Evolutionary loss of inflammasomes in the Carnivora and implications for the carriage of zoonotic infections

Abstract

Zoonotic pathogens, such as COVID-19, reside in animal hosts before jumping species to infect humans. The Carnivora, like mink, carry many zoonoses, yet how diversity in host immune genes across species affect pathogen carriage is poorly understood. Here, we describe a progressive evolutionary downregulation of pathogen-sensing inflammasome pathways in Carnivora. This includes the loss of nucleotide-oligomerization domain leucine-rich repeat receptors (NLRs), acquisition of a unique caspase-1/-4 effector fusion protein that processes gasdermin D pore formation without inducing rapid lytic cell death, and the formation of a caspase-8 containing inflammasome that inefficiently processes interleukin-1β. Inflammasomes regulate gut immunity, but the carnivorous diet has antimicrobial properties that could compensate for the loss of these immune pathways. We speculate that the consequences of systemic inflammasome downregulation, however, can impair host sensing of specific pathogens such that they can reside undetected in the Carnivora.

Keywords: Carnivora; NLRC4; NLRP3; caspase 1; caspase 11; caspase 4; inflammasome.

Graphical abstract

Figure 1

The activity of the Carnivora caspase-1/-4 fusion protein is markedly restricted in cells infected with S. Typhimurium (A) Representative table showing the evolutionary conservation of the inflammatory hybrid caspase-1/-4 and apoptotic caspases in species belonging to order Carnivora. (B) Mouse WT iBMDM and dog WT DH82 cells were infected with S. Typhimurium MOI of 10, and the amount of lactate dehydrogenase (LDH) released in the supernatant was measured via colorimetric assay over time. (C and E) Mouse WT, Nlrc4^−/−, Casp1^−/−/11^−/− iBMDM, and dog WT DH82 cells were infected with S. Typhimurium MOI of 10, and the amounts of LDH and IL-1β released in the supernatant were measured over time via colorimetric assay and ELISA, respectively. (D) Lysates from dog WT DH82 cells infected with S. Typhimurium MOI of 1 for 24 h in the supernatant were blotted against dog-specific IL-1β alongside non-infected controls (medium only). (F and G) Primary dog MNCs and dog WT DH82 cells were infected with S. Typhimurium MOI of 10, and the amounts of LDH and IL-1β released in the supernatant were measured over time, as in (C) and (E). (H and I) Mouse primary WT and DogMo BMDM together with dog WT DH82 cells were infected with S. Typhimurium MOI of 10, and the amounts of LDH and IL-1β in the supernatant were measured over time, as in (C) and (E). (J and K) Mouse primary WT and DogMo BMDMs were infected with S. Typhimurium at an MOI of 10 for 6 h, and GSDMD cleavage was determined by western blot analysis of cell lysates using anti-mouse GSDMD antibody. Data are shown as means ± SEMs in (B), (C), and (E)–(I). Data are pooled from 3 independent experiments in (B), (C), and (E), from 1 representative of 3 independent experiments in (D), from 1 representative of 2 independent experiments in (H)–(K), and from 1 single experiment in (F) and (G) for which cells were pooled from 4–6 dogs (Table S1). Statistical significance in (C) and (E) was calculated by a 1-way ANOVA for each time point individually, followed by Tukey’s multiple comparison test; ns, not significant; ^∗p < 0.05, ^∗∗p < 0.01, ^∗∗∗p < 0.001. See also Figures S1 and S2 and Table S1.

Figure 2

The Carnivora caspase-1/-4 fusion protein induces substrate cleavage in vitro, but fails to activate the non-canonical inflammasome pathway in vivo (A and B) Mouse primary WT and DogMo BMDM together with dog WT DH82 cells were primed with Pam3CSK4 (10 μg/mL for 4 h) and then transfected with LPS (5 μg/mL for 16 h) using Fugene HD. The amounts of LDH and IL-1β released in the supernatant were measured over time via colorimetric assay and ELISA, respectively. (C and D) Primary dog MNCs and dog WT DH82 cells were stimulated and analyzed as in (A) and (B). (E) Lysates from dog WT DH82 cells primed and stimulated as in (A) and (B) were blotted against dog-specific IL-1β alongside the appropriate controls. (F) Upper: in vitro protein cleavage. Recombinant mouse gasdermin D and pro-IL-1β were submitted to cleavage with a dilution series of dog caspase-1/-4 or mouse caspase-11, incubated for 30 min, and followed with SDS-PAGE analysis of cleavage products. Black arrows indicate bands corresponding to intact substrate, and gray arrows correspond to the biologically active form of the protein originated by caspase cleavage. Lower: catalytic efficiency represented as kcat/KM (M-1 s-1) values for peptidyl fluorogenic substrates (1) general inflammatory caspase substrate, (2) caspase-1 selective substrate, or (3) caspase-11 selective substrate. Average of 3 determinations ± SDs. Data are shown as means ± SEMs in (A)–(D). Data are pooled from 2 independent experiments in (A) and (B), from 1 representative of 2 independent experiments in (E), and from 1 single experiment in (C) and (D) for which cells were pooled from 4–6 dogs. See also Table S1 and Figure S3.

