Characterization of methicillin-resistant Staphylococcus pseudintermedius isolated from dogs and cats

Authors

  • Marjorie Bardiau,

    Corresponding author
    • Bacteriology, Department of Infectious Diseases, Faculty of Veterinary Medicine, University of Liège, Belgium
    Search for more papers by this author
  • Kazuko Yamazaki,

    1. Bacteriology, Department of Infectious Diseases, Faculty of Veterinary Medicine, University of Liège, Belgium
    2. Division Bioenvironmental Science, Frontier Science Research Centre, Faculty of Medicine, University of Miyazaki 5200 Kihara, Kiyotake, Miyazaki
    Search for more papers by this author
  • Isabelle Ote,

    1. Bacteriology, Department of Infectious Diseases, Faculty of Veterinary Medicine, University of Liège, Belgium
    Current affiliation:
    1. Isabelle Ote, Coris BioConcept, Gembloux, Belgium
    Search for more papers by this author
  • Naoaki Misawa,

    1. Laboratory of Veterinary Public Health, Department of Veterinary Science, Faculty of Agriculture, University of Miyazaki, Miyazaki, Japan
    Search for more papers by this author
  • Jacques G. Mainil

    1. Bacteriology, Department of Infectious Diseases, Faculty of Veterinary Medicine, University of Liège, Belgium
    Search for more papers by this author

Correspondence

Marjorie Bardiau, Bacteriology, Department of Infectious Diseases, Faculty of Veterinary Medicine, University of Liège, Sart-Tilman, Bât 43a, B-4000 Liège, Belgium.

Tel.: +32 4 366 93 88; fax: +32 4 366 42 61; email: mbardiau@ulg.ac.be

ABSTRACT

The aim of this study was to explore the presence of methicillin-resistant Staphylococcus pseudintermedius (MRSP) in a collection of S. pseudintermedius strains isolated from dogs and cats with dermatitis in Japan and to compare their genotypic and phenotypic characteristics. Clonal relationships were determined by pulse field gel electrophoresis (PFGE), staphylococcal chromosomal cassette mec (SCCmec) typing, and multilocus sequence typing (MLST). Biofilm formation assay was performed using safranin staining in microplates. Three virulence genes coding for S. intermedius exfoliative toxin and Panton-Valentine leukocidin (siet, lukS-PV and lukF-PV) were searched for in a collection of strains. Antimicrobial resistance against 15 antibiotics was studied by a disc diffusion method. Twenty-seven MRSP were isolated. According to PFGE results the isolates were not closely related except for a few strains. MLST showed that the strains belonged to five groups, ST71 and ST26 being the two most prevalent. Three types of SCCmec (II, II–III and V) were identified. All isolates were siet-positive but PVL-negative. Most strains (except for two) produced strong biofilm in tryptic soy broth with glucose. Seventy-eight percent of the isolates were resistant or intermediate to twelve or more antibiotics. Our study demonstrates that the ST71 lineage is widespread in Japan and that ST26 could represent an emerging lineage. Moreover, most of our strains are capable of forming strong biofilm and possess siet gene, two virulence characteristics that probably help the bacteria to persist and spread. Finally, our MRSP strains show a strong resistance profile to antibiotics commonly used in veterinary medicine.

List of Abbreviations
Luk-PV

Panton−Valentine leukocidin

MLST

multilocus sequence typing

MRSP

methicillin-resistant Staphylococcus pseudintermedius

OD

optical density

PFGE

pulsed field gel electrophoresis

RFLP

restriction fragment length polymorphism

S. pseudintermedius

Staphylococcus pseudintermedius

SCCmec

staphylococcal cassette chromosome mec

siet

Staphylococcus intermedius exfoliative toxin

ST

sequence type

TC

tissue culture

TSB

tryptic soy broth

TSBglc

TSB containing 0.25% glucose

UV

ultraviolet

In 2005, Devriese et al. first described S. pseudintermedius as a new species mainly found in cats and dogs [1]. This bacterium is commonly confused with S. intermedius in routine diagnostic practice and, rather than S. intermedius, is probably the major cause of canine pyoderma [2]. This commensal bacterium lives in many different animal species and can be an opportunistic pathogen that is mainly responsible for skin infections, such as pyoderma, in dogs and cats [3]. S. pseudintermedius rarely causes infections in humans; when it does, they are usually wound infections after animal contact [4, 5].

