Computational Modeling of a Cytokine Interaction Network in Microglia

Abstract

Neuroinflammation due to glial activation has been linked to many CNS conditions. To examine the mechanistic underpinnings of how intercellular cytokine interactions coordinate the neuroinflammatory milieu, we employed a dynamic modeling approach. We developed a microglial cytokine interaction model using differential equations. The model fit to published data from LPS‐stimulated cultured microglia demonstrates that the model captures the dynamics of cytokine regulatory interactions. Results from a variance‐based parametric sensitivity analysis suggest that TNFα is highly sensitive to negative feedback from TGFß and IL‐10. Based on this analysis, we examined the relative roles of these anti‐inflammatory cytokines in controlling TNFα levels in the contexts of adaptation and tolerance to LPS stimuli. Simulations examining adaptation of the TNFα response to LPS indicate that TGFß controls TNFα adaptation by modulating the steady‐state response to LPS, with negligible effects on the peak TNFα response. In contrast, IL‐10 primarily controlled the TNFα peak amplitude following LPS with less effect on adaptation. Consistent with experimental data, the model simulations also showed tolerance in the TNFα response to repeated doses of LPS. A simulated IL‐10 knockout (KO) phenotype resulted in enhanced TNFα tolerance whereas simulated TGFß KO hyper‐sensitized the TNFα response to repeated LPS stimuli. The contrasting effects of TGFß versus IL‐10 occurred because the IL‐10 KO enhanced the initial TNFα peak response, which resulted in subsequent TGFß activation and robust negative feedback, thereby rendering TNFα unresponsive to a second LPS stimulus.