1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

Formation of the Z ring is the first known event in bacterial cell division. However, it is not yet known how the assembly and contraction of the Z ring are regulated. Here, we identify a novel cell division factor ZapB in Escherichia coli that simultaneously stimulates Z ring assembly and cell division. Deletion of zapB resulted in delayed cell division and the formation of ectopic Z rings and spirals, whereas overexpression of ZapB resulted in nucleoid condensation and aberrant cell divisions. Localization of ZapB to the divisome depended on FtsZ but not FtsA, ZipA or FtsI, and ZapB interacted with FtsZ in a bacterial two-hybrid analysis. The simultaneous inactivation of FtsA and ZipA prevented Z ring assembly and ZapB localization. Time lapse microscopy showed that ZapB–GFP is present at mid-cell in a pattern very similar to that of FtsZ. Cells carrying a zapB deletion and the ftsZ84ts allele exhibited a synthetic sick phenotype and aberrant cell divisions. The crystal structure showed that ZapB exists as a dimer that is 100% coiled-coil. In vitro, ZapB self-assembled into long filaments and bundles. These results raise the possibility that ZapB stimulates Z ring formation directly via its capacity to self-assemble into larger structures.


  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

Bacterial cell division is a highly regulated developmental process that requires the co-ordinated action of many different factors in time and space. In the rod-shaped, Gram-negative bacterium Escherichia coli, more than 10 different proteins are required for the completion of cell division (Lutkenhaus, 2002; Goehring et al., 2005; Rothfield et al., 2005; Vicente and Rico, 2006). The tubulin homologue FtsZ (Löwe and Amos, 1998) forms the Z ring at mid-cell (Bi and Lutkenhaus, 1991) and is the first factor known to be recruited to the future division site (Addinall et al., 1996; Pichoff and Lutkenhaus, 2002). In many cells, the Z ring exists in more than half of the cell cycle and it has been suggested that the assembly of the division machinery occurs in two steps (Aarsman et al., 2005). The first step invokes formation of the Z ring that initially consists of FtsZ polymers linked to the membrane by FtsA and ZipA (Pichoff and Lutkenhaus, 2002). In the absence of both FtsA and ZipA, no Z rings are formed at mid-cell (Pichoff and Lutkenhaus, 2002). Complete Z rings can form without either FtsA or ZipA, but both are required for recruitment of the next factor, FtsK (Pichoff and Lutkenhaus, 2002). In the second step, addition of FtsQ and the FtsBL complex allow for the progressive addition of remaining factors (FtsW, FtsI, FtsN, AmiC and EnvC) that all have functions related to septal cell wall synthesis (Goehring et al., 2005; Vicente and Rico, 2006). Except for FtsZ, all of the above-mentioned factors are transmembrane proteins, associated with the inner membrane (FtsA) or periplasmic (AmiC and EnvC) (Bernhardt and de Boer, 2003; 2004; Goehring and Beckwith, 2005). The ABC transporter complex FtsEX is also associated with the septal ring and FtsE interacts directly with FtsZ (Corbin et al., 2007; Reddy, 2007).

Z ring assembly is the first known event in bacterial cell division and is a major regulatory process determining initiation of the cell division cascade. Besides, Z ring assembly functions as a cell cycle check point at which both positive and negative factors act to regulate or modulate Z ring assembly. SulA is an SOS-induced cell division inhibitor that prevents Z ring formation (Jones and Holland, 1984; Bi and Lutkenhaus, 1993). SulA dimers interact directly with the T7 loop of FtsZ, thereby inhibiting FtsZ GTPase activity and protofilament formation (Cordell et al., 2003). MinC, the division inhibitor carried by the oscillating MinD ATPase, also prevents cell division by interacting with FtsZ (Marston et al., 1998; Hu et al., 1999). However, it is less clear how MinC functions and two different modes have been proposed: in the first model, MinC inhibits polymerization of FtsZ (Hu et al., 1999; Pichoff and Lutkenhaus, 2001), whereas in the second model, MinC acts later to prevent FtsA incorporation into the FtsZ filaments (Justice et al., 2000).

Previous observations of septum formation suggested the presence of a mechanism that prevented septum formation over the nucleoid – a phenomenon termed ‘nucleoid occlusion’ (Woldringh et al., 1991). Nucleoid occlusion prevents the formation of aberrantly located septa coinciding with nucleoid regions that would otherwise lead to DNA breakage and daughter cells with incomplete chromosomes. Interestingly, nucleoid occlusion factors have since been identified in both Bacillus subtilis and E. coli (Wu and Errington, 2004; Bernhardt and de Boer, 2005) and, also here, Z ring formation functions as a regulatory check point. In E. coli, the slmA gene codes for a TetR repressor-like nucleoid-binding protein that interacts directly with FtsZ in vitro (Bernhardt and de Boer, 2005). FtsZ probably is also the target of the Noc nucleoid occlusion factor of B. subtilis (Wu and Errington, 2004).

Positive modulaters of Z ring formation have also been identified. In a screening for multi-copy suppression of lethality conferred by MinCD overexpression in B. subtilis, an FtsZ-associated protein ZapA was identified (Gueiros-Filho and Losick, 2002). GFP-ZapA localizes to the septum and promotes FtsZ bundling in vitro. The zapA gene became essential when FtsZ activity was reduced and, interestingly, deletion of zapA was synthetic lethal with a deletion of ezrA, which may be the functional analogue of E. coli ZipA in B. subtilis (Gueiros-Filho and Losick, 2002; Errington et al., 2003; Haeusser et al., 2004). ZapA is widely conserved and the E. coli homologue also localizes to the division ring in an FtsZ-dependent manner (Gueiros-Filho and Losick, 2002; Goehring et al., 2005). ZapA contains a coiled-coil domain that dimerizes ZapA dimers in vitro, as well as a small globular domain (Low et al., 2004).

Here we describe the discovery and characterization of a novel, abundant cell division factor ZapB in E. coli. Overproduction of ZapB confers condensation of the nucleoids and the formation of ectopic cell division septa. By contrast, the absence of ZapB leads to the formation of ectopic, helical Z rings, and delayed cell division. Recruitment of functional ZapB–GFP fusions to the Z ring depended on FtsZ but not on FtsA, ZipA or FtsI. In a bacterial two-hybrid analysis, ZapB interacted with FtsZ, and zapB had a synthetic detrimental phenotype with the conditional ftsZ84 mutation. The crystal structure showed that ZapB forms anti-parallel coiled-coil dimers that may interact at their termini.


  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

Identification of zapB

The par2 locus of E. coli plasmid pB171 is a prototypical type I partitioning locus that stabilizes heterologous replicons very efficiently (Ebersbach and Gerdes, 2001). par2 encodes an oscillating Walker box ATPase (ParA), a centromere binding protein (ParB) and two centromere-like sites (Gerdes et al., 2000; Ebersbach and Gerdes, 2001). The MinD-like ParA ATPase forms filaments that oscillate over the nucleoid and, by an unknown mechanism, positions plasmids regularly over nucleoids (Ebersbach and Gerdes, 2004; Ebersbach et al., 2006).

Host factors specific for plasmid-partitioning mechanisms have not yet been identified (Ebersbach et al., 2006). Using a pBR325-derived plasmid vector, we generated an E. coli K-12 chromosome library and screened for genes that, in multiple copies, would interfere specifically with the function of par2 of plasmid pB171. The screening method was a modification of the ‘split’ colony assay previously developed by the Cohen and Hiraga groups (see Fig. 1A and B) (Biek and Cohen, 1986; Niki et al., 1988). The procedure is described in the legend to Fig. 1 and in more detail in Experimental procedures. One plasmid that consistently destabilized a par2 plasmid contained a 3662 bp chromosomal DNA fragment (pGE73 in Fig. 1C). This region, located at 88.7 min on the E. coli K-12 chromosome, contains three genes, hslU, menA and menG (rraA) transcribed in the counterclockwise direction, and a small open reading frame, yiiU, transcribed in the opposite direction. To further pin down the region responsible for plasmid destabilization, derivatives of pGE73 were constructed that contained smaller segments. In this way, a fragment carrying only the yiiU open reading frame was found to be sufficient to destabilize the par2 plasmid. For reasons described later, yiiU was called zapB (Z ring-associated protein B). Finally, to prove that overexpression of the zapB-encoded protein (ZapB) destabilizes the test plasmid, zapB was cloned downstream of the arabinose-inducible PBAD promoter of pBAD33. Indeed, the resulting plasmid destabilized the test plasmid only in the presence of arabinose (pGE604 in Fig. 1C). These results strongly indicate that the product of the zapB gene was the factor responsible for par2-specific plasmid destabilization. Surprisingly, however, deletion of zapB did not have any measurable effect on the function of par2 in liquid media, not even during long-term plasmid stabilization tests (data not shown). We also used the bacterial two-hybrid system (BTH) as described previously (Karimova et al., 1998) to test if ZapB interacts with ParA or ParB encoded by par2. We found no indication for such interaction (data not shown). We conclude that ZapB only plays a minor role as a factor in plasmid partitioning by par2.


