Development and characterisation of a 3D cell culture model of the tuberculosis granuloma

Tuberculosis (TB) is a disease caused by a group of genetically related bacteria known as the Mycobacterium tuberculosis complex (MTBC) and causes 1.6 million deaths yearly. The TB granuloma is the hallmark cellular structure of latent TB that contains the spread of infection. More comprehensive in vitro models of TB that better resemble the cellular and immunoregulatory complexity of the granuloma would facilitate the study of the interplay between the bacteria and the different immune system cells. We aim to generate an in vitro, 3D cell culture model of the TB granuloma that can be easily implemented using readily available commercial reagents and materials. A commercial encapsulation system was used to generate small capsules containing human cell lines in the presence of GFP-expressing M. bovis BCG strain and maintained in culture for several weeks. The 3D structure formed by the cells inside the spheres was evaluated by microscopy and flow cytometry to verify how they are organised inside the sphere and to measure cell survival and bacteria replication. Antibiotics were used to compare the influence of the model on the resistance/susceptibility to common anti-TB drugs. The results show that cells readily form 3D cellular aggregates around infected cells. The model could be maintained for several weeks before bacteria-induced cell necrosis. Cell viability remained stable, with more than 80 % live cells following two weeks of culture. The usage of different cell lines to generate these granulomas, being those monocytes (THP-1), lymphocytes (Jurkat) or epithelial cells (A549) inside or outside the capsules, resulted in distinct bacterial replication, demonstrating the contribution of each cell type, and suggesting relevant communication between cells inside and outside the spheres to control the infection. Moreover, when comparing the infection using our 3D model versus the usual 2D monolayer model, we observed a distinct resistance profile against rifampicin and isoniazid. In conclusion, the generated 3D infection model resembles some of the structural and cellular features of the TB granuloma and significantly improves the duration of infection experiments. Our preliminary results demonstrate the stability of this model and differential bacterial growth kinetics and drug-resistance profiles of the bacteria, enabling its future use in immunological and drug discovery studies.