点击显示 收起
Division of Infectious Diseases, Department of Medicine, Department of Pathology, College of Physicians and Surgeons, Department of Statistics
Department of Epidemiology, Mailman School of Public Health, Columbia University, Panna Technologies and Division of Infectious Diseases, Department of Medicine, and the Department of Epidemiology and Population Health, Montefiore Medical Center and the Albert Einstein College of Medicine, New York, New York
Public Health Research Institute, Newark, New Jersey
Nasal carriage of Staphylococcus aureus is often a prelude to infection with the same strain. The prevailing assumption has been that colonized individuals carry a single strain. The present study investigated the frequency of simultaneous nasal carriage of multiple strains of S. aureus. Three bacterial colonies from plated samples from colonized subjects were initially compared by pulsed-field gel electrophoresis. Fourteen of 148 S. aureuspositive samples demonstrated at least a difference of a single band; 7 of these 14 samples contained different strains, and 3 of these 7 also belonged to different accessory gene regulator (agr) types. The remaining 7 samples contained clonally related isolates; 3 of these 7 contained pairs that differed by the presence or absence of the staphylococcal chromosomal cassette mec type IV. A mathematical model that we developed predicted that 6.6% of S. aureuscolonized individuals carry >1 strain. The present study demonstrates that carriage of discordant S. aureus strains in individuals with nasal colonization occurs regularly and suggests that the nares are likely sites for horizontal genetic exchange among strains.
Staphylococcus aureus is a transient or persistent part of the resident flora in the anterior nares of 20%50% of the population [1, 2]. A relatively small proportion of colonized individuals appear to remain colonized with the same strain for prolonged periods [3]. When either mucosal or cutaneous barriers are breached, individuals may develop life-threatening infections with their resident strains of S. aureus [4, 5]. Therefore, the S. aureus strains that reside in the anterior nares serve as a reservoir for future infection. Despite its importance, we have only a limited understanding of the biological nature of S. aureus nasal colonization [6].
The underlying presumption of virtually all studies of the nasal carriage of S. aureus is that carriage is clonal; individuals colonized by S. aureus are colonized by a single strain. This is in contrast to Hemophilus species and Streptococcus pneumoniae, for which carriage is frequently polyclonal [7].
The issue of the clonality of carriage is important. Biologically, the presence of >1 strain of S. aureus increases the potential for the horizontal exchange of genetic information, including antimicrobial-resistance genes or virulence determinants that may contribute to the invasiveness of a particular isolate. In light of the increasing evidence that conjugal transfer of genetic information is more common than has been suspected, the anterior nares might be an important site for this type of exchange if multiple strains are present [8].
Clinically, screening a single isolate from patients infected with S. aureus may identify an antibiotic-susceptible strain rather than a second, more resistant strain; selection of an antibiotic regimen may therefore be incorrect. This has been demonstrated to be a concern for coagulase-negative staphylococci, for which carriage is often polyclonal [9, 10].
From an epidemiologic perspective, the vast majority of studies of carriage have relied on sampling a single colony from a positive plate [2]. As a result, conclusions regarding a change in colonizing strains may reflect the selection of a different co-colonizer.
In an effort to enhance our understanding of the biological nature and dynamics of nasal colonization, we investigated the clonality of S. aureus carriage among subjects participating in an ongoing survey of nasal colonization with S. aureus. The issues addressed include the frequency of carriage of discordant S. aureus strains, the characteristics of discordant strains, and the potential effect of accessory gene regulator (agr) type on carriage. As a part of this analysis, a mathematical model to predict the frequency of discordance in individuals colonized by S. aureus was developed. Epidemiologic factors that might potentially contribute to discordance, such as a history of antibiotic exposure and hospitalization, were also examined.
SUBJECTS, MATERIALS, AND METHODS
Study population.
