215 Veterinaria Italiana 2021, 57 (3), 215-226. doi: 10.12834/VetIt.2497.15848.1 Accepted: 01.09.2021 | Available on line: 31.12.2021 1National and OIE Reference Laboratory for Brucellosis, Istituto Zooprofilattico Sperimentale dell'Abruzzo e del Molise “G. Caporale”, Teramo, Italy. 2Direzione generale della sanità animale e dei farmaci veterinari, Ministero della Salute, Roma, Italy. 3Istituto Zooprofilattico Sperimentale dell'Umbria e delle Marche “Togo Rosati”, Perugia, Italy. *Corresponding author at: National and OIE Reference Laboratory for Brucellosis, Istituto Zooprofilattico Sperimentale dell'Abruzzo e del Molise “G. Caporale”, Campo Boario, 64100 Teramo, Italy. E-mail: d.averaimo@izs.it. Fabrizio De Massis1, Flavio Sacchini1, Daniela Averaimo1*, Giuliano Garofolo1, Pierdavide Lecchini2, Luigi Ruocco2, Roberto Lomolino2, Ugo Santucci2, Elisa Sgariglia3, Silvia Crotti3, Antonio Petrini1, Giacomo Migliorati1, Nicola D'Alterio1, Stefano Gavaudan3 and Manuela Tittarelli1 Keywords Brucella canis, Isolation, Dogs, Serology, cgMLST. Summary Brucella canis has been isolated for the first time in Italy in a commercial breeding kennel. It was diagnosed after a deep investigation related to the onset of reproductive disorders. Animals were tested with direct and indirect techniques. The agent was first detected in two Chihuahua aborted foetuses by direct culture. Further, it was also isolated from blood samples of dogs hosted in the kennel, which also showed reaction to conventional serological tests (microplate serum agglutination test). The isolates were identified as B.  canis by standard microbiological methods and a Bruce-ladder multiplex PCR. To investigate the genomic diversity, whole genome sequencing was used, applying the core genome Multilocus Sequence Typing (cgMLST ). In a first round of serological testing performed on 598 animals, 269 (46.1%) tested positive. In the second round of laboratory testing carried out 4-5 weeks apart, the number of serologically positive dogs was 241 out of 683 tested (35.3%), while the number of dogs positive to isolation was 68 out of 683 tested (10.0%). The PCR showed a lack of sensitivity when compared to direct isolation. The epidemiological investigation did not identify the source of the infection, given the time elapsed from the onset of abortions to the definitive diagnosis of B. canis infection in the kennel. The genomic analyses featured the strains as ST21 and, according to the cgMLST, revealed the presence of a tight cluster with a maximum diversity of four allelic differences. The observed limited genomic variation, largely within the known outbreak cut-offs, suggests that the outbreak herein described was likely caused by a single introduction. Moreover, in a broader scale comparison using the public available genomes, we found that the closest genome, isolated in China, differed by more than 50 alleles making not possible to find out the likely origin of the outbreak. The lack of updated data on B. canis genome sequences in the public databases, together with the limited information retrieved from the epidemiological investigations on the outbreak, hampered identification of the source of B. canis infection. First Isolation of Brucella canis from a breeding kennel in Italy First identified in 1966 (Carmichael 1966), B.  canis is a Gram-negative non-motile aerobic intracellular coccobacillus with rough colony morphology when grown on artificial medium. Dogs and wild Canidae are the only animal species that act as reservoirs of B.  canis under natural conditions (Shin and Carmichael 1999). Canine Brucellosis due to B.  canis is a contagious disease characterized by abortions in females and Introduction Brucellosis is a zoonosis, widespread all over the world, caused by bacteria belonging to the genus Brucella. The most clinically relevant Brucella species are Brucella melitensis, B. abortus, B. suis and B. canis. They tend to be host-adapted, although infections may occur in other animal species, including humans (Michaux-Charachon et al. 2002). 216 Veterinaria Italiana 2021, 57 (3), 215-226. doi: 10.12834/VetIt.2497.15848.1 First isolation of Brucella canis in Italy De Massis et al. Unlike many bacterial infections, canine brucellosis is not amenable to treatment with antibiotics (NASPHV 2012, Carmichael and Greene 2013). Its intracellular persistence makes B.  canis a difficult target for antibiotics, despite its susceptibility in  vitro. Treatment failures or relapses are common (Carmichael and Greene 2013), leading some entities to recommend that infected dogs, regardless of whether they are treated or not, should not be rehomed from infected facilities (USDA 2015). Canine brucellosis due to B.  canis is considered a zoonosis (Blankenship and Sanford 1975). The disease is underestimated in man due to general lack of serological testing facilities and misconceptions concerning its prevalence. Confirmed cases of human illness due to B.  canis are relatively uncommon, with roughly 50 cases identified in the US since 1973 (Daly et  al. 2020). However, because of the vague symptoms and effects in infected people, it is likely that human cases of canine brucellosis are under-diagnosed and underreported (NASPHV 2012). Those working closely with potentially infected animals, such as in breeding kennels or with stray animals, are considered at higher risk than others (Hensel et  al. 2018). Culture-positive cases have been reported in laboratory personnel, animal technicians and persons known to have close and frequent contact with infected dogs (Carmichael et al. 1980). B.  canis is considered endemic in Southern USA, in Central and South America and in Mexico (Flores-Castro et al. 1977, Brower et al. 2007, Küster de Paula Dreer et al. 2013, Krueger et al. 2014, Keid et al. 2017). It has been also reported in Canada (Cosford et  al. 2018). Infections with B.  canis have been reported from Asian countries such as China (Di et al. 2014), Japan (Hayashi and Ysayama 1977), India (Yoak et al. 2014) and from African countries such as Nigeria (Cadmus et  al. 2011) and Zimbabwe (Chinyoka et  al. 2014). New Zealand and Australia appear to be free of B.  canis; however, Australia has reported some B. suis infections in dogs, mainly in animals used to hunt feral pigs. (Mor et al. 2016, Rovid Spickler 2018, Gardner and Reichel 1997). Cases have been reported also from European countries, such as Austria (Hofer et  al. 2012), Germany (Von Kruedener 1976, Nöckler et al. 2003), Hungary (Gyuranecz et al. 2011), Sweden (Holst et  al. 2012, Kaden et  al. 2014), Switzerland (Egloff et  al. 2018), and the United Kingdom (Taylor 1980, Whatmore et al. 2017). The only report recorded in Italy so far has been a presumptive B. canis infection in a dog with chronic prostatitis and discospondylitis, detected by PCR (Corrente et al. 2010). Aims of this paper are to report the first isolation of B. canis from a commercial breeding kennel in Italy, to describe the case, and to