SEARCH

SEARCH BY CITATION

Summary

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

Bacterial cytokinesis is orchestrated by an assembly of essential cell division proteins that form a supramolecular structure known as the divisome. DivIB and its orthologue FtsQ are essential members of the divisome in Gram-positive and Gram-negative bacteria respectively. DivIB is a bitopic membrane protein composed of an N-terminal cytoplasmic domain, a single-pass transmembrane domain, and a C-terminal extracytoplasmic region comprised of three separate protein domains. A molecular dissection approach was used to determine which of these domains are essential for recruitment of DivIB to incipient division sites and for its cell division functions. We show that DivIB has three molecular epitopes that mediate its localization to division septa; two epitopes are encoded within the extracytoplasmic region while the third is located in the transmembrane domain. It is proposed that these epitopes represent sites of interaction with other divisomal proteins, and we have used this information to develop a model of the way in which DivIB and FtsQ are integrated into the divisome. Remarkably, two of the three DivIB localization epitopes are dispensable for vegetative cell division; this suggests that the divisome is assembled using a complex network of protein–protein interactions, many of which are redundant and likely to be individually non-essential.


Introduction

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

Bacterial cytokinesis is a complex subcellular differentiation process that requires elaborate molecular mechanics as well as precise spatiotemporal regulation (Rothfield et al., 2001). During vegetative growth, Bacillus subtilis assembles a division apparatus at the cell centre that is responsible for physically dividing the rod-shaped parent bacterium into two identical daughter cells. In contrast, under conditions of severe nutrient deprivation, the parent cell forms a septum near one of the cell poles (Stragier and Losick, 1996). This leads to an asymmetric division that gives rise to a large mother cell and a smaller forespore. The mother cell engulfs the forespore and promotes its maturation, before eventually lysing to release the mature spore (Stragier and Losick, 1996).

The invagination and subsequent fusion of the cell membrane and cell wall layers during cytokinesis is orchestrated by a multiprotein complex known as the divisome. Divisome assembly is initiated by the formation of a circumferential ring of polymerized FtsZ, the prokaryotic precursor of tubulin, around the inner face of the plasma membrane (Bi and Lutkenhaus, 1991; Margolin, 2005). The Z ring may provide the contractile force for invagination (Romberg and Levin, 2003) and, in addition, it provides a molecular scaffold for recruitment of the remaining divisomal proteins (Errington et al., 2003; Weiss, 2004; Goehring and Beckwith, 2005). A series of predominantly cytoplasmic proteins (FtsA, ZipA and ZapA in Escherichia coli) are the first to be recruited to the Z ring; these proteins are responsible for tethering FtsZ to the membrane and stabilizing the Z ring. These early recruits are followed by a group of single- or multipass membrane proteins. Most of these are simple bitopic membrane proteins that contain a short cytoplasmic domain, a single transmembrane (TM) domain, and a larger extracytoplasmic domain (i.e. FtsL, FtsQ, FtsB, FtsI and FtsN in E. coli). Although it has been established that each of these proteins is essential for cell division, their biochemical functions, with the exception of FtsI, are essentially unknown (Goehring and Beckwith, 2005).

The B. subtilis divisomal protein DivIB localizes to midcell prior to septum formation and remains there throughout the entire invagination process. Although its E. coli orthologue FtsQ was discovered 27 years ago (Begg et al., 1980), the role of DivIB and FtsQ in cytokinesis remains enigmatic. The absence of these proteins from bacteria without cell walls implies a role in the synthesis or remodelling of septal peptidoglycan. Consistent with this hypothesis, it was recently demonstrated that divIB null cells form a polar septum with a thicker cell wall than wild-type cells during sporulation and fail to complete the forespore engulfment process (Thompson et al., 2006). Intriguingly, however, there are several significant differences between DivIB and FtsQ. FtsQ is essential in E. coli (Carson et al., 1991), while DivIB is only required for growth of B. subtilis at temperatures above 37°C (Beall and Lutkenhaus, 1989; Harry and Wake, 1989). E. coli contains only 25–50 copies of FtsQ per cell (Buddelmeijer et al., 1998) compared with 5000–13 000 copies of DivIB in B. subtilis (Katis et al., 1997; Rowland et al., 1997). Curiously, less than 2% of the cellular complement of DivIB is required for vegetative growth at temperatures ≤ 37°C (Rowland et al., 1997) (i.e. similar to the level of FtsQ in E. coli), whereas > 70% is required for efficient sporulation (Real et al., 2005).

DivIB is a bitopic membrane protein composed of a short cytoplasmic N-terminal domain (Cyto), a single TM segment and a larger 24 kDa extracytoplasmic region. Ablation of the chromosomal copy of either E. coli ftsQ or B. subtilis divIB can be complemented by chimeric constructs in which the native extracytoplasmic domain is fused to the Cyto-TM region of a heterologous bacterial membrane protein. For example, the Cyto-TM region of E. coli FtsQ can be replaced with the equivalent region from E. coli MalF (Guzman et al., 1997), while the Cyto-TM region of B. subtilis DivIB can be replaced with the corresponding region from E. coli TolR (Katis and Wake, 1999). These experiments suggest that although the Cyto-TM region of FtsQ and DivIB is presumably critical for anchoring these proteins to the cell membrane (Dopazo et al., 1992), it does not play a specific role in localizing them to the division site. However, septal localization of the DivIB chimera was not examined, and the FtsQ chimera does not localize to incipient division sites as efficiently as wild-type FtsQ (Chen et al., 1999). Thus, it remains uncertain which regions of FtsQ and DivIB are responsible for their septal localization and cell division functions.

It was recently demonstrated that the extracytoplasmic region of DivIB is comprised of three distinct domains: a membrane proximal α domain, a central β domain and a C-terminal γ domain that appears to be unstructured in the absence of other divisomal proteins (Robson and King, 2006) (Fig. 1A). This structural information has enabled us to perform a precise molecular dissection of DivIB in order to determine those regions of the protein that are critical for its vegetative cell division functions and for its recruitment to incipient division sites. We demonstrate that B. subtilis DivIB contains multiple septal localization signals that are located in the TM and extracytoplasmic regions of the protein, making it the first divisomal protein known to contain such a complex, multidentate localization signal. Moreover, we show that there is not a straightforward relationship between localization and cell division function; some DivIB domains have a septal localization signal but are nonetheless dispensable for vegetative cell division.

image

Figure 1. A. Schematic representation of the domain architecture of DivIB. Cyto and TM refer to the cytoplasmic and transmembrane domains respectively. The three extracytoplasmic domains are designated α, β and γ, with α proximal to the cell membrane. B. Cartoon showing the various GFP fusion constructs used in the current study. The numbers above the first construct correspond to the domain boundaries in B. subtilis DivIB.

Download figure to PowerPoint

Results

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

The transmembrane domain of DivIB contains a septal localization determinant

Molecular dissection is a technique that is often used to isolate motifs important for a protein's function and/or cellular localization. However, it is often difficult to interpret the results of such experiments because in many cases it is not known a priori whether N- or C-terminal truncations, point mutations or internal excisions of the protein are likely to perturb its three-dimensional structure. Our recent determination of the domain architecture of DivIB was thus critical for functional analysis as it enables us to make precise molecular dissections at domain boundaries which are much less likely to perturb the protein's fold than truncations within structural domains.

In a previous study it was demonstrated that a chimeric protein in which the Cyto-TM region of B. subtilis DivIB (DivIBBs) is replaced with the equivalent region from E. coli TolR retains full cell division function (Katis and Wake, 1999). This implies that only the extracytoplasmic domains of DivIBBs are essential for its septal localization. Thus, in order to determine which of these domains contain epitopes responsible for directing DivIB to incipient division sites, we constructed a panel of GFP fusion proteins containing various combinations of DivIBBs extracytoplasmic domains linked to the Cyto-TM region of DivIBBs (Fig. 1B).

