Blood microRNA signatures point to potential tuberculosis biomarkers

Research on small RNA molecules called microRNAs (miRNAs) has produced varying and sometimes conflicting results when individual studies use small-RNA sequencing. To address these inconsistencies, Bharanidharan Devarajan and colleagues carried out a metadata analysis, pooling data from multiple publicly available studies. They downloaded 131 samples from seven different datasets housed in the Sequence Read Archive (SRA) database to look for miRNAs that are consistently altered in people with systemic tuberculosis (TB). The pooled samples included 45 healthy controls, 47 people with active TB, and 39 people with Latent TB (LTB). By comparing active TB and LTB samples with controls, the team aimed to identify differentially expressed (DEs) miRNAs that could help distinguish disease states and reveal molecules linked to progression from latent to active infection. The study focused strictly on small-RNA sequencing data drawn from public repositories and used metadata analysis as a way to increase statistical power and find patterns that single studies might miss.

The investigators first identified differentially expressed miRNAs in active TB and LTB samples versus controls. From those comparisons they filtered a total of 52 miRNAs based on categories described as TB-specific, active TB, LTB-specific, and related to disease progression. The team then performed gene enrichment and network analysis separately for upregulated and downregulated miRNAs to understand the biological pathways these small RNAs might influence. From this analysis they selected eight miRNAs as candidate biomarkers: hsa-miR-155-5p, hsa-miR-223-3p, hsa-miR-32-3p, hsa-miR-374a-3p, hsa-miR-374a-5p, hsa-miR-582-5p, hsa-miR-320d, and miR-122-5p. These eight were highlighted because their predicted or known roles tie into TB-related processes and pathways, including the PI3K-Akt signaling pathway, TNF-signaling pathway, phagosome, and NOD-like signaling pathway. The authors note this is the first metadata analysis using small-RNA sequencing data from publicly available datasets to nominate miRNAs for systemic TB and that these candidates require further experimental confirmation.

If validated, these miRNA signatures could become blood-based biomarkers to help detect systemic TB or to indicate when latent infection is shifting toward active disease. The emphasis on pathways such as PI3K-Akt signaling, TNF-signaling, phagosome, and NOD-like signaling suggests the miRNAs identified may reflect immune and cellular processes central to TB pathogenesis, making them biologically plausible markers rather than random statistical hits. Performing metadata analysis across seven datasets also demonstrates a way to reconcile conflicting findings from small individual studies by increasing sample diversity and size. The authors are careful to note that the work is a computational and integrative step: experimental studies will be needed to confirm whether the eight miRNAs reliably indicate TB status in different patient groups, and whether they add value beyond existing diagnostic tools. Nonetheless, the study provides a prioritized list of miRNAs for follow-up laboratory and clinical research.