Brucellosis in Kenya - understanding epidemiology and informing control at regional scales

粮食安全和畜牧业生产力是肯尼亚政府的战略目标。布鲁氏菌病是撒哈拉以南非洲最重要的动物疾病之一，是实现这些目标的主要障碍，因为它降低了牛、绵羊、山羊和骆驼等多种牲畜的生育力和产奶量。在肯尼亚，有证据表明多种牲畜广泛接触布鲁氏菌，并且已知它是人类疾病的一个重要原因。已有针对布鲁氏菌的疫苗，并且有针对性地设计和实施干预措施已被证明可以减少受该疾病影响最严重的贫困社区中布鲁氏菌病的影响。然而，布鲁氏菌病是由几种不同的布鲁氏菌引起的，每种布鲁氏菌都具有复杂的多宿主流行病学，目前的疫苗是针对宿主和病原体物种的。通常在牛中存在abortus，在绵羊和山羊中存在melitensis，但在撒哈拉以南非洲，据报道在牛、绵羊和山羊等多种宿主中都存在abortus和melitensis。准确了解谁感染了谁，用什么布鲁氏菌种类感染了谁，对于为设计具有成本效益的控制布鲁氏菌病规划提供信息至关重要，但这仍然是包括肯尼亚在内的大多数低收入和中等收入国家的一个主要知识差距。实现布鲁氏菌病控制的重大挑战之一是血清学诊断测试无法区分由不同布鲁氏菌种类引起的感染，以及缺乏能够准确识别感染布鲁氏菌种类的基于分子或培养的诊断数据。对肯尼亚来说，这意味着我们不知道牛、骆驼、绵羊或山羊是否感染了melitensis、B. abortus或两者兼而有之。该项目将利用现有的样本收集和血清学数据，以解决我们对布鲁氏菌病多宿主传播动力学的理解中的关键知识空白。具体而言，我们将：1)在KEMRI建立qPCR检测布鲁氏菌及其种类类型，并应用该方法测试肯尼亚牲畜样本的现有档案。将使用qPCR检测200多只牲畜和另外400多只骆驼样本的档案，以确定感染布鲁氏菌的种类。2)对现有的多物种血清学数据进行建模，量化宿主物种之间的传播动态及其发生的空间尺度。将开发贝叶斯潜过程模型并应用于利用2012-2013年在肯尼亚三个不同地区进行的横断面调查获得的血液学数据重建布鲁氏菌传播动力学。3)将研究成果传达给包括决策者在内的利益相关者，以促进肯尼亚布鲁氏菌病控制策略的设计。通过研究会议和研讨会，将与当地合作伙伴分享本项目产生的见解，以设计肯尼亚的国家布鲁氏菌病控制政策。该项目建立并扩展了RCUK和DFiD资助的坦桑尼亚北部布鲁氏菌病分子流行病学研究（BB/L018845/1）。先前的研究结果表明，在坦桑尼亚北部存在着羊分枝杆菌、流产分枝杆菌和鹅分枝杆菌，羊分枝杆菌是导致人类布鲁氏菌病的主要人畜共患布鲁氏菌种，绵羊和山羊是人类和牛最有可能接触布鲁氏菌的动物来源。通过将数据资源和分析专业知识的新组合联系起来，该项目为扩展以前开发的分子诊断和建模方法提供了机会。我们的项目将填补肯尼亚关于布鲁氏菌病流行病学的关键知识空白。重要的是，它还将使共同方法能够应用于邻国坦桑尼亚和肯尼亚的数据集，提供一种机制来提高对布鲁氏菌流行病学更广泛的区域模式和变化的理解

Food security and livestock productivity are strategic objectives of the Government of Kenya. Brucellosis, one of the highest priority animal diseases in sub-Saharan Africa, is a major impediment to achieving these objectives as it reduces the fertility and milk production of multiple livestock species, including cattle, sheep, goats and camels. In Kenya, there is evidence of widespread exposure to Brucella in multiple livestock species and it is known to be an important cause of human illness. Vaccines against Brucella exist and targeted design and implementation of interventions have been shown to reduce the impacts of brucellosis in the poor communities most affected by this disease. However, brucellosis is caused by several different Brucella species, each of which has complex multi-host epidemiology and current vaccines are host and pathogen species specific. B. abortus is typically maintained in cattle and B. melitensis in sheep and goats, but in sub-Saharan Africa both B. abortus, B. melitensis have been reported in a variety of hosts including cattle, sheep and goats. Accurate understanding of who infects who, with what Brucella species is critical to inform the design of cost-effective programmes for the control of brucellosis but this remains a major knowledge gap in most low and middle income countries, including Kenya.One of the big challenges in achieving brucellosis control has been the inability of serological diagnostic tests to differentiate infections caused by different Brucella species and the scarcity of molecular or culture based diagnostic data that enable accurate identification of the infecting Brucella species. For Kenya, this means that we do not know whether cattle, camels, sheep or goats, are infected with B. melitensis, with B. abortus or with both. This project will exploit existing sample collections and serological data to address key knowledge gaps in our understanding of the multi host transmission dynamics of brucellosis. Specifically, we will:1) Establish qPCR assays to detect and species-type Brucella at KEMRI and apply this methodology to test an existing archive of Kenyan livestock samples. An archive of 200+ livestock and another 400+ camel samples will be tested using qPCR to identify the infecting Brucella species. 2) Model existing multi-species serological data to quantify transmission dynamics between host species and the spatial scales at which they occur. Bayesian latent process models will be developed and applied to reconstruct Brucella transmission dynamics using serology data obtained from a cross-sectional survey conducted in 2012-2013 in three distinct regions of Kenya. 3) Communicate research findings to stakeholders including policy makers to facilitate the design of brucellosis control strategies in Kenya. Through research meetings and a workshop the insights produced in this project will be shared with local partners to design national brucellosis control policy in Kenya.This project builds from and extends the RCUK and DFiD funded study The Molecular Epidemiology of Brucellosis in Northern Tanzania (BB/L018845/1). The results from this previous study indicate that B. melitensis, B. abortus and B. ovis are present in northern Tanzania, that B. melitensis is the predominant zoonotic Brucella species causing human brucellosis and that sheep & goats are the most likely animal source of Brucella exposure in both humans and cattle. By linking new combinations of data resources and analytical expertise, this project offers opportunities to extend molecular diagnostic and modelling approaches developed previously. Our project will fill critical knowledge gaps about brucellosis epidemiology in Kenya. Importantly, it will also enable application of common approaches to datasets from neighbouring Tanzania and Kenya, offering a mechanism to improve understanding of broader scale regional patterns and variation in Brucella epidemiology