CERS3 (ceramide synthase 3; formerly LASS3) is an endoplasmic reticulum membrane enzyme that catalyzes the N-acylation step of de novo ceramide biosynthesis: transfer of a fatty acyl group from acyl-CoA onto the amino group of a sphingoid base (dihydrosphingosine/sphinganine in the de novo pathway, sphingosine in the salvage pathway) to form dihydroceramide/ceramide (sphingosine N-acyltransferase, EC 2.3.1.24). Among the six mammalian ceramide synthases it is distinguished by its selectivity for very-long- and ultra-long-chain acyl-CoAs (chain length greater than C22, up to C26 and longer). CERS3 produces the ultra-long-chain ceramides, including omega-hydroxy species and acylceramides, that form the extracellular lipid lamellae and cornified lipid envelope of the epidermis and are essential for the skin permeability barrier. It is expressed most prominently in the epidermis (at the interface between the stratum granulosum and stratum corneum) and in testis. It is a multi-pass ER membrane protein whose catalytic activity resides in a TLC (TRAM-LAG1-CLN8) domain; it also carries a degenerate homeobox-like region that has lost DNA-binding residues and is not functional. Loss-of-function variants cause autosomal recessive congenital ichthyosis 9 (ARCI9), a keratinization/skin-barrier disorder.
| GO Term | Evidence | Action | Reason |
|---|---|---|---|
| GO:0005783 endoplasmic reticulum | IBA GO_REF:0000033 | ACCEPT | Summary: Phylogenetic (IBA) inference that CERS3 is active in the endoplasmic reticulum. Correct: CERS3 is an ER membrane enzyme. The more precise curated location is ER membrane (GO:0005789); this ER term is accurate but less specific. Reason: CERS3 and the whole LAG1/LASS/CerS family are ER-resident enzymes, and UniProt records ER membrane as the subcellular location. The IBA is consistent with the experimental/curated localization. Supporting Evidence: Reactome:R-HSA-428185 Ceramide synthase enzymes (CerS, LASS) associated with the endoplasmic reticulum membrane catalyze the reaction of a sphingoid and a long-chain fatty acyl CoA such as stearyl-CoA to form a dihydroceramide and CoASH |
| GO:0046513 ceramide biosynthetic process | IBA GO_REF:0000033 | ACCEPT | Summary: Phylogenetic (IBA) inference that CERS3 is involved in ceramide biosynthesis. This is a core function of CERS3, independently supported by direct experimental evidence. Reason: CERS3 catalyzes the committed N-acylation step of de novo ceramide synthesis; ceramide biosynthetic process is a core biological process. Supporting Evidence: PMID:23754960 CERS3 generates epidermis-specific ceramides by N-acylating dihydrosphingosine with acyl-CoAs ranging from long to very long aliphatic chains |
| GO:0050291 sphingosine N-acyltransferase activity | IBA GO_REF:0000033 | ACCEPT | Summary: Phylogenetic (IBA) inference of sphingosine N-acyltransferase (ceramide synthase) activity. This is the core molecular function of CERS3. Reason: This is the exact core catalytic activity of CERS3 (transfer of an acyl chain from acyl-CoA onto a sphingoid base). Supporting Evidence: PMID:17977534 ceramide is synthesized by N-acylation of a sphingoid long-chain base |
| GO:0003677 DNA binding | IEA GO_REF:0000002 | REMOVE | Summary: Automated InterPro2GO annotation triggered by the homeodomain-like signature (IPR001356) that CERS3 shares with the LAG1/LASS/CerS family. CERS3 does not bind DNA: its homeobox-like region is degenerate and lacks the residues required for DNA binding, and the protein is an ER membrane enzyme. Reason: This is a demonstrably wrong electronic (IEA) inference. UniProt explicitly cautions that the predicted homeobox domain is degenerated and lacks DNA-binding residues, and that ER localization argues against homeobox function. The catalytic activity resides in the TLC domain. Supporting Evidence: file:human/CERS3/CERS3-uniprot.txt Contains a predicted homeobox domain which is degenerated, lacking residues important for DNA-binding. |