Figure 1. Identification of zapB of E. coli and effects of ZapB overexpression on nucleoid morphology. A. Genetic set-up of plasmids used in the screen for multi-copy suppression of par2-mediated plasmid partitioning. A chromosomal Sau3AI library, established in a pBR325 derivative (pKG325) yielding resistance to chloramphenicol and tetracycline but not ampicillin, was transformed into strain KG61 already harbouring the R1 test plasmid pGE271 (par2+). Cells were plated on Xgal plates, selecting for the incoming plasmid only. KG61 carries lacZYA but lacks lacI. Plasmid pGE271 carries par2 and lacIq. Thus, KG61 cells harbouring pGE271 form white colonies on Xgal because of repression of the lac operon while loss of the plasmid leads to formation of blue colonies or white colonies with blue sectors. B. Xgal plates showing the results of transformation of pGE73 (zapB) into (a) strain KG61/pGE271 (par2+) and (b) KG61/pJV2 (hok/sok). In the former situation, destabilization of pGE271 resulted in the presence of blue sectors in almost all colonies while in the latter situation, the hok/sok-carrying test plasmid remained stable, thus resulting in white colonies. C. Illustration showing restriction map (S, Sau3AI; H, HpaI; B, BglII) and deletion analysis of the 3662 bp chromosomal fragment originally isolated in plasmid pGE73. The names of the different deletion constructs are shown to the left and the ability of these constructs to destabilize the pGE271 R1 (par2+) test plasmid is shown to the right. In the bottom construct, destabilization depended on the presence of arabinose in the plates. The pGE73 series of plasmids are all derivatives of pKG325. Plasmid pGE604 in the bottom of the illustration is a pBAD33 derivative, in which the zapB open reading frame was cloned downstream of the arabinose-inducible PBAD promoter. Numbers above the line in the top illustration are NCBI database co-ordinates of the chromosome of E. coli K-12 strain MG1655. D and E. Combined phase-contrast and fluorescence (DAPI) microscopy images of cells of strain MC1000 carrying pGE604 grown in liquid culture in the absence (D) or presence of arabinose (E). Samples were collected at time 0 (G0) and after four generations of growth (G4). In addition, samples were collected from overnight cultures (TON). Bar = 2 μm.

Download figure to PowerPoint

Overexpression of ZapB condenses the nucleoid

As ParA encoded by par2 of pB171 is a nucleoid-associated protein (Ebersbach and Gerdes, 2004), we inspected the nucleoid morphology of cells overproducing ZapB. Figure 1D shows cells of strain MC1000/pGE604 (pBAD::zapB) grown in Luria–Bertani (LB) medium at 30°C in the absence of inducer. As can be seen, the nucleoids filled almost the entire cellular space, indistinguishable from cells carrying the control vector (not shown). When an exponentially growing culture of MC1000/pGE604 was diluted into medium containing arabinose, the cells apparently grew normally but, after three generations, the nucleoid-free dark areas at the cell poles and in-between separated nucleoids increased in size as if the nucleoids had started to condense. After four generations, the nucleoids were clearly compacted (Fig. 1E, e–j). The contour of the nucleoids no longer followed the rod shape of the cell, but instead adopted a more rectangular profile. In most cells with two nucleoids, these were clearly separated from each other (Fig. 1E, g–j). In some cells, both nucleoids were centred around the quarter cell positions (Fig. 1E, h and j) while in other cells, one nucleoid was at the quarter-cell position and the other one at the opposite pole (Fig. 1E, g and i). Cells were also found in which the chromosomes looked as if they were in the process of segregation (Fig. 1E, f and g). After six generations of growth, the picture was the same, nucleoids were condensed but cell shape appeared normal (data not shown).

Figure 1E, k–x shows images of a culture of MC1000/pGE604 that was left to grow with arabinose overnight. In this culture, the cells looked very heterogeneous compared with cells of an overnight control culture (Fig. 1D, e). Not only nucleoid localization but also cell division were clearly disturbed, resulting in the formation of very small cells (Fig. 1E, k–n) in addition to more or less normal-sized cells (Fig. 1E, p–t) and cell filaments (Fig. 1E, v–x). The filamentous cells had clear constrictions and appeared to be blocked in a late state of septation, similar to cells carrying mutations in factors important for cell–cell separation, such as the periplasmic amidase AmiC or the putative peptidase EnvC (Hara et al., 2002; Bernhardt and de Boer, 2003). Apparently, these cell chains consisted of normal-sized cells, elongated cells and mini-cells linked to each other. In addition, some cells appeared to be linked by what looked like remnants of dead or lysed cells (Fig. 1E, v–x), a phenotype somewhat similar to cell chains formed by certain strains carrying mutations or deletions in the cell division protein FtsK, which acts to couple chromosome segregation and cell division (Diez et al., 1997; Liu et al., 1998; Bigot et al., 2004). The significance, if any, of this resemblance is not known. Similar results were obtained with cells harbouring pGE73, the original plasmid identified in the genetic screen (data not shown). Therefore, it is possible that the ZapB-induced nucleoid condensation and/or cell division defects might be the actual cause of the pGE271 instability observed in the genetic screen (Fig. 1B) (see Discussion).

Deletion of zapB results in cell elongation

The zapB gene was deleted as described earlier (Datsenko and Wanner, 2000). Strains MC1000 and MC1000ΔzapB had similar mass doubling times under three different growth conditions (Fig. 2C). At an OD450 of about 0.4, cells grown in glucose minimal medium at 30°C were stained with 4,6-diamidino-2-phenylindole (DAPI) and subjected to fluorescence- and phase-contrast microscopy. Wild-type (wt) cells appeared relatively homogeneous in size (Fig. 2A, a). This is also evident from Fig. 2B, which shows the results of cell length measurements of 250 wt cells (blue curve). Most cells had a length between 2 and 4 microns with the average cell length being 3 microns (Fig. 2C). The DAPI stain revealed that in general, the shorter cells harboured one nucleoid while longer cells, with or without constriction, appeared to have two separated nucleoids (Fig. 2A, a).


Figure 2. Deletion of zapB increases the average cell length. A. Combined phase-contrast and fluorescence microscopy images of living cells of strains MC1000 (wt) (a and d) and MC1000ΔzapB (b, c and e) respectively. Cells were grown in M9 glucose medium supplemented with casamino acids. Samples were collected at OD450∼0.4 (a–c) and OD450∼2.0 (d and e). Nucleoids were stained with DAPI. Bar = 2 μm. B. Diagram showing the cell length distribution of wt and ΔzapB cells grown as described in A. A total of 250 cells of each strain were measured. C. Table showing mass doubling time and average cell length of cells of strains MC1000 and MC1000ΔzapB grown in different media and at different temperatures, as indicated.

Download figure to PowerPoint

The ΔzapB cells had a more heterogeneous cell length distribution and a significant number of cells formed short filaments (Fig. 2A, b and c). Cells of this strain varied in length from about 2 to 15 microns (Fig. 2B, red curve), with an average cell length of 4 microns (Fig. 2C). Many of the longer cells did not appear to contain a constriction. The experiment was repeated under different growth conditions, i.e. in LB at 37°C and 30°C respectively. Under these conditions, the cells grew with mass doubling times of 22 and 38 min respectively. Again, cells of the ΔzapB strain were longer than wt cells (Fig. 2C). A somewhat similar result was obtained when cells were allowed to grow out of the exponential phase and into early stationary phase [Fig. 2A, d (wt) and e (ΔzapB)]. We thought possible that the long cells seen in the ΔzapB culture could be dead and therefore unable to divide. To test this, we used the LIVE/DEAD staining method to discriminate between live and dead cells. We found no difference in viability of the two strains (data not shown), and even the filamentous cells present in the ΔzapB culture appeared to be alive (Fig. S1). Thus, deletion of zapB does not lead to the formation of a subpopulation of dead cells.

