... 84...... . ffmuhkmi r. dowry? ”Mix/m9 firm. .. 1.. 3. i . . . . l on). 3 Gus.“ dfl . 4 . ‘ r I v . . h. c. .....n . Q1: . . . ‘ ,. ...L: goknfirfiflflqfinflmz . 3.1“”..1 DR. , . . ..i. ‘ . w .%fi.flfl.flr.mbflu I...» ...x. ,. . , r . ....Jwfi , . .w . u . .........~.,...u..>. Ft (7 HaflimvmWn ... Q :«mi. .,.. 1.. ..m. W ’ pi." . . snuff-I ‘ ‘ H‘.‘ r‘ I f ... km...’m.ht.r......r.. .. 2: . 33¢ . . . ., ~ xi: v.91... fl)? . . ‘ . . , ..‘wnkmfib- humufiafiqu ”Hanoi. .. ... . . ...-v“. I; 0h .31 ‘11, N. ‘ ['1 nfl L1. 1 . 5 :1 I . . v k 1 u: 0. u A . . n......d~..... ... a... . .4 .. . . - .-..MR‘... . § . 9.0% . .1. ...... w 3L. a c Hus... .ul w .l 6. v. I. . . «9...... w. 5.... NM! 1 .u‘mKM..vW5flJ.P ......miunnfi. . . I i O V’ . I h ‘ P t -5 V 9.94. .3. .. . . “w. .0 .w .l I; 3 . , 3‘ (I .r . 0 o A O o “wry...“- fla. .omn. v1. (finfihfifim - Krocftrv‘vll‘ . . . . a. 9.33...." Hard. .. . .c 3. ”C «.... ‘ fjjt'ifzpufifigpm, . ...om v. . V . .. h ”rumba! .454 . ......uflmuu . I . .. .. . . .Pfiuuwnn . ...- V . . .34.”... :5. . . .....finwr. <11. 51-325. .8 .(‘uvtu‘ . .1. .F}Y.F‘No'c.lt 1 . - .1381! 0.4.5.4"? . ... Ltd... ntl‘o‘nu 51;:6‘» uvt.nl I! 32“ 4 1- "3‘34"". '0“ HIE-Win. it. ;' . {33's}: ‘ iii “3';;:, l; '41 .n . k rl‘ .é, .p “any 2 I1. ‘ . ... AI ii. at... ”I?" r‘llfo‘ r? . .. (......B. 11qu $14.!thmunli. I . ng. . l . . .1518...“ ‘-.. . :|.. (.9. L .P .... r {u‘ hunocrm I of?! .r 1H“ bl. ... PI. .81.. t . .. .vrcvhnuvautioo- Ir... . .un ‘lua IV..|,. who, ‘. Vuol..t.'!l..ut.. . ‘Iw, . ~ 0_I...o.:|.. L ‘ ... .1 ...... 5...»..wmu ‘.u....s. ‘ «I’lbt. t}: 10. ‘ .V . r . . . ‘ '1“ 507.3. t ‘Kv. ‘ . ,v. In)“. to!!! ...l 1*! I! TH E SlS lllllllllllllllllllllllllllllllllllllllllllllll 293 01397 2975 LIBRARY Mlchlgan State Unlversity This is to certify that the dissertation entitled DEVELOPMENTAL GENE EXPRESSION REGULATED BY A CASCADE OF SIGMA FACTORS IN BACILLUS SUBTILIS presented by BIN ZHANG has been accepted towards fulfillment of the requirements for Ph .1) degree in Biochemistry (ink/mm. Major professor Date l1/’7I/q7 MSU is an Affirmative Action/Equal Opportunity Institution 042771 PLACE IN RETURN BOX to remove this checkout from your record. To AVOID FINES return on or before date due. DATE DUE DATE DUE 1/” WWW“ DEVELOPMENTAL GENE EXPRESSION REGULATED BY A CASCADE OF SIGMA FACTORS IN BACILLUS SUBTILIS By Bin Zhang A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Biochemistry 1997 ABSTRACT DEVELOPMENTAL GENE EXPRESSION REGULATED BY A CASCADE OF SIGMA FACTORS IN BACILLUS SUBTILIS By Bin Zhang Upon starvation, Bacillus subtilis undergoes sporulation that culminates with the formation of a dormant spore. Initiation of sporulation is governed by the phosphorylation of SpoOA through a phosphorelay. An elevated level of SpoOA~P leads to formation of an asymmetric septum that divides the sporulating cell into two compartments of unequal size, the mother cell and the forespore. Temporal and spatial regulation of gene expression during sporulation is achieved, in part, through the sequential synthesis and activation of compartment-specific sigma factors of RNA polymerase (RNAP) and DNA-binding proteins. Key regulators of mother cell gene expression are the sigma factors OE and OK, and the DNA-binding proteins SpoIIID and GerE. One function of SpoIIID is to switch on O'K- dependent gene expression, because SpoIIID is required for the appearance of OK. SpoIIID also inhibits transcription of some late (SK-dependent genes. A rapid decrease in the SpoIIID level is thought to be critical for the expression of these late genes. It is shown that the decrease in the level of SpoIIID is accompanied by a decrease in transcription of the SpoIIID gene, and this depends upon the appearance of CK. Transcription of SpoIIID depends on (SE. 6K negatively regulates transcription of SpoIIID by inhibiting the transcription of sigE, which encodes O'E. Transcription of sigE is carried out by RNAP containing the vegetative sigma factor GA, and is activated by SpoOA~P and repressed by SinR. It is shown that UK does not change sigE expression by increasing the level or activity of SinR. Rather, it appears that 6K affects the phosphorelay, lowering the level of SpoOA~P. (SK may also inhibit 0" activity, perhaps by competing directly for binding to core RNAP. Hence, the appearance of 0'" both turns on late mother cell gene expression and turns off early (IE-directed gene expression through a negative feedback loop. UK is first synthesized as an inactive precursor protein called pro-OK. Activation involves a proteolytic cleavage of the N-terminal pro-sequence from pro-OK. The putative protease is localized in the mother cell membrane surrounding the forespore. Subcellular fractionation studies show that the majority of pro-(SK is membrane-associated in cell extracts, and is not associated with the core subunits of RNAP. Immunolocalization of pro-(3'K suggests that pro-0’K interacts with both the membrane surrounding the mother cell and the membrane surrounding the forespore in sporulating cells. Pro- O'K fails to bind to core RNAP in vitro under conditions that permit O’K binding. These results suggest that the pro-sequence of pro-0'K inhibits the core-binding activity of OK and promotes its association with the membrane, where processing may occur. To My family iii ACKNOWLEDGMENTS I am very grateful to my thesis advisor, Dr. Lee Kroos for his encouragement and guidance, both scientific and general, during the years. He has provided me with every available opportunity to become a better scientist. I enjoyed all the discussions we had over the years, which have helped me tremendously on my scientific thinking and writing. I thank my committee members, Dr. Frans de Bruijn, Dr. Robert Hausinger, Dr. Jon Kaguni, and Dr. Steve Triezenberg, for their criticisms and advice. I also thank Dr. David Amosti for reading my dissertation and attending my defense. I would like to thank all the past and present members of the Myxo squad and Bacillus squad in the Kroos’ laboratory, Richard Halberg, Sijie Lu, Makda Fisseha, Jannine Brandner, Dvora Biran, Michele Anderson, Hiroshi Ichikawa, Tong Hao, Greg Velicer, Nicco Yu, and many undergraduates for sharing excitement and frustrations in research and in life. They have made the lab an enjoyable place to work. I thank Dr. Antje Hofmeister for sharing information and collaborating on immunofluorescence microscopy experiments, Dr. Pappan for help with the computers. I would like to thank my parents and my brothers for their love, support and their faith in me, my parents-in-law for their help and encouragement. Finally, but not least, I would like to thank my wife Yusong for everything, my sons Shawn and Jason for making life more pleasant and colorful. They made this long journey through graduate school possible and meaningful. TABLE OF CONTENTS LIST OF TABLES ................................................................................. viii LIST OF FIGURES ............................................................................... ix LIST OF ABBREVIATIONS ..................................................................... xi INTRODUCTION ................................................................................. 1 CHAPTER I. Literature Review Morphological changes during sporulation ............................................. 4 Initiation of sporulation ................................................................... 6 Polar division and activation of CF in the forespore ................................... 12 Signal transduction pathways leading to the activation of compartment-specific sigma factors ............................................................................... 16 GP to CE ........................................................................... 17 (SE to 60 ........................................................................... 18 0’0 to OK ........................................................................... 20 Hierarchical regulatory cascade of the mother-cell line of gene expression ....... 22 Summary ................................................................................... 27 CHAPTER II. A Feedback Loop Regulates the Switch from One Sigma Factor to the Next .> (I) in the Cascade Controlling Bacillus subtilis Mother-cell Gene Expression I“ , . )L Abstract ..................................................................................... 29 Materials and Methods ................................................................... 29 V/ Results ..................................................................................... 30 Stability of spoIIID mRNA in wild-type and sigK mutant cells ........... 30 GE level in wild-type and sigK mutant cells ................................... 31 05 level in cells that produce 6K earlier ....................................... 32 Turn-over of pro-GE and 0'5 in wild-type and sigK mutant cells ........... 32 Expression of a sigE-lacZ transcriptional fusion in wild-type cells, sigK mutant cells, and cells producing 0K earlier ................................... 32 x// Discussion ................................................................................. 33 / (i’f‘l‘ \M) ').\ ,g. )b “)7 pf, gal/.1 CHAPTER III. Characterization of the Mechanism by which 6K Negatively Regulates sigE Transcription during Sporulation of Bacillus subtilis { Abstract ..................................................................................... 37 \/ Introduction ............................................................................... 38 \/ Materials and Methods ................................................................... 4O Results and Discussion ................................................................... 43 v (3 '1" 4‘ 6K does not inhibit sigE transcription by increasing the level or activity of SinR ................................................................... 43 Bypassing the phosphorelay leading to the activation of SpoOA partially relieves the negative effect of OK on sigE transcription ..................... 45 Expression of a SpoOA-independent gene remains high late during sporulation in sigK mutant cells ............................................... 48 OK has little effect on expression of (SH-dependent genes ................... 51 Effect of inducing GK during vegetative growth on expression of ald and spa V64 2 ......................................................................... 54 CHAPTER IV. The Pro-sequence of Pro-(1'K Promotes Membrane Association and Inhibits the RNA Polymerase Core Binding Abstract ..................................................................................... 60 Introduction ............................................................................... 6 1 Materials and Methods ................................................................... 63 Results ..................................................................................... 68 vi The majority of pro-0K is membrane-associated ............................. 68 Effect of detergent and salt treatment on the membrane association of pro-OK ............................................................................. 72 Membrane association of pro-(5'K does not depend upon sporulation- specific gene products ........................................................... 75 Pro-GK binding sites are not saturated on the membranes of sporulating cells OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO Pro-GK localizes to the mother cell membranes that surround the forespore and the mother cell of the postengulfrnent sporangium ..................... 78 Pro-(3'K does not bind to exogenous core RNAP in vitro ................... 82 Discussion ................................................................................. 86 CHAPTER V. Summary and Perspectives Summary and Perspectives ............................................................... 93 APPENDIX. Bacillus subtilis SpoIIID protein binds to two sites in the spa VD promoter and represses transcription by GE RNA polymerase Abstract ..................................................................................... 97 Introduction ............................................................................... 97 spa VD is transcribed by EOE in vitro and SpoIIID represses this transcription ... 97 SpoIIID binds to two sites in the spa VD promoter region ........................... 98 BIBLIOGRAPHY ............................................................................... 1 01 vii LIST OF TABLES CHAPTER II Table l. B. subtilis strains used ................................................................. 4] viii LIST OF FIGURES CHAPTER I Figure l. Morphological stages during B. subtilis sporulation ............................. 5 Figure 2. Integration of multiple signals and conditions through the phosphorelay ....... 8 Figure 3. Positive and negative feedback loops controlling production and accumulation of SpoOA~P ............................................................... 1 1 Figure 4. Model for the forespore-specific activation of CF ................................. 15 Figure 5. Models for the intercompartrnental signaling pathways leading to the activation of pro-(3'E and pro-0'K ..................................................................... 19 Figure 6. Diagram of gene regulation in the mother-cell cascade ........................... 25 CHAPTER II Figure l. The production of OK does not change the stability of spoIIID mRNA ......... 30 Figure 2. GE persists at a higher level during sporulation of cells defective in OK production ................................................................................. 31 Figure 3. SpoIIID persists at a higher level during sporulation of cells defective in OK production ................................................................................. 31 Figure 4. GE disappears earlier during sporulation of cells that produce 6K earlier than normal ................................................................................. 32 Figure 5. The production of OK does not alter the stability of pro-(1'E and GE ............. 33 Figure 6. The effects of altered GK production on sigE-lacZ expression are similar to the effects on the GE level ................................................................... 33 Figure 7. Model for gene regulation in the mother-cell cascade ............................. 34 ix CHAPTER 111 Figure 1. The effect of a sinR null mutation on sigE-lacZ expression ................... 44 Figure 2. The effect of bypassing the phosphorelay on sigE-lacZ expression ........... 46 Figure 3. The effects of altered 6K production on ald-lacZ expression ................. 50 Figure 4. The effects of altered GK production on spoIlA-lacZ and spa VG42-lacZ expression ............................................................................... 53 Figure 5. The effect of making OK during vegetative grth on ald-lacZ and spa VG42- lacZ expression ......................................................................... 56 CHAPTER IV Figure 1. Diagram of subcellular fractionation of sporulating B. subtilis cells ............. 65 Figure 2. Subcellular fractionation of extracts of sporulating wild-type cells. ............. 70 Figure 3. Effects of detergent and salt treatment on fractionation of pro-0'K and 6K ... 