The damage-associated molecular pattern (DAMP) protein HMGB1 (high mobility group box 1) and Gram-negative bacterial lipids (lipid A and its precursor lipid IVa) individually fail to activate macrophages, but their complex engages TLR4–TRIF–RIPK3 signaling that produces MLKL-dependent necroptosis, caspase-8-dependent apoptosis, release of the proinflammatory cytokine interleukin-1α (IL-1α), and maturation of interleukin-1β (IL-1β) in mice (Meng et al., JBC 2019). What determines which death mode an individual RIPK3-activated cell executes, how the apoptotic branch nevertheless releases IL-1α (the untested proposal being caspase-3-mediated cleavage of the pore-forming protein gasdermin E), how far the axis generalizes across species and bacterial lipid structures, and whether it is protective or pathological during infection all remain unresolved. The problem asks for the molecular determinants of death-mode execution and IL-1α efflux in this pathway, its species and lipid-structure generality, and its in vivo balance between pathogen clearance and non-resolving inflammation — jointly, a mechanistic and physiological account of how a DAMP and a pathogen-associated molecular pattern (PAMP) cooperate to switch RIPK3-dependent inflammatory cell death on (DAMP–PAMP co-sensing).

Background

Necroptosis is a lytic, inflammatory form of regulated cell death in which the kinase RIPK3 (receptor-interacting protein kinase 3) phosphorylates the pseudokinase MLKL (mixed lineage kinase domain-like); phosphorylated MLKL oligomerizes and ruptures the plasma membrane, releasing intracellular content. RIPK3 is activated by several upstream adaptors, including TRIF (TIR-domain-containing adapter inducing interferon-β), which is engaged by Toll-like receptor 4 (TLR4), the receptor for bacterial endotoxin. Apoptosis is the quieter, non-lytic death program executed by caspase proteases: the initiator caspase-8 activates the executioners caspase-3 and caspase-7. RIPK3 signaling can terminate in either mode — active caspase-8 cleaves RIPK1 and RIPK3 and thereby suppresses necroptosis — so the outcome reflects a balance whose molecular determinants in individual cells remain incompletely defined.

TLR4 recognizes lipid A, the membrane-embedded glucosamine-phosphate lipid core of Gram-negative lipopolysaccharide (LPS), in the binding pocket of the co-receptor MD-2 (myeloid differentiation 2). Lipid IVa, the biosynthetic precursor of lipid A, acts as a TLR4 antagonist in humans but as an agonist in mice, a difference traced to a handful of species-specific MD-2 residues. HMGB1 is a nuclear DNA-binding protein released by activated immune cells and damaged tissue; extracellularly it acts as a prototypical damage-associated molecular pattern (DAMP) and can bind LPS-related lipids and shuttle them to innate immune receptors. Meng et al. (2019) showed that HMGB1–lipid IVa or HMGB1–lipid A complexes, but neither component alone, trigger TLR4–TRIF–RIPK3-dependent macrophage death with both necroptotic and apoptotic features, IL-1α release, and IL-1β maturation; genetic deletion of Ripk3, Trif, or Il1r (the gene encoding the interleukin-1 receptor), or monoclonal-antibody-mediated HMGB1 neutralization, uniformly attenuated Gram-negative-bacteria-induced inflammatory infiltration in a mouse air-pouch model.

