Previous Article | Next Article ![]()
Antimicrobial Agents and Chemotherapy, March 2006, p. 943-948, Vol. 50, No. 3
0066-4804/06/$08.00+0 doi:10.1128/AAC.50.3.943-948.2006
Copyright © 2006, American Society for Microbiology. All Rights Reserved.
Department of Global Health, Rollins School of Public Health,1 Division of Infectious Diseases, School of Medicine, Emory University,6 Respiratory Diseases Branch, Centers for Disease Control and Prevention,3 Georgia Emerging Infections Program, Atlanta, Georgia,5 Department of Respiratory Medicine, Hannover Medical School, Hannover, Germany,2 Connecticut Department of Public Health, Hartford, Connecticut4
Received 4 July 2005/ Returned for modification 31 October 2005/ Accepted 5 December 2005
|
|
|---|
|
|
|---|
Several factors contributing to the emergence and spread of FQ resistance have been identified in streptococci, such as spontaneous mutations within the QRDRs (11), acquisition of resistance via horizontal gene transfer (1), and consequent clonal spread of resistant strains (13).
Whereas penicillin resistance is well documented among Streptococcus pneumoniae strains, which recently have also exhibited increasing resistance to fluoroquinolones, Streptococcus pyogenes (Lancefield group A) is considered to be universally susceptible to penicillin, and to date only three isolated cases of FQ-resistant S. pyogenes have been described worldwide (15, 16, 24). However, in a recent Belgian surveillance study, the prevalence of fluoroquinolone resistance among S. pyogenes strains isolated from patients with pharyngotonsillitis was 5.4% (6). Because S. pyogenes strains are probably exposed to antibiotics as frequently as pneumococci, the reason for the difference in antimicrobial resistance is not clear. Several hypotheses have been suggested to explain the low resistance rates for S. pyogenes (except that to macrolides) compared to those for the closely related pneumococci. One hypothesis assumes that, in contrast to pneumococci, there is a lack of natural transformability in S. pyogenes, which would exclude acquisition of resistance by horizontal gene transfer.
In this paper we describe two cases of FQ-resistant S. pyogenes identified through the Active Bacterial Core surveillance (ABCs), a collaboration between the Centers for Disease Control and Prevention (CDC) and the Emerging Infections Program Network. In contrast to the previous reports of FQ-resistant S. pyogenes, we were able to investigate the origin of FQ resistance in these strains. Strain 741 evolved from a FQ-susceptible ancestor by spontaneous mutations in the QRDR of parC, and isolate 747 acquired the mutations by horizontal gene transfer, probably from Streptococcus dysgalactiae.
|
|
|---|
Antimicrobial susceptibility and characterization of isolates. MICs were determined at CDC by broth microdilution according to CLSI (formerly NCCLS) guidelines (10). The MICs for levofloxacin and ciprofloxacin were confirmed by Etest according to the manufacturer's instructions. The presence of an efflux pump was investigated by determination of the MICs of ciprofloxacin by the agar dilution method in the presence of reserpine (10 mg/liter) (2). A fourfold decrease in the MIC in the presence of reserpine (2 dilution steps) was considered evidence for the presence of an efflux mechanism.
emm typing was performed at CDC's Streptococcal Laboratory as described previously (3).
PCR and DNA sequencing of the QRDR. The QRDRs of the topoisomerase type II genes gyrA and parC were amplified from extracted chromosomal DNA by PCR using the primers and cycling conditions described by Richter et al. (16). The amplification products were purified with ExoSAP-IT (USB Corp., Cleveland, Ohio). DNA sequencing was performed using the BigDye Terminator 1.1 Cycle (Applied Biosystems) with the ABI 3100 automated sequencer.
Sequence analysis. The National Center for Biotechnology Information Blastx program (http://www.ncbi.nlm.nih.gov/blast/bl2seq/wblastl2.cgi) was used for comparison of the QRDR nucleotide sequences with the DNA and protein sequences of wild-type S. pyogenes (S. pyogenes ATCC 12344). The following sequence accession numbers were used as references: for gyrA, GI:37998981; for parC, GI:37999013.
BLAST search. The nucleotide sequence of the QRDR of parC (408 bp; codons 12 through 168) was used as the query for the GenBank (BLAST) search (http://www.ncbi.nlm.nih.gov/BLAST/BLAST.cgi; nonredundant database limited to bacterial sequences). The 19 closest matches were selected for further analysis. The cutoff for accepting sequences was an E value of e140, because the next sequence exhibited a markedly increased E value of e86.
Alignment. The parC sequences (DNA) of the 19 closest matches were aligned with ClustalX 1.8 (21) using the Gonet 250 similarity matrix with a gap opening penalty of 10.0 and a gap extension penalty of 0.1 for the pairwise alignment stage, and a gap opening penalty of 10.0 and a gap extension penalty of 0.2 for the multiple alignment stage.
