Cell Division Frederico Gueiros-Filho

4

Abstract Cell division in rod-shaped bacteria like Bacillus subtilis is carried out by a contractile protein ring, known as the divisome or septalsome, which is made up of about a dozen different polypeptides. This sophisticated macromolecular machine, which is centered around the tubulin-like protein FtsZ, is capable of promoting the coordinated invagination of the cell membrane and cell wall to create the so-called division septum. The goal of this chapter is to provide an overview of the mechanism of septum formation in B. subtilis. Emphasis will be placed on describing the properties of the individual division proteins and how they assemble into the divisome complex, and on a discussion of the regulatory mechanisms that ensure that septum formation will happen with great spatial and temporal precision at every cell cycle. In addition, the peculiar asymmetric division that happens during B. subtilis sporulation will be described. Introduction Cell division, in its most encompassing sense, is the process of generating two viable descendants from a progenitor cell. This involves two main coordinated events: the replication and segregation of the bacterial chromosome and the splitting of the progenitor cell by cytokinesis, which in bacteria is also known as septum formation. This chapter focuses on the mechanism of septum formation. For a detailed description of chromosome replication and segregation, see the Chapters 1 to 3 in this volume. Not surprisingly, given the importance of efficient reproduction for evolutionary success, bacterial cells have developed a remarkably sophisticated protein machinery capable of precisely splitting a progenitor cell at the right place and time in every cell cycle. This machine, which is often referred to as divisome or septalsome (septosome) (Nanninga, 1998), is based on a contractile protein ring, much as in the case of their more complex eukaryotic relatives. In contrast with eukaryotic cells, however, which frequently use actin and myosin as the base of their contractile protein ring, in bacteria the contractile machinery is based on the tubulin-like protein FtsZ.

Uncorrected proofs — not for distribution

94  | Gueiros-Filho

Understanding the composition and functioning of the divisome started almost 40 years ago when the first mutants impaired on division were isolated (Hirota et al., 1968; Nukushina and Ikeda, 1969). These mutants provided a first handle on the components of the division pathway. It was not until the pioneering work of Bi and Lutkenhaus in the early 90s, which used electron microscopy to determine the distribution of FtsZ during the cell cycle and discovered the Z ring (Bi and Lutkenhaus, 1991), that the function of these components began to be understood. Since then, the field has blossomed, advanced by the powerful genetic tools available for E. coli and B. subtilis, as well as the intensive use of cell biological approaches. In this regard, it is noteworthy that the study of cytokinesis, by popularizing the use of protein localization methods, has helped to establish a new view of cellular microbiology, centered on the idea that bacterial cells possess bona fide cytoskeletal elements and discretely localized proteins and thus are far more architecturally complex than previously anticipated. The goal of this chapter is to review the mechanism of cytokinesis in B. subtilis. This mechanism seems to be generally conserved between B. subtilis and E. coli and often E. coli information will be used to fill gaps in the B. subtilis literature. Whenever appropriate, the distinctions between B. subtilis and E. coli will be highlighted. This chapter starts with a brief description of the anatomy of the division site and an overview of cytokinesis. Next, it will present information on the components of the B. subtilis divisome, followed by a discussion of the mechanisms involved in regulating the timing and placement of the division septum. Finally, the peculiar asymmetric division that happens during sporulation will be discussed. Anatomy of B. subtilis and E. coli envelopes: implications for cytokinesis Cytokinesis requires the coordinated remodeling of the different layers that make up the bacterial envelope. Even though the overall mechanism of septum formation is highly conserved in B. subtilis and E. coli, there are some differences in the architecture of their envelopes and the way in which the septum is constructed that are worth highlighting. The envelope of B. subtilis, typical of Gram-positive bacteria, consists of the cytoplasmic membrane plus a thick cell wall made of peptidoglycan and associated anionic polymers such as teichoic acid (Fig. 4.1A) (see Chapter 10 by Scheffers). The situation in E. coli is more complex, since it has an LPS-containing outer membrane besides the cell wall and the cytoplasmic membrane (Fig. 4.1C). The cell wall in E. coli is also much thinner than in B. subtilis, probably a single peptidoglycan molecule thick. The presence of an outer membrane does not seem to create a need for special division mechanisms, since it seems to passively track with the cell wall. The difference in cell wall, on the other hand, may be related to a marked difference in the anatomy of the B. subtilis and E. coli developing division sites. In B. subtilis, division happens by the synthesis of a complete cross wall of peptidoglycan sandwiched by membrane bilayers. Initially, there is no obvious invagination or

Uncorrected proofs — not for distribution

Cell Division |  95

Figure 4.1 Anatomy of septation in B. subtilis and E. coli. (A) Schematic representation of the B. subtilis envelope and (B) fluorescence microscopy image of a septating B. subtilis cell. (C) Schematic representation of the E. coli envelope and (D) fluorescence microscopy image of a septating E. coli cell. Images B and D represent live cells that were treated with a fluorescent membrane dye. Note the absence of constriction at the B. subtilis septum. (E) Electron micrograph of a growing septum in B. subtilis, highlighting the simultaneous invagination of the two envelope layers (CW and PM). (F) Typical chain of cells in an exponentially growing culture of B. subtilis. Chain formation is due to the delay that exists between septation and cell separation. Membrane stains are often used to assess septation in B. subtilis cultures since septa that have not yet undergone constriction are difficult to detect by phase contrast or DIC methods. CW, cell wall; PM, plasma membrane; OM, outer membrane.

constriction at the division site (Fig. 4.1A, B, E, F). Invagination and maturation of the septa into new hemispherical poles will eventually occur by the degradation of the inner part of the septal peptidoglycan. In contrast, in E. coli synthesis of the division septum is concomitant with constriction (Fig. 4.1C, D). This is probably due to the thinner peptidoglycan septal wall in E. coli, which gets split in the middle as it is synthesized (Holtje, 1998). The thick cell wall of B. subtilis is probably also to blame for the time it takes for the wall in between cells to get degraded and the poles to mature. A consequence of this delay between cell division and cell separation is that fast growing cultures of this bacterium reveal mostly cells associated into long chains (Fig. 4.1F). Each cell in these chains is fully separated from the next by a complete septum. Overview of cytokinesis Cytokinesis in B. subtilis, and probably for any other rod-shaped bacterium, can be summarized in the following series of events. First, FtsZ, which is normally a cytoplasmic protein (Fig. 4.2A), assembles into a ring like structure associated

Uncorrected proofs — not for distribution

96  | Gueiros-Filho

Figure 4.2  The steps of cytokinesis. (A) In a newborn cell, FtsZ (gray circles) and the other divisome proteins, here represented as a membrane associated complex (black “paddle”), are dispersed throughout the cell. (B) Upon the decision to divide, FtsZ selfassociates into the Z ring at the cell center. (C) The Z ring acts as a scaffold and recruits the other division proteins to the division site, forming the divisome. (D) The assembly of a complete divisome triggers the ingrowth of the peptidoglycan cell wall and the invagination of the plasma membrane required for the synthesis of the new septum. Septum formation is followed by constriction of the divisome and disassembly of its components. (E) The ingrowth of the cell wall and membrane will eventually produce a complete septum separating the progenitor cell into two independent daughter cells. Finishing of the septum requires a membrane fusion event to separate the membranes of the daughter cells. This step has been omitted from the figure for simplicity. (F) Separation of the daughter cells is accomplished by the degradation of the inner part of the septal peptidoglycan through the action of murein hydrolases.

with the inner face of the cytoplasmic membrane (Fig. 4.2B). This structure, known as the Z ring, is a cytoskeletal element that serves as a scaffold onto which the divisome complex will assemble. Z ring formation is the target of careful regulatory mechanisms, described in detail below, that ensure that division will happen in coordination with other cell cycle events and at the proper place along the rod-shaped cell. Next, the Z ring nucleates the assembly of the divisome by attracting about ten other proteins to the division site. These components of the divisome are mostly integral membrane proteins and their recruitment to the Z ring occurs through both direct and indirect interactions with FtsZ (Fig. 4.2C). Once a mature divisome has been formed, it will promote

Uncorrected proofs — not for distribution

Cell Division |  97

the coordinated synthesis of a transversal cell wall and the invagination of the cytoplasmic membrane (Fig. 4.2D). The ingrowth of the cell wall and membrane will eventually produce a complete septum separating the progenitor cell into two independent daughter cells. Completion of the septum requires a membrane fusion event to separate the membranes of the daughter cells (Fig. 4.2E). Septum formation is accompanied by the constriction and disassembly of the divisome into its elementary parts (Fig. 4.2D) such that division proteins will have been redistributed back to the cytoplasm and cell membrane at the end of cytokinesis. No divisome proteins remain associated with the newly formed septum (Fig. 4.2E). The last step of cytokinesis is separation of the daughter cells by the action of murein hydrolases that degrade the inner part of the septal peptidoglycan cross wall (Fig. 4.2F). B. subtilis division proteins There are currently 18 proteins known to be associated with division in B. subtilis: ClpX, DivIB, DivIC, DivIVA, EzrA, FtsA, FtsL, FtsW, FtsZ, MinC, MinD, Noc, PBP2B, SepF, SpoIIE, SpoIIIE, YneA and ZapA. These can be divided in two main groups: proteins that make up the divisome and are directly involved in the construction of the division septum (DivIB, DivIC, FtsA, FtsL, FtsW, FtsZ, PBP2B, SepF) and proteins that regulate the assembly of the divisome (ClpX, DivIVA, EzrA, MinC, MinD, Noc, YneA, and ZapA). Some of the regulatory proteins, like EzrA and ZapA, are tightly associated with the divisome and thus can be considered as part of this complex. SpoIIIE, a protein whose function is to coordinate septum assembly with chromosome segregation, represents a third functional group. A summary of the features of these division proteins can be found in Table 4.1. The table also points to the correspondence between the B. subtilis and E. coli proteins, since the nomenclature is not unified in both species. Most of the components of the B. subtilis divisome have counterparts in E. coli and many other bacteria. These genes seem to make up a core division machinery that was probably already present in the last common ancestor of all bacteria (Robson F. Souza, Sandro de Souza and Frederico Gueiros-Filho, unpublished observations). The genes for many cell division proteins are often found next to each other as well as to some cell wall biosynthetic genes in a conserved region that has been called the dcw (for “divison and cell wall”) cluster. The remarkably conserved gene order of this region in diverse bacterial genomes has led some authors to suggest that there are functional consequences of this arrangement for division and the maintenance of a rod shape (Mingorance et al., 2004). The dcw cluster of B. subtilis is located at around 135 degrees in the chromosome. The ftsA, ftsZ, divIB (ftsQ) pbpB (ftsI), ftsL and ftsW genes are some of the genes commonly present at the dcw cluster. In B. subtilis and several other Gram-positives, the dcw cluster seems to extend further on one side to include also the gene for the division site selection protein DivIVA and several conserved “ylm” genes (Massidda et al., 1998), which, based on recent evidence, are also likely to participate in

Uncorrected proofs — not for distribution

E. coli ortholog

ClpX

FtsQ

FtsB

–

–

FtsA

FtsL

FtsW

FtsZ

MinC

B. subtilis ortholog

ClpX

DivIB

DivIC

DivIVA

EzrA

FtsA

FtsL

FtsW

FtsZ

MinC

25 kDa cytoplasmic protein. Site-specific FtsZ inhibitor. Prevents polar Z ring formation (with MinD)

40 kDa cytoplasmic protein. Structurally and functionally related to tubulin. Self-assembles to forms cytoskeletal scaffold (Z ring) for divisome assembly

44 kDa multipass transmembrane protein. Putative peptidoglycan precursor transporter

13 kDa membrane protein with extracytoplasmic domain. Part of multiprotein complex connecting the peptidoglycan synthesizing proteins to the early divisome

48 kDa membrane associated protein. Structurally similar to actin. Stabilizes and tethers Z ring to membrane. Recruits downstream divisome proteins

65 kDa membrane protein with cytoplasmic domain. Negative regulator of FtsZ polymerization. Prevents formation of multiple Z rings

19 kDa cytoplasmic protein. Topological determinant of Min system. Localizes MinCD to cell poles thus restricting division inhibition to this location. Analogous function to MinE in E. coli

15 kDa membrane protein with extracytoplasmic domain. Part of multiprotein complex connecting the peptidoglycan synthesizing proteins to the early divisome

30 kDa membrane protein with extracytoplasmic domain. Part of multiprotein complex connecting the peptidoglycan synthesizing proteins to the early divisome

46 kDa cytoplasmic protein. Substrate recognition subunit of the ClpXP protease complex. Negative modulator of Z-ring formation

Features and function

Table 4.1 Summary of B. subtilis division genes and their E. coli counterparts

?1

FtsI or PBP3 (ftsI or pbpB)

–

–

FtsK

–

ZapA (ygfE)

Noc

PBP2b (pbpB)

SepF or YlmF

SpoIIE

SpoIIIE

YneA

ZapA

10 kDa cytoplasmic protein. Positive modulator of FtsZ. Binds to FtsZ and promotes polymer bundling in vitro. Stabilizes Z ring in vivo. Dispensable for septum formation

12 kDa predicted membrane protein. SOS-induced septation inhibitor. Analogous function to SulA in E. coli

87 kDa protein with four transmembrane segments and a large cytoplasmic domain. DNA pump involved in chromosome segregation and dimer resolution. Dispensable for septum formation in B. subtilis but required in E. coli

92 kDa transmembrane protein with cytoplasmic domain. Sporulation protein needed for efficient polar septum formation. Binds to FtsZ. Exhibits also phosphatase activity required for the activation of foresporespecific gene expression

17 kDa cytoplasmic FtsZ-binding protein. Absence reduces division frequency and affects septum structure. Overexpression can compensate for the absence of FtsA

79 kDa membrane protein with large extracytoplasmic domain. Transpeptidase involved in septal peptidoglycan biosynthesis. Likely to be associated with DivIB, DivIC and FtsL in a multiprotein complex comprising a subassembly of the divisome

33 kDa DNA-associated FtsZ inhibitor. Effector of nucleoid occlusion. Analogous function to SlmA in E. coli

29 kDa membrane associated ATPase. Interacts with MinC and brings it to membrane. Prevents polar Z-ring formation

Gene names that differ from the corresponding protein name are presented in parentheses. 1Noc is a member of the ParB family of DNA binding proteins that have several orthologs in E. coli. It is unclear which of the E. coli family members is the true ortholog of Noc.

