J Med Microbiol 58 (2009), 222-227; DOI: 10.1099/jmm.0.001560-0
© 2009 Society for General Microbiology
ISSN 0022-2615
Prevalence of erythromycin and clindamycin resistance among clinical isolates of the Streptococcus anginosus group in Germany
Nadine Asmah1,
Bettina Eberspächer2,
Thomas Regnath3 and
Mardjan Arvand4
1 Hygiene Institut, Universität Heidelberg, Heidelberg, Germany
2 Institut für Laboratoriumsdiagnostik, Abt. Mikrobiologie, Vivantes Kliniken, Berlin, Germany
3 Labor Enders und Partner, Medizinisch-diagnostisches Gemeinschaftslabor, Stuttgart, Germany
4 Institut für Medizinische Mikrobiologie, Virologie und Hygiene, Universität Rostock, 18057 Rostock, Germany
Correspondence
Mardjan Arvand
marvand{at}freenet.de
Received February 28, 2008
Accepted October 14, 2008
Members of the Streptococcus anginosus group (SAG) are frequently involved in pyogenic infections in humans. In the present study, the antimicrobial susceptibility of 141 clinical SAG isolates to six antimicrobial agents was analysed by agar dilution. All isolates were susceptible to penicillin, cefotaxime and vancomycin. However, 12.8 % displayed increased MIC values (0.12 mg l–1) for penicillin. Resistance to erythromycin was detected in eight (5.7 %) isolates. Characterization of the erythromycin-resistant isolates with the double-disc diffusion test revealed Macrolide-Lincosamide-StreptograminB and M-type resistance in six and two isolates, respectively. The erythromycin-resistant isolates were further characterized by PCR for the resistance genes ermA, ermB and mefA. Resistance and intermediate resistance to ciprofloxacin were detected in two and six isolates, respectively. Molecular typing by PFGE revealed a high genetic heterogeneity among the SAG isolates and no evidence for a clonal relationship between the erythromycin-resistant isolates. Our data show that resistance to erythromycin, clindamycin and ciprofloxacin has emerged among SAG isolates in Germany. The implications of these findings for susceptibility testing and antimicrobial therapy of SAG infections are discussed.
Abbreviations: MLSB, Macrolide-Lincosamide-StreptograminB; SAG, Streptococcus anginosus group; UPGMA, unweighted pair group method with arithmetic averages.
 |
INTRODUCTION
|
|---|
The Streptococcus anginosus group (SAG), synonym Streptococcus milleri group, is composed of the species S. anginosus, Streptococcus constellatus and Streptococcus intermedius (Facklam, 2002). SAG isolates are part of the normal flora of the human respiratory, gastrointestinal and genitourinary tracts. However, the organisms can also cause serious invasive infections, including infections of the liver and lung, brain abscesses, bacteraemia, endocarditis and intra-abdominal infections (Clarridge et al., 2001; Rashid et al., 2007). In a recent study, members of the SAG were identified as the most frequent cause of invasive pyogenic streptococcal infections in Canada (Laupland et al., 2006).
SAG are generally considered to be susceptible to penicillin, other β-lactam antibiotics and macrolides, but resistant strains have been reported (Aracil et al., 1999; Limia et al., 1999; Tracy et al., 2001). Moreover, a significant increase in the antimicrobial resistance of viridans and β-haemolytic streptococci has been noticed in recent decades (Seppala et al., 2003). The present study was performed to determine the susceptibility in vitro of clinical SAG isolates from Germany to penicillin, erythromycin, clindamycin, ciprofloxacin, cefotaxime and vancomycin. Although erythromycin does not represent a first choice antibiotic for the therapy of SAG infections, it was included in this study because resistance to erythromycin may be associated with co-resistance to clindamycin and streptogramins. Macrolide-Lincosamide-StreptograminB (MLSB) resistance has been detected in different bacterial species, including Staphylococcus aureus, Streptococcus pyogenes and Streptococcus pneumoniae (Livermore, 2000; Sutcliffe et al., 1996b). The erythromycin-resistant isolates were therefore further characterized with the double-disc diffusion test and PCR to determine the MLSB resistance phenotype and to identify the erythromycin resistance genes involved. Two main mechanisms are known to account for streptococcal resistance to erythromycin. The MLSB phenotype is associated with the erm genes, which induce the methylation of a ribosomal protein resulting in reduced affinity between MLSB antibiotics and the ribosome (Seppala et al., 2003). The other resistance type, the so-called M-type resistance, is caused by the mefA gene, which induces the active efflux of erythromycin and other 14- and 15-membered macrolides (Sutcliffe et al., 1996b). All isolates were further subjected to PFGE analysis in order to evaluate the genetic heterogeneity among the SAG isolates and the clonal relationship between the erythromycin-resistant isolates.