74 Figure 4. Specificity of the membrane-association of pro-GK ............................... 76 Figure 5. Immunolocalization of pro-0'K and 6K in sporulating cells ....................... 80 Figure 6. OK, but not pro-6K, reassociated with core RNAP after being dissociated by salt treatment ............................................................................. 84 Figure 7. Model depicting association of pro-(15'K with the outer forespore membrane and signal transduction between the forespore and the mother cell leading to the processing of pro-6K ..................................................................... 87 APPENDIX Figure 1. In vitro transcription of spa VD by E65 and effect of SpoIIID ................... 98 Figure 2. SpoIIID footprints in the spa VD promoter region. ............................... 98 Figure 3. Position of SpoIIID binding sites in the spa VD promoter region and alignment of sequences within these sites with the consensus sequence for SpoIIID binding. ............................................................................................... 99 BSA dATP DNA DSM EDT A FITC HCl Hepes IPTG KCl KOH kb LB LIST OF ABBREVIATIONS adenosine-5’-diphosphate adenosine-5’-triphosphate base pair bovine serum albumin deoxyadenosine-S ’-triphosphate deoxyribonucleic acid Difco sporulation medium dithiothreitol (ethylenedinitrilo)tetraacetic acid fluorescein isothiocyanite hydrochloric acid N-(2-hydroxyethyl)piperazine-N’-(2-ethanesulf6nic acid) isopropyl B-D thiogalactopyranoside potassium chloride potassium hydroxide kilobases kilodalton Luria-Bertani molar milliliter millirnolar xi MgCl2 mRN A NaCl "8 ONPG PAGE PI pmole psi PMSF RNAP SDS SM Tricine Tris TSS T X #8 pl wt/vol magnesium chloride messanger ribonucleic acid sodium chloride nanogram nanomolar o-nitrophenol-B-D-galactoside polyacrylamide gel electrophoresis propidium iodide picomole pounds per square inch phenylmethylsulfonyl fluoride RNA polymerase sodium dodecylsulfate Sterlini-Mandelstam N-tris(Hydroxymethyl)methylglycine tris(hydroxymethyl)aminomethane transcriptional start site x hours after the onset of sporulation microgram microliter weight per volume xii INTRODUCTION Sporulation of Bacillus subtilis in response to starvation provides an excellent model system to study the fundamental problems of gene expression and the regulation of development of a living organism. Sporulation involves a series of morphological changes. Key to the initiation of sporulation is the phosphorylation of SpoOA, an event controlled by a phosphorelay system that is regulated by multiple signals. An elevated level of SpoOA~P leads to formation of an asymmetric septum that divides the sporulating cell into a larger mother cell compartment and a smaller forespore compartment. Each compartment receives a copy of the genome, but each follows a different pathway of development. Temporal and spatial regulation of gene expression during sporulation is achieved, in part, through the sequential synthesis and activation of compartment-specific sigma factors of RNA polymerase (RNAP). DNA-binding proteins also contribute to the proper regulation of gene expression. This dissertation focuses on the regulation of key transcription factors in the mother cell. Key regulators of gene expression in the mother cell are the sigma factors 6E and GK, and the DNA-binding proteins SpoIIID and GerE. SpoIIID activates and represses transcription by both GE RNAP and GK RNAP. One function of SpoIIID is to switch on (SK-dependent gene expression, because SpoIIID is required for the appearance of OK. SpoIIID also inhibits transcription of some late (SK-dependent genes. A rapid decrease in the SpoIIID level is thought to be critical for the expression of these late genes. Chapter 2 describes a study which demonstrates that the appearance of CK initiates a negative 2 feedback loop that turns off the transcription of sigE and the whole O'E regulon, includingspoIIID. This work was published in the Journal of Bacteriology. Transcription of sigE is carried out by RNAP containing the vegetative sigma factor GA, activated by SpoOA~P, and repressed by SinR. The experiments in Chapter 3 further characterize the negative effect of OK on sigE transcription. It is shown that UK appears to lower the level of SpoOA~P and inhibit 0" activity. A revised version of this chapter will be submitted to the Journal of Bacteriology. Experiments described in Chapter 2 and 3 contribute to our understanding of the developmental gene regulation in general by demonstrating that replacement of transcription factors acting early during development involves a feedback regulation by those acting late during development. This finding provides an alternative to the model that late-acting transcription factors simply accumulate enough amount and passively take over the earlier 01188. UK is first synthesized as an inactive precursor protein called pro-6K. Activation involves the proteolytic cleavage of the N-terminal pro-sequence from pro-(SK in response to a forespore signal. Chapter 4 describes a study which demonstrates that the pro- sequence of pro-(3'K inhibits the core-binding activity of OK and promotes its association with the membrane, where processing may occur. The immunolocalization data presented in this chapter was obtained through a collaboration with A. Hofrneister at Harvard University. A manuscript based on this chapter has been submitted to the Journal of Bacteriology. The Appendix describes my first project in the lab, a study which demonstrates that spo VD, a gene then newly cloned by R. Daniel and J. Errington at the University of Oxford, is transcribed by 0'5 RNA polymerase in vitro and that SpoIIID represses spo VD 3 transcription. Two strong SpoIIID-binding sites were mapped in the spo VD promoter region. J. Errington and R. Daniel shared results prior to publication and provided a plasmid containing the spo VD promoter region for this study, which was published in the Journal of Bacteriology. Chapter 1. LITERATURE REVIEW Understanding the temporal and spatial regulation of gene expression during the development of a living organism is a fundamental problem in developmental biology. An excellent system to address this problem is the sporulation process of the Gram-positive bacterium Bacillus subtilis because of its relatively simple cellular organization, its experimental tractability, and its excellent genetics. Morphological Changes during Sporulation. The successive morphological stages of sporulation are shown in Figure 1 (27, 87, 137). Vegetative cells are defined as stage 0 with respect to the sporulation process. 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 with the long axis of the cell (stage I). The first easily observed morphological change during sporulation is the formation of an asymmetrically positioned septum (stage II) that divides the developing cell (sporangium) into two unequal compartments: the larger one is called the mother cell and the smaller one is called the forespore. Each compartment receives a chromosome. The septum then migrates around the forespore, engulfs it in a double membrane, and eventually pinches it off as a free protoplast within the mother cell (stage III). The inner membrane surrounding the forespore is referred to as “the forespore membrane”. The outer membrane surrounding the forespore is called “the mother cell membrane surrounding the forespore”, because it is derived from the mother cell membrane. A modified form of cell wall, known as the cortex, is synthesized between the two Figure l. Morphological stages during B. subtilis sporulation. The compartment- specific sigma factors are shown in order of their appearance during sporulation. Reprinted from ref. 137. 6 membranes surrounding the forespore (stage IV). At about the time of cortex formation, more than twenty proteins are made in the mother-cell compartment and deposited around the forespore to form a multilayered protein shell called the spore coat (stage V). The final period of spore development, termed maturation (stage VI), happens with no dramatic change in morphology, but during this stage the properties of resistance, dormancy, and ability to germinate appear in sequence. The sporulation process culminates with lysis of the mother cell and release of the mature spore (stage VII). When conditions are favorable, the spore can germinate and resume vegetative growth. Over 125 genes have been identified that are induced and/or required for the sporulation process (27, 137). Initiation of Sporulation Initiation of sporulation depends upon activation of the SpoOA transcription factor by phosphorylation, and a threshold concentration of SpoOA~P appears to be required (34, 46). SpoOA receives phosphate from three histidine protein kinases (KinA, KinB and KinC) through a multicomponent phosphorelay (Figure 2). The phosphorelay is a modified bacterial two-component signal transduction system. Two-component systems are involved in numerous adaptive responses, including chemotaxis and nitrogen utilization (101 , 133). All three histidine protein kinases in the B. subtilis phosphorelay belong to a conserved family of proteins known as sensor kinases (2). Proteins that receive phosphate from sensor kinases belong to a conserved family of proteins known as response regulators. KinA, KinB and KinC first autophosphorylate and donate phosphate to the response regulator SpoOF. The phosphate is then transferred from SpoOF to SpoOB and finally to SpoOA (16). Among the three histidine protein kinases, KinA and KinB appear to contribute the most to the phosphorelay (144). KinC contributes the least to the phosphorelay (76, 77). KinC is required for the activation of altered forms of SpoOA (sofl, WM”, and surOBZO) (67, 76). A major function of the phosphorelay seems to be to integrate multiple developmental signals that regulate the initiation of sporulation. Signals generated by conditions of Figure 2. Integration of multiple signals and conditions through the phosphorelay. Sensor kinases KinA, KinB and KinC autophosphorylate and phosphate is transferred to the response regulator SpoOF, then to the phosphotransferase SpoOB and finally to an aspartate residue in the N-terminal domain of the response regulator SpoOA (16). It is not yet known which component of the phosphorelay is the direct target for a given physiological signal. Adapted from ref. 34. $368 2966 320523 228 mmmEmEm_mm£c>m cozmzamn mEoonoEo 39v. E6252 L L E L <83 m. 987%.“ "58m %H_ M%H :con 9432...... .5on 4 My 4 a $33383 229m 8365qu mwmmwwmwofi mamméamm _ $38096 . 83809.3 £83 :8 x88 9 nutrient depletion, high cell density, the Krebs cycle, DNA replication, DNA damage, and the chromosome partitioning machinery all modulate activation of SpoOA. Mutations in SpoOA (rvtAI I) that bypass the need for SpoOF and SpoOB (67, 76) can at least partially bypass the DNA synthesis and damage (52, 53), Krebs cycle (56), and the chromosome partitioning controls on sporulation (54), indicating that these signals inhibit the phosphorelay and limit production of SpoOA~P. Regulation by these signals appears to serve as a developmental checkpoint, ensuring that sporulation does not begin unless it seems likely that it can be completed. However, little is known about the nature of the signals and the signal transduction pathways that control the activity of the phosphorelay. Nutrient depletion causes a drop in the intracellular GTP level (84). Inhibiting GTP biosynthesis induces sporulation (28, 85, 96). The rvtAIl mutation in SpoOA does not bypass the need for nutrient depletion. However, expression of a constitutively active SpoOA (SpoOAsad) is sufficient to bypass the requirement for nutrient depletion (55). These results suggest that the target of nutrient depletion signal is probably the sensor kinases (Kin). Mutations in kinA and kinB cause different phenotypes depending on the starvation condition, indicating that there might be multiple ways of sensing nutrient depletion (55). However, it is not clear how sensor kinases are activated. In addition to contributing to activation of the sensor kinases, nutrient depletion also induces several genes required for sporulation, independent of the phosphorelay, including citrate synthase genes (cit) (57), ald, encoding alanine dehydrogenase (128), and spoOJ, a chromosome partitioning gene (54). Factors controlling expression of these genes have not been characterized. Production of SpoOA‘m" also bypasses the need for high cell density in sporulation, indicating the cell density signal regulates the SpoOA~P level (55). However, high cell density also induces the mutually exclusive process of competence development. Competence refers to the ability of cells to take up exogenous DNA. According to a simple 10 model (34), a low concentration of SpoOA~P in vegetative cells inhibits both the competence and the sporulation pathways. During the transition state (late exponential growth phase to early stationary phase), an intermediate level of SpoOA~P is optimal for competence development. A further increase of SpoOA~P crosses a threshold and induces sporulation. One of the mechanisms that cells use to respond to changes in cell density is sensing of extracellular peptide factors that accumulate in the culture medium (quorum sensing) (29). Three such peptide factors have been identified. The ComX pheromone mainly stimulates competence development (131). PhrA mainly stimulates sporulation (75, 107). CSF (competence and sporulation factor) has at least three distinct firnctions: stimulating competence gene expression at low concentrations, and inhibiting competence gene expression and stimulating sporulation at high concentrations (130). CSF appears to have three different targets in cells corresponding to its three functions (75). Both PhrA and CSF appear to be pentapeptides that are produced by secretion and processing of precursor molecules (75, 107, 130). They are transported back into the cell by the oligopeptide permease encoded by spoOK. One target of CSF is RapC, which it negative regulates (130), whereas PhrA negatively regulates the activity of RapA (107). RapA and RapC are members of the response-regulator aspartate phosphatase family that also includes RapB (105). RapA and RapB negatively regulate the phosphorelay by specifically dephosphorylating the response regulator SpoOF~P (Fig. 1.2) and thus reducing the level of SpoOA~P in cells (105). Interestingly, PhrA is derived from the product of a small gene downstream of rapA (107) and CSF is derived from the product of a small gene downstream of rapC (130). CSF and PhrA appear to represent an emerging class of cell-cell signaling molecules that are actively imported and function intracellularly. SpoOE is another phosphatase that regulates the flow of the phosphorelay. It specifically dephosphorylates SpoOA~P (98) (Figure 3). Transcription of spoOE is repressed by a DNA-binding protein, Aer, and is derepressed due to SpoOA~P repression Egg; SpoOF —> SpoOB $poOA~PK+ SPOHA EoH SpoOB f Aer I phosphatase spoOH Figure 3. Positive and negative feedback loops controlling production and accumulation of SpoOA~P. Lines with arrowheads indicate positive effects on synthesis or activition. Lines with barred ends indicate negative effects on synthesis or activation. See text for details. Reprinted from ref. 34. 