key: cord-0950089-qzf425lw
authors: Andrés-Lasheras, Sara; Martín-Burriel, Inma; Mainar-Jaime, Raúl Carlos; Morales, Mariano; Kuijper, Ed; Blanco, José L.; Chirino-Trejo, Manuel; Bolea, Rosa
title: Preliminary studies on isolates of Clostridium difficile from dogs and exotic pets
date: 2018-03-09
journal: BMC Vet Res
DOI: 10.1186/s12917-018-1402-7
sha: e990cc7120ae9e3ef809a225842f9d110f38b042
doc_id: 950089
cord_uid: qzf425lw

BACKGROUND: Clostridium difficile infection (CDI) is recognised as an emerging disease in both humans and some animal species. During the past few years, insights into human CDI epidemiology changed and C. difficile is also considered as an emerging community-acquired pathogen. Certain ribotypes (RT) are possibly associated with zoonotic transmission. The objective of this study was to assess the presence of C. difficile in a population of pets and to characterise the isolates. RESULTS: Faecal samples from a total of 90 diarrhoeic dogs and 24 from exotic animal species (both diarrhoeic and non-diarrhoeic) were analysed. Clostridium difficile was isolated from 6 (6.7%) dogs and one reptile sample (4.2%). Four (66.7%) of the six dog strains were capable of producing toxins. Four known different RTs were detected in dogs (010, 014, 123 and 358) and a new one was found in a faecal sample of an exotic animal. This new RT isolate was negative for all toxin genes tested and belonged to sequence type 347 which has been proposed as a Clade-III member. Importantly, two dog strains showed a stable resistance to metronidazole (initial MIC values: 128 and 48 μg/ml). CONCLUSIONS: The results obtained in this study suggest the implementation of antimicrobial susceptibility surveillance programs to assess the prevalence of metronidazole resistance in dogs; molecular studies to elucidate C. difficile metronidazole resistance mechanisms are warranted. Based on the similarity between the ribotypes observed in dogs and those described in humans, the zoonotic transmission should be further explored. Furthermore, exotic animals have shown to harbor uncommon C. difficile strains which require further genomic studies. ELECTRONIC SUPPLEMENTARY MATERIAL: The online version of this article (10.1186/s12917-018-1402-7) contains supplementary material, which is available to authorized users.

Clostridium difficile is a bacterium capable of producing enteric disease in different animal species included humans. Toxigenic C. difficile strains are the most common cause of antibiotic-associated diarrhoea in people from developed countries through the synthesis of toxins A and B, its main virulence factors [1] . The epidemiology of C. difficile has changed in the last 15 years and it is now recognised as an emerging pathogen in both humans and animals. It is also considered an emerging community-acquired pathogen likely associated with a zoonotic and/or foodborne transmission [2] .

Animals are an important source of many infectious diseases for humans. About 75% of emerging infectious diseases are zoonoses [3] , and it is thought that pets could be implicated in the transmission of C. difficile to humans since similar genotypes have been recovered from them and humans [4] . Among the strains isolated from dogs, there are several ribotypes of international interest, such as RT078, RT014/020 and RT045 [5] . Likewise, exotic animals can act as vectors of many zoonotic diseases, including enteric diseases, and several zoonotic outbreaks have been associated with the trade of this type of animal species (either in a legal or illegal way) as their international trade has increased recently [6] . However, there is limited information about toxigenic C. difficile carriage in exotic animals in the scientific literature [7] .

Dogs can develop diarrhoea due to different bacteria or virus infections (e.g. Salmonella spp., Campylobacter spp., Clostridium perfringens, C. difficile, or coronavirus), intestinal parasites (e.g. Giardia spp.), nutritional factors, inflammation, allergies or neoplasia [8, 9] . So far, the role of C. difficile in canine enteric disease is still unclear due to the presence of toxigenic strains or their toxins in asymptomatic animals and the failure to reproduce CDI in healthy dogs with and without antibiotic treatment [9, 10] . It can be isolated from 0 to 57% of healthy dogs with no diarrhoea [11] . However, there are several studies which have reported an association between the presence of C. difficile toxins in faeces with diarrhoea, as well as with outbreaks of haemorrhagic diarrhoea in veterinary hospitals [9, [12] [13] [14] . But it is still unknown whether C. difficile represents an opportunistic pathogen, or simply a fortuitous finding in this animal species [10] .

