INTRODUCTION

Top Abstract Introduction Materials and Methods Results and Discussion References

Macrolide antibiotics are active mainly against gram-positive bacteria, but erythromycin (EM) and clarithromycin are active against Helicobacter, Legionella, and Mycoplasma spp. (6, 31). Furthermore, it is becoming clear that some of these antibiotics have various other pharmacological activities, for example, as anti-inflammatory agents in addition to antibacterial agents (11). Thus, the medical use of macrolide antibiotics has increased significantly in recent years.

Resistance to macrolide antibiotics is usually due to modification of the target site (12, 29), active efflux of the antibiotic (24), or inactivation of the antibiotic (13). However, macrolide-inactivating enzymes, namely, EM esterases (1, 20) and macrolide 2'-phosphotransferases (10, 19), that mediate such resistance have been found in clinical isolates of Escherichia coli, which is naturally resistant to macrolides. Furthermore, almost all macrolide-inactivating enzymes are constitutively produced in E. coli (1, 2, 10, 20). However, the production of macrolide 2'-phosphotransferase I [Mph(A); formerly MPH(2')I] (22), which is a strong inactivator of 14-member ring macrolides, such as EM and oleandomycin (OL), is induced by EM in the original strain E. coli Tf481A (19). Furthermore, although the mph(A) gene confers low-level resistance to EM, cells that carry the mph(A) and mrx genes come to exhibit high-level resistance to EM. The deduced product of the mrx gene, Mrx, is a hydrophobic protein, but its function remains to be determined.

As an inducible macrolide resistance gene, the ermC gene, which encodes an rRNA methylase, is been well known (29). Inducible expression of ermC is regulated at the posttranscriptional level by attenuation of translation (30). However, not only the mechanism of the macrolide resistance conferred by ermC but also the structure of the ermC gene is completely different from that of the mph(A) gene.

In the present study of the inducible expression of mph(A), we identified the gene that regulates its expression and characterized the gene product. We also developed a model that explains the inducible expression of mph(A).

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results and Discussion References

Bacterial strains, plasmids, and antibiotics. The bacterial strains and plasmids used in this study are listed in Table 1. Expression of the mph(A) gene was induced by treatment with a subinhibitory concentration of EM (25 or 50 µg/ml) (16, 19).

Analysis of protein expression in minicells. Minicells were prepared from E. coli TH1219 that carried the indicated plasmid, and plasmid-encoded proteins were analyzed as described previously (17). Standard proteins for determination of molecular weight were purchased from Bio-Rad Laboratories.

Enzymatic inactivation of antibiotics. A crude preparation of enzymes was used as the source of Mph(A). The preparation of enzymes and the enzymatic inactivation of macrolide antibiotics were performed as described previously (18).

DNA manipulation and sequencing. All DNA manipulations were performed by standard methods (26). DNA amplification by PCR was performed using Gene Taq polymerase (Wako Pure Chemical Industries Ltd., Osaka, Japan). Nucleotide sequences were determined with an automated DNA sequencer (PRIZM 377) and a dye terminator cycle sequencing kit (both from PE Applied Biosystems, Foster City, Calif.).

Northern blotting and primer extension analysis. RNA was prepared from E. coli as described previously (18). A 508-bp fragment that had been amplified by PCR with primers mphA+835 (5'-GCTCGACTATAGGATCGTGATCGC) and mphA1343 (5'-CGTAGAGATCGCCATGCACCAC) was used as the probe for mph(A) transcripts. The probe was labeled with an AlkPhos Direct labeling kit (Amersham Pharmacia Biotech Inc.) and allowed to hybridize with RNA in accordance with the instructions from the manufacturer of the labeling kit. Primer extension analysis was performed as described previously (18), using primer mphA795 (5'-CCATGTCGGGCTGCAAGTGCGTACAGTTGGG), which was end labeled with [-³²P]ATP and T4 polynucleotide kinase.

Construction of a plasmid carrying a fusion gene for MphR(A)-GST and purification of MphR(A). A DNA fragment containing the mphR(A) structural gene was amplified by PCR with two primers that included EcoRI and SalI restriction sites (underlined), namely, mphR+1 (5'-AAGGTGAGAATTCATGCCCCGCCCCAAGCT) and mphR1 (5'-GGACTCTGTCGACCTCCGTTTACGCATGTG), respectively. Plasmid pGEX-mphR(A) was constructed by subcloning an EcoRI-SalI mphR(A) fragment from pTZ3509 between the EcoRI and SalI sites of pGEXT4-1. The synthesis of the glutathione S-transferase (GST)-MphR(A) fusion protein and purification of MphR(A) were performed as described by Smith and Johnson (27). The concentration of protein was estimated by Bradford's method (3) with bovine serum albumin as the standard.

