- Email: [email protected]
Bacterial differentiation: Sizing up sporulation
Dispatch
111
Bacterial differentiation: Sizing up sporulation Urs Jenal* and Craig Stephens
*Present address: Department of Microbiology, Biozentrum, University of Basel, Klingelbergstrasse 70, CH-4056 Basel, Switzerland. Current Biology 1996, Vol 6 No 2:111–114 © Current Biology Ltd ISSN 0960-9822
Asymmetric cell division is fundamental to the development of multicellular organisms, and is inevitably accompanied by the induction of distinct gene expression programs in the progeny cells. Understanding the mechanisms that initiate and propagate differential gene expression in dissimilar progeny cells is one of the ultimate goals of developmental biology. Asymmetric cell division is observed in unicellular organisms as well as metazoans. The process of starvation-induced endospore formation — sporulation — in Gram-positive bacteria is a dramatic example of such an event. One of the progeny is a compact and extremely durable spore, and the other is a ‘mother cell’ that aids in construction of the spore but is lysed at the completion of the process. The bacterium Bacillus subtilis has proven to be a fascinating model system for understanding molecular mechanisms of temporal and spatial regulation of gene expression during sporulation, and recent work from several laboratories [1–3] has revealed signalling pathways that connect early morphological changes in the sporulating cell to differential gene expression in the nascent forespore and mother cell compartments. The developmental programs of the forespore and mother cell compartments in B. subtilis are executed through a set of related transcription factors: sF, sE, sG and sK. The s subunit of bacterial RNA polymerase controls its specificity for promoter sequences. After the initial commitment to sporulation is made, these s factors are sequentially activated, and consequently new sets of promoters are turned on to express a variety of proteins necessary for the complex task of spore assembly [4]. These s factors control both timing and spatial location of gene expression, with sF and sE being responsible for initial forespore and mother-cell specific gene expression, respectively. Immediately following completion of the septum separating the two compartments, sF becomes active in the forespore; sE
As sF and sE are both synthesized prior to septum formation, the mere presence of these factors does not account for compartmentalized gene expression. Losick and colleagues [5] have shown that sF activity is controlled by the SpoIIAA and SpoIIAB proteins. SpoIIAB, an ‘antisigma factor’, binds sF and blocks its function. However, SpoIIAB can also bind to SpoIIAA, which prevents SpoIIAB from inhibiting sF. The affinity of SpoIIAB for either partner is modulated by adenosine nucleotides. When bound to ATP, SpoIIAB binds preferentially to sF; when bound to ADP, it binds SpoIIAA. A further twist to this regulatory pathway was revealed by work from the laboratories of Yudkin and Errington [6,7]. SpoIIAB was shown to possess a protein serine kinase activity which phosphorylates SpoIIAA. The resulting SpoIIAA–P (phosphorylated SpoIIAA) is unable to bind SpoIIAB and prevent sF sequestration. SpoIIAB kinase activity is dependent on ATP and is inhibited by ADP. From these various observations, a model was developed [5,7] in which the relative levels of ATP and ADP act as a switch controlling sF activity via SpoIIAB: at a high ATP/ADP ratio, proposed to exist in the mother cell, sF would be held inactive, whereas at a lower ATP/ADP ratio, somehow generated in the forespore compartment, sF would be free to activate transcription. Now evidence has been obtained, from exciting work in the laboratories of Losick and Stragier [1–3], suggesting that there may be an additional, nucleotide-independent mechanism by which septation regulates the phosphorylation state of SpoIIAA. Genetic evidence had earlier identified the spoIIE locus as a positive regulator of SpoIIAA function; spoIIE mutants form a septum, albeit an abnormal one, but fail to activate sF and proceed no further in sporulation. SpoIIE is a membrane-spanning protein with a large cytoplasmic domain, which Duncan et al. [2] have now shown has serine phosphatase activity. SpoIIE dephosphorylates SpoIIAA–P, thereby allowing SpoIIAA to bind SpoIIAB and free sF for transcription. The pathway that regulates sF activity, as currently understood [2,5,7], can thus be summarized as: SpoIIE
SpoIIAA–P + SpoIIAB:sF
➝ SpoIIAA:SpoIIAB + sF
SpoIIAB/ATP
⊥
Address: Department of Developmental Biology, Beckman Center for Molecular and Genetic Medicine, Stanford University, Stanford, California 94305-5427, USA.
activation in the mother cell depends on sF activity in the forespore, through an intercompartmental signalling pathway described below. Activation of sF in the forespore is thus a pivotal step in the generation of asymmetry.