There is limited information about antimicrobial susceptibilities of C. difficile strains obtained from dogs or exotic animal species [7, 15] . Metronidazole, and less frequently vancomycin, are used to treat CDI in dogs following the recommendations for human episodes [16] . These drugs are also commonly used to treat different kinds of exotic animal infections as well. The isolation of metronidazole-resistant strains or strains showing low susceptibility to this drug is growing in both humans and animals [17] , particularly in ribotypes 010 and 001 [18, 19] . In addition, resistance to metronidazole is heterogeneous and therefore C. difficile can show reduced susceptibility to this drug, which can be related to recurrent CDI cases [20] .

The objective of this study was to assess the presence of C. difficile in a population of diarrhoeic dogs and exotic animal species (both diarrhoeic and non-diarrhoeic), and to characterise the C. difficile isolates by the presence of toxin genes, their PCR-ribotype and toxinotype, and also their antimicrobial susceptibility pattern.

Diarrhoeic stool samples from dogs, submitted to a veterinary diagnostic laboratory located in the Zaragoza, NE of Spain were used. The samples were collected between October 2011 and November 2012. All of them were analysed for the presence of C. difficile by microbiological culture in our laboratory. The specimens were collected using commercial swabs with Amies transport medium (Deltalab, Barcelona, Spain) by veterinary staff at the time of the animal clinical examination. The samples originated from primary veterinary clinics located at different Spanish regions ( Table 1 ). All of them were sent under refrigerated conditions to the diagnostic laboratory, arrived within the 24 h after collection, and were kept at − 80°C until C. difficile culture was performed. In addition to C. difficile, faecal samples were analysed for other enteric pathogens upon request of the veterinarian in charge of the case (analysis carried out in the diagnostic laboratory; data not published). Dogs were classified by their age (0-4 months puppy; 5-12 months juvenile; 13-72 months adult; ≥73 months mature), gender, breed, location and presence/absence of dysbiosis. In addition, other laboratory data were recorded for each case when available.

Additionally, faecal samples were cultured from exotic animals from a veterinary clinic laboratory specialised in that kind of animals located in Barcelona, NE of Spain. The sampling period covered June and July 2013. The specimens derived from diarrhoeic and non-diarrhoeic animals. Each sample was taken by the veterinary staff during the examination of the animals and submitted to our laboratory under refrigerated conditions within the first 24 h after collection. As in the dog's study, the samples were kept at − 80°C until the culture for C. difficile was performed. Exotic animals were identified by species and grouped into three main animal categories (i.e. reptiles, small mammals and birds). No other data were available except being sick or not.

The isolation of C. difficile, molecular characterisation of the strains obtained (i.e. tpi housekeeping and toxin genes detection by PCR, identification of non-toxigenic strains, and PCR-ribotyping) and antimicrobial susceptibility testing was performed as described elsewhere [21] . The variability of the genes coding for toxins A and B (A3 and B1 fragments respectively) of toxigenic strains obtained was assessed through toxinotyping by PCR-RFLP [22] . Besides, those strains which showed unexpected results, i.e. negative PCR results for tcdA and tcdB genes and also non-toxigenic assay, were further studied by toxinotyping as well (A1, A2, A3, B1, B2 and B3 fragments) to test the possible presence of toxin gene fragments.

Multilocus sequence typing (MLST) technique was used for characterisation of strains which did not belong to a known PCR ribotype. By this technique, seven housekeeping loci are studied by PCR and subsequent sequenced, providing a sequence type (ST) profile and a clade [23] based on these genes alleles [24] . A maximum likelihood tree with 1000 bootstrap replicates was constructed based on the Kimura 2-parameter model [25] . A total of 37 strains were used for this purpose: one from this study (E6), 17 from previous works carried out by our team (Hu, RC and RF isolates; data not published for Hu isolates) [26] , and 19 from the PubMLST database to provide a context for C. difficile population (ST1, ST3, ST5, ST11, ST32, ST37-39, ST41, ST67, ST96, ST122, ST177-181, ST200 and ST206). Initial tree(s) for the heuristic search were obtained by applying the Neighbour-Joining method to a matrix of pairwise distances estimated using the Maximum Composite Likelihood approach. Evolutionary analyses were conducted in MEGA7 [27] .

Additionally, seven serial passages over 14 days were performed on Brucella blood agar plates without antibiotics in order to assess the stability of the initial metronidazoleresistant isolates, i.e. those which showed a breakpoint ≥32 μg/ml [28] . Then, the MIC to metronidazole was tested again by Etest as described above. Resistance was considered stable when the MIC of metronidazole against C. difficile remained (within ±1 dilution) after the passages. Resistance was considered unstable when resistant strains became susceptible (˂32 μg/ml) after the passages.

