Oncofusion

start / graph / gene-bcl2

Proteinstrong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.Q5c-interactome2

BCL2 — the one co-dependency in this beat that already has an approved drug pointed at it

also called BCL2 · Bcl-2 · apoptosis regulator Bcl-2 · B-cell lymphoma 2

BCL2 is the brake on the cell's self-destruct switch, and KMT2A::AFF1 leukemias turn it up, which is why they die when you block it. Venetoclax already blocks it and is approved, so this is the shortest path from a dependency measurement in this project to a medicine a child could actually be given.

The same thing, for a scientist

BCL2 (P10415, 239 aa) is the prototypical anti-apoptotic BCL-2-family protein and the target of venetoclax. In KMT2A-rearranged ALL it is directly upregulated by the fusion through DOT1L-dependent H3K79me2/3, making it a mechanistic output of the fusion rather than a coincidental dependency. It correlates with MEN1 at r = 0.43 across 54 leukemia lines and its most depleted leukemia lines are strongly KMT2A-enriched. The major caveat is lineage-specific: in AML, monocytic differentiation loses BCL2 expression and switches to MCL1, and KMT2A-rearranged AML is frequently monocytic.

loading 6O0K…
PDB 6O0K
Venetoclax sits in the same groove that BCL2 normally uses to hold the cell's self-destruct proteins captive.
Tell me more about this picture

6O0K: human BCL-2 with venetoclax, 1.62 A X-ray.

Tell me more about this picture

The BH1, BH2 and BH3 motifs together line the hydrophobic groove; 2XA0 shows the natural BAX BH3 helix occupying the same surface at 2.70 A.

Tell me more about this picture

Three generations of chemistry have targeted this one groove: ABT-737 analogues (6QGG), navitoclax (4LVT) and venetoclax (6O0K), with selectivity over BCL-XL being the property that made venetoclax tolerable.

What it looks like
BCL2239 aa
BH4 motifBH4 motif · residues 10–30Disordered loopDisordered loop · residues 39–85BH3 BH3 motif · residues 93–107BH1 motifBH1 motif · residues 136–155BH2 BH2 motif · residues 187–2020100200
loading AF-P10415-F1…
AlphaFold AF-P10415-F1 ↗
colourshape

This is a prediction, not a measurement. No one has seen this protein in this shape. The colours say how much the software trusts itself, and that is worth reading before anything else — the orange stretches are places the model is effectively saying this part has no fixed shape.

very high (90+)confident (70–90)low (50–70)very low (<50)
I have never read one of these pictures before

A protein is one long chain of amino acids that folds up. The ribbon traces that chain from one end to the other; it is not a wire, it is a simplification of where the backbone runs. Drag to rotate, scroll to zoom.

Domains are the stretches of chain that do a particular job, and they are what the coloured blocks above the picture mark out. Click one and the view flies to it. In this project the important question about any domain is usually whether the fusion still has it — the labels say lost where it does not.

Anything in yellow sticks is a bound small molecule, which is usually the drug.

Go deeper
Why this is the most actionable entry in the whole beat

Everything else in this domain is either pan-essential, structurally undruggable, or both. BCL2 is neither. It is a small, well-folded, deeply characterised protein with 55 PDB entries, an FDA-approved inhibitor, and a co-crystal structure of that inhibitor bound at 1.62 Å (6O0K) 4.

And the link to this disease is mechanistic, not correlative. t(4;11) patient cells express high BCL2 and are highly sensitive to venetoclax; MLL/AF4 specifically upregulates BCL2 — and not other BCL-2-family members — through DOT1L-mediated H3K79me2/3; a t(4;11) line is sensitive to venetoclax combined with a DOT1L inhibitor; and venetoclax synergises with standard induction therapy in a xenotransplant model 2. That is a chain running from the fusion, through an epigenetic mark this graph already has an entry for, to an approved drug.

The DepMap data agree. BCL2's mean Chronos across 1,258 lines is -0.036, with only 3.3% of lines depleted — it is not a general dependency at all — yet across 57 leukemia lines it falls to -0.351, and the most depleted set is KMT2A-heavy 1:

LineLesionChronos
OCI-AML2NPM1-mutant AML-1.457
EOL-1FIP1L1::PDGFRA AML-1.238
SEMKMT2A::AFF1-0.993
HB11;19KMT2A::MLLT1-0.896
JM1B-ALL-0.876
MOLM-14KMT2A::MLLT3-0.797
MV4-11KMT2A::AFF1-0.773
THP-1KMT2A::MLLT3-0.628
MOLM-13KMT2A::MLLT3-0.625

Six of the nine are KMT2A-rearranged.