| GO:0005789 endoplasmic reticulum membrane | IEA GO_REF:0000120 | ACCEPT | Summary: Automated (ARBA/UniProtKB-SubCell) annotation to ER membrane. This is the precise, correct subcellular location of CERS3. Reason: CERS3 is a multi-pass ER membrane protein; ER membrane is the accurate curated location and the site of its catalytic activity. Supporting Evidence: file:human/CERS3/CERS3-uniprot.txt SUBCELLULAR LOCATION: Endoplasmic reticulum membrane |
| GO:0016020 membrane | IEA GO_REF:0000002 | MARK AS OVER ANNOTATED | Summary: Automated InterPro2GO annotation to the generic 'membrane' term. Correct in substance (CERS3 is an integral membrane protein) but far less informative than the ER membrane annotation. Reason: 'membrane' is a high-level, uninformative location. The specific ER membrane term (GO:0005789) is already present and should be preferred. Supporting Evidence: file:human/CERS3/CERS3-uniprot.txt SUBCELLULAR LOCATION: Endoplasmic reticulum membrane |
| GO:0046513 ceramide biosynthetic process | IEA GO_REF:0000120 | ACCEPT | Summary: Automated (ARBA/ortholog/InterPro) annotation to ceramide biosynthetic process, redundant with the experimental and IBA annotations. Correct core process. Reason: Ceramide biosynthesis is the core process CERS3 participates in. Supporting Evidence: PMID:23754960 CERS3 generates epidermis-specific ceramides by N-acylating dihydrosphingosine with acyl-CoAs ranging from long to very long aliphatic chains |
| GO:0050291 sphingosine N-acyltransferase activity | IEA GO_REF:0000120 | ACCEPT | Summary: Automated (ARBA/InterPro/RHEA/EC 2.3.1.24) annotation of the core ceramide synthase activity, redundant with the experimental, ISS, IBA and TAS annotations. Reason: Correctly captures the core catalytic function via EC 2.3.1.24 / RHEA mapping. Supporting Evidence: PMID:17977534 ceramide is synthesized by N-acylation of a sphingoid long-chain base |
| GO:0005515 protein binding | IPI PMID:32296183 A reference map of the human binary protein interactome. | MARK AS OVER ANNOTATED | Summary: Bare 'protein binding' from a genome-scale binary interactome (HuRI, Y2H) screen, one of five partners reported (here UniProtKB:Q8N138 / ORMDL3). Reason: Per curation guidelines, bare 'protein binding' is uninformative and is not retained as a core function. Experimental IPI is not removed. Supporting Evidence: PMID:32296183 we present a human 'all-by-all' reference interactome map of human binary protein interactions, or 'HuRI' |
| GO:0005515 protein binding | IPI PMID:32296183 A reference map of the human binary protein interactome. | MARK AS OVER ANNOTATED | Summary: Bare 'protein binding' from the HuRI binary interactome screen (partner UniProtKB:Q96FB2). Uninformative molecular function term. Reason: Bare 'protein binding' is not retained as a core function. Experimental IPI is not removed. Supporting Evidence: PMID:32296183 we present a human 'all-by-all' reference interactome map of human binary protein interactions, or 'HuRI' |
| GO:0005515 protein binding | IPI PMID:32296183 A reference map of the human binary protein interactome. | MARK AS OVER ANNOTATED | Summary: Bare 'protein binding' from the HuRI binary interactome screen (partner UniProtKB:Q99519 / NEU1). Uninformative molecular function term. Reason: Bare 'protein binding' is not retained as a core function. Experimental IPI is not removed. Supporting Evidence: PMID:32296183 we present a human 'all-by-all' reference interactome map of human binary protein interactions, or 'HuRI' |
| GO:0005515 protein binding | IPI PMID:32296183 A reference map of the human binary protein interactome. | MARK AS OVER ANNOTATED | Summary: Bare 'protein binding' from the HuRI binary interactome screen (partner UniProtKB:Q9H0N5 / PCBD2). Uninformative molecular function term. Reason: Bare 'protein binding' is not retained as a core function. Experimental IPI is not removed. Supporting Evidence: PMID:32296183 we present a human 'all-by-all' reference interactome map of human binary protein interactions, or 'HuRI' |