inform about results of testing. epididymitis, testicular atrophy, prostatitis and infertility in males (Wanke 2004). The disease is insidious, and many dogs do not have prominent signs, otherwise some dogs with generalized infection could show uveitis, discospondylitis and other more chronic conditions such as lymphadenomegaly, osteomyelitis, polyarthritis, meningoencephalitis, and pyogranulomatous dermatitis (Carmichael and Greene 2013). The disease is of particular importance for dog breeders since infection with B.  canis usually ends a dog's reproductive career (Carmichael 1990). The most common routes of transmission of B. canis to humans are through contact with infected dogs, which may disseminate the bacteria with their secretions for many months after bacteraemia has ceased, and through direct laboratory exposure (Carmichael and Shin 1996). Dog to dog transmission occurs during breeding, or through oronasal contact with reproductive discharges following abortions. B.  canis may also be shed with urine, feces, and nasal and ocular secretions. Pups may be infected in utero or perinatally (Carmichael and Greene 2013). These bacteria may be transmitted either by the venereal or oral route, more frequently infection occurs following contact with abortive material. In males, urine and seminal fluid represent an important source of infection (George et  al. 1979). Prolonged bacteraemia is a typical sign of canine brucellosis that persists from 6 months to 5 years (Carmichael et al. 1984). Prevalence and spreading of B.  canis infection in canine populations depend upon several individual factors, including age and reproductive status (Carmichael and Greene 2013, Kaufmann and Petersen 2019). In endemic countries, infection rates are typically higher in stray animals, as they are more likely to be reproductively intact and active compared to owned pets. Breeding kennels can also be sites of higher-than-normal infection rates (Carmichael and Greene 2013). Identifying brucellosis-infected dogs is often challenging. Definitive diagnosis requires the culture of B.  canis from the blood of infected dogs, a process that has a relatively low sensitivity, is time-consuming, and somewhat technically impractical to apply to large populations of dogs. Some serological methods have been developed as commercial kits and test services are available at diagnostic laboratories. False-positive and false-negative results may occur with these tests (Keid et  al. 2009), however, in many situations they are the tests of choice for screening larger populations, such as stray dogs or those entering shelters (Carmichael and Greene 2013). 217Veterinaria Italiana 2021, 57 (3), 215-226. doi: 10.12834/VetIt.2497.15848.1 De Massis et al. First isolation of Brucella canis in Italy the NRC, IZSAM. Four-five weeks later, a second round of sampling involved 683 animals and this time both serum and EDTA blood were collected from each animal. Overall, serum samples were collected from 683 dogs and submitted to the NRC, IZSAM. Serological analyses were carried out using a microplate serum agglutination test (mSAT). The test was carried out modifying the tube agglutination test described by Alton and colleagues (Alton et  al. 1988), and volumes were adapted to be performed in 96-well U-shaped microplates. Briefly, B.  canis strain RM66 was used to prepare mSAT antigen as previously described (Alton et  al. 1988). Before testing, serum samples were diluted 1:10 in Tris-maleate buffer (TMB) pH 9.0 ± 0.5. The assay was performed by dispensing equal volumes (50 µl) of sera 2-fold serially diluted with TMB and B.  canis antigen in a 96 well U-shaped microplate, to obtain a final dilution ranging from 1:20 to 1:640. Plates were incubated at 37 ± 2 °C for 48 h. Samples displaying 100% agglutination at a dilution ≥ 1:20 were considered as positive. Serology titres were indicated as the highest dilution of serum showing 100% agglutination. Furthermore, in order to acquire information on specificity of mSAT, a panel of 143 samples from owned dogs non-related to the outbreak was also tested with this method. Bacteriological investigation A total of 683 dogs of different breed were sampled and a whole blood sample was taken from each of them and submitted to the NRC, IZSAM. Whole blood samples were cultured for detection of Brucella spp. according to the literature methods (Carmichael and Greene 2006, CFSPH 2018, GDA 2020). Briefly, samples were both streaked onto selective Farrell’s Brucella Agar (IZSAM) and inoculated in enrichment broth supplemented with equine serum (Brucella Broth, IZSAM). All media were incubated at 37 ± 1 °C. Subcultures in Farrell’s Brucella Agar of all the broth cultured samples were made weekly for a month. Plates were examined daily for evidence of growth and were considered negative when no colonies were seen neither from direct culture nor from all subcultures. Typical Brucella spp. colonies were subjected to specific PCR assay for species identification (OIE 2016). PCR assay DNA was extracted from all the samples (whole blood and suspected colonies) using the Maxwell® 16 Blood & Cell DNA Purification Kit (Promega Italia Srl, Milan, Italy), following the manufacturer’s specifications. After extraction, DNA was collected in DNA/RNA free tubes and stored at 4 ± 2 °C until Materials and methods Background and diagnostic screening In April 2020, two Chihuahua aborted foetuses were submitted to the Ancona Laboratory of the Istituto Zooprofilattico Sperimentale dell'Umbria e delle Marche (IZSUM) for the detection of canine reproductive pathogens. These specimens were collected from a kennel where several months of abortion, infertility and reproductive disorders were reported by the private veterinarians caring for the kennel. At time of sampling, the kennel hosted approximately 600 dogs, mostly Chihuahua breed, but also some Pomeranian, Maltese and Toy Poodle breeds. Samples collected from the two foetuses were subjected to standard bacteriological investigations and molecular tests for the following pathogens: Canine Herpes virus (CHV; real-time PCR); Brucella spp. (real-time PCR); Leptospira spp. (real-time PCR); Chlamydia spp. (real-time PCR); Mycoplasma spp. (PCR); Canine Parvovirus (CPV; real-time PCR). Brucella spp. were isolated in sheep blood agar, after incubation overnight at 37  °C (± 2 °C). Cultures were carried out from brain, spleen, stomach and lung. Identification of Brucella genus was confirmed by MALDI-TOF and PCR, both from colonies and pooled viscera. The strain isolated was submitted for identification to the National Reference Centre for Brucellosis (NRC), Istituto Zooprofilattico Sperimentale dell'Abruzzo e del Molise (IZSAM), where Bruce-ladder assay identified B.  canis (OIE 2016). Following the confirmation of B. canis isolation in the kennel, a bacteriological and serological survey was carried out, to investigate on the magnitude of the outbreak. Samples were collected by Local Veterinary