Oxacillin (Methicillin) Resistant Staphylococci in Domestic Animals in the Czech Republic

The aim of this study was to describe the prevalence of different Staphylococcus species isolated from pathological processes and lesions in domestic animals in the Czech Republic and to detect and describe oxacillin (methicillin)-resistant strains (MRS). During the years 2019–2020, a total of 5218 veterinary clinical samples from the Czech Republic were tested. Testing was performed by culture methods and typing by molecular phenotypic methods MALDI-TOF MS and PCR. Antimicrobial susceptibility testing of the strains was performed by the disk diffusion method. A total of 854 staphylococci strains were identified (16.37% prevalence), out of which 43 strains of 6 species of staphylococci were MRS (n = 43; 0.82% prevalence). Of the MRS strains, the most prevalent species were Staphylococcus pseudintermedius (n = 24; 0.46% prevalence) and Staphylococcus aureus (n = 7; 0.13% prevalence). Susceptibility testing showed resistance to beta-lactam antibiotics and, depending on the species, also to trimethoprim/sulfamethoxazole, gentamicin, tetracycline, erythromycin, clindamycin, and enrofloxacin. For further characterization of MRS, PCR assay for virulence factor genes was performed. Seven of the 14 target genes were observed only in S. aureus, except for the eno gene encoding laminin-binding protein, which was also detected in other staphylococci. It is necessary to emphasize the issue of correct using of antimicrobials in practice and antibiotic policy in university teaching and to create stricter legislation that would prevent the widespread use of antimicrobials in veterinary medicine, especially in livestock to reduce the emergence and spread of antimicrobial resistance.


Introduction
Several studies have shown that Gram-positive bacteria are apparently the most common microorganisms isolated from different human and veterinary clinical materials [1][2][3][4][5]. This group of bacteria includes staphylococci, which are mostly commensals of the skin and mucous membranes in animals and humans [6,7]. Many of them are opportunistic pathogens causing pyogenic infections [6,8]. Of the coagulase-positive staphylococci (CPS), the species of special human and veterinary importance is Staphylococcus aureus, which causes local purulent and systemic infections, as well as human and animal toxaemia [5,[9][10][11]. Other CPS such as Staphylococcus pseudintermedius, Staphylococcus intermedius, and Staphylococcus delphini, which are included in the Staphylococcus intermedius-group, as well as Staphylococcus schleiferi ssp. coagulans and S. hyicus, are significant primarily in the veterinary sphere [6,[12][13][14].
Staphylococci may have a number of other virulence factors that are involved in the pathogenesis of the disease. Some symptoms associated with S. aureus infection are caused by toxins, such as toxic shock syndrome toxin 1 (TSST-1), enterotoxins, and exfoliative toxins (ETs). More than 20 SEs have been identified to date and are one of the most frequent causes of food-borne disease. TSST-1 causes a serious illness with high mortality. Exfoliative toxins (ETs) (also known as "epidermolytic" toxins) cause staphylococcal scalded skin syndrome (SSSS) characterized by destruction of desmosomal cell attachments resulting in detachment of the epidermis [9,30] (Votava et al. 2006; Oliveira et al. 2018). Some strains of S. aureus can also produce Panton-Valentine leucocidin [4,16].
The presence of virulence factors in different staphylococcal strains varies depending on the location of the infection and the degree of virulence.
The aim of the present study was to describe species distribution of staphylococci isolated from pathological processes and lesions in domestic animals in the Czech Republic and to detect and characterize oxacillin/methicillin-resistant staphylococcal (MRS) strains, including their prevalence, site of infection, host specificity, and virulence factor determination.

