Oncofusion

start / graph / gene-skp2

Genestrong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.Q78-interactome-absent

SKP2 - the substrate receptor that grips the retained fragment; two nodes in this graph already pointed at it as a free-text string with nothing to point to

also called SKP2 · FBXL1 · S-phase kinase-associated protein 2 · p45 · Cyclin-A/CDK2-associated protein p45

SKP2 is the part of a cellular waste-disposal machine that decides which proteins get tagged for destruction. It grips the same front end of KMT2A that every leukaemia fusion keeps - and the fusions escape being destroyed.

The same thing, for a scientist

SKP2/FBXL1 (Q13309, 424 aa) is the F-box substrate-receptor subunit of the SCF(Skp2) E3 ubiquitin ligase. It reaches this project as a **KMT2A partner on the retained N-terminal fragment**: SCF(Skp2) degrades MLL in S phase through an entry point within residues 1-1400, prevalent MLL fusions are resistant to that degradation, and ATR phosphorylation of MLL **S516** disrupts the SKP2 contact and lets MLL accumulate [E2][E3]. This graph has held both halves of that mechanism since cycle 39 - `interface-skp2-kmt2a-n-terminal-degron` and `mechanism-atr-kmt2a-s516-skp2-switch` - and **both name SKP2 as a free-text string in `data.partner_a` with no node to resolve to**, so the contact has never been an edge. IntAct records **two KMT2A-SKP2 records, neither spoke-expanded**, both from Liu et al. 2010 [E1], which is the same publication the S516 mechanism rests on; the degradation half comes from a second, independent publication [E3]. This entry supplies the missing node and the identity, and does not restate the mechanism.

Go deeper
The interactome record, and a correction to the record count
IntAct ACtypedetectionexpansionsource
EBI-15875668physical association (MI:0915)anti-tag co-IPnonePMID 20818375
EBI-15875687physical association (MI:0915)anti-bait co-IPnonePMID 20818375

Both carry the interactor as Q13309-1 (isoform 1), not the bare accession 1. Neither is spoke-expanded, and the pair has both the tagged and the endogenous-antibody direction, which is the internal control most rows in this interactome lack.

The dispatch brief for this beat recorded SKP2 as 1 record; the REST endpoint returns 2. The discrepancy is in the counting, not in the data - the two rows are the two directions of one experiment pair - and it recurs across this beat's twelve partners (see ana-the-unwritten-interactome-partners-are-better-observed-than-the-written-ones).

What SKP2 is, outside this contact

SKP2 is the canonical F-box/LRR substrate receptor of SCF, best known for targeting the CDK inhibitor p27/CDKN1B for degradation at the G1/S transition. That gives the node a second reason to exist here: SKP2 is an oncogenic driver in its own right in several haematological malignancies, and a 'KMT2A interactor' label on a protein that is already a proposed drug target is exactly the kind of statement that gets over-read. What is documented for this disease is narrow - a degradation route on the retained fragment that the fusion escapes - and nothing more.

Why the fusion escaping degradation is the interesting half

A fusion protein that keeps the N-terminal 1,400 residues but loses the C-terminus, and is resistant to a degradation route that enters through that retained region, is a stability difference between the fusion and wild-type KMT2A that does not depend on the partner at all 3. If the escape is general across partners, it is a property of truncation, and every KMT2A fusion is more stable than the protein it came from for a reason nobody has had to invoke the partner to explain.

The graph's existing entry is careful about the other direction and this entry keeps that care: Liu et al.'s own conclusion is that MLL fusions act as dominant negatives that abrogate ATR phosphorylation of wild-type MLL, so the fusion is described as disabling this axis in trans rather than exploiting it 2.