Services according to Italian and European regulations for animal welfare. Samples were taken from the radial vessel with the Vacutainer™ system in anticoagulant tubes, preventing blood from clotting. After collection, blood samples were kept refrigerated until delivered to the laboratory and then stored at 4 ± 2 °C until analysis. Population data All dogs were identified with a microchip; data regarding age, gender, and breed were collected from the National and Regional Databases for the identification and registration of dogs. Serological investigation After the confirmation of B.  canis outbreak in the kennel, a first round of serological sampling was carried out on 598 animals and submitted to 218 Veterinaria Italiana 2021, 57 (3), 215-226. doi: 10.12834/VetIt.2497.15848.1 First isolation of Brucella canis in Italy De Massis et al. were ignored in the calculation of distance between pairs of sample profiles. All sequences were additionally typed using the Brucella 9 locus MLST scheme available at https://pubmlst.org/brucella/ (Whatmore et  al. 2007) accessible through Ridom SeqSphere+. Statistical analysis Data from Laboratory Information Management System were imported in MS Access® (Microsoft Access 2019, Redmond, Washington, DC, USA), which was used for cleaning and normalizing the dataset. To take into account the uncertainty of the proportion of positive laboratory results over the total tests performed, a beta distribution was used to define the 95% confidence interval of the proportion accuracy. The uncertainty interval was defined as the difference between upper and lower 95% confidence limits. The 95% lower and upper credibility levels (L.C.I. and U.C.L., respectively, composing the Credibility Interval, CI) of the distribution frequency of positive results were calculated using a Bayesian approach (Sivia 1996) with a beta distribution (n + 1; n - s + 1), where n is the total number of tested samples and s are the tested positive samples. Results Population data The total number of dogs considered in this study was 683. Out of them, 475 (69.5%) were females and 208 (30.5%) were males. The distribution of dogs by age at the time of sampling is shown in Figure  1. The dogs in the kennel were 2.8 years old in average, with the oldest dog being 8.5 year old. The most represented breed was Chihuahua (605 dogs, 88.6%), followed by Spitz (37 dogs, 5.4%), analysis. The reaction mix was prepared using the Brucella genus (all species) Genesig™ Advanced Kit (Genesig, York House, School Lane, Chandler's Ford, UK) in a final volume of 20 µl consisting of 5 µl of extracted DNA and 15 µl of master mix, according to the manufacturer's instructions. The samples were analyzed by real-time PCR. PCR was run in a QuantStudio™ 7 Pro real-time PCR System (Applied Biosystems, CA, USA) under the following conditions: enzyme activation step at 95 °C for 2 min, 50 cycles of denaturation at 95 °C for 10 s and data collection at 60 °C for 60 s. A positive control (K+) and a No-Template Control (NTC) were included in each run. Data were analyzed by Design and Analysis Software 2.4.3. Whole‑genome sequencing (WGS) A total of 67 B. canis strains each isolated from a single dog were submitted to WGS. The genomic DNA extracted from bacterial colonies was quantified with the Qubit fluorometer (QubitTM DNA HS assay; Life Technologies, Thermo Fisher Scientific Inc., Waltham, MA, USA). Sequencing libraries were prepared using Nextera XT library preparation kit (Illumina Inc., San Diego, CA, USA) according to the manufacturer’s instructions. The libraries were sequenced using the Illumina NextSeq 500 platform, producing 150-bp paired-end reads. After demultiplexing and removal of adapters, reads were trimmed from 50 and 30 ends using Trimmomatic tool version 0.36 to discard the nucleotides with quality scores of less than 25. Reads shorter than 36 bp were automatically discarded. Scaffolds were assembled with SPAdes version 3.11.1 with the careful option selected (Bankevich et  al. 2012). Read sequences were submitted to Sequence Read Archive of the National Center for Biotechnology Information (NCBI) under the BioProject accession number PRJNA748851. Multilocus sequence typing (MLST) and core genome (cg)MLST Analysis Genome assemblies produced in our study, along with 67 public genomes available at GenBank (accessed on 28 October 2020), were genotyped using cgMLST. The cgMLST profiles were assigned using the task Template B. melitensis/suis/canis/ abortus cgMLST with 2076 targets core genes in Ridom SeqSphere+ software, v4.1.1 (Ridom GmbH, Münster, Germany) as described by Janowicz and colleagues (Janowicz et al. 2018). Multiple spanning tree (MST) was generated by pairwise comparison of cgMLST target genes using default parameters. The Neighbour-Joining Tree was constructed by the distance allele table and circular tree was visualized using iTol (Letunic and Bork 2019). Missing values Pe rc en ta g e Age (Years) 30% 25% 20% 15% 10% 5% 0% 17.6% 0-1 17.6% 1-2 14.2% 3-4 7.9% 4-5 7.5% 5-6 5.3% 6-7 1.0% 7-8 0.6% 8-9 28.4% 2-3 Figure 1. Age distribution of dogs at the time of sampling. 219Veterinaria Italiana 2021, 57 (3), 215-226. doi: 10.12834/VetIt.2497.15848.1 De Massis et al. First isolation of Brucella canis in Italy identified 138 samples as negative, demonstrating a specificity of 96.5% (CI 92.1%-98.5%). The distribution of seropositivity was different between males and females. Overall, 48 males were positive out of 208 (23.1%), while 193 females were positive out of 475 (40.6%). This difference was found as significant (χ2 = 19.52; p < 0.01). When age classes were considered, the distribution of seropositivity was higher in the classes from 2 to 7 years. Lower rates of seropositivity were recorded above and under that age interval (Table II). When considering the credibility intervals, the number of seropositives under one year of age was significantly lower than the other age classes. The distribution of seropositivity according to breed is shown in Table III. The seropositivity in Poodle breed was significantly higher than the seropositivity recorded for Chihuahua breed (χ2 = 9.38; p < 0.01). Brucella isolation and identification Out of the 683 whole blood samples collected, 68  were positive for microbiological isolation. All strains isolated were confirmed as B.  canis by PCR assay. Moreover, 61 samples grew on direct sowing, while seven strains were detected only by subcultures from enrichment broth. Furthermore, real-time PCR allowed the identification of 32 positive dogs. Results are summarized in Table IV. When analysing the performances of the two tests, the Cohen’s Kappa value obtained was 0.381, Poodle (29 dogs, 4.2%), Maltese (7 dogs, 1.0%), Pug (3 dogs, 0.4%), Tibetan mastiff (1 dog, 0.1%), and a cross-breed dog (1 dog, 0.1%). Serology In the first round of sampling, 269 tested positive out of 598 animals sampled. Given that 15 sera were not analysed because haemolytic, the apparent seroprevalence was 