SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results and discussion
  5. Experimental procedures
  6. Acknowledgements
  7. References

The process of bacterial cell division involves the assembly of a complex of proteins at the site of septation that probably provides both the structural and the cytokinetic functions required for elaboration and closure of the septal annulus. During sporulation in Bacillus subtilis, this complex of proteins is modified by the inclusion of a sporulation-specific protein, SpoIIE, which plays a direct role in gene regulation and also has a genetically separable role in determining the gross structural properties of the specialized sporulation septum. We demonstrate by both green fluorescent protein (GFP) fusions and indirect immunofluorescence microscopy that SpoIIGA, a protein required for proteolytic cleavage of pro-σE, is also targeted to the sporulation septum. Septal localization of SpoIIGA–GFP occurred even in the structurally abnormal septum formed by a SpoIIE null mutant. We also report the isolation of a spoIIGA homologue from Bacillus megaterium, a species in which the cells are significantly larger than those of B. subtilis. We have exploited the physical dimensions of the B. megaterium sporangium, in conjunction with wide-field deconvolution microscopy, to construct three-dimensional projections of sporulating cells. These projections indicate that SpoIIGA–GFP is initially localized in an annulus at the septal periphery and is only later localized uniformly throughout the septa. Localization was also detected in a B. subtilis spo0H null strain that fails to construct a spore septum. We propose that SpoIIGA is sequestered in the septum by an interaction with components of the septation machinery and that this interaction begins before the construction of the asymmetric septum.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results and discussion
  5. Experimental procedures
  6. Acknowledgements
  7. References

A crucial step in endospore formation in Bacillus species is the construction of a sporulation septum at an asymmetric position proximal to a pole of the cell body. This polar septation event effectively partitions the sporangium into two unequally sized compartments with distinct developmental fates; the smaller prespore compartment develops into a mature spore, while the mother cell eventually lyses and releases the spore. These morphological changes are co-ordinated with changes in gene expression directed by RNA polymerase sigma factors that are activated in a compartment-specific manner (for reviews, see Errington, 1993; Stragier and Losick, 1996). Inactive forms of both σF, the first prespore-specific sigma factor, and σE, the first mother cell-specific sigma factor, are present in the cell before the synthesis of the polar septum (Gholamhoseinian and Piggot, 1989; Lewis et al., 1996). The mechanisms by which they become active differ but, in each case, compartment-specific activation requires a function attributable to sporulation proteins targeted to the asymmetric septum. The activation of σF requires a functional SpoIIE protein (Margolis et al., 1991), the septal localization of which has been previously reported (Arigoni et al., 1995; Barák et al., 1996; Levin et al., 1997), whereas the activation of σE requires SpoIIGA (Jonas et al., 1988; Peters and Haldenwang, 1994), the septal localization of which we now report.

The formation of a cell division septum in eubacteria is believed to involve the concerted action of a complex assemblage of proteins that has been termed the ‘septator’ or ‘septasome’ (Vincente and Errington, 1996). The most extensively characterized septasome component is the cell division protein FtsZ. FtsZ is widely conserved throughout the bacterial kingdom and appears to be essential for viability in most species, including B. subtilis (Beall et al., 1988; Dai and Lutkenhaus, 1991). FtsZ undergoes in vitro self-assembly into structural homologues of tubulin polymers and has slight sequence homologies to tubulin (Bramhill and Thompson, 1994). During the Escherichia coli division cycle, cytoplasmic FtsZ monomers aggregate in a circumferential ring at future division sites. This ring is progressively constricted and remains at the leading edge of the invagination of membrane and cell wall during cytokinesis, leading Bi and Lutkenhaus (1991) to propose that FtsZ performs a cytoskeletal function. The finding that FtsZ interacts directly with other important division proteins, such as FtsA (Ma et al., 1996; Wang et al., 1997) and ZipA (Hale and de Boer, 1997), and appears to co-localize with others, such as FtsI (J. Pogliano et al., 1997) and DivIB (Harry and Wake, 1997), has led to the suggestion that FtsZ may also act as a scaffold upon which the multiprotein septasome complex is built (Addinall et al., 1996).

The sporulation septum of endospore formers differs from the vegetative septum both in its site of assembly and in its physical characteristics. Like the vegetative division septum, the sporulation septum is initially filled with cell wall material. However, the quantity of murein deposited between the lipid bilayers of the sporulation septum is much reduced, resulting in a visibly thinner structure. In the normal course of sporulation, all detectable murein is removed from the polar septum to produce a pliable double membrane (Holt et al., 1975). This structure does not invaginate to form separate division products, but rather migrates toward the spore pole, eventually completely engulfing the forespore compartment. Despite these differences, it is clear that many of the basic components of the assembly machinery are the same in both septum types. Thus, it has been proposed that the sporulation septum should properly be viewed as a modified, specialized version of the cell division septum (Hitchins and Slepecky, 1969). How is the septasome scaffold modified during sporulation, and how is the special septal structure that forms during sporulation diverted to participate directly in establishing compartment-specific programmes of gene expression? The 92 kDa SpoIIE protein probably plays a central role. Biochemically, SpoIIE is a serine protein phosphatase and, in this capacity, it is directly responsible for modifying the interaction between SpoIIAA and SpoIIAB, which controls the activation of σF in the forespore (Duncan and Losick, 1993; Diederich et al., 1994; Duncan et al., 1995; 1996; Arigoni et al., 1996; Feucht et al., 1996). SpoIIE is localized at sites of septation (Arigoni et al., 1995; Barák et al., 1996) and is also required for the construction of morphologically normal polar septa; spoIIE null strains produce septa exhibiting a ‘straight and thick’ phenotype typically associated with murein-rich vegetative septa (Piggot, 1973; Illing and Errington, 1991). In addition, SpoIIE is required for the efficient initiation of polar septation, and this role in septum initiation and phenotypic determination is distinct and separable from its role in σF activation (Barák and Youngman, 1996; Feucht et al., 1996). Because mutations at no other genetic locus result in the thick polar septum phenotype, it has been suggested that SpoIIE could play a critical role in modifying the nature of the septum-synthesizing scaffold during sporulation (Barák et al., 1996). For example, it might be supposed that spoIIE null mutants would fail to recruit other sporulation-specific components of the septasome.

Mother cell gene expression is initially directed by the modified gene product of spoIIGB. The spoIIGB locus encodes a sigma factor precursor, pro-σE, that is activated by the proteolytic cleavage of 27 amino-terminal residues (Stragier et al., 1984; Trempy et al., 1985; LaBell et al., 1987). Genetic evidence indicates that the activating protease is encoded by spoIIGA, a gene located upstream in the same operon (Kenney and Moran, 1987; Stragier et al., 1988; Peters and Haldenwang, 1994). SpoIIGA has slight homology to a family of aspartic proteases and is predicted to be an integral membrane protein (Stragier et al., 1988; Peters and Haldenwang, 1991). The timing of SpoIIGA-mediated pro-σE cleavage is controlled by an intercellular signalling protein, SpoIIR, which is produced in the forespore under σF control and is secreted across the septum (Hofmeister et al., 1995; Karow et al., 1995; Londoño-Vallejo and Stragier, 1995). Topological analysis of SpoIIGA based solely on the B. subtilis sequence suggests a model with five amino-terminal membrane-spanning segments and a cytoplasmic carboxyl-terminal globular domain (Stragier et al., 1988). These observations make it attractive to suppose, as proposed by Stragier, that SpoIIGA is targeted to the sporulation septum and that obligatory targeting to the septum serves as a mechanism to co-ordinate morphogenesis with key changes in gene expression. To determine the subcellular localization of SpoIIGA, we have used fusions of SpoIIGA to an enhanced intensity variant of the green fluorescent protein (GFP) of Aequorea victoria, and here provide evidence that SpoIIGA from both B. subtilis and B. megaterium is targeted specifically to the sporulation septum. Moreover, we show that this localization is not dependent on the presence of functional SpoIIE. Previous studies have been unable to determine if localization of a protein at the sites of septation implies localization throughout the septum or only to an annulus at the periphery of the cell. To resolve this issue, we used immunofluorescent microscopy (IFM) in conjunction with wide-field deconvolution microscopy to create three-dimensional projections of sporulating B. megaterium cells. These computerized reconstructions can be rotated arbitrarily, allowing us to examine projections down the long axis of cells. Using this approach, we find that both SpoIIGA and SpoIIE are initially concentrated in an annulus corresponding to a site of septum initiation, but that both eventually spread throughout the septum. This is consistent with our observation that a low level of septal SpoIIGA localization can be observed in spo0H mutant strains in which sporulation is blocked before the construction of a polar septum.

