Distribution of the Most Prevalent Spa Types among Clinical Isolates of Methicillin-Resistant and -Susceptible Staphylococcus aureus around the World: A Review

Background: Staphylococcus aureus, a leading cause of community-acquired and nosocomial infections, remains a major health problem worldwide. Molecular typing methods, such as spa typing, are vital for the control and, when typing can be made more timely, prevention of S. aureus spread around healthcare settings. The current study aims to review the literature to report the most common clinical spa types around the world, which is important for epidemiological surveys and nosocomial infection control policies. Methods: A search via PubMed, Google Scholar, Web of Science, Embase, the Cochrane library, and Scopus was conducted for original articles reporting the most prevalent spa types among S. aureus isolates. The search terms were “Staphylococcus aureus, spa typing.” Results: The most prevalent spa types were t032, t008 and t002 in Europe; t037 and t002 in Asia; t008, t002, and t242 in America; t037, t084, and t064 in Africa; and t020 in Australia. In Europe, all the isolates related to spa type t032 were MRSA. In addition, spa type t037 in Africa and t037and t437 in Australia also consisted exclusively of MRSA isolates. Given the fact that more than 95% of the papers we studied originated in the past decade there was no option to study the dynamics of regional clone emergence. Conclusion: This review documents the presence of the most prevalent spa types in countries, continents and worldwide and shows big local differences in clonal distribution.


INTRODUCTION
Staphylococcus aureus, a leading cause of community-acquired and nosocomial infections, remains a major health problem around the world causing a variety of different conditions including wound infections, osteomyelitis, food poisoning, endocarditis, as well as more life-threatening diseases, such as pneumonia and bacteremia (Goudarzi et al., 2016b). Since the introduction of penicillin into medical therapy in the early 1940s, resistance against beta-lactams started to develop among staphylococcal isolates. To overcome this problem, a narrow spectrum semi-synthetic penicillin (methicillin) was introduced. However, soon after its first use in 1961, the first methicillinresistant S. aureus (MRSA) strain was identified (Turlej et al., 2011). Methicillin resistance is caused by the mecA gene product, a modified form of penicillin binding protein (PBP), called PBP2a or PBP2' , which has a lower affinity for all beta-lactam antibiotics (Hanssen and Ericson Sollid, 2006). The mecA gene is located within the mec operon of the staphylococcal cassette chromosome mec (SCCmec). SCCmec typing, which classifies SCCmec elements on the basis of their structural differences, is applied in several epidemiological studies of MRSA strains (Turlej et al., 2011). Molecular characterization of S .aureus is vital for the rapid identification of prevalent strains and will contribute to the control and prevention of S. aureus spread around healthcare settings if results are provided in real time (Siegel et al., 2007;Bosch et al., 2015;O'Hara et al., 2016). Phage typing was originally used for the formal typing of S. aureus isolates, but it was gradually replaced by pulsedfield gel electrophoresis (PFGE), the most recent gold standard method for the typing of S. aureus isolates (Bannerman et al., 1995;Murchan et al., 2003;Bosch et al., 2015). However, due to its laborious character and difficulties in exchanging data between laboratories, and the requirement for inter-laboratory standardization, PFGE was replaced by multi-locus sequence typing (MLST) and staphylococcal protein A (spa) typing (Harmsen et al., 2003). MLST is a great tool for evolutionary investigations and differentiates isolates according to nucleotide variations in 7 housekeeping genes. Spa typing, which relies only on the assessment of the number of and sequence variation in repeats at the x region of the spa gene, exhibits excellent discriminatory power and has become a useful typing tool for the sake of its ease of performance, cheaper procedure, and standardized nomenclature (Frenay et al., 1996;Koreen et al., 2004;Strommenger et al., 2008;Bosch et al., 2015;Darban-Sarokhalil et al., 2016;O'Hara et al., 2016). The spa gene contains three distinct regions: Fc, X, and C (Verwey, 1940;Harmsen et al., 2003;Goudarzi et al., 2016b). The polymorphic X region, which encodes a part of the staphylococcal protein A (Spa), contains variations in the number of tandem repeats and the base sequence within each repeat. In other words, each new sequence motif, with a length of 24 bp, found in any S. aureus strain is assigned a unique repeat code and the repeat succession and the precise sequences of the individual repeats for a given strain determines its spa type (Mazi et al., 2015). The primary binding site for protein A is the Fc region of mammalian immunoglobulins, most notably IgGs, which renders the bacteria inaccessible to opsonins, thus impairing phagocytosis by immune system attack (Graille et al., 2000).
According to the literature, the prevalence of spa types among S. aureus isolates varies in different areas around the world. According to the authors' knowledge, no comprehensive data, during the last decade, have been made available on the distribution of diverse spa types within different geographical areas, so the aim of the present study was to review the literature to report the most common clinical spa types which is important for discriminating S. aureus outbreak isolates and nosocomial infection control policies worldwide.
SeqNet.org has shown the 10 most frequent spa types on the seqNet during 2004-2008 which includes only the European countries plus Lebanon. These data seem to include MRSA from both human and veterinary sources which is different from the present review which includes only human clinical data and a larger geographic domain.

