Zihui Li1, Boping Du1, Jing Li2, Jinli Zhang2, Xiaojing Zheng1, Hongyan Jia1, Aiying Xing1, Qi Sun1, Fei Liu1, Zongde Zhang3. 1. Beijing Chest Hospital, Capital Medical University, Beijing Tuberculosis and Thoracic Tumor Research Institute, Beijing Key Laboratory for Drug Resistant Tuberculosis Research, Beijing 101149, China. 2. People's Liberation Army No. 263 Hospital, Beijing 101149, China. 3. Beijing Chest Hospital, Capital Medical University, Beijing Tuberculosis and Thoracic Tumor Research Institute, Beijing Key Laboratory for Drug Resistant Tuberculosis Research, Beijing 101149, China. Electronic address: zzd417@163.com.
Abstract
BACKGROUND: Tuberculous meningitis (TBM) is the most severe and frequent form of central nervous system tuberculosis. The current lack of efficient diagnostic tests makes it difficult to differentiate TBM from other common types of meningitis, especially viral meningitis (VM). Metabolomics is an important tool to identify disease-specific biomarkers. However, little metabolomic information is available on adult TBM. METHODS: We used 1H nuclear magnetic resonance-based metabolomics to investigate the metabolic features of the CSF from 18 TBM and 20 VM patients. Principal component analysis and orthogonal signal correction-partial least squares-discriminant analysis (OSC-PLS-DA) were applied to analyze profiling data. Metabolites were identified using the Human Metabolome Database and pathway analysis was performed with MetaboAnalyst 3.0. RESULTS: The OSC-PLS-DA model could distinguish TBM from VM with high reliability. A total of 25 key metabolites that contributed to their discrimination were identified, including some, such as betaine and cyclohexane, rarely reported before in TBM. Pathway analysis indicated that amino acid and energy metabolism was significantly different in the CSF of TBM compared with VM. CONCLUSIONS: Twenty-five key metabolites identified in our study may be potential biomarkers for TBM differential diagnosis and are worthy of further investigation.
BACKGROUND:Tuberculous meningitis (TBM) is the most severe and frequent form of central nervous system tuberculosis. The current lack of efficient diagnostic tests makes it difficult to differentiate TBM from other common types of meningitis, especially viral meningitis (VM). Metabolomics is an important tool to identify disease-specific biomarkers. However, little metabolomic information is available on adult TBM. METHODS: We used 1H nuclear magnetic resonance-based metabolomics to investigate the metabolic features of the CSF from 18 TBM and 20 VM patients. Principal component analysis and orthogonal signal correction-partial least squares-discriminant analysis (OSC-PLS-DA) were applied to analyze profiling data. Metabolites were identified using the Human Metabolome Database and pathway analysis was performed with MetaboAnalyst 3.0. RESULTS: The OSC-PLS-DA model could distinguish TBM from VM with high reliability. A total of 25 key metabolites that contributed to their discrimination were identified, including some, such as betaine and cyclohexane, rarely reported before in TBM. Pathway analysis indicated that amino acid and energy metabolism was significantly different in the CSF of TBM compared with VM. CONCLUSIONS: Twenty-five key metabolites identified in our study may be potential biomarkers for TBM differential diagnosis and are worthy of further investigation.
Authors: Pouya Akhbari; Matthew K Jaggard; Claire L Boulangé; Uddhav Vaghela; Gonçalo Graça; Rajarshi Bhattacharya; John C Lindon; Horace R T Williams; Chinmay M Gupte Journal: Bone Joint Res Date: 2021-01 Impact factor: 5.853