S. aureus isolates from 2 different study populations were analyzed. The majority of samples (from 121 of 500 subjects whose nasal cultures were positive for S. aureus) were collected from a cohort of current and former drug users who were participants in a hospital-affiliated methadone maintenance program (MMP) in the Bronx, NY, and were recruited into an ongoing longitudinal study of the natural history of HIV disease [11, 12]. Subjects were interviewed and, when possible, recultured every 6 months for up to 18 months. Additional strains were collected from patients who were admitted to the Montefiore Medical Center (New York, NY) and were suspected of having a S. aureus infection at the time of admission (28/98 subjects whose nasal cultures were positive for S. aureus). After providing informed consent, subjects from both study populations underwent (1) a standardized interview that included questions on sociodemographic information, drug use, and medical history and (2) a limited physical examination that included cultures of the anterior nares for S. aureus. The institutional review boards of Columbia University and the Montefiore Medical Center approved the present study.
Screening for nasal carriage of S. aureus.
Anterior nasal cultures were initially plated onto mannitol salt agar for 48 h at 37°C. The density of S. aureus colonization was graded on a scale of 0 to 3 (0 = 0 colonies, 1 = 1100 colonies, 2 = 1011000 colonies, and 3 = >1000 colonies). A sample (collected by swabbing the plate surface) of the primary culture plate was suspended in medium and stored in 20% glycerol at -80°C.
For each positive culture, 3 colonies were randomly selected from the primary culture plate, were replated onto sheep-blood agar plates (Becton Dickinson), and were incubated for 24 h at 37°C. Identification of S. aureus was confirmed by use of StaphAurex (Murex Biotech). These isolates were frozen, at -80°C, in 20% glycerol for future strain typing and analysis. Hereafter, each set of 3 colonies collected from 1 subject with nasal colonization is referred to as a "sample."
Antibiotic-susceptibility testing.
Antibiotic-susceptibility testing was performed at 35°C on Mueller-Hinton agar by use of the Kirby-Bauer disk-diffusion technique, as outlined by the NCCLS [13]. The antibiotic-susceptibility panel included vancomycin, trimethoprim-sulfamethoxazole, rifampin, penicillin, oxacillin, levofloxacin, gentamicin, erythromycin, clindamycin, cephalothin, ampicillin, and amikacin (all from Becton Dickinson).
Pulsed-field gel electrophoresis (PFGE).
All isolates were initially typed by PFGE. Preparation of DNA and resolution of the SmaI-digested fragments was performed as described elsewhere [14]. Samples were run on a CHEF-DR III System (BioRad) with 1.0% agarose gels in 50 mmol/L Tris, 50 mmol/L boric acid, and 1 mmol/L EDTA. Settings for the PFGE were as follows: initial switch time, 1.0 s; final switch time, 30.0 s; included angle, 120°; current, 6.0 V; and run time and temperature, 23 h at 14°C.
The PFGE results were entered into an archival, analytic program (Diversity Database software; version 2; BioRad). All DNA fragments were identified and sized by comparison with molecular size standards from each gel. The strain profiles were then compared by use of the analytic software. A tolerance of ±1% for band-size variations from different strains was used for determination of the similarity of strains [15].
Determination of the relatedness of strains.
A sample from the same subject with a single strain that, by PFGE, differed by even a single band was collected for additional analysis. Discordant samples were initially defined as differences of >3 bands. These strains were then reanalyzed by S. aureus protein A (spa) gene sequence typing and multilocus sequence typing (MLST). The low threshold for the selection of samples for further analysis was used to (1) assure the identification of all potentially discordant samples and (2) characterize the frequency and nature of both the samples with discordant strains and those with strains with minor clonal variations. For the statistical model, only isolates that belonged to different MLST and spa types were considered to be discordant. The remaining isolates that belonged to identical MLST and spa types were considered to be concordant.
spa typing.
spa typing was performed as described elsewhere [16]. This scheme relies on the polymorphism of the variable-number tandem repeat region of the spa gene. Strains were categorized as identical on the basis of an identical spa type designation.
MLST.
