Oncofusion

start / graph / gene-hdac3

Genestrong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.Q39-acetylation-cycle

HDAC3 - the only HDAC in this disease whose requirement was established genetically rather than by a drug, and the same experiment shows it is not fusion-specific

also called HD3 · RPD3-2 · histone deacetylase 3 · SMAP45

HDAC3 is an eraser enzyme that strips acetyl tags off histones. It is the one eraser whose importance in MLL-fusion leukaemia was shown by deleting the gene rather than by giving a drug - but the same experiment showed it is just as important in two completely unrelated blood cancers, so it is not something the fusion specifically needs.

The same thing, for a scientist

HDAC3 (O15379, 428 aa) is a class I deacetylase and the catalytic subunit of the NCoR/SMRT corepressor complexes. It is the **single case in this beat where an HDAC requirement in a KMT2A-fusion model rests on genetics rather than on pharmacology**. Matthews et al. depleted individual Hdacs in three genetically distinct murine malignancies - **MLL-AF9;Nras(G12D) AML**, PML-RARa APL, and Emu-Myc lymphoma - and found that **Hdac3-depleted cells were selected against in competitive assays for all three tumour types**; Hdac3 depletion in vivo significantly reduced tumour burden, the effect survived BCL-2 overexpression, caspase inhibition and Cdkn1a knockout, and the HDAC3-selective compound RGFP966 reproduced it [E1]. **That last phrase - 'for all three tumour types' - is the finding.** The design that gives HDAC3 a genuine per-enzyme attribution is the same design that shows the requirement is shared with an APL and a Myc lymphoma, i.e. it is not fusion-directed. This is the identical shape as `mechanism-kat8-mof-requirement-is-myeloid-not-fusion-specific`, arriving from the opposite half of the acetylation cycle, and it is stronger evidence than the MOF case because the comparators were actually run. Separately, combined HDAC1-3 inhibition in an MLL-AF4 reporter system **relieves an inhibitory function of wild-type MLL and overrides the dominant effect of MLL-AF4** - an allele-directed mechanism rather than a fusion-directed one [E2], the same logic this graph records for silmitasertib.

Go deeper
Why the Matthews experiment is the one that counts

Almost every HDAC claim in this disease comes from a compound. Panobinostat, vorinostat, romidepsin, chidamide and belinostat all hit several isoforms at once 1, so a positive result licenses the sentence "HDAC inhibition works here" and licenses nothing about which enzyme. Matthews et al. is the exception: individual Hdacs were depleted genetically, in vitro and in vivo, in three defined tumours 1.

What it establishes:

  • HDAC3 is required, in MLL-AF9;Nras AML and in two unrelated malignancies. Loss is not rescued by BCL-2 overexpression, by caspase inhibition, or by Cdkn1a knockout, so the mechanism is not simple apoptosis or p21-mediated arrest 1.
  • Single HDAC1 or HDAC2 depletion is not enough. The pro-apoptotic response required co-depletion of Hdac1 with Hdac2, phenocopied by the HDAC1/2-selective agent RGFP233 1. HDAC1 and HDAC2 behave as a redundant paralogue pair, which is why neither gets a dependency claim on its own.
  • A selective chemical tool exists for HDAC3. RGFP966 reduced Emu-Myc lymphoma proliferation and induced differentiation in APL 1.

What it forecloses: the reading that HDAC3 is a KMT2A-fusion dependency. The authors' framing is deliberately broad - HDAC3-selective inhibitors "could prove useful for the treatment of hematological malignancies", plural and unrestricted 1.

Applying this graph's own convergence test

mechanism-kat8-mof-requirement-is-myeloid-not-fusion-specific states the rule: a protein required by fusion family A and by something else is a convergent fusion dependency only if it is required more than in lineage-matched malignancy without the fusion.

requirementfusion armcomparator armsverdict
KAT8 / MOFMLL-AF9, NUP98-HOXA9none rununtestable from the paper; DepMap later gave d = 0.003
HDAC3MLL-AF9;NrasPML-RARa APL, Emu-Myc lymphoma - both runfails the test on the paper's own data

HDAC3 is the better-behaved case methodologically and the worse case therapeutically. MOF needed an external dataset to expose the lineage confound; HDAC3's confound is visible in the original figure legend. Emu-Myc lymphoma is B-lineage and PML-RARa is myeloid, so the requirement is not even lineage-restricted - it spans both.

