The only InterPro entry matched by ADAMTSL1 that carries a GO mapping is IPR013273 (ADAMTS/ADAMTS-like), and it maps to GO:0030198 extracellular matrix organization alone. Although the signature spans the catalytic ADAMTS metalloproteases, the mapping does not include any peptidase or hydrolase term, so no catalytic activity is propagated to the non-catalytic members of the family through this route.

Reaction of the O-glycosylation of TSR domain-containing proteins pathway in which B3GLCT, an ER membrane enzyme, adds glucose to the O-fucose already present on thrombospondin type-1 repeats. ADAMTSL1 is a member of the O-fuc-proteins participant set, which Reactome places in the endoplasmic reticulum lumen.

Reaction in which the ER-resident enzyme POFUT2 attaches fucose to the conserved Ser or Thr of thrombospondin type-1 repeats. ADAMTSL1 is a member of the POFUT2 substrates participant set, placed by Reactome in the endoplasmic reticulum lumen.

A Reactome FailedReaction modelling the Peters-plus syndrome disease state, in which the B3GALTL G393E variant fails to glucosylate O-fucosylated thrombospondin repeats. ADAMTSL1 appears as an input to a reaction that by construction does not proceed, yet the GO pipeline emits the same endoplasmic reticulum lumen location as for the working reaction.

Punctin/ADAMTSL-1 is a secreted glycoprotein resembling the ADAMTS proteases but lacking the pro-metalloprotease and disintegrin-like domains, and containing four thrombospondin type-1 repeats in the short form that was purified.

Recombinant punctin expressed in transfected COS-1 cells is deposited into the cell substratum in a punctate pattern and is excluded from focal contacts. This is the observation that gave the protein its name and the primary human evidence that it reaches the extracellular matrix.

The gene is expressed in adult skeletal muscle, and the authors present punctin as the founding member of the ADAMTS-like family whose matrix function was at that point prospective rather than demonstrated.

The thrombospondin repeats of punctin-1 carry POFUT2-type O-fucose, extended in TSR2, TSR3 and TSR4 to a fucose-glucose disaccharide, and mutating the modified Ser/Thr residues reduces the amount of protein secreted. This is the experimental basis for ADAMTSL1's participation in the two Reactome ER-lumen glycosylation reactions.

N-glycosylation and O-fucosylation together are required for efficient biosynthesis of the protein.

Trp39 and Trp42 of the first thrombospondin repeat carry C-linked mannose, adjacent to the O-fucose site, and substituting these tryptophans changes secretion efficiency. The authors conclude that the two modifications together act as a quality-control mechanism on export.

Mass spectrometry of recombinant human punctin-1 identified the TSR1 peptide spanning residues 36-49 bearing two mannose residues and a Glc-Fuc disaccharide.

Endogenous ADAMTSL1 was identified in human fibroblast secretomes by terminal amine isotopic labelling of substrates and shown to be a direct MMP10 substrate. This places native, untagged ADAMTSL1 in the extracellular compartment of a primary human cell type and shows it is subject to extracellular proteolysis rather than performing it.

The ADAMTS-like proteins are defined as comprising the ADAMTS ancillary domains while lacking the protease domain, and therefore lacking enzymatic activity.

Full-length ADAMTSL1 has thirteen thrombospondin type-1 repeats, four Ig-like C2-type domains and one PLAC domain over 1762 residues, and punctin-1 is the 525-residue short splice variant with four repeats.

The group that discovered the protein states that its tissue function is unknown and that a role in matrix assembly and turnover is an expectation drawn from other family members rather than an ADAMTSL1 result.

A heterozygous p.Trp42Arg allele at a C-mannosylation site cosegregates with a multisystem phenotype; the mutant protein is retained intracellularly and also reduces secretion of cotransfected wild-type protein.

All of the functional work was done on the short punctin-1 form because no full-length expression construct existed, which is why the experimental coverage of this gene is confined to isoform Q8N6G6-1.

MADD-4, the C. elegans protein most closely related to ADAMTSL1 and ADAMTSL3, is a secreted guidance cue acting through the netrin receptor UNC-40/DCC. The authors explicitly decline to extend a biological role to the mammalian orthologues.

Two rare ADAMTSL1 missense variants segregate with mandibular prognathism in Thai families, and mouse Adamtsl1 is expressed in the condensed mesenchyme of the condyle but not in long-bone cartilage. The mechanism the authors propose requires ADAMTSL1 to cleave aggrecan, an activity the protein's domain content does not support, and they label it a hypothesis.

Adamtsl1 is used as one of three transcriptional markers defining a myelinating Schwann cell subtype in mouse peripheral nerve that preferentially ensheathes motor axons. The gene's own function is not assayed.

