Attenuation of late-stage disease in mice infected by the Mycobacterium tuberculosis mutant lacking the SigF alternate sigma factor and identification of SigF-dependent genes by microarray analysis

Abstract

The Mycobacterium tuberculosis alternate sigma factor, SigF, is expressed during stationary growth phase and under stress conditions in vitro. To better understand the function of SigF we studied the phenotype of the M. tuberculosis DeltasigF mutant in vivo during mouse infection, tested the mutant as a vaccine in rabbits, and evaluated the mutant's microarray expression profile in comparison with the wild type. In mice the growth rates of the DeltasigF mutant and wild-type strains were nearly identical during the first 8 weeks after infection. At 8 weeks, the DeltasigF mutant persisted in the lung, while the wild type continued growing through 20 weeks. Histopathological analysis showed that both wild-type and mutant strains had similar degrees of interstitial and granulomatous inflammation during the first 12 weeks of infection. However, from 12 to 20 weeks the mutant strain showed smaller and fewer lesions and less inflammation in the lungs and spleen. Intradermal vaccination of rabbits with the M. tuberculosis DeltasigF strain, followed by aerosol challenge, resulted in fewer tubercles than did intradermal M. bovis BCG vaccination. Complete genomic microarray analysis revealed that 187 genes were relatively underexpressed in the absence of SigF in early stationary phase, 277 in late stationary phase, and only 38 genes in exponential growth phase. Numerous regulatory genes and those involved in cell envelope synthesis were down-regulated in the absence of SigF; moreover, the DeltasigF mutant strain lacked neutral red staining, suggesting a reduction in the expression of envelope-associated sulfolipids. Examination of 5'-untranslated sequences among the downregulated genes revealed multiple instances of a putative SigF consensus recognition sequence: GGTTTCX(18)GGGTAT. These results indicate that in the mouse the M. tuberculosis DeltasigF mutant strain persists in the lung but at lower bacterial burdens than wild type and is attenuated by histopathologic assessment. Microarray analysis has identified SigF-dependent genes and a putative SigF consensus recognition site.

FIG. 1.

Comparison of the survival of the ΔsigF mutant strain (○) and wild-type M. tuberculosis strains (CDC1551) (□) in the mouse tuberculosis model. Following intravenous inoculation of mice with 3.6 ×10⁵ (5.56 log) mutant or 6.1 × 10⁵ (5.79 log) wild-type bacteria, groups of six mice were sacrificed at the indicated times. A virulence analysis by whole organ CFU counts was done in the lung (A) and spleen (B) at each indicated time. Error bars, standard deviations.

FIG. 2.

Microscopic histopathology of mouse lung tissues during the course of infection by wild-type M. tuberculosis (A to C) and the ΔsigF mutant strain (D to F). Low-power views are representative of lung specimens after formalin fixation, paraffin embedding, and azure eosin staining, corresponding to an infection duration of 4 weeks (A and D) and 12 weeks (B and E). Representative high-power views of lung sections are shown after acid-fast staining at 20 weeks of infection (C and F). Also shown are the unfixed representative murine spleens (G) at necropsy after infection for 20 weeks: no infection (left), infection with wild type (middle), or infection with the ΔsigF mutant strain (right). See text for details.

FIG. 3.

The M. tuberculosis ΔsigF mutant produces reduced levels of cell envelope-associated sulfolipids as assessed by neutral red staining of whole bacteria. The wild-type and complemented mutant M. tuberculosis strains retained the dye, whereas the ΔsigF mutant strain is devoid of neutral red-retaining activity.