Oncofusion

start / graph / protein-huca-complex

Proteinstrong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.Q37-histone-chaperones

The HUCA complex (HIRA/UBN1/CABIN1/ASF1a) - four proteins, one job: put H3.3 into chromatin without waiting for S phase

also called HUCA · HIRA complex · HIRA/UBN1/CABIN1/ASF1a · HIR complex

Four proteins work together as one machine that inserts a replacement histone into genes that are actively being read, at any point in the cell cycle. Each part does something different, and knowing which part does what tells you which one you could interfere with.

The same thing, for a scientist

HUCA is the human counterpart of yeast HIR, assembled from recombinant HIRA, UBN1, CABIN1 and ASF1a, with **HIRA acting as the scaffold** that brings the other three together [E1]. The division of labour is unusually well mapped for a chaperone complex, and it matters because it says which subunit is the druggable specificity determinant rather than the structural core: **UBN1's Hpc2-related domain is the H3.3-versus-H3.1 discriminator**, reading H3.3 G90 [E2]; **UBN1 also carries a separate middle domain that dimerises and a DNA-binding region**, giving a route from an H3.3/H4 dimer to an (H3.3/H4)2 tetramer on DNA [E3]; **HIRA itself must trimerise** for the complex to work [E4]; and **CABIN1 is a functional member rather than a passenger**, regulating an overlapping, H3.3-enriched gene set with HIRA [E1]. This graph already files multi-subunit machines as their own entries (`protein-fact-complex`, `protein-nelf-complex`, `protein-little-elongation-complex`). HUCA is the H3.3 counterpart and it had no entry.

Go deeper
Who does what, and why that is the interesting part
subunitUniProtlengthrole in the complex
HIRAP541981,017scaffold; must trimerise for activity 14
UBN1Q9NPG31,134H3.3 specificity (HRD reads H3.3 G90); dimerisation middle domain; DNA binding 23
CABIN1Q9Y6J02,220functional member, yeast Hir3p orthologue; senescence heterochromatinisation 1
ASF1AQ9Y294204upstream H3-H4 donor shared with CAF-1 5

The specificity sits in UBN1, not HIRA. A reader who assumes the scaffold is the target would be aiming at the wrong subunit. And DAXX - the other H3.3-specific chaperone - contacts the histone at nearly identical points around H3.3 G90 by a distinct mechanism 2, which is the clearest structural statement available that the two H3.3 routes converge on the same recognition problem and solved it twice.

What this complex is NOT, in this disease

No subunit of HUCA has been studied in a KMT2A-rearranged model. HIRA AND (MLL OR KMT2A OR "mixed lineage leukemia") returns 0 including aliases (gene-hira E6), and the ASF1 leukaemia literature that does exist is AML and CML rather than KMT2A-rearranged (gene-asf1b).

And the DepMap picture is the opposite of a complex-level dependency: HIRA -0.082, UBN1 +0.029, CABIN1 -0.104, ASF1A +0.108 in KMT2A-rearranged versus other leukaemia lines, none of them selective, two of them pointing the wrong way 6. This project's analyses/depmap-complex-level-selectivity exists because subunits of a real functional module tend to move together. HUCA's subunits do not move together at all, which is either evidence that the complex is not limiting in these lines, or evidence that CRISPR viability is the wrong readout for a rate-setting chaperone. Both readings are in hyp-kmt2a-aff1-alters-h3-3-deposition-at-fusion-targets.

