Analysis of the macrophage biochemical network during activation and
infection with Mycobacterium tuberculosis

(Christian J. Ray and Denise E. Kirschner, Microbiology and Immunology,
University of Michigan, Ann Arbor, MI, USA, 48109 )

A complete understanding of the immune response to Mycobacterium tuberculosis will require an integrated approach accounting for the complex multiscale networks involved in the immune and pathogen interactions. One network of interest in M. tuberculosis infection is the macrophage biochemical network, as macrophages are the host cell for and primary defense against mycobacteria. We applied a local S-system representation to examine molecular scale activation, killing effectors and iron homeostasis mechanisms.  We represented a second physical scale by tracking an intracellular population of mycobacteria responsive to, and influencing, the macrophage state. Using this dual-scale model we examined system responses to different macrophage activation states due to stimulation by two activation signals. We used global statistical sensitivity analyses to show important model features with the partial rank correlation coefficient between model parameters and a given outcome variable. Local controlled comparisons between a wild-type model and cases with simulated inhibition of individual network parameters allowed us to find network properties determined by immune effectiveness criteria such as bacterial clearance and macrophage network robustness. We thus elucidated evolved properties of biochemical interactions that optimize the overall macrophage response. We found negative feedback by the killing mechanism on activation stimuli reduces bacterial clearance but is required to stabilize the positive regulation of killing effectors needed for effective clearance.  The negative feedback is required for a dependably quiescent state in the absence of sufficient activation signals. We also found that the interactions between killing mechanisms and iron regulation slightly undermine bacterial clearance but are necessary for overall functional effectiveness. We conclude that constraints on overall functional effectiveness sometimes reduce the efficacy of mycobacterial clearance by macrophages.