HDAC (Histone Deacetylase Family, Class I/II/IV)
Histone deacetylases (HDACs) are a family of eleven Zn2+-dependent enzymes that remove acetyl groups from lysine residues on histone tails and — critically — on ~3,600 non-histone substrates including p53, alpha-tubulin, HSP90, and FOXO transcription factors 1. They are the catalytic counterparts to histone acetyltransferases (HATs such as cbp-p300) and are central regulators of chromatin state, gene expression, and cellular stress responses. In aging, the global balance of histone acetylation shifts — the epigenetic-alterations hallmark — and HDAC activity is implicated in both protective and deleterious roles depending on cellular context and HDAC class.
Disambiguation: This page covers the Zn2+-dependent HDACs (Class I, IIa, IIb, IV; HDAC1–11). The NAD+-dependent Class III deacetylases — sirtuins (SIRT1–7) — are a structurally and mechanistically distinct family; see sirtuin (verified).
Canonical-ID note: UniProt frontmatter uses HDAC1 (Q13547) as the family representative — the founding and most-studied nuclear member. HDAC3 (O15379) is the most aging-relevant member for cardiac and inflammatory biology. Both are Swiss-Prot (manually curated) entries.
Family overview
Eleven classical HDACs are organized into four classes based on yeast homologs and subcellular distribution 2:
| Class | Members | Primary localization | Key features |
|---|---|---|---|
| Class I | HDAC1, 2, 3, 8 | Nuclear (constitutive) | Broad transcriptional repressor activity; form large multi-protein co-repressor complexes |
| Class IIa | HDAC4, 5, 7, 9 | Nucleocytoplasmic (signal-regulated shuttle) | Tissue-specific (heart, muscle, brain); shuttle in/out of nucleus in response to CaMKII/PKD signaling; low intrinsic deacetylase activity; require HDAC3 for activity in some complexes |
| Class IIb | HDAC6, 10 | Primarily cytoplasmic | HDAC6 has tandem catalytic domains; primary substrates are alpha-tubulin and HSP90 (non-histone); HDAC10 deacetylates polyamines |
| Class IV | HDAC11 | Nuclear | Shares features of Class I and II; deacetylates both histones and non-histone substrates; least studied |
UniProt IDs for each family member (for Dataview queries via complex-subunits:):
| Member | UniProt | NCBI Gene | HGNC symbol |
|---|---|---|---|
| HDAC1 | Q13547 | 3065 | HDAC1 |
| HDAC2 | Q92769 | 3066 | HDAC2 |
| HDAC3 | O15379 | 8841 | HDAC3 |
| HDAC4 | P56524 | 9759 | HDAC4 |
| HDAC5 | Q9UQL6 | 10014 | HDAC5 |
| HDAC6 | Q9UBN7 | 10013 | HDAC6 |
| HDAC7 | Q8WUI4 | 51564 | HDAC7 |
| HDAC8 | Q9BY41 | 55869 | HDAC8 |
| HDAC9 | Q9UKV0 | 9734 | HDAC9 |
| HDAC10 | Q969S8 | 83933 | HDAC10 |
| HDAC11 | Q96DB2 | 79885 | HDAC11 |
Mechanism of action
All Zn2+-dependent HDACs share a conserved DAC (deacetylase) catalytic domain that coordinates a single Zn2+ ion. The Zn2+ activates a water molecule for nucleophilic attack on the acetyl-lysine carbonyl, releasing acetate and regenerating the free amine on lysine 3. The deacetylation reaction is rapid and does not require a cofactor (distinguishing classical HDACs from sirtuins, which require NAD+).
Histone substrates: the core lysine residues on H3 and H4 tails — H3K9ac, H3K14ac, H3K27ac, H4K16ac — that are removed by Class I HDACs are marks of active transcription and open chromatin. Removal by HDACs promotes chromatin compaction and transcriptional repression.
Non-histone substrates: The Choudhary et al. 2009 quantitative mass-spectrometry survey identified ~3,600 acetylation sites across ~1,750 proteins — the majority non-histone 1. Key aging-relevant non-histone substrates include:
- p53 — HDAC1 and SIRT1 deacetylate p53 at K382, promoting MDM2-mediated degradation and modulating senescence/apoptosis fate. See p53.