Phylogenetic reconstruction. Phylogenies were constructed by the Bayesian method (7, 8, 14) as implemented by the program of Bayes (4). The evolutionary model used was the General Time Reversible model (20). Because evolutionary rates are not homogeneous for every site in a gene, the site variation in evolutionary rate was estimated separately for the first, second, and third positions of sites within codons. Four chains, with a "temperature" of 0.2 for the heated chains, were run for each tree. Trees were sampled every 100 generations. A total of 60,000,000 generations were run with a burn-in of 2,000 trees. The length of each burn-in was set at a value that exceeded twice the number of trees required for convergence of the ln likelihood. Because the consensus trees calculated by Bayes do not include the posterior probabilities of the clades, each entire set of trees was imported into the PAUP* (phylogenetic analysis using parsimony) program, version 4.0 (Sinauer Associates, Inc., Sunderland, Mass.), and the same trees used by Bayes to calculate a consensus were used to calculate a 50% majority rule consensus in PAUP* (19). The tree was rooted using the more distantly related parC gene of Streptococcus canis as an outgroup. The resulting tree shows the posterior probabilities of the clades, i.e., the percentage of time that those taxa are included in the clade. The consensus trees calculated by Bayes were imported into PAUP* for the purposes of displaying and printing the tree.
Test for recombination. The maximum chi-square test was used to confirm suspected recombination events between the two parC sequences of the S. pyogenes strain with the highest degree of homology according to the results of the BLAST search (MGAS8232, GI:19748133), the Streptococcus dysgalactiae subsp. dysgalactiae strain (GTC 431T, GI:37999015), and the putative derived parC sequence of isolate GAS 747. The test compares the distribution of polymorphic sites along such sequences with those expected to occur by chance (17). The test was performed using the START package, which is available for download at http://outbreak.ceid.ox.ac.uk/software.htm.
|
|
|---|
(ii) Case B (isolate GAS 741). An elderly white man, a resident of a nursing home, with a history of malignant lymphoma and diabetes was hospitalized for suspected pneumonia. Ciprofloxacin-resistant S. pyogenes was isolated from blood cultures. The patient was not taking antibiotics at the time the culture was obtained but had been treated with a brief course of levofloxacin (dosage unknown) for cold-like symptoms 2 weeks prior to admission. While hospitalized the patient was successfully treated with ceftriaxone, and he was discharged back to the nursing home, where ceftriaxone treatment was to be continued for a total of 14 days.
(iii) Case C (isolate GAS 784). The patient was an elderly black woman with underlying dementia and a history of bilateral above-the-knee amputations living in the same nursing home as the patient with case B. She presented to the emergency room with complaints of nausea, vomiting, and fever; she also had a large decubitus ulcer. For 6 months prior to admission she had been a resident of the nursing home, where she was bedridden, was fed by gastric tube, and used a Foley catheter. Initial therapy consisted of intravenously administered levofloxacin (250 mg once a day) and vancomycin (1 g every 36 h). After blood culture results revealed S. pyogenes (FQ susceptible), the antibiotic treatment was changed to cefazolin (1 g given three times a day). The patient recovered and was discharged to the nursing home 7 days later.
Since further analysis (see below) revealed that the FQ-susceptible strain GAS 785, isolated from another resident of the nursing home where the GAS 741 and GAS 784 patients lived, was not related to GAS 741, the case report for GAS 785 is not presented here.
MICs and characterization. emm types and MICs of ciprofloxacin and levofloxacin for the two ciprofloxacin-resistant isolates (strains 747 [case A] and 741 [case B]) and for the ciprofloxacin-susceptible isolates obtained from the same nursing home as strain 741 (strains 784 [case C] and 785 [case not presented]) are displayed in Table 1. All isolates were susceptible to penicillin, cefotaxime, erythromycin, clindamycin, and vancomycin.
|
View this table: [in a new window] |
TABLE 1. Clinical information and phenotypic characteristics of the S. pyogenes (GAS) isolates investigated
|
Comparison of the GAS 741 parC sequence with that of an S. pyogenes reference strain from GenBank (MGAS8232, GI:19748133) revealed 98.5% similarity. In contrast, comparison of the GAS 747 parC sequence with that of a reference strain revealed a decreased similarity of 95.0%, indicating the possibility of horizontal gene transfer.
Identification of the source of foreign DNA in GAS 741 parC. A search of GenBank using the parC sequence of GAS 747 revealed 19 very close matches (E values, 0.0 to e146) followed by less related sequences (starting at an E value of e86). The 19 close matches with their corresponding E values are displayed in Fig. 1. A very closely related (rank 7) and a less closely related (rank 18) S. dysgalactiae parC sequence were found among other S. pyogenes sequences (Fig. 1).
