CBX5 / HP1-alpha (P45973, 191 aa) - the H3K9me3 reader that recruits the writer, the writer's destroyer and the second writer; and the one member of that circuit with a published pro-leukaemic role, as part of an MLL leukaemia stem-cell programme that immortalises myeloid progenitors WITHOUT Hoxa or Meis

This is real experimental data — atoms placed by crystallography or cryo-EM, deposited by the people who did the experiment.

• THIS IS NOT A TARGET NOMINATION. CBX5 is the flattest dependency of the five H3K9-machinery genes measured here (mean -0.083, 0.4% of 1,258 lines below -0.5), it has three paralogues within eight residues that can dimerise with it, and its only pro-leukaemic evidence is murine over-expression of three genes together. A chromodomain reader of a mark whose own sign this graph disputes is the last thing in this beat that should be dosed.
• The reader is upstream of BOTH arms of the circuit, so any intervention against it moves the writer and the writer's destroyer in the same step. HP1-alpha recruits SUV39H1, ASB7 (which degrades SUV39H1) and SETDB1. Lowering HP1-alpha therefore removes a writer-recruiting activity and a writer-destroying activity at once, and the net effect on H3K9me3 at any locus is not predictable from this graph's model.
• It is a compartment-forming protein and forced local concentration is sufficient to nucleate PML nuclear bodies and induce ALT-like recombination biomarkers (PMID 41712379). Any strategy that raises or redistributes HP1-alpha carries a genome-instability liability at telomeres, in a patient population where the therapeutic goal is fewer rearrangements.
• The one MLL-fusion finding here is a MENIN-INHIBITOR RESISTANCE hypothesis, not a target. A programme that immortalises myeloid progenitors without Hoxa or Meis (PMID 19200802) describes a self-renewal route that the only approved targeted therapy in this disease does not act on. If it operates in KMT2A::AFF1 B-lineage cells, it is a mechanism of intrinsic resistance - and nobody has looked for it under a menin inhibitor.
• Do not write 'HP1' in an entry or a query. CBX5, CBX1 and CBX3 differ by eight residues, CBX5 and CBX1 both carry 'p25' in their protein names, the DSB literature names gamma and beta where this beat cares about alpha, and HP1[tiab] AND MLL[tiab] returns a Klebsiella oxytoca strain among three records. A paralogue swap here would put the wrong reader in a mechanism and the error would be invisible to every check this project runs.

1. Is the Myb/Hmgb3/Cbx5 Hoxa/Meis-independent self-renewal programme present in human KMT2A::AFF1 B-lineage cells, and does it change under a menin inhibitor? The programme is by construction outside a menin inhibitor's mechanism of action, and nobody has asked. Published transcriptional data under revumenib-class compounds would answer the second half without wet work.
2. Which HP1 paralogue is at the KMT2A locus, if any? ANSWERED negatively for all three this cycle: CBX5 (K562 and GM12878), CBX1 (K562) and CBX3 (K562 and GM12878) have zero called peaks in the gene or within 50 kb. The nearest H3K9-machinery address in either cell is G9a and TRIM28 at the KMT2A promoter, 45 kb away. What stays open is the damage state and a KMT2A-rearranged cell.
3. Does HP1-alpha's LSC function require its chromodomain? If it does not, then [E3] is not evidence about H3K9me3 at all and this graph should stop treating reader and mark as one axis. A chromodomain point mutant in the same immortalisation assay settles it.
4. Does human MLLT10/AF10 bind a human HP1 paralogue? The only published HP1-to-KMT2A-fusion-partner link is a 2001 Drosophila result on dAF10, and a PxVxL-motif inspection of human AF10 against the CBX5 chromoshadow domain costs nothing.
5. Why is CBX1/HP1-beta four times deeper than CBX5 and CBX3 in the CRISPR panel? If the three readers were functionally redundant they should score alike. This graph has no CBX1 or CBX3 entry and the asymmetry is unexplained.

