The Dissemination and Molecular Characterization of Clonal Complex 361 (CC361) Methicillin-Resistant Staphylococcus aureus (MRSA) in Kuwait Hospitals

Methicillin-resistant Staphylococcus aureus (MRSA) belonging to clonal complex 361 (CC361-MRSA) is rare among patients’ populations globally. However, CC361-MRSA has been isolated with an increasing trend among patients in Kuwait hospitals since 2010. This study investigated the molecular characteristics of CC361-MRSA isolated from patients in Kuwait hospitals in 2016–2018 to understand their genetic relatedness and virulence determinants. Of 5,223 MRSA isolates investigated by DNA microarray, 182 (3.4%) isolates obtained in 2016 (N = 55), 2017 (N = 56), and 2018 (N = 71) were identified as CC361-MRSA. The CC361-MRSA isolates were analyzed further using antibiogram, spa typing and multi locus sequence typing (MLST). Most of the isolates were resistant to fusidic acid (64.8%), kanamycin (43.4%), erythromycin (36.3%), and clindamycin (14.3%) encoded by fusC, aphA3, and erm(B)/erm(C) respectively. Nine isolates (4.9%) were resistant to linezolid mediated by cfr. The isolates belonged to 22 spa types with t3841 (N = 113), t315 (N = 16), t1309 (N = 14), and t3175 (N = 5) constituting 81.3% of the spa types, four genotypes (strain types), CC361-MRSA-[V/VT + fus] (N = 112), CC361-MRSA-IV, WA MRSA-29 (N = 36), CC361-MRSA-V, WA MRSA-70/110 (N = 33) and CC361-MRSA-[V + fus] variant (N = 1). MLST conducted on 69 representative isolates yielded two sequence types: ST361 (11/69) and ST672 (58/69). All CC361-MRSA isolates were positive for cap8, agr1, and the enterotoxin egc gene cluster (seg, sei, selm, seln, selo, and selu). The tst1 was detected in 19 isolates. The immune evasion cluster (IEC) genes type B (scn, chp, and sak) and type E (scn and sak) were detected in 20 and 152 isolates, respectively. The CC361-MRSA circulating in Kuwait hospitals consisted of two closely related sequence types, ST361 and ST672 with ST672-MRSA [V/VT + fus] as the dominant genotype. The dissemination of these newly emerged clones and the emergence of linezolid resistance limits therapeutic options, as well as present significant challenges for the control of MRSA infections in Kuwait hospitals.


INTRODUCTION
Methicillin-resistant Staphylococcus aureus (MRSA) remains a major universal healthcare problem as it causes a wide range of infections including skin and soft tissue infections (SSTI), pneumonia, bacteremia, endocarditis, and osteomyelitis (McCaig et al., 2006;Hassoun et al., 2017). The epidemiology of MRSA has changed significantly since its description in the 1960s. MRSA has evolved from being an exclusive healthcareacquired pathogen (healthcare-acquired MRSA (HA-MRSA) (McCaig et al., 2006), to being acquired outside the healthcare facilities in the communities (Community-associated MRSA, CA-MRSA) (Udo et al., 1993;McCaig et al., 2006;David and Daum, 2010). Recently, MRSA has also evolved to become a livestock-associated pathogen designated Livestock-associated MRSA (Armand-Lefevre et al., 2005;van Cleef et al., 2011). HA-MRSA are characteristically multi-resistant to antibiotics and carry SCCmec types I, II or III . In contrast, CA-MRSA are usually more susceptible to nonbeta-lactam antibiotics and carry SCCmec types IV, V, or VI (David and Daum, 2010). The Livestock-associated MRSA (LA-MRSA) initially caused major problems in agriculture and were the leading cause of bovine mastitis (Fluit, 2012) but have now become prominent among livestock, people associated with livestock and those with no previous contact with livestock (Graveland et al., 2011;Köck et al., 2013;Wagenaar et al., 2009;Boswihi et al., 2020a).
Molecular typing tools including staphylococcal protein A (spa) typing, multilocus sequence typing (MLST), pulsedfield gel electrophoresis, staphylococcal cassette chromosome mec (SCCmec) typing, DNA microarray, and whole genome sequencing (WGS) have been used in the epidemiologic surveillance of MRSA strains to detect and monitor emerging and reemerging infections as well as monitoring geographic spread and shifts of epidemic and endemic clones (Pfaller, 1999;van Belkum et al., 2001;Monecke et al., 2011;Boswihi et al., 2016Boswihi et al., , 2020aRebic et al., 2016). The application of these typing techniques to type MRSA from different geographic regions have shown that most of the MRSA infections reported worldwide were caused by a limited number of pandemic MRSA clones belonging to clonal complexes 5 (CC5), CC8/ST239, CC22, CC30, and CC45 (Robinson and Enright, 2003;Monecke et al., 2011;Guthrie et al., 2020) although CA-MRSA isolates belong to more diverse genetic backgrounds compared to HA-MRSA Udo, 2013;Tong et al., 2015;Guthrie et al., 2020).