The counter-evidence, which is specific and lineage-shaped

The strongest argument against BCL2 in this disease is not that BCL2 is unimportant, but that the cells most likely to survive venetoclax are exactly the ones KMT2A-rearranged leukemia tends to produce.

In AML, response to venetoclax plus azacitidine tracks developmental stage: phenotypically primitive AML is sensitive, monocytic AML is resistant. Resistant monocytic AML has a distinct transcriptome, loses BCL2 expression, and relies on MCL1 for oxidative phosphorylation and survival. That differential sensitivity drives selective outgrowth of monocytic subpopulations at relapse 3.

KMT2A-rearranged AML is frequently monocytic or myelomonocytic, and this graph already carries an entry on lineage switch in KMT2A-r disease. Put those together and venetoclax in KMT2A-r leukemia has a built-in escape route that is not a mutation and not a slow adaptation — it is a differentiation state the disease already visits. Worse, a menin inhibitor works by inducing differentiation, which is the direction that turns BCL2 off and MCL1 on. A menin-inhibitor/venetoclax combination may therefore be antagonistic in sequence even if synergistic in a dish, and that is a testable prediction rather than a rhetorical worry.

Tolerability: the mouse knockout is a real warning

Bcl2-null mice complete embryonic development but show growth retardation and early postnatal mortality. Haematopoiesis is initially normal, then thymus and spleen undergo massive apoptotic involution. The mice develop severe polycystic kidney disease with renal failure, and turn grey at the second hair-follicle cycle 5.

The lymphoid involution and the renal phenotype both matter here. Venetoclax's known clinical toxicities — cytopenias, neutropenia, tumour lysis syndrome — are recognisably the pharmacological echo of the first; the polycystic kidney phenotype has no clear clinical counterpart and appears to be developmental. In an infant, developmental phenotypes deserve more weight than they get from adult trial safety data.

Human constraint. pLI 0.995, LOEUF 0.333, with 1 observed loss-of-function variant against 14.2 expected 1. As with DPY30, the expected count is small for a 239-residue protein, so this LOEUF is weakly powered. The project's triage placed BCL2 AMBER on LOEUF 0.333; the honest reading is AMBER-with-low-confidence rather than a firm constraint call.

Expression. Thymus 17.0, spleen 16.9, lymph node 13.7, bone marrow 4.7 nTPM; classical monocytes register 0.0 1. That last number is the resistance mechanism visible in normal tissue: monocytes simply do not run on BCL2.

Structure and chemical matter

BCL2 is 239 residues with four BH motifs — BH4 at 10–30, BH3 at 93–107, BH1 at 136–155, BH2 at 187–202 — and a disordered loop at 39–85 6. The BH1–BH3 motifs together form the hydrophobic groove that binds the BH3 helices of pro-apoptotic partners, and that groove is the venetoclax site.

Verified structures spanning the relevant states 4: 2XA0 (BCL-2 with a BAX BH3 peptide, 2.70 Å) shows the natural ligand in the groove; 6O0K (venetoclax, 1.62 Å), 4LVT (navitoclax, 2.05 Å) and 6QGG (an ABT-737 analogue, 1.50 Å) show three generations of chemistry occupying the same groove.

This is the one place in this beat where a ChEMBL count would be meaningful, because the pocket is single and well-defined. We did not run one, and per this project's standing rule we do not quote a target-level number in its place.

