Abstract: | ![]() The correct interpretation of microbial sequencing data applied to surveillance and outbreak investigation depends on accessible genomic databases to provide vital genetic context. Our aim was to construct and describe a United Kingdom MRSA database containing over 1000 methicillin-resistant Staphylococcus aureus (MRSA) genomes drawn from England, Northern Ireland, Wales, Scotland, and the Republic of Ireland over a decade. We sequenced 1013 MRSA submitted to the British Society for Antimicrobial Chemotherapy by 46 laboratories between 2001 and 2010. Each isolate was assigned to a regional healthcare referral network in England and was otherwise grouped based on country of origin. Phylogenetic reconstructions were used to contextualize MRSA outbreak investigations and to detect the spread of resistance. The majority of isolates (n = 783, 77%) belonged to CC22, which contains the dominant United Kingdom epidemic clone (EMRSA-15). There was marked geographic structuring of EMRSA-15, consistent with widespread dissemination prior to the sampling decade followed by local diversification. The addition of MRSA genomes from two outbreaks and one pseudo-outbreak demonstrated the certainty with which outbreaks could be confirmed or refuted. We identified local and regional differences in antibiotic resistance profiles, with examples of local expansion, as well as widespread circulation of mobile genetic elements across the bacterial population. We have generated a resource for the future surveillance and outbreak investigation of MRSA in the United Kingdom and Ireland and have shown the value of this during outbreak investigation and tracking of antimicrobial resistance.Methicillin-resistant Staphylococcus aureus (MRSA) was first isolated in 1961 in the United Kingdom (UK), 1 yr after methicillin was introduced into clinical practice (Jevons 1961). The prevalence of MRSA gradually increased thereafter, and by 1971, 5% of S. aureus isolates referred to the National Staphylococcal Reference Laboratory were MRSA (Marples and Reith 1992). Outbreaks of gentamicin-resistant MRSA in several hospitals during the late 1970s (Shanson 1981) were followed by the emergence of MRSA with potential for epidemic spread (Johnson et al. 2005). By the mid-1980s, MRSA had spread across the UK, and the majority were epidemic (E)MRSA-1, later assigned as sequence type (ST) 239 by multilocus sequence typing (MLST) (Kerr et al. 1990; Johnson et al. 2005). A decline in EMRSA-1 in the late 1980s and early 1990s was associated with an increase in EMRSA-3 (ST 5) (Marples and Reith 1992; Richardson and Reith 1993; Cox et al. 1995; Enright et al. 2002). This dynamic process continued with the emergence in the early 1990s of EMRSA-15 (ST 22) and EMRSA-16 (ST 36) (Richardson and Reith 1993; Cox et al. 1995; Enright et al. 2002), which disseminated across the UK. These two clones continue to predominate, with EMRSA-15 accounting for ∼85% of MRSA bloodstream infections in the UK in 2007 and with trends suggesting that EMRSA-16 is in decline (Ellington et al. 2010; McAdam et al. 2012). Antimicrobial resistance is known to differ between EMRSA-15 and -16, with EMRSA-16 being the more resistant lineage of the two. However, for both lineages, the acquisition of the SCCmec element conferring methicillin resistance and the presence of mutations in gyrA and grlA conferring fluoroquinolone resistance are considered to be major contributors to the success of these epidemic lineages (Knight et al. 2012; McAdam et al. 2012; Holden et al. 2013).Bacterial genotyping using pulsed-field gel electrophoresis (PFGE), MLST, and spa typing has been used to identify epidemic clones and to give insights into the microevolutionary dynamics of predominant MRSA lineages in the UK. However, these methods have limited resolution and lack discriminatory power when one or a small number of clones predominate (McAdam et al. 2012; Miller et al. 2014; Bartels et al. 2015). This means that once widely established, the subsequent dynamics of clonal MRSA spread within and between healthcare facilities cannot be fully elucidated. As a result, bacterial typing does not form a central component of MRSA transmission and outbreak investigation. Several recent publications have confirmed the ability of whole-genome sequencing (WGS) to define transmission dynamics of a single clone at different geographic and temporal scales. This has identified global and local transmission routes and, when combined with epidemiological data, can confirm or refute putative MRSA outbreaks (Köser et al. 2012; Harris et al. 2013; Nubel et al. 2013; Miller et al. 2014; Török et al. 2014; Bartels et al. 2015). Similarly, while surveillance of MRSA has been carried over several years and a limited number of point prevalence studies of variable methodology have been undertaken in different settings, serial systematic prevalence studies of individual epidemic lineages are lacking (Johnson et al. 2012; Afshinnekoo et al. 2015; Bartels et al. 2015; Peng et al. 2015). WGS could potentially be used for national and local surveillance of MRSA lineages and to enhance the investigation of suspected outbreaks, but comprehensive genomic databases are required to provide the context that would allow robust epidemiological interferences. Here, we describe the analysis of over 1000 MRSA genomes drawn from across the UK and Ireland over a period of a decade and the first evaluation of this rich data set to describe the macroepidemiology of MRSA. |