MinD

MinD

100  | Gueiros-Filho

division (Fadda et al., 2003; Hamoen et al., 2006). The remaining genes are scattered throughout the genome. Curiously, in B. subtilis, the genes for the division inhibitor MinCD are in the same operon as the mre genes, which are the components of the actin-like cytoskeleton responsible for maintaining cell shape and chromosome segregation. It is unclear whether this organization implies a functional relationship between the division and morphogenesis systems. In the next sections, the current knowledge about the best-studied B. subtilis division proteins will be summarized. This will be presented in the order they assemble in the divisome. FtsZ and the Z ring The tubulin-like FtsZ is the earliest-acting division protein and a master coordinator of septum formation. This characterization is based on several observations. First, inactivation of FtsZ produces a block in division that seems to occur earlier than a block caused by inactivation of any other division protein (Taschner et al., 1988). Second, all known division proteins depend on FtsZ for their proper localization to the division site, while, conversely, localization of FtsZ to the division site is independent of any other division protein (Addinall et al., 1996; Errington et al., 2003). Third, cellular FtsZ levels seem to be the main determinant of division frequency (Ward and Lutkenhaus, 1985; Dai and Lutkenhaus, 1991). Fourth, FtsZ is the most widely distributed and conserved division protein. It is present in essentially all the bacterial genomes that have been sequenced to date, with a few notable exceptions, such as the obligate intracellular Chlamydia, the planctomycete Pirellula and the mollicutes Ureaplasma (Vaughan et al., 2004). Despite a low primary sequence identity with tubulin, FtsZ has been shown to be very similar to tubulin both at the structural and at the functional level. FtsZ is a 40 kDa globular protein that folds into two independent domains (Oliva et al., 2004) (Fig. 4.3A). Self-assembly of FtsZ involves interactions between the C-terminal domain of one subunit with the N-terminal domain of another subunit (Oliva et al., 2004) (Fig. 4.3B). Like tubulin, FtsZ polymer formation is regulated by guanine nucleotides. Binding of GTP to a pocket located in the N-terminal domain induces the assembly of polymers (Erickson et al., 1996; Mukherjee and Lutkenhaus, 1998) (Fig. 4.3C). An important consequence of FtsZ assembly is to activate its GTPase activity, which is essentially absent when the protein is monomeric. Structurally, activation of GTP hydrolysis is due to the positioning of a catalytic aspartate residue present in a C-terminal domain loop (the T7 loop) of one FtsZ monomer next to the GTP sitting in the binding pocket of another monomer. Hydrolysis leads to instability of the polymers and to disassembly. Thus, FtsZ polymers are highly dynamic assemblies in vitro (Mukherjee and Lutkenhaus, 1998), as would be expected of a cytoskeletal element that needs to undergo recurring assembly and disassembly during every cell cycle. Another similarity between the FtsZ and the tubulin cytoskeleton is

Uncorrected proofs — not for distribution

Cell Division |  101

Figure 4.3 FtsZ and the Z ring. (A) Crystal structure of FtsZ highlighting its main features. Note that the protein can be divided by the helix H7 into two independently folding domains (boundaries marked by dashed line): the N- and C-terminal domains. The nucleotide binding pocket is in the N-terminal domain whereas the T7 loop involved in catalyzing nucleotide hydrolysis is located in the C-terminal domain. (B) Model of an FtsZ protofilament showing the contacts involved in the self-association of monomers. The spacing between subunits is 42 Å, very similar to the spacing between subunits in a tubulin protofilament (40 Å). Molecules of GDP can be seen occupying the binding interface in between the monomers. Plus and minus signs indicate the polarity of the polymer. The plus end corresponds to the nucleotide binding surface of the protein. (C) Electron micrograph of a negatively stained sample of B. subtilis FtsZ polymerized by GTP. This sample exhibits predominantly straight protofilaments. (D) Fluorescence image of a Z ring in a dividing B. subtilis cell expressing GFP-tagged FtsZ. (E) A Z ring in three dimensions. 3D reconstruction of the Z ring was carried out from optical sections of a cell expressing FtsZ-GFP like the one in D. The dotted white line shows the boundaries of the imaged cell. Images in panels A and B adapted from Lowe et al., Annu. Rev. Biophys. Biomol. Struct. 2004;33:177–98 and reprinted with permission. Microscopy images in panels D and E adapted from Ben-Yehuda and Losick (2002) and reprinted with permission.

Uncorrected proofs — not for distribution

102  | Gueiros-Filho

that FtsZ assembly seems to be a cooperative process (Wang and Lutkenhaus, 1993; Huecas and Andreu, 2003; Chen et al., 2005). Structural studies of FtsZ polymerization using electron microscopy under different conditions have shown that FtsZ is capable of forming various types of polymers (Romberg and Levin, 2003). The simplest kind of polymer formed in the presence of GTP is the protofilament, a one-subunit thick head to tail string of FtsZ monomers (Fig. 4.3C). Protofilaments induced by GTP are straight polymers, indicating the absence of a tilt or angle in the bond between subunits. Straight protofilaments can also associate into two dimensional sheets or 3D bundles through lateral interactions (Gonzalez et al., 2003). The tendency of FtsZ polymers to form lateral interactions is greatly increased by the presence of cations such as Ca2+ (Mukherjee and Lutkenhaus, 1999), DEAE-dextran, or cationic phospholipids (Erickson et al., 1996) and by FtsZ binding proteins, such as ZapA discussed below. Polymerization can be achieved also in the presence of GDP as long as stabilizing factors such as DEAE-dextran are present in the reaction. In contrast with GTP-FtsZ, polymers formed by GDP-bound FtsZ are mostly curved. Helical tubes, mini rings and curved protofilaments have been detected in different preparations (Erickson et al., 1996; Lu et al., 1998). Both GTP-induced and GDP-induced FtsZ polymers are highly similar to tubulin polymers produced under equivalent in vitro conditions, reinforcing the notion that the two proteins are related. The main difference between FtsZ and tubulin is that FtsZ has never been seen to assemble as hollow microtubules like those formed by tubulin. Much less is known about the structure of the Z ring in live cells. The Z ring has never been detected using standard transmission electron microscopy. This is either due to inappropriate fixation conditions or because the FtsZ polymer that makes up the ring is not bulky enough to be distinguished above the high density of the surrounding bacterial cytoplasm. Based on the in vitro studies and the fact that in vivo FtsZ should be mostly bound to GTP, it is currently assumed that the physiologically relevant FtsZ polymer is some form of protofilament sheet. This sheet could contain between 3 and 12 protofilaments, based on estimates made in B. subtilis and E. coli that there are 5000 to 20 000 FtsZ molecules in a single cell (Feucht et al., 2001; Romberg and Levin, 2003). The actual number of protofilaments in a Z ring could be as low as one, if recent experiments indicating that the Z ring seems to contain only one third of the cellular FtsZ are taken into account (Stricker et al., 2002). Another important question that remains unanswered is whether the Z ring is made up of long continuous protofilaments or by an association of short protofilament segments. Advances, such as the use of cryoelectron tomography, which has already been used successfully for high resolution studies of other structural proteins in bacteria (Kurner et al., 2005), should facilitate the emergence of a more detailed description of the FtsZ polymer structure in vivo. Meanwhile, valuable information about the assembly and the dynamic behavior of the Z ring has been gathered by fluorescence microscopy. This

Uncorrected proofs — not for distribution

Cell Division |  103

approach revealed that Z ring assembly and disassembly are fast processes, occurring in a timescale of minutes (Addinall et al., 1997; Sun and Margolin, 1998). FtsZ assembly in vivo has been proposed to start from a discrete point on the cytoplasmic membrane and extend bidirectionally until a complete ring is formed (Addinall and Lutkenhaus, 1996). This has suggested the existence of a nucleation site at the membrane. The nature of this nucleation site is still elusive (see below). At any given time during cytokinesis, only about a third of the FtsZ in a cell is associated with the Z ring. The other two thirds are present in the cytoplasm. Photobleaching experiments with bacteria expressing GFP-tagged FtsZ have revealed that both in E. coli and in B. subtilis, FtsZ in the ring is constantly being exchanged with the cytoplasmic FtsZ pool (Stricker et al., 2002; Anderson et al., 2004). The rate of exchange is very fast (an average ring will turn over with a half life of between 8 and 30 seconds) and seems to correlate with the GTPase activity of the protein. The rate of exchange of ZipA, a membrane-associated FtsZ binding protein of E. coli, was as fast as that of FtsZ, indicating that not only the Z ring, but probably the whole divisome is a dynamic structure that is constantly recycling itself. The dynamic behavior of the Z ring and associated proteins has been proposed to be important to generate the mechanical force required for divisome constriction (Stricker et al., 2002). Z ring formation in vivo is affected by a number of proteins that interact with FtsZ and modulate its polymerization properties. These modulatory factors, which are capable of either inhibiting or stimulating the assembly of FtsZ, play key roles in the spatiotemporal regulation of Z ring formation (see section below on regulation). This is analogous to the regulation of the microtubule cytoskeleton by tubulin binding proteins such as the MAPs. FtsZ binding proteins—early divisome components In B. subtilis, there are four divisome proteins, FtsA, EzrA, ZapA and SepF (YlmF), which interact directly with FtsZ (Fig. 4.4A). Of the four, FtsA and ZapA are quite conserved, being present in diverse bacterial groups including E. coli. SepF, on the other hand, is present only in Gram-positive and cyanobacteria and EzrA is restricted to a subset of Gram-positive bacteria. A fifth well-studied FtsZ binding protein, ZipA, is absent from B. subtilis but present in E. coli and closely related enterobacteria. The FtsZ binding proteins of B. subtilis play different roles in division. FtsA seems to be important to connect the Z ring to the membrane and is required for the assembly of the later components of the divisome and septum formation. On the other hand, ZapA and EzrA affect the polymerization of FtsZ and are likely to play regulatory roles. A summary of the properties and function of FtsA and SepF are presented below. A detailed description of ZapA and EzrA can be found in the section on regulation of septum formation.

Uncorrected proofs — not for distribution

104  | Gueiros-Filho

Figure 4.4 Divisome components and assembly. (A) The B. subtilis divisome. This scheme represents a hypothetical divisome complex as it would be viewed in a cross section of the bacterium at the division site. Protein topologies and the interactions between FtsZ and the proteins that interact directly with it are accurately represented. Some of the other interactions are only meant to illustrate the existence of a complex and do not necessarily reflect reality (for example, there is no evidence for a direct interaction between FtsA and DivIB). (B) Pathway of divisome assembly in B. subtilis and E. coli. In each of these pathways, assembly of a given protein is dependent on all the proteins to its left and independent of the proteins to its right. Proteins that associate early with the divisome but that are not necessary for the recruitment of later proteins are shown in gray. Question marks represent placements for which experimental evidence is lacking. SepF has been omitted from the figure because there is little information on its interaction with other divisome components. CW, cell wall; CYT, cytoplasm.

Uncorrected proofs — not for distribution

Cell Division |  105

FtsA FtsA is a 45 kD cytoplasmic protein that belongs to the superfamily of ATPbinding proteins that includes actin, Hsp70 and hexokinase. Indeed, determination of the crystal structure of FtsA revealed that it is quite similar to actin (van den Ent and Lowe, 2000). Both proteins have a similar two-domain fold and nucleotide binding cleft. The main structural difference between actin and FtsA is in subdomain 1B, which in FtsA adopts an alternative topology and therefore has been named 1C (van den Ent and Lowe, 2000). In contrast with the other prokaryotic actin-like proteins MreB and ParM, FtsA does not seem to self-associate and form filaments under normal conditions. Studies in E. coli have shown that FtsA binds to a conserved 15 amino acid sequence present at the extreme C-terminus of FtsZ (Ma and Margolin, 1999; Wang et al., 1997). This is probably the case for B. subtilis as well given that several of the residues important for the FtsA-FtsZ interaction are present in B. subtilis FtsZ (Erickson, 2001). The C-terminal peptide of FtsZ has been shown to be also the binding site for the E. coli protein ZipA (Haney et al., 2001). The region of FtsA involved in binding to the C-terminus of FtsZ is still not known. It is also unknown whether the binding of FtsA to FtsZ is capable of promoting the assembly of FtsZ polymers, like ZapA and ZipA do. FtsA is required for septum formation in both B. subtilis and E. coli. Division inhibition due to the absence of FtsA is lethal in E. coli. In contrast, ftsA null mutants have been obtained in B. subtilis, although these are highly filamentous and unhealthy (Beall and Lutkenhaus, 1992). Several studies, predominantly in E. coli, have determined that FtsA seems to have two distinct roles in septum formation. The first role of FtsA is to promote the assembly of the Z ring. Assembly of the Z ring in E. coli requires both FtsA and ZipA, since mutants lacking either of these proteins still form Z rings (Addinall et al., 1996; Hale and de Boer, 1999; Liu et al., 1999). However, in an ftsA zipA double mutant, assembly of Z rings is hampered (Pichoff and Lutkenhaus, 2002). In B. subtilis, a mutation in ftsA alone seems to prevent Z ring formation ( Jensen et al., 2005). This is consistent with the fact that B. subtilis lacks ZipA. The primary role played by FtsA in B. subtilis is likely to be the rule rather than the exception, given that most bacteria have FtsA but lack a recognizable homolog of ZipA. FtsA promotes Z ring formation by bringing FtsZ polymers to the cytoplasmic membrane (Pichoff and Lutkenhaus, 2005). Although FtsA lacks a transmembrane segment, it associates with the membrane through a conserved C-terminal amphipathic alpha-helix (Pichoff and Lutkenhaus, 2005). Tethering of FtsZ to the membrane could facilitate Z ring formation by increasing the probability with which FtsZ polymers would associate into stable multifilament structures. Besides connecting FtsZ polymers to the membrane, binding of FtsA to FtsZ seems to stabilize FtsZ polymers. This is suggested by the observation that FtsA overexpression is capable of overcoming the division block mediated by the Min inhibitor (see below) both in E. coli ( Justice et al., 2000) and in B. subtilis (Alessandra Pancetti and Frederico Gueiros-Filho, unpublished).

