Brucella species circulating in wildlife in Serengeti ecosystem, Tanzania

Background Brucellosis is a bacterial zoonosis of public health and economic importance world-wide. It affects a number of domestic animals, wildlife and humans. This study was carried out to determine circulating Brucella species in wildlife in Serengeti ecosystem using molecular techniques. Methodology A total of 189 samples including EDTA blood, serum and amniotic fluid from buffalos, lions, wildebeest, impala, zebra and hyena that were collected in relation to different cross-sectional studies conducted in the Serengeti ecosystem in Tanzania were used. Multiplex polymerase chain reaction AMOS-PCR and quantitative Real-Time PCR (qPCR) targeting the genus specific surface protein bcsp31 gene and the insertion sequence IS711 element downstream of the alkB gene for B. abortus and BMEI1162 gene for B. melitensis were employed on the samples. Results Results indicated that out of 189 samples examined, 12 (6.4%) and 22 (11.6%) contained Brucella DNA as detected by AMOS-PCR and qPCR, respectively. Most of the positive samples were from lions (52.6%) and buffaloes (19.6%). Other animals that were positive included wildebeest, impala, zebra and hyena. Out of 22 positive samples, 16 (66.7%) were identified as B. abortus and the rest were B. melitensis. Conclusion Detection of zoonotic Brucella species in wildlife suggests that livestock and humans at the interface areas where there is high interaction are at risk of acquiring the infection. Therefore, public education to interrupt risky transmission practices is needed. The findings also shed light on the transmission dynamics around interface areas and the role of wildlife in transmission and maintenance of Brucella infection in the region.


Introduction
Brucellosis is a bacterial zoonosis of public health and economic importance world-wide. It affects a number of livestock and wildlife species including humans [1]. The disease is a challenging public health problem to control in many developing countries including Tanzania, especially in pastoral and agro-pastoral farming systems [2][3][4]. According to WHO, brucellosis is one of the important re-emerging neglected tropical zoonosis [5] largely due to lack of public awareness.
In wild animals, brucellosis can be a result of spill-over from infected livestock or as a natural sustainable infection within susceptible wild animal populations [6,7]. Wild ungulates could acquire infection by ingesting contaminated pasture [7]. Flesh-eaters such as wolves and foxes are thought to be exposed through the ingestion of infected animals, placentae or aborted foetuses [8].
The disease has been reported in wild animals in some African countries, which include Kenya [9], South Africa [7], Zimbabwe [10] and Tanzania [11][12][13][14]. In the later country wild animals, Brucella infections have been reported in topi, buffalo, impala, Thompson gazelle and wildebeest [15,16]. However, most of these studies have been serology based without indication of circulating Brucella spp. Other studies reported brucellosis in livestock-wildlife interfaces in the Ngorongoro Conservation Area and Mikumi Selous Ecosystem [12][13][14].
In recent years many African countries have prioritized zoonotic diseases under the global Health Security Agenda. In many countries to date, brucellosis has been ranked among important zoonoses. In Tanzania in particular, it ranks among the top six priority zoonoses that the country will focus control efforts [17]. Since the prioritization of brucellosis in 2017, there have been efforts for development of a control strategy. In implementation of that strategy, critically highlighted areas include the pattern and contribution of different hosts in the transmission and maintenance of brucellosis in the country. Studies have been done on the livestock and shed light Final draft to be submitted to PLOS ONE on Brucella species circulating in different regions of the country [18,19]. However, wildlife transmission dynamics remain a grey area. The aim of this study was to determine circulating Brucella species in wildlife in Serengeti ecosystem in Tanzania, using molecular techniques.

Study area
The study used samples of cross-sectional studies conducted so far in Serengeti ecosystem in Tanzania. The Serengeti is the world's most diverse ecosystem, located in the in north-west of the country between the Ngorongoro highlands and Lake Victoria. This ecosystem comprises of Serengeti National Park, the Ngorongoro Conservation Area, Maswa Game Reserve, Loliondo Game Controlled Area and Kenya's Masai Mara National Reserve (Fig. 1). The study area was selected because there is notable interaction between wild animals, livestock and humans. This area is mainly inhabited by the Maasai with livestock keeping being their main economic activity [20]. The study employed AMOS PCR and a quantitative Real-Time PCR (qPCR) in the detection of Brucella spp. from the samples. The PCR protocols used were as described elsewhere [21][22][23]. In detail, at the Microbiology labs, college of veterinary medicine and biosciences in Sokoine University of Agriculture (SUA) Tanzania, samples were subjected to DNA extraction using a commercial DNA extraction kit (Zymo Research, USA Genomic DNA™ Tissue Mini Prep kit) and a method as described by Navarro et al. (2002) [24]. Briefly, 40µl of genomic lysis buffer was added to 200µl of the source sample. The mixture was subjected to digestion, deactivation, washing and elution steps as per manufacturer's instructions. Stock DNA samples were stored at -20 ºC until the performance of PCR. Primers used in this analysis were obtained from Bioline Inc (Taunton, MA, USA) and are detailed in Table 1.  Final draft to be submitted to PLOS ONE Hercules, Calif.). Positivity criteria of the assay required that a sample amplifies in both targets and below a set amplification cycle time (<38).  USA v. 2010) then descriptive and crude analytical statistics done using R software (25). Chisquare (χ 2 ) or Fisher's exact test were used as appropriate for comparison of; age, sex, location, animal species and Brucella PCR positivity were done. Population differences with a p value < 0.05 were considered as significant. Cross tabulation was used to determine the diagnostic sensitivity and specificity of the AMOS and real-time qPCR using the qPCR speciation assay as the reference test.

