NLRP1-dependent activation of Gasdermin D in neutrophils controls cutaneous leishmaniasis

Abstract

Intracellular pathogens that replicate in host myeloid cells have devised ways to inhibit the cell's killing machinery. Pyroptosis is one of the host strategies used to reduce the pathogen replicating niche and thereby control its expansion. The intracellular Leishmania parasites can survive and use neutrophils as a silent entry niche, favoring subsequent parasite dissemination into the host. Here, we show that Leishmania mexicana induces NLRP1- and caspase-1-dependent Gasdermin D (GSDMD)-mediated pyroptosis in neutrophils, a process critical to control the parasite-induced pathology. In the absence of GSDMD, we observe an increased number of infected dermal neutrophils two days post-infection. Using adoptive neutrophil transfer in neutropenic mice, we show that pyroptosis contributes to the regulation of the neutrophil niche early after infection. The critical role of neutrophil pyroptosis and its positive influence on the regulation of the disease outcome was further demonstrated following infection of mice with neutrophil-specific deletion of GSDMD. Thus, our study establishes neutrophil pyroptosis as a critical regulator of leishmaniasis pathology.

Fig 1. NLRP1, but not NLRP3, protects against lesion exacerbation during L.

mexicana infection. (A) Nlrp3^-/- and C57BL/6 control mice were infected i.d. with 10⁶ L. mexicana metacyclic promastigotes. Lesion development was measured weekly over the indicated time frame, and lesion score was determined including lesion size, inflammation, and pathology status. (B) Eighteen weeks p.i., representative lesion pictures are shown, (C) lesion size was measured and (D) parasite load at the site of infection was determined by limiting dilution assay (LDA). (E) The number of CD45⁺ cells and the relevant immune populations, including CD45⁺CD11b⁺Ly6G⁺ neutrophils, CD45⁺CD11b⁺Ly6C⁺ monocytes, and CD45⁺CD11c⁺ dendritic cells in the infected ear, was determined by flow cytometry. (F) Nlrp1^-/- and C57BL/6 control mice were similarly infected i.d. with metacyclic L. mexicana promastigotes and lesion score was monitored over 8 weeks. (G) Representative pictures of Nlrp1^-/- and C57BL/6 ear lesions at 8 weeks p.i. (H) Representative ear lesion size and (I) parasite load as determined by LDA. (J) The number of CD45⁺ cells, CD45⁺CD11b⁺Ly6G⁺ neutrophils, CD45⁺CD11b⁺Ly6C⁺ monocytes, and CD45⁺CD11c⁺ dendritic cells in the infected ears was analyzed 8 weeks p.i. by flow cytometry. Data are shown as mean ± SD and are representative of ≥3 experiments, n≥4/group. Statistical differences in lesion development were analyzed by 2-way ANOVA, and cell numbers using a Mann-Whitney U-test. *p <0.05; **p <0.01. ***p <0.001.

Fig 2. Caspase-1 and Gasdermin D (GSDMD) reduce L. mexicana-induced pathology and parasite burden.

(A) Casp1^-/-and C57BL/6 control mice were infected i.d. with metacyclic L. mexicana promastigotes and lesion score was measured over time. (B) Representative pictures of ear lesion, (C) ear lesion size and (D) parasite load determined by LDA at the site of infection. (E) The number of CD45⁺ immune cells, CD45⁺CD11b⁺Ly6G⁺ neutrophils, CD45⁺CD11b⁺Ly6C⁺ monocytes, and CD45⁺CD11c⁺ dendritic cells at the infection site, was analyzed by flow cytometry (F) Gsdmd^-/- and C57BL/6 control mice were similarly infected with L. mexicana and lesion score assessed over time. (G) Representative pictures of ear lesions (H) lesion size and (I) parasite load determined by LDA at 8 weeks p.i. (J) Total number of CD45⁺ cells, CD45⁺CD11b⁺Ly6G⁺ neutrophils, CD45⁺CD11b⁺Ly6C⁺ monocytes, and CD45⁺CD11c⁺ dendritic cells at the infection site. Data are representative of ≥3 experiments, n≥4/group. Statistical differences in lesion development were analyzed with a 2-way ANOVA with repeated measures, and cell numbers using a Mann-Whitney U-test. *p <0.05; **p <0.01.

