Air Pollutants Toluene and Benzene Interact with Interferon Proteins, Upregulating a Five-Gene Tuberculosis Biomarker Signature

Exposure to air pollution, particularly particulate matter (PM), is a significant global health concern, increasing susceptibility to infectious diseases like Tuberculosis (TB). Despite this established epidemiological link, the precise molecular mechanisms by which pollutants exacerbate TB remain largely unknown. Current understanding suggests PM2.5 can induce inflammatory responses similar to endotoxins, but specific pollutant-protein interactions and their impact on host immunity, especially interferon-related signaling, are poorly characterized. Bridging this knowledge gap is crucial for developing targeted interventions and diagnostic tools for TB prevention and management.

Researchers employed a multi-omics and network toxicology approach, integrating human genes associated with seven common air pollutants with known TB-associated genes from public databases. They developed a diagnostic pipeline using 175 machine learning models across transcriptomic datasets to identify a core gene signature. This candidate signature was then validated using qRT-PCR in an independent clinical cohort of TB patients. To predict specific molecular interactions, molecular docking simulations were performed, focusing on pollutants like toluene and benzene with identified proteins. Further in silico single-cell knockout analyses were conducted to model transcriptional perturbations following key protein interactions.

The analysis identified 271 intersecting genes significantly enriched in critical inflammatory and immune-related pathways, including IL-17 signaling, TNF signaling, and Toll-like receptor signaling. Machine learning models converged on a robust five-gene candidate signature comprising STAT1, IFIH1, IFIT2, IFIT3, and CYBB. Clinical qRT-PCR validation confirmed the upregulation of these genes in TB patients, with individual AUCs ranging from 0.76 to 0.89, indicating strong diagnostic potential. Molecular docking simulations provided mechanistic insights:

Toluene was predicted to form significant hydrophobic interactions with STAT1, IFIT2, and IFIT3, suggesting direct modulation of these interferon-related proteins. In silico STAT1 perturbation in monocytes further predicted transcriptional alterations involving RETN and S100A9, with a notable enrichment in IFN-γ-related pathways, highlighting a direct link between pollutant exposure and altered interferon immunity.