- FOXO transcription factors — HDAC3 deacetylates FOXO3a; acetylation status affects nuclear localization and transcriptional activity of this longevity-linked TF.
- alpha-tubulin — HDAC6 (cytoplasmic) deacetylates K40 on alpha-tubulin, regulating microtubule stability and intracellular transport; relevant to neuronal aging and proteostasis.
- HSP90 — HDAC6 deacetylation promotes HSP90 chaperone activity; hyperacetylation upon HDAC6 inhibition reduces HSP90 client loading.
- NF-κB p65 (RelA) — HDAC3 and SIRT1 deacetylate K310, modulating NF-κB transcriptional output in the inflammatory SASP program. See nf-kb and cbp-p300.
Major co-repressor complexes
Class I HDACs do not act as isolated enzymes — they are obligate components of large multi-subunit co-repressor complexes that determine which genomic loci are silenced 3:
| Complex | HDAC subunit(s) | Other key components | Primary function |
|---|---|---|---|
| Sin3A/B | HDAC1, HDAC2 | SIN3A/B scaffold, SAP30, RBP1 | Transcriptional repression at promoters; p53 deacetylation |
| NuRD (nucleosome remodeling and deacetylase) | HDAC1, HDAC2 | CHD3/4 ATPase, MBD2/3, MTA1/2/3 | Coupled chromatin remodeling + deacetylation; developmental gene silencing |
| CoREST | HDAC1, HDAC2 | LSD1 (KDM1A), REST, RCOR1 | Neuronal gene silencing; REST is a key neuronal identity repressor |
| NCoR/SMRT | HDAC3 | NCoR1/2, GPS2, TBL1, TBLR1 | Nuclear receptor co-repressor; HDAC3 requires NCoR/SMRT DAD domain for full deacetylase activity |
The Class IIa HDACs (HDAC4, 5, 7, 9) have minimal intrinsic deacetylase activity — in many contexts, they function as scaffolds that recruit HDAC3 (via the NCoR/SMRT complex) to specific promoters. Their nucleocytoplasmic shuttle is controlled by phosphorylation: CaMKII and PKD phosphorylate specific serine residues, creating 14-3-3 binding sites that anchor Class IIa HDACs in the cytoplasm, de-repressing their target genes.
Role in aging
Aging-associated histone acetylation changes
Global histone acetylation patterns shift with age. H4K16ac — a mark of active transcription and a suppressor of higher-order chromatin compaction — declines in aged tissues. H3K9ac and H3K14ac at active gene promoters also fall, consistent with an overall drift toward chromatin compaction and transcriptional repression of stress-response and metabolic genes. The underlying mechanism involves both HDAC upregulation in certain contexts and loss of counterbalancing HAT activity (cbp-p300 decline has been documented in aged brains) 4.
| Dimension | Status |
|---|---|
| Pathway conserved in humans? | yes — HAT/HDAC balance is conserved; age-associated histone mark changes documented in human and mouse tissues |
| Phenotype conserved in humans? | partial — transcriptional drift is documented; HDAC-specific contribution vs. sirtuin decline vs. methyltransferase changes is not fully resolved |
| Replicated in humans? | no — no aging-endpoint human trial for HDAC-targeted intervention exists; evidence is mechanistic and model-organism |
needs-human-replication — direct evidence that modulating Zn2+-dependent HDACs improves human aging endpoints is absent.
HDAC1 in brain aging and DNA repair
HDAC1 interacts with OGG1 (the base excision repair enzyme that clears 8-oxoguanine) to regulate oxidative DNA damage repair in neurons. Pao et al. 2020 showed in aged mice (~17 months) and the 5XFAD Alzheimer model that:
- HDAC1 activity declines in aged neurons
- HDAC1 loss-of-function increases 8-oxoG accumulation at gene promoters via impaired OGG1 AP-lyase activity
- Pharmacological HDAC1 activation with exifone (50 mg/kg/day IP, 4 weeks) restored OGG1 activity and reduced hippocampal 8-oxoG in 17-month C57BL/6J mice; exifone also improved contextual fear conditioning and hippocampal LTP in 8-month-old 5XFAD mice (Fig. 5g–i) 5
Note on OGG1 acetylation site: the paper demonstrates HDAC1 deacetylates OGG1 to enhance AP-lyase cleavage activity; mass spectrometry identified the specific p300-acetylated lysine residues on OGG1. The K341 site attribution originates from Bhakat et al. 2006 (ref 47 in Pao 2020), not from Pao 2020 itself — the wiki’s original K341 claim should be cross-checked against Bhakat 2006 before propagating. needs-replication
This establishes a specific non-chromatin role for HDAC1 in brain aging: deacetylation of OGG1 is required for efficient base excision repair. See ogg1 (verified).
