A model of TLR4 signaling and tolerance using a qualitative, particle-event-based method: introduction of spatially configured stochastic reaction chambers (SCSRC).

INTRODUCTION: There have been great advances in the examination and characterization of intracellular signaling and synthetic pathways. However, these pathways are generally represented using static diagrams when in reality they exist with considerable dynamic complexity. In addition to the expansion of existing mathematical pathway representation tools (many utilizing systems biology markup language format), there is a growing recognition that spatially explicit modeling methods may be necessary to capture essential aspects of intracellular dynamics. This paper introduces spatially configured stochastic reaction chambers (SCSRC), an agent-based modeling (ABM) framework that incorporates an abstracted molecular 'event' rule system with a spatially explicit representation of the relationship between signaling and synthetic compounds. Presented herein is an example of the SCSRC as applied to Toll-like receptor (TLR) 4 signaling and the inflammatory response.
METHODS: The underlying rationale for the architecture of the SCSRC is described. A SCSRC model of TLR-4 signaling was developed after a review of the literature regarding TLR-4 signaling and downstream synthetic events. The TLR-4 SCSRC was implemented in the free-ware software platform, Netlogo. A series of in silico experiments were performed to evaluate the response of the TLR-4 SCSRC with respect to response to simulated administration of lipopolysaccharide (LPS). The pro-inflammatory response was represented by simulated secretion of tumor necrosis factor (TNF). Subsequent in silico experiments examined the response to of the TLR-4 SCSRC in terms of a simulated preconditioning effect represented as tolerance of pro-inflammatory signaling to a second dose of LPS.
RESULTS: The SCSRC produces simulated dynamics of TLR-4 signaling in response to LPS stimulation that are qualitatively similar to that reported in the literature. The expression of various components of the signaling cascade demonstrated stochastic noise, consistent with molecular expression data reported in the literature. There is a dose dependent pro-inflammatory response effect seen with increasing initial doses of LPS, and there was also a dose dependent response with respect to preconditioning effect and the establishment of tolerance. Both of these dynamics are consistent with published responses to LPS.
CONCLUSIONS: The particle-based, spatially oriented SCSRC model of TLR-4 signaling captures the essential dynamics of the TLR-4 signal transduction cascade, including stochastic signal behavior, dose dependent response, negative feedback control, and preconditioning effect. This is accomplished even given a high degree of molecular event abstraction. The component detail of the SCSRC may allow for sequential parsing of various preconditioning effects, something not possible without computational modeling and simulation, and may give insight into the expected consequences and responses resulting from manipulation of one or many of these modulating factors. The SCSRC is admittedly a work in evolution, and future work will sequentially incorporate additional regulatory mechanisms, both intracellular and paracrine/autocrine, and improved mapping between the spatial chamber configuration and molecular event rules, and experimentally define biochemical reaction rate constants. However, the SCSRC has promise as a highly modular and flexible modeling method that is suited to the dynamic knowledge representation of intracellular processes.