As overproduction of ZapB condensed the nucleoids and resulted in elongated cells (Fig. 1), we also analysed if deletion of zapB decondensed the nucleoids and/or affected nucleoid segregation. First, we analysed nucleoid lengths in wt and ΔzapB cells, but no significant difference was apparent (data not shown). Second, we analysed nucleoid sizes and segregation patterns in cells treated with cephalexin. Again, we did not observe any differences between wt and ΔzapB cells (Fig. S2).

ZapB localizes at mid-cell in a dynamic FtsZ-like pattern

To investigate the subcellular localization of zapB, we constructed a plasmid carrying PBAD::zapB::gfp (pEG3a) that expresses ZapB–GFP from an arabinose-inducible promoter. After a short pulse (5′) of transcription, almost all cells of MC1000/pEG3a (96%) showed a distinct, localized fluorescent signal (Fig. 3A). The GFP signal was present at mid-cell, either as a discrete transverse band, a single central dot or two dots close to the cell periphery. The zapB::gfp fusion complemented a ΔzapB strain, indicating that the fusion was functional (data not shown). The ZapB–GFP localization pattern in a ΔzapB strain was indistinguishable from that in the wt strain (data not shown). It should be noted that longer transcription pulses or expression of ZapB–GFP (and GFP–ZapB) from a synthetic lac promoter resulted in inclusion body formation (data not shown).


Figure 3. Dynamic subcellular localization of ZapB. Combined phase-contrast and fluorescence microscopy images showing the localization of ZapB–GFP in living cells of strain MC1000. Cells were grown at 37°C in M9 glucose medium supplemented with casamino acids. Expression of ZapB–GFP from plasmid pEG3a was induced by addition of arabinose as described in Experimental procedures. A. Fluorescent images using a GFP filter were taken immediately after mounting the cells under the microscope. B. Time lapse series. Fluorescent images were taken every 5 min with GFP filter. Scale bars are 2 μm. The different cell morphologies in A and B was due to differences in growth conditions (growth in liquid versus solid medium).

Download figure to PowerPoint

A typical time lapse microscopy of the ZapB–GFP signal is shown in Fig. 3B. As the pre-divisional cell became longer, the transverse band condensed into a central dot (0′ to 35′). Between 35′ and 40′, the central dot disappeared and new transverse bands formed at quarter-cell positions, corresponding to the mid-cell positions of the new daughter cells. Thus, the subcellular localization of ZapB is very similar to that of FtsZ (Addinall et al., 1996). In a few cases, we were able to capture a ZapB–GFP signal that connected the old dot at mid-cell with the new bands at quarter-cell positions (data not shown).

ZapB interacts with itself and FtsZ in a BTH assay

ZapB is a predicted coiled-coil protein (Fig. S3A), and coiled-coil regions are often involved in protein self-interaction. Therefore, we used a BTH assay to test whether ZapB interacts with itself in vivo (Karimova et al., 1998). ZapB was genetically fused to two complementary fragments, T25 and T18 that constitute the catalytic domain of Bordetella pertussis adenylate cyclase. As seen from Table 1A, plasmid pair pT18-ZapB/pT25-ZapB yielded a strong positive response in the BTH assay, indicating that ZapB interacts with itself.

Table 1.  Bacterial two-hybrid analysis of protein interactions.
 (A) BTH101
 (B) BTH101ΔzapB
  1. Interaction of indicated proteins fused to the T18 and T25 fragments, respectively, of the B. pertussis adenylate cyclase. Protein–protein interactions were tested in the cya strain BTH101 (Table 1A) and a derivative, called BTH101ΔzapB (Table 1A), which carries a kanamycin cassette inserted into zapB on the chromosome. Interaction of hybrid proteins resulted in formation of blue colonies on Xgal plates: +, weak interaction (heterogeneous colouring of colonies); ++, stronger interaction (most colonies had the same blue colour); +++, strong interaction (all colonies had an intense blue colour). Combinations of hybrid proteins that led to the formation of white colonies are indicated with −.


Test of plasmids expressing either T18-ZapB and T25-FtsZ or T18-FtsZ and T25-ZapB in strain BTH101 also resulted in a Lac+ phenotype (Table 1A), raising the possibility that ZapB and FtsZ interact in vivo. T18-ZapB and T25-ZapB also yielded positive responses with FtsA in the BTH assay (Table 1A). However, the responses were weaker with FtsA than with FtsZ and may reflect indirect interaction. No interaction could be detected between T18-ZapB or T25-ZapB and other known cell division proteins (i.e. ZipA, FtsK, FtsQ, FtsI, FtsB, FtsL, FtsN, FtsW, FtsX or YmgF) (data not shown). These results indicate that ZapB interacts with one of the divisome proteins, most likely FtsZ. As expected, a test of FtsZ against FtsA was positive in the BTH strain (Table 1A), and the FtsZ–FtsA interaction did not depend on ZapB (Table 1B). Several attempts (including co-sedimentation, co-immunoprecipitation and EM) were accomplished to determine if FtsZ interacts with ZapB in vitro. However, the high propensity for ZapB to self-aggregate made these experiments inconclusive (Jakob Møller-Jensen, unpublished) and in vitro interactions with ZapB will be the subject for further investigations.

Formation of ring-like ZapB structures at mid-cell depends on the Z ring but neither FtsA, ZipA nor FtsI

Strain KG22Z84 carries the temperature-sensitive ftsZ84 mutation. This mutation allows Z ring formation and cell division at 30°C. At 42°C, Z rings rapidly disintegrate, thereby leading to inhibition of cell division and formation of filaments (Addinall et al., 1997). Figure 4A shows examples of ZapB–GFP localization in the wt (KG22) strain at 30°C (a and b) and at 45 min after shift to 42°C (c and d). In all samples of this culture, cell division proceeded normally and ZapB–GFP showed the typical localization pattern described above. A similar localization pattern was seen in the ftsZ84 strain at 30°C (Fig. 4B, a–c), although the ftsZ84 cells generally were slightly elongated, as reported previously (Addinall et al., 1997). Figure 4B, d shows the localization pattern of ZapB–GFP in the ftsZ84 strain at the non-permissive temperature. As seen, the transverse bands disappeared and the GFP signal became diffuse in the elongated cells. Thus, ZapB–GFP localization in distinct transverse bands depended on a functional FtsZ protein.


Figure 4. Localization of ZapB depends on FtsZ but not FtsA, ZipA or FtsI. Fluorescence microscopy images showing the localization of ZapB–GFP in living cells of strains (A) KG22 (wt) and (B, e) KG22Z84 (ftsZ84), and in fixed cells of strains (B, a–d) KG22Z84, (C, a and b) PS236 (ftsA12), (D, a and b) PS223 (zipA1) and (E, a and b) PS234 (zipA1 ftsA12). Cells were grown at 30°C in M9 glucose medium supplemented with casamino acids (A and B) or in LB medium (C, D and E). Expression of ZapB–GFP from plasmid pEG3a was induced by the addition of arabinose as described in Experimental procedures. 1.5 h after the induction, exponentially growing cultures were diluted into fresh medium pre-heated to the non-permissive temperature (42°C). Samples were collected 45 min (A and B) or 60 min (C, D and E) after the shift to 42°C. The cultures in C, D and E had a normal morphology and localization of ZapB–GFP at 30°C and are not shown. In the cells in (A, e) and (B, e), cell division was inhibited by treatment with cephalexin, at 30°C. Bar = 2 μm.

Download figure to PowerPoint

To analyse the involvement of FtsA and ZipA in recruitment of ZapB to the division site, we made use of strains carrying temperature-sensitive alleles of ftsA and zipA (Table 2). Inactivation of FtsA did not affect the formation of the transverse bands formed by ZapB–GFP (Fig. 4C). Similarly, inactivation of ZipA only slightly reduced the number of ZapB–GFP bands (Fig. 4D), also excluding ZipA as a crucial factor for ZapB recruitment. By contrast, simultaneous inactivation of FtsA and ZipA prevented ZapB–GFP recruitment (Fig. 4E). It is known that the Z ring does not form when FtsA and ZipA are inactivated simultaneously (Pichoff and Lutkenhaus, 2002), and we conclude that ZapB recruitment requires Z ring formation.

