Investigation of The Presence of Brucella canis in Dogs in Western Part of Turkey

Brucella sp. is a gram-negative, facultative and intracellular bacterial species. Canine brucellosis has been widespread all over the world since the first isolation of the Brucella species, but in recent years, some countries have been reported as free from the disease (Wee et al., 2008). Generally, there are 10 different species in the Brucellaceae family that can cause brucellosis in animals: B. abortus, B. melitensis, B. canis, B. suis, B. ovis, B. neotomae, B. ceti, B. pinnipedalis, B. microti and B. inopinata. Among these, B. abortus, B. melitensis, B. canis and B. suis are zoonotic (OIE, 2018).
       
Brucella canis is the etiological agent of canine brucellosis in dogs and a very dangerous bacterium that causes infant death, infertility and mastitis in females and orchitis, epididymitis and testicular atrophy in males. Canine brucellosis is one of the most important causes of abortion and infertility in dogs worldwide (Carmicheal, 1990). The infection is transmitted to humans via direct contact with the abortion materials, placenta, foetal fluids or vaginal discharge (Alton et al., 1988; Lucero et al., 2005a). Natural infection with B. canis is seen only in dogs. Transmission between dogs occurs primarily via the venereal route (Brower et al., 2007; Kimura et al., 2008). Laboratory accidents and infectious dogs resulting in the transmission of the disease to humans also show the importance of B. canis in terms of public health (Lucero et al., 2005b). Mortality rates have been reported as 2-5% due to the complications of endocarditis and meningitis. Venereal transmission continues for 2 years in dogs that appear clinically healthy; for this reason, preventive measures and early-stage diagnosis are essential (Sayan et al., 2011). Stray dogs are the focus in preventing the spread of the disease. According to a previous study, a high prevalence of B. canis has been reported in stray and free-roaming dogs (Hubbard et al., 2018).
       
Conventional biotyping, serological tests and/or molecular techniques are generally used to diagnose canine brucellosis (Keid et al., 2009). The use of whole blood in the isolation of B. canis is quite effective, but this method has some disadvantages; for example, it requires a long incubation period due to the fastidious growth of the bacterium. Serological tests also have disadvantages, such as low sensitivity, variable specificity and cross reactions (Hollett, 2006). The common serological methods used in the diagnosis of canine brucellosis are the LAT, TAT, 2-ME TAT and 2-ME RSAT tests (Alton et al., 1988). Among these serological tests, the 2-ME RSAT (2-mercaptoethanol rapid slide agglutination test) test is the most sensitive and frequently used method for diagnostic procedures. In a modified version of the 2-ME RSAT test, the M (-) variant strain of B. canis, which is less mucoid, is used as the antigen (Carmicheal, 1990). In recent years, some molecular techniques have been developed using DNA isolation methods to identify B. canis (Kang et al., 2011; Aras et al., 2015). The molecular identification of Brucella species is performed via multiplex PCR; however, this technique may have low sensitivity in distinguishing clinical types because there are close genetic similarities among Brucella species (Romero et al., 1995; Koylass et al., 2010). Kang et al., (2014) performed a specific PCR study of canine blood and abortion-related material for B. canis and achieved a high sensitivity.
       
This study aimed to investigate the presence of canine brucellosis in the western part of the Aegean region of Turkey using 2-ME RSAT and species-specific PCR methods in pet dogs with or without contact with stray dogs.

Samples
 
A total of 200 blood samples (80 from Aydýn Province, 60 from Ýzmir Province and 60 from Muðla Province) were collected between March 2019 and February 2020. The samples were taken from owned dogs found in the city centre of Izmir province, which had no close contact with other stray dogs. The remaining dogs chosen for sample collection were selected from the Aydýn and Muðla provinces and they were owned dogs kept in the shelters of house gardens. The owned pet dogs had close contact with stray dogs residing in the open field.
       
The blood samples were brought to the research laboratory of the Microbiology Department of Aydýn Adnan Menderes University’s Veterinary Faculty in ice boxes. The authors of this study hereby declare that the collection of specimens was carried out in accordance with the guidelines of the National Institute of Health (NIH) in the USA regarding the care and use of animals for experimental procedures and with the European Communities Council Directive of 24 November 1986 (86/609/EEC).
 
Laboratory tests
 
Serological test antigen 2-ME RSAT was prepared as previously described (Carmichael, 1990). Each serum sample was mixed with 25 μl aliquots of 0.2 M 2-ME solution for at least 45 seconds. Then 25 μl of B. canis antigen was added and agglutination was observed after 2 minutes of circular shaking. The formation of agglutination was evaluated as a positive result (Kim et al., 2007).
 
PCR
 
The primers prepared from the BCAN-B0548 region of the B. canis DNA used in our study for B. canis identification were as follows: F: 5'-CCAGATAGACCTCTCTGGA-3, 5'-TGGCCTTTTCTGATCTGTTCTT-3'.
 