12 of aer (106). This is a negative feedback loop that inhibits the accumulation of SpoOA~P (Figure 3). SpoOB may represent an independent pathway for preventing the initiation of sporulation. Several positive feedback loops contribute to the accumulation of SpoOA~P (Figure 3). They are controlled initially by the level of SpoOA~P and can further increase the activity or expression of SpoOA. SpoOA~P directly activates transcription of spoOA and SpoOF, in combination with 0'“ RNAP (113, 140). In addition, SpoOA~P increases the level of 0'” by repressing aer (138), which encodes a repressor of several genes involved in sporulation including sigH (139). This causes derepression of sigH. An increase in G" then contributes to increased transcription of kinA as well as spoOF and spoOA (113). SinR is a DNA-binding protein that inhibits sporulation. It represses transcription of spoOA by (5'H RNA polymerase (91), as well as transcription of several key stage 11 genes (92). SpoOA~P activates transcription of sin], which encodes an inhibitor of SinR (8). Thus, inhibition of SinR by increased production of Sin] probably contributes to increased transcription of spoOA. In addition to stimulating the transcription of spoOA and spoOF, the accumulation of SpoOA~P activates the transcription of several key genes that govern entry into sporulation and the transition to a two-compartrnent sporangia. These include an unknown gene(s) that determines the switch from medial to polar division (78), the spolIE gene (153) and the spoIIA operon (14), which are responsible for switching on gene expression in the forespore, and the spoIIG operon (10, 12), which is responsible for gene transcription in the mother cell. SpoOA~P activates transcription by binding to promoters of these genes or operons and stimulating the rate of initiation by modifying the RNA polymerase preinitiation complex (9, 10, 12). Polar Division and Activation of (SF in the Forespore. A hallmark of l3 sporulation is the formation of the polar septum, which partitions the sporangium into two, dissimilar-sized cellular compartments. In vegetative cells that undergo binary fission, a ring of a tubulin-like, cell division protein, FtsZ, is formed at a medial position. Later in the cell cycle, a septum forms at the site of F tsZ assembly, resulting in cytokinesis (78). In sporulating cells, F tsZ ring formation switches to sites near both poles of the sporangium, and this is under the control of SpoOA~P (78). Next, a septum is formed at one of the polar rings of FtsZ. This event is controlled by 0'H (78). Another sporulation sigma factor, GE, is involved in suppressing septum formation at the distal pole, because mutants defective in GE production undergo septation at both poles, giving rise to “disporic” sporangia (79). The genes controlled by SpoOA~P, 6H and 6E that are responsible for polar septation have not yet been identified. Unlike binary fission in which chromosome segregation takes place prior to septum formation, during sporulation chromosome partitioning largely occurs only after the polar septum is formed. In spoIIIE mutant cells, about 30% of the chromosome corresponding to the region proximal to the replication origin becomes trapped in the forespore (151). SpoIIIE functions as a DNA translocase in wild-type cells, pumping 70% of the chromosome into the forespore compartment (151). SpoOJ, which itself is located in the origin region (80, 147), is required for the orientation of the origin-proximal region of chromosomes to the pole (123). Shortly after the formation of the polar septum, CF is activated and its activity is strictly confined to the forespore (42, 93). (SF is encoded by the third gene of the spoIIA operon, spoIIAC (142). Transcription of spoIIA is dependent upon SpoOA~P and 0'” RNAP (9, 14, 148, 150), and begins prior to septation (31). The mechanism by which 0F activity is confined to the forespore involves a pathway composed of proteins encoded by 14 the first two genes of the spoIIA operon, SpoIIAA and SpoIIAB, and SpoIIE (Figure 4). SpoIIAB is an anti-sigma factor that binds to CF and inhibits oF-directed gene expression (26, 95). SpoIIAA is an anti-anti-sigma factor that counteracts the inhibitory effect of SpoIIAB by binding to the SpoIIAB-6F complex and causing release of free and active 6F (3, 20). SpoIIAB is also a protein kinase that phosphorylates SpoIIAA on serine residue 58 (95), thereby impairing the capacity of SpoIIAA to bind to SpoIIAB (3, 20). Central to the cell-specific activation of CF is SpoIIE, a specific serine phosphatase that is responsible for dephosphorylating SpoIIAA~P, thereby reactivating it for binding to SpoIIAB (Figure 4) (25). SpoIIE is an integral membrane protein that is localized initially to the two potential polar division sites before septation, disappears from the distal pole, and persists at the sporulation septum after septation (5, 11). According to current thinking, the principle determinant of (SF activation is the cellular concentration of unphosphorylated SpoIIAA, which is governed by the opposing activities of the SpoIIE phosphatase and the SpoIIAB kinase. If SpoIIE is displayed equally on both sides of the septum, then SpoIIAA~P could be expected to be dephosphorylated at an equal rate in both compartments. But because the forespore is several-fold smaller than the mother cell, the concentration of unphosphorylated SpoIIAA would be higher in the forespore than in the mother cell. Once a critical concentration is reached, SpoIIAA would interact with SpoIIAB and discharge 6F from the SpoIIAB-OF complex (Figure 4), forming the SpoIIAB-SpoIIAA complex. In this model, activation depends on the completion of septum morphogenesis. The observation that ATP stimulates the formation of the SpoIIAB-0F complex and ADP enhances the formation of the SpoIIAB-SpoIIAA complex (3, 20) has also led to a speculation that perhaps there is a selective decrease in the ratio of ATP to ADP in the forespore compartment. This decrease would favor the binding of SpoIIAB to SpoIIAA, rather than to CF. It may also hinder the phosphorylation of 15 UP (a) HAB’O’F IIABOIIAA WP) IIAA OF IIAB IIE [MA-(D (b) IIABmF 0F IIAA-G IIAB'IIAA IIE Figure 4. Model for the forespore-specific activation of CF. (a) Reactions that contribute to the release of free OF from the SpoIIAB-oF complex. The heavy lines indicate reactions favored in the forespore. (b) The state and cellular locations of the proteins involved in the activation of (IF. Reprinted from ref. 137. l6 SpoIIAA by SpoIIAB, destabilizing the SpoIIAB-SpoIIAA complex (Figure 4). Anti-sigma factors and anti-anti-sigma factors are also involved in regulating the activity of the Bacillus subtilis and Staphylococcus aureus stress response sigma factors. In the case of B. subtilis GB, the activity of anti-sigma factor Ris is subject to regulation by multiple kinases and phosphatases that integrate multiple stress signals. FlgM is an anti- sigrna factor that inhibits the flagellar-specific 0'28 in Salmonella typhimurium. Activation of 0'28 involves secreting FlgM outside the cell by a type III secretion apparatus that assembles functional flagella (49, 62). In this case, the integrity of the flagellar hook-basal body complex serves as a morphogenetic cue to relieve the inhibition of 0'28 by FlgM (49, 62). Other anti-sigma factors include the T4 bacteriophage AsiA protein that binds and inhibits E. coli 0'70 activity, Myxococcus xanthus CarR that sequesters to the inner cell membrane sigma factor CarQ involved in carotenogenesis (33), and Pseudomonas aeruginosa MucA that inhibits AlgT (GE) involved in alginate synthesis (94, 127, 152). Signal Transduction Pathways Leading to the Activation of Compartment-specific Sigma Factors. The temporal and spatial pattern of gene expression is established, in part, by four compartment specific sigma factors: 6F and 0'6 in the forespore, GE and 0K in the mother cell. Each of the sigma factors is initially inactive. Activation depends on signals generated by the previously activated sigma factor in the opposite compartment (68, 86, 137). Thus, activation of CF in the forespore following the formation of the polar septum triggers the activation of (SE in the mother cell. (SE, in turn, in conjunction with the engulfment of the forespore by the mother cell, causes the activation of 0'0 in the forespore. Finally, 0'0 sets in motion a chain of events that leads to the activation of OK. 17 CF to GE. GE is synthesized as an inactive precursor, pro-6E, the primary product of the promoter-distal member (spoIIGB) of the two-cistron spoIIG operon (59, 64). Pro-0'E is converted to its active form by proteolytic removal of its pro-amino acid sequence (74), an N-terrninal extension of 27 residues (97). Synthesis of pro-0'E starts before asymmetric division due to SpoOA~P activated transcription of the spoIIG operon (10, 12). The first gene of the operon, spoIIGA, encodes a membrane protein that is sufficient for activation of pro- OE and is believed to be the processing enzyme (1 10, 135). Activation of pro-(rE is delayed until after septation and it requires a (SF-controlled gene spoIIR (48, 61, 83). Simultaneous expression of spoIIR, spoIIGA, and spoIlGB during exponential growth leads to efficient pro-(5E processing, which indicates that spoIIR is the only (SF-controlled gene needed for activation of pro- GE (83). SpoIIR contains a putative signal sequence and can be found in the supernatant fluid from a culture of B. subtilis cells engineered to express spoIIR during exponential growth (48). Thus, the simplest model is that SpoIIGA is a receptor/protease and that SpoIIR is a secreted signal protein that activates the intracellular protease domain of SpoIIGA by interacting with an extracellular receptor domain of this integral membrane protein (Figure 5a). Biochemical evidence in support of this hypothesis has come from the demonstration that partially purified SpoIIR can activate pro-(SE processing in protoplasts and intact cells of B. subtilis that have been engineered to produce SpoIIGA and pro-0'E during grth (48). Mutagenesis analysis revealed a residue near the N-terminus (D6) is required for function. This residue is in a region predicted to be exposed to the space between the mother cell and the forespore (82). SpoIIR can also activate SpoIIGA molecules in the same cell where it is synthesized (125, 158). This raises the question of how pro-(3'E is prevented from becoming active in the forespore. Recently, it was shown that 0E and pro-oE are absent in the forespore 18 compartment shortly after septation, suggesting that the loss of (SE/pro-oE from the forespore contributes to the compartmentalization of (SE-directed gene transcription (112). Interestingly, a null mutation in spoIlIE, but not a missense mutation, allows 0’13 and/or pro-OE to persist and GE to become active in the forespore. The same null mutant also allows SpoIIE phosphatase to persist at the mother-cell pole after septation, causing CF to become active in both compartments (1 12). Thus, SpoIIIE may be a dual function protein. It contributes to the proper spatial regulation of both 6E and CF activities, in addition to being a DNA translocase as noted above. (IE to CC. One hour afler CF is activated, it starts to transcribe spoIIIG, which encodes 0'0 (102). 0’0 directs the transcription in the engulfed forespore of a large set of genes, including a family of small, acid-soluble proteins (SASP) that protect the spore DNA from different types of environmental insult (122). 0'0 can also maintain its own synthesis by recognizing the spoIIIG promoter (14]). (SO-directed transcription in the forespore is controlled by mother-cell signals acting at two successive levels. First, transcription of spoIIIG is dependent on the presence of CE in the mother cell, suggesting that the mother cell generates a signal required for expression of spoIIIG (102). The molecular nature of this signal is unknown, but it does not appear to be related to the morphological development (137). Second, full activation of CG requires not only completion of engulfment, but also the eight products of the (IE-controlled spoIIIA operon (134). Genetic studies suggest that like 6F, 0'0 is also held inactive by SpoIIAB (114). The spoIIIA-dependence of (SO—controlled gene expression can be overcome by a mutation that impairs 0.0 in its binding to SpoIIAB (63). If SpoIIAB does repress both 6F and CG 19 6F -L1TpEIIT|- I SpoIIR Figure 5. Models for the intercompartrnental signaling pathways leading to the activation of pro-GE and pro-0K. (a) Activation of pro-6E. See text for details. (b) Activation of pro-6K. See text for details. Reprinted from ref. 137. 20 activity, then its repression of these sigma factors must be relieved by different mechanisms, because 6F and CG are activated sequentially. There is some evidence that SpoIIAB might be preferentially degraded in the forespore in a spoIIIA -dependent fashion (66). 0'0 to OK. OK, like 05, is first made as an inactive precursor. Pro-oK has 20 amino acid residues at its N-terminus that are not present in OK (69, 136). Pro-oK is first made at 3 hours into sporulation, while active 0K appears an hour later (89). Processing of pro-oK not only depends on several mother-cell-specific genes, but also on several forespore-specific genes, including spoIIIG, the gene that encodes the forespore-specific sigma factor, 00 (18, 89). A (SO-controlled gene, spoI VB, is involved in signalling the processing of pro-oK (17, 89, 145). Evidence shows that SpoIVB is the only protein produced under 0'0 control that is needed to trigger pro-0‘K processing (32). SpoIVB may, play a direct role in activating processing of pro-0K, possibly by interacting with the processing enzyme. Alternatively, SpoIVB may play a structural or enzymatic role in serving as part of the processing signal from the forespore. Genetic studies have suggested that SpoIVFB, the product of the promoter distal gene of the spoI VF operon, may encode the pro-<3K processing enzyme or, alternatively, a regulator of the processing event (1 8). SpoIVFA, the product of the promoter proximal gene of the spoI VF operon, is suggested to play dual roles in regulating the activity of SpoIVFB (19). In its positive-acting role, SpoIVF A is required to stabilize SpoIVFB, which is suggested to be thermolabile. In its negative-acting role, SpoIVF A inhibits the activity of SpoIVFB until a signal(s) from the forespore is received. BofA, the product of the (IE-dependent bofA gene, also plays a negative role in pro— 0K processing until a signal from the forespore is 21 received (116). Both spoI VF and bofA are transcribed by GE RNAP and thereby their expression is confined to the mother cell. The proteins encoded by these genes have potential membrane spanning domains. Indeed, SpoIVFA and SpoIVFB have been localized to the membrane surrounding the forespore (115). Based on these observations and the results from the genetic studies, it was proposed these proteins form an oligomeric complex in the outer membrane of the forespore and that from this position sense either a SpoIVB-dependent morphological change in the membrane and/or a spoIVB-dependent signal from the forespore (Figure 5b). It is of interest to compare the processing machinery of pro-(SK to that of pro-GE. SpoIVFB