This matters for how the graph reads a future trial. An HDAC3-selective agent tested in KMT2A-rearranged leukaemia would very likely show activity, and that activity would say nothing about the fusion.

The class I result in MLL-AF4 points at the wild-type allele, not the fusion

The one class I HDAC experiment done in an MLL-AF4 context reaches a conclusion the field does not usually quote. Ahmad et al. combined HDAC1-3 knockdown with HDAC inhibitors on the ALOX5 promoter and found that constitutive activation of ALOX5 by MLL-AF4 was inhibited by class I HDAC inhibitors by relieving inhibitory functions deriving from MLL, while knockdown of MLL increased the effects mediated by MLL-AF4; they summarise it as HDAC inhibitor treatment switching 'inactive MLL' into 'active MLL' and overwriting the dominant functions of MLL-AF4 2.

So the mechanism is allele-directed: the drug works by strengthening the wild-type KMT2A allele so that it outcompetes the fusion. This graph already holds one such mechanism, drug-silmitasertib, and the bite-back transfers exactly: an agent whose mechanism runs through the wild-type allele is defeated by losing that allele, and that escape needs no mutation in a drug-binding site, only loss of a copy.

The caveat is large and must be stated: the readout is a single target gene (ALOX5) in transfection and knockdown experiments, not a leukaemia phenotype, and the fusion is studied as a transcriptional effector on one promoter 2.