Results and discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results and discussion
  5. Experimental procedures
  6. Acknowledgements
  7. References

Localization of SpoIIGA–GFP in B. subtilis

To determine the subcellular localization of SpoIIGA, we constructed a temperature-sensitive integrational plasmid, pIIGA-GFP, which mediates chromosomal SpoIIGA–GFP translational fusions. Integrations of this plasmid result in a spoIIGA–GFP translational fusion under the control of the natural spoIIGA promoter, accompanied by a separation of the downstream spoIIGB gene from its promoter. Chromosomal integrations were made initially in a wild-type B. subtilis background, which established that SpoIIGA is targeted to sites of polar septum synthesis (strain PMF 9, Fig. 1A). When grown in liquid sporulation medium at 30°C, cells began to acquire a diffuse green fluorescence beginning approximately 1 h after the cessation of logarithmic growth (T1). The fluorescent signal slowly increased and, by T2, many cells exhibited localization at one or both cell poles (Table 1). This localization presumably corresponds to sites of septum synthesis and is stable for at least 8 h. Because the integrative plasmid mediating the SpoIIGA–GFP fusion separates spoIIGB from the spoIIG promoter, PMF 9 was abortively disporic. Apart from the 53% of PMF 9 cells that had localized signal at one or both cell poles, 29% of cells never developed more than a diffuse glow, while the remaining 18% developed a signal that was not associated with the cell poles (Table 1). This latter type of signal usually appeared as disorganized granular patches; examination of such cells by phase-contrast microscopy often revealed damage or lysis. We attribute the relatively low percentage of cells that developed a localized signal to the weakness of the GFP signal, reflecting the relative weakness of the spoIIG promoter. A slight alteration in the plane of focus often made a localized signal appear at one pole of the cell, while causing the signal at the other pole to diminish or disappear. Use of a 100 × objective lens in place of a 60 × lens also caused a decrease in apparent localized signal because of the reduced light-gathering ability of the instrument. Care was therefore taken to maximize the strength of GFP fluorescence (see Experimental procedures).

image

Figure 1. . Fluorescent and immunofluorescent microscopy. A–C. Localization of SpoIIGA–GFP-associated fluorescence at sites of polar septation in B. subtilis from culture samples harvested 2.5 h after cessation of logarithmic growth (T2.5) The scale bar in (A) represents 3 μm and applies to A–C. A. PMF 9 (spoIIGA–GFP ), wild-type background. B. PMF 10 (spoIIGA–GFP, spoIIE64 ), a spoIIE missense mutant background. C. PMF 14 (spoIIGA–GFP spoIIE ::Tn917ΩHU7 ), a spoIIE null background. D–F. Cells of B. megaterium PMF 50 (spoIIGA–GFP ) from T2 and stained for IFM using an anti-GFP primary antibody and an Oregon Green 488-conjugated secondary antibody. The scale bar in (D) represents 3 μm and applies to panels D–H. D. Both FITC and DAPI wavelengths shown. E. FITC wavelength only. F. DAPI wavelength only (nucleoids have been artificially coloured red). G. Computer-reconstructed projection down the long axis of a B. megaterium PMF 50 cell at T1.5, showing the early annular pattern of SpoIIGA localization. H. Projection down the long axis of a PMF 50 cell from T5, showing disc-like localization. I. B. megaterium PY1272 cells (spoIIE–GFP ) stained for IFM as described for D–F. In (I), the lefthand cell has the straight septum typical of early points, while the righthand cell shows the SpoIIE cage, which forms around the forespore nucleoid. The scale bar in (I) is 3 μm and refers to I–K. J. SpoIIE cages forming around the prespore of PY1272 cells as shown by GFP fluorescence alone. K. A composite micrograph of B. subtilis PMF 37 (spoIIGA–GFP, spo0H ::kan) cells. In (K), the top right cell shows normal localization (11% of cells), while the two bottom cells show the aberrant semi-circular pattern of localization seen only in this strain (approximately 5% of cells).

Download figure to PowerPoint

Table 1. . B. subtilis SpoIIGA–GFP localization patterns.a,b a. Cells were harvested 2.5 h after the cessation of logarithmic growth.b. All scoring was performed using cells from several independent experiments using a 60 × objective in conjunction with a 3.33 × side tube magnifier.c. Class A refers to cells with a disporic signal; class B refers to cells with a monosporic signal; class C refers to cells with unlocalized signal or no detectable signal; class D refers to cells with aberrant signal (see text). As signal visibility is dependent on correct focus, significantly out-of-focus cells were not scored.Thumbnail image of

To test the hypothesis that SpoIIGA localization might require a functional SpoIIE protein, we constructed spoIIGA–GFP fusions in spoIIE64 (PMF 10, Fig. 1B) and spoIIE ::Tn917ΩHU7 (PMF 14, Fig. 1C) genetic backgrounds. As a consequence of a missense mutation leading to a defect in phosphatase activity, spoIIE64 strains are defective in the expression of σF-dependent genes, but nevertheless produce polar septa of normal appearance (Barák and Youngman, 1996). The spoIIE ::Tn917ΩHU7 mutation is caused by a transposon insertion in spoIIE near the beginning of its coding sequence, and this produces a true null phenotype characterized by thick septa and a decreased frequency of septum initiation (Sandman et al., 1987; Barák and Youngman, 1996). Polar localization of SpoIIGA–GFP fusion proteins occurred in all three genetic backgrounds (Fig. 1A–C). We therefore conclude that SpoIIE plays neither a direct nor an indirect role in targeting SpoIIGA to the sporulation septum. An interesting possibility was raised by the extent of localization observed for PMF 14, the spoIIE null strain. While the total percentage of cells showing localization (classes A and B in Table 1) fell from 53% in PMF 9 to only 34% in PMF 14, the 34% of cells with localized signal in PMF 14 was greater than expected. While spoIIE null strains incubated for extended periods of time eventually septate at a frequency approaching that of other stage II mutants (I. Barák, personal communication), there is clearly a significant delay in the onset of septation. This delay was documented in the study of Feucht et al. (1996), which used phase-contrast microscopy to follow the kinetics of septum formation in a spoIIE ::ermC null mutant strain. This study established that only approximately 5% of spoIIE null mutant cells exhibited a disporic phenotype at T2.5. A similar delay was also reported by Barák and Youngman (1996), who used the direct enumeration of septation frequencies from electron micrographs to show that only approximately 5% of spoIIE ::Tn917ΩHU7 cells harvested at T10 exhibited class A localization. In contrast, we observed class A SpoIIGA localization in 21% of PMF 14 cells harvested at T2.5. We interpret this as consistent with the possibility that SpoIIGA–GFP is initially targeted to a preseptational component of the septasome rather than to the completed septum structure. We speculate that SpoIIGA may be interacting with a protein or complex associated with the septasome constructed shortly after the shift of the FtsZ ring from the mid-cell site to the site of asymmetrical division. The persistence of septal murein in spoIIE strains clearly argues against models requiring SpoIIGA recognition of the thin double-membrane configuration that results from the removal of murein. The septal localization of SpoIIGA in spoIIE backgrounds also implies that localization does not require σF- or σE-dependent transcription. While complementation studies have shown that the SpoIIGA–GFP fusion protein is inactive (H. Peters, personal communication), we do not believe that this influences localization; it seems improbable that the disruption of normal SpoIIGA targeting would be accompanied by a gain of function resulting in the precise localization of SpoIIGA–GFP at the highly specialized sporulation septum. SpoIIGA has a predicted membrane-spanning topology, and pro-σE, its proteolysis substrate, has been previously shown to localize at the sporulation septum (Stragier et al., 1988; Ju et al., 1997). It is therefore consistent to postulate that native SpoIIGA is also targeted to the polar septum.