MRSA Isolates
The MRSA isolates used in this study were obtained as part of routine diagnostic microbiology investigations. The MRSA were cultured and identified, using traditional diagnostic bacteriological methods including Gram stain, growth on Mannitol Salt Agar, positive DNAse and tube coagulase tests. The isolation and identification of the isolates were performed in the diagnostic microbiology laboratories where initial antibiotic susceptibility testing was also performed with VITEK MS (bioMérieux, Marcy l'Etoile, France). Pure cultures of isolates on blood agar plates were submitted to the Gram-Positive Bacteria Research laboratory, located at the Department of Microbiology, Faculty of Medicine, Kuwait University, where the isolates were retested for purity and preserved in 40% glycerol (v/v in brain heart infusion broth) at −80 • C for further analysis. The isolates were recovered by two subcultures on brain heart infusion agar at 35 • C before analysis. In total, 5,223 MRSA isolates were received from 13 different hospitals in Kuwait between 1 January, 2016 and 31 December, 2018. DNA microarray analysis performed on the 5,223 isolates revealed that 182 (3.4%) isolates were identified as CC361-MRSA. The 182 CC361-MRSA isolates were investigated further and reported in this study. The 182 CC361 isolates were collected from patients in Adan hospital (N = 37; 20.3%), Mubarak hospital (N = 32; 17.6%), Sabah hospital (N = 23; 12.4%), Maternity hospital (N = 20; 10.9%), Al-Amiri hospital (N = 19; 10.4%), Al-Razi hospital (N = 15; 8.2%), Chest Disease hospital (N = 13; 7.1%), Al-Jahra hospital (N = 8; 4.4%), Al-Farwaniya hospital (N = 8; 4.4%), KOC hospital (N = 3; 1.6%), Ibn-Sina hospital (N = 2; 1.1%), Dasman Diabetic Centre (N = 1; 0.5%) and Army Force hospital (N = 1; 0.5%).