➝
New results on Bacillus subtilis sporulation suggest that size differences between the post-septation compartments trigger differential gene expression, which is then coordinated by communication between the nascent mother cell and forespore compartments.
ADP
112
Current Biology 1996, Vol 6 No 2
Figure 1
(a)
The balance of these reactions is predicted to be shifted to the right in the forespore and to the left in the mother cell, though we await experimental confirmation of this. The activation of sF can be added to the growing list of phosphorylation-mediated signalling pathways in prokaryotes [8] and eukaryotes [9], in which the phosphorylation state of a critical component is controlled by a balance of kinase and phosphatase activities, each of which may serve as sites for transduction of intracellular signals.
Starvation, high cell density, other signals
B. subtilis cell
SpollE
pro-σE
SpollAB:σ
F
SpollAA–P
Septation (b) Mother cell
Forespore pro-σE F
pro-σE
σ
SpollAA–P
SpollAA:AB
SpollAB:σ
F
Intercompartmental communication (c)
SpollGA SpollAB:σ
F
pro-σE σ
F
pro-σE SpollAA–P
SpollR σ
E
? © 1996 Current Biology
A model for compartment-specific induction of sF activity and intercompartmental signal transduction early during B. subtilis sporulation. (a) Mid-cell septation is blocked at the onset of sporulation, and septum formation is redirected to one of the two possible sites near the pole. The SpoIIE phosphatase initially localizes to the polar sites. (b) SpoIIE is incorporated into the growing septum, where it catalyzes dephosphorylation of SpoIIAA–P, allowing SpoIIAA to bind SpoIIAB and free sF. (c) sF permits expression of the SpoIIR protein in the forespore. SpoIIR is secreted from the forespore, and activates the membrane-bound SpoIIGA protease. SpoIIGA processes pro-sE in the mother cell, initiating the mother-cell specific developmental program. Mechanisms that might account for restriction of sE activity to the mother cell are illustrated in Figure 2.
How does the SpoIIE phosphatase contribute to spatial regulation of sF activity? Like sF and sE, SpoIIE is synthesized before septum formation. During sporulation, the medial division site used during vegetative cell division is suppressed, and the sporulation septum forms near one of the poles. SpoIIE is initially concentrated near both poles of the cell, presumably at potential septation sites (Fig. 1a) [3]. Only one of these sites is actually used to form the septum during sporulation. The crucial observation of Arigoni et al. [3] is that, upon formation of the polar septum, SpoIIE localizes exclusively to the septum, and disappears from the other end of the cell (Fig. 1b). After forespore gene expression is activated, SpoIIE disappears completely (Fig. 1c). In ‘disporic’ mutants, which form forespores at both poles, both septal sites are used and both contain SpoIIE. The link between localization of SpoIIE to the septum and activation of sF in the forespore remains speculative. Nevertheless, one intriguing and elegantly simple idea has emerged. SpoIIE could be present in the membrane on both sides of the septum, but the size of the compartments might determine the extent of dephosphorylation of SpoIIAA–P by SpoIIE [2]. As cells increase in size, the surface area to volume ratio decreases; because the sporulation septum is positioned close to the pole, the volume of the forespore compartment is at least five-fold smaller than the mother cell, and the ratio of septal membrane surface area to cytoplasmic volume is at least five-fold higher in the forespore. Thus, even if the phosphatase domain of SpoIIE is equally abundant on both sides of the septum, it is more concentrated relative to cytoplasmic contents in the forespore, and could have a greater effect in shifting the balance from SpoIIAA–P to SpoIIAA. If localized variations in adenosine nucleotide ratios [5,7] do occur, they might also be linked to compartmental size differences, so that the SpoIIE phosphatase and nucleotide ‘sensing’ by SpoIIAB could work together to activate sF-dependent transcription in the forespore [2]. Gene expression dependent on sE is initiated in, and confined to, the mother cell after formation of the septum and activation of sF in the forespore [10,11]. Once sE is active, it blocks formation of a second potential polar septum in the mother cell that would otherwise generate an abortive disporic cell [12]. This is one reason why it
Dispatch
Figure 2 Septum
(a) Mother cell cytoplasm
Forespore cytoplasm pro-σE
pro-σE SpoIIGA
σF
SpoIIR
σE (b) pro-σE
pro-σE
σF σE
(c) pro-σE
pro-σE
σE
σF ?
