Susceptibility testing.

MICs were determined by the usual twofold agar dilution technique with MHA with an inoculum size of 10⁴ cells. All antimicrobial agents used in this study were obtained from commercial sources.

Molecular biology techniques.

Chromosomal DNA and plasmids were isolated using a DNeasy tissue kit and QIAfilter plasmid kit (Qiagen K.K., Tokyo, Japan), respectively. PCRs were performed with a Perkin-Elmer 480 thermal cycler using PfuTurbo DNA polymerase (Stratagene, La Jolla, Calif.). The thermal cycle profile for amplification of the mexX-mexY-flanking regions was 1 min at 96°C, 1 min at 59°C, 1 min at 72°C, and 30 cycles. The thermal cycle profile for amplification of the mexX-mexY region was 1 min at 96°C, 1 min at 68°C, 10 min at 72°C, and 30 cycles. Restriction endonucleases, alkaline phosphatase, and the DNA ligation kit were obtained from Takara Shuzo Co., Ltd., Kyoto, Japan. Restriction fragments were isolated, as required, from agarose gels using TaKaRa RECOCHIP (Takara). All molecular biology techniques were carried out according to the manufacturer's instructions or as described by Sambrook et al. (25). Transformation of E. coli (6) and P. aeruginosa (9) with plasmid DNA was performed as described previously.

Deletion of mexXY by gene replacement.

To construct a series of isogenic mutants lacking the mexX-mexY region, PCR primers for amplification of the mexX-mexY region and its flanking regions were synthesized based on nucleotide sequences of the Pseudomonas genome sequencing project database (Fig. ​(Fig.1A).1A). After amplifying a 1.2-kb region downstream of mexY on PAO1 chromosomal DNA as a template using GH3 (5′-TGTACTAGTTGATGCCCCTAGCGAAACTCTC-3′) and GH4 (5′-TTTAAGCTTGACCTACAGGACGCTGCTG-3′), a primer pair containing a newly added cutting site (underlined) for restriction nucleases, the region was ligated into the SpeI-HindIII site in a multicloning site of pMT5059 (27) to yield pNS001. Next, a 1.0-kb region upstream of the mexX gene was amplified by PCR using the primer pair GH1 (5′-TGTACGCGTATTCGGAACAAGGCGTCTGC-3′) and GH2 (5′-TTCTGCTAGCGATGTGCATGGGTGTCCCTC-3′), and this region was ligated into the MluI-NheI site of pNS001 to yield pNS002. (A 24-bp sequence derived from a multicloning site of pMT5059 was still left between the two DNA fragments inserted on pNS002.) After addition of a NotI-flanked Mob cassette from pMT5071 (28), the resulting plasmid pNS003 was mobilized from an E. coli strain S17-1 (26) to the P. aeruginosa strains to introduce deletion of the mexX-mexY region into the recipient chromosomes by allelic exchange (Fig. ​(Fig.1B).1B). The cell mixture was mated on MHA at 37°C for 3 h and then suspended in saline. Aliquots of the suspensions were plated on minimal agar plates supplemented with 30 to 300 μg of chloramphenicol per ml and incubated at 30°C for 2 days to isolate chloramphenicol-resistant pNS003 integrants. The obtained transconjugants were subsequently grown on drug-free MHA overnight for a second allelic exchange and then plated on MHA supplemented with 10% (wt/vol) sucrose. Chloramphenicol-susceptible and sucrose-resistant clones were selected and subjected to PCR analysis using the primer pair GH5 (5′-TCGCACTTGAGGTAGAGGATCTCCAGCACC-3′) and GH6 (5′-TCCTCACCGATCTGTCGAGCCTCTACTACG-3′). The sizes of the amplified DNA fragments obtained using these primers were 5.3 kb for the wild-type strain and 1 kb for the mexXY-deficient strain (data not shown). Thus, we constructed mutants lacking the mexXY region from MexAB-OprM-producing strain PAO1, MexAB-deficient- and OprM-producing mutant KG2225, and MexAB-OprM-deficient mutant KG2239, as described above, and we designated them N101, N102, and N103, respectively.

Open in a separate window

FIG. 1

Schematic representation of procedure for markerless deletion of the chromosomal mexXY gene. (A) Construction of plasmid for deletion of mexXY. The striped boxes represent the mexXY-flanking regions amplified by PCR. (B) Procedure for deletion of chromosomal mexXY. The thin line represents the plasmid sequences. The thick line represents the P. aeruginosa chromosome. The straight lines at the bottom represent the size of PCR products.

Production of polyclonal antisera specific to MexA and MexX.

