Ad

SJSa

Gb

Cc

a

ARRA

KBBDSZ

1

csScS1s

0

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

Contents lists available at ScienceDirect

Comparative Immunology, Microbiologyand Infectious Diseases

j o ur na l ho me pag e: www.elsev ier .com/ locate /c imid

new Brucella canis species-specific PCR assay for theiagnosis of canine brucellosis

ung-Il Kanga, Sang-Eun Leeb, Ji-Yeon Kima, Kichan Leea,ong-Wan Kima, Hyang-Keun Leea, So-Ra Sunga, Young-Ran Heoc,uk Chan Junga, Moon Hera,∗

The OIE Reference Laboratory for Brucellosis, Bacterial Disease Division, Animal and Plant Quarantine Agency, 480 Anyang,yeonggi-do 430-757, South KoreaDivision of Malaria & Parasitic Diseases, Korea National Institute of Health, Korea Centers for Disease Control & Prevention,hengwon-gun, South KoreaDepartment of Food and Nutrition, Chonnam National University, Yongbongdong, Gwanju, South Korea

r t i c l e i n f o

rticle history:eceived 25 February 2014eceived in revised form 30 June 2014ccepted 13 July 2014

eywords:rucella canisuffy coatiagnostics

a b s t r a c t

Brucellosis is a zoonotic disease that is transmitted from animals to humans, and the devel-opment of a rapid, accurate, and widely available identification method is essential fordiagnosing 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 setwas designed based on the BCAN B0548-0549 region in chromosome II of B. canis. The PCRdetection for B. canis included amplification of a 300-bp product that is, not found on otherBrucella species or, genetically or serologically related bacteria. The detection limit of BcSS-PCR assay was 6 pg/�l by DNA dilution, or 3 × 103 colony-forming units (CFU) in the buffycoats separated from whole blood experimentally inoculated with B. canis. Using the buffy

pecies-specific PCR assay

oonosis coat in this PCR assay resulted in approximately 100-times higher sensitivity for B. canis ascompared to detect directly from whole blood. This is the first report of a species-specificPCR assay to detect B. canis, and the new assay will provide a valuable tool for the diagnosisof B. canis infection.

© 2014 Elsevier Ltd. All rights reserved.

. Introduction

The genus Brucella is gram-negative, facultative intra-ellular pathogens. Since the first isolation of a Brucellatrain, this genus has been found to be prevalent inouth America, Africa, Southeast Asia, and some Europeanountries. This disease has also occurred sporadically in

outh Korea [1–3]. Currently, the genus Brucella consists of0 species preference of animal hosts: B. abortus, B. canis, B.uis, 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: herm@korea.kr (M. Her).

http://dx.doi.org/10.1016/j.cimid.2014.07.003147-9571/© 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 marinemammals are also known to have zoonotic potential[5].

With the rapid growth of the pet and companion animalindustries, Brucella infection could become a serious publichealth issue. Currently, canine brucellosis resulting fromB. canis infection is thought to be underestimated world-wide [6]. In dogs, B. canis can cause contagious abortion,

testicular atrophy, infertility and lymphadenitis. The infec-tious route is mainly through direct contact with the foetus,placenta, foetal fluids, and/or vaginal discharge after abor-tions [7]. Humans generally become infected through close

crobiology and Infectious Diseases 37 (2014) 237–241

Table 1Bacteria strains used in this study and comparison of the results of twoPCR assays.

Species Strains PCR results

16SrRNAa

B. canisPCRb

Brucella speciesB. abortus bv. 1 544 ATCC 23448 + −B. abortus bv. 2 86/8/59 ATCC 23449 + −B. abortus bv. 3 Tulya ATCC 23450 + −B. abortus bv. 4 292 ATCC 23451 + −B. abortus bv. 5 B3196 ATCC 23452 + −B. abortus bv. 6 870 ATCC 23453 + −B. abortus bv. 9 C68 ATCC 23455 + −B. canis RM6/66 ATCC 23365 + +B. suis bv. 1 1330 ATCC 23444 + −B. suis bv. 2 Thomsen ATCC 23445 + −B. suis bv. 3 686 ATCC 23446 + −B. suis bv. 4 40 ATCC 23447 + −B. suis bv. 5 513 NCTC 11996 + −B. ovis 63/290 ATCC 25840 + −B. neotomae 5K33 ATCC 23459 + −B. melitensis bv. 1 16M ATCC 23456 + −B. melitensis bv. 2 63/9 ATCC 23457 + −B. melitensis bv. 3 Ether ATCC 23458 + −B. ceti B1/94 NCTC 12891 + −B. pinnipedialis B2/94 NCTC 12890 + −B. microti CCM4915 BCCN 07-01 + −B. inopinata B01 BCCN 09-01 + −B. canis 94 isolates + +