seems to be active in its default state, requiring the SpoIVB signal protein to overcome the inhibitory effects of SpoIVFA and BofA, whereas its counterpart, SpoIIGA, appears to be inactive in its default state, requiring the SpolIR signal protein to become active (Figure 5). Consistent with SpoIVF B being an active protease, coexpression of SpoIVFB and sigK genes in growing B. subtilis or E. coli enhanced pro-0’K processing in the absence of other sporulation-specific gene products (88). Sequence analysis revealed that SpoIVFB contains a potential aspartyl protease motif. Mutation of a key amino acid residue in the motif inactivated SpoIVF B (155). Pro-t3K may also be subject to processing by a SpoIVFB-independent pathway. This was first suggested by an experiment in which overproducing pro-6K in spoI VF null mutant cells resulted in partial restoration of oK-dependent gene expression and sporulation by allowing accumulation of a small amount of OK (90). sop (suppressor of processing defect) mutants have now been isolated that are able to partially rescue (SK-dependent gene expression and sporulation in the absence of spoI VF gene products (4). In cells in which the need for pro-oK processing has been uncoupled from 22 dependence on a forespore signal(s), transcription of oK-dependent genes is advanced by 1 hour. This decreases the sporulation efficiency by 10-1,000 fold and the spores produced are germination defective (18, 32). In this case, coupling of UK activation in the mother cell to forespore morphogenesis is obviously of great biological importance. This is the only example of a cell-cell communication pathway during sporulation for which the physiological consequences of bypassing it are known. Like that of GE and GK, the activity of several eukaryotic transcription factors is regulated by proteolysis. NF-KB is a transcription factor that affects the expression of genes involved in immune function, inflammation, and cellular growth. Activation of NF- KB involves the proteasome-mediated destruction of an inhibitory protein, IKB, which masks the nuclear localization sequence of NF-KB (100). The sterol regulatory element-binding proteins (SREBPs) are derived from integral membrane proteins that are held inactive by virtue of being sequestered at the nuclear envelope and the endoplasmic reticulum. 1n sterol-depleted cells, the membrane-bound precursor is subject to two sequential cleavages, one within a transmembrane domain, to release a soluble fragment that translocates to the nucleus and activates transcription (117, 146). Finally, the Drosophila regulatory protein Ci (Cubitus interruptus) is initially inactive because it is tethered in the cytoplasm. It is cleaved to generate a form that lacks the tethering domain and migrates to the nucleus through a pathway that is governed by the Hedgehog signalling protein (6). Hierarchical Regulatory Cascade of the Mother-cell Line of Gene Expression. Regulation of mother cell gene expression is governed by a hierarchical regulatory cascade consisting of four key regulatory proteins, OE, SpoIIID, 6K, and GerE, in which each regulatory protein is responsible for the production of the next one (Figure 6). Shortly after cells commit to sporulation and before the septum forms, an increased 23 SpoOA-P level stimulates the transcription of spoIIG (sigE), which encodes SpoIIGA and pro-OE, by CIA-RNAP (10, 12). Transcription of sigE is repressed by SinR, a stationary phase regulator (129). SinR inhibits sporulation by repressing transcription of spoOA, sigE, SpoIIE and spoIIA (8, 91, 92). The activation of pro-0'E in the mother cell after the completion of septation sets in motion (SE-dependent gene transcription, including transcription of the regulatory gene spoIIID (70, 132). SpoIIID is a 10.8 kDa DNA=binding protein that activates or represses certain genes in the GE and OK regulon (40, 69, 156), regulating the timing and/or level of transcription of these genes (Figure 6). Among the (SE-dependent genes activated by SpoIIID are genes involved in the appearance of UK, including spoIVCA and sigK (39, 40, 69). The appearance of UK is subject to multiple levels of regulation (18, 89, 99, 136). First, OE acting in conjunction with SpoIIID turns on spoIVCA (39, 136), which encodes a site-specific recombinase (119). SpoIVCA recombinase catalyzes a chromosomal rearrangement event joining two truncated genes, spoIVCB (encoding the N-terrninal half of OK) and spoIIIC (encoding the C-terrninal half of OK) (71, 136), to form the composite sigK gene. Since 0'5 and SpoIIID are produced exclusively in the mother cell, the chromosomal rearrangement does not occur in the forespore. A second level of regulation is the transcription of sigK, which requires initially the concerted action of GE RNAP and SpoIIID, and then OK and SpoIIID (40, 69, 72). The appearance of GerE, the last regulatory protein in the cascade (discussed below), negatively regulates the transcription of sigK by GK RNAP (50, 159), possibly contributing to the maintenance of a proper level of OK. A third level of regulation is the activation of <5" by the removal of the pro-arnino acid sequence (69, 89). As we have seen, pro-0'K processing is coordinated with and 24 Figure 6. Diagram of gene regulation in the mother-cell cascade. Dashed lines with arrowheads represent gene-to-product relationships. Arrows and lines with barred ends indicate positive and negative effects, respectively, on expression. SpoOA~P activates the transcription of sigE by GA RNAP. SinR inhibits sigE expression. 0’5 RNAP transcribes spoIIID. SpoIIID regulates genes in both the GE and 0K regulons. Among them, it activates sigK transcription and represses transcription of certain cot genes. Transcription of sigK is directed first by GE and then by its own gene product, 0'". The cot D, C, and X genes are transcribed by GK RNAP. In one feedback loop, 6K negatively regulates the spoIIID mRNA level (not shown). A diminished SpoIIID level allows the previously repressed cot genes to be transcribed. GerE is a gene product of the 6K regulon. It represses transcription of sigK, forming another feedback loop in the mother-cell cascade of gene expression. GerE also activates transcription of the cot D, C, and X genes, reinforcing the switch in the mother-cell pattern of gene expression initiated by the decrease in the level of SpoIIID. \\ as 560% v / \n::o.n_m n_.. uoqm " \ WNW 55:60 $60 memo. 39m «Sinai/w Sol/a Dione muBTume /m < 26 controlled by events in the forespore. SpoIIID represses several late O’K-dependent genes, notably some of the spore coat genes (Figure 6). A decrease in the SpoIIID level late during sporulation is postulated to be critical for switching on these genes. The SpoIIID level decreases rapidly after reaching its maximum during sporulation (38). This decrease was found to depend on OK (3 8). Thus, the appearance of CK negatively regulates the level of SpoIIID (38). Chapters 2 and 3 of this dissertation discuss the continued investigation into the mechanism by which OK negatively regulates the level of SpoIIID. In addition, degradation of SpoIIID may involve the conversion of SpoIIID to an unstable 9 kDa form, apparently by removing 7 amino acid residues from its C-terminus (37). The conversion is a developmentally regulated event that is independent of OK (3 7). GerE is an 8.5 kDa protein that contains a putative helix-turn-helix DNA-binding motif. As we have seen, GerE represses transcription of sigK. It activates the expression of many spore coat genes transcribed by GK RNAP, some of which are repressed by SpoIIID (Figure 6). Thus, the appearance of GerE reinforces the switch in the mother cell pattern of gene expression initiated by the decrease in the level of SpoIIID. Interestingly, the SpoVT DNA-binding protein encoded by a (SO-dependent gene seems to play a similar role in the regulation of late gene expression in the forespore (7, 126). SpoVT is believed to bind to some regulatory DNA sequences and allow expression of the latest class of forespore-specific genes. Analogous to GerE repression of sigK transcription, SpoVT inhibits spoIIIG transcription, perhaps contributing to a progressive shut-off of 0'0 activity (7)- Hierarchical regulatory cascades are also found in other systems. An excellent example is flagellum biosynthesis in E. coli and S. typhimurium. The flagellar genes are 27 grouped into 13 operons. These operons have been divided into three classes. Class I genes are required for the expression of class 11 genes and class 11 genes are required for the expression of class 111 genes (1, 73, 149). Environmental cues trigger the expression of Class 1 genes, which encode the two master regulatory proteins Fth and FlhD. These proteins are required for the activation of CA RNAP for transcription of Class 11 genes, which encode the basal body-hook complex. Completion of this structure acts as the assembly checkpoint to regulate transcription of late (Sm-dependent Class 111 genes, by allowing anti-<528 protein FlgM to be secreted outside the cell (49, 62). In C. crescentus, the hierarchy of flagellar assembly involves four classes of genes and two assembly checkpoints (149). Summary. Gene expression is regulated temporally and spatially during development of B. subtilis through synthesis and activation of different sigma factors. Synthesis of late sigma factors depends on the activity of earlier ones in the same compartment. Activation of sigma factors involves signalling between the two compartments and in some cases appears to be coupled to the establishment of a morphological structure. Studies on the model system of B. subtilis sporulation are likely to continue to provide insight into the fundamental question of how genes are regulated temporally and spatially during development. Chapter II A Feedback Loop Regulates the Switch from One Sigma Factor to the Next in the Cascade Controlling Bacillus subtilis Mother-cell Gene Expression Reprinted from the Journal of Bacteriology 28 JOURNAL or Bacremowov. Oct 1997, p. 6138-6144 0021-9193/97/304.00+0 Copyright 0 1997. American Society for Microbiology 29 Vol. 179, No. 19 A Feedback Loop Regulates the Switch from One Sigma Factor to the Next in the Cascade Controlling Bacillus subtilis Mother Cell Gene Expression BIN ZHANG AND LEE KROOS’ Department of Biochemistry, Michigan State University, East Musing, Michigan 48824 Received 15 May 1997/Accepted 15 July 1997 Regulation of gene expression in the mother cell compartment of sporulating Bacillus subtilis involves sequential activation and inactivation of several transcription factors. Among them are two sigma factors, a” and o“, and a DNA-binding protein, SpoIIID. A decrease in the level of SpoIIID is thought to relieve its repressive elect on transcription by tr" RNA polymerase of certain spore coat genes. Previous studies showed that o" negatively regulates the level of spoIIID mRNA. Here, it is shown that 0" does not alect the stability of spoIIID mRNA. Rather, 0" appears to negatively regulate the synthesis of spoIIID mRNA by accelerating the disappearance of 0‘ RNA polymerase, which transcribes spoIIID. As 0" begins to accumulate by 4 h into sporulation, the 0' level drops rapidly in wild-type cells but remains twofold to fivefold higher in sigK mutant cells during the subsequent 4 h. In a strain engineered to produce tr" 1 h earlier than normal, twofold less oI than that in wild-type cells accumulates. 0 did not detectably alter the stability of o" in pulse-chase experiments. However, fl-galactosidase expression from a sigE-Incl transcriptional fusion showed a pattern similar to the level of a" protein in sigK mutant cells and cells prematurely expressing 0". These results suggest thattheappearanceofo‘initiatesanegativefeedhachloopcontrolling notonlytranscrlptlonof spolm2,buttheudno'regubmbydlnalywhdheaiyhhiflflngthemnsaipdmdsigfi Sporulation of the gram-positive bacterium Bacillus subtilis is a model system for studying developmental gene regulation (8). In response to starvation, B. subtilis undergoes a series of morphological changes that culminate in the formation of an endospore. Early during sporulation, an asymmetrically posi- tioned septum partitions the developing cell into two unequal compartments, the mother cell and the forespore, each of which carries a copy of the chromosome. The two compart- ments follow dilferent programs of gene expression that drive further morphological changes, including migration of the sep- tum to engulf the forespore, deposition of cell wall-like mate- rial called cortex between the two membranes surrounding the forespore, formation of a tough protein coat that encases the forespore, and lysis of the mother cell to release the endo- spore. Temporal and spatial gene regulation during sporula- tion is established by compartment-s ' c activation of a cas- cade of sigma factors, namely, 0", o , 0°, and tr", in order of their appearance (26, 34). The forespore-Opedfic program of gene expression is controlled (IF and o , while the mother cell program is controlled by and 0". Each sigma factor is initially inactive. trF is the first to become active, and this occurs only in the forespore (13, 32, 39). Activation of subse- quent sigma factors in the cascade is triggered by signal trans- duction between the two compartments (12, 34). The inactive forms of the mother cell-specific sigma factors are precursor proteins called pro-trE and pro-o". Each is synthesized about 1 h before it is activated by proteolysis (6, 30, 35). Temporal gene regulation in the mother cell is established primarily by the ordered appearance of a“: and then tr". Also involved is a transcription factor, SpoIIID, whose mRNA is synthesized by as RNA polymerase (28, 49, 52). SpoIIID is a sequence-specific DNA-binding protein that activates or re- ‘ Corresponding author. Mailing address: Department of Biochem- istry, Michigan State University, East Lansing, MI 48824. Phone: (517) 355-9726. Fax: (517) 353-9334. E-mail: kroos@pilot.msu.edu. 6138 presses many difierent genes transcribed by as andlor 0" RNA polymerase (10, 27, 55). One of the genes activated by SpoIIID is sigK, which encodes pro-tr". The sigK gene is constructed during Sporulation by a DNA rearrangement that joins spoIVCB (encoding the N-terminal part) and spolIlC (encod- ing the C-terrninal part) (51), and SpoIIID also activates tran- scription of spoIVCA (10, 45), the site-specific recombinase that catalyzes the rearrangement (29, 43, 46). Hence, SpoIIID plays a key role in progression from the early (re-directed pattern of gene expression to the late oK—directed pattern. Somewhat paradoxically. SpoIIID represses certain late genes in the o" regulon, apparently fine-tuning their timing and/or level of expression (10, 27, 56). How is the repressive effect of SpoIIID on late gene expression relieved? We showed previ- ously that the SpoIIID protein level decreases abruptly when (7" appears during sporulation (9). Also, in mutants that fail to make active 0", both SpoIIID and its mRNA persist at a higher level until later during 5 rulation compared to wild- type cells. This suggests that a negatively regulates the syn- thesis and/or stability of spoIlID mRNA. As the existing SpoIIID is degraded, the oK-dependent genes that were re- pressed by SpoIIID would begin to be transcribed. Here, we describe our continued investigation of the nega- tive feedback loop connecting the production of o" to the SpoIIID decrease during sporulation. We demonstrate that 0" does not affect the stability of spolIlD mRNA; therefore, it must exert its negative effect on spollID transcription. Indeed, a spolIlD-lacZ fusion is overexpressed in sigK mutant cells (28). Transcription of spollID is carried out by 05 RNA poly- merase (28, 49, 52). We show here that 0" also negatively regulates the o“: level, providing a simple explanation for the negative efiect of 0" on spolIlD transcription. 