Molecular characterization of Mycobacterium bovis infection in cattle and buffalo in Amazon Region, Brazil

Abstract The aim of this study was to characterize Mycobacterium bovis from cattle and buffalo tissue samples, from two Brazilian states, and to analyse their genetic diversity by spoligotyping. Tissue samples from tuberculosis suspect animals, 57 in Amazonas State (12 cattle and 45 buffaloes) and six from Pará State (5 cattle and one buffalo) from slaughterhouses under State Veterinary Inspection, were isolated in culture medium Stonebrink. The positive cultures were confirmed by PCR and analysed by the spoligotyping technique and the patterns (spoligotypes) were identified and compared at the Mycobacterium bovis Spoligotype Database (http://www.mbovis.org/). There was bacterial growth in 44 (69.8%) of the tissues of the 63 animals, of which PCR for region of differentiation 4 identified 35/44 (79.5%) as Mycobacterium bovis. Six different spoligotypes were identified among the 35 Mycobacterium bovis isolates, of which SB0295, SB1869, SB0121 and SB1800 had already been described in Brazil, and SB0822 and SB1608 had not been described. The most frequent spoligotype in this study (SB0822) had already been described in buffaloes in Colombia, a neighbouring country of Amazonas state. The other identified spoligotypes were also described in other South American countries, such as Argentina and Venezuela, and described in the Brazilian states of Rio Grande do Sul, Santa Catarina, São Paulo, Minas Gerais, Mato Grosso do Sul, Mato Grosso and Goiás, indicating an active movement of Mycobacterium bovis strains within Brazil.

Bovine tuberculosis predominantly affects cattle and buffaloes, but may occasionally infect other mammalian species, including humans (Etchechoury et al., 2010;Silva et al., 2018). It may spread through direct contact with infected animals, causing the spread of disease among herds or herds to wild animals and vice versa (Corner, Murphy, & Gormley, 2011;Reis, 2015), or being transmitted through indirect contact with contaminated equipment, water and food (Palmer, 2013;Smith, Tauer, Schukken, Lu, & Grohn, 2013).
Globally recognized, BTB persists in both developed and developing countries (Broughan et al., 2013;Michel, Müller, & Helden, 2010). In Brazil, the Ministry of Livestock and Food Supply was established in 2001 and modified in 2017 the National Program for the Control and Eradication of Brucellosis and Animal Tuberculosis, to reduce the prevalence and incidence of BTB (Carneiro & Kaneene, 2018). The regulation determines the slaughter of all bovines and buffaloes that present a positive reaction to the tuberculin test (ante mortem diagnosis) and as gold standard, isolation in culture medium for identification and confirmation of M. bovis infection (post-diagnosis-mortem) (Brasil, 2017).
Molecular techniques are increasingly used to support conventional methods, both for the identification and confirmation of M.
bovis strains, and for molecular epidemiology. Molecular genotyping by spoligotyping is a technique developed by Kamerbeek et al. (1997) which discriminates genotypes of M. bovis by amplification of the polymorphic DR (Direct Repeats) chromosomal locus in the MTC which contains DR sequences interspersed with variable spacer sequences, followed by a reversed line blot hybridization (RLBH). The presence or absence of the spacers is identified. Spoligotyping, by discriminating M. bovis genotypes, through RLBH patterns, may aid in BTB control programs, providing epidemiological data among isolates. Therefore, this study was designed to determine the spoligotypes of M. bovis isolates from cattle and buffalos in the Amazon region of Brazil.

| Sample collection
From July 2016 to February 2017, a total of 922 animals were inspected (635 cattle and 287 buffaloes), from those 63 samples of cattle (n = 17) and buffalo (n = 46) tissues were obtained. From the herds with report of TST reactive animal samples of all animals sent to the slaughterhouses, with or without lesions suggestive of tuberculosis (LST), were collected. From herds with unknown TST status samples were collected only from animals with LST ( Figure 1). Three abattoirs were selected based on logistics and willing to participate. The inspection of the animals was performed by trained officials of Amazonas State Veterinary Inspection Service, LST were defined as granulomas small, spherical, tan and firm nodules usually with a mineralized core. The same criteria for detection of lesions were used for cattle and buffaloes.
Two samples per animal were collected (one from the suspected tissue and one from the retropharyngeal lymph node) and the unit of analysis was the animal. For the analysis, herd was considered infected when it presented at least one animal confirmed positive by the PCR analysis. The animals were slaughtered for commercial purposes. Thus, there was no animal sacrifice due to this study.
Herds from 10 municipalities of Amazonas and four municipalities in Para were involved (Table 1). The median age group of inspected animals in both species was from 25 to 36 months old, the mean herd size was 142 for cattle and 84 for buffaloes. Unfortunately, we have only data available from Amazonas State, the municipalities on the study have 3,818 cattle herds and 1,315 buffalo herds.
The samples were shipped in a sterile plastic packages containing the Animal Transit Guide number, which has info about animal species, sex and age but no information about tissue and race.