Methicillin-resistant S. pseudintermedius have recently emerged as significant nosocomial pathogens in companion animals [6-8]. Because the therapeutic options are limited, both for animals and humans, their increasing incidence is an alarming problem. Moreover, in addition to having been isolated from cats and dogs, MRSP have also been isolated from humans, highlighting a public health issue for veterinarians and pet owners [9, 10].

The aim of this study was to explore the presence of MRSP in a collection of S. pseudintermedius isolated from dogs and cats in Japan and to compare the following genotypic and phenotypic characteristics: SCCmec-typing, PFGE, MLST, virulence-associated factor-encoding genes (siet, lukS-PV and luk-F-PV), biofilm formation and resistance profiles to 15 frequently used antibiotics.

1 MATERIALS AND METHODS

1.1 Bacterial isolates

Two hundred S. pseudintermedius isolates (22 isolated from cats and 178 from dogs) were collected from animals with dermatitis across south Japan (Kyushu area) between 2008 and 2010 by the Laboratory of Veterinary Public Health (Faculty of Agriculture, University of Miyazaki, Japan; import permit nr 604654). Identification of the species was confirmed by PCR assay based on amplification of the thermonuclease (nuc) gene of staphylococcal species as previously described [11]. All isolates were tested for methicillin resistance by streaking onto chromogenic chromID MRSA agar (bioMérieux, Marcy-l'étoile, France). After 24 hrs incubation at 37°C, blue green colonies, considered as positive methicillin-resistant colonies, were stored at −80°C in Luria–Bertani broth with 50% glycerol until further characterization.

1.2 Genotypic characterization

DNA extraction was carried out using the ChargeSwith gDNA Mini Bacteria Kit (Invitrogen, Carlsbad, CA, USA) according to the manufacturer's instructions for staphylococci. After purification, DNA was stored at −20°C.

Genotypic characterization on methicillin-resistant isolates was performed by PCR in two steps: (i) detection of mecA gene, coding for the penicillin-binding protein 2a; and (ii) detection on mecA-positive isolates of siet, lukS-PV, and lukF-PV genes coding respectively for S. intermedius exfoliative toxin and Panton-Valentine leukocidin (LukF-PV and LukS-PV). PCRs were performed on 2 µL of DNA using Taq DNA polymerase (New England BioLabs, USA) and 1 µM of each primer as previously described [12, 13]. All PCR products were separated by electrophoresis in 1.5% agarose gels. Gels were stained with ethidium bromide and photographed under UV light.

1.3 Staphylococcal cassette chromosome mec-typing

Typing of SCCmec was performed using multiplex PCR assays as previously described [14]. PCR products were separated by electrophoresis in 3% agarose gels. Gels were stained with ethidium bromide and photographed under UV light. Amplification patterns were compared to the patterns of prototype strains used as positive controls.

1.4 Pulsed field gel electrophoresis

All isolates were compared by PFGE. Briefly, lysostaphin pre-treated bacterial cells were embedded in 0.9% certified low-melt agarose (Bio-Rad, Hercules, CA, USA), lysed in a 0.5 M EDTA (pH 8) buffer containing 10% N-lauroylsarcosine and 2 mg/mL lysozyme, and treated with 1 mg/mL proteinase K (Sigma, St Louis, MO, USA). Genomic DNA was digested with 20 U of SmaI (Sigma). Restricted fragments were separated using CHEF Mapper (Bio-Rad) with 1% pulsed field certified agarose (Bio-Rad) gel in 0.5% Tris/borate/EDTA buffer at 6 V/cm for 20 hrs with pulse times ranging from 5 to 60 s (angle of 120° and linear ramp factor). The size of each DNA band was estimated by Biogene (Vilber Lourmat, Marne-la-Vallée, France). A dendrogram was prepared by the unweighted-pair group method using an arithmetic average algorithm, dice coefficient and optimization and position tolerance of 2%.

1.5 Multilocus sequence typing

Multilocus sequence typing based on the sequence of five housekeeping genes (pta, cpn60, tuf, 16SrRNA, and agrD) was performed on all methicillin-resistant strains as previously described [6]. Briefly, all five genes were amplified and sequenced, and allele number and ST assigned according to the scheme proposed by Bannoehr et al.[6].