Standard serotype
 
The B. canis RM666 ATCC 23365 standard strain was used as a positive control. The standard serotype was provided by the microbiology department of the Faculty of Veterinary Science at the University of Selcuk.
 
DNA extraction
 
Genomic DNA from blood samples and standard strains were isolated with a Genomic DNA Purification Kit^® (Thermo Fisher Scientific) according to the manufacturer’s protocol. The genomic DNA was used as a DNA template for PCR amplification. Genomic DNA samples were stored in a deep freeze (-20°C) prior to use for PCR.
 
B. canis species-specific PCR
 
B. canis-specific PCR procedures were performed according to the protocol reported by Kang et al., (2014). Genomic DNA extracted from each blood sample were amplified in a PCR reaction mixture (25 µl) that contained 10X reaction buffer with KCl, 1.75 mM MgCl₂, 0.2 mM each of dNTPs, 1 U of Taq DNA polymerase, 1 µl of forward primer B. canis (10 pmol), 1 µl of reverse primer B. canis (10 pmol) and 2 µl of template DNA (10 pg-1 µg). 
       
The reaction mixtures were amplified using a thermal cycler (Mastercycler Personal; Eppendorf, Netheler, Hinz GmbH, Hamburg, Germany). The PCR cycle parameters were as follows: initial denaturation at 94°C for 7 min, then 35 cycles of denaturation at 94°C for 35 sec, annealing at 59°C for 40 sec, extension at 72°C for 35 sec and final extension at 72°C for 5 min (Modified from Kang et al., 2014).
 
Agarose gel electrophoresis
 
The PCR products were analysed using 1.5% agarose gel electrophoresis (Agarose-ME, Classic Type; Nacalai Tesque, Inc.) with ethidium bromide. The DNA bands were visualised with a gel documentation system (Infinity VX2, France). The positive DNA samples were screened at 300 bp.

Twelve (6%) of the serum samples were agglutinated in 2 minutes with and were found positive via 2-ME RSAT test. The other 188 samples were found negative via 2-ME RSAT. PCR tests were performed with DNA samples of the same animals used in the serological test. The PCR indicated that 10 (5%) of the 200 samples were positive for B. canis-specific DNA (Fig 1). The serological 2 MR-SAT and PCR test results throughout the provinces are given in Table 1.
 

 

       
Twelve positive samples were detected via 2-ME RSAT and 10 positive samples were detected via molecular test. Two samples (one from Aydýn, another from Ýzmir) were found positive via serological test; however, the PCR test did not verify the positivity. This is considered a consequence of potential cross reactions in serological tests.
       
Although there is considerable information and evidence of Brucella sp. infections in the world, studies to determine the serological and molecular frequency of brucellosis in dogs are extremely limited. The seroprevalence of B. canis has been reported to vary between 1.1% and 60.6% in serological studies on dogs conducted in different countries (Greene and Carmichael, 2006). In a study conducted in Italy, 25 (1.1%) of 2328 sera (Ebani et al., 2003) were found to be positive and 20 (60.6%) of 33 sera were found to be positive in a study conducted in Canada (Brennan et al., 2008). B. canis antibodies were found to be positive in 12 of 485 sera (2.5%) in Japan (Kimura et al., 2008); in 5 of 102 (4.9%) samples in Iran (Mosallanejad et al., 2009); in 12 of 113 (10.6%) blood sera in another study conducted in Iran (Behzadi and Mohheiseh, 2012); in 100 of 2000 (5%) sera in another study conducted in Canada (Bosu and Prescott, 1980); in 16 of 219 (7.3%) sera in Argentina (Boeri et al., 2008); in 33 of 224 (14.7%) sera in another study conducted in Argentina (Lopez et al., 2009); in 72 of 280 (25.7%) sera and varying between 0.8% to 44.5% (Keid et al., 2017) in Brazil (Barrouin-Melo et al., 2007); in 85 of 317 (26.8%) sera in the USA (Brower, 2007) and in 181 of 463 (39.1%) samples in Korea (Kim et al., 2007) and 2.3% in the US State of Mississippi (Hubbard et al., 2018).
       
It has been reported that the seroprevalence in dogs changed between 5.4 and 7.7 in studies conducted in Turkey. In their study on 40 kennels in Van province, Ceylan et al., (2006) did not find B. canis antibodies in the serum samples of dogs. B. canis. antibodies were found positive in 6 of 111 (5.4%) sera by Yýlmaz and Gümüşsoy (2010); in 14 of 222 (6.3%) samples by Diker et al., (1987); in 9 of 134 (6.7%) sera by Ýstanbulluoðlu and Diker (1983) and in 28 of 362 (7.7%) samples by Oncel et al., (2005). In these studies, mercaptoethanol tube agglutination test was applied (Diker et al., 1987; Ýstanbulluoðlu and Diker, 1983; Yýlmaz and Gümüşsoy, 2010). In the present study, 200 canine sera collected from 3 different provinces of Turkey were examined using MAT and a positive rate of 0.8% was determined for B. canis antibody. When the results of the present study were compared with those of previous studies, B. canis antibodies were found to be positive in the similar number of samples.
       
