A Nonribosomal Peptide Synthase Gene Driving Virulence in Mycobacterium tuberculosis

Abstract

Nonribosomal peptide synthases produce short peptides in a manner that is distinct from classical mRNA-dependent ribosome-mediated translation. The Mycobacterium tuberculosis genome harbors a nonribosomal peptide synthase gene, nrp, which is part of a gene cluster proposed to be involved in the biosynthesis of isonitrile lipopeptides. Orthologous clusters are found in other slow-growing pathogenic mycobacteria and actinomycetes. To probe the role of the nrp gene in infection, we generated an nrp deletion mutant in M. tuberculosis H37Rv and tested its virulence in immunocompetent (C57BL/6) mice. The nrp mutant strain displayed lower initial growth rates in the lungs and a defective dissemination to the spleens of infected mice. Mice infected with the mutant strain also survived for twice as long as those infected with wild-type M. tuberculosis and, remarkably, showed subdued pathology, despite similar bacterial loads at later stages of infection. The differences in the course of infection between wild-type and nrp mutant strains were accompanied by distinct dynamics of the immune response. Most strikingly, the nrp mutant was highly attenuated in immunodeficient (SCID-, recombination activating 2 [RAG2]-, and gamma interferon [IFN-γ]-deficient) mice, suggesting that macrophages control the nrp mutant more efficiently than they control the wild-type strain. However, in the presence of IFN-γ, both strains were equally controlled. We propose that the nrp gene and its associated cluster are drivers of virulence during the early stages of infection.IMPORTANCE Over 10 million people developed tuberculosis (TB) in 2016, and over 1.8 million individuals succumbed to the disease. These numbers make TB the ninth leading cause of death worldwide and the leading cause from a single infectious agent. Therefore, finding novel therapeutic targets in Mycobacterium tuberculosis, the pathogen that causes most cases of human TB, is critical. In this study, we reveal a novel virulence factor in M. tuberculosis, the nrp gene. The lack of nrp highly attenuates the course of M. tuberculosis infection in the mouse model, which is particularly relevant in immune-deficient hosts. This is very relevant as TB is particularly incident in immune-suppressed individuals, such as HIV patients.

Keywords: immune deficiency; pathogenesis; tuberculosis; virulence factors.

FIG 1

The nrp cluster is highly conserved among the MTBC and slow-growing species of the genus Mycobacterium. (A) Sequence identity scores of the nrp cluster were obtained by querying the M. tuberculosis H37Rv sequences Rv0097-Rv0101 against a local database of 72 genomes of species representative of the phylogeny of the genus Mycobacterium (19). The genomics sequences include complete genomes and whole-genome shotgun sequences. The highest percentage (dark green) corresponds to the longest and most elevated sequence identities. (B) Synteny of the genome sections harboring the nrp cluster in M. tuberculosis, M. bovis, M. canettii, and M. leprae. The green shading indicate homologous DNA regions shared by the four chromosomes without major sequence rearrangements. The genes of the nrp cluster are highlighted in yellow. (C) Nucleotide diversity (π) level of the M. tuberculosis genes Rv0097, fcoT, fad10, Rv0100, and nrp and the complete genome across a set of 220 genomes, including strains from lineage 1 to 7 of the MTBC.

FIG 2

Loss of the nrp gene results in attenuation of M. tuberculosis H37Rv in mice. C57BL/6 mice were infected by aerosol exposure to H37Rv (WT), Δnrp, or nrp-comp strains. At the indicated time points, the lungs (A and D) and spleens (B) of infected mice were collected and the bacterial burden determined by CFU enumeration. (C) The weights of the animals were monitored to determine survival curves. On days 20, 30, and 90 postinfection, lung pathology (E and F) was determined by hematoxylin and eosin (H&E) staining and morphometric analysis of the right upper lobes of infected lungs. Data are shown as the means ± standard errors of the means (SEMs) from 5 independent animals in at least 2 independent experiments. The pictures in panel F are of one animal representative of the experimental group; ×40 magnification. Statistical analysis was performed with a two-way ANOVA using Sidak’s test for multiple comparisons (A, B, D, and E) or with log-rank (Mantel-Cox) test for the Kaplan-Meier curves (C). * and + refer to statistical differences between Δnrp and nrp-comp strains or Δnrp and WT strains, respectively. * or +, P < 0.05; **, P < 0.01; ***, P < 0.001; **** or ++++, P < 0.0001.

FIG 3

Lack of nrp alters the dynamics of the lung immune response to infection. (A to H) At the indicated time points postinfection, the lungs of C57BL/6 mice infected via aerosol exposure to Δnrp or nrp-comp strains were harvested, a cellular suspension was prepared, and the indicated immune cell populations were determined by flow cytometry. The gating strategy is shown in Fig. S2 in the supplemental material. (I) On day 20 postinfection, the expression of CCL2, CCL7, IFN-γ, and TNF was determined by real-time PCR, as described in Materials and Methods. Data are shown as the means ± SEMs from 5 independent animals in at least 2 independent experiments. The initial bacterial burdens were log₁₀ 1.86 ± log₁₀ 0.04460 and log₁₀ 1.931 ± log₁₀ 0.0822 for Δnrp and nrp-comp strains, respectively. Statistical analysis was performed with a two-way ANOVA using Sidak’s test for multiple comparisons. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001.

FIG 4

Growth of Δnrp mutant is attenuated in immune-deficient mice. C57BL/6 RAG2^−/− (A, C, and E) and IFN-γ^−/− (B, D, and F) mice were infected by aerosol exposure to a low dose of the Δnrp or nrp-comp strain. (A and B) The weights of the animals were monitored weekly up to day 30 and every 2 days thereafter to determine the survival curves. At the indicated time points (C) or at the time of death (D), the lungs of infected mice were collected and the bacterial burden was determined by CFU enumeration. The lung pathology was determined on day 40 postinfection for RAG2^−/− (E) or at the time of death for IFN-γ^−/− (F) animals. Shown are pictures from H&E staining for one animal representative of the experimental group; ×40 magnification. The initial bacterial burdens were log₁₀ 1.92 ± log₁₀ 0.069 and log₁₀ 1.701 ± log₁₀ 0.0707 (RAG2^−/− experiment) and log₁₀ 1.92 ± log₁₀ 0.069 and log₁₀ 1.748 ± log₁₀ 0.0707 (IFN-γ^−/− experiment) for Δnrp and nrp-comp strains, respectively. BMDM (G and H) or pMac (G) were infected with Δnrp or nrp-comp strains at an MOI of 1 in the absence (G) or presence (H) of exogenous IFN-γ. On day 4 postinfection, the intracellular bacterial load was determined by CFU enumeration. Data are shown as the means ± SEMs from 8 independent animals (A to F) or from 6 replicate wells in 2 independent experiments (G and H). Statistical analysis was performed with log-rank (Mantel-Cox) tests for the Kaplan-Meier curves (A and B) or two-way ANOVAs using Sidak’s tests for multiple comparisons (C to H). *, P < 0.05; ***, P < 0.001; ****, P < 0.0001.