A GntR family transcription factor positively regulates mycobacterial isoniazid resistance by controlling the expression of a putative permease

Abstract

Background: Bacteria use transcriptional regulation to respond to environmental stresses. Specifically, exposure to antibacterial drugs is deemed to be an atypical stress, and altering transcriptional regulation in response to such stress can increase bacterial drug resistance. However, only a few transcription factors that regulate drug resistance have been reported.

Results: In the present study, a GntR family transcription factor, encoded by the MSMEG_0535 (Ms0535) gene, was shown to be an isoniazid (INH) resistance regulator in Mycobacterium smegmatis. When the Ms0535 gene was overexpressed, cells showed a significant increase in INH resistance. First, the interaction between Ms0535 and its own promoter was determined, and a conserved 26-bp palindromic DNA binding motif was identified using electrophoretic mobility shift and DNaseI footprinting assays. Second, quantitative reverse transcription-PCR assays showed that Ms0535 acted as a transcriptional activator, and positively regulated its own expression, as well as that of a permease encoded by the MSMEG_0534 (Ms0534) gene. Similar to the case for the Ms0535 gene, a recombinant Ms0534-overexpressing strain also exhibited increased INH resistance compared with the wild-type strain. Furthermore, we showed that Ms0535 and Ms0534 deletion strains were more sensitive to INH than the wild-type strain. Interestingly, overexpressing Ms0534 in the Ms0535 deletion strain enhanced its INH resistance. In contrast, the Ms0534 deletion strain was still sensitive to INH even when Ms0535 was overexpressed. These findings suggest that Ms0534 is an effector protein that affects INH resistance in M. smegmatis.

Conclusions: In summary, the GntR transcriptional regulator Ms0535 positively regulates INH resistance by transcriptionally regulating the expression of the Ms0534 permease in M. smegmatis. These results improve our understanding of the role of transcriptional regulation in INH drug resistance in mycobacteria.

Fig. 1

The effect of Ms0535 on the drug resistance of M. smegmatis and a sequence alignment of the Ms0535 C-terminal domain. (a) Isoniazid (INH)-resistance of the M. smegmatis (Msm) strain overexpressing Ms0535. Msm/pMV261 and Msm/pMV261-Ms0535 were grown on medium containing 30 μg/mL kanamycin for 36 h. Then, different concentrations of freshly grown bacteria were streaked onto 7H10 plates containing kanamycin (30 μg/mL) and INH (20 μg/mL). The control plate did not contain INH. (b) Analysis of the structural characteristics of Ms0535. The Ms0535 amino acid sequence was analyzed using the online tool SWISS-MODEL. The N-terminal region of Ms0535 contains a winged helix-turn-helix GntR DNA-binding domain, while its C-terminus contains a domain that is similar to that of the FCD family. Amino acid residues present in α-helices are highlighted in gray, and spaces in the consensus sequence denote insertions within the alignment

Fig. 2

Ms0535 specifically binds to its own promoter. (a) Bacterial one-hybrid assays. Promoters of the Ms0535 and Ms0540 genes were cloned into the pBXcmT vector, and the Ms0535 gene was cloned into the pTRG vector. A pair of pBXcmT/pTRG plasmids was co-transformed into the reporter strain, and then its growth was tested together with the self-activation controls on a selective medium. Co-transformants containing the pBX-Rv2031/pTRG-Rv3133 plasmids [26] served as positive controls (CK+), and co-transformants containing the empty vectors pBXcmT and pTRG served as negative controls (CK−). (b) Electrophoretic mobility shift assays (EMSAs). The Ms0535p (lanes 1–4) and Ms0540p (lanes 5–8) DNA substrates were co-incubated with various amounts of the Ms0535 protein. The free DNA substrate and DNA-protein complexes are indicated. (c) EMSAs for the specific binding of Ms0535 to its own promoter. Then, 0.05 nM of fluorescein isothiocyanate-labeled Ms0535 promoter DNA substrate was co-incubated with the Ms0535 protein in the absence (lanes 1–5) or presence of non-labeled Ms0535p (0.25-1 nM) (lanes 6–8) or non-labeled Ms0540p (0.25-1 nM) (lanes 9–11). Unlabeled Ms0535 and unlabeled Ms0540 promoter DNA substrates were used to compete with the labeled Ms0535 promoter DNA. The Ms0535 promoter, but not the Ms0540 promoter, inhibited the binding of Ms0535 to the labeled Ms0535 promoter DNA substrate