46.1%. In the second round of sampling, out of the 683 serum samples collected, 241 were positive, for an apparent prevalence of 35.3%. Among the 241 seropositive animals, 64  were also positive to blood culture (26.6%). In four seronegative animals, B.  canis was isolated from blood cultures (Table I). When analysing the performances of the two tests, the Cohen’s Kappa value obtained was 0.307, considered as fair in the interpretation of strength of agreement suggested by Landis and Kock (Landis and Kock 1977). Comparing the results of serological testing during the first and second round of sampling, 220 animals remained serologically positive, 45 animals became negative, while 15 animals became seropositive. When considering the same group of animals for the first and second round of sampling, seroprevalence decreased from 46.1% to 40.0%. We also calculated the specificity of mSAT on 143 sera from owned dogs not-related to the outbreak. mSAT Table I. Comparison of positive and negative results obtained by B. canis microplate serum agglutination test (mSAT) and Brucella spp. isolation method. Serology (mSAT) Positive Negative Total Isolation Positive 64 4 68 Negative 177 438 615 Total 241 442 683 Table II. Distribution of seropositivity by age classes. Age class N. tested N. positive % positive L.C.L. U.C.L. 0 - 1 120 5 4.2% 1.8% 9.4% 1 - 2 120 40 33.3% 25.5% 42.2% 2 - 3 194 87 44.8% 38.0% 51.9% 3 - 4 97 43 44.3% 34.8% 54.3% 4 - 5 54 26 48.1% 35.4% 61.2% 5 - 6 51 22 43.1% 30.5% 56.8% 6 - 7 36 15 41.7% 27.1% 57.9% 7 - 8 7 2 28.6% 8.5% 65.1% 8 - 9 4 1 25.0% 5.3% 71.6% Total 683 241 35.3% 31.8% 38.9% Table III. Distribution of seropositivity by breed. Breed N. tested N. positive % positive L.C.L. U.C.L. Chihuahua 605 207 34.2% 30.5% 38.1% Spitz 37 13 35.1% 21.8% 51.4% Poodle 29 18 62.1% 43.9% 77.3% Maltese 7 1 14.3% 3.2% 52.7% Pug 3 2 66.7% 19.4% 93.2% Cross-Breed 1 0 0.0% 1.3% 84.2% Tibetan Mastiff 1 0 0.0% 1.3% 84.2% Total 683 241 35.3% 31.8% 38.9% Table IV. Comparison of positive and negative results obtained on EDTA blood samples by isolation and real-time PCR for B. canis identification. Real-time PCR Positive Negative Total Isolation Positive 21 47 68 Negative 11 604 615 Total 32 651 683 220 Veterinaria Italiana 2021, 57 (3), 215-226. doi: 10.12834/VetIt.2497.15848.1 First isolation of Brucella canis in Italy De Massis et al. strains fell in the ST21 group, identifying that closest genome was isolated in China on an unspecified date (Figure 3). Discussion and conclusions This is the first report of the isolation of B.  canis in dogs in Italy, as well as the first report of the occurrence of an outbreak in an Italian commercial breeding kennel. Moreover, and to the best of our knowledge, this is also the first report worldwide of an outbreak of B. canis infection involving very high number of dogs. Until the bacterial isolation described in this study, brucellosis due to B. canis was considered a foreign disease for Italy. Previous recordings were based on serological evidence only, with the exception of a single report of B.  canis identification by PCR, in a dog with chronic prostatitis and discospondylitis (Corrente et  al. 2010). Private veterinarians or considered as fair in the interpretation of strength of agreement suggested by Landis and Kock (Landis and Kock 1977). The distribution of dogs positive to B. canis isolation was different between males and females. Overall, 13 males were positive out of 208 (6.3%), while 55 females were positive out of 475 (11.6%). This difference was found as significant (χ2  =  4.58; p  <  0.05). When age classes were considered, the percentage of dogs positive to bacteriology was higher in the classes from 2 to 3 years, and from 7 to 8 years (Table V). However, when considering the credibility intervals, no significant difference was recorded between age classes. The distribution of positivity to isolation according to breed is shown in Table VI. The rate of isolation in Poodle breed was significantly higher than the rate recorded for Chihuahua breed (χ2 = 19.92; p < 0.01) Genomic analysis Genome assemblies were used to retrieve MLST profiles, which identified for all the strains analysed the Sequence Type (ST) 21. The assignation of the cgMLST profiles comprising 2074 targets retrieved at least 98.7% of them for all the strain analysed. The cluster analysis from the outbreak revealed that all the strains belong to the same cluster (Figure 2). The maximum distance observed between strains isolated in the outbreak was four allelic differences, thus confirming the hypothesis that the outbreak has been generated by a single introduction of B. canis in the kennel. The comparison of the cgMLST profiles against the public genomes revealed that the B. canis population was divided in two main groups corresponding to ST20, which was previously linked to South, Central and North America, and ST21 mostly linked to Asia and Europe (Vicente et  al. 2018). The outbreak Table V. Distribution of dogs positive to B. canis isolation according to age classes. Age class N. tested N. positive % positive L.C.L. U.C.L. 0 - 1 120 6 5.0% 2.4% 10.5% 1 - 2 120 13 10.8% 6.5% 17.7% 2 - 3 194 26 13.4% 9.3% 18.9% 3 - 4 97 11 11.3% 6.5% 19.2% 4 - 5 54 3 5.6% 2.0% 15.1% 5 - 6 51 5 9.8% 4.4% 21.0% 6 - 7 36 3 8.3% 3.0% 21.9% 7 - 8 7 1 14.3% 3.2% 52.7% 8 - 9 4 0 0.0% 0.0% 45.1% Total 683 68 Table VI. Distribution of dogs positive to B. canis isolation by breed. Breed N. tested N. positive % positive L.C.L. U.C.L. Chihuahua 605 54 8.9% 6.9% 11.5% Spitz 37 4 10.8% 4.4% 24.8% Poodle 29 10 34.5% 19.9% 52.8% Maltese 7 0 0.0% 0.3% 36.9% Pug 3 0 0.0% 0.6% 60.2% Cross-Breed 1 0 0.0% 1.3% 84.2% Tibetan Mastiff 1 0 0.0% 1.3% 84.2% Total 683 68 10.0% 7.9% 12.4% 1 1 1 1 1 1 1 1 1 1 1 2 2 Gt-9 Gt-6 b Gt-3 Gt-14 Gt-8 Gt-10 Gt-12 Gt-1a Gt-2 Gt-4 Gt-5 Gt-7 c Gt-11d Gt-13 1 1 1 1 1 1 1 1 1 1 1 2 2 Figure 2. Minimum spanning tree generated for 67 dog isolates. Each circle represents an allelic profile. The numbers on the connecting lines illustrate the numbers of target genes with differing alleles. 