HDAC3 in cardiac and inflammatory aging
HDAC3, as the enzymatic core of the NCoR/SMRT complex, is required for maintaining metabolic gene programs in cardiomyocytes. Cardiac-specific HDAC3 loss-of-function in mice produces severe cardiac hypertrophy and metabolic defects (impaired fatty acid oxidation gene expression). HDAC3 is also the principal deacetylase of NF-κB p65 at K310, modulating inflammatory gene transcription; HDAC3 loss in macrophages derepresses inflammatory cytokines relevant to the SASP 4. needs-replication — the aging-context specificity of HDAC3 cardiac effects requires further mechanistic confirmation.
HDAC6 in neurodegeneration and proteostasis
HDAC6 is the primary cytoplasmic deacetylase and is uniquely structured with two tandem catalytic domains and a C-terminal ubiquitin-binding ZnF-UBP domain. Its aging-relevant activities include:
- Tubulin deacetylation — K40 deacetylation of alpha-tubulin destabilizes microtubules; aged neurons show reduced tubulin acetylation, impairing axonal transport and contributing to neurodegeneration
- Aggresome formation — HDAC6 links ubiquitinated misfolded proteins to dynein/dynactin for retrograde transport to the aggresome; relevant to loss-of-proteostasis
- HSP90 regulation — HDAC6 inhibition hyperacetylates HSP90, disrupting its client-loading activity
HDAC6 inhibitors (tubastatin A, ricolinostat/ACY-1215) have been explored in preclinical neurodegeneration models. Ricolinostat entered Phase I/II trials in multiple myeloma; neurodegeneration/aging-specific trials have not been completed. needs-human-replication
HDAC inhibitors extend lifespan in model organisms
Pharmacological HDAC inhibition extends lifespan in invertebrate model organisms via multiple inhibitor classes. McDonald et al. (Exp Gerontology, 2013) showed that vorinostat (SAHA), a pan-HDAC inhibitor (Class I/II/IV), extended mid- and late-life longevity in Drosophila melanogaster 6. Sodium butyrate (Zhao 2005) and phenylbutyrate (Kang et al. 2002) independently also extend fly lifespan and are short-chain fatty acids, not hydroxamic acids. Valproic acid and beta-hydroxybutyrate (Class I-selective HDAC inhibitors) extend lifespan in C. elegans in multiple studies 4. The proposed mechanisms include derepression of stress-response, autophagy, and metabolic genes through increased histone acetylation at their promoters.
Important caveats:
- Invertebrate lifespan results with pan-HDAC inhibitors have not translated to mammalian aging models at approved doses
- Short-chain fatty acids like butyrate act on multiple targets beyond HDACs (GPR41/43 receptors, mTOR, etc.); HDAC-specificity of the lifespan effect is not established
- Beta-hydroxybutyrate (BHB), elevated during caloric restriction and ketogenic diet, is an endogenous HDAC inhibitor at 1–2 mmol/L — see ketogenic-diet (verified)
needs-human-replication needs-replication — no mammalian lifespan extension by selective HDAC inhibitors has been demonstrated.
Druggability — aging-context assessment
Frontmatter: druggability-tier: 2
Per CLAUDE.md R26/R27 aging-context convention: the tier reflects whether a clinical drug exists for an aging indication engaging this family, not the maximum-druggability of any member.
- Maximum druggability of the family = tier 1 — five FDA-approved pan-HDAC or HDAC-selective inhibitors exist: vorinostat (SAHA, CTCL), romidepsin (CTCL/PTCL), panobinostat (multiple myeloma), belinostat (PTCL), chidamide (PTCL, China). These are oncology drugs exploiting the hyperacetylation of tumor-suppressor gene promoters in cancer.