Table 2.  Strains and plasmids used and constructed.
Strain or plasmidRelevant genotype/descriptionResistanceReference/source
E. coli K-12 strains
 MC1000Δ(ara-leu) Δlac rpsL150 Casadaban and Cohen (1980)
 MC1000ΔzapBMC1000 strain in which zapB has been deleted This work
 MC1000ΔslmAMC1000 ttk::cmlCmlThis work
 MC1000ΔzapAMC1000 ygfE::cmlCmlThis work
 MC1000ΔminDMC1000 minD::kanKanThis work
 MC1000ΔzapBΔslmAMC1000 carrying the zapB deletion and ttk::cmlCmlThis work
 MC1000ΔzapBΔzapAMC1000 carrying the zapB deletion and ygfE::cmlCmlThis work
 MC1000ΔzapBΔminDMC1000 carrying the zapB deletion and minD::kanKanThis work
 KG22C600 lacIqlacZΔM15 Laboratory collection
 KG61lacZYA+lacI-recA::Tn10TcLaboratory collection
 KG22ΔzapBKG22 in which the zapB gene is deleted This work
 KG22Z84KG22 carrying the ftsZ84 mutationTcThis work
 KG22Z84ΔzapBKG22 carrying the ftsZ84 mutation and the zapB deletionTcThis work
 MG1655wt Xiao et al. (1991)
 BTH101F-, cya-99 araD139 galE15 galK16 rpsL1 hsdR2 μrA1 μrB1 Karimova et al. (1998)
 BTH101ΔzapBBTH101 strain carrying the zapB deletion This work
 AB1157Thr-1 leuB6 hisG4 proA2 argE3 thi-1 lacy1 galK2 ara-14 xyl-5 mtl-1 tsx-33 supE44 rpsL31 Bachmann (1972)
 AB1157Z84AB1157 carrying the ftsZ84 mutationTcGift from David J. Sherratt
 PB114minD::kanKande Boer et al. (1989)
 W3110wt Pichoff and Lutkenhaus (2002)
 PS223W3110 zipA1 Pichoff and Lutkenhaus (2002)
 PS236W3110 ftsA12 leu::Tn10TcPichoff and Lutkenhaus (2002)
 PS234W3110 zipA1 ftsA12 leu::Tn10TcPichoff and Lutkenhaus (2002)
 BL21AIArabinose-inducible gene expression from T7 promoter Invitrogen
 pGE2R1 LacZYA+par2+AmpEbersbach and Gerdes (2001)
 pGE271R1 lacIq+par2+AmpThis work
 pJV2R1 lacIq+hok-sok+AmpLaboratory collection
 pDD248R1 lacIq+parMRC+AmpLaboratory collection
 pGE73zapB+rraA+menA+hslU+CmlThis work
 pGE73-1zapB+rraA+menA+CmlThis work
 pGE73-2zapB+hslU+CmlThis work
 pGE73-3zapB+CmlThis work
 pGE73-4zapB+CmlThis work
 pGE620Plac::zapB::gfpAmpThis work
 pGE621Plac::gfp::zapBAmpThis work
 pGE622Plac::zapB::gfpTcThis work
 pGE624Plac::gfp::zapBTcThis work
 pGE626Plac::gfp::zapBAmpThis work
 pGE604PBAD::zapBCmlThis work
 pEG3aPBAD::zapB::gfpCmlThis work
 pUT18C-ParAPlac::T18::parAAmpEbersbach et al., 2006)
 pUT18C-ParBPlac::T18::parBAmpEbersbach et al., 2006)
 pUT18C-FtsZPlac::T18::ftsZAmpEbersbach et al., 2006)
 pUT18C-ZapBPlac::T18::zapBAmpThis work
 pUT18C-ZipPlac::T18::zipAmpKarimova et al., 1998)
 pUT18C-ZipAPlac::T18::zipAAmpThis work
 pUT18C-FtsAPlac::T18::ftsAAmpKarimova et al. (2005)
 pUT18C-FtsBPlac::T18::ftsBAmpKarimova et al. (2005)
 pUT18C-FtsIPlac::T18::FtsIAmpKarimova et al. (2005)
 pUT18C-FtsLPlac::T18::ftsLAmpKarimova et al. (2005)
 pUT18C-FtsNPlac::T18::ftsNAmpKarimova et al. (2005)
 pUT18C-FtsQPlac::T18::ftsQAmpKarimova et al. (2005)
 pUT18C-FtsWPlac::T18::ftsWAmpKarimova et al. (2005)
 pUT18C-FtsXPlac::T18::ftsXAmpKarimova et al. (2005)
 pUT18C-YmgFPlac::T18::ymgFAmpKarimova et al. (2005)
 pT25-ZipAPlac::T25::zipACmlThis work
 pKT25-FtsAPlac::KT25::ftsAKanKarimova et al. (2005)
 pKT25-FtsBPlac::KT25::ftsBKanKarimova et al. (2005)
 pKT25-FtsIPlac::KT25::FtsIKanKarimova et al. (2005)
 pKT25-FtsLPlac::KT25::ftsLKanKarimova et al. (2005)
 pKT25-FtsNPlac::KT25::ftsNKanKarimova et al. (2005)
 pKT25-FtsQPlac::KT25::ftsQKanKarimova et al. (2005)
 pKT25-FtsWPlac::KT25::ftsWKanKarimova et al. (2005)
 pKT25-FtsXPlac::KT25::ftsXKanKarimova et al. (2005)
 pKT25-YmgFPlac::KT25::ymgFKanKarimova et al. (2005)
 pT25-FtsZPlac::T25::ftsZCmlEbersbach et al. (2006)
 pT25-ZapBPlac::T25::zapBCmlThis work
 pLAU80PBAD::ftsZ::yfpAmpLau et al. (2003)
 pUT18CT18 fusion vectorAmpKarimova et al. (1998)
 pT25T25 fusion vectorCmlKarimova et al. (1998)
 pKG325Cloning vectorCml, TcGerdes and Molin (1986)
 pNDM220Cloning vector, contains Plac promoterAmpGotfredsen and Gerdes (1998)
 pBAD33Cloning vector, contains PBAD promoterCmGuzman et al. (1995)
 pKD13Template plasmid for construction of zapB deletion strainAmp, KanDatsenko and Wanner (2000)
 pKD3Template plasmid for construction of zapA and slmA deletion strainsAmp, CmlDatsenko and Wanner (2000
 pCP20FLP helper plasmidAmpDatsenko and Wanner (2000)
 Temperature-sensitive replicon  
 pKD46λ Red recombinase expression plasmidAmpDatsenko and Wanner (2000)
 Temperature sensitive replicon  
 pHis17Expression plasmid for C-terminal his-tagging. Contains the T7 promoterAmpvan den Ent and Löwe (2000)
 pZapB-hisPT7::zapB::His6AmpThis work

Cephalexin inhibits cell division by inhibiting PBP3 that is required for septal peptidoglycan synthesis (Hedge and Spratt, 1985). However, cephalexin does not prevent the formation of nascent Z rings (Pogliano et al., 1997). We inspected ZapB–GFP in wt and ftsZ84 cells treated with cephalexin. After about 2 h at 30°C, both wt and ftsZ84 cells contained band-like structures of ZapB–GFP (Fig. 4A, e and Fig. 4B, e). The wt and ftsZ84 strains responded differently to cephalexin, as the former contained many cells with three ZapB–GFP bands located in internucleoid regions (Fig. 4A, e) while cells of the ftsZ84 strain never had more than one band located at mid-cell (Fig. 4B, e). The reason for this difference is not known. However, it has been shown that cephalexin treatment delays Z ring assembly at future but not nascent division sites. Hence, it is possible that cells carrying ftsZ84 have fewer nascent division sites than wt cells as a result of a reduced level of FtsZ activity (Pogliano et al., 1997).

FtsZ forms arcs and helix-like structures in the zapB deletion strain

Next, we investigated FtsZ localization in the ΔzapB strain. Figure 5A shows localization of FtsZ-YFP in the wt strain. Consistent with previous reports, the Z ring appeared as a band or two spots at mid-cell (Fig. 5A, a–c). In cells with a clear constriction, FtsZ appeared more like a focus (Fig. 5A, d). Figure 5B shows FtsZ-YFP in the ΔzapB strain. In agreement with the results described above (Fig. 2), the cells were elongated. Compared with wt cells, it was generally more difficult to obtain good fluorescence microscopy images of the FtsZ-YFP signal in the ΔzapB cells even when the exposure time was increased. Apparently, this was due to the fact that only a minor population of cells contained a clear FtsZ-YFP band or focus at mid-cell (Fig. 5B, a and d). In most cells, the FtsZ signal appeared more diffuse, forming multiple weak band-like structures across the short cell axis. In some cells, these weaker structures appeared helical. The FtsZ structures were either centred around mid-cell (Fig. 5B, b and c) or covering a larger fraction of the cell and often overlapping the nucleoids (Fig. 5B, e–g). However, Z rings were only rarely found at the cell poles. In the longer cells, the FtsZ-YFP bands appeared to be present in doublets or triplets distributed along the length of the cell (Fig. 5B, h and i). These results raise the possibility that ZapB might play a role in the regulation of Z ring formation.