1. E1established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.database

   CBX5 is UniProt P45973, entry name CBX5_HUMAN, 191 amino acids, primary gene name CBX5 with synonym HP1A, alternative protein names Heterochromatin protein 1 homolog alpha, HP1 alpha and Antigen p25, HGNC:1555 at 12q13.13, Ensembl ENSG00000094916; its feature table annotates a chromo domain at 20-78, a chromoshadow-subtype chromo domain at 121-179 and a disordered region at 70-117, and it cross-references six PDB entries. Its coded locations are Nucleus, Chromosome and Chromosome/centromere, with notes that it is a component of centromeric and pericentromeric heterochromatin, associates with chromosomes during mitosis, associates specifically with chromatin during metaphase and anaphase, and colocalises with HNRNPU in the nucleus.

   UniProtKB P45973 (CBX5_HUMAN) - Chromobox protein homolog 5 / HP1 alpha, Homo sapiens; with HGNC:1555 · UniProt Consortium; HUGO Gene Nomenclature Committee · UniProt Knowledgebase / HGNC 2026
   ▸What this evidence cannot show

   Exactly one reviewed record, gene_primary read back. The length assertion gives almost no protection against a CBX1 or CBX3 swap (185 and 183 aa). The mitotic-chromosome note is cited to PubMed:40440427, which is the CUL5-ASB7 Science paper this graph already holds as gene-asb7 - so UniProt's CBX5 record and this graph's ASB7 entry rest on the same 2025 source for the mitotic behaviour.

   ▸Evidence that argues against this (2)

   • Rule 91, and it applies to this record directly: UniProt also states 'Localizes to sites of DNA damage (PubMed:28977666)', but that paper is about CHD3 and CHD4 forming distinct NuRD complexes, and its abstract's relevant sentence is that the NuRD complexes 'interact with HP1 and rapidly accumulate at UV-induced DNA repair sites'. The species accumulating at repair sites is NuRD; the HP1 paralogue is unspecified. This entry therefore declines to carry a DNA-damage localisation on that citation.
   • A 191-residue protein consisting of two small domains and a disordered hinge has no locus-targeting element of its own, so every statement about where HP1-alpha is, is a statement about the mark and the partner that brought it. That makes an occupancy measurement more informative than any amount of protein annotation - and it also means an occupancy null could reflect the mark's absence rather than the reader's.
2. E2strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

   HP1-alpha specifically recruits the writer SETDB1. The CHAMP1 complex - a highly conserved complex of CHAMP1, POGZ and HP1-alpha - is enriched in heterochromatin, promotes heterochromatin assembly at multiple chromosomal sites including centromeres and telomeres, and promotes homology-directed repair of double-strand breaks in those regions, including at the highly compacted telomeres of ALT-positive tumour cells; the complex binds and recruits the writer methyltransferase SETDB1 to heterochromatin regions of the genome and is required for efficient DSB repair at these sites; and peripheral blood lymphocytes from individuals with CHAMP1 syndrome, an inherited neurological disorder caused by heterozygous CHAMP1 mutations, exhibit defective heterochromatin clustering and defective repair of DSBs.

   CHAMP1 complex directs heterochromatin assembly and promotes homology-directed DNA repair. · Li F, Zhang T, Syed A, Elbakry A, et al. · Nature Communications 2025
   PMID 39962076DOI
   ▸What this evidence cannot show

   Abstract level; the cell lines used for the mechanism are not named in the abstract. The recruitment is to heterochromatin at centromeres and telomeres - not to a transcribed euchromatic loop anchor like the KMT2A exon-11 element. The human primary material is a repair-defect readout in CHAMP1-syndrome lymphocytes, not a SETDB1-occupancy measurement. 2025 paper, no replication known to this project.

   ▸Evidence that argues against this (2)

   • The compartment argues against transferring this to the KMT2A locus: an HP1-alpha complex that assembles heterochromatin at centromeres and telomeres has no reported activity at an active CTCF/cohesin anchor inside a highly transcribed gene, which is the opposite chromatin state.
   • A review of this literature notes that children with CHAMP1 syndrome have developed leukaemia and suggests CHAMP1 mutations may be a cancer risk factor (PMID 42748571, DNA Repair 2026, a review used only as a map). That is a prompt to look, not evidence, and no reported CHAMP1-syndrome leukaemia is described as KMT2A-rearranged.
3. E3moderate confidence 3 of 5: One strong study, or several smaller consistent ones.paper

