Oncofusion

start / graph / protein-rpa-heterotrimer

Proteinestablished confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.Q192-the-rpa-heterotrimer

The replication protein A heterotrimer (RPA1/RPA2/RPA3) - the ssDNA clamp that every resection and replication-stress claim in this corpus silently depends on, and which had no entry until cycle 160

also called RPA · RP-A · RF-A · replication protein A · RPA complex · RPA70/RPA32/RPA14

Whenever DNA is unzipped - to copy it, or to repair a break - the exposed single strand has to be covered immediately or it will fold, degrade or recombine wrongly. RPA is the three-part protein that does the covering, and it is also the flag that tells the cell's alarm kinase that exposed DNA exists.

The same thing, for a scientist

RPA is a **1:1:1 heterotrimer of RPA1 (P27694, 616 aa), RPA2 (P15927, 270 aa) and RPA3 (P35244, 121 aa)** [E1]. It binds ssDNA in **two modes - an initial 8-10 nt mode and a stable ~30 nt mode - and the switch is made by the trimerisation core engaging DNA**, demonstrated by time-resolved proteolysis showing the core contributes nothing at 10 nt but interacts at 24 nt [E2]. The 2.8 Angstrom human trimerisation-core structure comprises RPA1's DBD-C, RPA2's DBD-D and the whole of RPA3, each an OB fold flanked by a C-terminal helix, trimerised through three parallel C-terminal helices, with a zinc ribbon and helix-turn-helix embedded in DBD-C [E2]. The full-length trimer-on-ssDNA structure adds the mechanism by which **DNA affinity and protein affinity are allosterically coupled** - a four-way DBD-B/DBD-C/BC-linker/ssDNA interface, with the linker mimicking DNA in DBD-B's groove - but **that structure is Ustilago maydis, not human, and its coordinates must not be applied to human numbering** [E3]. Downstream, **RPA-coated ssDNA is the structure that recruits ATRIP-ATR and licenses Chk1 activation** [E4], and the **nuclear pool of RPA is finite: exhausting it converts stalled forks into nucleus-wide breakage, which is the mechanism by which ATR inhibition kills** [E5]. A composition caveat: RPA is not strictly obligate - **aRPA substitutes RPA4 for RPA2** [E1].

Go deeper
Why this entry is a gap worth closing, and what it is NOT allowed to imply

This corpus holds an extensive resection and replication-stress layer - gene-rbbp8, gene-exo1, gene-topbp1, gene-rad51, gene-atr, gene-chek1, cell-state-replication-stress-icl-repair, mech-transcription-replication-conflict-kmt2a-r, mechanism-replication-stress-endog-cleavage-kmt2a, and a hypothesis that the menin arm and the CTCF arm license consecutive steps of one resection reaction. Every one of them presupposes RPA, and none of them had an RPA entry to point at. RPA2 is named 21 times in the corpus, RPA1 seven, RPA3 five, and all three were named only as instruments or as list members.

This entry closes that structural gap and nothing more. It asserts no contact between RPA and any KMT2A fusion. The only fusion-adjacent evidence in the corpus is (i) a single declined 6-spectrum RPA2 prey row in the Skucha AP-MS dataset, and (ii) a genuine, published, locus-level measurement of RPA occupancy at the MLL 11q23 breakpoint cluster region after etoposide 6 - which is a measurement at the gene that breaks, in cells with no fusion present, and is therefore about translocation genesis and not about the fusion protein.

The finite-pool property, which is the one that has therapeutic teeth

RPA is usually described as a passive coating. The property that matters for a drug programme is that the coating can run out 5.

In ATR-deficient cells stalled forks are initially stable, but unscheduled origin firing generates so much ssDNA that the nuclear RPA pool is exhausted, and the forks then break nucleus-wide - 'replication catastrophe'. Partially reducing RPA accelerated breakage; forcing RPA up delayed it even without ATR activity; and driving unscheduled origin firing broke forks even in cells with active ATR 5.