In order to study the cellular localization of these GFP fusion proteins, we took advantage of the fact that DivIBBs is not essential for cell growth at temperatures below 37°C; all gfp–divIB constructs were inserted at the amyE locus of a divIB null strain under the control of an inducible Pxyl promoter and strains were grown at 30°C. Four hours after xylose induction, cells were examined using fluorescence microscopy. In this study, we considered GFP constructs to be localized to incipient division sites if fluorescence was visible at midcell in at least 80% of cells. We considered localization to be aberrant if less than 20% of cells showed fluorescence at midcell. As observed previously (Szeto et al., 2002), GFP alone does not localize to division sites but rather is uniformly distributed throughout the cell cytoplasm (Fig. 2H).

image

Figure 2. Fluorescence micrographs showing localization in B. subtilis of various constructs in which GFP was fused to the N-terminus of (A) CytoIB-TMIBIBβIBγIB (full-length DivIB), (B) CytoIB-TMIBIBγIB, (C) CytoIB-TMIBIBβIB, (D) CytoIB-TMIBIBγIB, (E) CytoIB-TMIBIB, (F) CytoIB-TMIBIB and (G) CytoIB-TMIBIB. (H) shows the localization of unfused GFP. The ‘IB’ subscript indicates that the domains are from DivIBBs. Septal localization is indicated by fluorescent stripes or dots at the centre of the long axis of the cell. In some cells that have recently divided (e.g. the cells marked with arrows in panels A and B) there is an obvious fluorescence at one of the cell poles, presumably resulting from incompletely disassembled divisomal complexes.

Download figure to PowerPoint

First, we examined the localization of full-length GFP–DivIBBs; we refer to this construct as GFP–CytoIB-TMIBIBβIBγIB, where IB indicates that the domain is from B. subtilis DivIB. This construct (which is fully functional; see below) localized to midcell in a similar pattern to that observed previously in immunofluorescence studies of wild-type DivIB (Harry and Wake, 1997; Fig. 2A). In addition, we observed strong fluorescence around the entire periphery of the cell, suggesting that much of the cellular complement of DivIB is not localized to incipient division sites.

Surprisingly, all of the other GFP constructs, each of which contained a deletion of one or two of the three extracytoplasmic domains, localized to incipient division sites (Fig. 2B–G). Septal localization was even evident in a construct consisting of the small 30-residue extracytoplasmic γ domain linked to the Cyto-TM region of DivIBBs (i.e. GFP–CytoIB-TMIBIB) (Fig. 2G). The midcell fluorescence was dimmer for this construct than each of the others even though Western blots using an anti-GFP antibody (data not shown) revealed that this fusion protein was not produced at significantly lower levels than the other GFP constructs. As all of the tested constructs contained at least one of the extracytoplasmic DivIBBs domains (i.e. α, β or γ), there are two potential interpretations of these data: (i) there is a septal localization determinant in each of the extracytoplasmic domains; and (ii) contrary to predictions in the literature, there is a localization determinant within the Cyto and/or TM domains of DivIBBs.

In order to determine which of these two hypotheses is correct, we examined whether a GFP–CytoIB-TMIB construct, which contains none of the extracytoplasmic domains of DivIBBs, could localize to incipient division sites. Surprisingly, we found that the Cyto-TM domains alone were sufficient for efficient septal localization (Fig. 3A), which indicates that a septal localization signal must be present in one or both of the Cyto and TM domains. Subsequent experiments revealed that a GFP–CytoIB construct was diffusely distributed throughout the cytoplasm (Fig. 3B) whereas a GFP–TMIB construct was clearly localized at midcell division sites (Fig. 3C). Thus, the TM domain of DivIB is sufficient to direct it to the septum. E. coli FtsI (an orthologue of B. subtilis PBP 2B) is the only other divisomal protein known to contain a TM localization determinant that is sufficient for targeting heterologous proteins such as GFP to cell division sites (Wissel et al., 2005).

image

Figure 3. Fluorescence micrographs showing localization in B. subtilis of GFP fused to the N-terminus of (A) CytoIB-TMIB, (B) CytoIB, (C) TMIB, (D) CytoTR-TMTR and (E) CytoTR-TMTRIBβIBγIB. The ‘IB’ and ‘TR’ subscripts indicate that the domains are from DivIBBs or E. coli TolR respectively. Septal localization is indicated by the fluorescent stripes or dots at the centre of the long axis of the cell in A, C and E.

Download figure to PowerPoint

The extracytoplasmic region of DivIB contains a septal localization signal

The presence of a septal localization signal in the TM region of DivIB explains why all members of the first panel of GFP–DivIB fusion proteins were able to localize to the septum (Fig. 2). However, it does not preclude the possibility that additional septal localization signals might be present in the extracytoplasmic region of DivIB. In order to address this possibility, we constructed a chimeric GFP fusion protein in which the entire extracytoplasmic region of DivIBBs was fused to the Cyto-TM region of the heterologous TolR protein from E. coli; we designate this fusion protein as GFP–CytoTR-TMTRIBβIBγIB, where the TR subscript indicates domains from TolR. Note that TolR localizes to the septal ring as part of the Tol–Pal complex in E. coli (Gerding et al., 2007). However, this septal localization requires the divisomal protein FtsN, which is not present in B. subtilis, as well the Tol–Pal components TolQ and possibly TolB, which are only present in Gram-negative bacteria (Gerding et al., 2007). Hence, we would not expect the TM region of TolR to localize to division sites in B. subtilis.

In contrast with the GFP–CytoIB-TMIB construct (Fig. 3A), a GFP–CytoTR-TMTR fusion protein was diffusely localized in the cytoplasm with no apparent localization to incipient division sites (Fig. 3D). The lack of peripheral fluorescence indicative of membrane localization is surprising, and it suggests that (i) the CytoTR-TMTR construct is unstable and/or (ii) the Cyto-TM region of TolR is not efficiently recognized and/or integrated into the cytoplasmic membrane by the B. subtilis membrane insertion machinery (Rubio et al., 2005). In contrast to GFP–CytoTR-TMTR, however, the GFP–CytoTR-TMTRIBβIBγIB chimera localized to midcell division sites (Fig. 3E) as efficiently as the non-chimeric GFP–CytoIB-TMIB–αIBβIBγIB construct (compare Fig. 3E with Fig. 2A). Thus, it is evident that the extracytoplasmic region of DivIB contains at least one signal for septal localization in addition to the septal localization signal present in the TM domain.

The extracytoplasmic region of DivIB contains multiple septal localization signals

In order to locate the septal localization signal(s) present in the extracytoplasmic region of DivIB, we constructed a second panel of domain deletion constructs that are analogous to those shown in Fig. 1B except that we replaced the Cyto-TM region of DivIBBs with the corresponding region from E. coli TolR (i.e. CytoTR-TMTR). In these constructs, any observed midcell localization must be due to a septal targeting signal located in one of the extracytoplasmic domains present in the construct.

All GFP constructs that contained the extracytoplasmic α domain of DivIBBs localized to midcell (Fig. 4A–C), including a construct that comprised only the α domain fused to the Cyto-TM region of TolR (i.e. CytoTR-TMTRIB) (Fig. 4C). Thus, the membrane-proximal α domain of DivIB must contain a septal localization signal. GFP constructs containing only the β domain (Fig. 4D) or only the γ domain (Fig. 4E) did not localize to midcell; curiously; however, a construct containing both the β and γ domains fused to the Cyto-TM region of TolR (i.e. CytoTR-TMTRIBγIB) showed weak but evident septal localization (Fig. 4F). Thus, in addition to the septal localization signal present in the α domain, there must be at least one septal localization determinant within the βγ region.

image

Figure 4. Fluorescence micrographs showing localization in B. subtilis of GFP fused to the N-terminus of (A) CytoTR-TMTRαIBβIB, (B) CytoTR-TMTRIBγIB, (C) CytoTR-TMTRIB, (D) CytoTR-TMTRβIB, (E) CytoTR-TMTRIB and (F) CytoTR-TMTRIBγIB. The ‘IB’ and ‘TR’ subscripts indicate that the domains are from DivIBBs or E. coli TolR respectively.

