Studies of the structure and putative complex of the bacterial cell division proteins FtsL, DivIC and DivIB

Bacterial cell division (cytokinesis) is a complex developmental process involving the coordinated septation of the cytoplasmic membrane in conjunction with the extracytoplasmic cell wall layers. During vegetative growth, this procedure is executed at midcell and generates two progeny from a single bacterium that are genetically and morphologically identical. In spore forming bacteria such as Bacillus subtilis, an alternative division pathway is executed during conditions of nutrient deprivation, which results in the development of a dormant spore. These division processes are performed by a macromolecular machine called the divisome. The divisome is composed of at least 10 different proteins. The first to assemble, FtsZ, does so as a circumferential ring on the inside of the cytoplasmic membrane. This ring permits the localization of the other divisomal proteins. Most of these proteins have recognized functions. In the B. subtilis divisome, however, at least three of these proteins, namely, FtsL, DivIC, and DivIB have thus far resisted functional classification. Despite this, genetic research has suggested that these three proteins may interact with each other in the divisome. In this doctoral thesis, we examine the structure and putative complex of these three proteins using biochemical and biophysical methods. We find that the extracytoplasmic domains of these three proteins are not sufficient to form a ternary complex and propose that insertion in the membrane is essential for ternary complex formation. The domain architecture of the extracytoplasmic region of DivIB protein is assessed and is composed of three structurally autonomous domains. We have named these domains α, β, and γ, from N- to C-terminus. The 3-dimensional structure of the β domain from Geobacillus stearothermophilus DivIB was determined by NMR. This domain exists in two structural states that are the result of cis-trans proline isomerization. Both cis and trans conformers were found to be composed of a unique protein fold. Mutagenesis showed that the γ domain appears to be more important for function in the divisome than the α domain. We propose a model in which the γ domain and the α domain (a putative POTRA domain) interact and stabilize an interaction between FtsL and DivIC.