Impact of mecA promoter mutations on mecA expression and β-lactam resistance levels

Abstract

The reason for the extremely low-level oxacillin resistance in a so-called ‘drug clone’, a methicillin-resistant Staphylococcus aureus circulating among injection drug users in Zurich, Switzerland, could be traced back to the mecA promoter sequence and particularly to the strain's genetic background. Sequencing of its mec complex identified a point mutation (TATACT to TATATT), creating a perfect palindrome in the −10 region of the mecA promoter/operator region containing the binding sites for the mecA repressors MecI and BlaI. Two strains with vastly different β-lactam resistance phenotypes, the low-level resistant drug clone type strain CHE482 and the highly homogeneously resistant strain COLn, were cured of their SCCmec elements and subsequently transformed with plasmids containing mecA under the control of either the wild-type or mutant promoter. Expression studies showed that this mutation had significant effects on both mecA transcription and corresponding PBP2a production, but only small effects on β-lactam resistance levels within a given genetic background. A further mutation in the mecA ribosomal binding site (GGAGG to GGAGT), common to SCCmec type IV strains, was found to have no discernable effect on mecA transcription and PBP2a content, and only minimal effects on β-lactam resistance. Factors associated with the genetic backgrounds into which these differently controlled mecA genes were introduced had a much higher impact on β-lactam resistance levels than the rates of mecA transcription. The tight repression of mecA expression in this drug clone in the absence of β-lactams could contribute to the apparent fitness of this fast growing strain.

Introduction

Methicillin-resistant Staphylococcus aureus (MRSA) can express highly variable levels of β-lactam resistance. Oxacillin minimal inhibitory concentrations (MIC) can vary from as low as 1 μg ml^–1 up to values over 1000 μg ml^–1. Extremely low-level resistant MRSA strains are dangerous as they often evade phenotypic detection. While they appear phenotypically susceptible, these strains still carry the mecA gene which encodes the β-lactam resistance protein penicillin binding protein (PBP) 2a, and express resistance heterogeneously. This means that upon β-lactam exposure they can segregate highly resistant subpopulations, at frequencies well above spontaneous mutation rates, and cause β-lactam treatment failure.

β-Lactam resistance relies on the presence of mecA and the expression of PBP2a; however, several other factors are known to influence resistance. Chromosomal mutations or specific genomic factors are required for expression of high-level resistance (Murakami and Tomasz, 1989; Ryffel et al., 1994). Resistance levels are also influenced by environmental conditions such as temperature, pH, osmolarity, divalent cations, and aerobiosis (Matthews and Steward, 1984). The actual genetic basis governing low-level, heterogeneous, or high homogeneous resistance is unknown.

Expression of mecA is inducible and can be controlled by either its cognate regulators MecI (DNA binding repressor protein) and MecR1 (sensor/signal transducer) or by the structurally and functionally similar β-lactamase regulators BlaI and BlaR1, respectively. Because of structural and functional similarity, both MecI and BlaI can bind as homodimers to the promoter/operator region of both mecA and blaZ (Gregory et al., 1997; Sharma et al., 1998). Repression of transcription by these two inhibitors is even stronger when both repressors are present (Rosato et al., 2003a). In the presence of β-lactams the MecR1/BlaR1 sensor-transducers promote the cleavage of their respective repressors allowing transcription of mecA. In the absence of both regulatory loci, mecA is constitutively expressed.

High conservation of the mecA gene sequence (Chambers, 1997), the promoter region, and the preferred presence of regulatory genes (blaI/blaR1 or mecI/mecR1) (Rosato et al., 2003a) highlights the need for controlled expression of mecA. In strains such as N315, a so-called pre-MRSA (Niemeyer et al., 1996; Weller, 1999), carrying both complete regulatory systems, mecA is repressed by both MecI and BlaI. This tight repression can prevent mecA transcription and lead to inhibition of β-lactam resistance and to the misinterpretation of these strains as methicillin-sensitive S. aureus (MSSA).

The amount of PBP2a produced has been shown to have no direct correlation with resistance level; however, mutations in mecA, mecI, the mecI ribosomal binding site, or mecI deletion can lead to increased resistance (Katayama et al., 2004; Rosato et al., 2003a; Niemeyer et al., 1996), whereas interruption of blaR1 results in constitutive repression and therefore decreased resistance (Hackbarth et al., 1994).

An MRSA strain with very low-level oxacillin resistance with the ability to generate highly resistant subclones is spreading among injection drug users in Zurich, Switzerland (Qi et al., 2005). Isolates of this clone having oxacillin MICs of 0.5–4 μg ml^–1 are susceptible or just at the breakpoint of resistance for oxacillin, according to CLSI guidelines. They fall within the classification of the recently described OS-MRSA (oxacillin-susceptible, methicillin-resistant S. aureus), defined as being oxacillin-sensitive but mecA-positive strains with MICs of 2 μg ml^–1 or below (Hososaka et al., 2007). A recent analysis of the SCCmec element of the drug clone type strain CHE482 identified a composite element carrying a mecB complex (IS1272–ΔmecR1–mecA) (Ender et al., 2007). Further investigation of the mec complex revealed that the mecA promoter differed from the consensus by a 1-bp substitution in the −10 region. Here we determined the influence of this nucleotide substitution on mecA transcription and β-lactam resistance levels in different genetic backgrounds, using MRSA cured from their SCCmec element as recipients.