Facts and figures
uniprotO15379
length AA428
gene symbolHDAC3
accession resolutiongene_exact:HDAC3 AND organism_id:9606 AND reviewed:true returned exactly 1 reviewed row; primary gene name read back as HDAC3 (rule 62 applied).
classclass I histone deacetylase; catalytic subunit of the NCoR and SMRT corepressor complexes
attribution basisgenetic (shRNA depletion of individual Hdacs) plus a selective chemical tool (RGFP966)
fusion specificfalse
fusion specificity basisthe same experiment ran two non-KMT2A comparator malignancies and the requirement held in both
lineage restrictedfalse
selective tool compoundRGFP966 (HDAC3-selective); RGFP233 is HDAC1/2-selective
mll af4 mechanismclass I HDAC inhibition relieves an inhibitory function of wild-type MLL and overrides the dominant effect of MLL-AF4 at the ALOX5 promoter - allele-directed, not fusion-directed
paralogue redundancy noteHDAC1 and HDAC2 required CO-depletion for a pro-apoptotic response; HDAC3 did not
subnuclear localizationchromatin, held there by SMRT (PMID 40966497), plus a distinct nuclear-periphery pool held by the integral nuclear-envelope protein LAP2beta (PMID 16129885). The unifying fact is that HDAC3 supplies the enzyme and its partners supply the address, so 'where HDAC3 is' in any cell is a question about which scaffold that cell expresses. The lamina pool makes HDAC3 one of the few entries in this graph with a mechanistic reason to sit in lamina-associated domains.
shown for fusionNO. Measured 2026-09-21 with NCBI esearch: (HDAC3[tiab] OR histone deacetylase 3[tiab]) AND (MLL[tiab] OR KMT2A[tiab] OR MLL-AF4[tiab] OR MLL-AF9[tiab] OR mixed lineage leukemia[tiab]) AND (localiz*[tiab] OR localis*[tiab] OR subcellular[tiab] OR nuclear body[tiab] OR nuclear bodies[tiab] OR speckle*[tiab] OR condensat*[tiab] OR nucleolar[tiab] OR immunofluorescence[tiab] OR compartment*[tiab]) returns 1 record, title-checked off-topic (PMID 22240255, menin knockdown and pre-mRNA processing at IRF1). Operands: gene clause 2,448, gene AND fusion 5 - the HDAC3-and-KMT2A literature is only five papers wide, which is itself the finding. Positive control through the identical shape with (MLLT1[tiab] OR ENL[tiab]) returns 23. Zero in titles and abstracts; supporting figures not searched (brief rule 80).
models requiring it
[
  "MLL-AF9;Nras(G12D) murine AML",
  "PML-RARa APL",
  "Emu-Myc B-lymphoma"
]
clinical agents reaching it
[
  "panobinostat",
  "vorinostat",
  "romidepsin",
  "chidamide",
  "belinostat - all multi-isoform"
]
essentiality
{
  "depmap_common_essential": null,
  "note": "DepMap not queried in this beat. Any selectivity claim must be lineage-split per this graph's standing DepMap rule and checked against the pan-essentiality bar."
}
localization
{
  "subcellular": "Nucleus and cytoplasm, both experimentally annotated. UniProt O15379 lists Nucleus, Chromosome, Cytoplasm and Cytoplasm cytosol, all at ECO:0000269. HDAC3 is the class I deacetylase with the best-documented cytoplasmic pool, which matters because the leukaemia rationale in this entry is entirely nuclear.",
  "subnuclear": "TWO tethered pools, neither of them intrinsic to HDAC3. (1) CHROMATIN, via the corepressor scaffold: SMRT anchors the HDAC3 complex to chromatin and uniquely controls the chromatin binding and nuclear localisation of NCOR, GPS2 and HDAC3 in macrophages. (2) THE NUCLEAR PERIPHERY, via the nuclear envelope: LAP2beta, an integral nuclear-membrane LEM-domain protein, binds HDAC3 at the envelope and induces histone H4 deacetylation. HDAC3 itself contributes no DNA-binding or targeting module to either, so its subnuclear address is whatever its partner supplies.",
  "how_shown": "corepressor depletion with cistrome, chromatin-accessibility and H3K27ac profiling plus nuclear-localisation readout in mouse macrophages (PMID 40966497); co-immunoprecipitation at the nuclear envelope, reporter repression and H4 deacetylation with TSA control (PMID 16129885); UniProt O15379 SUBCELLULAR LOCATION comment.",
  "systems": "mouse macrophages and cultured human non-haematopoietic cells; no KMT2A-rearranged system has ever been used to localise HDAC3",
  "node": "localization-nuclear-lamina-lads-kmt2a-r"
}
cell type
{
  "expressed_in": "Not resolved to a haematopoietic cell type in this entry. HDAC3 is broadly expressed; nothing here supports a cell-type-restricted expression claim and the entry should not imply one.",
  "required_in": "Murine MLL-AF9;NrasG12D AML - and NOT only that. The one genetic dissection of HDAC dependencies in this disease depleted individual Hdacs in three genetically distinct murine malignancies and found the HDAC3 requirement in MLL-AF9;NrasG12D AML alongside PML-RARa and lymphoid models [E1], which is why this entry's own title says the same experiment shows the requirement is not fusion-specific. In human MLL-AF4 cells, combined HDAC1-3 inhibition activates the wild-type MLL allele's ALOX5 programme [E2], but that is a class-wide pharmacological result and not an HDAC3-resolved one.",
  "not_required_in": "No cell type is recorded in which HDAC3 is dispensable, and no comparison against a non-KMT2A leukaemia establishes selectivity. This is a requirement without a window.",
  "how_shown": "individual Hdac depletion across three murine malignancies with survival and disease readouts (PMID 26447190); class I HDAC inhibition with allele-specific ALOX5 reporter readout in human MLL-AF4 cells (PMID 25402609)",