   In a MOUSE model of human MLL myeloid leukaemia, Cbx5 is a component of the leukaemia stem-cell maintenance programme and contributes to Hoxa/Meis-INDEPENDENT immortalisation. Leukaemia stem cells exhibit variable frequencies that correlate with the initiating MLL oncogene and are maintained in a self-renewing state by a transcriptional subprogramme more akin to embryonic stem cells than to adult stem cells; the transcription/chromatin regulatory factors Myb, Hmgb3 and Cbx5 are critical components of that programme and suffice for Hoxa/Meis-independent immortalisation of myeloid progenitors when co-expressed, establishing the cooperative and essential role of an ESC-like LSC maintenance programme ancillary to the leukaemia-initiating MLL/Hox/Meis programme; and enriched expression of LSC-maintenance and ESC-like programme genes in normal myeloid progenitors and poor-prognosis human malignancies links aberrantly self-renewing progenitor-like cancer stem-cell frequency to prognosis.

   Hierarchical maintenance of MLL myeloid leukemia stem cells employs a transcriptional program shared with embryonic rather than adult stem cells. · Somervaille TC, Matheny CJ, Spencer GJ, Iwasaki M, et al. · Cell Stem Cell 2009
   PMID 19200802DOI
   ▸What this evidence cannot show

   Abstract level. Murine, MYELOID, retroviral over-expression of three genes together at non-physiological levels; the abstract does not state which MLL fusion the LSCs came from, saying only that LSC frequency correlates with the initiating oncogene. The Cbx5 contribution is as one of three co-expressed factors, so this is not a single-gene sufficiency claim, and no loss-of-function arm for Cbx5 alone is reported in the abstract. Published 2009, before menin inhibitors existed, and it makes no claim about drug resistance. Confidence moderate because the load-bearing element for this graph - Hoxa/Meis independence - is one clause of a 2009 abstract and was not read at figure level.

   ▸Evidence that argues against this (2)

   • It points in the opposite direction to the mark CBX5 reads. mark-h3k9me3 argues H3K9 methylation restrains the fusion programme; here its reader sustains the leukaemia stem cell. The reconciliation may be that HP1-alpha's LSC function is not chromodomain-dependent at all - HP1 proteins have chromoshadow-mediated partner functions independent of H3K9me3 - in which case this result says nothing about the mark and should not be used to argue about it.
   • DepMap gives CBX5 a flat dependency (mean -0.083, 0.4% of lines below -0.5) while CBX1 is four times deeper [E6]. If HP1-alpha were broadly required for leukaemic self-renewal, a CRISPR panel containing many leukaemia lines would be expected to show something - so either the phenotype is transplant-specific and invisible to a proliferation screen, or the paralogues compensate, and neither possibility has been tested.
4. E4moderate confidence 3 of 5: One strong study, or several smaller consistent ones.paper

   In K562 leukaemia cells HP1-alpha is part of a SETDB1-dependent axis that compacts chromatin toward the nuclear periphery. HIRA expression is higher in CML patient samples than in healthy individuals; in HIRA-knockdown K562 cells, FRAP and FLIM-FRET show increased chromatin compaction and altered spatial distribution of chromatin toward the nuclear periphery; the compaction is attributed to increased H3K9me3 and HP1-alpha levels mediated by SETDB1; the increases in SETDB1 and HP1-alpha expression follow enrichment of H3.3 and loss of EZH2 recruitment with reduced H3K27me3 at the SETDB1 and HP1-alpha promoters; and the axis restricts cell proliferation and reduces expression of the BCR::ABL1 fusion protein.

   HIRA-SETDB1-H3K9me3 axis regulates chromatin architecture in leukemia cells. · Shirude MB, Devarajan A, Sahu SA, Mukherjee A, et al. · The Journal of Biological Chemistry 2026
   PMID 42055326DOI
   ▸What this evidence cannot show

   K562 is a CML-derived erythroleukaemia line and the fusion in question is BCR::ABL1, not a KMT2A fusion - this is a blood cell and a leukaemia and nothing more. Abstract level. Already cited in this graph in gene-hira, protein-huca-complex and gene-hdac7, so the finding is not new here; it is included because CBX5 had no entry to carry it and because it is the only measurement of HP1-alpha abundance in a leukaemia cell that this beat found. **Naming trap:** this paper's 'Breakpoint Cluster Region protein' is the BCR GENE at 22q11, not the KMT2A breakpoint cluster region that this graph and this beat abbreviate as the BCR.