Figure 5. FtsZ forms helices in ZapB-deficient cells. Subcellular localization of FtsZ-YFP and nucleoids (DAPI staining) in living cells of strains (A) MC1000 (wt) and (B) MC1000ΔzapB respectively. (A, a–d) shows FtsZ-YFP while (A, a′–d′) shows the DAPI signal of the same cells. Similar for (B, a–i) and (B, a′–i′). Cells were grown at 30°C in M9 glucose medium supplemented with casamino acids. Expression of FtsZ-YFP was induced by addition of arabinose as described in Experimental procedures.

Download figure to PowerPoint

Synthetic detrimental phenotype of ΔzapB and ftsZ84

The above-described results suggested that ZapB enhances FtsZ activity by facilitating Z ring formation. Therefore, we tested if the combination of the ftsZ84 and ΔzapB mutations would have a synthetic phenotype. Overnight colonies of an ftsZ84ΔzapB double-mutant strain were tiny and grew significantly slower than wt or single-mutant cells. In liquid medium, the double-mutant strain generated cells with highly abnormal morphologies, notable DNA-free mini-cells, blebs and branches (arrows in Fig. 6D). The ΔzapB and ftsZ84 single mutants also formed abnormal cells, but with considerably lower frequencies (Table 3; Fig. 6B and C). As expected, wt cells did not form aberrant morphologies to any significant extent (Fig. 6A). In a similar way, we combined ΔzapB with either Δmin, ΔslmA, ΔzapA or ftsA2ts and analysed the growth characteristics and morphologies of the corresponding cells. In none of these cases did we obtain synthetic sick or lethal phenotypes (data not shown).


Figure 6. Combination of ftsZ84 and ΔzapB mutations results in a synthetic detrimental phenotype. Cells of strains (A) KG22, (B) KG22ΔzapB, (C) KG22Z84 and (D) KG22Z84ΔzapB were grown at the permissive temperature (30°C) in M9 minimal medium supplemented with 0.2% glucose and 0.2% casamino acids and visualized by fluorescence microscopy. Nucleoids were stained with DAPI and the cells were mounted on glass slides covered with a thin layer of 1% agarose (in growth medium). Pictures are combined phase-contrast and fluorescence (DAPI) microscopy images. Arrows in C and D mark cells with aberrant cell shapes and mini-cells. Scale bar = 2 μm. E. Diagram showing the cell length distribution of cells in A–D, a total of 250 cells of each strain were measured.

Download figure to PowerPoint

Table 3.  Combining ftsZ84 and ΔzapB results in morphological defects.a
StrainAbnormal morphology (%, n)bMini-cells (%)
  • a.

    Cells were grown under permissive conditions (M9 minimal medium supplemented with glucose and casamino acids at 30°C). Under these conditions, the activity of FtsZ84 is sufficient to allow normal cell division in the majority of the cells.

  • b.

    All defects in shape visible in the light microscope, including branches, protrusions, bends and blebs. Cell elongations were not included.

wt0.1 (1073)0
ΔzapB0.8 (1118)0.2
Z843.5 (1007)0.9
Z84ΔzapB8.0 (1138)3.5

ZapB encodes an homodimeric, anti-parallel coiled-coil protein

The putative product of zapB is an 81 aa protein with a predicted coiled-coil secondary structure (Fig. S3A). Homologues of zapB were identified in a subset of the γ proteobacteria, including Salmonella typhimurium, Haemophilus influenzae and Vibrio cholerae (Fig. S3B). Using Western analysis, we determined that ZapB is present in approximately 13 000 copies per cell (Fig. S4). In E. coli B/r, FtsZ is present between 3200 and 15 000 molecules per cell (Lu et al., 1998; Rueda et al., 2003). Thus ZapB is present at a level comparable to or higher than that of FtsZ.

Using X-ray crystallography, we confirmed the predicted coiled-coil structure of ZapB. Figure 7A shows the structure of a ZapB dimer in side view and end-on. The structure consists of two alpha-helical polypeptide chains arranged in anti-parallel to form a coiled-coil of 116 Å. The amino acid residues are coloured in blue to red going from the N terminus to C terminus. ZapB dimers crystallized in space group P1 with two dimers in the asymmetric unit. The crystal packing of ZapB shown in Fig. 7B shows how coiled-coil dimers may form filaments through interactions near their termini. Adjacent dimers in the crystals are shifted by 80 Å along the polymer axis and interact with their ends.


Figure 7. Crystal structure of E. coli ZapB, forming an anti-parallel coiled-coil with all ordered residues participating in the coil structure. A. Crystal structure of an E. coli anti-parallel coiled-coil dimer. B. The crystal packing of ZapB shows how individual dimers may assemble into protofilaments. The left panel is tilted by 90° compared with the panel on the right. The cross-over points of the helices in A and B appear different because the molecules are rotated around their long axes. C–E. Electron microscopic imaging of negatively stained purified ZapB showing ZapB protein fibres that form spontaneously under the conditions given. ZapB was incubated in (C) buffer A (50 mM MES; 100 mM KCl; 5 mM MgCl2) or (D) in buffer A + 10 mM CaCl2 or (E) buffer A + 10 mM EDTA before application to EM grids and negative staining with 0.5% uranyl acetate. In C, D and E, scale bars are 100 nm.

Download figure to PowerPoint

Cell lysates containing overexpressed ZapB were unusually viscous, and purified ZapB formed a gel in solution making concentration of the protein beyond 2 mg ml−1 very difficult. We analysed purified ZapB by EM and found that the protein assembles into filaments under all conditions tested. In the presence of divalent cations, such as magnesium or calcium, ZapB further assembled into larger cables (Fig. 7C). By further addition of calcium, the cables assembled into a semi-crystalline bundle with a characteristic regularly striated pattern in the negative stain (Fig. 7D). Conversely, by addition of EDTA, the cables disintegrated into single protofilaments (Fig. 7E), indicating that divalent cations mediate intimate association of ZapB protofilaments.


  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

We show here that ZapB of E. coli is required for proper Z ring formation. The appearance of ectopic, helical Z rings in ZapB-deficient cells (Fig. 5) shows that ZapB is required for normal Z ring assembly and yields an explanation for the increased average cell length (Fig. 2). A ZapB–GFP fusion was functional and revealed a dynamic pattern that depended on FtsZ but not on FtsA, ZipA or FtsI (Figs 3 and 4). Importantly, ZapB required either functional FtsA or functional ZipA to be recruited to the Z ring, just like Z ring assembly itself (Pichoff and Lutkenhaus, 2002). Combined with the BTH data, our cytological observations indicate that ZapB is recruited early to the divisome by direct interaction with FtsZ, stimulates Z ring assembly and thereby promote cell division earlier in the cell cycle. Interestingly, a recent high-resolution electron cryotomography analysis showed that the Z ring consists of short (∼100 nm) filaments irregularly spaced near the division site (Li et al., 2007). Thus, it is possible that ZapB functions to bring together short Z filaments into a higher-order ring-like structure that can contract.

ZapB overproduction resulted in nucleoid condensation (Fig. 1E), formation of cell chains containing mini-cells and anucleate mini-cells. By contrast, deletion of zapB did not affect nucleoid morphology or its segregation (Fig. S2). Thus, we suspect that nucleoid condensation by ZapB overproduction may not be directly related to its biological function. The nucleoid condensation effect may, however, explain why we picked up the zapB gene in the plasmid destabilization assay. The par2 locus of pB171 encodes the ParA ATPase that oscillates on the nucleoid in spiral-shaped structures (Ebersbach and Gerdes, 2001; 2004). par2 positions plasmids regularly over the nucleoid in a process depending on ParA (Ebersbach et al., 2006). Neither deletion of zapB nor overproduction of ZapB resulted in any detectable reduction in the efficiency of par2-mediated plasmid stabilization in liquid media (data not shown). Thus, ZapB does not seem to play a pivotal role in plasmid partitioning. The multi-copy suppression assay (Fig. 1) is very sensitive and we infer that zapB was picked up as a result of a marginal effect on par2 efficiency that may be an indirect effect caused by ZapB-mediated nucleoid condensation.

