Horses are not considered as natural host for brucellosis but they can be infected as an by close contact with other animals, including cow, sheep and goats in some livestock farms (Hamdy and Amin, 2002; Tahamtan et al., 2015). Therefore, horses could be the carrier of infection without any clinical symptoms (Badiei et al., 2013). The brucellosis could be transmitted to human through direct contact with infected animals and their secretions, consumption of contaminated milk and dairy products or through inhalation of contaminated airborne particles (Tahamtan et al., 2010; Tuon et al., 2017). The mammary gland and accessory lymph nodes are considered as a target organ of Brucella localization, and viable bacterial cells may be excreted into milk (Shafei et al., 2012). In addition, during last year’s by increasing the tendency to consume raw horse milk in neonates and adults and contact of individuals with the animal and the possibility of transmitting disease, the present study was conducted with the aim of evaluating the prevalence of brucellosis in mare's milk and blood in Yazd city, Iran. In current study, 164 blood and milk samples screened by Rose Bengal and milk ring tests which were negative in terms of brucellosis. It was shown that 3 samples out of 82 milk samples were positive in PCR assay, while none of the blood samples was positive by PCR. The culture media of 3 positive samples in PCR showed no bacterial growth.
In a study conducted on 74 horses in Turkey in 2007, the sero-prevalence of anti-Brucella antibodies in horses was reported 9.5% using the Serum Tube Agglutination test (SAT) (Göz et al., 2007). Ghobadi et al (2013) investigated the prevalence of brucellosis antibodies in serum of 200 horses of Hamedan city by tube agglutination test and 2-mercaptoethanol test. The general prevalence in this study was reported 0.5 percent (only one sample was positive)(Ghobadi and Salehi, 2013). The prevalence of brucellosis in horses by Tel et al was reported 13.68% using the RBT method (Tel et al., 2011). In the another study prevalence of brucellosis in horses was reported 2.5% by RBT method (Tahamtan et al., 2010). Serologic tests are easy and fast, but false positive and negative results might be reported by these methods (Ilhan et al., 2008). In some studies, the prevalence of brucellosis was different from that of the present study, which might be due to the contamination of the horse after contact with brucellosis-infected animals and the transient and temporal presence of anti-Brucella antibodies in the horse, it has not had enough time to produce an antibody that can be at an acceptable level for the serological test (Rafiee, 2014).
Leary et al (2006) determined the Brucella abortus by PCR method in blood, milk, and lymphoid tissues of 21 positive cows. According to the results, the Brucella in the cow blood samples was not found. However, in the milk samples, 44% of positive cases were positive in culture method. Moreover, 75% of the extra-mammary lymph nodes and 66% of the retropharyngeal lymph nodes that were positive in culture were also reported to be positive by PCR. In present study, PCR assay did not show any superiority in comparison to conventional diagnostic methods in diagnosing the Brucella in blood samples. O'Learyet al suggested that one of the most important reasons for the inefficiency of PCR in their study was that the whole blood is not a good template for detection of B. abortus DNA (O’Leary et al., 2006).The results of the above-mentioned study are consistent with those of current study. Leal-Klevezas (2000) evaluated the efficiency of PCR method on blood and milk goat samples in comparison with milk culture, blood culture and Rose Bengal method. They showed that the sensitivity of the PCR method to culture and Rose Bengal test was higher and it could be used to diagnose the Brucella as a fast test (Leal-Klevezas et al., 2000). So far, there is not any research on the use of PCR in body fluids such as blood and milk of horse, simultaneously. In the current study, the PCR was used to evaluate the prevalence of brucellosis in milk and blood of the horse. The PCR method identified only 3 brucellosis positive in milk samples, although their blood samples were not positives. It should be noted that several factors can effect on the results of PCR and bacterial culture in clinical samples of the livestock, including the infection stage, the place of bacterium localized (depends on the stage of the infection) and the bacterial load present in the clinical sample, which may be one of the reasons for the differences between the results of the PCR assay in diagnosis of Brucella in serum and blood samples(Ilhan et al., 2008).Methods of DNA extraction are also effective in PCR ability to identification of bacteria. Brucella may be present in the collected samples at very small amounts, so extraction of DNA in animal samples should be maximized to increase the possibility of isolation and identification(O’Leary et al., 2006).
In the present study, whole blood was used for DNA extraction. The DNA of bacterium might be not detected in blood by inhibitory agents; therefore use of serum instead of whole blood could cover the problem. However, it should be noted that the negative false results could be attributed to specific characteristic of Brucella which is intracellular, and DNA extraction at chronic phase of the disease which is found at small amounts(Ilhan et al., 2008). Different studies have shown various results by using different methods such as PCR, serologic and bacteriological tests(O’Leary et al., 2006). In addition, PCR is preferred because of high speed, proper repeatability, non-interference with serum cross reaction agents bacteria, high sensitivity to low bacterial levels, no need to living bacteria and low risk for laboratory personnel, performing the test with a minimum sample and not affecting the presence of other opportunistic objects in the clinical sample, although observing the appropriate conditions to prevent infection of reaction with bacterium DNA or products propagated in previous reactions in PCR is necessary(Ilhan et al., 2008). In present research, although 3 samples of milk were positive in the PCR test, no sample was diagnosed positive in the MRT serological test. This may be due to the high sensitivity of PCR to serological and microbial methods (Leal-Klevezas et al., 2000). The stage of infection (acute or chronic phases) is also involved in the bacterial diagnosis (Ilhan et al., 2008). Finally, positive milk samples in PCR were cultured and the culture results were negative. The negative result of bacterial culture is attributed to difficulty of isolating the Brucella bacteria and low sensitivity of bacterial recovery from milk(Leal-Klevezas et al., 2000). It should be noted that the bacterial excretion in milk is not continuous in livestock and may not be excreted at the time of sampling (Leal-Klevezas et al., 2000).
In the present study, although the number of horses on the range of 1 to 5 years was higher, no significant difference was found between the age groups of horses and the prevalence of brucellosis. Badiei et al. (2012) showed that, there was no significant difference among various age groups in terms of presence of anti-Brucella antibody. However, in the study conducted by Safirollah et al. (2012) on the prevalence of anti-Brucella antibodies in horses, donkeys and mules in Pakistan, the prevalence of antibody was significantly higher in older animals (5–11) than younger ones. In the present study, clinical symptoms of brucellosis were not observed in any of horses. Tajik et al reported no clinical symptom on the sero-prevalence of anti-Brucella antibodies positive in Arabian horses(Rafiee, 2014). In the present study, no significant difference was found between the prevalence of brucellosis and coexistence with other animals (cow, sheep and goats). This result is consistent with the result of the study conducted by Badiei et al. (Badiei et al., 2013). However, Tahmtan et al who studied (2010) the prevalence of brucellosis in horses of Mashhad province showed the highest prevalence in horses due to the coexistence with other animals including cows, sheep and goats(Tahamtan et al., 2010). Unfortunately, horses may be infected with Brucellosis such as other species without any clinical manifestation which the reason is not clear. MacMillan et al infected horse with Brucella abortus to investigate the clinical changes. They showed that infected horses had not any sign of disease except for changes in rectal temperature. In biochemical observation the total serum amino acids, IgM and to lesser extent IgG were changed. They reported that the tendency of B. abortus toward localization in the lymph nodes was compatible with other species (MacMillan et al., 1982).