Three consequences for this project, stated as consequences and not as findings:

  1. ATR inhibition and RPA reduction are the same lever. They are not an obvious combination; they are one intervention approached from two sides. ATR/ATM kinase inhibition already shows antitumour activity in MLL-rearranged AML mouse models 7, and the corpus holds that work.
  2. A cell with high intrinsic replication stress has a smaller RPA reserve. This corpus argues that KMT2A-rearranged disease carries transcription-replication conflict and replication stress. If that is right, the RPA reserve is a quantitative property of these cells that nobody has measured.
  3. RPA abundance becomes a modifier, not a target. The therapeutic question is not 'inhibit RPA' - it is pan-essential - but 'how much reserve does this blast have'.
Facts and figures
representative subunitP27694
no bare uniprot fieldDeliberate, per brief rule 69b: a complex entry must not carry a single accession in a `uniprot` key, because an accession checker then validates the complex against one subunit. Subunit-level accessions are above and a representative is named explicitly.
stoichiometry1:1:1 (RPA1:RPA2:RPA3). Recorded as an inference from the trimerisation-core structure, in which one copy of each of DBD-C, DBD-D and RPA14 trimerises through three parallel C-terminal helices (PMID 11927569); UniProt states 'heterotrimer composed of RPA1, RPA2 and RPA3' without giving a ratio.
alternative compositionaRPA - RPA1 and RPA3 with RPA4 in place of RPA2 (UniProt P27694 and P35244, PubMed 19116208, 7760808). RPA4 has NO entry in this corpus and was not investigated. This is why 'obligate heterotrimer' is used with qualification throughout this beat.
PDB noteDELIBERATELY EMPTY. Two structures are cited by PMID in this entry - the human trimerisation core (PMID 11927569) and the fungal full-length trimer on ssDNA (PMID 23070815) - but no PDB accession was fetched and verified, and this project has a standing warning about the PDB giving confident wrong answers. A later beat should populate this by resolving the accessions and stating which is human.
domains noteComplex-level domain boundaries are held on the subunit entries (gene-rpa1, gene-rpa2, gene-rpa3), each sourced to its own fetched UniProt feature table. No DBD-A/B/C/E/F range is recorded anywhere in this beat, because the only full-length structural source is a fungal protein.
shown for fusionNO for the complex as a fusion partner - no KMT2A fusion, and no fusion oncoprotein of any family, has a published RPA interaction. The locus-level RPA measurement at 11q23 (PMID 21048951) is in cells with no fusion and concerns translocation genesis; the single RPA2 prey row in the MLL-fusion AP-MS dataset is declined elsewhere in this corpus.
epigenetic readoutNONE at complex level, class never-measured. RPA writes, reads and erases no histone mark. The published cascade that reaches it runs menin -> gamma-H2A.X -> SETD2 -> H3K36me3 -> LEDGF -> CtIP -> RPA2/RAD51 (PMID 42578372), placing RPA downstream of the mark. The KMT2A-locus RPA measurement was accompanied by an INO80 ChIP at the same site (PMID 21048951), which is a chromatin-remodeller readout rather than a mark.
absence classmeasured-not-searched
absence class noteFor the KMT2A question specifically. Two things are answerable with no wet work: (1) whether RPA1 and RPA3 were detected under MLL-AF4 below the top-300 rank cut in PXD009338, and (2) whether any deposited genome-wide ssDNA or RPA occupancy dataset covers 11q23 - the corpus's own absence entries pose that second question and it has not been run.
fetched2026-09-24
subunits
RPA3121.0RPA3: 121RPA2270.0RPA2: 270RPA1616.0RPA1: 616
measurements
count0.000count: 0occluded_site_size9.000occluded_site_size: 9occluded_site_size30.0occluded_site_size: 30
subunits — the numbers behind the chart
{
  "RPA1": {
    "accession": "P27694",
    "length_aa": 616,
    "kda_name": "RPA70",