Staphylococcus Protein A (spa) Typing
Spa typing was performed using protocol and primers published previously (Harmsen et al., 2003). Bacterial DNA isolation for amplification studies was performed as described previously (Boswihi et al., 2018). Three to five identical colonies of an overnight culture were picked using a sterile loop and suspended in a microfuge tube containing 50 µL of lysostaphin (150 µg/mL) and 10 µL of RNase (10 µg/mL) solution. The tube was incubated at 37 • C in the heating block (ThermoMixer, Eppendorf, Hamburg, Germany) for 20 min. To each sample, 50 µL of proteinase K (20 mg/mL) and 150 µL of Tris buffer (0.1 M) were added and mixed by pipetting. The tube was then incubated at 60 • C in the water bath (VWR Scientific Co., Shellware Lab, United States) for 10 min. The tube was transferred to a heating block at 95 • C for 10 min to inactivate proteinase K activity. Finally, the tube was centrifuged, and the supernatant containing extracted DNA was stored at 4 • C till used for PCR.
The PCR protocol consisted of an initial denaturation at 94 • C for 4 min, followed by 25 cycles of denaturation at 94 • C for 1 min, annealing at 56 • C for 1 min, and extension for 3 min at 72 • C, and a final cycle with a single extension for 5 min at 72 • C. Five µL of the PCR product was analyzed by 1.5% agarose gel electrophoresis to confirm amplification. The amplified PCR product was purified using Micro Elute Cycle-Pure Spin kit (Omega Biotek, Inc., United States) and the purified DNA was then used for sequencing PCR. The sequencing PCR product was then purified using Ultra-Sep Dye Terminator Removal kit (Omega Bio-tek, Inc., United States). The Purified DNA was sequenced in an automated 3130x1 genetic analyzer (Applied Biosystem, United States). The sequenced spa gene was analyzed using the Ridom Staph Type software (Ridom GmbH, Wurzburg, Germany).

DNA Microarray
DNA microarray analysis was performed using the Identibac S. aureus genotyping kit 2.0 and the ArrayMate reader (Alere Technology, Jena, Germany) as described previously by Monecke et al. (2011). The DNA microarray analysis was used for the simultaneous detection of SCCmec types, antibiotic resistance genotypes and virulence related genes, including PVL, genes encoding species markers, and to allocate clonal complex (CC). S. aureus genotyping array is presented in an ArrayStrip format which contains 336 probes printed onto an array located in the bottom of the ArrayStrip. MRSA isolates were grown on blood agar plates at 35 • C overnight. DNA extraction of the overnight culture was performed as described by the manufacturer using Identibac S. aureus genotyping kit 2.0 (Alere, GmbH, Germany). Linear amplification of the purified DNA was performed in a total of 10 µL of the reaction volume containing 4.9 µL of B1 (labeling reagent), 0.1 µL of B2 (DNA polymerase), and 5 µL of the purified DNA. The PCR protocol consisted of an initial denaturation for 5 min at 96 • C, followed by 50 cycles of denaturation for 60 s at 96 • C, annealing for 20 s at 50 • C, and extension for 40 s at 72 • C. hybridization and washing of the labeled arrays were performed as previously described . The array was scanned using the ArrayMate reader (CLONDIAG, Alere, Germany) and the image of the arrays was recorded and analyzed using IconoClust software plug-in (CLONDIAG). The result was interpreted as negative, positive, or ambiguous by the software.

Multilocus Sequencing Typing (MLST)
MLST was performed for representative isolates belonging to different spa types. The amplification of the seven housekeeping genes was performed using previously described M13-tailed primers (Tan et al., 2006). The amplified targets were sequenced with one pair of M13tailed primers: 5 -TGTAAAACGACGGCCAGT-3 and 3 -CAGGAAACAGCTATGACC-5 . The sequencing PCR protocol consisted of initial denaturation for 1 min at 94 • C, followed by 25 cycles of denaturation for 10 s at 96 • C, annealing at 55 • C for 5 s, and extension for 4 min at 66 • C. DNA sequencing was performed using a 313091 genetic analyzer (Applied Biosystems, Foster City, CA, United States) in accordance with the manufacturer's protocol. The sequences were submitted to http://www.pubmlst.net/ where an allelic profile was generated and the sequence type (ST) assigned.
The distribution of the antibiotic resistance genes of the CC361-MRSA isolates is presented in Table 2. There was general concordance between resistance phenotypes and genotypes. The penicillin resistance encoding genes blaZ and its regulatory genes, blaI and blaR were found in 177 isolates, while aminoglycoside (gentamicin and kanamycin) encoding genes aacA-aphD, aphA3, and aadD were detected in 12, 65, and 2 isolates, respectively. The five isolates that were negative for blaZ, blaI, and blaR, were susceptible to penicillin G and negative for penicillinase production but were positive for mecA and SCCmec genetic elements.
Whereas all of the CC361-MRSA [V/VT + fus) isolates were resistant to fusidic acid mediated by fusC, the CC361-MRRSA-V-WA-MRSA-70/110 isolates were more resistant to erythromycin mediated by erm(C), high-level mupirocin resistance mediated by mupA, and chloramphenicol and linezolid mediated by cfr.