Uncorrected proofs — not for distribution

106  | Gueiros-Filho

The second role of FtsA during septation is to recruit downstream divisome proteins to the Z ring. In the absence of FtsA none of the late division proteins localize properly to the division site (Errington et al., 2003). Recently, it has been shown that FtsA is indeed able to interact with late divisome proteins independently of FtsZ (Corbin et al., 2004). This same study implicated subdomain 1C as the important region for divisome protein recruitment (Corbin et al., 2004). The role of subdomain 1C was further supported by the observation that deletion of this region did not affect FtsA association with the Z ring but prevented the recruitment of several late proteins to the divisome (Rico et al., 2004). It is likely that FtsA also serves to connect the Z ring to the late division proteins in B. subtilis. Testing this will require finding specific mutants or using an assay that separates the two roles role of FtsA, since this protein seems to be strictly required for Z ring formation in B. subtilis ( Jensen et al., 2005). SepF (YlmF) SepF is a newly discovered component of the B. subtilis divisome (Hamoen et al., 2006; Ishikawa et al., 2006). The gene for SepF lies close to the gene for the division site selection protein DivIVA in an extension of the dcw cluster that is conserved among Gram-positive bacteria (see above). SepF is a 17 kDa protein that was shown to interact directly with FtsZ by two-hybrid analysis. Inactivation of the gene for SepF leads to mild division inhibition and to alterations in septum morphology. SepF is likely to complement the role of FtsA in promoting the assembly of the Z ring, since deletion of sepF exacerbates the division inhibition of an ftsA mutant, whereas SepF overexpression rescues the division block of the same mutant (Ishikawa et al., 2006). The alteration in septum morphology in a sepF mutant suggests that SepF may also play a role at a late stage of division, during the synthesis of the peptidoglycan crosswall (Hamoen et al., 2006). Late divisome components and the synthesis of septal peptidoglycan In B. subtilis, six integral membrane proteins are recruited to the divisome late in its assembly (Fig. 4.4A, B). None of these appear to bind directly to FtsZ. Thus, their recruitment must depend on interactions with FtsZ-binding proteins such as FtsA, although this has not been demonstrated experimentally in B. subtilis. One of these proteins, SpoIIIE, is involved in coordinating chromosome partitioning and septation. Most of the other late divisome proteins have large extracytoplasmic domains, suggesting that they function outside the cell in the construction of the septal peptidoglycan wall. Despite rather intense efforts by several laboratories, the actual functions of these proteins remain unknown. The only exception is PBP2B, which is a member of the penicillin binding protein family involved in the late stages of peptidoglycan assembly. In the following sections, an overview of what is known about each of these proteins is presented.

Uncorrected proofs — not for distribution

Cell Division |  107

SpoIIIE SpoIIIE is a member of a family of DNA translocases involved in chromosome segregation and conjugation of mobile elements (Iyer et al., 2004). The counterpart of SpoIIIE in E. coli is FtsK. These proteins generally have four transmembrane segments followed by a large cytoplasmic domain that functions as an ATP-driven DNA pump (Aussel et al., 2002; Bath et al., 2000) (see also Chapter 3). SpoIIIE localizes to the division complex through its membrane domain. Once constriction begins, SpoIIIE becomes organized as a focus at the leading edge of the closing septum. The SpoIIIE focus is presumed to represent a DNA translocation pore responsible for the correct distribution of chromosomes that occasionally fail to completely segregate and get trapped by the constricting septum (see Fig. 3.7) (Britton and Grossman, 1999). Thus, SpoIIIE represents functional link between the cytokinesis and the segregation systems in B. subtilis. The chromosome pumping role of SpoIIIE is crucial during sporulation, hence, the Spo name of the protein. This is because during sporulation, segregation of the chromosome destined for the developing spore always occurs after formation of the asymmetric septum (see Fig. 3.8, for a more detailed description of chromosome segregation during sporulation see Chapters 3 and 11). Besides pumping misaligned chromosomes, there is clear evidence for a role for FtsK in promoting chromosome dimer resolution by the Xer recombinase system in E. coli (Recchia et al., 1999; Steiner et al., 1999; Aussel et al., 2002). It is controversial whether SpoIIIE plays an equivalent role in B. subtilis (Lemon et al., 2001; Sciochetti et al., 2001). In E. coli, FtsK is required for septum formation, in addition to its role in chromosome segregation (Begg et al., 1995). Mutations in FtsK block divisome assembly by preventing the recruitment of other late division proteins such as FtsI, FtsL and FtsQ (Chen and Beckwith, 2001). The division-promoting activity of FtsK resides in the N-terminal membrane part of the protein and can be clearly separated from its DNA translocation activity (Draper et al., 1998; Liu et al., 1998; Wang and Lutkenhaus, 1998). In contrast, in B. subtilis, SpoIIIE is not required for divisome assembly. This is reflected in the fact that the N-terminal regions of the two proteins are quite divergent. DivIB, DivIC, and FtsL DivIB, DivIC, and FtsL are small bitopic membrane proteins, with a short N-terminal cytoplasmic region, a single transmembrane segment and a larger C-terminal extracytoplasmic region (Harry and Wake, 1989; Levin and Losick, 1994; Daniel et al., 1998). DivIB is the largest of the three (30 kDa), with a 210 amino acid globular extracytoplasmic region. DivIC and FtsL, on the other hand, have significantly shorter external regions (67 and 63 amino acids, respectively) that are predicted to adopt a coiled coil configuration. The E. coli counterparts of DivIB and DivIC are FtsQ and FtsB, respectively. FtsL follows the same nomenclature in both species.

Uncorrected proofs — not for distribution

108  | Gueiros-Filho

The three proteins localize to the division site at a late stage in cytokinesis (Harry and Wake, 1997; Katis et al., 1997; Daniel and Errington, 2000; Sievers and Errington, 2000b) and, with the exception of DivIB, are essential for septum formation (Levin and Losick, 1994; Daniel et al., 1998). DivIB mutants are also impaired in septation but this defect is mild when cells are grown at low temperatures (Beall and Lutkenhaus, 1989). Domain swap experiments demonstrated that only the extracytoplasmic domain of DivIB is required for localization and functioning in division (Chen et al., 1999; Katis and Wake, 1999). This seems also to be the case for DivIC (Katis and Wake, 1999). The function of FtsL, on the other hand, seems to require both its extracytoplasmic domain as well as its transmembrane segment (Ghigo and Beckwith, 2000; Sievers and Errington, 2000a). DivIB, DivIC, and FtsL are likely to interact to form a ternary complex, as recruitment of any one of these proteins to the divisome is dependent on the other two. Moreover, the intracellular levels of DivIC and FtsL, both intrinsically unstable proteins, are affected by the presence or absence of DivIB (Robson et al., 2002). Recent evidence obtained in E. coli suggests that the equivalent complex (FtsB-FtsL-FtsQ) is preassembled in the membrane before being recruited to the cytokinetic ring, thus representing a subcomplex of the divisome (Buddelmeijer and Beckwith, 2004). Interaction among the proteins in this complex is likely to involve the coiled coil domains of DivIC and FtsL. The interacting region in DivIB is not known. The DivIB-DivIC-FtsL subcomplex in B. subtilis might also include PBP2B, since this protein requires DivIB, DivIC, FtsL to localize to the divisome and, reciprocally, PBP2b is required to localize DivIB, DivIC, FtsL to the divisome (Daniel et al., 2000). Thus, the DivIB-DivIC-FtsL ternary complex seems to connect the early divisome proteins to the peptidoglycan synthesizing machinery. The instability of some of the proteins in this complex has led to the hypothesis that this may be a regulatory step in the maturation of the divisome (Errington et al., 2003). The association of DivIB-DivIC-FtsL with the early divisome is presumed to involve an interaction with FtsA, although no evidence is available to support this hypothesis. Finally, the relatively large size of the extracytoplasmic domain of DivIB suggests that this protein may carry out some other function in septum formation besides just promoting the recruitment of other divisome components. Interestingly, a recent report showed that DivIB affects the Spo0J/Soj proteins involved in the organization and segregation of chromosomal origins (Real et al., 2005). Thus, DivIB may provide another connection between the cytokinesis and segregation systems in B. subtilis, besides the one described above for SpoIIIE. PBP2B The assembly of peptidoglycan from its precursors requires the polymerization of GlucNAc-MurNAc sugar subunits into glycan strands and the connection of glycan strands through peptide chain cross links. The penicillin binding proteins

Uncorrected proofs — not for distribution

Cell Division |  109

(PBPs) are the family of enzymes that execute these reactions. For a more detailed account of the role of PBPs in peptidoglycan synthesis, see Chapter 10. There are over 10 different PBPs in B. subtilis and E. coli (see Table 10.1). High molecular weight PBPs are divided in two classes. Class A contains enzymes that possess both transglycosylase and transpeptidase activities. Class B, on the other hand, possess only transpeptidase activity (Popham and Young, 2003). There are two possible modes of peptidoglycan synthesis for a rod-shaped bacterium. During growth, elongation of the cell wall is required whereas during division, a transverse wall will be synthesized. Genetic evidence indicates that different class B PBPs are involved in the different kinds of peptidoglycan synthesis. Thus, in B. subtilis PBP2B, encoded by the pbpB gene, is responsible for septal wall synthesis. Mutation in pbpB blocks septum formation without affecting elongation (Daniel et al., 2000; Yanouri et al., 1993). In contrast, mutations in the genes encoding the other class 2 enzymes, PBP2a and PBPH, block elongation without affecting septum formation, and thus lead to the production of spherical cells (Wei et al., 2003). An equivalent situation exists in E. coli where PBP3 (encoded by ftsI) is essential for septum formation while PBP2 is required for elongation (Popham and Young, 2003). PBP2B is an 80 kDa protein with one transmembrane segment and two extracytoplasmic domains. One of these is the transpeptidase domain. The function of the other domain is unknown. In accordance with its role in septation, PBP2B and its E. coli counterpart localize to the division septum (see Fig. 10.4B) (Daniel et al., 2000; Weiss et al., 1999). As discussed above, localization of PBP2B to the divisome depends on DivIB, DivIC and FtsL, suggesting that these proteins associate cooperatively in a complex. How PBP2B interacts with its partner proteins in B. subtilis remains unclear, however. In E. coli, interaction with the divisome appears to involve the protein’s transmembrane segment (Weiss et al., 1999). Besides protein–protein interactions with divisome components, localization of PBP2B could also be promoted by the recognition of a specific substrate, such as a modified cell wall precursor, at the division site. A precedent for such a mechanism exists in the Gram-positive bacterium Staphylococcus aureus, where localization of the counterpart of PBP2B to the division site depends on the availability its transpeptidation substrates (Pinho and Errington, 2005). Other members of the PBP family besides PBP2B were recently found to localize preferentially, or at least partially, to the septum (Scheffers et al., 2004). One of these is PBP1, a class A enzyme, which is strongly targeted to the division complex (Pedersen et al., 1999; Scheffers et al., 2004). This suggests that these enzymes are also involved in septal synthesis, perhaps by being part of a peptidoglycan “factory” or holoenzyme complex (Holtje, 1998). Inactivation of PBP1 seems to cause only a slight inhibition of septation (Pedersen et al., 1999; Scheffers and Errington, 2004). The non-essential nature of PBPs other than PBP2B for septal peptidoglycan synthesis could be explained by the existence of extensive redundancy of activities.

Uncorrected proofs — not for distribution

110  | Gueiros-Filho

FtsW FtsW is a member of the SEDS (shape, elongation, division and sporulation) protein family. Proteins of this family are generally found associated with a cognate class B PBP. Thus, in E. coli, FtsW is a partner of the septation-specific PBP3, whereas the elongation-specific PBP2 is functionally associated with RodA, another SEDS protein in that organism. B. subtilis possesses three SEDS proteins: FtsW, RodA and SpoVE. FtsW is the partner of PBP2B, although they are not in the same operon, as are the orthologous genes in E. coli. RodA is a likely partner for the elongation PBPs, PBP2A and PBPH whereas SpoVE is a protein required for the synthesis of the spore peptidoglycan cortex. In E. coli, FtsW is required to localize PBP3 to the divisome (Mercer and Weiss, 2002). The localization of FtsW itself depends on FtsL and on FtsQ (DivIB in B. subtilis) (Mercer and Weiss, 2002). There are no published observations on either the essentiality or the requirements for FtsW localization in B. subtilis. Given that it seems to interact with PBP3 in E. coli, it is likely that FtsW will be part of a large complex with PBP2B, DivIB, DivIC and FtsL in B. subtilis. FtsW has 10 transmembrane domains (Gerard et al., 2002), leading to speculation that it will function as a transporter of peptidoglycan precursors that will be acted upon by PBP2B on the outside of the cell. FtsW could also provide a link between the Z ring and the membrane, since it has been reported that, in Mycobacteria, FtsW is capable of binding directly to FtsZ (Datta et al., 2002). This role of FtsW in Mycobacteria could have evolved to compensate for the absence of FtsA in these organisms. Assembly of the divisome An important issue in bacterial division is to define how the multiprotein complex of the divisome is formed. One of the approaches to this question has been to test how the elimination of one protein affects the recruitment of the remaining proteins to the complex. This type of experiment has produced a pattern of localization interdependency among the individual divisome proteins, which was discussed in the sections above, and that can be summarized in the form of a putative assembly pathway for divisome formation in both B. subtilis and E. coli (Fig. 4.4B) (Errington et al., 2003). These pathways revealed that the late proteins in B. subtilis seem to behave as a multiprotein complex—a subcomplex of the divisome—as shown by the fact that their recruitment is fully interdependent. The situation in E. coli is somewhat different, since the late proteins assemble following a predominantly linear dependency (Errington et al., 2003; Goehring and Beckwith, 2005). This could mean that the late proteins do not form a single subcomplex in this organism. The early proteins (FtsA, ZapA, EzrA in B. subtilis and FtsA, ZipA in E. coli), on the other hand, seem to assemble independently of each other, in line with their ability to interact directly with FtsZ. It is important to note that the pathways in Fig. 4.4B do not necessarily reflect a temporal sequence of events from Z ring to mature divisome. To deter-

