Anthrax lethal factor cleavage of Nlrp1 is required for activation of the inflammasome

Abstract

NOD-like receptor (NLR) proteins (Nlrps) are cytosolic sensors responsible for detection of pathogen and danger-associated molecular patterns through unknown mechanisms. Their activation in response to a wide range of intracellular danger signals leads to formation of the inflammasome, caspase-1 activation, rapid programmed cell death (pyroptosis) and maturation of IL-1β and IL-18. Anthrax lethal toxin (LT) induces the caspase-1-dependent pyroptosis of mouse and rat macrophages isolated from certain inbred rodent strains through activation of the NOD-like receptor (NLR) Nlrp1 inflammasome. Here we show that LT cleaves rat Nlrp1 and this cleavage is required for toxin-induced inflammasome activation, IL-1 β release, and macrophage pyroptosis. These results identify both a previously unrecognized mechanism of activation of an NLR and a new, physiologically relevant protein substrate of LT.

Figure 1. Comparison of the polymorphic regions of the rNlrp1constructs used in this study.

The top two constructs represent the full-length HA-tagged rNlrp1 proteins from the LT-sensitive rat (CDF) and the LT-resistant rat (LEW) used in this study. “S” denotes the rNlrp1^S sequence and “R” denotes the rNlrp1^R sequence. The next four constructs represent chimeric proteins, and chimeric proteins in which mutations were made at the putative LF cleavage site, (shown as a vertical dotted line below the filled arrow). These first six constructs were expressed as full-length proteins in cells. The last three constructs represent proteins where aa 3–100 of rNlrp1were expressed and purified from E. coli as 6His-GST tagged proteins. In the sequence alignments, residues identical to those in the construct listed above are indicated by quotations (”). The residue at which fusion of the CDF sequence to LEW or the LEW sequence to CDF occurs is shown with the open arrow. Grey shading denotes LEW sequence, while white denotes CDF sequence. An LF consensus cleavage site motif is shown at the top, where B represents basic (positively-charged) residues and h stands for hydrophobic residues – the position of each is shown relative to the cleavage site.

Figure 2. Cleavage of full length rNlrp1 by LF.

(A) Treatment of stably-transfected HT1080 cells overexpressing HA-tagged rNlrp1 proteins with LT (1 µg/ml) for 5 h followed by lysis and Western blotting with anti-HA and anti-Mek3 antibodies. Full length and cleaved HA-tagged rNlrp1 are shown. The central panel shows full length and cleaved MEK3 (in a reprobing of the same gel). Cleavage of rNlrp1 leads to appearance of the 6-kDa HA-reactive band. Cleavage of endogenous MEK3 leads to a shift in mobility. (B) Sucrose lysates from stably-transfected HT1080 cells overexpressing HA-tagged rNlrp1 proteins were treated with LF (1 or 10 µg/ml) for 3 h, and subjected to Western blotting with anti-HA (top row) or anti-MEK3 antibody (second row). Cleavage of rNlrp1 leads to loss of the N-terminal HA epitope, and cleavage of endogenous MEK3 leads to a shift in mobility. (C) IP (anti-HA pulldown) followed by anti-HA Western blotting of lysates from HT1080 cells expressing HA-tagged rNlrp1 following treatment with LF (10 µg/ml, 3.5 h). Anti-HA reactive cross-reactive bands not marked as HA-Nlrp1 also appear in vector-transfected controls (data not shown).

Figure 3. Cleavage of CDF100 by LF.

(A) In vitro cleavage of N-terminally 6His-GST-tagged aa 3–100 of rNlrp1^S (CDF100). Purified protein (1 mg/ml) was treated with the indicated molar ratios of LF for 6 h prior to SDS gel electrophoresis and Coomasie staining. (B) 6His-GST-tagged CDF100 protein or variants (1 mg/ml) were treated with recombinant LF or enzymatically inactive LF E687C (each at 128 µg/ml) for 3 h prior to SDS gel electrophoresis and Coomasie staining. F1 and F2 refer to two fragments generated following LF treatment of CDF100 and its mutated variants.

Figure 4. LT effects on cells expressing rNlrp1 constructs.

(A) BMAJ cells (normally resistant to LT) that were retrovirally transfected with the indicated rNlrp1 constructs were treated with LT (PA+LF) for 10 h. Percent viability was assessed by MTT staining and calculated relative to untreated controls. Graph shown is from a single experiment, which is representative of eight identically performed experiments. Each point represents the average of three wells. Insets show staining of representative sensitive and resistant cell lines following toxin treatment. (B) Supernatant levels of IL-1β measured by ELISA. Cells were pretreated for 2 h with LPS (0.5–1 µg/ml), prior to LT treatment (7 h). Results shown are from a single experiment, which is representative of three similar experiments. Each point represents the average of duplicate wells in this experiment. (C) Protection of the sensitized BMAJ cell line transfected with rNlrp1fusion CDF53-LEW against challenge with LT (5 µg/ml) following treatment with 50 µM of each drug, or heat shock. Results are average of three experiments, with duplicate wells per treatment in each experiment. (D) Average supernatant IL-1β levels associated with protective treatments from (C) are shown. Results are the average of three experiments, with duplicate wells per treatment in each experiment.