Facts and figures
complex nameHUCA (HIRA/UBN1/CABIN1/ASF1a)
yeast counterpartHIR complex (Hir1p, Hir2p, Hir3p, Hpc2p, Asf1p)
deposition modereplication-independent
cargoprotein-histone-h3-3
subunits
gene-asf1a204.0gene-asf1a: 204gene-hira1,017gene-hira: 1017gene-ubn11,134gene-ubn1: 1134gene-cabin12,220gene-cabin1: 2220
subunits — the numbers behind the chart
[
  {
    "gene": "gene-hira",
    "uniprot": "P54198",
    "length_aa": 1017,
    "role": "scaffold, trimerises"
  },
  {
    "gene": "gene-ubn1",
    "uniprot": "Q9NPG3",
    "length_aa": 1134,
    "role": "H3.3 specificity via HRD; DNA binding; dimerisation"
  },
  {
    "gene": "gene-cabin1",
    "uniprot": "Q9Y6J0",
    "length_aa": 2220,
    "role": "functional member; yeast Hir3p orthologue"
  },
  {
    "gene": "gene-asf1a",
    "uniprot": "Q9Y294",
    "length_aa": 204,
    "role": "upstream H3-H4 donor, shared with CAF-1"
  }
]
How this could bite us back
  • Every subunit of this complex is expressed in every dividing cell and the complex packages transcribed chromatin genome-wide; a general block on H3.3 deposition has no described window in any tissue.
  • UBN1 is the specificity subunit and is the most attractive target on that basis - but it is also the subunit with the LEAST evidence of being limiting (DepMap difference +0.029, pan-essentiality -0.130), so targeting the tractable part may do nothing.
  • Blocking replication-independent deposition slows nucleosome turnover, which by this graph's own clearance-clock argument would PROLONG H3K79 methylation rather than clear it - a chaperone inhibitor combined with a DOT1L inhibitor is predicted to be antagonistic, not synergistic.
Still open
  1. Is the HUCA complex present and intact in a KMT2A-rearranged blast? Nobody has blotted it, and the four subunits' DepMap profiles give no reason to assume they are co-limiting.
  2. Because UBN1's HRD and DAXX both read H3.3 G90 by distinct mechanisms, is there a small molecule or peptide that discriminates the two H3.3 routes? That would be the only way to ask the gene-body-versus-heterochromatin question pharmacologically.
  3. Does the elongation machinery this graph already holds (FACT, SPT6, PAF1C) hand histones to HUCA, or are the two networks independent at fusion target genes?
Every source behind this entry (6)
  1. E1established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper

    HIRA, UBN1, CABIN1 and ASF1a form a quaternary HUCA complex assembled from recombinant proteins, HIRA acts as the scaffold bringing UBN1, ASF1a and CABIN1 together, CABIN1 is the likely orthologue of yeast Hir3p, and HIRA and CABIN1 regulate overlapping gene sets enriched in H3.3.

    Human CABIN1 is a functional member of the human HIRA/UBN1/ASF1a histone H3.3 chaperone complex. · Rai TS, Puri A, McBryan T, et al. · Molecular and Cellular Biology 2011
    PMID 21807893DOI
    What this evidence cannot show

    Abstract read from efetch 2026-09-20; full text not read. Reconstitution from recombinant proteins plus co-immunoprecipitation at ectopic and endogenous levels in human fibroblasts. Establishes that the four proteins can form a complex, not the stoichiometry or the fraction of cellular HIRA that is in it. No haematopoietic cell.

    Evidence that argues against this (1)
    • A complex defined by co-purification is not a complex defined by requirement: functional activity needs HIRA trimerization, so parts that co-purify with a HIRA fragment are not thereby shown to be sufficient for deposition. Ray-Gallet D et al. 2018
  2. E2established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper

    The conserved UBN1 Hpc2-related domain is an H3.3-specific binding domain that preferentially binds H3.3/H4 over H3.1/H4; crystallography and mutation identify conserved UBN1-HRD residues and H3.3 G90 as the determinants, and comparison with the unrelated H3.3-specific chaperone DAXX shows nearly identical points of chaperone-histone contact near H3.3 G90 arrived at by a distinct mechanism.

    Ubinuclein-1 confers histone H3.3-specific-binding by the HIRA histone chaperone complex. · Ricketts MD, Frederick B, Hoff H, et al. · Nature Communications 2015
    PMID 26159857DOI
    What this evidence cannot show

    Abstract read from efetch 2026-09-20; full text not read. In vitro binding and X-ray crystallography of a domain fragment with a histone dimer - established for the isolated interaction, and a preference measured in vitro is not a demonstration that the intact complex discriminates variants in a cell. No residue coordinates are stored in this entry because they were not read from the deposit, and this project's rule is that a span must be derived by alignment to a freshly fetched UniProt sequence rather than read off a structure.