Non-Brucella organismsOchrobactrum anthropic Field strain + −Escherichia coli O157:H7c Field strain − −Pasteurella multocida ATCC 43017 − −Salmonella Typhimurium ATCC 14028 − −Campylobacter jejuni ATCC 33560 − −Yersinia enterocolitica O: 9 NCTC 11174 − −Staphylococcus aureusc Field strain + −Clostridium perfringens type A ATCC 13124 − −a Brucella genus specific PCR assay that targets the 16S rRNA [17].

238 S.-I. Kang et al. / Comparative Immunology, Mi

contact with infected dogs or abortion-related materials[8].

Generally, diagnosis of canine brucellosis is based onclassical biotyping methods, serological tests or molecu-lar techniques [4,9–11]. Whole blood is a useful samplefor isolation of B. canis because of the prolonged bacter-aemia, which characterizes the disease [6,12]. However,bacterial isolation has its disadvantages, because B. canisis a slow-growing organism that requires a long incuba-tion period and blind subcultures. This can also pose risks toresearchers working with the organism [6,13]. The serolog-ical tests routinely used for diagnosis of canine brucellosishave the disadvantages of low sensitivity and varyingspecificity [11]. Owing to these shortcomings, a variety ofmolecular techniques using Brucella DNA have been devel-oped to identify B. canis, such as a real-time PCR usingsingle nucleotide polymorphisms (SNPs) and a multiplexdifferential PCR [14,15]. These molecular techniques havetheir own shortcomings, including high expenditure andexpensive equipment. Currently using multiplex PCR it isonly possible to distinguish Brucella species by using a highquality and concentration genomic DNA from the bacteria.Therefore, this technique limits the choice of clinical speci-mens because of low sensitivity. On the hand, a high degreeof genetic similarity with other Brucella species, especiallyB. suis, has interrupted the development of a simple PCRassay to detect or identify B. canis until now [16,17].

Therefore, the aim of the current study was to developa specific PCR assay and evaluate its potential for detectingB. canis from blood and abortion-related materials of dogs.

2. Materials and methods

2.1. Bacterial strains

The strains used in this study included 22 Brucella ref-erence strains; 94 B. canis field isolates; and 8 non-Brucellastrains, including serologically cross-reacting strains orphylogenetically related strains (Table 1). All of the B. canisfield isolates were obtained from dog-breeding farms dur-ing 2002–2011 and were identified by classical biotypingassays and molecular methods [4,14]. Other non-Brucellastrains were identified by 16S rRNA sequence analysis andVITEK 2 Compact (Biomerieux, Seoul, South Korea).

The Brucella strains were cultured on tryptic soy agar(TSA) (BD, Franklin Lakes, NJ) supplemented with 5% foetalbovine serum (FBS) (GIBCO, Grand Island, NY, USA) for 3days at 37 ◦C under 5% CO2 or aerobic conditions. Othernon-Brucella organisms were grown on 5% sheep blood agaror MacConkey agar for 18–24 h at 37 ◦C.

2.2. PCR primer design and amplification

The PCR primer sets were designed at the BCAN B0548region of B. canis chromosome II. This site was care-fully analyzed with CLC Main Workbench software version6.0 (Insilicogen Inc., South Korea). The primer pairs were

used as follows: forward 5′-CCAGATAGACCTCTCTGGA-3′

and reverse 5′-TGGCCTTTTCTGATCTGTTCTT-3′. Amplifica-tion of specific DNA was performed using the amfiEco PCRpremix kit (GenDEPOT Inc., Barker, USA). The PCR reactions

b B. canis species-specific PCR assay developed in this study.c The field strains were identified as analysis of 16S ribosomal RNA gene

sequence.

were performed using a 20-�l reaction mixture contain-ing 2 �l of template DNA and 2 �l of each of the primers(10 pmol). The PCR conditions consisted of an initial dena-turation at 94 ◦C for 7 min; 35 cycles of denaturation at94 ◦C for 35 s, annealing at 59 ◦C for 40 s, and extensionat 72 ◦C for 35 s, followed by a final extension at 72 ◦Cfor 10 min. All amplicons were analyzed by electrophore-sis using 1.5% agarose gel with a 100-bp ladder (BioneerCo., Taejon, South Korea) as a molecular size marker andethidium bromide (0.02 ml/ml) for staining. Stained gelswere visualized and photographed under UV light with aUV transilluminator (Bio-Rad Laboratories, Milan, Italy).