|
METHODS
|
|---|
Bacterial isolates.
SAG isolates (141) were consecutively collected from March 2002 until March 2003 in three clinical microbiology laboratories located in different regions of Germany: Hygiene Institute, University of Heidelberg, Heidelberg (n=83), Vivantes Kliniken, Berlin (n=31), and Labor Enders und Partner, Stuttgart (n=27). The isolates were grown from clinical specimens from blood (n=8), abscess (35), abdomen (24), wounds (21), mediastinum (2), genitourinary tract (11), lower respiratory tract (7), upper respiratory tract (23) and skin (10). Taken together, 45 isolates (32 %) were associated with infections of normally sterile body sites, 45 isolates (32 %) were obtained from intra-abdominal sites or wounds and 51 isolates (36 %) were grown from body sites usually colonized with normal flora. Only one isolate from each patient was included in this study.
Identification and characterization of SAG isolates.
Identification was performed by routine laboratory methods including Gram stain, catalase reaction and biochemical analysis using the Api 20 Strep System (bioMérieux) for all isolates. Twelve isolates were additionally subjected to 16S rRNA gene sequencing for genetic identification. All isolates were further characterized by using the Streptococcal Grouping kit (Oxoid) to determine the Lancefield group antigens.
Antimicrobial susceptibility testing.
The MIC of several antibiotics was determined by using the agar dilution technique using Mueller–Hinton agar (Oxoid) with 5 % sheep blood. An inocula of 104 c.f.u. per spot was used, as described previously (Schoening et al., 2005). Plates were incubated in the presence of 5 % CO2 at 37 °C. The antimicrobial agents included in this study were penicillin G and erythromycin (Grünenthal), cefotaxime (Sigma), clindamycin (Pharmacia), ciprofloxacin (Bayer) and vancomycin (Lilly). S. pneumoniae ATCC 49619 and S. pyogenes CCUG 25570 and CCUG 25571 were used as controls. The CCUG strains have defined MIC values for erythromycin (Olsson-Liljequist et al., 1997). Clinical and Laboratory Standards Institute breakpoints (Clinical and Laboratory Standards Institute, 2008) were used for all antimicrobial agents except ciprofloxacin, for which the Deutsches Institut für Normung breakpoints (Deutsches Institut für Normung e. V, 1999) were used.
Double-disc diffusion test.
The double-disc diffusion test was performed as described previously (Schoening et al., 2005). Briefly, erythromycin (15 µg) and clindamycin (2 µg) (both from Oxoid) were placed 20 mm apart on Mueller–Hinton agar with 5 % sheep blood. Plates were incubated overnight at 37 °C in 5 % CO2 in air. Blunting of the clindamycin inhibition zone proximal to the erythromycin disc was interpreted as the inducible MLSB resistance type, resistance to both antimicrobials was considered as the constitutive MLSB resistance type and resistance to erythromycin with susceptibility to clindamycin was assigned to the M-type resistance.
16S rRNA gene sequencing and PCR detection of erythromycin resistance genes.
