The origins of directional persistence in amoeboid motility

Abstract

Directional persistence is crucial for graduate school as well as for life on Earth. Persistently directed motion enables neutrophils to target wound sites, growing axons to find their targets, migrating cells to create a correctly developed embryo, and <italic>Dictyostelium discoideum</italic> amoebas to hunt their bacterial prey. While chemical signals guide the cells in these examples, an underlying persistence in eukaryotic motility greatly enhances navigation toward a target. For amoeboid cells, this basal persistence results from a tendency of cells to zig-zag. Whence this zig-zag bias, and thereby whence directional persistence, is the question addressed by this thesis.

First, a new automated and unbiased pseudopod detection algorithm is presented. This algorithm is used to experimentally test a new model of pseudopod zig-zagging, in which pseudopods are excitable bursts of protrusive activity. The key assumption is that pseudopods produce a memory of their own activity, which serves to make the cortex locally more excitable in the future. This is a top-down, network level model that makes minimal assumptions and maintains an appropriate level of abstraction, while retaining biological interpretations and predictive power. The model's predictive power is evidenced by four novel results that are tested and supported by new experimental data.

Previous evidence suggested that pseudopod memory might be found in the specific delocalization of myosin II at growing pseudopods, though this evidence has been either low resolution or disconnected from local pseudopod dynamics. The possibility of memory via myosin delocalization is tested by relating pseudopod production to cortical protein localization. This approach reveals that myosin II delocalizes in growing pseudopods, myosin remains lower for a time after pseudopods stop, cortical regions with lower myosin are more excitable, and pseudopodial myosin delocalization is due neither to dilution from membrane extension nor to targeted disassembly through MHCK A. Rather, it appears that myosin is locally excluded by the dense actin meshwork created by a growing pseudopod. This suggests a model that intimately links the memory of a pseudopod to its very existence.