Individual RD1-region genes are required for export of ESAT-6/CFP-10 and for virulence of Mycobacterium tuberculosis

Abstract

The RD1 genomic region is present in virulent strains of Mycobacterium tuberculosis (MTB), missing from the vaccine strain M. bovis BCG, and its importance to virulence has been established experimentally. Based on in silico analysis, it has been suggested that RD1 may encode a novel secretion system, but the mechanism by which this region affects virulence is unknown. Here we examined mutants disrupted in five individual RD1 genes. Both in vitro and in vivo, each mutant displayed an attenuated phenotype very similar to a mutant missing the entire RD1 region. Genetic complementation of individual genes restored virulence. Attenuated mutants could multiply within THP-1 cells, but they were unable to spread to uninfected macrophages. We also examined export of two immunodominant RD1 proteins, CFP-10 and ESAT-6. Export of these proteins was greatly reduced or abolished in each attenuated mutant. Again, genetic complementation restored a wild-type phenotype. Our results indicate that RD1 genes work together to form a single virulence determinant, and argue that RD1 encodes a novel specialized secretion system that is required for pathogenesis of MTB.

Fig. 1

Bacterial growth (A) and cytotoxicity (B) by individual mutants in the human macrophage-like THP-1 cell line. Bacteria were quantified by luciferase assay or by plating for colony forming units. Macrophage metabolism was measured by a redox dye, and a decrease in metabolism is taken as an indication of cytotoxicity. Means and standard deviation (SD) for each time-point are from one infection done in triplicate, representative of (A) two or (B) three infections.

Fig. 2

Complementation of mutants in THP-1 cells. Cells were infected with control strains H37Rv (X), H37Rv:ΔRD1 (+), and complements (A) H37Rv:tn3870::KG18 (□) and H37Rv:tn3871::KG18 (◇) or (B) H37Rv:tn3874::MH408 (○), H37Rv: Δ3875::MH406 (△) and H37Rv:tn3876::MH429 ([unk]). Data for each time point are means and SD from one infection done in triplicate, representative of two infections.

Fig. 3

Enumeration of bacilli within THP-1 cells. Infected cells were stained with TB Quick and bacilli visualized by microscopy. Data are from one infection, representative of two. ***P < 0.0001.

Fig. 4

Bacterial growth in lungs after aerosol infection of 6–8 week-old C57BL/6 mice. Data at each time-point are the mean and SD of five mice per strain from one infection, representative of two.

Fig. 5

Lung histopathology of infected C57/BL6 mice at 4 weeks post infection. Mice were infected with (A) H37Rv (B) H37Rv:ΔRD1 (C) H37Rv:tn3870 (D) H37Rv:Δ3875 (E) H37Rv:tn3876.

Fig. 6

Detection of ESAT-6 in culture filtrates by Western immunoblot. Samples are (1) H37Rv (2) H37Rv: ΔRD1 (3) H37Rv:tn3870 (4) H37Rv:tn3870::KG18 (5) H37Rv:tn3874 (6) H37Rv:tn3874::MH408 (7) H37Rv:Δ3875 (8) H37Rv: Δ3875::MH406 (9) purified ESAT-6 protein.

Fig. 7

IFN-γ ELISPOT assay to determine the presence of CFP-10 protein. A. Standard curve with purified CFP-10 protein. Fewer than 10 spots were detected in the absence of antigen. Using recombinant protein, maximum achievable spot number was 234 ± 13. At a CFP-10 concentration of 2 ng ml⁻¹, 27 ± 4 spots were detected indicating a minimum sensitivity for the assay.B. ELISPOT results from cultured filtrate and cell pellet samples. Data are mean and standard error of duplicate determinations, representative of two independent experiments.