Isolation and Identification of Bacteria: Bacteriological Confirmation
A total of 5218 clinical samples from pathological processes and lesions of domestic animals between April 2019 and June 2020 underwent microbiological cultivation at the Veterinary Research Institute Brno (Czech Republic). Sampling was performed by instructed private veterinarians. Solid and slurry materials were sterile collected in 60 mL plastic containers (Dispolab Ltd. Brno, Czech Republic). Fluids were collected in 10 mL sterile closable tubes (Dispolab Ltd. Brno, Czech Republic). Swabs were taken using Transbak system containing Amies broth with active carbon (Dispolab Ltd. Brno, Czech Republic). All samples were stored and transported to the laboratory at 4 • C.

Samples from Digestive Tract
Feces, rectal swabs, and swabs taken from the stomach lining were examined routinely by conventional methods of cultivation on meat peptone blood agar (MPBA) (Trios Ltd., Prague, Czech Republic), and plates were incubated aerobically at 37 ± 1 • C for 24 h. The cultivation of hair; swabs and scrapings of skin; swabs of ear; urine and swabs of the urinary tract; swabs and the lavages of the respiratory tract, pharynx, conjunctiva and oral mucosa; the puncture of chest, lymph nodes and joints were performed on MPBA (Trios Ltd. Prague, Czech Republic), and the plates were again incubated aerobically at 37 ± 1 • C for 24 h.

Mammary Gland and Milk Samples
The milk samples were again inoculated on MPBA (Trios Ltd., Prague, Czech Republic) after thorough shaking, and incubation was carried out aerobically at 37 ± 1 • C for 42-48 h.

Bacteriological Confirmation and Susceptibility Determination
All types of colonies grown on plates were isolated, and suspected Gram-positive organisms were isolated and subsequently confirmed by the phenotypic molecular method using mass detector MALDI-TOF MS Microflex TM LT System (Bruker Daltonik GmbH, Bremen, Germany), on the basis of proteomics analyses and MALDI Biotyper software MBT Compass 4.1.100 (Bruker Daltonik GmbH, Bremen, Germany). In the plates with mixed bacterial cultures, the most prominent colony-forming unit agent was regarded as the leading pathogen. An identification score of 2.000 was set as the reliability threshold. The typing of strains with a lower score was specified by the MALDI-TOF method using a more exact library of spectra of the National Reference Laboratory for Staphylococci of the National Institute of Public Health in Prague or PCR method.

MRS Molecular Characterization
In 35 MRS isolates including S. pseudintermedius (n = 20), S. aureus (n = 7), S. haemolyticus (n = 4), S. intermedius (n = 2), and S. epidermidis (n = 2), polymerase chain reaction (PCR) for mecA gene detection was used to confirm methicillin resistance. Eight strains failed to resuscitate, and PCR was not performed. Gene mecA encodes the production of penicillin-binding protein PBP2A (or PBP2') and is considered the gold standard for methicillin resistance determination. For characterization of MRS isolates, the presence of virulence factor genes, including MSCRAMM, biofilm and the main toxin genes, were detected by PCR (polymerase chain reaction). The following genes were targeted: cna (encoding collagen-binding protein), eno (encoding laminin-binding protein), clfA and clfB (encoding clumping factors A and B), fib (encoding fibrinogen-binding protein), ebp (encoding elastin-binding protein), bbp (encoding bone sialoprotein-binding protein), fnbA and fnbB (encoding fibronectin-binding proteins A and B), icaA (encoding polysaccharide intercellular adhesin), etA (exfoliative toxin A), etB (exfoliative toxin B), and tsst (encoding toxic shock syndrometoxin) (see Table 2). TSST-2 ATCGAACTTTGGCCCATACTTT Quick boiling method was used for DNA isolation. A number of colonies of pure bacterial culture were resuspended in 50 µL of sterile distilled water. The suspension was incubated for 10 min at 100 • C and centrifuged for 10 min at 10,000× g at 4 • C. The supernatant was used in the PCR reaction as template DNA. Polymerase chain reactions were performed according to the protocols described in references in Table 2. Fragments were analyzed by electrophoresis in 2% agarose gel stained with ethidium bromide and visualized under ultraviolet light. Strains used as positive controls were S. aureus CCM 2353 for cna, eno, clfA, clfB, fib, ebp, bbp, fnbA, icaA; S. aureus CCM 2773 for fnbB; S. aureus CCM 7056 for etA and etB; and S. aureus RF122 (Fitzgerald J.R., University of Edinburgh, Edinburgh) [43] for tsst gene.