  "nodes": [
    "cell-type-gmp",
    "cell-type-monocyte"
  ]
}
cell state
{
  "maintains_state": "not established. No experiment in this entry shows HDAC3 maintaining a cell identity in KMT2A-rearranged leukaemia; the readouts are survival, disease latency and a single derepressed locus. Recorded as a bounded absence rather than inferred from what HDACs generally do.",
  "regulates_state": "Derepression of the wild-type MLL allele's programme: combined HDAC1-3 inhibition induces ALOX5 promoter activity in an MLL-dependent manner in MLL-AF4 cells [E2]. This is a locus-level derepression result, and reading it as a differentiation-state change would overstate it.",
  "regulates_state_nonleukaemic": "Inflammatory and metabolic transcriptional programmes in macrophages, split non-redundantly between the two scaffolds - SMRT depletion primarily derepresses inflammation, NCOR depletion primarily metabolism [E_q66_loc2]. Kept separate because it is a different species, a different cell type and a different readout from the leukaemia work.",
  "how_shown": "allele-specific promoter reporter under class I HDAC inhibition (PMID 25402609); corepressor-depletion transcriptome, epigenome and cistrome in mouse macrophages (PMID 40966497)",
  "nodes": [
    "cell-state-differentiation-block"
  ]
}
How this could bite us back
  • HDAC3 is required in the two comparator malignancies as strongly as in the MLL-AF9 model, so an HDAC3-selective agent has no fusion-directed therapeutic window on this evidence. It would be a general haematological-malignancy agent with all of that category's toxicity and none of the targeting argument.
  • HDAC3 is the catalytic subunit of the NCoR/SMRT corepressors, which run nuclear-receptor and circadian transcription in liver, heart and brain. In an infant these are developing systems, and constitutive Hdac3 loss in mice is embryonic lethal.
  • Because HDAC1 and HDAC2 are redundant, any agent that engages HDAC3 alone leaves the HDAC1/2 pair intact and vice versa - so the tempting move is a broader inhibitor, which returns the field to exactly the pan-class toxicity that motivated isoform selectivity in the first place.
  • An allele-directed class I mechanism is silently defeated by loss of the wild-type KMT2A allele, a common event that requires no drug-binding-site mutation and would not be detected by any resistance assay looking for one.
  • HDAC3 has no address of its own: SMRT anchors it to chromatin and controls its nuclear localisation, and LAP2beta holds a separate pool at the nuclear envelope (PMIDs 40966497, 16129885). A catalytic-site inhibitor therefore acts on every pool at once, and any attempt to gain selectivity by targeting the enzyme rather than the scaffold gives up the only axis on which HDAC3's contexts differ.
  • Scaffold identity is an untested resistance variable. SMRT and NCOR are non-redundant, and SMRT depletion alone repositions HDAC3 (PMID 40966497). A leukaemia with a shifted SMRT:NCOR balance would present a differently localised target to the same drug, and no HDAC3 study in this disease reports either scaffold's level.
  • The nuclear-periphery pool is an essential-function problem rather than a leukaemia one: LAP2beta-HDAC3 repression covers E2F members, p53 and NF-kappaB (PMID 16129885). Relieving repression at the lamina in a normal cell is not a benign off-target - it touches the p53 axis directly.
  • Selectivity: the genetic experiment that established the HDAC3 requirement in MLL-AF9 AML found requirements in PML-RARa and lymphoid malignancies too (this entry's E1). Every clinical agent reaching HDAC3 is class-wide, so both the target and the chemistry are non-specific, and a trial rationale built on 'HDAC3 in KMT2A-rearranged leukaemia' has neither specificity to stand on.
Still open
  1. Is HDAC3 required in a KMT2A-fusion LYMPHOID leukaemia? Every genetic HDAC3 result in this disease comes from MLL-AF9;Nras myeloid models, and the index disease is infant B-ALL.
  2. Has RGFP966 or any HDAC3-selective agent ever been tested against a KMT2A-rearranged ALL model? Nothing found in this beat.
  3. Does the pro-apoptotic HDAC1/HDAC2 co-depletion requirement mean a clinical HDAC1/2-selective agent needs both engaged at once, and is that achievable at a tolerable dose in an infant?
  4. Does the ALOX5 allele-directed result generalise beyond one promoter to genome-wide wild-type KMT2A occupancy, the way the silmitasertib result was shown to?
  5. Which scaffold carries HDAC3 in a KMT2A-rearranged blast, SMRT or NCOR? The macrophage result says the answer determines where HDAC3 sits and which programme is derepressed when it is removed, and nobody has measured either scaffold in this disease.
  6. Does HDAC3 occupy lamina-associated domains in a KMT2A-rearranged cell? This graph already carries localization-nuclear-lamina-lads-kmt2a-r as a counted negative; HDAC3 is the first entry with a published mechanism for being there, so it is the natural protein with which to test the compartment.
  7. Is the HDAC3 requirement in MLL-AF9 AML separable from the requirement in PML-RARa and lymphoid disease by any dose or schedule, or is it simply a shared dependency? No published experiment compares them in one system.
Every source behind this entry (5)
  1. E1established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper

