First isolation, identification and genetic characterization of Brucella abortus biovar 3 from dairy cattle in Bangladesh

Abstract Background Brucellosis is a zoonotic disease caused by bacteria Brucella spp. belonging to the genus Brucella. It is endemic in domesticated animals in Bangladesh. Isolation, identification and genetic characterization of Brucella spp. in dairy cattle are essential to undertake appropriate control and preventive measures. The study was conducted to isolate and characterize the Brucella spp. circulating in dairy cattle. Methods Uterine discharge (n = 45), milk (n = 115), vaginal swab (n = 71), placenta (n = 7) and aborted fetus (n = 2) were collected. Brucella selective agar plates were inoculated with samples and incubated at 37 ◦C for 14 days under 5% CO2 for isolation of Brucella spp. Brucella suspected colonies were recovered from samples were confirmed by genus and species specific PCR assays. Genetic characterization was performed by Multi Locus Variable number tandem‐repeat Analysis‐16 (MLVA‐16). Results The isolates of Brucella recovered from samples were confirmed as B. abortus by AMOS‐ERY PCR assay. The classical biotyping method confirmed all 10 B. abortus isolates belonged to the biovar 3. The MLVA‐16 assay indicated all B. abortus isolates identical and the same genotype 40, based on panel 1 MLVA‐8. Conclusion Dendrogram analysis revealed all B. abortus isolates of the study were identical to three isolates from Brazil, one isolate of France and closely related to Chinese isolates. This is the first report of isolation and genetic characterization of B. abortus from the dairy cattle in Bangladesh.

Brucellosis causes abortion, infertility, still birth and reduced milk production in animals. Animals get infected either through consumption of contaminated feed and water or contact with an infected animal. Routine bacteriological method and a classical biotyping scheme are used for characterization of Brucella both at species and subspecies levels (Alton, Jones, & Pietz, 1975;Whatmore et al., 2016). Polymerase chain reaction (PCR) assays are currently used for identification of Brucella at genus, species and biovar levels (Bricker & Halling, 1994;Ocampo-Sosa et al., 2005;Romero, Gamazo, & Pardo, 1995). Multi Locus Variable number tandem-repeat Analysis (MLVA) assay is used for genetic characterization of Brucella isolates (Le Fletch et al., 2006;Whatmore et al., 2006).
Isolation of Brucella at the genus level has been reported in milk sample of cattle (Islam et al., 2018). Brucella genus-specific DNA has been identified in the sera of humans by real time PCR assay (Rahman et al., 2012). The B. aborus species-specific DNA has been detected in the sera of cattle by real time PCR (Rahman et al., 2014). Isolation of Brucella from the infected host is considered as the gold standard for diagnosis of brucellosis (Rahman et al., 2012). However, identification of Brucella at species level and its biovar typing and genetic characterization of circulating Brucella spp. has not been reported in Bangladesh. The objectives of the present research work are: i) isolation of Brucella spp. from dairy cattle experiencing abortion, ii) identification of Brucella at species and biovar levels and iii) genetic characterization of circulating Brucella spp. by MLVA-16 assay.  (Figure 1).

| Study areas and samples
Each of the dairy farm consisted of 20-100 cattle which were indigenous breed and crossbreed of Friesian, Sahiwal and Red Chittagong.
The cattle of the study farms were not vaccinated against brucellosis. A total of 240 samples consisted of uterine discharges (n = 45), milk (n = 115), vaginal swabs (n = 71), placenta (n = 7) and aborted fetuses (n = 2) were collected from dairy cattle with the clinical sign of abortion (n = 55) or without the history of abortion (n = 185) during the period from August 2016 to December 2017. Samples were collected after 1 to 3 days of abortion. The abortion was occurred at third trimester of gestation from first to third pregnancy. Uterine discharge (10 ml) was collected by inserting a disposable artificial insemination pipette into the uterus. Midstream milk sample (20 ml) was collected from each quarter of the udder into a sterile 50 ml falcon tube. Vaginal swab was collected using a sterile applicator stick (HiMedia,, Mumbai, India). Placental cotyledons were collected from aborted cattle aseptically using sterile forceps and scissors and kept in a sterile plastic container. The aspirate of stomach content (50ml) was collected from the aborted fetus. The samples were transported to the Department of Microbiology and Hygiene, Bangladesh Agricultural University, using an ice box and kept at 4°C and cultured within 3 days.

| Isolation and biotyping of bacteria
Uterine discharge, vaginal swab and aspirate of fetal stomach content were streaked duplicate onto the Brucella selective agar supplemented with antibiotics (polymyxin B sulphate, bacitracin, nystatin, cycloheximide, nalidixic acid, vancomycin) (HiMedia, Mumbai, India) that inhibit growth of bacteria other than Brucella (Alton, Jones, Angus, & Veger, 1988). Milk was centrifuged at 3500rpm for 15 min. The cream and sediment were inoculated onto the Brucella selective agar (HiMedia, Mumbai, India) using a sterile cotton swab. Placental cotyledons were cut into small pieces and placed in a sterile plastic bag with equal volume of phosphate-buffered saline. The cotyledons were macerated by a stomacher for 5 min and tissue homogenate was inoculated onto the Brucella selective agar (HiMedia, Mumbai, India) using a sterile F I G U R E 1 Map of Bangladesh indicating study areas by colour highlights cotton swab. Inoculated plates were placed in an incubator supplied with 5% CO 2 at 37°C. The plates were observed daily up to 14 days for Brucella like colonies (smooth, small, translucent, glistening, dew drop like round and convex colony). Identification of bacteria in pure culture was performed by colony morphology, Gram's staining reaction, catalase, oxidase, H 2 S and urease tests (Alton et al., 1975).
Brucella spp. were subjected to classical biotyping described by Alton et al. (1988). A panel of biotyping tests such as CO 2 requirement for growth, H 2 S production and growth in presence of thionine and basic fuchsin were performed.