Figure 3

The NLR gene repertoire and NLRP3-dependent inflammasome pathway are severely compromised in the Carnivora (A) Representative table showing the evolutionary conservation of pattern recognition receptors (PRRs) in species belonging to order Carnivora. (B) Mouse WT iBMDM, mouse Casp1^−/−/11^−/− iBMDM, and dog WT DH82 cells were primed with LPS (200 ng/mL for 3 h) and then stimulated with nigericin (20–200 μM for 1 h). The amount of LDH released in the supernatant was measured over time via a colorimetric assay. (C) The same experiment as in (B); the amount of IL-1β in the supernatant was measured for mouse WT and dog WT DH82 cells only via ELISA. (D) Dog WT DH82 cells were primed with LPS (200 ng/mL for 3 h) and then stimulated with nigericin (200 μM for 1 h). Total protein was precipitated from the supernatant, and IL-1β cleavage was assessed by western blot analysis. (E) Mouse WT iBMDM and dog WT DH82 cells were primed with LPS (200 ng/mL for 3 h) and then stimulated with nigericin (10 μM for 24 h). The amount of LDH released in the supernatant was measured over time, as in (B). (F and G) Mouse primary WT and DogMo BMDM together with dog WT DH82 cells were primed with LPS (200 ng/mL for 3 h) and then stimulated with nigericin (20 μM for 1 h). The amounts of LDH and IL-1β were measured in the supernatant as in (B) and (C). (H and I) Mouse primary WT and DogMo BMDM were primed and stimulated as in (F) and (G). Total protein was precipitated from the lysate and gasdermin D cleavage was assessed by western blot analysis. Data are shown as means ± SEMs in (B) and (C) and (E)–(G). Data are pooled from 3 independent experiments in (B) and (C) and 2 independent experiments in (F) and (G), from 1 representative of 3 independent experiments in (D), from 1 representative of 2 independent experiments in (E), and from 1 single experiment in (H) and (I). Statistical significance was calculated by a 1-way ANOVA for each nigericin concentration individually followed by Tukey’s multiple comparison test in (B) and by 2-tailed unpaired t test for each nigericin concentration individually assuming equal or unequal variances in (C); ns, not significant; ^∗p < 0.05, ^∗∗p < 0.01, ^∗∗∗p < 0.001. See also Figure S4.

Figure 4

Caspase-8 mediates late cell death and IL-1β production in Carnivora cells infected with S. Typhimurium (A) Dog WT DH82 cells were primed with LPS (200 ng/mL) and stimulated with nigericin (20 μM for 1 h). Non-stimulated cells were left in cell culture medium. Live cells were stained for activated caspases (green, FLICA poly-caspase, FLICA caspase-1, or FLICA caspase-8). Following fixation, cells were stained for cytoplasmic ASC (red) and nuclei using DAPI staining (blue). White arrows point to ASC specks. (B) Dog WT, Casp8^−/− bulk edited, and cells from 2 individual Casp8^−/− DH82 clones were infected with S. Typhimurium MOI of 10, and the amount of LDH released in the supernatant was measured over time via a colorimetric assay. (C) Dog WT and Casp8^−/− bulk-edited DH82 cells were infected with S. Typhimurium MOI of 10, and the amount of IL-1β in the supernatant was measured over time by ELISA. (D and E) Dog WT and cells from 1 individual Ripk1^−/− DH82 clone were infected with S. Typhimurium MOI of 10, and the amounts of LDH and IL-1β were measured in the supernatant over time via colorimetric assay and ELISA, respectively. (F and G) Identical to (D) and (E), but with cells from a second individual Ripk1^−/− DH82 clone; LDH and IL-1β were measured in the supernatant at 24 h only. (H) Dog WT and cells from 1 individual Ripk1^−/− DH82 clone were infected with S. Typhimurium MOI of 1 and total protein extracts from cell lysates were subject to western blot analysis against dog-specific IL-1β. Uninfected controls (medium only) were also included for each cell line. Data are shown as means ± SEMs in (B)–(G). Data are pooled from 4 independent experiments in (C)–(E), from 1 representative of 3 independent experiments in (A), from 1 representative of 2 independent experiments in (B), and from 1 single experiment in (F)–(H). Statistical significance was calculated by the Mann-Whitney test in (C) and (D) and by 2-tailed unpaired t test assuming equal variances in (E); ns, not significant; ^∗p < 0.05. See also Figure S4.

Figure 5

Inflammasome activity induces pore formation but is uncoupled from cell lysis in the Carnivora (A) Representative table showing the evolutionary conservation of main components of different lytic cell death pathways in species belonging to the Carnivora order. (B) Dog WT DH82 cells were infected with S. Typhimurium MOI of 1, 10, and 50 or primed with LPS (200 ng/mL) for 3 h and stimulated with nigericin (200 μM) for 24 h. Propidium iodide (PI) fluorescence was measured every 10 min over a 24-h period and expressed as %PI uptake normalized against maximum PI uptake achieved after lysing the cells with Triton X-100. Data from 3 independent experiments are shown. A simplified schematic of the experimental design is also shown. (C) Dog WT DH82 cells were infected with S. Typhimurium MOI of 50 for 24 h, and bright-field images were taken on a confocal microscope every 5 min over the 24-h period (see also Video S1). (D) Dog WT DH82 cells primed with LPS (200 ng/mL) for 3 h were stimulated with nigericin (200 μM) for 1 h and bright-field images were taken on a confocal microscope every minute over the 1-h period (see also Video S2). Representative images of the different morphological stages are shown. Data are shown as mean of 3 replicate wells for each independent experiment in (B). Data have originated from a single experiment in (C) and (D). Scale bar, 10 μm.