Eight erythromycin-resistant and four erythromycin-susceptible isolates were subjected to 16S rRNA gene sequencing for genetic identification, and to PCR assays for detection of the erythromycin resistance genes ermA, ermB and mefA. Bacterial DNA was isolated with the mini tissue kit (Qiagen). 16S rRNA amplification (PCR product of 
510 bp) and sequencing was performed by using the eubacterial primers 5'-AGA GTT TGA TCM TGG CTC AG-3', 5'-GWA TTA CCG CGG CKG CTG-3' and standard protocols. The primer and PCR conditions for amplification of ermA (PCR product of 210 bp), ermB (PCR product of 640 bp) and mefA (PCR product of 350 bp) have been described previously (Morosini et al., 2003; Sutcliffe et al., 1996a).
Macrorestriction analysis of genomic DNA by PFGE.
A fresh overnight culture of each isolate was subjected to PFGE analysis following restriction with SmaI (Roche) as described previously (Bartie et al., 2000). Electrophoresis was conducted in a CHEF-DRII chamber (Bio-Rad). The data were analysed by Gelcompar II software (Applied Maths). Cluster analysis was performed by the unweighted pair group method with arithmetic averages (UPGMA). An optimization of 1 % and a tolerance of 2 % were applied.
|
RESULTS AND DISCUSSION
|
|---|
Characteristics of the SAG isolates
All isolates were assigned by biochemical methods to the SAG. Twelve isolates were additionally subjected to 16S rRNA gene sequencing for genetic identification. The sequence data confirmed the identification of all isolates as members of the SAG (data not shown). Discrepancies at the species level were observed in five cases, which were assigned by biochemical methods as S. constellatus (n=2), S. intermedius (2) and S. anginosus (1) and by 16S rRNA gene sequencing as S. anginosus (4) and S. constellatus (1). Previous studies based on 16S rRNA sequencing have shown the limitations of biochemical testing for identification of SAG isolates to the species level, but have clearly demonstrated the accuracy of biochemical tests for the assignment of streptococci to the SAG (Clarridge et al., 1999). Our data are in accordance with the latter findings and confirm that the assignment of SAG isolates to a distinct species within the SAG by biochemical methods is not reliable.
The Lancefield group antigens C, F and G were detected in 49 (34.8 %), 45 (31.9 %) and 7 (5 %) isolates, respectively. Thirty-nine (27.7 %) isolates did not express the grouping antigens A, C, F or G.
In vitro antimicrobial susceptibility results
The in vitro susceptibility of 141 clinical SAG isolates to penicillin, cefotaxime, erythromycin, clindamycin, ciprofloxacin and vancomycin was studied by agar dilution to determine the MIC values (Table 1
). All isolates were susceptible to penicillin, cefotaxime and vancomycin (Table 2). However, 18 (12.8 %) isolates displayed a MIC of penicillin of 0.12 mg l–1, which is the susceptible breakpoint for streptococci other than S. pneumoniae. Our results are in accordance with studies from the USA and Spain, which did not detect resistance against penicillin or third generation cephalosporines among SAG isolates (Aracil et al., 1999; Tracy et al., 2001). These data indicate that penicillin can still be considered as a first-choice drug for the therapy of SAG infections in Germany. However, the increased penicillin MIC in approximately 13 % of isolates underlines the need for determination of MIC values in isolates associated with severe and systemic infections. The relevance of antimicrobial resistance to the outcome of disease has been widely discussed for β-lactams (Pallares et al., 1998). However, little is known about the impact of reduced susceptibility to penicillin in SAG isolates. It is possible that SAG isolates with decreased in vitro susceptibility to penicillin may also display a decreased susceptibility in vivo, resulting in the need for higher penicillin doses or a switch to other antimicrobial agents, as has been recommended for severe S. pneumoniae infections (File, 2006). Further studies concentrating on the clinical outcome of severe SAG infections are required to address this point.
