NLRP3 Inflammasome Assembly Project

Bottom line: the NLRP3 inflammasome is a danger-sensing complex in which
the sensor NLRP3 recruits the adaptor PYCARD (ASC) to activate caspase-1, which
matures IL-1β and IL-18 and cleaves gasdermin D to open pyroptotic pores.
Scoped, with the pathway now modelled: this page lists ten candidate human genes
and the pathway architecture, and the assembly chain has been built as
canonical NLRP3 inflammasome,
a seven-part module (NLRP3 sensor activation → optional NEK7 licensing → ASC
speck → caspase-1 autoprocessing → caspase-1 substrate cleavage → GSDMD pore →
NINJ1 rupture) built from the NLRP3, PYCARD, CASP1, NEK7, GSDMD and NINJ1 gene
reviews and aligned with GO-CAMs
gomodel:635b1e3e00001811 and
gomodel:663d668500001246.
Three candidates already had complete reviews from other work (NLRP3,
CASP4 and GSDMD, 364 annotations between them; most removals are generic
protein binding rows, 24 of 25 on NLRP3), and the NLRP3 review proposes a
new GO term, inflammasome sensor activity, because GO:0140299 molecular
sensor activity requires binding the sensed molecule. The adaptor PYCARD, the effector
CASP1 and the licensing kinase NEK7 have since been reviewed (PR #4536); CASP5,
IL1B, IL18 and BRCC3 still have no gene folder. The draft
metazoan NLR signaling module keeps NLRP3,
PYCARD and CASP1 as a compact pathway-level view alongside the NOD2–RIPK2
branch; the mechanistic detail now lives in the dedicated inflammasome module.

We scoped this because NLRP3 is a major therapeutic target and drives
autoinflammatory disease (CAPS) and inflammation in gout, atherosclerosis and
Alzheimer disease, and recent work on activation sites, post-translational
control and structure is likely to be under-represented in GO.

Overview

The NLRP3 inflammasome is a multiprotein complex that activates inflammatory caspases in response to diverse danger signals, leading to IL-1β/IL-18 release and pyroptotic cell death. Recent discoveries have clarified activation mechanisms and identified new regulators.

Model Species

Primary: Homo sapiens (human)
- Major therapeutic target
- Autoinflammatory disease relevance

Core Pathway Architecture

1. Sensor

• NLRP3 - NOD-like receptor, sensor component

2. Adaptor

• PYCARD (ASC) - Adaptor with PYD and CARD domains

3. Effector Caspases

• CASP1 - Caspase-1, cleaves pro-IL-1β
• CASP4/5 - Non-canonical inflammasome

4. Substrates/Outputs

• IL1B - Pro-inflammatory cytokine
• IL18 - Pro-inflammatory cytokine
• GSDMD - Gasdermin D, pore-forming executioner

5. Critical Regulators

• NEK7 - Essential NLRP3 activator (discovered 2016)
• BRCC3 - Deubiquitinase
• Various negative regulators

6. Priming/Licensing

• NFKB pathway - Transcriptional priming

Candidate Genes (~12-15)

Gene	UniProt	Function
NLRP3	Q96P20	Sensor
PYCARD	Q9ULZ3	ASC adaptor
CASP1	P29466	Effector caspase
CASP4	P49662	Non-canonical
CASP5	P51878	Non-canonical
GSDMD	P57764	Pore formation
IL1B	P01584	Cytokine
IL18	Q14116	Cytokine
NEK7	Q8TDX7	Essential activator
BRCC3	P46736	Deubiquitinase

Mechanism as currently modelled

Reconciled against PMID:41062723
(Dubey et al., Cell Mol Immunol 2025, "Molecular mechanisms and regulation of
inflammasome activation and signaling") and the primary structural literature it
cites. Details and evidence quotes are in
the module.

