SMIM22, also called CASIMO1 (cancer-associated small integral membrane open reading frame 1), is an 83-amino-acid microprotein (about 10 kDa) encoded by a short open reading frame (sORF). It has a single predicted transmembrane helix (residues 32-52), followed by a cysteine-rich juxtamembrane stretch and a short, charged, disordered C-terminal tail. A splice isoform inserts five residues into this tail. The protein is membrane-associated; it partly overlaps with LAMP1-positive late endosomes and co-purifies with the ER membrane proteins calnexin and DDOST. Its best supported partner is squalene epoxidase (SQLE), the ER enzyme that catalyses the first oxygenation step of sterol synthesis. Raising SMIM22 levels increases SQLE protein without changing SQLE mRNA and causes lipid droplets to cluster. Lowering SMIM22 reduces SQLE protein and ERK phosphorylation. SMIM22 RNA is elevated in hormone receptor-positive breast tumours. In breast cancer cell lines, knockdown slows proliferation with a G0/G1 arrest, disorganises the actin cytoskeleton and reduces cell motility. SQLE knockdown reproduces this phenotype and SQLE overexpression rescues it, which places SQLE downstream of SMIM22. A non-translatable mutant does not reproduce the overexpression phenotype, so these effects are attributed to the protein rather than to the RNA. A single later study in lung cancer cells linked SMIM22 to glycolysis and proliferation. Whether SMIM22 has a role in non-transformed cells is unknown.
| GO Term | Evidence | Action | Reason |
|---|---|---|---|
| GO:0005515 protein binding | IPI PMID:29765154 The cancer-associated microprotein CASIMO1 controls cell pro... | MODIFY | Summary: IPI from the CASIMO1 discovery study, listing calnexin (CANX, P27824), the oligosaccharyltransferase subunit DDOST (P39656) and squalene epoxidase (SQLE, Q14534) as interactors. Bare protein binding says nothing about function. Reason: Of the three partners, only SQLE has functional support. SMIM22 raises SQLE protein levels, and SQLE knockdown and overexpression are epistatic to SMIM22 knockdown (PMID:29765154). CANX and DDOST are abundant ER-membrane proteins that co-purify with many ER-inserted transmembrane proteins, so they add no functional information. Enzyme binding (GO:0019899) is the most informative supported term for the SQLE interaction. A regulator term (for example, enzyme regulator activity) is not used, because the abstract reports changes in SQLE protein abundance rather than in its catalytic rate. Proposed replacements: enzyme binding Supporting Evidence: PMID:29765154 CASIMO1 microprotein interacts with squalene epoxidase (SQLE), a key enzyme in cholesterol synthesis and a known oncogene in breast cancer. PMID:29765154 Overexpression of CASIMO1 leads to SQLE protein accumulation without affecting its RNA levels and increased lipid droplet clustering, while knockdown of CASIMO1 decreased SQLE protein abundance and ERK phosphorylation downstream of SQLE. |
| GO:0005770 late endosome | IEA GO_REF:0000044 | KEEP AS NON CORE | Summary: UniProt subcellular-location mapping. UniProt records partial co-localisation with LAMP1 in late endosomes, based on the discovery paper (PMID:29765154). Reason: The underlying observation is a partial co-localisation in an imaging experiment that is not described in the cached abstract. The functional partner SQLE, and the co-purifying CANX and DDOST, are ER-membrane proteins, so the late endosome is unlikely to be the only or main site of SMIM22 action. The row is retained as curator-reported but is not treated as the defining location. |
| GO:0016020 membrane | IEA GO_REF:0000044 | ACCEPT | Summary: UniProt subcellular-location mapping to membrane, consistent with the single predicted helical transmembrane segment (residues 32-52) and the name "small integral membrane protein". Reason: Correct, though general. SMIM22 is a single-pass integral membrane microprotein. A more specific membrane (ER or endosome) cannot be assigned confidently from the available evidence. |
| GO:0030335 positive regulation of cell migration | IMP PMID:29765154 The cancer-associated microprotein CASIMO1 controls cell pro... | KEEP AS NON CORE | Summary: IMP from CASIMO1 knockdown in breast cancer cell lines, which inhibits cell motility. Reason: The knockdown phenotype is reported in the abstract, so the annotation is supported. The effect comes from cancer cell lines, and SQLE knockdown and rescue experiments place it downstream of SQLE, so it is more likely a downstream consequence than a direct activity of SMIM22. Supporting Evidence: PMID:29765154 Its loss disturbs the organization of the actin cytoskeleton, leads to inhibition of cell motility, and causes a G0/G1 cell cycle arrest. PMID:29765154 SQLE knockdown mimicked the CASIMO1 knockdown phenotype and in turn SQLE overexpression fully rescued the effect of CASIMO1 knockdown. |
| GO:0032956 regulation of actin cytoskeleton organization | IMP PMID:29765154 The cancer-associated microprotein CASIMO1 controls cell pro... | KEEP AS NON CORE | Summary: IMP from CASIMO1 knockdown, which disturbs actin cytoskeleton organisation in breast cancer cells. Reason: The phenotype is reported in the abstract. No mechanism links SMIM22 to actin-regulatory proteins. The effect is most plausibly indirect, through SQLE, sterol supply and ERK signalling, so it is kept as non-core. Supporting Evidence: PMID:29765154 Its loss disturbs the organization of the actin cytoskeleton, leads to inhibition of cell motility, and causes a G0/G1 cell cycle arrest. |
