A new Brucella canis species-specific PCR assay for the diagnosis of canine brucellosis

Comparative Immunology, Microbiology and Infectious Diseases 37 (2014) 237–241

Contents lists available at ScienceDirect

Comparative Immunology, Microbiology and Infectious Diseases journal homepage: www.elsevier.com/locate/cimid

A new Brucella canis species-specific PCR assay for the diagnosis of canine brucellosis Sung-Il Kang a , Sang-Eun Lee b , Ji-Yeon Kim a , Kichan Lee a , Jong-Wan Kim a , Hyang-Keun Lee a , So-Ra Sung a , Young-Ran Heo c , Suk Chan Jung a , Moon Her a,∗ a The OIE Reference Laboratory for Brucellosis, Bacterial Disease Division, Animal and Plant Quarantine Agency, 480 Anyang, Gyeonggi-do 430-757, South Korea b Division of Malaria & Parasitic Diseases, Korea National Institute of Health, Korea Centers for Disease Control & Prevention, Chengwon-gun, South Korea c Department of Food and Nutrition, Chonnam National University, Yongbongdong, Gwanju, South Korea

a r t i c l e

i n f o

Article history: Received 25 February 2014 Received in revised form 30 June 2014 Accepted 13 July 2014 Keywords: Brucella canis Buffy coat Diagnostics Species-specific PCR assay Zoonosis

a b s t r a c t Brucellosis is a zoonotic disease that is transmitted from animals to humans, and the development of a rapid, accurate, and widely available identification method is essential for diagnosing this disease. In this study, we developed a new Brucella canis species-specific (BcSS) PCR assay and evaluated its specificity and sensitivity. A specific PCR primer set was designed based on the BCAN B0548-0549 region in chromosome II of B. canis. The PCR detection for B. canis included amplification of a 300-bp product that is, not found on other Brucella species or, genetically or serologically related bacteria. The detection limit of BcSSPCR assay was 6 pg/␮l by DNA dilution, or 3 × 103 colony-forming units (CFU) in the buffy coats separated from whole blood experimentally inoculated with B. canis. Using the buffy coat in this PCR assay resulted in approximately 100-times higher sensitivity for B. canis as compared to detect directly from whole blood. This is the first report of a species-specific PCR assay to detect B. canis, and the new assay will provide a valuable tool for the diagnosis of B. canis infection. © 2014 Elsevier Ltd. All rights reserved.

1. Introduction The genus Brucella is gram-negative, facultative intracellular pathogens. Since the first isolation of a Brucella strain, this genus has been found to be prevalent in South America, Africa, Southeast Asia, and some European countries. This disease has also occurred sporadically in South Korea [1–3]. Currently, the genus Brucella consists of 10 species preference of animal hosts: B. abortus, B. canis, B. suis, B. ovis, B. neotomae, B. melitensis, B. ceti, B. pinnipedialis,

∗ Corresponding author. Tel.: +82 31 467 1776; fax: +82 31 467 1778. E-mail address: [email protected] (M. Her). http://dx.doi.org/10.1016/j.cimid.2014.07.003 0147-9571/© 2014 Elsevier Ltd. All rights reserved.

B. microti, and B. inopinata. Of these, B. melitensis, B. abortus, B. suis, and B. canis are transmitted to humans [4]. B. ceti and B. pinnipedialis originated from marine mammals are also known to have zoonotic potential [5]. With the rapid growth of the pet and companion animal industries, Brucella infection could become a serious public health issue. Currently, canine brucellosis resulting from B. canis infection is thought to be underestimated worldwide [6]. In dogs, B. canis can cause contagious abortion, testicular atrophy, infertility and lymphadenitis. The infectious route is mainly through direct contact with the foetus, placenta, foetal fluids, and/or vaginal discharge after abortions [7]. Humans generally become infected through close