(d) pro-σE
σE
pro-σE σF σE
© 1996 Current Biology
?
Models for the spatial control of sE activity. SpoIIR protein is exported from the forespore, and subsequently stimulates, either directly or indirectly, the protease activity of the membrane-bound SpoIIGA, leading to processing of pro-sE. Four models are shown that could explain how sE activity is restricted to the mother cell compartment. (a) SpoIIGA is specifically localized to the mother cell membrane. (b) SpoIIGA is present in forespore and mother cell membranes, but SpoIIR activates SpoIIGA only in the mother cell membrane. (c) SpoIIR interacts with SpoIIGA in both membranes, but SpoIIGA proteolytic activity is blocked in the forespore. (d) Pro-sE is cleaved in both compartments, but sE activity is blocked in the forespore.
might be important to delay the activation of sE until the first polar septum has successfully formed. As sE activation in the mother cell requires sF-dependent transcription in the forespore [13], there must be some form of
113
intercompartmental communication between the forespore and mother cell. sE is synthesized before septation as an inactive precursor protein, pro-sE, bearing an aminoterminal extension that is proteolytically removed in the activation step [10]. Pro-sE is encoded by the second gene, spoIIGB, of the spoIIG operon, the first gene of which, spoIIGA, encodes a membrane protein that appears to be the pro-sE protease. Losick and Stragier [4] had earlier postulated that SpoIIGA activity in the mother cell is controlled by the product of a gene that is synthesized under sF control in the forespore, and that this product acts in a directional way across the membrane to activate sE only in the mother cell. A candidate gene, spoIIR, has recently been isolated whose product is both required and sufficient for triggering the processing of pro-sE [14–15]. Hofmeister et al. [1] have now presented biochemical evidence that SpoIIR, which has a potential signal sequence at its amino terminus for export out of the cytoplasm, is an extracellular signaling protein. To mimic intercompartmental signaling during sporulation, an assay was designed using vegetative cells, with a donor strain constitutively expressing sF (and therefore SpoIIR), and recipient cells expressing pro-sE and SpoIIGA. The supernatant of the donor strain culture was able to activate pro-sE processing in recipient cells, suggesting that a secreted factor triggered the proteolytic event. Proteolysis of pro-sE in the recipient strain was dependent on a spoIIR copy in the donor, and could also be stimulated by SpoIIR produced in Escherichia coli, verifying that SpoIIR is the factor that activates SpoIIGA extracellularly. The current speculation is that SpoIIGA is both a receptor and a protease, able to process pro-sE after stimulation by interaction with SpoIIR, or a SpoIIR-dependent signal, outside of the cytoplasmic membrane (Fig. 1c). It seems likely that signalling occurs primarily via the fraction of SpoIIR secreted into the space between the septal membranes (as shown in Fig. 2), simply because SpoIIR could become concentrated in this space. So, we now have a tentative outline of the signalling pathway linking activation of sE to septation and early forespore gene expression. What possible mechanisms could confine sE activity to the mother cell? Pro-sE processing could be limited to the mother cell if the SpoIIGA protein is specifically localized to the mother cell membrane (Fig. 2a), or signalling by SpoIIR from the forespore to the mother cell compartment occurs unidirectionally through the sporulation septum (Fig. 2b). There is evidence, however, that at least some processed sE is found in the forespore as well [16]. If so, the SpoIIR–SpoIIGA pathway might be designed solely to regulate the timing of sE activation by making the process dependent on septum formation and early events in the forespore [13]. SpoIIR might activate SpoIIGA in both
114
Current Biology 1996, Vol 6 No 2
compartments, with additional unidentified processes inhibiting SpoIIGA protease activity (Fig. 2c) or sE itself (Fig. 2d) in the forespore. Such forespore-specific inhibition could be controlled by sF, in which case spatial regulation of sE activity could be accomplished solely by the already compartmentalized activity of sF, without further requirements for protein localization or vectorial signal transduction. Hence we return to asymmetric activation of sF as the ultimate arbiter of compartmentalized gene expression during sporulation. The model for sF activation discussed earlier implies that it would be impossible to compartmentalize gene expression properly if the septum were not asymmetrically positioned. Placement