2.3. Specificity and sensitivity of the BcSS-PCR assay

The BcSS-PCR assay was investigated using the Bru-cella and non-Brucella strains listed in Table 1. The resultswere compared to those of a 16S rRNA (F4/R2 primers)

PCR assay, which was used to identify Brucella species asdescribed previously [17]. The sensitivity of the BcSS-PCRassay was calculated to 10-fold serial diluted DNA samples.The concentration of B. canis DNA was determined by using

crobiology and Infectious Diseases 37 (2014) 237–241 239

aU

2

w(i

2

1w1cTw1

2c

cbsTbnmwTislc

2

1iosG

Table 2Direct detection from buffy coat of infected dogs by the B. canis species-specific PCR assay.

Specimen no. Isolationa Serological test BcSS-PCR

2-ME RSATb ICAc

1 + + − +2 + + − +3 + + − +4 + + + +5 + + + +6 + + + −7 + + + −8 + − + −9 + + + −

10 + + + −11 + + + +12 + + + +13 + − + +

a Isolation: there were identified using the classical biotyping assay anddifferential multiplex-PCR assay [14].

FMc

S.-I. Kang et al. / Comparative Immunology, Mi

n ND-1000 UV/VIS spectrophotometer (Nanodrop Tech.,SA).

.4. DNA extraction

Genomic DNAs from all the strains and the buffy coatas extracted using a commercial blood and tissue kit

Qiagen Ltd., South Korea) according to the manufacturer’snstructions.

.5. Buffy coat preparation

The buffy coat was prepared using the Histopaque®-083 solution (Sigma, South Korea). One millilitre ofhole-blood was layered onto 1 ml of the Histopaque®-

083 solution in a sterile 2-ml centrifuge tube. The tube wasentrifuged at 1500 × g for 30 min at room temperature.he agglomerative white band of leukocytes (buffy coat)as collected with a pipette and transferred to a sterile

.5 ml eppendorf tube.

.6. Comparison of the detection limit between buffyoat and whole-blood

To determine the detection limit of the assay, fresh B.anis reference strains (RM6/66) were diluted with saliney 10-fold serial dilution. One-hundred microliters of eacherially diluted solution was inoculated in triplicate ontoSA supplemented with 5% foetal bovine serum. After incu-ation for 3 days at 37 ◦C, the colonies were counted and theumber of inoculated bacteria was calculated. One hundredicroliters of the serially diluted B. canis reference strainsere inoculated into 900 �l of fresh canine whole blood.

he whole blood then settled for 2 h at 37 ◦C to permitnvasion of B. canis into peripheral blood mononuclear cellsuch as macrophages. The DNA from 200 �l of the inocu-ated whole blood or from the buffy coat was extracted andompared to the sensitivity of the BcSS-PCR assay.

.7. Clinical specimen

Specimens were collected from 2 aborted foetuses and3 whole blood samples from individual dogs in a breed-

ng farm (Table 2). The foetal tissues were ground in 1 mlf PBS buffer, and 0.1 ml was plated onto TSA mediumupplemented with 5% foetal bovine serum (FBS) (GIBCO,rand Island, NY, USA). Into the same medium as above

ig. 1. The detection limit of the new Brucella canis specific polymerase chain re: 100-bp DNA ladder, lane 1: 60 ng/�l, lane 2: 6 ng/�l, lane 3: 0.6 ng/�l, lane 4: 6

ontrol.

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 DNAextraction from tissues and buffy coat was performedusing the DNeasy blood and tissue kit (Qiagen Ltd., SouthKorea). Distilled water was used as a negative control. Anin house 2-mercaptoethanol rapid slide agglutination testkit (2-ME RSAT) with the B. canis M-strain and an immuno-chromatographic kit (Bionote Inc., South Korea) were usedfor serological tests [18].

3. Results

Specific primer sets for detecting B. canis were designedcarefully via alignments of a 12-bp deleted genetic sitewith those from other Brucella species. The forward andreverse primers were designed from BCAN B0548 encod-ing a hypothetical protein and from BCAN B0549 encodinga helix-turn-helix domain – containing protein. Amplifica-tion with these primers provided a 300-bp fragment.