    "contributes": "DBD-F (N-terminal, protein interaction), DBD-A, DBD-B, DBD-C (in the trimerisation core, carries the C4 zinc ribbon at 481-503)",
    "entry": "gene-rpa1"
  },
  "RPA2": {
    "accession": "P15927",
    "length_aa": 270,
    "kda_name": "RPA32",
    "contributes": "DBD-D (OB fold 74-148, in the trimerisation core), plus the disordered N-terminal phospho-tail and the C-terminal winged-helix partner-binding region 187-270",
    "entry": "gene-rpa2"
  },
  "RPA3": {
    "accession": "P35244",
    "length_aa": 121,
    "kda_name": "RPA14",
    "contributes": "DBD-E; the entire subunit sits in the trimerisation core",
    "entry": "gene-rpa3"
  }
}
ssdna binding
{
  "initial_mode_nt": "8-10",
  "stable_mode_nt": "~30",
  "switch_mechanism": "DNA binding by the trimerisation core; the core contributes nothing at 10 nt and interacts at 24 nt by time-resolved trypsin proteolysis (PMID 11927569)",
  "polarity_note": "RPA binds ssDNA with a defined 5'-to-3' polarity across its DBDs in the accepted model. NOT ASSERTED HERE as evidence: no source read by this beat states the polarity explicitly, and it is therefore recorded as a well-known claim this entry does not cite.",
  "allosteric_coupling": "Two conformational states with opposing DNA and protein affinities; the high-DNA-affinity state occludes the DBD-A/DBD-B protein-interaction site via a four-way DBD-B/DBD-C/BC-linker/ssDNA interface with the linker mimicking DNA (PMID 23070815). SPECIES: Ustilago maydis. Mechanism imported, coordinates NOT."
}
PDB IDS
[]
domains
[]
downstream
{
  "atr_activation": "RPA-coated ssDNA recruits ATRIP-ATR and licenses Chk1 activation (PMID 12791985). Corpus entries: gene-atr, gene-chek1 - do not duplicate them.",
  "alternative_activators_present_in_corpus": [
    "gene-topbp1"
  ],
  "alternative_activators_absent_from_corpus": [
    "ETAA1"
  ],
  "rad51_handoff": "RPA-coated ssDNA is displaced by RAD51 to form the recombinogenic filament; in the KMT2A-locus measurement the INO80 remodeller is implicated in that exchange (PMID 21048951). Corpus entry: gene-rad51."
}
locus level measurement
{
  "site": "MLL/KMT2A 11q23 breakpoint cluster region, exons 7-13, ~8.3 kb, around the ~20 bp therapy-related translocation breakpoint hotspot",
  "assay": "ChIP-qPCR with anti-RPA32/RPA34 antibody, six primer sets, normalised to a beta-globin control region at 11p15",
  "perturbation": "etoposide 100 uM, 10 min, +/- ATM (AT5BIVA versus chromosome-11-complemented 11-4; plus KU55933 in 11-4)",
  "result": "etoposide-induced RPA enrichment at the BCR in ATM-proficient cells, transient (back to normal by 30 min); in ATM-deficient cells more spatially extensive and persistent at 30 min",
  "fusion_present": false,
  "cells": [
    "AT5BIVA (ATM-deficient SV40-transformed human fibroblast)",
    "11-4 (AT5BIVA + chromosome 11)",
    "BV173 (human CML-derived B-cell line)"
  ],
  "pmid": "21048951",
  "why_this_matters_here": "It is the only published measurement placing RPA at this project's own locus, it is a DAMAGE-STATE measurement, and it was invisible to every subunit-symbol query because the paper writes 'Replication protein A (RPA)'. See ana-q192-the-complex-form-query-the-rpa2-decline-left-unrun-finds-rpa-at-the-kmt2a-breakpoint-hotspot."
}
cell type
{
  "answer": "NOT LINEAGE-SPECIFIC; required in every proliferating cell. No haematopoietic-specific measurement of the complex was found."
}
cell state
{
  "state": "S-PHASE, REPLICATION STRESS AND DAMAGE. The complex has no genomic address without single-stranded DNA, and its nuclear pool is FINITE and exhaustible under replication stress (PMID 24267891) - which makes cell state not merely relevant but quantitative for this complex."
}
measurements — the numbers behind the chart
[
  {
    "quantity": "occluded_site_size",
    "subject": "human RPA on ssDNA, initial mode",
    "value": 9,
    "units": "nt",
    "n": 1,
    "universe": "purified human RPA on naked oligonucleotides, time-resolved trypsin proteolysis, PMID 11927569",
    "null": null,
