Neutrophil Extracellular Traps are Involved in the Innate Immune Response to Infection with Leptospira

Abstract

NETosis is a process by which neutrophils extrude their DNA together with bactericidal proteins that trap and/or kill pathogens. In the present study, we evaluated the ability of Leptospira spp. to induce NETosis using human ex vivo and murine in vivo models. Microscopy and fluorometric studies showed that incubation of human neutrophils with Leptospira interrogans serovar Copenhageni strain Fiocruz L1-130 (LIC) resulted in the release of DNA extracellular traps (NETs). The bacteria number, pathogenicity and viability were relevant factors for induction of NETs, but bacteria motility was not. Entrapment of LIC in the NETs resulted in LIC death; however, pathogenic but not saprophytic Leptospira sp. exerted nuclease activity and degraded DNA. Mice infected with LIC showed circulating NETs after 2 days post-infection (dpi). Depletion of neutrophils with mAb1A8 significantly reduced the amount of intravascular NETs in LIC-infected mice, increasing bacteremia at 3 dpi. Although there was a low bacterial burden, scarce neutrophils and an absence of inflammation in the early stages of infection in the kidney and liver, at the beginning of the leptospiruric phase, the bacterial burden was significantly higher in kidneys of neutrophil-depleted-mice compared to non-depleted and infected mice. Surprisingly, interstitial nephritis was of similar intensity in both groups of infected mice. Taken together, these data suggest that LIC triggers NETs, and that the intravascular formation of these DNA traps appears to be critical not only to prevent early leptospiral dissemination but also to preclude further bacterial burden.

Fig 1. Leptospira interrogans induce NETs in a concentration-dependent manner.

Human neutrophils (2x10⁵/mL) were incubated with Leptospira interrogans serovar Copenhageni (LIC) (MOI = 50) for 180 min and (A) visualized with a Nikon E200 photomicroscope or (B) fixed (PF 4%) and stained with propidium iodide (red) or with the specific marker anti-neutrophil elastase (green), and analyzed by fluorescence microscopy (n = 10). Scale bar indicates 50 μm. (C) DNA or nucleosomes were measured by fluorometry (black bars) or ELISA kit (white bars) respectively in supernatants of LICa (bacteria alone) or unstimulated neutrophils (None) (500/μL) (negative control), stimulated with PMA (50 ng/mL) (positive control), or with LIC (MOI 50, 5 or 1) for 180 min Bars represent standard error of the mean (SEM) of assays from 3–10 independent assays; ***p <0.0001 vs. None. ##p<0.01 and ####p<0.0001 vs. LIC 50.

Fig 2. Pathogenicity and viability are relevant factors for NET formation.

(A) Human neutrophils (2x10⁵/mL) were incubated with live or inactivated with 4% PF or by heat Leptospira interrogans serovar Copenhageni (LIC) or Leptospira biflexa serovar Patoc (Patoc) (MOI = 50) for 180 min and then fixed (PF 4%), stained with propidium iodide (red) or with the specific marker anti-neutrophil elastase (green), and analyzed by fluorescence microscopy (n = 10). Scale bar indicates 50 μm. (B) Quantification of NETs released by fluorometry in the same conditions as in (A). Bars represent standard error of the mean (SEM) of assays from ten independent assays; *p <0.05, ** p <0.01.

Fig 3. Leptospira spp. motility is not a relevant factor for NET formation.

(A) Human neutrophils (2x10⁵/mL) were incubated with Leptospira interrogans serovar Manilae (LIM) and Leptospira biflexa serovar Patoc (Patoc) and their non-mobile mutants flaA2 and flaB (MOI = 50) for 180 min and then fixed (PF 4%) and stained with propidium iodide (red) or with the specific marker anti-neutrophil elastase (green) and analyzed by fluorescence microscopy (n = 10). Scale bar indicates 50 μm. (B) Quantification of NETs released by fluorometry in the same conditions as in (A).

Fig 4. NETs kill Leptospira sp.