View this table:
[in this window]
[in a new window]
|
Table 2. Range, MIC50, MIC90 and proportion of susceptibile and resistant isolates among 141 SAG isolates from Germany
All values are mg l–1.
|
|
Erythromycin resistance was detected in eight (5.7 %) isolates by agar dilution. The erythromycin MICs were confirmed by E-test in all erythromycin-resistant isolates (data not shown). The erythromycin resistance rate seen in our study is significantly higher than the resistance rate among SAG isolates from the Netherlands (2.6 %) (Jacobs & Stobberingh, 1996). It is, however, markedly lower than the resistance rate reported in Spain (17.1 %) (Limia et al., 1999). It is interesting to note that the prevalence of erythromycin resistance is markedly higher among S. pneumoniae isolates in Germany. In a recent study, the erythromycin resistance rate among invasive S. pneumoniae isolates from 2000 was 15.3 % (Reinert et al., 2002). Because of the high frequency of genetic exchange between different streptococcal species, it is likely that the increased macrolide resistance among viridans streptococci has contributed to an increased macrolide resistance among SAG isolates.
Two isolates (1.4 %) were resistant and six (4.3 %) showed intermediate resistance to ciprofloxacin, suggesting that quinolone resistance has emerged among SAG isolates in Germany. Fluoroquinolones are generally not considered as first-choice antibiotics for the treatment of streptococcal infections. However, they are frequently used for the therapy of purulent, multifactorial and intra-abdominal infections. Our data clearly show that even in ciprofloxacin-susceptible isolates, the ciprofloxacin MICs are very close to the resistance breakpoints and suggest that ciprofloxacin is inadequate for the treatment of SAG infections.
Determination of the erythromycin resistance types
The MLSB resistance phenotype of the erythromycin-resistant isolates (n=8) was analysed with the double-disc diffusion test (Arvand et al., 2000; Seppala et al., 1993). Two and four isolates displayed the constitutive and inducible MLSB resistance phenotypes, respectively. The constitutively MLSB-resistant isolates displayed MICs of 8–16 mg l–1 for clindamycin, which indicated that they were clindamycin resistant. In contrast, the isolates with inducible MLSB phenotype displayed clindamycin MICs of 0.06–0.25 mg l–1, which means that they were not recognized as being resistant to clindamycin. Two isolates displayed the M phenotype. All erythromycin-resistant isolates revealed positive PCR results for the erythromycin resistance genes ermA, ermB or mefA. Four isolates with the inducible MLSB resistance phenotype contained the ermA gene. One constitutively MLSB-resistant isolate contained ermA and the second one ermB and mefA. The two M-type isolates contained the mefA gene (data not shown).
Together, six of eight erythromycin-resistant isolates displayed MLSB resistance and should therefore be considered resistant to clindamycin and StreptograminB antibiotics. Hence, the overall frequency of clindamycin resistance was 5 % in the present study. This rate is lower than the clindamycin resistance rate (14 %) among SAG isolates in the USA (Tracy et al., 2001). Our data indicate that clindamycin-resistant SAG isolates are prevalent in Germany and underline the need for additional testing of the erythromycin-resistant isolates with the double-disc diffusion test to detect inducible clindamycin resistance.
Molecular fingerprinting of the SAG isolates by PFGE analysis
Analysis of the macrorestriction fragments of genomic DNA was performed by PFGE following restriction with SmaI. Twenty-two of 141 isolates, including two erythromycin-resistant isolates, did not reveal a sufficient number of bands, i.e. at least six bands, and were therefore considered to be untypable with SmaI. The remaining 119 isolates displayed a very high genetic heterogeneity, which is in agreement with the fact that three different species are represented among the SAG isolates. Most isolates displayed restriction patterns that differed by four or more bands from the other isolates. Cluster analysis was performed with the UPGMA method by using the dice coefficient to produce a dendrogram (Fig. 1). A few pairs of isolates displayed indistinguishable or very similar restriction patterns, suggesting a very close genetic relationship. Two erythromycin-resistant isolates (SAG 55 and 150) displayed a similarity of >90 % between their restriction patterns, which suggests that they may be clonally related. These isolates were grown from the blood and urine cultures of two different patients that were not epidemiologically related. The remaining four typable erythromycin-resistant isolates displayed different PFGE patterns and were randomly scattered among the SAG isolates. Although we did not obtain sufficient typing results for many isolates (22/141), probably because of the fastidious nature of SAG isolates, the PFGE results of the typable isolates suggest that the erythromycin-resistant isolates are genetically heterogeneous and not clonally related.