Surface plasmon resonance with recombinant human ADAMTSL-1, -2 and -3 against the N-terminal half of fibrillin-1 found that ADAMTSL-2 and -3 bound and ADAMTSL-1 did not. This is a direct, measured negative for ADAMTSL1 against the partner through which the other ADAMTS-like proteins reach the extracellular matrix organization term.

The negative covers fibrillin-1 in its entirety, not only the N-terminal half: binding to the C-terminal half was negative for every ADAMTSL protein in the panel. Held to a lower evidentiary standard than the N-terminal result, though: it is reported as data not shown, with no figure or table behind it, whereas the N-terminal negative is reported as an outcome of a panel that is displayed - Figure 5a shows sensorgrams for the binders only, so the paper contains no ADAMTSL-1 trace for either half.

In the same study ADAMTSL-1 also failed to bind the non-catalytic C-terminal region of ADAMTS-10, which ADAMTSL-3 binds with high affinity. This result is corroborating rather than discriminating and must not be added to the fibrillin-1 one: ADAMTSL-2 is equally negative against ADAMTS-10 and still reaches GO:0030198, ADAMTS-10 binding was shown only for ADAMTSL-3, and ADAMTSL-6/THSD4 was never tested against it, so ADAMTS-10 is not what places ADAMTSL-2 or -6 in the fibrillin-1 pathway. The discriminating result is the fibrillin-1 negative alone.

ADAMTSL1 was identified as a Hedgehog-responsive gene in primary chondrosarcoma xenografts treated with the SMO inhibitor IPI-926, and manipulating it altered chondrosarcoma cell proliferation. This is the only gene-specific functional perturbation result reported for ADAMTSL1, and Apte is a co-author.

Ablating the Schwann cell population co-marked by Adamtsl1, Cldn14 and Pmp2 causes loss of large-caliber motor axons. The perturbation is of the cell population, not of the gene, so the phenotype is a property of the cell type.

The feature table over the 1762-residue precursor lists a signal peptide, nine annotated thrombospondin type-1 repeats, four Ig-like C2-type domains and a PLAC domain, with no metalloprotease domain, no disintegrin-like domain, no prodomain and no zinc-binding site. The entry states the absence explicitly and records no catalytic activity and no EC number.

Despite that, the entry carries the molecular-function keyword Hydrolase (KW-0378), which is what generates the GO:0016787 hydrolase activity cross-reference in the entry's own GO list. This annotation is not present in GOA, because keyword-derived annotations were withdrawn for cellular organisms.

The keyword-derived hydrolase activity term appears in the entry's own GO cross-reference block with evidence IEA:UniProtKB-KW.

The entry records that PAN-GO assigns no annotations to this protein from evolutionary models, consistent with the complete absence of IBA rows in GOA.

The ADAMTS spacer region named in the description is called by Pfam and InterPro rather than by a UniProt FT DOMAIN line, which is why it does not appear in the feature-table enumeration above.

Tissue expression is highest in adult skeletal muscle.

Of the 26 human members of PANTHER family PTHR13723, 24 receive GO:0031012 extracellular matrix by IBA from node PTN000347317. PAPLN does not, but holds the term from its own evidence. ADAMTSL1 is the only human member of the family with no GO:0031012 annotation from any evidence code, while its mouse orthologue - same subfamily SF157 - does receive the IBA from the same node.

Within the ADAMTS-like branch, ADAMTSL1 and THSD4 carry the Hydrolase keyword while ADAMTSL2, ADAMTSL3, ADAMTSL4 and ADAMTSL5 do not, and none of the six has a catalytic activity comment.

PAINT records IKR and IRD loss calls at node PTN002673039 that block metalloendopeptidase activity and proteolysis from propagating down the ADAMTS-like branch, but in GOA the loss reaches only ADAMTSL2. ADAMTSL1 receives neither the catalytic terms nor the negation.

PAINT records an inferred-from-key-residues loss of metalloendopeptidase activity at node PTN002673039, negating the term that the family node PTN000347317 carries. A companion inferred-from-rapid-divergence row does the same for proteolysis.

The family node PTN000347317 carries GO:0031012 extracellular matrix as an IBD annotation, which is the source of the IBA rows that 24 of 26 human family members receive and ADAMTSL1 does not.

The provider record passed its own gates (gates_passed True, faith 100 percent, ten numeric PMIDs, no preprint identifiers in PMID-shaped fields) and its central framing - a secreted matrix glycoprotein without the catalytic domains, with no demonstrated enzymatic activity - agrees with UniProt and with the primary literature.