JAC Advance Access originally published online on January 29, 2008
Journal of Antimicrobial Chemotherapy 2008 61(3):533-540; doi:10.1093/jac/dkn008
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Original research |
Foggy D-shaped zone of inhibition in Staphylococcus aureus owing to a dual character of both inducible and constitutive resistance to macrolide–lincosamide–streptogramin B
College of Pharmacy and Research Institute of Pharmaceutical Sciences, Seoul National University, Seoul 151-742, Korea
* Corresponding author. Tel: +82-2-880-7874; Fax: +82-2-872-1795; E-mail: ecchoi{at}snu.ac.kr
Received 14 November 2007; returned 27 December 2007; revised 15 December 2007; accepted 31 December 2007
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and methodsResults and discussionFundingTransparency declarationsSupplementary dataReferences |
|---|
Objectives: We identified Staphylococcus aureus strains having blunt inhibitory zones around the clindamycin or josamycin discs proximal to the erythromycin disc filled with slight bacterial growth. This ambiguous phenotype was termed as ‘macrolide–lincosamide–streptogramin B antimicrobial (MLSB) resistance having a foggy D-shaped inhibitory zone’ (fMLSB), and we tried to analyse its molecular mechanisms.
Methods: Forty-one clinically isolated strains of fMLSB S. aureus were studied. The regulatory region of the erm(A) gene, which was found to be the only molecular mechanism of fMLSB, was sequenced. Then, β-galactosidase assays were performed to observe their expression patterns through the nucleotide sequential alteration.
Results: According to the sequencing electropherogram, a grouping was made of a homogeneous nucleotide sequence group (73.2%) and a heterogeneous nucleotide sequence group (26.8%). The former group was composed of a 25 bp tandem duplication type and a 25 bp tandem triplication type. Their β-galactosidase activity was similar to that of constitutive MLSB resistance due to its high basal level. Nevertheless, the predicted mRNA secondary structure of the regulatory region maintains the stem–loops of inducible wild-type erm(A), and thereby its inducible character might be expected in vivo. Strains in the latter group were proven to have two different erm(A) genes, and then dual effect of expression was observed.
Conclusions: The ambiguous phenotype of fMLSB is due to its possessing the dual character of inducible and constitutive expression of erm(A). The dual character is due to having one erm(A) gene of dual character or coexistence of two characterized erm(A) genes simultaneously.
Keywords: fMLSB , MLSB , S. aureus , erm(A)
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and methodsResults and discussionFundingTransparency declarationsSupplementary dataReferences |
|---|
Staphylococcal resistance to the macrolide–lincosamide–streptogramin B antimicrobials (MLSB) is mostly mediated by erythromycin ribosome methylase (erm) genes.1,2 In particular, the erm(A) gene is the most predominant molecular mechanism among clinically isolated MLSB-resistant Staphylococcus aureus forms.3 The erm(A) gene encodes ribosomal methylase and the expression is regulated by translational attenuation. Ribosomal RNA methylation leads MLSB antibiotics not to bind to the ribosome, cross-resistance and peculiar phenotypes, including inducible (iMLSB) and constitutive MLSB resistance (cMLSB).1 The phenotype of the wild-type erm(A) gene is iMLSB owing to an alteration of the mRNA secondary structure by induction and ribosomal stalling by erythromycin.4 Certain mutations in the regulatory region make the stem and loop of the mRNA structure unstable and allow translation of erm(A) transcripts independent of the presence or absence of inducers,3,5 which results in the phenotype cMLSB. In the case of staphylococcal MLSB resistance, these phenotypes predict the susceptibility to lincosamides, 16-membered macrolides and ketolides.3,5,6 Thus, the report of an accurate phenotype for MLSB resistance is important to determine the proper therapeutic strategy.7–9
In a previous study, we identified an ambiguous phenotype among clinically isolated S. aureus.10 There were blunt inhibitory zones around the clindamycin or josamycin discs proximal to the erythromycin disc filled with slight bacterial growth. We termed this phenomenon ‘MLSB resistance having a foggy D-shaped inhibitory zone’ (fMLSB). This phenotype was introduced previously in a case report.7,8 Furthermore, we have reported that the fMLSB phenotype was detected in 9% on average among the MLSB-resistant S. aureus strains in Korea; 10.8% in 1999, 8% in 2001, 4.8% in 2003, and 9% in 2005, and all the fMLSB S. aureus strains had only the erm(A) gene.10 The aim of this study was to define the fMLSB phenotype and analyse its molecular mechanisms.
