Molecular characterization of Brucella species detected in humans and domestic ruminants of pastoral areas in Kagera ecosystem, Tanzania

Abstract Brucellosis is a zoonotic disease of importance to both public health and the livestock industry. The disease is likely to be endemic in Tanzania and little is reported on molecular characterization of Brucella species in pastoral settings. This study aimed at characterizing Brucella species (targeting genus Brucella) infecting humans, cattle and goat in Kagera region (Ngara and Karagwe districts) using real‐time PCR, PCR amplification of 16S rRNA genes and Sanger sequencing. Brucella spp. were detected in 47 samples (19 sera and 28 milk) out of 125 samples (77 sera, 35 milk and 13 aborted materials) using real‐time PCR. All aborted materials (13 samples) were negative to real‐time PCR. Out of the 47 real‐time PCR positive samples (28 milk and 19 sera), 20 samples (10 milk and 10 sera) showed an expected 16S rRNA gene PCR product. Sequence analysis and blasting confirmed the presence of Brucella spp. in pastoral areas of Kagera region. The Brucella spp. from Kagera were phylogenetically grouped in two clades and three branches all closer to B. melitensis, B. abortus and B. suis from USA, Sudan and Iran. However, they were distinct from other species isolated also in USA, New Zealand, Germany and Egypt. This was expected based on the distance between the geographical regions from which the data (nucleotides sequences from 16S gene sequencing) for the phylogeny reconstruction were obtained. This is the first study to report Brucella species identified using 16S rRNA gene sequencing in East and Central Africa. A livestock vaccination program re‐inforced with a high index of Brucella diagnosis is needed to eradicate brucellosis in animals and minimize suffering from Brucella infections in humans in Tanzania.

However, due to the variations which can be exhibited in the con- This study aimed at molecular characterization of Brucella species in humans and livestock in pastoral areas of Kagera in Tanzania.
Real-time PCR was used on sera, aborted materials (from women, cattle, goats and sheep) and milk taken from cattle and goats. The 16S rRNA gene was amplified on positive samples from Real-time PCR: sera (from women, cattle and goats) and milk (from cattle and goats) while Sanger sequencing was done using the fragment amplified from the 16S rRNA gene.

| Study design
The samples were collected during previous cross-sectional and prospective cohort studies conducted in Ngara and Karagwe districts  (Ntirandekura, Matemba, Kimera, Muma, & Karimuribo, 2020). Five millilitres of venous blood were firstly taken from humans going for malaria checking (July 2017) (personal communication) and from cattle, goats and sheep in the same villages. Secondly, 5ml of venous blood were taken from pregnant women attending antenatal care (November 2017-April 2018 and gravid ruminants in the same pastoral area (Ntirandekura et al., 2020). Each blood samples reacting to the rose Bengal, C-Elisa and FPA tests was aliquoted and kept at −20°C for molecular analysis. For the prospective study, aborted materials (from women and ruminants) and milk (from cattle, goat and sheep) were collected and stored at −20°C until the DNA extraction. During the two studies, structured questionnaires were used to complete the information on brucellosis status in the study area. Domestic ruminants were apparently healthy.
This study used sera which were positive for Brucella by the previous serological tests (RBPT, c-Elisa and FPA), milk samples and aborted materials (Table 1).
The objective for this study was to characterize Brucella spp.
using 16S rRNA gene sequencing. However, 16S rRNA PCR would have given too many samples for sequencing which is rather too costly. Therefore, after the DNA extraction, a real-Time PCR (targeting genus Brucella) was first used to screen for Brucella positive samples only, then continued with a 16S rRNA PCR and sequencing.

| DNA extraction
Genomic DNA from serum samples and aborted materials were extracted using the QIAamp DNA Mini Kit (Qiagen kit Germany) according to the manufacturer's instructions. To obtain genomic DNA from milk, the samples were centrifuged for 10 min at 10,000 × g, following which the supernatant was discarded. The GeneJET Genomic DNA Purification Kit (Thermofisher Scientific-K0721) was used to extract DNA from the pellet, according to the manufacturer's instructions.

| Real-time PCR for Brucella spp
One hundred twenty-five samples (Table 1) were subjected to real-time PCR which was performed on the PikoReal machine (Thermo Fisher Scientific) for Brucella spp. detection. Targeting Vdcc gene, the forward GTGGCGATCTTGTCCG and the reverse ACGGCGATGGATTTCCG Brucella spp. specific primers were used (Winchell, Wolff, Tiller, Bowen, & Hoffmaster, 2010). A vaccine Brucella strain S19 was used as a positive control. The final reaction volume was 25 µl consisting of 2.5 µl DNA template, 1x of RealQ Plus 2x Master Mix Green (Low Rox), 10 µm of each primer, 10 µm of probe (5' FAM-AAATCTTCCACCTTGCCCTTGCCATCA-BHQ 3') and 5.5 µl of PCR grade water. The PCR reaction started with initial heating at 50°C for 2 min, then at 95°C for 7 min, followed by 35 cycles at 95°C for 5 s and 60°C for 30 s. Data were acquired at 60°C (the extension step).  Sequencing was done in duplicate for each primer. The sequencing conditions included incubation at 96°C for 1 min, followed by 25 cycles of denaturation at 96°C for 10 s, 50°C for 5 s and 60°C