221Veterinaria Italiana 2021, 57 (3), 215-226. doi: 10.12834/VetIt.2497.15848.1 De Massis et al. First isolation of Brucella canis in Italy brucellosis control and eradication in ruminants would not be acceptable for application in dogs, both for animal welfare and ethical reasons. On the other hand, due to possible health implications, either for dogs and their owners, actions are required to acquire data on distribution of this underhanded zoonosis of pets, as well as for controlling his spread in situations where it occurs. Thus, protocols are necessary for proper detection, control, and eradication of the infection in kennels and in owned dogs. Laboratory tests are a cornerstone for B.  canis diagnosis and control. Serological testing represents a valid screening tool for evaluating the presence of B.  canis in breeding kennels. Based on serological results, apparent seroprevalence during the first round of sampling (598 animals) was 46.1% while in the second round of sampling (683 animals) public health authorities have rarely investigated this zoonosis, also due to the lack of regulations on canine brucellosis surveillance or provisions for specific controls in dog international trade for this disease. This contributed to the current lack of data on B.  canis diffusion among the Italian dog population, and to the limited knowledge about the disease distribution in the European Union (EU) canine population. Furthermore, considering its zoonotic potential, uncontrolled spread of B.  canis may have important public health implications. In Italy, the veterinary regulations do not provide specific restrictive measures for canine brucellosis caused by B.  canis. Actually, brucellosis and agents thereof are included in the list of the zoonosis requiring surveillance on Annex I of the so-called “Zoonoses Directive” (EU 2003). However, in the context of canine brucellosis, current rules for GCF 900497665 1 09 369 776 1 genomic 21 FINLAND GC F 0 00 48 02 75 1 B ruc ca ni 96 72 58 V1 ge no mi c 2 0 ERR473754 20 ERR467400 20 ER R4 16 54 08 20 GCF 004127195 1 ASM412719v1 genomic 21 CHINA GC F 0 00 53 04 95 1 O live ri g en om ic 2 0 ERR467401 20 GCF 00 029218 5 1 ASM 29218v 2 geno mic 21 CHINA 2020-TE-128612-2-1 ST21 outbreak Italy GCF 00 069 158 5 1 AS M6 915 8v1 ge nom ic 2 0 S WE DEN GC F 0 01 07 83 35 1 B c an is S CL ge no mi c 2 0 C HIL E ERR485955 20 ERR473749 20 SRR 587 932 1 2 0 ER R2 13 65 46 2 0 Br az il SRR40390 06 20 GCF 0007 4033 5 1 A SM7 4033 v1 g enom ic 20GC F 0 00 36 68 25 1 Br uc ca ni UK 10 02 V1 ge no mi c 2 0 ER R4 16 54 10 20 ERR473748 20 ER R2 13 65 47 un kn ow n B ra zil GCF 003010715 1 ASM301071v1 genomic 21 CHINA GCF 00 259 125 5 1 AS M2 591 25v 1 g eno mic 20 ERR473747 20 GC F 0 01 71 53 65 1 A SM 17 15 36 v1 ge no mi c 2 0 U SA SRR 403 900 7 2 0 SRR5484406 20 U S GCF 000367305 1 Bruc cani F7 05A V1 genomic21 SOUTH AFRICA ERR473755 20 GCF 001715415 1 ASM171540v1 21 USA GC F 9 00 49 18 95 1 07 28 59 60 71 ge no m ic un kn ow n B RA SIL GCF 001 715 445 1 A SM1 715 44v 1 ge nom ic 20 USA ERR418017 20 GCF 00029 8575 1 V1 0 for Bruce lla canis str 118 genom ic 21 CHINA SRR5 4841 36 20 US GCF 000480295 1 Bruc cani 04 2330 1 V1 genomic 20 ER R2 13 65 49 2 0 Br az il ERR2136545 20 Spain ERR554821 20 ER R4 16 54 07 20 GCF 001758265 1 ASM175826v1 genomic 21 SWEDEN GC F 0 00 36 72 85 1 Br uc ca ni CN GB 13 24 V 1 g en om ic 20 AR GE NT IN A ERR473750 20 SR R6 42 80 9 2 0 A rg en tin a ERR473752 20ERR473753 20 SRR54 83825 20 US ER R4 16 54 09 20 GCF 004127225 1 ASM412722v1 genomic 21 CHINA ERR473746 20 GCF 000 018 525 1 A SM1 852 v1 g eno mic 20 GC F 0 00 37 06 05 1 Br uc ca ni CN GB 51 3 V 1 g en om ic 2 0 C HI LE ERR418019 20 ERR485956 20 ERR418018 20 GC F 9 00 49 19 15 1 0 7 28 59 6 07 0 ge no m ic 20 B RA SI L ERR473756 20 ERR473757 20ERR473751 20 GCF 000238195 1 ASM 23819v1 genomic 21 S OUTH KOREA ERR554818 20 ERR473758 20 ERR2136548 21 Finland ERR554819 20 LISARB 02 cimoneg 96401 1 529194009 FCG GCF 000370585 1 Bruc cani 79 122 V1 genomic 21 JAPAN GCF 900491905 1 96 9626 genomic 20 SPAIN ST21 ST20 Figure 3. Neighbour-Joining tree of B. canis. The tree was constructed using distance allele table sequences of 68 isolates and mid-point rooted. The label from the outbreak clone is reported in red. Sequence type groups are reported in green for ST20 and red for ST21. 222 Veterinaria Italiana 2021, 57 (3), 215-226. doi: 10.12834/VetIt.2497.15848.1 First isolation of Brucella canis in Italy De Massis et al. than in males, both for serology and isolation tests. Actually, in a free-living dog population and in a situation of uncontrolled spread of the disease, it would be expected to have a similar prevalence in both genders. However, giving that the outbreak occurred in a commercial breeding kennel, this finding could be the result of breeding practices. In fact, one hypothesis is that to maintain a good bloodline not all males were allowed to mate and then only a fraction of them was actually exposed to the infection from the venereal route. The rate of positivity was distributed evenly across age classes, except for dogs aged one year or under. This is suggestive of an infection that remained uncontrolled for a long period of time, favouring the disease spread among dogs of all age classes, both through mating and environmental contamination. Similar observations on the low susceptibility of animals under the age of reproduction to brucellosis were also reported in other animal species (Nicoletti 1980). However, it is not clear if this resistance or lower susceptibility of young animals is related to immature reproductive system, a reduced time of exposure to infection, or other factors that remain to be investigated. When looking at the breed involved in the B.  canis outbreak, positivity was found significantly higher in Poodles, even if this breed was low in numbers compared to Chihuahuas. This may suggest that the infection was introduced in the kennel with this breed and then was transmitted to other dogs due to environmental contamination after abortion. The fair agreement recorded between isolation and PCR testing procedures (Cohen’s Kappa value 0.381), considering that the isolation gave positive results in a higher number of specimens with respect to PCR, suggests that the PCR assay has difficulties in performing on blood matrix. In the literature the sensitivity of PCR for Brucella is very variable when performed on blood matrix, ranging from 3.26% (Kauffman et  al. 2012) to 97.14% (Keid et  al. 2010). This may be due to the presence of PCR-inhibitory molecules like haemoglobin (Sidstedt et  al. 2018), which may affect the amplification process through different mechanisms or lead to a failure of amplicons detection. Not only the presence of inhibitors, but also the use of antibiotics and heparin blood sampling could alter the sensitivity of the PCR results (Mol et al. 2020). To improve the performance of the method, it is possible to proceed with centrifugation of the blood sample and extraction of the DNA from buffy coat. However, larger quantities of blood are required for this purpose, and often they are not available from dogs due to the breed size. Also because Brucella is a facultative intracellular bacterium, the DNA yield during extraction may be low. This critical point may be solved by sowing decreased to 35.3%. These data resemble or even exceed the highest values reported in the literature (Hensel et al. 2018), giving a clear number of the impact the disease may have in a kennel. The high seroprevalence is probably related and consequent to the high density dog population hosted in the kennel, in addition to inadequate kennel management, both factors contributing to B. canis spreading. We observed a reduced apparent seroprevalence between the first and second sampling. This result is partially explained with the increased number of animal tested during the second