238

S.-I. Kang et al. / Comparative Immunology, Microbiology and Infectious Diseases 37 (2014) 237–241

contact with infected dogs or abortion-related materials [8]. Generally, diagnosis of canine brucellosis is based on classical biotyping methods, serological tests or molecular techniques [4,9–11]. Whole blood is a useful sample for isolation of B. canis because of the prolonged bacteraemia, which characterizes the disease [6,12]. However, bacterial isolation has its disadvantages, because B. canis is a slow-growing organism that requires a long incubation period and blind subcultures. This can also pose risks to researchers working with the organism [6,13]. The serological tests routinely used for diagnosis of canine brucellosis have the disadvantages of low sensitivity and varying specificity [11]. Owing to these shortcomings, a variety of molecular techniques using Brucella DNA have been developed to identify B. canis, such as a real-time PCR using single nucleotide polymorphisms (SNPs) and a multiplex differential PCR [14,15]. These molecular techniques have their own shortcomings, including high expenditure and expensive equipment. Currently using multiplex PCR it is only possible to distinguish Brucella species by using a high quality and concentration genomic DNA from the bacteria. Therefore, this technique limits the choice of clinical specimens because of low sensitivity. On the hand, a high degree of genetic similarity with other Brucella species, especially B. suis, has interrupted the development of a simple PCR assay to detect or identify B. canis until now [16,17]. Therefore, the aim of the current study was to develop a specific PCR assay and evaluate its potential for detecting B. canis from blood and abortion-related materials of dogs. 2. Materials and methods 2.1. Bacterial strains

Table 1 Bacteria strains used in this study and comparison of the results of two PCR assays. Species

Strains

PCR results 16S rRNAa

B. canis PCRb

Brucella species B. abortus bv. 1 544 B. abortus bv. 2 86/8/59 B. abortus bv. 3 Tulya B. abortus bv. 4 292 B. abortus bv. 5 B3196 B. abortus bv. 6 870 B. abortus bv. 9 C68 B. canis RM6/66 B. suis bv. 1 1330 B. suis bv. 2 Thomsen B. suis bv. 3 686 B. suis bv. 4 40 B. suis bv. 5 513 B. ovis 63/290 B. neotomae 5K33 B. melitensis bv. 1 16M B. melitensis bv. 2 63/9 B. melitensis bv. 3 Ether B. ceti B1/94 B. pinnipedialis B2/94 B. microti CCM4915 B. inopinata B01 B. canis

ATCC 23448 ATCC 23449 ATCC 23450 ATCC 23451 ATCC 23452 ATCC 23453 ATCC 23455 ATCC 23365 ATCC 23444 ATCC 23445 ATCC 23446 ATCC 23447 NCTC 11996 ATCC 25840 ATCC 23459 ATCC 23456 ATCC 23457 ATCC 23458 NCTC 12891 NCTC 12890 BCCN 07-01 BCCN 09-01 94 isolates

+ + + + + + + + + + + + + + + + + + + + + + +

− − − − − − − + − − − − − − − − − − − − − − +

Non-Brucella organisms Ochrobactrum anthropic Escherichia coli O157:H7c Pasteurella multocida Salmonella Typhimurium Campylobacter jejuni Yersinia enterocolitica O: 9 Staphylococcus aureusc Clostridium perfringens type A

Field strain Field strain ATCC 43017 ATCC 14028 ATCC 33560 NCTC 11174 Field strain ATCC 13124

+ − − − − − + −

− − − − − − − −

a

The strains used in this study included 22 Brucella reference strains; 94 B. canis field isolates; and 8 non-Brucella strains, including serologically cross-reacting strains or phylogenetically related strains (Table 1). All of the B. canis field isolates were obtained from dog-breeding farms during 2002–2011 and were identified by classical biotyping assays and molecular methods [4,14]. Other non-Brucella strains were identified by 16S rRNA sequence analysis and VITEK 2 Compact (Biomerieux, Seoul, South Korea). The Brucella strains were cultured on tryptic soy agar (TSA) (BD, Franklin Lakes, NJ) supplemented with 5% foetal bovine serum (FBS) (GIBCO, Grand Island, NY, USA) for 3 days at 37 ◦ C under 5% CO2 or aerobic conditions. Other non-Brucella organisms were grown on 5% sheep blood agar or MacConkey agar for 18–24 h at 37 ◦ C.

Brucella genus specific PCR assay that targets the 16S rRNA [17]. B. canis species-specific PCR assay developed in this study. c The field strains were identified as analysis of 16S ribosomal RNA gene sequence. b

were performed using a 20-␮l reaction mixture containing 2 ␮l of template DNA and 2 ␮l of each of the primers (10 pmol). The PCR conditions consisted of an initial denaturation at 94 ◦ C for 7 min; 35 cycles of denaturation at 94 ◦ C for 35 s, annealing at 59 ◦ C for 40 s, and extension at 72 ◦ C for 35 s, followed by a final extension at 72 ◦ C for 10 min. All amplicons were analyzed by electrophoresis using 1.5% agarose gel with a 100-bp ladder (Bioneer Co., Taejon, South Korea) as a molecular size marker and ethidium bromide (0.02 ml/ml) for staining. Stained gels were visualized and photographed under UV light with a UV transilluminator (Bio-Rad Laboratories, Milan, Italy).