Localization of SpoIIGA in a B. subtilis spo0H mutant

To clarify whether SpoIIGA localization occurs in predivisional sporangia, we constructed a SpoIIGA–GFP fusion strain in an a spo0H genetic background. The spo0H locus encodes a sigma factor, σH, that is most strongly expressed during late log phase growth and is involved in the transition to stationary phase (Weir et al., 1991). Upon entry to sporulation, spo0H null mutants undergo axial filament elongation, but are blocked before the construction of the sporulation septum. Our choice of a spo0H mutant to abolish septation during this experiment was influenced by the observation that the position of the FtsZ ring is correctly shifted from medial to polar positions in such mutants (Levin and Losick, 1996). The FtsZ ring is known to be essential for the localization of SpoIIE (Levin et al., 1997), and we considered it likely that predivisional SpoIIGA localization would also require this fundamental septasome component. This is in contrast to spo0A mutants, which fail to shift the FtsZ ring to the poles (Levin and Losick, 1996). The difficulty with this approach is that expression from the spoIIG promoter is significantly diminished in spo0H strains (Kenney and Moran, 1987). Localization of SpoIIGA–GFP was difficult to detect in the wild-type background and could not be detected in a spo0H mutant. To overcome this problem, it was necessary to construct an integrative spoIIGA–GFP plasmid containing the entire coding region of spoIIGA, as well as an upstream region required for maximal expression from this promoter. By training strains with chromosomal integrants of this plasmid on successively higher levels of chloramphenicol (up to a final concentration of 80 μg ml−1), we created a strain containing multiple tandem duplications of the spoIIGA–GFP translational fusion in a spo0H ::kan background (Jannière et al., 1985; Piggot and Curtis, 1987). These tandem repeats apparently partially compensated for reduced expression from the spoIIG promoter, allowing polar localization of SpoIIGA–GFP to be observed in this strain (Fig. 1K, cell at top righthand corner), albeit at a reduced frequency. We found that 11% (total n = 234) of the slightly elongated spo0H filaments contained one or more bands appearing at presumptive polar septation sites. We also observed that 21% of the filaments contained disorganized patterns of localization or normal localization together with disorganized patterns. As with the wild-type SpoIIGA–GFP strain, many of the cells with disorganized localization appeared to be damaged or lysed when examined by phase-contrast microscopy. However, we reproducibly observed one unusual pattern not seen in any of the other strains in this study: approximately 5% of the cells showed a semicircular pattern of localization (bottom two cells in Fig. 1K). We speculate that this structure might correspond to localization of SpoIIGA to incomplete or deformed septasomes that have formed at an oblique angle relative to the long axis of the cell. The supplementary ftsAZ promoter P2 is σH dependent, and the induction of this promoter at T0 may be required for the efficient production of normal septation sites (Gholamhoseinian et al., 1992; Gonzy-Tréboul et al., 1992). This localization pattern may also be a consequence of reduced levels of Spo0A or some other σH-dependent factor. This phenotype is in some ways reminiscent of the occasional oblique septation event that has been reported during minicell formation in divIVA mutants (Edwards and Errington, 1997).

Conserved features of SpoIIGA from B. subtilis, B. megaterium, Bacillus thuringiensis and Clostridium acetobutylicum

To achieve enhanced resolution of SpoIIGA localization, we characterized the SpoIIGA homologue from B. megaterium, a species in which the cell diameter is approximately two times greater than in B. subtilis cells. This homologue was successfully isolated using a degenerate primer/inverse polymerase chain reaction (PCR)-based approach and sequenced (GenBank accession number AF0171810; see Experimental procedures). This approach assumed a preservation of the gene order found in B. subtilis, and we indeed found that the organization of the B. megaterium genome near the spoIIG locus is similar to that of B. subtilis (gene order: bpr — spoIIGA — spoIIGB ). A search of the GenBank database revealed that in addition to the spoIIGA homologue of B. subtilis (Stragier et al., 1988; Masuda et al., 1990), those of C. acetobutylicum (Sauer et al., 1994) and B. thuringiensis (Adams et al., 1991) have also been cloned and sequenced (although only the carboxyl-terminus of the B. thuringiensis sequence is available). These sequences were aligned using the PILEUP module of the GCG suite (Devereux et al., 1984) to determine which regions of this protein are conserved and are therefore most likely to be of biological relevance (Fig. 2C). We found that the B. megaterium homologue encodes a 307 amino acid (aa) open reading frame (versus 309 aa for B. subtilis), corresponding to a predicted 35 kDa peptide with significant similarities to the B. subtilis protein (49% identity). This compares with a 42% identity between the B. subtilis and B. thuringiensis proteins (calculated over the available 162 aa sequence of the B. thuringiensis carboxyl-terminus) and 19% identity between the B. subtilis and C. acetobutylicum proteins. The predicted topological features of the three proteins are similar (Fig. 2A–C) and are consistent with the model proposed by Stragier et al. (1988).

image

Figure 2. . Topological model of the B. megaterium SpoIIGA protein and alignment of the peptide sequences deduced by translation of spoIIGA homologues from B. megaterium (GenBank accession number AF0171810; this study), B. subtilis (X17344; Masuda et al., 1990), B. thuringiensis (X56697; Adams et al., 1991) and C. acetobutylicum (Z23079; Sauer et al., 1994). A. Hydrophobicity profile of the 307-residue B. megaterium SpoIIGA peptide, based on the GES scale (Engelman et al., 1986). A sequence file translated using MACVECTOR (International Biotechnologies) was subjected to a GES hydropathy index calculation using TOPPREDII (Claros and von Heijne, 1994) with an optimal window size of 21 and a core window size of 11. B. Hypothetical membrane-spanning topology of the B. megaterium SpoIIGA protein. K + R-values represent the number of lysine and arginine residues in the region indicated. Numbers at the top and bottom of each loop indicate the first or last residue predicted to be within the membrane-spanning domain. C. Complete or partial translated sequences of homologues were aligned using the GCGPILEUP module (Devereux et al., 1984). The resulting multiple sequence file was output to BOXSHADE. TRANSVERTER PRO (TechTool Software) was used to convert Postscript format BOXSHADE output to ADOBEILLUSTRATOR format for final editing. In the edited version: black box, identical residues; dark grey box, conserved residues; light grey box, residues similar to conserved residues; white box, unconserved residue. ·, Gaps introduced by PILEUP; –, sequence undetermined; *stop codon; 1, a G100R change defines the SpospoIIGA49 allele (Stragier et al., 1988); 2, a ΔQ110-S132 variant is Spo+ (Londoño-Vallejo, 1997); 3, the conserved DS(T)GN aspartic protease motif (Pearl and Taylor, 1987); 4, a P259L change suppresses processing-negative mutations in pro-σE (Peters and Haldenwang, 1994); 5, G279 is predicted to be required for protease activity (Pearl and Taylor, 1987). The roman numerals and bars above and below the alignment define predicted membrane-spanning domains: black bars, B. megaterium; dark grey bars, B. subtilis; light grey bars, C. acetobutylicum.