Facts and figures
uniprotQ13309
length AA424
gene symbolSKP2
protein nameS-phase kinase-associated protein 2 (Cyclin-A/CDK2-associated protein p45, F-box/LRR-repeat protein 1)
uniprot resolutionaccession:Q13309, reviewed human, one record; primary gene name read back as SKP2 (entry SKP2_HUMAN); length 424 aa checked against UniProt 2026-09-21.
symbol collision checkdata/symbol-collisions.json greped 2026-09-21 for SKP2 and FBXL1: no row existed and none is proposed.
baitKMT2A (Q03164)
complexSCF(Skp2) - SKP1, CUL1, RBX1 and the SKP2 F-box substrate receptor
canonical substratep27/CDKN1B at the G1/S transition
shown for fusionPARTIAL - 'tested prevalent MLL fusions are resistant to degradation' (PMID 17908926), fusion identities not readable from the abstract; the S516 phospho-switch itself has NOT been tested on any fusion.
fetched2026-09-21
subnuclear localizationNONE ASSIGNED - A BOUNDED ABSENCE, and the productive axis for this protein is demonstrably the nucleocytoplasmic one rather than the subnuclear one. UniProt carries no subnuclear qualifier; the entire published localisation literature for SKP2 that this beat retrieved concerns nuclear-versus-cytoplasmic partitioning through the 67-73 NLS [E_q91_skp2_loc1][E_q91_skp2_loc2][E_q91_skp2_loc3]. SKP2 is a soluble 424-residue F-box protein with no membrane anchor, so a subnuclear answer is possible in principle and has simply never been sought - this is a bounded absence, not a NOT APPLICABLE.
localization shown for fusionNO - SKP2 has never been localised in the presence of any KMT2A fusion, in any lineage. The existing fusion-scoped statement in this entry is about DEGRADATION RESISTANCE, not compartment: prevalent MLL fusions are resistant to SCF(Skp2)-mediated degradation (its E3, PMID 17908926) and the ATR/S516 phospho-switch has not been tested on any fusion. The localisation question this opens is specific and new: if Akt1 activity exports SKP2 from the nucleus, then the degree to which KMT2A - or a KMT2A fusion - escapes SCF(Skp2) should depend on the PI3K/Akt state of the blast, and no experiment in this literature has controlled for that.
interaction with kmt2a
{
  "source_pmid": "20818375",
  "source_class": "focused mechanistic study",
  "n_intact_records": 2,
  "n_spoke_expanded": 0,
  "n_source_publications_in_intact": 1,
  "n_source_publications_in_literature": 2,
  "intact_acs": [
    "EBI-15875668",
    "EBI-15875687"
  ],
  "interaction_types": {
    "MI:0915 physical association": 2
  },
  "detection_methods": [
    "anti tag coimmunoprecipitation",
    "anti bait coimmunoprecipitation"
  ],
  "interactor_identifier_recorded": "Q13309-1 (isoform 1, not the bare accession)",
  "brief_stated_record_count": 1,
  "rest_record_count": 2,
  "shared_pdb_entries": []
}
localization
{
  "CONSTITUTIVE": "NUCLEUS, via a short annotated nuclear localisation signal at residues 67-73. UniProt Q13309 (fetched live 2026-09-21) carries both 'Nucleus' and 'Cytoplasm' and attributes BOTH to the same single paper, PubMed:22770219, with no note separating the conditions - so the database records the two compartments and discards the switch between them [E_q91_skp2_loc1].",
  "INDUCED_AND_THERE_ARE_TWO_INDEPENDENT_ROUTES_INTO_THE_SAME_SEVEN_RESIDUES": "SKP2 leaves the nucleus when its NLS is post-translationally modified, and two different enzymes do it. (1) PHOSPHORYLATION: Akt1 - but not Akt2 - phosphorylates Skp2 at SER 72, which lies within the putative NLS, and phosphorylation of Ser72 leads to CYTOPLASMIC TRANSLOCATION of Skp2; the same modification disrupts the Cdh1-Skp2 interaction and so also stabilises the protein [E_q91_skp2_loc2]. (2) ACETYLATION: p300 acetylates Skp2 at LYS 68 AND LYS 71, antagonised by the SIRT3 deacetylase, and acetylation of Skp2 IN THE NLS promotes its CYTOPLASMIC RETENTION [E_q91_skp2_loc3]. K68, K71 and S72 are three adjacent residues inside the 67-73 motif, and two signalling pathways converge on them.",
  "WHY_THIS_MATTERS_FOR_A_DEGRADATION_CONTACT": "SKP2 is in this graph because SCF(Skp2) degrades KMT2A through an entry point within residues 1-1400 and the fusions escape that degradation. A substrate receptor that can be exported to the cytoplasm by growth-factor signalling is a receptor whose access to a nuclear substrate is conditional - so 'the fusion escapes SCF(Skp2)' has a second, untested mechanism available to it alongside the structural one this entry already records."
}
cell type
{
  "where_the_localisation_switch_was_measured": "Breast and prostate carcinoma systems - cytoplasmic Skp2 correlates with more aggressive forms of breast and prostate cancer, and the acetylation work uses those models [E_q91_skp2_loc3]; the Akt1/Ser72 work is in cancer cell lines with elevated Akt activity [E_q91_skp2_loc2]. Neither is haematopoietic.",