0" directly or indirectly inhibits the transcription of sigE (encoding «13). based on the levels of expression from a sigE-Incl transcrip- tional fusion in wild-type and different mutant strains. Thus, VOL. 179, 1997 or" initiates a negative feedback loop that controls not only spoIlID expression but expression of the entire o‘h regulon. MATERIALS AND METHODS Bacterial strains. B. subtilis BK556 (spell/€823) (28). V048 (spell/(BAN car) (6). and $076 (ho/B8 car) (6). which are isogcnic with the wild-type Spo‘ strain PY79 (54). were provided by R. Lmiclt. Strain 82.536 (P,,,,,,-P,,x,,—sigKA19 spc) was constructed by first replacing the car allele of V0536 (PW, Pu“,- sigKAIO cal) (40) with a spectinomycin (spc) allele by using plasmid pCm::Sp (48) and then by using the chromosomal DNA of the resulting strain to transform competent PY79 cells and to select for a spectinomycin-rcsistant transformant. General methods. Preparation of competent cells for transformation with plasmid DNA or chromosomal DNA was described previously (14). Sporulation was induced by resuspending growing cells in SM medium as described previ- ously (14). The onset of sporulation (T0) was defined as the time of resuspension. Use of the specialized transducing phage SPB:;sr'gE-lacl (also called spoIIG- lacZ) has been described elsewhere (24). B-Galactosidasc activity was assessed qualitatively by placing cells on DSM agar (14) containing 5-bromo—4-chloro-3- indolyl-B-o—galactopyranoside (X—Gal) (20 ug/ml) and was determined quanti- tatively with toluene to permeabilize cells and o-nitrophenol-B-o-galactopyrano— side as the substrate (14). One unit of enzyme hydrolyzes 1 umol of substrate per min per Am of initial cell density. Measurement of the stability afapallID mRNA. At the fourth hour of sporu- lation. rifampin (75 ug/ml) was added to cultures to stop transcription initiation. Samples (9 ml) were taken before and immediately after the addition of rifampin and were centrifuged at 12.11!) X g for 1.5 min. Cell pellets were frozen in a dry ice-ethanol bath. Samples were also taken at 5 and 12 min after the addition of rifampin. The process of sample collection took about 3 min to complete; there- fore, the first time point immediately after rifampin treatment was designated as 3 min after the stoppage of new transcription in cells. Likewise, the ensuing time points were designated as 8 and 15 min after the stoppage of transcription. RNA was prepared by the acid guanidinium thiocyanate-phenol-chloroform method (3) with the following modifications. Cell pellets were resuspended in 1.5 m1 of denaturing buler (4 M guanidinium thiocyanate, 25 mM sodium citrate [pl-l 7.0]. 0.5% sarcosyl, 0.1 M 2-mercaptoethanol) and 1.5 ml of acid phenol-chloroform (5:1) (Ambion) and then mixed vigorously with l-ml glass beads (500 um. acid washed; Sigma) to break cells. 1he mixture was centrifuged at 10,111) x g for 20 min at 4‘C. After centrifugation. the aqueous phase was re-extracted with acid phenol-chloroform. RNA was precipitated by ethanol. Residual DNA was re. moved by digesting with RNase-free DNase. RNA (20 ug) was fractionated on a 1.2% (wt/vol) agarose gel containing 1.11% (vol/vol) formaldehyde, transferred to a nylon membrane. and hybridized to a random primed 1.1-lib DNA fragment containing the spolllD coding sequence purified from pBlCl9 (28) digested with Fall. The radioactive signals were quantified with a Phosphorlmager (Molecular Dynamics). Western blot analysis. Preparation of whole-cell proteins, electrophoresis. and electroblotting were described previously (9, 35). The membrane was with monoclonal anti-trI5 antibody (30) diluted 1:611) or polyclonal anti-SpoIIID antiserum (9) diluted 1:10,”. Chemiluminescence detection was performed according to the manufacturer’s instructions (ECL; Amersham). When neces- sary. the membrane was then stripped of the bound antibodies and reprobed with polyclonal anti-prov" antiserum (35) diluted 1:10.111). Signals were quantified with a computing densitometer (Molecular Dynamics). Exposure times that gave maximum signal intensities within the linear response range of the X-ray film. as determined by control experiments, were used. Pulse-chase and lmmunopreeipltatiaa. At the third hour after the onset of morulation. cells were pulse-labeled by adding [”SImethionine (ICN) (35 p.01 ml) to the culture and incubating for 5 min. Excess (Lino-fold) unlabeled methionine and cysteine were then added. and the incubation was continued at 37‘C. Cells (1 ml) were collected by centrifugation immediately following the pulse-chase and at 30-min intervals thereafter until the fifth hour after the crust of sporulation. The cells were froaen in a dry ice-ethanol bath and stored at -70'C Cell pellets were reauspended in 50 p1 of lysis bufier (10 mM Tris-Cl [pH 8.4], 1 mM EDTA. 10 mM MgCl, 1 mM phenylmethylsulfonyl fluoride, 0.5-mglml lymyme, 0.1-mg/ml DNase I) and incubated for 10 min at 37'C. Sodium dodqu sulfate (SDS) was added to a concentration of 1%. and samples were boiled for 3 min. Lysates were centrifuged at 12.11!) x g for 10 min. Lysates of dilerent samples contained approximately equal amounts of radioactivity as judged by SDS-polyacrylamide gel electrophoresis (PAGE) and autoradiography. After centrifugation, the supernatant was diluted 10-fold in immunoprecipitation buler (50 mM Tris-Cl, 150 mM NaCi, 1 mM EDTA, 1% Nonidet P-40, 0.2% deoxycholate. 1 mM phenylmethylsulfonyl fluoride). Monoclonal anti-trE anti- body (80 ul). which was sulicient to quantitatively precipitate pro-oIE and on from a 1-ml culture in a control experiment. was added. and the mixture was incubated at 0°C for 2 h. A slurry (20 id) of 1:1 (vol/vol) protein A-Scpharose CL—48 (PharmaciaHmmunoprecipitation buler was then added, and the incu- bation was continued with gentle mixing on a rotary shaker at 4'C for 1.5 h. Samples were centrifuged briefly. The pellets were washed three times with 1 ml of im ' ' lion buIer tried with 0.1% SDS and then resus- pcndcd in 30 ul of SDS sample Met (31), boiled for 5 min. and centrifuged FEEDBACK REGULATION IN A CASCADE 6130 (A) wr 0381503815 . r ‘I b. . a u .‘ " . . ‘._-’.::'” I“ '4 ‘1'“... 1‘ . .pA _ . , ., .z ' ” C". . ' ‘5 ' i. - ." ,-‘ 41. 1‘“ .1 :- . . .. . . '- t - . - , '.‘.' .~ J - 7‘- , F .rg‘v... '_ _ q. v-u’r‘ "- a .1 a. '.- " stanzas use... . .-. .91.. \ \. \ \ (B) "”1 80‘ El wr sigK \\\\\\\\\\\\ \ spolllD mRNA Laval \\§ \ / 0 3 I l 5 Time (min) FIG. 1. The production of o" does not change the stability of moIND mRNA (A) Wild-type (PY79) and sigK mutant (81656) strains were indiud to sporulate by resuspension in SM medium. Rifampin (75 uyml) was added to the medium at the fourth hour after the onset of sporulation. RNA was prepared from cells collected before and at the indicated number of minutes after the addition of rifampin. and equal amounts (20 pg) were analyzed by Northern blot armlysis (B) The level of 11101110 mRNA, as quantified by a Phosphorlmager. is plotted as a percentage relative to the level before rifampin treatment. WT, wild UPC- again to remove Sepharose beads. lmmunoprecipitates were analyzed by SDS- PAGE. Pro-oz and 0‘ bands were visualized by fluorography with EMENSIFY (Dupont) as enhancing floors and quantified by a Pliosphorlmager (Molecular ‘ Dynamic). with the background of each lane subtracted from the band intensity. RESULTS Stability of spoIIID mRNA in wild-type and sigK mutant cells. We showed previoqu that the spoIIID mRNA level reaches a higher maximum and remains higher late during sporulation of sigK (spolllC94; spoIlIC encodes the C-terrninal part of 0") (51) mutant cells compared to that of wild-type cells (9). Similar results were obtained when cells containing another sigK mutation, spoIVCBZ3 (spoIVCB encodes the N- terminal part of a“) (51) (both spoIIIC94 and spoil/€823 cells fail to make it") (35), were analyzed (data not shown). This sigK mutant was used in the studies reported here. The higher level of spolllD mRNA in the sigK mutants must be due to increased synthesis and/or stability of spoIIID mRNA. To measure the stability of spolllD mRNA, sporulat- ing wild-type and sigK mutant cells were treated with rifampin at T4 (i.e., 4 h after starvation initiated sporulation) to stop transcription initiation. Total cellular RNA was isolated from cells collected before and at different times after the rifampin treatment. Northern blot analysis was performed to detect spolllD mRNA. At T4, there was already more spoIIID mRNA in sigK mutant cells than in wild-type cells (Fig. 1A), and a considerable amount of o" was present in the wild-type cells (data not shown). The amount of spolllD mRNA remaining at difierent times after the rifampin treatment is shown in Fig. 1A and was quantified with a Phosphorlmager. The half-life of spolllD mRNA at T. is about 3.5 min in both wild-type and 31 6140 ZHANG AND KROOS (A) 12345678910 WT ‘§“-,- - .( (B) 1201 i g 100‘ so- u t e and . t z :2- so‘ . t 1‘ so- 0 fl I I I I I I 1 a s a s s 7 a s 10 Hours Altar Rasuapanslon (e) 2 3 4 s s 1 .‘T;§--o-v:( inn?“ 4.; a'- ‘ s1. FIG. 2. on persists at a higher level during sporulation of cells defective in 0" production. Whole-cell extracts were prepared from wild-type (PY79) and sigK mutant (81656) cells collected at the indicated numbers of hours after the onset of sporulation in SM medium. Proteins (5 pg) were fractionated on an SDS-12% polyacrylamide gel and subjected to Western blot analysis with either monoclo- nal anti-oE or polyclonal anti- pro-0'" antibodies (A) Levels of a“ in wild- -type (WT) and sigK mutant cells. Arrowheads 0“ signal (the faints inal of lesser mobility most apparent at T, is pro-GE). (B) Relative amounts of in wild-type (O) and sigK mutant (0) cells during sporulation. The oE signals in three experiments with both the wild-type strain and the sigK mutant and two exper- iments with just the wild~type strain were quantified with a computing densitom- eter. For each experiment, the signal intensrties were normalized to the maxi- mum signal in wild-type cells Points on the graph are averages of the normalized values, and error bars show one standard deviation of the data. (C) Levels of o" in wild-type cells. Arrowhead, tr" signal (the faint signal of lesser mobility first appearing at T, is pro-<7"). sigK mutant cells (Fig. 18). Since no substantial ditference in the stability of spollID mRNA was detected, the higher level of spoIlID mRNA in sporulating sigK mutant cells must be due to increased synthesis of spollID mRNA. In support of this idea and in agreement with the results of Kunkel et a1. (28), we found that a spoIIID-lacZ transcriptional fusion is overex- pressed by approximately 1.7- fold in sigK mutant cells com- pared to wild- -type cells (data not shown). a” level in wild-type and sigK mutant cells. Since spoIIlD 18 transcribed by oE RNA polymerase (28,49 52), we reasoned that increased spoIIlD transcription in sigK mutant cells might result from an elevated level of 0B .We measured the level of Ein extracts of wild- -type and sigK mutant cells using anti-o“; antibody (30) in Western blot analysis. To facilitate the com- parison, the two strains were induced to sporulate in parallel cultures and equal amounts of protein in whole-cell extracts were electrOphoresed in the same SDS- Epo-lyacrylamide gel. Figure 2A shows that in wild- -type cells, o‘i was first detected at T2 and reached a maximum level by T3, and the level decreased rapidly thereafter. in sigK mutant cells, the a” level remained .1. B~\(T1;‘Rl()l. (A) 12345678910 WT I. ....-,.. — u ( .‘IM -' umpires TEE-‘1’" 810K '---¢--aa'.-—‘n.. ... ( (B) E 260‘ o “ zoo- 9 =0 iso-l O. m 100‘ o .2 ‘6' so~ 3 C 04 a . . 12345078010 Hours Altar Rasuspsnalon FIG. 3. SpolllD persists at a higher level during sporulation of cells defective in 0" production. (A) Proteins (1.7 ing) in the same samples shown in Fig. 2A were fractionated on an SDS—18% polyacrylamide gel and subjected to Western blot analysis with anti-SpoIIID antiserum. Arrowheads, SpolllD signal. WT. wild type. (8) Relative amounts of SpolllD in wild-type (PY79 [0]) and sigK mutant (81656 [0]) cells during sporulation. The SpolllD signals in three experiments were quantified. normalized. and plotted as described in the legend to Fig. 2. high at T.. and thereafter its level decreased less rapidly than that in wild type cells. The experiment was repeated several times and the Western blot signals were quantitated. Figure ' 28 shows that, after T3, oE reproducibly persisted at a level in sigK mutant cells Ehigher than that in wild type cells. Between T5 and T8, the 05 level was twofold to fivefold higher in the sigK mutant than that in the wild type. Similar results were obtained when other mutants that fail to make 0" (i.e., cells containing a spolllC94 or a spoIVCAI33 mutation) (35) were tested (data not shown). We also tested mutants (i..,e those containing spoIIIGAI and spoIVFAAB: :car) that produce pro-0" but fail to process it to active 0" (35, 36). Again, similar results were observed (data not shown), indicating the pro-oK must be processed to active 0" in order to accelerate the disappearance of (IE from sporulating cells. Moreover, as shown in Fig. 2C, processing in wild- -type cells causes 0" to begin accumulating by T., which is the earliest time that the 0'5 level is lower in wild type cells than that in sigK mutant cells (Fig. 28) We conclude that the appearance of active 0" ac— celerates the disapp‘earance of 05 during sporulation. We note that 0 E15 not essential for the level of 05 to decrease, since the 0'“ level eventually declines in mutants that fail to make a“ (Fig. 28 and data not shown). Cell lysis IS not the explanation for the decrease in ch in the mutants or for the more rapid decrease in 05 in wild- -type cells. Although a small amount of cell lysis began to occur after T7 in both the wild- type and the mutant cultures, the ability to recover protein from sedimented cells never varied by more than 10% during the course of our experiments. We also measured SpoIIID levels in most of the samples used in the experiments summarized in Fig. 28. Figure 3A shows the results for the same samples used in the experiment shown in Fig. 2A. Figure 3B shows quantitation of several experiments. SpolllD accumulated by T5 to a level in sigK 32 VOL. 17‘), 1997 mutant cells that was twofold higher than that in wild-type cells. The level of SpolllD in the wild-type strain decreased threefold by T6, while in the sigK mutant the SpolllD level remained high until T6 and then declined thereafter. Thus, the levels of both SpolllD and 05 are significantly higher in sigK mutant cells than in wild-type cells between T4 and T8 of sporulation. The absence of 0" has a larger effect on the SpolllD level than on the as level (compare Fig. 28 and 3B). This difference might be explained by the fact that oE RNA polymerase acts enzymatically to increase spolIlD transcrip- tion and/or by the observation that SpoIIID positively auto- regulates spolllD transcription (23, 28, 49, 52). Thus, a rela- tively small effect on a could lead to a larger effect on SpolllD. Clearly, o" negatively regulates the 0 level during sporulation, providing a simple explanation for the negative elfect of 0" on the SpoIIID level. 