HMGB1 — High Mobility Group Box 1
A 215-aa nuclear architectural protein that doubles as a quintessential damage-associated molecular pattern (DAMP) when released into the extracellular space. Inside the nucleus, HMGB1 binds DNA in a sequence-nonspecific, structure-dependent manner via two HMG-box domains, bending DNA and facilitating chromatin remodeling, transcription, DNA repair, and V(D)J recombination. Outside the cell, HMGB1 is one of the most studied DAMPs in immunology and inflammation biology — released passively from necrotic/pyroptotic cells and actively secreted (via inflammasome activation, JAK/STAT signaling, or hyperacetylation-dependent nucleocytoplasmic shuttling) by activated immune cells, senescent cells, and stressed tissues. Listed as a canonical SASP factor 1 and a primary contributor to “sterile inflammation” / inflammaging.
The aging-relevant biology hinges on three redox-sensitive cysteines (C23, C45, C106) that determine which extracellular activity HMGB1 displays (Venereau 2012; Yang/Antoine; Schiraldi 2012):
| Redox state | Cysteine configuration | Receptor / activity |
|---|---|---|
| Fully reduced (ReHMGB1) | C23, C45, C106 all -SH | Chemoattractant via CXCR4 (HMGB1·CXCL12 heterocomplex; Schiraldi 2012); senescence propagation via rage → JAK/STAT + PI3K-AKT/NF-κB (per shin-2025-rehmgb1-paracrine-senescence) |
| Disulfide (DsHMGB1) | C23–C45 intramolecular disulfide; C106 -SH | Pro-inflammatory cytokine inducer via TLR4 (canonical macrophage activation; Venereau 2012) |
| Sulfonyl (OxHMGB1) | All three cysteines oxidized to -SO₃H | Inactive — fails to induce senescence (Shin 2025 Fig 2: SA-β-gal, EdU, p21 all non-significant vs PBS at 20 μg/mL) and fails to activate immune cells |
Because the cysteine oxidation states equilibrate with extracellular redox conditions, HMGB1’s biological activity is not a fixed property — it is gated by the redox environment. Serum half-life of reduced HMGB1 is ~17 min (Zandarashvili 2013 NMR spectroscopy in extracellular fluids; Shin 2025 ref [58]), meaning extracellular ReHMGB1 must engage receptors within a narrow active window before oxidative inactivation. Therapeutic strategies cannot stabilize ReHMGB1 in vivo — interventions must instead target downstream RAGE/JAK signaling or sequester all HMGB1 forms regardless of redox state.
Aged human serum carries elevated ReHMGB1. Shin 2025 Supp Fig 6 demonstrated that in human serum, older (70–80 yr) individuals have significantly elevated reduced HMGB1 compared to younger (~40 yr) individuals, while total HMGB1 trended up but was non-significant. The parallel finding in 24-mo C57BL/6 mice (Supp Fig 5): both total and reduced HMGB1 elevated in aged serum. This positions redox-form-specific HMGB1 measurement as a candidate plasma biomarker of inflammaging tempo, orthogonal to bulk total-HMGB1.
Aging-relevant roles
As a SASP factor
HMGB1 is constitutively released from senescent cells as part of the senescence-associated secretome 1 (HMGB1 listed in the original Coppé 2008 SASP characterization — see sasp). Davalos et al. 2013 mapped HMGB1 as a key SASP factor that is lost from the senescent nucleus and accumulates in the extracellular space, contributing to NF-κB activation in surrounding cells. needs-replication (Davalos 2013 not yet seeded as a study page on this wiki).
As a senescence propagator (Shin 2025)
shin-2025-rehmgb1-paracrine-senescence establishes that only the reduced redox form (ReHMGB1) propagates senescence to bystander cells, acting via rage → jak-stat-pathway (specifically JAK2 → STAT1) + PI3K-AKT/nf-kb → SASP gene expression + cell-cycle arrest. Key specifics from the paper:
- In vitro (20 μg/mL × 72 h): ReHMGB1 induces senescence-like phenotypes in WI-38, BJ foreskin, primary renal epithelial, and human skeletal-muscle (HSKM) cells (Supp Fig 3). OxHMGB1 inert across all four cell types.
- Transcriptomically: ReHMGB1 produces 1,087 DEGs that closely resemble IR-induced primary senescence (GSE13027 dataset); OxHMGB1 produces a divergent profile enriched for interferon response rather than senescence. p15^INK4b^ (CDKN2B) — a bystander senescence marker — is specifically upregulated by ReHMGB1.
