CADM1 (cell adhesion molecule 1; SynCAM1, nectin-like protein 2, TSLC1) is a single-pass type I transmembrane immunoglobulin-superfamily protein of the nectin-like (Necl) family with three extracellular Ig-like domains (one V-type, two C2-type). It mediates Ca(2+)-independent homophilic cell-cell adhesion and heterophilic adhesion with other nectin-like proteins. Binding of its V-type Ig domain to the receptor CRTAM on activated NK cells and CD8+ T cells makes CADM1-expressing cells targets of NK-cell cytotoxicity and promotes CD8+ T-cell cytokine responses. In neurons it acts as a synaptic adhesion molecule that drives synapse assembly, and it supports epithelial, mast-cell and spermatogenic cell adhesion. This entry is one of eight Ensembl products of the donkey CADM1 gene. It keeps the complete extracellular region, a short stalk and the transmembrane helix, but ends in an 11-residue alternative tail. That tail replaces the cytoplasmic region that binds protein 4.1 and the C-terminal PDZ-binding motif found in the canonical 473-residue product.
Summary: CADM1 binds the immune receptor CRTAM through its V-type Ig domain, acting as a ligand that triggers NK and CD8+ T cell responses.
Reason: Human CADM1 has IPI evidence for CRTAM binding (PMID:15811952), confirmed independently in PMID:15781451, and human CADM1 also receives it by IBA from PAINT node PTN002913382. The donkey row is a TreeGrafter transfer from PTN000589348 within the CADM1 subfamily (PTHR45889:SF2). The CRTAM-binding V-type Ig domain lies in the extracellular region, which this donkey product retains intact.
Summary: CADM1 is a single-pass type I plasma membrane protein; this donkey product has a predicted signal peptide and one transmembrane helix.
Reason: Cell-surface localization is experimentally established for human CADM1 and is inherent to its roles as an adhesion molecule and CRTAM ligand. The donkey product retains the signal peptide and transmembrane helix, so it is expected to reach the cell surface despite its altered cytoplasmic tail.
Summary: Homophilic, Ca(2+)-independent cell-cell adhesion through V-type Ig domain dimerization is the defining activity of CADM1.
Reason: Established experimentally for rodent SynCAM1/CADM1 (PMID:12202822) and human CADM1. PAINT placed the term at a deeper node shared with CADM3 (PTN001910345), so it is a conserved family function. The adhesive ectodomain is fully retained in this donkey product.
Summary: CADM1 is the cell-surface molecule by which activated NK and CD8+ T cells recognize CADM1-expressing target cells through CRTAM.
Reason: Human CADM1 carries this term by IDA from PMID:15811952 and by IBA from PAINT node PTN002913382. CADM1 takes part as the recognized partner, not as the detecting cell. The specific role is captured by the accepted susceptibility to NK cell mediated cytotoxicity and signaling receptor binding rows, so this broad term is kept as non-core.
they reveal Necl-2 as one of the molecular targets that allows the immunosurveillance network to distinguish tumor cells from normal cells
GO:0042271 susceptibility to natural killer cell mediated cytotoxicity
IEA GO_REF:0000118
ACCEPT
Summary: CADM1 on target cells engages CRTAM on NK cells, promoting NK-cell cytotoxicity and rejection of CADM1-expressing tumors.
Reason: The term describes the target-cell role of CADM1 (human IDA, PMID:15811952), and human CADM1 also receives it by IBA from PAINT node PTN002913382. The CRTAM-binding V-type domain is retained in this donkey product.
CRTAM-Necl-2 interactions promote cytotoxicity of NK cells and interferon gamma (IFN-gamma) secretion of CD8+ T cells in vitro as well as NK cell-mediated rejection of tumors expressing Necl-2 in vivo
Summary: CADM1 is present on axons and dendrites of neurons, where it acts in synapse formation and axon fasciculation.
Reason: Human CADM1 receives this location by IBA (PTN002913382) from rodent data, and it holds for the CADM1 gene. This donkey product lacks the cytoplasmic region that binds protein 4.1 and PDZ scaffolds. The canonical donkey product A0A9L0IJE9 keeps that region. Whether this short-tail product is targeted to neurites is therefore untested, and the term is kept as a non-core location.
