Abstract
Objectives:
Solithromycin is a fluoroketolide that is considered to be a noninducing antibiotic for macrolide–lincosamide–streptogramin B resistance mediated by erm genes. The exact activity of solithromycin to induce erm gene expression remains to be determined.
Materials and Methods:
The potential of solithromycin to induce erm(A), erm(C), and erm(B) gene expression was examined using a lacZ reporter assay, double-disk diffusion test, and determination of the minimal inhibitory concentration after incubation with subinhibitory concentration of different antibiotics.
Results:
Neither solithromycin nor the ketolides telithromycin and cethromycin induced erm(A) or erm(C) gene expression. However, solithromycin could significantly induce erm(B) gene expression at levels greater than that seen for cethromycin and clindamycin, but less than that for erythromycin, rokitamycin, and telithromycin.
Conclusion:
Solithromycin does not induce erm(A) and erm(C) gene expression, but does induce erm(B) gene expression, although to a weaker extent than that seen for macrolides.
Introduction
Ketolide antibiotics are derivatives of 14-membered-ring macrolides and are characterized by the substitution of a ketone group for the cladinose moiety. In contrast to macrolides, ketolides are thought to be incapable of inducing macrolide–lincosamide–streptogramin B (MLSB) resistance mediated by erm genes.1,2 However, the marketed ketolides telithromycin and cethromycin were shown to induce expression of the erm(B) gene, 3 the predominant MLSB resistance gene in streptococci. Moreover, expression of erm(C), a common MLSB resistance gene in staphylococci, was reported to be induced by telithromycin, cethromycin, and some experimental fluoroketolides in Escherichia coli.4,5
Solithromycin was the first fluoroketolide to enter clinical development, and has a fluorine at the C2 position of the 14-membered macrolactone ring that increases binding to 23S rRNA.6,7 Consequently, solithromycin has enhanced activity against pneumonia-causing bacteria, including penicillin-, macrolide-, and even telithromycin-resistant strains.6,7 Solithromycin is also thought not to trigger inducible MLSB resistance due to its lack of cladinose, like other ketolides. 6 Despite the importance of solithromycin for development of new antibiotics that do not induce MLSB resistance, few studies have evaluated the potential of solithromycin to induce expression of genes related to MLSB resistance. In this study, a lacZ reporter assay, double-disk test, and determination of the minimal inhibitory concentration (MIC) after incubation in the presence of a subinhibitory concentration of various antibiotics were used to examine the ability of solithromycin to induce expression of erm(A), erm(C), and erm(B), the three major erm genes involved in MLSB resistance.
Materials and Methods
β-Galactosidase induction assay
The regulatory region of erm(A) was amplified from pLS2008 using the primers 5′-TCTGAATTCGTCAAGGTGCAAAATTAC-3′ and 5′-GTGGGATCCAATATTTCTTTTACATGC-3′ to construct the erm(A)-lacZ reporter. The EcoRI/BamHI-digested PCR product was cloned into EcoRI/BamHI-cleaved pMM156 containing a promoterless lacZ gene. 9 The resulting plasmid was transformed into E. coli CSH26 and subsequently transformed into Bacillus subtilis BR151. The construction of erm(C)-lacZ was performed in the same manner using the template pE19410 and primers 5'-TCCGAATTCAGTATAAATTTAACGATC-3’ and 5'-CCTGGATCCATTTCAAAGATATTATCATG-3’. B. subtilis BR151 cultures carrying either erm(A)-lacZ, erm(C)-lacZ, or the previously constructed erm(B)-lacZ 11 were induced for 120 minutes with the indicated antibiotics at concentrations ranging from 1 ng/mL to 10 μg/mL, and the activities of β-galactosidase were measured, as described previously. 12
Double-disk diffusion test
The double-disk test was performed for 30 clinical isolates having an inducible MLSB resistance phenotype, and included 10 clinical isolates of Staphylococcus aureus harboring erm(A), 10 isolates of S. aureus harboring erm(C), and 10 isolates of Staphylococcus pneumoniae harboring erm(B) that were collected from two Korean hospitals during 2008–2011. The presence of the erm genes was determined by PCR using primers specific for erm(A), erm(C), and erm(B). 13 For S. aureus, solithromycin (0.02 μg; Cayman Chemical, Ann Arbor, MI) and rokitamycin (5 μg; Asahi Kasei, Tokyo, Japan) disks were placed on Mueller–Hinton (MH) agar (Becton Dickinson, Sparks, MD). For S. pneumoniae, solithromycin (0.02 μg) and rokitamycin (30 μg) disks were placed on MH agar supplemented with 5% sheep blood. An erythromycin (1 μg; Sigma, St. Louis, MO) disk was used as a control. The S. aureus strains were incubated at 35°C in ambient air for 18 hours, and the S. pneumoniae strains were incubated at 35°C in 5% CO2 for 24 hours. Blunting of the rokitamycin zone of inhibition proximal to the solithromycin disk was taken to indicate the induction potential of solithromycin.