    "note": "reported by the source as a range, 8-10 nt; recorded as the midpoint with the range in the field above, and it is an occluded-site size in vitro, not a footprint in chromatin"
  },
  {
    "quantity": "occluded_site_size",
    "subject": "human RPA on ssDNA, stable mode",
    "value": 30,
    "units": "nt",
    "n": 1,
    "universe": "purified human RPA on naked oligonucleotides, PMID 11927569",
    "null": 9,
    "note": "null column holds the initial mode, which is the comparison that matters: the switch is the finding, not either value"
  },
  {
    "quantity": "count",
    "subject": "RPA heterotrimer subunits with an entry in this corpus before cycle 160",
    "value": 0,
    "units": "subunits",
    "n": 3,
    "universe": "data/graph/nodes over 3,086 entries, scripts/is-symbol-covered.py run on RPA1 (9 candidate names), RPA2 (4) and RPA3 (5), 2026-09-24",
    "null": 3,
    "note": "null is what a complete corpus would have. All three were NAMED - RPA1 in 7 files, RPA2 in 21, RPA3 in 5 - and none was a subject. A protein used as an assay readout is invisible to a coverage audit."
  }
]
How this could bite us back
  • RPA is required in every dividing cell. Direct RPA inhibition has no plausible therapeutic window in a paediatric leukaemia, and the existence of published RPA inhibitors targeting its ssDNA-binding and protein-interaction surfaces makes it more important, not less, that this entry is not read as target validation.
  • ATR inhibition kills by exhausting RPA (PMID 24267891), so an ATR inhibitor and any agent that lowers RPA are the same lever pulled twice rather than a rational combination. Conversely, a cell with a large RPA reserve is predicted to resist ATR inhibition - a resistance mechanism nobody has looked for in this disease.
  • The one locus-level RPA result in this disease points at GENESIS, not treatment: RPA and RAD51 over-load at the KMT2A breakpoint hotspot after etoposide when ATM is deficient (PMID 21048951). Its therapeutic implication, stated by the authors, is that inhibiting HR might REDUCE therapy-related secondary leukaemia - the opposite direction from the usual 'inhibit repair to kill the tumour' argument, and a bite-back on any HR-directed agent given alongside a topoisomerase-II poison.
  • Because RPA2 phosphorylation is the field's standard replication-stress readout, any drug that raises it will appear to be engaging this mechanism whether or not it is. A combination trial reading out phospho-RPA2 alone cannot distinguish target engagement from general replication stress.
Still open
  1. Does any deposited genome-wide ssDNA-mapping dataset (END-seq, BLESS, ssDNA-seq, S1-END-seq) cover chr11q23, and specifically the 8.3 kb KMT2A breakpoint cluster region? This corpus's absence-no-damage-state-occupancy-measurement-for-any-resection-factor-at-the-kmt2a-locus poses this question and it has not been run; PMID 21048951 shows the locus-level answer is positive by candidate qPCR, which makes the genome-wide version much more likely to be informative than a null result would suggest.
  2. What is the RPA reserve of a KMT2A-rearranged blast? If replication stress is intrinsically high in this disease, the finite-pool result (PMID 24267891) makes that reserve a measurable determinant of ATR-inhibitor sensitivity, and single-cell chromatin-bound RPA imaging is the published assay for it.
  3. Does menin bind the low-DNA-affinity RPA conformer preferentially? The allosteric-coupling model predicts it must, because menin binds RPA2 aa 43-171 and had no detectable effect on RPA-DNA binding - and the prediction is testable with the 8-10 nt versus 30 nt substrates the 2002 human paper already used.
  4. Is RPA4 expressed in haematopoietic cells, and does an aRPA complex lacking RPA2 exist in blasts? RPA4 has no entry in this corpus and the question bears directly on every subunit-coherence argument made about RPA here.
Every source behind this entry (7)
  1. E1established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.database