Search Strategy and Selection Criteria
The PubMed, Google Scholar, Web of Science, Embase, Cochrane library, and Scopus databases were searched for original articles, reporting regionally prevalent spa types among S. aureus isolates. The search terms were "Staphylococcus aureus, spa typing." The articles were selected according to evaluations on titles, abstracts and the main text. The reasons for exclusion of certain articles were: non-human clinical isolates of S. aureus, old spa-typing methods (e.g., RFLP), non-English articles, and isolates from patients with certain specific diseases including HIV and other immune mediated afflictions. Following main text assessment, several articles were excluded for unspecified sample size or the lack of a predominant spa type in the study. Studies presented only in abstract form were also excluded. Papers occurring in more than a single database were cited once.

Data Extraction
The following data were extracted from each article: first author's name, year of publication, country, number of isolates, number of methicillin resistant and susceptible S. aureus isolates (MRSA and MSSA), the predominant spa type, SCCmec types of the predominant spa types, MLST and spa clonal complexes belonging to the most prevalent spa types.
The Spa server has identified 17625 different spa types until the 17th of December, 2017 1 . Table 2 illustrates the distribution of diverse spa types among various SCCmec types in different continents. In Europe, 52 studies performed SCCmec typing on 3208 spa types and SCCmec types IV (1830 isolates) and II (800 isolates) were most associated and SCCmec type V (126 isolates) was least associated with the most common spa types. In Asia, SCCmec typing was performed on 4179 spa types by 41 studies and the most common spa types were classified into SCCmec types III (2725 isolates) and II (677 isolates), whilst the least number of spa types were categorized into SCCmec type V (104 isolates). A total of 12 studies in America performed SCCmec typing on 531 spa types, showing that SCCmec types IV (238 isolates) and II (167 isolates) were most associated with the frequent spa types. In Africa, 5 studies assessed the SCCmec types of 615 spa types and the common spa types were classified into SCCmec types IV (217 isolates) and V (185 isolates) whilst the least number of spa types were categorized into SCCmec type I (37 isolates). Finally, in Australia SCCmec typing was performed on 107 spa types by 5 studies and the most common spa types were classified into SCCmec types IV (49 isolates) and III (40 isolates), whilst the least number of spa types were categorized into SCCmec type I and II.
The total number of MRSA and MSSA isolates of the 3 most common spa types among different continents are shown in Table 3. In Europe, all the isolates related to spa type t032 were MRSA. In addition, spa type t037 in Africa and t037 and t437 in Australia were MRSA as well.
Spa clonal complex (S-CC) and MLST clonal complex (M-CC), plus the sequence types (STs) of the most common spa types among different continents are illustrated in Table 4. The number of studies that reported spa clonal complex for the common spa types were 30 in Europe, 12 in Asia, 10 in America and 9 in Africa. Common spa types categorized into distinct MLST clonal complexes were 43 in Europe, 19 in Asia, 14 in America, and 9 in Africa. Forty eight studies in Europe, 29 in Asia, 18 in America, and 11 studies in Africa assessed the sequence types of the most common spa types. In Australia, no studies reported the spa or MLST complexes, nor any sequence types for the common spa types assessed.
The association of the most prevalent spa types with different countries among different continents is shown by Table S1 in Supplementary Material. The data exhibit that The Netherlands has reported the most diverse range of spa types (34 types), followed by China (22 spa types), Germany (16 types), UK (15 types), Spain (11 types), Sweden and USA (10 spa types each), Italy and Iran (8 spa types each), France and Portugal (7 spa types each) and Switzerland (6 spa types).
Dissemination of different spa types among different countries is illustrated by Table S2 in Supplementary Material. The spa types t008 and t002 were the most frequently repeated spa types among the others, each repeated in 16 countries among different continent. The next most frequently repeated spa types were respectively t037 (12 countries), t044 (11 countries), t084 (8 countries), t012 and 127 (7 countries each), t041 (6 countries), and t019, t011, t034, t355, t189, t304 (5 countries each). Almost 50% of the spa types (43 out of the 87) were only reported by 1 country.