    Individual Hdac depletion in three genetically distinct murine malignancies - MLL-AF9;Nras(G12D) AML, PML-RARa APL and Emu-Myc lymphoma - showed Hdac3-depleted cells were selected against in competitive assays for all three tumour types; decreased proliferation on Hdac3 knockdown was not prevented by BCL-2 overexpression, caspase inhibition or Cdkn1a knockout, Hdac3 depletion in vivo significantly reduced tumour burden, the HDAC3 inhibitor RGFP966 reduced Emu-Myc proliferation and induced differentiation in APL, and genetic co-depletion of Hdac1 with Hdac2 was pro-apoptotic and was phenocopied by the HDAC1/2-specific agent RGFP233.

    Functional-genetic dissection of HDAC dependencies in mouse lymphoid and myeloid malignancies. · Matthews GM, Mehdipour P, Cluse LA, Falkenberg KJ, Wang E, Roth M, et al. · Blood 126(21):2392-2403 2015
    PMID 26447190DOI
    What this evidence cannot show

    Mouse models throughout; the MLL-AF9 arm carries a cooperating Nras(G12D) mutation, so it is not a pure fusion model. shRNA depletion, not knockout, so residual protein is present and the magnitude of the requirement is a lower bound. No lymphoid KMT2A-fusion arm and no KMT2A::AFF1 arm. Only the abstract was read for this entry; the PMCID is PMC4653767 and the full text was not retrieved in this beat. Confidence is 'established' for the depletion results within these three models and no further.

    Evidence that argues against this (1)
    • The same paper is the counter-evidence to its own most useful reading. Because Hdac3 depletion selected against cells in all three tumour types - one KMT2A-fusion AML, one PML-RARa APL and one B-lineage Myc lymphoma - the requirement is neither fusion-specific nor lineage-specific, and an HDAC3-selective agent showing activity in KMT2A-rearranged leukaemia would be showing an effect it also has in unrelated malignancies. Matthews GM et al. 2015
  2. E2moderate confidence 3 of 5: One strong study, or several smaller consistent ones.paper

    Combined inhibition of HDAC1-3 induces ALOX5 promoter activity in an MLL-dependent manner; constitutive activation of ALOX5 by MLL-AF4 was inhibited by class I HDAC inhibitors by relieving inhibitory functions deriving from MLL; conversely a knockdown of MLL increased the effects mediated by MLL-AF4, so HDAC inhibitor treatment switches 'inactive MLL' into 'active MLL' and overwrites the dominant functions deriving from MLL-AF4.

    Inhibition of class I HDACs abrogates the dominant effect of MLL-AF4 by activation of wild-type MLL. · Ahmad K, Katryniok C, Scholz B, Merkens J, Loscher D, Marschalek R, et al. · Oncogenesis 3(11):e127 2014
    PMID 25402609DOI
    What this evidence cannot show

    The readout is transcription of a single target gene, ALOX5, in promoter-reporter and knockdown experiments, not a leukaemia phenotype, cell viability or an animal model. HDAC1, HDAC2 and HDAC3 were knocked down in combination for the load-bearing result, so the paper does not attribute to a single class I enzyme either. Trichostatin A, a pan-HDAC inhibitor, is the primary chemical tool. The mechanism proposed is allele-directed, so it predicts nothing in a cell that has lost or silenced the wild-type KMT2A allele. Abstract read; PMCID PMC4259963, full text not retrieved in this beat.

    Evidence that argues against this (1)
    • An allele-directed mechanism is defeated by loss of the wild-type allele, which requires no mutation in any drug-binding site. The same structural vulnerability applies to the CK2 route this graph records under drug-silmitasertib: pharmacological CKII inhibition blocks taspase1-dependent MLL1 processing, increases MLL1 stability, and results in the displacement of the MLL chimeras from chromatin - a mechanism that works only while an intact wild-type MLL1 allele remains to be stabilised. Zhao Z et al. 2019
  3. E3established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.database

    Human HDAC3 is UniProt O15379, primary gene name HDAC3, 428 amino acids, primary protein name 'Histone deacetylase 3'.

    UniProtKB O15379 (HDAC3_HUMAN), Histone deacetylase 3 · UniProt Consortium · UniProtKB, retrieved 2026-09-20 2026
    What this evidence cannot show

    Resolved per rule 62: gene_exact:HDAC3 AND organism_id:9606 AND reviewed:true returned exactly one reviewed row and the PRIMARY gene name was read back as HDAC3 before the accession was stored. Note that the sibling query for HDAC7 in this same beat returned TWO reviewed rows with HDAC9 first, so the single-row outcome here was confirmed rather than assumed.

