Function of the ATP-dependent Metalloprotease FtsH during sporulation in Bacillus subtilis

Function of the ATP-dependent metalloprotease FtsH during sporulation in Bacillus subtilis Dissertation zur Erlangung des Grades eines Doktors der Naturwissenschaften -Dr. rer. nat.- der Fakultät für Biologie, Chemie und Geowissenschaften der Universität Bayreuth vorgelegt von Ai Thi Thuy Le Bayreuth 2008 Die vorliegende Arbeit wurde in der Zeit von Januar 2005 bis Oktober 2007 an der Universität Bayreuth am Lehrstuhl für Genetik unter der Betreuung von Prof. Dr. Wolfgang Schumann angefertigt. Vollständiger Abdruck der von der Fakultät für Biologie, Chemie und Geowissenschaften der Universität Bayreuth genehmigten Dissertation zur Erlangung des akademischen Grades eines Doktors der Naturwissenschaften (Dr. rer. nat.). Promotionsgesuch eingereicht am: 09.01.2008 Tag des wissenschaftlichen Kolloquiums: 14.03.2008 1. Gutachter: Prof. Dr. Wolfgang Schumann 2. Gutachter: Prof. Dr. Franz X. Schmid i Acknowledgements This thesis was carried out in the group of Prof. Dr. Wolfgang Schumann at the Department of Genetics, University of Bayreuth. I would like to express my special gratitude to my supervisor Prof. Dr. Wolfgang Schumann who shared with me a lot of his expertise and research insight. I appreciate his excellent direction and useful comments on this work. I would like to thank PD. Dr. Thomas Wiegert for his helpful comments throughout this process. Thanks to Karin Angermann for the technical discussions and her kindness. I wish to thank all my friends and my country-mates in Bayreuth for inestimable friendship. Special thanks to Hue Nguyen, one of my great friends, for her assistance on the ministerial computer. My most sincere thanks for the financial support of the Deutsche Forschungsgemeinschaft (Schu 414/20-2) and the EU (LSGH-CT2004-503468). Last but not least, I am forever indebted to my parents, my husband and my daughter for their understanding, endless patience and encouragement when it was most required. ii Index 1. Summary 1 Zusammenfassung 3 2. Introduction 5 2.1. FtsH - an ATP-dependent membrane protease 5 2.1.1. Structure and function 5 2.1.2. Alteration of FtsH has pleiotropic effects on cell physiology 6 2.1.3. Mechanism of degradation by FtsH 10 2.1.4. Cytoplasmic protein degradation by FtsH 11 2.1.5. Membrane protein degradation by FtsH 12 2.1.6. Recognition strategy for FtsH 13 2.2. The sporulation in Bacillus subtilis 13 2.2.1. Morphological stages 14 2.2.2. Spo0A 15 2.2.2.1. Spo0A is a master regulator of sporulation initiation 15 2.2.2.2. Activation of Spo0A - The phosphorelay 17 2.2.2.3. The auto-stimulation of Spo0A 18 2.2.3. Regulation of the phosphate flow 19 2.2.3.1. Regulation of kinase 19 2.2.3.2. Regulation of response regulators 20 2.2.3.2.1. The Rap family of phosphatases 20 2.2.3.2.2. The Spo0E family of phosphatases 21 2.2.3.3. Control of the Rap phosphatase 22 2.2.4. Bistable outcoming in sporulation 23 2.2.5. SpoVM - an essential morphogenetic protein 25 2.3. Construction of a cold-inducible expression system for recombinant proteins in the B. subtilis 26 iii 3. Results and Discussion 28 Part A 28 3.1. Influence of FtsH on the synthesis and activation of the master regulator Spo0A 28 3.1.1. Synthesis of Spo0A is partly impaired in the ftsH knockout 28 3.1.2. Activation of Spo0A does not occur in the ftsH knockout 29 3.1.3. After artificial induction of active Spo0A, cells are able to sporulate in the absence of FtsH 30 3.2. The level of Spo0A is restored in the ftsH spo0E knockout, but not in the ftsH rap knockout 30 3.3. The activation of Spo0A in the phosphorelay phosphatase deletion strains 31 3.4. Sporulation frequency increasing in the phosphorelay phosphatase deletion strains 31 3.5. Spo0E is a target of FtsH 32 3.6. The C-terminal end of Spo0E is responsible for degradation by FtsH 33 3.7. The 25 C-terminal amino acids of Spo0E is considered as the proteolytic tag for FtsH-mediated degradation 34 Part B 35 3.1. Assumption that FtsH interacts with the small sporulation peptide SpoVM 35 3.2. Screening potential subtrates for FtsH late during sporulation 36 3.3. Transcription of spoVM during sporulation 36 3.4. The 5´untranslated region of spoVM acts as a negative regulator of its own translation 37 Part C 38 4. References 39 iv 5. Own Contribution 53 Appendix : Own publications Part A 54 Part B 85 Part C 101 6. Abbreviations 108 Erklärung I. Summary 1 I. Summary The analysis of the function of the ftsH gene of Bacillus subtilis started about ten years ago. It was shown at that time that an ftsH knockout was viable, but exhibited a pleiotropic phenotype. Cells are sensitive to salt and heat shock, exhibit filamentous growth, are difficult to be transformed and are almost unable to sporulate. Despite the severe phenotype caused by the absence of the ftsH gene, the precise functions of this protein remained unclear. This PhD thesis presents data to elucidate the function of ftsH during sporulation. Furthermore, it describes the construction of a cold-inducible expression system. The major finding of this thesis is that the FtsH protease interferes with the synthesis and/or phosphorylation of Spo0A, the master regulator during initiation of sporulation called phase 0. In the ftsH knockout, the amount of Spo0A is greatly reduced, and the small amounts present are inactive. When the wild-type spo0A allele was replaced by an IPTG-inducible allele coding for mutant Spo0A protein being fully active in the absence of phosphorylation (Spo0A-Sad67), spores were formed at a normal rate in an ftsH knockout. Again, this result indicates that FtsH is clearly involved in the formation of active Spo0A and that this protease is only essential during stage 0 of sporulation. To become active, Spo0A needs to be phosphorylated by the multi-component system called phosphorelay. Since no active Spo0A is present in an ftsH knockout, it was hypothesized that FtsH has to degrade one or more negative regulator(s) either preventing the phosphorylation of Spo0A or/and being involving in its rapid dephosphorylation. The further analysis focused on four antagonists of the phosphorelay, three Rap phosphatases being involved in the dephosphorylation of Spo0F~P, and Spo0E which targets Spo0A~P. When a null allele in any one of them was combined with the ftsH knockout, the wild-type amount of Spo0A was restored only in the case of the ftsH spo0E knockout and the sporulation frequency was increased by two to three orders of magnitude in all double knockouts, but remained below 1%. Since overexpression of Spo0E reduces the sporulation frequency and removal of the gene from the genome has an opposite effect, a direct interaction between FtsH and Spo0E was envisaged. In vitro proteolysis assays with purified GST-FtsH and GST-Spo0E showed that Spo0E is indeed a target of FtsH. In contrast, the two homologs of Spo0E, YisI and YnzD, remained stable upon incubation with FtsH. Since all three proteins are distinguished by a Cterminal extension of about 25 amino acids present in Spo0E, but not in the two other phosphatases, these additional amino acids could serve as a target for FtsH. When two mutant versions of Spo0E, Spo0E94 and Spo0E11, with truncated C-terminal ends were I. Summary 2 analyzed, they turned out to be stable in the presence of FtsH. When the C-terminal 25 amino acids was transferred to YnzD, this fusion protein became unstable when incubated with FtsH. In conclusion, the C-terminal end of Spo0E confers instability to this enzyme. Since a spo0E knockout in a wild-type background does not result in a sporulation frequency close to 100% and a combination of a spo0E and an ftsH knockout raises the sporulation frequency only close to 1%, it can be concluded that there are additional targets for FtsH interfering with the synthesis of active Spo0A. Moreover, it is likely that FtsH also exerts a function late during sporulation. It could be shown that SpoVM, a small peptide essential for spore morphogenesis, inhibits the proteolytic activity of the B. subtilis FtsH protease in vitro. It can be inferred that SpoVM also inhibits activity of FtsH during sporulation, and in the absence of SpoVM, FtsH will degrade at least one protein essential for successful completion of sporulation. When the intracellular proteomes of spoVM+ and spoVMcells were compared, a total of 83 proteins were identified being either completely absent or present in reduced amounts in the absence of the peptide. Analysis of the expression of the spoVM gene revealed that cells started to synthesize the spoVM transcript at stage 2 while the SpoVM peptide accumulated at stage 4. The 5´ untranslated region of the spoVM transcript has been identified to act as a negative regulator of its own transcription or translation. Furthermore, a cold-inducible expression system has been constructed allowing intraand extracellular production of recombinant proteins. This expression system makes use of a two-component signal transduction system, which senses changes in the fluidity of the cytoplasmic membrane. I. Zusammenfassung 3 I. Zusammenfassung Die Analyse der Funktion des ftsH-Gens von Bacillus subtilis begann vor etwa 10 Jahren. Damals konnte gezeigt werden, dass eine ftsH Knockout-Mutante lebensfähig ist, aber über einen pleiotropen Phänotyp verfügt. Die Zellen sind Salzund Hitzesensitiv, wachsen filamentös, sind schwierig zu transformieren und zeigen eine stark reduzierte Sporulationsfrequenz. Trotz dieser gravierenden Phänotypen blieb die Funktion von ftsH bislang im Dunkeln. Diese Doktorarbeit präsentiert Daten, die einige Funktionen von ftsH während der Sporulation aufdecken. Außerdem wird die Konstruktion eines Kälte-induzierbaren Expressionssystems beschrieben. Das besondere Ergebnis dieser Dissertation ist der Befund, dass die FtsH Protease mit der Synthese und/oder der Phosphorylierung von Spo0A, dem MasterRegulator während der Initiation, der Phase 0, interferiert. In einer ftsH-Knockout ist die Menge an Spo0A signifikant reduziert und die geringen Mengen sind inaktiv. Wenn das Wildtyp-Allel von spo0A durch ein IPTG-induzierbares Allel ersetzt wurde, welches für ein mutantes Protein codiert, das auch in Abwesenheit von Phosphorylierung voll aktiv ist (Spo0A-Sad67), dann wurde eine Sporulationsfrequenz gemessen, die der von Wildtyp-Zellen entsprach. Aus diesem Ergebnis ist zu folgern, dass ftsH nur während der Phase 0 essentiell ist. Um Aktivität zu erlangen, muss Spo0A phosphoryliert werde, und dies geschieht durch ein Phosphorelay. Da in einer ftsH-Mutante kein aktives Spo0A nachweisbar ist, ist zu vermuten, dass FtsH einen oder mehrere negative Regulatoren abbauen muss, die entweder die Phosphorylierung von Spo0A verhindern oder an einer schnellen Dephosphorylierung beteiligt sind. Die weitere Analyse konzentrierte sich auf vier verschiedene Antagonisten des Phosphorelays, drei Rap Phosphatasen, die Spo0F~P dephosphorylieren und Spo0E, welche Spo0A~P dephosphoryliert. Wenn NullAllele der vier Phosphatasen mit einer ftsH-Knockout kombiniert wurden, dann wurde nur im Fall von Δspo0E Wildtyp-Mengen an Spo0A detektiert. Die Sporulationsfrequenz wurde in allen vier Stämmen um 2-3 Größenordnungen erhöht gegenüber der ftsHKnockout, blieb aber in allen Fällen unter 1%. Da eine Überexpression von Spo0E die Sporulationsfrequenz reduziert und ein spo0E-Knockout den gegenteiligen Effekt hat, wurde eine direkte Interaktion zwischen FtsH und Spo0E in Betracht gezogen. In vitro Proteolysetests mit gereinigtem GST-FtsH und GST-Spo0E ergaben, dass Spo0E abgebaut wird. Im Gegensatz dazu erwiesen sich zwei Homologe von Spo0E, YisI und YnzD, als stabil. Da alle drei Phosphatasen sich nur in ihren N-Termini unterscheiden und nur Spo0E einen um etwa 25 Aminosäurereste verlängerten C-Terminus enthält, war zu vermuten, dass dieser Anhang von FtsH als Target erkannt wird. Wenn zwei 2. Introduction 10 Spo0A is responsible for the failure to sporulate in a ftsH null mutant strain (Deuerling, et al., 1997). Hence, the first function of FtsH proposed is related to the initiation stages of sporulation. Another study could show that a spoVM mutant is blocked in expression of sigmaK-dependent genes (as expected for a mutant blocked at stage II-III). Extragenic suppressors of spoVM mutants mapped in the ftsH gene and it was evident that the 26amino-acid SpoVM peptide could inhibit the proteolytic activity of FtsH (Cutting et al., 1997). 2.1.3. Mechanism of substrate degradation by FtsH All characterized ATP-dependent proteases exhibit multimeric ring-like structures in which the proteolytic active site is buried within a central cavity. A consequence of this configuration is that the active site is accessible only to unfolded and extended polypeptides. According to the prevailing model, proteases utilize specific degradation signals to recognize and bind substrates. Then, powered by ATP hydrolysis, they unfold these substrates and translocate them into the proteolytic chamber (Sauer et al., 2004). Within this chamber, the protein is hydrolyzed to small peptides, which are released into the cytoplasm. The evidence supporting this model is derived primarily from studies of the bi-partite cytoplasmic ClpAP and ClpXP proteases (Weber-Ban et al., 1999; Kim et al., 2000; Hoskins et al., 2000). These Clp-family proteases exhibit robust unfoldase activity, enabling them to degrade substrate proteins with intrinsic thermostabilities (Kenniston et al., 2003). A recent work indicates that FtsH operates through a different mode of degradation than these cytoplasmic proteases (Herman et al., 2003). In contrast to the Clp-family proteases, FtsH lacks a robust unfoldase activity which would be necessary to thread a substrate into the proteolytic cavity. It was proposed that the weak unfoldase activity allows the cell to control the rate at which FtsH degrades regulatory targets like σ32 and LpxC (Herman et al., 2003). Under appropriate conditions, it is likely that FtsH employs co-factors that accelerate the unfolding and the degradation of targets. Thus, FtsH can sense the folding state of regulatory substrates within the physiological range of the cell. Apart from its role in regulated proteolysis, it was speculated that the weak unfoldase activity employed by FtsH may be a central feature of membrane protein degradation in general (Herman et al., 2003). FtsH might be able to sense the folded state of proteins, and degrade only those that exhibit low intrinsic thermostability (e.g. unassembled or misfolded membrane proteins), independently of a specific degradation signal. 2. Introduction 11 2.1.4. Cytoplasmic protein degradation by FtsH Most of the FtsH substrates thus far characterized are from E. coli in which soluble substrates are the most part naturally short-lived. The bacteriophage lambda protein CII is short-lived transcription factor for genes required for the establishment of lysogenization. It is rapidly degraded by FtsH in vivo and in vitro (Kihara et al., 1997; Shotland et al., 1997; Shotland et al., 2000). The cIII gene product, a small membraneinteracting protein with an amphiphilic α-helix region, is slowly degraded by FtsH in vivo (Herman et al., 1997), and it antagonizes the FtsH-catalyzed in vitro proteolysis of CII (Shotland et al., 2000). Thus, the balance among these gene products is an important factor for the decision between the lytic growth and lysogenization/integration of the infecting λ genome (Hoyt et al., 1982). The heat shock sigma factor σ32 is rapidly degraded in the absence of a heat shock or other stress, with an half-life that is affected by the level of FtsH (Herman et al., 1995b; Tomoyasu et al., 1995) as well as by other ATP-dependent proteases (Kanemori et al., 1997). In a purified reaction system, FtsH degrades σ32 in Zn2+- and ATPdependent manners (Okuno et al., 2004, Toyomasu et al., 1995). FtsH may also contribute to the degradation of another short-lived transcription factor, SoxS (Griffith et al., 2004). LpxC is the most important FtsH substrate because degradation of LpxC renders FtsH essential. Since the same reaction precursor (R-3-hydroxymyristoyl-ACP) is used by the lpxC (envA) gene-encoded deacetylase for the biosynthesis of lipid A, a LPS component, and by the fabA gene-encoded dehydrase for fatty acid biosynthesis, the balance of these enzymes is important to maintain a proper LPS/phospholipids ratio in E. coli cells. The LpxC deacetylase is short-lived (half-life of 4 min) owing to FtsH-catalyzed degradation, dysfunction of which results in the lethal over-accumulation of LPS (Ogura et al., 1999). LpxC is a globular protein. Basing on the structure of LpxC from Aquifex aeolicus has been solved (Coggins et al., 2003; Whittington et al., 2003) suggesting that the final 14 residues are not structured. As the E. coli protein contains an extension of eleven residues, an unstructured C-terminus of 25 amino acids was predicted. In vivo degradation LpxC by FtsH was evident and a detailed mutational analysis revealed six non-polar residues in the C-terminus of LpxC that are critical for degradation (Führer et al., 2007). The SsrA-tag is a short sequence that is appended to the C-terminus of truncated proteins on stalled ribosomes to promote their proteolysis (Keiler et al., 1996). In E. coli, degradation occurs mainly through the ClpAP/XP proteases (Keiler et al., 1996; Gottesman et al., 1998), which recognize specific residues within the SsrA-tag (Flynn et 2. Introduction 12 al., 2001). In addition, FtsH also recognizes the SsrA-tag and efficiently degrades the λCI-SsrA model substrate (Herman et al., 1998) though the λCI N-terminal domain is a stable cytosolic protein. It appears that the SsrA-tag confers a relatively unspecific degradation signal to the protein at which it is attached (Herman et al., 1998). It remains to be elucidated how the different proteases, ClpAP/XP and FtsH, recognize the same degradation tag. 2.1.5. Membrane protein degradation by FtsH The major housekeeping function proposed for FtsH is the rapid removal of harmful membrane protein subunits when they fail to integrate into functional complexes (Ito et al., 2005). A study of membrane degradation has been carried out with two membrane substrates, the translocase subunit SecY, and YccA, a membrane protein of unknown function. SecY is one of the major components of the translocation apparatus. It becomes a substrate for FtsH mediated degradation when it fails to assemble into a complex with its partner, SecE and SecG (Kihara et al., 1995). This can occur either when SecY is overexpressed, or when the SecY-SecE interaction is weakened through a mutation (Chiba et al., 2000). YccA is inherently unstable, and it is unknown whether YccA forms any higher order complexes (Kihara et al., 1998). YccA is degraded by FtsH with protease recognition through its cytoplasmic tail. However, YccA associates with FtsH even when its degradation signal is absent (Kihara et al., 1998). FtsH can also degrade integral membrane proteins starting from their cytoplasmic ends, provided that there is an unfolded cytoplasmic tail, of any sequence composition, that is at least 20 amino acids in length (Chiba et al., 2002). Such degradation can begin at the Nor Cterminus and proceeds sequentially to the other end, apparently pulling the protein through the membrane, as the periplasmic domains are degraded in an FtsH-dependent manner (Chiba et al., 2002). Because the proteolytic active site of FtsH resides in the cytoplasm, substrates must be extracted from the lipid bilayer before entering the proteolytic cavity, a process called dislocation (Kihara et al., 1999). Dislocation of membrane proteins raises the intriguing question of how hydrophilic domains traverse the plasma membrane. Numerous studies on protein transport across cellular membranes have demonstrated the requirement of proteinaceous hydrophilic pores for the translocation process (Prakash et al., 2004). More specifically, the turnover of membrane proteins in the yeast endoplasmic reticulum, mediated by the 26S proteosome, involves the retrograde translocation of substrates into the cytosol via the Sec61 translocase machinery (Stirling et al., 2006; Groll et al., 2005). Sec61 is the yeast homolog of E. coli SecY. A recent in vitro study, using purified FtsH and YccA in reconstituted proteoliposomes, showed that 2. Introduction 13 FtsH is able to dislocate and degrade YccA in the absence of other components (Akiyama et al., 2003). 