| PCR Amplification Of 16S rRNA genes
for 4 min (Bio-Rad) on a heated lid. Following the cycle-sequencing protocol, the reactions were cleaned up by ethanol precipitation. In this procedure, 5 µl of freshly prepared 125 mM EDTA and 60 µl of 100% ethanol were added to each reaction tube containing the sequencing products. After vortexing, the mixture was incubated in the dark for 15 min at room temperature to precipitate the extension products. As the BigDye® reagent is light-sensitive, the precipitation was carried out in the dark. Following precipitation, the tubes were centrifuged at 13,000 × g for 30 min and the supernatant was discarded without disturbing the pellet. Subsequently, the pellets were washed with 60 µl of 70% ethanol and centrifuged at 13,000 × g for 30 min. After the supernatant had been removed, the pellets were shaded from direct light and dried in a vacuum drier until no ethanol was present. Before loading onto the ABI 3,730 DNA Analyser the samples were re-suspended in 20 µl of HiDi Formamide (Life Technologies) and analysis was done according to the manufacturer's instructions.

| Data analysis
Similarity searches for the 16S rRNA gene sequences obtained using Applied Biosystems 3,500 genetic analyzer (Thermo Fisher Scientific) were done using the blastn (NCBI) in GenBank databases.  (Kimura, 1980). All analyses were done in MEGA X (Kumar et al., 2018).

| PCR amplification of 16S rRNA genes
The DNA fragment of Brucella species amplified from 16S rRNA gene is of different size according to the primers used during the amplification. In this study, we used primers (according to Bricker et al., 2003) targeting a band size of 800 base pairs. An expected 800 bp PCR product was amplified in 20 out of 47 (Table 2)

| 16S rRNA gene sequencing and phylogeny reconstruction
Sanger sequencing was successful for 10 out of the 20 samples ( intermedium (AJ867325.1) deposited in GenBank by Lebhun et al. (2000) from Germany.
The evolutionary history was inferred using the Maximum Likelihood method and the Kimura 2-parameter model (Kimura, 1980). Initial tree(s) for the heuristic search were obtained by ap-   (Assenga et al., 2015;Mathew et al., 2015;Kassuku, 2017).

| D ISCUSS I ON
We detected Brucella species in cow milk, which is similar to previous reports in Egypt (Wareth, Melzer, Elschner, Neubauer, & Roesler, 2014), in Uganda (Hoffman et al., 2016) and in Tanzania (Assenga et al., 2015;Mathew et al., 2015). This calls for public health attention since some residents in the study area drink unboiled milk as it was reported previously (Ntirandekura, Matemba, Ngowi, Kimera, & Karimuribo, 2018a). We are also reporting the presence of Brucella in goat milk. This is in contrast to a previous study in Katavi region which could not detect any Brucella spp. in milk from goat (Assenga et al., 2015). However, Brucella abortus was detected in serum from goats in Morogoro region (Kassuku, 2017).
Even though people from the study area do not consume goat milk, brucellosis in goats still poses a risk of spill over to cattle and humans in a livestock pastoral setting.
Brucellosis is reported to be associated with abortions in humans and animals (Khan, Mah, & Memish, 2001;Kurdoglu, Cetin, Kurdoglu, & Akdeniz, 2015;Mathew, 2017;Megersa et al., 2011;Muma, Godfroid, Samui, & Skjerve, 2007;Nigro et al., 2011). This study detected Brucella in sera from abortive woman, cow and goat. Although there was a failure to detect Brucella from a limited sample size of aborted materials, the detection of Brucella from sera of aborted individuals could raise the same suspicions regarding the contribution of brucellosis to reproductive failure in this pastoral area as reported earlier (Ntirandekura et al., 2020). The failure to detect Brucella species in aborted materials could have been associated to the transport medium used (liquid nitrogen), long time for conservation of samples before analysis (9 months) and the low DNA concentration among other factors.
However, other factors associated with reproductive failures need to be ruled out for a causal relationship between brucellosis and abortions to hold (Ntirandekura, Matemba, Ngowi, et al., 2018).
NtirandekurReal-time PCR positive samples were used for PCR amplification of 16S rRNA genes to get an 800 bp fragment as reported earlier (Gee et al., 2004;Khan et al., 2018 (Corbel, 2006). Hence, for phylogeny reconstruction we retrieved 16S rRNA gene sequences for Brucella species from GenBank. These sequences belonged to Brucella spp. reported from USA, Germany, Iran, Sudan and Egypt.
We could not find 16S rRNA gene sequences for Brucella spp. isolated from East and central African regions in the DNA databases.
Studies that identified Brucella in the region by sequencing targeted other genes (Hoffman et al., 2016;Mathew, 2017;Mugizi et al., 2015). To the best of our knowledge, this is the first study to the species circulating in the same region. However, all clades with