Increase of genetic diversity and clonal replacement of epidemic methicillin-resistant Staphylococcus aureus strains in South-East Austria

Spa-typing and microarray techniques were used to study epidemiological changes in methicillin-resistant Staphylococcus aureus (MRSA) in South-East Austria. The population structure of 327 MRSA isolated between 2002 and 2012 was investigated. MRSA was assigned to 58 different spa types and 14 different MLST CC (multilocus sequence type clonal complexes); in particular, between 2007 and 2012, an increasing diversity in MRSA clones could be observed. The most abundant clonal complex was CC5. On the respective SCCmec cassettes, the CC5 isolates differed clearly within this decade and CC5/SCCmecI, the South German MRSA, predominant in 2002, was replaced by CC5/SCCmecII, the Rhine-Hesse MRSA in 2012. Whereas in many European countries MLST CC22-MRSA (EMRSA 15, the Barnim epidemic MRSA) is predominant, this clone occurred in Austria nearly 10 years later than in neighbouring countries. CC45, the Berlin EMRSA, epidemic in Germany, was only sporadically found in South-East Austria. The Irish ST8-MRSA-II represented by spa-type t190 was frequently found in 2002 and 2007, but disappeared in 2012. Our results demonstrate clonal replacement of MRSA clones within the last years in Austria. Ongoing surveillance is warranted for detection of changes within the MRSA population.


INTRODUCTION
Over the last six decades, methicillin-resistant Staphylococcus aureus (MRSA) spread over the whole world and has become a global public health threat. Primarily MRSA was restricted to hospitals (hospital-acquired MRSA), where few, multiresistant strains dominated (Grundmann et al. 2010;Kinnevey et al. 2014;Stryjewski and Corey 2014). For the last three decades, MRSA could also be found outside this setting (communityacquired MRSA, CA-MRSA) (Kock et al. 2014). Most of these CA-MRSA strains carry the Panton-Valentine leukocidin (PVL), a vir-ulence factor associated with soft tissue infection or skin infections (Chambers and Deleo 2009;Albrecht et al. 2011;Otto 2013;Kock et al. 2014). Another development was the occurrence of so-called livestock-associated MRSA (LA-MRSA) linked with mainly pig farming and is therefore found besides the animals, in people with contact to these. Spread from human to human is rarely seen in LA-MRSA, and LA-MRSA seems to have a reduced repertoire of virulence factors. In Austria, the first LA-MRSA was detected in 2004, with increasing numbers up to now (Huijsdens et al. 2006;Witte et al. 2007;Zarfel et al. 2013). While these changes within the MRSA population were observed, during the last 10 years MRSA predominantly belonging to only a few multilocus sequence typing (MLST) clonal complexes (CC) (CC5, CC22, CC8, CC1 or CC398) emerged worldwide, but appeared at different times, replacing other epidemic MRSA clones (Chambers and Deleo 2009;Wyllie, Paul and Crook 2011). In a recently described example from Germany, CC22-MRSA (Barnim Epidemic strain, UK-EMRSA15) was first detected in 2001 and increased up to 58.6% of the investigated MRSA isolates in 2010, in contrast to CC5/ST228-MRSA I decreased from nearly 50% in 2000 to 2.3% in 2010 (Albrecht et al. 2011). Further examples of a replacement of MRSA clones were reported from many countries all over the world, such as Ireland, UK, Portugal or the USA (Enright et al. 2002;Amorim et al. 2007;Chambers and Deleo 2009;Grundmann et al. 2010).
There are a few publications on MRSA epidemiology from Austria, but they mainly focused on CA-MRSA or LA-MRSA (Ruppitsch et al. 2006;Krziwanek et al. 2007Krziwanek et al. , 2008Grisold et al. 2009).
This study is the first study for Austria, investigating the genetic background of all detected MRSA primary isolates from 2002 up to 2012 in the South-East Austria to determine epidemiological changes. A total of 327 MRSA isolates detected in 2002, 2007 and 2012 in the South-East Austria was spa typed, analysed with microarray techniques and assigned to epidemic strains.

Spa typing
Spa typing, DNA purification and PCR were performed as described previously (Grisold et al. 2009). The spa types and appropriate BURP clusters were assigned by using Ridom StaphType software (http://www.ridom.de/staphtype).

