| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 ,2 ,3 Servei de Microbiologia1 , Servei de Medicina Interna2 and Unitat de Malalties Infeccioses3 , Hospital de la Santa Creu i Sant Pau, Avda Sant Antoni Ma Claret 167, 08025 Barcelona, Spain
4 Unitat de Microbiologia, Departament de Genètica i Microbiologia, Universitat Autònoma de Barcelona, 08193 Cerdanyola del Vallès (Bellaterra), Spain
Correspondence
Ferran Navarro
fnavarror{at}santpau.es
Received 21 December 2005
Accepted 18 April 2006
Abbreviations: GAS, group A streptococcus; NF, necrotizing fasciitis; SAg, superantigen; STSS, streptococcal toxic shock syndrome.
The GenBank/EMBL/DDBJ accession numbers for emm1.25, emm12.27, emm83.4, smeZ-35, smeZ-36, smeZ-37, smeZ-38 and smeZ-39 are AY686603, AY742805, AY742806, AY965265, AY965878, AY965879, DQ001538 and DQ001539, respectively.
 |
INTRODUCTION |
|---|
 TOP INTRODUCTION METHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
The resurgence and persistence of severe forms of GAS diseases reported since the mid 1980s (Hoge et al., 1993; Stevens, 2002) has motivated intensive research on epidemiological, microbiological and clinical aspects of invasive GAS disease. Several factors have been considered to explain differences in disease frequency and severity, including changes in the virulence of the bacterium (Banks et al., 2002; Musser et al., 1993) and the role of host immunity (Åkesson et al., 2004; Basma et al., 1999; Kotb et al., 2002; Norrby-Teglund et al., 2000).
Among the many factors involved in the virulence of the pathogen, the M protein and a group of exotoxins known as streptococcal superantigens (SAgs) have received considerable attention. Sequence analysis of the emm gene (encoding the M protein) has become an important surveillance tool for investigating the dynamics of GAS infection and more than 150 emm gene sequence types and several emm subtypes have been documented (Facklam et al., 2002; Li et al., 2003).
SAgs are thought to contribute to the pathogenesis of severe GAS infections by virtue of their potent immunostimulatory activity (Norrby-Teglund et al., 2001). The gene distribution of SAgs has been used as an additional epidemiological tool to explore genomic heterogeneity and the possible correlation between toxin gene content and disease type (Chatellier et al., 2000; Schmitz et al., 2003; Vlaminckx et al., 2003).
To improve our understanding of severe GAS disease in Europe, the Strep-EURO project was recently set up (Lamagni et al., 2005; http://www.strep-euro.lu.se). To our knowledge, no studies on GAS epidemiology and SAg distribution, including invasive and non-invasive GAS isolates, have been conducted to date in Spain. Surveillance that includes invasive and non-invasive GAS isolates is important to evaluate epidemiological changes in GAS diseases and to distinguish between virulence properties and the prevalence of a particular GAS strain in the general population (Johnson et al., 2002; O'Brien et al., 2002).
In this study, we report clinical and epidemiological data on invasive and non-invasive GAS infections from a tertiary care hospital in Barcelona. emm subtype distribution, SAg gene profiles (speA–C, speF–J, speL, speM, ssa and smeZ) and allelic variants of speA and smeZ genes were used to evaluate possible differences between GAS isolates causing invasive or non-invasive infections. The prevalence and mechanisms of macrolide, tetracycline and fluoroquinolone resistance were also determined.
|
METHODS |
|---|
|
TOPINTRODUCTIONMETHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
Invasive infection was defined as the recovery of GAS from sites that are usually sterile, including blood, cerebrospinal, peritoneal, pleural and joint fluids, deep tissue abscesses and a superficial site in clinical association with STSS or NF. STSS was defined according to the consensus definition of the Working Group on Severe Streptococcal Infections (1993). Non-invasive isolates were recovered from sites that are usually non-sterile (skin, wound, ear, genital, throat and other miscellaneous sites) and were included for genetic comparison with invasive isolates. Demographic and clinical data on patient characteristics, course of infection and outcome were recorded from the medical history. Medical records were also examined to assess underlying diseases or conditions that may have predisposed patients to invasive disease, such as skin trauma, injection drug use, varicella infection, alcohol abuse or decreased body defences (malignancy, HIV infection, diabetes and immunosuppressive therapy). The overall fatality rate was assessed 30 days after the date of specimen sampling.