Ethical consideration
This study was conducted in conformity with the ethical and animal welfare guidelines stipulated by Sokoine University of Agriculture research ethics. A research permit was provided  (Table 3).  Table 4. Two samples; from buffalo and impala were found to harbor more than one Brucella species each. The species that were found in a single animal included B. melitensis, B. suis and B. abortus suggesting multiple infections.   [12,15,16,18,25]. Therefore, detection of pathogenic DNA in samples collected from wildlife in the area confirm that Brucella is circulating in the studied ecosystem.
Most of the Brucella positive samples were detected in female animals 30 (16.4%) different from the previous report in Tanzania [18] and elsewhere in Africa (3). The current finding further stresses the role of female animals in the transmission and maintenance of Brucella infection.
During abortion or normal birth there is massive shedding of Brucellaceae in the environment which are likely to be picked by susceptible animals during grazing [20].
It was further observed that Brucella DNA were detected more in lions (52.6%) and buffaloes (19.6%) than in other wild animal species. The findings could probably be because lions are indiscriminate carnivores and are likely to prey on Brucella infected animals like buffaloes.
However, the high detection rates observed in buffaloes may be due to B. abortus being the common species in the ecosystem and is known to mostly infect bovine ungulates. Generally, detection of zoonotic Brucella in wildlife in this study, point to their possible involvement as the source of sustained Brucella transmission in livestock and humans in the interface areas of Serengeti ecosystem. It has been earlier reported that wildlife can act as a source of infection for livestock and humans [25][26][27].

Final draft to be submitted to PLOS ONE
It was also observed that there was higher Brucella infection in Serengeti National Park than in the Ngorongoro Conservation Areas probably because Serengeti is a niche habitat of lions and buffaloes and hence contamination of environment is likely to be high [13]. In addition, in the Ngorongoro conservation area there are close interactions between livestock and wild animals [28].
However, it is unclear whether sustained Brucella infection in wildlife is acquired from livestock or vice versa. Nevertheless, vaccination in livestock may minimize the reverse spread of the disease from livestock to wildlife and vice versa [28].
Wildebeest seasonally migrate from Serengeti to Masai Mara for pastures [29], a practice likely to spread Brucella in the Serengeti ecosystem. Zebra constantly intermingle with wildebeest during grazing [29]; living together in close association and this behavior could be the basis for the transmission of the pathogens amongst the wild animals. Detection of B. suis in hyena could be explained by the scavenging behavior in this species.
In this study some animals were detected to have more than one of Brucella spp. B. abortus and B. melitensis, and B. abortus and B. suis were detected in one individual indicating occurrence of multiple infection as it has been reported before [30]. This study also reported the occurrence of B. melitensis and B. abortus in Impala. Studies have associated high sero-prevalences in antelope in Kafue flat area in Zambia because cattle were sharing source of water with wild animals during dry season [2,27]. Infection of B. melitensis is reported to be less common in sub-Saharan African countries [3133]. Brucella melitensis preferably infect sheep and goat which are related with impala [34].
In this study, qPCR was observed to have a higher detection rate of Brucella spp. than AMOS PCR. This finding is supported by reports from other studies which reported qPCR as superior tool [21,23,35,36]. Previous studies that have compared the two platforms and reported similar Final draft to be submitted to PLOS ONE performance in the detection of Brucella DNA [21,36]. This could probably be because AMOS PCR is limited in the detection of all Brucella spp. biotypes. Depending on which biotypes are predominant in the region, example B. abortus biovar 3 which has previously been detected in Tanzania [19] cannot be detected in AMOS PCR. Although the assay sensitively detected Brucella DNA in these archived samples, we did not have sufficient quantities and quality to confirm genomic material harvested from such archived samples.