Gel mobility shift assay. Labeled DNA fragments were prepared by PCR using primers that had been end labeled with [-³²P]ATP. The 66-bp fragment designated DNA-1, containing the promoter region of mph(A), was generated with primers mphA+638 (5'-CTGCCTCATCGCTAACTTTG) and mphA703 (5'-CCTAAATGTAACAGTCA). The 77- and 60-bp fragments designated DNA-2 and DNA-3 were generated with primers mphA+591 (5'-GGTAAGCAGAGTTTTTGAAATGTAAGGCCT) and mphA667 (5'-GGCACTGTTGCAAAGTTA) and primers mphA+696 (5'-CATTTAGGTGGCTAAACCC) and mphA755 (5'-CGGTCGTGACTACGGTCATGA), respectively. The ³²P-labeled DNA fragments (approximately 5 × 10³ cpm/reaction) were incubated with MphR(A) in DNA binding buffer that contained 5 µg of bovine serum albumin, 1 µg of poly(dI-dC), and 10% glycerol in a total volume 30 µl (7). After a 30-min incubation at room temperature, the reaction mixtures were analyzed by 8% polyacrylamide gel electrophoresis in Tris-borate buffer.

DNase I footprinting. DNase I footprinting of the promoter region of mph(A) was performed using a 165-bp DNA fragment that had been amplified by PCR with primer mphA755, which had been end labeled with [-³²P]ATP, and primer mphA+591 (4). Reaction mixtures for analysis of binding contained approximately 1 pmol of labeled DNA and 10, 25, or 50 pmol of purified MphR(A) in 50 µl of a reaction buffer composed of 50 mM Tris-HCl (pH 7.4), 50 mM KCl, 1 mM EDTA, 1 mM 2-mercaptoethanol, 5 µg of bovine serum albumin, 0.2 µg of poly(dI-dC), and 10% glycerol. Each reaction mixture was incubated for 30 min at 37°C. Then, 0.5 U of DNase I (Promega) was added to the reaction mixture and the mixture was incubated for 1 min at room temperature. The DNA fragments were analyzed by 8% polyacrylamide gel electrophoresis sequencing with a G+A sequencing ladder (15).

Nucleotide sequence accession number. The nucleotide sequence reported here has been deposited in the DDBJ, EMBL, and GenBank databases under accession number AB038042.

	RESULTS AND DISCUSSION

Top Abstract Introduction Materials and Methods Results and Discussion References

Identification and nucleotide sequence of the mphR(A) gene. To confirm the production of Mph(A) directly and to identify the regulatory gene [mphR(A)] that controls the inducible expression of the mph(A) gene for Mph(A) in minicells, we recloned the inserts in pUC119 into pHSG399, in which the ampicillin resistance gene of pUC19 had been replaced with a chloramphenicol resistance gene, since the electrophoretic mobility of -lactamase (32 kDa), which was generated from the ampicillin resistance gene on the pUC119 vector, was similar to that of Mph(A) (16). We introduced various derivatives of pHSG399 into minicell-producing E. coli and analyzed the radiolabeled products (Fig. 1). In the case of plasmid pTZ3517, which contained a 3.3-kb PstI-BamHI fragment that included the mph(A) and mrx genes, high-level production of Mph(A) was recognized when cells were cultured in the presence and in the absence of EM (Fig. 1). However, in case of plasmid pTZ3510, which contained a 4.1-kb PstI fragment that consisted of 0.8 kb of the BamHI-PstI fragment that included the downstream region of the mrx gene and the above-mentioned 3.3-kb PstI-BamHI fragment, the level of Mph(A) produced in the absence of EM was lower than that produced in the presence of EM. This result indicated that the regulatory gene mphR(A) is located downstream from the mrx gene and that the expression of mph(A) is negatively regulated by mphR(A). However, no specific products encoded by mrx and mphR(A), respectively, were detected after autoradiography of the proteins produced in minicells. Furthermore, although the product of the mrx gene is required for expression of the high-level resistance to EM mediated by Mph(A), the production of Mph(A) in the minicells carrying pTZ3514, which included mrx with a nonsense mutation, was as enhanced in the presence of EM as that of the minicells carrying pTZ3510. This result indicated that mrx is not required for the inducible expression of mph(A).