Uncorrected proofs — not for distribution

Cell Division |  111

mine the order of events, experiments that address the timing of arrival of each of the proteins at the complex must be carried out. The data from the single report that has been done towards that end suggest that, in E. coli, the divisome assembles in two steps (Aarsman et al., 2005). In the first step, proteins up to FtsK will arrive at the complex. After a ~15 minute delay, the remaining proteins will assemble. The interval between the two waves of assembly could mean that the division site has to be modified by the early assembling proteins in some way, perhaps at the level of the local peptidoglycan structure, before the late proteins can recognize it. Finally, an important approach to understand divisome assembly will be to characterize the protein–protein interactions that hold the complex together. Several two-hybrid screens have identified potential interactions between divisome components (Di Lallo et al., 2003; Karimova et al., 2005). These screens reported a large number of potential interactions between the division proteins, several of them not anticipated by the dependency pathways. There are two potential technical caveats that may explain the surprisingly high number of interactions found. First, the screens were carried out in homologous systems (i.e. the interaction between a pair of E. coli proteins was measured in E. coli, in the presence of all the other cognate divisome proteins) such that a positive signal could be due to an indirect association between the tested proteins. Second, they often relied on overexpression of the proteins to be tested, a situation that may promote physiologically irrelevant interactions. The final steps of septum formation: divisome constriction, membrane fusion, and cell separation Once the divisome is fully assembled, it will trigger the synthesis of the septal peptidoglycan and subsequent invagination of the cell envelope. The signal that induces peptidoglycan synthesis and constriction remains unknown. Another important mystery is the force that drives constriction of the divisome and the invagination of the envelope. There are two main views about this problem. One view argues that constriction is an indirect consequence of the ingrowth of a rigid peptidoglycan wall. The other view is that the force for constriction results from the dynamic behavior of the FtsZ cytoskeleton. This view is supported by three types of observations. First, there are bacteria like Mycoplasma, which are devoid of cell walls and whose division machinery seems to be made up solely of FtsZ (Vaughan et al., 2004). Second, wall-less forms of bacteria that normally contain peptidoglycan, such as E. coli’s L-forms, are still capable of dividing (Siddiqui et al., 2006). And third, membrane invagination can be uncoupled from septal wall growth by mutations such as the inactivation of PBP2B (Daniel et al., 2000). How the FtsZ cytoskeleton would produce force is still a matter of debate. An actomyosin-like mechanism seems unlikely, given that there are no known motor proteins in bacteria that could power constriction by sliding FtsZ filaments against each other. Other potential mechanisms would include the regulated disassembly of the FtsZ polymer while maintaining the ring associ-

Uncorrected proofs — not for distribution

112  | Gueiros-Filho

ated with the membrane, or alterations in filament curvature induced by the hydrolysis of GTP to GDP in the polymer (Romberg and Levin, 2003). Even though constriction can occur in the absence of a cell wall, it cannot be ruled out that under normal conditions constriction will be powered by both FtsZ and cell wall ingrowth. Constriction of the cell envelope will eventually promote the formation of a complete septum. A required step in the completion of the septum is the fusion of the invaginating membranes. How septal membrane fusion is accomplished is not known. There are no obvious homologs of eukaryotic proteins specialized in membrane fusion such as dynamins and SNAREs in bacterial genomes. Recent findings, however, suggest that the DNA-translocation protein SpoIIIE could play a role in membrane fusion during the engulfment step of sporulation (see Fig. 11.1, and further details in Chapter 11) (Sharp and Pogliano, 2003). Sporulation membrane fusion promoted by SpoIIIE seems to be carried out by its N-terminal membrane domain. Since a deletion of SpoIIIE has no effect in septum formation, it remains unclear what is the contribution of this protein for septal membrane fusion during growth. Following completion of the septum, hydrolysis of the inner portion of the peptidoglycan cross-wall allows the final separation of the daughter cells. As discussed before, daughter cell separation in B. subtilis does not happen simultaneously to septum formation. Hydrolysis of septal peptidoglycan is carried out by murein hydrolases, a large family of enzymes involved in several aspects of peptidoglycan turnover and recycling necessary for growth and division. At least five of these enzymes have been implicated in cell separation in B. subtilis, since mutations in their genes lead to an increase in the average size of the chains of cells in the mutants (Smith et al., 2000). One way to restrict the hydrolase activity to the desired location during cell separation would be to target the enzymes to the septum. This seems to be the case for the LytF and LytE proteins (Yamamoto et al., 2003). In E. coli, the AmiC murein hydrolase was also shown to localize specifically to the periplasmic space around the septum (Bernhardt and de Boer, 2003). Localization of AmiC to the septum was dependent on recruitment by the late divisome protein FtsN. There is no information on how LytF and LytE are recruited to the B. subtilis septum. Regulation of septum formation As discussed above, several lines of evidence have demonstrated that Z ring formation is the earliest detectable event and the rate-limiting step in septum formation. Thus, any regulatory mechanism operating to ensure that cytokinesis will happen at the right place and time during the cell cycle must be affecting either FtsZ availability or its propensity to assemble into the Z ring. Timing of Z-ring formation during the cell cycle Restricting FtsZ expression to a period of the cell cycle would be the simplest explanation for the timing of Z-ring formation. This hypothesis has been tested

Uncorrected proofs — not for distribution

Cell Division |  113

using synchronized populations of E. coli with conflicting results. Even though there are at least two studies that showed fluctuations in the amount of ftsZ mRNA (Garrido et al., 1993; Zhou and Helmstetter, 1994), levels of FtsZ or of the FtsZ-binding proteins FtsA and ZipA do not seem to change during the E. coli cell cycle (Rueda et al., 2003). Because B. subtilis is not as amenable to synchronization as E. coli, studies of B. subtilis ftsZ expression have not emphasized cell cycle control and focused instead on the developmental regulation of this gene during sporulation. In B. subtilis, ftsZ forms an operon with the neighboring ftsA and both genes are expressed from three promoters located upstream of ftsA. Two of these promoters (P1 and P3) are controlled by the housekeeping sigma A factor and are “on” during vegetative growth. The third promoter (P2) is turned on by a stationary phase/sporulation specific sigma factor and is responsible for an increase in FtsZ concentration as cells enter sporulation (Gholamhoseinian et al., 1992; Gonzy-Treboul et al., 1992). This burst in FtsZ expression seems to be required for the switch in Z ring position that occurs during sporulation, as will be discussed below. Despite the absence of any evidence for fluctuation of ftsZ mRNA or protein levels during the cell cycle, the observation that the natural ftsZ promoters can be substituted by an IPTG-controlled promoter without a noticeable alteration in the pattern of division (Beall and Lutkenhaus, 1991; Weart and Levin, 2003) suggests that transcriptional regulation is not the main component of Z ring control in B. subtilis. The only bacterium for which there is clear evidence for cell cycle control of FtsZ levels is Caulobacter crescentus. In this bacterium, every division cycle produces two different types of progeny: the sessile stalked cell, which is competent for chromosome replication and cell division; and the flagellated swarmer cell, which has to differentiate back to a stalked cell before it can reproduce. Remarkably, FtsZ is absent from swarmer cells as a consequence of targeted proteolysis combined with gene repression by CtrA, the master cell cycle regulatory protein of C. crescentus (Kelly et al., 1998). As swarmer cells differentiate into replication-competent stalked cells, FtsZ levels increase concomitantly. However, the fluctuation of FtsZ levels during the C. crescentus cell cycle is likely to be an exceptional situation, given that it is associated with the peculiar differentiation that occurs at every C. crescentus cell cycle, which is absent from most bacteria. Thus, the current view is that the timing of Z ring formation in B. subtilis and most bacteria is not primarily determined by fluctuating FtsZ levels. Instead, the timing may depend on mechanisms operating at the level of the activity of preexisting FtsZ, such as factors that inhibit or stimulate FtsZ polymerization. Another aspect of the timing of cytokinesis has been to define whether Z ring formation is dependent on other cell cycle events, in particular DNA replication. Studies carried out with germinating spores as a means to obtain synchronous B. subtilis populations have suggested that Z ring formation depends on both replication initiation and the cells being capable of replicating at least the origin proximal region of their chromosome (Harry, 2001b). In contrast, Z ring

Uncorrected proofs — not for distribution

114  | Gueiros-Filho

formation in E. coli was shown to coincide with termination of DNA replication (Den Blaauwen et al., 1999). Despite these observations, the actual connection, if any, between DNA replication and Z ring formation remains unclear. In B. subtilis, it has been suggested that during the initial phases of DNA replication the replisome would be located at midcell where it would mask a putative FtsZ nucleation site. Upon replication progression, the replisome would move away from midcell, freeing the nucleation site for Z ring assembly. This model has been recently disfavored by the finding that replisome localization during early DNA replication is not restricted to the cell center and seems too dynamic to account for the masking of a Z nucleation site (Migocki et al., 2004). An alternative model to explain the coordination between DNA replication and cytokinesis invokes the relieving of nucleoid occlusion by chromosome segregation, as described in more detail below (Fig. 4.5A, B). Finally, it must be emphasized that, in contrast to eukaryotes, there is no strict interdependence or checkpoint tying cytokinesis to DNA replication in bacteria, because both in B. subtilis and in E. coli Z-ring formation is only delayed but will eventually occur even in the absence of DNA replication (Harry, 2001b). The converse is also true, as inhibition of septation will generate long filaments containing multiple, evenly spaced, nucleoids (that contain the chromosomes, see Fig. 3.2A), indicating that septum formation is not required for DNA replication or segregation to occur. Spatial control of septum formation One of the most intriguing questions in bacterial cytology is how can a bacterium determine where its middle is so precisely? Measurements of the positioning of Z rings in B. subtilis have shown that the division site is placed to within 6% of the cell middle, i.e. with very high precision (Migocki et al., 2002). Initial models to explain this exquisite site preference proposed that only a few sites along the rod-shaped cell (more specifically, the midcell and the cell poles) would be recognized as potential division sites. These potential division sites would be defined by the existence of a positive signal, also referred to as a nucleation site, at the place where the Z ring was supposed to form. These models have gradually lost their appeal because no one could demonstrate the existence, let alone reveal the nature, of the putative positive signal triggering Z-ring formation. Moreover, there has been mounting evidence that Z ring formation can potentially happen anywhere along the bacterial cell cylinder (Harry, 2001a; Margolin, 2001). Thus, the current view is that the division site is determined by two complementary inhibitory systems that prevent Z ring formation everywhere but at midcell (Fig. 4.5A). These two systems are the Min proteins and nucleoid occlusion. The Min system Initial insight into the control of septum position in bacteria came from the isolation of mutants in which division frequently occurred asymmetrically at one of the cell poles, generating a larger cell with two nucleoids and a small DNA-less minicell (Adler et al., 1967; Reeve et al., 1973). These mutants defined

Uncorrected proofs — not for distribution

Cell Division |  115

Figure 4.5 Spatio-temporal control of cytokinesis. (A) Positioning of the division septum at midcell by the combined action of two inhibitory systems. The MinCD proteins (gray crescents at the cell poles, top panel) prevent the use of polar sites for division. Nucleoid occlusion (N.O.) mediated by proteins such as Noc (middle panel) prevents Z-ring formation in the vicinity of the nucleoids, and thus in the region between the cell poles and the midcell; the combined action of both systems restricts Z ring formation to the middle of the cell (bottom panel). (B) A model for the cell cycle timing of Z ring formation. At birth, the combination of polarly localized MinCD and the central location of an unreplicated nucleoid prevent Z ring formation throughout the cell. As the cell progresses through the cell cycle, replication and concomitant segregation of the nucleoid will eventually create a zone free of inhibitor at the cell middle.

the so-called min genes, encoding a set of proteins responsible for suppressing division specifically at the cell poles. In B. subtilis, the Min system is composed of the proteins MinC, MinD and DivIVA (Cha and Stewart, 1997; Edwards and Errington, 1997; Levin et al., 1992). MinC and MinD interact to form a membrane-associated complex capable of inhibiting Z ring formation in a sitespecific manner. Inhibition of Z ring formation by the MinCD complex has been shown to be carried out by MinC, a 25 kDa protein that is organized in two functional domains: the N-terminal domain interacts with FtsZ while the C-terminal domain has been implicated in MinC dimerization and interaction with MinD (Hu et al., 1999). MinD, on the other hand, is a 29 kDa ATPase of the ParA family, which associates with the membrane through an amphipathic alpha helix at its C-terminus (Szeto et al., 2003). Binding of MinC to MinD is necessary to localize and concentrate the inhibitory complex at the inner face of the cytoplasmic membrane, where it ultimately has to act. Recent evidence suggests that the interaction between MinC and MinD is necessary to activate the inhibitory action of MinC beyond simply increasing its local concentration near the membrane.