of the septum near the pole of the cell is thus the essential morphogenetic step on the path to distinct developmental programs in the forespore and mother cell. It is critical to understand what controls placement of the polar septum, and how the medial septation site is suppressed. Polar and mid-cell septation are clearly related events, sharing at least some of the same players. The tubulin-like protein FtsZ, for example, is essential for cell division during vegetative growth, and is also required for the formation of the sporulation septum, and consequently for activation of the s factor cascade [17]. FtsZ has been shown in several bacterial species, including B. subtilis, to form ring structures at sites of cell division. There must be mechanisms that redirect FtsZ assembly from the mid-cell to the polar sites at the onset of sporulation. The propensity of SpoIIE to localize initially in the vicinity of the polar septation sites suggests that it recognizes and is targeted to these sites, where it might be incorporated continuously into the growing septum. If SpoIIE is targeted to potential polar septation sites, it presumably interacts with one or more proteins that mark these sites, such as FtsZ or its companions. An understanding of the nature and temporal sequence of these interactions should reveal the molecular nature of the polar septation sites, and help to unveil the initial morphogenetic events that ultimately lead to asymmetric cell division. References 1. Hofmeister AEM, Londono-Vallejo A, Harry E, Stragier P, Losick R: Extracellular signal protein triggering the proteolytic activation of a developmental transcription factor in B. subtilis. Cell 1995, 83:219–226. 2. Duncan L, Alper S, Arigoni F, Losick R, Stragier P: Activation of cellspecific transcription by a serine phosphatase at the site of asymmetric division. Science 1995, 270:641–644. 3. Arigoni F, Pogliano K, Webb CD, Stragier P, Losick R: Localization of protein implicated in establishment of cell type to sites of asymmetric division. Science 1995, 270:637–640. 4. Losick R, Stragier P: Crisscross regulation of cell-type-specific gene expression during development in B. subtilis. Nature 1992, 355:601–604. 5. Alper S, Duncan L, Losick R: An adenosine nucleotide switch controlling the activity of a cell type-specific transcription factor in B. subtilis. Cell 1994, 77:195–205.
6. Min KT, Hilditch CM, Diederich B, Errington J, Yudkin MD: Sigma F, the first compartment-specific transcription factor of B. subtilis, is regulated by an anti-sigma factor that is also a protein kinase.Cell 1993, 74:735–742. 7. Diederich B, Wilkinson JF, Magnin T, Najafi M, Errington J, Yudkin MD: Role of interactions between SpoIIAA and SpoIIAB in regulating cell-specific transcription factor sigma F of Bacillus subtilis. Genes Dev 1994, 8:2653–2663. 8. Perego M, Hanstein C, Welsh KM, Djavakhishvili T, Glaser P, Hoch JA: Multiple protein-aspartate phosphatases provide a mechanism for the integration of diverse signals in the control of development in B. subtilis. Cell 1994, 79:1047–1055. 9. Charbonneau H, Tonks NK: 1002 protein phosphatases? Annu Rev Cell Biol 1992, 8:463–493. 10. LaBell TL, Trempy JE, Haldenwang WG: Sporulation-specific sigma factor sigma 29 of Bacillus subtilis is synthesized from a precursor protein, P31. Proc Natl Acad Sci USA 1987, 84:1784–1788. 11. Driks A, Losick R: Compartmentalized expression of a gene under the control of sporulation transcription factor sigma E in Bacillus subtilis. Proc Natl Acad Sci USA 1991, 88:9934–9938. 12. Lewis PJ, Partridge SR, Errington J: Sigma factors, asymmetry, and the determination of cell fate in Bacillus subtilis. Proc Natl Acad Sci USA 1994, 91:3849–3853. 13. Shazand K, Frandsen N, Stragier P: Cell-type specificity during development in Bacillus subtilis: the molecular and morphological requirements for sigma E activation. EMBO J 1995, 14:1439–1445. 14. Karow ML, Glaser P, Piggot PJ: Identification of a gene, spoIIR, that links the activation of sigma E to the transcriptional activity of sigma F during sporulation in Bacillus subtilis. Proc Natl Acad Sci USA 1995, 92:2012–2016. 15. Londono-Vallejo JA, Stragier P: Cell-cell signaling pathway activating a developmental transcription factor in Bacillus subtilis. Genes Dev 1995, 9:503–508. 16. Kirchman PA, DeGrazia H, Kellner EM, Moran C Jr: Foresporespecific disappearance of the sigma-factor antagonist spoIIAB: implications for its role in determination of cell fate in Bacillus subtilis. Mol Microbiol 1993, 8:663–671. 17. Beall B, Lutkenhaus J: FtsZ in Bacillus subtilis is required for vegetative septation and for asymmetric septation during sporulation. Genes Dev 1991, 5:447–455.
