Mycobacterium abscessus Tetracycline-Modifying Monooxygenase MAB_1496c Appears Not to Be Sufficient to Cause Resistance to Tetracycline When Expressed in Mycobacterium smegmatis.

LETTER

Recently, a gene in Mycobacterium abscessus (MAB_1496c) coding for a FAD-binding monooxygenase protein was described (1, 2). It was shown that the protein bears resemblance to tetracycline destructases (TetX) from other bacteria (3), that deletion of the gene rendered M. abscessus substantially more sensitive to tetracycline, and that this resistance was unrelated to the whiB7 pathway (4, 5). The protein was also shown to modify tetracycline in vitro by monooxygenation and to be inhibited by a low concentration of the tetracycline analog anhydrotetracycline (ATc) (1, 6). These findings suggested that the protein coded by MAB_1496c can induce tetracycline resistance in those mycobacteria that are naturally more sensitive to tetracycline than M. abscessus, such as M. smegmatis and M. tuberculosis (7, 8). This could be used for research and biotechnology purposes, such as in vitro positive selection (9, 10).

We therefore opted to examine whether the recombinant expression of MAB_1496c (MabTetX protein) in M. smegmatis would substantially increase the bacterium’s MIC to tetracycline, to the point where it could be used for in vitro selection. We cloned MAB_1496c into the blunt HpaI site in pDB32—a multicopy, kanamycin-selected, episomal vector (11). The gene was PCR amplified from M. abscessus genomic DNA (gDNA) using the following primers: forward, 5′-ACAGTGGTGATCGCCGGGGCCGGCC-3′; reverse, 5′-CATCTAGACAACACGGGCGAGATA-3′. Cloning into the HpaI site of this vector places the gene in frame with a hemagglutinin (HA) tag at the N terminus and under a constitutive mycobacterial optimized promoter (MOP). As the native promoter of MAB_1496c is regulated by MAB_1497c (5), a tetR-like transcription regulator (and is induced by tetracycline), placing the recombinant construct under a constitutive promoter avoids problems resulting from regulation of expression. The genetic structure of the construct is shown in Fig. 1A for clarity. Correct in-frame cloning with the HA tag was confirmed by Sanger sequencing, and the plasmid (pDB451) was electroporated into wild-type (WT) M. smegmatis MC2 155 to produce mDB335. The expression of the full-length MabTetX protein was verified by Western blotting using anti-HA antibodies (Abcam), showing a protein at the expected size of 53.2 kDa (Fig. 1B). However, when we tested the TetX-expressing mutants for tetracycline sensitivity, we found no difference between their MICs and that of WT M. smegmatis with an empty vector on either 7H10 agar plates (Fig. 1Ci) or in 7H9/glycerol broth (Fig. 1Cii). The MIC also remained well below that of WT M. abscessus (Fig. 1D). Of note, when we examined the MIC to tetracycline in the E. coli bacteria used for the cloning (DH5-α), we again found bacteria with the plasmids to have the same MIC to tetracycline as “WT” E. coli bacteria, with the empty control vectors (data not shown).

FIG 1

(A) Genetic sequences of the MOP promoter (green) and untranslated region (UTR) (yellow) driving the expression of an HA-tagged (blue) TetX protein (red). (B) Western blot demonstrating the full-length (53.2-kDa) HA-TetX protein in mDB335 in comparison to WT bacteria. (Ci) Approximately 250 CFU of M. smegmatisempty kana vector (top) or mDB335 (bottom) was plated onto 7H10 agar plates with increasing concentrations of tetracycline. (Cii) WT M. smegmatis and M. smegmatis mDB335 were grown in 7H9/glycerol broth with the designated concentrations of tetracycline. (D) M. abscessus ATCC 19977 was plated onto 7H10 agar plates with the designated concentrations of tetracycline.

To summarize, despite the convincing results of the carefully performed experiments reported in the original description (1), it appears that the full picture is more complex, and expression of the MAB_1496c protein is, by itself, not sufficient to induce tetracycline resistance in M. smegmatis. This could be due to a requirement for another, yet unidentified cofactor, low activity of the protein in the M. smegmatis context, retained antimicrobial activity of the oxygenated tetracycline product against M. smegmatis, or misfolding of the protein in M. smegmatis, resulting in a non- or poorly functional, albeit full-size, protein.