A Rapid Active–Latent–Relapse Murine Model of Tuberculosis Based Blood Transcriptional Signature That Distinguishes Disease Stages

The lack of reliable diagnostic tools and relapse monitoring for latent tuberculosis infection (LTBI) constitutes a major obstacle to global tuberculosis (TB) control. This highlights an urgent need for robust animal models and predictive biomarkers. To address this, we report the successful establishment of a rapid murine model of recapitulating the active, latent, and relapse phases of TB within a compressed ten-week timeframe-hence termed the rapid multi-stage TB murine model. In this model, mice were first intravenously infected with <i>Mycobacterium tuberculosis</i>, followed by a four-week isoniazid (INH) regimen starting at two weeks post-infection. By week six, pulmonary bacterial loads in most mice dropped below the detection limit, signifying the establishment of latency. Reactivation was subsequently triggered by a four-week administration of anti-TNF-α (Tumor Necrosis Factor-α) monoclonal antibody. Leveraging this reproducible and time-efficient model, we performed transcriptomic profiling of peripheral blood and identified a distinct sixteen-gene signature (including <i>Ets2</i>, <i>Fam111a</i>, <i>Fosl2</i>, <i>Gadd45b</i>, <i>Nfkbid</i>, <i>Rgs1</i>, <i>Bhlhe40</i>, <i>Il1r2</i>, <i>Clec2d</i>, <i>Kmo</i>, <i>Lynx1</i>, <i>Papd4</i>, <i>Trim34a</i>, <i>Wrb</i>, <i>Nlrp12</i>, <i>Spns1</i>) that dynamically tracks disease progression. Collectively, these findings not only provide a valuable and efficient preclinical tool but also deliver transformable candidate biomarkers with immediate potential to guide the development of novel diagnostic strategies for LTBI surveillance and management.