Prevalence of Virulence-Related Genes
The prevalence of the virulence factors amongst the CC361 MRSA isolates are presented in Table 3. All 182 isolates were positive for genes encoding the accessory gene regulator type I (agrI) and capsular polysaccharide type 8 (cap8) but lacked genes for Panton Valentine Leukocidin (PVL).
The isolates varied in the carriage of genes for the immune evasion cluster (IEC). Twenty isolates carried immune evasion cluster genes of type B (scn, chp, and sak), whilst 152 isolates carried the immune evasion cluster genes of genes for type E (scn and sak). Ten isolates were negative for the IEC genes.
The hemolysin encoding genes hlgA, hl, and hlIII, were detected in all 182 isolates. However, hlb was detected in 181 of the 182 isolates while hla was detected in 172 isolates (94.5%). In addition, all 182 isolates were positive for genes encoding clumping factors A and B (clfA and clfB), fibronectin-binding proteins A and B (fnbA and fnbB), Staphylococcus aureus surface protein G (sasG) and major histocompatibility complex class II analog protein (map). All isolates lacked cna that codes for the collagen-binding adhesin.

DISCUSSION
This study has demonstrated a steady expansion of the CC361-MRSA lineage among patients in Kuwait hospitals in recent years. Until recently, CC361-MRSA was reported sporadically from human patients in Oman (Udo et al., 2014), Abu Dhabi (Weber et al., 2010), Australia , Ireland (Kinnevey et al., 2014), Bangladesh (Afroz et al., 2008), Kuwait (Boswihi et al., 2016), Saudi Arabia (Senok et al., 2019), and Switzerland (Etter et al., 2020) as well as in monkeys in Nepal (Roberts et al., 2019), in cattle in Czech Republic (Tegegne et al., 2017), and in ready-to-eat food in Bangladesh (Islam et al., 2019). However, this report represents the largest number of CC361-MRSA reported to date. The number of CC361-MRSA obtained from human patients increased from two isolates identified for the first time in Kuwait in 2010 (Boswihi et al., 2016) (Table 1) suggesting its higher transmission capacity. The acquisition of the variant V/VT genetic element represents an evolutionary event that probably confers greater capacity to spread. Although this clone appears to be restricted to the countries of the Gulf at this time, it has the potential to spread widely as the highly virulent USA300 MRSA clone that emerged initially as a less virulent and less resistant clone (Strauß et al., 2017). The USA300 MRSA clone gradually acquired multiple antibiotic resistance, SCCmec Iva genetic element, genes for PVL, arginine catabolic mobile element, and a specific mutation in capsular polysaccharide gene, capSE and became highly transmissible following its introduction to North America from Central Europe (Strauß et al., 2017).
The CC361-MRSA strains appear to have appeared independently in the GCC countries. It was first reported in Abu Dhabi, UAE, from patients in 2009 (Weber et al., 2010) followed by the report in two patients in Kuwait in 2010 (Boswihi et al., 2016) and in two patients in Oman in 2011 (Udo et al., 2014). However, prior to their emergence in the GCC countries, an isolate of ST361-MRSA-IV was isolated from a patient in Bangladesh in 2004 (Afroz et al., 2008) and two isolates of ST672-MRSA-V were reported among MRSA isolates obtained in 2004in India (Shambat et al., 2012 supporting the recent emergence of CC361-MRSA in the GCC countries. The isolates belonged to 22 different spa types with t3841 (N = 112; 61.51%), t315 (N = 16; 8.8%), and t1309 (N = 14; 7.7%) constituting 78 percent of the isolates. Whereas t3841 and t1309 were associated with ST672 distributed in three strain types, t315 was associated only with ST361 in a single strain type, CC361-MRSA-V, WA MRSA 70/110, in this study. Similarly, t315 has been associated only with ST361 in studies conducted in Australia , Ireland (Kinnevey et al., 2014), Austria (Zarfel et al., 2016), and from cattle in Czech Republic (Tegegne et al., 2017) where spa typing were also reported. In contrast, ST672 isolates were associated with t1309 in studies that reported spa types in Australia  and India (Shambat et al., 2012). These studies support the recent acquisition of the ST672-MRSA [V/VT + Fus]/t3841 strain in Kuwait. The origin of the dominant t3841 isolates in this study is not clear since previous ST672 isolates have largely been associated with t1309 Shambat et al., 2012). However, a previous study in Kuwait detected t3841 among MSSA isolates in the country (Vali et al., 2017). It is possible that the t3841 MRSA isolates in this study emerged from a locally circulating MSSA isolate that acquired mecA. An isolate of t3841 MSSA had also been reported in India (Shambat et al., 2012).
Besides the diversity in spa types, the insolates in this study harbored different SCCmec types consisting of SCCmec types IV, V and V/VT that are usually associated with communityassociated genotypes. Apart from SCCmec V/VT, SCCmec types IV and V were also reported in CC361-MRSA from human patients in Western Australia , Ireland (Kinnevey et al., 2014) and Bangladesh (Afroz et al., 2008), Abu Dhabi (Weber et al., 2010), Oman (Udo et al., 2014) and Saudi Arabia (Senok et al., 2019) and in animals in Nepal (Roberts et al., 2019) and Czech Republic (Tegegne et al., 2017). CC361 isolates were also reported to harbor SCCmec VIII in Australia where it is known as WA MRSA-28 (Nimmo and Coombs, 2008). These observations demonstrate the ability of CC361 to acquire different SCCmec elements. The detection of SCCmec V/VT together with spa type t3841 only in the current isolates supports the recent emergence of these strains in the GCC countries.
The CC361-MRSA isolates were resistant to different antibiotics including gentamicin, kanamycin, erythromycin, clindamycin, tetracycline, trimethoprim fusidic acid and highlevel mupirocin and harbored aacA-aphD, aphA3, msrA, tet(K), dfrS1, fusC, and mupA mediating resistance to the corresponding  Toxic shock syndrome toxin-1 (TSST-1) antibiotics as has been reported in isolates from Saudi Arabia (Senok et al., 2019), Abu Dhabi , and Australia . All the 112 ST672-MRSA-[V/VT + fus] isolates harbored fusC and was responsible for the high prevalence of fusidic acid resistance in this study. Fusidic acid resistance mediated by fusC was also reported in CC361-MRSA isolates obtained from dental room environment and human patients in Saudi Arabia (Senok et al., 2019). Fusidic acid resistance has remained a major problem in MRSA isolates in Kuwait for some years, and has been suggested to be due to independent acquisition of fusidic acid determinants in isolates belonging to diverse backgrounds, and the consumption of over the counter fusidic acid preparations which are available without prescription in the country (Boswihi et al., 2018). We detected linezolid resistance in nine CC361-MRSA-V, WA MRSA-70/110 isolates that were resistant to chloramphenicol and linezolid, and were positive for cfr that encodes resistance to linezolid, chloramphenicol, lincosamides, and streptogramin A (Pillai et al., 2002;Long et al., 2006;Morales et al., 2010). This is the first report of linezolid resistance in S. aureus in Kuwait. Linezolid resistance mediated by cfr was recently reported in a recent MRSA isolate obtained in the United Arab Emirate (Senok et al., 2020) indicating that linezolid resistance is emerging in these countries and should raise awareness of an emerging problem. In contrast, none of the CC361-MRSA-V, WA MRSA-70 reported in Australia expressed resistance to linezolid or high-level mupirocin . Since its approval for clinical use, Linezolid has remained an important treatment option for treating infections caused by MRSA and glycopeptide-resistant enterococci (Pillai et al., 2002). Therefore, the emergence of linezolid resistance in MRSA observed in this study is concerning because it will limit treatment options available for MRSA infections. The 10 high-level mupirocinresistant isolates harbored mupA that codes for this resistance. The mupA mediated high-level mupirocin resistance have been previously in MRSA belonging to other genetic backgrounds in Kuwait (Udo et al., 2001;Udo and Sarkhoo, 2010).