| Molecular identification and genotyping of B. abortus
The genomic DNA was extracted from suspect Brucella colonies by a genomic DNA extraction kit using manufacturer's protocol (GeneJet Genomic DNA Purificaion Kit, Thermo Fisher Scientific, Vilnius, Lithuania).
To confirm Brucella spp. at molecular level a genus specific PCR assay targeting 905 bp fragment of the 16S rRNA gene was performed (Romero et al., 1995). Identification of B. abortus biovar 1, 2 and 4 was performed by AMOS PCR assay with oligonucleotide primers and PCR conditions described by Bricker and Halling (1994). Amplification of MLVA-16 loci was performed using multiplex PCRs as described previously (Garofolo, Ancora, & Giannatale, 2013). PCR amplifications were performed in a total volume of 10 μl containing 1.50 ng DNA, 1 × Type-it microsatellite PCR Master Mix (QiagenSrl, Milan, Italy), and proper concentration of each fluorescent primer pairs (Garofolo et al., 2013). Thermal cycling was conducted on a GeneAmp 9700 thermal cycler (Applied Biosystems) following thermal reaction profiles: initial heating at 95°C for 5 min, 30 cycles denaturation at 95°C for 30 s, annealing at 60°C for 90 s and extension at 72°C for 30 s. A final extension step at 60°C for 45 min and 20°C for 120 min was run to reduce artefacts such as stutter and non-templated 3' A nucleotide additions. Fragments were then separated through capillary electrophoresis on an ABI 3500 instrument with POP 7 polymer. Data analysis was done using

| Isolation and biotyping characteristics
Ten Brucella spp. were isolated from uterine discharge (n = 7, sample They grew in a 5% CO 2 atmosphere after 3-14 days incubation at 37°C. Bacterial colonies were small, convex and regular with smooth surface, honey coloured, shiny and translucent. The organisms appeared to be Gram negative, small coccobacilli arranged singly or in pairs. The isolates were catalase, oxidase, H 2 S and urease positive. The isolates grew in the presence of thionin and basic fuchsin dyes suggesting that all isolates belonged to the biovar 3 (Table 1).

| D ISCUSS I ON
Brucellosis causes abortion in the third trimester of bovine pregnancy (Megid, Mathias, & Robles, 2010 (Capparelli et al., 2009). In F I G U R E 2 An extract from an unweighted dendrogram constructed from the profiles of 1633 Brucella isolates submitted to the international MLVA database (MLVA-NET) using UPGMA analysis, categorical coefficient plus the 10 isolates from Bangladesh. The MLVA-16 profiles of the 10 isolates from Bangladesh are shown to cluster with isolates from Brazil. The columns following the data represent sample ID, country of origin, species/biovar and year of isolation. The coloured boxes denote country of origin brucellosis endemic areas, transmission of Brucella to humans can occur through consumption of unpasteurized milk (Deshmukh et al., 2015). In Bangladesh milk ring tests are not routinely practiced for screening B. abortus specific antibodies in milk (Islam et al., 2018).
The presence of B. abortus in cattle milk constitutes a public health hazard as people in Bangladesh mostly purchase unpasteurized milk.
In the study, AMOS PCR assay failed to amplify a 498 bp PCR amplicon (data not shown) indicating none of the Brucella isolates belonged to the B. abortus biovar 1, 2 and 4 (Bricker & Halling, 1994).
The AMOS-ERY assay identifies B. abortus biovar 3b, 5, 6 and 9 (Ocampososa et al., 2005) suggesting that the B. abortus isolates from cattle might be any one of these four biovars.
In the present study, all B. abortus isolates of cattle belonged to the biovar 3, indicating this biovar is being transmitted in the dairy cattle in the study areas. In this study, B. abortus was isolated from dairy farms located in two neighbouring districts of Bangladesh; Dhaka (Savar) and Gazipur. Therefore, it is very difficult to draw a conclusion that the B. abortus biovar 3 is predominately circulating in dairy population in Bangladesh as this study screened only a small number of samples obtained from aborted cows in a limited geographical area of Bangladesh. All isolates were designated MLVA panel-1 genotype 40 and had full MLVA-16 profiles identical to three isolates from Brazil (Minharro et al., 2013) and one isolate from France (Vergnaud et al., 2018). While Bangladesh and Brazil are clearly geographically well separated, it has previously been reported that most of the cattle imported into Brazil are from Europe or India (Minharro et al., 2013). This provides a plausible explanation The data of species and biovar identification and genetic characterization of Brucella field isolates of the present work may be useful to formulate policy and strategies for the control of bovine brucellosis in Bangladesh.

ACK N OWLED G EM ENTS
The authors thank Dr. Stephen M. Boyle (Virgi nia Tech, Blacksburg, VA, USA) for editorial comments on this manuscript.

CO N FLI C T O F I NTE R E S T
The authors declare no competing interest.

E TH I C A L S TATEM ENT
The protocol for field studies and collection of animal sample was approved by Animal Welfare and Ethical Committee, Bangladesh Agricultural University, Mymensingh-2202. Farmers were informed and their verbal consent was taken previously for the collection of samples from their animals.