    Evidence that argues against this (1)
    • A reviewed database record fixes the canonical sequence on the date it was read and nothing more. It does not establish which isoform is expressed in KMT2A-rearranged blasts, and this beat's own findings turn on distinctions a canonical record hides - HDAC9's catalytically inactive MITR form, and the nomenclature collision in which HDAC9's protein name reads 'Histone deacetylase 7B'. Accession and length are asserted here so a swap fails the validator, not because they carry biological weight. Vega-Garcia N et al. 2018
  4. E_q66_loc1moderate confidence 3 of 5: One strong study, or several smaller consistent ones.paper

    HDAC3 has a NUCLEAR-PERIPHERY pool held by an integral nuclear-membrane protein: LAP2beta, an integral nuclear-envelope protein carrying the chromatin-binding LEM domain, interacts with HDAC3 AT THE NUCLEAR ENVELOPE, LAP2beta represses a range of transcription factors including E2F members, p53 and NF-kappaB, trichostatin A abrogates that repressive activity, and LAP2beta induces histone H4 deacetylation - a repressive complex composed of an integral nuclear membrane protein and a histone modifier at the nuclear periphery.

    PMID 16129885DOI
    What this evidence cannot show

    Cultured human cells in a nuclear-envelope-biology setting, not leukaemia and not KMT2A-rearranged. The interaction is shown biochemically and by reporter repression rather than by quantitative imaging of the endogenous HDAC3 pool, so what fraction of HDAC3 sits at the periphery is unmeasured. The TSA result establishes that the repression is deacetylase-dependent, not that it is HDAC3-dependent specifically - TSA is pan-class-I/II. Read: abstract only. Identifiers copied field by field from the efetch record fetched 2026-09-21, ArticleId anchored per brief rule 81.

    Evidence that argues against this (1)
    • The dominant determinant of where HDAC3 is turns out not to be the envelope but the corepressor scaffold, and specifically SMRT rather than NCOR: in macrophages SMRT uniquely controls the chromatin binding AND THE NUCLEAR LOCALISATION of NCOR, GPS2 and HDAC3, acting as the chromatin anchor for the whole corepressor complex. An HDAC3 compartment claim made without stating which scaffold is present is therefore underdetermined, and the two homologous scaffolds are not interchangeable - SMRT depletion primarily derepresses inflammatory pathways, NCOR depletion primarily metabolic ones. Efthymiadou A; Gu C; Wang C; Wang H; Li Z; Garcia-Irigoyen O; Fan R; Treuter E; Huang Z 2025
  5. E_q66_loc2strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

    HDAC3's chromatin address belongs to SMRT, not to HDAC3: in mouse macrophages and bone-marrow-derived macrophages, SMRT uniquely controls the chromatin binding and nuclear localisation of NCOR, GPS2 and HDAC3, acting as the chromatin anchor for the corepressor complex, and SMRT and NCOR are non-redundant - SMRT depletion primarily upregulates inflammation-related pathways while NCOR depletion primarily upregulates metabolism-related pathways.

    SMRT anchors the HDAC3 corepressor complex to chromatin to regulate inflammatory and metabolic pathways in macrophages. · Efthymiadou A; Gu C; Wang C; Wang H; Li Z; Garcia-Irigoyen O; Fan R; Treuter E; Huang Z · Nucleic Acids Res 2025
    PMID 40966497DOI
    What this evidence cannot show

    Mouse RAW264.7 macrophages and mouse bone-marrow-derived macrophages: murine, myeloid-differentiated, and not a leukaemia. Macrophages are a reasonable proxy for none of the index disease's cell types, and the claim is written with that system first (brief rule 73). The localisation result is a cistrome/fractionation readout in depletion conditions, so it shows dependence rather than the steady-state distribution. Read: abstract only.

    Evidence that argues against this (1)
    • A second, scaffold-independent tether exists at the nuclear periphery - LAP2beta, an integral nuclear-membrane protein, binds HDAC3 there and drives H4 deacetylation. So 'SMRT is the anchor' is a statement about the corepressor-complex pool and not about all HDAC3, and at the lamina the anchor is a membrane protein instead. Two anchors means a scaffold-directed drug leaves one pool in place. Somech R; Shaklai S; Geller O; Amariglio N; Simon AJ; Rechavi G; Gal-Yam EN 2005
Connected to