High-Resolution Plasma Metabolomic Profiling to Identify Biomarkers for Tuberculosis Disease and Response to Therapy

结核病（TB）是全球头号传染病杀手， 2014年因病死亡。由于缺乏生物学知识， 与TB发作、进展和消退相关的途径和预测性生物标志物。的改进的诊断 结核病检测将对全球控制结核病和降低死亡率产生重大影响。不幸的是， 没有经过验证的结核病发作、进展和消退的生物标志物; 生物标志物是迫切需要的。目前的超高分辨率代谢组学（HRM）方法具有相当大的优势。 为结核病相关生物标志物的发展提供了希望。几项研究发现， 活动性分枝杆菌感染者血液和其他生物样本中的分子代谢谱 结核病（Mtb）诱导的结核病与未感染的对照相比，但没有研究探讨是否 代谢组学可预测结核病结果，即痰中结核分枝杆菌清除率。此外，迄今为止的所有研究都使用了 低分辨率代谢组学方法。质谱仪机器能力的进步，加上 最近先进的数据提取/分析方法已经大大增加了 通过HRM在生物样品中进行代谢物检测，目前已超过20，000种（> 100，000种离子）。我们 HRM处理方案提供了检测非常低丰度代谢物的能力，包括Mtb细胞 包膜糖脂，并探索人体代谢途径的调节。在试点研究中，我们成功地 鉴定了61种血浆代谢物，用于区分活动性肺结核成人患者和家庭接触者 没有TB。分化的物种包括特定的结核分枝杆菌来源的细胞壁糖脂和内源性脂质 介体分解素。我们最近的试验数据显示，特定的代谢物和人体代谢途径， 包括参与药物、氨基酸和脂质代谢的那些，与以下倾向有关： 痰Mtb培养物随时间清除。这一探索性提议的最终目标是获得新颖的数据 这可能会导致新的结核病生物标志物的发展。我们假设血浆HRM分析可以：1） 通过鉴定结核分枝杆菌衍生代谢物和内源性代谢物预测成功抗结核治疗的倾向 （生物标志物）和与痰Mtb清除相关的人代谢途径（病理生理学）;和2） 成功区分活动性TB患者与无潜伏性TB感染（LTBI）或TB疾病的对照组。 为了验证这些假设，我们在这个探索性的R21项目中提出了以下具体目标： 具体目标1：确定血浆HRM分析是否可以鉴别结核分枝杆菌来源和内源性 预测痰中Mtb清除的代谢物（从阳性培养物变为阴性培养物）。 具体目标2：确定血浆HRM分析是否可以区分成人药物依赖性疾病。 敏感或MDR-TB ± HIV合并感染，无LTBI的无症状对照。

Summary/abstract: Tuberculosis (TB) is the number one infectious disease killer in the world with 1.5 million deaths attributable to the disease in 2014. A critical knowledge gap exists given the lack of biological pathways and predictive biomarkers related to TB onset, progression and resolution. An improved diagnostic test for TB would have a major impact on global control of the disease and reduce mortality. Unfortunately, there are no validated biomarkers for TB onset, progression and resolution; the development of a TB biomarker is urgently needed. Current ultra-high-resolution metabolomics (HRM) methods have considerable promise for the development of TB-associated biomarkers. Several studies have identified distinguishing small molecule metabolic profiles in blood and other bio samples in individuals with active Mycobacterium tuberculosis (Mtb)-induced TB disease compared to uninfected controls, but no studies have explored whether metabolomics is predictive of TB outcomes, Mtb clearance from sputum. Further, all studies to date have used low-resolution metabolomics methods. Advances in mass spectrometer machine capability, coupled with recent advanced data extraction/analysis methods has considerably increased the dynamic range of metabolite detection by HRM in biologic samples, which now exceeds 20,000 species (>100,000 ions). Our HRM processing scheme provides capability to detect very low abundance metabolites, including Mtb cell envelope glycolipids, and explore regulation of human metabolic pathways. In a pilot study, we successfully identified 61 plasma metabolites that differentiated adults with active pulmonary TB from household contacts without TB. Differentiating species included specific Mtb-derived cell wall glycolipids and endogenous lipid mediator resolvins. Our recent pilot data shows that specific metabolites and human metabolic pathways, including those involved in drug, amino acid, and lipid metabolism, are associated with the propensity for sputum Mtb culture clearance over time. The ultimate goal of this exploratory proposal is to obtain novel data that may lead to the development of new TB biomarkers. We hypothesize that plasma HRM analysis can: 1) predict the propensity for successful anti-TB treatment by identifying Mtb-derived and endogenous metabolites (biomarkers) and human metabolic pathways (pathophysiology) associated with sputum Mtb clearance; and 2) successfully differentiate patients with active TB from controls without latent TB infection (LTBI) or TB disease. To test these hypotheses, we propose the following Specific Aims in this exploratory R21 project: Specific Aim 1: To determine whether plasma HRM analysis can identify Mtb-derived and endogenous metabolites that predict clearance of Mtb from sputum (change from positive to negative culture). Specific Aim 2: To determine whether plasma HRM analysis can differentiate adults with drug- sensitive or MDR-TB ± HIV co-infection from asymptomatic controls without LTBI.