Facts and figures
uniprotP10415
gene symbolBCL2
ensemblENSG00000171791
length AA239
alphafold IDAF-P10415-F1
cytoband18q21.33
PDB count55
roleanti-apoptotic output of the KMT2A fusion, upregulated via DOT1L-dependent H3K79me2/3; not a fusion-complex subunit
druggabilityExcellent and already realised. Venetoclax is FDA-approved and co-crystallised at 1.62 A (6O0K). This is the only entry in this beat where a pocket-level ChEMBL count would be meaningful; we did not run one and do not quote a target-level number in its place.
subnuclear localizationnuclear envelope only, in a patchy pattern the founding study likens to nuclear pore complexes; no nucleoplasmic or nuclear-body BCL2 has been reported
shown for fusionNO - the founding localisation study is in a t(14;18) lymphoma line and no study has localised BCL2 in a KMT2A-rearranged cell. This matters more than usual here: venetoclax is used in KMT2A-rearranged disease on the strength of BCL2 dependency measured by BH3 profiling and killing, and BH3 profiling reads the mitochondrial pool only.
domains
BH4 motif10.0BH4 motif: 10Disordered loop39.0Disordered loop: 39BH3 motif93.0BH3 motif: 93BH1 motif136.0BH1 motif: 136BH2 motif187.0BH2 motif: 187
most depleted leukemia lines
OCI-AML2-1.457OCI-AML2: -1.457EOL-1-1.238EOL-1: -1.238SEM-0.993SEM: -0.993HB11;19-0.896HB11;19: -0.896JM1-0.876JM1: -0.876MOLM-14-0.797MOLM-14: -0.797MV4-11-0.773MV4-11: -0.773THP-1-0.628THP-1: -0.628MOLM-13-0.625MOLM-13: -0.625
PDB verified
[
  "2XA0",
  "4LVT",
  "6O0K",
  "6QGG"
]
domains — the numbers behind the chart
[
  {
    "name": "BH4 motif",
    "start": 10,
    "end": 30,
    "pfam": null
  },
  {
    "name": "Disordered loop",
    "start": 39,
    "end": 85,
    "pfam": null
  },
  {
    "name": "BH3 motif",
    "start": 93,
    "end": 107,
    "pfam": null
  },
  {
    "name": "BH1 motif",
    "start": 136,
    "end": 155,
    "pfam": null
  },
  {
    "name": "BH2 motif",
    "start": 187,
    "end": 202,
    "pfam": null
  }
]
essentiality
{
  "depmap_common_essential": false,
  "chronos_mean": -0.0364,
  "chronos_mean_leukemia_lines": -0.3505,
  "n_lines": 1258,
  "n_leukemia_lines": 57,
  "fraction_lines_depleted": 0.0326,
  "leukemia_selectivity": -0.0917,
  "gnomad_pLI": 0.995,
  "gnomad_loeuf": 0.333,
  "gnomad_obs_lof": 1,
  "gnomad_exp_lof": 14.2,
  "gnomad_power_note": "Only 14.2 expected LoF variants; LOEUF weakly powered for a 239-residue protein.",
  "mouse_ko": "Bcl2-null mice survive embryogenesis but show growth retardation, early postnatal mortality, massive apoptotic involution of thymus and spleen, severe polycystic kidney disease with renal failure, and hair greying",
  "hpa_ntpm": {
    "thymus": 17,
    "spleen": 16.9,
    "lymph_node": 13.7,
    "heart_muscle": 5,
    "bone_marrow": 4.7,
    "naive_B_cell": 3.1,
    "liver": 1.9,
    "classical_monocyte": 0
  },
  "hpa_specificity": "Low tissue specificity"
}
codependency men1
{
  "r_all_lines": 0.1844,
  "r_leukemia_lines": 0.4289,
  "ci_leukemia": [
    0.1476,
    0.6302
  ],
  "p_permutation": 0.0012,
  "p_holm": 0.0228,
  "n_leukemia_lines": 54
}
most depleted leukemia lines — the numbers behind the chart
[
  {
    "line": "OCI-AML2",
    "lesion": "NPM1-mutant AML",
    "chronos": -1.457
  },
  {
    "line": "EOL-1",
    "lesion": "FIP1L1::PDGFRA AML",
    "chronos": -1.238
  },
  {
    "line": "SEM",
    "lesion": "KMT2A::AFF1",
    "chronos": -0.993
  },
  {
    "line": "HB11;19",
    "lesion": "KMT2A::MLLT1",
    "chronos": -0.896
  },
  {
    "line": "JM1",
    "lesion": "B-ALL",
    "chronos": -0.876
  },
  {
    "line": "MOLM-14",
    "lesion": "KMT2A::MLLT3",
    "chronos": -0.797
  },
  {
    "line": "MV4-11",
    "lesion": "KMT2A::AFF1",
    "chronos": -0.773
  },
  {
    "line": "THP-1",
    "lesion": "KMT2A::MLLT3",
    "chronos": -0.628
  },
  {
    "line": "MOLM-13",
    "lesion": "KMT2A::MLLT3",
    "chronos": -0.625
  }
]
localization
{
  "subcellular": "integral membrane protein of the OUTER MITOCHONDRIAL MEMBRANE, the endoplasmic reticulum (preferentially light-density, smooth microsomes) and the NUCLEAR ENVELOPE - not one compartment but three, established together in the founding fractionation-plus-microscopy study and never since reduced to one. Absent from cytosol in that study.",
  "subnuclear": "the only subnuclear element is the nuclear ENVELOPE, a membrane rather than a nucleoplasmic body: anti-Bcl-2 immunoreactivity overlies the nuclear envelope in a patchy high-magnification distribution the authors describe as reminiscent of nuclear pore complexes. There is no nucleoplasmic, nuclear-body or chromatin-associated BCL2 pool in any source read for this beat.",