Uncorrected proofs — not for distribution

116  | Gueiros-Filho

In B. subtilis, DivIVA is the protein that imparts site selectivity to the Min inhibitor (Cha and Stewart, 1997; Edwards and Errington, 1997). Thus, this protein, and its functional analog MinE in E. coli, are commonly known as the topological specificity factors of the Min system. DivIVA is a small protein with a long tropomyosin-like coiled-coil region and that is conserved in several Grampositive species. How does DivIVA restrict the action of the MinCD inhibitor to the poles? Insight into this question came from studies using GFP fusions to determine the subcellular distribution of the Min and DivIVA proteins in B. subtilis (Marston et al., 1998). These fusions revealed that the MinCD complex is predominantly associated with the cell poles, exhibiting a crescent like staining pattern at the old pole and strong staining at the recently formed septum, which will become the new pole once cell separation happens. This localization was dependent on DivIVA, since in its absence MinCD appeared evenly distributed throughout the cell membrane. DivIVA itself was also shown to localize to the septum and the poles. Initial experiments suggested that polar localization of DivIVA was dependent on the assembly of the divisome, as it was blocked in the absence of FtsZ or other cell division proteins. This led to a model in which the anchoring of the MinCD complex to the cell poles would occur by a DivIVA-mediated recruitment of this complex to the divisome at a late stage in its assembly. Upon completion of division, the DivIVA-MinCD complex would be retained at the newly formed poles (Marston et al., 1998). Using the divisome to recruit a division inhibitor to its proper location may sound paradoxical. It must be remembered, however, that MinCD inhibits the first step of divisome assembly such that its association with a mature divisome would be unlikely to affect its stability. The recent observation that DivIVA and MinD localize to the poles of outgrowing spores even before the assembly of the division complex challenges the notion that recruitment of DivIVA, and consequently of MinCD, requires a direct interaction between DivIVA and the division complex (Harry and Lewis, 2003). Although this does not rule out the existence of affinity between DivIVA and the division complex, it suggests that DivIVA would be also attracted by some physical property of the division site and cell poles, such as membrane curvature. This view is supported by the fact that DivIVA is capable of associating with the cell poles when expressed heterologously in E. coli and even in S. pombe, a eukaryote whose division machinery is completely unrelated to the bacterial one (Edwards et al., 2000). Besides piloting MinCD to the cell poles, DivIVA is also involved in the anchoring of the origin regions of chromosomes to the cell poles during sporulation (Thomaides et al., 2001; Ben-Yehuda et al., 2003). This is important to ensure correct segregation of a chromosome copy to the developing spore and will be discussed in more detail below. Thus, the ability of DivIVA to recognize and mark the poles of a bacterial cell has been co-opted during evolution to play two distinct roles in the reproductive cycle of B. subtilis. In E. coli, the MinCD complex is also restricted to the cell poles. However, the way in which the polar localization of MinCD is achieved is remarkably

Uncorrected proofs — not for distribution

Cell Division |  117

different. In contrast to the static localization of MinCD in B. subtilis, MinCD in E. coli exhibit a dynamic oscillation in which the proteins move from one pole of the cell to the other approximately every minute (Raskin and de Boer, 1999; Rothfield et al., 2005). During this oscillation, MinD (and the associated MinC) initially forms a polar zone, a MinD-rich region on the membrane periphery extending from one of the poles toward the cell middle. MinD in the polar zone exists as a helical polymer created by the self-assembly of ATP-bound MinD subunits. Once the polar zone has grown enough, the topological specificity protein MinE will associate with it, creating the so-called MinE ring at the leading edge of the MinD polymers. The MinE ring will then promote the disassembly of the polar zone by inducing the ATPase activity of MinD. Next, the disassembled MinD, which now is bound to ADP, will diffuse away from the pole, exchange the bound ADP for ATP and then reassemble into a new polar zone at the opposite cell pole, until the next round of MinE-mediated disassembly occurs (Rothfield et al., 2005). The net effect of this fast alternation between cell poles is that the average occupancy of these sites by the MinCD proteins during the division cycle will be higher than that of the cell middle. Thus, over time, this dynamic system promotes the same effect as the static anchoring of MinCD in B. subtilis. An intriguing evolutionary question is why do two mechanistically quite different solutions for the problem of targeting proteins to the cell poles exist? Despite its well-established capacity to block polar Z-ring formation, the Min system cannot explain by itself the use of the midcell site with high precision during bacterial division. This is supported by work in B. subtilis and E. coli showing that the midcell site is still favored even in the absence of the Min proteins under certain conditions, and by the fact that several bacteria, like C. crescentus, lack min homologs in their genomes (Levin et al., 1998; Margolin, 2001; Migocki et al., 2002). The residual spatial preference in the absence of the Min proteins is likely to be a consequence of the nucleoid occlusion effect, as described below. Nucleoid occlusion In the absence of the Min system cells divide only at midcell and at the poles, even though any position along the length of a rod-shaped cell can sustain Z ring formation. So, there must be a second inhibitory system preventing the use of the space between the cell middle and the poles. Since this space ordinarily corresponds to the position occupied by the nucleoid in the cell, an attractive possibility is that the nucleoid would play a role in defining the site of division. A precedent for such a mechanism exists in the case of eukaryotic cells like S. pombe where the nucleus defines the correct plane of division, and recent evidence described below has shown that the bacterial “nucleus” is indeed important for determining where cells will divide. The recognition that bacteria avoid forming a septum over their nucleoid under conditions that perturb DNA replication and segregation led to the for-

Uncorrected proofs — not for distribution

118  | Gueiros-Filho

mulation of the hypothesis that nucleoids were capable of exerting a “veto effect” on septum formation in their vicinity (Woldringh et al., 1991). This veto by the nucleoid has been called nucleoid occlusion (NO) and was shown to happen at the level of Z ring formation by subsequent studies (Harry, 2001a; Margolin, 2001). These studies produced two other important observations regarding NO. First, blocking of Z rings by the nucleoid seemed to be relieved as soon as the nucleoid had undergone partial replication and segregation and assumed a bilobed shape. And second, mutations that reduced the general organization or condensation state of nucleoids, such as those in the smc or mukB genes, greatly reduced the efficacy of NO. Therefore, NO seems to be affected by the topology and/or local DNA concentration in the vicinity of the cytoplasmic membrane. The molecular basis of NO remained mysterious until recent reports showed that in both B. subtilis and E. coli NO could be exerted by the DNAbinding proteins Noc (in B. subtilis) and SlmA (in E. coli) (Bernhardt and de Boer, 2005; Wu and Errington, 2004). These proteins were implicated in NO because mutations in their genes promoted a severe division block and synthetic lethality when combined with a min mutation. Microscopic observation of conditional noc min or slmA min double mutants revealed that instead of forming the typical rings, FtsZ assembled in a haphazard pattern of arcs, foci and other irregular structures often present across the nucleoids. The explanation for this initially surprising phenotype is that in the absence of its two main negative regulatory systems, there are too many sites available for FtsZ to polymerize and this prevents formation of a productive Z ring. In line with the idea that Noc and SlmA are nucleoid-associated inhibitors of Z ring formation, both proteins localize diffusely on the nucleoids and promote division inhibition when overexpressed. Moreover, deletion of noc or slmA led to frequent Z ring formation over nucleoids, and eventually nucleoid bisecting by a septum in situations in which chromosome replication or segregation were perturbed. Interestingly, however, Z ring formation and division continued to occur preferentially between segregated nucleoids in noc and slmA mutants when replication was unperturbed. This observation implies that nucleoids still exhibit some NO activity even in the absence of Noc and SlmA and suggests that there may be other nucleoid-associated protein inhibitors of Z ring formation awaiting discovery. Alternatively, there could be a protein-independent component to the NO effect, as originally predicted in the “transertion” model. As stated by this model, the NO effect would be a consequence of the molecular crowding effects arising from a physical connection between the nucleoid and the adjacent membrane promoted by the simultaneous transcription, translation and insertion (hence, “transertion”) of integral membrane proteins into the cytoplasmic membrane (Rabinovitch et al., 2003). Determining the precise contributions of Min and NO to the spatial control of septum formation will require a better understanding of the molecular mechanism of NO. Besides regulating the position of the septum, nucleoid occlusion could also provide a mechanism for temporal coordination between chromosome replica-

Uncorrected proofs — not for distribution

Cell Division |  119

tion and segregation, and septum formation. According to this hypothesis, a newborn cell will not form a Z ring immediately because the combined distribution of Min and NO would inhibit Z ring formation throughout the cell. Thus, at the beginning of the cell cycle there would be no place available for a Z ring to form. As cells progress through the division cycle, chromosome replication and segregation would occur, relieving nucleoid occlusion at midcell and creating an inhibitor-free region in the cell for the first time. Thus, the timing of Z ring formation would be determined by a change in the pattern of NO promoted by the concomitant replication and segregation of the bacterial nucleoid (Fig. 4.5B). Other regulators of Z-ring formation Even though the Min system and nucleoid occlusion seem to account for most of the spatial control of Z-ring formation, there are at least four other proteins in B. subtilis that affect the general propensity of FtsZ to assemble and thus contribute to the regulation of division. Three of these modulators (EzrA, ZapA, ClpX) play roles in normal cell division whereas YneA is an SOS-induced protein important to prevent division in cells that suffered DNA damage. ZapA ZapA is a positive modulator of FtsZ, which was first identified in B. subtilis as a protein capable of overcoming the division block produced by overexpression of the MinD inhibitor (Gueiros-Filho and Losick, 2002). ZapA is a small cytoplasmic protein (10 kDa) widely conserved in bacteria and that is recruited to the divisome at an early stage in its assembly through a direct interaction with FtsZ. Binding of FtsZ by ZapA in vitro promotes the formation of higher order assemblies of FtsZ (Gueiros-Filho and Losick, 2002; Low et al., 2004). If GTP is present in the reactions, bundles of straight protofilaments are formed. ZapA can also promote the assembly of FtsZ structures in the presence of GDP. In this case, curved FtsZ assemblies, such as minirings, are formed. The crystal structure of Pseudomonas aeruginosa ZapA was recently solved and revealed a two-domain protein with a globular head and an elongated stalk corresponding to a long C-terminal alpha helix (Low et al., 2004). The crystal structure also showed that the protein oligomerizes, predominantly through a coiled coil association between the C-terminal alpha helices of monomers. P. aeruginosa ZapA is a tetramer in the crystal but biochemical experiments in P. aeruginosa and B. subtilis suggest that the protein is likely a dimer in solution (Low et al., 2004) (Aaron Handler, Glenn King and Frederico Gueiros-Filho, unpublished). Despite the availability of 3D structures for both proteins, the details of the interaction between ZapA and FtsZ remain unclear at this point. In contrast to FtsA (and ZipA in E. coli), ZapA is not essential for Z ring or septum formation under normal laboratory conditions. The frequency and pattern of division of zapA mutants is essentially identical to wild type cells. ZapA becomes required, however, if secondary mutations are used to weaken the tendency of Z rings to form, either by reducing the available pool of cytoplasmic

Uncorrected proofs — not for distribution

120  | Gueiros-Filho

FtsZ or by deregulating a division inhibitor (Gueiros-Filho and Losick, 2002). This suggests that ZapA facilitates the assembly of the Z ring, perhaps through its ability to bundle FtsZ protofilaments. Whatever its exact contribution, the dispensability of ZapA indicates that it plays a redundant role in the assembly of the Z-ring. Likely candidates to overlap with ZapA are the FtsZ binding proteins FtsA and SepF described above. It remains a possibility that ZapA becomes required under specific physiological conditions. At present, there is no hint as to what these conditions could be. EzrA EzrA is a negative regulator of Z-ring formation that was identified in a genetic screen for suppressors of an ftsZ temperature-sensitive allele (Levin et al., 1999). It is a 65 kDa single-span membrane protein with a short N-terminal transmembrane segment and a large cytoplasmic domain. EzrA is distributed homogenously throughout the cytoplasmic membrane, but localizes preferentially to the Z-ring once it is formed (Levin et al., 1999). In vitro, EzrA binds to FtsZ and prevents the growth of polymers from unassembled protein (Haeusser et al., 2004). However, EzrA is incapable of disassembling preformed polymers. This inability of EzrA to disassemble FtsZ polymers might explain why EzrA can localize to the Z ring without disturbing it. The precise role of EzrA during division remains unclear. In contrast to the site-specific action of MinCD, EzrA seems to generally decrease the propensity of FtsZ to polymerize inside the cell. Thus, EzrA could be in charge of controlling that Z rings would not form prematurely or too frequently during the cell cycle. This is supported by the observation that inactivation of ezrA leads to cells that have more than one Z ring (Levin et al., 1999). Interestingly, the extra rings are frequently present at one of the cell poles, presumably because in the absence of EzrA, FtsZ polymers become hyperstable and therefore less sensitive to inhibitors, such as MinCD, which prevent their assembly at aberrant places in the cell. An alternative view of the function of EzrA is that it is necessary for the complete disassembly of the Z ring at the end of division. According to this view, the extra Z rings present at cell poles would be derived from incompletely disassembled polymer from the previous round of division. The idea that EzrA is involved in the disassembly of FtsZ polymers is also supported by its tendency to associate with the Z ring. Another interesting feature of EzrA is that it seems to functionally interact with several of the other regulators of FtsZ polymerization. Two well-documented examples are the observation that a mutation in ezrA is capable of suppressing the overexpression of MinCD (Levin et al., 2001), and the observation that a mutation in ezrA exhibits a synergistic effect and severely blocks division when combined with a mutation in zapA (Gueiros-Filho and Losick, 2002). The first observation is easy to rationalize—the elimination of one destabilizing factor compensating for the overexpression of another destabilizing factor—and has led to the idea that the overall balance of stabilizing versus destabilizing

Uncorrected proofs — not for distribution

Cell Division |  121

activities is key for the proper functioning of the FtsZ polymer during division (Romberg and Levin, 2003). The second case, however, seems paradoxical since the simultaneous elimination of a negative (EzrA) and a positive (ZapA) regulator would be expected to have compensatory effects on FtsZ polymer assembly and be benign for division. The division inhibition of the ezrA zapA double mutant suggests that the interplay between positive and negative regulators of FtsZ may be more complex than the simple addition or subtraction of opposite activities. ClpX ClpX is member of the HSP100 family of AAA+ ATPases, which play a role in protein degradation by serving as the substrate recognition portion of the ClpXP protease (see Fig. 7.1, and see Chapter 7 for more details). In addition, ClpX can function as a molecular chaperone in the absence of ClpP, preventing protein aggregation and promoting the remodeling and disassembly of multiprotein complexes (Burton and Baker, 2005). Recent observations have implicated ClpX as a negative regulator of Z ring formation in B. subtilis (Weart et al., 2005). Inactivation of clpX suppresses the division defect of a temperature sensitive ftsZts mutant. Conversely, ClpX overexpression inhibits Z-ring formation. The effect of ClpX on Z ring formation is independent of ClpP and does not involve proteolysis of FtsZ. Moreover, purified ClpX inhibits FtsZ polymerization in vitro. Thus, ClpX must be preventing the assembly or promoting the disassembly of FtsZ polymers through its chaperone activity. In E. coli, ClpX was shown to bind FtsZ suggesting that the modulatory role of this chaperone in cell division may be generally conserved among bacteria (Burton and Baker, 2005). YneA Induction of the SOS response by DNA damage in bacteria (see also Chapter 2 by Alonso) often leads to a block in division. This block in division offers cells time to repair all damage and resume replication of their genome, and prevents the formation of anucleate cells or the fragmentation of nucleoids by the formation of a septum over unreplicated chromosomes. Suppression of division by the SOS response in E. coli is caused by the induction of SulA, a protein that binds tightly to the T7 loop surface of FtsZ and inhibits its polymerization (Huisman et al., 1984; Cordell et al., 2003). In B. subtilis, suppression of division during the SOS response is caused by YneA (Kawai et al., 2003). The gene encoding YneA is located next to the gene for DinR, the repressor that controls the SOS response in B. subtilis. Induction of YneA causes filamentation and a reduction in the frequency of Z rings. No direct interaction between YneA and FtsZ could be demonstrated and no evidence is available on the effect of YneA on FtsZ polymerization (Kawai et al., 2003). Thus, it is still unclear whether YneA blocks division at the level of Z ring formation or at a later step. The absence of

