Differential regulation of the two-component regulatory system senX3-regX3 in Mycobacterium tuberculosis

Abstract

The highly successful pathogen Mycobacterium tuberculosis (Mtb) has evolved strategies to adapt to various stress conditions, thus promoting survival within the infected host. The two-component regulatory system (2CRS) senX3-regX3, which has been implicated in the Mtb response to inorganic phosphate depletion, is believed to behave as an auto-regulatory bicistronic operon. Unlike other 2CRS, Mtb senX3-regX3 features an intergenic region (IR) containing several mycobacterium interspersed repetitive units (MIRU) of unknown function. In this study, we used a lacZ reporter system to study the promoter activity of the 5' untranslated region of senX3, and that of various numbers of MIRUs in the senX3-regX3 IR, during axenic Mtb growth in nutrient-rich broth, and upon exposure to growth-restricting conditions. Activity of the senX3 promoter was induced during phosphate depletion and nutrient starvation, and IR promoter activity under these conditions was directly proportional to the number of MIRUs present. Quantitative reverse transcriptase (qRT)-PCR analysis of exponentially growing Mtb revealed monocistronic transcription of senX3 and regX3, and, to a lesser degree, bicistronic transcription of the operon. In addition, we observed primarily monocistronic upregulation of regX3 during phosphate depletion of Mtb, which was confirmed by Northern analysis in wild-type Mtb and by RT-PCR in a senX3-disrupted mutant, while upregulation of regX3 in nutrient-starved Mtb was chiefly bicistronic. Our findings of differential regulation of senX3-regX3 highlight the potential regulatory role of MIRUs in the Mtb genome and provide insight into the regulatory mechanisms underlying Mtb adaptation to physiologically relevant conditions.

Fig. 1.

Construction of promoter-containing Mtb recombinant strains. (a) Illustration of the Mtb senX3-regX3 2CRS and the strategy utilized to generate promoter-containing recombinant strains. The senX3-regX3 IR consists of two identical 77 bp (white arrows) and a 53 bp (black arrow) MIRUs. (b) The 5′-UTR of senX3 (shaded arrow) was cloned into the promoter-less plasmid pYUB76 upstream of the reporter gene lacZ, generating pSR. A similar cloning strategy was used for the senX3-regX3 IR, yielding constructs pIR-2, pIR-1 and pIR-0.5 containing two full 77 bp MIRUs, one 77 bp MIRU and one 53 bp MIRU, respectively. (c) Nucleotide sequences are shown of the full 77 bp MIRU (white arrow) and the 53 bp MIRU (black arrow) within the senX3-regX3 IR of wild-type CDC1551. Underlined sequences are common sequences shared by MIRUs and bold type indicates putative DTGA insertion sites.

Fig. 2.

Promoter activity of the Mtb senX3 promoter and the Mtb senX3-regX3 IR, as measured by β-galactosidase assay. Fluorescence units represent β-galactosidase activity produced by LacZ, whose expression is driven by the promoters of interest, after 24 h exposure to Middlebrook 7H9 broth (black bars), P_i depletion (0 µM P_i; white bars) and nutrient starvation (NS; grey bars). C₂FDG was used as fluorescent substrate. pYUB76-containing Mtb CDC1551 showed similar background activity, regardless of condition. (a) pSR showed stronger promoter activity following Mtb exposure to NS (P<0.01) and 0 µM P_i (P<0.05) relative to exponential growth in 7H9 broth. (b) Promoter activity of the IR increased with increasing number of MIRUs during NS (P<0.01). pIR-2 showed stronger promoter activity during Mtb exposure to NS (P<0.01) and 0 µM P_i (P<0.01) than during growth in 7H9 broth. Error bars represent sd of the mean. *P<0.05; **P<0.01.

Fig. 3.

qRT-PCR reveals that bicistronic and monocistronic expression of Mtb senX3 and regX3 is condition-dependent. Total RNA was purified from wild-type CDC1551 grown in Middlebrook 7H9 broth to mid-exponential phase, and under nutrient starvation (NS) and phosphate depletion (0 µM P_i) conditions for 24 h. Abundance of transcripts of senX3 and regX3, and of senX3-regX3 co-transcripts was calculated relative to that of sigA. (a) Abundance of senX3-regX3 co-transcripts was significantly lower relative to that of senX3 (P<0.01) and regX3 (P<0.01) during growth in 7H9 broth. (b) Expression of senX3 (black bars) and regX3 (white bars) was upregulated, while that of the senX3-regX3 co-transcript (grey bars) was downregulated, during Mtb exposure to 0 µM P_i relative to 7H9 broth (P<0.01 for both comparisons). By contrast, expression of senX3, regX3 and the senX3-regX3 co-transcript was upregulated to a similar extent during Mtb exposure to NS (P>0.05 for both comparisons). The level of expression of each gene was normalized to that of the housekeeping gene sigA under each condition prior to comparison between individual stress conditions and 7H9 broth. Positive values in these graphs represent increased gene expression and negative values represent decreased expression under each stress condition relative to 7H9 broth. Samples were prepared in triplicate under each experiment condition. Error bars represent sd of the mean. ±, P>0.05; **P<0.01.

Fig. 4.

Northern analysis confirms predominantly independent expression of Mtb regX3 during P_i depletion. Biotinylated single-strand regX3 probe (247 bp) was used for hybridization. Lane 1, BrightStar biotinylated RNA millennium marker (Ambion). Lane 2, total Mtb RNA following 72 h of P_i depletion. White arrow indicates a band corresponding to the independently expressed regX3 transcript (~0.8 kb). The experiment was repeated using two biological samples with the same results.

Fig. 5.

Preserved expression of regX3 by RT-PCR in a senX3-disrupted Mtb mutant strain during P_i depletion. Total RNA from 24 h P_i-depleted cultures of a mutant strain containing a Tn insertion in senX3 (senX3 : : Tn) and the isogenic wild-type was used for cDNA synthesis with oligo(dT)₂₀ primer followed by PCR amplification. (a) Illustration of primer pairs used to amplify cDNA corresponding to the senX3 transcript (senX3-F/senX3-R) and regX3 transcript (regX3-F/regX3-R), and the senX3-regX3 co-transcript (SR-F/SR-R). The Tn insertion is at bp 162 in the senX3 gene of senX3 : : Tn (grey triangle). (b) RT-PCR results. Lanes: 1100 bp DNA marker (Fermentas); 2, senX3 expression in wild-type; 3, co-transcript expression in wild-type; 4, regX3 expression in wild-type; 5, senX3 expression in senX3 : : Tn; 6, co-transcript expression in senX3 : : Tn; 7, regX3 expression in senX3 : : Tn.