To obtain antibodies specific to MexA and MexX, the oligopeptides (C)YQIDPATYEADYQSA (amino acids 92 to 106 of MexA) and (C)EDSPTPLTRVEQID (amino acids 197 to 210 of MexX) were synthesized and conjugated to keyhole limpet hemocyanin by Chiron Technologies Pty., Ltd. (Clayton Victoria, Australia). A cysteine residue was added to each N terminus for conjugation. Rabbit antiserum raised against each antigen was prepared by Takara Shuzo Co., Ltd. (Otsu, Japan).

Inducible expression of MexX in PAO1.

To confirm the expression of MexXY in wild-type strains, we prepared rabbit antiserum specific to MexX. For this purpose, a synthetic oligopeptide based on the deduced amino acid sequences of this protein was synthesized and used to immunize rabbits. The increase in susceptibility to tetracycline, erythromycin, and gentamicin caused by the deletion of mexXY from PAO1 suggests that MexXY is expressed in PAO1. However, an immunoblot assay using the antiserum as the primary antibody showed no production of MexX in PAO1 (Fig. ​(Fig.3,3, lane 1). To elucidate the discrepancy between the contribution of MexXY on intrinsic resistance and absence of MexX in PAO1, we examined the expression of MexX in PAO1 exposed to antimicrobial agents. MHB containing subinhibitory concentrations of antimicrobial agents and approximately 5 × 10⁵ CFU of PAO1 per ml were incubated for 18 h at 37°C with shaking, and then total membranes were prepared. The exposure to subinhibitory concentrations of tetracycline, erythromycin, and gentamicin induced the production of MexX in PAO1 (Fig. ​(Fig.3,3, lanes 2 to 4). The amounts of MexA and OprM in the presence of these agents were comparable to those in the absence of the agents (data not shown). Production of MexX was also induced by subinhibitory concentrations of tetracycline in KG2225 and KG2239 (Fig. ​(Fig.3,3, lanes 5 and 6) but not in N101, N102, and N103 (data not shown). These results suggest that the MexXY induced by these agents is associated with spontaneously expressed OprM and contributes to the intrinsic resistance to these agents.

Open in a separate window

FIG. 3

Detection of MexX with antiserum directed against synthetic oligopeptide containing part of the amino acid sequence of MexX. Each lane contains 25 μg of total membrane protein. PAO1 (lanes 1 to 4, and 9), KG2225 (lane 5), KG2239 (lane 6), N135 (lane 7), N136 (lane 8), and N126 (lane 10) were grown in MHB (lanes 1, 7, and 8) or MHB containing 8 (lane 2), 1 (lane 5), or 0.25 (lane 6) μg of tetracycline per ml, 128 μg of erythromycin per ml (lane 3), 0.5 μg of gentamicin per ml (lane 4), or 1 (lane 9) or 0.25 (lane 10) μg of ofloxacin per ml. The MICs of ofloxacin for PAO1 and N126 in MHB were 2 and 0.5 μg/ml, respectively.

MexXY-OprM is the primary system for intrinsic resistance to tetracycline and erythromycin.

Paradoxically, PAO1 lacking mexXY (refer to N101) and PAO1 lacking mexAB (refer to KG2225) showed eight- and fourfold increases in susceptibility to tetracycline and erythromycin, respectively (Table ​(Table2),2), while deletion of either mexXY or mexAB should cause a twofold increase in susceptibility if the two systems contribute equally to intrinsic resistance to these agents. To elucidate the reason for this discrepancy, we tested the susceptibility of N126 (N. Masuda, E. Sakagawa, S. Ohya, N. Gotoh, T. Tsujimoto, and T. Nishino, submitted for publication), which is a markerless mexAB-deficient mutant of a MexAB-OprM-overproducing mutant of PAO1 called OCR1, because the decreased level of OprM (Fig. ​(Fig.2,2, lane 4) in KG2225 might affect the susceptibility. The amount of OprM produced in N126 was lower than that in OCR1 but almost the same as that in PAO1 (Masuda et al., submitted) (Fig. ​(Fig.2,2, lane 3). N126 had almost the same susceptibility to tetracycline and erythromycin as PAO1, whereas N101 and KG2225 showed susceptibilities to these agents four to eight times lower than that of PAO1 (Table ​(Table2).2). In addition, N128, an N126-derived-mexXY-deficient strain, showed a susceptibility to tetracycline and erythromycin lower than those of N126 and N101 and almost equal to those of N102 and N103. These results suggest that the decreased level of OprM diminishes the efflux activity of MexXY-OprM in KG2225 and increases the susceptibility of KG2225 to tetracycline and erythromycin. The data also suggest that MexXY-OprM is the primary system to extrude tetracycline and erythromycin in the wild-type strain and that MexAB-OprM supports this process as a supplementary system.

This research was partially supported by a Grant-in-Aid for Scientific Research from the Ministry of Education, Science, Sports, and Culture of Japan and by a grant from the Ministry of Health and Welfare of Japan.

We are grateful to K. Okamoto for providing KG2225.