METHICILLIN-RESISTANT STAPHYLOCOCCUS PSEUDINTERMEDIUS, A NEW THREAT IN HUMAN AND VETERINARY MEDICINE?

In this paper we briefl y described the worldwide distribution of methicillin-resistant Staphylococcus pseudintermedius (MRSP) in dogs and cats. Th e most common sequence type of MRSP strains in dogs is ST71 as it was detected in isolates on four continents (Europe, Asia, North and South America). However, several diff erent MRSP sequence types are detected in small animals, and the presence of new genetic variants is continually reported. Sometimes isolates belonging to the same sequence type (ST) are detected in dogs, cats, their owners, and veterinarians. MRSP is oft en multidrug-resistant and its resistance patterns are usually linked to certain sequence types. Th e resistance to non-beta-lactam antibiotics such as erythromycin, clindamycin, tetracycline, gentamicin, enrofl oxacin and sulfamethoxazole/trimethoprim is also recorded. Taking into account that MRSP tends to confer multidrug-resistant phenotype, it is quite challenging for veterinarians to give adequate therapy in clinically ill animals. It would seem as if the signifi cance of MRSP in the clinical epidemiology of humans is not fi rmly established. However, the importance of MRSP in human medicine should not be underestimated given the fact that all methicillinresistant Staphylococcus spp. carry resistance and virulence genes and have the potential to share their genetic elements with other bacteria.


INTRODUCTION
Shortly aft er methicillin was introduced into medical practice, the resistance to this antibiotic occurred in Staphylococcus aureus isolates. Not only did MRSA become one of the most important nosocomial pathogens worldwide, but it was also a signifi cant community-acquired and livestock-associated pathogen (Gordon and Lowy, 2008).
Th e resistance to methicillin occurs due to the presence of mecA gene that encodes penicillin-binding proteins (PBPs). Th is gene is integrated into a staphylococcal cassette chromosome mec (SCCmec) locus. Th e integration of the cassette is accomplished by the recombinase encoded by ccr gene. Th e sequence of mec-scc is used for the determination of SCCmec cassette type and the establishment of its epidemiological status (Gordon andLowy, 2008, rev. Velhner et al., 2016). Other additional methods of MRSA molecular typing include spa typing, multilocus sequence typing (MLST) and pulsed-fi eld gel electrophoresis (PFGE). Isolates causing nosocomial infections health careassociated MRSA (HA-MRSA) belong to SCCmec types I, II or III and less frequently to types IV and V, while community-acquired MRSA (CA-MRSA) belongs to SCCmec types V and VI (Kasai et al., 2016, rev. Velhner et al., 2016. Staphylococcus pseudintermedius has been recognized as a relatively new species since 2005 when it was isolated from a dog, a cat, a horse and a parrot by Devriese et al. (2005). Methicillin-resistant Staphylococcus pseudintermedius (MRSP) was fi rst reported in Europe in 2006 and the whole genome sequence was available by 2013 (Moodley et al., 2013). MRSP belongs to Staphylococcus intermedius group and causes opportunistic infections in dogs and cats. Th ere are six coagulase-positive staphylococcus species including S. aureus, S. intermedius, S. schleiferi subsp. coagulans, S. hyicus, S. lutrae, S. delphini and S. pseudintermedius. Among them, S. intermedius, S. pseudintermedius and S. delphini are very closely related (Sasaki et al., 2007).
MRSP produces coagulase the same as MRSA strains, and therefore laboratory identifi cation can be a bit challenging (Sewid et al., 2018). However, molecular phylogenetic analysis (Sasaki et al., 2007) In this brief review we described the worldwide distribution of MRSP specifi c sequence types, their antimicrobial resistance and the signifi cance of MRSP in clinical epidemiology of humans and small animals, mainly dogs and cats.