Download figure to PowerPoint

It cannot be ascertained from these experiments alone whether the second extracytoplasmic localization signal is located within the β or γ domain, or whether this epitope spans both domains. It is theoretically possible that both the β and γ domains contain septal localization signals but that neither the GFP–CytoTR-TMTRIB nor the GFP–CytoTR-TMTRIB constructs localize because these fusion proteins are unstable and are extensively degraded relative to the construct containing both domains (i.e. GFP–CytoTR-TMTRIBγIB). However, Western blot analysis of all strains expressing TolR-DivIB chimeras revealed similarly intense bands of the expected size (Table 1). Thus, it seems unlikely that the lack of septal localization of the GFP–CytoTR-TMTRIB and GFP–CytoTR-TMTRIB constructs is due to their inherent instability. Rather, we tentatively conclude that the second extracytoplasmic localization signal comprises an epitope spanning both the β and γ domains and that both domains must be present and correctly juxtaposed for this signal to be functional.

Table 1.  Septal localization of GFP–DivIB constructs.
StrainConstructa,bLocalizationcWestern blot bandd
  • a.

    The subscripts IB and TR indicate that the domains are from B. subtilis DivIB and E. coli TolR respectively.

  • b

    α, β and γ refer to the three extracytoplasmic domains of DivIB (Robson and King, 2006).

  • c.

    The symbols ‘+’ and ‘−’ indicate midcell localization in > 80% and < 20% of cells respectively.

  • d.

    A ‘+’ symbol indicates that a band of the appropriate size was detected on Western blots and that the intensity of the protein band was not significantly different from that seen in KDW1.

  • ND, not determined.

KDW1GFP–CytoIBTMIBαIBβIBγIB++
KDW2GFP–CytoIBTMIBαIBβIB++
KDW3GFP–CytoIBTMIBαIB++
KDW4GFP+
KDW5GFP–CytoIBTMIBβIB++
KDW6GFP–CytoIBTMIBγIB++
KDW7GFP–CytoIBTMIBβIBγIB++
KDW8GFP–CytoIBTMIBαIBγIB++
KDW9GFP–CytoIBTMIB++
KDW10GFP–CytoTRTMTRαIBβIBγIB++
KDW11GFP–CytoTRTMTRαIBβIB++
KDW12GFP–CytoTRTMTRαIB++
KDW13GFP–CytoTRTMTRαIBγIB++
KDW14GFP–CytoTRTMTRβIBγIB++
KDW15GFP–CytoTRTMTRβIB+
KDW16GFP–CytoTRTMTRγIB+
KDW17GFP–CytoIB+
KDW18GFP–TMIB++
KDW19GFP–CytoTRTMTR+
KDW27GFP–CytoTRTMTRαIB(E122A)+ND
KDW28GFP–CytoTRTMTRβIBγIB(Y208A)+ND
KDW29GFP–CytoTRTMTRβIBγIB(P192A)+ND
KDW30GFP–CytoTRTMTRβIBγIB(P225A)+ND
KDW31GFP–TMIB(M37R/I41R)+ND
KDW32GFP–TMIB(I41R/L44R)+ND
KDW33GFP–TMIB(M37R/I41R/L44R)+ND
KDW34GFP–CytoTRTMTRαIB(I59P)+ND

Molecular definition of the septal localization epitopes of DivIB

In an attempt to define more precisely the septal targeting determinants in DivIB we examined the cellular localization of a panel of constructs with point mutations in specific DivIB domains. It was recently reported that the ability of E. coli FtsQ to localize to the division site is severely impaired when residues Leu29 and Leu32 in the TM domain are both mutated to Arg (Scheffers et al., 2007). In striking contrast, the septal localization of GFP–TMIB was not diminished by the introduction of either two (M37R/I41R and I41R/L44R) or three (M37R/I41R/L44R) non-conservative mutations at similar locations in the DivIB TM domain (Fig. 5A–C). Similarly, while an L60P mutation in the α domain of FtsQ leads to a septal localization defect (Goehring et al., 2007), an I59P mutation at the equivalent position in the α domain of DivIB had no effect on septal localization of GFP–CytoTR-TMTRIB (Fig. 5D), nor did mutation to Ala of the Glu residue at position 122 (Fig. 5E), which is highly conserved throughout the FtsQ/DivIB family (see fig. 1 in Robson and King, 2006). These results indicate that while the septal localization determinants in DivIB and FtsQ might be embedded within similar domains (although this remains to be determined), the molecular details of the localization epitopes are nevertheless likely to be significantly different. This is perhaps not surprising given that E. coli FtsQ and B. subtilis DivIB share only ∼18% sequence identity (Harry et al., 1994) and that the divisomal proteins with which they interact to form a ternary complex (i.e. FtsL and FtsB/DivIC) have a similarly low level of conservation between Gram-positive and Gram-negative eubacteria. This highlights the fact that one needs to be circumspect in drawing conclusions about the B. subtilis divisome from experiments performed with E. coli, and vice versa.

image

Figure 5. Fluorescence micrographs showing localization in B. subtilis of GFP fused to the N-terminus of (A) TMIB (M37R, I41R), (B) TMIB (I41R, L44R), (C) TMIB (M37R, I41R, L44R), (D) CytoTR-TMTRαIB (I59P), (E) CytoTR-TMTRαIB (E122A), (F) CytoTR-TMTRIBγIB (P192A), (G) CytoTR-TMTRIBγIB (Y208A) and (H) CytoTR-TMTRIBγIB (P225A). The ‘IB’ and ‘TR’ subscripts indicate that the domains are from DivIBBs or E. coli TolR respectively. Point mutations are indicated in parentheses. Septal localization is evident in all panels from the fluorescent stripes or dots at midcell.

Download figure to PowerPoint

In an attempt to identify the localization epitope that appears to span the β and γ domains, we examined the effect of mutating conserved residues located near the β–γ domain boundary. Neither mutation of two conserved Pro residues (i.e. P192A and P225A; Fig. 5F and G) nor a semi-conserved Tyr residue (Y208A; Fig. 5H) near the C-terminal end of the β domain diminished the septal localization of a GFP–CytoTR-TMTRIBγIB fusion protein. We conclude that these residues in the β domain are not part of the β/γ localization epitope or alternatively that this epitope is complex and that mutation of a single residue within the localization determinant is insufficient to cause a septal targeting defect.

The α domain of DivIB is dispensable for vegetative cell division

It has often been assumed, perhaps naively, that if a divisomal protein domain contains a septal localization signal then that domain is most likely critical for the protein's cell division function. The localization studies presented herein indicate that this is not necessarily the case, as the TM domain of DivIBBs contains a septal localization signal and yet it can be replaced without phenotypic consequence by the TM region from the E. coli TolR protein, which we demonstrated in the current study does not localize to division septa in B. subtilis. Thus, we decided to determine which domains of DivIB are essential for its role in vegetative cell division in order to further examine whether the presence of a septal localization signal in a particular protein domain correlates with an absolute requirement for that domain in cytokinesis.