| GO:0005515 protein binding | IPI PMID:32296183 A reference map of the human binary protein interactome. | MARK AS OVER ANNOTATED | Summary: Bare 'protein binding' from the HuRI binary interactome screen (partner UniProtKB:Q9NUM3 / SLC39A9). Uninformative molecular function term. Reason: Bare 'protein binding' is not retained as a core function. Experimental IPI is not removed. Supporting Evidence: PMID:32296183 we present a human 'all-by-all' reference interactome map of human binary protein interactions, or 'HuRI' |
| GO:0005783 endoplasmic reticulum | IEA GO_REF:0000107 | ACCEPT | Summary: Automated ortholog transfer (Ensembl Compara, from mouse Cers3 Q1A3B0) of ER localization. Correct but less precise than ER membrane. Reason: CERS3 is ER-localized; consistent with all other localization evidence. Supporting Evidence: file:human/CERS3/CERS3-uniprot.txt SUBCELLULAR LOCATION: Endoplasmic reticulum membrane |
| GO:0008544 epidermis development | IEA GO_REF:0000107 | KEEP AS NON CORE | Summary: Automated ortholog transfer (from mouse Cers3) of a role in epidermis development. CERS3-made ultra-long-chain ceramides are essential for the epidermal permeability barrier, a genuine physiological role downstream of the core molecular activity. Reason: Epidermis development is a real, well-supported downstream role of CERS3-produced ULC-ceramides, but the core function is the ceramide synthase activity. Supporting Evidence: PMID:22038835 are key components of extracellular lipid lamellae |
| GO:0070268 cornification | IEA GO_REF:0000107 | KEEP AS NON CORE | Summary: Automated ortholog transfer (from mouse Cers3) of a role in cornification. Loss of CERS3 produces deficient cornification; a downstream barrier outcome rather than the core molecular function. Reason: CERS3-derived ULC-ceramides contribute to the cornified lipid envelope; retain as a non-core downstream role. Supporting Evidence: PMID:22038835 deficient cornification are hallmarks of mutant skin |
| GO:0030148 sphingolipid biosynthetic process | TAS Reactome:R-HSA-1660661 | KEEP AS NON CORE | Summary: Reactome (TAS) placement of CERS3 in sphingolipid de novo biosynthesis. Correct but broader than the specific ceramide biosynthetic process. Reason: Ceramide biosynthesis is part of sphingolipid biosynthesis; this parent term is accurate but less specific than GO:0046513. Supporting Evidence: Reactome:R-HSA-1660661 Glycosphingolipid biosynthesis is based on salvage of sphingolipids and de novo sphingolipid synthesis. |
| GO:0006665 sphingolipid metabolic process | IEA GO_REF:0000041 | KEEP AS NON CORE | Summary: Automated UniPathway (UPA00222) mapping. Correct but the broadest process term for CERS3; the specific biosynthetic branch is captured by GO:0046513. Reason: Accurate but very general; ceramide biosynthetic process is the more informative term for CERS3. Supporting Evidence: file:human/CERS3/CERS3-uniprot.txt PATHWAY: Lipid metabolism; sphingolipid metabolism. |
| GO:0050291 sphingosine N-acyltransferase activity | TAS Reactome:R-HSA-428185 | ACCEPT | Summary: Reactome (TAS) annotation of the core ceramide synthase / sphingosine N-acyltransferase activity, redundant with the experimental and IBA annotations. Reason: Correctly captures the core catalytic activity of CERS3. Supporting Evidence: Reactome:R-HSA-428185 Ceramide synthase enzymes (CerS, LASS) associated with the endoplasmic reticulum membrane catalyze the reaction of a sphingoid and a long-chain fatty acyl CoA such as stearyl-CoA to form a dihydroceramide and CoASH |
| GO:0005654 nucleoplasm | IDA GO_REF:0000052 | MARK AS OVER ANNOTATED | Summary: HPA immunofluorescence (IDA) annotation to nucleoplasm. This conflicts with the well-established ER membrane localization of CERS3 and with the UniProt CAUTION that the protein is not nuclear. Likely nonspecific antibody signal. Reason: CERS3 is a multi-pass ER membrane enzyme, and UniProt explicitly states it localizes in the ER and not the nucleus; a nucleoplasmic location is not biologically supported. The experimental IDA is not removed. Supporting Evidence: file:human/CERS3/CERS3-uniprot.txt the protein localizes in the endoplasmic reticulum and not in the nucleus, which also argues against homeobox function |