Antimicrobial susceptibility testing. The susceptibility to penicillin G, erythromycin, clindamycin, tetracycline and levofloxacin was determined by using the disc-diffusion method, according to Clinical and Laboratory Standards Institute guidelines (National Committee for Clinical Laboratory Standards, 2003) using commercial discs (Bio-Rad). Macrolide resistance phenotypes [constitutive (cMLSb), inducible (iMLSb) and M phenotype] were determined by using the double-disc test, as described by Seppälä et al. (1993). MICs of penicillin G, cefotaxime and levofloxacin were determined for all isolates by using Etest (ABBiodisk), according to the manufacturer's recommendations.
emm typing. The emm gene was amplified and sequenced as described previously (Rivera et al., 2005) and by Beall et al. (1996). emm type and subtype assignments were determined as described on the Centers for Disease Control and Prevention (CDC) website (http://www.cdc.gov/ncidod/biotech/strep/doc.htm).
Detection of toxin genes.
GAS isolates were tested for the presence of speA, speC, speH, speI, speJ, speL, speM, ssa and smeZ by PCR, using the primers listed in Table 1
. With the exception of speL, speM and smeZ, two different primer pairs were used. PCRs were performed using the following conditions: denaturing for 5 min at 94 °C, 30 cycles for 30 s at 94 °C, annealing for 30 s at the temperature determined for each primer pair (Table 1) and elongation for 1 min at 72 °C, followed by a final elongation step at 72 °C for 10 min.
|
The presence of speB, speF and speG was determined for all isolates by slot-blot analysis (Schleicher & Schuell). DNA samples were obtained as described by Pan & Fisher (1996). DNA probes used in hybridization assays were obtained by PCR with the same conditions as described above using the primers for speB, speF and speG given in Table 1
. Labelling of probes and hybridization and detection were performed using ECL Direct Nucleic Acid Labelling (Amersham Biosciences). Hybridization results were confirmed by PCR in 38 randomly selected isolates. Previously published GAS strains, kindly provided by Dr T. Proft, were used as positive controls for the detection of toxin genes, including FP 4223 (positive for speG, speJ, speM and smeZ), FP 5417 (positive for speA, speG, speJ, speL and smeZ), FP 5971 (positive for speA, speC, speG, speH, speJ and smeZ), 1/5045 (positive for speA, speC, speG, speJ, ssa and smeZ) and 85/167 (positive for speG, speH, speI, speJ and smeZ) (Proft et al., 2003).
Identification of resistance genes.
Resistance determinants erm(A) subclass erm(TR) [hereafter designated erm(TR)], erm(B) and mef(A) were detected in erythromycin-resistant isolates by PCR with the primers listed in Table 1, using the same conditions as described for the toxin genes, except for mef(A) (denaturing for 5 min at 94 °C, 30 cycles for 1 min at 94 °C, annealing for 1 min at 60 °C and elongation for 1 min 30 s at 72 °C, followed by 10 min at 72 °C). In tetracycline-resistant isolates, tet(M) and tet(O) were also detected by PCR using primers described by Ng et al. (2001). Characterization of mutational changes in parC, parE, gyrA and gyrB were determined previously in isolates with a levofloxacin MIC of 
2 µg ml–1 (Rivera et al., 2005).
Statistical analysis. Chi square or Fisher's exact test was used for statistical analysis. A P value of <0.05 was considered significant. The SPSS 11.5 statistical package was used for all analyses.
|
RESULTS AND DISCUSSION |
|---|
|
TOPINTRODUCTIONMETHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
The median age of patients with invasive infections was 55 years (mean 51.2 years; range 10 months to 89 years), whereas in the non-invasive group the median age was 8 years (mean 22.9 years; range 1 month to 91 years). The male to female ratio was 1.5 : 1 and 1 : 1 for the invasive and non-invasive groups, respectively. At least one comorbid disease or local predisposing factor was found in 20 (80 %) out of 25 patients with available data in the invasive group. In one of the paediatric patients, invasive infection was associated with varicella, whereas, among adult patients, underlying conditions included skin trauma (eight cases), intravenous drug use and HIV infection (four cases, two with STSS and NF), cancer (four cases), peripheral vascular disease (three cases), immunosuppressive therapy (two cases), alcohol abuse (two cases) and diabetes mellitus (one case). Mortality occurred in three of 27 (11 %) patients with invasive infections, two of whom had STSS. Therefore, in agreement with previously published reports (Kao et al., 2005; O'Brien et al., 2002), most patients with invasive GAS infection were adults and a large proportion had underlying conditions, with HIV infection and injection drug use among the most frequent. Invasive GAS infection associated with injection drug use appears to be increasing in some countries (Efstratiou et al., 2003) and epidemics among this group have been documented (Léchot et al., 2001). It should be noted that, during the period studied, a local outbreak of invasive disease was detected among injecting drug users (Sierra et al., 2006). In this work, all invasive emm25.2 isolates were isolated from injecting drug users and were part of the local outbreak. Interestingly, an outbreak of invasive infection among drug users in Switzerland was reportedly caused by a specific emm25 clone (Bohlen et al., 2000).