Fig 3. L. mexicana leads to inflammasome activation in neutrophils.

(A) Bone marrow-derived C57BL/6 neutrophils (BMNs) were isolated, primed with LPS and infected with L. mexicana promastigotes at a multiplicity of infection (MOI) of 2, 5, and 10 for 16 hours, or exposed to Nigericin for 4 hours as a positive control. Immunoblotting of GSDMD and beta-actin was performed on cell extracts. (B) BMNs were similarly infected with other Leishmania species at the MOI 5. (C) Comparison of GSDMD cleavage between L. mexicana M379 and Tab3 strains at the indicated MOI. (D) C57BL/6 BMNs were isolated, primed with LPS and infected at a MOI of 10 with L. mexicana promastigotes or axenic amastigotes, as indicated. Immunoblotting of GSDMD and beta-actin was performed on cell extracts and Nigericin was used as a positive control (same blot and exposure). (E) LPS-primed C57BL/6 BMNs were infected with L. mexicana promastigotes and axenic amastigotes at MOI of 5 and propidium iodide (PI) uptake was quantified over 18 hours of infection. (F) The corresponding IL-1β release in supernatants was assessed by ELISA. NS (non-stimulated), LPS (LPS 100ng/ mL priming alone), MOI (multiplicity of infection), Pro (promastigotes), Ama (amastigotes). Data are representative of >2 independent experiments. Differences between populations are analyzed by 2-way ANOVA (E) and Unpaired t-test (F).

Fig 4. mexicana leads to NLRP1-dependent GSDMD-cleavage and pyroptosis in neutrophils.

L. BMNs were primed with LPS and infected with L. mexicana promastigotes at the indicated MOI for 16 hours or exposed to Nigericin for 4 hours as a positive control. (A) Immunoblot of GSDMD and beta-actin of infected Nlrp1^-/- and C57BL/6 BMNs. (B) Immunoblot for GSDMD and beta-actin of Casp1^-/-, Casp11^-/- and C57BL/6 BMNs after infection with L. mexicana. Images are representative of n>3 independent experiments. (C) IL-1β release was measured by ELISA in LPS-primed Gsdmd^-/- and C57BL/6 BMN supernatant after 16 hours of infection. (D) LPS-primed Gsdmd^-/- and C57BL/6 BMNs were infected with L. mexicana promastigotes at MOI of 5 and propidium iodide (PI) uptake was quantified over 24 hours of infection. (E) LPS-primed BMNs were infected with L. mexicana for 16 hours and lactate dehydrogenase (LDH) release was analyzed, with values expressed as a percentage of total LDH upon full lysis of BMNs. Data are shown as mean ± SD and representative of n>3 independent experiments. (F) Human neutrophils from healthy donors were primed with LPS and infected with L. mexicana at MOI of 5 and PI uptake analyzed in presence or absence of the caspase-1 inhibitor ac-YVAD-cmk (G). Human neutrophils were similarly primed and exposed to the NLRP1 activator Val-boroPro in presence or absence of caspase-1 inhibitor. (H). IL-1β release was measured in human neutrophils exposed to L. mexicana at the indicated MOI or to Val-boroPro, in the presence or absence of caspase-1 inhibitor. These results are representative of three experiments. NS (non-stimulated), NI (non-infected), LPS (LPS 100ng/ mL for mouse neutrophils, 500ng/mL for human neutrophils), VbP (Val-boroPro at 10μM). (D, F, G) 2-way ANOVA with Tukey’s multiple comparisons test, (C, H) Unpaired t-test. *p <0.05; **p <0.01. ***p <0.001, ****p<0.0001.

Fig 5. GSDMD regulates the L. mexicana neutrophil niche during the first days of infection.