|
Materials and methods |
|---|
|
TopAbstractIntroductionMaterials and methodsResults and discussionFundingTransparency declarationsSupplementary dataReferences |
|---|
Bacterial strains
Forty-one clinically isolated S. aureus strains exhibiting foggy D-shaped blunt zone of inhibition around the disc containing clindamycin (2 µg) proximal to the erythromycin (20 µg) disc were studied. The strains were from Severance Hospital in Seoul and Kosin University Gospel Hospital in Busan, Korea, from 1999 to 2005; 6 isolates in 1999, 3 isolates in 2003 and 32 isolates in 2005. Multiple isolates from the same patient were avoided.
Determination of the phenotype of D-shaped zone of inhibition
All 41 strains were tested by a disc diffusion method. An aqueous suspension of bacterial growth was adjusted to an optical density of 0.5 at 600 nm (OD600) and inoculated by swab on Mueller–Hinton (MH) agar (Difco, USA; 20 mL in a plate with a diameter of 89 mm). Each plate was set with six discs as shown in the right-hand panels of Figure 1. This arrangement allowed observation of both the growth inhibition under various concentrations (0.1, 1, 10, 100 and 1000 µg) of clindamycin and its resistance inducibility by erythromycin (20 µg). For another observation, an aqueous bacterial suspension was adjusted to an OD600 of 0.5, and then 0.1 mL of 105-fold diluted bacterial suspension was inoculated on MH agar. Each plate was set with erythromycin (20 µg) and clindamycin (100 µg) discs, as shown in the left-hand panels of Figure 1. This arrangement was made in order to compare the size of the bacterial colonies inside and outside of the D-zone. All the aspects were identified after overnight incubation at 37°C.
|
Induction assay
To test inducibility, the growth rate of the strains under clindamycin was determined after induction by erythromycin. Bacterial cultures grown overnight in MH broth without antibiotic were diluted to an OD600 of approximately 0.05 in MH broth. After induction by incubation with 0.2 mg/L erythromycin for 2 h, the cultures were diluted again to an OD600 of approximately 0.05 in MH broth. Growth under 100 mg/L clindamycin at 37°C was monitored by measuring the turbidity of the culture at OD600 for up to 8 h. Cultures under no challenge or induction were also observed for comparison. Strains resistant to MLSB, i.e. iMLSB and cMLSB strains, were observed under the same conditions for comparison.
Analysis of the regulatory regions of the erm(A) gene
The sequences of the erm(A) regulatory regions of 41 strains were analysed. DNA fragments were amplified by PCR using the primers, ERMA-F (5'-gaagtcgtcaaggtgcaaaattac-3') and ERMA-R (5'-gactagctctttggtaaaatgtcc-3'), designed to amplify DNA fragments of the regulatory region including the promoter and the first 147 bases of the erm(A) gene. PCR experiments were carried out as follows: 25 cycles of amplification including 10 s of denaturation at 97°C, 30 s of annealing at 47°C and 40 s of elongation at 74°C. The PCR products were purified with a PCR purification kit (INtRON, Korea) and sequenced by using the primer ERMA-F with an ABI 3730XL DNA Analyzer (Applied Biosystems, USA). According to the sequencing electropherograms, we divided the fMLSB strains into two groups: a homogeneous nucleotide sequence group and a heterogeneous nucleotide sequence group (Figure 3). These two groups were then analysed further in different aspects.