The mechanistic logic of this pathway was left open in four respects. First, the authors proposed a 'stochastic availability' model — that the local availability or activation state of MLKL versus caspase-8 dictates whether an individual RIPK3-activated cell dies by necroptosis or apoptosis — but did not identify the molecular determinants. Second, surprisingly, macrophages on the apoptotic branch still released the alarmin IL-1α (a danger signal released by damaged or dying cells): deletion of Mlkl only partially reduced IL-1α release and the residual release was caspase-8-dependent, contradicting the textbook view that apoptotic cells do not release danger signals. The authors raised, without testing, the hypothesis that executioner caspase-3 cleaves gasdermin E (GSDME, also called DFNA5; a pore-forming member of the gasdermin family) into a membrane-lysing fragment that provides the IL-1α efflux route. This hypothesis has mechanistic precedent: caspase-3-mediated GSDME cleavage converts apoptosis into secondary necrosis in GSDME-expressing cells (Wang et al. 2017; Rogers et al. 2017), and a caspase-8 → caspase-3 → GSDME cascade was later shown to let human TH17 cells release IL-1α while remaining viable (Chao et al. 2023). Third, the complex of HMGB1 with lipid A, but not with lipid IVa, stimulated human peripheral blood mononuclear cells (PBMCs), consistent with known species-specific antagonism of lipid IVa at human MD-2 (Meng et al. 2010); the generality of the axis across species and across the structural diversity of Gram-negative lipid A variants is unestablished. Fourth, RIPK3-dependent signaling protects mice against Yersinia pestis infection (Weng et al. 2014) yet drives non-resolving IL-1-dependent inflammation during Escherichia coli infection, so the protective-versus-pathological balance of the HMGB1–lipid axis in vivo — and hence the therapeutic window for blocking HMGB1, TRIF, RIPK3, or the IL-1 receptor — is undefined.

Problem Statement

Give a mechanistic and physiological account of inflammatory cell death downstream of HMGB1–bacterial lipid co-sensing, answering the questions left open by Meng et al. (2019):

(1) Death-mode determinants. Identify the molecular determinants, regulatory checkpoints, and cell-context factors that dictate whether RIPK3, activated by the HMGB1–lipid IVa/lipid A complex acting through TLR4–TRIF, executes MLKL-dependent necroptosis or caspase-8-dependent apoptosis in an individual macrophage — moving beyond the authors' untested 'stochastic availability' proposal to establish which components (for example, the relative abundance or activation state of MLKL versus caspase-8, or upstream regulators of their balance) quantitatively predict the outcome, and whether the choice is genuinely stochastic or deterministically specified by cell state.

(2) IL-1α efflux in the apoptotic branch. Determine the mechanism by which macrophages committed to the apoptotic branch of this pathway release IL-1α, in particular testing whether executioner caspase-3 cleaves GSDME into its pore-forming fragment such that GSDME pores constitute the IL-1α release conduit, and identifying the actual conduit if GSDME proves dispensable.

(3) Species and lipid-structure generality. Establish how far the HMGB1–bacterial lipid → TLR4–TRIF–RIPK3 axis extends across species and across the structural diversity of Gram-negative lipid A variants, beyond the demonstrated mouse activity and the human PBMC lipid A response (in which the lipid IVa mouse–human difference is imposed by species-specific MD-2 residues), and identify which features of the lipid A structure and of the host receptor set these boundaries.

(4) In vivo protective-versus-pathological balance. Determine when this signaling promotes pathogen clearance versus driving non-resolving IL-1-dependent inflammation in vivo — RIPK3-dependent signaling protects mice against Yersinia pestis infection yet drives non-resolving IL-1-dependent inflammation during Escherichia coli infection — and thereby define the overall protective-versus-pathological balance and the therapeutic window for targeting HMGB1, TRIF, RIPK3, or the interleukin-1 receptor (IL-1R).

These questions jointly support one objective: to explain how a host-derived danger signal and bacterial endotoxin lipids cooperate to switch RIPK3-dependent inflammatory cell death on, and to shape its execution mode and inflammatory output.

All four numbered sub-questions are mandatory and jointly required: the task counts as resolved only when each is answered, and settling any proper subset while leaving the remaining sub-questions open does not constitute a complete solution.

