Investigating early innate immune signalling in Tuberculosis using advanced microphysiological models

Tuberculosis (TB) remains one of the world's deadliest infectious diseases, causing over 1.2 million deaths every year. Yet we still understand surprisingly little about the very first moments of infection when the TB bacterium lands deep in the lungs and encounters our immune cells. Studying this in living patients is impossible, and animal models don't faithfully reproduce what happens in human lungs. To overcome this, our lab at the Medical Faculty of Heidelberg University has spent the past 18 months refining a lung-on-chip model that we have developed. This tiny device, roughly the size of a USB stick, which recreates the key features of the human lung's deepest air sacs using living human cells. The device is optically transparent, allowing us to visualise the cellular interactions in real-time with microscopy.

A central challenge we have tackled is to generate and incorporate the right type of immune cell. The lungs contain specialized sentinels called alveolar macrophages which are the first defenders that encounter inhaled bacteria. Normal lab-grown immune cells don't behave like these lung-resident cells. We developed a method to convert cells from the blood of human donors into cells closely resembling real alveolar macrophages. We confirmed this by analysing which genes they switch on and by showing that they take a thin oily film that coats our airways known as lung surfactant just as their natural counterparts do. These "alveolar macrophage-like" cells were successfully incorporated into the lung-on-chip, where they sit atop epithelial cells that line the lung and blood vessel cells. This co-culture of cells can be exposed to breathing-like stretching motions, and an air–liquid boundary, much like in a real lung.

With this improved model, we tackled a question with direct clinical relevance: can we deliver drugs to the lung more effectively by harnessing the natural surfactant film? In real life, inhaled medicines must travel along this surfactant layer to reach the deepest parts of the lung. Standard lab methods simply flood cells with molecules dissolved in liquid, which is nothing like how cells at the airway are exposed to foreign substances. We adapted a technique using a small paper bridge to mimic the natural transport of drugs along the surfactant film into our chip (Fig. 1a). The results were striking: drugs delivered this way ended up predominantly inside the macrophages and were retained and stayed active for much longer (Fig.1b). In contrast, the conventional liquid method deposited drugs mainly in the lung epithelial cells below the macrophages, who quickly removed the drug.

As a practical demonstration, we showed that the TB antibiotic bedaquiline, delivered via the surfactant film, could prevent bacterial growth for at least two days after treatment (Fig. 1c). This proof-of-concept suggests that specially formulated inhaled antibiotics could one day deliver

medicine directly to infected cells in the lung, potentially reducing the heavy pill burden and side effects that TB patients currently endure over months of treatment.

Our work also revealed that alveolar macrophage-like cells are more susceptible to TB infection than standard lab-grown immune cells and respond with a gene activation pattern matching what is known in vivo. This further validating our model as a more faithful mimic for the human lung. We have reported these findings in a manuscript currently under peer review. In the final months of the project, we are investigating how additional immune cells including eosinophils, another type of immune cell recently linked to TB defence, interact with infected macrophages. Early data shows they can help control bacterial growth, and we are working to understand the mechanisms. Overall, this project has produced a new, more realistic laboratory model of the human lung that can be used to study the earliest stages of TB infection and to develop better strategies for delivering inhaled therapies without the need for animal experiments.

Publications

C Garcia-Mouton, R Mishra, K Sharma, T Nass, M Sommers, J Perez-Gil&, VV Thacker&. Vehiculation and functional delivery of lipophilic therapeutics and antibiotics via pulmonary surfactant in a lung-on-chip model. (&senior, corresponding author), doi: 10.64898/2025.12.18.695116 .