(A) Percentage viability after 180 minutes of Leptospira interrogans serovar Copenhageni (LIC) (MOI = 50) alone, after being incubated with human neutrophils (Neu) (2x10⁵/mL), or in the presence of DNase (0.25 U/mL). (B) Percentage viability of LIC (MOI = 50) after 60 minutes of incubation with different concentrations of recombinant histone H4. Bars represent standard error of the mean (SEM) of assays from five independent assays; *p <0.05, **p <0.01, ****p <0.0001.

Fig 5. Pathogenic but not saprophyte Leptospira spp. degrade DNA.

Representative analysis of DNA digestion by gel electrophoresis. From left to right: plasmid DNA (100 ng/μL) after incubation with PBS (negative control), DNase I (positive control), and live Leptospira interrogans serovar Copenhageni (LIC) or Leptospira biflexa serovar Patoc (Patoc) (1x10⁸/mL) after 60 minutes of incubation at 37°C.

Fig 6. Early role of NETs in murine experimental model.

(A) Granulocytes were quantified in blood samples taken from C57BL/6J mice treated or not with mAb1A8 antibody with a veterinarian hematology counter. Bars represent the mean ± SEM. * p <0.05. (B) C57BL/6J male weanlings were inoculated intraperitoneally with 200 μL of 1x10⁷/mL pathogenic Leptospira interrogans serovar Copenhageni (LIC) and 0.25 mg of non-immune rat IgG or purified anti-Ly6G rat mAb1A8 every 48 h. Blood was collected by retro-orbital venous puncture and circulating nucleosomes were measured by ELISA *p <0.05, ***p <0.001, ****p <0.0001. (C) Blood, kidney, and liver samples of non-depleted and mAb1A8-depleted LIC-infected mice were collected at 3 days post-infection. Leptospiral DNA was quantified by real-time PCR and normalized to host cell number. Bars represent standard error of the mean (SEM) of assays from two independent assays; *p <0.05. (D) Kidney, and liver tissues samples of non-depleted animals were immunostained for neutrophils using anti-Ly6G rat mAb1A8 at 3 days post-infection. Representative hematoxylin and eosin (H&E) stained kidney and liver tissues samples of non-depleted animals at 3 days post-infection showing histology (E) and inflammation score (F) do not showed any alteration after analysis (n = 6–10 mice). Scale bar indicates 50 μm.

Fig 7. Renal bacterial burden at the beginning of the leptospiruric phase and immunohistochemistry analysis.

(A) C57BL/6J male weanlings received 0.25 mg of non-immune rat IgG or purified anti-Ly6G rat mAb1A8 every 48 h and 200 μL of 1x10⁷/mL pathogenic Leptospira interrogans serovar Copenhageni (LIC). Kidney samples of non-depleted and neutrophil-depleted LIC-infected mice were collected at 14 days post-infection. Leptospiral DNA was quantified by real-time PCR and normalized to host cell number. Bars represent mean ± SEM of two independent experiments; **p <0.01. (B) Positive renal tubules were counted under a 40x field. Bars represent mean ± SEM of three independent observations; **p <0.01. (C) Representative positive slides of immunohistochemical staining of LipL32 in kidney from LIC-infected mice rat IgG-treated or mAb1A8-treated. The results are representative of two different experiments (n = 6–10 mice). Slides were counterstained with hematoxylin. Scale bar indicates 50 μm.

Fig 8. Renal histopathology at the beginning of the leptospiruric phase.

C57BL/6J male weanlings received 0.25 mg of non-immune rat IgG or purified anti-Ly6G rat mAb1A8 every 48 h and 200 μL 1x10⁷/mL pathogenic Leptospira interrogans serovar Copenhageni (LIC). Representative hematoxylin and eosin (H&E) stained kidney sections from 14 days post-infection showing histology (A) and inflammation score (B) from uninfected mAb1A8-treated mice, rat IgG-treated LIC-infected mice, and mAb1A8-treated LIC-infected mice (n = 6–10 mice). Scale bar indicates 50 μm.