BCG sharply cuts tuberculosis spread between mouse lungs

Tuberculosis remains a major scientific challenge, in part because the ways Mycobacterium tuberculosis (Mtb) moves and grows inside the body are complex. To study how infection spreads within the lungs, researchers led by Vitaly V. Ganusov analyzed data from mice infected with ultra-low doses and compared animals that had received BCG vaccination with those that had not. The study focused on colony-forming unit (CFU) counts in the right and left murine lungs and on whether infection appeared in one lung (unilateral) or both (bilateral). Rather than relying only on descriptive statistics, Ganusov and colleagues built mathematical models to represent the dynamics of Mtb within and between lungs. By fitting those models to CFU data from vaccinated and unvaccinated mice, the researchers aimed to quantify how BCG changes the course of infection — not just whether it lowers bacterial numbers, but whether it changes the pattern and rate of spread between the two lungs. The work links experimental infection measurements directly with model-based estimates of replication and dissemination, providing a clearer picture of what BCG does early and later in infection.

The team developed several mathematical models of Mtb dynamics and of dissemination between murine right and left lungs, then fitted those models to the CFU data from unvaccinated and BCG-vaccinated mice. They tested alternative models that incorporated either a direct lung-to-lung pathway or an indirect lung-intermediate-tissue-lung pathway for dissemination. Both kinds of models fit the unvaccinated data equally well, which suggests multiple plausible routes for Mtb spread in this system. Across models, the fits predicted rapid Mtb replication during early infection, a transient period of control within 1–2 months, and then continued bacterial growth in the chronic phase. When the models were fitted to data from BCG-vaccinated animals, the estimated rate of Mtb dissemination between the lungs was reduced by 89% while the within-lung replication rate was reduced by about 9%. The vaccination also corresponded to fewer infected mice, lower lung CFU burden, and more frequent unilateral rather than bilateral lung infection. Using these parameterized models, the authors calculated how many mice would be needed to detect vaccine effects on Mtb clearance or dissemination, extending previous sample-size estimates.

Taken together, this modeling work clarifies how BCG exerts its dominant effect in this mouse model: by reducing Mtb replication early enough to produce fewer infected animals, lower CFU counts in lungs, and a much lower chance that both lungs become infected. The result is a large drop in the modeled rate of lung-to-lung dissemination, even though the direct impact on replication rates within an already infected lung was modest. By providing a rigorous, quantitative framework that links experimental CFU data to modeled rates of replication and spread, the study offers a practical tool for pre-clinical vaccine evaluation. Researchers designing and interpreting ultra-low-dose (ULD)-infected mouse experiments can use these models to estimate necessary sample sizes and to separate effects on replication from effects on dissemination, which should help prioritize candidate vaccines and refine experimental designs for next-generation TB vaccine development.