    Evidence that argues against this (1)
    • In mouse ES cells the deposition specificity attributed to UBN1/2 is for cis-regulatory elements specifically, not for H3.3 generically, so 'UBN1 confers H3.3 specificity' and 'UBN1 determines where H3.3 goes' are different claims and only the first is structural. Xiong C et al. 2018
  3. E3strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

    A conserved central middle domain of UBN1 forms dimers and binds H3/H4 without discriminating H3.1 from H3.3, and a separate nearby DNA-binding region binds DNA through conserved lysines, suggesting HIRA-mediated deposition proceeds by UBN1 associating with DNA and dimerising to form an (H3.3/H4)2 tetramer before chromatin deposition.

    The HIRA histone chaperone complex subunit UBN1 harbors H3/H4- and DNA-binding activity. · Ricketts MD, Dasgupta N, Fan J, et al. · Journal of Biological Chemistry 2019
    PMID 31040182DOI
    What this evidence cannot show

    Abstract read from efetch 2026-09-20; full text not read. Domain-level biochemistry on fragments; the mechanism at the end of the abstract is explicitly a proposal ('suggest a mechanism'), which is why this is graded strong and not established. Secondary structure of the middle domain is predicted, not solved.

    Evidence that argues against this (1)
    • A later study of the same system attributes the driving force to histone tetramerization and histone-DNA binding rather than to a UBN1-led ordering of events, so the step order in the proposed mechanism is not settled. Vogt A et al. 2024
  4. E4strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

    Functional activity of the H3.3 histone chaperone complex HIRA requires trimerization of the HIRA subunit.

    Functional activity of the H3.3 histone chaperone complex HIRA requires trimerization of the HIRA subunit. · Ray-Gallet D, Ricketts MD, Sato Y, et al. · Nature Communications 2018
    PMID 30082790DOI
    What this evidence cannot show

    Title and record read from efetch 2026-09-20; the abstract was not analysed in depth and the full text was not read, so the assay establishing 'functional activity' is not characterised here. Cited for the architectural point only. Note that `gene-hira` cites this same PMID as counter-evidence rather than support; that is deliberate and not a contradiction - the trimerization requirement supports the complex's architecture here and simultaneously qualifies any reading of HIRA as a simple scaffold there.

    Evidence that argues against this (1)
    • No published disagreement was found. The relevant limit is scope: a requirement for oligomerisation established in a reconstituted or cell-line system says nothing about whether the complex is limiting in a leukaemic blast. Pchelintsev NA et al. 2013
  5. E5established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper

    The purified H3.1 and H3.3 deposition machineries contain distinct chaperones - CAF-1 and HIRA respectively - and both are fed by the same upstream H3-H4 handling, so ASF1 sits upstream of the branch between replication-coupled and replication-independent deposition.

    PMID 14718166DOI
    What this evidence cannot show

    Abstract read from efetch 2026-09-20; full text not read. The branch-point framing of ASF1 is this entry's summary of the two-pathway result plus the subsequent HUCA reconstitution, not a sentence in the 2004 abstract. Tagged-histone purification from HeLa.

    Evidence that argues against this (1)
    • Interactomics of the whole H3-H4 chaperone network shows it is a network with cross-talk rather than a clean binary branch - including a previously uncharacterised ASF1-SPT2 co-chaperone complex and a DAXX route that methylates histones before deposition. Carraro M et al. 2023
  6. E6moderate confidence 3 of 5: One strong study, or several smaller consistent ones.dataset

    In DepMap CRISPR gene effect across 11 KMT2A-rearranged and 43 other leukaemia lines, the four HUCA subunits do not behave as a module: HIRA -0.082, UBN1 +0.029, CABIN1 -0.104 and ASF1A +0.108 (index minus other-leukaemia difference), with pan-essentialities of -0.814, -0.130, -0.151 and -0.187 respectively.

    What this evidence cannot show

    Run 2026-09-20. Moderate because no permutation null was computed and none of these numbers is a hit under this project's conventions; they are reported to show the ABSENCE of a coherent module signal, which is a weaker claim than a positive one and correspondingly more robust. HIRA's pan-essentiality of -0.81 versus UBN1's -0.13 also means the four subunits are not comparably measured: only one of them is near the Chronos floor.

    Evidence that argues against this (1)
    • Non-coherence across subunits is also what you see when one subunit has an essential job outside the complex. HIRA's much deeper pan-essentiality than UBN1's is compatible with HIRA mattering for something other than H3.3 deposition, in which case the complex-coherence test does not apply. Shirude MB et al. 2026
Connected to