SEQUENCE TYPE DISTRIBUTION OF THE MRSP
All MRSP isolates from dogs described until 2016 belong to 16 diff erent sequence types and the most frequent clone worldwide is ST71. Th is clone was detected on four continents (Europe, Asia, North and South America). Th e most prevalent clone in the USA is ST68, while in Asia it is ST45/ST112. Other clones that are less abundant belong to sequence types ST45, ST258, ST261, ST112, ST265, ST68, ST169 and ST181. Interestingly, the genetic diversity among methicillin-sensitive Staphylococcus pseudintermedius (MSSP) is higher compared to MRSP and seven STs are present in both groups of isolates , dos Santos et al., 2016. Clonal complex (CC) 71 is detected only in MRSP, while CC75 is exclusively found in MSSP strains. In addition, various SCCmec types were found in MRSP clonal lineages due to their independent acquisition (dos Santos et al., 2016). All MRSP isolates collected from felines in diff erent European countries belonged to ST71, spa type 02 and SCCmec type II-III. A single MRSP isolate from Canada harbored SCCmec type V element . A research work carried out in Iran showed that MRSP can be isolated from the nostrils and perianal area of healthy dogs and cats and those isolates carried the cassette chromosome SSCmec II and V (that have been detected in two and 10 isolates respectively) (Tabatabaei et al., 2019).
It was established that the presence of MRSP-ST71 is slowly decreasing in dogs with skin and soft tissue infections in France (Bergot et al., 2018). Even though this trend was not statistically signifi cant in the 2012-2013 period, the expanding decreasing trend was reported in the following years (2015-2016). Th e majority of MRSP isolates have their specifi c genetic background, except for a few clones such as 258. Also, of great importance is an emerging ST496 clone in France, previously identifi ed in Australia, primarily in Sidney. In Finland, the examination of 1958 clinical isolates of Staphylococcus pseudintermedius showed that 266 isolates were oxacillin-resistant and 321 multidrugresistant. Th e total number of 42 diff erent sequence types was identifi ed and among those 19 STs were new in a database. MRSP was predominantly diagnosed in private clinics in Finland since more patients with dermatological problems were admitted there compared to the Veterinary teaching hospital  (Grönthal et al., 2017). Th e collection of MRSP isolates from dogs (No 28) and cats (No 11) from Th ailand was studied thoroughly in order to identify resistance genes, spa and dru types, MLST and PFGE. Th e most frequent MLST among the same PFGE cluster was ST45 followed by ST112, ST155, ST282, including three novel MLST types ST432, ST433 and ST434. Th e ST45, previously identifi ed in Th ailand, was found in 30 isolates that belong to the two most common PFGE patterns (Kadlec et al., 2016). In a research work from Japan (Kasai et al., 2016), clear diff erences were found between HA-MRSP and CA-MRSP isolates from dogs. Th e most homogenous group of isolates was SCCmec type III and most of them belong to the pulsotype A. Additionally, they have similar resistotypes and belong to the endemic clone ST71. SCCmec type V isolates were heterogeneous as they were classifi ed into 25 pulsotypes. Furthermore, there were diff erences between MRSP and MSSP strains since MRSP was usually identifi ed in older dogs aft er hospital admission. A survey conducted in a small animal hospital in Germany also contributed to the identifi cation of SCCmec type II-III strains among MRSP isolates (60 out of 814 dogs included in the study were MRSP positive). MRSP identifi cation has proved signifi cant in dogs hospitalized and treated with antibiotics within six months before sampling. Samples had been collected before they entered the clinic. All isolates conferred high resistance rates to antibiotics (Nienhoff et al., 2011). A research carried out in Italy at the University of Bari revealed that out of 175 clinical samples from dogs, 151 were culture positive and 63 were identifi ed as S. pseudintermedius by PCR detection of nuc gene. Th e most dominant clone was ST71, identifi ed in 48% isolates. Single-locus variants of ST71 were ST410 (found in four isolates) and ST261 (two isolates), while one isolate was identifi ed as a double-locus variant ST290. Th e new allelic profi le, unrelated to ST71, was detected in one isolate and assigned as ST477. Th e SC-Cmec type II-III was identifi ed in 67% isolates, while 33% isolates, including ST258, were SCCmec type IV (Ventrella et al., 2017). Th ree diff erent sequence types ST71, ST252 and ST305 were identifi ed among healthy dogs examined in small animal clinics in Oslo. Th is research also included 49 clinical isolates of MRSP that were classifi ed into 15 diff erent MLST groups. Th e majority of isolates belonged to the ST258 and ST71 and they have been found in the same geographical area in Norway (Kjellman et al., 2015).

