The interaction between NLRP1 and oxidized TRX1 involves a transient disulfide bond

Abstract

NLRP1 is an innate immune receptor that detects pathogen-associated signals, assembles into a multiprotein structure called an inflammasome, and triggers a proinflammatory form of cell death called pyroptosis. We previously discovered that the oxidized, but not the reduced, form of thioredoxin-1 directly binds to NLRP1 and represses inflammasome formation. However, the molecular basis for NLRP1's selective association with only the oxidized form of TRX1 has not yet been established. Here, we leveraged AlphaFold-Multimer, site-directed mutagenesis, thiol-trapping experiments, and mass spectrometry to reveal that a specific cysteine residue (C427 in humans) on NLRP1 forms a transient disulfide bond with oxidized TRX1. Overall, this work demonstrates how NLRP1 monitors the cellular redox state, further illuminating an unexpected connection between the intracellular redox potential and the innate immune system.

Keywords: AlphaFold-Multimer; NLRP1; disulfide; inflammasome; pyroptosis; redox; thioredoxin.

Figure 1.. AlphaFold-Multimer predicts the protein-protein interface between NLRP1 and TRX1.

(A) Domain architecture of human NLRP1 and mouse NLRP1B, indicating the sites of autoproteolysis, sequence alignment at the TRX1-binding interface, and the conserved cysteine residue. (B) Model of the NACHT-LRR-TRX1 heterodimer predicted by AlphaFold-Multimer. Inset: close-up view of the cysteine residues potentially involved in an intermolecular disulfide bond.

Figure 2.. Mutation of C427 (human) or C225 (mouse) abrogates TRX1 binding to NLRP1.

(A-C) HEK 293T cells were transiently transfected with plasmids encoding the indicated FLAG-tagged proteins before lysates were subjected to anti-FLAG IP and immunoblotting analyses. In A and B, an asterisk indicates a background band. Immunoblots are representative of three or more biological replicates.

Figure 3.. A transient intermolecular disulfide forms between NLRP1 and TRX1.

(A) Schematic of the IAA- and NEM-conjugate trapping experiment. (B,C) HEK 293T cells were transiently transfected with plasmids encoding the indicated FLAG- or HA-tagged proteins before cells were lysed in IAA or NEM and proteins were enriched by anti-FLAG IP and analyzed by immunoblotting. In C, cells were treated with H₂O₂ as indicated prior to lysis. H.E., high exposure; L.E., low exposure. (D) CARD8^−/− THP-1 cells containing plasmids (pInducer-20 vector) encoding doxycycline-inducible, HA-tagged proteins were treated with DOX (1 μg/mL) for 16 h before being lysed in IAA (10 mM) and visualized by immunoblotting. (E) HEK 293T cells were transiently transfected with plasmids encoding the indicated proteins were lysed in IAA (10 mM) and visualized by immunoblotting. (F) N/TERT-1 keratinocytes were lysed in IAA (10 mM) and visualized by immunoblotting. In B-F, reducing refers to the presence of DTT (50 mM) in the loading dye for SDS-PAGE analysis. Immunoblots are representative of three or more biological replicates.

Figure 4.. Mass spectrometry-based identification of the modified cysteine residue.

(A) Schematic of the workflow used to identify cysteine residues involved in the NLRP1-TRX1 disulfide bond by LC-MS/MS. The excised gel band is shown. (B) Ratio of extracted ion current (XIC) for peptides with IAA-modified cysteine residues compared to the total XIC for the corresponding identified peptide sequence, following the schematic in A.