  "how_shown": "subcellular fractionation on Nycodenz and Percoll continuous gradients plus discontinuous sucrose gradients, laser-scanning microscopy, pre- and post-embedding immuno-electron microscopy with SIX different anti-Bcl-2 antibodies, and in vitro import assays on isolated mitochondria (PMID 8402648)",
  "systems": "a t(14;18)-containing lymphoma cell line - a B-lineage malignancy driven by a different oncogenic mechanism (IGH-BCL2 juxtaposition), NOT a KMT2A-rearranged cell",
  "node": "localization-kat8-mof-has-a-mitochondrial-pool"
}
depmap kmt2a selectivity
{
  "chronos_mean_kmt2a_rearranged": -0.6657,
  "chronos_mean_other_leukaemia": -0.2698,
  "difference": -0.3959,
  "pan_essentiality_all_lines": -0.0362,
  "selectivity_rank": 36,
  "rank_denominator": 17728,
  "provenance": "Joined 2026-09-21 from analyses/depmap-kmt2a-selective-scan-full (generated 2026-09-19T10:12:06Z), the project's own genome-wide scan over 11 KMT2A-rearranged and 43 other leukaemia DepMap lines. NOT re-fetched and NOT independent of that scan: every caveat it carries applies here, including that the KMT2A-rearranged arm is a small number of lines drawn from nine individuals, that the rank denominator is 17,728 genes, and that a difference is not a dependency. Added because only 26 of 308 gene entries carried this number, which made any KMT2A-selectivity question answerable for 8% of the corpus.",
  "how_to_read": "A negative difference means MORE required in KMT2A-rearranged lines than in other leukaemia lines. A rank near 17,728 is ANTI-selective. Neither is evidence of a fusion-driven dependency on its own: check pan_essentiality_all_lines before calling anything selective, and note that a paralogue-redundant gene can score near zero for reasons that have nothing to do with this disease - see ana-the-paralogue-redundancy-excuse-is-not-tested-and-my-proxy-decided-the-answer."
}
How this could bite us back
  • Compensation and lineage escape: monocytic cells lose BCL2 and switch to MCL1, so venetoclax selects for exactly the differentiation state KMT2A-rearranged disease is prone to occupying.
  • Combination antagonism: a menin inhibitor works by inducing differentiation, which is the direction that turns BCL2 off. The most natural combination in this graph may be self-defeating if given in the wrong order.
  • Essential-function toxicity: Bcl2-null mice suffer massive apoptotic involution of thymus and spleen, which is the developmental version of the cytopenias venetoclax already causes, in a patient who is building an immune system.
  • Tumour lysis: venetoclax's on-mechanism toxicity scales with how well it works, and hyperleukocytosis is already an emergency in this population.
  • Interpretive risk: BCL2 is a downstream output of the fusion, so blocking it does not disable the fusion. It buys apoptotic sensitivity, not disease reversal, and resistance does not need to touch the tether at all.
  • Venetoclax engages the mitochondrial pool of BCL2, and the mitochondrial pool is one of three (PMID 8402648). Compartment-restricted Bcl-2 mutants show that the ER and mitochondrial pools serve spatially distinct apoptotic pathways in a cell-type-dependent way (PMID 8861942), so an ER-resident or nuclear-envelope-resident reservoir could survive a BH3 mimetic, and the fractional distribution in a KMT2A-rearranged blast has never been measured. BH3 profiling, the assay the clinical case rests on, reads mitochondria by construction and cannot see this.
Still open
  1. Does menin-inhibitor-induced differentiation turn BCL2 off and MCL1 on, making a menin/venetoclax combination antagonistic in sequence even where it is synergistic in simultaneous exposure? This is directly testable in a KMT2A-r line with a differentiation time course and a BH3 profile, and the answer changes how the combination should be scheduled.
  2. Does the monocytic venetoclax-resistance mechanism apply to KMT2A-rearranged B-ALL that undergoes lineage switch to myeloid under CD19-directed therapy? This graph already records that switch as a risk; if switched blasts are also venetoclax-resistant, two of the project's leading strategies share one failure mode.
  3. Is BCL2 dependency in KMT2A-r ALL a consequence of DOT1L activity specifically, such that H3K79me2 level predicts venetoclax response? Benito and colleagues show the mechanism but not that the mark is a usable biomarker.
  4. What is the paediatric-specific safety of sustained BCL2 inhibition during lymphoid development? The mouse knockout involutes thymus and spleen, and infants are building both.
Every source behind this entry (7)
  1. E1strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.database