111
Bacterial differentiation: Sizing up sporulation Urs Jenal* and Craig Stephens
*Present address: Department of Microbiology, Biozentrum, University of Basel, Klingelbergstrasse 70, CH-4056 Basel, Switzerland. Current Biology 1996, Vol 6 No 2:111–114 © Current Biology Ltd ISSN 0960-9822
Asymmetric cell division is fundamental to the development of multicellular organisms, and is inevitably accompanied by the induction of distinct gene expression programs in the progeny cells. Understanding the mechanisms that initiate and propagate differential gene expression in dissimilar progeny cells is one of the ultimate goals of developmental biology. Asymmetric cell division is observed in unicellular organisms as well as metazoans. The process of starvation-induced endospore formation — sporulation — in Gram-positive bacteria is a dramatic example of such an event. One of the progeny is a compact and extremely durable spore, and the other is a ‘mother cell’ that aids in construction of the spore but is lysed at the completion of the process. The bacterium Bacillus subtilis has proven to be a fascinating model system for understanding molecular mechanisms of temporal and spatial regulation of gene expression during sporulation, and recent work from several laboratories [1–3] has revealed signalling pathways that connect early morphological changes in the sporulating cell to differential gene expression in the nascent forespore and mother cell compartments. The developmental programs of the forespore and mother cell compartments in B. subtilis are executed through a set of related transcription factors: sF, sE, sG and sK. The s subunit of bacterial RNA polymerase controls its specificity for promoter sequences. After the initial commitment to sporulation is made, these s factors are sequentially activated, and consequently new sets of promoters are turned on to express a variety of proteins necessary for the complex task of spore assembly [4]. These s factors control both timing and spatial location of gene expression, with sF and sE being responsible for initial forespore and mother-cell specific gene expression, respectively. Immediately following completion of the septum separating the two compartments, sF becomes active in the forespore; sE
As sF and sE are both synthesized prior to septum formation, the mere presence of these factors does not account for compartmentalized gene expression. Losick and colleagues [5] have shown that sF activity is controlled by the SpoIIAA and SpoIIAB proteins. SpoIIAB, an ‘antisigma factor’, binds sF and blocks its function. However, SpoIIAB can also bind to SpoIIAA, which prevents SpoIIAB from inhibiting sF. The affinity of SpoIIAB for either partner is modulated by adenosine nucleotides. When bound to ATP, SpoIIAB binds preferentially to sF; when bound to ADP, it binds SpoIIAA. A further twist to this regulatory pathway was revealed by work from the laboratories of Yudkin and Errington [6,7]. SpoIIAB was shown to possess a protein serine kinase activity which phosphorylates SpoIIAA. The resulting SpoIIAA–P (phosphorylated SpoIIAA) is unable to bind SpoIIAB and prevent sF sequestration. SpoIIAB kinase activity is dependent on ATP and is inhibited by ADP. From these various observations, a model was developed [5,7] in which the relative levels of ATP and ADP act as a switch controlling sF activity via SpoIIAB: at a high ATP/ADP ratio, proposed to exist in the mother cell, sF would be held inactive, whereas at a lower ATP/ADP ratio, somehow generated in the forespore compartment, sF would be free to activate transcription. Now evidence has been obtained, from exciting work in the laboratories of Losick and Stragier [1–3], suggesting that there may be an additional, nucleotide-independent mechanism by which septation regulates the phosphorylation state of SpoIIAA. Genetic evidence had earlier identified the spoIIE locus as a positive regulator of SpoIIAA function; spoIIE mutants form a septum, albeit an abnormal one, but fail to activate sF and proceed no further in sporulation. SpoIIE is a membrane-spanning protein with a large cytoplasmic domain, which Duncan et al. [2] have now shown has serine phosphatase activity. SpoIIE dephosphorylates SpoIIAA–P, thereby allowing SpoIIAA to bind SpoIIAB and free sF for transcription. The pathway that regulates sF activity, as currently understood [2,5,7], can thus be summarized as: SpoIIE
SpoIIAA–P + SpoIIAB:sF
➝ SpoIIAA:SpoIIAB + sF
SpoIIAB/ATP
⊥
Address: Department of Developmental Biology, Beckman Center for Molecular and Genetic Medicine, Stanford University, Stanford, California 94305-5427, USA.
activation in the mother cell depends on sF activity in the forespore, through an intercompartmental signalling pathway described below. Activation of sF in the forespore is thus a pivotal step in the generation of asymmetry.