References

  1. Meng R, Gu L, Lu Y, Zhao K, Wu J, Wang H, Han J, Tang Y, Lu B. High mobility group box 1 enables bacterial lipids to trigger receptor-interacting protein kinase 3 (RIPK3)-mediated necroptosis and apoptosis in mice. Journal of Biological Chemistry 294(22):8872-8884, 2019. doi:10.1074/jbc.ra118.007040, https://doi.org/10.1074/jbc.ra118.007040
  2. Wang Y, Gao W, Shi X, Ding J, Liu W, He H, Wang K, Shao F. Chemotherapy drugs induce pyroptosis through caspase-3 cleavage of a gasdermin. Nature 547(7661):99-103, 2017. doi:10.1038/nature22393, https://doi.org/10.1038/nature22393
  3. Rogers C, Fernandes-Alnemri T, Mayes L, Alnemri D, Cingolani G, Alnemri ES. Cleavage of DFNA5 by caspase-3 during apoptosis mediates progression to secondary necrotic/pyroptotic cell death. Nature Communications 8:14128, 2017. doi:10.1038/ncomms14128, https://doi.org/10.1038/ncomms14128
  4. Chao YY, Puhach A, Frieser D, Arunkumar M, Lehner L, Seeholzer T, Garcia-Lopez A, van der Wal M, Fibi-Smetana S, Dietschmann A, Sommermann T, Ćiković T, Taher L, Gresnigt MS, Vastert SJ, van Wijk F, Panagiotou G, Krappmann D, Groß O, Zielinski CE. Human TH17 cells engage gasdermin E pores to release IL-1α on NLRP3 inflammasome activation. Nature Immunology 24(2):295-308, 2023. doi:10.1038/s41590-022-01386-w, https://doi.org/10.1038/s41590-022-01386-w
  5. Kulkarni M, Bourne CM, Mahale AB, Exconde PM, Murphy C, Goodrow HT, Cervantes S, Kardhashi M, Kambayashi M, Yoo W, Wrong TJ, Patio RC, Discher BM, Taabazuing CY. Human non-canonical inflammasomes activate CASP3 to limit intracellular Salmonella replication in macrophages. PLoS Pathogens 22(4):e1014178, 2026. doi:10.1371/journal.ppat.1014178, https://doi.org/10.1371/journal.ppat.1014178
  6. Meng J, Drolet JR, Monks BG, Golenbock DT. MD-2 residues tyrosine 42, arginine 69, aspartic acid 122, and leucine 125 provide species specificity for lipid IVA. Journal of Biological Chemistry 285(36):27935-27943, 2010. doi:10.1074/jbc.M110.134668, https://doi.org/10.1074/jbc.M110.134668
  7. Weng D, Marty-Roix R, Ganesan S, Proulx MK, Vladimer GI, Kaiser WJ, Mocarski ES, Pouliot K, Chan FK-M, Kelliher MA, Harris PA, Bertin J, Gough PJ, Shayakhmetov DM, Goguen JD, Fitzgerald KA, Silverman N, Lien E. Caspase-8 and RIP kinases regulate bacteria-induced innate immune responses and cell death. Proceedings of the National Academy of Sciences USA 111(20):7391-7396, 2014. doi:10.1073/pnas.1403477111, https://doi.org/10.1073/pnas.1403477111
  8. Wang X, Li Z, Bai Y, Zhang R, Meng R, Chen F, Wang H, Billiar TR, Xiao X, Lu B, Tang Y. A small molecule binding HMGB1 inhibits caspase-11-mediated lethality in sepsis. Cell Death & Disease 12(4):402, 2021. doi:10.1038/s41419-021-03652-5, https://doi.org/10.1038/s41419-021-03652-5
  9. Tamai R, Mashima I, Kiyoura Y. Alendronate augments lipid A-induced IL-1α release via activation of ASC but not caspase-11. Inflammation 44(5):2132-2141, 2021. doi:10.1007/s10753-021-01489-w, https://doi.org/10.1007/s10753-021-01489-w
  10. Chang NP, DaPrano EM, Lindman M, Estevez I, Chou TW, Evans WR, Nissenbaum M, McCourt M, Alzate D, Atkins C, Kusnecov AW, Huda R, Daniels BP. Neuronal DAMPs exacerbate neurodegeneration via astrocytic RIPK3 signaling. JCI Insight 9(11):e177002, 2024. doi:10.1172/jci.insight.177002, https://doi.org/10.1172/jci.insight.177002