Inflammasome-Independent NLRP3 Restriction of a Protective Early Neutrophil Response to Pulmonary Tularemia

Abstract

Francisella tularensis (Ft) causes a frequently fatal, acute necrotic pneumonia in humans and animals. Following lethal Ft infection in mice, infiltration of the lungs by predominantly immature myeloid cells and subsequent myeloid cell death drive pathogenesis and host mortality. However, following sub-lethal Ft challenge, more mature myeloid cells are elicited and are protective. In addition, inflammasome-dependent IL-1β and IL-18 are important for protection. As Nlrp3 appears dispensable for resistance to infection with Francisella novicida, we considered its role during infection with the virulent Type A strain SchuS4 and the attenuated Type B live vaccine strain LVS. Here we show that both in vitro macrophage and in vivo IL-1β and IL-18 responses to Ft LVS and SchuS4 involve both the Aim2 and Nlrp3 inflammasomes. However, following lethal infection with Francisella, IL-1r-, Caspase-1/11-, Asc- and Aim2-deficient mice exhibited increased susceptibility as expected, while Nlrp3-deficient mice were more resistant. Despite reduced levels of IL-1β and IL-18, in the absence of Nlrp3, Ft infected mice have dramatically reduced lung pathology, diminished recruitment and death of immature myeloid cells, and reduced bacterial burden in comparison to wildtype and inflammasome-deficient mice. Further, increased numbers of mature neutrophil appear in the lung early during lethal Ft infection in Nlrp3-deficient mice. Finally, Ft infection induces myeloid and lung stromal cell death that in part requires Nlrp3, is necrotic/necroptotic in nature, and drives host mortality. Thus, Nlrp3 mediates an inflammasome-independent process that restricts the appearance of protective mature neutrophils and promotes lethal necrotic lung pathology.

Fig 1. F. tularensis and F. novicida activate both Nlrp3- and Aim2-inflammasomes.

(A-C) Levels of IL-1β, IL-18 or IL-6 measured in culture supernatants of BMDM infected with Ft LVS, SchuS4 or F. novicida at MOI = 100 for 24 h (mean ± SD of three independent experiments, Student’s t-test, *p<0.05, **p<0.01 or ***p<0.001 indicates the significant difference from wildtype BMDM).

Fig 2. Nlrp3^-/- mice are less susceptible to lethal pulmonary Ft infection.

(A) Survival of mice following lethal Ft LVS (1000 cfu) infection (% survival of three independent experiments, n = 18 mice, Log-rank (Mantel-Cox) test, ***p<0.001) and lung bacterial burden (mean ± SD of three independent experiments, n = 9 mice, Student’s t-test, *p<0.05). (B) Histological section of lung from wildtype mice infected with Ft LVS shows inflammatory foci and massive necrosis (‘necrotizing inflammation’) at 6 dpi, while these pathological changes were less extensive in lung section of Nlrp3^-/- mice (HE, 400x). (C) Survival of mice following Ft SchuS4 (10 cfu) infection (% survival of three independent experiments, n = 15 mice, Log-rank test, *p<0.05) and lung bacterial burden (mean ± SD of two independent experiments, n = 6, Student’s t-test, *p<0.05). (D) Survival of mice following lethal F. novicida (20 cfu) infection (% survival of two independent experiments, n = 12 mice, Log-rank test, *p<0.05) and lung bacterial burden (mean ± SD, n = 5 mice, Student’s t-test, *p<0.05).

Fig 3. Nlrp3 mediates pathogenesis of pulmonary tularemia in an inflammasome-independent manner.