2.2. PCR primer design and amplification 2.3. Specificity and sensitivity of the BcSS-PCR assay The PCR primer sets were designed at the BCAN B0548 region of B. canis chromosome II. This site was carefully analyzed with CLC Main Workbench software version 6.0 (Insilicogen Inc., South Korea). The primer pairs were used as follows: forward 5 -CCAGATAGACCTCTCTGGA-3 and reverse 5 -TGGCCTTTTCTGATCTGTTCTT-3 . Amplification of specific DNA was performed using the amfiEco PCR premix kit (GenDEPOT Inc., Barker, USA). The PCR reactions

The BcSS-PCR assay was investigated using the Brucella and non-Brucella strains listed in Table 1. The results were compared to those of a 16S rRNA (F4/R2 primers) PCR assay, which was used to identify Brucella species as described previously [17]. The sensitivity of the BcSS-PCR assay was calculated to 10-fold serial diluted DNA samples. The concentration of B. canis DNA was determined by using

S.-I. Kang et al. / Comparative Immunology, Microbiology and Infectious Diseases 37 (2014) 237–241

an ND-1000 UV/VIS spectrophotometer (Nanodrop Tech., USA). 2.4. DNA extraction

239

Table 2 Direct detection from buffy coat of infected dogs by the B. canis speciesspecific PCR assay. Specimen no.

Isolationa

Serological test b

Genomic DNAs from all the strains and the buffy coat was extracted using a commercial blood and tissue kit (Qiagen Ltd., South Korea) according to the manufacturer’s instructions. 2.5. Buffy coat preparation The buffy coat was prepared using the Histopaque® 1083 solution (Sigma, South Korea). One millilitre of whole-blood was layered onto 1 ml of the Histopaque® 1083 solution in a sterile 2-ml centrifuge tube. The tube was centrifuged at 1500 × g for 30 min at room temperature. The agglomerative white band of leukocytes (buffy coat) was collected with a pipette and transferred to a sterile 1.5 ml eppendorf tube. 2.6. Comparison of the detection limit between buffy coat and whole-blood To determine the detection limit of the assay, fresh B. canis reference strains (RM6/66) were diluted with saline by 10-fold serial dilution. One-hundred microliters of each serially diluted solution was inoculated in triplicate onto TSA supplemented with 5% foetal bovine serum. After incubation for 3 days at 37 ◦ C, the colonies were counted and the number of inoculated bacteria was calculated. One hundred microliters of the serially diluted B. canis reference strains were inoculated into 900 ␮l of fresh canine whole blood. The whole blood then settled for 2 h at 37 ◦ C to permit invasion of B. canis into peripheral blood mononuclear cells such as macrophages. The DNA from 200 ␮l of the inoculated whole blood or from the buffy coat was extracted and compared to the sensitivity of the BcSS-PCR assay. 2.7. Clinical specimen Specimens were collected from 2 aborted foetuses and 13 whole blood samples from individual dogs in a breeding farm (Table 2). The foetal tissues were ground in 1 ml of PBS buffer, and 0.1 ml was plated onto TSA medium supplemented with 5% foetal bovine serum (FBS) (GIBCO, Grand Island, NY, USA). Into the same medium as above

1 2 3 4 5 6 7 8 9 10 11 12 13

+ + + + + + + + + + + + +

BcSS-PCR c

2-ME RSAT

ICA

+ + + + + + + − + + + + −

− − − + + + + + + + + + +

+ + + + + − − − − − + + +

a Isolation: there were identified using the classical biotyping assay and differential multiplex-PCR assay [14]. b 2-ME RSAT, 2-mercaptoethanol rapid slide agglutination test. c ICA, immuno-chromatographic assay [18].

was directly inoculated 0.1 ml of whole blood. The DNA extraction from tissues and buffy coat was performed using the DNeasy blood and tissue kit (Qiagen Ltd., South Korea). Distilled water was used as a negative control. An in house 2-mercaptoethanol rapid slide agglutination test kit (2-ME RSAT) with the B. canis M-strain and an immunochromatographic kit (Bionote Inc., South Korea) were used for serological tests [18].