MLST was performed as outlined by Enright et al. [17]. Strains were analyzed on the basis of the partial sequences of 7 S. aureus housekeeping genes. Polymerase chain reaction (PCR) was performed with 50-L reaction volumes, including 5 L of chromosomal DNA, 20 pmol of each primer, 1.25 U of Platinum Taq polymerase (Invitrogen), 5 L of 10× PCR buffer, and 10 mmol/L of each dNTP. A GeneAmp PCR System 9600 Thermocycler (Perkin-Elmer) was used. Primers were as described by Enright et al. [17], except that the original forward primer for the gene arcC was replaced by the sequence 5-CCTTTATTTGATTCACCAGCG-3 [18]. PCR products were purified by use of the QIAquick PCR Purification Kit (Qiagen) and submitted to the Rockefeller University sequencing facility. Sequences were analyzed by use of DNASTAR software (version 5.05) and submitted to an Internet database, http://www.mlst.net, which assigned an allelic profile and sequence type to each isolate.
Additional analysis of methicillin-resistant strains.
Methicillin-resistant S. aureus (MRSA) isolates were further analyzed for their specific staphylococcal chromosomal cassette (SCC) mec type by use of the multiplex PCR assay developed by Oliveira et al. [19] and were validated by use of primers for the pls region of the SCCmec complex (forward, 5-CAACAGGAGCAGATGGCTG-3; reverse, 5-GGTGAGCCTGGTTCTAGAG-3) as well as primers for the ccr gene complex, as described by Okuma et al. [20]. H. de Lencastre (Rockefeller University, New York, NY) provided control strains containing the different SCCmec types.
Southern-blot analyses of discordant isolates from the same subject colonized by both MRSA and methicillin-susceptible S. aureus (MSSA) strains were compared as described elsewhere [21]. An internal sequence of mecA (GenBank accession number AB033763; nt 32,57332,734) was biotin labeled by use of the NEBlot Phototope Kit (New England BioLabs); hybridization was performed as recommended by the manufacturer. Generation and detection of signals were done according to the protocol of the Phototope-Star Detection Kit (New England BioLabs).
agr typing.
Isolates were analyzed for agr type by use of the method of Jarraud et al. [22]. agr specificity groups were identified by PCR amplification of the hypervariable domain of the agr locus with oligonucleotide primers specific for each of the 4 major specificity groups, as described by Shopsin et al. [23].
Statistical analysis.
A case-control comparison that used S. aureuscolonized subjects from the 2 study populations as control subjects (18/117) was established. Every tenth subject with an S. aureuspositive sample and with 3 identical profiles by PFGE, as well as all subjects colonized by MRSA isolates (because of limited numbers), were selected for comparison with the discordant group. Subjects with discordant isolates from the Montefiore Medical Center study population were also part of this analysis.
To examine differences in sociodemographic characteristics, drug use, and medical history between subjects with discordant and concordant samples, the 2 test was used for analysis of categorical variables, and Student's t test was used for analysis of continuous variables. Testing was 2-sided, and P < .05 was considered to be significant; Fisher's exact P values were reported for analyses in which cell counts were expected to be <5. SAS software was used for all statistical analyses (version 8.02; SAS Institute).
Development of a mathematical model to estimate the frequency of discordance among S. aureuscolonized individuals.
A mathematical model was developed to estimate the frequency of discordance among S. aureuscolonized individuals. The data for the model were based on the results of PFGE performed on 3 colonies from the primary culture plate of each subject's nasal culture, regardless of the total number of colonies on a plate. The distribution of the number of colonies per plate was similar between plates with concordant samples and those with discordant samples (table 1).
The mathematical model, which was developed by use of standard probability theory, was based on the 121 S. aureuspositive samples (of which 4 were discordant) from the 500 baseline respondents who participated in the MMP study [24]. To further refine the model and to increase the accuracy of the point estimate of discordance, the frozen aliquots from the original plates were subject to repeat analysis (by use of the methods outlined aboveagain selecting 3 colonies and performing PFGE), to determine the reproducibility of initial observations. The results from the repeat analysis were incorporated into the final probability model.