The prevalence of C. difficile was estimated for dogs and exotic pets separately. When possible, comparisons of C. difficile prevalence among the factors considered for dogs were assessed by univariable logistic regression. Thus, the outcome variable was the presence/absence of C. difficile and the explanatory variable was the corresponding factor (age, gender, breed, location and dysbiosis). Significance was set at p-values ≤0.05.

Faecal samples from a total of 90 dogs were analysed. The samples came from 42 different veterinary clinics located in 11 different Autonomous Communities in Spain, most of them (86.7%) located in the north/northeast of the country. The median age of dogs was 9.5 months (range 1-156) and 52.2% of the animals were females. A total of 29 dog breeds were included and grouped as large (e.g. German shepherd), medium (e.g. Beagle) and small size (e.g. Yorkshire). Clostridium difficile was isolated from 6 (6.7%) out of the 90 dogs analysed. All these results are summarised in Table 1 . CD Clostridium difficile; a Univariable logistic regression; b percentages calculated from the total number of samples; c age unknown for 14 samples (two of them are positive to non-toxigenic C. difficile); d breed unknown for eight samples (one of them is positive to toxigenic C. difficile and another one to non-toxigenic C. difficile)

Four (66.7%) strains with toxin genes (Table 2) yielded an A + B + CDT-genotype in all cases. Non-toxigenic strains showed a positive result with lok3/1 primers. Four different ribotypes were detected, and all toxigenic strains belonged to toxinotype 0 ( Table 2) . The four (100%) toxigenic C. difficile positive dogs showed dysbiosis and faeces without mucus, while only 57 (67.8%) out of 84 of the C. difficile negative dogs showed this problem. However, this difference was not statistically significant (p = 0.3). No relationship was observed between the presence of C. difficile and any of the other factors considered ( Table 1) . None of the dogs presenting toxigenic C. difficile yielded a positive result for another enteric pathogen (data not shown).

Twenty-four faecal samples derived from exotic species, 10 (41.7%) from diarrhoeic animals, 11 (45.8%) from nondiarrhoeic animals and three with unknown clinical data (12.5%). Fifteen (62.5%) were birds, mainly composed of different species of psittacines, 3 (12.5%) small mammals (lagomorphs, mustelids and rodents) and 6 (25%) sauropsid reptiles. Only one sample (4.2%; E6 strain, Table 2 ) belonging to a reptile (Pogona vitticeps) yielded a positive culture result for C. difficile. It was negative for all toxin genes tested (tcdA, tcdB, cdtA and cdtB) but also negative to the non-toxigenic PCR (see Additional file 1) and to the toxinotyping scheme (A1, A2, A3, B1, B2 and B3 fragments) (see Additional files 2 and 3). Interestingly, this isolate represented a new PCR-ribotyping type ( Table 2) and belonged to ST347, which was recently proposed as a Clade C-III member [29] (Fig. 1) .

All dog strains were susceptible to tetracycline and vancomycin (range between 0.064-0.38 μg/ml for vancomycin), whereas resistance to clindamycin, erythromycin, metronidazole and moxifloxacin varied (50%, 33.3%, 33.3% and 16.7% respectively) ( Table 3) . Two strains showed a MDR phenotype to clindamycin, erythromycin and metronidazole; resistance to metronidazole was unchanged and stable after a serial number of passages on antibiotic-free medium ( Table 2 ). The MICs of the remaining strains (n = 4) to metronidazole varied between 0.19-0.38 μg/ml.

One strain (D24) was considered resistant (and stable) to metronidazole with a heterogeneous pattern. After the 24 h incubation time, the isolate showed a fully susceptible phenotype to metronidazole (MIC 0.38 μg/ml) (Fig. 2 , growth II or GII; improved contrast image), but a subpopulation of tiny colonies (Fig. 2 , growth I or GI) was present inside the initial inhibition halo exhibiting a higher MIC (8 μg/ml). MIC results remained unharmed after three extra days of incubation. After observing these results, the growth I and II were sub-cultured separately in blood agar without antibiotics in order to repeat the susceptibility test to metronidazole following the protocol described above. After 48 h of incubation, both isolates showed metronidazole resistance (MICs 128 and 192 μg/ml, GI and GII respectively). The metronidazole resistance stability test was performed only for GI (Table 2) . A similar phenomenon was observed for this D24 isolate regarding its susceptibility to erythromycin and clindamycin. However, the subpopulations growing inside the initial halo reached a fully resistant phenotype to both antibiotics after 48 h of incubation. When the susceptibilities were repeated for the two different growths, both showed a fully resistant phenotype to these antimicrobials after 24 h.