2.1.6. Recognition strategy for FtsH The ability of FtsH to discriminate between correct and incorrect protein substrates is critical to both its housekeeping and regulatory functions. FtsH recognizes and degrades cytoplasmic proteins that contain C-terminal non-polar tails (Herman et al., 1998). This tail-specific recognition is physiologically relevant, and is used for degrading the λCII activator, SsrA-tagged proteins and the LpxC deacetylase (Herman et al., 1998; Kobiler et al., 2002). The SsrA-tag consists of 11 residues that is appended to the Cterminus of truncated proteins on stalled ribosomes to promote their proteolysis (Keiler et al., 1996). The aforementioned tail-specific recognition, however, the C-terminal region is not important for degradation in case of the heat shock factor σ32 (Toyomasu et al., 2001); Instead, an internal region may be important for the FtsH-mediated degradation of this protein (Bertani et al., 2001). A mode of σ32 proteolysis is also described, in which it is degraded by FtsH in the Nto Cterminus direction (Okuno et al., 2004; Obrist et al., 2007). It is clear that FtsH readily degrades functional native protein substrates that are susceptible to the proteolytic reaction of FtsH in initiation-signal-dependent manners. This mode of proteolysis likely involves sequential substrate unfolding that could be initiated at an initiation signal and then propagated along the polypeptide chain (Ito et al., 2005). In addition to specific sequence recognition, the substrate protein thermostability plays an important role in the decision to degrade it; the more thermostable the protein, the less likely that FtsH will degrade it, even when it carries a good recognition tag. This is due to the fact that FtsH does not possess a robust unfoldase activity, which would be necessary to thread a thermostable substrate into the proteolytic cavity. Lacking of a robust unfoldase allows FtsH to carry out a second regulatory step in the decision to degrade a protein by enabling FtsH to sense the folding state of proteins within the physiological range of the cell, and degrade only those that display low thermostability (Herman et al., 2003). This second regulatory step in substrate selection may be crucial for the degradation of regulatory proteins and membrane proteins. σ32, a natural substrate of FtsH, indeed contains region of low thermodynamic stability (Ito et al., 2005). 2.2. The sporulation in Bacillus subtilis Under conditions of nutrient deprivation, cells of B. subtilis can undergo a process of development that leads to the formation of dormant, environmentally resistant spores. Sporulation takes approximately six to eight hours and involves extensive changes in 2. Introduction 14 gene expression and morphology (Errington, 2003; Hilbert et al., 2004; Piggot et al., 2004). The hallmark of endospore formation is an asymmetric cell division that produces two cell types: a larger cell called the mother cell, and a smaller cell called the forespore. 2.2.1. Morphological stages The series of complex morphological changes that occur during the sporulation process in B. subtilis have been extensively studied (Fig. 4). Entry into sporulation is characterized by the formation of a so-called axial filament in which two chromosomes from the last round of DNA replication become aligned across the long axis of the cell. Next, a septum is formed at an extreme polar position. This partitions the developing cell (hereafter referred to as the sporangium) into large and small compartments known as the mother cell and the forespore (or prespore), respectively, where each receives a chromosome (Stragier and Losick, 1996). Initially, the large and the small compartments lie side-by-side, but in the next stage of development the forespore becomes engulfed by the mother cell. During engulfment, the membrane on the mother cell face of the polar septum migrates around the membrane surrounding the forespore and eventually the forespore is completely pinched off as a free protoplast within the mother cell, such that the sporangium becomes a cellwithin-a-cell (for this reason spores of Bacillus and related genera are more properly known as endospores) (Stragier and Losick, 1996). In subsequent morphogenesis, the forespore produces large amounts of a family of small acid-soluble proteins known as SASP. Some of these proteins bind to and coat the forespore chromosome, packaging it into a doughnut-like structure and conferring on it resistance to ultraviolet radiation. Meanwhile, in the intermembrane space between the forespore and mother cell, a thin layer of peptidoglycan known as the germ cell wall is produced on the surface of the forespore membrane. This is followed by the synthesis of a thick layer of peptidoglycan known as the cortex, which is thought to be involved in attaining or maintaining the dehydrated and heat-resistant state of the spore. The mother cell produces a proteinaceous coat that assembles on the outside surface of the mothercell membrane around the forespore. The coat consists of a lamellar inner layer and an electron-dense outer layer and provides a thick, protective barrier that encases the mature spore. Eventually, after about 6–8 hours of development, when maturation is complete, the fully ripened spore is liberated by lysis of the mother cell. Thus, the mother cell is mortal in that it undergoes programmed cell death, whereas the forespore is immortal in that it becomes the spore and gives rise to subsequent progeny (Stragier and Losick, 1996). 2. Introduction 15 FIG. 4. Schematic representation of the stages of sporulation in Bacillus subtilis This sequence of morphological events is divided into different stages: Stage 0 represents cells that have not entered the sporulation pathway. Stage I represents cells that have entered the pathway and have formed an axial filament; Stage II and III refer to sporangia that have reached the stages of polar septation and engulfment, respectively. Synthesis of a distinctive form of peptidoglycan between the membranes surrounding the prespore is defined as stage IV. Deposition o f spore coat around the prespore is defined as stage V. Stage VI is maturation, when the spore acquires its full resistant properties spore. Stage VII represents lysis o f the mother cell and release of the mature spore (Hilbert and Piggot, 2004) 0Growth IAxial Filamentation II Asymmetric Septum IIIEngulfment IVCortex Synthesis VCoat Synthesis VIIMother Cell Lysis V IMaturation 2.2.2. Spo0A 2.2.2.1. Spo0A is a master regulator of sporulation initiation The master regulator for entry into sporulation in B. subtilis is the DNA-binding protein Spo0A, which is a member of the response regulator family of transcription factors (Perego and Hoch, 2002). The activation of this key transcriptional regulatory protein occurs through environmental and physiological signals, triggered by nutrient depletion and cell density. Activation of Spo0A proceeds through several phases. Initial activation at the end of exponential growth leads to the ‘transition state’, which is associated with such phenomena as protease production, motility, competence for transformation (Sonenshein, 2000), biofilm formation (Branda et al., 2001; Hamon and 2. Introduction 16 Lazazzera, 2001) and even cannibalism (Gonzalez-Pastor et al., 2003) and spore formation which is thought to require increased the phosphorylation of Spo0A. The molecular details of the interaction of Spo0A with its target DNA, the ‘Spo0A box’, a consensus 7-bp sequence (5′-TGNCGAA-3′, with a preference for N = T), have now been analyzed with a crystal structure (Zhao et al., 2002). Spo0A has been found to influence, directly or indirectly, the expression of over 500 genes during the early stages of development. An approach in combination with transcriptional profiling using gene microarrays, gel electrophoretic mobility shift assays, using the DNA-binding domain of Spo0A, and bioinformatics enabled to assign a total 121 genes, which are organized as 30 single-gene units and 24 operons, are likely to be under the direct control of Spo0A. About one-third of these genes are activated and the remainder are repressed (Molle et al., 2003). Among the identified members of the regulon where transcription was stimulated by Spo0A are genes for metabolic enzymes and genes for efflux pumps (Molle et al., 2003). Among the members where transcription that was inhibited by Spo0A are genes encoding components of the DNA replication machinery and genes that govern flagellum biosynthesis and chemotaxis. During stage 0 of sporulation, the active form of Spo0A (Spo0A~P) acts as a repressor of certain vegetatively expressed genes (e.g. abrB) (Perego et al., 1988; Strauch et al., 1989; Strauch and Hoch, 1993; Fujita and Sadaie, 1998) and an activator of genes directly involved in sporulation (Piggot and Losick, 2002). So far, a total of 10 transcription units which are organized in six single-gene units (abrB, kinA, kinC, spo0A, spo0F and spoIIE) and four operons (dlt, sin, spoIIA and spoIIG) are controlled by Spo0A~P. Among the genes activated by Spo0A~P are those involved in remodeling the sister chromosomes of the sporulating cell into an ‘axial filament’ (Pogliano et al., 2002; Ben-Yehuda et al., 2003) and in the formation of a polar septum that divides the developing cell into a small forespore compartment and a large mother cell compartment (Levin and Losick, 1996; Ben-Yehuda and Losick, 2002). Spo0A~P is also responsible for activating genes that lead to the appearance of the cell-specific regulatory proteins σF and σE which act in the forespore and the mother cell, respectively (Stragier and Losick, 1996; Piggot and Losick, 2002). Recent work indicates that Spo0A~P continues to function after the polar septum is formed, when it accumulates to high levels and directs transcription in the mother cell (Fujita and Losick, 2003). Clearly, Spo0A has a profound effect on the global pattern of gene expression (Molle et al., 2003). Importantly, cells require a high threshold of active Spo0A to initiate sporulation (Fujita et al., 2005). Mutations within the phosphorelay, leading to lower concentrations of intracellular Spo0A~P, caused a smaller population of cells initiating sporulation 2. Introduction 17 (Molle, et al., 2003). There are four categories of genes within the Spo0A regulon that respond to different thresholds of Spo0A as follows: (i) those that require a high level of Spo0A to become activated, (ii) those that required a high level of Spo0A to be repressed, (iii) those that were activated at a low level of the regulator, and (iv) those that were repressed at a low dose of the regulator. Genes that required a high dose of Spo0A to be activated were found to have low binding constants for Spo0A~P. Some genes that were turned on at a low dose of Spo0A either had a high binding constant for the regulatory protein or were activated by an indirect mechanism involving Spo0A-mediated relief of repression by repressor protein AbrB (Fujita et al., 2005). Moreover, Spo0A~P also sets in motion several positive and negative regulatory loops that govern the rate of expression of spo0A and other relay and phosphatase genes. Detailed accounts of the functioning of the phosphorelay, the nature of these regulatory loops, and the modulation of phosphate flow by specific phosphatase are indicated below. 2.2.2.2. Activation of Spo0A - The phosphorelay The activity of Spo0A is governed by a multicomponent phosphorelay - an extended version of the typical two-component system, which consists of five histidine autokinases (KinA, KinB, KinC, KinD and KinE) and two phosphorelay proteins (Spo0F and Spo0B) (Jiang et al., 2000) (Fig. 5). Differential signals activate multiple histidine kinases to autophosphorylate and then transfer their phosphoryl group to the intermediate response regulator Spo0F (Burbulys et al., 1991; Jiang et al., 2000). Spo0F∼P is the substrate for a phosphotransferase, Spo0B, which transfers the phosphoryl group to the Spo0A response regulator and transcription factor (Burbulys et al., 1991). Signal integration is the responsibility of the phosphorelay, whose structural complexity reflects the requirement for precise coordination of numerous cellular events. The multicomponent structure of the phosphorelay provides multiple entries for regulatory signals affecting the final goal of producing the appropriate level of Spo0A~P. These regulatory mechanisms are exerted both on the level of transcription of the phosphorelay components and on their enzymatic activity. KinA is the primary kinase in the phosphorelay and it has the major role at the onset of spore formation (Stephenson and Hoch, 2001). It has been demonstrated that the fraction of cells that initiate sporulation is decreased in a kinA mutant background (Chung et al., 1994). The most amino-terminal end of the domains in KinA is important for spore formation and has been shown to bind ATP, but is unlikely to be regulated directly by ATP levels (Stephenson and Hoch, 2001). 2. Introduction 18 2.2.2.3. The auto-stimulation of Spo0A The activity of Spo0A is subject to several auto-stimulatory loops (Strauch et al., 1992; Strauch et al., 1993; Fujita and Sadaie, 1998). These loops involve transcription of spo0A and phosphorylation of Spo0A. Transcription of spo0A is directly activated by Spo0A~P (Strauch et al., 1992) and indirectly activated by induced expression of spo0H. First of all, Spo0A~P represses the expression of abrB, a gene encoding a transcriptional FIG. 5. The phosphorelay signal transduction system for sporulation initiation. In the phosphorelay, two cytoplasmic kinases (KinA and KinE) and three membranebound kinases (KinB, KinC, KinD) phosphorylate the Spo0F (0F) response regulator in response to differential signals. Spo0F∼P transfers the phosphoryl group to the Spo0B (0B) phosphotransferase that, in turn, transfers it to the Spo0A (0A) response regulator and transcription factor for sporulation initiation. KapB is a lipoprotein essential for KinB activity. The Rap phosphatases dephosphorylate the Spo0F∼P intermediate while Spo0E, YisI and YnzD dephosphorylate Spo0A∼P. Transcription of the phosphatase coding genes is activated by physiological conditions antithetical to sporulation such as growth and competence to DNA transformation (Perego, 2001). 2. Introduction 19 regulator that inhibits various stationary phase processes (Robertson et al., 1989). This results in an indirect autostimulatory loop actived via the transcriptional regulator AbrB. During exponential growth, AbrB represses various stationary phase processes, including the transcription of genes required for sporulation (e.g. kinA) (Strauch et al., 1989). Importantly, AbrB represses gene expression of the alternative RNA polymerase sigma factor σH that recognizes an alternative promoter upstream of spo0A, and in addition, activates genes required for phosphorylation of Spo0A such as kinA and spo0F (Predich et al., 1992). Thus, when Spo0A is phosphorylated, alleviation of AbrB repression by Spo0A~P stimulates both transcription of spo0A and indirectly phosphorylation of Spo0A. A simplified scheme of the autostimulation of Spo0A is described in Fig. 6. FIG. 6. Simplified schematic representation of the autostimulatory loop involving transcription and activation of Spo0A. Perpendiculars and arrows represent the negative and positive regulations, respectively (Smits et al., 2006). 2.2.3. Regulation of the phosphate flow 2.2.3.1. Regulation of kinase The discovery that the level of Spo0A~P is crucial in determining the cell fate led to the discovery of a series of mechanisms that modulate the flux of phosphate in the phosphorelay in response to specific signals. The first level of control is on the histidine kinases, KinA and KinB. Although mechanisms of activation of these kinases are predictable, a negative regulator of phosphate input has been described as an inhibitor of the kinase activity of KinA. KipI is a potent inhibitor of the autophosphorylation reaction of kinase A but does not inhibit phosphate transfer to the Spo0F response regulator once 2. Introduction 26 assembly of the coat around the forespore. GFP-SpoIVA has been shown to surround the forespore in a shell-like structure that is believed to serve as a basement layer for the coat (van Ooij and Losick, 2003; Ramamurthi et al., 2006). Genetic, biochemical and cytological evidence indicates that this mutual dependence is mediated in part by contact between an amino acid side-chain located near the extreme C-terminus of SpoIVA and an amino acid side-chain on the hydrophilic face of the SpoVM helix. SpoVM serving as a membrane anchor, it adheres to the outer forespore membrane via the hydrophobic face of the helix first and then tethers SpoIVA (Ramamurthi et al., 2006). 2.3. Construction of a cold-inducible expression system for recombinant proteins in B. subtilis One of the major drawbacks during high-level production of recombinant proteins in bacteria is the inability of many proteins to reach their native conformation. Under conditions of overproduction, proteins tend to accumulate within refractile aggregates designated inclusion bodies (Mogk et al., 2002). Several strategies have been described to reduce the formation of inclusion bodies including cultivation of the cells at low temperatures (Thomas and Baneyx, 1996). Besides reducing formation of inclusion bodies, low-temperature expression lowers the degradation of proteolytically sensitive proteins (Emerick et al., 1984;Chesshyre and Hipkiss, 1989). To ensure high level production of recombinant proteins at low temperature, two different strategies can be used: (i) Fusion of the coding region of the protein of interest to an inducible promoter followed by growth at a low temperature, e.g. 20°C. (ii) Fusion of the gene of interest to a cold-inducible promoter, growth of the expression strain at the physiological temperature first followed by induction at the appropriate low temperature. Such cold-inducible expression systems have already been developed for E. coli (Mujacic et al., 1999;Qing et al., 2004). Here, the promoter region of the cold-inducible cspA (for cold-shock proteins A) gene has been used. This gene is expressed at all temperatures, but the transcript is extremely unstable at physiological temperatures and greatly stabilized after a temperature downshift to 20°C (Fang et al., 1997). A sudden decrease in temperature affects membrane fluidity, and to restore its fluidity B. subtilis cells increase the level of a membrane-bound desaturase (Aguilar et al., 1998). This enzyme (called Δ5-Des) is encoded by the des gene and catalyzes the introduction of a cis double bond at the Δ5 position of a wide variety of fatty acids (Aguilar et al., 1998). While the des transcript is barely detectable at 37°C, its synthesis is transiently induced upon a temperature downshift (Aguilar et al., 1999). Expression of the des gene does not depend on de novo protein synthesis, but on a two-component signal transduction system which consists of the sensor kinase DesK and the response 2. Introduction 27 regulator DesR (Aguilar et al., 2001). It is assumed that the transmembrane domain of the kinase senses a temperature downshift through changes in the physical state of the cytoplasmic membrane (Hunger et al., 2004). The C-terminal kinase domain of DesK undergoes autophosphorylation, and the phosphoryl group is then transferred to the response regulator DesR. Phosphorylated DesR binds to two adjacent DNA-binding sites leading to the recruitment of RNA polymerase to the des promoter and activation of transcription (Cybulski et al., 2004). The Δ5-desaturase directly introduces double bonds into membrane lipids leading to a return to the original fluidity of the membrane. This is sensed by DesK which changes from a kinase to phosphatase activity leading to a dephosphorylation of DesR with a concomitant turn off of the des gene (Mansilla and De Mendoza, 2005). Based on these data, a cold-inducible expression system for B. subtilis was developed. 