It then ends in an 11-residue alternative tail (LFSLTSSPRIK) in place of the 38-residue cytoplasmic tail that carries the protein 4.1-binding region and the C-terminal EYFI PDZ-binding motif.
Summary: CADM1 (SynCAM1) is a synaptic adhesion molecule.
Reason: Synaptic localization and function are established for rodent SynCAM1 (PMID:12202822), and human CADM1 receives this location by IBA (PTN002913382). The same study showed that the cytoplasmic tail is functionally important at synapses, since the isolated tail inhibited synapse assembly. This donkey product replaces that tail with an 11-residue alternative sequence, so the term is kept as a non-core location for this product.
It then ends in an 11-residue alternative tail (LFSLTSSPRIK) in place of the 38-residue cytoplasmic tail that carries the protein 4.1-binding region and the C-terminal EYFI PDZ-binding motif.
Summary: CADM1 is the ligand recognized by CRTAM; the detecting entity is the NK or T cell carrying CRTAM, not CADM1.
Reason: Human CADM1 carries this term by IDA from PMID:15811952, and by IBA from PAINT node PTN002913382, so it is not removed. In that study, however, the receptor-bearing NK and CD8+ T cells detect Necl-2. CADM1 is what is detected. Its role is captured precisely by the accepted susceptibility to natural killer cell mediated cytotoxicity and signaling receptor binding rows. As a process that CADM1 itself performs, detection of stimulus is an over-annotation.
ISS PMID:12202822 SynCAM, a synaptic adhesion molecule that drives synapse ass...
NEW
Summary: CADM1 on one cell binds CADM1 (or other nectin-like proteins) on an adjacent cell through its V-type Ig domain, directly mediating cell-cell adhesion.
Reason: This is the molecular activity that CADM1 itself performs in the accepted homophilic cell-cell adhesion process. Donkey GOA has the process but no corresponding molecular function. The adhesive ectodomain is 96% identical to human CADM1, and the donkey protein sits in the CADM1 PANTHER subfamily (PTHR45889:SF2).
Q: Is the 11-residue donkey CADM1 tail of A0A9L0IX39 (ending LFSLTSSPRIK) a genuine alternative terminal exon or an intron read-through in the Ensembl model? The canonical donkey product A0A9L0IJE9 and several other donkey products keep the EYFI PDZ-binding tail. Is the short-tail product expressed at the protein level?
Q: Does a CADM1 product without the protein 4.1- and PDZ-binding cytoplasmic region still localize to synapses and support synapse assembly, or does it act only in trans as an adhesion partner and CRTAM ligand?
Suggested Experiments
Experiment: RT-PCR or long-read RNA-seq across the CADM1 3' exons in donkey brain and lung, to test whether the LFSLTSSPRIK terminal sequence is spliced and expressed.
External Prediction Reviews
These computational predictions are reviewed separately from the GOA annotation set used for this review. The assessments below are from this project and do not constitute official GO annotations or endorsement by GO/UniProt. They are not included in the existing annotation review above.
Assessment of the single ProtNLM2 GO prediction for donkey CADM1 (A0A9L0IX39), a single-pass type I transmembrane Ig-superfamily adhesion molecule 96% identical to human CADM1 over its extracellular and transmembrane regions. The prediction membrane (GO:0016020) is correct but less precise than the donkey GOA plasma membrane annotation (LSP). ProtNLM2 emitted no function text or subcellular location for this accession.
LSP — Less precise than existing annotation Review score: 2/2
Prediction method: ProtNLM2 · Version: UniProt API snapshot 2026-10-14
Review rationale: Human CADM1 is experimentally localized to the cell membrane as a single-pass type I protein, and the donkey sequence conserves the signal peptide, Ig-like ectodomain and transmembrane helix, so membrane localization is well supported. Donkey GOA already holds the more specific plasma membrane (GO:0005886) annotation, which is also the GO term named in the ProtNLM2 evidence record. The prediction is correct but less precise than an existing annotation.