The reptile C. difficile isolate (E6) was susceptible to all antimicrobial drugs (MICs range 0.19-2 μg/ml) tested except to tetracycline, for which a decreased susceptibility was found (MIC 6 μg/ml).

Clostridium difficile was isolated from 6.7% (6/90) of dogs with diarrhoea, but only four strains were toxigenic (66.7%). Overall, this prevalence was somewhat lower than that reported in previous studies of diarrheic dogs or dogs with various digestive disorders (29% - [30] , 12.1% - [15] and 10% - [31] ). However, these differences can be explained by variations of the study design (i.e. number and type of dogs included, geographical location, age distribution, etc.). They may be also due to the fact that samples were collected in veterinary clinics all around Spain and submitted to a veterinary diagnostic laboratory before arriving at our laboratory. It is well known that the survival rate of C. difficile may be compromised if faecal samples are stored under aerobic conditions [32] . It has also been reported that toxin-positive samples can yield a culture-negative result for C. difficile from dog faecal samples [9] . In addition, there were several limitations in this study to consider: the lack of the information on previous or ongoing antibiotic treatment data at the time of sampling. Among the ribotypes detected in this study the most frequent was the epidemic RT014 (3/6) which has been widely isolated from CDI cases in humans in several European countries, including Spain [33] . This ribotype has been also isolated from different animal species and types of samples, including dogs, retail meat and water [34] [35] [36] . Only one out of the six isolates from this study belonged to the non-toxigenic ribotype 010 which has been largely associated with dogs [18, 37, 38] and occasionally isolated from humans [5] . These results add more evidence to a The breakpoints for resistance established by the Clinical and Laboratory Standards Institute (CLSI) for anaerobic bacteria are those marked by b [50] . The breakpoint for tetracycline was ≥8 μg/ml [51] . The remaining breakpoints were based on the literature [7] possible inter-species C. difficile transmission as previously observed other authors [39, 40] . However, to the author's knowledge, there are no available data regarding the epidemiology of ribotypes 123 and 358 in Europe, which would suggest that they are not very common genotypes neither in humans nor in animals.

Metronidazole is the first choice to treat non-severe CDI in humans [41] and a very common treatment for CDI and Giardia spp. infections in dogs [16, 42] . Overall, the frequency of resistant C. difficile isolates to metronidazole is still low (0%-18%) [17] , and even lower after laboratory strain manipulation [28] . In this study, a higher proportion of strains (33.3%, ribotypes 010 and 014) revealed stable resistance to metronidazole. An increase in the number of metronidazole-resistance C. difficile isolates has been observed for the last years in both humans and animals strains [17] . Interestingly, higher MIC averages to this drug are frequently associated with the non-toxigenic RT010 [19] . Thus, despite the low number of positive samples, these results could reflect this trend in dogs in Spain. In addition, these two isolates were also multi-drug resistant to clindamycin, erythromycin and metronidazole, a pattern already described in dogs [15, 18] . Overall, these results suggest that antimicrobial susceptibility surveillance programs should be implemented in C. difficile strains isolated from dogs since they could be a possible source of metronidazole-resistant and MDR C. difficile. However, in vitro results should be interpreted with caution due to the limited information available regarding breakpoints for anaerobic microorganisms of veterinary relevance, and the fact that the antimicrobial intestinal concentrations are not known for all cases.

The number of resistant isolates to moxifloxacin in this study was relatively low (1/6), which was expected due to the low use of this drug in companion animals in Spain and the low prevalence of resistance to this drug reported in previous studies in dogs and other animal species [43] . Although the use of clindamycin and erythromycin in companion animals in Spain is low [44] , 50% and 33.3% of the isolates, respectively, were resistant to them. The resistance to clindamycin and erythromycin is common among C. difficile isolates in dogs [18] . This feature is mainly associated to ermB genes which are located in mobile elements, so these results probably reflect the spread of this determinant among C. difficile isolates regardless of their origin.

