Exploratory proteomic and metabolomic profiling of pleural effusions identifies histone H4 and alanine as promising complementary markers for pleural tuberculosis - 01/09/26

Abstract |
The diagnosis of pleural tuberculosis (Pl-TB) remains challenging. Histopathological analysis and pathogen detection in pleural biopsies are informative but limited. We investigated differentially expressed proteins and metabolites in pleural effusions from patients with Pl-TB, malignancies, and other pathologies. A proteomic analysis of pooled pleural effusions identified 45 proteins exclusively detected or upregulated in Pl-TB samples, many linked to infectious processes. Conversely, 18 proteins were uniquely found or upregulated in malignant pleural effusions, mainly associated with detoxification and hemostasis. To validate these findings, we employed targeted proteomics in individual samples. Eight proteins were validated: S100-A9, histone H4, insulin-like growth factor-binding protein 2, fibrinogen beta chain, ficolin-3, immunoglobulin heavy constant alpha 1, sulfhydryl oxidase 1, and histidine-rich glycoprotein. Additionally, NMR-based metabolomics identified 13 metabolites with differential abundance between Pl-TB and non-TB samples. Notably, N-acetyl-glycoprotein and the branched-chain amino acids, alanine and lysine differed between groups. Proteomic and metabolomic analyses revealed distinct molecular profiles between Pl-TB and non-TB patients, despite intra-group variability. To address this, we applied classification models. Histone H4 and alanine consistently emerged as discriminative features. Overall, this study provides novel insights into the molecular landscape of Pl-TB. The combined quantification of proteins and metabolites may improve differential diagnosis, although should be further validated in larger, independent cohorts before clinical application.
Le texte complet de cet article est disponible en PDF.Graphical abstract |
Highlights |
• | Pleural tuberculosis differential diagnosis remains challenging. |
• | 63 proteins differ; 8 validated by targeted proteomics. |
• | 13 metabolites differ by NMR-based metabolomics. |
• | Models identify histone H4 and alanine as biomarkers. |
• | Combined protein–metabolite profiling may improve diagnosis. |
Keywords : Multiomic, Infection, Differential diagnosis, Bacteria, Neoplasia
Plan
Vol 160
Article 102801- septembre 2026 Retour au numéroBienvenue sur EM-consulte, la référence des professionnels de santé.
L’accès au texte intégral de cet article nécessite un abonnement.
Déjà abonné à cette revue ?