DNA microarray
For genetic characterization, diagnostic DNA microarray (Identibac, UK; StaphyType, Alere Technologies GmbH, Germany) was used, analysing the MRSA isolates for the presence of over 300 different genes. Protocols have been previously described in detail and are also provided by the kit's manufacturer (Monecke et al. 2007).

Statistical analyses
The statistical analyses were conducted using R R Version 3.2.1, a free software environment for statistical computing (www.r-project.org). Group-specific proportions were tested on their equality by a two-sided binomial test. On the respective SCCmec cassettes, the CC5 isolates differed clearly within the three years. In 2002, the CC5/SCCmecI (South German EMRSA) was predominant with 63 isolates, represented by the spa types t001, t041, t12555 and t12898 and in 2007 with 24 isolates, represented by the spa types t001, t002, t041 and t1301. In 2012, CC5/SCCmecI was found in four isolates (t001, t010, t041). The decline of the South German EMRSA was significant in both observed time periods, with P < 0.01 for each period. The CC5/SCCmecII (Rhine-Hesse EMRSA) was found in two isolates (t002) in 2002 and four isolates in 2007 (t002, t003, t014). In 2012, CC5/SCCmecII was found as the dominant clone, with 58/159 (37.18%) isolates (t002, t003, t014, t626, t10303), representing a significant increase between 2007 and 2012 (P < 0.01). In 2012, also three CC5-MRSA with SCCmecIV were detected, two of them t002, which were also positive for the PVL toxin (the only PVL-positive CC5 isolates) and one t579 without PVL. In 2012, also two CC5-MRSA isolates harbouring SCCmecV were detected, both of which were assigned to spa-type t688.
All other sporadic PVL-negative MLST CCs were represented by one or two isolates and were all detected in 2012: two CC361/SCCmecV (t315), one CC59/SCCmecIV (t216), one CC6/SCCmecV (t701) and one isolate CC130/SCCmecXI (t4335). This isolate was the only one with SCCmecXI, so therefore it did not harbour the mecA gene but the mecC gene.