All of the GFP constructs used in the localization studies (see Fig. 1B) were streaked onto duplicate LB plates supplemented with 0.5% xylose. One set of plates was incubated at the permissive temperature for the divIB null (30°C) while the second set was incubated at the non-permissive temperature (48°C). After approximately 16 h growth at 30°C, all strains (including the divIB null) had grown to levels comparable with the wild-type parent strain. However, at the non-permissive temperature, the only viable strains were those that produced DivIB constructs containing both the β and γ domains tethered to the Cyto-TM region of either DivIBBs or E. coli TolR (Fig. 6 and Table 2). Remarkably, even though the α domain contains a septal localization signal, it appears to be completely dispensable for vegetative cell division at the non-permissive temperature. As demonstrated previously (Katis and Wake, 1999), we also found that the Cyto-TM region of DivIBBs, which contains a septal localization signal within the TM domain, could be replaced with the corresponding region from E. coli TolR without any phenotypic consequences at the non-permissive temperature. These latter two observations reveal that the presence of a septal localization signal within a specific domain of a divisomal protein does not necessarily indicate that the domain is essential for the protein's cell division function(s).

image

Figure 6. Growth on LB plates of the wild-type parental strain (SU5), a divIB null (RSA8) and strains containing an ectopic variant of divIB at the amyE locus. Cyto and TM refer to the cytoplasmic and transmembrane domains respectively. The ‘IB’ and ‘TR’ subscripts indicate that the domains are from DivIBBs or E. coli TolR respectively. The extracytoplasmic α domain of DivIB is dispensable for growth at the non-permissive temperature (see sectors 5 and 6 on the 48°C plate).

Download figure to PowerPoint

Table 2.  Summary of plate assays.
StrainConstructa,bGrowth at 30°CGrowth at 48°C
  • a.

    The subscripts IB and TR indicate that the domains are from B. subtilis DivIB and E. coli TolR respectively.

  • b.

    α, β and γ refer to the three extracytoplasmic domains of DivIB (Robson and King, 2006).

SU5Wild type++
RSA8DivIB null+
KDW1GFP–CytoIBTMIBαIBβIBγIB++
KDW2GFP–CytoIBTMIBαIBβIB+
KDW3GFP–CytoIBTMIBαIB+
KDW4GFP+
KDW5GFP–CytoIBTMIBβIB+
KDW6GFP–CytoIBTMIBγIB+
KDW7GFP–CytoIBTMIBβIBγIB++
KDW8GFP–CytoIBTMIBαIBγIB+
KDW9GFP–CytoIBTMIB+
KDW10GFP–CytoTRTMTRαIBβIBγIB++
KDW11GFP–CytoTRTMTRαIBβIB+
KDW12GFP–CytoTRTMTRαIB+
KDW13GFP–CytoTRTMTRαIBγIB+
KDW14GFP–CytoTRTMTRβIBγIB++
KDW15GFP–CytoTRTMTRβIB+
KDW16GFP–CytoTRTMTRγIB+
KDW17GFP–CytoIB+
KDW18GFP–TMIB+
KDW19GFP–CytoTRTMTR+

Discussion

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

DivIB contains multiple septal localization epitopes

DivIB is a bitopic membrane protein that comprises a small N-terminal cytoplasmic domain, a single-pass TM domain and an extracytoplasmic region that comprises three separate domains designated α, β and γ. The architecture of FtsQ is expected to be similar based on sequence homology with DivIB (Robson and King, 2006). The membrane-proximal α domain is coincident with the polypeptide transport associated (POTRA) domain which is also found in the Toc75 subunit of the chloroplast import translocon and in the outer-membrane proteins of some Gram-negative bacteria where it is involved in the integration of proteins into the outer membrane or the export of virulence factors (Sanchez-Pulido et al., 2003; Gentle et al., 2005). The centrally located β domain of DivIB has a unique 3D structure that dynamically interconverts between two distinct conformations (Robson and King, 2006). In contrast, the small C-terminal γ domain appears to be largely unstructured in the absence of other divisomal proteins (Robson and King, 2006). The aim of the current study was to determine which domains of DivIB mediate its divisomal recruitment and which domains are essential for DivIB function during vegetative division. An unanticipated outcome was the finding that some domains which mediate divisomal recruitment of DivIB are completely dispensable for cytokinesis during vegetative growth.

In E. coli, most divisomal proteins, with the exception of FtsZ, are present at 50–300 molecules per cell, and therefore they are physically incapable of forming a contiguous structure that extends completely around the plasma membrane (Goehring and Beckwith, 2005). It therefore seems likely that late recruits to the divisome, which are predominantly extracytoplasmic, assemble into a finite number of discrete complexes that are attached to the Z ring at specific locations. We imagine that a similar scenario prevails in B. subtilis. However, in contrast with E. coli FtsQ, DivIBBS is a highly abundant divisomal protein that is present at ∼10 000 copies/cell, and therefore it is theoretically capable of forming a circumferential ring of DivIBBs molecules around the entire cell membrane. If DivIBBs assembles into a finite number of divisomal complexes rather than forming a contiguous circumferential ring around the plasma membrane then we might expect a large residual of extra-divisomal DivIB, consistent with the observation that < 2% of the cellular complement of DivIBBs is required for normal vegetative division (Rowland et al., 1997). The studies described herein support this hypothesis. In addition to the striking midcell fluorescence seen in cells expressing GFP–DivIBBs, we observed distinct additional fluorescence around the entire periphery of the cell (see the cell ‘halo’ in Fig. 2A). We interpret this as an indication that while a significant proportion of the cellular DivIBBs is localized to the septum during vegetative cell division, a considerable amount of extra-divisomal DivIBBs is distributed throughout the plasma membrane at the time of cytokinesis. This ‘non-septal’ DivIB might have an additional function that is unrelated to the role of DivIB during cytokinesis.

Analysis of a variety of DivIB truncations and chimeras (Figs 2–5) indicated that DivIBBs contains three separate molecular epitopes that mediate its recruitment into divisomal complexes. The TM domain and the α/POTRA domain each contain a septal localization signal, while a third localization epitope appears to span the β and γ domains. As far as we are aware, DivIB is the first example of a cell division protein that has a tripartite septal localization signal. While it is possible that one or more of these molecular epitopes recognize a septal-specific lipid or murein component, we favour the alternative hypothesis that they each represent points of contact with other divisomal proteins. This hypothesis is consistent with the fact that two- and three-hybrid assays consistently indicate that FtsQ and DivIB interact with numerous other divisomal proteins (Di Lallo et al., 2003; Daniel et al., 2006; D'Ulisse et al., 2007). Indeed, FtsQ has been described as a ‘connector protein’ that nucleates the assembly of higher-order divisomal complexes (Vicente et al., 2006).

Modelling the extracytoplasmic architecture of the bacterial divisome

The development of a model for the architecture of the divisome would facilitate our understanding of the way in which this large protein assembly physically divides the cell and, moreover, it should provide insight into key protein–protein interactions that could be targeted to produce novel antibacterial agents. While it is not possible to construct such a model at the current time, the present study, in combination with recent genetic and biochemical assays, at least allows us to begin to assemble a testable model of the way in which DivIB and FtsQ are integrated into the divisome (Fig. 7). There are two caveats to this model-building exercise. The first is our assumption that DivIB and FtsQ are directed to the division site through similar protein–protein interactions with other members of the divisome. The second caveat is the widely held, but as yet unproven, assumption that divisomal complexes are completely homogeneous. It is entirely possible that heterogeneous divisomal complexes containing different protein components and performing different functions are attached to the Z ring at different times throughout the division process.

image

Figure 7. Model of topological interrelationships between the divisomal proteins FtsW, DivIB, FtsL, DivIC and PBP 2B. Red stars highlight putative protein–protein interaction motifs of DivIBBs that likely correspond to the septal targeting motifs identified in the current study. The model of PBP 2B (blue) is based on the crystal structure of Streptococcus pneumoniae PBP2x (PDB Accession No. 1K25; Dessen et al., 2001) while the schematic of the β domain of DivIB is from the recently determined NMR solution structure (PDB Accession No. 2ALJ) (Robson and King, 2006). The putative coiled-coil motif formed by association of the leucine zipper motifs of FtsL and DivIC was modeled using a superhelical pitch of 137 Å, which is typical for left-handed parallel coiled coils (Junius et al., 1996). The final two TM helices of FtsW are omitted for the sake of clarity.