RESISTANCE TO NON-BETA-LACTAM ANTIBIOTICS IN MRSP
What should be pointed out is that the resistance phenotype to non-betalactam antibiotics is oft en related to their clonal group. For instance, the iso- lates from the most abundant clonal group CC71 were less resistant to amikacin, chloramphenicol, tetracycline and trimethoprim-sulfametoxasole while CC258 was less likely to be resistant to amikacin, enrofl oxacin, gentamycin and chloramphenicol (dos Santos et al., 2016). In a research from Norway, it was shown that the clonal lineage ST71 was more resistant to antibiotics than other clonal groups. Th e highest resistance rates in ST71 included ciprofl oxacin and gentamicin antibiotics as well (Kjellman et al., 2015). Th e clone ST496 is susceptible only to fl orfenicol and fusidic acid (Bergot et al., 2018). According to a study carried out in Finland, MRSP was more oft en multidrug-resistant than MSSP. Th e only exception is fusidic acid, to which both strains showed similar levels of resistance (around 24%). Additionally, a total number of 219 MRSP isolates (100%) were susceptible to amikacin (Grönthal et al., 2017).
In a research from Kasai et al. it was determined (2016) that therapy with minocycline was eff ective in the case of MRSP infection with SCCmec type III strains. It was found that the number of isolates belonging to the type V strains was susceptible to amoxicillin-clavulanic acid, cephalexin and cefazolin even if they were mecA positive, and in most cases, multidrug-resistant. SCCmec type III strains were more frequently resistant to oxacillin compared to type V isolates. In a research work conducted in Germany, it was established that MRSP isolates with high MIC to oxacillin were also frequently resistant to non-betalactam antibiotics such as erythromycin and other macrolides, clindamycin, tetracycline, enrofl oxacin, sulfamethoxazole/trimethoprim (Nienhoff et al., 2011). In Poland, a total of 51 S. pseudintermedius isolates from the veterinary clinic and breeding kennels, mostly from dogs with pyoderma and bitches with reproductive disorders, were susceptible to vancomycin, daptomycin and linezolid. More than 50% isolates were resistant to penicillin, tetracycline and macrolides. Resistance to ciprofl oxacin was detected in 31.4% isolates (Ruzauskas et al., 2016). In a research work from Italy (Ventrella et al., 2017) resistance rates to sulfamethoxasole/trimethoprim, clindamycin, ciprofl oxacin and doxycyline were much more pronounced in MRSP isolates than in MSSP. Almost all isolates (except one) were susceptible to gentamicin. MRSP isolates from dogs and cats that were treated at a veterinary clinic in Iran, when resistant to oxacillin/ cefoxitin, tend to develop additional resistance to gentamicin, tetracycline, lincomycin and erythromycin (Tabatabaei et al., 2019). MRSP isolates from two small clinics in Sydney showed increasing rates of resistance to fl uoroquinolones, trimethoprim/sulfamethoxazole and erythromycin compared to MRSA isolates from the same clinic. It was concluded that the clonal lineage ST496 is the area of concern for public health since those isolates carry transferable genetic resistance determinants (Worthing et al., 2018). MRSP rarely causes health problems in cats, unlike in dogs. In a research work conducted by Kadlec et al. (2010) isolates were collected from European countries and North America and antimicrobial resistance data was compared. Th e data showed that the isolates from Canada were resistant only to β-lactam antibiotics and tetracycline, while isolates from Europe were also resistant to macrolide/lincosamide, gentamicin, kanamycin, trimethoprim and ciprofl oxacin. Th e majority of isolates were also resistant to chloramphenicol and tetracycline.