The results regarding strain D24 were unexpected. Heterogeneous and stable resistance to metronidazole was observed with high MIC values. The in vitro detection of subpopulations with reduced susceptibility to this antibiotic may be related to treatment failure observed in humans [28] . A similar phenomenon was observed on erythromycin and clindamycin susceptibility tests. The reason for these results is not clear and warrants further molecular studies.

The strain isolated from the lizard (E6) represents a new ribotype and it belongs to ST347 for which a new clade (C-III) was proposed [29] . A maximum likelihood tree was constructed using the E6 strain ST, sequence types from our C. difficile strain collection and from PubMLST database (Fig. 1) , and it showed that E6 strain clade is completely apart from the rest of defined and proposed new clades described until the termination of this study [29, 45, 46] . E6 seems to be a non-toxigenic isolate (tcdA, tcdB, cdtA and cdtB negative results by PCR), but it yielded also a negative result in nontoxigenic PCR which allows thinking that it has not the traditional non-coding region which replaces the PaLoc as it has been described before [45, 47, 48] . Besides, the possible presence of toxin A and B genes remnants was analysed by toxinotyping with negative results as well. Therefore, it seems appropriate to perform whole genome sequencing analysis to verify the relationship of this isolate with C. difficile population and to examine its PaLoc insertion site organisation [48] . It is reasonable to assume that a new genotype could be introduced in Spain by an exotic animal species as has happened in the past with other C. difficile genotypes which have spread among different continents [49] .

In this study, 4.4% toxigenic C. difficile strains were isolated from diarrhoeic dogs. Since the ribotypes found in dogs are also commonly found in humans, it is possible Fig. 2 Metronidazole susceptibility test of Clostridium difficile D24 strain after 48 h of incubation. GI, growth I; GII, growth II that dogs may act as a source of contamination for human beings or vice versa. However, further studies are needed in particular to define the likely role of pets as vectors, especially in light of the spread of pet therapy practices. There is also a potential risk for the emergence of C. difficile strains resistant to the first line of antimicrobial drugs used for the treatment of CDI in people. Metronidazole is widely used in dogs, thus animals treated with this drug could be a possible source of MDR strains for humans. Clostridium difficile antimicrobial susceptibility surveillance programs should be advised in this animal species. Exotic animals can harbour uncommon C. difficile strains, different than those already established in the country of destination. The molecular results observed for E6 C. difficile strain, which can be also found in companion animal species, prompt the need for genome sequencing analyses to study this unusual isolate.

Additional file 1: cdu1-cdd1 PCR (700 bp) agarose gel image from Clostridium difficile field isolates. Additional file 3: Clostridium difficile field isolates and Clostridium difficile ATCC 43255 (positive control) toxinotyping in agarose gel. Lanes 1-3 A2 fragment, 2 kb: ATTC strain, E6 strain and negative control respectively; lanes 4-6 B2 fragment, 2 kb: ATTC strain, E6 strain and negative control respectively; lanes 7-9 B3 fragment, 2 kb: ATTC strain, E6 strain and negative control respectively. M: 1 kb molecular mass (upper arrows point to 3 kb fragment and lower arrows point to 1 kb fragment) (JPEG 839 kb) 

Clostridium difficile infection: new developments in epidemiology and pathogenesis

Clostridium difficile in foods and animals: history and measures to reduce exposure

Risk factors for human disease emergence

PCR ribotyping of Clostridium difficile isolates originating from human and animal sources

Underdiagnosis of Clostridium difficile across Europe: the European, multicentre, prospective, biannual, point-prevalence study of Clostridium difficile infection in hospitalised patients with diarrhoea (EUCLID)

One health: zoonoses in the exotic animal practice

Shedding of Clostridium difficile PCR ribotype 078 by zoo animals, and report of an unstable metronidazole-resistant isolate from a zebra foal (Equus quagga burchellii)

Evaluation of a routine diagnostic fecal panel for dogs with diarrhea

The roles of Clostridium difficile and enterotoxigenic Clostridium perfringens in diarrhea in dogs

Enteropathogenic bacteria in dogs and cats: diagnosis, epidemiology, treatment, and control

Prevalence and distribution of Clostridium difficile PCR ribotypes in cats and dogs from animal shelters in Thuringia

Chronic diarrhoea in dogs associated with Clostridium difficile infection

Genotypic and phenotypic characterization of Clostridium perfringens and Clostridium difficile in diarrheic and healthy dogs

Outbreak of Clostridium difficile-associated disease in a small animal veterinary teaching hospital