   ▸Evidence that argues against this (2)

   • The direction is that MORE H3K9me3 and HP1-alpha restricts proliferation and lowers fusion-protein expression - consistent with mark-h3k9me3's restraining limb, but in a different fusion, a different lineage and with the fusion's expression rather than its target programme as the readout. It cannot be transferred to KMT2A::AFF1, where the fusion is not transcriptionally repressed by heterochromatin in any published experiment.
   • It is a knockdown of HIRA, so the HP1-alpha change is two steps downstream (HIRA to H3.3/EZH2 to the SETDB1 and HP1-alpha promoters to protein) and the chromatin-compaction readout is a biophysical measurement of bulk chromatin, not an occupancy measurement at any locus.
5. E5strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

   Tethering HP1-alpha is sufficient to create a subnuclear compartment. Using a targeted system to locally modulate heterochromatin features at telomeres, telomeric heterochromatin promotes telomere clustering and multiple hallmarks of APB-associated telomere processing in ALT-positive cells; molecular tethering of HP1-alpha at telomeres is sufficient to nucleate PML nuclear bodies in NON-ALT cells and, in specific contexts, to induce biomarkers of ALT-like recombination; and heterochromatin-driven PML-telomere colocalisation is inhibited by ATRX and DAXX, factors frequently mutated in ALT-positive tumours.

   Local heterochromatin enrichment promotes telomere clustering and PML nuclear body assembly at telomeres. · Taylor ER, Proctor B, Vaurs M, Wu G, et al. · Cell Reports 2026
   PMID 41712379DOI
   ▸What this evidence cannot show

   Abstract level; ALT-positive and non-ALT tumour lines, telomeres, no haematopoietic cell and no KMT2A fusion. The tethering is a synthetic recruitment system at an engineered target, so 'sufficient to nucleate' is a statement about forced local concentration and not about endogenous HP1-alpha at an endogenous locus. A 2026 paper with no replication known to this project.

   ▸Evidence that argues against this (2)

   • A forced-tethering result is the weakest kind of sufficiency claim for this graph's purposes, and it is the same family of concern as this project's condensate caveats: local over-concentration of a multivalent chromodomain protein can produce a compartment without that being the protein's endogenous function. No 1,6-hexanediol or calibrated half-bleaching control is mentioned in the abstract.
   • It is inhibited by ATRX and DAXX, both of which have entries in this graph as H3.3 depositors maintaining H3K9me3 - so the same two proteins that this graph records as supporting the mark oppose the compartment HP1-alpha nucleates here. That is a sign inconsistency inside the heterochromatin machinery itself and nothing in this beat resolves it.
6. E6established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.database

   CBX5 is not a dependency, and the three HP1 paralogues are not interchangeable in the CRISPR panel. Flattening every tissue block of the DepMap essentiality payload gives 1,258 screens with a gene effect for CBX5, mean -0.083, median -0.0808, minimum -0.6567, with 0.4% of lines below -0.5 and none below -1.0; in the 147 haematopoietic and lymphoid screens the mean is -0.0804 with 0.68% below -0.5. In the same call and data version, CBX1 gives mean -0.3494 with 16.6% of lines below -0.5 and CBX3 gives -0.0817 with 1.1% - so CBX1/HP1-beta is four times deeper than either paralogue.

   Open Targets Platform GraphQL API, target.depMapEssentiality with all tissue blocks flattened, plus meta.dataVersion · Open Targets Platform; Broad Institute DepMap · Open Targets Platform 2026
   ▸What this evidence cannot show

   Rule 83: no DepMap release in the payload, only the Open Targets data version (year 26, month 06). Rule 84: no median-split proportions; absolute thresholds only. No infant KMT2A::AFF1 line in the panel. Panel-maximum guard cleared at 1,258 screens and POLR2A was run as a live zero-screen control in the same call, returning 0 blocks and 0 screens with a passing approvedSymbol readback.

