Computational models of naturally acquired immunity to falciparum malaria

项目摘要/摘要 对疟疾的免疫是复杂的，随着时间的推移，涉及到免疫之间的良好相互作用。最早的 了解免疫力发展的努力仅限于一组狭窄的测量或 简化论的动物或人类挑战模型未能捕捉到反复感染的复杂性 自然暴露的个体。我们建议对先天和适应性进行综合评估和建模 随着时间的推移，对反复感染恶性疟原虫(PF)的免疫反应。这个项目利用了一个独特的 乌干达的疟疾队列研究，参与者每月在我们的诊所接受检查，并针对所有疾病，使我们能够 捕捉有症状和无症状的感染。通过利用我们特征良好的队列，详细 宿主反应的免疫学特征，以及最先进的免疫计算模型，我们 将1)全面描述对有症状和无症状P. 恶性疟原虫感染。我们假设有症状但不是无症状的感染将是 以先天和适应性炎症反应减弱为特征的。我们将介绍先天的 儿童在多个时间点对有症状和无症状感染的适应性免疫反应 在PF感染后的几周内。数据来自转录图谱，深层细胞表型，抗体 分析和刺激分析将用于构建灵活的计算模型，捕获交互 在免疫系统的不同部分之间和免疫反应的轨迹之间 单一感染。2)确定免疫状态如何演变以响应复发的恶性疟原虫 感染。我们假设，反复感染将导致免疫状态从偏向的 对一种以更稳定、更调节的状态和 获得功能性抗体。我们将模拟已确定的关键免疫学参数的演变 在目标1中，随着抗寄生虫体液和细胞功能的检测，多年反复感染和 通过在两年的时间内生成纵向数据，跨越各个年龄段。这一目标与目标1相辅相成， 重要信息用于定义来自累积感染的免疫反应的紧急特性 更长的时间尺度，跨越免疫获得的时期。3)确定免疫状态的关键方面 导致对恶性疟原虫感染的抗寄生虫和抗病免疫。我们假设 功能性抗体反应将与抗寄生虫免疫最密切相关，而这种减弱 与生俱来的反应将与抗病免疫最密切相关。由AIMS的调查结果指导 1和2，我们将开发计算模型来确定临床免疫表型的关键决定因素， 通过评估随后一年感染的临床结果而获得。所有型号都将经过验证 并使用来自独立个人的数据、外部数据和基于实验室的数据进行迭代优化 实验。数据和模型将通过适当的公共储存库提供和查找。

PROJECT SUMMARY/ABSTRACT Immunity to malaria is complex, involving a fine interplay between immune compartments over time. Most prior efforts to understand the development of immunity have been limited to a narrow set of measurements or reductionist animal or human challenge models that fail to capture the complexity of repeated infection in naturally exposed individuals. We propose to comprehensively evaluate and model the innate and adaptive immune response to repeated P. falciparum (Pf) infections over time. This project takes advantage of a unique malaria cohort study in Uganda, with participants seen in our clinic monthly and for all illnesses, allowing us to capture both symptomatic and asymptomatic infections. By leveraging our well-characterized cohort, detailed immunological characterization of host responses, and state-of-the-art computational models of immunity, we will 1) Comprehensively characterize the immune response to symptomatic and asymptomatic P. falciparum infections. We hypothesize that symptomatic – but not asymptomatic – infections will be characterized by an attenuation of the innate and adaptive inflammatory response. We will profile the innate and adaptive immune response to symptomatic and asymptomatic infections in children at multiple time points in the weeks following Pf infection. Data from transcriptional profiling, deep cellular phenotyping, antibody profiling, and stimulation assays will be used to build flexible computational models, capturing interactions between different compartments of the immune system and the trajectory of the immune response after a single infection. 2) Determine how the immune state evolves in response to recurrent P. falciparum infections. We hypothesize that recurrent infection will result in a shift of the immune state from one biased towards dynamic, inflammatory immune responses to one characterized by a more stable, regulatory state and the acquisition of functional antibodies. We will model the evolution of key immunological parameters identified in Aim 1, along with assays of anti-parasitic humoral and cellular function, over years of repeated infection and across ages by generating longitudinal data over a period of 2 years. This aim complements Aim 1 in providing important information to define emergent properties of the immune response from cumulative infections over longer time scales, spanning the period of immune acquisition. 3) Identify key aspects of the immune state leading to anti-parasite and anti-disease immunity to P. falciparum infection. We hypothesize that functional antibody responses will be most strongly associated with anti-parasite immunity, and that attenuation of innate responses will be most strongly associated with anti-disease immunity. Guided by findings from Aims 1 and 2, we will develop computational models to identify the key determinants of clinical immune phenotypes, obtained by evaluating the clinical outcomes of infection over the subsequent year. All models will be validated and iteratively refined using data from independent individuals, external data and laboratory-based experiments. Data and models will be made available and findable through appropriate public repositories.