3. Results and Discussion 28 3. Results and Discussion Part A: Function of FtsH during initiation of sporulation 3.1. Influence of FtsH on the synthesis and activation of the master regulator Spo0A 3.1.1. Synthesis of Spo0A is partly impaired in the ftsH knockout Based on previous studies, ftsH insertion mutants were completely deficient in sporulation where the sporulation frequency of the mutant was less than 10-8 as compared to the wild-type strain (Deuerling et al., 1997). The stage of the sporulation program, which was impaired in the ftsH mutants had been determined based on transcriptional fusions of the lacZ reporter to genes expressed during two different sporulation stages: spoIIA-lacZ and spoIID-lacZ representing stage 0 and stage II, respectively. Both transcriptional fusions were induced during the transition phase in the wild-type strain, but their induction was completely abolished in the ftsH mutants (Deuerling et al., 1997). These results clearly demonstrated that ftsH is required at an early stage of the sporulation process. Spo0A is a master regulator of the initiation of sporulation (Errington, 1993). Here, I asked whether ftsH interferes with the synthesis or activity of Spo0A. An ftsH null mutant, in which the ftsH gene was completely replaced by an erm-cassette (Wehrl et al., 2000) was used throughout this study. The amount of Spo0A was measured by Western blotting (Fig. 1, Part A). In the wild-type strain, Spo0A started to be present from stage 0 on and continued to be produced to at least stage 3. On the contrary, Spo0A was present in greatly reduced amounts in the ftsH knockout. The spo0A gene is transcribed from the two promoters Pv (for vegetative) and Ps (for stationary). During exponential growth, the transcription of spo0A is under the σAdependent Pv promoter and occurs at a low level. Under sporulation conditions, so far unknown metabolic signals trigger the phosphorylation of Spo0A through the phosphorelay. The phosphorylated Spo0A (Spo0A~P) directly activates the Ps promoter through binding to a Spo0A box located adjacent to this promoter which is recognized by σH. It indirectly stimulates its own transcription by inhibiting the synthesis of AbrB, which is a repressor of the gene coding for σH (Hoch, 1991). Furthermore, kinA (gene coding for the sensor kinase A of the phosphorelay) and spo0A belong to the category of lowthreshold activated Spo0A genes, and both genes are transcribed by the σH-containing RNA polymerase holoenzyme (Predich et al., 1992; Fujita and Sadaie, 1998). Low- 3. Results and Discussion 29 threshold activation of kinA and spo0A is likely to be mediated by the positive feedback loop as indicated above (see Introduction - 2.2.3) involving synthesis and activation of Spo0A. Thus, the transcription of spo0A during stationary phase is stimulated by a certain amount of phosphorylated Spo0A (Spo0A~P), and in turn, the phosphorylation of Spo0A is accelerated through the Spo0A auto-stimulatory loop. 3.1.2. Activation of Spo0A does not occur in the ftsH knockout There is likely to be at least a partial correlation between the level of Spo0A and the level of its phosphorylation as Spo0A is part of a positive feedback loop, in which the response regulator directly and indirectly stimulates the expression of genes involved its phosphorylation (Hoch, 1991). We asked whether the small amount of Spo0A in the ftsH knockout is present in its active or inactive form. Recently, it was reported that many of the genes of the Spo0A regulon respond to the transcription factor in a dose-dependent manner. Four different categories of responses to active Spo0A were distinguished: (i) genes that require a high level of Spo0A~P to be activated (e.g. spoIIA), (ii) those that require a high level of Spo0A~P to be repressed (e.g. rapA), (iii) those that are activated at a low level of the regulator (e.g. skf), and (iv) those that are repressed at a low dose of the regulator (e.g. abrB) (Fujita et al., 2005). Indeed, I could show progressive increases in the level of Spo0A~P in the wild-type strain leading to the transcription of skf-lacZ and spoIIA-lacZ from the onset of the transition phase on defined as t0. While further accumulation of abrB-lacZ was successfully repressed in the wild-type strain, the ftsH knockout failed to activate transcription of skf-lacZ and spoIIA-lacZ or to repress expression of abrB-lacZ (Fig. 8). These results agree with previous findings that expression of spoIIE-lacZ was prevented in the ftsH::spc mutant (Lysenko et al., 1997) and suggest that Spo0A is present in its inactive form in the ftsH mutant. The absence of sufficient amounts of active Spo0A are responsible for the failure to sporulate. It can be hypothesized that either Spo0A is not phosphorylated at all or immediately dephosphorylated after phosphorylation. Therefore, it has to be assumed that FtsH has to degrade one or more proteins which act, directly or indirectly, as an negative regulators of the synthesis or/and activation of Spo0A. 3. Results and Discussion 30 Fig.8. Transcription from the Spo0Acontrolled promoters at various level of Spo0A in sporulating growth condition. β-galatosidase activity was measured in strains containing the fusions integrated at amyE locus. Symbols (■) the wild-type ftsH strain, (Ο) the ftsH::erm strain Beta-galactosidase activity [units] 0 5 10 15 20 25 t-2 t-1 t0 t1 t2 t3 t4 spoIIA -lacZ t-2 t-1 t0 t1 t2 t3 t4 Time [h] 0 10 20 30 t-2 t-1 t0 t1 t2 t3 t4 abrB-lacZ t-2 t-1 t0 t1 t2 t3 t4 Time [h] 0 20 40 60 80 100 120 t-1t0t1t2t3t4 skf -lacZ t-1 t0 t1 t2 t3 t4 3.1.3. After artificial induction of active Spo0A, cells are able to sporulate in the absence of FtsH To examine whether FtsH is essential for sporulation only because of its affect on the synthesis or/and phosphorylation of Spo0A, the wild-type spo0A allele was replaced by an IPTG-inducible variant of Spo0A (spo0A-sad67D56N), which is active in the complete absence of phosphorylation (Ireton et al., 1993). The sporulation frequencies upon induction of the constitutively active form of Spo0A in the presence or absence of ftsH were comparable and reached 37.2% and 33.5%, respectively (Table 2, Part A), while the sporulation frequency was low in the absence of IPTG. These results strongly suggest that ftsH is only essential during stage 0 of sporulation. 3.2. The level of Spo0A is restored in the ftsH spo0E knockout, but not in the ftsH rap knockout Since FtsH has been clearly shown to be involved in the synthesis and activation of Spo0A, it can be assumed that FtsH can influence one of the different components of the phosphorelay, thereby preventing phosphorylation of Spo0A. I focused on a series of antagonists of the phosphorelay, each of which responds to a particular environmental 3. Results and Discussion 31 signal thereby altering the rate of accumulation of Spo0A~P. Of particularly interest are the phosphorelay-associated phosphatases Rap (RapA, RapB and RapE) and the Spo0E family (Spo0E, YisI and YnzD), which dephosphorylate Spo0F~P and Spo0A~P, respectively. They were first examined whether they are involved in the production of active Spo0A in the presence and absence of FtsH. Eight strains were analysed for the production of Spo0A by Western blotting. While in the absence of rapA, rapB and spo0E the amount of Spo0A was increased at t0 as compared to the wild-type situation, its amount was reduced at both stage 0 and stage 1 in the rapE knockout (Fig. 1, Part A). When the ftsH null allele was added, the amount of Spo0A dropped in all strains carrying rap disruptant alleles as already observed for the wild-type strain in the absence of ftsH, but not in the case of the Δspo0E. In the Δspo0E ΔftsH strain, the amount of Spo0A was reduced at stage 0 and further increased to levels comparable to those present in the ftsH+ strain. This result suggests an interaction between FtsH and Spo0E, either directly or indirectly, thereby influencing on the expression of spo0A. 3.3.The activation of Spo0A in the phosphorelay phosphatase deletion strains The transcriptional skf-lacZ fusion, which responds to a low threshold of Spo0A∼P for activation, was used as a reporter system to study the activity of Spo0A in strains with mutant rap and spo0E alleles. While the expression started in all strains at stage 0 and reached its plateau value at stage 2 in the wild-type and in the ΔrapB strain, it accelerated in the Δspo0E earlier and further increased in the ΔrapA and the ΔrapE strains (Fig. 2, Part A). However, in the absence ftsH, the failure to activate the skf promoter was observed in all strains. In conclusion, the absence of rap or spo0E alleles in an ftsH knockout leads to an expression of the spo0A gene, but the protein remained inactive. These data indicate that ftsH influences production of active Spo0A by either allowing its phosphorylation or preventing its rapid dephosphorylation. 3.4. Sporulation frequency increasing in the phosphorelay phosphatase deletion strains In order to assess the relative contribution of each Rap and of the Spo0E phosphatase in modulating the activity of Spo0A, the sporulation efficiency in each phosphatase mutant in the presence or absence of FtsH was measured. The sporulation frequencies in all rap and spo0E knockouts were higher than that of the wild-type strain and ranged from 67% to 75%, where the sporulation frequency of wild-type cells was determined to be 58%. A similar observation has been published for RapA, RapE (Jiang et al., 2000a) and Spo0E (Perego and Hoch, 1991). If a null allele in any one of them 3. Results and Discussion 32 was combined with an ftsH knockout, the sporulation frequency was increased by about two to three orders of magnitude as compared to a single ftsH mutant, but always remained below 1% (Table 1, Part A). The sporulation frequency in the phosphatase and ftsH double knockouts was only partly restored and still about 100-fold lower than the wild-type level. It can be concluded that ftsH is somehow involved in the synthesis of active Spo0A. This result agrees with the previous studies that deletion of the rap genes or spo0E gene results in increased sporulation frequency (Jiang et al., 2000a; Perego, 2001), while over-production of these genes results in inhibition of sporulation. It is likely that the phosphatases act as negative regulators of the developmental process and respond to inand external signals to influence the amount of active Spo0A. Members of the Rap family of phosphatases are known to be differentially activated by physiological processes alternative to sporulation, e.g. competence development induces RapA and RapE, while vegetative growth conditions induce RapB (Perego et al., 1994; Jiang et al., 2000). Since vegetative growth and competence are processes that cannot occur in a sporulating cell, the induction of rap phosphatases prevents sporulation from interfering with these processes. Transcription of the Spo0E phosphatase is affected by signals that are still unknown (Perego and Hoch, 1991; Ohlsen et al., 1994), but it is induced at the end of the exponential growth phase as repression by AbrB is relieved owing to accumulation of Spo0A P (Strauch et al., 1989; Perego, 2001). Apparently, cells that initiate spo0A autoactivation also induce the Spo0E levels. This co-expression suggests that Spo0E serves as some kind of a ‘safety lid’ to prevent over-stimulation of spo0A autoactivation. 3.5. Spo0E is a target of FtsH in vitro The observation that FtsH interferes with the synthesis and phosphorylation of Spo0A resulted in the hypothesis that FtsH has to degrade one or more negative regulators either preventing the phosphorylation of Spo0A or being involved in its rapid dephosphorylation. Since Spo0E directly targets Spo0A∼P and represses sporulation when overproduced and increases sporulation as its absence, one possibility to explain these data is a direct interaction between FtsH and Spo0E resulting in its degradation. To test this hypothesis, FtsH was purified with a GST-tag as reported before (Kotschwar et al., 2005). This purification tag keeps the protein soluble in the absence of any added detergent. Furthermore, GST has a low thermodynamic stability, which is efficiently degraded by FtsH (Okuno et al., 2003). The C-terminus of GST contains an alpha-helix as a last structured element (Andujar-Sanchez et al., 2005), which is followed by an unstructured spacer; this tag was expected to facilitate exposure of fused 3. Results and Discussion 33 polypeptides. Indeed, GST is generally stable in E. coli but when the proteolytic SsrA-tag is added, the fusion protein is rapidly degraded (Okuno et al., 2004). From our data, GST-FtsH is stable in B. subtilis (data not shown). In the in vitro degradation assay with both purified GST-FtsH and GST-Spo0E, it could be shown that Spo0E was degraded by FtsH (Fig 3, Part A). Since no antibodies against Spo0E were available, antibodies against GST were used to confirm the instability of Spo0E. As mentioned above, two homologs of Spo0E, YisI and YnzD, are also able to dephosphorylate Spo0A in vitro (Perego, 2001). These two phosphatases are distinguished from Spo0E by two characteristics: First, their genes are expressed during the vegetative growth phase and second, they lack a C-terminal extension of about 25 amino acid residues. It could be shown that both GST-YisI and GST-YnzD remained stable upon incubation with FtsH up to at least 5 h (Fig. 4, Part A). These data clearly demonstrated that neither YisI nor YnzD are substrates of FtsH. They further suggest that the C-terminal extension of Spo0E renders this phosphatase unstable. Does FtsH fully degrade Spo0E in all cells or does it modulate its steady-state level? Based on my data, I would like to suggest that FtsH regulates the steady-state level of Spo0E rather than completely degrading it. Alternatively, though less likely, it might fully degrade Spo0E is some cells and not attack it at all in others. The first assumption is based on two observations:(i) a spo0E knockout leads to an increase in the sporulation frequency, which never reaches 100%; (ii) the cellular amount of Spo0E is low as the protein is barely detectable in cellular extracts (data not shown). A similar observation has been published for LpxC of E. coli, where only a few hundred molecules per cell are present (Führer et al., 2006). Here, a tight control of the amount of this enzyme by FtsH is essential to prevent the accumulation of abnormal membranes in the periplasm (Ogura et al., 1999) leading to cell death (Sullivan and Donachie, 1984). This finding further indicates that FtsH has to degrade or regulate the steady-state level of one or more proteins interfering negatively with successful sporulation. 3.6. The C-terminal end of Spo0E is responsible for degradation by FtsH A remarkable feature of the FtsH protease is its specificity for target recognition. To initiate cytoplasmic protein degradation, FtsH recognizes a tail, normally located at Cterminus, which contains a critical signal for degradation initiation. The published observation that the two Spo0E homologs, the proteins YisI and YnzD, lack about 25 amino acid residues at their C-terminus and that the two truncated versions of Spo0E, Spo0E94 and Spo0E11, are still able to dephosphorylate Spo0A∼P strongly suggest that 3. Results and Discussion 34 the C-terminus of Spo0E may have an inhibitory role rather than being required for its enzymatic activity. Spo0E11 and Spo0E94 were tagged with GST, overproduced in E. coli and purified. When these two purified proteins were incubated with FtsH, both remained stable for at least 5 h, while β-casein as a control was degraded under these conditions (Fig. 4, Part A). Since the full-length Spo0E protein is unstable when incubated with FtsH, I infer that the 25 C-terminal amino acid residues is responsible for this instability. 3.7. The 25 C-terminal amino acids of Spo0E is considered as the proteolytic tag for FtsH-mediated degradation Since Spo0E serves as a target for FtsH, but two its homologs YisI and YnzD not, I fused the coding region for the Spo0E C-terminal 25 amino acids to ynzD designated YnzD-0E. The GST-tagged hybrid protein was overproduced in E. coli, purified by affinity chromatography and incubated with GST-FtsH. As can be seen from Fig. 6, Part A (lane 6), the YnzD-0E is largely degraded over time. In conclusion, the C-terminal 25 amino acids of Spo0E contain the residues recognized by FtsH and it was considered as proteolytic tag for B. subtilis FtsH-mediated degradation. Similar observations were made with two different E. coli FtsH substrates, the phage λ CII protein, whose C-terminus is required for degradation by FtsH (Kobiler et al., 2002), and the SsrA-degradation-tag (Keiler et al., 1996), in which the λ CI protein has been converted into an FtsH substrate by attachment of the SsrA-tag to its C-terminus (Herman et al., 1998). Which amino acids are recognized by the FtsH protease? It has been suggested that the FtsH protease recognizes flexible tail specificity or even the length of the target protein (Herman et al., 1998; Führer et al., 2007). The tail-specific recognition is physiologically relevant, and is used for degrading the λ CII activator, SsrA-tagged proteins (Herman et al., 1998; Kobiler et al., 2002) and LpxC, the key enzyme in lipopolysacharide formation by controlling the ratio between LPS and phospholipids (Sorensen et al., 1996). These sequences are enriched in non-polar amino acids at their very C-terminus. Comparison of the C-termini from LpxC and λ CII revealed no similarities. In contrast, the SsrA-tag exhibits a remarkable similarity to the final eleven residues of LpxC. The exact LpxC degradation-tag was determined by mutational analysis. Six non-polar amino acids within the C-terminal eleven residues of LpxC turned out to be required for degradation (Führer et al., 2007). From several studies with FtsH and its eukaryotic homologs, it was concluded that the protease prefers hydrophobic and non-polar residues at cleavage sites whereas acidic residues abolish degradation (Ito and Akiyama, 2005; Koppen and Langer, 2007). In the case of Spo0E, there is no similarity to the B. subtilis SsrA-tag (Wiegert and Schumann, 2001). Therefore, the amino 3. Results and Discussion 35 acid sequence recognized by the FtsH protease is different from that of the SsrA-tag. Experiments are in progress to identify the amino acid residues of Spo0E recognized by FtsH. Part B: Function of FtsH late during sporulation 3.1. Assumption that FtsH interacts with the small sporulation peptide SpoVM spoVM is a developmental gene essential for sporulation. Some spoVM mutants arrest sporulation at stage IV-V and allow the formation of the forespore but impair synthesis and assembly of the spore cortex (Levin et al., 1993). SpoVM was considered as a morphogenetic protein since it is synthesized in the mother cell compartment and almost quantitatively localized to the engulfing membrane (Levin et al., 1993; van Ooij and Losick, 2003). Furthermore, spore morphogenesis is dependent on the proper localization of SpoVM (van Ooij and Losick, 2003). A transposon insertion within spoVM leading to sporulation-deficient cells was used to select for extragenic suppressors; such extragenic suppressors were mapped within ftsH (Cutting et al., 1997). Furthermore, it could be shown that chemically synthesized SpoVM was able to inhibit degradation of σ32 by purified E. coli FtsH (Cutting et al., 1997). I could show in an in vitro degradation assay with purified B. subtilis FtsH that βcasein, an unstructured protein strongly degraded by FtsH, turned out to be stabilized in the presence of SpoVM (Fig. 1, Part B). These findings strongly suggest that, first, FtsH and SpoVM interact functionally and that, second, SpoVM inhibits the B. subtilis FtsH protease late during sporulation. This assumption is sustained by two observations: First, FtsH-GFP has been shown to accumulate within the asymmetric septum (Wehrl et al., 2000) and, second, SpoVM-GFP colocalized with the polar septum, too (van Ooij and Losick, 2003). It can be inferred that, in the absence of SpoVM, FtsH will degrade at least one protein essential to complete successful sporulation or, alternatively, regulate the steady-state level of SpoVM; both possibilities are not mutually exclusive. In some degradation assays, the partial disappearance of SpoVM was observed (data not shown) as described for the bacteriophage λ CIII peptide, which is known to inhibit λ CII protein degradation by FtsH, but being unstable when FtsH is overproduced (Herman et al., 1997). A short domain (residues 16-37) of CIII may form an amphipathic α-helix which is essential