Download figure to PowerPoint

The sequence alignment also revealed the conservation of several critical bases. The B. subtilis spoIIG49 mutation, which causes a 104-fold reduction in sporulation efficiency, maps to residue 100 of the B. subtilis sequence (Stragier et al., 1988). This residue (indicated by ‘1’ in Fig. 2C) is a perfectly conserved glycine in all of the organisms for which we have sequence and is predicted to occur within a transmembrane domain in all three organisms. Another perfectly conserved residue is the proline (indicated by ‘4’ in Fig. 2C) at position 259 of the B. subtilis sequence; strains with a Pro-to-Leu change at this position have been isolated as suppressors of a processing-negative allele of pro-σE, suggesting that this amino acid contributes to the specificity of the processing reaction (Peters and Haldenwang, 1994). We also note the universal conservation, beginning at residue 183 of the B. subtilis sequence (indicated by ‘3’ in Fig. 2C), of the Asp-Thr(Ser)-Gly core sequence characteristic of aspartic/retroviral proteases, as first noted by Stragier (Pearl and Taylor, 1987; Stragier et al., 1988). This motif occurs in the region of the alignment with the greatest overall conservation, and the only variation in this sequence is the permitted Ser-to-Thr substitution in the second position of the C. acetobutylicum motif. Aspartic proteases also require a glycine preceded by two hydrophobic residues positioned approximately 100 bases downstream of the Asp-Thr(Ser)-Gly motif (Pearl and Taylor, 1987). Occurring at residue 279 of the B. subtilis sequence (indicated by ‘5’ in Fig. 2C), this glycine is conserved, except in C. acetobutylicum, which has a Gly-to-Ala substitution. In each case, the two preceding residues are appropriately hydrophobic. All topological predictions indicate an extracellular location for the first six residues of the amino-terminus, as well as for the large loop between the fourth and fifth membrane-spanning domains. These regions have therefore been considered to be possible interaction sites required for the transduction of the activating signal to the proteolytic domain. The alignment in 2Fig. 2C shows that, while the six exterior-exposed residues of the amino-terminus are highly conserved (with the exception of position 2), the large extracellular loop between the fourth and fifth membrane-spanning domains is not. This is consistent with a recent report indicating that the entire poorly conserved loop (indicated by a ‘2’ in Fig. 2C) is dispensable for SpoIIGA function (Londoño-Vallejo, 1997). This report has also shown that a Tyr-to-Ala substitution at residue four or an Val-to-Ala substitution at residue seven leads to no decrease in sporulation efficiency. However, an Asp-to-Ala substitution at residue six causes a 2–3 order of magnitude inhibition of sporulation in B. subtilis. The effect of an alteration to the conserved isoleucine at residue three or the conserved leucine at residue five has not been evaluated (Londoño-Vallejo, 1997). The alignment provided here should prove useful for identifying conserved residues to target in future directed-mutagenesis experiments.

Localization of B. megaterium SpoIIGA–GFP by deconvolution microscopy

The large size of the B. megaterium cell makes this species an attractive model system for protein localization studies. In particular, we hoped to exploit the size of this organism to address the issue of whether SpoIIGA is targeted to the entire septum or only to a peripheral annulus. We were able to approach this problem by using wide-field deconvolution microscopy, a technique that requires taking multiple optical sections through a cell using standard epifluorescence optics, then using computational methods to remove out-of-focus and artefactual information from each image (Agard and Sedat, 1983; Agard, 1984). Optimized data sets may then be used to create projections of the sample from any perspective. A sequence of many such projections is used to generate a movie simulating rotation of the sample around an arbitrary axis. With this in mind, we constructed a spoIIGA–GFP translational fusion in B. megaterium analogous to our B. subtilis construction PMF 9. This construction used the intensity-enhanced F64L, S65T variant of GFP (Cormack et al., 1996). PMF 50 yielded good results in conjunction with standard microscopic techniques and showed septal localization of SpoIIGA–GFP in > 85% of cells. However, the 60 or more long exposures required for collecting a data set suitable for deconvolution microscopy required an alternative to GFP providing a more favourable signal-to-noise ratio and less photobleaching. It has been determined recently that the IFM technique is sensitive enough to localize proteins with as few as 100 molecules per cell (Weiss et al., 1997). We therefore used a modification of the IFM technique of Pogliano and Harry (Pogliano et al., 1995) that employed a primary antibody raised against GFP and a secondary antibody conjugated to the fluorophore Oregon Green 488. When used in conjunction with an antifade reagent, we were able to obtain fluorescent signal of an intensity and stability sufficient for the collection of deconvolution data sets. Figure 1(D–F) shows that the B. megaterium homologue of SpoIIGA is also localized to the sporulation septum. However, the level of detail that we obtained with this technique is best appreciated in three-dimensional projections; QuickTime format movies of these and other images can accessed on the Molecular Microbiology Web site (http://www.blackwell-science.com. products/journals/mole.htm). One insight provided by deconvolution microscopy is that the initially diffuse fluorescent signal detected at the onset of SpoIIGA–GFP expression is preferentially membrane associated. It is therefore likely that SpoIIGA targeting involves an initial non-specific interaction or insertion of the hydrophobic domain of SpoIIGA into any membrane it contacts. In this model, newly synthesized SpoIIGA rapidly becomes associated with membrane, then diffuses laterally until contacting some other septum-specific protein or component of the septasome. The interaction of SpoIIGA with this unknown factor would sequester or anchor SpoIIGA into the nascent septal structure. While passive, this mechanism for SpoIIGA targeting could, nevertheless, result in a rapid preferential accumulation of SpoIIGA in the septum. This general model is supported by 1Fig. 1G, which is a computer-reconstructed projection down the long axis of a fluorescently stained PMF 50 cell. At the earliest time points after localized signal began to develop, such projections revealed that the localization of signal throughout the septum was often incomplete (60% of cells), with the most intense signal being observed in an annulus at the cell periphery and with the centre of the cell exhibiting very little or no signal above background levels (Table 2, and Fig. 1G[link]). This pattern also occurs in cells that appear to have completed translocation of DNA into the prespores (approximately 40% of the subclass of cells with an annular pattern) and have therefore presumably finished constructing a polar septum. The annular pattern of localization is not an artefact of fixation; PMF 50 cells grown until T5 have invariably progressed to a disc-like pattern of localization throughout the septum (Fig. 1H). The disc-like pattern was distinguished by the significant distribution of signal throughout the septum, with the region of highest intensity staining being located in the centre of the septum.

Table 2. . Initial appearance of septum in projections down the long axis of PMF 50 cells processed for IMF.a,b a. This table reflects the appearance of the septa approximately 1.5 h after the cessation of logarithmic growth and indicates whether or not translocation of DNA into the prespore appeared to be taking place. By T5, the septa become uniformly disc-like, and translocating chromosomes are very rare.b. See Experimental procedures for details.Thumbnail image of

Localization of B. megaterium SpoIIE–GFP

We have also observed an initial annular localization pattern in cells of immunofluorescently stained spoIIE–GFP B. megaterium strain PY1272 (data not shown). However, the progression of localization after the annular stage differs from that seen in PMF 50. Unlike PMF 50 cells, PY1272 cells sporulate at normal frequency and consequently pass through an engulfment stage. As shown in the leftmost cell visible in 1Fig. 1I, SpoIIE localization at early times appears as a band across the cell at the site of septum formation (Barák et al., 1996). At later times, both immunofluorescence and GFP micrographs (rightmost cell of Fig. 1I and J respectively) show that the engulfment of PY1272 prespores leads to the formation of a cage surrounding the prespore (QuickTime movies illustrating the three-dimensional nature of the cage are available at the Web site). We also observed the formation of cages in a B. subtilis spoIIE–GFP strain constructed in this laboratory (data not shown) and found that cages in both of these strains are stable for several hours. It should be noted that this observation of SpoIIE stability contradicts results from other groups; it has been reported previously that SpoIIE is quickly removed from the sporangium, first from the prespore distal division site, followed shortly thereafter by its removal from the prespore pole (Arigoni et al., 1995; K. Pogliano et al., 1997). While it is possible that the GFP tag may confer resistance to degradation, the normal sporulation of this strain implies that the disappearance of SpoIIE from the forespore is not required for the establishment of cell-specific gene transcription.