The BcSS-PCR assay was only positive for B. canis fieldstrains and yielded a negative reaction for other Brucellaspecies 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]. Inour BcSS-PCR, these strains were not amplified. The opti-mal conditions of this species-specific PCR assay were

action assay determined by the use of 10-fold serially diluted DNA. Lane0 pg/�l, lane 5: 6 pg/�l, lane 6: 0.6 pg/�l, lane 7: 60 fg/�l, lane 8: negative

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. canisstrain or buffy coat. DNA extracted from whole blood directly (A) or from buffy coat concentrated by a commercial solution, Histopaque®-1083 (B). Lane

e 4: 3 ×

M: 100-bp DNA ladder, lane 1: 3 × 108, lane 2: 3 × 107, lane 3: 3 × 106, lan9: 3 × 100 CFU/ml; lane 10, negative control.

established, and the sensitivity was determined using 10-fold serially diluted genomic DNA.

The sensitivity of the BcSS-PCR assay was 6 pg/�l ofDNA (Fig. 1). The detection limit for bacterial cells was3 × 105 CFU/ml when whole blood samples spiked with B.canis were used (Fig. 2a). However, using the buffy coatextracted by Histopaque from whole blood, the detectionlimit was 3 × 103 CFU/ml (Fig. 2b). The buffy coat was 5times more concentrated than the whole blood, but thesensitivity of the PCR assay using the buffy coat was approx-imately 100 times greater than that of whole blood (Fig. 2).

The B. canis strains were isolated from 2 aborted foe-tuses and 13 whole blood samples. These isolated strainshave been characterized using the classical Brucella biotyp-ing assay previously described, i.e., CO2 requirement, ureahydrolysis, oxidase, catalase test, dye medium test (basicfuchsin and thionin), and phage typing (Tbilisi [Tb], 104 Tb,Weybridge [Wb], and R/C) [4].

Serologically, 11 out of 13 specimens were positive byRSAT with 2-mercaptoethanol (2-ME RSAT). However, 3specimens from the 2-ME RSAT positive dogs were negativeby immuno-chromatographic assay (ICA). In contrast, two2-ME RSAT negative dogs were positive by ICA (Table 2).The BcSS-PCR assay was also applied to these specimens totest its clinical utility. Abortion materials gave strong posi-tive band of 300-bp for B. canis (data not shown). Comparedwith the isolation, 8 of 13 blood samples tested to be pos-itive by the BcSS-PCR assay. Abortion materials were alsopositive for B. canis (data not shown).

4. Discussion

Canine brucellosis is a zoonotic disease that poses a riskfor public health, and it is also an important disease in petdogs [6]. Canine brucellosis causes reproductive problemsin dogs and economic losses to dog breeders. As the numberof pet dogs increases, the risk of disease also increases dueto direct or indirect transfer from dogs to people. Quick

and accurate diagnostic tools are required to prevent thetransfer of this disease between pets and humans and tominimize public health hazards and economic loss. A PCRassay is known to be the most effective diagnostic methods

105, lane 5: 3 × 104, lane 6: 3 × 103, lane 7: 3 × 102, lane 8: 3 × 101, lane

for the detection of Brucella strains. However, a one-stepPCR assay for the detection of the B. canis strain has notbeen developed so far.

Our BcSS-PCR assay provided specific amplification forall B. canis isolates. This result was compared to the per-formance of a previously reported genus specific BrucellaPCR assay based on the 16S rRNA gene. The field strainsof Ochrobactrum anthropi and Staphylococcus aureus werefalsely positive using PCR assay with the 16S rRNA gene(F4/R2 primer sets) [20,21].

The sensitivity of the PCR assay using buffy coat wasapproximately 100 times higher than using DNA extractedfrom whole blood. This finding can be attributed to a con-centration of peripheral blood mononuclear cells and theabsence of the inhibitory materials found in whole blood[22,23]. The inhibitory materials of DNA amplification mayinclude anticoagulants and haemoglobin resulting fromthe concentration of peripheral blood mononuclear cells inwhole blood [13]. According to previous reports, the detec-tion limit of the BcSS-PCR described in the current study isequal to the sensitivity of the BCSP31 PCR with primer pairsof B4/B5, and less than that of the 16S rRNA PCR. The greatersensitivity of this PCR could be because it amplifies a regionof the 16S rRNA gene, present in several copies in the bacte-rial genome [19]. In addition, the BcSS-PCR showed highersensitivity than the PCR assay using JPF/JPR primer pairs ofa 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. canisinfection by results by 2-ME RSAT and 3 by ICA. In a serodi-agnosis report by Keid et al. [24], among the blood-culturepositive dogs, only 82.81% by RBT (rapid slide agglutina-tion) and 39.06% by 2-ME RSAT had positive reaction. Inaddition, as reported by Abernethy et al. [25], the sensi-tivity of the RSAT was 76.6%. Kim et al. [18] reported thatthe kappa value between 2-ME RSAT and ICA was 0.89.Any single serological method cannot definitively diagnosecanine brucellosis owing to low specificity and sensitivity,

so substitutive or complementary diagnostic methods arerequired.