| GO:0042127 regulation of cell population proliferation | IBA GO_REF:0000033 | KEEP AS NON CORE | Summary: PAINT phylogenetic assertion of regulation of cell population proliferation. SMIM22 itself is the experimental source (it appears in its own WITH/FROM, as expected). Reason: Consistent with the IMP evidence on SMIM22 itself (PMID:29765154). The node placement is not contested. Proliferation is a cell-level phenotype, though, not the molecular role of SMIM22, so the row is kept as non-core, matching the IMP row. |
| GO:0042127 regulation of cell population proliferation | IMP PMID:29765154 The cancer-associated microprotein CASIMO1 controls cell pro... | KEEP AS NON CORE | Summary: IMP from the discovery study. CASIMO1 knockdown reduces proliferation in several breast cancer lines, and overexpression of the protein, but not of a non-translatable mutant, produces a proliferation phenotype. Reason: This is the best supported phenotype of SMIM22. It is attributed to the protein, and SQLE overexpression rescues it. It is a phenotype in transformed cells that acts through SQLE, so it is retained as non-core rather than as the molecular function. A separate lung cancer study (PMID:39215037) also reports that SMIM22 overexpression promotes proliferation, but that study is of low quality. Supporting Evidence: PMID:29765154 Its knockdown leads to decreased proliferation in multiple breast cancer cell lines. PMID:29765154 The proliferation phenotype upon overexpression is observed only with CASIMO1 protein expression, but not with a non-translatable mutant attributing the effects to the sORF-derived protein rather than a lncRNA function. PMID:29765154 SQLE knockdown mimicked the CASIMO1 knockdown phenotype and in turn SQLE overexpression fully rescued the effect of CASIMO1 knockdown. |
| GO:0051726 regulation of cell cycle | IMP PMID:29765154 The cancer-associated microprotein CASIMO1 controls cell pro... | KEEP AS NON CORE | Summary: IMP from the discovery study. CASIMO1 loss causes a G0/G1 cell cycle arrest. Reason: The arrest is reported in the abstract, but it is a readout of the reduced proliferation rather than evidence that SMIM22 acts on the cell cycle machinery. The row is retained as non-core and adds little beyond the proliferation annotation. Supporting Evidence: PMID:29765154 Its loss disturbs the organization of the actin cytoskeleton, leads to inhibition of cell motility, and causes a G0/G1 cell cycle arrest. |
| GO:0140042 lipid droplet formation | IMP PMID:29765154 The cancer-associated microprotein CASIMO1 controls cell pro... | KEEP AS NON CORE | Summary: IMP from the discovery study. CASIMO1 overexpression increases lipid droplet clustering, and UniProt states that the SMIM22-SQLE interaction modulates lipid droplet formation. Reason: The abstract reports increased lipid droplet clustering on overexpression. Quantitative lipid droplet formation data, which the curator may have seen in the full text, are not available to this review, so the experimental annotation is not overruled. The effect is most plausibly indirect, through stabilisation of SQLE and increased sterol synthesis. Formation of the droplets themselves is carried out by other proteins (seipin and the neutral-lipid synthesising enzymes). GO has no 'regulation of lipid droplet formation' term, so the row cannot be refined to a regulation term and is kept as non-core. Supporting Evidence: PMID:29765154 Overexpression of CASIMO1 leads to SQLE protein accumulation without affecting its RNA levels and increased lipid droplet clustering, while knockdown of CASIMO1 decreased SQLE protein abundance and ERK phosphorylation downstream of SQLE. PMID:29765154 CASIMO1 microprotein interacts with squalene epoxidase (SQLE), a key enzyme in cholesterol synthesis and a known oncogene in breast cancer. |
Loading supporting content…
Download this section (compressed HTML)Q: Does SMIM22 stabilise SQLE against degradation? SQLE turnover is controlled by cholesterol-induced, MARCHF6-dependent ER-associated degradation. If SMIM22 shields SQLE from MARCHF6, protein stabilization (GO:0050821) or a negative regulation of ERAD-type term would be justified. The abstract shows only steady-state protein accumulation.
Suggested experts: Sven Diederichs, Andrew J Brown
Q: Where does SMIM22 reside at steady state: ER, lipid droplet surface, or late endosome? Its partners (SQLE, CANX, DDOST) are ER proteins, yet the only localisation recorded in UniProt is a partial late-endosome overlap.
Suggested experts: Sven Diederichs
Q: Does SMIM22 have a function outside transformed cells, for example in normal mammary epithelium or in tissues where it is highly expressed?
Experiment: Measure SQLE half-life by cycloheximide chase in SMIM22-knockdown and SMIM22-overexpressing cells, under low and high sterol conditions and with or without MARCHF6 depletion. Test whether a SQLE-binding-deficient SMIM22 mutant (for example, with the TM helix or cysteine-rich juxtamembrane motif altered) loses the effect.
Hypothesis: SMIM22 extends the half-life of SQLE by protecting it from MARCHF6-mediated ERAD.
Experiment: Map the subcellular distribution of endogenous or endogenously tagged SMIM22 by co-staining with ER, lipid droplet and LAMP1 markers, and determine its membrane topology by protease protection.
Hypothesis: SMIM22 is an ER-resident single-pass membrane protein.
Loading supporting content…
Download this section (compressed HTML)Loading supporting content…
Download this section (compressed HTML)