Tuberculosis (TB) remains one of the world's most serious public health threats, with approximately one-quarter of the global population infected with Mycobacterium tuberculosis. While most infections remain latent (LTBI), individuals with LTBI are at risk of progressing to active TB and potentially becoming new sources of transmission. Effective TB control, therefore, requires not only diagnosing active disease but also predicting disease progression. Although existing biomarkers can distinguish between LTBI and active TB, their ability to dynamically monitor disease progression remains unknown, representing a critical scientific challenge for early intervention and transmission control.
Previous work by the research team revealed that alternative splicing events are particularly sensitive to changes in the intracellular microenvironment, making them ideal biomarkers for monitoring disease development. Among various types of splicing, intron retention (IR) affects nearly 80% of protein-coding genes and represents a crucial biological process in disease pathogenesis. The dynamic and adaptive nature of certain IR events has made them valuable indicators for assessing processes such as cancer progression and aging. However, the role of IR in TB progression had not been explored until now.
In a groundbreaking study published in Chinese Medical Journal on January 30, 2026, researchers have systematically mapped the molecular landscape of IR-mediated splicing reprogramming during TB progression. The study identifies deoxyribonuclease 1 like 2 IR (DNASE1L2-IR) as a novel dynamic biomarker for monitoring TB progression and elucidates its mechanism in modulating host immune responses through the regulation of DNA degradation by different protein isoforms.
The research team conducted high-throughput sequencing analysis of 1,729 clinical samples and discovered genome-wide intron splicing reprogramming in host cells across different stages—from healthy controls to LTBI and active TB. From this comprehensive analysis, the researchers identified four core IR events significantly associated with both latent infection and disease progression. Among these, the IR event of DNASE1L2 (a gene encoding a deoxyribonuclease) demonstrated the most striking pattern: its level was elevated in LTBI compared to healthy controls, significantly reduced in patients with active TB, and most notably, markedly higher in "progressors" (those who progressed from LTBI to active TB) than in "non-progressors." This pattern was consistently validated in three different M. tuberculosis antigen-stimulated cell models, showing a characteristic biphasic "rise-and-fall" response, suggesting a close association with immune status changes under infection pressure.
Mechanistic investigation revealed that DNASE1L2-IR generates two distinct transcript isoforms: a long isoform (DNASE1L2-L) and a short isoform (DNASE1L2-S). Following M. tuberculosis stimulation, the long isoform predominantly localized to the cytoplasm, whereas the short isoform remained membrane-anchored. Functional experiments demonstrated significant differences in DNase activity between these isoforms: DNASE1L2-L exhibited substantially higher efficiency in degrading both M. tuberculosis genomic DNA and supercoiled plasmid DNA compared to DNASE1L2-S. Crucially, cell-based functional assays confirmed that overexpression of the highly active DNASE1L2-L isoform effectively suppressed the release of pro-inflammatory cytokines (TNF-α and IL-1β) induced by M. tuberculosis stimulation, whereas DNASE1L2 knockout exacerbated inflammatory responses.
These findings indicate that upregulation of DNASE1L2-IR during early TB progression promotes the generation of the highly active DNASE1L2-L isoform, which assists the host in clearing pathogen DNA and mitigating excessive inflammation. Conversely, downregulation of DNASE1L2-IR in some individuals who progress to active TB weakens this antibacterial defense mechanism, allowing persistent infection and exacerbation of disease.
This study establishes the first connection between post-transcriptional regulation via intron retention and the dynamic monitoring of TB progression. It not only identifies DNASE1L2-IR as a promising clinical biomarker but also reveals a new mechanism by which M. tuberculosis remodels host splicing patterns to influence disease progression. These findings provide a solid theoretical foundation for developing RNA splicing-based tools for predicting TB progression and novel therapeutic strategies, potentially transforming TB control through precision early warning and targeted intervention.
Reference
DOI: http://doi.org/10.1097/CM9.0000000000003974
About Professor Ying Binwu from Sichuan University
Professor Ying Binwu is a Cheung Kong Scholar Distinguished Professor, Chief Physician, and Doctoral Supervisor at West China Hospital of Sichuan University, where he directs both the Department of Science and Technology and the Department of Laboratory Medicine. With expertise in molecular diagnostics of infectious diseases, he has led over 20 national research projects, published more than 300 papers, and been granted 10 national patents. His work has earned prestigious honors, including the Sichuan Province Scientific and Technological Progress Award (First Class).