  "HAEMATOPOIETIC_EVIDENCE_EXISTS_AND_IT_IS_ABOUT_LEVEL_NOT_COMPARTMENT": "In BCR-ABL-positive cells (Mo7e-p210, BaF/3-p210) and in PRIMARY CML CD34+ CELLS, BCR-ABL induces Skp2 expression through the PI3K pathway to promote p27(Kip1) degradation and proliferation; imatinib inhibition of BCR-ABL produces a PI3K-dependent G1 arrest with NUCLEAR ACCUMULATION OF p27(Kip1) [E_q91_skp2_loc4]. That is SKP2 abundance and its substrate's compartment in primary human leukaemic progenitors - the closest haematopoietic anchor available, and it is the same PI3K/Akt axis that moves SKP2 itself.",
  "never_in_this_disease": "SKP2 protein has never been localised, or even quantified, in a KMT2A-rearranged cell of either lineage."
}
cell state
{
  "CELL_CYCLE_IS_A_REAL_STATE_VARIABLE_HERE": "SCF(Skp2) degrades MLL IN S PHASE (this entry's E3), so the contact that puts SKP2 in this graph is itself cell-cycle-restricted. A single resting-state statement about SKP2 would be wrong twice over - once for the cycle and once for the signalling-driven export below.",
  "GROWTH_FACTOR_PI3K_AKT_STATE": "Elevated Akt1 activity moves SKP2 out of the nucleus and stabilises it [E_q91_skp2_loc2]; BCR-ABL drives the same pathway to raise SKP2 levels in CML progenitors [E_q91_skp2_loc4]. In a cell with active PI3K/Akt signalling, SKP2 is simultaneously more abundant and less nuclear - two effects of one input pointing in opposite directions for a nuclear substrate.",
  "METABOLIC_STATE": "SIRT3 is the deacetylase antagonising p300 at K68/K71 [E_q91_skp2_loc3]; SIRT3 is NAD+-dependent and mitochondrial, so SKP2's nuclear residence is coupled to a metabolic sensor. This graph already holds cell-state-oxphos-dependence, and the connection is untested in any leukaemia."
}
localization bounded absence
{
  "date": "2026-09-22",
  "tool": "NCBI esearch db=pubmed",
  "run_by": "Q91",
  "queries": [
    {
      "query": "Skp2[tiab] AND \"Ser72\"[tiab]",
      "querytranslation": "\"Skp2\"[Title/Abstract] AND \"Ser72\"[Title/Abstract]",
      "count": 5,
      "role": "the phospho-switch literature, small and specific"
    },
    {
      "query": "Skp2[tiab] AND phosphorylation[tiab] AND \"cytoplasmic translocation\"[tiab]",
      "querytranslation": "\"Skp2\"[Title/Abstract] AND \"phosphorylation\"[Title/Abstract] AND \"cytoplasmic translocation\"[Title/Abstract]",
      "count": 2
    },
    {
      "query": "Skp2[tiab] AND \"nuclear localization signal\"[tiab]",
      "querytranslation": "\"Skp2\"[Title/Abstract] AND \"nuclear localization signal\"[Title/Abstract]",
      "count": 6
    },
    {
      "query": "SKP2[tiab] AND \"S phase\"[tiab] AND nuclear[tiab]",
      "querytranslation": "\"SKP2\"[Title/Abstract] AND \"S phase\"[Title/Abstract] AND \"nuclear\"[Title/Abstract]",
      "count": 86,
      "role": "operand control - the cell-cycle literature is large"
    },
    {
      "query": "SKP2[tiab] AND (cytoplasmic[tiab] OR nuclear[tiab]) AND leukemia[tiab]",
      "querytranslation": "\"SKP2\"[Title/Abstract] AND (\"cytoplasmic\"[Title/Abstract] OR \"nuclear\"[Title/Abstract]) AND \"leukemia\"[Title/Abstract]",
      "count": 7,
      "note": "screened; the productive one is the BCR-ABL/CML CD34+ paper cited as E_q91_skp2_loc4, and it measures SKP2 LEVEL and p27's compartment rather than SKP2's own"
    }
  ],
  "reading": "The absence bounded here is a subnuclear compartment for SKP2 and any localisation of SKP2 in KMT2A-rearranged disease. It is NOT an absence of localisation biology - SKP2 has an unusually well-worked nucleocytoplasmic switch. Titles and abstracts only (rule 80). The alias FBXL1 and the protein name p45 were not run as separate operands.",
  "absence_class": "never-measured"
}
How this could bite us back
  • SKP2 inhibitors exist and 'KMT2A interactor' would make SKP2 look like a target here. The documented direction is the opposite of useful: SCF(Skp2) DEGRADES wild-type MLL, so inhibiting SKP2 stabilises the wild-type protein and does nothing obvious to a fusion that already escapes the route.
  • SKP2's dominant substrate is p27. Anything done to SKP2 in a proliferating marrow moves the G1/S brake in every cell, which is a cell-cycle toxicity and not an on-target effect in this disease.
  • The graph has carried this contact for several cycles as text inside two other entries' data blocks. A contact that cannot be reached by an edge is invisible to every degree-based and coverage-based measurement this project runs - which is how it stayed unwritten.
Still open
  1. Is the SCF(Skp2) escape general across KMT2A fusion partners, or partner-specific? If it is general it is a property of truncation and every fusion is intrinsically more stable than wild-type KMT2A - a fact that would apply to KMT2A::AFF1 without anyone having tested KMT2A::AFF1.
  2. Is S516 phosphorylated on the fusion protein itself? The residue is retained in every fusion and the question has never been asked; a phospho-specific blot on a KMT2A::AFF1 line under ATR inhibition would answer it.
  3. Where on KMT2A residues 1-1400 does SKP2 actually bind? No residues, no structure and no degron motif are mapped, so 'the retained fragment' is as precise as this contact currently gets.
Every source behind this entry (7)
  1. E1established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.database