Results
Out of 854 (16.37% prevalence) isolated staphylococci strains, 613 strains belonged to the group of CPSs (11.75% prevalence) and 241 strains to the group of CNSs (4.62% prevalence). The detailed numbers and prevalence of different species of staphylococci are shown in Table 3. Of these, 43 strains of 6 species were methicillin-resistant staphylococci (MRS) (n = 43; 0.82% prevalence). Of them, 36 strains with 4 species (n = 36; 0.69% prevalence) belonged to the CPS group, and 7 strains of 2 species (n = 7; 0.13% prevalence) to the CNS group. The most prevalent species among the MRS strains were S. pseudintermedius (n = 24; 0.46% prevalence) and S. aureus (n = 7; 0.13% prevalence). Table 4 shows the distribution of individual species of staphylococci in animal population in the observed period. The exact distribution of MRS strains in different species and groups of domestic animals is shown in Table 5. It follows from this table that MRS of all six detected species were isolated from domestic carnivores and their prevalence was the highest in these animals (n = 34; 1.38% prevalence). Surprisingly, three of the six MRS species were detected in rodents (S. aureus, S. haemolyticus, and S. pseudintermedius) (n = 3; 1.27% prevalence). In ruminants, pigs, solipeds, and exotic birds, MRS of one species was detected in each (n = 1, prevalence: 0.16%, 1.41%, 1.10%, and 0.97%, respectively). Table 6 shows the distribution of MRS isolates in different organs and organ systems. The greatest species diversity and prevalence was found in skin (n = 21; 4 species; 4.79% prevalence) and the respiratory system (n = 8; 4 species; 2.72% prevalence). All MRS strains (100%) showed resistance to oxacillin, cephalothin, cefoxitin, cefovecin, piperacillin and amoxicillin/clavulanic acid. Most of these isolates were co-resistant to enrofloxacin (93%). In contrast, the isolates showed high susceptibility to florfenicol and nitrofurantoin (100%). Susceptibility to novobiocin and rifaximin was also high (97.7% of susceptible strains). The detailed results are shown in Table 7.       A total of 35 suspected MRS strains (according to disc diffusion method) were characterized by molecular methods, and all were positive for mecA gene. All MRS tested (except of two S. aureus isolated from cat urine and cow milk) were positive for eno gene encoding laminin-binding protein. Genes cna, clfA, clfB, and icaA were detected in all of seven S. aureus isolates, and fnbB was detected in six S. aureus isolates. None of these genes were detected in non-S. aureus isolates. None of MRS tested strain was positive for toxin genes etA, etB or tsst (see Table 8).  mecA-gene encoding methicillin resistance, can-collagen-binding protein gene, eno-laminin-binding protein gene, clfA and clfB-genes encoding clumping factors, fib-fibrinogen-binding protein gene, ebp-elastin-binding protein gene, bbp-bone sialoprotein-binding protein gene, fnbA and fnbB-genes encoding fibronectin-binding proteins, icaA-polysaccharide intercellular adhesin gene, etA and etB-exfoliative toxin genes, tsst-toxic shock syndrome toxin gene.