DISCUSSION
Staphylococcus aureus is capable of adapting to a variety of conditions and successful clones can be epidemic and even pandemic as can be concluded by their spreading from one continent to another . The current review reports the prevalence of spa types among clinical isolates, both as carriage and infectious isolates, across the world. Our analysis showed that t032 was the most prevalent spa type in Europe, predominantly centered in the UK and Germany (Figure 3), and among the 5 most predominant spa types in Austria. No other countries in Europe have reported t032 among its most frequent spa types. Moreover, t008 was the second most prevalent and, nonetheless, the most frequently identified spa type in the various European countries, distributed among 11 out of the 22 of them investigating local spa types while also being the predominant type in France and Italy (Table S2 in Supplementary Material). Germany and UK principally provided a larger sample size compared to other European countries, looking over 10081 and 2644 isolates, respectively. Despite the fact that a larger sample size could be a proof to the validity of acquired data, it might also be that the disparity of the sample size among different countries has caused deviance in the report of the most prevalent spa type in Europe by the present study. Sweden appeared to be the only European country to have t002 as its most predominant spa type, even though t002 was disseminated in 9 out of the 22 European countries included in this analysis. A comprehensive molecular-epidemiological analysis, investigating the geographical distribution of invasive S. aureus isolates in Europe , revealed that the 3 most common spa types in Europe were t032, t008, and t002, respectively; which was in agreement with the results  of the current meta-analysis. In Asia, t030 was the predominant spa type mainly located in China (Figure 3), while also reported by Iran as the fifth most common spa type. Moreover, t037, as the second most common spa type in Asia, was reported by more Asian countries compared to other spa types (Korea, China, Taiwan, Iran and Malaysia out of 10 Asian countries under this survey). Similarly in Africa, t037 was the most prevalent and t084 and t064 the most frequently repeated spa types, reported by 3 African countries each. Even though t008 was the most prevalent spa type in America, it was only reported by the USA and Canada. Then again, t002, as the second most common spa type was distributed among the USA, Canada and Brazil. Again, for these 3 continents the distinct sample size variation within the conforming countries might account for the different reports of the prevalent spa types among the associated continents.
In Australia no precise information was revealed about the distribution of spa types. The spa typing method, although being one of the valid schemes for the epidemiological surveillance of S. aureus, only considers a very limited portion of the whole genome and, therefore, could not possibly reflect the mutational events occurring in other parts of this organism's genome. Since certain spa types are still restricted to particular geographic locations, it might be considered that the polymorphic X region and, hence, the type of protein A have possible associations with the organism's adaptations to diverse conditions such as different host populations, the weather and geographical diversity.  (7), t202 (42) t437 (12), t011 (6) *Unknown spa type.   As a vital virulence factor which enables the escape of S. aureus from innate and adaptive immune responses, the Spa protein may be an important target for adaptive evolution by means of host specialization and other environmental factors (Santos-Júnior et al., 2016). The plasticity of the spa gene, as a result of intragenic recombination, non-synonymous mutations as well as duplications events, can indeed influence the pathogenicity of S. aureus 1 . It has been shown that the mosaic spa gene is composed of different segments, each with a distinct evolutionary histories which could provide S. aureus with increased fitness to colonize the host surfaces or bind the immunoglobulin subunits. This diversity of Spa domains has contributed to the epidemic phenotype of S. aureus strains implying that they represent selected adaptations to their environment (Santos-Júnior et al., 2016).
Considering the fact that the primary binding site for protein A is the Fc region of mammalian immunoglobulins, and most notably IgGs (Graille et al., 2000), one possible justification for such an association might be the likely difference in the incidence rates of immunoglobulin subclasses among different geographical populations and, hence, the different binding strength of protein A types to these immunoglobulins. This might consequently cause a difference in the extent of opsonization and phagocytosis and, hence, the survival rates of particular S. aureus spa types within different populations (Sasso et al., 1991) Overall, t008 (2692 MRSA, 258 MSSA, 222 uncertified) and t002 (9364 MRSA, 189 MSSA and 1441 uncertified) were the most widely distributed spa types worldwide, disseminated each through 16 out of the 34 countries assessing spa types, followed by t037 (1971 MRSA, 51 MSSA, 62 uncertified) and t044 (590MRSA, 0 MSSA, 77 uncertified) respectively occurring in 12 and 11 countries worldwide. Almost half of the spa types (43 out of the 87) were yet localized and limited to 1 country each (Table S2 in Supplementary Material). Migrations from one country/continent to