As described in the Introduction, several proteins regulate Z ring assembly. Even though ZapB does not show homology to any of these proteins, ZapB does appear to share some of their functional features and phenotypes. For example, like the negative regulator EzrA and the positive regulator ZapA, ZapB is a non-essential protein that interacts with itself and localizes to the septum in an FtsZ-dependent manner (Levin et al., 1999; Gueiros-Filho and Losick, 2002; Haeusser et al., 2004; Low et al., 2004). Like the zapB deletion, an ezrA deletion also makes extra FtsZ rings (Levin et al., 1999). However, in the ΔezrA strain, the extra Z rings were located near the cell poles (Levin et al., 1999) whereas this was clearly not the case in ΔzapB cells. ZapA and ZapB of E. coli both promote FtsZ assembly, and deletion of either gene yields synthetic sick cells under conditions where the activity of FtsZ is reduced (Gueiros-Filho and Losick, 2002). It will be interesting to learn if ZapA and ZapB have similar functions or if they promote FtsZ assembly by different mechanisms.

In addition to ZapA and EzrA, B. subtilis has yet a third non-essential cell division factor called SepF (YlmF) (Hamoen et al., 2006; Ishikawa et al., 2006). SepF is also recruited to the divisome in an FtsZ-dependent manner. SepF may act early (Ishikawa et al., 2006) or late in septum formation (Hamoen et al., 2006). Interestingly, deletion of sepF is synthetic lethal with both ezrA and ftsA (Hamoen et al., 2006; Ishikawa et al., 2006). Like ZapB, SepF may function in Z ring formation.

The structure of ZapB presented here shows 100% coiled-coil without any globular domains. The coiled-coil is anti-parallel. The 100% coiled-coil content is in contrast to the other FtsZ-modulating proteins (ZapA, EzrA and SepF) where one could imagine the globular domains to be the interaction domains. Whatever ZapB is interacting with, it will have to interact with the coiled-coil or with the few residues disordered in the crystal structure. ZapB in vitro tends to form oligomers as demonstrated by electron microscopy. The significance of this is not known, but it raises the possibility that the protein forms polymers in the cell, especially as the copy number of ZapB in the cell is relatively high (≈13 000 per cell), present in approximately two to four copies per FtsZ monomer (Pla et al., 1991; Rueda et al., 2003). The crystal packing suggests how the polymerization is accomplished by the ends of the coiled-coil dimer binding to each other. However, we could not relate any of the distances found in the EM pictures with those observed in the crystals. This might be due to a fundamental difference in the way these interactions form or it might reflect an offset rotation/orientation of the polymer in the EM pictures. Further studies will be needed to verify the polymeric nature of ZapB in cells.

We show here that E. coli encodes a novel cell division factor, ZapB that is required to maintain a normal cell length, and that ZapB acts at mid-cell to promote Z ring assembly. The discovery of ZapB adds a novel component to the complex cell division machinery. There are several possibilities for how ZapB might stimulate Z ring assembly. The most straightforward possibility that is supported by our two-hybrid analysis and consistent with the in vitro filamentation and bundling is that ZapB connects smaller assemblies of FtsZ into the mature Z ring simply via a bridging mechanism. Although our cytological data point to a direct interaction, we can not rule out that FtsZ and ZapB interact indirectly. We are now performing in vitro experiments to directly address this question.

Experimental procedures

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

Strains and plasmids are described in the supplementary online material and oligonucleotides are listed in Table S4.

Fluorescence microscopy

Live cells (2–3 μl) were mounted on microscope slides covered with a thin layer of 1% agarose solution (in growth medium or water). When required, cells were stained with 1 μg ml−1 DAPI before mounting. Fixated cells (20–30 μl) were placed on an 8-well microscope slide coated with Poly L-Lysine (Sigma). Slides were mounted with Slow-Fade (Molecular Probes) and 1 μg ml−1 DAPI before microscopy. Microscopy was carried out by using a 100× Leica PL APO phase-contrast objective on a Leica DMRA fluorescence microscope, equipped with a Leica DC500 colour CCD camera. Images were acquired and analysed by using the Leica IM500 computer software. Image processing was carried out by using Corel Photo-Paint and Corel Draw 10 software.

Localization of ZapB–GFP

Cells of strain MC1000 carrying the ZapB–GFP fusion plasmid pEG3a were grown at 37°C in M9 minimal medium supplemented with 0.2% glucose, 1 μg ml−1 thiamine, 0.1% casamino acids and 30 μg ml−1 cloramphenicol. The generation time of the strain under this growth condition was ∼45–50 min. At an OD450 value of about 0.2, the cells were harvested by centrifugation followed by resuspension in pre-heated medium without glucose. 0.2% arabinose was added in order to induce expression of ZapB–GFP. After 5 min, induction was stopped by adding 0.2% glucose. The cells were allowed to grow for at least another 30 min before the samples were collected for microscopy.

Growth of zapB deletion strains

The growth of strains KG22ΔzapB and MC1000ΔzapB in addition to the corresponding wt strains were tested under three different growth conditions: LB at 37°C, LB at 30°C and M9 minimal medium supplemented with 0.2% glucose, 1 μg ml−1 thiamine and 0.1% casamino acids at 30°C. Overnight cultures were diluted 500-fold into fresh medium. To follow cell growth, samples were collected at regular intervals for optical density OD450 measurements. At OD450 values of about 0.4 and 2.0, cells were mounted for microscopy.

Localization of FtsZ-YFP in zapB deletion strains

Cells of strains MC1000 and MC1000ΔzapB carrying the FtsZ-YFP expression plasmid pLAU80 were grown in M9 minimal medium supplemented with 0.2% glucose, 1 μg ml−1 thiamine, 0.1% casamino acids and 100 μg ml−1 ampicillin at 30°C. The generation time under these growth conditions was ∼50 min. Overnight cultures were diluted about 150-fold into fresh medium. At an OD450 value of about 0.2, the cells were harvested by centrifugation followed by resuspension in pre-heated medium without glucose. 0.2% arabinose was added in order to induce expression of FtsZ-YFP. After 6–8 min, induction was stopped by addition of 0.2% glucose. The cells were allowed to grow for another 60–80 min before samples were collected for microscopy.

Localization of ZapB–GFP in an ftsZ84(ts) strain

Cells of strains KG22 and KG22Z84 carrying the ZapB–GFP expression plasmid pEG3a were grown at the permissive temperature (30°C) in M9 minimal medium supplemented with 0.2% glucose, 1 μg ml−1 thiamine, 0.1% casamino acids and 30 μg ml−1 cloramphenicol. The generation time of strains under these growth conditions was ∼60 min. Overnight cultures were diluted 200 times into fresh medium. At an OD450 value of about 0.2, the cells were harvested by centrifugation followed by resuspension in pre-heated medium without glucose. 0.2% arabinose was added in order to induce expression of ZapB–GFP. After 5 min, induction was stopped by addition of 0.2% glucose. A sample of each culture was then withdrawn and cephalexin was added to a final concentration of 10 μg ml−1. What remained of each culture was left to grow for another 60 min at 30°C before it was divided into two: one was diluted twofold into fresh medium at 30°C and one was diluted twofold into fresh medium pre-heated to non-permissive temperature (42°C). Samples were collected from KG22/pEG3a and KG22Z84/pEG3a cultures, both at 30°C and 42°C after 45 min, whereas from the cephalexin-treated cultures, samples were taken after 90 min. Samples of KG22Z84/pEG3a at 30°C and 42°C were immediately fixed with 2% formaldehyde and 0.2% glutaraldehyde in 32 mM sodium phosphate buffer pH 7.5 for 20 min at room temperature followed by 40 min on ice. The fixed cells were washed three times in PBS (140 mM NaCl, 3 mM KCl, 8 mM Na2HPO4, 1.5 mM KH2PO4) and finally dissolved in 100–300 μl of GTE (50 mM glucose, 20 mM Tris-HCl pH 7.5, 10 mM EDTA).

Localization of ZapB–GFP in PS223, PS234, PS236

Cells of strains PS223 [zipA1(ts)], PS234 [zipA1(ts) ftsA12(ts)] and PS236 [ftsA12(ts)] carrying the ZapB–GFP expression plasmid pEG3a were grown at the permissive temperature (30°C) in LB medium. Overnight cultures were diluted 200 times into fresh medium. At an OD450 of ≈0.2, 0.2% arabinose was added in order to induce expression of ZapB–GFP and, after 5 min, induction was stopped by addition of 0.2% glucose. Each culture was left to grow for another 60 min at 30°C before it was divided into two: one was diluted twofold into fresh medium at 30°C and one was diluted twofold into fresh medium pre-heated to non-permissive temperature (42°C). Samples were collected both at 30°C and 42°C after 60 min and immediately fixed, as previously described.

