Dual oxidase 1 is dispensable during Mycobacterium tuberculosis infection in mice

Abstract

Introduction: Mycobacterium tuberculosis (Mtb) is the primary cause of human tuberculosis (TB) and is currently the second most common cause of death due to a singleinfectious agent. The first line of defense against airborne pathogens, including Mtb, is the respiratory epithelium. One of the innate defenses used by respiratory epithelial cells to prevent microbial infection is an oxidative antimicrobial system consisting of the proteins, lactoperoxidase (LPO) and Dual oxidase 1 (Duox1), the thiocyanate anion (SCN-) and hydrogen peroxide (H2O2), which together lead to the generation of antimicrobial hypothiocyanite (OSCN-) in the airway lumen. OSCN- kills bacteria and viruses in vitro, but the role of this Duox1-based system in bacterial infections in vivo remains largely unknown. The goal of this study was to assess whether Duox1 contributes to the immune response against the unique respiratory pathogen, Mtb.

Methods: Duox1-deficient (Duox1 KO) and wild-type (WT) mice were infected with Mtb aerosols and bacterial titers, lung pathology, cytokines and immune cell recruitment were assessed.

Results and discussion: Mtb titers in the lung, spleen and liver were not different 30 days after infection between WT and Duox1 KO mice. Duox1 did not affect lung histology assessed at days 0, 30, and 90 post-Mtb infection. Mtb-infected Duox1 KO animals exhibited enhanced production of certain cytokines and chemokines in the airway; however, this response was not associated with significantly higher numbers of macrophages or neutrophils in the lung. B cell numbers were lower, while apoptosis was higher in the pulmonary lesions of Mtb-infected Duox1 KO mice compared to infected WT animals. Taken together, these data demonstrate that while Duox1 might influence leukocyte recruitment to inflammatory cell aggregates, Duox1 is dispensable for the overall clinical course of Mtb lung infection in a mouse model.

Keywords: Duox1; Mycobacterium tuberculosis; antimicrobial; epithelium; inflammation.

Figure 1

Mtb growth in vivo in Duox1 KO mice. Mature Duox1 KO and WT mice were aerogenically infected with 200 CFU of Mtb (Erdman). (A) On 30 and 90 days post-Mtb infection, mice from each genotype were euthanized and lungs (n=9), spleens (n=14) and livers (n=14) were removed for CFU burden assessment by colony counting. (B) On 1 day post-infection, lungs from mice (n=5) of each genotype were removed for CFU counts. Mann-Whitney U test was used to compare CFUs between animal groups. Mean+/-S.E.M. CFU, colony-forming unit; Duox1 KO: Duox1-deficient; WT, wild-type.

Figure 2

Duox1 does not affect Mtb proliferation in the lung. Lungs from Mtb-infected Duox1 KO and WT mice were harvested at 1 and 30 day(s) post-infection, respectively. Fixed and processed lung tissues were stained with primary anti-Mtb antibody followed by a secondary AF488 antibody. (A) Representative fluorescent images of six similar results are shown. White scale bars indicate 25 μm. Mean green fluorescent intensities (MFI) indicative of Mtb burden were quantified for (B) 1 day post-infection and (C) 30 days post-infection in Duox1 KO (n=6) and WT mice (n=6). Error bars represent ± SEM. Mann Whitney U-test. Duox1 KO: Duox1-deficient; WT, wild-type; RFU, relative fluorescence unit.

Figure 3

Lung pathology during Mtb infection is independent of Duox1. Mtb-infected lung lobes from both Duox1 KO and WT mice were isolated and fixed in 10% buffered formalin at 1, 30, and 90 day(s) post-infection. Fixed lung tissues were processed for histology, sectioned, and subjected to hematoxylin and eosin staining. Lungs collected at 1 day post-infection from both mouse background (A, D) do not show any evidences of lung pathology while granuloma-like lesions in both groups are visible at 30 (B, E) and 90 (C, F) days post-infection. (G) Comparative percentage of lung tissue affected (0% to 80%). (H) The number of inflammatory aggregates inside the lung tissues (from 0 to 40 maximum). The lung histological score was assessed with a scale from 0 to 4 and included the (I) alveoli, (J) the alveolar edema, (K) necrosis, and (L) polymorphonuclear neutrophils as well as (M) perivascular score. (N) The vascular, (O) IP score and (P) pleuritis are also shown. Sample sizes: n=8 Duox1 KO; n=8 WT mice for histopathological scoring at 1 day post-infection. For 30 days post-infection: n=9 Duox1 KO and n=9 WT mice were used. For 90 days post-infection, n=6 WT and n= 7 Duox1 KO animals were assessed. ANOVA with multiple comparison was used for statistical data analysis. *p<0.05, **p<0.01, ****p<0.0001. IP, intraperitoneal, WT, wild-type; day post infection, days post-infection; PMN, polymorphonuclear neutrophil granulocyte; Duox1 KO, Duox1-deficient; WT, wild-type.