for its activity (Kornitzer et al., 1991). Interestingly, SpoVM was also predicted to form such an amphipathic α-helix, though it displays no sequence similarity with λ CIII (Prajapati, et al., 2000; Ramamurthi et al., 2006). We infer from these data that the essential SpoVM 4. References 42 Fleischmann,R.D., Adams,M.D., White,O., Clayton,R.A., Kirkness,E.F., Kerlavage,A.R. et al. (1995) Whole-genome random sequencing and assembly of Haemophilus influenzae Rd. Science 269: 496-512. Flynn,J.M., Levchenko,I., Seidel,M., Wickner,S.H., Sauer,R.T., and Baker,T.A. (2001) Overlapping recognition determinants within the ssrA degradation tag allow modulation of proteolysis. Proc Natl Acad Sci USA 98: 10584-10589. Fraser,C.M., Gocayne,J.D., White,O., Adams,M.D., Clayton,R.A., Fleischmann,R.D. et al. (1995) The minimal gene complement of Mycoplasma genitalium. Science 270: 397-403. Führer,F., Langklotz,S., and Narberhaus,F. (2006) The C-terminal end of LpxC is required for degradation by the FtsH protease. Mol Microbiol 59: 1025-1036. Führer,F., Müller,A., Baumann,H, Langklotz,S., Kutscher,B., and Narberhaus,F. (2007) Sequence and length recognition of the C-terminal turnover element of LpxC, a soluble substrate of the membrane-bound FtsH protease. J Mol Biol 372: 485-496 Fujita,M., Amemura,A., and Aramaki,H. (1998) Transcription of the groESL operon in Pseudomonas aeruginosa PAO1. FEMS Microbiol Lett 163: 237-242. Fujita,M., and Losick,R. (2003) The master regulator for entry into sporulation in Bacillus subtilis becomes a cell-specific transcription factor after asymmetric division. Genes Dev 17: 1166-1174. Fujita,M., and Sadaie,Y. (1998) Feedback loops involving Spo0A and AbrB in in vivo transcription of the genes involved in the initiation of sporulation in Bacillus subtilis. J Biochem 124: 98-104. Fujita,M., Gonzalez-Pastor,J.E., and Losick,R. (2005) Highand low-threshold genes in the Spo0A regulon of Bacillus subtilis. J Bacteriol 187: 1357-1368. Geisler,U., and Schumann,W. (1993) Isolation of stress mutants of Bacillus subtilis by a novel genetic method. FEMS Microbiol Lett 108: 251-254. Gonzalez-Pastor,J.E., Hobbs,E.C., and Losick,R. (2003) Cannibalism by sporulating bacteria. Science 301: 510-513. Gottesman,S., Roche,E., Zhou,Y.N., and Sauer,R.T. (1998) The ClpXP and ClpAP proteases degrade proteins with carboxy-terminal peptide tails added by the SsrA-tagging system. Genes Dev 12: 1338-1347. Griffith,K.L., Shah,I.M., and Wolf,R.E. (2004) Proteolytic degradation of Escherichia coli transcription activators SoxS and MarA as the mechanism for reversing the 4. References 43 induction of the superoxide (SoxRS) and multiple antibiotic resistance (Mar) regulons. Mol Microbiol 51: 1801-1816. Groll,M., Bochtler,M., Brandstetter,H., Clausen,T., and Huber.R. (2005) Molecular machines for protein degradation. Chembiochem 6: 222-256. Grossman,A.D. (1995) Genetic networks controlling the initiation of sporulation and the development of genetic competence in Bacillus subtilis. Annu Rev Genet 29: 477508. Guex,N., and Peitsch,M.C. (1997) SWISS-MODEL and the Swiss-PdbViewer: an environment for comparative protein modeling. Electrophoresis 18: 2714-2723. Halder,S., Datta,A.B., and Parrack,P. (2007) Probing the antiprotease activity of λCIII, an inhibitor of the Escherichia coli metalloprotease HflB (FtsH). J Bacteriol 189: 8130-8138. Hamon,M.A., and Lazazzera,B.A. (2001) The sporulation transcription factor Spo0A is required for biofilm development in Bacillus subtilis. Mol Microbiol 42: 1199-1209. Hasty,J., Pradines,J., Dolnik,M., and Collins.J.J. (2000) Noise-based switches and amplifiers for gene expression. PNAS 9: 2075-2080. Herman,C., and D'Ari,R. (1998) Proteolysis and chaperones: The destruction/reconstruction dilemma. Curr Opin Microbiol 1: 204-209. Herman,C., Ogura,T., Tomoyasu,T., Hiraga,S., Akiyama,Y., Ito,K. et al. (1993) Cell growth and lambda phage development controlled by the same essential Escherichia coli gene, ftsH/hflB. Proc Natl Acad Sci USA 90: 10861-10865. Herman,C., Prakash,S., Lu,C.Z., Matouschek,A., and Gross,C.A. (2003) Lack of a robust unfoldase activity confers a unique level of substrate specificity to the universal AAA protease FtsH. Mol Cell 11: 659-669. Herman,C., Thévenet,D., Bouloc,P., Walker,G.C., and D'Ari,R. (1998) Degradation of carboxy-terminal-tagged cytoplasmic proteins by the Escherichia coli protease HflB (FtsH). Genes Dev 12: 1348-1355. Herman,C., Thévenet,D., D'Ari,R., and Bouloc,P. (1995) Degradation of σ32, the heat shock regulator in Escherichia coli, is governed by HflB. Proc Natl Acad Sci USA 92: 3516-3520. Herman,C., Thévenet,D., D'Ari,R., and Bouloc,P. (1997) The HflB protease of Escherichia coli degrades its inhibitor λCIII. J Bacteriol 179: 358-363. 4. References 44 Hilbert,D.W., and Piggot,P.J. (2004) Compartmentalization of gene expression during Bacillus subtilis spore formation. Microbiol Mol Biol Rev 68: 234-262. Hoskins,J.R., Kim,S.Y., and Wickner,S. (2000) Substrate recognition by the ClpA chaperone component of ClpAP protease. J Biol Chem 275: 35361-35367. Hoyt,M.A., Knight,D.M., Das,A., Miller,H.I., and Echols,H. (1982) Control of phage λ development by stability and synthesis of cII protein: Role of the viral cIII and host hflA, himA, and himD genes. Cell 31: 565-573. Hunger,K., Beckering,C.L., and Marahiel,M.A. (2004) Genetic evidence for the temperature-sensing ability of the membrane domain of the Bacillus subtilis histidine kinase DesK. FEMS Microbiol Lett 230: 41-46. Ireton,K., Rudner,D.Z., Siranosian,K.J., and Grossman,A.D. (1993) Integration of multiple developmental signals in Bacillus subtilis through the Spo0A transcription factor. Genes Dev 7: 283-294. Isaacs,F.J., Hasty,J., Cantor,C.R., and Collins,J.J. (2003) Prediction and measurement of an autoregulatory genetic module. Proc Natl Acad Sci USA 100: 7714-7719. Ito,K., and Akiyama,Y. (2005) Cellular functions, mechanism of action, and regulation of ftsH protease. Annu Rev Microbiol 59: 211-231. Jarosch,E., Taxis,C., Volkwein,C., Bordallo,J., Finley,D., Wolf,D.H., and Sommer,T. (2002) Protein dislocation from the ER requires polyubiquitination and the AAAATPase Cdc48. Nat Cell Biol 4:134-139. Jiang,M., Grau,R., and Perego,M. (2000a) Differential processing of propeptide inhibitors of Rap phosphatases in Bacillus subtilis. J Bacteriol 182: 303-310. Jiang,M., Shao,W., Perego,M., and Hoch,J.A. (2000b) Multiple histidine kinases regulate entry into stationary phase and sporulation in Bacillus subtilis. Mol Microbiol 38: 535-542. Kanemori,M., Nishihara,K., Yanagi,H., and Yura,T. (1997) Synergistic roles of Hs1VU and other ATP-dependent proteases in controlling in vivo turnover of σ32 and abnormal proteins in Escherichia coli. J Bacteriol 179: 7219-7225. Karzai,A.W., Roche,E.D., and Sauer,R.T. (2000) The SsrA-SmpB system for protein tagging, directed degradation and ribosome rescue. Nat Struct Biol 7: 449-455. Keiler,K.C., and Sauer,R.T. (1996) Sequence determinants of C-terminal substrate recognition by the Tsp protease. J Biol Chem 271: 2589-2593. 4. References 45 Kenniston,J.A., Baker,T.A., Fernandez,J.M., and Sauer,R.T. (2003) Linkage between ATP consumption and mechanical unfolding during the protein processing reactions of an AAA+ degradation machine. Cell 114: 511-520. Kihara,A., Akiyama,Y., and Ito,K. (1995) FtsH is required for proteolytic elimination of uncomplexed forms of SecY, an essential protein translocase subunit. Proc Natl Acad Sci USA 92: 4532-4536. Kihara,A., Akiyama,Y., and Ito,K. (1997) Host regulation of lysogenic decision in bacteriophage lambda: Transmembrane modulation of FtsH (HflB), the cII degrading protease, by HflKC (HflA). Proc Natl Acad Sci USA 94: 5544-5549. Kihara,A., Akiyama,Y., and Ito,K. (1998) Different pathways for protein degradation by the FtsH/HflKC membrane-embedded protease complex: An implication from the interference by a mutant form of a new substrate protein, YccA. J Mol Biol 279: 175-188. Kihara,A., Akiyama,Y., and Ito,K. (1999) Dislocation of membrane proteins in FtsHmediated proteolysis. EMBO J 18: 2970-2981. Kim,Y.I., Burton,R.E., Burton,B.M., Sauer,R.T., and Baker.T.A. (2000) Dynamics of substrate denaturation and translocation by the ClpXP degradation machine. Mol Cell 5: 639-648. Kim,K.I., Cheong,G.W., Park,S.C., Ha,J.S., Woo,K.M., Choi,S.J., and Chung,C.H. (2000) Heptameric ring structure of the heat-shock protein ClpB, a proteinactivated ATPase in Escherichia coli. J Mol Biol 303: 655-666. Kobiler,O., Koby,S., Teff,D., Court,D., and Oppenheim,A.B. (2002) The phage lambda CII transcriptional activator carries a C-terminal domain signaling for rapid proteolysis. Proc Natl Acad Sci USA 99: 14964-14969. Korat,B., Mottl,H., and Keck,W. (1991) Penicillin-binding protein 4 of Escherichia coli: molecular cloning of the dacB gene, controlled overexpression, and alterations in murein composition. Mol Microbiol 5: 675-684. Kornitzer,D., Altuvia,S., and Oppenheim,A.B. (1991) The activity of the CIII regulator of lambdoid bacteriophages resides within a 24-amino acid protein domain. Proc Natl Acad Sci USA 88: 5217-5221. Kotschwar,M., Harfts,E., Ohanjan,T., and Schumann,W. (2005) Construction and analyses of mutant ftsH alleles of Bacillus subtilis involving the ATPaseand Znbinding domains. Curr Microbiol 49: 180-185. 4. References 46 Krzywda,S., Brzozowski,A.M., Verma,C., Karata,K., Ogura,T., and Wilkinson,A.J. (2002) The crystal structure of the AAA domain of the ATP-dependent protease FtsH of Escherichia coli at 1.5 Å resolution. Structure 10: 1073-1083. Levin,M.E., Hendrix,R.W., and Casjens,S.R. (1993a) A programmed translational frameshift is required for the synthesis of a bacteriophage lambda tail assembly protein. J Mol Biol 234: 124-139. Levin,P.A., and Losick,R. (1996) Transcription factor Spo0A switches the localization of the cell division protein FtsZ from a medial to a bipolar pattern in Bacillus subtilis. Genes Dev 10: 478-488. Levin,P.A., Fan,N., Ricca,E., Driks,A., Losick,R., and Cutting,S. (1993b) An unusually small gene required for sporulation by Bacillus subtilis. Mol Microbiol 9: 761-771. Lysenko,E., Ogura,T., and Cutting,S.M. (1997) Characterization of the ftsH gene of Bacillus subtilis. Microbiology 143: 971-978. Obrist,M., Milek,S., Klauck,E., Hengge,R., and Narberhaus,F. (2007) Region 2.1 of the Escherichia coli heat-shock sigma factor RpoH (sigma32) is necessary but not sufficient for degradation by the FtsH protease. Microbiology 153: 2560-2571 Makino,S., Qu,J.N., Uemori,K., Ichikawa,H., Ogura,T., and Matsuzawa,H. (1997) A silent mutation in the ftsH gene of Escherichia coli that affects FtsH protein production and colicin tolerance. Mol Gen Genet 254: 578-583. Mansilla,M.C., and De Mendoza,D. (2005) The Bacillus subtilis desaturase: a model to understand phospholipid modification and temperature sensing. Arch Microbiol 183: 229-235. Michaels,M.L., Cruz,C., Grollman,A.P., and Miller,J.H. (1992) Evidence that MutY and MutM combine to prevent mutations by an oxidatively damaged form of guanine in DNA. Proc Natl Acad Sci USA 89: 7022-7025. Mogk,A., Mayer,M.P., and Deuerling,E. (2002) Mechanisms of protein folding: molecular chaperones and their application in biotechnology. Chembiochem 3: 807-814. Molle,V., Fujita,M., Jensen,S.T., Eichenberger,P., Gonzalez-Pastor,J.E., Liu,J.S., and Losick,R. (2003) The Spo0A regulon of Bacillus subtilis. Mol Microbiol 50: 1683-1701. Mueller,J.P., and Sonenshein,A.L. (1992) Role of the Bacillus subtilis gsiA gene in regulation of early sporulation gene expression. J Bacteriol 174: 4374-4383. 4. References 47 Mueller,J.P., Bukusoglu,G., and Sonenshein,A.L. (1992) Transcriptional regulation of Bacillus subtilis glucose starvation-inducible genes: Control of gsiA by the ComPComA signal transduction system. J Bacteriol 174: 4361-4373. Mujacic,M., Cooper,K.W., and Baneyx,F. (1999) Cold-inducible cloning vectors for lowtemperature protein expression in Escherichia coli: application to the production of a toxic and proteolytically sensitive fusion protein. Gene 238: 325-332. Nilsson,D., Lauridsen,A.A., Tomoyasu,T., and Ogura,T. (1994) A Lactococcus lactis gene encodes a membrane protein with putative ATPase activity that is homologous to the essential Escherichia coli ftsH gene product. Microbiology 140: 2601-2610. Niwa,H., Tsuchiya,D., Makyio,H., Yoshida,M., and Morikawa,K. (2002) Hexameric ring structure of the ATPase domain of the membrane-integrated metalloprotease FtsH from Thermus thermophilus HB8. Structure 10: 1415-1423. Ogura,M., Hirao,S., Ohshiro,Y., and Tanaka,T. (1999a) Positive regulation of Bacillus subtilis sigD by C-terminal truncated LacR at translational level. FEBS Lett 457: 112-116. Ogura,M., Liu,L., Lacelle,M., Nakano,M., and Zuber,P. (1999b) Mutational analysis of ComS: evidence for the interaction of ComS and MecA in the regulation of competence development in Bacillus subtilis. Mol Microbiol 32: 799-812. Ogura,T., and Wilkinson,A.J. (2001) AAA+ superfamily ATPases: common structurediverse function. Genes to Cells 6: 575-597. Ohlsen,K.L., Grimsley,J.K., and Hoch,J.A. (1994) Deactivation of the sporulation transcription factor Spo0A by the Spo0E protein phosphatase. Proc Natl Acad Sci USA 91: 1756-1760. Okuno,T., Yamada-Inagawa,T., Karata,K., Yamanaka,K., and Ogura,T. (2004) Spectrometric analysis of degradation of a physiological substrate σ32 by Escherichia coli AAA protease FtsH. J Struct Biol 146: 148-154. Okuno,T., Yamanaka,K., and Ogura,T. (2006) An AAA protease FtsH can initiate proteolysis from internal sites of a model substrate, apo-flavodoxin. Genes Cells 11: 261-268. Parker,G.F., Daniel,R.A., and Errington,J. (1996) Timing and genetic regulation of commitment to sporulation in Bacillus subtilis. Microbiology 142: 3445-3452. Perego,M. and HochJ.A. (2002) Two-component systems, phosphorelays and regulation of their activities by phosphatases, pp.473-481. In A.L. Sonenshein, et. 4. References 48 al. (ed.) Bacillus subtilis and Its Closest Relatives: From Genes to Cells. ASM Press, Washington, D.C. Perego,M. (2001) A new family of aspartyl phosphate phosphatases targeting the sporulation transcription factor Spo0A of Bacillus subtilis. Mol Microbiol 42: 133143. Perego,M., and Hoch,J.A. (1991) Negative regulation of Bacillus subtilis sporulation by the spo0E gene product. J Bacteriol 173: 2514-2520. Perego,M., and Hoch,J.A. (1996) Cell-cell communication regulates the effects of protein aspartate phosphatases on the phosphorelay controlling development in Bacillus subtilis. Proc Natl Acad Sci USA 93: 1549-1553. Perego,M., Hanstein,C., Welsh,K.M., Djavakhishvili,T., Glaser,P., and Hoch,J.A. (1994) Multiple protein-aspartate phosphatases provide a mechanism for the integration of diverse signals in the control of development in B. subtilis. Cell 79: 1047-1055. Perego,M., Spiegelman,G.B., and Hoch,J.A. (1988) Structure of the gene for the transition state regulator, abrB: regulator synthesis is controlled by the spo0A sporulation gene in Bacillus subtilis. Mol Microbiol 2: 689-699 Piggot,P.J., and Losick,R. (2002) Sporulation genes and intercompartmental regulation. In Bacillus subtilis and its Closest Relatives: from Genes to Cells. Sonenshein, A.L., Hoch, J.A., and Losick, R. (eds). Washington, DC: American Society for Microbiology Press: 483-517. Piggot,P.J., and Hilbert,D.W. (2004) Sporulation of Bacillus subtilis. Curr Opin Microbiol 7: 579-586. Pogliano,J., Sharp,M.D, and Pogliano,K. (2002) Partitioning of chromosomal DNA during establishment of cellular asymmetry in Bacillus subtilis. J Bacteriol 184: 1743-1749. Prajapati,R.S., Ogura,T., and Cutting,S.M. (2000) Structural and functional studies on an FtsH inhibitor from Bacillus subtilis. Biochim Biophys Acta Gen Subj 1475: 353-359. Prakash, S. and Matouschek,A. (2004) Protein unfolding in the cell. Trends Biochem Sci 29: 593-600. Predich,M., Nair,G., and Smith,I. (1992) Bacillus subtilis early sporulation genes kinA, spo0F, and spo0A are transcribed by the RNA polymerase containing σH. J Bacteriol 174: 2771-2778. 4. References 49 Qing,G.L., Ma,L.C., Khorchid,A., Swapna,G.V.T., Mal,T.K., Takayama,M.M. et al. (2004) Cold-shock induced high-yield protein production in Escherichia coli. Nat Biotechnol 22: 877-882. Qu,J.N., Makino,S., Adachi,H., Koyama,Y., Akiyama,Y., Ito,K. et al. (1996) The tolZ gene of Escherichia coli is identified as the ftsH gene. J Bacteriol 178: 34573461. Ramamurthi,K.S., Clapham,K.R., and Losick,R. (2006) Peptide anchoring spore coat assembly to the outer forespore membrane in Bacillus subtilis. Mol Microbiol 62: 1547-1557. Robertson,J.B., Gocht,M., Marahiel,M.A., and Zuber,P. (1989) AbrB, a regulator of gene expression in Bacillus subtilis, interacts with the transcription initiation regions of a sporulation gene and an antibiotic biosynthesis gene. Proc Natl Acad Sci USA 86: 8457-8461. Santos,D., and Almeida,D.F. (1975) Isolation and characterization of a new temperature-sensitive cell division mutant of Escherichia coli K-12. J Bacteriol 124: 1502-1507. Sauer,R.T., Bolon,D.N., Burton,B.M., Burton,R.E., Flynn,J.M., Grant,R.A. et al. (2004) Sculpting the proteome with AAA plus proteases and disassembly machines. Cell 119: 9-18. Shotland,Y., Koby,S., Teff,D., Mansur,N., Oren,D.A., Tatematsu,K. et al. (1997) Proteolysis of the phage lambda CII regulatory protein by FtsH (HflB) of Escherichia coli. Mol Microbiol 24: 1303-1310. Shotland,Y., Shifrin,A., Ziv,T., Teff,D., Koby,S., Kobiler,O., and Oppenheim,A.B. (2000a) Proteolysis of bacteriophage lambda CII by Escherichia coli FtsH (HflB). J Bacteriol 182: 3111-3116. Shotland,Y., Teff,D., Koby,S., Kobiler,O., and Oppenheim,A.B. (2000b) Characterization of a conserved λ-helical, coiled-coil motif at the C-terminal domain of the ATP-dependent FtsH (HfIB) protease of Escherichia coli. J Mol Biol 299: 953-964. Silvaggi,J.M., Perkins,J.B., and Losick,R. (2006) Genes for small, noncoding RNAs under sporulation control in Bacillus subtilis. J Bacteriol 188: 532-541. Smits,W.K., Kuipers,O.P., and Veening,J.W. (2006) Phenotypic variation in bacteria: the role of feedback regulation. Nat Rev Microbiol 4: 259-271. 4. References 50 Sonenshein,A.L. (2000) Control of sporulation initiation in Bacillus subtilis. Curr Opin Microbiol 3: 561-566. Sorensen,P.G., Lutkenhaus,J., Young,K., Eveland,S.S., Anderson,M.S., and Raetz,C.R. (1996) Regulation of UDP-3-O-[R-3-hydroxymyristoyl]-Nacetylglucosamine deacetylase in Escherichia coli. The second enzymatic step of lipid a biosynthesis. J Biol Chem 271: 25898-25905. Sowell,M.O., and Buchanan,C.E. (1983) Changes in the penicillin binding proteins during sporulation of Bacillus subtilis. J Bacteriol 153: 1331-1337. Stephenson,K., and Hoch,J.A. (2001) PAS-A domain of phosphorelay sensor kinase A: A catalytic ATP-binding domain involved in the initiation of development in Bacillus subtilis. Proc Natl Acad Sci USA 98: 15251-15256. Stirling,C.J. and Lord,J.M. (2006) Quality control: linking retrotranslocation and degradation. Curr Biol 16: 1035-1047. Stragier,P., and Losick,R. (1996) Molecular genetics of sporulation in Bacillus subtilis. Annu Rev Genet 30: 297-341. Strauch,M.A., and Hoch,J.A. (1993a) Signal transduction in Bacillus subtilis sporulation. Curr Opin Genet Dev 3: 203-212. Strauch,M.A., and Hoch,J.A. (1993) Transition-state regulators: sentinals of Bacillus subtilis post-exponential gene expression. Mol Microbiol 7: 337-342. Strauch,M.A., Spiegelman,G.B., Perego,M., Johnson,W.C., Burbulys,D., and Hoch,J.A. (1989) The transition state transcription regulator abrB of Bacillus subtilis is a DNA binding protein. EMBO J 8: 1615-1621. Strauch,M.A., Trach,K.A., and Hoch,J.A. (1992) Spo0A activates and represses its own synthesis by binding at its dual promoters. Biochimie 74: 619-626. Strauch,M.A., Wu,J.-J., Jonas,R.H., and Hoch,J.A. (1993) A positive loop controls transcription of the spo0F gene, a component of the sporulation phosphorelay in Bacillus subtilis. Mol Microbiol 7: 967-974. Sullivan,N.F., and Donachie,W.D. (1984) Transcriptional organization within an Escherichia coli cell division gene cluster: direction of transcription of the cell separation gene envA. J Bacteriol 160: 724-732. Teff,D., Koby,S., Shotland,Y., Ogura,T., and Oppenheim,A.B. (2000) A colicin-tolerant Escherichia coli mutant that confers Hfl phenotype carries two mutations in the region coding for the C-terminal domain of FtsH (HflB). FEMS Microbiol Lett 183: 115-117. 