(A) Levels of IL-1β and IL-18 measured in lung homogenates at indicated days post-Ft LVS infection (mean ± SD of two independent experiments, n = 6 mice, Student’s t-test, **p<0.01 indicates difference from wildtype mice). (B) Levels of IL-1β and IL-18 measured in lung homogenates at indicated days post-Ft LVS infection (mean ± SD of two independent experiments, n = 6 mice, Student’s t-test, *p<0.05 or ***p<0.001 indicates difference from wildtype mice). (C) Level of IL-6 and TNFα measured in lung homogenates at indicated days post-Ft LVS infection (mean ± SD of two independent experiments, n = 6 mice, Student’s t-test). (D) Survival of mice following lethal Ft LVS (1000 cfu) infection (% survival of two independent experiments, n = 12, Log-rank test). (E) Lung bacterial burden (mean ± SD of two independent experiments, n = 6 mice, Student’s t-test, **p<0.01). (F) Survival of mice following with sub-lethal Ft LVS (500 cfu) infection (% survival of two independent experiments, n = 12, Log-rank test). (G) Survival of mice following with lethal (1000 cfu) Ft LVS infection and treatment with MCC950 (250μg/mouse daily at 2–7 dpi) or with Glyburide (500μg/mouse daily at 2–6 dpi) by i.p route (% survival of two independent experiments, n = 12 mice, Log-rank test). (H) Survival of mice following with sub-lethal Ft LVS (500 cfu) infection and MCC950 treatment (250μg/mouse daily at 2–7 dpi) by i.p route (% survival of two independent experiments, n = 10 mice, Log-rank test).

Fig 4. Innate antibody response is dispensable for protection during acute phase of tularemia.

(A-B) Levels of IL-1β and IL-18 measured in serum (mean ± SD of two independent experiments, n = 6 mice, Student’s t-test, *p<0.05 or **p<0.01 indicates the difference from wildtype mice at 6 dpi). (C) Level of anti-Ft LPS IgM antibody titer measured in serum (mean ± SD of OD₄₅₀ from two independent experiments, n = 6 mice, Student’s t-test). (D) Levels of anti-Ft IgG and IgA antibodies measured in serum (mean ± SD of OD₄₅₀ from two independent experiments, n = 6, Student’s t-test).

Fig 5. An early neutrophil response in Nlrp3^-/- mice is protective during pulmonary tularemia.

(A-E) Total numbers of CD11b⁺ myeloid cells, Ly6G⁺ neutrophils, polymorphonucleated-MDSC (pMDSC), F4/80⁺macrophages or mono-nucleated-MDSC (mMDSC) in LVS-infected lungs (mean ± SD of two independent experiments, n = 6 mice, Student’s t-test, *p<0.05 indicates difference from wildtype mice at specific dpi). (F) Survival of mice following Ft LVS (1000 cfu) infection and treatment with anti-Gr-1 antibody (200μg/mouse, i.p route) at -1 and 1 dpi (% survival of two experiments, n = 6 mice, Log-rank test, *p<0.05). (G) Total numbers of Ly6G^high neutrophils in lungs of mice following i.n. instillation of LPS and infected 48 h later with Ft LVS (1000 cfu) (mean ± SD of three mice, Student’s t-test, *p<0.05 and **p<0.01 indicate the difference from those mice receiving no LPS but infected with LVS). (H) Survival of mice following Ft LVS (1000 cfu) infection. Naïve mice were first adoptively transferred with PMN (1 x10⁶ cells/mouse; isolated from bone marrow cells) or CD3+ T cells (1 x10⁶ cells/mouse; isolated from spleen) by intratracheal intubation and infected with Ft LVS on the following day. Other groups of naïve mice were treated with LPS (100 or 10 μg/mouse, i.n instillation) to elicit PMN/myeloid cell response, infected with Ft LVS after 48 h, and monitored for survival and mortality (% survival of two independent experiment, n = 10 mice, Log-rank test, *p<0.05 or ***p<0.001 indicates difference from naïve mice infected with LVS alone). (I) Lung bacterial burden in mice receiving i.n. instillation of LPS and infected 48 h later with Ft LVS (1000 cfu) (mean ± SD of four mice, Student’s t-test, *p<0.05 and **p<0.01 indicate the difference from those mice receiving no LPS but infected with LVS). (J) Lung bacterial burden in mice receiving PMN or CD3+ T cells post-infection with Ft LVS (1000 cfu) (mean ± SD of four mice, Student’s t-test, *p<0.05 and **p<0.01). (K) Levels of IL-17, KC and MCP-1 in lung homogenates (mean ± SD of two independent experiments, n = 6 mice, Student’s t-test). (L) Survival of mice following Ft LVS (1000 cfu) infection and treatment with anti-IL-17 antibody (200μg/mouse, i.p route) at 1 and 3 dpi (% survival of two independent experiments, n = 8 mice, Log-rank test). (M) Total numbers of Ly6G^high neutrophils in lungs of wildtype and IL-1r^-/- mice infected with lethal (1000 cfu) Ft LVS (mean ± SD of two independent experiments, n = 6 mice, Student’s t-test, *p<0.05 and **p<0.01).