sampling. Actually, the additional population tested in second sampling was composed by young animals, and most of them resulted seronegative. Someone may also argue that the decreased prevalence observed was related to the antibody isotype switching that does occur as the disease progresses, when the most prevalent B. canis specific antibodies found in serum after infection shift from IgMs to IgGs. IgGs are characterised by lower agglutination properties compared to IgMs. In the hypothesis that mSAT mainly detect IgMs, mSAT would have a reduced sensitivity on sera with high concentrations of B. canis specific IgGs but low IgMs. In our opinion, this was not the case and previous studies in human, performed on similar test, showed that even when IgMs are removed, serum agglutination capacity, despite reduced, was maintained by IgG and IgA antibodies (Marrodan et al. 2001). This suggests that also animals with lower amount of IgMs compared to IgGs should test positive to mSAT. On the other hand, these observations support the use in parallel of serological tests that evaluate IgG antibodies, in addition to IgMs, this in order to maximize the chance of detecting seropositive animals. Finally, the non-optimal specificity observed for mSAT may have influenced the different prevalence observed between the first and second sampling. In fact, 32  animals that tested positive to mSAT during the first sampling, with an antibody titer around the cut-off value, turned negative few weeks later. A fair agreement was recorded between isolation and mSAT testing procedures (Cohen’s Kappa value 0.307). During the outbreak investigation, we identified four cases of active infection and bacteraemia that tested negative to serological tests. As previously described (Carmichael 1990, Carmichael and Greene 2006) antibody response only appears 5-8 weeks after infection and animals with ongoing bacteraemia may result negative to serological tests for 3-4 weeks post infection. Similarly, animals with chronic infection may also result negative to serological screening (Carmichael and Greene 2006). The study also evaluated the disease prevalence among gender and age. The rate of positivity to laboratory tests was significantly higher in females 223Veterinaria Italiana 2021, 57 (3), 215-226. doi: 10.12834/VetIt.2497.15848.1 De Massis et al. First isolation of Brucella canis in Italy Further circumstances prevented the identification of the source of infection in the kennel under study. Dogs were kept in a commercial breeding kennel, where epidemiological investigation on movements of breeding dogs were not conclusive. Furthermore, from the occurrence of abortions to the diagnosis of canine brucellosis several months elapsed, allowing the infection spreading largely in the kennel, thus hampering the possibility to identify the origin of the infection and, therefore, to trace back its source. Public and private veterinarians should be trained in canine brucellosis diagnosis, prevention and control, and should investigate the causes of abortion in dogs considering also B.  canis. More research on the proper use of antimicrobials in affected dogs is advisable, at least until a reliable vaccine will become available. This highlights also the importance of reminding breeders of the clinical signs of canine brucellosis and their responsibility to prevent intra-population and inter-population spread of the disease and possible human infections. Surviving puppies can be an important source of B.  canis infection, as they can become permanent carriers and shedders of the pathogen. All cases of canine abortion should be examined for brucellosis by bacterial culture of the foetuses and placentas. Commercial breeding kennels should be regularly checked for causes of abortion and international trade rules should foresee testing for B.  canis of imported breeding dogs. Finally, more researches are required in order to enhance the performances of the diagnostic tests in terms of sensitivity and specificity, as well as more survey data are needed to determine the current spread of B. canis in Italy, in the light to identify the related epidemiological patterns and burden of the disease on Italian canine populations, as well as on public health. in enrichment broth and extracting DNA from the one-week incubation culture. The genomic analysis of strains isolated from the outbreak revealed that they all belong to the same cluster (Figure 1). The maximum distance observed between strains isolated in the outbreak was only of four allelic differences, thus confirming the hypothesis that the outbreak under study has been generated by introduction of a single of B.  canis strain in the kennel. The number of allelic differences was within the published cut-off of genomic cluster found out in a B. melitensis outbreak (Janowicz et al. 2018). Moreover, these limits were also reasonable if compared with several bovine brucellosis outbreaks. These observations strengthen the idea that a single B.  canis introduction led to unrestrained spread within the kennel, affecting many dogs and puppies. Importantly, by surveilling B.  canis sequences from the worldwide database, it was not possible to trace back this strain; however, the availability of the genomes generated by the present study will allow the trace forward and the surveillance of the infection spread from this outbreak. This study highlights the need of further data for an updated genomic surveillance with the aim to avoid public health consequences related to a poor mitigation strategy in fighting the spread of this disease at national and international level. The scarce information about B.  canis contained in the international Brucella genomic databases has made not possible to give further details on the genomic differences of the strains isolated, which would have been useful to trace back the infection source. Indeed, to better help epidemiological investigations in B.  canis outbreaks, it would be needed to genotype as much B.  canis isolates as possible worldwide, in order to improve the quantity and quality of information stored in the international genomic databases. 