Download figure to PowerPoint

One of the most surprising results of the current study is the observation that the TM domain of DivIBBs, which is not essential for cytokinesis, contains a strong septal localization epitope that is sufficient for targeting heterologous proteins such as GFP to incipient division sites. The only other divisomal proteins that have been shown to contain a TM septal localization signal are E. coli FtsI (Weiss et al., 1999; Piette et al., 2004; Wissel and Weiss, 2004; Wissel et al., 2005) and FtsL (Ghigo and Beckwith, 2000). It is likely that DivIB and FtsL interact through their TM domains (Fig. 7), in addition to possible interactions between their extracytoplasmic regions (see below). While this TM interaction is clearly not essential in vivo (as the Cyto-TM region of both DivIB and FtsQ is dispensable for vegetative cell division), it would nevertheless increase the affinity of these proteins for one another. This might explain why it has not been possible to isolate the putative DivIB:DivIC:FtsL ternary complex in vitro using only the extracytoplasmic regions of these proteins (Noirclerc-Savoye et al., 2005; Robson and King, 2006) even though the ternary complex can be detected using the full-length proteins in yeast three-hybrid experiments (Daniel et al., 2006).

The identification of a localization epitope in the α domain of DivIBBs is consistent with the recent observation that L60P and V92P mutations in the α domain of FtsQ cause localization defects (Goehring et al., 2007). Moreover, these mutants point to a likely location for the septal targeting signal within the α domain of FtsQ. Note, however, that this epitope most likely has a quite different molecular fingerprint in DivIB as an I59P mutation in DivIBBs (equivalent to L60P in FtsQEc) does not diminish its ability to localize to the septum.

A recent B2H study of the divisomal interactions of FtsQ truncations (D'Ulisse et al., 2007) indicates that the α domain of FtsQ and DivIB most likely contains motifs for interaction with FtsI and FtsW. In the case of FtsW, the only significant extracytoplasmic portion of the protein is the ∼75-residue loop between TM helices 7 and 8 (Gérard et al., 2002; Lara and Ayala, 2002), and hence this is the most likely site for interaction with the α domain of FtsQ and DivIB (Fig. 7). As the α domain of FtsQ is proximal to the membrane, the FtsI interaction epitope is most likely contained within the membrane-proximal non-catalytic module rather than the membrane-distal catalytic domain (Fig. 7). Mutational analysis of FtsI revealed previously that this region is involved in homotypic and/or heterotypic protein–protein interactions (Marrec-Fairley et al., 2000; Wissel and Weiss, 2004).

The second septal extracytoplasmic septal targeting epitope in DivIBBs appears to span the β and γ domains. Consistent with this finding, B2H analyses indicate that a region comprising residues 202–276 of FtsQ, which encompasses the γ domain (residues 232–276) and the C-terminal end of the β domain (residues 202–231), is engaged in interactions with FtsL, FtsI and FtsN. Topological considerations dictate that this region of FtsQ most likely associates with the membrane-distal catalytic domain of FtsI rather than the membrane-proximal non-catalytic module (Fig. 7). Similarly, as argued previously (Robson and King, 2006), the C-terminal region of FtsQ and DivIB most likely engages the C-terminal region of FtsL, which is thought to be located distal to the membrane by virtue of its leucine zipper motif; it is presumed (although not proven experimentally) that this motif associates with the corresponding region of FtsB (DivIC) to form a rod-like coiled-coil domain (Ghigo and Beckwith, 2000; Sievers and Errington, 2000; Robson et al., 2002).

While the model in Fig. 7 is necessarily speculative, it is nevertheless compatible with virtually all available experimental data and, most importantly, it provides testable hypotheses for future experiments. For example, the model predicts that the functionally critical C-terminal domains of DivIBBs, DivIC and FtsL are all positioned close to the catalytic domain of PBP 2B, raising the possibility that all of these proteins function primarily to allosterically regulate PBP 2B activity.

The role of DivIB/FtsQ in cytokinesis

The absence of the bitopic membrane proteins DivIB/FtsQ, FtsL and DivIC/FtsB from bacteria without cell walls implicates them in the synthesis and/or remodelling of septal peptidoglycan. Although no experimental data confirming this function have been obtained thus far, this hypothesis is supported by our recent demonstration that during sporulation divIB null cells form an abnormally a thick wall and are unable complete the forespore engulfment process (Thompson et al., 2006).

Curiously, the current study revealed that the α/POTRA domain of DivIBBs is dispensable for vegetative cell division, even though it contains a septal localization signal. The POTRA domain has been proposed to have a chaperone function (Sanchez-Pulido et al., 2003) and we previously suggested that the α/POTRA domain of DivIBBs might stabilize the unfolded or partially folded form of FtsL, thereby aiding its assembly into the divisome and protecting it from degradation (Robson and King, 2006). However, if the α domain does have such a role it is clearly dispensable for normal cytokinesis, at least during vegetative division. In marked contrast, neither of the two POTRA domains of the FhaC outer-membrane transporter of Bordetella pertussis can be deleted without abolishing the ability of FhaC to secrete virulence factors (Clantin et al., 2007). It will be interesting in future studies to determine whether the α domain plays an essential role in polar division during sporulation, as this process is clearly more sensitive to defects in DivIB (Thompson et al., 2006).

In summary, we have shown that DivIBBs contains multiple signals that mediate its recruitment to incipient division sites. We have interpreted these localization epitopes as sites of interaction with other divisomal proteins and have used this information to build a model of how DivIB and FtsQ are integrated into the divisome. Surprisingly, two of the septal localization signals in DivIBBs are contained within the TM and α domains, both of which are dispensable for vegetative cell division. This finding is consistent with the hypothesis that the Cyto-TM regions of the predominantly extracytoplasmic divisomal proteins engage in a network of protein–protein interactions that are individually non-essential, such that loss of a single cytoplasmic or TM domain often does not have a noticeably deleterious effect on cytokinesis (Geissler and Margolin, 2005).

Experimental procedures

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

Strain construction

Standard molecular biology methods were used for plasmid construction and transformation into E. coli DH5α (Sambrook and Russell, 2001). B. subtilis transformation was performed as described (Jensen et al., 2005). B. subtilis RSA8 (Robson and King, 2006) and SU321 (Harry and Wake, 1997), which are both divIB null strains derived from SU5 (168), were used for strain constructions (see Table 3). Plasmids pKDW1 through pKDW18 (Table 3) were made by cloning target DNA into the ApaI and EcoRI restriction sites of pSG1729 (Lewis and Marston, 1999), while plasmids pKDW29 through pKDW36 (Table 3) were cloned into the BamHI and EcoRI restriction sites of pSG1729. pSG1729 is a B. subtilis-E. coli shuttle vector that allows insertion of cloned DNA by double cross-over at the B. subtilis chromosomal amyE site. Directional in-frame cloning into the ApaI/EcoRI or BamHI/EcoRI sites generates an N-terminal GFP fusion that is selectable with spectinomycin, and driven by Pxyl, upon correct chromosomal insertion. After pSG1729-derived plasmids were transformed into RSA8, disruption of the amyE locus was verified by plating transformants on LB/starch (1%) plates, growing them overnight, then checking for inability to digest starch (Sekiguchi et al., 1975). Further details of plasmid construction are given in the Supplementary Information section.