➝
New results on Bacillus subtilis sporulation suggest that size differences between the post-septation compartments trigger differential gene expression, which is then coordinated by communication between the nascent mother cell and forespore compartments.
ADP
112
Current Biology 1996, Vol 6 No 2
Figure 1
(a)
The balance of these reactions is predicted to be shifted to the right in the forespore and to the left in the mother cell, though we await experimental confirmation of this. The activation of sF can be added to the growing list of phosphorylation-mediated signalling pathways in prokaryotes [8] and eukaryotes [9], in which the phosphorylation state of a critical component is controlled by a balance of kinase and phosphatase activities, each of which may serve as sites for transduction of intracellular signals.
Starvation, high cell density, other signals
B. subtilis cell
SpollE
pro-σE
SpollAB:σ
F
SpollAA–P
Septation (b) Mother cell
Forespore pro-σE F
pro-σE
σ
SpollAA–P
SpollAA:AB
SpollAB:σ
F
Intercompartmental communication (c)
SpollGA SpollAB:σ
F
pro-σE σ
F
pro-σE SpollAA–P
SpollR σ
E
? © 1996 Current Biology
A model for compartment-specific induction of sF activity and intercompartmental signal transduction early during B. subtilis sporulation. (a) Mid-cell septation is blocked at the onset of sporulation, and septum formation is redirected to one of the two possible sites near the pole. The SpoIIE phosphatase initially localizes to the polar sites. (b) SpoIIE is incorporated into the growing septum, where it catalyzes dephosphorylation of SpoIIAA–P, allowing SpoIIAA to bind SpoIIAB and free sF. (c) sF permits expression of the SpoIIR protein in the forespore. SpoIIR is secreted from the forespore, and activates the membrane-bound SpoIIGA protease. SpoIIGA processes pro-sE in the mother cell, initiating the mother-cell specific developmental program. Mechanisms that might account for restriction of sE activity to the mother cell are illustrated in Figure 2.
How does the SpoIIE phosphatase contribute to spatial regulation of sF activity? Like sF and sE, SpoIIE is synthesized before septum formation. During sporulation, the medial division site used during vegetative cell division is suppressed, and the sporulation septum forms near one of the poles. SpoIIE is initially concentrated near both poles of the cell, presumably at potential septation sites (Fig. 1a) [3]. Only one of these sites is actually used to form the septum during sporulation. The crucial observation of Arigoni et al. [3] is that, upon formation of the polar septum, SpoIIE localizes exclusively to the septum, and disappears from the other end of the cell (Fig. 1b). After forespore gene expression is activated, SpoIIE disappears completely (Fig. 1c). In ‘disporic’ mutants, which form forespores at both poles, both septal sites are used and both contain SpoIIE. The link between localization of SpoIIE to the septum and activation of sF in the forespore remains speculative. Nevertheless, one intriguing and elegantly simple idea has emerged. SpoIIE could be present in the membrane on both sides of the septum, but the size of the compartments might determine the extent of dephosphorylation of SpoIIAA–P by SpoIIE [2]. As cells increase in size, the surface area to volume ratio decreases; because the sporulation septum is positioned close to the pole, the volume of the forespore compartment is at least five-fold smaller than the mother cell, and the ratio of septal membrane surface area to cytoplasmic volume is at least five-fold higher in the forespore. Thus, even if the phosphatase domain of SpoIIE is equally abundant on both sides of the septum, it is more concentrated relative to cytoplasmic contents in the forespore, and could have a greater effect in shifting the balance from SpoIIAA–P to SpoIIAA. If localized variations in adenosine nucleotide ratios [5,7] do occur, they might also be linked to compartmental size differences, so that the SpoIIE phosphatase and nucleotide ‘sensing’ by SpoIIAB could work together to activate sF-dependent transcription in the forespore [2]. Gene expression dependent on sE is initiated in, and confined to, the mother cell after formation of the septum and activation of sF in the forespore [10,11]. Once sE is active, it blocks formation of a second potential polar septum in the mother cell that would otherwise generate an abortive disporic cell [12]. This is one reason why it
Dispatch
Figure 2 Septum
(a) Mother cell cytoplasm
Forespore cytoplasm pro-σE
pro-σE SpoIIGA
σF
SpoIIR
σE (b) pro-σE
pro-σE
σF σE
(c) pro-σE
pro-σE
σE
σF ?