There was no positivity in Ýzmir Province, whereas in Aydýn Province, the positivity rate was 7.5% and it was 6.6% in Muðla Province. In Ýzmir, where there is no positivity, all of the samples were taken from owned dogs from the city centre. These dogs remained in the house and had no contact with stray dogs. However, in Aydýn and Muðla, the samples were collected from coastal areas. In these regions, the owned dogs were usually in contact with stray dogs in the gardens of summer houses or in the streets. The rates of positivity are higher in these 2 cities. In addition, these pet dogs, which are in contact with stray dogs, can act as a reservoir for the transmission of brucellosis to their owners. Stray dogs are more likely to have a higher documented level of B. canis seropositivity then pet dogs (Hensel et al., 2018; Alshehabat et al., 2019).
       
Recent studies have shown that agglutination tests using sera treated with 2-mercaptoethanol, agar gel immunodiffusion tests and immunochromatographic tests are highly specific, but a significant proportion of false-negative results were observed in bacteremic dogs (Keid et al., 2009; 2017). A PCR test can show negative results in cases of chronic infection due to the lack of bacteria circulation and localisation in organs, such as the mesenteric lymph nodes. In such cases, serology may be positive because antibodies were previously produced, as reported by Keid et al., (2009).
       
Limitations of this research are that the results refer to samples from 200 selected dogs and do not reflect regional occurrence. The samples were collected from a heterogeneous group of dogs, so the proportions of positive tested samples must be interpreted critically and cannot measure a real regional prevalence of canine brucellosis. Two PCR negative samples were found serologically positive as indicated above. This is regarded as another limitation since dogs are susceptible to infections with B. abortus, B. suis and B. melitensis. In the present study, the laboratory database only specified the three provinces of Turkey, but otherwise the precise origin of the samples remained unknown. Furthermore, the diagnosis of B. canis infection with antibodies faces several other limitations, such as false positive results can result from cross-reactions with other bacteria that offer the same antigenic determinants. Cross reactions are frequently seen due to the presence of Pseudomonas aeruginosa, mucoid Staphylococcus spp. and Bordetella bronchiseptica-specific antibodies (Barrouin-Melo et al., 2007). PCR is useful to detect B. canis in clinical samples; however, it is preferable to include the 2ME-RSAT test because this improves the accuracy of the diagnosis. Blood culture is the gold standard for the diagnosis of brucellosis (Sánchez-Jiménez et al., 2014). However, serological and molecular tests are very useful in detecting the prevalence of the disease and offering rapid results. It is thought that the reason for the positivity in the serum of 2 animals that are serologically positive but not detected in the PCR test may be due to differing periods of the current infection; in addition, different immunoglobulins or cross reactions may be present in the serum.

In recent years, many people have owned dogs acquired from shelters or from the streets. Rapid tests are preferred to evaluate the presence of brucellosis in dogs, particularly when dogs are adopted from a shelter or from the streets. This study provides insights into the regional occurrence of B. canis in dogs in Aydin, Izmir and Manisa, which are western provinces of Turkey. B. canis DNA was present in 5% of all submitted samples and B. canis antibodies in 6% of samples collected for diagnostic testing for canine Brucella infection from 200 specimen. Yet, the true distribution of B. canis infection in dogs in Turkey remains unknown. Data presented here reveal only occurrence in samples in which testing for B. canis was randomly performed by the authors of this research, thus, are not representative of the exact prevalence in the dog population of the respective provinces.  In conclusion, infection with B. canis needs to be estimated in dogs in Turkey. PCR for diagnosis of Brucella infection may be included in the routine diagnostic procedures of dogs. In summary, reasons for the occurrence of B. canis in Turkey could be poor health in selected dogs and presence of stray dog populations in selected provinces. Keeping dogs as pets in very poor housing conditions without veterinary care may constitute additional risks. Therefore, effective follow-up concerning brucellosis in dogs is required. For this purpose, the routine use of rapid diagnostic test kits should be expanded in shelters and veterinary clinics. Dogs adopted from shelters or the streets are frequently evaluated for parasitic agents, whereas zoonotic bacterial diseases are often overlooked. However, these animals should also be evaluated for zoonotic bacterial diseases.

This study was funded by Aydýn Adnan Menderes University BAP by VTF-16015 project code.

The authors declare that they have no conflict of interest.