DISCUSSION
Infections caused by MRSA have globally reached epidemic proportions and are well established both in the healthcare setting and in the community (Diederen and Kluytmans 2006;Chambers and Deleo 2009;Stryjewski and Corey 2014). In some countries such as the United States, MRSA is among the most common causes of nosocomial infections (Uhlemann et al. 2014). The prevalence rate of MRSA in bloodstream infections (BSIs) in Europe is monitored by the European Antimicrobial Resistance Surveillance Network (EARS-Net). Annual rates of MRSA from BSIs reach from 0.7% for Norway to 64.5% for Romania. For Austria, MRSA in BSI was reported with 9.2% in 2013 (European Centre for Disease Prevention andControl 2010, 2014).
With the occurrence of MRSA strains in different settings, hospitals as well as in outpatients or livestock associated, an increasing number of circulating MRSA clones was observed, the clones obviously in competition with each other (Huijsdens et al. 2006;Chambers and Deleo 2009;Otto 2013).
At the beginning of this century, the South German type (CC5) was very common in different parts of Europe, Germany or Croatia; also for Austria, the South German MRSA was the dominant clone in 2002 (Budimir et al. 2010). But within the next five years, the South German type disappeared and was replaced mainly by the Rhine-Hesse type; this clone was already common in other parts of Europe such as Great Britain or Ireland (UK-EMRSA-3) (Hookev, Richardson and Cookson 1998;Aucken et al. 2002) For Austria, as found in this study, the South German type was still the predominant clone until 2007 but disappeared later, with only four MRSA isolates in 2012. It could be noticed that the South German type was replaced by the Rhine-Hesse type, which was the predominant clone in 2012 with 38.18% of all investigated MRSA isolates (Enright et al. 2002;Chambers and Deleo 2009;Albrecht et al. 2011).
A similar delay could be observed for EMRSA 15. In many European countries, such as Germany, Ireland or Great Britain, EMRSA 15 is the predominant clone since the middle of the last decade (Amorim et al. 2007;Marchese et al. 2009;Grundmann et al. 2010;Albrecht et al. 2011). In Austria, the EMRSA 15 was found in one isolate only until 2012, but was the second most common clone in 2012, with 22.64%.
Beside CC5 and CC22, the CC8-MRSA, including the Irish ST8-MRSA-II and the USA300 clone, was continuously found throughout the study period. The Irish ST8-MRSA-II, represented by spa-type t190, was frequently found in 2002 (10.42%) and 2007 (20.83%); this spa type could not be found in any isolate in 2012. The high occurrence of spa190 in Austria was confirmed by a study from Ruppitsch et al. (2006) and Schmid et al. (2013), investigating MRSA from other parts of Austria, whereas there is no report of spa190 from neighbouring countries such as Germany (Ruppitsch et al. 2006;Cookson and HARMONY participants 2008;Albrecht et al. 2011;Schmid et al. 2013). In contrast CC45, the Berlin EMRSA clone which is epidemic in Germany was only found sporadically in Austria throughout the study period.
Although some other PVL-negative MRSA clones occurred sporadically throughout the study period, e.g. the USA600 (CC45) clone. The highest diversity of PVL-negative MRSA clones was found in 2012, including one CC130 isolate harbouring SCCmecXI with the new mecC variant, up to date only reported once for Austria. Isolates with SCCmecXI are described since only a few years, found both in humans and animals, such as cows (Smith and Cook 2005;Albrecht et al. 2011;García-Álvarez et al. 2011;Loncaric et al. 2013;Kerschner et al. 2014).
CA-MRSA showed high clonal diversity in this study with the majority of the strains belonging to a few predominant clones. The most epidemic clonal type among CA-MRSA in Europe is the European CA-MRSA clone (CC80/SCCmecIV), found in at least 11 European countries, as described by Rolo et al. For Austria, this clone was described first in 2001 and was the dominant CA-MRSA clone throughout the observed time period in this study (Krziwanek et al. 2007;Rolo et al. 2012).
The USA300 clone (CC8) is epidemic in the USA, where it became a predominant strain. Additionally to the USA, the USA300 spread all over the world, with reports from Asia, Australia, South America and throughout Europe, but this clone did not become the predominant CA-MRSA in these countries. For Austria, the first report was published in 2003 and USA300 is found up to now constantly, but in annually low numbers (Krziwanek et al. 2007;Chambers and Deleo 2009;Grisold et al. 2009). Since 2004, LA-MRSA/CC398 is also constantly found, mainly concentrated in two regions in Austria (Upper Austria and South-East Austria) with high density of pig farming (Krziwanek et al. 2009;Grisold et al. 2010). Eight CC1/ SCCmecIV with spa-type t127 and PVL toxin negative were detected. This LA-MRSA clone is known for harbouring numerous virulence factors and was first described in Austria in 2008, in that report associated with horses. In other European countries, the MRSA t127 clone was also isolated from pigs and cattle (Cuny et al. 2008;Franco et al. 2011;Alba et al. 2015).
This study investigated for the first time the molecular epidemiology of MRSA isolates in Austria within a 10-year period, but there are some limitations of this study. Although all primary MRSA isolates of a huge region in the South-East Austria were investigated, data from other Austrian regions could not be included. It was also not aim of this study to focus in detail on the clinical background or investigate the outcome of hospital hygiene regimes.
To summarise the diversity within the MRSA, population in South-East Austria increased within the observed period from 2002 to 2012. The results showed that the occurrence of some MRSA clones follows the European trends, but with a delay of five to ten years. CC22-MRSA (the Barnim clone), mostly represented in neighbouring countries such as Austria, did not appear before 2012, which means nearly 10 years later than in Germany. Local specifics like the occurrence of the spa-type t190 could be observed. Following the dynamics of MRSA, this study documents that new MRSA clones with new features could emerge at any time highlighting the importance of closely monitoring the epidemiology and genetic background of MRSA.

FUNDING
This study was supported by Pfizer Inc. (Pfizer reference number #W1182197).

ETHICAL STATEMENT
Isolates were obtained as part of routine diagnostic testing and were analysed anonymously. All data were collected in accordance with the European Parliament and Council decision for the epidemiological surveillance and control of communicable diseases in the European community (The European Parliament and the Council of the European Union. 1998; European Commission. 1999). Ethical approval and informed consent were thus not required. Patients were not physically involved and privacy of patients was provided by coding the tested specimens.