Silencing Mycobacterium smegmatis by using tetracycline repressors

Abstract

Many processes that are essential for mycobacterial growth are poorly understood. To facilitate genetic analyses of such processes in mycobacteria, we and others have developed regulated expression systems that are repressed by a tetracycline repressor (TetR) and induced with tetracyclines, permitting the construction of conditional mutants of essential genes. A disadvantage of these systems is that tetracyclines function as transcriptional inducers and have to be removed to initiate gene silencing. Recently, reverse TetR mutants were identified that require tetracyclines as co-repressors. Here, we report that one of these mutants, TetR r1.7, allows efficient repression of lacZ expression in Mycobacterium smegmatis in the presence but not the absence of anhydrotetracycline (atc). TetR and TetR r1.7 also allowed efficient silencing of the essential secA1 gene, as demonstrated by inhibition of the growth of a conditional mutant and dose-dependent depletion of the SecA1 protein after the removal or addition, respectively, of atc. The kinetics of SecA1 depletion were similar with TetR and TetR r1.7. To test whether silencing of secA1 could help identify substrates of the general secretion pathway, we analyzed the main porin of M. smegmatis, MspA. This showed that the amount of cell envelope-associated MspA decreased more than 90-fold after secA1 silencing. We thus demonstrated that TetR r1.7 allows the construction of conditional mycobacterial mutants in which the expression of an essential gene can be efficiently silenced by the addition of atc and that gene silencing permits the identification of candidate substrates of mycobacterial secretion systems.

FIG. 1.

Regulation of β-galactosidase expression in M. smegmatis by wt TetR and revTetR. (A) Schematic of the assay system. wt TetR is depicted by gray ovals and revTetR by striped ovals. The tetO operator sites that bind the repressors are shown as black boxes in the TetR-responsive promoter P_myc1tetO. β-Galactosidase was expressed from a lacZ gene located downstream of P_myc1tetO and integrated into the M. smegmatis chromosome at the mycobacteriophage L5 att site. Expression of wt TetR and TetR r1.7 was achieved using episomal plasmids containing the respective tetR genes downstream of a strong (P_smyc) or weaker (P_imyc) promoter. For wt TetR, the expression of lacZ is switched on after the addition of atc, depicted by dark gray hexagons (left side). For revTetR, the addition of atc mediates repression of lacZ transcription (right side). (B) Repression and induction with TetR, TetR r1.7, and atc. β-Galactosidase activities in relative fluorescence units (RFU) were measured in the absence (−) or presence (+) of 30 ng/ml atc as described in Materials and Methods. The light gray bars represent β-galactosidase activities in the absence of atc, whereas the dark gray bars show β-galactosidase activities after the addition of atc. Promoters P_smyc and P_imyc were used to achieve high or low expression, respectively, of TetRs. Error bars represent the standard errors of the means. (C) Dose responsiveness of β-galactosidase expression regulated by wt TetR and TetR r1.7 measured in relative fluorescence units (RFU). Regulation mediated by low-level expression of TetR is depicted by filled circles for wt and filled gray triangles for TetR r1.7. Regulation mediated by high-level expression of TetR is shown as open circles for wt and open gray triangles for TetR r1.7. atc concentrations ranging from 1 to 200 ng/ml were used to regulate the transcription of lacZ. Error bars represent the standard deviations of the means.

FIG. 2.

Steady-state protein levels of TetR and TetR r1.7 in M. smegmatis. Total lysates from M. smegmatis strains transformed with either pTE-1S0X (expressing wt TetR from the strong promoter P_smyc) or pTE-4S0X (expressing TetR r1.7 from the strong promoter P_smyc) were loaded on a 10% SDS-PAGE gel in 1:4 serially decreasing amounts for quantitative Western blot analysis using antibodies against TetR (α-TetR). Purified wt TetR and TetR r1.7 proteins were also loaded in 1:4 serially decreasing amounts on the same gel for the calibration of standard curves. Different amounts of wt TetR- and TetR r1.7-containing lysates were used to obtain signal strengths in or near the linear range comparable with that of the purified standards so that quantification could be performed. AhpC was used as the loading control. Positions of the molecular weight markers are indicated on the left.

FIG. 3.

