FtsZ ring formation without subsequent cell division after replication runout in Escherichia coli.

In this report, we have investigated cell division after inhibition of initiation of chromosome replication in Escherichia coli. In a culture grown to the stationary phase, cells containing more than one chromosome were able to divide some time after restart of growth, under conditions not allowing initiation of chromosome replication. This shows that there is no requirement for cell division to take place within a certain time after initiation of chromosome replication. Continued growth without initiation of replication resulted in filamented cells that generally did not have any constrictions. Interestingly, FtsZ rings were formed in a majority of these cells as they reached a certain cell length. These rings appeared and were maintained for some time at the cell quarter positions on both sides of the centrally localized nucleoid. These results confirm previous findings that cell division sites are formed independently of chromosome replication and indicate that FtsZ ring assembly is dependent on cell size rather than on the capacity of the cell to divide. Disruption of the mukB gene caused a significant increase in the region occupied by DNA after the replication runout, consistent with a role of MukB in chromosome condensation. The aberrant nucleoid structure was accompanied by a shift in FtsZ ring positioning, indicating an effect of the nucleoid on the positioning of the FtsZ ring. A narrow cell length interval was found, under and over which primarily central and non-central FtsZ rings, respectively, were observed. This finding correlates well with the previously observed oscillatory movement of MinC and MinD in short and long cells.

The ftsZ gene of Haloferax mediterranei: sequence, conserved gene order, and visualization of the FtsZ ring.

We sequenced the ftsZ gene region of the halophilic archaeon Haloferax mediterranei and mapped the transcription start sites for the ftsZ gene. The gene encoded a 363-amino-acid long FtsZ protein with a predicted molecular mass of 38 kDa and an isoelectric point of 4.2. A high level of similarity to the FtsZ protein of Haloferax volcanii was apparent, with 97 and 90% identity at the amino acid and nucleotide levels, respectively. Structural conservation at the protein level was shown by visualization of the FtsZ ring structure in H. mediterranei cells using an antiserum raised against FtsZ of H. volcanii. FtsZ rings were observed in cells in different stages of division, including cells with pleomorphic shapes and cells that appeared to be undergoing asymmetric division. Cells were also observed that displayed constriction-like invaginations in the absence of an FtsZ ring, indicating that morphological data are not sufficient to determine whether pleomorphic Haloferax cells are undergoing cell division. Both the upstream and downstream gene order in the ftsZ region was found to be conserved within the genus Haloferax. Furthermore, the downstream gene order, which includes the secE and nusG genes, is conserved in almost all euryarchaea sequenced to date. The secE and nusG genes are likely to be transcriptionally and translationally coupled in Haloferax, and this co-expression may have been a selective force that has contributed to keeping the gene cluster intact.

On the origin of branches in Escherichia coli.

Some Escherichia coli strains with impaired cell division form branched cells at high frequencies during certain growth conditions. Here, we show that neither FtsI nor FtsZ activity is required for the development of branches. Buds did not form at specific positions along the cell surface during high-branching conditions. Antibiotics affecting cell wall synthesis had a positive effect on branch formation in the case of ampicillin, cephalexin, and penicillin G, whereas mecillinam and D-cycloserine had no substantial effect. Altering the cell morphology by nutritional shifts showed that changes in morphology preceded branching, indicating that the cell's physiological state rather than specific medium components induced branching. Finally, there was no increased probability for bud formation in the daughters of a cell with a bud or branch, showing that bud formation is a random event. We suggest that branch formation is caused by abnormalities in cell wall elongation rather than by aberrant cell division events.