- Aging-context druggability = tier 2 — no FDA-approved drug exists for an aging indication engaging HDACs. HDAC6-selective inhibitors (ricolinostat/ACY-1215, tubastatin A) have entered clinical trials for neurological diseases but have not completed aging trials. Pan-HDAC inhibitors at oncology doses carry significant toxicity risk (cytopenias, GI, thrombocytopenia) that makes them unsuitable for healthy-aging indications.
The aging-context tier-2 rating reflects: high-quality probes and clinical candidates exist (ricolinostat, vorinostat), but no aging-validated drug is approved. The dosing, selectivity, and safety profile required for chronic aging-prevention use has not been established.
Pharmacology summary
| Drug | Selectivity | Status (aging context) | Primary mechanism |
|---|---|---|---|
| Vorinostat (SAHA) | Pan (Class I/II) | FDA-approved oncology; not aging-indicated | Competitive Zn-chelator; hydroxamic acid |
| Romidepsin | Class I-selective (prodrug; thiol-based) | FDA-approved oncology; not aging-indicated | Cyclic depsipeptide; Class I > II |
| Panobinostat | Pan (Class I/II) | FDA-approved oncology; not aging-indicated | Hydroxamic acid |
| Ricolinostat (ACY-1215) | HDAC6-selective | Phase I/II (myeloma, neurodegeneration); no aging trial | HDAC6 > Class I; hydroxamic acid |
| Tubastatin A | HDAC6-selective | Preclinical (neurodegeneration models) | Hydroxamic acid; structural selectivity |
| Sodium butyrate | Pan (Class I/IIa) | Preclinical / food-grade; endogenous metabolite class | Short-chain fatty acid; competitive |
| Valproic acid | Pan (Class I/IIa) | FDA-approved epilepsy; off-label aging research | Short-chain fatty acid derivative |
| Entinostat (MS-275) | Class I-selective | Clinical (oncology + immunotherapy); no aging trial | Benzamide; competitive |
Pathway and hallmark connections
- epigenetic-alterations — direct mechanistic contributor; HDAC activity determines histone acetylation state at age-regulated loci
- dna-damage-response — HDAC1 deacetylates OGG1 to promote base excision repair; HDAC inhibition impairs DNA repair in some contexts
- cellular-senescence — p53 deacetylation by HDAC1/SIRT1 modulates p53 stability and senescence threshold; HDAC3 deacetylation of NF-κB p65 modulates SASP output
- loss-of-proteostasis — HDAC6-driven aggresome formation; HDAC6 inhibition disrupts ubiquitinated-protein clearance
- cbp-p300 — the principal HDAC counterpart (HAT enzyme); CBP/p300 writes acetyl marks that HDACs remove; the HAT/HDAC balance is the core regulatory axis
- p53 — HDAC1 deacetylates K382; cooperative with SIRT1; affects MDM2-mediated p53 degradation
- sirtuin — NAD+-dependent Class III deacetylases; functionally overlapping substrates (H3K9ac, H4K16ac, p53 K382, NF-κB p65 K310) but mechanistically distinct; both decline with age through different mechanisms (HDAC: complex disassembly / expression changes; sirtuins: NAD+ depletion)
- scfa-signaling — gut-derived short-chain fatty acids (butyrate, propionate) are endogenous pan-HDAC inhibitors; links microbiome to epigenome
- ketogenic-diet — beta-hydroxybutyrate (BHB) is an endogenous HDAC I/IIa inhibitor at nutritional ketosis concentrations
Disambiguation: classical HDACs vs. sirtuins
| Feature | Class I/II/IV HDACs (this page) | Class III (Sirtuins; sirtuin) |
|---|---|---|
| Members | HDAC1–11 | SIRT1–7 |
| Cofactor | Zn2+ (no cofactor consumed) | NAD+ (consumed stoichiometrically) |
| Structural fold | Arginase/deacetylase fold | Rossmann fold |
| Inhibited by | Hydroxamic acids, benzamides, short-chain fatty acids | Splitomicin, EX-527; activated by NAD+ precursors |
| Age-related change | Variable by member/tissue; often upregulated in inflammation | SIRT1/3/6 decline with age as NAD+ falls |
| Primary aging relevance | Epigenetic drift; non-histone deacetylation; complex disassembly | NAD+-sensing; mitochondrial homeostasis; genome stability |
Limitations and gaps
- No aging-endpoint human trial data. All longevity-relevant HDAC inhibitor evidence is from invertebrate models or human oncology. needs-human-replication
- Family-level page limitation. HDAC1, HDAC3, and HDAC6 have the strongest individual aging literatures. If any accumulates sufficient primary literature (>3 dedicated aging primary studies), a dedicated per-member page should be seeded. needs-canonical-id is not flagged here — all UniProt IDs are confirmed Swiss-Prot entries.