    RPA is a heterotrimer composed of RPA1 (UniProt P27694, 616 aa), RPA2 (P15927, 270 aa) and RPA3 (P35244, 121 aa); RPA1 and RPA3 are additionally components of aRPA, an alternative trimeric complex in which they associate with RPA4 instead of RPA2, and UniProt's own note on RPA1 states that the DNA-binding activity may reside exclusively on the RPA1 subunit.

    What this evidence cannot show

    Three accessions fetched individually and each primary gene name read back (rules 62, 65); md5s recorded in the per-subunit entries. Per rule 69b this complex entry carries subunit-level accessions in data.subunits and NO bare `uniprot` field, so an accession checker cannot mistake one subunit for the complex. Stoichiometry: UniProt says 'heterotrimer composed of RPA1, RPA2 and RPA3' and does not state 1:1:1 in those words; the 1:1:1 reading follows from the trimerisation-core structure [E2] and is recorded as an inference from it, not as a UniProt statement. 'DNA-binding activity may reside exclusively on RPA1' is UniProt's own hedge and is inconsistent with RPA2 having an OB fold and a low-affinity ssDNA site - the databases are not internally tidy here and this entry does not resolve it.

    Evidence that argues against this (1)
    • The word 'obligate', used about this complex throughout this corpus including in the adjudication that declined the Skucha RPA2 row, is not supported at full strength by this record: aRPA exists, so the composition is at least two complexes, and the founding menin paper reports binding to FREE RPA2 in preference to the trimer. Any argument of the form 'subunit X must have co-purified with subunit Y' inherits both caveats. Sukhodolets KE et al. 2003
  2. E2established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper

    Human RPA binds DNA in at least two modes, initial (8-10 nucleotides) and stable (approximately 30 nucleotides), with a large conformational change between them; the 2.8 Angstrom structure of the trimerisation core - RPA70's DBD-C, RPA32's DBD-D and the entire RPA14 subunit - shows three OB folds each flanked by a C-terminal helix with trimerisation mediated by three parallel C-terminal helices and a zinc ribbon plus helix-turn-helix embedded in DBD-C; and time-resolved trypsin proteolysis shows the core does not contribute to binding 10-nt substrates but does interact with 24-nt oligonucleotides, indicating the 8-10 to 30 nt switch is mediated by DNA binding with the trimerisation core.

    Structure of the RPA trimerization core and its role in the multistep DNA-binding mechanism of RPA. · Bochkareva E, Korolev S, Lees-Miller SP, Bochkarev A · EMBO Journal 21(7):1855-63 2002
    PMID 11927569DOI
    What this evidence cannot show

    Abstract level; full text not opened. HUMAN protein - this is the reason it, and not the more complete 2012 structure, is this entry's architectural reference. The nucleotide occluded-site sizes are from purified protein on naked oligonucleotides, not from chromatin. The zinc-ribbon location is cross-checked only against UniProt's C4-type zinc finger annotation at P27694 residues 481-503; no PDB coordinate file was downloaded and no residue range from the paper is quoted, so this entry's data.pdb_ids is empty rather than populated from recall.

    Evidence that argues against this (1)
    • A core structure cannot describe the parts of RPA that do the recruiting: RPA1's flexible N-terminal DBD-F, which is the ATRIP-binding module, and RPA2's disordered phospho-tail are both outside it. So this structure explains the DNA-binding switch and explains none of the protein-recruitment behaviour that makes RPA a signalling hub, and it should not be cited as a structure of RPA. Fan J et al. 2012
  3. E3established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper

    RPA binds ssDNA in two conformational states with opposing affinities for DNA and for proteins: RPA-protein interactions are compatible with a low-DNA-affinity state involving DBD-A and DBD-B but not with the high-DNA-affinity state that additionally engages DBD-C and DBD-D; the high-affinity RPA-ssDNA structure is a compact quaternary structure held together by a four-way interface between DBD-B, DBD-C, the intervening BC linker and ssDNA, with the linker binding into DBD-B's DNA-binding groove and mimicking DNA, which partially occludes the DBD-A/DBD-B protein-protein interaction site and establishes a mechanism for allosteric coupling of RPA's DNA and protein interactions.

    Structure and conformational change of a replication protein A heterotrimer bound to ssDNA. · Fan J, Pavletich NP · Genes & Development 26(20):2337-47 2012
    PMID 23070815DOI
    What this evidence cannot show

    CRITICAL SPECIES CAVEAT, and it is the reason no residue number from this paper appears anywhere in this beat: the protein crystallised is a USTILAGO MAYDIS heterotrimer. PubMed's MeSH terms for this record are 'Fungal Proteins' and 'Ustilago', and there is no 'Humans' term. The MECHANISM - two conformational states, DNA affinity anticorrelated with protein affinity, a DNA-mimicking BC linker - is what is imported here; the coordinates are not, and a later agent must not populate human domain boundaries from this paper. Abstract level only.

