High-Level Macrolide Resistance Due to the Mega Element [ mef(E)/ mel] in Streptococcus pneumoniae

Abstract

Transferable genetic elements conferring macrolide resistance in Streptococcus pneumoniae can encode the efflux pump and ribosomal protection protein, mef(E)/mel, in an operon of the macrolide efflux genetic assembly (Mega) element- or induce ribosomal methylation through a methyltransferase encoded by erm(B). During the past 30 years, strains that contain Mega or erm(B) or both elements on Tn2010 and other Tn916-like composite mobile genetic elements have emerged and expanded globally. In this study, we identify and define pneumococcal isolates with unusually high-level macrolide resistance (MICs > 16 μg/ml) due to the presence of the Mega element [mef(E)/mel] alone. High-level resistance due to mef(E)/mel was associated with at least two specific genomic insertions of the Mega element, designated Mega-2.IVa and Mega-2.IVc. Genome analyses revealed that these strains do not possess erm(B) or known ribosomal mutations. Deletion of mef(E)/mel in these isolates eliminated macrolide resistance. We also found that Mef(E) and Mel of Tn2010-containing pneumococci were functional but the high-level of macrolide resistance was due to Erm(B). Using in vitro competition experiments in the presence of macrolides, high-level macrolide-resistant S. pneumoniae conferred by either Mega-2.IVa or erm(B), had a growth fitness advantage over the lower-level, mef(E)/mel-mediated macrolide-resistant S. pneumoniae phenotypes. These data indicate the ability of S. pneumoniae to generate high-level macrolide resistance by macrolide efflux/ribosomal protection [Mef(E)/Mel] and that high-level resistance regardless of mechanism provides a fitness advantage in the presence of macrolides.

Keywords: Mega; Streptococcus pneumoniae; Tn2010; erm(B); macrolide resistance; mef(E)/mel; pneumococcus.

FIGURE 1

Macrolide resistance phenotypes and genotypes of S. pneumoniae. Erythromycin minimum inhibitory concentrations (MICs) were determined under uninduced conditions (black bars) and cultures induced with 0.1 μg/ml erythromycin (white bars). Each bar is the average MIC for a macrolide resistance genotype and strains in each group are detailed in Table 1.

FIGURE 2

The competitive index of isogenic GA44288 Tn2010 mutants grown in vitro with erythromycin (0.5 μg/ml) for approximately 50 generations: (A) MS41 [Tn2010Δmef(E)/mel] versus Tn2010, (B) MS32 [Tn2010Δerm(B)] versus Tn2010, and (C) MS32 [Tn2010Δerm(B)] versus MS41 [Tn2010Δmef(E)/mel]. ^∗p < 0.05, ^∗∗p < 0.01, ^∗∗∗p < 0.001.

FIGURE 3

The competitive index of high-level macrolide resistance strains with distinct mechanisms [erm(B) and Mega-2.Iva] in vitro with erythromycin (0.5 μg/ml) grown for approximately 50 generations: (A) GA44288 [erm(B)-and (mef(E)/mel-containing] versus GA16242 (Mega-2.IVa-containing), (B) MS41 [GA44288 Δmef(E)/mel, erm(B)-containing] versus GA16242 (Mega-2.IVa-containing), and (C) MS32 [GA44288 Δerm(B)] versus GA16242 (Mega-2.IVa-containing). ^∗p < 0.05, ^∗∗p < 0.01.)