Discussion
Even though the time span of collection and testing of clinical samples in our study was rather short, a quite large species diversity of the isolated Staphylococcus strains was shown and confirmed in domestic animals. A total of 28 Staphylococcus species were detected. In our study, of the CPS, the major species identified was S. pseudintermedius (n = 336; 6.44% prevalence), which was predominantly found in domestic carnivores, especially dogs. The second most frequently encountered CPS was S. aureus (n = 205; 3.93% prevalence), surprisingly most often found in ruminants and solipeds, followed by S. intermedius (n = 40; 0.77% prevalence), which was also predominant in carnivores, especially dogs. Out of the CNS, the major species was S. haemolyticus (n = 68; 1.30% prevalence), most often isolated from ruminants and carnivores, followed by S. chromogenes (n = 45; 0.86% prevalence), which was most prevalent in ruminants, then followed by S. felis (n = 32; 0.61%), having a clear affinity for cats, and S. epidermidis (n = 23; 0.44% prevalence), which was most frequently isolated from clinical material from domestic carnivores, ruminants, and pigs. S. chromogenes has the ability to form biofilms and in veterinary medicine is also a common pathogen of the mammary gland of cattle (8.8% to 51.4% of isolated CNS strains) [45,46]. The literature sources confirmed that all the above-mentioned four CNS species may be pathogenic to animals, and some of them also to humans. S. haemolyticus is an opportunistic pathogen infecting debilitated human patients [18] and is often detected in veterinary laboratories in association with mastitis in cattle (12.2% to 20.3% of isolated CNS strains) [45,46].
Not surprisingly, most MRS isolates originate in the skin and the respiratory system of animals, which is consistent with the literature. Our findings of MRS strains of staphylococci in domestic animals confirmed that S. aureus, S. pseudintermedius, S. intermedius, S. epidermidis, and S. haemolyticus species may have resistance genes to methicillin, as previously reported by Votava et al. (2006) and Oreiby et al. (2019) [9,23]. In our study, the strain S. coagulans was also detected, which in addition to methicillin resistance also showed resistance to all beta-lactam antibiotics and to enrofloxacin. The prevalence of MRSA strains in our study was relatively low (n = 43; 0.82%) in comparison with 15.5% in the human clinical material of hospitalized patients [47] and the prevalence of veterinary MRSA strains isolated, for example, from clinical material from domestic carnivores (16.1%) in the Czech Republic [12], as well as from milk of cows with signs of mastitis in the Czech Republic and Egypt where the prevalence of MRSA ranged from 31.7% to 50% [24,25]. The prevalence of MRS strains of S. pseudintermedius (n = 24; 0.46%) and S. intermedius (n = 4; 0.08%) in our study is also very low in comparison with other Czech studies dealing with S. pseudintermedius (25%) and S. intermedius (50%) [12].
The MRS strains of S. haemolyticus and S. epidermidis are no exception, with their prevalence in our study of 0.1% (n = 5) and 0.04% (n = 2), respectively, while literature sources reported prevalence, for example in humans, of up to 45.4% for both of the above bacterial species [48]. This can be attributed to the fact that our strains were collected for 14 months only and that relatively large numbers of various materials from different animals were examined and tested, while other studies focused on narrow spectra of human patients, animals, and clinical materials such as cow's milk samples or clinical specimens from dogs and cats. Our study also shows that the detected strains display a certain species-specificity in terms of antimicrobial resistance.
In addition to beta-lactams, MRS S. aureus (MRSA) strains show 100% resistance also to tetracycline; S. epidermidis MRS strains also to gentamicin, tetracycline, erythromycin, and enrofloxacin; S. haemolyticus MRS strains also to trimethoprim/sulfamethoxazole, gentamicin, erythromycin, and enrofloxacin; MRS strains of S. intermedius also to gentamicin, erythromycin, clindamycin, and enrofloxacin; and MRS strains of S. pseudintermedius and S. coagulans also to enrofloxacin.
Due to the diversity of the clinical material in our study, a diverse capture of adherence factors could be expected, as the first step of successful infection is adherence to different surfaces, depending on the site of infection. However, genes encoding MSCRAMMs were detected only in S. aureus, none in other staphylococci (CNS or CPS).
The exception was the eno gene encoding laminin-binding protein. This gene was detected in all MRS isolated from clinical material except for two S. aureus isolated from cat urine and cow milk. Thus, this virulence factor was shown to be unrelated to the location of the infection, as different isolates from different sites of infection carried this gene. Moreover, other studies described a very common occurrence of the eno gene in both CPS and CNS (73-100%) [39,49,50]. Consistent with our results, rare occurrences of other MSCRAMM genes in CNS isolates were confirmed by other studies [50]. In contrast, in CPS including S. intermedius, a noticeably higher occurrence was described (ebp-73%, fib-91%, and fnbA-7%) [49]. In our study, neither S. intermedius nor S. pseudintermedius harbored MSCRAMM genes.
Isolates of S. aureus from different animals and different sites of infections (cat, pig, cow, horse, dog; urine, joint, milk, skin) were included in our study, yet all these isolates showed very similar MSCRAMM gene profiles: eno+ (except two isolates), cna+, clfA+, clfB+, and fnbB+ (except one isolate from cat urine).
The icaA gene encoding polysaccharide intercellular adhesin was detected in all MRSA isolates. The ica operon is considered to be the main operon responsible for biofilm formation in the Staphylococcus genus. However, there is not an absolute correlation between the presence or absence of ica genes and biofilm production [51]. In our study, the icaA gene was not detected in other staphylococci (non-S. aureus), consistent with other studies where icaA was only rarely detected in these strains [51][52][53]. Genes encoding etA, etB, and tsst toxins were not detected in our MRS isolates and also in other publications, only the rare occurrence of these genes in staphylococci of animal origin was described [51,54]. Although these virulence factors can be found in animal strains, they are generally associated with human clinical pathogens [51,55].
It is well known that the growth of antibiotic resistance is a global problem. Veterinary medicine undoubtedly contributes to its origin and spread due to the application of antimicrobials in water and feed, especially in livestock, and also due to the use of local antibiotics (sprays, suppositories, ointments, dusting powders), where accurate dosing of these substances is not guaranteed. It is therefore necessary to tighten up the legislation governing the use of these substances in animals and to set up a system for its control. At the same time, there is a need to expand and improve university teaching on the use of antimicrobials in the veterinary field.