    Evidence that argues against this (1)
    • Importing a mechanism across a fungal-to-human boundary is exactly the move this project has been burned by, in the opposite direction, when murine coordinates were used against a human junction. The allosteric-coupling model has not, in the sources this beat read, been demonstrated on human RPA at structural resolution. Treat it as a strong hypothesis about human RPA rather than a fact about it - and note it makes a testable human prediction: menin, which binds RPA2 aa 43-171, should bind the LOW-DNA-affinity state preferentially, which would explain why menin had no effect on RPA-DNA binding in vitro. Bochkareva E et al. 2002
  4. E4established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper

    RPA is required for recruitment of ATR to sites of DNA damage and for ATR-mediated Chk1 activation in human cells; RPA stimulates ATRIP binding to ssDNA in vitro; ATRIP binding to RPA-coated ssDNA enables the ATR-ATRIP complex to associate with DNA and stimulates phosphorylation of DNA-bound Rad17; and the budding-yeast ATRIP homologue Ddc2 is recruited to double-strand breaks in an RPA-dependent manner.

    Sensing DNA damage through ATRIP recognition of RPA-ssDNA complexes. · Zou L, Elledge SJ · Science 300(5625):1542-8 2003
    PMID 12791985DOI
    What this evidence cannot show

    Abstract level. HeLa plus purified proteins plus budding yeast; no haematopoietic cell and no fusion context. The corpus already holds gene-atr and gene-chek1 and this item exists to connect them to RPA rather than to restate them - see mechanism-rpa-ssdna-recruits-atrip-atr-and-activates-chek1 for the mechanism entry and do not duplicate the ATR or CHEK1 entries.

    Evidence that argues against this (1)
    • ATR has activators that do not work this way, and the corpus should not present RPA-ssDNA as the only licence. TOPBP1 is an independent ATR activator with its own entry here, and ETAA1 activates ATR through a direct RPA contact that is distinct from the ATRIP route and has NO entry anywhere in this corpus. So 'RPA licenses ATR' is right about a necessary condition and incomplete about the mechanism, and ETAA1 is a named gap. This repository 2026
  5. E5established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper

    The nuclear pool of RPA is finite and exhaustible: stalled forks in ATR-deficient cells are initially stable but undergo nucleus-wide breakage after unscheduled origin firing generates an excess of ssDNA that exhausts the nuclear RPA pool; partial reduction of RPA accelerated fork breakage; forced elevation of RPA delayed such replication catastrophe even in the absence of ATR activity; and unscheduled origin firing induced breakage of stalled forks even in cells with active ATR.

    ATR prohibits replication catastrophe by preventing global exhaustion of RPA. · Toledo LI, Altmeyer M, Rask MB, Lukas C, Larsen DH, Povlsen LK, Bekker-Jensen S, Mailand N, Bartek J, Lukas J · Cell 155(5):1088-103 2013
    PMID 24267891DOI
    What this evidence cannot show

    Abstract level; the full text was not opened. Human tumour cell lines (MeSH 'Cell Line, Tumor'), single-cell imaging of chromatin-bound RPA as the primary readout; no haematopoietic system and no fusion. The paper's own closing move is to offer this as 'a molecular rationale for the hypersensitivity of cancer cells to ATR inhibitors', which is a rationale rather than a result, and the distinction should be kept when this is cited toward a combination.