| GO:0005789 endoplasmic reticulum membrane | ISS GO_REF:0000024 | ACCEPT | Summary: Curator sequence-similarity (ISS, from mouse Cers3 Q1A3B0) annotation to ER membrane. This is the precise, correct location of CERS3. Reason: ER membrane is the accurate curated location and the site of CERS3 catalytic activity. Supporting Evidence: file:human/CERS3/CERS3-uniprot.txt SUBCELLULAR LOCATION: Endoplasmic reticulum membrane |
| GO:0050291 sphingosine N-acyltransferase activity | ISS GO_REF:0000024 | ACCEPT | Summary: Curator sequence-similarity (ISS, from mouse Cers3) annotation of the core ceramide synthase activity, redundant with the experimental, IBA, TAS and IEA annotations. Reason: Correctly captures the core catalytic function of CERS3. Supporting Evidence: PMID:17977534 ceramide is synthesized by N-acylation of a sphingoid long-chain base |
| GO:0005783 endoplasmic reticulum | ISS GO_REF:0000024 | ACCEPT | Summary: Curator sequence-similarity (ISS, from mouse Cers3) annotation to ER. Correct but less specific than ER membrane. Reason: CERS3 is ER-resident; consistent with the more precise ER membrane annotation. Supporting Evidence: file:human/CERS3/CERS3-uniprot.txt SUBCELLULAR LOCATION: Endoplasmic reticulum membrane |
| GO:0008544 epidermis development | ISS GO_REF:0000024 | KEEP AS NON CORE | Summary: Curator sequence-similarity (ISS, from mouse Cers3) annotation to epidermis development, consistent with the essential role of CERS3-made ULC-ceramides in the skin barrier. A genuine downstream role, not the core molecular function. Reason: Epidermis development is a real physiological outcome of CERS3 activity but is downstream of its ceramide synthase function. Supporting Evidence: PMID:22038835 are key components of extracellular lipid lamellae |
| GO:0046513 ceramide biosynthetic process | IDA PMID:17977534 Kinetic characterization of mammalian ceramide synthases: de... | ACCEPT | Summary: Direct experimental (IDA) annotation from biochemical characterization of mammalian ceramide synthases, including CERS3, N-acylating sphinganine to form (dihydro)ceramide. Core biological process. Reason: Ceramide biosynthesis is the core process CERS3 participates in; established directly by enzyme assays measuring ceramide synthase activity and K(m) toward sphinganine. Supporting Evidence: PMID:17977534 ceramide is synthesized by N-acylation of a sphingoid long-chain base PMID:17977534 K(m) value of each CerS towards sphinganine |
| GO:0046513 ceramide biosynthetic process | IDA PMID:22038835 Loss of ceramide synthase 3 causes lethal skin barrier disru... | ACCEPT | Summary: Direct experimental (IDA) annotation showing CERS3 is required for synthesis of ultra-long-chain ceramides; CerS3 deficiency abolishes ULC-ceramides. Core biological process. Reason: CERS3 is exclusively required for ultra-long-chain ceramide synthesis, a specialized part of ceramide biosynthesis; strong direct evidence. Supporting Evidence: PMID:22038835 we identified human and mouse ceramide synthase 3 (CerS3), among PMID:22038835 Deficiency of CerS3 in mice results in complete loss of |
| GO:0050291 sphingosine N-acyltransferase activity | IDA PMID:17977534 Kinetic characterization of mammalian ceramide synthases: de... | ACCEPT | Summary: Direct experimental (IDA) determination of ceramide synthase / sphingosine N-acyltransferase activity, including kinetic characterization (K(m) toward sphinganine). This is the core molecular function of CERS3. Reason: Biochemical assays directly demonstrate the N-acyltransferase activity of CERS3, with each CerS showing chain-length-specific acyl-CoA preference. Supporting Evidence: PMID:17977534 specificity towards acyl CoAs of different chain lengths PMID:17977534 K(m) value of each CerS towards sphinganine |