    IntAct's REST record holds two KMT2A-SKP2 interaction records, EBI-15875668 (anti-tag co-immunoprecipitation) and EBI-15875687 (anti-bait co-immunoprecipitation), both typed MI:0915 physical association, neither carrying an expansion method, both from PMID 20818375, with the SKP2 side recorded as isoform Q13309-1.

    What this evidence cannot show

    One database on one day, REST endpoint (brief rule 93). Two records are two curated experiments from ONE publication, not two independent findings. IntAct curates no rows at all from PMID 17908926, the paper that established the degradation itself, so the curated record understates the literature here rather than overstating it.

    Evidence that argues against this (2)
    • These two rows are not IntAct's own curation. Both carry the annotation 'comment (DIP original experiment: DIP-254396X)' and 'DIP-254397X' respectively - they are imports of the Database of Interacting Proteins' reading of this paper, re-exported under IntAct accessions. So 'IntAct's REST record holds two KMT2A-SKP2 records' is one curation team's reading surfaced through a second database, and treating IntAct as an instrument independent of DIP would double-count it. Two of the other four rows from this publication are filed against DIP-29329N, an identifier with no UniProt accession at all, which is what a cross-database import looks like when the mapping fails. EMBL-EBI IntAct 2026
    • SKP2 is the accession in this beat where the two endpoints AGREE most closely - PSICQUIC id:Q13309 returns 300 against REST's 365, 82% - while for P36956 the recovery is 33% and for O60477 26%. That matters in the opposite direction from the usual caveat: it means the two-record count here is close to complete, and it also means the record counts this beat compares across partners were produced by an instrument whose blind spot varies by a factor of three between them. A partner ranking built on those counts is not comparing like with like. EMBL-EBI IntAct and PSICQUIC 2026
  2. E2strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

    In human cells under genotoxic stress in S phase (no KMT2A fusion in the arm that maps the site), ATR phosphorylates MLL at serine 516, the phosphate disrupts MLL's interaction with SCF(Skp2) and blocks its degradation, and Mll-null MEFs are rescued by wild-type but not S516A MLL.

    PMID 20818375DOI
    What this evidence cannot show

    Already cited five times in this graph; this entry adds no new reading of it and exists to give the SKP2 side a node (brief rule 82 - the conclusion is written, in mechanism-atr-kmt2a-s516-skp2-switch). MEFs and human lines; no infant B-lineage system, no KMT2A::AFF1 (brief rule 73). The paper's own conclusion is that MLL fusions act as dominant negatives on this axis rather than using it.

