Transcriptional response to the host cell environment of a multidrug-resistant Mycobacterium tuberculosis clonal outbreak Beijing strain reveals its pathogenic features

Abstract

Tuberculosis is a global public health problem with emergence of multidrug-resistant infections. Previous epidemiological studies of tuberculosis in Thailand have identified a clonal outbreak multidrug-resistant strain of Mycobacterium tuberculosis in the Kanchanaburi province, designated "MKR superspreader", and this particular strain later was found to also spread to other regions. In this study, we elucidated its biology through RNA-Seq analyses and identified a set of genes involved in cholesterol degradation to be up-regulated in the MKR during the macrophage cell infection, but not in the H37Rv reference strain. We also found that the bacterium up-regulated genes associated with the ESX-1 secretion system during its intracellular growth phase, while the H37Rv did not. All results were confirmed by qRT-PCR. Moreover, we showed that compounds previously shown to inhibit the mycobacterial ESX-1 secretion system and cholesterol utilisation, and FDA-approved drugs known to interfere with the host cholesterol transportation were able to decrease the intracellular survival of the MKR when compared to the untreated control, while not that of the H37Rv. Altogether, our findings suggested that such pathways are important for the MKR's intracellular growth, and potentially could be targets for the discovery of new drugs against this emerging multidrug-resistant strain of M. tuberculosis.

Figure 1

Principle component analysis. Samples from the same experimental conditions tend to cluster together according to their gene expression profiles. The first principle component (PC1) appears to capture the differences of the gene expression profiles across the three experimental conditions (log-phase, T = 0 and T = 4), and the second component appears to capture the differences of the gene expression profiles among H37RV and MKR superspreader.

Figure 2

Differential gene expression analyses. (a) differentially expressed genes were determined among the three biological conditions (baseline VS experimental group: log-phase VS T = 0, log-phase VS T = 4, and T = 0 VS T = 4). The numbers of significantly up-regulated genes are shown on the top of the arrows while the numbers of significantly down-regulated genes are shown on the bottom. (b) Volcano plots (red: down-regulated genes; and blue: up-regulated genes). (c) Venn diagrams illustrating the numbers of differentially expressed genes common and unique among different strains of M. tuberculosis.

Figure 3

Gene functional classification analyses. (a) up-regulated genes were annotated with various functional terms, and gene functional classification analyses were performed by using DAVID Bioinformatics Resources 6.8 under the default settings. Baseline VS experimental group: log-phase VS T = 0 (left), log-phase VS T = 4 (middle), and T = 0 VS T = 4 (right). (b) down-regulated genes were annotated with functional terms and gene functional classification analyses were performed as in (a).

Figure 4

Expression levels of genes in the cholesterol degradation pathway. (a–i) THP-1 macrophages were infected with the M. tuberculosis reference strain H37Rv (H) or the MKR superspreader (M) for 4 h. The expression levels of various genes in the cholesterol degradation pathway were quantified by qRT-PCR. 2^-∆∆ct is used for normalisation and relative quantification. One-way ANOVA with Tukey's multiple comparison test was used to determine which pairs of expression levels were significantly different (*: p < 0.05; **: p < 0.01; and ***: p < 0.001).

Figure 5

Intracellular survival odds of the MKR superspreader and H37Rv upon the treatments of V-59, aripiprazole, manidipine and ethoxzolamide. (a–d) THP-1 cells were infected with mCherry-expressing MKR superspreader or H37Rv for 30 min, and were treated with media containing various concentrations of (a) V-59, (b) aripiprazole, (c) manidipine and (d) ethoxzolamide for 3 days. Cells were then fixed and processed for high-content image analysis to quantify the number of mycobacteria per host cell. Percent mycobacterial survival at each concentration was calculated relative to that of the non-treated control group set to 100%. Two-way ANOVA with Bonferroni post-tests were used to determine if the intracellular survival levels were significantly different from that of the control group (*: p < 0.05; **: p < 0.01; and ***: p < 0.001).

Figure 6

Expression levels of genes involved in the ESX-1 secretion system: espC, espD, esxB, and esxN. (a–d) THP-1 macrophages were infected with the M. tuberculosis reference strain H37Rv (H) or the MKR superspreader (M) for 4 h. The expression levels of various genes involved in the ESX-1 secretion system were quantified by qRT-PCR. 2^-∆∆ct is used for normalisation and relative quantification. One-way ANOVA with Tukey's multiple comparison test was used to determine which pairs of expression levels were significantly different (*: p < 0.05; **: p < 0.01; and ***: p < 0.001).

Figure 7

Intracellular survival odds of the MKR, H37Rv, and different clinical isolates upon the treatments of V-59, aripiprazole, manidipine and ethoxzolamide. (a–d) THP-1 cells were infected with Alexa-568-labelled MKR, H37Rv, or different clinical isolates for 30 min, and were treated with media containing various concentrations of (a) V-59, (b) aripiprazole, (c) manidipine and (d) ethoxzolamide for 3 days. Cells were then fixed and processed for high-content image analysis to quantify the number of mycobacteria per host cell. Percent mycobacterial survival at each concentration was calculated relative to that of the non-treated control group set to 100%. Two-way ANOVA with Bonferroni post-tests were used to determine if the intracellular survival levels were significantly different from that of the control group (*: p < 0.05; **: p < 0.01; and ***: p < 0.001).