| GO:0050291 sphingosine N-acyltransferase activity | IDA PMID:22038835 Loss of ceramide synthase 3 causes lethal skin barrier disru... | ACCEPT | Summary: Direct experimental (IDA) demonstration that CERS3 has ceramide synthase activity with selectivity for ultra-long-chain acyl-CoAs, required for ULC-ceramide synthesis. Core molecular function. Reason: CERS3 is the ceramide synthase exclusively responsible for ULC-ceramide synthesis in vitro, establishing its N-acyltransferase activity with ultra-long-chain substrate selectivity. Supporting Evidence: PMID:22038835 to be exclusively required for the ULC-Cer synthesis in vitro |
| GO:0070268 cornification | ISS GO_REF:0000024 | KEEP AS NON CORE | Summary: Curator sequence-similarity (ISS, from mouse Cers3) annotation to cornification. CerS3 loss causes deficient cornification; a downstream barrier outcome rather than the core molecular activity. Reason: CERS3-derived ULC-ceramides contribute to the cornified lipid envelope; retain as a non-core downstream role. Supporting Evidence: PMID:22038835 deficient cornification are hallmarks of mutant skin |
| GO:0030216 keratinocyte differentiation | IMP PMID:23754960 Mutations in CERS3 cause autosomal recessive congenital icht... | KEEP AS NON CORE | Summary: Mutant-phenotype (IMP) annotation: CERS3 loss-of-function mutations in humans (ARCI9) disturb the epidermal sphingolipid profile and impair terminal keratinocyte differentiation. Genuine downstream physiological role. Reason: CERS3-produced ceramides are required for proper terminal differentiation of keratinocytes; a real but downstream role, secondary to the core ceramide synthase activity. Supporting Evidence: PMID:23754960 these observations suggest that mutated CERS3 affects the terminal differentiation process in human skin |
| GO:0046513 ceramide biosynthetic process | IMP PMID:23754960 Mutations in CERS3 cause autosomal recessive congenital icht... | ACCEPT | Summary: Mutant-phenotype (IMP) annotation: CERS3 loss-of-function reduces epidermis-specific very-long-chain ceramides, establishing CERS3's role in ceramide biosynthesis in human skin. Core biological process. Reason: Patient keratinocytes carrying CERS3 mutations show a marked decrease in very-long-chain ceramides, directly demonstrating CERS3's involvement in ceramide biosynthesis. Supporting Evidence: PMID:23754960 marked decrease of very long-chain (VLC) ceramides |
| GO:0005789 endoplasmic reticulum membrane | TAS Reactome:R-HSA-428185 | ACCEPT | Summary: Reactome (TAS) annotation to ER membrane, the precise and correct subcellular location of CERS3 and site of its catalytic activity. Reason: ER membrane is the accurate curated location; Reactome describes ceramide synthases as ER-membrane-associated enzymes. Supporting Evidence: Reactome:R-HSA-428185 Ceramide synthase enzymes (CerS, LASS) associated with the endoplasmic reticulum membrane catalyze the reaction of a sphingoid and a long-chain fatty acyl CoA such as stearyl-CoA to form a dihydroceramide and CoASH |
Loading supporting content…
Download this section (compressed HTML)Q: What are the specific in vivo acyl-chain-length preferences of human CERS3, and how do partner proteins such as ORMDL3, fatty acid elongases, and hydroxylases shape the ultra-long-chain acylceramide output in epidermis?
Experiment: Reconstitute human CERS3 in a defined membrane system and measure activity across a panel of acyl-CoA chain lengths with sphinganine and sphingosine acceptors to quantify its ultra-long-chain selectivity.
Experiment: Test whether the CERS3-ORMDL3 interaction detected in HuRI modulates CERS3 activity or sphingolipid flux in differentiating keratinocytes.
Loading supporting content…
Download this section (compressed HTML)Loading supporting content…
Download this section (compressed HTML)