RESULTS
Molecular typing of the discordant samples.
One hundred twenty-one of 500 and 28 of 98 anterior nares cultures from the MMP and Montefiore Medical Center study populations, respectively, were positive for S. aureus. In the follow-up evaluation of the MMP subjects, 50 (18%) of 282, 21 (21%) of 98, and 5 (56%) of 9 had cultures that were positive for S. aureus at 6, 12, and 18 months, respectively.
Two different patterns of discordant colonization were demonstrated. The first pattern consisted of isolates (from 7 of the samples) that differed by >3 bands (all 7 differed by 12 bands), were <70% related by Dice coefficient analysis, and belonged to different spa and MLST types (table 1). The second pattern consisted of isolates (from the remaining 7 samples) that differed by 3 bands (figure 1A) and belonged to the same spa and MLST type; these therefore represented clonal variants (table 1). Two of these latter samples contained both MRSA and MSSA isolates. Among both the discordant and concordant samples, the most common strain identified belonged to spa type 7 and MLST type 8. All 12 subjects colonized by a spa type 7 strain (including subjects with both concordant and discordant samples) were infected with HIV, whereas 16 (59%) of 27 known HIV-infected subjects were colonized with a strain other than spa type 7 (P = .009).
Antibiotic susceptibility.
Eight (57%) of the 14 subjects with discordant samples were colonized by at least 1 MRSA isolate. This is in marked contrast to the overall prevalence of MRSA strains in the MMP study population, for whom only 12 (10%) of 121 S. aureus isolates were MRSA strains (P < .001). The prevalence of MRSA strains in the community that the Montefiore Medical Center serves, which is the same community in which the MMP study population resides, is 13% [25]. Although the majority of the MRSA isolates in both the discordant and concordant groups were highly resistant (i.e., were also resistant to clindamycin, levofloxacin, gentamicin, and trimethoprim-sulfamethoxazole), 1 discordant and 3 concordant MRSA isolates did appear to fit the "community MRSA phenotype," because they were resistant only to penicillin, oxacillin, and erythromycin [2628].
Southern-blot hybridization in which an internal sequence from the mecA gene was used to probe the PFGE of these samples revealed that the MSSA isolates lacked the band containing the mecA gene (figure 1B). In particular, the 7 MRSA/MSSA pairs differed by the presence or absence of the band containing the mecA gene.
agr typing.
There was no overall difference in the prevalence of agr types between the strains from the discordant and concordant groups (table 1). In both groups, agr type 1 was the most prevalent. Of the 7 discordant samples with a >3-band difference, 3 pairs belonged to different agr types, and 4 belonged to the same agr type. Of the strains with 3-band difference, the agr type was the same in all instances.
Background characteristics of the subjects.
Demographic and medical characteristics were compared between the subjects with discordant samples and the subjects with concordant samples (table 2). None of the differences between the concordant and discordant groups were significant. The results were virtually identical when the subgroups with MRSA isolates were compared (data not shown).
Mathematical model to estimate the frequency of discordance.
The 121 S. aureuspositive baseline samples from the MMP study population were used to develop our model. Only those isolates from the same subject that differed by >3 bands and also belonged to different spa or MLST types were considered to be discordant (n = 4). For this analysis, the remaining isolates that differed by 3 bands were considered to be concordant. Therefore, there were 4 discordant and 117 concordant samples at baseline. When the frozen aliquots taken from the original plates of the 4 discordant samples were independently retested, 2 were again discordant, and the other 2 were concordant. This latter information is expressed as P(sample concordant/originally discordant in the population) 2/4, or P(SC/OD).
Eighteen concordant samples were randomly sampled from the group of 117 baseline-concordant samples. When these samples were again screened, 16 remained concordant, and 2 failed to grow. If = P(originally discordant in the population), or P(OD), then 1 - = P(originally concordant in the population), or P(OC).
