Search

In mating experiments, the erythromycin resistance methylase gene erm(A) was successfully transferred from erm(A)-positive clinical isolates of Streptococcus pyogenes to macrolide-susceptible recipients of S. pyogenes, Enterococcus faecalis and Listeria innocua. Compared with the SmaI macrorestriction pattern of the S. pyogenes recipient, the patterns of S. pyogenes transconjugants shared the lack of a fragment and the appearance of a new, larger fragment. This is the first experimental evidence that the erm(A) gene can be transferred from erythromycin-resistant S. pyogenes to macrolide-susceptible S. pyogenes as well as to other Gram-positive recipients.

Laboratory differentiation of erythromycin resistance phenotypes is poorly standardized for pneumococci. In this study, 85 clinical isolates of erythromycin-resistant (MIC > or = 1 microg/ml) Streptococcus pneumoniae were tested for the resistance phenotype by the erythromycin-clindamycin double-disk test (previously used to determine the macrolide resistance phenotype in Streptococcus pyogenes strains) and by MIC induction tests, i.e., by determining the MICs of macrolide antibiotics without and with pre-exposure to 0.05 microg of erythromycin per ml. By the double-disk test, 65 strains, all carrying the erm(AM) determinant, were assigned to the constitutive macrolide, lincosamide, and streptogramin B resistance (cMLS) phenotype, and the remaining 20, all carrying the mef(E) gene, were assigned to the recently described M phenotype; an inducible MLS resistance (iMLS) phenotype was not found. The lack of inducible resistance to clindamycin was confirmed by determining clindamycin MICs without and with pre-exposure to subinhibitory concentrations of erythromycin. In macrolide MIC and MIC-induction tests, whereas homogeneous susceptibility patterns were observed among the 20 strains assigned to the M phenotype by the double-disk test, two distinct patterns were recognized among the 65 strains assigned to the cMLS phenotype by the same test; one pattern (n = 10; probably that of the true cMLS isolates) was characterized by resistance to rokitamycin also without induction, and the other pattern (n = 55; designated the iMcLS phenotype) was characterized by full or intermediate susceptibility to rokitamycin without induction turning to resistance after induction, with an MIC increase by more than three dilutions. A triple-disk test, set up by adding a rokitamycin disk to the erythromycin and clindamycin disks of the double-disk test, allowed the easy differentiation not only of pneumococci with the M phenotype from those with MLS resistance but also, among the latter, of those of the true cMLS phenotype from those of the iMcLS phenotype. While distinguishing MLS from M resistance in pneumococci is easily and reliably achieved, the differentiation of constitutive from inducible MLS resistance is far more uncertain and is strongly affected by the antibiotic used to test inducibility.

High rates of erythromycin resistance among Streptococcus pyogenes strains have been reported in Italy in the last few years. In this study, 370 erythromycin-resistant (MIC, > or = 1 microg/mL) Italian isolates of this species obtained in 1997-1998 from throat swabs from symptomatic patients were typed by analyzing SmaI macrorestriction fragment patterns by pulsed-field gel electrophoresis (PFGE). Among the typable isolates (n = 341; the genomic DNA of the remaining 29 isolates was not restricted by SmaI), 48 distinct PFGE types were recognized, of which 31 were recorded in only one isolate (one-strain types). Fifty-two percent of typable isolates fell into three type clusters and 75% into six, suggesting that erythromycin-resistant group A streptococci circulating in Italy are polyclonal, but the majority of them probably derives from the spread of a limited number of clones. In parallel experiments, the 370 test strains were characterized for the macrolide resistance phenotype: 80 were assigned to phenotype cMLS, 89 to phenotype iMLS-A, 33 to phenotype iMLS-B, 11 to phenotype iMLS-C, and 157 to phenotype M. There was a close correlation between these phenotypic data and the genotypic results of PFGE analysis, the vast majority of the isolates assigned to individual PFGE classes belonging usually to a single phenotype of macrolide resistance. All of the 29 untypable isolates belonged to the M phenotype. Further correlations were observed with tetracycline resistance.

Two ketolides, telithromycin and HMR 3004, were evaluated for their in vitro activity against erythromycin-susceptible and -resistant strains of Streptococcus pyogenes and Streptococcus pneumoniae. On the basis of their resistance to macrolide, lincosamide and streptogramin (MLS) antibiotics, erythromycin-resistant test strains were assigned to the constitutive resistance (cMLS) phenotype, the inducible resistance (iMLS) phenotype or the M phenotype. iMLS S. pyogenes strains were further subdivided into the three recently described subtypes iMLS-A, -B and -C. Telithromycin and HMR 3004 were uniformly and highly active against pneumococci (regardless of their susceptibility or resistance to erythromycin and/or penicillin), erythromycin-susceptible S. pyogenes and erythromycin-resistant S. pyogenes strains of the M phenotype (in which resistance is mediated by an efflux system) or iMLS-B or -C phenotype (in which resistance is mediated by a methylase encoded by the ermTR gene). Both ketolides were less active against erythromycin-resistant S. pyogenes strains with the cMLS phenotype or the iMLS-A subtype (where resistance is mediated by a methylase encoded by the ermAM gene), these strains ranging in phenotype from the upper limits of susceptibility to low-level resistant.