Fig 6. Tissue pathology and necrotic death of myeloid cells is reduced in Ft infected Nlrp3^-/- mice.

(A) Lung pathology score for mice infected with lethal Ft LVS (1000 cfu). Pathology score was calculated by analysis of lung sections (n = 6 mice) for location/type/and extent of inflammation and necrosis (see Methods) (mean ± SD of two experiments, Mann-Whitney test, *p<0.05 indicates difference from wildtype mice). (B) Individualized pathology scores for inflammatory site and cellular types (mean ± SD of two experiments as shown in panel A). (C-D) Individualized pathology scores for number/size of inflammatory foci and extent of necrosis (mean ± SD of two experiments, n = 6 mice, Mann-Whitney test, *p<0.05). (E) LDH level in BAL fluid following LVS-infection (mean ± SD of two independent experiments, Student’s t-test, **p<0.01). (F) Positive immunoreaction for localization of LDH (asterisk), as an indicator of necrosis, in representative lung section at 6 dpi (IHC with hematoxylin counterstaining, 400x). (G) Per cent 7-AAD⁺ neutrophils (PMN) and macrophages (MØ) in Ft-infected lungs at 3 and 6 dpi (mean ± SD of two independent experiments, Student’s t-test, *p<0.05 indictaes difference from wildtype mice). (H) Total number of 7-AAD⁺ cells in Ft-infected lungs at 6 dpi (mean ± SD of two independent experiments, Student’s t-test, *p<0.05). (I) Per cent necrotic cells calculated by LDH release in BMDM infected with Ft LVS at increased MOI for different time points (mean ± SD of three independent experiments, Student’s t-test, **p<0.01 indicates difference from 3 and 6 h at specified MOI). (J) Per cent necrosis (LDH release) of BMDM pre-treated (30 min) with z-DEVD-fmk (50 μM) or Nec-1 (50 μM) and then infected with Ft LVS (MOI = 100) for 24 h (mean ± SD of three independent experiments, Student’s t-test, *p<0.05 indicates difference from LVS-infected cells treated with DMSO). (K) Per cent necrosis (LDH release) of PMN or lung macrophages infected with Ft LVS (MOI = 100) for 24 h (mean ± SD of two independent experiments, Student’s t-test, *p<0.05 indicates difference from wildtype cells). (L) Per cent necrosis (LDH release) of LA-4 cells infected with Ft LVS (MOI = 100) for 24 h (mean ± SD of three independent experiments, Student’s t-test, *p<0.05 indicates difference from LVS-infected cells treated with DMSO).

Fig 7. Necrostatin-1s treatment protects Ft infected mice.

(A) Survival of wildtype mice infected with Ft LVS and then treated i.p. with z-DEVD-fmk (200 μg/mouse) or Nec-1 (200 μg/mouse) daily between 2–6 dpi (%survival of two independent experiments, n = 10 mice, Log rank test). (B) Lung pathology scores wildtype mice infected with Ft LVS and then treated i.p. with z-DEVD-fmk (200 μg/mouse) or Nec-1 (200 μg/mouse) as above (mean ± SD of two experiments, n = 4 mice, Student’s t-test, *p<0.05. (C) Survival of wildtype and Nlrp3^-/- mice infected with Ft LVS and then treated i.p. with and without Nec-1s (200 μg/mouse) daily between 2–6 dpi (% survival of two independent experiments, n = 10 mice, Log rank test, n.s. = p>0.5). (D). Lung bacterial burden in wildtype mice infected with Ft LVS and then treated i.p. with Nec-1s as above (mean ± SD of four mice, Student’s t-test**p<0.01).