Serum metabolomics profile identifies patients with community-acquired pneumonia infected by bacteria, fungi, and viruses

Purpose: Patients with bacterial, fungal, and viral community-acquired pneumonia (CAP) were studied to determine their metabolic profiles. Methods：Loop-mediated isothermal amplification technology and nucleic acid sequence-dependent amplification combined with microfluidic chip technology were applied to screen multiple pathogens from respiratory tract samples. Eighteen patients with single bacterial infection (B-CAP), fifteen with single virus infection (V-CAP), twenty with single fungal infection (F-CAP), and twenty controls were enrolled. UHPLC-MS/MS analysis of untargeted serum samples for metabolic profiles. Multiple linear regression and Spearman's rank correlation analysis were used to determine associations between metabolites and clinical parameters. The sensitivity and specificity of the screened metabolites were also examined, along with their area under the curve. Results：The metabolic signatures of patients with CAP infected by bacteria, viruses, and fungi were markedly different from those of controls. The abundances of 48, 57, and 81 metabolites were significantly unbalanced. Among these differential metabolites, 13, 14, and 30 were unique to the B-CAP, V-CAP, and F-CAP groups, respectively. Bacterial infections were the only known causes of disturbances in the pentose and glucuronate and aldarate and ascorbate metabolism interconversions metabolic pathway. Conclusions: Serum metabolomic techniques based on UHPLC-MS/MS may identify differences between individuals with CAP who have been infected by various pathogens, and they can also build a metabolite signature for early detection of the origin of infection and prompt care.研究目的：本研究针对细菌、真菌及病毒感染的社区获得性肺炎（community-acquired pneumonia, CAP）患者展开探究，以明确其代谢组特征。 研究方法：采用环介导等温扩增技术（loop-mediated isothermal amplification technology）联合核酸序列依赖性扩增技术与微流控芯片技术，对呼吸道样本中的多种病原体进行筛查。本研究共纳入18例单一细菌感染型社区获得性肺炎（B-CAP）患者、15例单一病毒感染型社区获得性肺炎（V-CAP）患者、20例单一真菌感染型社区获得性肺炎（F-CAP）患者及20例健康对照者。采用超高效液相色谱-串联质谱（ultra high performance liquid chromatography-tandem mass spectrometry, UHPLC-MS/MS）对血清样本进行非靶向代谢组学分析。通过多元线性回归与斯皮尔曼秩相关分析，明确代谢物与临床参数之间的关联；同时检测筛选出的代谢物的灵敏度、特异性及其曲线下面积（area under the curve, AUC）。 研究结果：细菌、病毒及真菌感染型社区获得性肺炎患者的代谢特征与健康对照者存在显著差异。其中，B-CAP、V-CAP及F-CAP组分别有48、57及81种代谢物的丰度呈现显著失衡。在上述差异代谢物中，分别有13、14及30种为对应组别所特有。细菌感染是目前已知唯一可干扰戊糖与葡糖醛酸、半乳糖醛酸及抗坏血酸代谢相互转化通路的病因。 研究结论：基于超高效液相色谱-串联质谱（UHPLC-MS/MS）的血清代谢组学技术可有效区分不同病原体感染的社区获得性肺炎患者，同时可构建特异性代谢物特征谱，用于早期识别感染病原体类型并指导及时诊疗。