Isolation of Clostridium difficile from dogs with digestive disorders, including stable metronidazole-resistant strains

Antimicrobial susceptibilities of canine Clostridium difficile and Clostridium perfringens isolates to commonly utilized antimicrobial drugs

Recent advances in the understanding of antibiotic resistance in Clostridium difficile infection

Antibiotic resistance patterns and PCR-ribotyping of Clostridium difficile strains isolated from swine and dogs in Italy

Analysis of metronidazole susceptibility in different Clostridium difficile PCR ribotypes

Comparison of Clostridium difficile isolates from individuals with recurrent and single episode of infection

Presence of Clostridium difficile in pig faecal samples and wild animal species associated with pig farms

An update on Clostridium difficile toxinotyping

Multilocus sequence typing of Clostridium difficile

A simple method for estimating evolutionary rate of base substitutions through comparative studies of nucleotide sequences

Molecular analysis of three Clostridium difficile strain genomes isolated from pig farm-related samples

MEGA7: molecular evolutionary genetics analysis version 7.0 for bigger datasets

Metronidazole resistance in Clostridium difficile is heterogeneous

Highly divergent Clostridium difficile strains isolated from the environment

Clostridium difficile in Dutch animals: their presence, characteristics and similarities with human isolates

Clostridium difficile in faeces from healthy dogs and dogs with diarrhea

Survival of Clostridium difficile and its toxins in equine feces: implications for diagnostic test selection and interpretation

Pan-European longitudinal surveillance of antibiotic resistance among prevalent Clostridium difficile ribotypes

International Clostridium difficile animal strain collection and large diversity of animal associated strains

Possible seasonality of Clostridium difficile in retail meat

The occurrence and high diversity of Clostridium difficile genotypes in rivers

Faecal shedding of antimicrobial-resistant Clostridium difficile strains by dogs

Prevalence of PCR ribotypes among Clostridium difficile isolates from pigs, calves, and other species

Clostridium difficile in food and domestic animals: a new foodborne pathogen

More than 50% of Clostridium difficile isolates from pet dogs in flagstaff, USA, carry toxigenic genotypes

European Society of Clinical Microbiology and Infectious Diseases: update of the treatment guidance document for Clostridium difficile infection

Update on the diagnosis and management of Giardia spp infections in dogs and cats

Antimicrobial susceptibility of animal and human isolates of Clostridium difficile by broth microdilution

Sales of veterinary antimicrobial agents in 26 EU/EEA countries in 2013. In: European surveillance of veterinary antimicrobial consumption

Evolutionary history of the Clostridium difficile pathogenicity locus

Comparative analysis of an expanded Clostridium difficile reference strain collection reveals genetic diversity and evolution through six lineages

A new type of toxin A-negative, toxin B-positive Clostridium difficile strain lacking a complete tcdA gene

Clostridium difficile: new insights into the evolution of the pathogenicity locus

Emergence and global spread of epidemic healthcare-associated Clostridium difficile

Performance standards for antimicrobial susceptibility testing; twentyfourth informational supplement. CLSI document M100-S24

Tetracycline resistance gene tet(W) in the pathogenic bacterium Clostridium difficile

We very much appreciate the collaboration of Laboratorios Albéitar (Zaragoza) and Lourdes Molina from Veterinary Diagnostic (Barcelona). We also thank Anabel Sánchez Bellido, Eloisa Sevilla, Pablo Magallón Verde and Yovana Bolea Albero for their technical assistance.

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

The datasets generated and/or analysed during the current study are available in the TreeBASE repository, http://purl.org/phylo/treebase/phylows/ study/TB2:S22326. The datasets generated and/or analysed during the current study are not publicly available due to the full clinical files from the animals whose samples were included in this work are confidential information (private diagnostic center), but are available from the corresponding author on reasonable request.Authors' contributions RB and MCT designed the study. MM analyzed and interpreted clinical dog data. SAL performed bacterial culture and molecular analysis, interpreted the results and wrote the manuscript. EK conducted PCR-ribotyping analysis. IMB performed molecular analysis. RCMJ performed the statistical analysis and interpreted the results. RB, MCT, RCMJ, MM, JLB, EK and IMB reviewed the work critically for important intellectual content. All authors read and approved the final manuscript.Ethics approval and consent to participate Not applicable.

Not applicable.

The authors declare that they have no competing interests.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.• We accept pre-submission inquiries • Our selector tool helps you to find the most relevant journal Submit your next manuscript to BioMed Central and we will help you at every step: