Oncofusion

start / graph / mechanism-nothing-known-makes-cd65s-and-the-enzymes-that-might-are-excluded

Mechanismmoderate confidence 3 of 5: One strong study, or several smaller consistent ones.

CD65s has no known synthesising enzyme, and the three fucosyltransferases nearest to it are explicitly excluded

also called VIM-2 biosynthesis · CD65s enzyme · sialyltransferase bounded absence

CD65s is a sugar structure on the surface of these leukaemia cells, not something a gene makes directly, so the only way to connect it to the rest of this graph is to find the enzyme that builds it. The surface beat resolved two candidate enzymes and deliberately refused to name either, because no paper attributes the structure to them. Checking that properly, the refusal was right, and the nearest enzymes are not merely unproven - a 2018 paper tested three of them and reports that none of them makes it.

The same thing, for a scientist

Q48 resolved ST3GAL4 (Q11206) and ST3GAL6 (Q9Y274) as candidate sialyltransferases for the VIM-2/CD65s epitope and **deliberately did not assert either**, because no source attributes the epitope to them. A validated search confirms that: `"VIM-2" AND (ST3GAL4 OR ST3GAL6)` returns **3 records, none attributing**, against controls of 1,146 and 2,622. And the adjacent question is answered in the negative: **PMID 29593094 reports that neither FT4 (FUT4), FT6 (FUT6) nor FT9 (FUT9) synthesises the internally fucosylated sialyllactosamine VIM-2 (CD65s)** — so the fucosyltransferase route that gives CD15 its gene-keyed bridge is **closed for CD65s by direct test**, not merely unexplored.

Go deeper
The correlation this disciplines

This graph's surface-antigen entries carry a real observation from myeloid maturation: Fuc-TIV (FUT4) mRNA peaks at the HL-60 promyelocyte stage and that peak coincides with upregulation of Lewis-x, VIM-2 and Lewis-y. Those entries are careful — they say correlated with, never synthesises — and that care turns out to be load-bearing.

PMID 29593094 tested the causal reading directly and excluded it. In primary human cells rather than recombinant enzymes on synthetic substrates, it reports that neither FT4, FT6 nor FT9 synthesises VIM-2 (CD65s). So the co-regulation is real and the attribution is not available, which is a sharper position than 'unstudied'.

Why this matters beyond one antigen. CD15 has a gene-keyed bridge — the fucosyltransferases that build Lewis-x — and that bridge is what lets a gene-symbol instrument reach a carbohydrate epitope at all. CD65s does not have one, and now it is known not to have the obvious one. An antigen with no gene and no enzyme is invisible to every symbol-keyed tool in this repository, and that is a property of the antigen rather than a gap in our searching.

A new identifier collision, and it is the worst kind

VIM-2 is the antibody clone that defines CD65s. It is also the Verona integron-encoded metallo-beta-lactamase 2, one of the major carbapenemases in antimicrobial-resistance literature.

A bare "VIM-2" search returns 2,622 records, of which 8 of the top 10 are beta-lactamase or antimicrobial-resistance papers — Pseudomonas aeruginosa co-producing KPC-2 and VIM-2, metallo-beta-lactamase active-site chemistry, tigecycline resistance gene clusters. The haematology clone is buried.

This is a fifth shape for rule 69. The registry already carries: an acronym meaning a complex not a gene (PRC1), a previous official symbol of a different thing (CD15→FUT4), ordinary technical vocabulary (REST, PDF), and a clinical endpoint (EFS). This one is a non-gene antigen's defining ANTIBODY CLONE colliding with a gene in an unrelated field — and it is worse than the others because the collision is not in a gene database at all, so no lookup discipline catches it. Only the search results reveal it.

How this could bite us back
  • CD65s is used diagnostically in this disease, and nothing here changes that. What it changes is the graph's ability to reason about it mechanistically: an antigen with no gene and no identified enzyme cannot be linked to a transcriptional programme, so any claim that the fusion regulates CD65s would have to be made at the level of the antigen, not the pathway.
  • Do not cite FUT4 as the enzyme behind CD65s. The co-regulation in HL-60 is real and the causal reading is excluded by direct test in primary human cells.
Still open
  1. Do FT3, FT5 or FT7 synthesise VIM-2? PMID 29593094 excludes three of the six alpha(1,3)-fucosyltransferases by name; the other three are not addressed in the abstract.
  2. Which sialyltransferase adds the sialic acid to the VIM-2 precursor? Still unattributed.
  3. Is there a glycobiology literature indexed on the structure rather than the CD number or the clone, which these searches cannot reach?
Every source behind this entry (2)
  1. E1strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

    A 2018 study of alpha(1,3)-fucosyltransferase specificity in primary human cells reports that neither FT4, FT6 nor FT9 synthesises the internally fucosylated sialyllactosamine VIM-2, also known as CD65s, while characterising which of the six human alpha(1,3)-fucosyltransferases generate Lewis-X and sialyl Lewis-X.

    Distinct human alpha(1,3)-fucosyltransferases drive Lewis-X/sialyl Lewis-X assembly in human cells · Mondal N, ..., Sackstein R · Journal of Biological Chemistry 2018
    PMID 29593094DOI
    What this evidence cannot show

    Abstract fetched and read back on 2026-09-20; full text not read. **Strong as a negative within the enzymes and cells tested** — the paper's stated advance is using primary human cells rather than recombinant enzymes on synthetic substrates, which is what makes the negative meaningful. **It does not test the sialyltransferases**, so it closes the fucosyltransferase route and says nothing about ST3GAL4 or ST3GAL6. FT3, FT5 and FT7 are not named in the excluded set in the abstract, so the exclusion is of three of six, not all six.

    Evidence that argues against this (1)
    • A negative in primary human cells does not exclude the enzyme acting in a leukaemic blast, where glycosylation machinery and acceptor availability both differ from normal myeloid cells - and the observed co-regulation of FUT4 mRNA with VIM-2 in HL-60 is exactly such a system. The honest position is that the attribution is unavailable, not that the enzyme is irrelevant. Schwartz S et al. 2003
  2. E2moderate confidence 3 of 5: One strong study, or several smaller consistent ones.database

    No published source attributes the VIM-2/CD65s epitope to ST3GAL4 or ST3GAL6: a Europe PMC search for '"VIM-2" AND (ST3GAL4 OR ST3GAL6)' on 2026-09-20 returned 3 records, none of which makes the attribution, against operand controls of 1,146 for ST3GAL4 alone and 2,622 for "VIM-2" alone.

    What this evidence cannot show

    Positive controls run through the identical code path per brief rule 67, so the instrument demonstrably returns results. **The 2,622 control is itself contaminated** — most of it is metallo-beta-lactamase literature — so it demonstrates the query works without bounding the haematology literature well. Only the first page of each result set was read, making this an absence of *findable attribution* rather than proof none exists.

    Evidence that argues against this (1)
    • Biosynthetic attributions for glycan epitopes frequently live in enzymology and glycobiology journals that index the structure rather than the CD number or the clone name, so a search keyed on 'VIM-2' and 'CD65s' may systematically miss the literature that would answer this. The absence is partly a statement about nomenclature. Mondal N et al. 2018
Connected to