Uncorrected proofs — not for distribution

122  | Gueiros-Filho

sequence similarity between YneA and SulA is consistent with the possibility that these proteins function by different mechanisms. Asymmetric division during sporulation When faced with nutrient deprivation, B. subtilis and related bacteria undergo a developmental process that leads to the formation of a spore, a highly specialized cell capable of resisting a number of environmental stresses and staying dormant for long periods of time. Once conditions become favorable again, spores germinate and give rise to vegetative cells capable of resuming growth. Spore formation starts with the asymmetric division of a progenitor cell into two unequal cellular compartments by the formation of a polar septum. The smaller cell will become the resistant spore and is called the forespore, whereas the larger compartment is called the mother cell. Establishment of morphological asymmetry by polar septation is a key event in triggering distinct programs of gene expression in each of the compartments and, thus, to promote spore differentiation. For a detailed description of sporulation see Chapter 11. The polar septum made during sporulation uses essentially the same division machinery responsible for septum formation in growing cells. Mutations that block division in vegetative cells also block formation of the sporulation septum (Beall and Lutkenhaus, 1991; Beall and Lutkenhaus, 1992). Thus, the alteration in septum position during sporulation must be due to the interaction of the division machinery with sporulation-specific proteins. In addition to the change in positioning, vegetative and sporulation septa also differ in septal peptidoglycan content. Sporulation septa have a thinner peptidoglycan layer than vegetative septa, a likely adaptation to the requirement that septal peptidoglycan be completely removed to permit the engulfment of the forespore by the mother cell (see Chapter 11). The only sporulation protein with a well-established role in polar septation is SpoIIE. This protein seems to play a role both in controlling the thickness of the peptidoglycan in the polar septum as well as in the switching of the location of the division site (Barak and Youngman, 1996; Feucht et al., 1996). How SpoIIE affects the septal peptidoglycan is not known. The role of SpoIIE in the change of division site is discussed below. The mechanism of the switch of division site The switch from medial to asymmetric division during sporulation is a consequence of reprogramming the sites allowed for Z ring formation in the cell (Levin and Losick, 1996). In fact, the pattern of Z ring formation during sporulation is exactly the opposite of that during growth: whereas in vegetative cells the midcell site is always used and the cell poles are forbidden, in sporulating cells, Z rings avoid the midcell site and instead form close to each of the cell poles generating a typical bipolar pattern. Subsequently, only one of these polar rings will develop into a mature division complex to generate the asymmetric septum (Fig. 4.6).

Uncorrected proofs — not for distribution

Cell Division |  123

Figure 4.6 Asymmetric division during sporulation. Left: During sporulation, cells will deliberately divide at the pole as a consequence of switching the position of Z rings. Repositioning of the Z ring from midcell to the cell poles involves a transient FtsZ spiral intermediate, which will eventually produce a cell with two polar Z rings. Only one of these Z rings will become a septum. Right: Representative fluorescence microscopy images of the different stages of asymmetric division. The top three panels show two cells with medial Z rings, a cell with a Z spiral, and a cell with bipolar Z rings, respectively, as detected by immunofluorescence with anti-FtsZ antibodies. The bottom panel shows the fluorescently labeled membranes of a typical sporulating cell after polar septation. At this later stage, the second FtsZ ring will have dissipated, due to inhibition of septum formation at the slower-forming Z ring. Microscopy images adapted from Ben-Yehuda and Losick (2002) and reprinted with permission.

Through careful cytological analysis, it is now recognized that the switch of Z rings to the cell poles involves the dynamic redeployment of FtsZ structures across the cell (Ben-Yehuda and Losick, 2002) according to the sequence of events depicted in Fig. 4.6. Initially, a cell that is ready to sporulate will construct a medial Z ring, which, in contrast to the rings in vegetative cells, will not mature into a complete divisome. Instead, this “metastable” Z ring seems to be converted into spiral-like structures that emanate from the midcell towards the cell poles. Eventually, the FtsZ spirals reach the poles where they are converted back to stable bipolar Z rings (Ben-Yehuda and Losick, 2002). FtsZ spirals had been previously detected in E. coli when FtsZ was overexpressed or in strains

Uncorrected proofs — not for distribution

124  | Gueiros-Filho

bearing certain mutant ftsZ alleles, but the detection of a spiral intermediate in sporulating B. subtilis was the first time in which these structures were shown to be physiologically relevant. In fact, these observations have led to the suggestion that the Z ring is actually a tightly wound helical polymer whose interconversion between ring and spiral structures could be controlled by changing the pitch of the FtsZ helices. Concomitant with the above-described FtsZ choreography, at the onset of sporulation, there is also an important change in nucleoid organization. The pair of nucleoids present in the sporulating cell fails to fully segregate and adopt an extended structure that stretches from pole-to-pole of the cell, and has been named the “axial filament” (see Fig. 3.8B). This change in nucleoid organization is key to allow the efficient trapping of a complete genome in the future spore through a specialized form of chromosome segregation (Errington, 2003). The axial filament is also likely to be an important factor in the Z ring switch, as discussed below. Even though a rather refined description of the events from commitment to sporulation to polar septation exists, a clear understanding of the molecular mechanism responsible for the change in preference for the division site is still lacking. Asymmetric division is controlled by two transcriptional regulators that orchestrate the first steps of sporulation: the response regulator Spo0A and the alternative sigma factor σH (see Fig. 11.2). Mutants lacking each of these fail to switch the position of Z rings in sporulation (Ben-Yehuda and Losick, 2002; Levin and Losick, 1996). Despite the fact that Spo0A and σH control about one hundred genes at the onset of sporulation (Britton et al., 2002; Molle et al., 2003), genetic experiments suggest that there are only two genes whose activity seem to be required for the Z ring switch. One of these is ftsZ itself, whose expression increases about two-fold during sporulation due to the activation of a supplementary σH-responsive (P2) promoter (Gholamhoseinian et al., 1992; Gonzy-Treboul et al., 1992). The other one is spoIIE, which encodes a multifunctional protein expressed exclusively during sporulation. SpoIIE is a membrane-bound protein phosphatase (see Fig. 11.5A) whose activity is required for triggering the gene expression program in the developing forespore compartment (Duncan et al., 1995; Errington, 2003). SpoIIE is also capable of binding directly to FtsZ, and this interaction has been hypothesized to contribute to the stability and the anchoring of FtsZ polymers to the cytoplasmic membrane (Lucet et al., 2000). Mutations inactivating either the ftsZ P2 promoter or spoIIE lead to a mild inhibition of the Z ring switch. The combination of both mutations, however, practically abolishes polar septation, indicating that these are the main factors required for the change in Z ring position (Ben-Yehuda and Losick, 2002). Further support for this hypothesis came from experiments in which simultaneously boosting the expression of ftsZ and artificially expressing SpoIIE in vegetative cells led to polar septum formation during growth (BenYehuda and Losick, 2002). How do the increase in FtsZ levels and the expression of SpoIIE promote the switch? In order to answer this question we should take into account the

Uncorrected proofs — not for distribution

Cell Division |  125

known division site selection mechanisms operating in growing cells. As discussed above, asymmetric division in vegetative cells is prevented by the presence of the MinCD inhibitor at the cell poles. The localization of DivIVA, the protein that guides MinCD to the cell poles, does not change significantly as cells enter sporulation (Thomaides et al., 2001). Although similar experimental evidence is lacking for MinCD, it seems reasonable to assume that at the onset of sporulation the Min proteins will be present and active at the cell poles. How, than, is the inhibitory action of MinCD at the poles overridden? One possibility is that the SpoIIE interaction with FtsZ makes polymer formation refractory to inhibition. Alternatively, MinCD would move away from the cell poles during sporulation. According to this model, displacement of MinCD from the cell poles would be promoted by RacA, a DNA binding protein induced early on during sporulation that promotes the anchoring of the replication origin region of nucleoids to the cell poles (Ben-Yehuda et al., 2003). Anchoring of the nucleoids to the cell poles depends on an interaction between RacA and DivIVA. Thus, the RacA-DivIVA interaction could alter the properties of DivIVA such that it would no longer be capable of maintaining MinCD at the cell poles (BenYehuda et al., 2003). Even though there is a compelling logic behind this model, further experiments will be necessary to elucidate how MinCD is neutralized during sporulation. Another important unanswered question about the mechanism of the Z ring switch is what is the role, if any, of nucleoid occlusion? During vegetative growth, nucleoid occlusion delays Z ring formation at the midcell position until the replicating nucleoids have undergone at least partial segregation. Given that the sister nucleoids remain unsegregated and instead form the “axial filament” structure at the onset of sporulation, nucleoid occlusion may explain why the cell stops using the midcell site. The formation of polar Z rings over the nucleoid in sporulation would seem like an argument against nucleoid occlusion being active at the time. It is possible, however, that some special property of the edges of the nucleoids, such as a special chromatin organization (Pogliano et al., 2002) or lowered local DNA concentration would reduce the effectiveness of nucleoid occlusion at the ends of the nucleoid without affecting it at midcell. It would be interesting to determine if inactivation of noc, the only currently known effector of nucleoid occlusion, would have any effect on the position of the sporulation septum. Why does only one of the bipolar Z rings becomes a septum? Another striking feature of division during sporulation is that only one of the two polar Z rings matures into a polar septum (Levin and Losick, 1996). Formation of a single polar septum is important to ensure that a cell pair consisting of a large mother cell and a small forespore is always formed, thus maintaining the morphological asymmetry required for proper spore development. The use of a single Z ring seems to result from the inability of the cell to simultaneously assemble two complete division complexes. This is suggested by observations

Uncorrected proofs — not for distribution

126  | Gueiros-Filho

that division proteins like FtsA and PBP2b localize to a single polar ring in sporulating cells (Daniel et al., 2000; Feucht et al., 2001). The choice of which of the two Z rings to use is probably stochastic. Thus, in contrast with vegetative cells, where Z-ring formation seems to be the limiting step for division, in sporulating cells septum formation seems to be limited by the availability of division proteins other than FtsZ. The Z ring that does not normally become a septum is also competent to assemble a mature divisome, as shown by the identification of mutants that divide at both poles, generating “disporic” cells. Formation of the two polar septa in a disporic mutant is sequential, in line with the idea that one or more divisome proteins are limiting (Pogliano et al., 1999). Maturation of the second septum probably depends on divisome components being made available by the disassembly of the complex involved in the synthesis of the first septum. Time-lapse microscopic observation of septum formation during early sporulation revealed that even normal cells tend to start the synthesis of the second septum, although these septa seem to regress and are rarely completed (Pogliano et al., 1999). These observations suggest the existence of a mechanism to actively suppress the formation of a second polar septum. Suppression of the second polar septum depends on the activation of three proteins, SpoIID, SpoIIM and SpoIIP, in the mother cell (Pogliano et al., 1999; Eichenberger et al., 2001). These three proteins have features of peptidoglycan hydrolases and are likely to inhibit the second polar septum by preventing the growth of the peptidoglycan crosswall. Perspectives During the past 15 years remarkable progress has been made in understanding the molecular mechanism of septum formation. In this period researchers went from a list of loci which affected division to our current picture of the divisome as a intricate multiprotein complex capable of promoting the ingrowth of the bacterial envelope with high spatial precision and in perfect synchrony with DNA replication and segregation. Furthermore, it has been clearly demonstrated that the assembly of the division machinery is coordinated by a bacterial cytoskeleton based on the tubulin-like protein FtsZ. Despite these advances there are many discoveries still to be made and throughout the chapter I have tried to highlight specific gaps in our knowledge. Among the things we still do not know, I have singled out a few themes there are likely to pose the greatest challenges and to generate the most exciting discoveries in this field in the near future. First, it will be necessary to refine our understanding of the mechanisms controlling Z-ring and divisome assembly. Even though we know a number of proteins that affect Z-ring formation, the details of how they function remain largely unknown. Another remaining mystery is how the timing of Z-ring formation is achieved. Current proposals invoking the relieving of nucleoid occlusion by chromosome segregation, although attractive, need to be experimentally validated. Furthermore, the recent discovery that DnaA plays a broader role than anticipated as a transcriptional regulator during the cell cycle (Goranov et

Uncorrected proofs — not for distribution

Cell Division |  127

al., 2005; Hottes et al., 2005) suggests that oscillations in the cellular concentration of division proteins might play a role in the timing of division after all. Perhaps the biggest challenge for the future will be to improve our understanding of the late events of septum formation, including how the synthesis of the septal peptidoglycan is controlled, the specific role of the late divisome proteins in this process and the source of the driving force for divisome constriction. Dealing with these questions will benefit from the recent development of reagents, such as a fluorescent derivative of vancomycin (Daniel and Errington, 2003), capable of probing the local structure of the peptidoglycan in intact cells. It will probably also require the reconstitution in vitro of at least some of the steps of septal peptidoglycan synthesis, a daunting task because the synthesis of peptidoglycan involves large membrane-embedded protein complexes acting on an insoluble polymeric template. Finally, if we are to understand the process of bacterial division in all its complexity and detail it will be important to eventually identify the full complement of cell division proteins in model organisms like B. subtilis or E. coli. The steady reporting of novel cell division proteins every year and the substantial number of predicted proteins still with unknown function in the B. subtilis genome suggests that there is no end in sight for this quest. Novel cell division proteins will likely also include previously known proteins with an unforeseen role in division, such as ClpX. If we take into account that the recently discovered division proteins were not required for septum formation or had very mild loss-of-function phenotypes it becomes evident that the comprehensive identification of new cell division proteins will not be trivial and shall require the combination of approaches such as clever genetic screens (gain-of-function, synthetic lethal) and large scale protein interaction and localization studies. Acknowledgments I would like to thank Jonathan Dworkin for critical reading of the manuscript, and Sigal Ben-Yehuda, Ulysses Lins and Fernanda Abreu for images. References

Aarsman, M.E., Piette, A., Fraipont, C., Vinkenvleugel, T.M., Nguyen-Disteche, M., and den Blaauwen, T. (2005). Maturation of the Escherichia coli divisome occurs in two steps. Mol. Microbiol. 55, 1631–1645. Addinall, S.G., Bi, E., and Lutkenhaus, J. (1996). FtsZ ring formation in fts mutants. J. Bacteriol. 178, 3877–3884. 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, 4277–4284. Addinall, S.G., and Lutkenhaus, J. (1996). FtsZ-spirals and -arcs determine the shape of the invaginating septa in some mutants of Escherichia coli. Mol. Microbiol. 22, 231–237. Adler, H.I., Fisher, W.D., Cohen, A., and Hardigree, A.A. (1967). Miniature Escherichia coli cells deficient in DNA. Proc. Natl. Acad. Sci. USA 57, 321–326. Anderson, D.E., Gueiros-Filho, F.J., and Erickson, H.P. (2004). Assembly dynamics of FtsZ rings in Bacillus subtilis and Escherichia coli and effects of FtsZ-regulating proteins. J. Bacteriol. 186, 5775–5781.