Identification of zapB as a gene that specifically destabilizes a par2-carrying plasmid

We used the colony ‘split’ assay devised by the Cohen and Hiraga laboratories (Biek and Cohen, 1986; Niki et al., 1988) to screen for genes that, when present on a high-copy-number plasmid, would interfere with the function of the par2 locus of plasmid pB171. The screen was carried out using strain KG61 harbouring pGE271, an R1 test plasmid carrying par2 of pB171 (Table 3 and Fig. 1A). KG61 carries the lacZYA operon but lacks a functional lacI. As pGE271 carries lacIq, KG61/pGE271 cells form uniform white colonies on Xgal plates. However, conditions that destabilize pGE271 result in the formation of blue-sectored colonies on Xgal media. We screened a chromosomal E. coli gene library of MG1655 established as Sau3AI DNA fragments of sizes 3500–8000 bp in a pBR325-derived vector pKG325. Approximately 36 000 colonies were screened and 82 of these were either blue (56) or white with blue sectors (26) (see Table S1).

The 82 positive clones obtained in the screen were purified by two sequential single-colony isolations; furthermore, plasmid DNA was prepared. To test for reproducibility and to eliminate false positive arising from spontaneous plasmid loss or spontaneous mutations in, for example, the lacIq gene on the test plasmid or the lacZ gene on the chromosome, plasmid DNA of the 82 clones was re-transformed in duplicate experiments into strain KG61/pGE271. In addition, the plasmid DNA was transformed into a strain of KG61 harbouring another R1 test plasmid, pJV2. Like pGE271, pJV2 contains lacIq but, instead of par2, this plasmid is stabilized by the hok/sok gene locus, which promotes plasmid stability by killing of plasmid-free cells. Transformation into this strain enabled us to exclude clones that mediated plasmid instability by interfering with parameters other than par2 that influence the stability of the test plasmid, the most obvious being plasmid replication.

Table S1 shows the results of the re-transformation experiments. Of the 82 original clones, 44 did not destabilize pGE271 when re-transformed into strain KG61/pGE271. Another 36 clones resulted in uniformly blue colonies (33) or sectored colonies (3) in both the KG61/pGE271 and the KG61/pJV2 strains. Thus, only two potentially interesting clones remained. One was clone number 73 (pGE73), which apparently had the desired phenotype, i.e. the capability to destabilize only the par2+ plasmid (blue sectors in almost all colonies) (Fig. 1B, a) but not the hok/sok plasmid (white colonies) (Fig. 1B, b). The other clone, number 79 (pGE79), was less convincing as this construct gave blue sectors in colonies of both the pGE271-harbouring and the JV2-containing strains. However, because of an apparent predominance, at least in some experiments, of sectored colonies in the former strain compared with the latter, clone number 79 was also selected for one final transformation test, this time against a third test plasmid, pDD248, carrying the natural parMRC system of R1. In contrast to par2, which belongs to the family of par systems expressing deviant Walker-type ATPases, the unrelated parMRC system of R1 encodes an actin-like ATPase. The two par systems both act to stabilize plasmids by specific subcellular positioning, but probably by different means. Furthermore, there is no indication that the parMRC system requires host-encoded factors. Thus, pDD248 was chosen as a final test for par2 specificity. pGE79 evidently destabilized pDD248 as efficiently as it destabilized pGE271 and was therefore eliminated. Plasmid pGE73, on the other hand, did not destabilize plasmid pDD248 (Table S1).


  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

We thank Leendert Hamoen for critical reading of the manuscript, Jeff Errington for stimulating discussions and Joe Lutkenhaus for donating strains. We would like to thank staff at beamline ID14eh4 (ESRF, Grenoble, France) for excellent support during data collection. This work was supported by the Danish Natural Science Foundation (SNUF/FNU) by a grant to K.G. J.M.J. was supported by an EMBO long-term fellowship.


