Oncofusion

start / graph / gene-mybbp1a

Genemoderate confidence 3 of 5: One strong study, or several smaller consistent ones.Q133-lin28b-fetal-ontogeny

MYBBP1A - the let-7-independent effector through which LIN28B suppresses MLL-rearranged AML, and a nucleolar MYB co-repressor that reads H3K9me2

also called MYB binding protein 1a · p160 · PAP2 · Pol5 · MBB1A_HUMAN

MYBBP1A is a protein that sits on the MYB growth switch and turns it down. LIN28B works by making more of it, and that - not the famous let-7 pathway - is how LIN28B blocks one kind of MLL-rearranged leukaemia in mice.

The same thing, for a scientist

MYBBP1A (HGNC:7546, 17p13.2, ENSG00000132382, UniProt Q9BQG0, 1328 aa) is a large nucleolar protein annotated as a transcriptional co-regulator that may activate or repress transcription through sequence-specific DNA-binding partners, with repression mediated at least in part through histone-deacetylase activity, a role in rRNA biogenesis, and **preferential binding to dimethylated histone H3 Lys-9 (H3K9me2)**; its experimental subcellular locations are cytoplasm, nucleus and **nucleolus** (all ECO:0000269, PubMed 12429849) [E1]. It enters this graph because Eldeeb et al. identified it as the **direct, let-7-independent effector of LIN28B's tumour-suppressive action in MLL-ENL AML**: LIN28B binds MYBBP1A mRNA, MYBBP1A protein rises, MYB signalling is suppressed, leukaemia-initiating-cell activity is compromised; MYBBP1A overexpression **phenocopies** LIN28B and MYBBP1A perturbation **abolishes** its effect [E2]. `gene-myb` already exists here; this is the co-repressor that acts on it.

Go deeper
Why a MYB co-repressor is the interesting part of the LIN28B story

This graph's LIN28B reasoning has been let-7-shaped. The one KMT2A-fusion study that traced a mechanism end to end found a different chain, with both sufficiency and necessity arms:

LIN28B --binds--> MYBBP1A mRNA --raises--> MYBBP1A protein --co-represses--> MYB --> LIC activity down

MYB is not a peripheral node in KMT2A biology. It is a documented dependency across MLL-rearranged leukaemia and gene-myb is already in this corpus. So the Eldeeb result is not an exotic side-mechanism: it is LIN28B acting on the MYB axis this project already tracks, through a protein the project had no entry for.

Two things that follow and are worth stating as facts rather than hopes.

  1. If the same LIN28B-MYBBP1A link operates in the human fetal B-lineage KMT2A::AFF1 system, then the same molecular event is protective in myeloid and permissive in lymphoid. That is a testable, cheap, high-information experiment - a western blot for MYBBP1A in LIN28B-null versus LIN28B-intact CRISPR-KMT2A::AFF1 cells.
  2. Ling et al. report LIN28B stabilising 'epigenetic regulators' in the index fusion. MYBBP1A reads H3K9me2 and is annotated for HDAC-mediated repression, so it would fit that description. Whether it is on their list cannot be checked, because the paper is paywalled and not in PMC.
The nucleolar coincidence, and why it is only a coincidence so far

Both partners in this link are nucleolar: LIN28B's route-1 pool sequesters pri-let-7 in the nucleolus 3, and MYBBP1A's UniProt experimental annotation includes nucleolus alongside nucleus and cytoplasm, with a documented role in rRNA biogenesis 1.

Resist the inference. LIN28B's action on MYBBP1A is reported as binding the mRNA, which is a cytoplasmic or nucleoplasmic event on a transcript, not a protein-protein meeting in the nucleolus. A shared compartment annotation between a regulator and its target's protein product is not evidence of a shared site of action. This graph has a standing worry - recorded in its own proximity-interactome analyses - that nucleolar and ribosome-biogenesis proteins are an AP-MS and BioID contaminant signature, and a nucleolar coincidence should trip that alarm rather than supply a mechanism.