1.6 Biofilm production assay

Biofilm formation assay was performed using safranin staining in microplates. Briefly, isolates were incubated in TSB overnight at 37°C. Cultures were then diluted 1:100 in TSBglc and cell suspensions inoculated into wells of sterile 96-well polystyrene TC plates in triplicate for each tested isolate. TSBglc without bacteria served as negative control. The TC plates were incubated at 37°C for 24 hrs. Each well was then carefully washed twice with sterile PBS, dried and stained with safranin 0.1% (w/v) for 10 min. The TC plates were further washed twice with distilled water and dried again at 37°C before adding a mixture of 50% ethanol–50% acetic acid to each well. Finally, absorbance (OD) of adherent biofilm was measured at 490 nm using a microplate reader. Results are reported according to published recommendations based on average OD values and on the cut-off value, named ODc (ODc = average OD of negative control + [3 × SD of negative control]) [15]. Strains were divided into the following four categories: not a biofilm producer (OD ≤ ODc); weak biofilm producer (ODc < OD ≤ 2 × ODc); moderate biofilm producer (2 × ODc < OD ≤ 4 × ODc); and strong biofilm producer (4 × ODc < OD). Biofilm formation assays were performed twice.

1.7 Antimicrobial susceptibility testing

Susceptibility tests were carried out on MSRP isolates by the disc diffusion method of Bauer et al. [16] on Mueller-Hinton agar (Oxoid, Cambridge, UK). Inhibition zones were measured (in mm) after overnight incubation at 37°C and were interpreted according to the 2010 recommendations of the Comité de l'Antibiogramme de la Société Française de Microbiologie. Fifteen antibiotics were tested, namely: penicillins (penicillin 10 IU, ampicillin 10 µg), cephems (cephalexin 30 µg, cefoperazone 30 µg, ceftiofur 30 µg), aminoglycosides (gentamicin 15 µg, kanamycin 30 IU, streptomycin 10 IU), macrolide (erythromycin 15 IU), tetracycline (tetracycline 30 IU), fluoroquinolone (ciprofloxacin 5 µg), lincosamide (clindamycin 2 IU), folate pathway inhibitor (trimethoprim-sulfamethoxazole 1.25 − 23.75 µg), aminocyclitol (spectinomycin 100 µg) and phenicol (chloramphenicol 30 µg; (ceftiofur, kanamycin, ciprofloxacin; Becton Dickinson, Sparks, MD, USA; The remaining antibiotics listed; I2A, Perols, France).

2 RESULTS

2.1 Methicillin-resistant Staphylococcus pseudintermedius identification

Of the 200 S. pseudintermedius isolates collected from cats and dogs with dermatitis, 27 (13.5%) were characterized as MRSP after growth on chromID MRSA agar plates. The mecA gene presence was confirmed by PCR for all MRSP isolates. Most of the identified MRSP were isolated from dogs (n = 25).

2.2 Clonal relationship

After SmaI digestion of total DNA, 22 different pulsotypes were obtained for the 26 isolates (Fig. 1). DNA from one isolate could not be digested with SmaI. Using a cut-off of 80% similarity, only a few isolates could be grouped together, they formed four distinct clusters. The first (A), second (B), third (C), and fourth (D) clusters comprised four, two, four and two MRSP isolates, respectively. Interestingly, clusters A and B included isolates collected from different cities and at different times (Fig. 1). According to the PFGE results, the other 14 isolates were not genetically related (Fig. 1).

Figure 1.

PFGE dendrogram, MLST and SCCmec-typing results for MRSP strains. Strains in italics were isolated from cats. The dendrogram was constructed using UPGMA (Dice coefficient, 2% position tolerance). The strain JAP170 was untypeable by PFGE.

Multilocus sequence typing showed that the 27 isolates belonged to five different types: ST2, ST26, ST29, ST71 and ST115 (Fig. 1). Most MRSP belonged to ST71 group (18 isolates). Six MRSP isolates belonged to ST26. The three remaining isolates belonged to ST29 (MRSP isolate untypeable by PFGE), ST2 and ST115 types.

Staphylococcal chromosomal cassette mec-typing identified three types of SCCmec cassette, namely the II, II–III and V types (Fig. 1). A good correlation was observed between the MLST results and SCCmec-typing. Indeed, all isolates belonging to ST71 were assigned to type II-III SCCmec and all isolates belonging to ST26 were assigned to type II SCCmec, except for one isolate (type V SCCmec). Isolates ST115, ST29 and ST2 were assigned to types V, V and II SCCmec, respectively.

2.3 Virulotyping and biofilm formation

All but one isolate from a cat (JAP144) tested positive for the siet gene, which codes for the exfoliative toxin specific to S. intermedius. Conversely, none of the isolates harbored the lukS-PV and lukF-PV genes, coding for LUK–PV.