Distribution of emm types
In total, 29 emm types were identified among the 126 isolates (Table 2), of which emm1 (22 isolates, 17.5 %) was the most prevalent, followed by emm3 (11 isolates, 8.7 %), emm4 (11 isolates, 8.7 %), emm12 (nine isolates, 7.1 %), emm28 (nine isolates, 7.1 %), emm11 (eight isolates, 6.3 %) and emm77 (eight isolates, 6.3 %). These seven emm types accounted for 61.9 % of isolates and have been reported to be among the most prevalent in USA, Canada and Europe (O'Brien et al., 2002; Schmitz et al., 2003; Tyrrell et al., 2002; Vlaminckx et al., 2003).
|
A total of 15 different emm types were identified in the 27 invasive isolates. All but two represented by single isolates were also present among non-invasive isolates. The most prevalent emm types were emm3 (14.8 %), emm4 (11.1 %), emm25 (11.1 %), emm77 (11.1 %), emm1 (7.4 %), emm6 (7.4 %) and emm9 (7.4 %). A diversity of emm types was detected in isolates from the six patients with the most severe forms of infection (NF and SSTS), including two emm25 and one each of emm1, emm4, emm11 and emm22. In the 99 non-invasive isolates, 27 different emm types were found, the most prevalent being emm1 (20.2 %), emm4 (8.1 %), emm12 (8.1 %), emm28 (8.1 %), emm3 (7.1 %) and emm11 (7.1 %).
No significant differences in emm type distribution were found between invasive and non-invasive isolates, as has been described recently (Descheemaeker et al., 2000; Haukness et al., 2002; Ho et al., 2003; Johnson et al., 2002; Kao et al., 2005). However, earlier reports identified a shift in dominant serotypes coincident with an increase in severe GAS infections, with M1 and M3 being those most frequently isolated from invasive cases (Johnson et al., 1992; Schwartz et al., 1990). The emergence of clones with increased virulence was proposed to explain the observed changes in GAS epidemiology (Musser et al., 1993, 1995).
Over the study period, there were no annual variations in the prevalence of the most common emm types. Certain subtypes such as emm11.0, emm25.2 and the newly identified emm1.25 were only found in years 2002 and 2003.
Distribution of SAgs and toxin gene profiles
Among the toxin genes that are thought to be chromosomally encoded, speB and speF were found in all isolates. As shown in Table 2, speG was found in 106 (84.1 %) isolates and speJ was only found in 42 (33.3 %) isolates. smeZ was identified in 88 (91.7 %) of 96 isolates tested. DNA sequence analysis revealed the presence of 23 different smeZ alleles. Five smeZ alleles designated smeZ-35–smeZ-39 were identified for the first time in this study. As described by Proft et al. (2000), there was a strong association between the smeZ allele and the emm type. Five smeZ alleles (smeZ-3, smeZ-4, smeZ-13, smeZ-17 and smeZ-37) were each found in isolates of two different emm types. Isolates within an emm type carried the same smeZ allele except for the emm50 and emm4 isolates (Table 2). The genes speA, speC, speH, speI, speL, speM and ssa, which have been found to be associated with prophage elements, were present at variable frequencies (Table 2). Significant differences were only found for ssa, which was more frequent in isolates from invasive cases (P<0.05, Odds' ratio 2.5; 95 % confidence interval, 1.27–4.91). This correlation probably reflects the association of the ssa gene with the most prevalent emm types found in invasive infections (emm3, emm25 and emm4). Sequence analysis of the speA gene revealed four different alleles (Nelson et al., 1991), with a limited distribution of speA2, speA3 and speA4 in emm1, emm3 and emm6 isolates, respectively. The speA1 allele was identified in single isolates of three distinct emm types (emm18, emm4 and emm49).
The analysis revealed that isolates of the same emm type shared a common toxin gene profile (emm types 1, 3, 6, 9, 59, 64, 87 and 89) or a predominant numerical profile (Table 2). Differences in toxin gene profiles between isolates of the same emm type involved the presence or absence of only one or two toxin genes compared with the dominant profile, except for the emm4.2 isolate, which differed in four toxin genes from emm4.0 isolates.
Although the number of isolates studied was limited, no significant difference in toxin gene profile was found between invasive and non-invasive isolates. Two recent European studies involving a larger number of isolates found that, despite the presence of several different toxin gene profiles within an emm type, each emm type was characterized by a predominance of one or two toxin gene profiles, reflecting the spread of few invasive clones throughout the world (Schmitz et al., 2003; Vlaminckx et al., 2003).
Antimicrobial susceptibility
All isolates were susceptible to penicillin G and cefotaxime. Erythromycin resistance was found in 35 isolates (27.8 %), tetracycline resistance in 32 isolates (25.4 %) and resistance to both agents in 15 isolates (11.9 %). Four isolates (3.2 %) showed reduced susceptibility to levofloxacin (MIC 2–3 µg ml–1).