View larger version (16K):
[in this window]
[in a new window]
|
Fig. 1. Cluster analysis of PFGE patterns of 119 clinical SAG isolates, as determined with the dice coefficient and the UPGMA method. Erythromycin-resistant isolates have been marked (R).
|
|
In conclusion, this is the first study to address the antimicrobial susceptibility of SAG isolates in Germany. Although we did not find any penicillin-resistant isolates, the increased penicillin MIC of >0.1 mg l–1 in approximately 13 % of the isolates underlines the need for determination of penicillin MIC in SAG isolates associated with invasive or complicated infections. Our data show that erythromycin, clindamycin and ciprofloxacin resistance are prevalent among SAG isolates in Germany and further emphasize the requirement for additional testing of erythromycin-resistant isolates with the double-disc diffusion test in order to detect inducible MLSB-resistant isolates.
|
ACKNOWLEDGEMENTS
|
|---|
This work includes parts of the Doctoral Thesis of Nadine Asmah. We thank Andrea Koch and Constanze Wendt for taking care of the strain collection and Yvonne Humboldt for technical assistance.
|
References
|
|---|
Aracil, B., Gomez-Garces, J. L. & Alos, J. I. (1999). A study of susceptibility of 100 clinical isolates belonging to the Streptococcus milleri group to 16 cephalosporins. J Antimicrob Chemother 43, 399–402.[Abstract/Free Full Text]
Arvand, M., Hoeck, M., Hahn, H. & Wagner, J. (2000). Antimicrobial resistance in Streptococcus pyogenes isolates in Berlin. J Antimicrob Chemother 46, 621–624.[Abstract/Free Full Text]
Bartie, K. L., Wilson, M. J., Williams, D. W. & Lewis, M. A. (2000). Macrorestriction fingerprinting of "Streptococcus milleri" group bacteria by pulsed-field gel electrophoresis. J Clin Microbiol 38, 2141–2149.[Abstract/Free Full Text]
Clarridge, J. E., III, Attorri, S., Musher, D. M., Hebert, J. & Dunbar, S. (2001). Streptococcus intermedius, Streptococcus constellatus, and Streptococcus anginosus ("Streptococcus milleri group") are of different clinical importance and are not equally associated with abscess. Clin Infect Dis 32, 1511–1515.[CrossRef][Medline]
Clarridge, J. E., III, Osting, C., Jalali, M., Osborne, J. & Waddington, M. (1999). Genotypic and phenotypic characterization of "Streptococcus milleri" group isolates from a Veterans Administration hospital population. J Clin Microbiol 37, 3681–3687.[Abstract/Free Full Text]
Clinical and Laboratory Standards Institute (2008). Performance standards for antimicrobial susceptibility testing; eighteenth informational supplement M100–S13 28.