Fig. 3

DNA-binding motif assays for Ms0535. (a) Dye primer sequencing based on a DNase I footprinting assay. The Ms0535 promoter DNA was digested with DNaseI in the presence of increasing amounts of Ms0535. The protected regions are indicated by a black frame. (b) Sequence and structural characteristics of the protected Ms0535 promoter region. The regions protected by Ms0535 are underlined. The binding motif is a 26-bp sequence containing invert repeats (IRs) with a 4-bp spacer. The translational start codon of Ms0535 is indicated in bold. (c) Electrophoretic mobility shift assays of the DNA-binding activity of Ms0535 towards DNA substrates with (lanes 1–4) or without (lanes 13–16) the IR sequences. The DNA substrates were individually incubated with 0–4 μM of the Ms0535 protein. (d) Analysis of the domain structure of Ms0535 and the genomic location of its DNA binding motif. The Ms0535 gene encodes a typical GntR regulator, and it shares a common upstream promoter region with the major facilitator superfamily permease Ms0534. The distance between the motif and the coding sequence of Ms0535 is 18 bp

Fig. 4

Assays for Ms0534-Ms0535 co-transcription by reverse transcription-PCR. (a) The schematic operon structure of Ms0534-Ms0535. Primers designed for the assays are indicated by black arrows; F1 primers are complementary to the sequences of the two adjacent genes. (b) Reverse transcription-PCR assays for Ms0534-Ms0535 co-transcription. In this assay, the resulting cDNA was transcribed by the F3 reverse primer. For the F1 fragment, mRNA (DNA-free) was used as a negative control (lane 1). In lane 2, a 189-bp reverse transcription-PCR product indicates the co-transcription of Ms0534-Ms0535. The 146-bp F2 and 100-bp F3 extension products are shown in lanes 2 and 3. The negative result in lane 5 reveals that the F4 fragment cannot be amplified from the cDNA template. The PCR procedure was as follows: the reactions underwent 35 cycles of denaturation at 95 °C for 30 s, annealing at 60 °C for 30 s, and extension at 72 °C for 30 s

Fig. 5

Quantitative real time-PCR assays. Gene expressions were determined in the M. smegmatis Ms0535 deletion strain (a), in the Ms0535-overexpressing strain (b), and in wild type strain with or without isoniazid (c). The relative expression levels of the genes were normalized using the sigA gene as an invariant transcript. An unrelated promoter of the Ms0540 gene was used as a negative control. Data were analyzed using the 2^−ΔΔCt method as described in the Methods. P-values of the relative expression data were calculated by an unpaired two-tailed Student’s t-test

Fig. 6

Determinations of growth curves of M. smegmatis strains. Mycobacterial strains were grown in 7H9 medium in the presence of 10 μg/ml isoniazid, and growth curves were determined. (a) Left: the wild-type (WT/pMV261) and Ms0535-overexpressing (WT/pMV261-Ms0535) strains; Right: the WT (WT/pMV261) and Ms0534-overexpressing (WT/pMV261-Ms0534) strains. To avoid potential side effects, the empty pMV261vector was included in the WT strain. (b) Left: the WT (WT/pMindD), Ms0535 deletion (ΔMs0535/pMindD), and ΔMs0535 complemented (ΔMs0535/pMindD-Ms0535) strains; Right: the WT (WT/pMindD), Ms0534 deletion (ΔMs0534/pMindD), and ΔMs0534 complemented (ΔMs0534/pMindD-Ms0534) strains. To avoid potential side effects, the empty pMindD vector was included in the WT and Ms0534/Ms0535 deletion strains. (c) Left: the Ms0535 deletion (ΔMs0535/pMV261) strain and the ΔMs0535 strain complemented with Ms0534 (ΔMs0535/pMV261-Ms0534); Right: the Ms0534 deletion strain (ΔMs0534/pMV261) and the ΔMs0534 strain complemented with Ms0535 (ΔMs0534/pMV261-Ms0535). To avoid potential side effects, the empty pMV261 vector was included in the Ms0534/Ms0535 deleted strains. Error bars represent the standard deviation of three biological replicates