After growth in TSBglc, biofilm production was estimated by spectrophotometry using safranin staining. All but two isolates produced strong biofilm in TSBglc. The remaining two isolates (JAP001 and JAP178) were moderate and weak biofilm producers, respectively.

2.4 Antibiotic susceptibility profiles

All isolates were resistant to at least five of the fifteen tested antibiotics (Fig. 2): gentamicin, kanamycin, spectinomycin, erythromycin and trimethoprim-sulfamethoxazole. Moreover, all but one isolate were also resistant to penicillin, clindamycin and ciprofloxacin. By contrast, more than 40% of the isolates were sensitive to cefoperazone, cephalexin and ceftiofur. One isolate was resistant to all tested antibiotics.

Figure 2.

Antibiotic susceptibility profiles to fifteen antibiotics.

3 DISCUSSION

Staphylococcus pseudintermedius is one of the bacterial species that causes skin infections, primarily in dogs and cats. MRSP emergence could therefore complicate the treatment of pet infections, leading to recurrent disease. Both animal-to-animal and animal-to-human transmission are potential risks that have to be considered. The purpose of this study was to isolate MRSP from cats and dogs with dermatitis in south Japan and to characterize several of their genotypic (PFGE, MLST, SCCmec-typing and virulotyping) and phenotypic (biofilm formation, antibiotic resistance) features.

Twenty-seven of the 200 tested isolates (13.5%) were methicillin-resistant. According to PFGE results, only the four MRSP isolates with pulsotype C were closely related and clonal. Not surprisingly, these strains were obtained from specimens from the same city (Miyazaki) and over the same period of time (May–July 2008). On the other hand, most other MRSP isolates did not form any homogeneous groups and were not clonal. Nevertheless, because two-third of the strains belonged to ST71 and almost one-third to ST26, the MLST results did not reflect this major clonal variation. To our knowledge, this is the first report of the ST71 lineage in Japan, supporting the contention that this clone is disseminated worldwide [17]. The ST71 lineage is highly prevalent in Europe and has also been identified in China, USA and Canada [7, 18-21]. Similarly to the findings of other studies [20], the ST71 lineage we isolated harbored the II-III SCCmec-type. Such a widespread clone presents a potential risk for pet owners. Indeed, Stegmann et al. recently reported one case of human infection associated with the MRSP ST71 strain [4]. Contrastingly, the second more prevalent lineage in our study (ST26) is not so much widespread, having been only sporadically identified [4, 18, 20, 21]. Most of our ST26 lineage isolates were associated with II SCCmec-type (except for one isolate that was associated with V SCCmec-type).

The ability of MRSP to form biofilm has rarely been studied [22]. In agreement with previously published results on Norwegian MRSP, we found that MRSP are good biofilm producers [22]. Moreover, Osland et al. showed that MRSP belonging to the ST71 lineage have the ability to stronger produce biofilm than isolates of other MLST lineages. This feature could confer advantages in persistence and spreading on strains of this lineage [22]. The other main MLST lineage identified in our study (ST26) also showed a strong biofilm formation. Concerning virulotyping, all strains possessed the siet gene. This exfoliative toxin may also help the bacteria to be more virulent and to persist.

Concerning the antibiotic sensitivity tests, all strains but one were multiresistant (resistant to more than eight antibiotics and three antibiotic classes), 78% of multiresistant strains being resistant or intermediate to twelve or more of the fifteen tested antibiotics. On average, strains were resistant to 12 antibiotics (78% of the tested antibiotics). Our study represents one piece of supplemental evidence for the necessity to regulate and control the use of antibiotics in veterinary medicine, not only in farm but also in pet animals, as was recently concluded by the Council of the European Union [23].

In conclusion, this study provides an insight into MRSP by demonstrating that: (i) ST71 is one of the main MLST lineages in Japan; (ii) ST26 MRSP could represent an emerging MLST lineage in Japan; (iii) most MRSP in our study are strong biofilm producers and possess the exfoliative toxin gene (siet); and (iv) all strains are resistant to a large number of antibiotics.

ACKNOWLEDGMENTS

This work was supported by the Japanese Society for the Promotion of Science JSPS (grant “Strategic Young Researcher Overseas Visits Program for Accelerating Brain Circulation” to K.Y.) and by grants of the Fonds de la Recherche Scientifique (FRS-FNRS, Crédits aux chercheurs, nos. 1642 and 1363 to J.M.).

4 DISCLOSURE

The authors declare that they have no conflict of interest.

Ancillary