HUMANS AS MRSP HOSTS?
Some methicillin-resistant Staphylococcus (MRS) clones are spreading worldwide and this is the main reason why much attention is devoted to decreasing MRS infection in human and veterinary medicine. Furthermore, human infections with S. pseudintermedius have been recorded sporadically (Stegmann et al., 2010). Dogs have been indicated as a natural reservoir of S. pseudintermedius since this species can be isolated from healthy animals as much as the ill ones. Hanselman et al. (2009) pointed out that household hygiene (i.e. hand washing) can be a crucial step in preventing human infection with bacteria originating from pets, providing less chance for bacteria to switch from their commensal to pathogenic state and become a threat, both to humans and animals. In their research, indistinguishable S. pseudintermedius isolates were found in 4/9 households in humans and their pets (dogs). In another study conducted during a dog show in Berlin, nasal swabs were taken from dogs and their owners and it was elucidated that six owners and 13 dogs carried S. pseudintermedius. In addition, one human isolate was MRSP. For a total of 21 S. pseudintermedius isolates sequence type was defi ned and it was evident that isolates were heterogeneous and nine new sequence types were identifi ed (Walther et al., 2012). On the other hand, in a research work in Sydney, conducted in two small veterinary clinics, MRSP was identifi ed in 8% of personnel-owned dogs and 8% of veterinary personnel. Th ree MRSP isolates from personnel-owned dogs admitted to the clinic B were ST64 but the same ST was not identifi ed in their owners. Th is fi nding implied that dogs got infected with S. pseudintermedius when being taken to clinic B but did not transfer it to their owners. Dogs with no visible skin lesions and generally regarded as healthy were included in the study. In these two clinics the most frequent MRSP isolate from dogs was ST469, but this clone was not identifi ed in personnel-owned dogs, unlike ST64. Lack of human host tropism by MRSP was explained by a small sample size and the fact that dogs had to meet special selection criteria (Worthing et al., 2018). A research was conducted in a small animal hospital in Germany. It included swabs from employees and the hospital environment for isolation and characterization of Staphylococcus spp. isolates resistant to oxacillin. MRSP was detected in nasal swabs of two persons, from the hands of two persons and in three environmental swabs (fl oor in the waiting area, clippers in the emergency room and a table in the intensive care unit). Two spa types (t02 and t06) were identifi ed among MRSP isolates. In addition, all isolates had related PFGE patterns, the same antibiotic resistance profi les and the same resistance genes indicating that the transfer of MRSP between dogs and humans had occurred (Feβler et al., 2018). Even though zoonotic transmission of S. pseudintermedius from dogs to humans is rarely documented, there are many pieces of evidence that such transmissions can happen. It was documented in a case report from Scotland that Clostridium perfringens and S. pseudintermedius were isolated from a patient with an ecthyma-like lesion in a forehead. Th is patient was also an owner of three Siberian husky dogs and therefore swabs from nostrils, mouth, ear, forelimb, hind limb axillae between toes and anal margin were taken from his pets. S. pseudintermedius was isolated from dog swabs but these isolates were not in concordance with isolates from the patient. Moreover, they were genetically unrelated which was confi rmed by MLST. Th e human isolate was resistant to penicillin while isolates from dogs were susceptible. All isolates were susceptible to oxacillin (Robb et al., 2017). However, in a case study from Spain, S. pseudintermedius isolated from patients and dogs of the same household had identical PFGE patterns, sequence types and antimicrobial phenotype and genotype even if isolates were susceptible to methicillin (Lozano et al., 2017). Early studies, conducted in Denmark by Guardabassi et al. (2004), have shown that S. intermedius isolates from dogs and their owners show identical or close related PFGE patterns, but between households these isolates were genetically unrelated. It was not certain if these strains were transmitted from dogs to owners or vice versa. However, authors postulated that resistance genes can be transferred from S. intermedius to human S. aureus and, therefore, present a serious risk for human health. In a study conducted in the Netherlands, MRSP was found in 15/20 households in 18 dogs and six cats. In addition, MRSP was identifi ed in 5/14 contact dogs and 4/13 contact cats in six households. Nasal swabs were taken from 45 humans who owned pets and only two swabs taken from the same household were MRSP positive. Th ese owners had a cat with cystitis caused by MRSP. A total number of 141 personnel employed in the veterinary clinic (13 clinics were included in the study) agreed to participate in the second part of the study and only four persons had MRSP. Th irty-one (16%) of environmental swabs was positive aft er the fi rst sampling and aft er the usual hygiene protocol only 14% were still MRSP positive. Th e authors concluded that there is a small likelihood of humans being colonized by MRSP in contaminated household and hospital environments (Duijkeren et al., 2011). Scientists drew a comparison between MRSP isolates from dogs, veterinarians and dog owners in Th ailand based on sequence types and the result has shown that ST45, 112, 169, 178, 181 and 183 are shared between them. A new cassette SCCmec 57395 was identifi ed in isolates from dogs and veterinarians which belong to ST45. However, this cassette was not found in isolates from dog owners (Chanchaithong et al., 2014).

CONCLUSION
In conclusion, MRSP therapy is diffi cult to undergo since most of the isolates are multidrug-resistant and resistant to beta-lactam antibiotics. Th e data about successful antibiotic therapy is scarce. Local treatment with fusidic acid should be considered, where possible, including the use of non-antibiotic substances for local and surgical wound treatment. Albeit there is little information about the role of MRSP in human infections, the nosocomial transmission and widespread contamination of the environment is an existing threat. Th erefore, it is important to prevent contact with infected animals and have personal hygiene. In that way, the spread of MRSP between animals and humans will be restricted and antibiotic therapy, which is assigned as critical in human medicine, avoided.

ACKNOWLEDGMENT
Th is work was supported by a grant from the Ministry of Education, Science and Technological Development of the Republic of Serbia, Grant number TR 31071.