Table 3.  Strains and plasmids.
Strain/plasmidGenotypea,bSource/Construction
  • a.

    The subscripts IB and TR indicate that the domains are from B. subtilis DivIB and E. coli TolR respectively.

  • b.

    The notations ‘C’ and ‘T’ refer to the cytoplasmic and transmembrane domains respectively.

B. subtilis
 SU5 (168)trpC2E. Harry
 RSA8trpC2 divIB::cat::ermCRobson and King (2006)
 SU321trpC2 divIB::catHarry and Wake (1997)
 KDW2trpC2 divIB::cat::ermC amyE::Pxyl-gfpmut1-CIBTIBαIBβIB spcpKDW2 [RIGHTWARDS ARROW] RSA8
 KDW3trpC2 divIB::cat::ermC amyE::Pxyl-gfpmut1-CIBTIBαIB spcpKDW3 [RIGHTWARDS ARROW] RSA8
 KDW4trpC2 divIB::cat::ermC amyE::Pxyl-gfpmut1 spcpSG1729 [RIGHTWARDS ARROW] RSA8
 KDW5trpC2 divIB::cat::ermC amyE::Pxyl-gfpmut1-CIBTIBβIB spcpKDW4 [RIGHTWARDS ARROW] RSA8
 KDW6trpC2 divIB::cat::ermC amyE::Pxyl-gfpmut1-CIBTIBγIB spcpKDW5 [RIGHTWARDS ARROW] RSA8
 KDW7trpC2 divIB::cat::ermC amyE::Pxyl-gfpmut1-CIBTIBβIBγIB spcpKDW6 [RIGHTWARDS ARROW] RSA8
 KDW8trpC2 divIB::cat::ermC amyE::Pxyl-gfpmut1-CIBTIBαIBγIB spcpKDW7 [RIGHTWARDS ARROW] RSA8
 KDW9trpC2 divIB::cat::ermC amyE::Pxyl-gfpmut1-CIBTIB spcpKDW8 [RIGHTWARDS ARROW] RSA8
 KDW10trpC2 divIB::cat::ermC amyE::Pxyl-gfpmut1-CTRTTRαIBβIBγIB spcpKDW9 [RIGHTWARDS ARROW] RSA8
 KDW11trpC2 divIB::cat::ermC amyE::Pxyl-gfpmut1-CTRTTRαIBβIB spcpKDW10 [RIGHTWARDS ARROW] RSA8
 KDW12trpC2 divIB::cat::ermC amyE::Pxyl-gfpmut1-CTRTTRαIB spcpKDW11 [RIGHTWARDS ARROW] RSA8
 KDW13trpC2 divIB::cat::ermC amyE::Pxyl-gfpmut1-CTRTTRαIBγIB spcpKDW12 [RIGHTWARDS ARROW] RSA8
 KDW14trpC2 divIB::cat::ermC amyE::Pxyl-gfpmut1-CTRTTRβIBγIB spcpKDW13 [RIGHTWARDS ARROW] RSA8
 KDW15trpC2 divIB::cat::ermC amyE::Pxyl-gfpmut1-CTRTTRβIB spcpKDW14 [RIGHTWARDS ARROW] RSA8
 KDW16trpC2 divIB::cat::ermC amyE::Pxyl-gfpmut1-CTRTTRγIB spcpKDW15 [RIGHTWARDS ARROW] RSA8
 KDW17trpC2 divIB::cat::ermC amyE::Pxyl-gfpmut1-CIB spcpKDW16 [RIGHTWARDS ARROW] RSA8
 KDW18trpC2 divIB::cat::ermC amyE::Pxyl-gfpmut1-TIB spcpKDW17 [RIGHTWARDS ARROW] RSA8
 KDW19trpC2 divIB::cat::ermC amyE::Pxyl-gfpmut1-CTRTTR spcpKDW18 [RIGHTWARDS ARROW] RSA8
 KDW27trpC2 divIB::cat amyE::Pxyl-gfpmut1-CTRTTRαIB(E122A) spcpKDW29 [RIGHTWARDS ARROW] SU321
 KDW28trpC2 divIB::cat amyE::Pxyl-gfpmut1-CTRTTRβIBγIB(Y208A) spcpKDW30 [RIGHTWARDS ARROW] SU321
 KDW29trpC2 divIB::cat amyE::Pxyl-gfpmut1-CTRTTRβIBγIB(P192A) spcpKDW31 [RIGHTWARDS ARROW] SU321
 KDW30trpC2 divIB::cat amyE::Pxyl-gfpmut1-CTRTTRβIBγIB(P225A) spcpKDW32 [RIGHTWARDS ARROW] SU321
 KDW31trpC2 divIB::cat amyE::Pxyl-gfpmut1-TIB(M37R/I41R) spcpKDW33 [RIGHTWARDS ARROW] SU321
 KDW32trpC2 divIB::cat amyE::Pxyl-gfpmut1-TIB(I41R/L44R) spcpKDW34 [RIGHTWARDS ARROW] SU321
 KDW33trpC2 divIB::cat amyE::Pxyl-gfpmut1-TIB(M37R/I41R/L44R) spcpKDW35 [RIGHTWARDS ARROW] SU321
 KDW34trpC2 divIB::cat amyE::Pxyl-gfpmut1-CTRTTRαIB(I59P) spcpKDW36 [RIGHTWARDS ARROW] SU321
Plasmids
 pSG1729B. subtilis amyE integration vector; Pxyl bla spcLewis and Marston (1999)
 pKDW1pSG1729 containing CIBTIBαIBβIBγIB; Pxyl bla spcThis study
 pKDW2pSG1729 containing CIBTIBαIBβIB; Pxyl bla spcThis study
 pKDW3pSG1729 containing CIBTIBαIB; Pxyl bla spcThis study
 pKDW4pSG1729 containing CIBTIBβIB; Pxyl bla spcThis study
 pKDW5pSG1729 containing CIBTIBγIB; Pxyl bla spcThis study
 pKDW6pSG1729 containing CIBTIBβIBγIB; Pxyl bla spcThis study
 pKDW7pSG1729 containing CIBTIBαIBγIB; Pxyl bla spcThis study
 pKDW8pSG1729 containing CIBTIB; Pxyl bla spcThis study
 pKDW9pSG1729 containing CTRTTRαIBβIBγIB; Pxyl bla spcThis study
 pKDW10pSG1729 containing CTRTTRαIBβIB; Pxyl bla spcThis study
 pKDW11pSG1729 containing CTRTTRαIB; Pxyl bla spcThis study
 pKDW12pSG1729 containing CTRTTRαIBγIB; Pxyl bla spcThis study
 pKDW13pSG1729 containing CTRTTRβIBγIB; Pxyl bla spcThis study
 pKDW14pSG1729 containing CTRTTRβIB; Pxyl bla spcThis study
 pKDW15pSG1729 containing CTRTTRγIB; Pxyl bla spcThis study
 pKDW16pSG1729 containing CIB; Pxyl bla spcThis study
 pKDW17pSG1729 containing TIB; Pxyl bla spcThis study
 pKDW18pSG1729 containing CTRTTR; Pxyl bla spcThis study
 pKDW29pSG1729 containing CTRTTRαIB(E122A); Pxyl bla spcThis study
 pKDW30pSG1729 containing CTRTTRβIBγIB(Y208A); Pxyl bla spcThis study
 pKDW31pSG1729 containing CTRTTRβIBγIB(P192A); Pxyl bla spcThis study
 pKDW32pSG1729 containing CTRTTRβIBγIB(P225A); Pxyl bla spcThis study
 pKDW33pSG1729 containing TIB(M37R/I41R); Pxyl bla spcThis study
 pKDW34pSG1729 containing TIB(I41R/L44R); Pxyl bla spcThis study
 pKDW35pSG1729 containing TIB(M37R/I41R/L44R); Pxyl bla spcThis study
 pKDW36pSG1729 containing CTRTTRαIB(I59P); Pxyl bla spcThis study

Bacterial growth conditions

Bacterial strains used in this study are listed in Table 3. For microscopic and Western blot analyses, strains were grown in LB at 30°C overnight (∼16 h) with shaking. Overnight cultures were supplemented with spectinomycin (100 μg ml−1) as required. Cultures were then diluted to an OD600 of ∼0.05 with LB and induced with a final concentration of 0.5–1% xylose. Cultures were allowed to grow at 30°C with shaking to a final OD600 of ∼1 before samples were taken for analysis.