3. Results Specific primer sets for detecting B. canis were designed carefully via alignments of a 12-bp deleted genetic site with those from other Brucella species. The forward and reverse primers were designed from BCAN B0548 encoding a hypothetical protein and from BCAN B0549 encoding a helix-turn-helix domain – containing protein. Amplification with these primers provided a 300-bp fragment. The BcSS-PCR assay was only positive for B. canis field strains and yielded a negative reaction for other Brucella species and non-Brucella bacterial species (Table 1). O. anthropi and Staphylococcus aureus were amplified by the 16S rRNA gene PCR assay as reported previously [19]. In our BcSS-PCR, these strains were not amplified. The optimal conditions of this species-specific PCR assay were

Fig. 1. The detection limit of the new Brucella canis specific polymerase chain reaction assay determined by the use of 10-fold serially diluted DNA. Lane M: 100-bp DNA ladder, lane 1: 60 ng/␮l, lane 2: 6 ng/␮l, lane 3: 0.6 ng/␮l, lane 4: 60 pg/␮l, lane 5: 6 pg/␮l, lane 6: 0.6 pg/␮l, lane 7: 60 fg/␮l, lane 8: negative control.

240

S.-I. Kang et al. / Comparative Immunology, Microbiology and Infectious Diseases 37 (2014) 237–241

Fig. 2. Comparison of the detection limit of the new Brucella canis specific polymerase chain reaction assay using whole blood inoculated with a B. canis strain or buffy coat. DNA extracted from whole blood directly (A) or from buffy coat concentrated by a commercial solution, Histopaque® -1083 (B). Lane M: 100-bp DNA ladder, lane 1: 3 × 108 , lane 2: 3 × 107 , lane 3: 3 × 106 , lane 4: 3 × 105 , lane 5: 3 × 104 , lane 6: 3 × 103 , lane 7: 3 × 102 , lane 8: 3 × 101 , lane 9: 3 × 100 CFU/ml; lane 10, negative control.

established, and the sensitivity was determined using 10fold serially diluted genomic DNA. The sensitivity of the BcSS-PCR assay was 6 pg/␮l of DNA (Fig. 1). The detection limit for bacterial cells was 3 × 105 CFU/ml when whole blood samples spiked with B. canis were used (Fig. 2a). However, using the buffy coat extracted by Histopaque from whole blood, the detection limit was 3 × 103 CFU/ml (Fig. 2b). The buffy coat was 5 times more concentrated than the whole blood, but the sensitivity of the PCR assay using the buffy coat was approximately 100 times greater than that of whole blood (Fig. 2). The B. canis strains were isolated from 2 aborted foetuses and 13 whole blood samples. These isolated strains have been characterized using the classical Brucella biotyping assay previously described, i.e., CO2 requirement, urea hydrolysis, oxidase, catalase test, dye medium test (basic fuchsin and thionin), and phage typing (Tbilisi [Tb], 104 Tb, Weybridge [Wb], and R/C) [4]. Serologically, 11 out of 13 specimens were positive by RSAT with 2-mercaptoethanol (2-ME RSAT). However, 3 specimens from the 2-ME RSAT positive dogs were negative by immuno-chromatographic assay (ICA). In contrast, two 2-ME RSAT negative dogs were positive by ICA (Table 2). The BcSS-PCR assay was also applied to these specimens to test its clinical utility. Abortion materials gave strong positive band of 300-bp for B. canis (data not shown). Compared with the isolation, 8 of 13 blood samples tested to be positive by the BcSS-PCR assay. Abortion materials were also positive for B. canis (data not shown). 4. Discussion Canine brucellosis is a zoonotic disease that poses a risk for public health, and it is also an important disease in pet dogs [6]. Canine brucellosis causes reproductive problems in dogs and economic losses to dog breeders. As the number of pet dogs increases, the risk of disease also increases due to direct or indirect transfer from dogs to people. Quick and accurate diagnostic tools are required to prevent the transfer of this disease between pets and humans and to minimize public health hazards and economic loss. A PCR assay is known to be the most effective diagnostic methods