Construction of erm(A)-lacZ reporter plasmids and β-galactosidase assay
Through sequencing, the homogeneous nucleotide sequence group was proven to be composed of two mutation types: a 25 bp tandem duplication and a 25 bp tandem triplication. To analyse them, in-phase fusions of the erm(A) regulatory region with lacZ were constructed; pIND having the wild-type erm(A) regulatory region of Tn554, pTD having the 25 bp tandem duplicate erm(A) regulatory region and pTT having the 25 bp tandem triplicate erm(A) regulatory region. The regulatory region of the erm(A) gene was amplified with the primers ERMA-ECORIF (5'-atctgaattcgtcaaggtgca-3') for the 5' end and ERMA-BAMHIR (5'-acataagcctggatccatttc-3') for the 3' end. The PCR products were purified and ligated to the EcoRI and BamHI sites of pMM156 containing a promoterless lacZ gene.11 The resultant ligation products were introduced into Escherichia coli NM522, and transformants were selected on Luria–Bertani agar plates containing chloramphenicol (10 mg/L). Bacillus subtilis BR151 strains harbouring each confirmed recombinant plasmid were grown to early log phase at 37°C in SPII medium, and β-galactosidase assays were carried out as described previously.12 Cultures were induced for 50 min with either erythromycin or clindamycin under various concentrations, and the optimal concentration for the induction was determined. The β-galactosidase assays were carried out after induction by 0.1 mg/L erythromycin, and Miller units at 0, 5, 10, 20, 30, 45 and 60 min were measured.
In the case of the heterogeneous nucleotide sequence group, we could not determine the exact nucleotide sequence directly. Thus, multiple in-phase fusions of the erm(A) regulatory region with lacZ from each isolate were constructed. The regulatory region of the erm(A) gene was amplified with the primers ERMA-ECORIF for the 5' end and ERMA-BAMHIR for the 3' end. The purified PCR products were ligated to pMM156 and the recombinant plasmid was introduced into E. coli NM522. Numerous transformants were selected and plasmids were extracted. Also, the plasmids were introduced, respectively, into B. subtilis BR151. Then, β-galactosidase assays were carried out and Miller units at 0, 15, 30, 60 and 90 min after induction by 0.1 mg/L erythromycin were measured. For further sequencing, we classified the plasmids into several groups by the expression pattern of the β-galactosidase assay, and then sequencing of more than three randomly selected plasmids for each expression pattern was conducted.
|
Results and discussion |
|---|
|
TopAbstractIntroductionMaterials and methodsResults and discussionFundingTransparency declarationsSupplementary dataReferences |
|---|
Expression of fMLSB resistance
All the 41 fMLSB S. aureus strains exhibited a foggy D-shaped zone of inhibition around the disc containing clindamycin proximal to the erythromycin disc. The size of the D-shaped zone was dependent on the concentrations of clindamycin, as in the case of iMLSB (Figure 1, the right-hand column). Also, we could observe slight growth inside the D-shaped zone. This growth was confirmed by comparing the colony size inside and outside of the D-shaped zone (Figure 1, the left-hand column). The colony size inside was smaller than that outside of the D-shaped zone. To observe the influence of clindamycin challenge before and after induction by erythromycin, the growth rate of the representative fMLSB strain was measured in broth under various conditions (Figure 2). When the fMLSB strain was not induced by erythromycin, the growth rate was inhibited slightly by clindamycin. However, its repressed growth under clindamycin recovered to the level of the cMLSB strain after induction by erythromycin. Similar results were obtained with other fMLSB strains (data not shown).
|
|
The foggy D-shaped inhibitory zone of the fMLSB strain appeared gradually on the agar plate. Initially, it seemed that only the clear D-shaped inhibitory zone had appeared. However, with the lapse of time, the clear D-shaped inhibitory zone became foggy (data not shown). With this phenomenon, we propose that the formation of the foggy D-shaped zone proceeds step by step, in the following order. First, the fMLSB strain grows in a manner similar to the iMLSB strain; its growth is inhibited by clindamycin and the resistance is induced by erythromycin, and naturally, it makes a clear D-shaped zone. After that, the growth around the clindamycin disc recovers from the repression like the cMLSB strain, and small colonies start to appear in the D-shaped zone. The observation of the phenomenon of the similar growth curve in broth of the fMLSB strain may also support this suggestion. Growth was repressed under clindamycin in the early stage of the curve, but ultimately recovered back to the level of the cMLSB strain.