   ▸Evidence that argues against this (2)

   • Paralogue redundancy makes a flat single-gene dependency uninterpretable here: three HP1 proteins share a chromodomain and a chromoshadow domain and can dimerise with each other, so knocking out one is not a test of HP1 function. The CBX1 result argues the paralogues are NOT equivalent, which weakens the redundancy explanation and deepens the puzzle rather than resolving it.
   • The phenotype in [E3] is self-renewal measured by immortalisation and LSC frequency, which a 21-day proliferation screen does not measure. A flat Chronos score is therefore not counter-evidence to [E3], and this entry does not use it as such.

• binds →H3K9me3 - measured in KMT2A::AFF1 cells, in a supplementary figure, and it rises GENERALLY where H3K27ac falls SPECIFICALLY in the same experiment

  chromodomain reader of H3K9me2/me3 (aa 20-78); the reader whose occupancy, not just writer activity, sets the mark's level at a locus

• recruits →SUV39H1 (O43463, KMT1A) - the H3K9me3 writer: a menin effector whose binding region OVERLAPS the drug pocket at its N-terminus, a tumour suppressor in MLL-AF9 AML, recruited to double-strand breaks by CTCF, and never localised at the KMT2A locus in any cell type

  HP1 recognises pre-existing H3K9me3 and recruits SUV39H1 to methylate adjacent new histones - the self-templating loop

• recruits →ASB7 (Q9H672) - the brake on H3K9me3's own positive feedback loop, released by CDK1 in mitosis, and it reached this graph as the flattest dependency in the read queue

  HP1/CBX5 recruits ASB7 to heterochromatin, where CUL5-ASB7 ubiquitinates SUV39H1 for degradation - the same reader delivering the writer and the writer's destroyer

• recruits →SETDB1 (Q15047, KMT1E/ESET) - the second H3K9me3 writer: it represses the fusion's own targets via the PAF1 complex, it is required in fetal liver HSPCs and in leukaemic stem cells, and it is a broad dependency in half the CRISPR panel - three published directions that do not agree

  the CHAMP1-POGZ-HP1-alpha complex binds and recruits SETDB1 to heterochromatin and is required for homology-directed repair there

• supports →KMT2A::MLLT3 (MLL-AF9), t(9;11)(p21.3;q23.3)

  Cbx5 with Myb and Hmgb3 is a critical component of the MLL leukaemia stem-cell maintenance programme and the three co-expressed suffice for Hoxa/Meis-INDEPENDENT immortalisation of myeloid progenitors (mouse)

• related to →MLLT10 (AF10) — DOT1L cofactor and KMT2A fusion partner

  the only published HP1-to-fusion-partner link and it is a fly: dAF10, the Drosophila homolog of human AF10/MLLT10, associates with HP1 in vitro and in vivo and suppresses position-effect variegation

• ← regulatesHIRA - the chaperone that puts new H3.3 into transcribed gene bodies, and therefore the machine that runs this graph's clearance clock

  HIRA knockdown in K562 raises HP1-alpha via SETDB1 and shifts chromatin toward the nuclear periphery

• ← inhibitsATRX - DAXX's partner and the other arm of H3.3 deposition; in this disease its only signal is in the four lymphoid cell lines

  ATRX/DAXX inhibit the heterochromatin-driven PML-telomere colocalisation that tethered HP1-alpha nucleates

• ← related toNo H3K9me3 writer and no H3K9me3 reader has ever been localised at the KMT2A locus IN THE PUBLISHED LITERATURE, in any cell type: 20 queries, 5 alias spellings, 0 on-topic records - and CORRECTED 2026-09-22: the SUV39H1 half is no longer unanswerable from archives. ENCODE still holds zero SUV39H1 experiments (verified by three routes with controls), but GEO holds two deposits, and Q114 read both at this locus and found the writer ABSENT

  measured-not-analysed, and uniquely in TWO blood cells including the lymphoblastoid line GM12878