- In vivo (3-mo C57BL/6J, single IV 5 mg/kg, 7 dpt): systemic ReHMGB1 elevates serum IL-6 + IL-1β (TNF-α ns) and induces p21 upregulation in gastrocnemius, tibialis anterior, and liver (heart + kidney trend up but ns). p16+/p21+ dual-positive cells increase in TA myofibers (p < 0.001); HMGB1+ myonuclei decrease (p < 0.01) — consistent with nuclear-to-extracellular HMGB1 export.
- Pharmacological rescue: FPS-ZM1 (100 nM RAGE antagonist) and Momelotinib (0.3 μM JAK2 inhibitor — FDA-approved for myelofibrosis) both abrogate ReHMGB1-induced senescence in WI-38, confirming RAGE → JAK2 → STAT1 is functionally necessary.
- Therapeutic blockade (15-mo C57BL/6J, BaCl₂ TA injury model): single IV dose of anti-HMGB1 mAb clone 3E8 (BioLegend 651402) at 0.1 mg/kg on Day −1 reduces senescent-cell burden, increases MyoD+ myogenic progenitor recruitment, partially restores myofiber CSA, and fully restores grip strength to uninjured levels (rotarod + treadmill also improved). The mAb is not isoform-specific — it targets all HMGB1 redox forms; the paper notes isoform-selective tools are an outstanding need.
Conboy MJ + IM are co-authors on the paper, reflecting the conceptual lineage from heterochronic-parabiosis biology (Conboy 2005, Rebo 2016) to Jeon’s single-blood-exchange senescence-transfer paper (Jeon 2022 Nat Metab) to this redox-form-resolved propagator identification.
As an NLRP3-inflammasome–coupled DAMP
Pyroptotic cell death releases HMGB1 along with IL-1β and IL-18 (see hallmarks/chronic-inflammation.md and gsdmd). Released HMGB1 then signals through TLR4 (disulfide form) and RAGE (reduced form) on neighboring cells, amplifying inflammaging. The HMGB1 → RAGE → NF-κB loop is a positive-feedback amplifier on top of the IL-1β/NF-κB loop already operating in aged tissue.
As a substrate for sirtuin / acetyltransferase regulation
HMGB1 nuclear retention vs. cytoplasmic release is regulated by lysine acetylation (hyperacetylation drives release) and deacetylation (SIRT1-mediated deacetylation retains HMGB1 in the nucleus). needs-replication (the SIRT1-HMGB1 link is implied by sirtuin-loss-of-function studies but the primary-source mapping has not been added to this wiki — links to sirt1 should be developed when this evidence is seeded).
Receptors
- rage (AGER) — primary receptor for the reduced (ReHMGB1) form per Shin 2025; also receives AGEs and S100 proteins, making RAGE a multi-DAMP convergence node.
- TLR4 — primary receptor for the disulfide form; drives macrophage NF-κB activation.
- TLR2 — minor; context-dependent.
- CXCR4 (in complex with CXCL12) — HMGB1 forms heterocomplex with CXCL12 that signals through CXCR4 for monocyte recruitment (Schiraldi 2012). needs-replication
Druggability
druggability-tier: 2 — multiple high-quality probes exist but no aging-validated clinical drug:
- Glycyrrhizin (licorice triterpenoid) — binds HMGB1 directly, blocks chemoattractant + cytokine activity. Probe-quality; not aging-validated. Used in Japan as a hepatitis treatment.
- Anti-HMGB1 monoclonal antibodies (e.g., the 2G7 mAb from Tracey lab; Shino-Test 2D6) — preclinical only.
- Recombinant thrombomodulin (ART-123, recomodulin) — sequesters and inactivates HMGB1 via Lys-cleavage in its lectin-like domain. Approved in Japan for sepsis-associated DIC.
- Soluble RAGE (sRAGE) decoys — preclinical only.
- RAGE antagonists (azeliragon, FPS-ZM1) — small-molecule receptor blockers; azeliragon failed Phase 3 in Alzheimer’s disease (TTP488); FPS-ZM1 is a research-grade probe.
The Shin 2025 result revives interest in repurposing these agents for aging indications; the question is whether redox-form-selective interventions (i.e., targeting only ReHMGB1) can preserve any beneficial roles of HMGB1 in development, wound healing, or regeneration. dose-response-unclear.
Extrapolation
| Dimension | Status |
|---|---|
| Pathway conserved in humans? | yes (HMGB1, RAGE, TLR4, JAK/STAT, NF-κB all canonical in humans) |
| Phenotype conserved in humans? | unknown (no human in-vivo aging-rejuvenation evidence) |
| Replicated in humans? | no (no clinical trial of HMGB1 blockade for aging indication; azeliragon Alzheimer’s failure noted) |
Cross-references
shin-2025-rehmgb1-paracrine-senescence · rage · sasp · cellular-senescence · chronic-inflammation · altered-intercellular-communication · advanced-glycation-end-products · nf-kb · jak-stat-pathway · pyroptosis · gsdmd