The Spatiotemporal Dynamics of Mycobacterial Infection



Journal Title

Journal ISSN

Volume Title



Host response to Mycobacterium tuberculosis (Mtb) is distinctive in the use of a spatial immunological response to limit the progression of infection. This results in the formation granulomas, aggregations of immune cells that isolate invading microbes, a hallmark of the adaptive immune response to Mtb infection. Traditional in vivo studies have investigated the mature granuloma, but relatively fewer studies investigate how these structures form during the early stages of infection nor how spatial organization impacts control, resolution, or dissemination of the bacterium. Research has shown that initial aggregation of macrophages during innate immune response influences the progression of disease and formation of granulomas during adaptive immunity. However, current experimental methods for studying cellular interactions during early stages of infection are ill adapted for concurrent spatial and temporal quantification of host-pathogen dynamics, which is necessary for a quantitative understanding of the innate spatial immune response to Mtb and to inform the development of accurate computational models of tuberculosis disease. To address this, we developed a three-dimensional (3D) ex vivo model of mycobacterium infected macrophages cultured in reconstituted basement membrane and characterized the structural impact of 3D structure on infection dynamics in comparison to standard two-dimensional (2D) infection models. We quantified temporal immune response using standard biological sampling methodologies and long-term time-lapse confocal imaging to quantify the early spatiotemporal dynamics of macrophage response to mycobacterium infection. Our studies using Mycobacterium smegmatis indicate that the 3D environment induces a shift in dynamics. In 3D we see significantly higher cellular velocities in infected conditions as compared to control non-infected conditions, whereas the converse occurs in 2D. This may impact computational models that utilize 2D assumptions. We developed a data analysis pipeline to quantify macrophage state with respect to infection and cellular microenvironment. Results show non-infected and non-active macrophages within infected environments present dynamics comparable to controls, while infected and activated macrophages exhibit comparable spatiotemporal dynamics in 2D and 3D. Using the more virulent Mycobacterium bovis BCG, we observe a greater distinction between control and infected conditions and preliminary evidence of a more distinct 3D immune response resulting in increased cell death and extracellular bacteria.



Tuberculosis, Granuloma, Mycobacterium, Infection, Spatiotemporal Dynamics, Host-Pathogen Interactions, Machine learning, Data analytics


Portions of this document appear in: Stolley, Danielle L., and Elebeoba E. May. "Spatiotemporal Analysis of Mycobacterium-Dependent Macrophage Response." In 2018 40th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), pp. 2390-2393. IEEE, 2018.