View larger version (40K): [in this window] [in a new window]   FIG. 1.   Autoradiogram after sodium dodecyl sulfate-polyacrylamide gel electrophoresis showing plasmid-encoded proteins labeled with [14C]leucine that were produced in minicells. The autoradiograms show radioactive proteins synthesized in the absence of EM ( lanes) and in the presence of EM (+ lanes). The mobilities and molecular weights (in thousands) of standard proteins are indicated on the left. CAT, chloramphenicol acetyltransferase; Mph(A), macrolide 2'-phosphotransferase I.

View larger version (49K): [in this window] [in a new window]   FIG. 2.   Structures of the mph(A), mrx, and mphR(A) genes. The nucleotide sequence from nt 521 to nt 4070 is shown. The 35 and 10 sequences that define the promoter and ribosome-binding site (RBS), as well as various restriction sites, are underlined. The arrows indicate the positions and directions of primers used in this study. The initiation transcription site and IS174 are indicated by hooked arrows. The binding site of MphR(A) is shown by a dotted line. The putative -helices and turn in MphR(A) are boxed and underlined, respectively.

Repression of transcription of mph(A) by the MphR(A) protein. To determine whether the expression of the mph(A) gene is regulated at the transcriptional level, we analyzed the transcription of mph(A) by Northern blotting (Fig. 3). Although there was no difference in the transcript of mph(A) from pTZ3519 in the presence and in the absence of EM, as observed in our minicell experiment, the level of the transcripts from pTZ3509 was raised when cells were cultured with EM. Additionally, we found an mRNA of 2.9 kb among transcripts in cells that harbored pTZ3509 but not in those that harbored pTZ3519. The level of the 2.9-kb mRNA was markedly elevated in the presence of EM. Therefore, the expression of mph(A) was clearly regulated at the transcriptional level.

View larger version (64K): [in this window] [in a new window]   FIG. 3.   Northern blotting analysis of transcripts from E. coli cells harboring pTZ3519 and pTZ3509. Cells were grown in the absence of EM ( lanes) and in the presence of EM (+ lanes). The mobilities and sizes (kilobases) of standards are indicated on the left. Specific transcripts of interest are indicated on the right, along with their lengths in kilobases.

View larger version (48K): [in this window] [in a new window]   FIG. 4.   Mapping of the transcription initiation site of mph(A). Size markers (lanes G, A, T, and C) were generated in sequencing reactions performed with pTZ3519 DNA. The arrow on the right indicates the product of the extension reaction (PE).

Specific binding of MphR(A) to the promoter region of mph(A). An analysis of the putative product of the mphR(A) gene revealed that the first 64 amino acids of MphR(A) exhibit strong homology to proteins in the TetR/AcrR family. The proteins in this family bind in the vicinity of the promoters of their respective target genes (8, 14). To examine whether MphR(A) is the repressor protein that controls expression of mph(A), we constructed an mphR(A) expression system and purified MphR(A). We performed gel mobility shift assays using the purified protein and a 66-bp fragment (DNA-1) that includes the promoter sequence of mph(A). As shown in Fig. 5, the mobility of the DNA-1 fragment was shifted in the presence of MphR(A). To analyze the binding specificity of MphR(A) in further detail, we performed a competition experiment. The shifted DNA band was almost completely abolished when the unlabeled DNA-1 fragment was added as a competitor to the reaction mixture before addition of the labeled DNA. By contrast, an excess of individual nonspecific competitors that included regions either upstream (DNA-2) or downstream (DNA-3) from the promoter of mph(A), as well as an excess of poly(dI-dC), had no effect on the binding of MphR(A) to radiolabeled DNA-1. These results indicated that MphR(A) bound specifically to the promoter region of mph(A).

View larger version (82K): [in this window] [in a new window]   FIG. 5.   Gel mobility shift assay with the promoter region of the mph(A) gene. A 66-bp fragment (DNA-1) including the promoter of the mph(A) gene was amplified with end-labeled primers mphA+638 and mphA667 (see Materials and Methods) and used as the probe. Reaction mixtures contained no MphR(A) protein (lane 1); no competitor (lane 2); unlabeled competitor DNA-1 (lane 3); the DNA-2 fragment, amplified with primers mphA+591 and mphA667 (lane 4); and the DNA-3 fragment, amplified with primers mphA+638 and mphA703 (lane 5).