(A) C57BL/6 and Gsdmd^-/- mice were infected i.d with metacyclic dsRed⁺ L. mexicana promastigotes and the number and frequency of neutrophils in the bone marrow, blood, and at the infection site was determined by flow cytometry at the indicated timepoints. The flow cytometry gating strategy is shown. (B) The frequency of CD45⁺CD11b⁺Ly6G⁺ neutrophils at 24 hours p.i. in bone marrow, blood, and infected ear is represented (C) The number of CD45⁺CD11b⁺Ly6G⁺ neutrophils present at the infection site at 24, 48, and 72 hours p.i., as analyzed by flow cytometry and (D) the number of L. mexicana-DsRed⁺ CD45⁺CD11b⁺Ly6G⁺ C57BL/6 and Gsdmd^-/- infected neutrophils is given for each genotype and analyzed time point p.i. (E). The parasite burden in the infected C57BL/6 and Gsdmd^-/- ear was determined by limiting dilution (LDA) analysis 48 hours p.i. (F) C57BL/6 and Gsdmd^-/- mice were similarly infected with L. mexicana and 21 days p.i, the parasite burden assessed by LDA in infected ears. Data are shown as mean ± SD and are representative of n>2 experiments per time point, with n>4 mice/group. Mann-Whitney U-test *p<0.05, **p<0.01.

Fig 6. NLRP1 regulates the neutrophil niche through caspase-1 and GSDMD-dependent pyroptosis.

(A) BMNs isolated from naïve Gsdmd^-/-, Casp1^-/-, Nlrp1^-/- mice or C57BL/6 control neutrophils were stained with the indicated fluorescent dyes, mixed at a 1:1 ratio and transferred into Genista neutropenic mice at the time of infection with L. mexicana metacyclic promastigotes, as represented in the experimental design. Representative flow cytometry plot showing the lack of mature CD11b⁺Ly6G⁺ neutrophils in (B) peripheral blood and (C) uninfected ear of Genista mice. (D) The gating strategy to analyse the genotype of the two transferred neutrophil populations present in Genista mice is shown. Bar-plot showing the relative frequency of (E) Gsdmd ^-/- (F), Casp1^-/- (G) and Nlrp1^-/- neutrophils compared to the simultaneously transferred C57BL/6 neutrophils at the site of infection 24 hours p.i. Data are shown as mean ± SD and are representative of n>2 experiments, with n>4 ears/group. Wilcoxon signed-rank test, *p<0.05, **p<0.01.

Fig 7. Early GSDMD activation in neutrophils reduces L. mexicana-driven pathology and parasite burden.

(A) Gsdmd^-/-, Gsdmd^ΔPMN, and WT, mice were infected i.d. with L. mexicana promastigotes, and lesion development was measured. (B) Ear lesion size at 11 weeks p.i. and (C) representative pictures of WT, Gsdmd^-/- and Gsdmd^ΔPMN infected ears. (D) Parasite load was determined by LDA. (E) Representative images of immunohistology of ear cryosections at 8 weeks p.i., stained for Ly6G⁺ neutrophils (green), dsRed-expressing L. mexicana parasites (red), and DAPI (grey). Scale bar, 50 μm. (F) Quantification of immunofluorescence images. (G) Eleven weeks p.i. the number of CD45⁺ cells, CD45⁺CD11b⁺Ly6G⁺ neutrophils, CD45⁺CD11b⁺Ly6C⁺ monocytes, and CD45⁺CD11c⁺ dendritic cells at the infection site, was analyzed by flow cytometry (H) C57BL/6 and Gsdmd^-/- BMNs were isolated and adoptively transferred at the time of infection in two separate groups of Genista mice as indicated in the experimental design. (I) Ear lesion score was measured over the indicated time. (J) Ten weeks p.i., representative pictures of the indicated infected ear are shown and (K) the number of living parasites was determined by LDA. Data are shown as mean ± SD and are representative of 2 experiments, n≥3/group. Statistical differences in lesion development were analyzed with a 2-way ANOVA repeated measures, and differences in parasite load and cell populations were analyzed with Mann-Whitney U test. *p <0.05; **p <0.01.

Fig 8. The role of NLRP1-dependent neutrophil pyroptosis during L. mexicana infection.

Neutrophils are rapidly recruited upon infection with L. mexicana and become infected with parasites. Infection leads to activation of the NLRP1 inflammasome and caspase-1 in neutrophils, which cleaves and activates the pore-forming protein GSDMD. GSDMD cleavage leads to neutrophil pyroptosis, which contributes to reduction of the parasite load. GSDMD regulates the early neutrophil parasite niche limiting the subsequent pathology. In absence of GSDMD, more neutrophils provide increased parasite niche and enhance pathology. Figure created with biorender.com.