another provides a reasonable justification as to why some spa types are common between certain countries/continents. In Europe, all the isolates related to spa type t032 were MRSA isolates. In addition, spa type t037 in Africa and t037and t437 in Australia consisted only of MRSA isolates; however, as shown in Table 3, the majority of predominant spa types consist of both MRSA and MSSA isolates (Adler et al., 2012;Jiménez et al., 2013;Aiken et al., 2014). Here again, a notable number of studies have not deduced whether the predominant spa types are MRSA or MSSA and there is therefore some missing points in the data regarding the association of prevalent spa types and methicillin resistance among different continents. Furthermore, results are dependent on the original sample collection to be spa typed. Most studies have, in the first place, spa typed methicillin resistant S. aureus isolates because of their epidemiological importance among clinical settings (Ruppitsch et al., 2006;Zhang et al., 2008;Miller et al., 2011) and therefore no specific conclusion is to be invoked as to whether MRSA/MSSA isolates belong to specific spa types or vice versa. In Europe, SCCmec types IV and II were most associated with the common spa types. In Asia, the most common spa types were classified into SCCmec types III and II. In America, SCCmec types IV and II were most associated with frequent spa types. In Africa, the common spa types were classified into SCCmec types IV and V and finally, in Australia the most common spa types were classified into SCCmec types IV and III. The spa and SCCmec typing methods focus on two distinct locations within the genome of S. aureus. The last SCCmec type reported in 2015 in Germany was the SCCmec type XII (Wu et al., 2015), whereas the studies assessed in this review, have only ascertained limited SCCmec types (I, II, II, IV, and V). Moreover, a significant number of spa types have not been associated to any specific SCCmec type and the number of studies which have assessed SCCmec typing for the prevalent spa types are limited. For the above mentioned reasons, the association between certain spa and SCCmec types found in this review might be of questionable reliability.
Data relating to the spa and MLST clonal complexes, and sequence types of the most common spa types revealed that the spa clonal complexes (S-CC) 001 and 002 were common among Europe and Asia and had the highest association with prevalent spa types in this continents. Similarly, S-CC012 contained some frequent spa types reported by Europe, America and Africa while S-CC84 was only common among America and Africa. This means that some related spa types exist among different continents. On the other hand, MLST clonal complex (M-CC) 5 was associated with prevalent spa types in Europe, Asia, America and Africa. Meanwhile, some of the most frequently encountered spa types were associated to M-CC 8 which were common among Asia, America and Africa. It seems that there is a virtually sustained association between the spa and sequence types irrelevant of the continent. For example, t032 has almost always been associated with ST22 across all continents; the same is true for t008 which has been associated with either ST8 or ST247, among all the studies being assessed in this review. As some of the most prevalent spa types reported by many different studies, t002, t030, and t037 have been constantly associated with ST5; ST239 and ST22; and ST239, respectively. In Australia, no studies reported the spa, MLST complexes or sequence types for the common spa types assessed. S. aureus, as an organism with a relatively stable genome, tends to present as clones which are relatively stable and generally diversify by the accumulation of single nucleotide substitutions without frequent inter-strain recombination Shittu et al., 2011;van der Donk et al., 2013b). It is also noteworthy to mention that S. aureus clones might vary among different clinical settings within the same country or even among different wards of the same hospital (Shittu et al., 2011;van der Donk et al., 2013b;Seidl et al., 2015). Since a majority of studies under this review did not specifically discern the exact location of sampling, the data presented in this review presents a general information about the prevalent spa types and the associated clonal complexes in each country/continent; so, it would have been valuable to provide information on the exact sampling time and location within each country among different continents.

CONCLUSION
This review shows the spread of the most prevalent spa types in countries, continents and worldwide. Such data can be used for epidemiological purposes, such as defining the geographical spread of the predominant spa types of S. aureus, the interpretation of relative frequencies, comparing the worldwide diverse evolutionary trajectories of S. aureus lineages, and the understanding of molecular epidemiological dynamics of S. aureus transmission.

AUTHOR CONTRIBUTIONS
DD-S designed the first concept, helped in the literature review, data extraction, and preparation of the manuscript.
PA prepared the manuscript, interpreted the data, and helped in the literature review and data extraction. NF helped in the literature review and data extraction. MM, SSK, and MD participated in the manuscript preparation. AvB made critical revision and helped in the preparation of the manuscript. KA helped in the analysis, interpretation of data, and preparation of the manuscript. All authors read and confirmed the content of the paper.