Therefore,

E. coli can divide at the midpoint of the cell without an oscillating Min system. So far we don’t know why AtMinD is localized to the polar region in E. coli cells. Compared with chloroplasts, E. coli cells are much smaller and have a rod shape. By just localized to the polar region, AtMinD may keep the FtsZ ring and the division site at the midpoint of the cell. Since EcMinD depolymerize the FtsZ filaments at the non-division site through its interacting protein EcMinC [8], it is also likely that AtMinD interacts with and functions through EcMinC. To test this prediction, GFP-EcMinC and AtMinD were coexpressed at 50 μM IPTG in RC1 mutant (Figure Bortezomib in vitro 2J and 2K). The mutant phenotype was rescued and GFP-EcMinC was localized to

puncta at cell ends except that there was some signal in the cytosol. Without AtMinD, GFP-EcMinC was distributed evenly throughout the cell in RC1 mutant (Figure 2L and 2M). These data further suggest that AtMinD may interact with EcMinC and helps interpret the complementation of HL1 mutant by AtMinD. To get an idea of the levels of GFP-AtMinD, GFP-EcMinD and other GFP fusion proteins, an immuno-blot was done (Figure 2N). The levels of these proteins were very close at the same concentration of IPTG. This is probably is because their coding genes are in similar vectors and under the control of the same promoter. The level of GFP-EcMinD probably was a little higher than that of GFP-AtMinD. This BMS-354825 concentration could be due to a better codon usage, higher stability etc. EcMinD rescues the mutant phenotype best at 20 μM IPTG, while AtMinD and its GFP fusion proteins rescues the mutant phenotype best at 50 μM IPTG. This probably is because their working mechanisms or (and) their activities are different. AtMinD interacts with EcMinC To further explore the function of AtMinD, we studied the protein-protein interaction

between AtMinD and EcMinC. First, we tested this by yeast two hybrid (Figure 3). In the yeast strain AH109 we used, certain genes for the biosynthesis of histidine, leucine and tryptophan are not expressed. If two proteins fused to the bait and prey respectively interact, the genes for the synthesis of histidine, leucine and tryptophan will be induced Rebamipide and the yeast cell will be able to grow without histidine, leucine and tryptophan. Because this system is leaky, 3-AT was used to reduce the basal level. As shown in Figure 3, full length AtMinD can interact with EcMinC no matter whether it is fused to the activation domain or the binding domain. The presence or the absence of the chloroplast transit peptide had no effect on the interaction between AtMinD and EcMinC (Figure 3). Both AtMinD and EcMinC can self-interact (Figure 3). Figure 3 Interactions of EcMinC and AtMinD examined by yeast two hybrid analysis. Yeast cells grown without Leucine (L), Tryptophan (T) and Histidine (H), but with 3-AT. ΔTP, deletion of the chloroplast transit peptide. SD, synthetic defined.