The BcSS-PCR assay was evaluated with specimensincluding a buffy coat from whole blood and aborted foetal

crobiolo

mirciw

BbahaaBthc

C

e

A

AAA

R

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

S.-I. Kang et al. / Comparative Immunology, Mi

aterials. Our PCR assay was not able to detect B. canisn some samples (number 6–10) irrespective of serologicalesults. This may have been due to the low number of B.anis bacteria in blood or due to a loss of buffy coat dur-ng separation, which is dependent on the condition of the

hole blood.In conclusion, the BcSS-PCR assay can specifically detect

. canis strains by a simple and easy method that coulde available in a diagnostic centre, inspection agency ornimal hospital. This BcSS-PCR assay has a comparativelyigh sensitivity when using a buffy coat sample from livenimals. Therefore, our PCR assay could be used as anlternative diagnostic method to culture and serology. ThiscSS-PCR assay could also be used as a routine screeningool for the diagnosis of B. canis in infected animals orumans, particularly in underdeveloped and developingountries.

onflict of interest

The authors declare that they have no conflict of inter-st.

cknowledgements

This study was supported by a grant (Project No: C-D13-2010-12-01) from the Animal and Plant Quarantinegency (QIA) of the Ministry of Agriculture, Food and Ruralffairs (MAFRA) of the South Korea.

eferences

[1] Memish ZA, Balkhy HH. Brucellosis and international travel. J TravelMed 2004;11(1):49–55.

[2] Pappas G, Papadimitriou P, Akritidis N, Christou L, Tsianos EV. Thenew 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 theRepublic of Korea. Vet Rec 2008;162(17):556–7.

[4] Office International des Épizooties O.I.E. Manual of diagnostic testsand vaccines for terrestrial animals. 6th ed. Paris, France: Office Inter-national 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. Therioge-nology 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.

[

gy and Infectious Diseases 37 (2014) 237–241 241

[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 Brucellamelitensis and Brucella abortus by DNA amplification. J Trop Med Hyg1992;95(4):271–5.

10] Bricker BJ. PCR as a diagnostic tool for brucellosis. Vet Microbiol2002;90(1–4):435–46.

11] Keid LB, Soares RM, Vasconcellos SA, Megid J, Salgado VR, Richtzen-hain LJ. Comparison of agar gel immunodiffusion test, rapid slideagglutination test, microbiological culture and PCR for the diagnosisof 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, Richtzen-hain LJ. Comparison of a PCR assay in whole blood and serumspecimens 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 mul-tiplex PCR assay for differentiation of Brucella species. Appl EnvironMicrobiol 2011;77(18):6726–8.

15] Winchell JM, Wolff BJ, Tiller R, Bowen MD, Hoffmaster AR. Rapididentification and discrimination of Brucella isolates by use ofreal-time PCR and high-resolution melt analysis. J Clin Microbiol2010;48(3):697–702.

16] Koylass MS, King AC, Edwards-Smallbone J, Gopaul KK, Perrett LL,Whatmore AM. Comparative performance of SNP typing and ‘Bruce-ladder’ in the discrimination of Brucella suis and Brucella canis. VetMicrobiol 2010;142(3–4):450–4.

17] Romero C, Gamazo C, Pardo M, López-Goni I. Specific detection ofBrucella 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 ofimmunochromatographic assay for serodiagnosis of Brucella canis. JVet Med Sci 2007;69(11):1103–7.

19] Navarro E, Escribano J, Fernández J, Solera J. Comparison ofthree different PCR methods for detection of Brucella spp. inhuman blood samples. FEMS Immunol Med Microbiol 2002;34(2):147–51.

20] Al-Ajian HH, Ibrahim AS, Al-Salamah AA. Comparison of different PCRmethods for detection of Brucella spp. in human blood samples. Pol JMicrobiol 2011;60(1):27–33.

21] Horvat RT, El Atrouni W, Hammoud K, Hawkinson D, Cow-den S. Ribosomal RNA sequence analysis of Brucella infectionmisidentified as Ochrobactrum anthropi infection. J Clin Microbiol2011;49(3):1165–8.

22] Goto-Koshino Y, Ohno K, Nakajima M, Mochizuki H, KanemotoH, Tsujimoto H. A rapid and simple method to obtain canineperipheral 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 ofvisceral 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 serologi-

cal and microbiological tests and a PCR based on primers to 16S-23SrDNA 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.