Induction of telithromycin resistance in Streptococcus pneumoniae

Sir, Ketolides are a new class of semisynthetic macrolide derivatives that show excellent activity against Streptococcus pneumoniae, even against erythromycin-resistant isolates. Previous investigators have not found any telithromycin-resistant isolates among S. pneumoniae with the constitutive macrolide–lincosamide–streptogramin B (cMLSB) phenotype, even though staphylococci or Streptococcus pyogenes with the cMLSB phenotype can develop resistance to this drug.1–3 The lack of induction of methylase production by this drug is one reason for such a difference.4 Recently, Hamilton-Miller & Shah5 reported that resistance to the ketolide cethromycin (formerly ABT-773) could be induced in S. pneumoniae with the cMLSB phenotype by erythromycin or other related antibiotics. In this study, we examined whether 55 isolates of erythromycin-resistant S. pneumoniae (MIC ≥ 1 mg/L) carrying the mef(A) and/or erm(B) genes could develop resistance to telithromycin, a ketolide with different substitutions from cethromycin (ABT-773), when exposed to erythromycin. Fifty-five clinical isolates of erythromycin-resistant S. pneumoniae (obtained from the sputum of patients with lower respiratory tract infections between 1998 and 2000) were used. These isolates were identified by their sensitivity to optochin and the bile solubility test, and by PCR amplification of the lytA gene. S. pneumoniae ATCC 6305 was used as the quality control strain. Reference samples of the following antimicrobial agents of known potency were kindly supplied in powder form by the indicated manufacturers: erythromycin (Shionogi Pharmaceutical Co., Osaka, Japan), clarithromycin (Taisho Pharmaceutical Co., Tokyo, Japan), azithromycin (Pfizer Laboratories, Groton, CT, USA), rokitamycin (Asahi Kasei, Tokyo, Japan), telithromycin (Nippon HoechstMarion-Roussel, Tokyo, Japan) and clindamycin (Upjohn, Tokyo, Japan). All isolates were tested for susceptibility to the six antibiotics listed above at concentrations between 0.015 and 128 mg/L. MICs were determined by the two-fold agar dilution method using susceptibility test agar (Mueller–Hinton agar medium; Eiken Chemicals, Tokyo, Japan) with 8% Strepto Haemo supplement (SHS; Eiken Chemicals). DNA was obtained as previously reported, and the presence of macrolide resistance genes was investigated by PCR using primers and amplification conditions that have been described previously.6 Induction of telithromycin resistance was examined by the following two methods. (i) Agar dilution method: induction of telithromycin resistance was evaluated by comparing the MICs for telithromycin in the presence and absence of a subinhibitory concentration of erythromycin (0.1 mg/L). Two susceptibility agar plates were prepared for each isolate: one contained only telithromycin and the other contained telithromycin plus erythromycin. The MIC was determined in accordance with the susceptibility test. (ii) Disc diffusion method: A bacterial suspension (2 mL of ∼108 cfu/mL) was inoculated onto 10 mL of susceptibility test agar containing 8% SHS and spread over the surface. After excess suspension was removed, paper discs (8 mm diameter high discs; Tokyo Roshi Kaisha, Tokyo, Japan) containing erythromycin or rokitamycin at 20 μg/disc or telithromycin at 5 μg/disc were placed on the surface of the agar plate. Then the plates were incubated overnight at 35°C and induction of telithromycin resistance was assessed from the shape of the zone of inhibition around the telithromycin disc nearest either the erythromycin or the rokitamycin disc. Fifteen isolates carrying only the mef(A) gene had the M phenotype. Among the other 40 isolates, 25 carried the erm(B) gene, and 15 had both the mef(A) and erm(B) genes. All 40 isolates showed a high