4. References 51 Thomas,J.G., and Baneyx,F. (1996) Protein misfolding and inclusion body formation in recombinant Escherichia coli cells overexpressing heat-shock proteins. J Biol Chem 271: 11141-11147. Tomoyasu,T., Arsène,F., Ogura,T., and Bukau,B. (2001) The C terminus of σ32 is not essential for degradation by FtsH. J Bacteriol 183: 5911-5917. Tomoyasu,T., Gamer,J., Bukau,B., Kanemori,M., Mori,H., Rutman,A.J. et al. (1995) Escherichia coli FtsH is a membrane-bound, ATP-dependent protease which degrades the heat-shock transcription factor σ32. EMBO J 14: 2551-2560. Tomoyasu,T., Yamanaka,K., Murata,K., Suzaki,T., Bouloc,P., Kato,A. et al. (1993a) Topology and subcellular localization of FtsH protein in Escherichia coli. J Bacteriol 175: 1352-1357. Tomoyasu,T., Yura,T., Morimura,S., Mori,H., Yamanaka,K., Niki,H. et al. (1993b) The Escherichia coli FtsH protein is a prokaryotic member of a protein family of putative ATPases involved in membrane functions, cell cycle control, and gene expression. J Bacteriol 175: 1344-1351. Van Ooij,C., and Losick,R. (2003) Subcellular localization of a small sporulation protein in Bacillus subtilis. J Bacteriol 185: 1391-1398. Veening,J.-W., Hamoen,L.W., and Kuipers,O.P. (2005) Phosphatases modulate the bistable sporulation gene expression pattern in Bacillus subtilis. Mol Microbiol 56: 1481-1494. Wang,C.C., and Tsou,C.L. (1998) Enzymes as chaperones and chaperones as enzymes. FEBS Lett 425: 382-384. Weber-Ban,E.U., Reid,B.G., Miranker,A.D., and Horwich,A.L. (1999) Global unfolding of a substrate protein by the Hsp100 chaperone ClpA. Nature 401: 90-93. Wehrl,W., Niederweis,M., and Schumann,W. (2000) The FtsH protein accumulates at the septum of Bacillus subtilis during cell division and sporulation. J Bacteriol 182: 3870-3873. Whittington,D.A., Rusche,K.M., Shin,H., Fierke,C.A., and Christianson,D.W. (2003) Crystal structure of LpxC, a zinc-dependent deacetylase essential for endotoxin biosynthesis. Proc. Natl. Acad. Sci. USA 100: 8146-8150. Wiegert,T., and Schumann,W. (2001) SsrA-mediated tagging in Bacillus subtilis. J Bacteriol 183: 3885-3889. Woodman,P.G. (2003) A protein coping with multiple identities. J Cell Sci 116: 42834290. protein is regulated by phosphorylation through the phosphorelay signal transduction system (Burbulys et al., 1991;Hoch, 1993). The transfer of the phosphate to Spo0A involves a complex network consisting of several kinases (KinA, KinB, KinC, KinD and KinE), where each probably responds to a different stimulus (Jiang et al., 2000b). Upon autophosphorylation, the phosphate is transferred by two intermediates, Spo0F and Spo0B and finally to Spo0A (Burbulys et al., 1991). The phosphotransfer reactions or the phosphoproteins are subject to regulation by phosphatases, where one group, the Rap phosphatases, are regulated by pentapeptides (Perego, 1998). The Rap phosphatases specifically dephosphorylate Spo0A~P, while another three phosphatases (Spo0E, YisI and YnzD) attack Spo0A~P. The ftsH gene coding for a membrane-anchored metalloprotease is present in most if not all bacterial species (Schumann, 1999;Ogura and Wilkinson, 2001). The FtsH protein and its biochemical and biological functions have been studied in detail in E. coli. It carries two transmembrane segments close to its N-terminal end which anchor this protein into the cytoplasmic membrane in such a way that both its short Nand its long Cterminus are exposed into the cytoplasm (Tomoyasu et al., 1993a). The C-terminal part contains a Walker A and B box, involved in binding and hydrolysis of ATP (Tomoyasu et al., 1993b), and a binding site for Zn2+. The Aquifex aeolicus FtsH protein devoid of its transmembrane segments has been crystallized and shown to form a ringlike hexameric structure (Suno et al., 2006). While the ftsH gene in E. coli is essential (Ogura et al., 1999), a B. subtilis ftsH knockout is viable, but displays a pleiotropic phenotype (Deuerling et al., 1997). Cells with an ftsH null allele are sensitive to heatand osmotic stress, grow largely as filaments and last, but not least, exhibit a significantly reduced sporulation frequency. Here, we started to analyze the role of ftsH during sporulation. So far, we could show that ftsH interferes with the synthesis or/and phosphorylation of Spo0A. Based on this result we hypothesize that the FtsH protease has to degrade one or more proteins involved, directly or indirectly, in the production of a sufficient amount of active Spo0A. To this end, we have identified the Spo0E phosphatase as one of the targets of FtsH, and our observations indicate that the Cterminus of Spo0E is necessary for degradation. Furthermore, we show that FtsH is needed only during stage 0. Results In the absence of the FtsH metalloprotease only small amounts of inactive Spo0A are present during the onset of sporulation Based on the analysis of transcriptional fusions, we concluded that ftsH interferes with the synthesis or activity of Spo0A (Deuerling et al., 1997). Next, we attempted to 2 identify the gene(s) responsible for this effect. We first measured the sporulation frequencies in the wild-type and the ftsH knockout strain (Table 1). While about 59% of the cells in our wild-type strain were able to form heat-resistant spores, the sporulation frequency dropped by five orders of magnitude in the absence of the ftsH allele confirming earlier data (Deuerling et al., 1997) and further underlining the importance of the metalloprotease for the sporulation process. Next, we measured the amount of Spo0A in both strains by Western blotting. As can be seen from Fig. 1, Spo0A started to be present from stage 0 on and continued to be produced to at least stage 3. On the contrary, Spo0A is present in greatly reduced amounts in this knockout (Fig. 1). Are the small amounts of Spo0A present in the ftsH knockout active that means present in the phosphorylated form? To answer this question, we constructed a transcriptional fusion between the promoter of the skf operon and the lacZ reporter gene and integrated this fusion ectopically at the amyE locus. It has been reported that small amounts of active Spo0A (Spo0A~P) are sufficient to activate the skf operon (Fujita et al., 2005). When this fusion was analyzed in the wild-type background, the β-galactosidase activity started to increase from stage 0 on (Fig. 2A). When the same operon fusion was tested in the ftsH knockout, only a very low background activity was measured without any increase at least up to t3 (Fig. 2A). We conclude from these results that ftsH interferes with the synthesis or/and activation of Spo0A, where both are interwoven (Strauch et al., 1992). We further assume that FtsH has to degrade one or more proteins which act, directly or indirectly, as negative regulators of the synthesis or/and activation of Spo0A. The ftsH interferes with the expression or activity of three Rap phosphatases Expression and activation of Spo0A is embedded in a sophisticated network involving a plethora of regulators among them three phosphatases termed RapA, RapB and RapE which specifically dephosphorylate Spo0F~P, the second component of the phosphorelay (Perego, 1998). First, we asked whether ftsH influences the sporulation frequencies in the presence or absence of one of the three phosphatases. We constructed knockouts in all three genes as described in the Experimental procedure section. Then, these null alleles were combined with an ftsH knockout each, and all six strains were analyzed for their sporulation frequencies, for the amount of Spo0A present and for its activity status. As can be seen from Table 1, the sporulation frequencies in all three rap knockouts are higher than that of the wild-type strain and range from 67% to 72%, where the sporulation frequency of wild-type cells was determined to be 59%. A similar observation has been published for RapA and RapE (Jiang et al., 2000a). When the sporulation frequencies in the double knockouts were measured, it turned out to be increased by two to three orders of magnitude as compared to a single ftsH null mutant, 3 but remained below 1% (Table 1). These data clearly indicate an influence of the ftsH allele on all three Rap phosphatases. Next, we analyzed all six strains for the production of Spo0A by Western blotting. Fig. 1 shows the results from a representative assay. The absence of any of the three phosphatases in the otherwise wild-type background exhibited a different outcome. While in the absence of both rapA and rapB the amount of Spo0A was increased at t0 as compared to the wild-type situation, its amount was reduced at both t0 and t1 in the rapE knockout (Fig. 1). When the ftsH null allele was added, the amount of Spo0A dropped as already observed for the wild-type strain in the absence of ftsH (Fig. 1). Is the Spo0A protein present in the double knockouts active? To answer this question, the Pskf-lacZ fusion was introduced in all six strains followed by measurement of the β-galactosidase activities of the strains grown in sporulation medium. Expression of the Pskf-lacZ fusions is somewhat different within the six strains. While expression started in all strains at t0 and reached its plateau value at t2 in the wild-type and in the ΔrapB strains, it further increased in the ΔrapA and the ΔrapE strains (Fig. 2). In the presence of ΔftsH, the expression of the operon fusion did not increase over the basal level with the exception of ΔrapE where a slight increase to about 20 units was observed (Fig. 2D). We conclude from these results that both the rapA and the rapE genes are involved in shutting off PskflacZ at t2. The ftsH gene interferes with the phosphorylation status of Spo0A through Spo0E Besides the Rap phosphatases, another set of three phosphates is involved in the specific dephosphorylation of Spo0A~P designated Spo0E, Yis and YnzD, where only the first is active during sporulation (Perego, 2001). While overproduction of Spo0E reduced the sporulation frequency, deletion of spo0E resulted in an increase (Perego and Hoch, 1991). We constructed a spo0E knockout, combined it with the ftsH null allele and measured the sporulation frequencies in both strains. As to be expected the sporulation frequency raised in the absence of the spo0E gene above the level observed in the wild-type strain (Table 1). If combined with an ftsH knockout, the sporulation frequency was increased 1000-fold over the level measured in the ΔftsH strain, but was still about 100-fold lower than the wild-type level (Table 1). Next, we analyzed for the production of Spo0A in both mutant strains. The Western-blot analysis revealed that Spo0A is present in large amounts already at t0 in the Δspo0E strain followed by no significant further increase when cells entered the sporulation pathway (Fig. 1). When the ΔftsH allele was added, the amount of Spo0A was reduced at t0 and further increased to levels comparable to those present in the ftsH+ strain (Fig. 1). This result suggests an 4 interaction between both proteins, either directly or indirectly, thereby influencing expression of spo0A. When we tested for the activity of Spo0A in both mutant strains, it turned out to result in a higher activation of the skf promoter, but completely failed to activate this promoter in the absence of ftsH (Fig. 2E). In conclusion, the absence of an active spo0E allele in an ftsH knockout leads to an expression of the spo0A gene, but the protein remained inactive. These data indicate that ftsH influences production of active Spo0A by either allowing its phosphorylation or preventing for its rapid dephosphorylation. Spo0E is a target protein for FtsH One possibility to explain the interaction between FtsH and Spo0E is a direct one whereby FtsH degrades Spo0E. To test that possibility, we decided to purify both proteins and to incubate them under conditions where FtsH is able to degrade β-casein (Kotschwar et al., 2005). FtsH was purified with a GST-tag as reported before where the purification tag keeps the protein soluble in the absence of any detergent (Kotschwar et al., 2005). Since we failed to overproduce Spo0E equipped with a His-tag (unpublished data), we decided to add the GST-tag as well. Next, both proteins were incubated in the presence and absence of ATP. While in the absence of ATP, the GST-Spo0E remained stable during a 4 h incubation time, it was largely degraded in the presence of the nucleotide (Fig. 3A). This could be verified by probing some lanes with αGST (Fig. 3B). To rule out the possibility that cleavage occurs at or within the GST tag rather than within Spo0E, this tag was purified and incubated with GST-FtsH. It could be shown that GST remained stable for at least 5 h (data not shown). As mentioned above, two homologues of Spo0E, YisI and YnzD, are also able to phosphorylate Spo0E (Perego, 2001). Are these two phosphatases also a substrate of FtsH? While β-casein was completely degraded within 5 h of incubation, both GST-YisI and GST-YnzD remained stable under these conditions (Fig. 4). These data clearly demonstrate that neither YisI nor YnzD are substrates of FtsH. It further confirms that the GST-tag is not recognized by FtsH. The mutant proteins Spo0E11 and Spo0E94 are not degraded by FtsH The spo0E11 and spo0E94 gain-of-function mutations encode overactive phosphatases that inhibit sporulation by specifically dephosphorylating Spo0A~P (Perego and Hoch, 1991;Ohlsen et al., 1994). Both mutations resulted in a stop codon reducing the length of the proteins from 85 to 71 (spo0E11) and 59 amino acids (spo0E94) (Perego and Hoch, 1987;Ohlsen et al., 1994). To find out whether these two shortened versions of the Spo0E protein are still a target for FtsH, both were tagged with 5 GST, overproduced in E. coli and purified. When these two purified proteins were incubated with FtsH, both remained stable for at least 5 h, while β-casein as a control was degraded under these conditions (Fig. 4). Since the full-length Spo0E protein is unstable when incubated with FtsH, we infer that the C-terminal 25 amino acid residues are responsible for this instability. The C-terminal end of Spo0E confer target specificity to FtsH As already mentioned the Spo0E phosphatase is distinguished from the YisI and YnzD phosphatases by a C-terminal extension of about 25 amino acid residues (Perego, 2001). Since Spo0E serves as a target for FtsH, but YisI and YnzD not, we asked whether the C-terminal extension of Spo0E is responsible for recognition by FtsH. To answer this question, we fused the coding region for the C-terminal 25 amino acids to ynzD (YnzD-0E). The GST-tagged hybrid protein was overproduced in E. coli, purified by affinity chromatography and incubated with GST-FtsH. As can be seen from Fig. 6 (lane 6), the YnzD-0E is largely degraded over time. We conclude from this experiment that indeed the C-terminal end of Spo0E contains the recognition sequence for the FtsH protease. Does the absence of spo0E influence expression of yisI or/and ynzD? The yisI and ynzD genes have been reported to be expressed during the vegetative growth phase while spo0E is induced at around t0 (Perego and Hoch, 1987). We asked whether there is a crosstalk between these genes concerning their expression level. Is there increased expression of either yisI or/and ynzD in a spo0E knockout? First, we fused the promoters of the two genes to lacZ and integrated both transcriptional fusions at the amyE locus. Next, the spo0E knockout was introduced into both strains. Then, all four strains (see Table 3) were grown in DSM, samples were taken from t0 up to t3, and the β-galactosidase activity was determined. While no difference was measured for the ynzD promoter independent of the presence or absence of the spo0E allele (Fig. 7B), there was a slight increase in the transcription of the yisI gene (Fig. 7A) in the absence of spo0E. To conclude removal of the spo0E gene did not influence expression of the two other genes significantly excluding a crosstalk at the transcriptional level. Does the spo0A-sad67 allele allow successful sporulation in the ftsH knockout? Several spo0A mutations have been isolated and analyzed among them those which are active in the absence of phosphorylation. One of these mutations, spo0Asad67D56N, carries an internal in-frame deletion removing amino acids 63 through 81 and a point mutation exchanging the aspartate to an asparagine (Ireton et al., 1993). The 6 aspartate residue at position 56 of Spo0A acts as the phosphorylation site (Burbulys et al., 1991) and is dispensable in the spo0A-sad67 allele (Ireton et al., 1993). We asked whether an ftsH knockout strain is able to form spores in the presence of the spo0Asad67D56N allele. We measured the sporulation frequencies in strain SIK190 which carries the spo0A-sad67D56N allele fused to an IPTG-inducible promoter. While a sporulation frequency of 0.06% was measured in the absence of IPTG (Table 2), induction of the mutant allele at t0 resulted in 37% heat-resistant cells. Then, the ftsH::tet knockout was introduced into SIK190 (SIK190F) and the sporulation frequency was determined. While the sporulation frequency was low in the absence of IPTG, it was high after IPTG-induction and both values were comparable to those measured in the ftsH wild-type strain (Table 2). These results strongly suggest that ftsH is needed only during phase 0 and that the spo0A-sad67D56N allele can be expressed in the absence of ftsH. Measure activation of the two promoters (Ps and Pv) preceding the spo0A gene Transcription of the spo0A gene is initiated at two different promoters termed Pv and Ps (Ferrari et al., 1985;Kudoh et al., 1985). While Pv is recognized by the housekeeping sigma factor σA and functions as a low-level promoter to produce a maintenance level of the Spo0A protein during exponential growth (Yamashita et al., 1989), the second promoter, Ps, is recognized by the stationary sigma factor σH. This promoter is required for induction of the protein at the end of exponential growth and during stage 0 (Ferrari et al., 1985;Yamashita et al., 1989) and is further activated indirectly by Spo0A~P, which represses expression of abrB, a negative regulator of sigH (Perego et al., 1988). Therefore, phosphorylation of Spo0A at the onset of sporulation activates an autoregulatory loop leading to an increase in activated Spo0A. We asked whether ftsH or/and spo0E influences transcription at either promoter. Both promoters were separately fused to lacZ and the transcriptional fusions were ectopically integrated at the amyE locus. Then, either the ftsH or the spo0E knockouts or both were added and the β-galactosidase activities were determined in all eight strains. While the βgalactosidase activity initiated at Pv increased slightly up to t1 followed by a modest decrease in the wild-type strain, its activity was reduced to about 50% in the ftsH knockout (Fig. 8A). While the enzymatic activity in the spo0E null mutant was comparable to that measured in the wild-type strain, addition of the spo0E null allele to that of ftsH resulted in a slight increase in the β-galactosidase activity (Fig. 8A). In summary, the influence of both spo0E and ftsH on the Pv promoter is minor. Next, we measured the β-galactosidase activity of lacZ fused to the Ps promoter. In the wild-type background, this promoter is induced about 7-fold between t -1 and t1 (Fig. 8B). In the absence of spo0E, it is induced about 10-fold, while an only 3-fold 7 induction was measured in the ftsH knockout which was not increased in the double knockout (Fig. 8B). To conclude transcription at Ps is strongly reduced in ΔftsH which is not compensated by Δspo0E. In total, ftsH influences only transcription at Ps most probably through the strongly reduced level of active Spo0A which is needed as part of the autoregulatory loop. Discussion When B. subtilis cells enter the transition phase, several different genetic programs are activated including the production of extracellular enzymes and peptide antibiotics, cells become motile and competent, and, as the response of last resort, initiate the process of spore formation. It has been shown that cells either become competent or sporulate, never both together (Errington, 1993). It is also known that never 100% of the cells sporulate. The decision to sporulate or not to sporulate is dependent on the amount of active Spo0A at the end of stage 0 which takes about 2 h. Sporulating cells sense a multitude of mostly unknown signals including the metabolic state, the fate of the