Therefore,
The point estimate of discordance in the cohort is therefore = 6.6%. In summary, the model suggests that, at 6.6%, the frequency of discordance among S. aureuscolonized individuals is much higher than expected.
DISCUSSION
The present study has investigated the clonality of S. aureus nasal carriage. The results demonstrated surprising strain diversity among subjects colonized with S. aureus. On the basis of our model, 6.6% of S. aureuscolonized individuals would be expected to carry >1 strain. Data from several sequenced S. aureus genomes have already demonstrated extensive horizontal transfer of genetic information among bacterial species [8, 29]. What has been lacking to date is the identification of an environmental location at which this genetic exchange occurs. The present study, by demonstrating the coexistence of multiple strains of S. aureus in the anterior naresthe primary reservoir for persistent colonization in humanshas identified the likely site for the horizontal exchange of genetic information among strains of the same species. Our finding therefore provides a further explanation for the unique adaptability and continued pathogenicity of S. aureus.
In the present study, 2 distinct groups of discordant samples were identified. The first consisted of 7 samples that contained isolates that differed by >3 bands by PFGE and that also belonged to distinct MLST and spa types. This group reflected co-colonization with different strains and formed the basis of our mathematical model to estimate the frequency of discordant colonization with S. aureus. In contrast, the second group consisted of samples that contained minor genetic variations.
In addition to its biological relevance, this observation on strain diversity has clinical and epidemiological importance. Earlier studies discussing persistent colonization with 1 strain or the replacement of colonizing strains with another may reflect a sampling error, with less abundant co-colonizing strains having been missed. As a result, the replacement of resident strains may be less common than has been suggested. Depending on the results of subsequent studies, physicians may need to be cautious when selecting prophylactic therapy if their selection is based on the antibiotic susceptibility of a single colonizing strain.
It is unclear whether polyclonal S. aureus infections occur; however, with coagulase-negative staphylococci, bacteremias are occasionally polyclonal [10, 30]. This would clearly affect the selection of antimicrobial therapy. Although a recent investigation by Khatib et al. [31] has suggested that polyclonal bacteremia does not occur with S. aureus, given our findings, additional studies of infections at different tissue sites are warranted.
The second group of discordant samples reflected minor genetic variations of clonal types. Two samples from this group contained MRSA-MSSA pairs that differed by the presence or absence of a band containing the mecA gene. These samples illustrate the genetic mobility of SCCmec type IV, which has been found in a wider array of genetic backgrounds than have the other SCCmec types. The small size of SCCmec type IV (<25 kb) and the presence of 2 functional recombinase (ccr) genes facilitate the movement and site-specific integration of this element into different genetic backgrounds [29]. The genotype for several of the MRSA clonal strains has been identified in MSSA strains [32, 33]. Several authors have now reported clinical settings in which MRSA/MSSA homologues have been described, either in individual patients or in outbreak settings [3437]. Our study is the first to identify the simultaneous presence of these MRSA/MSSA pairs in the same subject and suggests that the process of acquisition or deletion of these genes from either other S. aureus strains or other species colonizing the anterior nares is a common occurrence. The SCCmec type IV isolates found in the present study also illustrate the diversity of this element, in that the genetic backgrounds contained highly antimicrobial-resistant profiles that were typical of hospital isolates as well as the more-susceptible profiles reported here as community MRSA types [28].
More than 50% of the 14 discordant samples contained at least 1 MRSA strain. This was in marked contrast to the percentage of MRSA strains in the overall cohort that gave rise to the discordant samples. All SCCmec types in the discordant samples were type IV, and the majority of these strains belonged to spa type 7, also known as the "archaic" or "type V" clone. It is noteworthy that all of the subjects infected with a spa type 7 strain were also infected with HIV. This clone, which has been described in immunocompromised subjects in a prevalence survey of New York City hospitals, appears to have an unexplained association with carriage and infection in HIV-infected individuals [38].
