Date of Award

August 2018

Degree Type


Degree Name

Doctor of Philosophy (PhD)




Roy D. Welch


biofilm, fruiting body formation, Myxococcus xanthus, Ostwald ripening, phase separation, self-organization

Subject Categories

Life Sciences


The primary goal of systems biologists is to understand the mechanics underlying complex, collective, self-organizing behaviors displayed by all living systems, from biofilm formation and wound healing to embryogenesis. Myxococcus xanthus is a soil bacterium used as a model organism to study biofilm self-organization. It has a relatively large genome and a complex life cycle that involves two distinct phases. M. xanthus cells can move on agar, and a few million cells will organize to form a predatory biofilm or “swarm” that grows and expands if placed on a nutrient-rich agar surface. If placed on a nutrient-poor agar surface, the same swarm will turn inward and contract, aggregating to form spore-filled multicellular fruiting bodies designed to survive periods of starvation. Extensive progress has been made in identifying genes and genetic pathways that regulate fruiting body formation. However, an accurate description of the dynamics that underlie the process of aggregation is still lacking, and there is still debate and disagreement on the subject. This dissertation provides some explanation regarding individual M. xanthus cell behavior during fruiting body formation, as well as the behavior of aggregates.

In the first part of this work, we show that the transition from individual cells to the formation of multicellular aggregates can be controlled through relatively small changes in M. xanthus cell behavior; complicated cell-to-cell signaling, stigmergy (where a trace formed by a cell on an agar surface influences the movement of nearby cells that contact the trace), and cell differentiation are not required for aggregation. We propose that M. xanthus aggregation matches a physical phenomenon that has been characterized in non-living systems, called motility induced phase separation (MIPS). By studying non-reversing mutant cells and manipulating their velocities, we show that cell movement can be made to fall within the boundary of the phase region so that cells succeed in forming aggregates. Alternately, cell movement can be made to fall outside the boundary of the phase region so that cells fail to form aggregates.

After the initial stages of aggregation, an M. xanthus swarm actively rearranges the number and relative positions of aggregates by causing some of them to move, merge, or disappear. In the second part of this work, we demonstrate that equations describing Ostwald ripening within a thin liquid film were able to predict aggregate behavior with high accuracy. Consistent with this theory, both relative aggregate size and the distance between aggregates influence the likelihood that a given aggregate will disappear. In general, in neighboring aggregates, the ones that are small and in close proximity will tend to shrink and disappear, while the larger, more isolated aggregates will likely persist and become permanent. By tracking individual cells around aggregates, we show that more cells are leaving shrinking, disappearing aggregates than entering them, while more cells are entering growing, persistent aggregates than leaving them. All of these data are in good agreement with the Ostwald ripening equations. In the last part of this work, we show that aggregation can break down in only a certain number of ways by analyzing single gene disruption of four paralogous gene families representing almost 400 genes. We found that gene families correlate with phenotype, suggesting possible redundancy. In conclusion, my dissertation data provides possible answers to some of the persistent questions regarding M. xanthus developmental dynamics.


Open Access

Included in

Life Sciences Commons