Implications for models of protein targeting

The septal localization of SpoIIGA in B. subtilis spoIIE strains with ‘straight and thick’ septa argues against any model proposing that initial targeting is delayed until the stage of murein removal. It therefore seems unlikely that the targeting of proteins to the polar septum depends upon recognition of the characteristic double-membrane topology of normal sporulation septa. As the GFP fusion constructs effectively separate spoIIGB from the spoIIG promoter, we also conclude that SpoIIGA localization does not depend on the prior localization of pro-σE, which is itself localized at the spore septum until it is processed (Ju et al., 1997). The localization of SpoIIGA in spoIIE and spo0H backgrounds also indicates that localization does not depend on any σF- or σE-dependent transcription. We have not attempted to determine the localization of SpoIIGA in the context of spo0A mutations, and it will be interesting to determine if targeting takes place in cells that fail to reposition the FtsZ ring. The observation in B. megaterium of an initial annular pattern of SpoIIGA and SpoIIE localization appears to be consistent with the observation of Levin et al. (1997) that SpoIIE is co-localized with FtsZ rings in a B. subtilis divIC strain grown at temperatures non-permissive for septum formation. The next obvious goal is to determine whether SpoIIGA recognizes some mark at future sites of septation, or if SpoIIGA localization depends on some protein (such as FtsZ, FtsA or a possible ZipA homologue) known or suspected to be a component of the septasome.

It is interesting to speculate on the reasons why an annular pattern of SpoIIGA localization is sometimes observed in B. megaterium cells in which DNA translocation has apparently been completed. It is possible that the co-ordination of septum construction and DNA translocation is different in this species than in B. subtilis, and DNA translocation may be essentially completed before the closure of the sporulation septum. However, there are other possibilities. Although it seems unlikely, exclusion of SpoIIGA from the centre of the annulus might be caused by the steric hindrance of SpoIIGA by another protein complex. SpoIIIE, which acts as a DNA translocating pore in B. subtilis (Wu and Errington, 1994; Wu et al., 1995), is a possible candidate. It is also possible that the transition from annular to disc-like localization is triggered by the presence of a later appearing protein. For instance, the transition may require the production and membrane translocation of SpoIIR. We have also considered that, while localization to the annulus does not depend on the removal of murein from the septum, the transition to the disc-like pattern of localization may not occur until murein degradation has taken place. All of our SpoIIGA–GFP fusion strains are abortively disporic, but we do not believe that this influences localization. In B. subtilis, septal SpoIIGA–GFP localization is first observed at times corresponding to the initiation of SpoIIGA synthesis, and well before any unusual accumulation of SpoIIGA could occur. Although the timing of spoIIG expression in B. megaterium has not been described, the appearance of localized protein occurs at the same time as in B. subtilis. Moreover, a significant proportion of septa examined at the earliest times at which localized signal can be observed have already transitioned from the annulus to the disc, suggesting that the kinetics of this transition are rapid relative to any possible abnormal SpoIIGA–GFP accumulation.

It is important to note that our results also suggest that there may be more than one mechanism of targeting. When PMF 9 or PMF 50 cells were grown for > 5 h after the initial development of fluorescent signal, the GFP signal remained restricted to the septum. This is in marked contrast to the localization pattern of SpoIIQ, a σF-dependent prespore protein (Londoño-Vallejo et al., 1997). Initially targeted to the polar septum, SpoIIQ spreads progressively around the prespore as engulfment occurs. However, this progression of SpoIIQ distribution occurs even in a spoIIGB mutant blocked at stage IIi of sporulation, suggesting that SpoIIQ (at least after its initial sequestration at the septum) is targeted differently from SpoIIGA (Londoño-Vallejo et al., 1997). Determining the nature of the interactions required for protein targeting and understanding how these proteins are integrated into the septasome remains a long-term goal.

Experimental procedures

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results and discussion
  5. Experimental procedures
  6. Acknowledgements
  7. References

Bacterial strains, culture media, genetic techniques and in vitro manipulation of DNA

B. subtilis and B. megaterium strains used in this work are listed in Table 3. E. coli strains MM294 (Bachmann, 1987) and DH5α-mcr (Gibco BRL) were used for routine genetic manipulations, and E. coli strain S17-I (ATCC 47055) was used as the donor strain for B. megaterium conjugations (Simon et al., 1983). Isolation of genomic, plasmid and phage DNA, methods for the transformation of B. subtilis and E. coli strains, selection of antibiotic resistance markers and other standard genetic manipulations were carried out as described previously (Cutting and Youngman, 1994; Provence and Curtiss, 1994). Enzymatic reagents for DNA modification, including restriction enzymes, DNA ligase and Taq DNA polymerase were purchased from Boehringer Mannheim or New England Biolabs and were used as specified by the manufacturers or according to standard protocols (Ausubel et al., 1994).

Table 3. . Bacillus strains used in this study. a. Strains with a spoIIGA–GFP fusion are also spoIIGB, as the integrating plasmid disrupts the operon. (Youngman et al. (1984)).Thumbnail image of

DNA sequencing of the B. megaterium SpoIIGA homologue

The B. megaterium spoIIGA homologue was identified and sequenced using a PCR approach based on homology to conserved regions in previously identified spoIIGA homologues and the downstream sigma factor encoded by spoIIGB. Amplification from a QM B1551 chromosomal DNA template using degenerate primers (mrBmIIGA2R 5′-GGNAAYCARYTNTAYGAYCC-3′ and sigE2.2L 5′-ACNGCYTTDATNARNCCDAT-3′; DNagency) resulted in a single product of 1.4 kb that was sequenced using the fmol thermal cycling sequencing system (Promega) and determined to be a spoIIGA homologue. An inverse PCR approach was then used to recover the entire spoIIGA gene. Primers (bmIIGA67Ls: 5′-CCTTACCATATCTGGC-3′; and bmIIGB33Rs 5′-GCTGTTGATTGAGCG-3′) internal to the original PCR product were used to amplify an approximately 3.0 kb product from a QM B1551 chromosomal DNA template that had been digested with StyI (as determined by a preliminary Southern blot using the original PCR product as probe against QM B1551 DNA digested with various enzymes) and self-ligated. This PCR product was T/A cloned into the SmaI site of pUC19. The resulting 5.7 kb plasmid, pBm33673, was found to contain the complete spoIIGA homologue and was sequenced using a set of nine internal primers.