chromosomes, the cell density and others, integrate and process these signals by the phosphorelay which controls the level of phosphorylated Spo0A. Cells able to synthesize active Spo0A above a threshold value have been called Spo0A-ON and those which fail to do so Spo0A-OFF (Chung et al., 1994). The formation of two subpopulations of otherwise isogenic cells is designated bistability (Smits et al., 2006;Dubnau and Losick, 2006). But it has to be questioned whether cells in the transition phase exhibit multirather than bistability based on the observation that so different genetic programs are activated. This can be tested by double and triple labelling using transcriptional fusions between program-specific promoters and gfp and its derivatives (Margolin, 2000). Several years ago, we discovered that the sporulation frequency in an ftsH knockout is reduced by about five orders of magnitude (Deuerling et al., 1997). The ftsH gene codes for a membrane-anchored ATP-dependent metalloprotease which seems to be present in all bacterial species (Schumann, 1999;Ogura and Wilkinson, 2001). The objective of this ongoing research project is to elucidate the role of the FtsH protease during sporulation of B. subtilis cells. It is based on the assumption that FtsH has to degrade or to regulate the steady-state level of one or more proteins negatively interfering with the sporulation program. To this end we could show that the activity of FtsH is needed only during phase 0 which culminates in the Spo0A-ON status. Cells carrying an ftsH knockout synthesize a significantly reduced amount of Spo0A which, based on genetic data, is inactive. This observation explains why ftsH null mutants exhibit a dramatically reduced sporulation frequency. But ftsH could also play a role 8 during subsequent sporulation stages. This possibility could be ruled out by introduction of an ftsH null allele in a strain carrying a mutant spo0A allele which is active in the absence of phosphorylation. Upon expression of the spo0A-sad67D56N allele, cells exhibited a normal sporulation frequency. Therefore, the role of ftsH is exclusively confined to the synthesis or/and activation of wild-type Spo0A. The next question to be raised is why the amount of Spo0A is significantly reduced and why this reduced amount is inactive? So far, we used the candidate approach to identify putative substrate proteins of FtsH. First, we constructed knockouts for four different phosphatases where it has already been published that the sporulation frequencies is slightly, but reproducibly enhanced in their absence of two of them (Jiang et al., 2000a); these results could be confirmed and extended for two additional phosphatases. Upon introduction of an ftsH null allele into these four mutant strains, the sporulation frequencies were increased by two to three orders of magnitude, but still remained very low. Not surprisingly, no increase in the amount of active Spo0A could be measured. We would like to conclude that ftsH interferes directly or indirectly with expression or activity of these four phosphatases. To explain the increases in the sporulation frequencies, we further suggest that a few cells within the whole population are in the Spo0A-ON state. To investigate this possibility, we will fuse the promoter of the skf and the spoIIA operons to gfp and analyze single cells under the fluorescence microscope. Whereas the skfA promoter needs a low amount of active Spo0A, the spoIIA one requires a high amount to become activated (Fujita et al., 2005). M. Perego identified two homologs of the Spo0E phosphatase, YisI and YnzD (Perego, 2001). She could show that both phosphatases are able to dephosphorylate Spo0A~P in vitro. These two phosphatases are distinguished from Spo0E by two characteristics: First, their genes are expressed during the vegetative growth phase and second, they lack a C-terminal extension of about 25 amino acid residues. She suggested that this C-terminal extension could be recognized by a protease. Based on her assumption and our finding that an ftsH spo0E double knockout exhibited the highest sporulation frequency of all tested phosphatase null alleles, we asked the question whether the Spo0E protein is a substrate for FtsH. Both proteins were overproduced in E. coli with a GST immobilization tag and incubated under appropriate conditions. We could show that indeed Spo0E is degraded by FtsH. When GST-tagged YisI or YnzD were incubated with FtsH, these proteins turned out to be stable under the same conditions. In the last experiment, we fused the C-terminal extension of Spo0E to YnzD. This fusion protein was shown to be unstable in the presence of FtsH. To conclude, the C-terminal 25 amino acids of Spo0E contain the residues recognized by FtsH. 9 Which amino acids are recognized by the FtsH protease? One of the substrate proteins of the E. coli FtsH protease is LpxC enzyme (Ogura et al., 1999). This enzyme represents the key enzyme in lipopolysacharide (LPS) formation and controls the ration between LPS and phospholipids (Sorensen et al., 1996). Since overproduction of LpxC causes accumulation of abnormal membranes in the periplasm (Ogura et al., 1999) leading to cell death (Sullivan and Donachie, 1984), the amount of LpxC must be carefully regulated which is done by FtsH. Here, the C-terminus has been identified, too, to be responsible for being recognized and degraded by FtsH (Führer et al., 2006). The authors pointed to the about ten amino acids being present at the immediate C-terminus which resemble the SsrA-tag. In the case of Spo0E, there is no similarity to the B. subtilis SsrA-tag (Wiegert and Schumann, 2001). Therefore, the amino acid sequence recognized by the FtsH protease is different from that of the SsrA-tag. Experiments are in progress to identify the amino acid residues of Spo0E recognized by FtsH. Does FtsH fully degrade Spo0E in all cells or does it modulate its steady-state level? Base on our data, we would like to suggest that FtsH regulates the steady-state level of Spo0E rather than completely degrading it. This assumption is based on the observation that a knockout of spo0E leads to in increase in the sporulation frequency which should not occur when Spo0E is completely degraded by FtsH. In conclusion, our results strongly suggest that regulation of stability of several proteins involved directly or indirectly in the synthesis of active Spo0A exerts a new level of posttranslational regulation through the FtsH protease. Another protease has been identified yielding a comparable phenotype. Inactivation of clpP resulted in cells deficient in sporulation initiation and in competence and in a highly filamentous morphology (Msadek et al., 1998;Gerth et al., 1998). In such a mutant, the expression of spo0A and spo0H, coding for the stationary sigma factor σH, was significantly decreased (Nanamiya et al., 2000). Introduction of a mutant spo0E allele into the clpP knockout restored the expression of spo0A, but not sporulation. Based on our results, additional genes have to be identified to influence the synthesis of active Spo0A. These genes will be identified with three different experimental strategies: First, the candidate strategy; second, saturated transposon mutagenesis using pMarA (Le Breton Y. et al., 2006); and third, construction of an ftsH trap mutant (Flynn et al., 2003). Identification of these additional targets might shed some light on the molecular mechanism of bistability. Experimental procedures Bacterial strains, plasmids, media and growth conditions 10 All strains used in this study are listed in Table 3. E. coli DH10B was used for plasmid construction and propagation. E. coli A8926 is a derivative of W3110 used for the expression of GST-tagged proteins. The B. subtilis strain 1012 was used in most of the experiments. All strains were either grown in Luria-Bertani (LB) or in Difco Sporulation medium (DSM). Antibiotics were added when appropriate at the following concentrations: ampicillin, 100 μg ml-1; chloramphenicol, 10 μg ml-1; erythromycin, 50 μg ml–1; neomycin, 10 μg ml–1; kanamycin, 20 μg ml –1; spectinomycin, 100 μg ml–1. Construction of plasmids and recombinant strains All transcriptional fusions were constructed using the integration vector pDG1728 (Guérout-Fleury et al., 1996). This vector contains a promoter-less lacZ and allows insertion of the operon fusions ectopically at the amyE locus. Three different promoters were fused to lacZ generated by PCR using chromosomal DNA of strain 1012 DNA as template. These promoters are Pskf (amplified by primers ON1 and ON2; see Table 4) preceding the skf operon which is activated by a low amount of active Spo0A (Fujita et al., 2005), the vegetative and the stationary phase induced promoters Pv (ON3/ON4) and Ps (ON5/ON6), respectively, of the spo0A gene (Chibazakura et al., 1991). While Pskf was inserted between the EcoRI and HindIII sites of pDG1728, Pv and Ps were ligated into the EcoRI and BamHI sites. Knockouts in the four genes rapA, rapB, rapE and spo0E were constructed as follows. First, the two flanking regions of each gene (about 300 bp) were amplified (see Table 4 for the primer sequences) and inserted into pBluescript SKII+. Next, the chloramphenicol resistance cassette was amplified using pDG364 as template and inserted between the flanking regions of the three rap genes. In the case of the spo0E gene, a phleomycin resistance marker generated plasmid pBlueSKII+-phleo was ligated between the two flanking regions. In the last steps, PCR fragments containing the resistance marker and the flanking regions were transformed into B. subtilis 1012 followed by selection on LB plates containing either chloramphenicol or phleomycin. Chromosomal DNA was prepared from several transformants each and checked by Southern blotting for replacement of the wild-type alleles. One knockout mutant each was kept for further studies. Recombinant vectors allowing overexpression and purification of GST-tagged proteins were prepared using pGEX-2T. The genes ftsH (ON21/22), spo0E (ON23/ON24), yisI (ON25/26) and ynzD (ON29/ON30) were amplified using chromosomal DNA of strain 1012. The mutants spo0E11 and spo0E94 carry stop codons at positions 72 and 60, respectively (Perego and Hoch, 1987). The two truncated versions were generated by amplification of the appropriate coding region (spo0E11: 11 Table 2. Sporulation frequencies of strains expressing spo0A active in the absence of phosphorylationa Strain ftsH IPTGb Viable Spore % of genotype added cell count count sporulation SIK190 + - 6.0 x 106 3.7 x 103 0.06 SIK190 + + 2.5 x 1010 9.3 x 109 37.2 SIK190F - - 4.8 x 106 2.4 x 103 0.05 SIK190F - + 1.7 x 1010 5.7 x 109 33.5 See legend to Table 1 for technical details. a Representative of three different experiments b IPTG was added at a final concentration of 1 mM 18 Table 3. Bacterial strains and plasmids used in this study Plasmid or strain Relevant genotype Source Strains E. coli DH10B mcrA Δ(mrr hsdRMS mcrBC) φ80d lacZM15 ΔlacX74 deoR recA1 araD139 Δ(ara leu)7697 galU galK rpsL endA1 nupG Bethesda Research Laboratories, Inc. A8296 sfhC zad-220::Tn10 ΔftsH3::kan (Tatsuta et al., 1998) B. subtilis 1012 leuA8 metB5 trpC2 hsrM1 (Saito et al., 1979) WW01 1012 ΔftsH::erm (Wehrl et al., 2000) ED04 1012 ΔftsH::tet (Deuerling et al., 1997) AL31 ΔrapA::cat This study AL32 ΔrapA::cat ΔftsH::erm This study AL33 ΔrapB::cat This study AL34 ΔrapB::cat ΔftsH::erm This study AL35 ΔrapE::cat This study AL36 ΔrapE::cat ΔftsH::erm This study AB07 Δspo0E::bleo A. Brandl AB08 Δspo0E::bleo ΔftsH::erm A. Brandl AL37 amyE::Pskf –lacZ spc This study AL38 amyE::Pskf –lacZ spc ΔftsH::erm This study AL39 amyE::Pskf –lacZ spc ΔrapA::cat This study AL40 amyE::Pskf –lacZ spc ΔrapA::cat ΔftsH::erm This study AL41 amyE::Pskf –lacZ spc ΔrapB::cat This study AL42 amyE::Pskf –lacZ spc ΔrapB::cat ΔftsH::erm This study AL43 amyE::Pskf –lacZ spc ΔrapE::cat This study AL44 amyE::Pskf –lacZ spc ΔrapE::cat ΔftsH::erm This study AL45 amyE::Pskf –lacZ spc Δspo0E::bleo This study AL46 amyE::Pskf –lacZ spc Δspo0E::bleo ΔftsH::erm This study 19 AL47 amyE::Pv(spo0A)-lacZ spc This study AL48 amyE::PBv(spo0A)-lacZ spc ΔftsH::erm This study AL49 amyE::PBv(spo0A)-lacZ spc Δspo0E::bleo This study AL50 amyE::Ps(spo0A)-lacZ spc This study AL51 amyE::Ps(spo0A)-lacZ spc ΔftsH::erm This study AL52 amyE::Ps(spo0A)-lacZ spc Δspo0E::bleo This study AL53 amyE::PyisI-lacZ spc This study AL54 amyE::PyisI-lacZ Δ spo0E::bleo This study AL55 amyE::PynzD-lacZ spc This study AL56 amyE::PynzD-lacZ Δ spo0E::bleo This study AL57 amyE::Pv(spo0A)-lacZ spc Δspo0E::bleo ΔftsH::erm This study AL58 amyE::Ps(spo0A)-lacZ spc Δspo0E::bleo ΔftsH::erm This study SIK190 amyE::(Pspac-spo0A-sad67D56N cat), spo0A::erm, Emr, Cmr (Ireton et al., 1993) SIK190F ftsH::tet in SIK190 This study Plasmids pDG1728 Permits transcriptional fusion to lacZ (Guérout-Fleury et al., 1996) p1728-Pskf amyE::Pskf –lacZ This study p1728-Pv amyE::Pv(spo0A)-lacZ This study p1728-Ps amyE::Ps(spo0A)-lacZ This study pGex-2t Expression vector Pharmacia pGST-ftsH Ptac-GST-ftsH This study pGST-spo0E Ptac-GST-spo0E This study pGST-yisI Ptac-GST-yisI This study pGST-ynzD Ptac-GST-ynzD This study pGST-spo0E94 Ptac GST-spo0E94 This study pGST-spo0E11 Ptac-GST-spo0E11 This study pGST-ynzD-C0E Ptac-GST-ynzD-C0E This study pBluecript SKII+Cloning vector Stratagene prapA-cat pBluescript SKII+ with 300 bp upand downstream flanking regions of rapA gene and cat cassette This study prapB-cat pBluescript SKII+ with 300 bp upand This study 20 downstream flanking regions of rapB gene and cat cassette prapE-cat pBluescript SKII+ with 300 bp upand downstream flanking regions of rapE gene and cat cassette This study 21 Table 4. Oligonucleotides used in this study 0ligonucleot ide primer SequencesaDescription ON1 ggccatGAATTCttacaggagacttcattcatt (EcoRI) 5´ skfA promoter ON2 ggccatAAGCTTaagtaaacctcctctcaattttt (HindIII) 3´ skfA promoter ON3 ggccatGAATTCgaaaagtgatcggtgctgtcac (EcoRI) 5´ spo0A vegetative promoter ON4 ggccatGGATCCatcttcttttgtatattttaccgta (BamHI) 3´ spo0A vegetative promoter ON5 ggccatGAATTCattcacgtttccttgtttgtcaaa (EcoRI) 5´ spo0A sporulation promoter ON6 ggccatGGATCCgtttcttcctccccaaatgtagtt (BamHI) 3´ spo0A sporulation promoter ON7 ggccatAAGCTTttgaggatgaagcagacgattccg (HindIII) 5´ rapA upstream ON8 ggccatGAATTCttcgagaagccctgtcagcttgta (EcoRI) 3´ rapA upstream ON9 ggccatGGATCCcgaagcgcaaaaaaagtatcgtga (BamHI) 5´ rapA downstream ON10 ggccatTCTAGAatttcatataaacaatctcctctc (XbaI) 3´ rapA downstream ON11 ggccatAAGCTTatggccgcgtacgagatcccgtca (HindIII) 5´ rapB upstream ON12 ggccatGAATTCatactcagataatccggagatgct (EcoRI) 3´ rapB upstream ON13 ggccatGGATCCgccgcgatacggtactatgaaaa (BamHI) 5´ rapB downstream ON14 ggccatCCGCGGtacttcatataaacaatcgttcct (SacII) 3´ rapB downstream ON15 ggccatAAGCTTttgatatcaatcacatcagctgaa (HindIII) 5´ rapE upstream ON16 ggccatGAATTCgaagttataataatatgcccgcat (EcoRI) 3´ rapE upstream ON17 ggccatGGATCCcaagcaatggattgcttccgcaaa (BamHI) 5´ rapE downstream ON18 ggccatTCTAGAgatttcatacatgcattcccctttcg (XbaI) 3´ rapE downstream ON19 ggccatGAATTCcggatttttcgctacgctcaaatcc (EcoRI) 5´ cat cassette ON20 ggccatGGATCCatcttcaactaacggggcaggtta (BamHI) 3´ cat cassette ON21 ggccatGGATCCaatcgggtcttgcgtaatacc (BamHI) 5´ ftsH 22 ON22 ggccatAGATCTttactctttcgtatcgtctttcttttc (BglII) 3´ ftsH ON23 ggccatGGATCCggcggttcttctgaacaagaaa (BamHI) 5´ spo0E ON24 ggccatCCCGGGattatttatttgcatcatatgctggc (SmaI) 3´ spo0E ON25 ggccatGGATCCaacagtaaaattgaagaaatga (BamHI) 5´ yisI ON26 ggccatCCCGGGattacatacgggagttttcaagat (SmaI) 3´ yisI ON27 ggccatCCCGGGattattgggaatgttcgttttcttgcata (SmaI) 3´ 75-nucleotides-shorten spo0E ON28 ggccatCCCGGGattaccacaagcctaatttctttacaaggc (SmaI) 3´ 39-nucleotides-shorten spo0E ON29 ggccatGGATCCattagagagcatctattaaaag (BamHI) 5´ ynzD ON30 ggccatCCCGGGttcatcacccgctactgctcga (SmaI) 3´ ynzD ON31 ggccatGAATTCaattcaaggccttgtaaagaaattag (EcoRI) 5´ Cteminus of spo0E ON32 ggccatGAATTCattatttatttgcatcatatgctggc (EcoRI) 3´ Cteminus of spo0E a restriction endonuclease sites are shown in capital letters 23 Fig. 1. Amount of Spo0A in different B. subtilis strains. The strains were grown in DSM at 37°C, and aliquots were taken after entry into the transition phase (t0) and up to 3 h later (t1 to t3). Cells were lysed by sonication and equal amounts of proteins were applied per lane (5 µg). First line: wild-type 1012 and WW01 (ΔftsH); second line: AL31 (ΔrapA) and AL32 (ΔrapA ΔftsH); third line: AL33 (ΔrapB) and AL34 (ΔrapB ΔftsH); fourth line: AL35 (ΔrapE) and AL36 (ΔrapE ΔftsH); fifth line: AB07 (Δspo0E) and AB08 (Δspo0E ΔftsH). 24 Fig. 2. Transcription from the Spo0A~P-activated promoter skf. Cells containing the Pskf-lacZ fusion integrated ectopically at the amyE locus were grown in DSM at 37°C, and aliquots (5 OD578 units) were withdrawn for measurement of β-galactosidase activities. (A) AL37 (ftsH+) and AL38 (ΔftsH); (B) AL39 (ΔrapA ftsH+) and AL40 (ΔrapA ΔftsH); (C) AL41 (ΔrapB ftsH+) and AL42 (ΔrapB ΔftsH); (D) AL43 (ΔrapE ftsH+) and AL44 (ΔrapE ΔftsH); (E) AL45 (Δspo0E ftsH+) and AL46 (Δspo0E ΔftsH). ■ ftsH+, ○ ΔftsH. 25 A B Fig. 3. Spo0E acts as a substrate for FtsH. Purified GST-FtsH was incubated with GSTSpo0E under conditions described in the Experimental procedures. (A) Aliquots of the reaction mixtures were separated by SDS-PAGE and stained with Coomassie brilliant blue. (B) Western blot using αGST and showing GST-Spo0E. The incubation time for all three samples was 4 h. 26 A B Fig. 4. The phosphatases YisI and YnzD are stable in the presence of FtsH. GST-FtsH was incubated with (A) GST-YisI and (B) GST-YnzD. β-casein served as a control. Fig. 5. Two mutant Spo0E proteins are stable in the presence of FtsH. GST-tagged Spo0E94 and Spo0E11 proteins were incubated with FtsH. The reaction products were resolved by SDS-PAGE and stained with Coomassie blue. 27 Besides serving as a membrane anchor, a second function has been suggested for SpoVM. When a transposon insertion within spoVM which is biologically inactive was used to identify possible interacting proteins, the membrane-bound FtsH metalloprotease was identified (3). Mutations in ftsH suppressed the sporulation defect of certain spoVM mutants but not others, another case of allele-specific extragenic suppressors. Furthermore, it could be shown that chemically synthesized SpoVM is able to inhibit degradation of σ32 by purified E. coli FtsH (3). Based on these findings, one can assume that a second function of SpoVM is to inhibit the B. subtilis FtsH protease late during sporulation. This assumption is sustained by two observations: First, FtsH-GFP has been shown to accumulate within the asymmetric septum (28) and, second, SpoVM-GFP colocalizes with the polar septum, too (26). It can be inferred that, in the absence of SpoVM, FtsH will degrade at least one protein essential to complete successful sporulation or, alternatively, regulate the steady-state level of SpoVM. The objective of the current work was to analyze whether SpoVM inhibits the protease activity and to study expression of the spoVM gene. Purified GST-FtsH was incubated with or without the SpoVM peptide for up to 4 h (Fig. 1). While the band of β-casein partly disappeared after 3 h of incubation in the presence of GST-FtsH (lane 2), it was almost completely absent 4 h after of incubation (lane 4). Most interestingly, the β-casein was not completed degraded, but converted into distinct degradation products. If the SpoVM peptide was present, the β-casein turned out to be stabilized (lanes 3 and 5). In the absence of GST-FtsH, β-casein remained stable (lane 6). In conclusion, GST-FtsH is able to degrade the substrate protein β-casein into specific fragments, and this proteolytic activity can be inhibited by the SpoVM peptide. In some experiments, we observed the partial disappearance of SpoVM (data not shown) as described for the bacteriophage λ encoded CIII peptide (10, 12), where a short domain (residues 16-37) may form an amphipathic α-helix which is essential for CIII activity (13). Interestingly, SpoVM was also predicted to form such an amphipathic αhelix, though it displays no sequence similarity with CIII (16, 17). We infer from these data that the essential SpoVM peptide, by interaction with FtsH, either prevents degradation of one or more proteins essential for successful spore formation or that FtsH influences the steady-state level of SpoVM. Both possibilities are not mutually exclusive. To construct a spoVM knockout, about 300 bp each of its