Analysis of the regulatory region of the erm(A) gene
Using the oligonucleotides ERMA-F and ERMA-R, we obtained amplicons of the erm(A) regulatory regions from all the 41 strains and they were sequenced as described previously. The sequencing electropherograms revealed that two groups of the erm(A) translational attenuator were available: a homogeneous nucleotide sequence group and a heterogeneous nucleotide sequence group. The homogeneous nucleotide sequence group showed clear single peaks, but the heterogeneous nucleotide sequence group displayed the coexistence of two peaks in such a manner that two nucleotide sequences were merged (Figure 3).
Among all the 41 fMLSB strains investigated, 30 isolates (73.2%) belonged to the homogeneous nucleotide sequence group. The homogeneous nucleotide sequence group was composed of two mutation types: the 25 bp tandem duplication type and the 25 bp tandem triplication type. The 25 bp tandem duplication type [25 bp nucleotides from the third Shine Dalgarno (SD3) to the three codons of the erm(A) gene were duplicated; Figure 4 (b2)] proved to be the most predominant alteration, 26 isolates (63.4%, Table 1). Four isolates (9.8%) were of the 25 bp tandem triplication type, in which the same site of the 25 bp duplication type was triplicated [Figure 4(b3)]. These alterations, i.e. the 25 bp tandem duplication and triplication, did not result in a frame-shift. Duplication or triplication of the 25 bp nucleotides generated one successive stop codon after three valid codons, resulting in a truncated peptide of three amino acids (Figure 5). Thus, the erm(A) gene would start to translate again and be completely formed into the methylase.
|
|
|
Among the 41 fMLSB strains, 11 isolates (26.8%) displayed merged nucleotide sequences starting from a specific base pair (Figure 3). We speculated that the coexistence of the two peaks was due to the coexistence of two different DNA sequences in a single isolate. Therefore, we tried to separate the nucleotides and analysed the constructs individually.
Analysis of a homogeneous nucleotide sequence group
In the case of the homogeneous nucleotide sequence group, we performed β-galactosidase assays with reporter constructs. The regulatory region and truncated methylase genes were translationally fused with E. coli lacZ, and three kinds of recombinant plasmids were obtained, pIND, pTD and pTT, which were constructed as described previously. β-Galactosidase activity was measured with B. subtilis BR151 carrying pIND, pTD and pTT, respectively, under various concentrations of erythromycin or clindamycin. For plasmid pIND, erythromycin was a potent inducer at concentrations of 0.01 and 0.1 mg/L. The basal level of the erm(A) expression of pIND was low, and the expression level increased 3.9- and 4.6-fold during 50 min of induction at 0.01 and 0.1 mg/L erythromycin, respectively. In contrast to pIND, the basal levels of expression of pTD and pTT were drastically high; 21.3- and 24.7-fold higher than pIND, respectively (data not shown). β-Galactosidase activity at various times after treatment with the optimal concentration of 0.1 mg/L erythromycin also showed that pTD and pTT were not induced by erythromycin (Figure 6).
|
The 25 bp tandem duplicate and triplicate nucleotide sequences did not include part of the inverted repeats, and thus the mRNA secondary structures were maintained (Figure 5). Because their mRNA secondary structures are the same as the wild-type erm(A) regulatory region,13,14 the same process of denaturing the secondary structure and inducible exhibition of the primary SD3 might be a possibility in vivo, even though we could not observe the phenomenon in the β-galactosidase assay in vitro.