Uncorrected proofs — not for distribution

128  | Gueiros-Filho

Aussel, L., Barre, F.X., Aroyo, M., Stasiak, A., Stasiak, A.Z., and Sherratt, D. (2002). FtsK Is a DNA motor protein that activates chromosome dimer resolution by switching the catalytic state of the XerC and XerD recombinases. Cell 108, 195–205. Barak, I., and Youngman, P. (1996). SpoIIE mutants of Bacillus subtilis comprise two distinct phenotypic classes consistent with a dual functional role for the SpoIIE protein. J. Bacteriol. 178, 4984–4989. Bath, J., Wu, L.J., Errington, J., and Wang, J.C. (2000). Role of Bacillus subtilis SpoIIIE in DNA transport across the mother cell-prespore division septum. Science 290, 995–997. 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, 6821–6834. Beall, B., and Lutkenhaus, J. (1991). FtsZ in Bacillus subtilis is required for vegetative septation and for asymmetric septation during sporulation. Genes Dev. 5, 447–455. Beall, B., and Lutkenhaus, J. (1992). Impaired cell division and sporulation of a Bacillus subtilis strain with the ftsA gene deleted. J. Bacteriol. 174, 2398–2403. Begg, K.J., Dewar, S.J., and Donachie, W.D. (1995). A new Escherichia coli cell division gene, ftsK. J. Bacteriol. 177, 6211–6222. Ben-Yehuda, S., and Losick, R. (2002). Asymmetric cell division in B. subtilis involves a spiral-like intermediate of the cytokinetic protein FtsZ. Cell 109, 257–266. Ben-Yehuda, S., Rudner, D.Z., and Losick, R. (2003). RacA, a bacterial protein that anchors chromosomes to the cell poles. Science 299, 532–536. 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, 1171–1182. 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, 555–564. Bi, E.F., and Lutkenhaus, J. (1991). FtsZ ring structure associated with division in Escherichia coli. Nature 354, 161–164. Britton, R.A., Eichenberger, P., Gonzalez-Pastor, J.E., Fawcett, P., Monson, R., Losick, R., and Grossman, A.D. (2002). Genome-wide analysis of the stationary-phase sigma factor (sigma-H) regulon of Bacillus subtilis. J. Bacteriol. 184, 4881–4890. Britton, R.A., and Grossman, A.D. (1999). Synthetic lethal phenotypes caused by mutations affecting chromosome partitioning in Bacillus subtilis. J. Bacteriol. 181, 5860–5864. Buddelmeijer, N., and Beckwith, J. (2004). A complex of the Escherichia coli cell division proteins FtsL, FtsB and FtsQ forms independently of its localization to the septal region. Mol. Microbiol. 52, 1315–1327. Burton, B.M., and Baker, T.A. (2005). Remodeling protein complexes: insights from the AAA+ unfoldase ClpX and Mu transposase. Protein Sci. 14, 1945–1954. Cha, J.H., and Stewart, G.C. (1997). The divIVA minicell locus of Bacillus subtilis. J. Bacteriol. 179, 1671–1683. Chen, J.C., and Beckwith, J. (2001). FtsQ, FtsL and FtsI require FtsK, but not FtsN, for colocalization with FtsZ during Escherichia coli cell division. Mol. Microbiol. 42, 395–413. 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, 521–530. Chen, Y., Bjornson, K., Redick, S.D., and Erickson, H.P. (2005). A rapid fluorescence assay for FtsZ assembly indicates cooperative assembly with a dimer nucleus. Biophys. J. 88, 505–514. Corbin, B.D., Geissler, B., Sadasivam, M., and Margolin, W. (2004). Z-ring-independent interaction between a subdomain of FtsA and late septation proteins as revealed by a polar recruitment assay. J. Bacteriol. 186, 7736–7744. Cordell, S.C., Robinson, E.J., and Lowe, J. (2003). Crystal structure of the SOS cell division inhibitor SulA and in complex with FtsZ. Proc. Natl. Acad. Sci. USA 100, 7889–7894. Dai, K., and Lutkenhaus, J. (1991). ftsZ is an essential cell division gene in Escherichia coli. J. Bacteriol. 173, 3500–3506.

Uncorrected proofs — not for distribution

Cell Division |  129

Daniel, R.A., and Errington, J. (2000). Intrinsic instability of the essential cell division protein FtsL of Bacillus subtilis and a role for DivIB protein in FtsL turnover. Mol. Microbiol. 36, 278–289. Daniel, R.A., and Errington, J. (2003). Control of cell morphogenesis in bacteria: two distinct ways to make a rod-shaped cell. Cell 113, 767–776. Daniel, R.A., Harry, E.J., and Errington, J. (2000). Role of penicillin-binding protein PBP 2B in assembly and functioning of the division machinery of Bacillus subtilis. Mol. Microbiol. 35, 299–311. Daniel, R.A., Harry, E.J., Katis, V.L., Wake, R.G., and Errington, J. (1998). Characterization of the essential cell division gene ftsL(yIID) of Bacillus subtilis and its role in the assembly of the division apparatus. Mol. Microbiol. 29, 593–604. Datta, P., Dasgupta, A., Bhakta, S., and Basu, J. (2002). Interaction between FtsZ and FtsW of Mycobacterium tuberculosis. J. Biol. Chem. 277, 24983–24987. Den Blaauwen, T., Buddelmeijer, N., Aarsman, M.E., Hameete, C.M., and Nanninga, N. (1999). Timing of FtsZ assembly in Escherichia coli. J. Bacteriol. 181, 5167–5175. 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, 3353–3359. Draper, G.C., McLennan, N., Begg, K., Masters, M., and Donachie, W.D. (1998). Only the N-terminal domain of FtsK functions in cell division. J. Bacteriol. 180, 4621–4627. Duncan, L., Alper, S., Arigoni, F., Losick, R., and Stragier, P. (1995). Activation of cell-specific transcription by a serine phosphatase at the site of asymmetric division. Science 270, 641–644. Edwards, D.H., and Errington, J. (1997). The Bacillus subtilis DivIVA protein targets to the division septum and controls the site specificity of cell division. Mol. Microbiol. 24, 905–915. Edwards, D.H., Thomaides, H.B., and Errington, J. (2000). Promiscuous targeting of Bacillus subtilis cell division protein DivIVA to division sites in Escherichia coli and fission yeast. EMBO J. 19, 2719–2727. Eichenberger, P., Fawcett, P., and Losick, R. (2001). A three-protein inhibitor of polar septation during sporulation in Bacillus subtilis. Mol. Microbiol. 42, 1147–1162. Erickson, H.P. (2001). The FtsZ protofilament and attachment of ZipA—structural constraints on the FtsZ power stroke. Curr. Opin. Cell. Biol. 13, 55–60. Erickson, H.P., Taylor, D.W., Taylor, K.A., and Bramhill, D. (1996). Bacterial cell division protein FtsZ assembles into protofilament sheets and minirings, structural homologs of tubulin polymers. Proc. Natl. Acad. Sci. USA 93, 519–523. Errington, J. (2003). Regulation of endospore formation in Bacillus subtilis. Nat. Rev. Microbiol. 1, 117–126. Errington, J., Daniel, R.A., and Scheffers, D.J. (2003). Cytokinesis in bacteria. Microbiol. Mol. Biol. Rev. 67, 52–65. Fadda, D., Pischedda, C., Caldara, F., Whalen, M.B., Anderluzzi, D., Domenici, E., and Massidda, O. (2003). Characterization of divIVA and other genes located in the chromosomal region downstream of the dcw cluster in Streptococcus pneumoniae. J. Bacteriol. 185, 6209–6214. Feucht, A., Lucet, I., Yudkin, M.D., and Errington, J. (2001). Cytological and biochemical characterization of the FtsA cell division protein of Bacillus subtilis. Mol. Microbiol. 40, 115–125. Feucht, A., Magnin, T., Yudkin, M.D., and Errington, J. (1996). Bifunctional protein required for asymmetric cell division and cell-specific transcription in Bacillus subtilis. Genes Dev. 10, 794–803. Garrido, T., Sanchez, M., Palacios, P., Aldea, M., and Vicente, M. (1993). Transcription of ftsZ oscillates during the cell cycle of Escherichia coli. EMBO J. 12, 3957–3965. Gerard, P., Vernet, T., and Zapun, A. (2002). Membrane topology of the Streptococcus pneumoniae FtsW division protein. J. Bacteriol. 184, 1925–1931. 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, 116–129.

Uncorrected proofs — not for distribution

130  | Gueiros-Filho

Gholamhoseinian, A., Shen, Z., Wu, J.J., and Piggot, P. (1992). Regulation of transcription of the cell division gene ftsA during sporulation of Bacillus subtilis. J. Bacteriol. 174, 4647–4656. Goehring, N.W., and Beckwith, J. (2005). Diverse paths to midcell: assembly of the bacterial cell division machinery. Curr. Biol. 15, R514–526. Gonzalez, J.M., Jimenez, M., Velez, M., Mingorance, J., Andreu, J.M., Vicente, M., and Rivas, G. (2003). Essential cell division protein FtsZ assembles into one monomer-thick ribbons under conditions resembling the crowded intracellular environment. J. Biol. Chem. 278, 37664–37671. Gonzy-Treboul, G., Karmazyn-Campelli, C., and Stragier, P. (1992). Developmental regulation of transcription of the Bacillus subtilis ftsAZ operon. J. Mol. Biol. 224, 967–979. Goranov, A.I., Katz, L., Breier, A.M., Burge, C.B., and Grossman, A.D. (2005). A transcriptional response to replication status mediated by the conserved bacterial replication protein DnaA. Proc. Natl. Acad. Sci. USA 102, 12932–12937. Gueiros-Filho, F.J., and Losick, R. (2002). A widely conserved bacterial cell division protein that promotes assembly of the tubulin-like protein FtsZ. Genes Dev. 16, 2544–2556. 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, 801–814. Hale, C.A., and de Boer, P.A. (1999). Recruitment of ZipA to the septal ring of Escherichia coli is dependent on FtsZ and independent of FtsA. J. Bacteriol. 181, 167–176. Hamoen, L.W., Meile, J.C., de Jong, W., Noirot, P., and Errington, J. (2006). SepF, a novel FtsZ-interacting protein required for a late step in cell division. Mol. Microbiol. 59, 989–999. Haney, S.A., Glasfeld, E., Hale, C., Keeney, D., He, Z., and de Boer, P. (2001). Genetic analysis of the Escherichia coli FtsZ.ZipA interaction in the yeast two-hybrid system. Characterization of FtsZ residues essential for the interactions with ZipA and with FtsA. J. Biol. Chem. 276, 11980–11987. Harry, E.J. (2001a). Bacterial cell division: regulating Z-ring formation. Mol. Microbiol. 40, 795–803. Harry, E.J. (2001b). Coordinating DNA replication with cell division: lessons from outgrowing spores. Biochimie 83, 75–81. Harry, E.J., and Lewis, P.J. (2003). Early targeting of Min proteins to the cell poles in germinated spores of Bacillus subtilis: evidence for division apparatus-independent recruitment of Min proteins to the division site. Mol. Microbiol. 47, 37–48. 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, 6835–6839. 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, 275–283. Hirota, Y., Ryter, A., and Jacob, F. (1968). Thermosensitive mutants of E. coli affected in the processes of DNA synthesis and cellular division. Cold Spring Harb. Symp. Quant. Biol. 33, 677–693. Holtje, J.V. (1998). Growth of the stress-bearing and shape-maintaining murein sacculus of Escherichia coli. Microbiol. Mol. Biol. Rev. 62, 181–203. Hottes, A.K., Shapiro, L., and McAdams, H.H. (2005). DnaA coordinates replication initiation and cell cycle transcription in Caulobacter crescentus. Mol. Microbiol. 58, 1340–1353. 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, 14819–14824. Huecas, S., and Andreu, J.M. (2003). Energetics of the cooperative assembly of cell division protein FtsZ and the nucleotide hydrolysis switch. J. Biol. Chem. 278, 46146–46154. Huisman, O., D’Ari, R., and Gottesman, S. (1984). Cell-division control in Escherichia coli: specific induction of the SOS function SfiA protein is sufficient to block septation. Proc. Natl. Acad. Sci. USA 81, 4490–4494.