![]() View larger version (63K): [in a new window] |
FIG. 1. Results of the GenBank (BLAST) search for the parC sequence (408 bp) of isolate 747. The 20 closest matches, including 2 S. dysgalactiae sequences, were included in further analyses. The sequence with the lowest homology (S. canis) was used to root the tree.
|
![]() View larger version (19K): [in a new window] |
FIG. 2. Phylogenetic tree calculated by the Bayesian method. Branching between different species indicates recombination. Besides strain 747, which branches with S. dysgalactiae 431, another S. pyogenes strain (NIHRO1) branches with S. dysgalactiae subsp. equisimilis 842.
|
![]() View larger version (11K): [in a new window] |
FIG. 3. Alignment of parC of strain 747, the S. pyogenes reference strain MGAS8232, and the recombination partner S. dysgalactiae. The D91N mutation, which does not confer FQ resistance, and the nucleotide changes in strain 747 match S. dysgalactiae 432.
|
|
|
|---|
tyrosine or phenylalanine) and parC (serine79
tyrosine or phenylalanine) mutations. Additional mutations in gyrA and parC were also identified among these strains, but their role in FQ resistance in S. pyogenes was not determined. The emergence of FQ resistance in S. pyogenes in vivo has not yet been addressed. FQ resistance is conferred by mutations in the QRDRs of the topoisomerase class II enzymes. Those mutations can arise spontaneously, or they can be transferred by horizontal gene transfer from the same (intraspecies recombination) or a different (interspecies recombination) species. Interspecies recombination occurs more frequently between closely related species, because higher homology between the two parental sequences carries a greater likelihood of successful recombination.
In this paper we present clinical isolates as examples for two paths for the emergence of FQ resistance. (i) Isolate GAS 741 (case B) was recovered from a nursing home resident who had previously been exposed to levofloxacin. An additional invasive S. pyogenes strain was recovered from the same nursing home (isolate GAS 784) and was demonstrated to be the susceptible ancestor of strain GAS 741. This case demonstrates the emergence of FQ resistance due to spontaneous mutation and consequent selection of the resistant mutant by exposure to FQs. (ii) Isolate GAS 747 (case A) showed decreased similarity of the parC sequence to that of the S. pyogenes reference strain. We could demonstrate that this was due to horizontal gene transfer between S. pyogenes and S. dysgalactiae.
The impact of interspecies recombination on the emergence of FQ resistance in S. pyogenes depends on two parameters: (i) the natural frequency of horizontal gene transfer with the donor species (i.e., S. dysgalactiae) and (ii) the prevalence of FQ resistance genes within the donor species. We believe that interspecies recombination with S. dysgalactiae may be a frequent event for a number of reasons. Both species colonize the human nasopharynx, which provides physical contact between the two species as a conditio sine qua non for recombination. There is also a high degree of homology between S. pyogenes and S. dysgalactiae, which increases the likelihood of successful recombination, and our GenBank research and consequent phylogenetic analysis revealed another case of probable S. pyogenes-S. dysgalactiae recombination. Interestingly, this particular S. pyogenes strain (NIH-RO1-GAS) has also been shown to be FQ resistant (24). The parental S. dysgalactiae subsp. equisimilis strain was also FQ resistant and had mutations in parC and gyrA. However, the origin of the isolate could not be obtained from the literature (5). Some alleles of the S. dysgalactiae gki gene, determined by multilocus sequence typing, are identical to those of S. pyogenes, indicating frequent recombinational events between the two species (B. Beall, CDC, personal communication). Finally, other examples of interspecies recombination between group A and group G streptococci have been described (18, 22).
S. dysgalactiae may serve as a resistance gene pool for S. pyogenes. However, the prevalence of FQ resistance among S. dysgalactiae strains is not known. The species S. dysgalactiae consists of two subpopulations of human and animal (particularly porcine and bovine) origin. A study by Meunier and coworkers addresses the prevalence of FQ resistance in bovine S. dysgalactiae (9). MICs of marbofloxacin, an FQ for individual administration to animals, were determined. In that study about 1.6% of S. dysgalactiae strains causing bovine mastitis were not susceptible to marbofloxacin. Since FQs are frequently used for cattle and particularly for pigs, one could postulate that resistance may spread to human strains.
In conclusion, horizontal gene transfer of FQ resistance has not previously been described for S. pyogenes. In this study we show that, in addition to the selection of FQ-resistant mutants by FQ exposure and consequent spread (isolates GAS 784 and 741), horizontal gene transfer may be another mechanism of acquisition of FQ resistance in S. pyogenes. This mechanism of acquisition of resistance has been described for the closely related species S. pneumoniae. The frequency of recombination events leading to FQ resistance in S. pneumoniae ranges from 0% in the United States (12) to 11% in Spain (1). The extent to which horizontal gene transfer will contribute to emerging FQ resistance in S. pyogenes remains to be seen.
ABCs team members include Tami Skoff, Carolyn Wright, Delois Jackson, and Bernie Beall (all of the Centers for Disease Control and Prevention, Atlanta, Ga.); James Hadler, Connecticut Department of Public Health, Hartford; and Monica M. Farley, Emory University, Atlanta, Ga.
M.W.R.P. was supported by a scholarship from the German Research Foundation (Deutsche Forschungsgemeinschaft).
|
|
|---|
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»