upand downstream region were amplified by PCR and ligated into pBR322 (2). The upstream region was generated using the primer pair ON01 and ON02 (Table 1), the downstream region ON03 and ON04 and chromosomal DNA of strain 1012 (18) as template. While the amplicon representing the upstream region was flanked by EcoRI and SmaI sites, the 2 downstream amplicon was flanked by SmaI and HindIII sites allowing their insertion into EcoRI and HindIII cleaved pBR322 resulting into pMB02. In the next step, a spectinomycin resistance cassette was amplified using the primer pair ON05 and ON06 and pK2-spec (9) as template and ligated into SmaI linearized pMB02 resulting in pMB03. Then, pMB03 was transformed into B. subtilis 1012 where the plasmid is unable to replicate. Transformed cells were plated on LB agar plates containing spectinomycin to select for those cells where the spoVM wild-type allele has been replaced by the knockout allele. Chromosomal DNA from several candidates was prepared and checked by Southern blotting for successful replacement (data not shown). Strain MB03 was kept for further studies. Next, we determined the sporulation frequency of the knockout strain by growing cells in DSM, a sporulation medium (19), at 37°C for 36 h, heated them for 20 min to 80°C and then plated for survivors as described (5). Whereas about 82% of the cells of the wild-type strain 1012 were able to sporulate, cells of strain MB03 turned out to be completely deficient in sporulation (less than 10-6). These data are in agreement with previously published results where B. subtilis strains with mutant spoVM alleles were unable to produce heat-resistant spores (3, 14). Since SpoVM interferes with the proteolytic activity of FtsH, we examined the possibility that FtsH will degrade one or more proteins essential to complete spore formation. This protein(s) could be located either in the outer spore membrane or in the cytoplasm of the mother cell. In a first attempt to identify this protein(s), strains 1012 and MB03 were grown in DSM to t5. In one experiment, the membranes of the mother cell and of the prespore were isolated as described (1) and the membrane proteins resolved by SDS-PAGE. A few protein bands could be identified which are either absent or present in the spoVM knockout (data not shown). Experiments are in progress to identify these proteins in collaboration with the group of Dr. M. Hecker, University of Greifswald. In a second experiment, we compared the cytoplasmic proteomes of strains 1012 and MB03 using the 2D-gel electrophoresis technique. Here, a total of 83 protein spots were present in the wild-type, but completely absent or present in reduced amounts in the spoVM null mutant. One of these proteins turned out to be SpoIVA which seems to be completely absent in the spoVM knockout (Fig. 2). This raises the interesting question whether SpoIVA is unstable in the absence of SpoVM and which ATP-dependent protease is responsible for its degradation. If SpoIVA is a substrate for FtsH, it will be interesting to find out whether tethering of SpoIVA to membrane-bound SpoVM will be sufficient to prevent its degradation or whether SpoVM has to directly interact with the protease acting as an antiprotease as described for the phage λ CIII protein (8). 3 The spoVM gene forms a monocistronic operon, and its transcription is controlled by the sporulation-specific sigma factor σE which acting in conjunction with the DNAbinding protein SpoIIID, a 93-amino-acid protein (14). To find out when transcription of spoVM is initiated, cells of strain 1012 were grown in DSM at 37°C into stationary phase. Aliquots were taken immediately upon entering the stationary phase (t0) and up to 5 h later (t1 to t5). As can be seen from Fig. 3, the spoVM transcript (about 200 nucleotides in length) started to appear after t2 and continued to be produced in large quantities at least until t5. Using a spoVM-lacZ transcriptional fusion, a similar expression pattern was described (14). When total RNA isolated from the spoVM knockout at t5 was analyzed, no signal was obtained as expected (Fig. 3). Next, we wanted to find out when the SpoVM peptide can be detected in sporulating B. subtilis cells. Cells of strain 1012 were grown again as described before and aliquots where prepared for a Western blot. As shown in Fig. 4, the SpoVM peptide started to accumulate at t4 (lane 4) and further increased during the next 2 h (lane 5 and 6). No SpoVM peptide could be detected in the null mutant strain at t6 (lane 7). Chemically synthesized SpoVM served as a positive control (lane 8). All aliquots were also checked for the presence of the heat shock protein HtpG (20) which served as a control for a protein not subject to sporulation regulation (Fig. 4). In summary, these results clearly demonstrate that while the spoVM transcript is present about 3 h after entry into the transition phase, synthesis of the SpoVM peptide is delayed by 2 h suggesting posttranscriptional regulation of spoVM (see below). The lag-phase of about 2 h between the onset of spoVM transcription and translation suggests a so far unknown mechanism that prevents early appearance of the peptide. This could involve stabilization of the transcript, delayed translation initiation or stability of the peptide. A close inspection of the spoVM transcript reveals an 87 nucleotide - untranslated region (5' UTR) preceding the coding region (Fig. 5A). We asked whether this 5' UTR is involved in expression of spoVM. Two different translational fusions using lacZ as a reporter gene were constructed one with and the other without the untranslated region. In both cases, the 9th codon of spoVM was fused to the 8th codon of lacZ. To discriminate between cisand trans-acting factors, we first analyzed the two fusions in E. coli. This decision was based on the assumption that trans-acting factors are completely absent from this host. When both fusions were analyzed in E. coli, between 30 and 50 units of β-galactosidase activity were measured during the exponential growth phase with no difference between the two fusions (Fig. 5B). This result indicates that the 5' UTR does not act as a cis-acting factor in E. coli, e.g. by influencing the stability of the transcript. 4 Next, we investigated the influence of the 5' UTR in B. subtilis grown in DSM. In the presence of the untranslated region the β-galactosidase activity started to rise at t2 and increased from about 7 to 20 units at t5 and dropped thereafter (Fig. 5C). When the translational fusion without the 5' UTR was analyzed, expression of the reporter gene started at t2, too, but at a significantly higher level (about 5-fold) and further increased at later sporulation times (Fig. 5C). In summary, the 5' UTR of spoVM negatively influences its own transcription or translation in B. subtilis. Since this effect was not observed in E. coli, we conclude that a trans-active regulator is involved in transcription or translation of spoVM which is not present in E. coli. Yet another possibility is that the 5' UTR affects the stability of the transcript. We regard this mechanism as rather unlikely because such an effect could not be observed in E. coli. The factor influencing regulation at the 5' UTR could be a translational repressor protein binding within the untranslated region, e.g., at the inverted repeat (Fig. 5A). Alternatively, a non-coding (nc) RNA may interact with the spoVM transcript thereby reducing its translation. Recently, several ncRNAs have been described which are under sporulation control (21). In conclusion, the SpoVM peptide exerts at least two functions. First, as shown by R. Losick and coworkers, it adheres to the outer forespore membrane via hydrophilic amino acid side-chains on the hydrophobic face of the helix (17). Then, it will recruit the SpoIVA protein, a morphogenetic protein that forms the basement layer of the spore coat. Whether really SpoVM first adheres to the membrane and then tethers SpoIVA is not clear. Alternatively, both components could interact in the cytoplasm and then binds to the membrane. Second, SpoVM, at least in vitro, inhibits the proteolytic activity of FtsH, and we can assume that it does the same in vivo. Since we could show that FtsH is essential for cells to enter the sporulation program (5), an early synthesis of SpoVM could prevent cells go beyond stage 0. Why FtsH has to be inhibited by SpoVM? At least two possibilities can be considered, which are not mutually exclusive. First, FtsH might fine-tune the amount of SpoVM and thereby prevent accumulation of increased amounts which might be deleterious for the cells as has been shown for the LpxC protein (15). Second, by binding to FtsH SpoVM may prevent degradation of at least one protein needed to complete the sporulation program. Since SpoIVA is absent in a spoVM knockout as revealed by a 2D-gel analysis, this morphogenetic protein might be a substrate for FtsH and its interaction with SpoVM will protect it from degradation. We would like to thank Prof. M. Hecker and his group for their generous help with the 2D-gel electrophoresis and the identification of protein spots by mass spectrometry. We also appreciate the help of Monika Batzer with some of the experiments. This project was supported by the Deutsche Forschungsgemeinschaft (Schu 414/20-2). 5 References 1. Bagyan, I., M. Noback, S. Bron, M. Paidhungat, and P. Setlow. 1998. Characterization of yhcN, a new forespore-specific gene of Bacillus subtilis. Gene 212:179-188. 2. Bolivar, F., R. Rodriquez, M. Betlach, and H. Boyer. 1977. Construction and characterization of new cloning vehicles I: Ampicillin-resistant derivatives of the plasmid pMB9. Gene 2:75-93. 3. Cutting, S., M. Anderson, E. Lysenko, A. Page, T. Tomoyasu, K. Tatematsu, T. Tatsuta, L. Kroos, and T. Ogura. 1997. SpoVM, a small protein essential to development in Bacillus subtilis, interacts with the ATP-dependent protease FtsH. J. Bacteriol. 179:5534-5542. 4. Cutting, S. M. and P. B. Vander Horn. 1990. Genetic analysis, p. 27-60. In C. R. Harwood and S. M. Cutting (ed.), Molecular biological methods for Bacillus. John Wiley & Sons, Chichester. 5. Deuerling, E., A. Mogk, C. Richter, M. Purucker, and W. Schumann. 1997. The ftsH gene of Bacillus subtilis is involved in major cellular processes such as sporulation, stress adaptation and secretion. Mol. Microbiol. 23:921-933. 6. Driks, A. 1999. Bacillus subtilis spore coat. Microbiol. Mol. Biol. Rev. 63:1-20. 7. Eymann, C., A. Dreisbach, D. Albrecht, J. Bernhardt, D. Becher, S. Gentner, l. T. Tam, K. Buttner, G. Buurman, C. Scharf, S. Venz, U. Volker, and M. Hecker. 2004. A comprehensive proteome map of growing Bacillus subtilis cells. Proteomics. 4:2849-2876. 8. Halder, S., A. B. Datta, and P. Parrack. 2007. Probing the antiprotease activity of CIII, an inhibitor of the Escherichia coli metalloprotease HflB (FtsH). J. Bacteriol. 189:8130-8138. 9. Härtl, B., W. Wehrl, T. Wiegert, G. Homuth, and W. Schumann. 2001. Development of a new integration site within the Bacillus subtilis chromosome and construction of compatible expression cassettes. J. Bacteriol. 183:2696-2699. 10. Herman, C., D. Thévenet, R. D'Ari, and P. Bouloc. 1997. The HflB protease of Escherichia coli degrades its inhibitor lambdacIII. J. Bacteriol. 179:358-363. 11. Homuth, G., S. Masuda, A. Mogk, Y. Kobayashi, and W. Schumann. 1997. The dnaK operon of Bacillus subtilis is heptacistronic. J. Bacteriol. 179:1153-1164. 12. Kobiler, O., S. Koby, D. Teff, D. Court, and A. B. Oppenheim. 2002. The phage lambda CII transcriptional activator carries a C-terminal domain signaling for rapid proteolysis. Proc. Natl. Acad. Sci. USA 99:14964-14969. 13. Kornitzer, D., S. Altuvia, and A. B. Oppenheim. 1991. The activity of the CIII regulator of lambdoid bacteriophages resides within a 24-amino acid protein domain. Proc. Natl. Acad. Sci. USA 88:5217-5221. 6 14. Levin, P. A., N. Fan, E. Ricca, A. Driks, R. Losick, and S. Cutting. 1993. An unusually small gene required for sporulation by Bacillus subtilis. Mol. Microbiol. 9:761-771. 15. Ogura, T., K. Inoue, T. Tatsuta, T. Suzaki, K. Karata, K. Young, L.-H. Su, C. A. Fierke, J. E. Jackman, C. R. H. Reatz, J. Coleman, T. Tomoyasu, and H. Matsuzawa. 1999. Balanced biosynthesis of major membrane components through regulated degradation of the committed enzyme of lipid A biosynthesis by the AAA protease FtsH (HflB) in Escherichia coli. Mol. Microbiol. 31:833-844. 16. Prajapati, R. S., T. Ogura, and S. M. Cutting. 2000. Structural and functional studies on an FtsH inhibitor from Bacillus subtilis. Biochim. Biophys. Acta Gen. Subj. 1475:353-359. 17. Ramamurthi, K. S., K. R. Clapham, and R. Losick. 2006. Peptide anchoring spore coat assembly to the outer forespore membrane in Bacillus subtilis. Mol. Microbiol. 62:1547-1557. 18. Saito, H., T. Shibata, and T. Ando. 1979. Mapping of genes determining nonpermissiveness and host-specific restriction to bacteriophages in Bacillus subtilis Marburg. Mol. Gen. Genet. 170:117-122. 19. Schaeffer, P., J. Millet, and J. P. Aubert. 1965. Catabolic repression of bacterial sporulation. Proc. Natl. Acad. Sci. USA 54:704-711. 20. Schulz, A., S. Schwab, S. Versteeg, and W. Schumann. 1997. The htpG gene of Bacillus subtilis belongs to class III heat shock genes and is under negative control. J. Bacteriol. 10:3103-3109. 21. Silvaggi, J. M., J. B. Perkins, and R. Losick. 2006. Genes for small, noncoding RNAs under sporulation control in Bacillus subtilis. J. Bacteriol. 188:532-541. 22. Stragier, P. and R. Losick. 1996. Molecular genetics of sporulation in Bacillus subtilis. Annu. Rev. Genet. 30:297-341. 23. Tam, T., H. Antelmann, C. Eymann, D. Albrecht, J. Bernhardt, and M. Hecker. 2006. Proteome signatures for stress and starvation in Bacillus subtilis as revealed by a 2-D gel image color coding approach. Proteomics. 6:4565-4585. 24. Teff, D., S. Koby, Y. Shotland, T. Ogura, and A. B. Oppenheim. 2000. A colicintolerant Escherichia coli mutant that confers Hfl phenotype carries two mutations in the region coding for the C-terminal domain of FtsH (HflB). FEMS Microbiol. Lett. 183:115-117. 25. Tomoyasu, T., J. Gamer, B. Bukau, M. Kanemori, H. Mori, A. J. Rutman, A. B. Oppenheim, T. Yura, K. Yamanaka, H. Niki, S. Hiraga, and T. Ogura. 1995. Escherichia coli FtsH is a membrane-bound, ATP-dependent protease which degrades the heat-shock transcription factor 32. EMBO J. 14:2551-2560. 26. Van Ooij, C. and R. Losick. 2003. Subcellular localization of a small sporulation protein in Bacillus subtilis. J. Bacteriol. 185:1391-1398. 27. Warrens, A. N., M. D. Jones, and R. I. Lechler. 1997. Splicing by overlap extension by PCR using asymmetric amplification: an improved technique for the generation of hybrid proteins of immunological interest. Gene 186:29-35. 7 28. Wehrl, W., M. Niederweis, and W. Schumann. 2000. The FtsH protein accumulates at the septum of Bacillus subtilis during cell division and sporulation. J. Bacteriol. 182:3870-3873. 8 TABLE 1. Oligonucleotides used Primer Sequence (5' to 3')a ON01 GGCCAT GAATTC GAGCTGATCATTTTTTAGGAAAC; EcoRI ON02 GGCCAT CCCGGG AAACGAAAAAGTACCTCGTGAAT; SmaI ON03 GGCCAT CCCGGG TTCAAAGCCCTCTTTCACCACAT; SmaI ON04 GGCCAT AAGCTT TGAAAGATGATGAAACAATAGTTGC; HindIII ON05 GGCCAT CCCGGG CGATTTGACATTTTTCTTGTG, SmaI ON06 GGCCAT CCCGGG ATCAATAGTTACAAATTCTTTCA; SmaI ON07 GGCCAT GGATCC CTGGCCGTCGTTTTACAACGT; BamHIII ON08 GGCCAT GGATCC TTATTTTTGACACCAGACCAACTGGTAAT; BamHI ON09 GGCCAT GGATCC AATATCCTCTAAATAATTGTCATAT; BamHIII ON10 GGCCAT AAGCTT CGGCAATTTAATGGTGTAAAATTT; HindIII ON11 AAAGCCATATTAATAATGATAAGTATAGGAGGGGACAAAAATG ON12 CTTATCATTATTAATATGCCTTTT a G/C clamps are shown in boldface; restriction sites are underlined 9 FIG. 1. The SpoVM peptide retards degradation of β-casein by the FtsH protease. The SpoVM peptide has been prepared by chemical synthesis (KLH; Peptide Speciality Laboratories, Heidelberg, Germany) and FtsH has been purified as a GST-tagged protein as described (24). Incubation of the different components followed a published method (25). The complete reaction mixture (30 µl) consisted of the following components: 50 mM Tris-acetate (pH 8.0), 5 mM magnesium acetate, 12.5 µM zinc acetate, 80 mM NaCl, 1.4 mM βmercaptoethanol, 5 mM ATP, 100 µg/ml bovine serum albumin (BSA), 200 µg/ml of purified SpoVM peptide, 50 µg/ml β-casein and 50 µg/ml of purified B. subtilis GSTFtsH. SpoVM peptide, β-casein, and GST-FtsH were present at a molar ratio of 132:4:1. Reactions were performed at 40°C for the time points indicated. Aliquots of the reaction mixtures were analyzed by 15% SDS-PAGE followed by staining with Coomassie blue. 10 FIG. 2. The cytoplasmic proteomes of B. subtilis wild-type (A) and ΔspoVM::spec (B) strains. Strains 1012 and MB03 were grown in DSM at 37°C into stationary phase. Cytoplasmic proteins were separated by two-dimensional (2D) gel electrophoresis using immobilized pH gradients (IPG) in the range 4-7 as described (23). For identification of the proteins by mass spectrometry, the 2D gels were stained with Colloidal Coomassie brilliant blue (Amersham Biosciences). Spot cutting, tryptic digestion of the proteins and spotting of the resulting peptides onto the MALDI targets were performed as described (7). Here, only a small part of the gel is shown. The white arrow indicates the position of the SpoIVA protein. 11 leading to a dephosphorylation of DesR with a concomitant turn off of the des gene [13]. Based on these data, we developed a cold-inducible expression system for B. subtilis making use of the des promoter. We show here that cold-induction results in a significant induction of reporter genes largely preventing formation of aggregates of an aggregation-prone protein. Cold-inducible expression systems have also been developed for Escherichia coli which are based on a different principle [14,15]. Cold-inducible expression systems provide an inexpensive alternative technology especially for industrial production of recombinant proteins complementing the widely used IPTGand xylose-inducible systems. Materials and methods Materials Bacteria, plasmids and growth conditions Bacterial strains and plasmids used are listed in Table 1. Cells were grown in Luria Broth (LB) medium at 37 or 25 °C under aeration. Antibiotics were added where appropriate (ampicillin at 100 lg/ml, neomycin at 10 lg/ml and chloramphenicol at 10 lg/ml). Methods Construction of a transcriptional fusion between the des promoter and the lacZ reporter gene The promoter region of the des gene was fused to the lacZ reporter gene using the integration vector pDG1728 [16]. The resulting transcriptional fusion is sandwiched between amyE-front and amyE-back allowing its integration into the B. subtilis chromosome at the amyE locus. The des promoter region was amplified using oligonucleotides (ON) ON1 (GGCCATGAATTCTCCGGCATCCC GATCATCGC; restriction site underlined) and ON2 (GGCCATAAGCTTTCTCATTGTGTGTCTCGGTTC AG). The amplicon was cleaved with EcoRI and HindIII and inserted into pDG1728 cut with the same enzymes resulting in pDG1728-des. This recombinant plasmid was transformed into strain WW02, and transformants were selected on LB plates containing chloramphenicol and screened for the loss of the neomycin resistance marker, and one positive transformant (AL03) was kept for further studies. Construction of a des null mutant To construct a des knockout, the gene including flanking regions was amplified using the primer pairs ON3/ON4 (GGCCATGTCGACTGAACCGAGACACACAATG; GGCCATGAGCTCATAGTTGAGCACCTTTGG), and the amplicon was cleaved with SalI and SacI and cloned into pBluescript SKII + treated with the same enzymes. Next, the recombinant plasmid was treated with HindIII and BclI to remove a 61-bp internal fragment of des which was replaced by the neo marker using pBgaB as template and the primer pair ON5/ON6 (GGCCATAAGCTT AGGTCGAGATCAGGGAATGAGTT; GGCCATTGA TCAGATCAATTCTGACAGCCATG). Using the primer pair ON3/ON4, the modified gene was amplified and transformed into B. subtilis 1012. Neomycin-resistant transformants were selected and checked by Southern-blot for chromosomal replacement of the des by the neo gene (data not shown). One strain (AL02) was kept for further studies. Table 1 Strains and plasmids used in this study Strains Genotype Reference/source E. coli DH10B F  mcrA D(mrr hsdRMS mcrBC)u80d lacZ DM15 deoR recA1 araD139 D(ara leu) 7697 galU galK k  rpsL endA1 nupG BRL B. subtilis 1012 leuA8 metB5 trpC2 hsrM1 [26] WW02 1012 amyE::neo [27] AL02 1012 des::neo This work AL03 1012 amyE::Pdes-lacZ This work AL04 AL02 des::neo amyE::Pdes-lacZ This work AL05 AL02 htpG::erm This work AL06 AL02 pbpE::erm This work Plasmids pBluescript SK + Cloning vector Stratagene pBgaB Integration vector containing the bgaB gene [28] pDG1728 Vector allowing integration of DNA sequences at the amyE locus [16] pHT01 Derivative of pNDH33 without a direct repeat [22] pKTH10 Recombinant vector containing the amyQ gene [17] pAL10 Expression vector allowing cold-inducible intracellular production of recombinant proteins This work pAL12 Expression vector allowing cold-inducible secretion of recombinant proteins This work pNDH33-htpG htpG fused to an IPTG-inducible promoter [22] A.T. Thuy Le, W. Schumann / Protein Expression and Purification 53 (2007) 264–269 265 The Pdes–lacZ fusion was introduced into AL02 by transformation resulting in AL04. Construction of the two expression vectors pAL10 and pAL12 We started from the vector pHT01, where lacI,Pgrac and bgaB were removed by SacI/BamHI digestion followed by religation with the des promoter region including the binding sites for DesR generated by PCR using ON7/ON8 (GGCCATGAGCTCTCCGGCATCCCGAT CATCGC; GGCCATGGATCCTCTTGATCGCCTCCT CATTGTGTGTCTCGG) and resulting in the new expression vector pAL10 (Fig. 1a). This vector allows intracellular production of recombinant proteins. A second vector allowing secretion of proteins was obtained by fusing the signal sequence of the amyQ gene [17] to the des promoter (ON9/ON10 (GGGCCCATGGATCCATGATTCAA AAACGAAAGCGGACAG; GGCCATTCTAGATTTT TCTGAACATAAATGGAGACG) and pKTH10 as template) resulting in the expression-secretion vector pAL12 (Fig. 1b). To test suitability and efficacy of the new expression vectors, different genes were fused to the des promoter and the synthesis was monitored after temperature downshock to 25 °C. pAL10 was tested by insertion of htpG, coding for a heat shock protein of unknown function [18],andpbpE encoding the penicillin-binding protein Pbp4 * [19]. The htpG gene was amplified using ON11/12 (GGCCATGGATCCATGGCGAAAAAAGAGTTTAAA GC; GGCCATTCTAGATTACACCATGACCTTGCAA ATATTGTTCG), pbpE ON13/14 (GGCCATGGATCCA TGAAGCAGAATAAAAGAAAGC; GGCCATGGATC CTTACTACTTCGTACGGACCGCTTCT) and chromosomal DNA of B. subtilis 1012 as template. To analyse for the versatility of pAL12, the coding region for amyQ [17] was inserted (ON15/16 (GGCCATTCTAGAGTAA ATGGCACGCTGATGCAGT; GGCCATCCCGGGTT ATTTCTGAACATAAATGGAGACG) and pKTH10 as template). Determination of enzymatic activities and Western blot analysis The b-galactosidase activities encoded by lacZ was determined as described elsewhere [20], with the exception that LacZ activity was measured kinetically in a microplate reader (VersaMax, Molecular Devices) at 405 nm at 28 °C. One unit was defined as DE405 min1OD1 578 103,in which OD 578 is the optical density of the growth culture when samples were drawn. Western blot analyses were carFig. 