Seventeen (48.6 %) of the erythromycin-resistant isolates expressed the M phenotype and all of them harboured mef(A). The cMLSb phenotype was observed in 15 (42.8 %) isolates, in association with erm(B) in all but one, which carried erm(TR). Three (8.6 %) isolates presented iMLSb; two of these carried erm(B) and the third carried erm(TR). The most prevalent emm types of erythromycin-resistant isolates were emm4, emm11, emm12 (four subtypes), emm25 and emm75. Together, these five emm types represented 80 % of the erythromycin-resistant isolates, but only 7.7 % of the susceptible isolates. All the emm4, emm12 and emm75 erythromycin-resistant isolates harboured the mef(A) gene, whereas all emm11 and emm25 isolates carried the erm(B) gene. These data suggested that the high rate of erythromycin resistance was caused by the spread of a limited number of clones. An association between certain emm types and erythromycin resistance or macrolide resistance mechanisms has been documented (Albertí et al., 2003; Zampaloni et al., 2003) and may possibly reflect the spread of resistant clones. In Spain, a strong association has been reported between emm types 4 and 75 and erythromycin resistance (Albertí et al., 2003; Pérez-Trallero et al., 1999).
A total of 32 tetracycline-resistant isolates were detected, belonging to 14 different emm types. The resistance determinant tet(M) was present in 18 isolates (56.2 %), tet(O) in six isolates (18.7 %), and one isolate carried both genes. None of the remaining seven tetracycline-resistant isolates carried either tet(M) or tet(O). tet(O) was found exclusively in emm77.0 isolates.
Among the 15 isolates that were resistant to both erythromycin and tetracycline, erm(B) and tet(M) were found in six isolates, erm(TR) and tet(O) in one isolate and erm(TR) and tet(M) in one isolate. Seven isolates, all emm11, carried erm(B), but none of the tetracycline resistance genes studied. None of the strains presenting the M phenotype were resistant to tetracycline.
The frequency of erythromycin-resistant isolates showed a rising trend over the study period (16.6 % in 1999, 19 % in 2000, 21 % in 2001, 34.6 % in 2002 and 38.8 % from January to June 2003). A high frequency of erythromycin resistance has been found in Spain (29.7 %) (Alós et al., 2003) and other European countries, including Italy (35.8 %; Dicuonzo et al., 2002) and France (22.4 %; Bingen et al., 2004), in contrast with the low rates observed in other countries such as the USA (6.8 %; Richter et al., 2005). The frequency of tetracycline resistance fluctuated year-to-year (33.3 % in 1999, 19 % in 2000, 15.8 % in 2001, 34.6 % in 2002 and 24.2 % from January to June 2003). Multiclonal dissemination and high rates of tetracycline resistance have been reported from other countries (de Melo et al., 2003; Jasir et al., 2000).
The proportions of resistant isolates were similar in the invasive and non-invasive groups for both erythromycin (22.2 % invasive versus 29.3 % non-invasive) and tetracycline (33.3 % invasive versus 23.2 % non-invasive).
A number of reports during the last few years have described an increased prevalence of isolates with reduced susceptibility to fluoroquinolones (ciprofloxacin MIC 2 µg ml–1) (Albertí et al., 2005; Malhotra-Kumar et al., 2005; Powis et al., 2005). In the present study, four isolates of three different emm types (emm3.1, emm6.0 and emm12.27) showed reduced susceptibility to levofloxacin (MIC 2–3 µg ml–1) and carried mutational changes in parC. A detailed characterization of these isolates has been published previously (Rivera et al., 2005).
In conclusion, the overall emm-type distribution found in this study was comparable with previous reports from North America and Europe. We found no differences in emm type distribution among isolates causing invasive or non-invasive infection. In addition, isolates of the same emm type presented restricted genetic variation with respect to SAg profiles. The erythromycin resistance rate presented a rising trend over the period studied and involved the spread of a limited number of clones. The SAg and antibiotic resistance genes thus appeared to be associated with the emm type and were independent of the disease type.
|
ACKNOWLEDGEMENTS |
|---|
|
REFERENCES |
|---|
|
TOPINTRODUCTIONMETHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  A. Friaes, M. Ramirez, J. Melo-Cristino, and and the Portuguese Group for the Study of Streptoc Nonoutbreak Surveillance of Group A Streptococci Causing Invasive Disease in Portugal Identified Internationally Disseminated Clones among Members of a Genetically Heterogeneous Population J. Clin. Microbiol., June 1, 2007; 45(6): 2044 - 2047. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| INT J SYST EVOL MICROBIOL | J MED MICROBIOL | MICROBIOLOGY | J GEN VIROL | ALL SGM JOURNALS |