Construction of SpoIIGA–GFP fusions

B. subtilis spoIIGA–GFP fusions.

Using PY79 chromosomal DNA as a template, PCR was used to amplify an 834 bp region corresponding to the 3′ end of the spoIIGA gene. The rightward primer (IIGA391R: 5′-TTCATCGGATCCAGTATTGTCC-3′) converts a cryptic BamHI site into a unique BamHI site. The leftward primer (IIGA1225L: 5′-GATAAGGTACCGACATTTGCGAACATTTTGAAACG-3′) alters the spoIIGA stop codon to a lysine and generates a unique Asp718I site. This product was digested with BamHI and Asp718I and ligated into the Asp718I/BamHI backbone of plasmid pKSV7 (Smith and Youngman, 1992) to create plasmid pIIGA. The GFP cassette used in this construction was from pMutGFP, a variant of plasmid TU65 (Chalfie et al., 1994) incorporating the intensity-enhancing S65T mutation (Heim et al., 1995). A 740 bp Asp-718I/EcoRI GFP-containing fragment was cloned into pIIGA, resulting in plasmid pIIGA-GFP. In this plasmid, the spoIIGA fragment is joined to the GFP cassette by a linker encoding KCSQMSVPVGK, the initial lysine of which was previously the spoIIGA stop codon. This plasmid was transformed into strains PY79, PY180 and PY507. Chromosomal integration and clone selection were as described previously (Barák et al., 1996). All integrations were confirmed by PCR (data not shown). H. Peters (personal communication) has determined by complementation that the GFP fusion protein is processing negative.

spo0H B. subtilis SpoIIGA–GFP fusions.

spo0H ::kan strain M0165 was obtained from P. Stragier (the integrative plasmid used to create this strain is described in Guérout-Fleury et al., 1995). Chromosomal DNA of this strain was prepared and transformed into PY79 with selection for kanamycin resistance. The resulting strain was then transformed with plasmid pWIIGA-GFP. This plasmid is identical to pIIGA-GFP, except that it includes a 155 bp region upstream of spoIIGA required for full expression and duplication of the gene. We amplified a product from PMF 9 chromosomal DNA using a rightward primer (IIGA104R: 5′-GCTTTTTCTAGATCCTCTCATTATACTTCC-3′), incorporating a unique XbaI site, in conjunction with a leftward primer (GFP-cterm: 5′-GGCTGCAGGAATTCTACGAATGCTATTTGTATAGTTCATCC-3′), incorporating an EcoRI site. This PCR product was digested with XbaI/EcoRI and cloned into pKSV7. Initial transformation and chromosomal integration of this plasmid in the spo0H background used 5 μg ml−1 chloramphenicol, and the resulting strain was trained on increasing levels of chloramphenicol until a concentration of 80 μg ml−1 was reached (Jannière et al., 1985; Piggot and Curtis, 1987).

B. megaterium spoIIGA–GFP fusion.

PCR was used to amplify a 326 bp product corresponding to the 3′ end of the B. megaterium spoIIGA homologue, using QM B1551 chromosomal DNA as a template. The rightward primer (BmGA21R: 5′-GTTACAGCGGATCCTTTAAAAGAAATATTGC-3′) introduced a unique BamHI site. The leftward primer (BmGA347L: 5′-CGCATCGGTACCGGTGTGAACAAATAAAGG-3′) converted the spoIIGA stop codon to valine and introduced a unique Asp718I site. This product was digested with BamHI and Asp718I and ligated into the BamHI/Asp718I backbone of plasmid pCON1 (Barák et al., 1996) to create plasmid pBmIIGA. The GFP cassette used in this construction was from pMutGFP2, a variant of plasmid TU65 (Chalfie et al., 1994) incorporating the intensity enhancing F64L, S65T mutations (Cormack et al., 1996). The Asp718I/EcoRI GFP cassette from pMutGFP2 was then cloned into pBmIIGA, resulting in plasmid pBmIIGA-GFP. In this construct, the spoIIGA gene fragment is connected to the GFP cassette by a linker encoding VPVQK. This plasmid was conjugated into QM B1551 as described previously, except that strain S17-I was used as the E. coli donor, kanamycin was not used, and only T7 phage was used for donor counterselection (Barák and Youngman, 1996). Chromosomal integration and clone selection were as described previously and confirmed by PCR (data not shown).

Preparation of samples for fluorescence and immunofluorescence microscopy

Strains were grown overnight on Difco sporulation medium (DSM) plates containing 5 μg ml−1 chloramphenicol and inoculated into 30 ml of fresh liquid DSM with no antibiotic to a density of approximately 50–70 Klett units (green filter). Samples were periodically withdrawn and incubated at 4°C in preservation buffer (Barák et al., 1996). These samples were either examined directly for GFP fluorescence or processed for IFM using the method of Pogliano et al. (1995) with slight modifications. Initial fixation was in DSM containing 2.5% (v/v) formaldehyde and 0.01% (v/v) glutaraldehyde, and 30 mM NaPO4 buffer. Incubation in lysozyme was for 3 min, and no acetone or methanol was used during subsequent fixation. Labelling reactions used a polyclonal rabbit anti-GFP primary antibody (Clontech) diluted 1:2000 in conjunction with a goat anti-rabbit secondary antibody conjugated to Oregon Green 488 (Molecular Probes) diluted 1:200. Nucleoids were counterstained with 2 μg ml−1 4,6-diamidino-2-phenylindole (DAPI; Sigma). Fluorescent signal from Oregon Green 488 was stabilized by treating slides using the SlowFade Lite antifade reagent (Molecular Probes) according to the manufacturer's instructions. Appropriate control strains (QM B1551 and PS 1) that do not express GFP were used. For microscopy, we used a customized Olympus IMT2 microscope fitted out with a Nanomotion wide-field deconvolution package (Applied Precision) and equipped with fluorescein isothiocyanate (FITC) and DAPI excitation and emission filters (Chroma). Images were captured using a cooled CCD (Princeton Instruments) interfaced to a Silicon Graphics workstation with DeltaVision software, which was used to correct, deconvolve and scale data sets of B. megaterium images. Images were typically taken with a 30 s FITC channel exposure followed by a 10 s DAPI channel exposure through 30–35 iterations of a programme altering the focus by 0.1 μm after each pair of exposures. Optimized images were then used to create three-dimensional projections and QuickTime format movies or flat TIFF format files. Fluorescence micrographs were prepared for printing using an Apple Macintosh running ADOBE PHOTOSHOP 3.05 (Adobe Systems) and printed on a Tektronix Phaser IISDX dye sublimation printer.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results and discussion
  5. Experimental procedures
  6. Acknowledgements
  7. References