| Preparation and culture of samples
Lesions suggestive of tuberculosis (10-25 mg) were macerated in 2 ml tubes containing ceramic beads (MagNA Lyser green beads) and 1 ml of sterile water in a MagNA Lyser Instrument (Roche) for three cycles of 30 s at 6.000 rpm. Later, 1 ml of 1N NaOH was added, and the tube was incubated at 37°C for 15 min. The tube was centrifuged at 3,000 rpm for 15 min, and the supernatant was discarded. The decontamination by Petroff method was performed. Briefly, the pellet was suspended in 1 ml of sterile distilled water and 100 µl of 0.2% phenol red solution was added.
After that, 50-100 µl of 1% HCl was added until change of colour was visualized-from pink to amber yellow. The pH was adjusted to 7.0 with neutralizing solution and 300 µl of the material was inoculated in duplicate into the Stonebrink medium (de Kantor et al., 1998). The Stonebrink medium has the same composition as Lowenstein-Jensen, except that glycerol is replaced by 0.5% sodium pyruvate, further incubated at 37°C, and evaluated weekly for 90 days to verify bacterial growth. One medium per sample was used. The colonies with characteristics suggestive of M. bovis were submitted to DNA extraction.

| DNA extraction
The bacterial colonies were washed with 500 μl of Tris-EDTA buffer in microtubes and inactivated in a dry bath for 1 hr at 87°C, with subsequent centrifugation at 14,000 rpm for 2 min. The pellet that formed was discarded and the supernatant containing the mycobacterial DNA was transferred to new microtubes and stored at −20°C for subsequent analysis.
This DNA extraction method has been reported by our labora-

| Spoligotyping
The spoligotyping was performed on M. bovis isolates, following the instructions of Kamerbeek et al. (1997). Hybridization of the PCR product was performed on a spoligotyping membrane with oligonucleotides of spacer sequences, using a miniblotter according to the manufacturer's instructions (MapMyGenome). The membrane was incubated with streptavidin-peroxidase and the spacers were

| RE SULTS AND D ISCUSS I ON
The study was based on a convenience sampling of adult animals sent to three major slaughterhouses in Amazonas State, those slaughterhouses represent more than 70% of all regional cattle and buffaloes slaughtered on the region. All herds for the two species, but one, on this study came from the Geographical region known as hand, during the Rainy season, although less exposed to BTB transmission between herds, the probability of transmission within the herd can increase due to less favourable offer of pasture and proximity of animals during dairy offer of food supplementation.
The regional herds, cattle and buffaloes, are mainly managed in extensive system characterized by farms with low technological level and productivity and few managed in semi-intensive systems characterized by farms with good technological level and productivity.
In common, the two systems have the influence of the river floods.   Other spoligotypes considered unusual were observed in this study, as SB1869, which had already been described in São Paulo by Rocha (Rocha, 2013), was identified in five buffaloes (14.3%) in in Autazes/Novo Ceu district (Table 1). Also considered unusual, SB1608 and SB1800 spoligotypes were isolated in this study, the spoligotype SB1608 isolated in buffalo in the municipality of Parintins had not previously been described in Brazil and was described in wild animals in Portugal (Carvalho et al., 2016), SB1800 was described in Brazil (www.mbovis.org) and identified in cattle in Apuí-AM (Table 1).
SB0295 was the second most frequent spoligotype observed, with nine buffaloes in Amazonas, and was most frequently found in animals without LST (7/9). This spoligotype is considered the second most frequent in Brazil (Zumárraga et al., 2013), described in the state of Paraíba (Higino et al., 2011), Bahia (Filho et al., 2014, Mato Grosso, Goiás (Carvalho et al., 2016), Mato Grosso do Sul, Santa Catarina (Parreiras et al., 2012) (Reis, 2015) and France (Hauer et al., 2015). In addition, the spoligotype SB0121 was identified as an agent of human tuberculosis in England (Stone, Brown, & Drobniewski, 2011)  and SB1869 (lack of spacers N°1 and 2). The spoligotype SB0121 could be the founder of SB0822 and SB1800; however, to explain their relationship, the presence of other spoligotypes not detected in this study should be considered (Figure 2).

| CON CLUS IONS
• A high genetic diversity of M. bovis isolates were found in the Brazilian Amazon.
• The data corroborate with the previous information that buffaloes are more infected than cattle in the region.
• Genotype isolates in this study were reported in neighbouring countries suggesting the need for more studies to clarify the routes of transmission between regions.

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

E TH I C S S TATEM ENT
The authors confirm that the ethical policies of the journal, as noted on the journal's author guidelines page, have been adhered to and the appropriate ethical review committee approval has been received. The US National Research Council's guidelines for the Care and Use of Laboratory Animals were followed.