There was considerable diversity among agr types in the subgroup of discordant isolates that differed by >3 bands. Consistent with the results of previous studies, the MRSA isolates all belonged to agr type 1 [39]. Of the 7 discordant samples that contained different strains, 3 of the strains belonged to different agr types. The simultaneous presence of 2 strains belonging to different agr types argues against the hypothesis that the different autoinducing peptides secreted by the different agr types interferes with the ability of strains belonging to different agr types to coexist [40]. Our results are in agreement with those of Kahl et al. [41], who reported that agr-dependent bacterial interference had no effect on long-term S. aureus colonization in patients with cystic fibrosis. Alternatively, the discordance found in the present study might reflect a transitional state of colonization. By stimulating the expression of secreted proteins involved in tissue invasion and inhibiting the expression of surface proteins (most of which are adhesins involved in the attachment of bacteria to host cells), agr may foster the switching of colonizing strains [42]. Therefore, if one agr type inhibits a second agr type's response, the result would be an agr-negative phenotype of the second group, which would lead to an increase in the expression of surface proteins and consequently enhance colonization by the second type.
To our knowledge, the present study is the first to develop a mathematical model to estimate the frequency of pathogen discordance. Moreover, to account for the possibility of sampling error in our study design, the specimens were resampled, to determine the reproducibility of our initial observations. Because the study was performed in a unique cohortcurrent and former drug users and hospitalized S. aureuscolonized patientsthe possibility exists that the results may not be reflective of a more general population. Nevertheless, our results suggest that 6.6% of S. aureuscolonized individuals carry >1 strain. The model has the potential to define diversity in other settings in which an estimate of the clonality of carriage is important and sample size may be limited.
Understanding the clonality of S. aureus carriage has the potential to contribute to our understanding of the biological nature of nasal colonization. Greater strain diversity enhances the opportunity for the horizontal transfer of genetic material, including virulence determinants and antibiotic-resistance genes. It is unclear whether individuals colonized with multiple strains are more or less susceptible to prolonged carriage, whether the state is transient or prolonged, and whether it reflects a predisposition to carriage or not. Discordance also creates greater clinical uncertainty about the predictive value of nasal cultures for the empiric treatment of clinical infections. The present study establishes an estimate for the prevalence of the discordance of S. aureus carriage and identifies a host tissue site that fulfills the criteria of allowing the exchange of genetic information that likely occurs among co-colonizing strains.
References
1. Maslow JN, Brecher S, Gunn J, Durbin A, Barlow MA, Arbeit RD. Variation and persistence of methicillin-resistant Staphylococcus aureus strains among individual patients over extended periods of time. Eur J Clin Microbiol Infect Dis 1995; 14:28290. First citation in article
2. Kluytmans J, van Belkum A, Verbrugh H. Nasal carriage of Staphylococcus aureus: epidemiology, underlying mechanisms, and associated risks. Clin Microbiol Rev 1997; 10:50520. First citation in article
3. VandenBergh MF, Yzerman EP, van Belkum A, Boelens HA, Sijmons M, Verbrugh HA. Follow-up of Staphylococcus aureus nasal carriage after 8 years: redefining the persistent carrier state. J Clin Microbiol 1999; 37:313340. First citation in article
4. Williams RE, Jevons MP, Shooter RA, et al. Nasal staphylococci and sepsis in hospital patients. BMJ 1959; 5153:65862. First citation in article