0‘ level in cells that produce 0" earlier. it was shown previously that earlier production of 0" during sporulation resulted in less accumulation of SpoIIID and earlier disappear- ance of SpolllD (9). To examine whether these effects might also be explained by a negative efiect of 6" on 05, we moni- tored the level of 0'5 in spoIVCBAI9 mutant cells. In these cells, codons 2 through 20 of sigK, which encode the N-tenni- nal prosequence of pro-0", are missing, resulting in production of active 0" 1 to 2 h earlier than normal (Fig, 4A) (6. 9). As documented in Fig. 4B and C, the maximum level of o'E in spoIVCBAI9 mutant cells reached only about 50% of the wild- type maximum. These results support the idea that the appear- ance of o" negatively regulates the (IE level during sporulation. Turnover of pro-0": and o": in wild-type and sigK mutant cells. 0" might negatively regulate the (IE level by destabilizing as, possibly by directly competing with oE for core RNA poly- merase. it has been suggested that (IE is unstable in cells when it is not bound to core RNA polzmerase (21). A complication in measuring the stability of o is that it is generated from pro-oIE by proteolytic processing (30). However, since process- ing of pro-oE occurs normally in sigK mutant cells (Fig. 2A , we reasoned that a comparison of the total amounts of pro-o and (IE remaining at different times after pulse-labeling of sigK mutant cells and wild-type cells should reveal a difference in as stability, if it exists. Sporulating’wild-typc and sigK mutant cells were pulse-labeled at T3 with [ 5S]methionine and chased with an excess amount of unlabeled methionine. We chose T3 to perform the labeling because during the subsequent hours of sporulation large difl'erences in the levels of crE between wild- type and sigK mutant cells were observed (Fig. 28). Samples were collected every half hour after the pulse-labeling, and pro-oE and 0‘3 in crude cell extracts were irnmunoprecipitated with monoclonal anti-oE antibody. The pro-0‘5 and as signals were revealed by SDS-PAGE and fluorography (Fig. 5A). The [3581methionine was first incorporated into pro-«rE through protein synthesis and then appeared as oE upon roteolytic cleavage of the N-terminal sequence from pro-o . A small portion of the pro-oE had already been processed into 11'5 at the end of the 5 min of pulse-labeling (labeled 0 min in Fig. SA). Upon incubation, the 35S label was chased into 0’5, and eventually oE was de raded. Figure SB shows that the decay rate of pro-o‘a plus a was similar in wild-type and sigK mutant cells, as judged by the quantification of the combined signal intensities of pro-oE and GE. When the experiment was re- peated and samples were collected at 45-min intervals after pulse-labeling, again no substantial difference between wild- type and sigK mutant cells was observed (data not shown . Therefore, destabilization of (re upon the appearance of 0 cannot explain the level of o“ in wild-type cells being lower than that in sigK mutant cells at T4 to T5 of sporulation (Fig. 2). FEEDBACK REGULATION IN A CASCADE hl~il (A) 2 3 4 5 6 7 a d‘fi'wj" ‘5 ‘ "' ‘ ‘ ’._t . < P ingrw. -‘ .- (B) 2 3 4 5 6 7 a spoIVCBA 19 “-5- < (C) Relative (:5 Level 1 2 3 4 5 G 7 8 Hours After Reeuepenelon FIG. 4. a": disappears earlier during sporulation of cells that produce 0" earlier than normal. Whole-cell extracts were prepared from wild-type (PY79) and spoIVCBA/9 mutant (V048) cells collected at the indicated numbers of hours after the onset of sporulation in SM medium. Proteins (5 pg) were fractionated on an SDS—12% polyacrylamide gel and subjected to Western blot analysis with either monoclonal anti-oE or polyclonal anti-pro-o" antibodies. (A) The level of o" in the spoIVCBAIO mutant. Arrowhead, 0" signal. (8) Levels of o5 in wild-type (WT) and spoIVCBAI9 mutant cells. Arrowheads, 6E signal (the faint signal of lesser mobility most apparent at T, is pro-OE). (C) Relative amounts of o'E in wild-type (O) and spoIVCBAl9 ([3) cells during sporulation. For the wild-type strain. the data shown in Fig. 28 are also shown here. In three of the experiments with wild-type cells. the spoil/CHAN mutant was induced to sporulate in a parallel culture. The oE signals were quantified. normalized. and plotted as described in the legend to Fig. 2. Expression of a sigE-Incl transcriptional fusion in wild-type cells, sigK mutant cells, and cells producing it" earlier. Since 0'" did not appear to affect the stability of are, we tested the possibility that 0" may affect the transcription of the sigE gene that encodes pro-OE. sigE (also called spoIIGB) is the second gene in the spoIIG operon (22, 24). The first gene of the operon, spoIIGA, encodes a putative protease that processes pro-oE to 0'5 (16, 41, 50). First, we tried to directly compare the levels of sigE mRNA in sporulating wild-type and sigK mutant cells by Northern blot analysis. In agreement with a previous report (24), we found that 3135 mRNA was unstable and subject to processing or breakdown. Despite the difficulty in detecting sigE mRNA, we noticed that slightly more sigE mRNA appeared to be present in sigK mutant cells than that in wild-type cells at T3 and later times during development (data not shown). We then examined expression of a sigE-lacZ tran- scriptional fusion as a simple, albeit indirect. measure of sigE transcription. We introduced a sigE-lacZ transcriptional fusion 33 6142 ZHANG AND KROOS (A) WT SIQK 0 30 60 90120 0 30 60 90120 - .. ~ _- -.. __, PrEeoE ---___---“’6 (B) 2100 V a W 2 El WT E Z 7% sng I a In 0 / // E so % / "‘ 20 Q/ 0 / 2 /// a. it so so 90 120 Time (min) .The roduction ofo I‘does not alter the stability of pro-oE and 0" (A) Wild type (PY79) and sigK mutant (BK556) cells were labeled at the third IIuIII qucI ltd chased with excess amounts of unlabeled methionine and qsteine. Cells were collected immediately and at the indicated numbers of minutes following pulse- labeling. Whole-cell extracts were prepared and pro-on and immunoprecipitated withinonoc nal anti-oE antibody lm munoprecipitatcs from 300 pic of the sporulating cell culture were separated by SIDS-P ACE and detected by fluorognphy. (B) Pro-oE and a“: were quantifiedw a-Phosphor lmager and plotted as percentages relative to the levels immediately after pulse labeling. Wl‘, wild type. carried on an SPB phage (24), via specialized transduction, into the chromosomes of wild- -type cells, sigK (spoIVCBZ3) mutant cells, and mutants (carrying spoIVCBAI9, bost, or Punt-P r sigKA19) that produce active oK earlier than normal ffiie mutant, like the spoIVCBAI9 mutant, produces active 0 about I h earlier than normal becau use processing of pro-oK is uncoupled from its normal dependence on a signal from the forespore compartment (6). The P,” PK-sigKAl9 mutant contains in its chromosome the isopropyl- -D-thiogalactopyr- anoside (IPTG)- -inducible promoter, spac, fused to a copy of the sigK gene (sigKA19) that permits production of active 0" without the need for the site-specific recombination event that normally joins the two parts (i. e., spoIVCB and spolllC) of the sigK gene and without the need for processing (40). Thus sigKA19 cells produce UK upon the addition of IfiG 6‘40) sigE- -lacZ expression was highest' In sigK mutant cells, lower' In wild-type cells, even lower In spoIVCBA A19 and bof38 mutant cells, and lowest in PM - ,grsigKAI9 cells, as judged by the intensity of blue of colonies sporulating on DSM agar containing X-Gal and IPTG (data not shown). In agree- ment with these qualitative results were the results of quanti- tative B-galactosidase assays of cells sporulating in SM liquid. Figure 6 shows that in wild-type cells, sigE-directed B-galacto— sidase activity increased at T,, reached its peak level at T2 or T3, and decreased thereafter. In sigK mutant cells, B-galacto— sidase activity rose to a slightly higher level and remained higher late in sporulation B-Galactosidase activity was re- duced In spoIVCBAI9 and b01138 mutant cells that produce 0" about 1 h earlier than normal. The elfects on sigE— lacZ expres- sion in the sigK mutant and the spoIVCBAI9 mutant were similar to the effects on the 0'": level (Fig. 2 and 4). When J. Il.\<'II.RI0i.. I20 >. ‘m .2 2 I; so < 0 so z E 40 e t: 20 G I I I I I I w a 0 t 2 3 4 5 S 7 Hours Atter Reauepension FIG. 6. The elfccts of altered a“ production( on slgE- 4ch expression are similar to the elect son teh 0'3 level. Wild -(typc Y79 [0]) cells, II‘ng mutant (BK556 [0]) cells. and spoIVCBAl9 (V048 Ill), (Io/BIN SC77hD [D [).a lid-Pym P‘rmKAIO (37.536 [A]) cells that produce it earlier than norma were lyso- geniaed with phage SP3: LtigEr- [41:2 and the resulting strains were induced to I ' ' ' ' " ‘ ' ‘ " For Jpcrimcnts reperfonned at least twice for each strain. For each experiment, the specific activities were normalized to the maximum 5 cific activity in wild type cells (typically U.) Points on the graph are averages of the normalized values and error bars show one standard deviation of thed production of a" was induced 30 min before cells were resus- cnded to initiate sporulation, sigE- -lacZ expression was even lower (Fig.6). Taken together, these results suggest that o" negatively regulates the (7'5 level by affecting the transcription of: sigE. DISCUSSION We have demonstrated that a Knegatively regulates the level of GE during sporulation. ln wild-type cells, the level of 0'3 begins to decrease when active 0" begins to accumulate (Fig. 2) In mutants defective In a" production, 0'3 persists at an elevated level for several hours '(Fig 2 and data not shown). In cells engineered to produce a earlier than normal, twofold less (IE than that in wild-type cells accumulates (Fig. 4). A similar pattern of effects in sigK mutant cells and cells prema- turely expressing a" 15 observed for expression of a sigE- -lacZ fusion (Fig. 6), suggesting that 0.x exerts its negative efiect at the level of sigE transcription. The finding that 0'" negatively regulates 0E provides a sim- ple explanation for the previous observation that o" negatively regulates SpoIIID (9). As depicted in Fig. 7, 0E RNA poly- asetranscribes the spoIIID gene (28, 49, 52). A decrease In the [GE level brought about by a negative effect of oK on sigE transcription (Fig. 7) would reduce the synthesis of spolllD mRNA, assuming that 0'5 RNA polymerase becomes limiting for spolllD transcription. It seems likely that spolllD transcrip- tion is limited by the availability of GE, because earlier produc- tio on of 0" reduces the one level (Fig. 4), and the level of SpoIIIDK is likewise reduced (9). Conversely, the failure to make tr" results in an elevated 115 level beginning at T of sporulation (Fig. 2), and the level of KpolllD Is also elevated (Fig.3). We found no evidence that 0' affects spolIlD expres- sion at the level Kof mRNA stability (Fig. I). Also, there Is no evidence that 0" regulates the SpoIIID level via a positran- scriptional mechanism. The difference in the SpoIIID protein level between wild-type and sigK mutant cells (Fig. 3) is similar VOL. I79, I997 0' K ..-4/\\. 3195 apolitD ang gorE cotD / 5 s eotc SpoOA-P T l; / \i /corx SinR SpoIIID GerE FIG. 7. Model for gene regulation in the mother cell cascade. Dashed lines with arrowheads, gene-to-product relationships; arrows and lines with barred ends, positive and negative effects, respectively, on expression. The hallmark of initiation of sporulation is an increase in the level of SpoOA-P. SpoOA-P activates transcription of sigE by (TA RNA polymerase. SinR directly or indirectly inhibits sigE expression. oE RNA polymerase transcribes spoIIID. SpoIIID regulates genes in both the oE and o regulons. Among them, it activates 513K transcrip- tion and represses transcription of certain cor genes. Transcription of sigK is directed first by GE and then by its own gene product. 0". The corD, -C. and -X genes are transcribed by or" RNA polymerase. o" negatively regulates spoIllD and the entire 05 regulon by negatively regulating sigE transcription. A dimin~ ished SpoIIID level allows the previously repressed cor genes to be transcribed. GerE is a gene product of the tr" regulon. It represses transcription of sigK. forming another feedback loop in the mother cell cascade of gene expression. GerE also activates transcription of the cab, -C, and -X genes, reinforcing the switch of the mother cell gene expression pattern initiated by the decrease in the level of SpoIIID. to the difference in the spoIIID mRNA level (9) (data not shown). In addition, the pattern of overaccumulation of B- galactosidase activity from a spoIIID-lacZ transcriptional fu- sion in sporulating sigK mutant cells (28) (data not shown) was similar to the pattern of overaccumulation of SpoIIID (Fig. 3). Therefore, we propose that 0" directly or indirectly inhibits sigE transcription, reducing synthesis of 05, which in turn re- duces transcription of spolllD, and, as the level of SpoIIID declines, its repressive effect on excependcnt genes such as cotD, corX, and corC is relieved (9, 10, 19, 56, 58) (Fig. 7). Transcription of sigE is carried out by 0" RNA polymerase and requires SpoOA phosphate (25, 47) (Fig. 7). 