    BCL2 is not DepMap common-essential (mean Chronos -0.036 across 1,258 lines, 3.3% depleted) but falls to -0.351 across 57 leukemia lines, with six of the nine most depleted leukemia lines KMT2A-rearranged; gnomAD reports pLI 0.995 and LOEUF 0.333 with 1 observed against 14.2 expected loss-of-function variants; HPA reports thymus 17.0, spleen 16.9 and classical monocyte 0.0 nTPM.

    What this evidence cannot show

    Per-line Chronos values retrieved through Open Targets and means computed by us; DepMap release version unverified. Cell-line lesion assignments taken from Cellosaurus, not from DepMap. With only 14.2 expected loss-of-function variants, LOEUF for BCL2 is weakly powered and should not be treated as a firm constraint estimate. CRISPR knockout of an anti-apoptotic gene measures a different thing from pharmacological BH3 mimicry.

    Evidence that argues against this (1)
    • A Chronos mean mixed across a KMT2A-rearranged subset and a comparator arm must be lineage-split before it is stated: the index arm is myeloid-enriched, so a difference between the two mixes fusion status with lineage. Ten or eleven index lines and four KMT2A::AFF1 lines also put wide intervals on every figure quoted. Six of nine most-depleted leukaemia lines being KMT2A-rearranged is also an enrichment computed on nine lines. Oncofusion knowledge graph (internal analysis) 2026
  2. E2strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

    t(4;11) patient cells express high BCL2 and are highly sensitive to the BCL-2-specific BH3 mimetic ABT-199 (venetoclax); MLL/AF4 specifically upregulates BCL2 but not other BCL-2-family members via DOT1L-mediated H3K79me2/3; a t(4;11) line is sensitive to ABT-199 combined with DOT1L inhibition; and ABT-199 synergises with standard induction-type therapy in a xenotransplant model.

    MLL-Rearranged Acute Lymphoblastic Leukemias Activate BCL-2 through H3K79 Methylation and Are Sensitive to the BCL-2-Specific Antagonist ABT-199. · Benito JM, Godfrey L, Kojima K, Hogdal L, Wunderlich M, Geng H, et al., Milne TA, Konopleva M · Cell Reports 2015
    PMID 26711339DOI
    What this evidence cannot show

    Cell lines, primary samples and xenografts, all preclinical; no randomised clinical data in KMT2A-rearranged infant ALL supports this. The DOT1L-inhibitor combination is shown in a single t(4;11) line. The paper advocates adding venetoclax to therapy, which is an authors' recommendation rather than a demonstrated clinical result.

    Evidence that argues against this (1)
    • Monocytic AML loses BCL2 expression and becomes venetoclax-resistant, and KMT2A-rearranged AML is frequently monocytic, so the sensitivity demonstrated in lymphoid t(4;11) may not transfer to the myeloid side of KMT2A-rearranged disease. Pei S et al. 2020
  3. E3strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

    Response to venetoclax plus azacitidine in AML tracks developmental stage: primitive AML is sensitive while monocytic AML is resistant, loses BCL2 expression, relies on MCL1 for oxidative phosphorylation and survival, and is selected for at relapse.

    Monocytic Subclones Confer Resistance to Venetoclax-Based Therapy in Patients with Acute Myeloid Leukemia. · Pei S, Pollyea DA, Gustafson A, Stevens BM, Minhajuddin M, Fu R, et al., Jordan CT · Cancer Discovery 2020
    PMID 31974170DOI
    What this evidence cannot show

    Older, previously untreated AML patients on venetoclax plus azacitidine; neither the population nor the regimen is paediatric KMT2A-rearranged ALL. The extension to KMT2A-rearranged disease rests on the epidemiological observation that KMT2A-r AML is often monocytic, which this paper does not itself test. The prediction that menin-inhibitor-induced differentiation would antagonise venetoclax is our inference and is untested.