We thank Kelly Dawe and Michael Bender for the use of their microscopes, Howard Peters for his evaluation of pro-σE processing, and Dave Brown for assistance with computers and in preparing figures for this manuscript. We also thank Pat Vary and Patrick Stragier for supplying strains, Petra Levin for providing immunofluorescence protocols, and Imro Barák and Antje Hofmeister for commenting on early versions of this manuscript. This work was supported by Public Health Services grant GM35495 from the National Institutes of Health.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results and discussion
  5. Experimental procedures
  6. Acknowledgements
  7. References
  • 1
    Adams, L.F., Brown, K.L., Whiteley, H.R. (1991) Molecular cloning and characterization of two genes encoding sigma factors that direct transcription from a Bacillus thuringiensis crystal protein gene promoter. J Bacteriol 173: 28462854.
  • 2
    Addinall, S.G., Bi, E., Lutkenhaus, J. (1996) FtsZ ring formation in fts mutants. J Bacteriol 178: 38773884.
  • 3
    Agard, D.A. (1984) Optical sectioning microscopy: cellular architecture in three dimensions. Annu Rev Biophys Bioeng 13: 191219.
  • 4
    Agard, D.A. & Sedat, J.S. (1983) Three-dimensional architecture of a polytene nucleus. Nature 302: 676681.
  • 5
    Arigoni, F., Pogliano, K., Webb, C.D., Stragier, P., Losick, R. (1995) Localization of protein implicated in establishment of cell type to sites of asymmetric division. Science 270: 637640.
  • 6
    Arigoni, F., Duncan, L., Alper, S., Losick, R., Stragier, P. (1996) SpoIIE governs the phosphorylation state of a protein regulating transcription factor σF during sporulation in Bacillus subtilis. Proc Natl Acad Sci USA 93: 32383242.
  • 7
    Ausubel, F.M., Brent, R., Kingston, R.E., Moore, D.D., Seidman, J.G., Smith, J.A. et al (eds). (1994) Current Protocols in Molecular Biology. New York: John Wiley and Sons.
  • 8
    Bachmann, B.J. (1987) Derivations and genotypes of some mutant derivatives of Escherichia coli K-12. In Escherichia coli and Salmonella typhimurium: Cellular and Molecular Biology. Neidhardt, F.C., Ingraham, J.L., Low, K.B., Magasanik, B., Schaechter, M., and Umbarger, H.E. (eds). Washington, DC: American Society for Microbiology Press, pp. 11901219.
  • 9
    Barák, I. & 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: 49844989.
  • 10
    Barák, I., Behari, J., Olmedo, G., Guzmán, P., Brown, D.P., Castro, E. et al (1996) Structure and function of the Bacillus SpoIIE protein and its localization to sites of sporulation septum assembly. Mol Microbiol 19: 10471060.
  • 11
    Beall, B., Lowe, M., Lutkenhaus, J. (1988) Cloning and characterization of Bacillus subtilis homologs of Escherichia coli cell division genes ftsZ and ftsA. J Bacteriol 170: 48554864.
  • 12
    Bi, E.F. & Lutkenhaus, J. (1991) FtsZ ring structure associated with division in Escherichia coli. Nature 354: 161164.
  • 13
    Bramhill, D. & Thompson, C.M. (1994) GTP-dependent polymerization of Escherichia coli FtsZ protein to form tubules. Proc Natl Acad Sci USA 91: 58135817.
  • 14
    Chalfie, M., Tu, Y., Euskirchen, G., Ward, W.W., Prasher, D.C. (1994) Green fluorescent protein as a marker for gene expression. Science 263: 802805.
  • 15
    Claros, M.G. & Von Heijne, G. (1994) TopPredII: improved software for membrane protein structure predictions. Comput Appl Biosci 10: 685686.
  • 16
    Cormack, B.P., Valdivia, R., Falkow, S. (1996) FACS-optimized mutants of the green fluorescent protein (GFP). Gene 173: 3338.
  • 17
    Cutting, S.M. & Youngman, P. (1994) Gene transfer in Gram-positive bacteria. In Methods for General and Molecular Bacteriology. Gerhardt, P., Murray, R.G.E., Wood, W.A., and Krieg, N.R. (eds). Washington, DC: American Society for Microbiology Press, pp. 348364.
  • 18
    Dai, K. & Lutkenhaus, J. (1991) ftsZ is an essential cell division gene in Escherichia coli. J Bacteriol 173: 35003506.
  • 19
    Devereux, J., Haeberli, P., Smithies, O. (1984) A comprehensive set of sequence analysis programs for the VAX. Nucleic Acids Res 12: 387396.
  • 20
    Diederich, B., Wilkinson, J.F., Magnin, T., Najafi, M., Errington, J., Yudkin, M.D. (1994) Role of interactions between SpoIIAA and SpoIIAB in regulating cell-specific transcription factor σF of Bacillus subtilis. Genes Dev 8: 26532663.
  • 21
    Duncan, L. & Losick, R. (1993) SpoIIAB is an anti-sigma factor that binds to and inhibits transcription by regulatory protein σF from Bacillus subtilis. Proc Natl Acad Sci USA 90: 23252329.
  • 22
    Duncan, L., Alper, S., Arigoni, F., Losick, R., Stragier, P. (1995) Activation of cell-specific transcription by a serine phosphatase at the site of asymmetric division. Science 270: 641644.
  • 23
    Duncan, L., Alper, S., Losick, R. (1996) SpoIIAA governs the release of the cell-type specific transcription factor σF from its anti-sigma factor SpoIIAB. J Mol Biol 260: 147164.
  • 24
    Edwards, D.H. & Errington, J. (1997) The Bacillus subtilis DivIVA protein targets to the division septum and controls the site specificity of cell division. Mol Microbiol 24: 905915.
  • 25
    Engelman, D.M., Steitz, T.A., Goldman, A. (1986) Identifying non-polar transbilayer helices in amino acid sequences of membrane proteins. Annu Rev Biophys Chem 15: 321353.
  • 26
    Errington, J. (1993) Bacillus subtilis sporulation: regulation of gene expression and control of morphogenesis. Microbiol Rev 57: 133.
  • 27
    Feucht, A., Magnin, T., Yudkin, M.D., Errington, J. (1996) Bifunctional protein required for asymmetric cell division and cell-specific transcription in Bacillus subtilis. Genes Dev 10: 794803.
  • 28
    Gholamhoseinian, A. & Piggot, P.J. (1989) Timing of spoII gene expression relative to septum formation during sporulation of Bacillus subtilis. J Bacteriol 171: 57475749.
  • 29
    Gholamhoseinian, A., Shen, Z., Piggot, P.J. (1992) Regulation of transcription of the cell division gene ftsA during sporulation of Bacillus subtilis. J Bacteriol 174: 46474656.
  • 30
    Gonzy-Tréboul, G., Karmazyn-Campelli, C., Stragier, P. (1992) Developmental transcription of the Bacillus subtilis ftsAZ operon. J Mol Biol 224: 967979.
  • 31
    Guérout-Fleury, A.M., Shazand, K., Frandsen, N., Stragier, P. (1995) Antibiotic-resistance cassettes for Bacillus subtilis. Gene 167: 335336.
  • 32
    Hale, C.A. & De Boer, P.A.J. (1997) Direct binding of FtsZ to ZipA, an essential component of the septal ring structure that mediates cell division in E. coli. Cell 88: 175185.
  • 33
    Harry, E.J. & Wake, R.G. (1997) The membrane-bound cell division protein DivIB is localized to the division site in Bacillus subtilis. Mol Microbiol 25: 275283.
  • 34
    Heim, R., Cubitt, A.B., Tsien, R.Y. (1995) Improved green fluorescence. Nature 373: 663664.
  • 35
    Hitchins, A.D. & Slepecky, R.A. (1969) Bacterial spore formation as a modified prokaryotic cell division. Nature 223: 804807.
  • 36
    Hofmeister, A.E., Londoño-Vallejo, A., Harry, E., Stragier, P., Losick, R. (1995) Extracellular signal protein triggering the proteolytic activation of a developmental transcription factor in B. subtilis. Cell 83: 219226.
  • 37
    Holt, S.C., Gauthier, J.J., Tipper, D.J. (1975) Ultrastructural studies of sporulation in Bacillus sphaericus. J Bacteriol 122: 13221338.
  • 38
    Illing, N. & Errington, J. (1991) Genetic regulation of morphogenesis in Bacillus subtilis: roles of σE andσF in prespore engulfment. J Bacteriol 173: 31593169.
  • 39
