Global analysis of mRNA stability in Mycobacterium tuberculosis

Abstract

Mycobacterium tuberculosis (MTB) is a highly successful pathogen that infects over a billion people. As with most organisms, MTB adapts to stress by modifying its transcriptional profile. Remodeling of the transcriptome requires both altering the transcription rate and clearing away the existing mRNA through degradation, a process that can be directly regulated in response to stress. To understand better how MTB adapts to the harsh environs of the human host, we performed a global survey of the decay rates of MTB mRNA transcripts. Decay rates were measured for 2139 of the ~4000 MTB genes, which displayed an average half-life of 9.5 min. This is nearly twice the average mRNA half-life of other prokaryotic organisms where these measurements have been made. The transcriptome was further stabilized in response to lowered temperature and hypoxic stress. The generally stable transcriptome described here, and the additional stabilization in response to physiologically relevant stresses, has far-ranging implications for how this pathogen is able to adapt in its human host.

Figure 1.

Histogram of transcript half-lives in log phase MTB. The distribution of MTB mRNA half-lives follows a normal distribution (dark gray), with a mean half-life of 9.5 min. This is substantially longer than that previously measured for E. coli [light gray, data from (2), shown with permission of the authors]. The fast growing mycobacterium M. smegmatis had a decay rate similar to previously described bacteria (black). No decay rate was measured for genes that were not expressed at least 4-fold above background or did not follow a logarithmic decay.

Figure 2.

Impact of transcript attributes on mRNA half-life. (A) Half-lives of the 5′ and 3′ genes from operons show generally similar mRNA stabilities, with no trend toward either end of the transcript. (B) Histogram of the differences between first and last half-lives in minutes. (C) Graph showing the poor correlation between mRNA stability and the length of single gene transcripts. (D) Graph showing the poor correlation between mRNA stability and the %GC of mRNAs.

Figure 3.

Transcript abundance and transcript decay. (A) Graph relating the abundance of a transcript at the time of transcription inhibition with mRNA half-life shows a very strong negative correlation, which follows a power regression line (R² > 0.8). (B) Transcript abundance versus mRNA half-life for genes of the DosR regulon either before (open squares) or after (filled diamonds) ectopic induction.

Figure 4.

mRNA degradation during stress conditions. mRNA decay curves for MTB in aerobic log phase (squares), hypoxia (triangles) or room temperature (20°C, diamonds) measured by microarray. Shown are averages over all genes above background in four replicate experiments.