0" is the major sigma factor present in growing cells, in which it directs transcription of most genes (12). SpoOA is also present in growing cells, and an increase in the level of phosphorylated SpoOA (SpoOA-P) initiates sporulation gene yression, in- cluding directly activating sigE transcription by RNA poly- merase, in response to nutritional, extracellular, and cell cycle signals (1, 20). Transcription of sigE is also subject to negative control by SinR (37, 38) (Fig. 7). We are currently trying to determine whether oK exerts its negative effect on sigE tran- scription by afiecting 0", S A-P, or SinR. The negative effect of a on sigE transcription may explain why a sigE-lacZ fusion is overexpressed in 3:55 mutant cells (24). Since the sigE mutant fails to make , the negative feedback on sigE transcription would not occur, resulting in sigE overexpression. Similarly, the elevated level of o5 found in sigK mutant cells might cause overexpression of other a":- dependent genes in addition to spoIIID. The promoter of spoIID is a well-known example of a (IE-dependent promoter that is independent of SpoIIID for transcription (4, 44). We found that spolID-lacZ is overexpressed in sigK mutant cells (data not shown). The negative effect of tr" on (1E and SpoIIID is not the only example of a feedback loop in the cascade of transcription factors controlling mother cell gene expression. As illustrated in Fig. 7, 0" RNA polymerase transcribes the gerE gene and GerE limits the 0" level by repressing sigK transcription (19, 58). It was attractive to think that in addition to repressing sigK 34 FEEDBACK REGULATION IN A CASCADE 6143 transcription, GerE might repress the transcription of sigE and/or spoIIID. However, expression of sigE-Incl and spoIIID- lacZ transcriptional fusions is indistinguishable in wild-type and gerE mutant cells (data not shown). The finding that a" negatively regulates sigE transcription provides an alternative to the model that each subsequent o in a cascade competes more effectively for a limiting amount of core RNA polymerase (33). In vitro studies with phage 0 factors involved in cascade regulation support the direct a competition model in some cases (2, 18), but not in others (53). Recently, Hicks and Grossman (15) presented in vivo experi- ments that suggest that o" competes with a" for binding to core RNA polymerase. If 0" could outcompete 0‘3 for core binding, it seemed likely that the appearance of 0" in cells would destabilize 05, since it had been suggested that free (IE is unstable (21). However, 0" did not affect the stability of as (Fig. 5). In the a cascade controlling B. subtilis sporulation gene expression, each a is either made as an inactive precursor or is initially held inactive by an anti-o factor (12, 34). This ensures that later-acting 0 factors accumulate sufficiently before neg- atively regulating earlier-acting 0 factors that control their synthesis. Regulation of a factor activity also appears to couple the program of gene expression in the mother cell and fore- spore during B. subtilis sporulation (26, 34). For example, pro- teolytic processing of inactive pro—oK to active 0 in the mother cell is governed by a signal transduction pathway that emanates from the forespore and may depend on a morpho- logical feature of the developing sporangium (5—7, 35). In this case, the primary event responding to morphological and/or cell—cell signals is pro—0'" processing. Loss of (TE and SpoIIID is a secondary event brought about by the negative effect of a" on sigE transcription. In contrast, loss of a transcription factor from cells due to secretion is the primary event regulated by morphological cues or cell-cell interactions in a few examples that have emerged recently (17, 42, 57). These examples high- light the importance of considering the disappearance of ex- isting transcription factors, as well as the appearance of new ones, during adaptive processes. ACKNOWLEDGMENTS We thank V. Oke and R. Losick for providing bacterial strains, C. Buckner and C. Moran for providing SPBrsI'gE-lacl, and W. Halden- wang for providing monoclonal anti-o": antibody. We thank A. Hofrneister and W. Haldenwang for helpful comments on the manu- script. This research was supported by the Michigan Agricultural Experi- ment Station and grant GM43585 from the National Institutes of Health. REFERENCES 1. MD”KLMMJ.A. Koch. 1991. Initiationofcporulation in B. subtilis is controlled by a multicomponent phosphorelay. Cell 64:545-552. 2. Chelm, 3. IL, J. J. Duly, and E. P. Gelduadiet. 1982. Interaction of Bacillus nibtilis RNA polymerase core with two specificity-determining subunits. J. Biol. Chem. 257:6501-6508. 3. Che-mush, P. and N. Sacehl. 1987. Single-step method of RNA isolation by guanidinium thiocyanate-phenol-chloroform extraction. Anal. Biochem. [62:156—159. 4. Clarke, 8., I. lapel-Din. and J. Mandelstam. I986. Use of («Z fusions to determine the dependence pattern of the sporulation gene spollD in spa mutants of Bacillus subtilis. J. Gen. Microbiol. 132:2987-2994. 5. Cutting. S., A. Drika, R. Schmidt, B. Kunkel, and R. Loalek. 1991. Forc- apore-spccific transcription of a gene in the signal transduction pathway that governs pro-<1" procesing in Bacillus subtilis. Genes Dev. 5:456—466. 6. Cutting. 8., V. Ole, A. Driks, R. Loslclt, S. La, and L Kroos. 1990. A forespore checkpoint for mother-cell gene expression during development in Bacillus subtilis. Cell 62:239-250. 7. Cutting. 8., S. Rods, and R. butch. I991. Sporulation operon spoIVF and the characterization of mutations that uncouplc mother-cell from forespore 6144 II. 12. 13. 17. 18. 21. 22. 23. 24. 26. 27. 31. 32. ZHANG AND KROOS gene expression in Bacillus subtilis. J. Mol. Biol. 221:1237-1256. . Errington, J. 1993. Bacillus subtilis sporulation: regulation of gene expression and control of morphogenesis. Microbiol. Rev. 57:1—33. . Halberg, R., and L Kroos. 1992. Fate of the SpoIIID switch protein during Bacillus subtilis sporulation depends on the mothencell sigma factor, a". J. Mol. Biol. 28:840-849. . Halberg, R., and L Kroos. 1994. Sporulation regulatory protein SpoIIID from Bacillus subtilis activates and represses transcription by both mother- cell-specific forms of RNA polymerase. J. Mol. Biol. 243:425—436. Halberg, R., V. Oke, and L Kroos. 1995. Eliects of Bacrllus subtilis sporula- tion regulatory protein SpoIIID on transcription by a" RNA polymerase in vivo and in vitro. J. Bacteriol. 177:1888—1891. Haidenwang, W. G. 1995. The sigma factors of Bacillus subtilis. Microbiol. Rev. 59:1—30. Harry, E. J., K. legible. and R. Loaiek. 1995. Use of immunofiuorescence to visualize cell-specific gene expression during sporulation in Bacillus sub- tilis. J. Bacteriol. [773386—3391 . Hal-wood, C. R., and S. M. Cutting. 1990. Molecular biological methods for Bacillus. John Wiley & Sons. Chichester, England. . Hicks, K. A.. and A. D. Grossman. 1996. Altering the level and regulation of the major sigma subunit of RNA polymerase aflects gene expression and development in Bacillus subtilis. Mol. Microbiol. 20:201-212. . Hoheister, A, A. Londono-Vaileio. E. Harry. P. Stragier, and R. Loeicir. 1995. Extracellular signal protein triggering the proteolytic activation of a developmental transcription factor in B. subtilis. Cell 83:219—226. Hughes, K. T. K. L Giiien, M. J. Season. and J. E. Karlinsey. 1993. Sensing structural intermediates in bacterial flageliar assembly by export of a nega- tive regulator. Science 262:1277—1280. Hyde, E. L, M. D. Hilton, and H. R. Whiteley. 1986. Interactions of Bacillus subtilis RNA rase with subunits determining the specificity of initia- tion. J. Biol. Chem. 261:16565—16570. . iehlhawa. ii. and L Kroos. Unpublished data. . lretoa. K., D. Z. Rodner. K. J. Siranoataa. and A. D. Grossmaa. 1993. Integration of multiple developmental signals in Bacillus subtilis through the spoOA transcription factor. Genes Dev. 7:283—294. Jonas. R. M., ii. K. Peters, and W. G. Baideowang. 1990. Phenotypes of Bacillus subtilis mutants altered in the precursor-specific region of as. J. Bacteriol. 172:4178-4186. Jonas, R. M., E. A. Weaver, 1‘. J. Ken-q, C. P. Moran, and W. G. [hideo- waag. 1988. The Bacillus subtilis spoIIG operon encodes both a5 and a gene necessary for an activation. J. Bacteriol. 110:507—51 1. Jones, C. IL and C. P. Moran. 1992. Mutant a factor blocks transition between promoter binding and initiation of transcription. Proc. Natl. Acad. Sci. USA 89:1958-1962. Kearney. T. J., and C. P. Moran. 1987. Organization and regulation of an operon that encodes a sporulation-essential sigma factor in Bacillus subtilis. J. Bacteriol. [69:3329—3339. . Keaney. T. J.. K. York, P. You-glue, -d C. P. Moran. 1989. Genetic evidence that RNA polymerase associated with it" uses a sporulation-spe- cific promoter in Bacillus subtilis. Proc. Natl. Acad. Sci. USA “:91W9113. Kroos, L. and S. Cutting. 1994. lntercelluiar and intereompartmental com- munication during Bacillus subtilis sporulation. p. 155—180. In P. J. Piggot. C. P. Moran. and P. Youngman (ed). Regulation of bacterial differentiation. American Society for Microbiology, Washington. DC. Kroos, L, I. Kniel, and R. Losick. 1989. Switch protein alters specificity of RNA polymerase containing a compartmenbspecific sigma factor. Science 143:526—529. Kaakei, I. L Kroos. [1. Path. P. Yong-n, and R. bookie 1989. Temporal and spatial control of the mother-cell regulatory gene spoIIlD of Bacillus was. Genes Dev. 3:1735—1744. leMandP.Stngla.1990.1‘heBacillius-ubrilirgeneforthe developmental transcription factor a" '3 generated by excision of a dispens- able DNA element containing a sporulation recombinase gene. Genes Dev. 4:525—535. . knell. 1'. L. J. I. This", and W. G. Baldenwang. 1987. Sporulation- specific sigma factor. a”. of Bacillus subtilis is synthesized from a precursor protein, P '. Proc. Natl. Acad. Sci. USA 84:1784—1788. Denali, U. K. 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227:680—685. hewis,P.J-.C.E.Nwoguh,M.R.Rarer,C.J.Harwood,andJ.Err|agtoa. 1994. Use of digitized video microscopy with a fluorogenic enzyme substrate to demonstrate celi- and compartment-specific gene expression in Salma nella auermdi's and Bacillus subtilis. Mol. Microbiol. 13:3849—3853. 33. 34. 35. 3 . \i 39. 4 . fl 42. 43. 45. 47. 49. 50. 51. 52. 53. 55. 57. 58. J. BAtTliRIOL. Loslclt, R.. and J. Peru. 1981. Cascades of sigma factors. Cell 25:582-584. Losick, R., and P. Stragier. 1992. Crisscross regulation of cell-lypc-specific gene expression during development in B. subtilis. Nature 355:601-604. Lu, 8., R. Halberg, and L Kroos. 1990. Processing of the mother-cell a factor, 0", may depend on events occurring in the forespore during Bacillus subtilis development. Proc. Natl. Acad. Sci. USA 87:9722—9726. . Lu, 8., and L Kroos. 1994. Overproducing the Bacillus sublilis mother-cell sigma factor precursor. pro-0". unmuples a"-dcpendent gene expression from dependence on intercompartmental communication. J. Bacteriol. 176: 3936—3943. Maudie-Melee, I.. L Donkhan, and 1. Smith. 1995. The Bacillus subtilis SinR protein is a repressor of the key sporulation gene spoOA. J. Bacteriol. 177: 4619—4627. Mandie-Muiee, I., N. Gaur. U. Bat. and I. Smith. 1992. Sin. 3 stage-specific repressor of cellular differentiation. J. Bacteriol. 174:3561—3569. Margolia. P., A. Drllts. and R. hutch. 1991. Establishment of cell type by compartmentalized activation of a transcription factor. Science 254:562-565. . Olse, V.. and R. Losiclt. 1993. Multilevel regulation of the sporulation tran- scription factor a" in Bacillus subtilis. J. Bacteriol. 175:7341—7347. Peters, ii. IL, and W. G. Haldenwang. 1994. Isolation of a Bacillus subtilis spollGA allele that suppresses processing-negative mutations in the pro-aE gene (sigE). J. Bacteriol. l76:7763~7766. Pettersson, J., R. Nordfelth, E. Dublnina. 'l'. Bergman. M. Gustafssou, K. E. Magansson, and H. Walt-Wan. 1996. Modulation of virulence factor expres- sion by pathogen target cell contact. Science 273:1231—1233. Pophan, D. L, and P. Stragier. 1992. Binding of the Bacillus subtilis spoIVCA product to the recombination sites of the element interrupting the o"-encoding gene. Proc. Natl. Acad. Sci. USA 89:5991-5995. . Rang. 8., M. S. Rosenkrantz. and A. L Sonenshein. 1986. Transcriptional control of the Bacillus subtilis spollD gene. J. Bacteriol. [65:771-779. Sato, 1‘. K. Harada, Y. Ohta, and Y. Kobayashi. 1994. Expression of the Bacillus subtilis spoIVCA gene, which encodes a site~specific recombinase. depends on the spollGB product. J. Bacteriol. 176:935-937. . Sato, ‘11, Y. Samori, and Y. Kobayashl. I990. The CM cistron of Bacillus subtilis sporulation gene spoIVC encodes a protein homologous to a site- specific recombinase. J. Bacteriol. [72:1092—1098. Satoia, S. W., J. M. Baldus, and C. P. Moran. 1992. Binding of SpoOA stimulates spoIIG promoter activity in Bacillus subtilis. J. Bacteriol. 174: 1448-1453. . Steinmetz, M., and R. Richter. 1994. Plasmids designed to alter the antibiotic resistance expression by insertion mutations in Bacillus subtilis. through in viva recombination. Gene 142:79—83. Stevens. C. M., and J. En'lngton. 1990. Diflerential gene expression during sporulation in Bacillus subtilis: structure and regulation of the spoIlID gene. Mol. Microbiol. 4:543—552. Stragler, P., C. Bonamy, and C. Karmazyn-Canipelli. 1988. Processing of a sporulation sigma factor in Bacillus subtilis: how morphological structure could control gene expression. Cell 52:697—704. Stragier. P., B. Kunitel, L Kroos, and R. Losiclt. 1989. Chromosomal rear- rangement generating a composite gene for a developmental transcription factor. Science 243:507-512. Tatti, K. M., C. H. Jones, and C. P. Moran. 1991. Genetic evidence for interaction of 05 with the spolllD promoter in Bacillus subtilis. J. Bacteriol. [73:7828—7833. Williams, K. P., G. A. Kassavetis. and E. P. Geiduacheh. 1987. Interaction of the bacteriophage T4 gene 55 product with Escherichia coli RNA polymer- ase. J. Biol. Chem. 262:12365-12371. . Yeoman. P., J. B. Perkins. and R. [article 1984. Construction of a cloning site near one end of Tn9l7 into which foreign DNA may be inserted without alecting transposition in Bacillus nibrili‘s or expression of the transposon- borne cm gene. Plasmid 12:1—9. Zhang, B. R. Daniel. J. Errington, and L Kroos. 1997. Bacillus subtilis SpoIIID protein binds to two sites in the spoVD promoter and represses transcription by GE RNA polymerase. J. Bacteriol. [79:972-975. Zhang.J-.I-LIchikawa,R.iialberg,LKrooe.andA.l.Amnaon. 1994. Regulation of the transcription of a cluster of Millus subtilis spore coat genes. J. Mol. Biol. 240:405-415. Zhang, J. P., and S. Normarit. 1996. Induction of gene expression in E coli after pilus-mediated adherence. Science 273:1234—1236. Zheng, L, R. Halberg, S. Roels, H. Ichikawa, L. Kroos. and R. loaiek. 1992. Spomlation regulatory protein GerE from Bacillus subtilis binds to and can activate or repress transcription from promoters for mother-eelI-specific genes. J. Mol. Biol. 226:1037-1050. Chapter III Characterization of the Mechanism by which OK Negatively Regulates sigE Transcription during Sporulation of Bacillus subtilis 36 37 ABSTRACT Temporal and spatial gene regulation during B. subtilis sporulation involves the activation and inactivation of multiple sigma subunits of RNA polymerase in a cascade. It was shown previously that transcription of the sigE gene encoding the earlier-acting sigma E factor 0’ is negatively regulated by the later-acting sigma factor OK in the mother cell compartment of sporulating cells. Here, it is shown that UK does not do so by increasing the level or activity of SinR, which is an inhibitor of sigE transcription. A null mutation in SinR did not change the expression of sigE-IacZ during sporulation in otherwise wild-type cells, in sigK mutant cells, or in cells engineered to produce 0K earlier than normal. A mutation in the spoOA gene that bypasses the phosphorelay leading to the phosphorylation of SpoOA partially relieved the negative effect of OK on sigE transcription. This suggests that OK affects the phosphorylation of SpoOA, an activator of sigE transcription. 0K also affected expression of the SpoOA-independent alanine dehydrogenase promoter late in sporulation. This promoter, like the sigE promoter, is thought to be recognized by 0'" RNA polymerase, suggesting that GK inhibits 6A activity. In contrast, O'K did not inhibit (SH-dependent gene expression as strongly. We propose that the product(s) of a GK- dependent gene(s) lowers the level of phosphorylated SpoOA and that UK competes with GA for binding to core RNAP. By negatively regulating both positive factors required for sigE transcription, the appearance of OK would facilitate the switch from early (SE-directed gene expression to late (SK-directed gene expression in the mother cell compartment of sporulating B. subtilis. 