP
What is known: Established: deleting either paralogue raises the other's protein level, and this happens without a change in the other's mRNA, so it is post-transcriptional. Established too that the consequences are real, since it is why the single knockouts give only partial cristae phenotypes and why two careful human studies reached opposite conclusions about whether MIC27 depletion affects cristae at all. Not established: whether the surviving paralogue is stabilised by occupying the vacated position in the MIC10 module, protected from a specific protease, or regulated at translation.
Significance: Beyond the mechanism itself, this is a standing warning for curation of this gene pair. Any single-depletion phenotype, positive or negative, is measured in a cell that has compensated. The one negative cristae report for MIC27 in the literature is best read in that light rather than as a contradiction.
What would resolve it: Cycloheximide chase and proteasome or AAA-protease inhibition in each single knockout would separate degradation from synthesis; expressing MIC26 from a heterologous promoter in MIC27 knockout cells would test whether the compensation is dose-dependent; and pulse-labelling would settle whether translation rate changes.
Provenance (the field's own admissions):
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)