1. Genetic and restriction map of the two vectors pAL10 and pAL12 allowing intraand extracellular expression of recombinant proteins, respectively. (a) pAL10 and the DNA sequence of the Pdes promoter (in capital letters); (b) pAL12 and the DNA sequence of Pdes, the ribosome-binding site (underlined) and the coding region for the signal sequence (highlighted in grey). Unique restriction sites which can be used for insertion of recombinant genes are also presented. 266 A.T. Thuy Le, W. Schumann / Protein Expression and Purification 53 (2007) 264–269 ried out as published [21]. Blots were developed with polyclonal antibodies against HtpG and Pbp4 * used at a dilution of 1:10,000. Results and discussion Construction and analysis of an operon fusion between the des promoter and lacZ Based on published data [11], we devised a cold-inducible expression system consisting of the regulatory region of the des gene consisting of the des promoter and its upstream region serving as binding sites for DesR P. This region was cloned into the integration vector pDG1728 [16] followed by insertion at the amyE locus. Strain AL03 was first grown at 37 °C to the early exponential growth phase. Then, the culture was divided into two subcultures where one was further kept at 37 °C while the second was challenged with a cold shock to 25 °C. Aliquots were withdrawn just before dividing the culture (t= 0) and up to 12 h post-induction for determination of the b-galactosidase activities. As shown in Fig. 2, the enzymatic activity of the unshocked culture started with about 4 units and increased to about 15 units over time. In contrast, the coldshocked culture exhibited an increase to about 105 units after 1.5 h followed by a decrease to 50 units 5 h after induction (Fig. 2). The decrease can be explained by induction of the des gene from the chromosome restoring the fluidity of the membrane followed by turning off expression of des gene [11]. It should be possible to prevent turning off expression of the des gene by deleting this gene from the chromosome as published [11]. This has been done as described under Materials and methods, and the operon fusion was integrated at the amyE locus of strain AL03 where the des gene has been replaced by a neomycin resistance marker. When this strain AL04 was analysed, the b-galactosidase activity was even lower during growth at 37 °C, while the activity increased from 7 to about 100 units within the first 12 h after cold challenge (Fig. 2). This result clearly demonstrates that no turn off of the lacZ expression occurs in the absence of the desaturase. Based on this finding we asked whether a further increase in the enzymatic activity can be obtained upon prolonged incubation. As shown in Fig. 3, while the OD 578 continued to increase steadily for at least 58 h, the b-galactosidase activity increased up to about 12 h and decreased thereafter. This result indicates that either the half-live of the enzyme or/and the synthesis capacity of the cells decrease during prolonged incubation at 25 °C. We also measured the b-galactosidase activity after temperature downshift to 20 and 15 °C. While about 100 units were measured after 5 h of growth at 25 °C (Fig. 2), 60 units and 5 units were determined at 20 and 15 °C, respectively (data not shown). This results suggests that growth at 20 °C reduces the expression level to 60%, while expression of the lacZ gene is completely abolished at 15 °C. The expression vector pAL10 allows production of recombinant proteins to a significant level Next, we attempted to directly visualize the amount of recombinant proteins produced. To accomplish this goal, two different genes were fused to Pdes in the expression vector pAL10, namely the htpG and the pbpE gene coding for a heat shock protein of unknown function and a penicillinbinding protein, respectively [18,19]. Both strains (the chromosomal copies of htpG and pbpE have been deleted) were grown in LB medium to the mid-exponential growth phase, divided into two subcultures where one was further incubated at 37 °C, while the second was cold-shocked to 25 °C. Aliquots were withdrawn at different time points for the analysis of the presence of the HtpG or Pbp4 * protein as indicated. While no HtpG protein was visible when the strain AL05 containing the plasmid pAL10-htpG was incubated at 37 °C, this protein became apparent already after 3 h and increased in its amount up to 9 h (Fig. 3a) Fig. 2. Induction of b-galactosidase activity in two B. subtilis strains grown at two different temperatures. (a) B. subtilis strains AL03 and AL04 (Ddes::neo) were grown in LB medium at 37 °C to the early logarithmic growth phase. Then, the cultures were divided into two subcultures (at t= 0) where one was further grown at 37 °C, while the second was challenged with 25 °C. Aliquots were removed for determination of b-galactosidase activities at the time points indicated. The complete experiments were repeated three times and yielded comparable results. Data from one of these experiments are presented. AL03 grown at 37 °C (d)or25°C(s); AL04 grown at 37 °C(j)or25°C(h). (b) B. subtilis strain AL04 was grown up to 58 h after the temperature downshock. The OD 578 was measured during growth (r) and the b-galactosidase activities as indicated (white columns). A.T. Thuy Le, W. Schumann / Protein Expression and Purification 53 (2007) 264–269 267 where it represented about 10% of the total cellular protein. When the htpG gene was expressed from an IPTG-inducible promoter for 6 h at 25 °C, only tiny amounts of the HtpG protein became visible (Fig. 3a, lane 10). We conclude from this result that the expression level is rather low at 25 °C from the IPTG-inducible promoter and can be compensated from a cold-inducible promoter. In parallel, we visualized HtpG by Western blot. While small amounts were present at 37 °C most probably due to the leakiness of the promoter, it increased dramatically up to 9 h after cold-shock (Fig. 3b). The Pbp4 * protein has been reported to be membraneattached due to one or more hydrophobic patches [19]. We could already show that overproduction of this protein at 37 °C leads mainly to insoluble Pbp4 * [22]. Therefore, we wondered whether overproduction at low temperature will influence the folding of Pbp4 * leading to mainly soluble protein. We analysed the amount of Pbp4 * by Western blot from three different fractions: total cellular content, soluble and insoluble fraction obtained after a centrifugation step. While some Pbp4 * protein was present already at 37 °C incubation as reported for HtpG, its amount increased significantly 6 h after incubation of the cells at 25 °C(Fig. 4). As can be seen, most of the recombinant protein stayed soluble indicating that the lower temperature favours formation of folded polypeptides as described for the aggregation-prone fusion protein preS2-S0-b-galactosidase in E. coli [23]. A comparable result has been obtained during constitutive high level production of the DnaK and GroE chaperone systems [22]. We conclude that production of aggregation-prone recombinant proteins at low temperatures is alternative way to largely prevent formation of aggregates. The expression-secretion vector pAL12 allows regulated secretion of exoproteins To test the secretion capability at low temperature, the amyQ gene coding for an a-amylase [17] was inserted into pAL12 resulting in pAL12-amyQ. Strain AL02 carrying pAL12-amyQ was grown in LB medium at 37 and 25 °C, and aliquots were taken at the time points indicated in Fig. 5. If the amount of a-amylase present at 37 and 25 °C were compared, significantly more enzyme was present at 25 °C as compared to 37 °C(Fig. 5). We also meaFig. 4. Immunoblot analysis of Pbp4 * .B. subtilis strain 1012 carrying pAL10-pbpE was grown as described in the legend to Fig. 3. Cells were lysed by sonification and the cellular lysate was applied directly (T) or after a centrifugation step to separate soluble (S) from insoluble (P) protein. 0.3 lg of protein were applied per lane. Fig. 5. Detection of extracellular a-amylase by SDS–PAGE. Strain AL02 carrying pAL12-amyQ was grown as described in the legend to Fig. 3. Aliquots were taken from the supernatant of both cultures at the time points indicated. Purified a-amylase was added to one lane. 11 lg of protein were applied per lane. Fig. 3. Identification of the htpG gene product. Cells of strain AL05 carrying the plasmid pAL10-htpG were grown in LB medium at 37 °C to mid-log (t= 0), divided into two cultures, where one was further incubated at 37 °C and the second cold-shocked to 25 °C. As a control, cells of strain 1012 carrying pNDH33-htpG were grown at 37 °C to mid-log and then cold-shocked and induced by addition of 1 mM IPTG for 6 h. Cells were lysed by sonification and 0.5 lg of protein was loaded per lane on an 10% SDS–PAGE. (a) After gel electrophoresis, the proteins were stained with Coomassie blue. 37 °C culture: lanes 1, 2, 4, 6 and 8; 25 °C culture: lanes 3, 5, 7 and 9; lane 10, IPTG-treated cells grown at 25 °C. (b) Immunoblot analysis of HtpG. Cells were grown and treated as described. After separation of the proteins and Western blot, HtpG was detected using antibodies raised against this protein. Molecular weight markers are indicated. 268 A.T. Thuy Le, W. Schumann / Protein Expression and Purification 53 (2007) 264–269 sured the a-amylase activities within supernatant and compared it to those produced at 37 °C. While the activities were comparable during the first 5 h, higher activities were measured at later times in accordance with the results obtained by gel analysis (data not shown). At 20 °C, we have been unable to detect any a-amylase indicating that secretion of this enzyme and most probably many others is severely impaired under these growth conditions. It has been reported that the SecA abundance in E. coli was 3-fold higher at 20 °C than at 37 °C[24], in accordance with the notion that the E. coli protein export includes some intrinsically cold-sensitive element [25]. Based on these observations we can only speculate that at least one component of the Sec pathway does not function properly at 20 °C. This could be the SecA motor protein or/and the SecYEG translocons or/and a so far unknown component. Acknowledgments This work was financially supported by the Deutsche Forschungsgemeinschaft (Schu 414/20-2). The two expression vectors pAL10 and pAL12 can be ordered from MoBiTec (www.mobitec.com). References [1] A. Mogk, M.P. Mayer, E. Deuerling, Mechanisms of protein folding: molecular chaperones and their application in biotechnology, Chembiochem 3 (2002) 807–814. [2] A. Mitraki, J. King, Protein folding intermediates and inclusion body formation, Bio/Technology 7 (1989) 690–697. [3] J.G. Thomas, F. Baneyx, Protein misfolding and inclusion body formation in recombinant Escherichia coli cells overexpressing heatshock proteins, J. Biol. Chem. 271 (1996) 11141–11147. [4] C.H. Schein, M.H.M. Noteborn, Formation of soluble recombinant proteins in Escherichia coli is favored by lower growth temperature, Bio/Technology 6 (1988) 291–294. [5] A.W. Emerick, B.L. Bertolani, A. Ben-Bassat, T.J. White, M.W. Konrad, Expression of a b-lactamase preproinsulin fusion protein in Escherichia coli, Bio/Technology 2 (1984) 165–168. [6] J.A. Chesshyre, A.R. Hipkiss, Low temperatures stabilize interferon a-2 against proteolysis in Methylophilus methylotrophus and Escherichia coli, Appl. Microbiol. Biotechnol. 31 (1989) 158–162. [7] B. Ha ¨rtl, W. Wehrl, T. Wiegert, G. Homuth, W. Schumann, Development of a new integration site within the Bacillus subtilis chromosome and construction of compatible expression cassettes, J. Bacteriol. 183 (2001) 2696–2699. [8] M.H.W. Weber, M.A. Marahiel, Bacterial cold shock responses, Science Progress 86 (2003) 9–75. [9] P.S. Aguilar, J.E. Cronan Jr., D. De Mendoza, A Bacillus subtilis gene induced by cold shock encodes a membrane phospholipid desaturase, J. Bacteriol. 180 (1998) 2194–2200. [10] P.S. Aguilar, P. Lopez, D. De Mendoza, Transcriptional control of the low-temperature-inducible des gene, encoding the Delta5 desaturase of Bacillus subtilis, J. Bacteriol. 181 (1999) 7028–7033. [11] P.S. Aguilar, A.M. Hernandez-Arriaga, L.E. Cybulski, A.C. Erazo, D. De Mendoza, Molecular basis of thermosensing: a two-component signal transduction thermometer in Bacillus subtilis,EMBOJ.20 (2001) 1681–1691. [12] L.E. Cybulski, G. Del Solar, P.O. Craig, M. Espinosa, D. De Mendoza, Bacillus subtilis DesR functions as a phosphorylationactivated switch to control membrane lipid fluidity, J. Biol. Chem. 279 (2004) 39340–39347. [13] M.C. Mansilla, D. De Mendoza, The Bacillus subtilis desaturase: a model to understand phospholipid modification and temperature sensing, Arch. Microbiol. 183 (2005) 229–235. [14] M. Mujacic, K.W. Cooper, F. Baneyx, Cold-inducible cloning vectors for low-temperature protein expression in Escherichia coli: application to the production of a toxic and proteolytically sensitive fusion protein, Gene 238 (1999) 325–332. [15] G.L. Qing, L.C. Ma, A. Khorchid, G.V.T. Swapna, T.K. Mal, M.M. Takayama, B. Xia, S. Phadtare, H.P. Ke, T. Acton, G.T. Montelione, M. Ikura, M. Inouye, Cold-shock induced high-yield protein production in Escherichia coli, Nat. Biotechnol. 22 (2004) 877–882. [16] A.M. Gue ´rout-Fleury, N. Frandsen, P. Stragier, Plasmids for ectopic integration in Bacillus subtilis, Gene 180 (1996) 57–61. [17] I. Palva, Molecular cloning of alpha-amylase gene from Bacillus amyloliquefaciens and its expression in B. subtilis, Gene 19 (1982) 81–87. [18] A. Schulz, S. Schwab, S. Versteeg, W. Schumann, The htpG gene of Bacillus subtilis belongs to class III heat shock genes and is under negative control, J. Bacteriol. 10 (1997) 3103–3109. [19] D.L. Popham, P. Setlow, Cloning, nucleotide sequence, and regulation of the Bacillus subtilis pbpE operon, which codes for penicillinbinding protein 4 * and an apparent amino acid racemase, J. Bacteriol. 175 (1993) 2917–2925. [20] T. Wiegert, G. Homuth, S. Versteeg, W. Schumann, Alkaline shock induces the Bacillus subtilis r W regulon, Mol. Microbiol. 41 (2001) 59–71. [21] G. Homuth, S. Masuda, A. Mogk, Y. Kobayashi, W. Schumann, The dnaK operon of Bacillus subtilis is heptacistronic, J. Bacteriol. 179 (1997) 1153–1164. [22] T.T.P. Phan, H.D. Nguyen, W. Schumann, Novel plasmid-based expression vectors for intraand extracellular production of recombinant proteins in Bacillus subtilis, Protein Expr. Purif. 46 (2006) 189–195. [23] J.A. Vasina, F. Baneyx, Expression of aggregation-prone recombinant proteins at low temperatures: a comparative study of the Escherichia coli cspA and tac promoter systems, Protein Express. Purif. 9 (1997) 211–218. [24] A. Murakami, H. Nakatogawa, K. Ito, Translation arrest of SecM is essential for the basal and regulated expression of SecA, Proc. Natl. Acad. Sci. USA 101 (2004) 12330–12335. [25] K.J. Pogliano, J. Beckwith, The Cs sec mutants of Escherichia coli reflect the cold sensitivity of protein export itself, Genetics 133 (1993) 763–773. [26] H. Saito, T. Shibata, T. Ando, Mapping of genes determining nonpermissiveness and host-specific restriction to bacteriophages in Bacillus subtilis Marburg, Mol. Gen. Genet. 170 (1979) 117–122. [27] W. Wehrl, M. Niederweis, W. Schumann, The FtsH protein accumulates at the septum of Bacillus subtilis during cell division and sporulation, J. Bacteriol. 182 (2000) 3870–3873. [28] A. Mogk, R. Hayward, W. Schumann, Integrative vectors for constructing single-copy transcriptional fusions between Bacillus subtilis promoters and various reporter genes encoding heat-stable enzymes, Gene 182 (1996) 33–36. A.T. Thuy Le, W. Schumann / Protein Expression and Purification 53 (2007) 264–269 269 6. Abbreviations 108 6. Abbreviations σ Sigma factor ∆ deletion µg microgram µl microliter 0C degree centigrade 2D-Gel two-dimensional gel 5' UTR 5' UnTranslated Region aa amino acid(s) ADP Adenosine-5’-diphosphate ATP Adenosine-5’-triphosphate B. subtilis Bacillus subtilis bp base pairs BSA Bovine Serum Albumin cat gene coding for chloramphenicol-acetytransferase cfu colony forming units DNA Deoxyribonucleic acid DSM Difco Sporulation medium E. coli Escherichia coli EDTA Ethylene diamine tetraacetic acid erm Gene coding for erythromycine resistance et al. et alteri G gram GFP Green Fluorescent Protein GSH Glutathione GST Glutathione-S-transferase H hour(s) IPG Immobilized pH gradient IPTG Isopropyl-β-D-thiogalacto pyranoside 6. Abbreviations 109 kDa kilo-Dalton l liter lacZ beta-galactosidase gene LB Luria-Bertani (growth medium) LPS Lipopolysacharide M Molar MALDI-TOF Matrix-assisted Laser Desorption-ionization Time-of-flight MS Mass Spectrometry mg milligram min minute(s) ml mililiter mM milimole mRNA messenger RNA OD Optical Density PBP Penicillin binding protein PBS Phosphate-buffer saline PCR Polymerase Chain Reaction RBS Ribosome Binding Site RNA Ribonucleic acid SDS-PAGE Sodium Dodecyl Sulfate Polyacrylamide Gel Electrophoresis sec second spec Gene coding for Spectinomycin txstage x of sporulation program tet Gene coding for tetracycline resistance WT Wild-type Erklärung Hiermit versichere ich, die vorliegende Arbeit selbstständig verfasst und keine anderen als die von mir angegebenen Quellen und Hilfsmittel benutzt zu haben. Ferner erkläre ich, dass ich weder an der Universität Bayreuth, noch an einer anderen Hochschule versucht habe, eine Dissertation einzureichen, oder mich einer Promotionsprüfung zu unterziehen. Ai Thi Thuy Le Bayreuth, December 2007