for the detection of Brucella strains. However, a one-step PCR assay for the detection of the B. canis strain has not been developed so far. Our BcSS-PCR assay provided specific amplification for all B. canis isolates. This result was compared to the performance of a previously reported genus specific Brucella PCR assay based on the 16S rRNA gene. The field strains of Ochrobactrum anthropi and Staphylococcus aureus were falsely positive using PCR assay with the 16S rRNA gene (F4/R2 primer sets) [20,21]. The sensitivity of the PCR assay using buffy coat was approximately 100 times higher than using DNA extracted from whole blood. This finding can be attributed to a concentration of peripheral blood mononuclear cells and the absence of the inhibitory materials found in whole blood [22,23]. The inhibitory materials of DNA amplification may include anticoagulants and haemoglobin resulting from the concentration of peripheral blood mononuclear cells in whole blood [13]. According to previous reports, the detection limit of the BcSS-PCR described in the current study is equal to the sensitivity of the BCSP31 PCR with primer pairs of B4/B5, and less than that of the 16S rRNA PCR. The greater sensitivity of this PCR could be because it amplifies a region of the 16S rRNA gene, present in several copies in the bacterial genome [19]. In addition, the BcSS-PCR showed higher sensitivity than the PCR assay using JPF/JPR primer pairs of a gene encoding an outer membrane protein (omp-2) that, is used to detect Brucella species [19]. Of the 13 dogs studied, 2 were found negative for B. canis infection by results by 2-ME RSAT and 3 by ICA. In a serodiagnosis report by Keid et al. [24], among the blood-culture positive dogs, only 82.81% by RBT (rapid slide agglutination) and 39.06% by 2-ME RSAT had positive reaction. In addition, as reported by Abernethy et al. [25], the sensitivity of the RSAT was 76.6%. Kim et al. [18] reported that the kappa value between 2-ME RSAT and ICA was 0.89. Any single serological method cannot definitively diagnose canine brucellosis owing to low specificity and sensitivity, so substitutive or complementary diagnostic methods are required. The BcSS-PCR assay was evaluated with specimens including a buffy coat from whole blood and aborted foetal

S.-I. Kang et al. / Comparative Immunology, Microbiology and Infectious Diseases 37 (2014) 237–241

materials. Our PCR assay was not able to detect B. canis in some samples (number 6–10) irrespective of serological results. This may have been due to the low number of B. canis bacteria in blood or due to a loss of buffy coat during separation, which is dependent on the condition of the whole blood. In conclusion, the BcSS-PCR assay can specifically detect B. canis strains by a simple and easy method that could be available in a diagnostic centre, inspection agency or animal hospital. This BcSS-PCR assay has a comparatively high sensitivity when using a buffy coat sample from live animals. Therefore, our PCR assay could be used as an alternative diagnostic method to culture and serology. This BcSS-PCR assay could also be used as a routine screening tool for the diagnosis of B. canis in infected animals or humans, particularly in underdeveloped and developing countries. Conflict of interest The authors declare that they have no conflict of interest. Acknowledgements This study was supported by a grant (Project No: CAD13-2010-12-01) from the Animal and Plant Quarantine Agency (QIA) of the Ministry of Agriculture, Food and Rural Affairs (MAFRA) of the South Korea. References [1] Memish ZA, Balkhy HH. Brucellosis and international travel. J Travel Med 2004;11(1):49–55. [2] Pappas G, Papadimitriou P, Akritidis N, Christou L, Tsianos EV. The new global map of human brucellosis. Lancet Infect Dis 2006;6(2): 91–9. [3] Wee SH, Nam HM, Kim CH. Emergence of brucellosis in cattle in the Republic of Korea. Vet Rec 2008;162(17):556–7. [4] Office International des Épizooties O.I.E. Manual of diagnostic tests and vaccines for terrestrial animals. 6th ed. Paris, France: Office International des Épizooties; 2008. p. 624–9. [5] Sohn AH, Probert WS, Glaser CA, Gupta N, Bollen AW, Wong JD, et al. Human neurobrucellosis with intracerebral granuloma caused by a marine mammal Brucella spp. Emerg Infect Dis 2003;9(4):485–8. [6] Hollett RB. Canine brucellosis: outbreaks and compliance. Theriogenology 2006;66(3):575–87. [7] Greene CE, Carmichael LE. Infectious Diseases of Dog and Cat. 3rd ed. Philadelphia: WB Saunders Co.; 2006. p. 369–81.