• ← depends onA published MLL leukaemia stem-cell programme immortalises progenitors WITHOUT Hoxa or Meis - Myb, Hmgb3 and Cbx5 - and nobody has ever asked what a menin inhibitor does to it, although this graph reads the drug's effect almost entirely through HOXA9 and MEIS1

  one of the three genes; the H3K9me3 reader, and the route by which this hypothesis arrived from an H3K9 beat

• ← testsRUN: ten ENCODE peak sets and fourteen signal tracks read at the KMT2A locus in two blood cells - not one called peak for SETDB1, CBX5, CBX1, CBX3 or TRIM28 anywhere in the gene, while G9a/EHMT2 and TRIM28 both sit on the KMT2A PROMOTER and neither touches the breakpoint cluster region

  reads CBX5 peak sets in K562 AND the lymphoblastoid line GM12878: zero called peaks in the KMT2A gene or within 50 kb

• ← testsThe deposit this graph filed as 'four SUV39H1 ChIP-seqs' is actually 77 samples in MCF-7 carrying the ENTIRE H3K9 axis in one cell type and one pipeline - writer SUV39H1, writer G9a/EHMT2, reader HP1-alpha/CBX5 (wild-type AND two phospho-mutants), co-repressor KAP1/TRIM28, erasers KDM4B and KDM4C, the mark H3K9me3, and an shKDM4B+shKDM4C eraser-knockdown arm

  the only published HP1-alpha mutant (4A, 4E) occupancy maps in a human cell

• ← related toTHE POSITIVE RESULT: the writer G9a, the reader HP1a and the scaffold KAP1 all sit on the KMT2A PROMOTER in MCF-7 at the 87.6th-98.9th percentile of 2,000 ACTIVE promoters - surviving the control that says an active promoter is ChIP-hot for everything - while none of the three is above the 75th percentile at the exon-11 anchor 52 kb away

  HP1a at the KMT2A promoter, 91.1st-94.5th percentile of active promoters - the first CBX5 occupancy at this locus that is not a zero

• ← related toRUN: endogenous HP1a/CBX5 is DEPLETED at the exon-11 anchor (median 23.94th percentile) and has the lowest KMT2A gene-body reading of any protein in the panel (30.16) - while the phosphomimetic Bio-ChIP mutant 4E reads higher than the phospho-compromised 4A at all seven intervals AND that difference falls inside the genome-wide interquartile range of the same comparison, so it carries no information

  first HP1a occupancy measurement at the KMT2A locus outside blood: DEPLETED at the exon-11 anchor (median 23.94) and the lowest protein gene-body reading in a 28-track panel (30.16)

• ← related toCBX3 / HP1-gamma (Q13185, 183 aa) - the EUCHROMATIC HP1 paralogue, and the one CTCF actually prefers: the reader that carries BRCA1/BARD1 onto H3K9-methylated break chromatin, flat in DepMap, and a fusion partner in one KMT2A::AFDN AML patient

  paralogue within 8 residues; CBX3 is euchromatic, CBX5 pericentromeric - do not substitute one for the other

• ← related toCBX1 / HP1-beta (P83916, 185 aa) - the HP1 paralogue with the fitness cost and no role in this graph's cascade: DepMap mean -0.349 with 16.6% of 1,257 lines below -0.5, four times deeper than HP1-gamma or HP1-alpha, and zero papers place it anywhere near KMT2A

  HP1 paralogue; 185 vs 191 aa, and 4x the DepMap dependency

• ← testsThe HP1 paralogue asymmetry is a guide-cross-reactivity artefact until somebody shows it is not - and if it is real, HP1-beta is doing a job the other two cannot, which is the only HP1 question in this graph with a fitness phenotype attached

  the second flat paralogue

• ← related toOutside ENCODE the resection layer is not one absence but four: 31 BRCA1, 45 CBX3, 22 KDM5A, 14 MRE11 and 8 CtIP ChIP-seq experiments exist in ChIP-Atlas, DNA2 and RAD50 have one each, and only BARD1 and EXO1 have none in any of the three archives

  29 ChIP-Atlas SRX in 12 cell types, all peak-read here and all negative at the breakpoint region - Q35 listed CBX5 as never read