5. von Eiff C, Becker K, Machka K, Stammer H, Peters G. Nasal carriage as a source of Staphylococcus aureus bacteremia. N Engl J Med 2001; 344:116. First citation in article
6. Peacock SJ, de Silva I, Lowy FD. What determines nasal carriage of Staphylococcus aureus Trends Microbiol 2001; 9:60510. First citation in article
7. St Sauver J, Marrs CF, Foxman B, Somsel P, Madera R, Gilsdorf JR. Risk factors for otitis media and carriage of multiple strains of Haemophilus influenzae and Streptococcus pneumoniae. Emerg Infect Dis 2000; 6:62230. First citation in article
8. Kuroda M, Ohta T, Uchiyama I, et al. Whole genome sequencing of meticillin-resistant Staphylococcus aureus. Lancet 2001; 357:122540. First citation in article
9. Van Wijngaerden E, Peetermans WE, Van Lierde S, Van Eldere J. Polyclonal staphylococcus endocarditis. Clin Infect Dis 1997; 25:6971. First citation in article
10. Van Eldere J, Peetermans WE, Struelens M, Deplano A, Bobbaers H. Polyclonal staphylococcal endocarditis caused by genetic variability. Clin Infect Dis 2000; 31:2430. First citation in article
11. Miller M, Cespedes C, Vavagiakis P, Klein RS, Lowy FD. Staphylococcus aureus colonization in a community sample of HIV-infected and HIV-uninfected drug users. Eur J Clin Microbiol Infect Dis 2003; 22:4639. First citation in article
12. Schoenbaum EE, Hartel D, Selwyn PA, et al. Risk factors for human immunodeficiency virus infection in intravenous drug users. N Engl J Med 1989; 321:8749. First citation in article
13. NCCLS. Performance standards for antimicrobial disk susceptibility tests [NCCLS document M2-A6]. Wayne, PA; NCCLS, 1997. First citation in article
14. Cespedes C, Miller M, Quagliarello B, Vavagiakis P, Klein RS, Lowy FD. Differences between Staphylococcus aureus isolates from medical and nonmedical hospital personnel. J Clin Microbiol 2002; 40:25947. First citation in article
15. Quagliarello B, Cespedes C, Miller M, et al. Strains of Staphylococcus aureus obtained from drug-use networks are closely linked. Clin Infect Dis 2002; 35:6717. First citation in article
16. Shopsin B, Gomez M, Montgomery SO, et al. Evaluation of protein A gene polymorphic region DNA sequencing for typing of Staphylococcus aureus strains. J Clin Microbiol 1999; 37:355663. First citation in article
17. Enright MC, Day NP, Davies CE, Peacock SJ, Spratt BG. Multilocus sequence typing for characterization of methicillin-resistant and methicillin-susceptible clones of Staphylococcus aureus. J Clin Microbiol 2000; 38:100815. First citation in article
18. Crisostomo MI, Westh H, Tomasz A, Chung M, Oliveira DC, de Lencastre H. The evolution of methicillin resistance in Staphylococcus aureus: similarity of genetic backgrounds in historically early methicillin-susceptible and -resistant isolates and contemporary epidemic clones. Proc Natl Acad Sci USA 2001; 98:986570. First citation in article
19. Oliveira DC, de Lencastre H. Multiplex PCR strategy for rapid identification of structural types and variants of the mec element in methicillin-resistant Staphylococcus aureus. Antimicrob Agents Chemother 2002; 46:552161. First citation in article
20. Okuma K, Iwakawa K, Turnidge JD, et al. Dissemination of new methicillin-resistant Staphylococcus aureus clones in the community. J Clin Microbiol 2002; 40:428994. First citation in article
21. Sambrook J, Fritsch EF, Maniatis T. Molecular cloning: a laboratory manual. 2nd ed. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press, 1989. First citation in article
22. Jarraud S, Mougel C, Thioulouse J, et al. Relationships between Staphylococcus aureus genetic background, virulence factors, agr groups (alleles), and human disease. Infect Immun 2002; 70:63141. First citation in article
23. Shopsin B, Mathema B, Alcabes P, et al. Prevalence of agr specificity groups among Staphylococcus aureus strains colonizing children and their guardians. J Clin Microbiol 2003; 41:4569. First citation in article