Fluorescence microscopy

Cells were grown as described above, then a 500 μl sample was harvested by centrifugation and re-suspended in ∼50 μl LB. Cells were viewed unfixed using either 1% agarose pads in LB or LB alone. Dabco (1,4-diazabicyclo[2.2.2]octane) (2.5% in 0.1 M Tris, pH 7.5) was sometimes used at a 1:1 ratio to increase fluorescence intensity. All fluorescence images, except those of DivIB point mutants, were obtained using an Olympus BX40 microscope equipped with a MagnaFire digital camera (Optronics International, Chelmsford, MA). Micrographs of point mutants were obtained using a Zeiss Axioplan 2 fuorescence microscope equipped with an AxioCam MRm cooled CCD camera (Carl Zeiss, Oberkochen, Germany).

Western blot analysis

The cell pellet from a 4.5 ml culture was harvested by centrifugation, re-suspended in lysis buffer (10 mM Tris-HCl pH 8, 1 mM EDTA, 10 mM MgCl2, 0.3 mg ml−1 PMSF, 0.5 mg ml−1 lysozyme and 0.1 mg ml−1 DNase 1), then incubated at 37°C for 10 min. The sample was then boiled at 100°C for 2 min following addition of 1% SDS. Total protein concentration was determined with a bicinchoninic acid assay kit (Pierce) using bovine serum albumin as a standard. SDS/polyacrylamide gels were loaded with 130 μg of total cell protein for each GFP fusion construct. Following electrophoresis, gels were electroblotted onto Hybond ECL nitrocellulose membrane (Amersham Biosciences), and GFP fusion proteins were probed using a rabbit anti-GFP primary antibody (kindly provided by R. Losick, Harvard University, Cambridge, MA) and an anti-rabbit horseradish peroxidase-conjugated secondary antibody (Amersham Biosciences). Immunoblots were developed on Hyperfilm ECL (Amersham Biosciences) using a chemiluminescent substrate.

Plate assay

All strains were streaked onto duplicate LB/Xylose (0.5%) plates; one plate was incubated at 30°C (permissive temperature) while the other was incubated at 48°C (non-permissive temperature). Colonies were grown overnight prior to photography. Assays were performed in triplicate. All GFP fusion strains were compared with the wild-type parental strain (SU5).

Acknowledgements

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

We thank members of the King lab for comments and suggestions, Phoebe Peters and Joana Santos for help with microscopy, Dr Rich Losick for the anti-GFP antibody, Dr Jon Beckwith for valuable discussions and the NIH for financial support (AI48583).