- Class IIa HDAC activity mechanism. Class IIa HDACs (HDAC4/5/7/9) are often described as catalytically inactive because their active-site tyrosine is replaced with histidine. The requirement for HDAC3 recruitment vs. intrinsic activity is not fully resolved for all members. contradictory-evidence
- HDAC3 cardiac aging data. The McKinsey-lab cardiac HDAC3 biology is well-established in development; its specific contribution to aged heart function (rather than congenital/developmental defects) requires clearer primary source. needs-replication
- Non-histone substrate aging relevance. The Choudhary 2009 acetylome identified ~3,600 sites, but which of these are HDAC-family-specific (vs. sirtuin-specific) and which change in an aging-relevant manner is largely unresolved. no-mechanism
- Aging-context safety of HDAC inhibitors. Pan-HDAC inhibitors at oncology doses cause cytopenias and GI toxicity; low-dose or selective use for aging has not been pharmacokinetically characterized. long-term-unknown
Footnotes
Footnotes
-
doi:10.1126/science.1175371 · Choudhary C et al. · Science 2009 · in-vitro (MS proteomics; HeLa + Jurkat cells treated with HDAC inhibitors) · 3985 citations (FWCI 115.3; 100th percentile) · quantitative acetylome: identified ~3,600 acetylation sites on ~1,750 proteins; majority non-histone; established scale of HDAC substrate landscape · archive: not OA ↩ ↩2
-
doi:10.1038/nrm2346 · Yang XJ, Seto E · Nature Reviews Molecular Cell Biology 2008 · review · 1191 citations (FWCI 36.8; 100th percentile) · Rpd3/Hda1 family classification from bacteria to humans; canonical class I/II/IV taxonomy · archive: download failed (green OA) ↩
-
doi:10.1101/cshperspect.a018713 · Seto E, Yoshida M · Cold Spring Harbor Perspectives in Biology 2014 · review · 1905 citations · foundational HDAC family review covering all 11 class I/II/IV members, mechanism, complex composition, and substrates · archive: downloaded (bronze OA) ↩ ↩2
-
doi:10.15252/emmm.201809854 · McIntyre RL, Daniels EG, Molenaars M, Houtkooper RH, Janssens GE · EMBO Molecular Medicine 2019 · review · 112 citations (FWCI 6.3; 100th percentile) · HDAC inhibitors in aging — covers lifespan extension in C. elegans (valproic acid, BHB via hda RNAi) and Drosophila (sodium butyrate, phenylbutyrate, TSA, vorinostat); preclinical results in neurodegeneration, cardiac, metabolic disease · archive: downloaded (gold OA) ↩ ↩2 ↩3
-
doi:10.1038/s41467-020-16361-y · Pao PC et al. · Nature Communications 2020 · in-vivo · model: Hdac1 cKO (Nestin-Cre; neurons + astrocytes) + 5XFAD AD model + aged (17-month) C57BL/6J mice · 199 citations · HDAC1 deacetylates OGG1 (on p300-acetylated lysine residues), stimulating AP-lyase activity; HDAC1 loss → elevated 8-oxoG at gene promoters → transcriptional repression; exifone (50 mg/kg/day IP, 4 wk) rescued OGG1 activity in 17-month WT mice and improved contextual fear conditioning + hippocampal LTP in 8-month 5XFAD mice · archive: downloaded (gold OA) ↩
-
doi:10.1016/j.exger.2012.09.006 · McDonald P, Maizi BM, Arking R · Experimental Gerontology 2013 · in-vivo · model: Drosophila melanogaster · 35 citations · vorinostat (SAHA) treatment during mid- and late-life extended longevity; mechanism attributed in part to HDAC inhibition and altered histone acetylation state · archive: download failed (green OA); drug identity confirmed via McIntyre 2019 Table 1 ↩