24. Feller W. An introduction to probability theory and its applications. Vol. 1. New York: John Wiley and Sons, 1968. First citation in article
25. Antibiotic susceptibility patterns 2000. New York, NY: Department of Pathology, Montefiore Medical Center, 2000. First citation in article
26. Herold BC, Immergluck LC, Maranan MC, et al. Community-acquired methicillin-resistant Staphylococcus aureus in children with no identified predisposing risk. JAMA 1998; 279:5938. First citation in article
27. Hiramatsu K, Cui L, Kuroda M, Ito T. The emergence and evolution of methicillin-resistant Staphylococcus aureus. Trends Microbiol 2001; 9:48693. First citation in article
28. Said-Salim B, Mathema B, Kreiswirth BN. Community-acquired methicillin-resistant Staphylococcus aureus: an emerging pathogen. Infect Control Hosp Epidemiol 2003; 24:4515. First citation in article
29. Baba T, Takeuchi F, Kuroda M, et al. Genome and virulence determinants of high virulence community-acquired MRSA. Lancet 2002; 359:181927. First citation in article
30. Sharma M, Riederer K, Johnson LB, Khatib R. Molecular analysis of coagulase-negative Staphylococcus isolates from blood cultures: prevalence of genotypic variation and polyclonal bacteremia. Clin Infect Dis 2001; 33:131723. First citation in article
31. Khatib R, Sharma M, Naqvi SA, Riederer K, Almoujahed MO, Fakih MG. Molecular analysis of Staphylococcus aureus blood isolates shows lack of polyclonal bacteremia. J Clin Microbiol 2003; 41:17179. First citation in article
32. Dominguez MA, de Lencastre H, Linares J, Tomasz A. Spread and maintenance of a dominant methicillin-resistant Staphylococcus aureus (MRSA) clone during an outbreak of MRSA disease in a Spanish hospital. J Clin Microbiol 1994; 32:20817. First citation in article
33. Shopsin B, Mathema B, Zhao X, Martinez J, Kornblum J, Kreiswirth BN. Resistance rather than virulence selects for the clonal spread of methicillin-resistant Staphylococcus aureus: implications for MRSA transmission. Microb Drug Resist 2000; 6:23944. First citation in article
34. Inglis B, el-Adhami W, Stewart PR. Methicillin-sensitive and -resistant homologues of Staphylococcus aureus occur together among clinical isolates. J Infect Dis 1993; 167:3238. First citation in article
35. Donnio PY, Louvet L, Preney L, Nicolas D, Avril JL, Desbordes L. Nine-year surveillance of methicillin-resistant Staphylococcus aureus in a hospital suggests instability of mecA DNA region in an epidemic strain. J Clin Microbiol 2002; 40:104852. First citation in article
36. Wielders CL, Vriens MR, Brisse S, et al. In-vivo transfer of mecA DNA to Staphylococcus aureus . Lancet 2001; 357:16745. First citation in article
37. Mongkolrattanothai K, Boyle S, Kahana MD, Daum RS. Severe Staphylococcus aureus infections caused by clonally related community-acquired methicillin-susceptible and methicillin-resistant isolates. Clin Infect Dis 2003; 37:10508. First citation in article
38. Roberts RB, de Lencastre A, Eisner W, et al. Molecular epidemiology of methicillin-resistant Staphylococcus aureus in 12 New York hospitals. MRSA Collaborative Study Group. J Infect Dis 1998; 178:16471. First citation in article
39. van Leeuwen W, van Nieuwenhuizen W, Gijzen C, Verbrugh H, van Belkum A. Population studies of methicillin-resistant and -sensitive Staphylococcus aureus strains reveal a lack of variability in the agrD gene, encoding a staphylococcal autoinducer peptide. J Bacteriol 2000; 182:57219. First citation in article
40. Ji G, Beavis R, Novick RP. Bacterial interference caused by autoinducing peptide variants. Science 1997; 276:202730. First citation in article
41. Kahl BC, Becker K, Friedrich AW, et al. agr-Dependent bacterial interference has no impact on long-term colonization of Staphylococcus aureus during persistent airway infection of cystic fibrosis patients. J Clin Microbiol 2003; 41:5199201. First citation in article
42. Novick RP, Muir TW. Virulence gene regulation by peptides in staphylococci and other gram-positive bacteria. Curr Opin Microbiol 1999; 2:405. First citation in article