Throat swab specimens were obtained from 3,227 children with symptoms of acute pharyngotonsillitis. After 14 to 21 days, a second throat swab specimen was obtained at a follow-up visit. Over 42% of the 934 strains of Streptococcus pyogenes isolated in the primary study were resistant to erythromycin, azithromycin, and clarithromycin. Eradication rates among the 668 patients who entered the follow-up study were as follows: 84.1%, penicillin recipients; 82.7%, cephalosporin recipients; and 71.7%, macrolide recipients. Among patients treated with macrolides, the eradication rate was approximately 80% when the infecting organisms were erythromycin-susceptible and approximately 60% when they were erythromycin-resistant. These results indicate substantial in vitro macrolide resistance among Italian isolates of S. pyogenes. However, at least for a minor self-limiting condition such as acute S. pyogenes pharyngitis, our findings point to a limited overall correlation between in vitro susceptibility (to penicillins, cephalosporins, or macrolides) and eradication in patients treated with these drugs and an even weaker correlation between in vitro resistance (to macrolides) and noneradication in patients receiving macrolide therapy.

A total of 387 clinical strains of erythromycin-resistant (MIC, >/=1 microg/ml) Streptococcus pyogenes, all isolated in Italian laboratories from 1995 to 1998, were examined. By the erythromycin-clindamycin double-disk test, 203 (52.5%) strains were assigned to the recently described M phenotype, 120 (31.0%) were assigned to the inducible macrolide, lincosamide, and streptogramin B resistance (iMLS) phenotype, and 64 (16.5%) were assigned to the constitutive MLS resistance (cMLS) phenotype. The inducible character of the resistance of the iMLS strains was confirmed by comparing the clindamycin MICs determined under normal testing conditions and those determined after induction by pregrowth in 0.05 microg of erythromycin per ml. The MICs of erythromycin, clarithromycin, azithromycin, josamycin, spiramycin, and the ketolide HMR3004 were then determined and compared. Homogeneous susceptibility patterns were observed for the isolates of the cMLS phenotype (for all but one of the strains, HMR3004 MICs were 0.5 to 8 microg/ml and the MICs of the other drugs were >128 microg/ml) and those of the M phenotype (resistance only to the 14- and 15-membered macrolides was recorded, with MICs of 2 to 32 microg/ml). Conversely, heterogeneous susceptibility patterns were observed in the isolates of the iMLS phenotype, which were subdivided into three distinct subtypes designated iMLS-A, iMLS-B, and iMLS-C. The iMLS-A strains (n = 84) were highly resistant to the 14-, 15-, and 16-membered macrolides and demonstrated reduced susceptibility to low-level resistance to HMR3004. The iMLS-B strains (n = 12) were highly resistant to the 14- and 15-membered macrolides, susceptible to the 16-membered macrolides (but highly resistant to josamycin after induction), and susceptible to HMR3004 (but intermediate or resistant after induction). The iMLS-C strains (n = 24) had lower levels of resistance to the 14- and 15-membered macrolides (with erythromycin MICs increasing two to four times after induction), were susceptible to the 16-membered macrolides (but resistant to josamycin after induction), and remained susceptible to HMR3004, also after induction. The erythromycin resistance genes in 100 isolates of the different groups were investigated by PCR. All cMLS and iMLS-A isolates tested had the ermAM (ermB) gene, whereas all iMLS-B and iMLS-C isolates had the ermTR gene (neither methylase gene was found in isolates of other groups). The M isolates had only the macrolide efflux (mefA) gene, which was also found in variable proportions of cMLS, iMLS-A, iMLS-B, and iMLS-C isolates. The three iMLS subtypes were easily differentiated by a triple-disk test set up by adding a josamycin disk to the erythromycin and clindamycin disks of the conventional double-disk test. Tetracycline resistance was not detected in any isolate of the iMLS-A subtype, whereas it was observed in over 90% of both iMLS-B and iMLS-C isolates.

An erythromycin-resistant (Emr) Listeria innocua and an Emr Listeria monocytogenes isolate both carried ermC genes, which code for rRNA methylases. The ermC genes were transferable by conjugation to recipient L. monocytogenes, Listeria ivanovii, and Enterococcus faecalis but did not appear to be associated with conjugative plasmids.

The usefulness of macrolides in treating respiratory infections has been established for over thirty years. Currently, a great deal of interest is being focused on roxithromycin, a new semisynthetic derivative of erythromycin which is more stable than erythromycin under acidic conditions and exhibits improved pharmacokinetic properties. In this study, special attention is paid to the results of recent multicenter studies in Italy aimed at evaluating the in vitro activity of roxithromycin versus erythromycin against a wide range of respiratory pathogens. Considering that a high degree of overlap was observed between the roxithromycin-susceptible and the erythromycin-susceptible strains, whereas a significant proportion of erythromycin-resistant strains shifted to the intermediate category with roxithromycin, there appeared to be cross-susceptibility rather than cross-resistance between the two macrolides.