Facts and figures
hgnc IDHGNC:7546
symbolMYBBP1A
ensembl gene IDENSG00000132382
uniprotQ9BQG0
length AA1328
chromosomal location17p13.2
identifier provenanceHGNC REST fetch/symbol/MYBBP1A (numFound 1) and UniProtKB Q9BQG0 fetched live 2026-09-23; PMIDs 36763502, 38321106 and 22118463 fetched from PubMed the same day and their titles read back before citation. data/symbol-collisions.json has no MYBBP1A row (word-boundary check). UniProt's ECO:0000269 source PMID 12429849 was OPENED and read (Scherl et al., Mol Biol Cell 2002, human nucleolar proteome, 213 proteins from HeLa) rather than taken on trust.
subnuclear locationnucleolus - the only one of the three UniProt location calls its cited experiment can support, and it is a mass-spectrometry co-purification from a HeLa nucleolar fraction, not an imaging or fractionation study of MYBBP1A itself
epigenetic rolePreferentially binds H3K9me2 (UniProt, By similarity, from the CRY1/PER2 circadian work) and repression is annotated as at least partly HDAC-mediated. Not demonstrated in a haematopoietic or leukaemic cell.
role in the lin28b axisDirect mRNA target of LIN28B; its protein product is a MYB co-repressor. Sufficiency (overexpression phenocopies) and necessity (perturbation abolishes) both shown in the MLL-ENL mouse.
cell type of the observationmouse MLL-ENL AML and leukaemia-initiating cells
cell statenot a fetal-restricted gene itself - it is the downstream effector of one
shown for fusionKMT2A::MLLT1 (MLL-ENL, mouse, transgenic). NOT shown for KMT2A::AFF1, not shown in any human cell, not shown in any lymphoid cell.
measured in kmt2a aff1false
depmapnot queried by this beat
prev and alias symbols
[
  "P160",
  "PAP2",
  "FLJ37886",
  "Pol5"
]
subcellular location uniprot
[
  "Cytoplasm (ECO:0000269, PubMed 12429849 - CITATION CANNOT SUPPORT THIS CALL, see E1)",
  "Nucleus (ECO:0000269, PubMed 12429849 - CITATION CANNOT SUPPORT THIS CALL, see E1)",
  "Nucleus, nucleolus (ECO:0000269, PubMed 12429849 - a HeLa nucleolar-fraction proteomics catalogue; supports nucleolus only, and as a co-purification not an imaging result)"
]
How this could bite us back
  • MYBBP1A is a MYB co-repressor and MYB is required for normal haematopoiesis. Raising MYBBP1A therapeutically means suppressing MYB in a patient who needs it.
  • MYBBP1A is nucleolar with a role in rRNA biogenesis; perturbing it risks a ribosome-biogenesis phenotype with the p53 stress response that usually accompanies one, which would confound any interpretation of an anti-leukaemic effect.
  • The entire sufficiency-and-necessity chain is in mouse myeloid disease with a transgene at supraphysiological LIN28B. Building an intervention on it for human infant B-ALL would be importing a mechanism across species, lineage, partner and expression regime at once.
Still open
  1. Is MYBBP1A protein raised in LIN28B-intact versus LIN28B-null human fetal liver CRISPR-KMT2A::AFF1 cells? A western blot on existing material. If MYBBP1A rises in a cell where LIN28B is PERMITTING transformation, then the same molecular link has opposite consequences in the two lineages and the lineage - not the link - is the explanation.
  2. Is MYBBP1A among the 'epigenetic regulators' Ling et al. report LIN28B stabilising? Invisible behind the paywall; the authors could be asked directly.
  3. Does MYBBP1A have any dependency signal in KMT2A::AFF1 lines in DepMap? This beat did not query it, and it is a one-call check that would say whether the co-repressor matters in the index disease at all.
  4. Does MYBBP1A's H3K9me2 preference operate in a leukaemic cell? The annotation is 'By similarity' from circadian biology, and this graph already tracks H3K9 methylation at the KMT2A locus for unrelated reasons.
Every source behind this entry (3)
  1. E1established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.database

    MYBBP1A is HGNC:7546 at 17p13.2, ENSG00000132382, UniProt Q9BQG0, 1328 aa; UniProt annotates it as a transcriptional co-regulator acting through sequence-specific DNA-binding proteins with repression mediated at least in part by HDAC activity, a role in rRNA biogenesis, preferential binding to H3K9me2 on the PER2 promoter as a CRY1 co-repressor, and experimental subcellular locations cytoplasm, nucleus and nucleolus (ECO:0000269, PubMed 12429849).