Figure 4

Myeloid cell populations in the lung tissues of Mtb-infected WT and Duox1 KO mice. (A) Mean observed cell proportions determined by flow cytometry in lung lysates from two independent experiments (n= 5 per mouse genotype) of Mtb-infected WT mice at 1 day post-infection. (B) Marker expression by t-sne in Mtb-infected WT mice at 1 day post-infection. (C-F) Marker expressions by t-sne in Mtb-infected Duox1 KO and WT mice at 1 and 30 day(s) post-infection as indicated. Duox1 KO, Duox1-deficient; WT, wild-type.

Figure 5

Neutrophil markers in the blood and airways of Mtb-infected Duox1 KO mice. The concentrations of MPO and NE were measured in the BAL fluid, sera and lung lysates from Mtb-infected WT and Duox1 KO mice at 1 and 30 days post-infection. NE enzymatic activities in cell-free BAL fluid and sera are also shown. The following parameters were measured at 1 and 30 day(s) post-infection. (A) BAL MPO, (B) serum MPO, (C) MPO in lung lysates, (D) BAL NE, (E) serum NE, (F) NE in lung lysates, (G) NE activity in BAL and (H) NE activity in sera. Statistical comparisons were done with ANOVA followed by multiple comparisons. No significant differences were found. The concentrations of both MPO and NE are expressed in pg/ml. BAL, bronchoalveolar lavage fluid; Duox1 KO, Duox1-deficient; MPO, myeloperoxidase; NE, neutrophil elastase; WT, wild-type.

Figure 6

MPO and NE expressions in the lungs of Duox1 KO mice infected with Mtb. Mtb-infected lungs from Duox1 KO and WT mice were harvested at 1 or 30 day(s) post-infection, fixed and processed for staining with rabbit anti-MPO or rabbit anti-NE antibodies followed by anti-rabbit secondary antibody (HRP-linked) and DAB amplification. Sections are counterstained with hematoxylin and representative micrographs are shown in panels. (A) Representative micrograph of MPO staining of Mtb-infected lungs (bronchiolar area) of Duox1 KO and WT mice at 1 and 30 day(s) post-infection. (B) Representative photomicrographs of NE in Mtb-infected lungs at 30 days post-infection, at low (200 µm) and high magnification (20 µm), respectively. Duox1 KO, Duox1-deficient; MPO, myeloperoxidase; NE, neutrophil elastase; WT, wild-type.

Figure 7

Duox1-depndent B cell accumulation in Mtb-infected lungs. Duox1 KO and WT mice were aerogenically infected with Mtb Erdman (200 CFU). Lung tissues were obtained at 1 and 30 day(s) post-infection, fixed and immunofluorescence staining was performed using anti-mouse CD45R specific for B cells (red). DAPI (blue) was used to depict cellular DNA. (A) Representative images of five similar results (n = 5 WT and n = 5 Duox1 KO mice) are shown. Quantitation of CD45R fluorescence at (B) 1 day post-infection and (C) 30 days post-infection. Scale bars indicate 50 μm. Mean fluorescent intensity (MFI) ± SEM in arbitrary unit (AU) is shown (Mann–Whitney U test). Significance level is indicated as *, p < 0.05. Duox1 KO, Duox1-deficient; WT, wild-type.

Figure 8

Caspase-3 activation in Mtb-infected Duox1 KO lungs. Mtb-infected lungs of WT and Duox1 KO mice were harvested and fixed at 1 or 30 day(s) post-infection. (A) Representative immunofluorescence staining of cleaved caspase-3 (CC3, red), CD68 (macrophages, green) and DAPI (DNA, blue). Scale bar: 25 µm). (B) Representative results of cleaved caspase-3 immunohistochemistry images at 1 (n=3 per genotype) and 30 days post-infection (n=8 per genotype). Scale bar indicates 25 µm. (C) Comparative non-stochiometric DAB signal quantification with ImageJ shows that the cleaved caspase-3 signal is significantly higher in the lung inflammatory aggregates of Duox1 KO mice compared to WT mice at 1 (n=3) and 30 (n=8) day(s) post-infection. A 2-way Anova with Sidak’s multiple comparison test was used for statistical data analysis. Significance levels are indicated as **P ≤ 0.01. AU, arbitrary unit; Br, bronchus; CC3, cleaved caspase-3; Duox1 KO, Duox1-deficient; WT, wild-type.