The CC361-MRSA isolates in this study were negative for genes encoding PVL similar to CC631-MRSA isolates reported previously in Oman (Udo et al., 2014), UAE (Senok et al., 2020), Australia , and most of the isolates in Saudi Arabia (Senok et al., 2019). However, two isolates consisting of, an isolate of CC361-MRSA-V/VT (PVL+) isolated in Saudi Arabia (Senok et al., 2019) and an isolate of ST361-MRSA-/t315 isolated from ready-to-eat food sample in Bangladesh (Islam et al., 2019), were positive for PVL suggesting that PVL is rare among CC361-MRSA isolates.
Most of the CC361-MRSA isolates in this study harbored type E (scn, sak) (N = 152; 83.5%) or type B (scn, chp, and sak) (N = 20; 11.0%) immune evasion cluster genes. Likewise, CC361-MRSA reported in Australia  Saudi Arabia (Senok et al., 2019), and United Arab Emirates (Senok et al., 2020) also carry either type E or type B immune evasion cluster (IEC) genes. The immune evasion cluster genes binds specifically with compounds of the human innate immune system to protect bacteria from the human innate immune system (Resch et al., 2013;Sieber et al., 2020) and are therefore used to distinguish S. aureus of human from those of animal origin since human isolates harbor the IEC genes (Resch et al., 2013). The presence of IEC carrying S. aureus in animals usually suggests human contamination (Resch et al., 2013;Sieber et al., 2020). Viewed from this perspective, the presence of type E (scn and sak) IEC in CC361-MRSA that were recovered from monkeys (Roberts et al., 2019) which is similar to the IEC content of 94.5% of our isolates may suggest a human origin for the monkey isolates. Although CC361-MRSA have also been isolated from cattle (Tegegne et al., 2017), and ready to-eatfood (Islam et al., 2019) the IEC genes contents of these isolates were not reported. Hence it is not possible to speculate their origin. Ten of the isolates in this study consisting of five ST672-MRSA-V/VT + Fus/t3841 and five ST672-MRSA-IV, WAMRSA 29/t14690/t14271/t1309/t3175/3841 were negative for IEC genes suggesting the loss of the bacteriophage that bears the IEC genes.

CONCLUSION
In conclusion, this study reports an increase in the prevalence of the CC361-MRSA isolates with the dominance and transmission of a newly emerged ST672-MRSA [V/VT + fus] genotype in Kuwait hospitals. The CC361-MRSA isolates expressed resistance to different antibiotics including linezolid resistance observed for the first time in Kuwait. The isolates were negative for genes encoding PVL but harbored common enterotoxins encoding genes (e.g., egc gene cluster). The detection of the various virulence genes in these isolates and their isolation from different clinical samples indicate their capacity to cause serious infections like other virulent MRSA lineages. Continuous surveillance is necessary to monitor and evaluate emerging MRSA clones in Kuwait to assist in developing better means of prevention and management.

DATA AVAILABILITY STATEMENT
The original contributions presented in the study are included in the article/supplementary material, further inquiries can be directed to the corresponding author/s.

AUTHOR CONTRIBUTIONS
SB carried out the laboratory work. ES, SM, EM, and RE performed the data analysis. EU performed the experimental design. ES, SM, SB, and EU carried out the manuscript writing and editing. All authors read and approved the final manuscript.