Construction and verification of conditional secA1 mutants of M. smegmatis. (A) Schematic overview of strain construction. The suicide plasmid pKIsecA1 was electroporated into M. smegmatis mc²155 and integrated at the secA1 locus via homologous recombination, inserting the TetR-responsive promoter P_myc1tetO before secA1 and generating strain MSE10. wt TetR or TetR r1.7 was subsequently introduced into MSE10 to generate the conditional secA1 mutants MSE13 and MSE14, respectively; TetR or TetR r1.7 was also introduced into mc²155 to generate the control strains MSE11 and MSE12, respectively. 1881 and 1883 stand for MSMEG1881 and MSMEG1883, genetic loci upstream and downstream of secA1 on the genome of M. smegmatis mc², according to The Institute for Genomic Research. PCR primers 154, 155, 117, and 120 were used for verifying the genotype of MSE10. (B) Confirmation of the MSE10 genotype by PCR. Genomic DNA prepared from mc²155 or MSE10 was used in PCRs with the indicated primer pairs, and the products were analyzed on a 1% agarose gel. No PCR product was expected using primers 117 plus 120 and the genomic DNA of mc²155; the size of the expected product from primers 117 plus 120 and MSE10 was 3.4 kb. Primers 154 plus 155 were expected to give a 1.0-kb product from mc²155 and a 4.3-kb product from MSE10. The arrows indicate the locations of the products. DNA molecular weight markers are shown on the left.

FIG. 4.

Growth of secA1 conditional silencing mutants on agar plates in the absence (−) and presence (+) of atc. wt-TetR-regulated strain MSE13, TetR r1.7-regulated strain MSE14, and the respective control strains MSE11 and MSE12 were streaked onto 7H11 plates without atc or with 50 ng/ml atc, and pictures of the plates were taken after 3 days of incubation at 37°C.

FIG. 5.

atc dose dependency of growth inhibition and SecA1 depletion in conditional secA1 mutants. (A) Growth inhibition. Cultures of M. smegmatis strains were initially obtained in media allowing secA1 expression, washed, and diluted into fresh media containing different atc concentrations at the starting OD₅₈₀ of 0.02 AU/cm. The cultures were agitated at 250 rpm at 37°C for 21 h, and bacterial growth was measured by the OD₅₈₀. (B) SecA1 Western blot analysis. Total lysates of M. smegmatis were prepared from cultures harvested at 21 h from the above experiment and analyzed by 10% SDS-PAGE, followed by Western blotting using antibodies against SecA1 and SecA2 (α-SecA1 and α-SecA2). SecA2 was used as the loading control. Positions of the protein molecular weight markers are indicated on the left. (C) Quantification of the Western blot signals. The signal strengths of the SecA1 and SecA2 bands were determined by densitometry, and the loading-adjusted SecA1 expression levels were calculated by dividing the SecA1 level by the SecA2 level. In order to compare the two groups of strains that were regulated by wt TetR or TetR r1.7 on the same graph, the loading-adjusted SecA1 levels for all samples of MSE11 and MSE13 were expressed as the percentages of the level for MSE11 at 0 ng/ml atc, and the loading-adjusted SecA1 levels for all samples of MSE12 and MSE14 were expressed as the percentages of the level for MSE12 at 0 ng/ml atc. RU, relative units.

FIG. 6.

Kinetics of growth inhibition and SecA1 depletion. (A) Growth inhibition. Cultures of M. smegmatis strains were grown in media allowing secA1 expression, washed, and diluted into fresh media containing no atc (−) or 40 ng/ml atc (+). Bacterial growth was monitored in the following 48 h by measuring the OD₅₈₀. (B) SecA1 Western blot analysis. Samples of bacterial cultures in the experiment described above were harvested at the times indicated, and total lysates were prepared and analyzed by 10% SDS-PAGE, followed by Western blot analysis using antibodies against SecA1 and SecA2 (α-SecA1 and α-SecA2). SecA2 was used as the loading control. Positions of the protein molecular markers are indicated on the left. (C) Quantification of the Western blot signals, performed as described in the legend to Fig. 5C.

FIG. 7.

Impact of secA1 silencing on steady-state levels of cell wall-associated MspA. (A) MspA Western blot analysis. Cell wall-associated proteins were prepared from MSE12 and MSE14 cultures grown for 21 h with different atc concentrations as described in Materials and Methods and analyzed by 10% SDS-PAGE, followed by Western blot analysis using antibodies against MspA and AhpC (α-MspA and α-AhpC). AhpC was used as the loading control. Positions of the protein molecular markers are indicated on the left. (B) Quantification of Western blot signals. The signal strengths of the MspA and AhpC bands were determined by densitometry, and the loading-adjusted MspA levels were calculated by dividing the MspA level by the AhpC level. RU, relative units.