Conclusions
The present study demonstrated high prevalence of some CPS and CNS species detected in animals. These strains should be taken seriously from both the epizootiological and epidemiological point of view, as they can pose health risk to both animals and humans not only in terms of potential pathogenicity but in that they can also confer resistance genes and pathogenicity factors to other veterinary and human strains of staphylococci. In terms of antimicrobial susceptibility, it has been confirmed that methicillin-resistant microorganisms are resistant to all other beta-lactam antibiotics, even in isolated veterinary strains and also to amoxicillin/clavulanic acid. All MRS strains of staphylococci isolated in the present study except for three S. aureus strains were resistant to enrofloxacin. Due to the occurrence of these microorganisms in farm animals and animals kept for hobby, it would be very appropriate to set up research projects aimed at the detection of oxacillin/methicillinresistant staphylococci in these animal groups and to map their occurrence at least in Europe, because the microecosystems of animals, humans, plants, and the macroecosystem are interconnected. It is therefore necessary to emphasize the issue of correct using of antimicrobials in practice and antibiotic policy in university teaching and to create stricter legislation that would prevent the widespread use of antimicrobials in veterinary medicine, especially in livestock to reduce the emergence and spread of antimicrobial resistance.  Institutional Review Board Statement: Not applicable. All animals were patients, not experimental animals.
Informed Consent Statement: Not applicable. Our study is strictly anonymous and the rights of the animal patients and their owners have not beenviolated in any way.All laboratory tests and data processing were performed on request and with the express consent of the animal owners. Data Availability Statement: All data supporting reported results of this work are archived in the personal database of the first author (Bzdil, J.) and partial data concerning the genomic characterization of bacterial strains are stored in the database of the Veterinary Research Institute Brno (Czech Republic).