    What this evidence cannot show

    Fetched live 2026-09-23; HGNC numFound was 1, so no symbol ambiguity, and `data/symbol-collisions.json` has no row for MYBBP1A (checked with word-boundary matching per rule 88). **RULE 91 INSTANCE, CHECKED AND FAILING: all three of MYBBP1A's location annotations - cytoplasm, nucleus AND nucleolus - carry the same single ECO:0000269 citation, PubMed 12429849, and that paper is Scherl et al., 'Functional proteomic analysis of human nucleolus' (Mol Biol Cell 2002), a catalogue of 213 proteins identified by mass spectrometry in HeLa nucleolar preparations.** A nucleolar-fraction proteomics paper can support the NUCLEOLUS call and nothing else; it contains no experiment capable of establishing a cytoplasmic or nucleoplasmic pool, and finding a protein in a purified nucleolar fraction is the classic way to acquire a compartment annotation that is really a statement about what co-purifies. So two of MYBBP1A's three experimental location calls rest on a citation that cannot bear them, and the third rests on a bulk fraction rather than an imaging or fractionation experiment on this protein. The founding localisation study for MYBBP1A, whatever it is, is not cited by UniProt. Several functional statements above additionally carry UniProt's 'By similarity' tag rather than a human experiment - notably the HDAC-mediated repression and the H3K9me2 preference, which come from the circadian PER2 work.

    Evidence that argues against this (1)
    • Two of MYBBP1A's three UniProt experimental location calls cite a paper that cannot support them - a HeLa nucleolar-fraction proteomics catalogue. The gene's identity fields are reliable; its localisation record is not, and this entry's nucleolar framing rests on a mass-spectrometry co-purification rather than on any imaging of this protein. Mol Biol Cell 2002
  2. E2strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

    In a dual-transgenic MLL-ENL / LIN28B mouse, LIN28B directly binds MYBBP1A mRNA resulting in elevated MYBBP1A protein; MYBBP1A acts as a MYB co-repressor, suppressing MYB signalling and leukaemia-initiating cell activity; overexpression of MYBBP1A phenocopies the tumour-suppressor effect of LIN28B while perturbing MYBBP1A omits that effect.

    A fetal tumor suppressor axis abrogates MLL-fusion-driven acute myeloid leukemia. · Eldeeb M, Yuan O, Guzzi N, Thi Ngoc PC, Konturek-Ciesla A, Kristiansen TA, Muthukumar S, Magee J, Bellodi C, Yuan J, Bryder D · Cell Rep 2023
    PMID 36763502DOI
    What this evidence cannot show

    Abstract read; full text not opened. MOUSE, transgenic MLL-ENL (KMT2A::MLLT1 analogue), MYELOID disease, LIN28B supplied by transgene at levels a companion paper from the same collaboration says exceeded physiological ones. The binding assay behind 'directly binds' is not named in the abstract. The paper's own abstract calls the influence of prenatal signals on leukaemogenesis 'controversial'. No measurement in KMT2A::AFF1, in a lymphoid cell, or in a human cell.

    Evidence that argues against this (1)
    • The necessity arm for the whole fetal-protection framing failed in the companion study: conditional Lin28b deletion in fetal progenitors did NOT accelerate MLL::ENL AML initiation, so whatever MYBBP1A does downstream of LIN28B, LIN28B is not the effector of fetal leukaemia suppression. Leukemia 2024
  3. E3established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper

    LIN28B's let-7-repressive activity is nucleolar: it sequesters primary let-7 transcripts in the nucleolus away from the Microprocessor, by a ZCCHC11-independent mechanism distinct from LIN28A's cytoplasmic TUTase route.

    Lin28A and Lin28B inhibit let-7 microRNA biogenesis by distinct mechanisms. · Piskounova E, Polytarchou C, Thornton JE, LaPierre RJ, Pothoulakis C, Hagan JP, Iliopoulos D, Gregory RI · Cell 2011
    PMID 22118463DOI
    What this evidence cannot show

    Abstract read. Cited here ONLY to show that the shared nucleolar annotation between LIN28B and MYBBP1A is a coincidence of compartment, not an observed co-localisation or contact. No study has reported LIN28B and MYBBP1A protein in the same compartment in the same cell.

    Evidence that argues against this (1)
    • The founding characterisation paper reports LIN28B as PREDOMINANTLY CYTOPLASMIC at steady state, so the nucleolar pool invoked here is a minority pool. Two proteins each having a minority nucleolar annotation is very weak ground for any shared-compartment reasoning, which is why this entry declines to draw one. Gene 2006
Connected to