  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information
  • Aarsman, M.E., Piette, A., Fraipont, C., Vinkenvleugel, T.M., Nguyen-Disteche, M., and Den, B.T. (2005) Maturation of the Escherichia coli divisome occurs in two steps. Mol Microbiol 55: 16311645.
  • Addinall, S.G., Bi, E., and Lutkenhaus, J. (1996) FtsZ ring formation in fts mutants. J Bacteriol 178: 38773884.
  • Addinall, S.G., Cao, C., and Lutkenhaus, J. (1997) Temperature shift experiments with an ftsZ84 (ts) strain reveal rapid dynamics of FtsZ localization and indicate that the Z ring is required throughout septation and cannot reoccupy division sites once constriction has initiated. J Bacteriol 179: 42774284.
  • Bachmann, B.J. (1972) Pedigrees of some mutant strains of Escherichia coli K-12. Bacteriol Rev 36: 525557.
  • Bernhardt, T.G., and De Boer, P.A. (2003) The Escherichia coli amidase AmiC is a periplasmic septal ring component exported via the twin-arginine transport pathway. Mol Microbiol 48: 11711182.
  • Bernhardt, T.G., and De Boer, P.A. (2004) Screening for synthetic lethal mutants in Escherichia coli and identification of EnvC (YibP) as a periplasmic septal ring factor with murein hydrolase activity. Mol Microbiol 52: 12551269.
  • Bernhardt, T.G., and De Boer, P.A. (2005) SlmA, a nucleoid-associated, FtsZ binding protein required for blocking septal ring assembly over Chromosomes in E. coli. Mol Cell 18: 555564.
  • Bi, E., and Lutkenhaus, J. (1993) Cell division inhibitors SulA and MinCD prevent formation of the FtsZ ring. J Bacteriol 175: 11181125.
  • Bi, E.F., and Lutkenhaus, J. (1991) FtsZ ring structure associated with division in Escherichia coli. Nature 354: 161164.
  • Biek, D.P., and Cohen, S.N. (1986) Identification and characterization of recD, a gene affecting plasmid maintenance and recombination in Escherichia coli. J Bacteriol 167: 594603.
  • Bigot, S., Corre, J., Louarn, J.-M., Cornet, F., and Barre, F.X. (2004) FtsK activities in Xer recombination, DNA mobilization and cell division involve overlapping and separate domains of the protein. Mol Microbiol 54: 876886.
  • De Boer, P.A.J., Crossley, R.E., and Rothfield, L.I. (1989) A division inhibitor and a topological specificity factor coded for by the minicell locus determine proper placement of the division septum in E. coli. Cell 56: 641649.
  • Casadaban, M.J., and Cohen, S.N. (1980) Analysis of gene control signals by DNA fusion and cloning in Escherichia coli. J Mol Biol 138: 179207.
  • Corbin, B.D., Wang, Y., Beuria, T.K., and Margolin, W. (2007) Interaction between cell division proteins FtsE and FtsZ. J Bacteriol 189: 30263035.
  • Cordell, S.C., Robinson, E.J.H., and Löwe, J. (2003) Crystal structure of the SOS cell division inhibitor SulA and in complex with FtsZ. Proc Natl Acad Sci USA 100: 78897894.
  • Datsenko, K.A., and Wanner, B.L. (2000) One-step inactivation of chromosomal genes in Escherichia coli K-12 using PCR products. Proc Natl Acad Sci USA 97: 66406645.
  • Diez, A.A., Farewell, A., Nannmark, U., and Nyström, T. (1997) A mutation in the ftsK gene of Escherichia coli affects cell-cell separation, stationary-phase survival, stress adaptation, and expression of the gene encoding the stress protein UspA. J Bacteriol 179: 58785883.
  • Ebersbach, G., and Gerdes, K. (2001) The double par locus of virulence factor pB171: DNA segregation is correlated with oscillation of ParA. Proc Natl Acad Sci USA 98: 1507815083.
  • Ebersbach, G., and Gerdes, K. (2004) Bacterial mitosis: partitioning protein ParA oscillates in spiral-shaped structures and positions plasmids at mid-cell. Mol Microbiol 52: 385398.
  • Ebersbach, G., Ringgaard, S., Moller-Jensen, J., Wang, Q., Sherratt, D.J., and Gerdes, K. (2006) Regular cellular distribution of plasmids by oscillating and filament-forming ParA ATPase of plasmid pB171. Mol Microbiol 61: 14281442.
  • Van Den Ent, F., and Löwe, J. (2000) Crystal structure of the cell division protein FtsA from Thermotoga maritima. EMBO J 19: 53005307.
  • Errington, J., Daniel, R.A., and Scheffers, D.J. (2003) Cytokinesis in bacteria. Microbiol Mol Biol Rev 67: 5265. table.
  • Gerdes, K., and Molin, S. (1986) Partitioning of plasmid R1. Structural and functional analysis of the parA locus. J Mol Biol 190: 269279.
  • Gerdes, K., Møller-Jensen, J., and Jensen, R.B. (2000) Plasmid and chromosome partitioning: surprises from phylogeny. Mol Microbiol 37: 455466.
  • Goehring, N.W., and Beckwith, J. (2005) Diverse paths to midcell: assembly of the bacterial cell division machinery. Curr Biol 15: R514R526.
  • Goehring, N.W., Gueiros-Filho, F., and Beckwith, J. (2005) Premature targeting of a cell division protein to midcell allows dissection of divisome assembly in Escherichia coli. Genes Dev 19: 127137.
  • Gotfredsen, M., and Gerdes, K. (1998) The Escherichia coli relBE genes belong to a new toxin-antitoxin gene family. Mol Microbiol 29: 10651076.
  • Gueiros-Filho, F., and Losick, R. (2002) A widely conserved bacterial cell division protein that promotes assembly of the tubulin-like protein FtsZ. Genes Dev 16: 25442556.
  • Guzman, L.M., Belin, D., Carson, M.J., and Beckwith, J. (1995) Tight regulation, modulation, and high-level expression by vectors containing the arabinose pBAD promoter. J Bacteriol 177: 41214130.
  • Haeusser, D.P., Schwartz, R.L., Smith, A.M., Oates, M.E., and Levin, P.A. (2004) EzrA prevents aberrant cell division by modulating assembly of the cytoskeletal protein FtsZ. Mol Microbiol 52: 801814.
  • Hamoen, L.W., De Meile, J.C.J.W., Noirot, P., and Errington, J. (2006) SepF, a novel FtsZ-interacting protein required for a late step in cell division. Mol Microbiol 59: 989999.
  • Hara, H., Narita, S., Karibian, D., Park, J.T., Yamamoto, Y., and Nishimura, Y. (2002) Identification and characterization of the Escherichia coli envC gene encoding a periplasmic coiled-coil protein with putative peptidase activity. FEMS Microbiol Lett 212: 229236.
  • Hedge, P.J., and Spratt, B.G. (1985) Amino acid substitutions that reduce the affinity of penicillin-binding protein 3 of Escherichia coli for cephalexin. Eur J Biochem 151: 111121.
  • Hu, Z., Mukherjee, A., Pichoff, S., and Lutkenhaus, J. (1999) The MinC component of the division site selection system in Escherichia coli interacts with FtsZ to prevent polymerization. Proc Natl Acad Sci USA 96: 1481914824.
  • Ishikawa, S., Kawai, Y., Hiramatsu, K., Kuwano, M., and Ogasawara, N. (2006) A new FtsZ-interacting protein, YlmF, complements the activity of FtsA during progression of cell division in Bacillus subtilis. Mol Microbiol 60: 13641380.
  • Jones, C.A., and Holland, I.B. (1984) Inactivation of essential division genes, ftsA, ftsZ, suppresses mutations at sfiB, a locus mediating division inhibition during the SOS response in E. coli. EMBO J 3: 11811186.
  • Justice, S.S., Garcia-Lara, J., and Rothfield, L.I. (2000) Cell division inhibitors SulA and MinC/MinD block septum formation at different steps in the assembly of the Escherichia coli division machinery. Mol Microbiol 37: 410423.
  • Karimova, G., Pidoux, J., Ullmann, A., and Ladant, D. (1998) A bacterial two-hybrid system based on a reconstituted signal transduction pathway. Proc Natl Acad Sci USA 95: 57525756.
  • Karimova, G., Dautin, N., and Ladant, D. (2005) Interaction network among Escherichia coli membrane proteins involved in cell division as revealed by bacterial two-hybrid analysis. J Bacteriol 187: 22332243.
  • Lau, I.F., Filipe, S.R., Soballe, B., Okstad, O.A., Barre, F.X., and Sherratt, D.J. (2003) Spatial and temporal organization of replicating Escherichia coli chromosomes. Mol Microbiol 49: 731743.
  • Levin, P.A., Kurtser, I.G., and Grossman, A.D. (1999) Identification and characterization of a negative regulator of FtsZ ring formation in Bacillus subtilis. Proc Natl Acad Sci USA 96: 96429647.
  • Li, Z., Trimble, M.J., Brun, Y.V., and Jensen, G.J. (2007) The structure of FtsZ filaments in vivo suggests a force-generating role in cell division. EMBO J 26: 46944708.
  • Liu, G., Draper, G.C., and Donachie, W.D. (1998) FtsK is a bifunctional protein involved in cell division and chromosome localization in Escherichia coli. Mol Microbiol 29: 893903.
  • Low, H.H., Moncrieffe, M.C., and Löwe, J. (2004) The crystal structure of ZapA and its modulation of FtsZ polymerisation. J Mol Biol 341: 839852.
  • Löwe, J., and Amos, L.A. (1998) Crystal structure of the bacterial cell-division protein FtsZ. Nature 391: 203206.
  • Lu, C., Stricker, J., and Erickson, H.P. (1998) FtsZ from Escherichia coli, Azotobacter vinelandii, and Thermotoga maritima – quantitation, GTP hydrolysis, and assembly. Cell Motil Cytoskeleton 40: 7186.
  • Lutkenhaus, J. (2002) Dynamic proteins in bacteria. Curr Opin Microbiol 5: 548552.
  • Marston, A.L., Thomaides, H.B., Edwards, D.H., Sharpe, M.E., and Errington, J. (1998) Polar localization of the MinD protein of Bacillus subtilis and its role in selection of the mid-cell division site. Genes Dev 12: 34193430.
  • Niki, H., Ichinose, C., Ogura, T., Mori, H., Morita, M., Hasegawa, M., et al. (1988) Chromosomal genes essential for stable maintenance of the mini-F plasmid in Escherichia coli. J Bacteriol 170: 52725278.
  • Pichoff, S., and Lutkenhaus, J. (2001) Escherichia coli division inhibitor MinCD blocks septation by preventing Z-ring formation. J Bacteriol 183: 66306635.
  • Pichoff, S., and Lutkenhaus, J. (2002) Unique and overlapping roles of ZipA and FtsA in septal ring assembly in Escherichia coli. EMBO J 21: 685693.
  • Pla, J., Sanchez, M., Palacios, P., Vicente, M., and Aldea, M. (1991) Preferential cytoplasmic location of FtsZ, a protein essential for Escherichia coli septation. Mol Microbiol 5: 16811686.
  • Pogliano, J., Pogliano, K., Weiss, D.S., Losick, R., and Beckwith, J. (1997) Inactivation of FtsI inhibits constriction of the FtsZ cytokinetic ring and delays the assembly of FtsZ rings at potential division sites. Proc Natl Acad Sci USA 94: 559564.
  • Reddy, M. (2007) Role of FtsEX in cell division of Escherichia coli: viability of ftsEX mutants is dependent on functional SufI or high osmotic strength. J Bacteriol 189: 98108.
  • Rothfield, L., Taghbalout, A., and Shih, Y.L. (2005) Spatial control of bacterial division-site placement. Nat Rev Microbiol 3: 959968.
  • Rueda, S., Vicente, M., and Mingorance, J. (2003) Concentration and assembly of the division ring proteins FtsZ, FtsA, and ZipA during the Escherichia coli cell cycle. J Bacteriol 185: 33443351.
  • Vicente, M., and Rico, A.I. (2006) The order of the ring: assembly of Escherichia coli cell division components. Mol Microbiol 61: 58.
  • Woldringh, C.L., Mulder, E., Huls, P.G., and Vischer, N. (1991) Toporegulation of bacterial division according to the nucleoid occlusion model. Res Microbiol 142: 309320.
  • Wu, L.J., and Errington, J. (2004) Coordination of cell division and chromosome segregation by a nucleoid occlusion protein in Bacillus subtilis. Cell 117: 915925.
  • Xiao, H., Kalman, M., Ikehara, K., Zemel, S., Glaser, G., and Cashel, M. (1991) Residual guanosine 3′,5′-bispyrophosphate synthetic activity of relA null mutants can be eliminated by spoT null mutations. J Biol Chem 266: 59805990.

Supporting Information

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information
MMI_6190_sm_Figures_S1-S4_and_Tables_S1-S4.pdf2984KSupporting info item

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.