224 Veterinaria Italiana 2021, 57 (3), 215-226. doi: 10.12834/VetIt.2497.15848.1 First isolation of Brucella canis in Italy De Massis et al. Alton G.G., Jones L.M., Angus R.D. & Verger J.M. 1988. Techniques for the brucellosis laboratory. I.N.R.A., Paris, 169-174. Auwera G.A., Carneiro M.O., Hartl C., Poplin R., Levy-Moonshine A., Jordan T., Shakir K., Roazen D., Thibault J., Banks E., Garimella K.V., Altshuler D., Gabriel S. & DePristo M.A. 2013. From FastQ data to high-confidence variant calls: the genome analysis toolkit best practices pipeline. Curr Protoc Bioinformatics, 43, 11.10.1-11.10.33. Bankevich A., Nurk S., Antipov D., Gurevich A.A., Dvorkin M., Kulikov A.S., Lesin V.M., Nikolenko S.I., Pham S., Prjibelski A.D., Pyshkin A.V., Sirotkin A.V., Vyahhi N., Tesler G., Alekseyev M.A. & Pevzner P.A. 2012. SPAdes: a new genome assembly algorithm and its applications to single-cell sequencing. J Comput Biol, 19, 455-477. https://doi.org/10.1089/cmb.2012.0021. Blankenship R.M. & Sanford J.P. 1975. Brucella canis: a cause of undulant fever. Am J Med, 59, 424-426. Brower A., Okwumabua O., Massengill C., Muenks Q., Vanderloo P., Duster M., Homb K. & Kurth K. 2007. Investigation of the spread of Brucella canis via the U.S. interstate dog trade. Int J Infect Dis, 11, 454-458. doi: 10.1016/j.ijid.2006.12.009. Cadmus S.I.B., Adesokan H.K., Ajala O.O., Odetokun W.O., Perrett L.L. & Stack J.A. 2011. Seroprevalence of Brucella abortus and B. canis in household dogs in southwestern Nigeria: a preliminary report. J S Afr Vet Assoc, 82, 56-57. Carmichael L.E., Zoha S.J. & Flores-Castro R. 1984. Problems in the serodiagnosis of canine brucellosis: dog responses to cell wall and internal antigens of Brucella canis. Dev Biol Stand, 56, 371-383. Carmichael L.E. & Greene G.E. 2013. Canine brucellosis. In Infectious diseases of the dog and cat (Greene G.E., ed.). 4 ed. Saunders, Philadelphia, 398-410. Carmichael L.E. & Shin S.J. 1996. Canine brucellosis: a diagnostician’s dilemma. Semin Vet Med Surg (Small Anim), 11, 161-165. Carmichael L.E. 1966. Abortion in 200 beagles. J Am Vet Med Assoc, 149, 1126. Carmichael L.E. 1990. Brucella canis. In Animal Brucellosis (Nielsen K. & Duncan J.R., eds). CRC Press, Boca Raton, Florida, 335-350. Carmichael L.E., Flores-Castro R. & Zoha S. 1980. Brucellosis caused by Brucella canis: an update of infection in animals and in humans. Geneva, World Health Organization. Document WHO/BRUC./80.361 WHO/ ZOON./80.135. Carriço J.A., Silva-Costa C., Melo-Cristino J., Pinto F.R., de  Lencastre H., Almeida J.S. & Ramirez M. 2006. Illustration of a common framework for relating multiple typing methods by application to macrolide-resistant Streptococcus pyogenes. J Clin Microbiol, 44, 2524-2532. Center for Food Security and Public Health (CFSPH). 2020. The canine brucellosis. Brucella canis. Ames, References Iowa. http://www.cfsph.iastate.edu/Factsheets/pdfs/ brucellosis_canis.pdf. Chinyoka S., Dhliwayo S., Marabini L., Dutlow K., Matope G. & Pfukenyi D.M. 2014. Serological survey of Brucella canis in dogs in urban Harare and selected rural communities in Zimbabwe. J S Afr Vet Assoc, 85, 1087. Corrente M., Franchini D., Decaro N., Greco G., D’Abramo M., Greco M.F., Latronico F., Crovace A. & Martella V. 2010. Detection of Brucella canis in a dog in Italy. New Microbiol, 33, 337-341. Cosford K.L. 2018. Brucella canis: an update on research and clinical management. Can Vet J, 59, 74-81. Daly R., Willis K.C., Wood J., Brown K., Brown D., Beguin-Strong T., Smith R. & Ruesch H. 2020. Seroprevalence of Brucella canis in dogs rescued from South Dakota Indian reservations, 2015-2019. Prev Vet Med, 184, 105157. Di D., Cui B., Wang H., Zhao H., Piao D., Tian L., Tian G., Kang J., Mao X., Zhang X., Du P., Zhu L., Zhao Z., Mao L., Yao W., Guan P., Fan W. & Jiang X. 2014. Genetic polymorphism characteristics of Brucella canis isolated in China. PLoS ONE, 9, 1-7. Dreer M.K.P., Dib Gonçalves D.D., Caetano I.C., Gerônimo E., Menegas P.H., Bergo D., Lopes-Mori F.M.R., Benitez A., de Freitas J.C., Evers F., Navarro T.I. & de Almeida Martins L. 2013. Toxoplasmosis, leptospirosis, and brucellosis in stray dogs housed at the shelter in Umuarama municipality, Paraná, Brazil. J Venom Anim Toxins Incl Trop Dis, 19, 23. Egloff S., Schneeberger M., Gobeli Brawand S., Krudewig C., Schmitt S., Reichler I. & Peterhans S. 2018. Brucella canis infection in a young dog with epididymitis and orchitis. Schweiz Arch Tierheilkd, 160 (12), 743-748. European Union (EU). 2003. Directive 2003/99/EC of the European Parliament and of the Council of 17 November 2003 on the monitoring of zoonoses and zoonotic agents, amending Council Decision 90/424/ EEC and repealing Council Directive 92/117/EEC. OJ, L 325, 12/12/2003. Flores-Castro R., Suarez F., Ramirez-Pfeiffer C. & Carmichael L.E. 1977. Canine brucellosis: bacteriological and serological investigation of naturally infected dogs in Mexico City. J Clin Microbiol, 6, 591-597. Gardner D. & Reichel M. 1997. No evidence of Brucella canis infection in New Zealand dogs. Surveillance, 24, 17-18. Garofolo G., Ancora M. & Di Giannatale E. 2013. MLVA-16 loci panel on Brucella spp. using multiplex PCR and multicolor capillary electrophoresis. J Microbiol Methods, 92, 103-107. George L.W., Duncan J.R. & Carmichael L.E. 1979. Semen examination in dogs with canine brucellosis. Am J Vet Res, 40 (11), 1589-1595. Georgia Department of Agriculture (GDA). 2020. Canine Brucellosis (Brucella canis). Atlanta, Georgia. http://agr. georgia.gov/Data/Sites/1/media/ag_animalindustry/ animal_health/files/caninebrucellosis.pdf. 225Veterinaria Italiana 2021, 57 (3), 215-226. doi: 10.12834/VetIt.2497.15848.1 De Massis et al. First isolation of Brucella canis in Italy among adults with occupational exposure to dogs. Zoonoses Public Health, 61, 509-518. Landis J.R. & Koch G.G. 1977. The measurement of observer agreement for categorical data. Biometrics, 33, 159-174. Le Flèche P., Jacques I., Grayon M., Al Dahouk S., Bouchon P., Denoeud F., Nockler K., Neubauer H., Guilloteau L.A. & Vergnaud G. 2006. Evaluation and selection of tandem repeat loci for a Brucella MLVA typing assay. BMC Microbiol, 6, 9. Letunic I. & Bork P. 2019. Interactive tree of life (iTOL) v4: recent updates and new developments. Nucleic Acids Res, 47(W1), W256-W259. Li H. 2013. Aligning sequence reads, clone sequences, and assembly contigs with BWA-MEM. arXiv, 13033997v1. https://arxiv.org/abs/1303.3997. Marrodan T., Nenova-Poliakova R., Rubio M., Ariza J., Clavijo E., Smits H.L. & Diaz R. 2001. Evaluation of three methods to measure anti-Brucella IgM antibodies and interference of IgA in the interpretation of mercaptan-based tests. J Med Microbiol, 50 (8), 663-666. Mellmann A., Harmsen D., Cummings C.A., Zentz E.B., Leopold S.R., Rico A., Prior K., Szczepanowski R., Ji Y., Zhang W., McLaughlin S.F., Henkhaus J.K., Leopold B., Bielaszewska M., Prager R., Brzoska P.M., Moore R.L., Guenther S., Rothberg J.M. & Karch H. 2011. Prospective genomic characterization of the German enterohemorrhagic Escherichia coli O104:H4 outbreak by rapid next generation sequencing technology. PLoS One, 6, e22751. Michaux-Charachon S., Jumas-Bilak E., Allardetservent A., Bourg G., Boschirolim L. & O’Callaghan D. 2002. The Brucella genome at the beginning of the post-genomic era. Vet Microbiol, 90, 581-585. Mol J.P.S., Guedes A.C.B., Eckstein C., Quintal A.P.N., Souza T.D., Mathias L.A., Haddad J.P.A., Paixao T.A. & Santos R.L. 2020. Diagnosis of canine brucellosis: comparison of various serologic tests and PCR. J Vet Diagn Invest, 32, 77-86. Mor S.M., Wiethoelter A.K., Lee A., Moloney B., James D.R. & Malik R. 2016. Emergence of Brucella suis in dogs in New South Wales, Australia: clinical findings and implications for zoonotic transmission. BMC Vet Res, 12, 199. Nascimento M., Sousa A., Ramirez M., Francisco A.P., Carrico J.A. & Vaz C. 2017. PHYLOViZ 2.0: providing scalable data integration and visualization for multiple phylogenetic inference methods. Bioinformatics, 33, 128-129. National Association of State Public Health Veterinarians (NASPHV). 