Integrative bioinformatics and machine learning reveal an association of LTF and MMP8 with systemic inflammation and lung injury

BACKGROUND: Acute lung injury and its severe form, acute respiratory distress syndrome (ARDS), represent life-threatening conditions with high mortality rates. Since the lung is the primary target organ in systemic inflammatory conditions like sepsis, this study aimed to identify key regulatory genes by analyzing systemic inflammation-related transcriptomic data, and to explore their roles in lung injury. METHODS: We performed bioinformatic screening on the GSE32707 dataset (systemic inflammation patients vs. controls) using differential expression analysis, WGCNA, and machine learning. Functional enrichment, PPI network, immune infiltration, ceRNA network, and molecular docking analyses were conducted. Validation was carried out using human samples, cell-based assays, and murine lung injury models. RESULTS: We identified 506 differentially expressed genes, predominantly enriched in immune-related pathways. Machine learning algorithms prioritized lactoferrin (LTF) and matrix metalloproteinase-8 (MMP8) as key genes. Co-IP assays demonstrated a physical association between LTF and MMP8. A predicted ceRNA network suggested post-transcriptional regulation of LTF, while molecular docking indicated binding potential of LTF/MMP8 with methylprednisolone. Experimental validation confirmed that both genes are upregulated during lung injury and can be modulated by anti-inflammatory treatment. CONCLUSION: LTF and MMP8 play critical roles in lung injury, potentially serving as molecular connectors between systemic inflammation and pulmonary damage, and represent promising therapeutic targets for further investigation.