The genetic requirements for fast and slow growth in mycobacteria

Abstract

Mycobacterium tuberculosis infects a third of the world's population. Primary tuberculosis involving active fast bacterial replication is often followed by asymptomatic latent tuberculosis, which is characterised by slow or non-replicating bacteria. Reactivation of the latent infection involving a switch back to active bacterial replication can lead to post-primary transmissible tuberculosis. Mycobacterial mechanisms involved in slow growth or switching growth rate provide rational targets for the development of new drugs against persistent mycobacterial infection. Using chemostat culture to control growth rate, we screened a transposon mutant library by Transposon site hybridization (TraSH) selection to define the genetic requirements for slow and fast growth of Mycobacterium bovis (BCG) and for the requirements of switching growth rate. We identified 84 genes that are exclusively required for slow growth (69 hours doubling time) and 256 genes required for switching from slow to fast growth. To validate these findings we performed experiments using individual M. tuberculosis and M. bovis BCG knock out mutants. We have demonstrated that growth rate control is a carefully orchestrated process which requires a distinct set of genes encoding several virulence determinants, gene regulators, and metabolic enzymes. The mce1 locus appears to be a component of the switch to slow growth rate, which is consistent with the proposed role in virulence of M. tuberculosis. These results suggest novel perspectives for unravelling the mechanisms involved in the switch between acute and persistent TB infections and provide a means to study aspects of this important phenomenon in vitro.

Figure 1. Growth of wild type and glpK mutant strains.

Colonial morphology of 3 week grown (i) wild type M. bovis BCG (ii) a glpK deletion mutant and (iii) glpK mutant carrying a copy of the glpK gene in the attB site on the chromosome on (A) 7H11 media and (B) Roisin's glycerol minimal media. The glpK mutant exhibited dysgonic growth on 7H11 medium and was unable to grow on Roisin's minimal media. The wild type phenotype was restored in the complemented strain.

Figure 2. TraSH screen for mutants with reduced fitness at different growth rates in a carbon limited chemostat.

A transposon mutant library of Mycobacterium bovis BCG was inoculated into the chemostat and after an initial phase in batch culture was grown at a doubling time of 23 h (D = 0.03 h⁻¹) and samples removed for analysis by transposon site hybridisation (TraSH). The transposon mutant library was similarly inoculated into the chemostat at a growth rate corresponding to a doubling time of 69 h (D = 0.01 h⁻¹) and samples again removed for TraSH analysis. Finally, cultures established at the fast growth rate were switched to the slow growth rate, and visa versa, to identify mutants unable to switch between the different growth rates.

Figure 3. Mutants with reduced fitness at (A) fast (t_d = 23 h) and (B) slow (t_d = 69 h) growth rate in a carbon limited chemostat.

Venn diagram shows the overlapping genes.

Figure 4. Competitive chemostat experiments: Wild type Vs Δhspr.

Average CFU ratios of (i) wild type H37Rv+ΔhspR (closed circles) and wild type BCG+ΔhspR (closed triangles) during continuous culture at a dilution rate of 0.01 h⁻¹ (t_d = 69 h) were plotted as a function of time. Data were normalized so that the ratio at the start of chemostat culture equaled 1. Only the BCG ΔhspR strain was attenuated for slow growth rate in the chemostat.

Figure 5. Competitive chemostat experiments: Wild type M. tuberculosis Vs Δmce1 operon.

(A) Average CFU ratios of wild type H37Rv and Δmce1 operon during continuous culture at a dilution rate of 0.01 h⁻¹ (t_d = 69 h) were plotted as a function of time. Data were normalized so that the ratio at the start of chemostat culture equaled 1. The mce1 operon mutant had a competitive advantage over the wild type strain at slow growth rate. The graph is representative of two independent experiments. (B) At day 53 the dilution rate was altered to 0.03 h⁻¹ (t_d = 23 h). At fast growth the Δmce1 did not have a competitive advantage over the wild type strain and the ratio returned to 2 after only 11 days at a dilution rate of 0.03 h⁻¹ (t_d = 23 h).