(556h) Multiscale Image-Based Simulation of Transient PDGF-BB Gradient Formation Explains How Fibroblasts Affect Each Other in Making Directional Decision during Chemotaxis

Authors: 
Pham, L. Q., New Jersey Institute of Technology NJIT
Voronov, R., New Jersey Institute of Technology NJIT
Chandran, V. D., New Jersey Institute of Technology
Chege, D., New Jersey Institute of Technology
Tong, N. A. N., New Jersey Institute of Technology NJIT
Rodrigues, L. N., New Jersey Institute of Technology NJIT
Directed fibroblast migration is central to highly proliferative processes in regenerative medicine and developmental biology, such as wound healing and embryogenesis. However, the mechanisms by which single fibroblasts affect each other’s directional decisions, while chemotaxing in microscopic tissue pores, are not well understood. Here we use image-based multiscale modeling in order to explain an experimentally observed contradiction to the classical chemotaxis theory: fibroblasts alternating their directional decisions in microscopic tissue-sized confinement, instead of following the steepest chemotactic gradient. The chemotaxis experiment was performed in a microfluidic maze, which contains micron-sized paths (i.e. a long path and a short path) that the cells preferred depending on the sequence in which they entered the maze. The maze had two ports, one entrance and one exit. The entrance was connected to a cell reservoir, while the exit was connected to a chemoattractant source of 50 ng/ml of PDGF-BB. NIH-3T3 fibroblasts were allowed to migrate along the PDGF-BB gradient, towards the exit of the maze. Real-time imaging of the migration was performed using an automated Olympus IX83 microscope. Experimental data, such as directional decisions of the cells, was manually extracted from time-lapse images acquired at 15-min intervals. An image-based multiscale COMSOL model was performed in order to visualize the gradient of PDGF-BB inside of the microfluidic maze. The macroscopic scale of the model simulated the global chemoattractant gradient established in the maze, while the microscopic scale was used to represent the localized gradient disturbances caused by the presence of the cells. In the former case, we used a transient two-dimensional transport-of-diluted-species physics with two constant concentration boundary conditions applied at the maze ports. Moreover, the model took into account depletion of the PDGF-BB via natural decay. On the microscopic level, on the other hand, we modeled endocytosis of the PDGF-BB by the individual cells. The simulation results showed that the reason why the fibroblasts alternate the paths that they take through the maze can be explained by the localized PDGF-BB consumption by the preceding cell. Namely, the leading cell follows the shortest path (i.e. the steepest gradient), as expected from classical chemotaxis theory. The subsequent cell, however, sees a gradient altered by the leading cell ahead of it, and gets steered into the long path. Overall, these finding supports the conclusion that the presence of the individual cells in microconfinement can modify the external chemotactic cues via self-generation of local gradients. Consequently, accounting for such effects could lead to a better understanding of tissue generation in vivo, and result in more advanced engineered tissue products in vitro.