Uncorrected proofs — not for distribution

Cell Division |  131

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, 1364–1380. Iyer, L.M., Makarova, K.S., Koonin, E.V., and Aravind, L. (2004). Comparative genomics of the FtsK-HerA superfamily of pumping ATPases: implications for the origins of chromosome segregation, cell division and viral capsid packaging. Nucleic Acids Res. 32, 5260–5279. Jensen, S.O., Thompson, L.S., and Harry, E.J. (2005). Cell division in Bacillus subtilis: FtsZ and FtsA association is Z-ring independent, and FtsA is required for efficient midcell Z-Ring assembly. J. Bacteriol. 187, 6536–6544. 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, 410–423. 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, 2233–2243. 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, 1047–1055. 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, 2710–2718. Kawai, Y., Moriya, S., and Ogasawara, N. (2003). Identification of a protein, YneA, responsible for cell division suppression during the SOS response in Bacillus subtilis. Mol. Microbiol. 47, 1113–1122. Kelly, A.J., Sackett, M.J., Din, N., Quardokus, E., and Brun, Y.V. (1998). Cell cycle-dependent transcriptional and proteolytic regulation of FtsZ in Caulobacter. Genes Dev. 12, 880–893. Kurner, J., Frangakis, A.S., and Baumeister, W. (2005). Cryo-electron tomography reveals the cytoskeletal structure of Spiroplasma melliferum. Science 307, 436–438. Lemon, K.P., Kurtser, I., and Grossman, A.D. (2001). Effects of replication termination mutants on chromosome partitioning in Bacillus subtilis. Proc. Natl. Acad. Sci. USA 98, 212–217. 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, 9642–9647. Levin, P.A., and Losick, R. (1994). Characterization of a cell division gene from Bacillus subtilis that is required for vegetative and sporulation septum formation. J. Bacteriol. 176, 1451–1459. Levin, P.A., and Losick, R. (1996). Transcription factor Spo0A switches the localization of the cell division protein FtsZ from a medial to a bipolar pattern in Bacillus subtilis. Genes Dev. 10, 478–488. Levin, P.A., Margolis, P.S., Setlow, P., Losick, R., and Sun, D. (1992). Identification of Bacillus subtilis genes for septum placement and shape determination. J. Bacteriol. 174, 6717–6728. Levin, P.A., Schwartz, R.L., and Grossman, A.D. (2001). Polymer stability plays an important role in the positional regulation of FtsZ. J. Bacteriol. 183, 5449–5452. Levin, P.A., Shim, J.J., and Grossman, A.D. (1998). Effect of minCD on FtsZ ring position and polar septation in Bacillus subtilis. J. Bacteriol. 180, 6048–6051. 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, 893–903. Liu, Z., Mukherjee, A., and Lutkenhaus, J. (1999). Recruitment of ZipA to the division site by interaction with FtsZ. Mol. Microbiol. 31, 1853–1861. Low, H.H., Moncrieffe, M.C., and Lowe, J. (2004). The crystal structure of ZapA and its modulation of FtsZ polymerisation. J. Mol. Biol. 341, 839–852.

Uncorrected proofs — not for distribution

132  | Gueiros-Filho

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, 71–86. Lucet, I., Feucht, A., Yudkin, M.D., and Errington, J. (2000). Direct interaction between the cell division protein FtsZ and the cell differentiation protein SpoIIE. EMBO J. 19, 1467–1475. Ma, X., and Margolin, W. (1999). Genetic and functional analyses of the conserved C-terminal core domain of Escherichia coli FtsZ. J. Bacteriol. 181, 7531–7544. Margolin, W. (2001). Spatial regulation of cytokinesis in bacteria. Curr. Opin. Microbiol. 4, 647–652. 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, 3419–3430. Massidda, O., Anderluzzi, D., Friedli, L., and Feger, G. (1998). Unconventional organization of the division and cell wall gene cluster of Streptococcus pneumoniae. Microbiology 144, 3069–3078. Mercer, K.L., and Weiss, D.S. (2002). The Escherichia coli cell division protein FtsW is required to recruit its cognate transpeptidase, FtsI (PBP3), to the division site. J. Bacteriol. 184, 904–912. Migocki, M.D., Freeman, M.K., Wake, R.G., and Harry, E.J. (2002). The Min system is not required for precise placement of the midcell Z ring in Bacillus subtilis. EMBO Rep. 3, 1163–1167. Migocki, M.D., Lewis, P.J., Wake, R.G., and Harry, E.J. (2004). The midcell replication factory in Bacillus subtilis is highly mobile: implications for coordinating chromosome replication with other cell cycle events. Mol. Microbiol. 54, 452–463. Mingorance, J., Tamames, J., and Vicente, M. (2004). Genomic channeling in bacterial cell division. J. Mol. Recognit. 17, 481–487. Molle, V., Fujita, M., Jensen, S.T., Eichenberger, P., Gonzalez-Pastor, J.E., Liu, J.S., and Losick, R. (2003). The Spo0A regulon of Bacillus subtilis. Mol. Microbiol. 50, 1683–1701. Mukherjee, A., and Lutkenhaus, J. (1998). Dynamic assembly of FtsZ regulated by GTP hydrolysis. EMBO J. 17, 462–469. Mukherjee, A., and Lutkenhaus, J. (1999). Analysis of FtsZ assembly by light scattering and determination of the role of divalent metal cations. J. Bacteriol. 181, 823–832. Nanninga, N. (1998). Morphogenesis of Escherichia coli. Microbiol. Mol. Biol. Rev. 62, 110–129. Nukushina, J.I., and Ikeda, Y. (1969). Genetic analysis of the developmental processes during germination and outgrowth of Bacillus subtilis spores with temperature-sensitive mutants. Genetics 63, 63–74. Oliva, M.A., Cordell, S.C., and Lowe, J. (2004). Structural insights into FtsZ protofilament formation. Nat. Struct. Mol. Biol. 11, 1243–1250. Pedersen, L.B., Angert, E.R., and Setlow, P. (1999). Septal localization of penicillin-binding protein 1 in Bacillus subtilis. J. Bacteriol. 181, 3201–3211. Pichoff, S., and Lutkenhaus, J. (2002). Unique and overlapping roles for ZipA and FtsA in septal ring assembly in Escherichia coli. EMBO J. 21, 685–693. Pichoff, S., and Lutkenhaus, J. (2005). Tethering the Z ring to the membrane through a conserved membrane targeting sequence in FtsA. Mol. Microbiol. 55, 1722–1734. Pinho, M.G., and Errington, J. (2005). Recruitment of penicillin-binding protein PBP2 to the division site of Staphylococcus aureus is dependent on its transpeptidation substrates. Mol. Microbiol. 55, 799–807. Pogliano, J., Osborne, N., Sharp, M.D., Abanes-De Mello, A., Perez, A., Sun, Y.L., and Pogliano, K. (1999). A vital stain for studying membrane dynamics in bacteria: a novel mechanism controlling septation during Bacillus subtilis sporulation. Mol. Microbiol. 31, 1149–1159. Pogliano, J., Sharp, M.D., and Pogliano, K. (2002). Partitioning of chromosomal DNA during establishment of cellular asymmetry in Bacillus subtilis. J. Bacteriol. 184, 1743–1749.

Uncorrected proofs — not for distribution

Cell Division |  133

Popham, D.L., and Young, K.D. (2003). Role of penicillin-binding proteins in bacterial cell morphogenesis. Curr. Opin. Microbiol. 6, 594–599. Rabinovitch, A., Zaritsky, A., and Feingold, M. (2003). DNA-membrane interactions can localize bacterial cell center. J. Theor. Biol. 225, 493–496. Raskin, D.M., and de Boer, P.A. (1999). MinDE-dependent pole-to-pole oscillation of division inhibitor MinC in Escherichia coli. J. Bacteriol. 181, 6419–6424. 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, 349–367. Recchia, G.D., Aroyo, M., Wolf, D., Blakely, G., and Sherratt, D.J. (1999). FtsK-dependent and -independent pathways of Xer site-specific recombination. EMBO J. 18, 5724– 5734. Reeve, J.N., Mendelson, N.H., Coyne, S.I., Hallock, L.L., and Cole, R.M. (1973). Minicells of Bacillus subtilis. J. Bacteriol. 114, 860–873. Rico, A.I., Garcia-Ovalle, M., Mingorance, J., and Vicente, M. (2004). Role of two essential domains of Escherichia coli FtsA in localization and progression of the division ring. Mol. Microbiol. 53, 1359–1371. 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, 663–674. 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, 125–154. Rothfield, L., Taghbalout, A., and Shih, Y.L. (2005). Spatial control of bacterial division-site placement. Nat. Rev. Microbiol. 3, 959–968. 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, 3344–3351. Scheffers, D.J., and Errington, J. (2004). PBP1 is a component of the Bacillus subtilis cell division machinery. J. Bacteriol. 186, 5153–5156. Scheffers, D.J., Jones, L.J., and Errington, J. (2004). Several distinct localization patterns for penicillin-binding proteins in Bacillus subtilis. Mol. Microbiol. 51, 749–764. Sciochetti, S.A., Piggot, P.J., and Blakely, G.W. (2001). Identification and characterization of the dif site from Bacillus subtilis. J. Bacteriol. 183, 1058–1068. Sharp, M.D., and Pogliano, K. (2003). The membrane domain of SpoIIIE is required for membrane fusion during Bacillus subtilis sporulation. J. Bacteriol. 185, 2005–2008. Siddiqui, R.A., Hoischen, C., Holst, O., Heinze, I., Schlott, B., Gumpert, J., Diekmann, S., Grosse, F., and Platzer, M. (2006). The analysis of cell division and cell wall synthesis genes reveals mutationally inactivated ftsQ and mraY in a protoplast-type L-form of Escherichia coli. FEMS Microbiol. Lett. 258, 305–311. Sievers, J., and Errington, J. (2000a). Analysis of the essential cell division gene ftsL of Bacillus subtilis by mutagenesis and heterologous complementation. J. Bacteriol. 182, 5572–5579. Sievers, J., and Errington, J. (2000b). The Bacillus subtilis cell division protein FtsL localizes to sites of septation and interacts with DivIC. Mol. Microbiol. 36, 846–855. Smith, T.J., Blackman, S.A., and Foster, S.J. (2000). Autolysins of Bacillus subtilis: multiple enzymes with multiple functions. Microbiology 146 (Pt 2), 249–262. Steiner, W., Liu, G., Donachie, W.D., and Kuempel, P. (1999). The cytoplasmic domain of FtsK protein is required for resolution of chromosome dimers. Mol. Microbiol. 31, 579–583. Stricker, J., Maddox, P., Salmon, E.D., and Erickson, H.P. (2002). Rapid assembly dynamics of the Escherichia coli FtsZ-ring demonstrated by fluorescence recovery after photobleaching. Proc. Natl. Acad. Sci. USA 99, 3171–3175. Sun, Q., and Margolin, W. (1998). FtsZ dynamics during the division cycle of live Escherichia coli cells. J. Bacteriol. 180, 2050–2056.

Uncorrected proofs — not for distribution

134  | Gueiros-Filho

Szeto, T.H., Rowland, S.L., Habrukowich, C.L., and King, G.F. (2003). The MinD membrane targeting sequence is a transplantable lipid-binding helix. J. Biol. Chem. 278, 40050–40056. Taschner, P.E., Huls, P.G., Pas, E., and Woldringh, C.L. (1988). Division behavior and shape changes in isogenic ftsZ, ftsQ, ftsA, pbpB, and ftsE cell division mutants of Escherichia coli during temperature shift experiments. J. Bacteriol. 170, 1533–1540. Thomaides, H.B., Freeman, M., El Karoui, M., and Errington, J. (2001). Division site selection protein DivIVA of Bacillus subtilis has a second distinct function in chromosome segregation during sporulation. Genes Dev. 15, 1662–1673. van den Ent, F., and Lowe, J. (2000). Crystal structure of the cell division protein FtsA from Thermotoga maritima. EMBO J. 19, 5300–5307. Vaughan, S., Wickstead, B., Gull, K., and Addinall, S.G. (2004). Molecular evolution of FtsZ protein sequences encoded within the genomes of archaea, bacteria, and eukaryota. J. Mol. Evol. 58, 19–29. Wang, L., and Lutkenhaus, J. (1998). FtsK is an essential cell division protein that is localized to the septum and induced as part of the SOS response. Mol. Microbiol. 29, 731–740. Wang, X., Huang, J., Mukherjee, A., Cao, C., and Lutkenhaus, J. (1997). Analysis of the interaction of FtsZ with itself, GTP, and FtsA. J. Bacteriol. 179, 5551–5559. Wang, X., and Lutkenhaus, J. (1993). The FtsZ protein of Bacillus subtilis is localized at the division site and has GTPase activity that is dependent upon FtsZ concentration. Mol. Microbiol. 9, 435–442. Ward, J.E., Jr., and Lutkenhaus, J. (1985). Overproduction of FtsZ induces minicell formation in E. coli. Cell 42, 941–949. Weart, R.B., and Levin, P.A. (2003). Growth rate-dependent regulation of medial FtsZ ring formation. J. Bacteriol. 185, 2826–2834. Weart, R.B., Nakano, S., Lane, B.E., Zuber, P., and Levin, P.A. (2005). The ClpX chaperone modulates assembly of the tubulin-like protein FtsZ. Mol. Microbiol. 57, 238–249. Wei, Y., Havasy, T., McPherson, D.C., and Popham, D.L. (2003). Rod shape determination by the Bacillus subtilis class B penicillin-binding proteins encoded by pbpA and pbpH. J. Bacteriol. 185, 4717–4726. 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, 508–520. 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, 309–320. Wu, L.J., and Errington, J. (2004). Coordination of cell division and chromosome segregation by a nucleoid occlusion protein in Bacillus subtilis. Cell 117, 915–925. Yamamoto, H., Kurosawa, S., and Sekiguchi, J. (2003). Localization of the vegetative cell wall hydrolases LytC, LytE, and LytF on the Bacillus subtilis cell surface and stability of these enzymes to cell wall-bound or extracellular proteases. J. Bacteriol. 185, 6666–6677. Yanouri, A., Daniel, R.A., Errington, J., and Buchanan, C.E. (1993). Cloning and sequencing of the cell division gene pbpB, which encodes penicillin-binding protein 2B in Bacillus subtilis. J. Bacteriol. 175, 7604–7616. Zhou, P., and Helmstetter, C.E. (1994). Relationship between ftsZ gene expression and chromosome replication in Escherichia coli. J. Bacteriol. 176, 6100–6106.

Uncorrected proofs — not for distribution