    Evidence that argues against this (1)
    • The exhaustion model makes RPA abundance a rate-limiting variable, which predicts that RPA levels should be a resistance mechanism to ATR inhibitors - and that prediction is not, in the sources this beat read, confirmed in any patient material. The model is also cell-line based, and the RPA pool in a primary quiescent progenitor is not obviously the same quantity as in a cycling tumour line. Searched for a paper disputing the exhaustion model itself and found none; 'RPA exhaustion' as a phrase returns 18 title/abstract records, which is a real but small literature and every one this beat saw builds on the model rather than challenging it. NCBI PubMed E-utilities 2026
  6. E6strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

    RPA is loaded onto the MLL/KMT2A breakpoint cluster region after etoposide: chromatin immunoprecipitation with anti-RPA antibodies in ATM-proficient 11-4 cells showed etoposide-induced enrichment of RPA in the BCR immediately after etoposide removal which returned to normal within 30 minutes, while in ATM-deficient AT5BIVA cells the enrichment was more spatially extensive and remained high at 30 minutes; the authors conclude ATM controls the appropriate loading and release of RPA on damaged chromatin in the BCR, and they describe the BCR as the first native human chromosomal DNA locus where recombination proteins were shown to accumulate after damage by a specific agent.

    ATM modulates the loading of recombination proteins onto a chromosomal translocation breakpoint hotspot. · Sun J, Oma Y, Harata M, Kono K, Shima H, Kinomura A, et al. · PLoS One 5(10):e13554 2010
    PMID 21048951DOI
    What this evidence cannot show

    Full text read from PMC 2026-09-24. Systems: SV40-transformed ATM-deficient fibroblast line AT5BIVA and its chromosome-11-complemented derivative 11-4, plus the CML-derived B-cell line BV173 - so a fibroblast pair does the ATM comparison and NO arm carries a KMT2A fusion. Etoposide 100 uM for 10 min, a pharmacological dose. ChIP-qPCR at six primer sets around the ~20 bp hotspot inside the 8.3 kb BCR, normalised to a beta-globin control region at 11p15 - a candidate-locus assay with one internal control locus, not a genome-wide measurement, so it says nothing about specificity relative to the rest of the genome. RPA antibody: 'anti-RPA34 (LabVision)' in Methods and 'anti-RPA32' in figure legends - both are P15927 synonyms so the target is RPA2, but no clone is named (rule 74) and the paper's own supplementary figure shows the antibody immunoprecipitates both phosphorylated and unphosphorylated RPA32. Three independent experiments for the RPA panels. Single group, and the result has attracted little follow-up.

    Evidence that argues against this (1)
    • Occupancy is not translocation, and the paper's own control undermines the most attractive mechanism: there were no significant differences in BrdU incorporation after etoposide between the BCR and the beta-globin control locus, so replication-fork stalling appears similar at the two loci and the BCR-specific RPA loading is NOT explained by preferential fork stalling there. The link from 'more RPA and RAD51 load here' to 'therefore a translocation forms' is the authors' proposal, not a demonstrated causal chain - which is how this corpus's gene-rad51 entry already grades the same paper. Sun J et al. 2010
  7. E7moderate confidence 3 of 5: One strong study, or several smaller consistent ones.paper

    Targeting the kinase activities of ATR and ATM exhibits antitumoral activity in mouse models of MLL-rearranged AML.

    Targeting the kinase activities of ATR and ATM exhibits antitumoral activity in mouse models of MLL-rearranged AML. · Morgado-Palacin I, Day A, Murga M, Lafarga V, Anton ME, Tubbs A, et al. · Science Signaling 9(445):ra91 2016
    PMID 27625305DOI
    What this evidence cannot show

    Abstract-level title claim only; this paper is already cited four times in this corpus (gene-atr, cell-state-replication-stress-icl-repair, exp-replication-stress-readout-kmt2a-r, mech-transcription-replication-conflict-kmt2a-r) and those entries hold its content. It is cited HERE solely to mark that the druggable end of the RPA axis has already been taken into this disease, so that this complex entry is not read as proposing a fresh target. Mouse models; MLL-rearranged AML, not the index B-lineage KMT2A::AFF1 disease. Author list beyond the first author was not re-verified by this beat and is taken from PubMed's record.

    Evidence that argues against this (1)
    • An ATR-inhibitor result in MLL-rearranged AML does NOT validate RPA as a node, and joining the two would be the error this beat is most at risk of. ATR has RPA-independent activation routes (TOPBP1, and ETAA1 which has no entry here), and the antitumoral activity could run through either. Reading a downstream drug result backwards onto an upstream protein with no measured involvement is the same shape as reading a prey row as an interaction. This repository 2026
Connected to