(d) pro-σE
σE
pro-σE σF σE
© 1996 Current Biology
?
Models for the spatial control of sE activity. SpoIIR protein is exported from the forespore, and subsequently stimulates, either directly or indirectly, the protease activity of the membrane-bound SpoIIGA, leading to processing of pro-sE. Four models are shown that could explain how sE activity is restricted to the mother cell compartment. (a) SpoIIGA is specifically localized to the mother cell membrane. (b) SpoIIGA is present in forespore and mother cell membranes, but SpoIIR activates SpoIIGA only in the mother cell membrane. (c) SpoIIR interacts with SpoIIGA in both membranes, but SpoIIGA proteolytic activity is blocked in the forespore. (d) Pro-sE is cleaved in both compartments, but sE activity is blocked in the forespore.
might be important to delay the activation of sE until the first polar septum has successfully formed. As sE activation in the mother cell requires sF-dependent transcription in the forespore [13], there must be some form of
113
intercompartmental communication between the forespore and mother cell. sE is synthesized before septation as an inactive precursor protein, pro-sE, bearing an aminoterminal extension that is proteolytically removed in the activation step [10]. Pro-sE is encoded by the second gene, spoIIGB, of the spoIIG operon, the first gene of which, spoIIGA, encodes a membrane protein that appears to be the pro-sE protease. Losick and Stragier [4] had earlier postulated that SpoIIGA activity in the mother cell is controlled by the product of a gene that is synthesized under sF control in the forespore, and that this product acts in a directional way across the membrane to activate sE only in the mother cell. A candidate gene, spoIIR, has recently been isolated whose product is both required and sufficient for triggering the processing of pro-sE [14–15]. Hofmeister et al. [1] have now presented biochemical evidence that SpoIIR, which has a potential signal sequence at its amino terminus for export out of the cytoplasm, is an extracellular signaling protein. To mimic intercompartmental signaling during sporulation, an assay was designed using vegetative cells, with a donor strain constitutively expressing sF (and therefore SpoIIR), and recipient cells expressing pro-sE and SpoIIGA. The supernatant of the donor strain culture was able to activate pro-sE processing in recipient cells, suggesting that a secreted factor triggered the proteolytic event. Proteolysis of pro-sE in the recipient strain was dependent on a spoIIR copy in the donor, and could also be stimulated by SpoIIR produced in Escherichia coli, verifying that SpoIIR is the factor that activates SpoIIGA extracellularly. The current speculation is that SpoIIGA is both a receptor and a protease, able to process pro-sE after stimulation by interaction with SpoIIR, or a SpoIIR-dependent signal, outside of the cytoplasmic membrane (Fig. 1c). It seems likely that signalling occurs primarily via the fraction of SpoIIR secreted into the space between the septal membranes (as shown in Fig. 2), simply because SpoIIR could become concentrated in this space. So, we now have a tentative outline of the signalling pathway linking activation of sE to septation and early forespore gene expression. What possible mechanisms could confine sE activity to the mother cell? Pro-sE processing could be limited to the mother cell if the SpoIIGA protein is specifically localized to the mother cell membrane (Fig. 2a), or signalling by SpoIIR from the forespore to the mother cell compartment occurs unidirectionally through the sporulation septum (Fig. 2b). There is evidence, however, that at least some processed sE is found in the forespore as well [16]. If so, the SpoIIR–SpoIIGA pathway might be designed solely to regulate the timing of sE activation by making the process dependent on septum formation and early events in the forespore [13]. SpoIIR might activate SpoIIGA in both
114
Current Biology 1996, Vol 6 No 2
compartments, with additional unidentified processes inhibiting SpoIIGA protease activity (Fig. 2c) or sE itself (Fig. 2d) in the forespore. Such forespore-specific inhibition could be controlled by sF, in which case spatial regulation of sE activity could be accomplished solely by the already compartmentalized activity of sF, without further requirements for protein localization or vectorial signal transduction. Hence we return to asymmetric activation of sF as the ultimate arbiter of compartmentalized gene expression during sporulation. The model for sF activation discussed earlier implies that it would be impossible to compartmentalize gene expression properly if the septum were not asymmetrically positioned. Placement of the septum near the pole of the cell is thus the essential morphogenetic step on the path to distinct developmental programs in the forespore and mother cell. It is critical to understand what controls placement of the polar septum, and how the medial septation site is suppressed. Polar and mid-cell septation are clearly related events, sharing at least some of the same players. The tubulin-like protein FtsZ, for example, is essential for cell division during vegetative growth, and is also required for the formation of the sporulation septum, and consequently for activation of the s factor cascade [17]. FtsZ has been shown in several bacterial species, including B. subtilis, to form ring structures at sites of cell division. There must be mechanisms that redirect FtsZ assembly from the mid-cell to the polar sites at the onset of sporulation. The propensity of SpoIIE to localize initially in the vicinity of the polar septation sites suggests that it recognizes and is targeted to these sites, where it might be incorporated continuously into the growing septum. If SpoIIE is targeted to potential polar septation sites, it presumably interacts with one or more proteins that mark these sites, such as FtsZ or its companions. An understanding of the nature and temporal sequence of these interactions should reveal the molecular nature of the polar septation sites, and help to unveil the initial morphogenetic events that ultimately lead to asymmetric cell division. References 1. Hofmeister AEM, Londono-Vallejo A, Harry E, Stragier P, Losick R: Extracellular signal protein triggering the proteolytic activation of a developmental transcription factor in B. subtilis. Cell 1995, 83:219–226. 2. Duncan L, Alper S, Arigoni F, Losick R, Stragier P: Activation of cellspecific transcription by a serine phosphatase at the site of asymmetric division. Science 1995, 270:641–644. 3. Arigoni F, Pogliano K, Webb CD, Stragier P, Losick R: Localization of protein implicated in establishment of cell type to sites of asymmetric division. Science 1995, 270:637–640. 4. Losick R, Stragier P: Crisscross regulation of cell-type-specific gene expression during development in B. subtilis. Nature 1992, 355:601–604. 5. Alper S, Duncan L, Losick R: An adenosine nucleotide switch controlling the activity of a cell type-specific transcription factor in B. subtilis. Cell 1994, 77:195–205.
6. Min KT, Hilditch CM, Diederich B, Errington J, Yudkin MD: Sigma F, the first compartment-specific transcription factor of B. subtilis, is regulated by an anti-sigma factor that is also a protein kinase.Cell 1993, 74:735–742. 7. Diederich B, Wilkinson JF, Magnin T, Najafi M, Errington J, Yudkin MD: Role of interactions between SpoIIAA and SpoIIAB in regulating cell-specific transcription factor sigma F of Bacillus subtilis. Genes Dev 1994, 8:2653–2663. 8. Perego M, Hanstein C, Welsh KM, Djavakhishvili T, Glaser P, Hoch JA: Multiple protein-aspartate phosphatases provide a mechanism for the integration of diverse signals in the control of development in B. subtilis. Cell 1994, 79:1047–1055. 9. Charbonneau H, Tonks NK: 1002 protein phosphatases? Annu Rev Cell Biol 1992, 8:463–493. 10. LaBell TL, Trempy JE, Haldenwang WG: Sporulation-specific sigma factor sigma 29 of Bacillus subtilis is synthesized from a precursor protein, P31. Proc Natl Acad Sci USA 1987, 84:1784–1788. 11. Driks A, Losick R: Compartmentalized expression of a gene under the control of sporulation transcription factor sigma E in Bacillus subtilis. Proc Natl Acad Sci USA 1991, 88:9934–9938. 12. Lewis PJ, Partridge SR, Errington J: Sigma factors, asymmetry, and the determination of cell fate in Bacillus subtilis. Proc Natl Acad Sci USA 1994, 91:3849–3853. 13. Shazand K, Frandsen N, Stragier P: Cell-type specificity during development in Bacillus subtilis: the molecular and morphological requirements for sigma E activation. EMBO J 1995, 14:1439–1445. 14. Karow ML, Glaser P, Piggot PJ: Identification of a gene, spoIIR, that links the activation of sigma E to the transcriptional activity of sigma F during sporulation in Bacillus subtilis. Proc Natl Acad Sci USA 1995, 92:2012–2016. 15. Londono-Vallejo JA, Stragier P: Cell-cell signaling pathway activating a developmental transcription factor in Bacillus subtilis. Genes Dev 1995, 9:503–508. 16. Kirchman PA, DeGrazia H, Kellner EM, Moran C Jr: Foresporespecific disappearance of the sigma-factor antagonist spoIIAB: implications for its role in determination of cell fate in Bacillus subtilis. Mol Microbiol 1993, 8:663–671. 17. Beall B, Lutkenhaus J: FtsZ in Bacillus subtilis is required for vegetative septation and for asymmetric septation during sporulation. Genes Dev 1991, 5:447–455.