    Evidence that argues against this (1)
    • Developmental-stage-dependent venetoclax response was established in ADULT AML, and the resistant compartment is monocytic. Infant KMT2A::AFF1 disease is B-lymphoid and pro-B, so neither pole of that axis is the beat's patient — though lineage switch to monocytic AML is exactly what this graph records under CD19-directed pressure, which makes the finding relevant for a different reason than the entry gives. Journal of Clinical Oncology 2025
  4. E4established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.database

    Verified co-crystal structures show the BCL-2 hydrophobic groove occupied by a BAX BH3 peptide (2XA0, 2.70 A), by venetoclax (6O0K, 1.62 A), by navitoclax (4LVT, 2.05 A) and by an ABT-737 analogue (6QGG, 1.50 A).

    What this evidence cannot show

    We verified accession existence, title, method and resolution only. Many BCL-2 crystal structures use chimeric or loop-deleted constructs, which we did not check entry by entry; construct details should be read from each entry before structural conclusions are drawn.

    Evidence that argues against this (1)
    • Counting PDB entries measures how tractable a protein was to CRYSTALLISE, which correlates with being small, ordered and well-behaved rather than with being druggable. What decides design value is whether the deposited density covers the site that matters, at a resolution that supports modelling, in the relevant state — and this graph has repeatedly found structures described in the direction that overstates what is resolved. The BCL2 groove is the best-characterised protein-protein interface in oncology and venetoclax's existence proves the point; it also means every structural fact here is downstream of a solved problem and contributes nothing to the fusion question. RCSB PDB 2026
  5. E5established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper

    Bcl-2-deficient mice complete embryonic development but show growth retardation and early postnatal mortality, massive apoptotic involution of thymus and spleen after initially normal haematopoiesis, severe polycystic kidney disease with renal failure, and hair greying at the second follicle cycle.

    PMID 8402909DOI
    What this evidence cannot show

    Germline knockout from conception, which models a developmental requirement rather than time-limited pharmacological inhibition in a child. Decades of venetoclax clinical use have not produced a polycystic kidney signal, so the renal phenotype is probably developmental and not translatable; the lymphoid involution, by contrast, has a clear clinical counterpart in venetoclax-associated cytopenias.

    Evidence that argues against this (1)
    • This is a mouse knockout describing what the gene does in normal development or normal haematopoiesis, over the animal's whole life, with the protein entirely absent. A drug gives partial reversible inhibition in an adult for weeks. The experiment bounds the biological requirement and not the therapeutic window. Bcl-2-null mice completing development then involuting thymus and spleen is also a prediction of on-target lymphoid toxicity, which is what venetoclax delivers clinically. Nature 2012
  6. E6established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.database

    UniProt P10415 annotates human BCL2 as 239 residues with BH4 at 10-30, BH3 at 93-107, BH1 at 136-155 and BH2 at 187-202, plus a disordered region at 39-85 and a 92-107 region required for interaction with the SEPTIN4 ARTS isoform.

    What this evidence cannot show

    UniProt motif boundaries are curated annotations; the hydrophobic groove that binds BH3 helices is formed by residues from several motifs and is not itself a single annotated feature, so the venetoclax site cannot be read off these coordinates directly.

    Evidence that argues against this (1)
    • A UniProt domain table is an annotation, and the BH-domain boundaries it gives are consensus assignments that differ by a few residues between sources. Nothing here is measured in a KMT2A-rearranged cell. UniProt Consortium 2026
  7. E7established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper

    The 26-kDa Bcl-2 protein resides in the nuclear envelope, the endoplasmic reticulum and the OUTER mitochondrial membrane in a non-uniform, patchy distribution, and is absent from cytosol - shown together by gradient fractionation, laser-scanning and immuno-electron microscopy with six independent antibodies, and by import assays showing transmembrane-domain-dependent association with mitochondria without import into a protease-resistant compartment.

    PMID 8402648
    What this evidence cannot show

    1993, and the index system is a t(14;18) lymphoma line, not a KMT2A-rearranged leukaemia. The six-antibody design is unusually strong for its era and is why the three-compartment result has held. What the study does NOT give is the fractional distribution between the three compartments, which is the number a BH3-mimetic argument actually needs.

    Evidence that argues against this (1)
    • The three compartments are not interchangeable. Bcl-2 mutants engineered to sit in only one location reveal spatially distinct apoptotic pathways, and which compartment is protective depends on the cell type. So 'BCL2 is at the mitochondrion, the ER and the nuclear envelope' is not a single fact, and a venetoclax argument that reasons only from the mitochondrial pool is reasoning about a subset of the protein whose size in this disease nobody has measured. Zhu W et al. 1996
Connected to