    Evidence that argues against this (1)
    • UniProt annotates the adjacent S518 rather than S516, and the PRIDE/PTMeXchange re-analysis localises S516 at probability 0.82 - short of certainty. This is recorded in the graph's existing S516 entry and is not resolved here. UniProtKB 2026
  3. E3moderate confidence 3 of 5: One strong study, or several smaller consistent ones.paper

    MLL is degraded biphasically by SCF(Skp2) in S phase and APC(Cdc20) in late M phase, with the reported entry point within the N-terminal approximately 1,400 residues that KMT2A fusions retain, and tested prevalent MLL fusions are resistant to that degradation.

    PMID 17908926DOI
    What this evidence cannot show

    The graph's existing interface-skp2-kmt2a-n-terminal-degron entry records this paper as abstract-only after four retrieval routes failed on 2026-09-19. Every field of this citation block was taken from a live PubMed efetch on 2026-09-21 and not from that entry (brief rule: never from a sibling entry, never from memory) - and that mattered: the DOI this agent would have written from recall, 10.1101/gad.1594107, is wrong; the fetched value is 10.1101/gad.1574507. The residues of the contact, whether it is a classical phosphodegron, and why the fusion escapes are all unmapped, and the identities of the fusions tested are not readable from the abstract.

    Evidence that argues against this (2)
    • SKP2's documented relevance in KMT2A-rearranged leukaemia runs through p27, not through MLL stability, which removes the main reason to read this degron as a disease mechanism. Meyer et al. 2018 identify Cdkn1b/p27Kip1 as a direct miR-196b target in MLL-AF9 leukaemia, show that raising p27 reduces MLL-r self-renewal, promotes monocytic differentiation and induces cell death, and that pharmacologic inhibition of the Cks1-Skp2-containing SCF complex raises p27 and inhibits human AML growth. So a SKP2 perturbation in this disease has a well-supported alternative explanation with no KMT2A in it, and any phenotype from targeting SKP2 here would not be evidence for the N-terminal degron. Meyer SE et al. 2018
    • MLL1 abundance is set by competition for Skp2 as well as by the substrate, so 'fusions are resistant to that degradation' is one of several routes to MLL accumulation and not a property that can be read off a fusion-versus-wild-type comparison alone. Swarnalatha et al. show that the hepatitis B HBx oncoprotein interferes with E2F1 binding to the Skp2 E3 ligase and that transcriptionally active E2F1 and MLL1 protein accumulate as a consequence. A third protein occupying Skp2 produces the same readout - more MLL1 - as a substrate that escapes it, and the 2007 experiment does not distinguish them. Swarnalatha M et al. 2013
  4. E_q91_skp2_loc1established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.database

    UniProtKB Q13309 (SKP2_HUMAN, primary gene SKP2, 424 aa, read back live 2026-09-21) carries TWO subcellular location values, 'Cytoplasm' and 'Nucleus', and attributes BOTH to the same single paper, ECO:0000269 PubMed:22770219. There is no location note and no condition scoping of any kind. The feature table carries a NUCLEAR LOCALIZATION SIGNAL motif at RESIDUES 67-73.

    What this evidence cannot show

    Two compartments recorded and the switch between them discarded. The paper cited is a genuine localisation source - unlike several in this tranche - but the annotation reduces a regulated, modification-dependent partition to a flat list of two places.

    Evidence that argues against this (1)
    • The flat two-value record is worse than under-informative here because it loses the one fact that makes SKP2's compartment mechanistically interesting: BOTH regulatory modifications land inside the seven residues UniProt does annotate. K68 and K71 are acetylated by p300 and S72 is phosphorylated by Akt1, and the NLS is 67-73 - so the feature table and the location comments contain, separately, everything needed to state the switch, and no field in the record joins them. Reading the location comments alone gives 'SKP2 is in the nucleus and the cytoplasm', which is true of a large fraction of the proteome and predicts nothing. Gao D et al. 2009
  5. E_q91_skp2_loc2strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

    The serine/threonine kinase Akt1, BUT NOT Akt2, directly controls Skp2 stability through the APC-Cdh1 ubiquitin ligase pathway: Akt1 phosphorylates Skp2 at SER 72, which is required to disrupt the Cdh1-Skp2 interaction. Ser72 lies WITHIN A PUTATIVE NUCLEAR LOCALIZATION SEQUENCE, and phosphorylation of Ser72 by Akt LEADS TO CYTOPLASMIC TRANSLOCATION OF SKP2.