Analysis of a heterogeneous nucleotide sequence group
The typical Tn554 containing the erm(A) gene is known to exist as one or two copies per chromosome.15,16 To confirm that the erm(A) gene does exist in the fMLSB S. aureus strain in the typical mode, we performed Southern blotting.17 Visual inspection and densitometric analysis revealed that the copy numbers of the erm(A) gene were all similar and not affected by the phenotype (data not shown). Thus, the analysis of strains in the heterogeneous nucleotide sequence group was performed on the premise that there is/are one or two erm(A) copies per chromosome. We tried to verify the individual genes by multiple molecular cloning and respective characterization. The typical molecular cloning method is performed to obtain information on one specific gene on the assumption that only one type of gene exists per isolate. However, in this case, we performed the molecular cloning to characterize one or two genes of one isolate. Thus, we named this ‘the multiple molecular cloning method’ and carried it out. First, we selected numerous transformed E. coli NM522 for each fMLSB isolate and extracted their corresponding plasmids. We had analysed all the recombinant plasmids in B. subtilis by β-galactosidase assay and grouped all the plasmids into several types according to their expression patterns. Also, more than three plasmids in each expression pattern were sequenced. Some plasmids in the different expression patterns displayed the same sequencing results, so that ultimately only three alterations were identified (Figure 4). According to the results, three mixture types were determined (Table 1): a mixture of the wild-type erm(A) attenuator and the two point mutations T5361G and G5288T, eight cases (19.5% among 41 fMLSB strains); a mixture of the 25 bp tandem duplication and the 25 bp tandem triplication, two cases (4.9%); and a mixture of the wild-type erm(A) attenuator and the 25 bp tandem duplication, one case (2.4%). The coexistence of the two expression types is represented in Figure 7 as the result of β-galactosidase assays. When the alteration of two nucleotides, T5361G and G5288T, was in the homogeneous nucleotide sequence group, the phenotype was typical cMLSB (data not shown). Accordingly, the coexistence of two genes, a wild-type erm(A) attenuator and the two point mutations, may result in the dual characteristic phenotype, fMLSB. The 25 bp tandem duplication or triplication types were introduced, respectively, in the homogeneous nucleotide sequence group showing the phenotype fMLSB. Thus, the coexistence of these two different nucleotide sequences and the mixture of a wild-type erm(A) attenuator and a 25 bp tandem duplication result in having two expression characters simultaneously, and it might be revealed as the phenotype of fMLSB.
|
Conclusions
We observed the fMLSB isolates among the clinically isolated forms of S. aureus. Furthermore, the alteration of the erm(A) regulatory region, which was regarded as the only mechanism, was analysed. Two groups of mutation types were identified according to the sequencing electropherograms. One was the homogeneous nucleotide sequence group having the alteration of a 25 bp tandem duplication or triplication. Although their expression pattern seemed to be of a cMLSB type in a β-galactosidase assay in vitro, their predicted mRNA secondary structures suggest that they might possess both inducible and constitutive characters simultaneously in vivo. The other was the heterogeneous nucleotide sequence group, in which two different erm(A) regulatory regions were found in a single isolate. The strains in this group were considered to have dual characters, one from each gene. These two different groups display both inducible and constitutive expression. In conclusion, the phenotype fMLSB is due to possession of the inducible and constitutive resistance expression characters at once.
|
Funding |
|---|
|
TopAbstractIntroductionMaterials and methodsResults and discussionFundingTransparency declarationsSupplementary dataReferences |
|---|
This study was supported by a grant of the Korea Centers for Disease Control and Prevention (2007-E00035-00).
|
Transparency declarations |
|---|
|
TopAbstractIntroductionMaterials and methodsResults and discussionFundingTransparency declarationsSupplementary dataReferences |
|---|
None to declare.
|
Supplementary data |
|---|
|
TopAbstractIntroductionMaterials and methodsResults and discussionFundingTransparency declarationsSupplementary dataReferences |
|---|
Colour versions of Figures 3 and 5 are available as Supplementary data at JAC Online (http://jac.oxfordjournals.org/).