• Priming vs activation. Canonical activation is two-step. Priming is
  TLR/NF-κB transcriptional licensing of NLRP3 and pro-IL-1β plus poising
  modifications (PKD phosphorylation, ZDHHC5 palmitoylation, TRIM28 SUMOylation);
  activation is the second signal. The module covers activation onward and treats
  priming as upstream context, so priming belongs with the TLR modules.
• Inactive cage → active disc. Inactive NLRP3 is not a monomer: human NLRP3
  is an ADP-bound decamer (PMID:35114687)
  and endogenous mouse NLRP3 a membrane-bound 12–16-mer double-ring cage with the
  pyrin domains shielded inside
  (PMID:34861190). The active form is
  a disc whose PYDs form a filament
  (PMID:36442502).
• NEK7's actual role. NEK7 is kinase-independent here: it binds the NLRP3 LRR
  (PMID:26642356), its catalytic
  activity is dispensable
  (PMID:26814970), it bridges
  adjacent NLRP3 subunits
  (PMID:31189953), and it is proposed
  to break the inactive cage. It is therefore modelled as a licensing annoton
  inside the sensor part, not as a step in the sequence.
• TGN and MTOC. NLRP3 binds PtdIns4P on the dispersed trans-Golgi network
  (PMID:30487600); assembly and
  caspase activation occur at the microtubule organizing center, with HDAC6 as the
  dynein adapter (PMID:32943500).
• ASC speck. One polymerization mechanism serves both families: the sensor
  nucleates ASC PYD filaments, clustered ASC CARDs nucleate caspase-1 CARD
  filaments, and the protease is activated by induced proximity
  (PMID:24630722).
• GSDMD pores. Caspase-1 and caspase-4/5/11 cleave the interdomain linker
  (PMID:26375003); GSDMD-NT binds
  inner-leaflet lipids and oligomerizes into pores
  (PMID:27383986) whose negatively
  charged conduit electrostatically favours mature cytokines over their acidic
  precursors (PMID:33883744).
  Terminal membrane rupture additionally requires NINJ1, which has its own review.
• Non-canonical caspase-4/5/11. These bind cytosolic LPS directly through
  their CARD and cleave GSDMD themselves; the resulting K⁺ efflux is what
  activates NLRP3. So the non-canonical route is a separate sensing module that
  joins this one at a single feedback edge, not an alternative sensor inside it.
  CASP4's review already models it with GO:0160074/GO:0160075.

Other inflammasomes in the review (not modelled here)

PMID:41062723 also covers NLRP1, NLRP6, NLRP7, NLRP9, NLRP10, NLRP12,
NAIP–NLRC4, AIM2, Pyrin, IFI16, CARD8 and MxA. Two patterns matter for future
modelling:

• Shared effector core. Several differ from NLRP3 only in the sensor and reuse
  the ASC → caspase-1 → GSDMD chain verbatim. These are best added later as sensor
  variants over a shared effector core rather than as duplicate modules. AIM2 is
  the cleanest case (HIN200 binds dsDNA sequence-independently, PYD nucleates ASC;
  GO even has GO:0097169 for the complex).
• Mechanisms that do not fit this module's shape. NLRP1 and CARD8 are
  activated by functional degradation — FIIND autoproteolysis leaves two
  non-covalently associated fragments, and proteasomal destruction of the
  N-terminal fragment liberates a UPA–CARD that oligomerizes; DPP8/DPP9 restrain
  this and Val-boroPro releases it. NAIP–NLRC4 puts ligand recognition on NAIP
  (mouse NAIPs are specialised for flagellin vs T3SS needle/rod, humans have a
  single NAIP) while NLRC4 builds the platform and recruits caspase-1 with or
  without ASC. Both need their own module shapes.

Disease Relevance

• CAPS (Cryopyrin-associated periodic syndromes)
• Gout, atherosclerosis
• Alzheimer's disease
• COVID-19 severity

Modules

• Canonical NLRP3 inflammasome —
  the assembly chain in seven parts (NLRP3 sensor activation, optional NEK7
  licensing, ASC speck, caspase-1 autoprocessing, caspase-1 substrate cleavage,
  GSDMD pore, NINJ1 rupture), built from the NLRP3, PYCARD, CASP1, NEK7, GSDMD and
  NINJ1 gene reviews and grounded on GO-CAMs gomodel:635b1e3e00001811 and
  gomodel:663d668500001246. This is the fold of the former four-part overview and
  the seven-part step-level decomposition into one module.
• Metazoan NLR signaling — compact pathway-level
  view, NOD2–RIPK2 plus NLRP3–ASC–caspase-1.
• Plant NLR resistosome signalling
  — the plant counterpart, split out so the metazoan module's scope is unambiguous.
  Not part of this project, listed because the scope boundary was drawn here.

Project Status

• [ ] Stub - needs gene folder setup
• [x] Canonical assembly chain modelled and reconciled with the 2025 CMI review
• [x] Gene reviews for PYCARD, CASP1 and NEK7 (PR #4536; the PYCARD review
  adjudicates its NAS GO:0070269 pyroptotic inflammatory response rows)
• [ ] Gene folders for IL1B and IL18
• [x] Folded the overview and step-level NLRP3 modules into one
  canonical_nlrp3_inflammasome (PR #5013)
• [ ] Decide whether a shared ASC→caspase-1→GSDMD effector core module is worth
  factoring out before adding AIM2/NLRC4/NLRP1 sensor variants
• [ ] Ontology follow-up: the inflammasome sensor activity term proposed in the
  NLRP3 review is still the blocker for typing NLRP3's sensing function; the
  module works around it with GO:0035591 signaling adaptor activity