TissueSim

ABSTRACT

Turnover in Epithelia is a Mechanism for the Formation of Clonal Multifocal Lesions

Multi-focal neoplasms are often observed that are clonally related.  Two current hypotheses for this phenomenon are local metastases and expansion of a premalignant clone from which the foci later emerge.  Here we introduce a third hypothesis in which clones can fragment due to normal background turnover of cells and tissues.  We developed an agent-based model that simulates clonal expansion on a patch of tissue containing crypts arranged in a hexagonal grid.  Each crypt possesses a single gene which, when mutated, may confer a selective advantage (reproductive or survival) for that crypt.  Crypts die as a result of background turnover or wounding. Mutants with survival advantage are less susceptible to wounding, to a degree controlled by a survival advantage parameter. Crypts divide in a wound healing response to the death of a neighbor after a stochastic waiting time that represents the time required to double the number of cells in the crypt and bifurcate.  Mutants with a reproductive advantage benefit from shorter waiting times which is governed by the reproductive advantage parameter.  Additional model parameters control the frequency and intensity of wounding.  We performed a sweep of the area of parameter space that is most biologically plausible and discovered that the probability of fragmentation is relative to the degrees of reproductive and survival advantage but only for very low fitness or neutral clones. We also found that a clone was most likely to fragment under moderate amounts of wounding and turnover. Similar to disturbance theory in ecology, too little turnover results in the stable, slow growth of the mutant clone. Too much disturbance often results in the extinction of the mutant clone. These results suggest a more parsimonious explanation for clonally related foci of neoplasms than has previously been recognized. Multifocal lesions may emerge from turnover and the expansion of clones with weak fitness advantages over neighboring epithelial cells.

The images to the left are sample tissue configurations that were generated from my tissue simulation.  The red dots are mutants, the white dots are wild-type, and the black spaces are dead areas on the tissue.  My simulation suggests that multifocal lesions can arise from a simple geometrical process.

View my poster presented at the NIH Multi-scale Modeling Consortium

Also here is a PowerPoint presentation explaining the simulation in greater detail.

The source code is available at SourceForge.