|
Footnotes |
|---|
Present address. Department of Herbal Skin Care, Daegu Haany University, Gyeongsan 712-715, Korea. 
|
References |
|---|
|
TopAbstractIntroductionMaterials and methodsResults and discussionFundingTransparency declarationsSupplementary dataReferences |
|---|
1 Weisblum B. Erythromycin resistance by ribosome modification. Antimicrob Agents Chemother (1995) 39:577–85.[Web of Science][Medline]
2 Nakajima Y. Mechanisms of bacterial resistance to macrolide antibiotics. J Infect Chemother (1999) 5:61–74.[CrossRef][Medline]
3
Schmitz FJ, Petridou J, Jaqusch H, et al. Molecular characterization of ketolide-resistant erm(A)-carrying Staphylococcus aureus isolates selected in vitro by telithromycin, ABT-773, quinupristin and clindamycin. J Antimicrob Chemother (2002) 49:611–7.
4 Sandler P, Weisblum B. Erythromycin-induced stabilization of ermA messenger RNA in Staphylococcus aureus and Bacillus subtilis. J Mol Biol (1988) 203:905–15.[CrossRef][Web of Science][Medline]
5
Werckenthin C, Schwarz S. Molecular analysis of the translational attenuator of constitutively expressed erm(A) gene from Staphylococcus intermedius. J Antimicrob Chemother (2000) 46:785–8.
6
Davis KA, Crawford SA, Fiebelkorn KR, et al. Induction of telithromycin resistance by erythromycin in isolates of macrolide-resistant Staphylococcus spp. Antimicrob Agents Chemother (2005) 49:3059–61.
7
Fiebelkorn KR, Crawford SA, McElmeel ML, et al. Practical disk diffusion method for detection of inducible clindamycin resistance in Staphylococcus aureus and coagulase-negative staphylococci. J Clin Microbiol (2003) 41:4740–4.
8
O'Sullivan MVN, Cai Y, Kong F, et al. Influence of disk separation distance on accuracy of the disk approximation test for detection of inducible clindamycin resistance in Staphylococcus spp. J Clin Microbiol (2006) 44:4072–6.
9 Siberry GK, Tekle T, Carroll K, et al. Failure of clindamycin treatment of methicillin-resistant Staphylococcus aureus expressing inducible clindamycin resistance in vitro. Clin Infect Dis (2003) 37:1257–60.[CrossRef][Web of Science][Medline]
10 Yoon E-J, Kim H, Kwon A-R, et al. Characterization of S. aureus showing MLS resistance with foggy D shape (fMLS). Yakhakhoeji (2006) 50:199–203.
11
Min YH, Jeong JH, Choi YJ, et al. Heterogeneity of macrolide–lincosamide–streptogramin B resistance phenotypes in enterococci. Antimicrob Agents Chemother (2003) 47:3415–20.
12 Miller JH. Experiments in Molecular Genetics (1972) Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press.
13
Zuker M. Mfold web server for nucleic acid folding and hybridization prediction. Nucleic Acids Res (2003) 31:3406–15.
14
Doktor SZ, Shortridge V. Differences in the DNA sequences in the upstream attenuator region of erm(A) in clinical isolates of Streptococcus pyogenes and their correlation with macrolide/lincosamide resistance. Antimicrob Agents Chemother (2005) 49:3070–2.
15
Tillotson LE, Jenssen WD, Moon-McDermott L, et al. Characterization of a novel insertion of the macrolides–lincosamides–streptogramin B resistance transposon Tn554 in methicillin-resistant Staphylococcus aureus and Staphylococcus epidermidis. Antimicrob Agents Chemother (1989) 33:541–50.
16 Murphy E, Löfdahl S. Transposition of Tn554 does not generate a target duplication. Nature (1984) 307:292–4.[CrossRef][Medline]
17
Olsen JM, Christensen H, Aaerestrup FM. Diversity and evolution of blaZ from Staphylococcus aureus and coagulase-negative staphylococci. J Antimicrob Chemother (2006) 57:450–60.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||