Facklam, R. (2002). What happened to the streptococci: overview of taxonomic and nomenclature changes. Clin Microbiol Rev 15, 613–630.[Abstract/Free Full Text]
File, T. M., Jr (2006). Clinical implications and treatment of multiresistant Streptococcus pneumoniae pneumonia. Clin Microbiol Infect 12, 31–41.[CrossRef][Medline]
Jacobs, J. A. & Stobberingh, E. E. (1996). In vitro antimicrobial susceptibility of the "Streptococcus milleri" group (Streptococcus anginosus, Streptococcus constellatus and Streptococcus intermedius). J Antimicrob Chemother 37, 371–375.[Abstract/Free Full Text]
Laupland, K. B., Ross, T., Church, D. L. & Gregson, D. B. (2006). Population-based surveillance of invasive pyogenic streptococcal infection in a large Canadian region. Clin Microbiol Infect 12, 224–230.[CrossRef][Medline]
Limia, A., Jiménez, M. L., Alarcón, T. & López-Brea, M. (1999). Five-year analysis of antimicrobial susceptibility of the Streptococcus milleri group. Eur J Clin Microbiol Infect Dis 18, 440–444.[CrossRef][Medline]
Livermore, D. M. (2000). Antibiotic resistance in staphylococci. Int J Antimicrob Agents 16, S3–S10.[CrossRef][Medline]
Morosini, M. I., Cantón, R., Loza, E., del Campo, R., Almaraz, F. & Baquero, F. (2003). Streptococcus pyogenes isolates with characterized macrolide resistance mechanisms in Spain: in vitro activities of telithromycin and cethromycin. J Antimicrob Chemother 52, 50–55.[Abstract/Free Full Text]
Deutsches Institut für Normung e. V. (1999). Normenausschuss Medizin in DIN. Susceptibility testing of pathogens to antimicrobial agents, part 4. DIN 5894–4.
Olsson-Liljequist, B., Larsson, P., Walder, M. & Miörner, H. (1997). Antimicrobial susceptibility testing in Sweden. III. Methodology for susceptibility testing. Scand J Infect Dis Suppl 105, 13–23.[Medline]
Pallares, R., Viladrich, P. F., Liñares, J., Cabellos, C. & Guidol, F. (1998). Impact of antibiotic resistance on chemotherapy for pneumococcal infections. Microb Drug Resist 4, 339–347.[Medline]
Rashid, R. M., Salah, W. & Parada, J. P. (2007). ‘Streptococcus milleri’ aortic valve endocarditis and hepatic abscess. J Med Microbiol 56, 280–282.[Abstract/Free Full Text]
Reinert, R. R., Al-Lahham, A., Lemperle, M., Tenholte, C., Briefs, C., Haupts, S., Gerards, H. H. & Lütticken, R. (2002). Emergence of macrolide and penicillin resistance among invasive pneumococcal isolates in Germany. J Antimicrob Chemother 49, 61–68.[Abstract/Free Full Text]
Schoening, T. E., Wagner, J. & Arvand, M. (2005). Prevalence of erythromycin and clindamycin resistance among Streptococcus agalactiae isolates in Germany. Clin Microbiol Infect 11, 579–582.[CrossRef][Medline]
Seppälä, H., Nissinen, A., Yu, Q. & Huovinen, P. (1993). Three different phenotypes of erythromycin-resistant Streptococcus pyogenes in Finland. J Antimicrob Chemother 32, 885–891.[Free Full Text]
Seppälä, H., Haanperä, M., Al-Juhaish, M., Järvinen, H., Jalava, J. & Huovinen, P. (2003). Antimicrobial susceptibility patterns and macrolide resistance genes of viridans group streptococci from normal flora. J Antimicrob Chemother 52, 636–644.[Abstract/Free Full Text]
Sutcliffe, J., Grebe, T., Tait-Kamradt, A. & Wondrack, L. (1996a). Detection of erythromycin-resistant determinants by PCR. Antimicrob Agents Chemother 40, 2562–2566.[Abstract/Free Full Text]
Sutcliffe, J., Tait-Kamradt, A. & Wondrack, L. (1996b). Streptococcus pneumoniae and Streptococcus pyogenes resistant to macrolides but sensitive to clindamycin: a common resistance pattern mediated by an efflux system. Antimicrob Agents Chemother 40, 1817–1824.[Abstract/Free Full Text]
Tracy, M., Wanahita, A., Shuhatovich, Y., Goldsmith, E. A., Clarridge, J. E., III & Musher, D. M. (2001). Antibiotic susceptibilities of genetically characterized Streptococcus milleri group strains. Antimicrob Agents Chemother 45, 1511–1514.[Abstract/Free Full Text]
TOP
INTRODUCTION
METHODS
