An In Vivo Model of Separate M. tuberculosis Phagocytosis by Neutrophils and Macrophages: Gene Expression Profiles in the Parasite and Disease Development in the Mouse Host

Abstract

The role of neutrophils in tuberculosis infection remains less well studied compared to that of the CD4⁺ T-lymphocytes and macrophages. Thus, alterations in Mycobacterium tuberculosis transcription profile following phagocytosis by neutrophils and how these shifts differ from those caused by macrophage phagocytosis remain unknown. We developed a mouse model that allows obtaining large amounts of either neutrophils or macrophages infected in vivo with M. tuberculosis for mycobacteria isolation in quantities sufficient for the whole genome RNA sequencing and aerosol challenge of mice. Here, we present: (i) the differences in transcription profiles of mycobacteria isolated from liquid cultures, neutrophils and macrophages infected in vivo; (ii) phenotypes of infection and lung inflammation (life span, colony forming units (CFU) counts in organs, lung pathology, immune cells infiltration and cytokine production) in genetically TB-susceptible mice identically infected via respiratory tract with neutrophil-passaged (NP), macrophage-passaged (MP) and conventionally prepared (CP) mycobacteria. Two-hour residence within neutrophils caused transcriptome shifts consistent with mycobacterial transition to dormancy and diminished their capacity to attract immune cells to infected lung tissue. Mycobacterial multiplication in organs did not depend upon pre-phagocytosis, whilst survival time of infected mice was shorter in the group infected with NP bacilli. We also discuss possible reasons for these phenotypic divergences.

Keywords: M. tuberculosis transcriptome; immune response; infection progression; macrophages; mouse model; neutrophils.

Figure 2

M. tuberculosis genes differentially expressed after phagocytosis by macrophages (MP) and neutrophils (NP). (A) Scatterplot comparing transcript measurements (FPM, fragments per million mapped) for MP and NP transcriptomes. Each dot represents one transcript. Genes that are differentially expressed in MP and NP are marked by blue dots. Scatter plot built using ViDGER package [54]. (B) Heat map of DEGs. Heat map was visualized by TPM (transcripts per kilobase million) approach using heat-mapper web tool [55]. The log₂FC numbers of genes differentially expressed in MP and NP versus CP are provided; NS—non significant. (C) Validation of DEGs by qRT-PCR. mRNA expression levels of Rv0440, Rv3131, and Rv3133 genes were estimated in MP, NP, and CP cDNA samples. mRNA expression was normalized to that of 16S rRNA. Statistical analysis was performed using GraphPad Prism 6.0 (GraphPad Software Inc., La Jolla, CA). The data are presented as the mean ± SD of three biological replicates for each sample, Mann–Whitney U-test. Differences were considered statistically significant at p < 0.05. * p < 0.05, ** < 0.01, and *** < 0.001, respectively.

Figure A1

Lung cell numbers in I/St mice 3 weeks post infection with culture-passaged (CP) or neutrophil-passaged (NP) M. tuberculosis. Total numbers (A) and FACS analysis of main immune cell populations (B) counted per right lung.

Figure A2

Production of major regulatory cytokines by lung cells incubated for 48 h with mycobacterial antigens. ELISA assessment of lung cell supernatants. 5 mice per group, p > 0.05 for all cytokines, ANOVA.

Figure A3

Mycobacterial CFU counts in organs at the stages of infection corresponding to establishment (A,B) and advanced (C,D) stages of adaptive immune response. Results of the second experiment are displayed as mean CFU ± SD per the right lung. * p < 0.05, ANOVA.

Figure 1

Differential RNA-seq of M. tuberculosis cultured in Dubos medium (CP), engulfed by neutrophils (NP) or macrophages (MP). Heat map of M. tuberculosis RNAs differentially expressed across individual biological replicates (p < 0.05; log2FC ≥ 1.5).

Figure 3

Analysis of cell numbers in lungs of I/St mice 8 weeks post infection by culture-passaged (CP), neutrophil-passaged (NP) or macrophage-passaged (MP) M. tuberculosis. Total numbers (A) and FACS analysis of main populations (B) of cells counted per right lung. Five mice in each group were assessed individually and the results presented as mean ± SD, * p < 0.05, ** < 0.01, ANOVA.

Figure 4

Mycobacterial CFU counts in organs at the stages of infection corresponding to establishment (A,B) and advanced (C,D) stages of adaptive immune response. Mice in groups of 4–5 were infected via respiratory tract with indicated types of mycobacteria. Results of one of two similar experiments are displayed as mean CFU ± SD per the right lung. * p < 0.05, ANOVA. Results of the second experiment are displayed in Figure 3A.

Figure 5

Lung pathology in mice infected with NP mycobacteria (A) and survival time of infected mice (B–D). (A) At week 4 post-challenge, mice infected with CP and NP mycobacteria developed condensed inflammatory foci in the lungs surrounded by zones of non-inflamed, porous breathing tissue. (B) NP- and CP-infected, (C) NP- and MP-infected, (D) NP-infected neutrophil-depleted and control animals. Mice in groups of 8–10 were infected with ~80 mycobacteria from indicated stocks. Survival was monitored daily starting day 50 post-challenge and survival curves compared using log-rank (Mantel-Cox) test.

Figure 6

The approach for obtaining large amounts of ex vivo neutrophils or macrophages containing engulfed mycobacteria and avoiding contamination with bacilli that escaped phagocytosis (modification from Ref. [66] with permission). The layer of Percoll with ρ = 1.1 serves as an additional protector isolating the interphase from pellet.