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information
  • Beall, B., and Lutkenhaus, J. (1989) Nucleotide sequence and insertional inactivation of a Bacillus subtilis gene that affects cell division, sporulation, and temperature sensitivity. J Bacteriol 171: 68216834.
  • Begg, K.J., Hatfull, G.F., and Donachie, W.D. (1980) Identification of new genes in a cell envelope-cell division gene cluster of Escherichia coli: cell division gene ftsQ. J Bacteriol 144: 435437.
  • Bi, E., and Lutkenhaus, J. (1991) FtsZ ring structure associated with division in Escherichia coli. Nature 354: 161164.
  • Buddelmeijer, N., Aarsman, M.E.G., Kolk, A.H.J., Vicente, M., and Nanninga, N. (1998) Localization of cell division protein FtsQ by immunofluorescence microscopy in dividing and nondividing cells of Escherichia coli. J Bacteriol 180: 61076116.
  • Carson, M.J., Barondess, J., and Beckwith, J. (1991) The FtsQ protein of Escherichia coli: membrane topology, abundance, and cell division phenotypes due to overproduction and insertion mutations. J Bacteriol 173: 21872195.
  • Chen, J.C., Weiss, D.S., Ghigo, J.-M., and Beckwith, J. (1999) Septal localization of FtsQ, an essential cell division protein in Escherichia coli. J Bacteriol 181: 521530.
  • Clantin, B., Delattre, A.-S., Rucktooa, P., Saint, N., Méli, A.C., Locht, C., et al. (2007) Structure of the membrane protein FhaC: a member of the Omp85-TpsB transporter family. Nature 317: 957961.
  • D'Ulisse, V., Fagioli, M., Ghelardini, P., and Paolozzi, L. (2007) Three functional domains of the Escherichia coli FtsQ protein are involved in its interaction with the other division proteins. Microbiology 153: 124138.
  • Daniel, R.A., Noirot-Gros, M.F., Noirot, P., and Errington, J. (2006) Multiple interactions between the transmembrane division proteins of Bacillus subtilis and the role of FtsL instability in divisome assembly. J Bacteriol 188: 73967404.
  • Dessen, A., Mouz, N., Gordon, E., Hopkins, J., and Dideberg, O. (2001) Crystal structure of PBP2x from a highly penicillin-resistant Streptococcus pneumoniae clinical isolate: a mosaic framework containing 83 mutations. J Biol Chem 276: 4510645112.
  • Di Lallo, G., Fagioli, M., Barionovi, D., Ghelardini, P., and Paolozzi, L. (2003) Use of a two-hybrid assay to study the assembly of a complex multicomponent protein machinery: bacterial septosome differentiation. Microbiology 149: 33533359.
  • Dopazo, A., Palacios, P., Sánchez, M., Pla, J., and Vicente, M. (1992) An amino-proximal domain required for the localization of FtsQ in the cytoplasmic membrane, and for its biological function in Escherichia coli. Mol. Microbiol 6: 715722.
  • Errington, J., Daniel, R.A., and Scheffers, D.J. (2003) Cytokinesis in bacteria. Microbiol Mol Biol Rev 67: 5265.
  • Geissler, B., and Margolin, W. (2005) Evidence for functional overlap among multiple bacterial cell division proteins: compensating for the loss of FtsK. Mol Microbiol 58: 596612.
  • Gentle, I.E., Burri, L., and Lithgow, T. (2005) Molecular architecture and function of the Omp85 family of proteins. Mol Microbiol 58: 12161225.
  • Gérard, P., Vernet, T., and Zapun, A. (2002) Membrane topology of the Streptococcus pneumoniae FtsW division protein. J Bacteriol 184: 19251931.
  • Gerding, M.A., Ogata, Y., Pecora, N.D., Niki, H., and De Boer, P.A. (2007) The trans-envelope Tol-Pal complex is part of the cell division machinery and required for proper outer-membrane invagination during cell constriction in E. coli. Mol Microbiol 63: 10081025.
  • Ghigo, J.-M., and Beckwith, J. (2000) Cell division in Escherichia coli: role of FtsL domains in septal localization, function, and oligomerization. J Bacteriol 182: 116129.
  • 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., Petrovska, I., Boyd, D., and Beckwith, J. (2007) Mutants, suppressors, and wrinkled colonies: mutant alleles of the cell division gene ftsQ point to functional domains in FtsQ and a role for domain 1C of FtsA in divisome assembly. J Bacteriol 189: 633645.
  • Guzman, L.M., Weiss, D.S., and Beckwith, J. (1997) Domain-swapping analysis of FtsI, FtsL, and FtsQ, bitopic membrane proteins essential for cell division in Escherichia coli. J Bacteriol 179: 50945103.
  • Harry, E.J., and Wake, R.G. (1989) Cloning and expression of a Bacillus subtilis division initiation gene for which a homolog has not been identified in another organism. J Bacteriol 171: 68356839.
  • Harry, E.J., and Wake, R.G. (1997) The membrane-bound cell division protein DivIB is localized to the division site in Bacillus subtilis. Mol Microbiol 25: 275283.
  • Harry, E.J., Partridge, S.R., Weiss, A.S., and Wake, R.G. (1994) Conservation of the 168 divIB gene in Bacillus subtilis W23 and B. licheniformis, and evidence for homology to ftsQ of Escherichia coli. Gene 147: 8589.
  • Jensen, S.O., Thompson, L.S., and Harry, E.J. (2005) Cell dvision in Bacillus subtilis: FtsZ and FtsA association is Z-ring independent, and FtsA is required for efficient midcell Z-ring assembly. J Bacteriol 187: 65366544.
  • Junius, F.K., O'Donoghue, S.I., Nilges, M., Weiss, A.S., and King, G.F. (1996) High resolution solution structure of the leucine zipper domain of the c-Jun homodimer. J Biol Chem 271: 1366313667.
  • Katis, V.L., and Wake, R.G. (1999) Membrane-bound division proteins DivIB and DivIC of Bacillus subtilis function solely through their external domains in both vegetative and sporulation division. J Bacteriol 181: 27102718.
  • Katis, V.L., Harry, E.J., and Wake, R.G. (1997) The Bacillus subtilis division protein DivIC is a highly abundant membrane-bound protein that localizes to the division site. Mol Microbiol 26: 10471055.
  • Lara, B., and Ayala, J.A. (2002) Topological characterization of the essential Escherichia coli cell division protein FtsW. FEMS Microbiol Lett 216: 2332.
  • Lewis, P.J., and Marston, A.L. (1999) GFP vectors for controlled expression and dual labelling of protein fusions in Bacillus subtilis. Gene 227: 101109.
  • Margolin, W. (2005) FtsZ and the division of prokarytotic cells and organelles. Nat Rev Mol Cell Biol 6: 862871.
  • Marrec-Fairley, M., Piette, A., Gallet, X., Brasseur, R., Hara, H., Fraipont, C., et al. (2000) Differential functionalities of amphiphilic peptide segments of the cell-septation penicillin-binding protein 3 of Escherichia coli. Mol Microbiol 37: 10191031.
  • Noirclerc-Savoye, M., Le Gouellec, A., Morlot, C., Dideberg, O., Vernet, T., and Zapun, A. (2005) In vitro reconstitution of a trimeric complex of DivIB, DivIC and FtsL, and their transient co-localization at the division site in Streptococcus pneumoniae. Mol Microbiol 55: 413424.
  • Piette, A., Fraipont, C., Den Blaauwen, T., Aarsman, M.E., Pastoret, S., and Nguyen-Disteche, M. (2004) Structural determinants required to target penicillin-binding protein 3 to the septum of Escherichia coli. J Bacteriol 186: 61106117.
  • Real, G., Autret, S., Harry, E.J., Errington, J., and Henriques, A.O. (2005) Cell division protein DivIB influences the Spo0J/Soj system of chromosome segregation in Bacillus subtilis. Mol Microbiol 55: 349367.
  • Robson, S.A., and King, G.F. (2006) Domain architecture and structure of the bacterial cell division protein DivIB. Proc Natl Acad Sci USA 103: 67006705.
  • Robson, S.A., Michie, K.A., Mackay, J.P., Harry, E., and King, G.F. (2002) The Bacillus subtilis cell division proteins FtsL and DivIC are intrinsically unstable and do not interact with one another in the absence of other septasomal components. Mol Microbiol 44: 663674.
  • Romberg, L., and Levin, P.A. (2003) Assembly dynamics of the bacterial cell division protein FtsZ: poised at the edge of stability. Annu Rev Microbiol 57: 125154.
  • Rothfield, L.I., Shih, Y.-L., and King, G.F. (2001) Polar explorers: membrane proteins that determine division site placement. Cell 103: 1316.
  • Rowland, S.L., Katis, V.L., Partridge, S.R., and Wake, R.G. (1997) DivIB, FtsZ and cell division in Bacillus subtilis. Mol Microbiol 23: 295302.
  • Rubio, A., Jiang, X., and Pogliano, K. (2005) Localization of translocation complex components in Bacillus subtilis: enrichment of the signal recognition particle receptor at early sporulation septa. J Bacteriol 187: 50005002.
  • Sambrook, J., and Russell, D.W. (2001) Molecular Cloning: A Laboratory Manual. New York: Cold Spring Harbor Laboratory Press.
  • Sanchez-Pulido, L., Devos, D., Genevrois, S., Vicente, M., and Valencia, A. (2003) POTRA: a conserved domain in the FtsQ family and a class of β-barrel outer membrane proteins. Trends Biochem Sci 28: 523526.
  • Scheffers, D.J., Robichon, C., Haan, G.J., Den Blaauwen, T., Koningstein, G., Van Bloois, E., et al. (2007) Contribution of the FtsQ transmembrane segment to localization to the cell division site. J Bacteriol 189: 72737280.
  • Sekiguchi, J., Takada, N., and Okada, H. (1975) Genes affecting the productivity of α-amylase in Bacillus subtilis Marburg. J Bacteriol 121: 688694.
  • Sievers, J., and Errington, J. (2000) The Bacillus subtilis cell division protein FtsL localizes to sites of septation and interacts with DivIC. Mol Microbiol 36: 846855.
  • Stragier, P., and Losick, R. (1996) Molecular genetics of sporulation in Bacillus subtilis. Annu Rev Genet 30: 297241.
  • Szeto, T.H., Rowland, S.L., Rothfield, L.I., and King, G.F. (2002) Membrane localization of MinD is mediated by a C-terminal motif that is conserved across eubacteria, archaea, and chloroplasts. Proc Natl Acad Sci USA 99: 1569315698.
  • Thompson, L.S., Beech, P.L., Real, G., Henriques, A.O., and Harry, E.J. (2006) Requirement for the cell division protein DivIB in polar cell division and engulfment during sporulation in Bacillus subtilis. J Bacteriol 188: 76777685.
  • Vicente, M., Rico, A.I., Martinez-Arteaga, R., and Mingorance, J. (2006) Septum enlightenment: assembly of bacterial division proteins. J Bacteriol 188: 1927.
  • Weiss, D.S. (2004) Bacterial cell division and the septal ring. Mol Microbiol 54: 588597.
  • Weiss, D.S., Chen, J.C., Ghigo, J.-M., Boyd, D., and Beckwith, J. (1999) Localization of FtsI (PBP3) to the septal ring requires its membrane anchor, the Z ring, FtsA, FtsQ, and FtsL. J Bacteriol 181: 508520.
  • Wissel, M.C., and Weiss, D.S. (2004) Genetic analysis of the cell division protein FtsI (PBP3): amino acid substitutions that impair septal localization of FtsI and recruitment of FtsN. J Bacteriol 186: 490502.
  • Wissel, M.C., Wendt, J.L., Mitchell, C.J., and Weiss, D.S. (2005) The transmembrane helix of the Escherichia coli division protein FtsI localizes to the septal ring. J Bacteriol 187: 320328.

Supporting Information

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information
FilenameFormatSizeDescription
MMI_6114_sm_Tables_S1-S2.pdf127KSupporting info item
MMI_tables_s1-s2.pdf127KSupporting 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.