Reconsidering the bacterial division-apparatus positioning system

The Min system is one of the two known mechanisms that are responsible for the accurate positioning of the division apparatus in Escherichia coli. Quite fascinatingly, the Min proteins achieve this by forming a dynamic concentration gradient that shuttles constantly from pole to pole. It is believed that this behavior depends only on the mutual interaction of three components - two proteins and the cell membrane. One of the important features of the Min system is its typical length scale of a couple of microns. Extensive research in the past years resulted in a detailed mathematical model that can reproduce the in vivo Min system behavior quite accurately.

However, reconstituted in vitro studies have resulted in the formation of patterns such as surface waves and oscillations that – though fascinating in its own right – have a typical length scale that is an order of magnitude larger that the in vivo observed one. It is highly important to understand this difference, both from a basic-science point of view and if one wants to use this system in the context of synthetic cells.

We study the Min system behavior in fully enclosed microfluidic compartments (coated with a supported lipid membrane). This setup enables us for the first time a full control on the characteristics of the system. We show that the compartment geometry is a major determinant of the dynamical mode that the system will adopt. We find that dominant mode in the major part of the phase diagram are spirals. Waves is the dominate behavior only for long or large chambers while oscillations is the dominate mode for narrow chambers or these with a large aspect ratio. Notably, the geometrical phase-diagram that we discover does not correspond to the in vivo observed behavior. Various additional parameters such as the temperature and crowding of the bulk media or of the membrane do not bridge the gap between the in vivo vs. in vitro pattern formation differences.

Our results strongly suggests that in spite of the good correspondence between the mathematical model and the in vivo behavior of the Min system, we still lack a real understanding of this important model system. In particular, we suggest that additional components probably contribute to the Min system dynamical pattern formation in vivo.