2012. Public health implications of Brucella canis infections in humans. http://nasphv.org/ Documents/BrucellaCanisInHumans.pdf. Nicoletti P. 1980. The epidemiology of bovine brucellosis. Adv Vet Sci Comp Med, 24, 69-98. Nöckler K., Kutzer P., Reif S., Rosenberger N., Draeger A., Bahn P., Göllner C. & Erlbeck C. 2003. Canine brucellosis - a case report. Berl Munch Tierarztl Wochenschr, 116, 368-372. Nurk S., Bankevich A., Antipov D., Gurevich A.A., Korobeynikov A., Lapidus A., Prjibelski A.D., Pyshkin A., Sirotkin A., Sirotkin Y., Stepanauskas R., Clingenpeel S.R., Gyuranecz M., Szeredi L., Rónai Z., Dénes B., Dencso L., Dán Á., Pálmai N., Hauser Z., Lami E., Makrai L., Erdélyi K. & Jánosi S. 2011. Detection of Brucella canis-induced reproductive diseases in a Kennel. J Vet Diagnostic Investig, 23, 143-147. Hayashi T.T.A & Ysayama Y. 1977. Detection of Brucella canis infection in dogs in Hokkaido. Microbiol Immunol, 21 (5), 295-298. Hensel M.E., Negron M. & Arenas-Gamboa A.M. 2018. Brucellosis in dogs and public health risk. Emerg Infect Dis, 24, 1401-1406. Hofer E., Bagó Z., Revilla-Fernández S., Melzer F., Tomaso H., López-goñi I., Fasching G. & Schmoll F. 2012. First detection of Brucella canis infections in a breeding kennel in Austria. New Microbiol, 35, 507-510. Holst B.S., Löfqvist K., Ernholm L., Eld K., Cedersmyg M. & Hallgren G. 2012. The first case of Brucella canis in Sweden: background, case report and recommendations from a northern European perspective. Acta Vet Scand, 54, 18. Janowicz A., De Massis F., Ancora M., Cammà C., Patavino C., Battisti A., Prior K., Harmsen D., Scholz H., Zilli K., Sacchini L., Di Giannatale E. & Garofolo G. 2018. Core genome multilocus sequence typing and single nucleotide polymorphism analysis in the epidemiology of Brucella melitensis infections. J Clin Microbiol, 56 (9), e00517-18. Junemann S., Sedlazeck F.J., Prior K., Albersmeier A., John U., Kalinowski J., Mellmann A., Goesmann A., von Haeseler A., Stoye J. & Harmsen D. 2013. Updating benchtop sequencing performance comparison. Nat Biotechnol, 31, 294-296. Kaden R., Ågren J., Båverud V., Hallgren G., Ferrari S., Börjesson J., Lindberg M., Bäckman S. & Wahab T. 2014. Brucellosis outbreak in a Swedish kennel in 2013: determination of genetic markers for source tracing. Vet Microbiol, 174, 523-530. Kauffman L.K., Bjork J.K., Gallup J.M., Boggiatto P.M., Bellaire B.H. & Petersen C.A. 2014. Early detection of Brucella canis via quantitative polymerase chain reaction analysis. Zoonoses Public Health, 61, 48-54. Kaufmann L.K. & Petersen C.A. 2019. Canine brucellosis: old foe and reemerging scourge. Vet Clin North Am Small Anim Pract, 49 (4), 763-779. Keid L.B., Soares R.M., Vasconcellos S.A., Salgado V.R., Megid J. & Richtzenhain L.J. 2010. Comparison of a PCR assay in whole blood and serum specimens for canine brucellosis diagnosis. Vet Rec, 167, 96-99. Keid L.B., Chiebao D.P., Batinga M.C.A., Faita T., Diniz J.A., Oliveira T.M.F. de S., Ferreira H.L. & Soares R.M. 2017. Brucella canis infection in dogs from commercial breeding kennels in Brazil. Transbound Emerg Dis, 64, 691-697. Keid L.B., Soares R.M., Vasconcellos S.A., Megid J., Salgado V.R. & Richtzenhain L.J. 2009. Comparison of agar gel immunodiffusion test, rapid slide agglutination test, microbiological culture and PCR for the diagnosis of canine brucellosis. Res Vet Sci, 86, 22-26. Krueger W.S., Lucero N.E., Brower A., Heil G.L. & Gray G.C. 2014. Evidence for unapparent Brucella canis infections 226 Veterinaria Italiana 2021, 57 (3), 215-226. doi: 10.12834/VetIt.2497.15848.1 First isolation of Brucella canis in Italy De Massis et al. United States Department of Agriculture Animal and Plant Health Inspection Service (USDA). 2015. Best practices for Brucella canis prevention and control in dog breeding facilities. U.S. Department of Agriculture. h t t p s : / / w w w. a p h i s . u s d a . g o v / a n i m a l _ w e l f a r e / downloads/brucella_canis_prevention.pdf. Vicente A.F., Girault G., Corde Y., Mioni M.S.R., Keid L.B., Jay M., Megid J. & Mick V. 2018. New insights into phylogeography of worldwide Brucella canis isolates by comparative genomics-based approaches: focus on Brazil. BMC Genomics, 19 (1), 636. Von Kruedener R.B. 1976. Outbreak of a Brucella canis infection in a beagle colony in West Germany. Dev Biol Stand, 31, 251-253. Wanke M.M. 2004. Canine brucellosis. Anim Reprod Sci, 82‑83, 195-207. Whatmore A., Perret L. & Friggens M. 2017. Second UK isolation of Brucella canis. Vet Rec, 180 (25), 617. Whatmore A.M., Perrett L.L. & MacMillan A.P. 2007. Characterisation of the genetic diversity of Brucella by multilocus sequencing. BMC Microbiol, 7, 34. World Organization for Animal Health (OIE). 2019. Brucellosis (Brucella abortus, B. melitensis and B. suis) (infection with B. abortus, B. melitensis and B. suis) In Manual of Diagnostic Tests and Vaccines for Terrestrial Animals. Paris, Office International des Epizooties, 1-44. Yoak A.J., Reece J.F., Gehrt S.D. & Hamilton I.M. 2014. Disease control through fertility control: secondary benefits of animal birth control in Indian street dogs. Prev Vet Med, 113, 152-156. Woyke T., McLean J.S., Lasken R., Tesler G., Alekseyev M.A. & Pevzner P.A. 2013. Assembling single-cell genomes and minimetagenomes from chimeric MDA products. J Comput Biol, 20, 714-737. Rovid Spickler A. 2018. Brucellosis: Brucella canis. United States Department of Agriculture Animal and Plant Health Inspection Service, 1-10. http://www.cfsph. iastate.edu/Factsheets/pdfs/brucellosis_canis.pdf. Sahl J.W., Beckstrom-Sternberg S.M., Babic-Sternberg J., Gillece J.D., Hepp C.M., Auerbach R.K., Tembe W., Wagner D.M., Keim P.S. & Pearson T. 2015. The in  silico genotyper (ISG): an open-source pipeline to rapidly identify and annotate nucleotide variants for comparative genomics applications. bioRxiv, 1-7. https://doi.org/10.1101/015578. Severiano A., Pinto F.R., Ramirez M. & Carriço J.A. 2011. Adjusted Wallace coefficient as a measure of congruence between typing methods. J Clin Microbiol, 49, 3997-4000. Shin S. & Carmichael L.E. 1999. Canine brucellosis caused by Brucella canis. In Recent advances in canine infectious disease (Carmichael L.E., ed). New York International Veterinary Information Service, Ithaca NY, 1-4. Sidstedt M., Hedman J., Romsos E.L., Waitara L., Wadsö L., Steffen C.R., Vallone P.M. & Rådström P. 2018. Inhibition mechanisms of hemoglobin, immunoglobulin G, and whole blood in digital and real-time PCR. Anal Bioanal Chem, 410, 2569-2583. Sivia D.S. 1996. Data Analysis: a bayesian Tutorial. Oxford University Press, Oxford, UK. Taylor D.J. 1980. Serological evidence for the presence of Brucella canis infection in dogs in Britain. Vet Rec, 106, 102-104.