    PMID 19270695DOI
    What this evidence cannot show

    Isoform-specific for Akt1 over Akt2, which is an unusual and therefore fragile claim; the systems are cancer cell lines with elevated Akt activity, not haematopoietic. Read from the abstract. The Ser72 site was contested in the contemporaneous literature on the stability half of the mechanism, though not on the localisation half.

    Evidence that argues against this (1)
    • The two halves of this result pull against each other in a way the paper does not resolve and that matters for how this graph uses SKP2. Phosphorylation at Ser72 STABILISES Skp2 by blocking Cdh1 AND EXPORTS it to the cytoplasm. More Skp2 protein and less nuclear Skp2 is not obviously more SCF(Skp2) activity on a nuclear substrate like KMT2A - it could be less. Any prediction this graph makes about fusion escape from SCF(Skp2) under PI3K/Akt signalling therefore has an undetermined sign, and saying so is more useful than picking one. Inuzuka H et al. 2012
  6. E_q91_skp2_loc3strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

    A second, independent modification of the same motif: Skp2 is ACETYLATED BY p300 AT K68 AND K71, a process antagonised by the SIRT3 deacetylase. Inactivation of SIRT3 raises Skp2 acetylation and Skp2 stability by impairing Cdh1-mediated proteolysis; and ACETYLATION OF SKP2 IN THE NUCLEAR LOCALIZATION SIGNAL PROMOTES ITS CYTOPLASMIC RETENTION, with cytoplasmic Skp2 enhancing cellular migration through ubiquitination and destruction of E-cadherin. Cytoplasmic Skp2 correlates with more aggressive forms of breast and prostate cancer.

    Acetylation-dependent regulation of Skp2 function. · Inuzuka H, Gao D, Finley LW, Yang W et al. · Cell 2012
    PMID 22770219DOI
    What this evidence cannot show

    Epithelial tumour biology throughout - breast, prostate, E-cadherin, migration - and none of it haematopoietic. This is the paper UniProt cites for BOTH of Skp2's compartments, and it earns that citation, which is worth noting in a tranche where most such citations do not.

    Evidence that argues against this (1)
    • The convergence of two enzymes on one seven-residue motif reads as strong corroboration and should be discounted somewhat, because K68, K71 and S72 are adjacent residues in a basic NLS and ANY bulky or charge-neutralising modification there would be expected to impair importin binding. Two mechanisms hitting the same motif is partly a statement about the motif's chemistry rather than two independent discoveries of a regulated switch. The stronger evidence is that both groups independently observed the functional consequence - cytoplasmic Skp2 - in different systems, and that is what this entry rests on rather than the site coincidence. Gao D et al. 2009
  7. E_q91_skp2_loc4strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

    In BCR-ABL-expressing cells (Mo7e-p210 and BaF/3-p210) and in PRIMARY HUMAN CML CD34+ CELLS, BCR-ABL induces Skp2 expression through the PI3K pathway to promote p27(Kip1) degradation and proliferation; STI571/imatinib inhibition of BCR-ABL kinase activity produces a PI3K-mediated G1 cell-cycle arrest associated with NUCLEAR ACCUMULATION OF p27(Kip1) and downregulation of cyclins D and E.

    PMID 15833859DOI
    What this evidence cannot show

    BCR-ABL CML, not KMT2A-rearranged disease (rule 73). The compartment measured is p27's, not SKP2's; SKP2 appears here as a transcriptionally induced level. Cited as the haematopoietic anchor for the PI3K/Akt-to-SKP2 axis, not as a localisation of SKP2.

    Evidence that argues against this (1)
    • Using this to argue that SKP2 will behave the same way in a KMT2A-rearranged blast would ignore the mechanism by which the two diseases differ. BCR-ABL is a constitutively active tyrosine kinase driving PI3K directly; KMT2A::AFF1 is a transcription-elongation lesion with no kinase activity, and the PI3K/Akt tone of an infant t(4;11) blast is not a published quantity anywhere this beat could find. The SKP2 level and the SKP2 compartment in this disease are both open, and the CML result establishes only that the axis is live in a leukaemic cell. Liu H et al. 2007
Connected to