241

[8] Lucero NE, Jacob NO, Ayala SM, Escobar GI, Tuccillo P, Jacques I. Unusual clinical presentation of brucellosis caused by Brucella canis. J Med Microbiol 2005;54(Pt 5):505–8. [9] Baily GG, Krahn JB, Drasar BS, Stoker NG. Detection of Brucella melitensis and Brucella abortus by DNA amplification. J Trop Med Hyg 1992;95(4):271–5. [10] Bricker BJ. PCR as a diagnostic tool for brucellosis. Vet Microbiol 2002;90(1–4):435–46. [11] Keid LB, Soares RM, Vasconcellos SA, Megid J, Salgado VR, Richtzenhain LJ. Comparison of agar gel immunodiffusion test, rapid slide agglutination test, microbiological culture and PCR for the diagnosis of canine brucellosis. Res Vet Sci 2009;86(1):22–6. [12] Carmichael LE, Shin SJ. Canine brucellosis: a diagnostician’s dilemma. Semin Vet Med Surg (Small Anim) 1996;11(3):161–5. [13] Keid LB, Soares RM, Vasconcellos SA, Salgado VR, Megid J, Richtzenhain LJ. Comparison of a PCR assay in whole blood and serum specimens for canine brucellosis diagnosis. Vet Rec 2010;167(3): 96–9. [14] Kang SI, Her M, Kim JW, Kim JY, Ko KY, Ha YM, et al. Advanced multiplex PCR assay for differentiation of Brucella species. Appl Environ Microbiol 2011;77(18):6726–8. [15] Winchell JM, Wolff BJ, Tiller R, Bowen MD, Hoffmaster AR. Rapid identification and discrimination of Brucella isolates by use of real-time PCR and high-resolution melt analysis. J Clin Microbiol 2010;48(3):697–702. [16] Koylass MS, King AC, Edwards-Smallbone J, Gopaul KK, Perrett LL, Whatmore AM. Comparative performance of SNP typing and ‘Bruceladder’ in the discrimination of Brucella suis and Brucella canis. Vet Microbiol 2010;142(3–4):450–4. ˜ I. Specific detection of [17] Romero C, Gamazo C, Pardo M, López-Goni Brucella DNA by PCR. J Clin Microbiol 1995;33(3):615–7. [18] Kim JW, Lee YJ, Han MY, Bae DH, Jung SC, Oh JS, et al. Evaluation of immunochromatographic assay for serodiagnosis of Brucella canis. J Vet Med Sci 2007;69(11):1103–7. [19] Navarro E, Escribano J, Fernández J, Solera J. Comparison of three different PCR methods for detection of Brucella spp. in human blood samples. FEMS Immunol Med Microbiol 2002;34(2): 147–51. [20] Al-Ajian HH, Ibrahim AS, Al-Salamah AA. Comparison of different PCR methods for detection of Brucella spp. in human blood samples. Pol J Microbiol 2011;60(1):27–33. [21] Horvat RT, El Atrouni W, Hammoud K, Hawkinson D, Cowden S. Ribosomal RNA sequence analysis of Brucella infection misidentified as Ochrobactrum anthropi infection. J Clin Microbiol 2011;49(3):1165–8. [22] Goto-Koshino Y, Ohno K, Nakajima M, Mochizuki H, Kanemoto H, Tsujimoto H. A rapid and simple method to obtain canine peripheral blood-derived macrophages. J Vet Med Sci 2011;73(6): 773–8. [23] Salam MA, Khan MG, Bhaskar KR, Afrad MH, Huda MM, Mondal D. Peripheral blood buffy coat smear: a promising tool for diagnosis of visceral leishmaniasis. J Clin Microbiol 2012;50(3):837–40. [24] Keid LB, Soares RM, Vieira NR, Megid J, Salgado VR, Vasconcellos SA, et al. Diagnosis of canine brucellosis: comparison between serological and microbiological tests and a PCR based on primers to 16S-23S rDNA interspacer. Vet Res Commun 2007;31(8):951–65. [25] Abernethy DA, Menzies FD, McCullough SJ, McDowell SWJ, Burns KE, Watt R, et al. Field trial of six serological tests for bovine brucellosis. Vet J 2012;191(3):364–70.