MIC determination after induction
The MICs of MLSB antibiotics and ketolides were determined after incubation in the presence of a subinhibitory concentration of solithromycin to assess the influence of the presence of solithromycin on MLSB and ketolide resistance, as previously described. 14 Three clinical isolates with inducible erm genes [S. aureus SH20 with erm(A), S. aureus SH58 with erm(C), and S. pneumoniae KH67 with erm(B)] were grown on MH agar (for S. aureus) or MH agar supplemented with 5% sheep blood (for S. pneumoniae). The initial suspensions of the bacteria were prepared from colonies in MH II broth (for S. aureus; Becton Dickinson, Sparks, MD) or MH II broth supplemented with 5% lysed horse blood (for S. pneumoniae) to a density of McFarland standard 0.5. Bacterial suspensions were prepared by a 10-fold dilution of the initial suspensions and then incubated in the presence of a subinhibitory concentration (0.01 μg/mL) of telithromycin (Sanofi-Aventis Korea, Seoul, South Korea), cethromycin (Sino-Standards, Chengdu, China), or solithromycin at 35°C for 2.5 hours. Erythromycin was used as a control (0.1 μg/mL). After washing with fresh medium, the MICs of the MLSB and ketolide antibiotics were determined using broth microdilution. 15 Clindamycin was purchased from Sigma-Aldrich (St. Louis, MO). Quinupristin was obtained from Handok Pharmaceuticals (Seoul, South Korea). Linezolid (Dong-A Pharmaceutical, Yongin, South Korea) was included as a negative control.
Results
Expression of erm(A) or erm(C) was not induced by solithromycin or by the 16-membered-ring macrolide rokitamycin, the lincosamide clindamycin, or the ketolides telithromycin and cethromycin (Fig. 1). Only erythromycin could induce erm(A) or erm(C) expression (Fig. 1). In contrast, erm(B) gene expression was induced by all of the MLSB and ketolide antibiotics tested, including solithromycin, which exhibited significantly higher inducing activity than cethromycin and clindamycin, but less than that for erythromycin, rokitamycin, or telithromycin. This result is consistent with an earlier study showing that erm(B) expression is induced by all MLSB antibiotics and the ketolides telithromycin and cethromycin. 3 In the double-disk test, no induction potential for solithromycin was seen for any of the S. aureus isolates with either erm(A) or erm(C), but induction was seen for all S. pneumoniae isolates with erm(B) (Table 1). In accordance with the lacZ reporter assay and double-disk test results, none of the ketolide antibiotics tested, including solithromycin, altered the susceptibility of the S. aureus isolates with either erm(A) or erm(C) after incubation in the presence of a subinhibitory concentration (0.01 μg/mL) of ketolides (Table 2). In contrast, for the erm(B)-harboring S. pneumoniae isolate, pregrowth in a subinhibitory concentration (0.01 μg/mL) of ketolides did increase the MICs of MLSB and the ketolide antibiotics tested. In particular, solithromycin induced a two- to eightfold increase in the MICs of the MLSB and ketolide antibiotics.

Expression of erm(A)-lacZ, erm(C)-lacZ, and erm(B)-lacZ fusions after incubation with various concentrations of erythromycin (●), rokitamycin (○), clindamycin (■), telithromycin (□), cethromycin (♦), and solithromycin (◊) for 120 minutes. The experiment was performed in triplicate and mean values are plotted.
Induction Potential of Erythromycin and Solithromycin Measured by a Double-Disk Diffusion Test
Determination of Susceptibility to MLSB and Ketolide Antibiotics After Incubation with Subinhibitory Concentration of Erythromycin, Telithromycin, Cethromycin, or Solithromycin
CET, cethromycin; CLI, clindamycin; ERY, erythromycin; LIN, linezolid; MIC, minimal inhibitory concentration;
Discussion
In a previous study, the ketolide antibiotic telithromycin induced no erm(A) or erm(C) expression in S. aureus or S. pyogenes strains, but did induce erm(B) in S. pyogenes by disk diffusion. 14 Furthermore, this study found that upon pregrowth in the presence of subinhibitory concentrations of telithromycin, the MICs of MLSB against the S. aureus strain harboring inducible erm(C) were unchanged, whereas MICs against the S. pyogenes strain harboring inducible erm(B) increased. 14 This study showed that the ketolide antibiotics solithromycin and cethromycin both exhibited the same induction specificity as telithromycin.
Meanwhile, other earlier studies reported that ketolides induced expression of erm(C) in E. coli,4,5 which is in contrast to results of this study. This discrepancy could be due to the difference between ribosomes of Gram-positive bacteria and Gram-negative E. coli. For example, the uridine at position 746 in domain II (hairpin 35) of 23S rRNA, a telithromycin binding site, is modified to pseudouridine in E. coli.16,17 The potential role of this modification in erm induction requires further investigation to determine whether it is in fact involved in the action of ketolide antibiotics.
Taken together, the data for this study demonstrate that solithromycin, like telithromycin and cethromycin, does not induce expression of erm(A) and erm(C), but does induce erm(B) expression. These findings suggest that the substitution of a ketone for cladinose, a key feature of ketolides, is sufficient to abolish erm(A) and erm(C) induction. Ketolides could not completely suppress induction of erm(B) expression, but induced lower expression levels than those seen for macrolides. Additional modifications to the ketolide structure will be needed to minimize induction of erm(B) expression.
Footnotes
Acknowledgment
The author is grateful to Dr. Byoungduck Park (College of Pharmacy, Keimyung University, Republic of Korea) for his helpful advice and review.
Disclosure Statement
No competing financial interests exist.
Funding Information
This research received no specific grant from any funding agency.
