Development of muti-agent-simulation models for intercellular communication via cytokines and extracellular matrices

BackgroundIntercellular communication is a critical innovation during multicellular organismal evolution. Cells release / receive cytokines and utilize them as intercellular signal entities. How a well-orchestrated communication emerges from individual cell behavior remains largely unknown. Here we abstracted the biological phenomenon and developed multi-agent-simulation to investigate the intracellular communication. MethodsTwo dimensional MAS platform was developed using Artisoc 4.2.1 standard software. We focused on intercellular communication via cytokines and extracellular matrices. Three agents, "cells", "cytokines" and "extracellular matrices" were defined and the interaction rules among the agents were designed. Two different mathematical models of cytokine-gradient determination were tested: spatial derivative and temporal derivative models. As a case study, neutrophil swarming was modeled and the cell swarming was defined as an evaluation criterion. Moreover, a surgically injured mouse model and a fluorescent time-lapse imaging were used to observe neutrophil swarming. ResultsWe performed multiple simulations with six different virtual conditions, changing multiple parameters simultaneously and randomly. After 400 simulations for each condition, we counted the number of successful trials (i.e. neutrophil swarming within 10000 steps). Spatial derivative model showed more successes compared to temporal derivative model. Among eight parameters randomly assigned, cells exploration speed by random walk most remarkably influenced on success rate of neutrophil swarming. In in vivo model, bone marrow derived neutrophils gather towards the clumps with various migration speed. The mode of approaching resembles to spatial derivative model, rather than temporal derivative model. ConclusionsMAS could be a useful approach to investigate the emergence of intercellular communication during multicellular evolution. Neutrophil could adopt the spatial derivative model as a sensing mechanism of cytokine gradient.