Identification of the MphR(A) binding site. DNA-binding proteins often bind sequences displaying dyad symmetry. However, no repeated sequence was present in the region of the promoter of the mph(A) gene. We identified the exact location of the binding site of MphR(A) in a DNase I footprinting experiment with purified MphR(A) and an end-labeled 165-bp DNA fragment that included the promoter region of mph(A). Relative to the transcription initiation site of mph(A), the region protected by MphR(A) extended from nt 32 (674 nt) to nt 3 (703 nt) on the coding strand and from nt 37 (669 nt) to nt 9 (698 nt) on the noncoding strand (Fig. 6). Thus, the protected region (from nt 37 to nt 3) on the two strands corresponds exactly to the promoter region of the mph(A) gene. These results indicated that MphR(A) represses the initiation of transcription of the mphA operon by blocking the binding of RNA polymerase to the promoter of the mph(A) gene.

View larger version (100K): [in this window] [in a new window]   FIG. 6.   DNase I footprinting analysis of protection by MphR(A) of the promoter region of mph(A). (A) Footprinting analysis of the coding strand. (B) Footprinting analysis of the noncoding strand. G+A indicates patterns of Maxam-Gilbert chemical cleavage reactions. Lanes 0 through 50 indicate digestion with DNase I of DNA in reaction mixtures that contained increasing concentrations (in micrograms per milliliter) of DNase I. The thick vertical lines delineate regions protected by MphR(A). The transcription initiation site and the promoter region (10 and 35) are also indicated. The values to the right are lengths in kilobases.

Effects of macrolides on binding of MphR(A) to the promoter. In the presence of a subinhibitory concentration of EM, namely, 25 µg/ml (19), the rate of transcription of mph(A) is elevated. To determine whether macrolides can directly inhibit the binding of MphR(A) to the promoter of mph(A) and to identify the kinds of macrolide that can act as inducers, we performed gel mobility shift assays using a variety of macrolides at various concentrations. Included were EM and OL as representative 14-member ring macrolides and kitasamycin and josamycin as representative 16-member ring macrolides. Each macrolide in the reaction mixture abolished the binding of purified MphR(A) to the promoter fragment as the concentration of the macrolide was increased. However, the concentration at which 14-member ring macrolides inhibited the binding of MphR(A) to the DNA was 100-fold lower than at which 16-member ring macrolides inhibited such binding (Fig. 7). This result is supported by the profiles of inactivation of macrolides by Mph(A) and the pattern of susceptibility to macrolides of E. coli that carried the mphA operon. Although the inhibition of the binding of MphR(A) to the DNA by EM was slightly stronger than that of OL, we found no difference in the ability to induce mph(A) in the presence of EM and OL by monitoring the inactivation of OL (data not shown). These results indicated that 14-member ring macrolides are the inducers of the mphA operon and directly inhibited the binding of MphR(A) to the promoter of the mph(A) gene.

View larger version (45K): [in this window] [in a new window]   FIG. 7.   Gel mobility shift assays showing the binding of MphR(A) to the promoter region of mph(A) in the presence of various macrolides. Lanes 0 through 100 contained increasing concentrations (in micrograms per milliliter) of macrolides. LM, kitasamycin; JM, josamycin.

A model for regulation of expression of the mphA operon. Based on our results, a model of the inducible expression of the macrolide resistance determinant mphA is presented in Fig. 8. Thus, the mechanism of inducible expression of the mphA determinant is entirely different from that of the typical macrolide resistance determinant ermC.

View larger version (26K): [in this window] [in a new window]   FIG. 8.   A model for regulation of the expression of the mph(A) operon in E. coli. P represents the promoter of the mph(A) gene. Induced and uninduced expression is indicated by dotted and solid lines, respectively. The mph(A) gene encodes the repressor protein MphR(A) and the macrolide resistance determinant consisted of the mphA operon, which is made up of the mph(A), mrx, and mph(A) genes. The expression of the mphA operon is negatively regulated at the transcriptional level by the binding of MphR(A) to the promoter of the mph(A) gene. In the presence of a subinhibitory concentration of a 14-member ring macrolide, transcription of the mphA operon is activated by blockage of the binding of MphR(A) to the mph(A) promoter.