    Jannière, L., Niaudet, B., Pierre, E., Ehrlich, S. (1985) Stable gene amplification in the chromosome of Bacillus subtilis. Gene 40: 4755.
  • 40
    Jonas, R.M., Weaver, E.A., Kenney, T.J., Moran, Jr, C., Haldenwang, W.G. (1988) The Bacillus subtilis spoIIG operon encodes both σE and a gene necessary for σE activation. J Bacteriol 170: 507511.
  • 41
    Ju, J., Luo, T., Haldenwang, W.G. (1997) Bacillus subtilis pro-σE fusion protein localizes to the forespore septum and fails to be processed when synthesized in the forespore. J Bacteriol 179: 48884893.
  • 42
    Karow, M.L., Glaser, P., Piggot, P.J. (1995) Identification of a gene, spoIIR, that links the activation of σE to the transcriptional activity of σF during sporulation in Bacillus subtilis. Proc Natl Acad Sci USA 92: 20122016.
  • 43
    Kenney, T.J. & Moran, Jr, C. (1987) Organization and regulation of an operon that encodes a sporulation-essential sigma factor in Bacillus subtilis. J Bacteriol 169: 33293339.
  • 44
    LaBell, T.L., Trempy, J.E., Haldenwang, W.G. (1987) Sporulation-specific σ factor σ29 of Bacillus subtilis is synthesized from a precursor protein, P31. Proc Natl Acad Sci USA 84: 17841788.
  • 45
    Levin, P.A. & 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: 478488.
  • 46
    Levin, P.A., Losick, R., Stragier, P., Arigoni, F. (1997) Localization of the sporulation protein SpoIIE in Bacillus subtilis is dependent upon the cell division protein FtsZ. Mol Microbiol 25: 839846.
  • 47
    Lewis, P.J., Magnin, T., Errington, J. (1996) Compartmentalized distribution of the proteins controlling the prespore-specific transcription factor σF of Bacillus subtilis. Genes Cells 1: 881894.
  • 48
    Londoño-Vallejo, J.A. (1997) Mutational analysis of the early forespore/mother-cell signalling pathway in Bacillus subtilis. Microbiology 143: 27532761.
  • 49
    Londoño-Vallejo, J.A. & Stragier, P. (1995) Cell-cell signalling pathway activating a developmental transcription factor in Bacillus subtilis. Genes Dev 9: 503508.
  • 50
    Londoño-Vallejo, J.A., Fréhel, C., Stragier, P. (1997) spoIIQ, a forespore-expressed gene required for engulfment in Bacillus subtilis. Mol Microbiol 24: 2939.
  • 51
    Ma, X., Ehrhardt, D.W., Margolin, W. (1996) Colocalization of cell division proteins FtsZ and FtsA to cytoskeletal structures in living Escherichia coli cells by using green fluorescent protein. Proc Natl Acad Sci USA 93: 1299813003.
  • 52
    Margolis, P., Driks, A., Losick, R. (1991) Establishment of cell type by compartmentalized activation of a transcription factor. Science 254: 562565.
  • 53
    Masuda, E.S., Anaguchi, H., Sato, T., Takeuchi, M., Kobayashi, Y. (1990) Nucleotide sequence of the sporulation gene spoIIGA from Bacillus subtilis. Nucleic Acids Res 18: 657.
  • 54
    Pearl, L.H. & Taylor, W.R. (1987) A structural model for the retroviral proteases. Nature 329: 351354.
  • 55
    Peters, Iii, H.K. & Haldenwang, W.G. (1991) Synthesis and fractionation properties of SpoIIGA, a protein essential for pro-σE processing in Bacillus subtilis. J Bacteriol 173: 78217827.
  • 56
    Peters, Iii, H.K. & Haldenwang, W.G. (1994) Isolation of a Bacillus subtilis spoIIGA allele that suppresses processing-negative mutations in the pro-σE gene (sigE ). J Bacteriol 176: 77637766.
  • 57
    Piggot, P.J. (1973) Mapping of asporogenous mutations of Bacillus subtilis: a minimum estimate of the number of sporulation operons. J Bacteriol 114: 12411253.
  • 58
    Piggot, P.J. & Curtis, C.A. (1987) Analysis of the regulation of gene expression during Bacillus subtilis sporulation by manipulation of the copy number of spo–lacZ fusions. J Bacteriol 169: 12601266.
  • 59
    Pogliano, J., Pogliano, K., Weiss, D.S., Losick, R., Beckwith, J. (1997) Inactivation of FtsI inhibits constriction of the FtsZ cytokinetic ring and delays the assembly of FtsZ rings at potential division sites. Proc Natl Acad Sci USA 94: 559564.
  • 60
    Pogliano, K., Harry, E., Losick, R. (1995) Visualization of the subcellular location of sporulation proteins in Bacillus subtilis using immunofluorescence microscopy. Mol Microbiol 18: 459470.
  • 61
    Pogliano, K., Hofmeister, A.E.M., Losick, R. (1997) Disappearance of the σE transcription factor from the forespore and the SpoIIE phosphatase from the mother cell contributes to establishment of cell-specific gene expression during sporulation in Bacillus subtilis. J Bacteriol 179: 33313341.
  • 62
    Provence, D.L. & Curtiss, Iii, R. (1994) Gene transfer in Gram-negative bacteria. In Methods for General and Molecular Bacteriology, Gerhardt, P., Murray, R.G.E., Wood, W.A., and Krieg, N.R. (eds). Washington, DC: American Society for Microbiology Press, pp. 317347.
  • 63
    Sandman, K., Losick, R., Youngman, P. (1987) Genetic analysis of Bacillus subtilis spo mutations generated by Tn917-mediated insertional mutagenesis. Genetics 117: 603617.
  • 64
    Sauer, U., Treuner, A., Buchholz, M., Santangelo, J.D., Dürre, P. (1994) Sporulation and primary sigma factor homologous genes in Clostridium acetobutylicum. J Bacteriol 176: 65726582.
  • 65
    Simon, R., Priefer, U., Pühler, A. (1983) A broad host range mobilization system for in vivo genetic engineering: transposon mutagenesis in Gram-negative bacteria. Biotechnology 1: 784791.
  • 66
    Smith, K. & Youngman, P. (1992) Use of a new integrational vector to investigate compartment-specific expression of the Bacillus subtilis spoIIM gene. Biochimie 74: 705711.
  • 67
    Stragier, P. & Losick, R. (1996) Molecular genetics of sporulation in Bacillus subtilis. Annu Rev Genet 30: 297341.
  • 68
    Stragier, P., Bouvier, J., Bonamy, C., Szulmajster, J. (1984) A developmental gene product of Bacillus subtilis homologous to the sigma factor of Escherichia coli. Nature 312: 376378.
  • 69
    Stragier, P., Bonamy, C., Karmazyn-Campelli, C. (1988) Processing of a sporulation sigma factor in Bacillus subtilis: how morphological structure could control gene expression. Cell 52: 697704.
  • 70
    Trempy, J.E., Bonamy, C., Szulmajster, J., Haldenwang, W.G. (1985) Bacillus subtilis sigma factor σ29 is the product of the sporulation-essential gene spoIIG. Proc Natl Acad Sci USA 82: 41894192.
  • 71
    Vincente, M. & Errington, J. (1996) Structure, function and controls in microbial division. Mol Microbiol 20: 17.
  • 72
    Wang, X., Huang, J., Mukherjee, A., Cao, C., Lutkenhaus, J. (1997) Analysis of the interaction of FtsZ with itself, GTP, and FtsA. J Bacteriol 179: 55515559.
  • 73
    Weir, J., Predich, M., Dubnau, E., Nair, G., Smith, I. (1991) Regulation of spo0H, a gene coding for the Bacillus subtilisσH factor. J Bacteriol 173: 521529.
  • 74
    Weiss, D.S., Pogliano, K., Carson, M., Guzmán, L., Fraipont, C., Nguyen-Distèche, M. et al (1997) Localization of the Escherichia coli cell division protein FtsI (PBP3) to the division site and cell pole. Mol Microbiol 25: 671681.
  • 75
    Wu, L.J. & Errington, J. (1994) Bacillus subtilis SpoIIIE protein required for DNA segregation during asymmetric cell division. Science 264: 572575.
  • 76
    Wu, L.J., Lewis, P.J., Allmansberger, R., Hauser, P.M., Errington, J. (1995) A conjugation-like mechanism for prespore chromosome partitioning during sporulation in Bacillus subtilis. Genes Dev 9: 13161326.
  • 77
    Youngman, P., Perkins, J.B., Losick, R. (1984) Construction of a cloning site near one end of Tn917 into which foreign DNA may be inserted without affecting transposition in Bacillus subtilis or expression of the transposon-borne erm gene. Plasmid 12: 19.