38 INTRODUCTION In response to nutrient depletion, Bacillus subtilis undergoes a developmental process that culminates with the formation of a dormant spore (137). Two compartments, the mother cell and the forespore, are formed early during the sporulation process due to the formation of an asymmetric septum. The forespore is later engulfed within the mother cell, being completely surrounded by the two membranes of the septum. The mother cell contributes to the synthesis of many components necessary for forespore maturation, including a thick layer of peptidoglycan called cortex and a tough proteinaceous spore coat, and is discarded by lysis at the end of sporulation, releasing the mature spore. Sporulation involves highly ordered programs of gene expression in the two compartments that are regulated primarily by the ordered appearance of two series of alternate sigma factors (68, 137). Upon starvation, multiple signals impinge on a phosphorelay system composed of kinases, phosphotransferases, and phosphatases (16, 34, 104). The result is an elevated level of phosphorylated SpoOA (SpoOA~P), a transcription factor that activates (3" RNA polymerase (RNAP) and 0’” RNAP to transcribe the genes encoding OE and (IF, respectively (10, 12). After formation of the asymmetric septum, 6F becomes active in the forespore and directs transcription of the gene encoding 0’0 (42, 72, 93). Similarly, 6E becomes active in the mother cell and directs transcription 'of the gene encoding 6K (23, 42, 103, 141). Communication between the mother cell and the forespore compartments regulates sigma factor activity. All the compartment-specific sigma factors are initially inactive. In the forespore, 6F and perhaps 0'0 are held inactive by an anti-sigma factor, SpoIIAB (26, 66, 95). In the mother cell, GE and UK are first synthesized as inactive precursor proteins, pro-(rE and pro-oK (18, 74, 89). Compartmentalized activation of these sigma factors, 39 except for CF, depends on intercompartrnental signal transduction (137). In this way, the programs of gene expression in the two compartments are coupled. In addition to controlling the synthesis and activation of subsequent sigma factors in the cascade, each 6 directs core RNAP to transcribe different genes whose products drive morphogenesis (137). Although the synthesis and activation of sigma factors during B. subtilis sporulation has been relatively well-studied, little is known about how later sigma factors replace the earlier ones. We showed previously that in the mother cell compartment, the appearance of OK accelerates the disappearance of GE and facilitates the switch of gene expression from the (SE-dependent pattern to the (SK-dependent pattern (157). In mutants that fail to produce GK, a sigE-lacZ transcriptional fusion is overexpressed late in sporulation. In mutants that produce OK earlier than normal, sigE-IacZ expression is reduced. These results suggest that 0K negatively regulates sigE transcription. Transcription of sigE is carried out by GA RNAP, and is activated by SpoOA~P (12, 65, 120) and repressed by SinR (91 , 92). We show here that UK does not affect the activity of SinR. OK, or, more likely, the product of a gene(s) under its control, does appear to inhibit the phosphorylation of SpoOA. 0K also appears to inhibit 0" activity, perhaps by competing directly for binding to core RN AP. 40 MATERIALS AND METHODS Bacterial strains. The B. subtilis strains used in this study are listed in Table 1. To introduce gene fusions and mutations into the wild-type strain PY79 and its derivatives BK556, V048 and BZS36, chromosomal DNA was prepared from a strain containing the desired fusion or mutation, and used to transform competent cells of the recipient strain (43). Transforrnants were selected on LB plates containing appropriate antibiotics. Chloramphenicol was used at 5 ug/ml and spectinomycin was used at 100 ug/ml. Resistance to macrolide-lincosamide-streptogramin B (MLS) antibiotics, encoded by Tn91 7, was selected by using a combination of erythromycin (1 ug/ml) and lincomycin (25 ug/ml). Colonies of cells containing the sinR null mutation displayed a characteristic “rough” phenotype (30). The rvtAI I mutation in AG919 is 80-90% linked by co-transformation to a downstream Chloramphenicol resistance gene marker (35). To verify the presence of the rvtAII mutation in a Chloramphenicol-resistant transformant, chromosomal DNA was used to transform competent AGl431 cells. DNA from isolates containing the NM] 1 mutation rescued the Sp0' and Pig' AGl43l cells to Spo+ and Pig+ at a frequency of 80-90%. Specialized transduction was used to move lacZ firsions carried on SPB phages into various strains (43). Cell growth and sporulation. Sporulation was induced by resuspending growing cells in SM medium as described previously (43). The onset of sporulation (T O) is defined as the time of resuspension. The sporulation efficiency was measured as described (43). For experiments involving the induction of PspaC-PsigK-sigKAm in strain BZ536 during vegetative growth, cells were first grown overnight (12-14 h) in LB medium at 37 °C. The overnight culture was used to inoculate fresh LB medium to an optical density (CD) at 600 nm of 0.05. The culture was divided into aliquots of equal volume 41 Table l. B. subtilis strains used. Strain Relevant Genotype Sources PY79 wild type (154) BK556a spoIVCBZ3 (70) V048a spoIVCBA19 cat ( 1 8) BZS36a Pspac-PsigK-sigKAI 9 spc (157) AG919b rvtAll cat (35) K1220b ald.'.'Tn9l 7106 MLS (128) K1126]b amyEzzspOOA-lacZ cat (52) AGI431b spoOFA, spoOBA (35) 184320 SinR cat (30) ZB307b SPB::spo VG42-lacZ (160) DZR67b amyEzzspOOK-lacZ cat A. Grossman KH566b spoOH-lacZ cat A. Grossman 1(11202b SPflzzspoIIA-lacZ A. Grossman, (150) KY9b SPflzzspoIIE-lacZ (36) a, derived from PY79. b, derived from JH642. c, derived from 1875. Abbreviations: cat, Chloramphenicol resistance; spc, spectinomycin resistance; MLS, resistance to macro]ide-lincosamide-streptograrnin B antibiotics. 42 when it reached an OD600 of 0.3-0.5. Different concentrations of isopropyl- B-D-thiogalactopyranoside (IPTG) were then added to the aliquots. Incubation was continued and samples were taken every half hour. B-galactosidase assays. B-galactosidase activity was assessed qualitatively by placing cells on DSM agar (43) containing 5-bromo-4-chloro-3-indolyl- B-D-galactopyranoside (X-gal) (20 rig/m1) and was determined quantitatively using toluene to perrneabilize cells and o-nitrophenolfi-D-galactopyranoside as the substrate (43). One unit of enzyme hydrolyses l umol of substrate per minute per A 600 of initial cell density. 43 RESULTS AND DISCUSSION OK does not inhibit sigE transcription by increasing the level or activity of SinR. The gene that encodes pro-OE, sigE (also called spoIIGB), is the second gene in the spoIIG operon (59, 64). The first gene of the operon, spoIIGA, encodes a putative protease that processes pro-(5'E to (SE (48, 110, 135). Using a transcriptional fusion between the spoIIG promoter and lacZ (64), which we referred to as sigE-lacZ since it provided an indirect measure of sigE transcription, we showed previously that UK appears to negatively regulate transcription of sigE (157). In spoIVCBZ3 (spoIVCB encodes the N-terminal part of 0") mutant cells that fail to produce 0'", sigE-lacZ was overexpressed late in sporulation. In spoIVCBAI9 cells that make active OK 1 h earlier than normal due to a deletion in the pro-sequence of pro- 0'", sigE-lacZ expression was reduced. To further explore the mechanism by which GK inhibits sigE transcription, we introduced additional mutations into the mutants with altered O’K production and measured sigE-IacZ expression. The first mutation we tested in this way was a null mutation in SinR (30). SinR is a transcription factor that inhibits the transcription of some early sporulation genes, including sigE (91 , 92). A simple mechanism by which 0'" might decrease the transcription of sigE is by increasing the level or activity of SinR. To test this hypothesis, we introduced a SinR null mutation into wild type cells, and spoIVCBZ3 and spoIVCBA19 mutant cells. We then introduced the sigE-lacZ transcriptional fusion carried on an SPB phage into the chromosome of each strain via specialized transduction. Expression of sigE-lacZ was monitored in these strains by measuring the B—galactosidase activity of samples collected at hourly intervals during sporulation. 44 1207 100‘ Relative Activity O d -| - - —l —r d d Hours After Resuspension Figure l. The effect of a sink null mutation on sigE-lacZ expression. The sinR mutation was introduced into wild-type (PY79; O) cells, sigK (spoIVCBZ3) mutant cells (BK556, O), and spoIVCBA19 (V048, A) cells that produce O'K earlier than normal. The resulting strains were lysogenized with phage SP f3::sigE-lacZ and induced to sporulate in SM media. Samples were collected at the indicated times following the initiation of sporulation and assayed for B-galactosidase activity. Experiments were performed at least twice for each strain. For each experiment, the specific activities were normalized to the maximum specific activity in cells containing the rvtAI 1 mutation in the wild-type PY7 9 background (typically 130 units). Points on the graph are averages of the normalized values and error bars show one standard deviation of the data. 45 As shown in Figure 1, the pattern of sigE-lacZ expression was preserved in these strains relative to the parental strains without the SinR mutation (157). In cells containing only the sink mutation, sigE-lacZ expression increased and decreased with similar timing as in wild-type cells, but reached a 2-fold higher maximum level [130 units versus 70 units (157)], consistent with the finding reported previously that SinR inhibits sigE expression (91, 92). In SinR spoIVCBZ3 mutant cells that fail to make 6K, sigE-lacZ expression was higher late in sporulation (i.e. at T 3 - T7) than in cells containing only the sinR mutation. In sinR spoIVCBAI9 cells that make (5" earlier than normal, sigE-lacZ expression was reduced at T2 - T5 compared to that in SinR mutant cells. Thus, SinR is not required for the negative effect of OK on sigE transcription. We conclude that CK does not affect the transcription of sigE by increasing the level or activity of SinR. Bypassing the phosphorelay leading to the activation of SpoOA partially relieves the negative effect of OK on sigE transcription. The second mutation we tested for an effect on the O‘K-dependent inhibition of sigE transcription was a missense mutation in spoOA called WM]! (124). At the onset of sporulation, multiple signals activate a milticomponent phosphorelay system to phosphorylate SpoOA (34, 46, 104). Only afier it is phosphorylated can SpoOA activate the transcription of sigE and other early sporulation genes (10, 12). The rvtAIl mutation bypasses the need for the phosphorelay and renders SpoOA able to be phosphorylated by an alternate kinase (76). If OK inhibits sigE transcription by affecting a component of the phosphorelay so as to lower the level of SpoOA~P, then the rvtAI 1 mutation might bypass this effect and relieve the rvtAl 1 mutation was introduced into wild-type (PY79; 0) cells, sigK (spoIVCBZ3) mutant cells (BK556, O), and spoIVCBAI 9 (V048, A) cells that produce (3" earlier than normal. The resulting strains were lysogenized with phage SPflxsigE—lacZ and induced to sporulate 46 Ivity Relative act Hours After Resuspension Figure 2. The effect of bypassing the phosphorelay on sigE-lacZ expression. The in SM media. Samples were collected at the indicated times following the initiation of sporulation and assayed for B—galactosidase activity. Experiments were performed at least twice for each strain. For each experiment, the specific activities were normalized to the maximum specific activity in cells containing the NM] 1 mutation in the wild-type PY79 background (typically 70 units). Points on the graph are averages of the normalized values and error bars show one standard deviation of the data. 47 inhibition of sigE transcription by OK. We introduced the WM]! mutation into wild—type cells, and spoIVCBZ3 and spa] V CBA19 mutant cells. The sigE-lacZ transcriptional fusion, carried on phage SPB, was then integrated into the chromosomes of these strains. sigE-lacZ expression was first examined by allowing the strains to sporulate on DSM agar containing X-gal. The pattern of expression was preserved to some extent as judged by the intensity of blue color of colonies (data not shown). That is, expression appeared to be highest in rvtAI I spoIVCBZ3 cells, lower in cells containing only the rvtAI I mutation, and lowest in NM] 1 spoIVCBAI9 cells. However, the differences did not appear to be as great as when the corresponding strains without the th1 1 mutation were examined on DSM agar containing X-gal (data not shown). To test these qualitative observations more carefully, the thl 1 mutant strains were induced to sporulate in SM liquid and the B—galactosidase activity was quantitatively assayed. In agreement with the qualitative observations, Figure 2 shows that the pattern of sigE—IacZ expression was preserved in these strains, but the differences were not as great as for the parental strains without the WM] 1 mutation (157) or the strains with the sinR mutation (Figure 1). For example, sigE-lacZ expression in WM] 1 spoIVCBA19 cells reached 80% of the maximum level observed (at T2) in cells containing only the thI 1 mutation, whereas sigE-lacZ expression in spoIVCBA19 cells reached only 55% of the maximum level observed in wild-type cells (157 ), and expression in sinR spoIVCBA19 cells reached only 65% of the maximum observed in cells containing only the sinR mutation. Thus, the negative effect of OK on sigE expression was partially relieved by bypassing the phosphorelay, suggesting that UK negatively regulates the activity of SpoOA by interfering with the phosphorelay. How might OK affect the phosphorylation of SpoOA? The phosphorelay is a complicated variation of the two-component signal transduction system (34, 46). Two 48 independent histidine kinases, KinA and KinB, phosphorylate a response regulator, SpoOF. The phosphate group of SpoOF~P is then transferred to another response regulator, SpoOA, by the phosphotransferase SpoOB. SpoOF~P can also be dephosphorylated by a family of phosphatases (104). The phosphorelay provides many regulatory sites to integrate a large variety of intracellular and extracellular signals that regulate sporulation initiation. The rvtAIl mutation in spoUA bypasses the need for the phosphorelay (124). The observation that the negative effect of UK on sigE transcription is partially relieved in the presence of the rvtAI 1 mutation (Figure 2) suggests that OK targets a component(s) of the phosphorelay and/or a SpoOF~P phosphatase(s). It seems likely that this is an indirect effect of OK due to the transcription of one or more cK-dependent genes. Clearly though, the thI 1 mutation did not completely relieve the inhibition of sigE transcription by GK. Compared to cells containing only the rvtAI 1 mutation, cells that in addition contained a spoIVCBZ3 mutation and therefore failed to make O’K overexpressed sigE-lacZ at T5 - T7, whereas rvtAII spoIVCBA19 cells that make 0" earlier than normal exhibited slightly reduced sigE-lacZ expression at T2 - T 4 (Figure 2). This implies that OK can also affect SpoOA~P activity in a way that is not bypassed by the WM]! mutation and/or that 0‘K affects 0A RNAP activity, since O'A RNAP transcribes the sigE gene (12, 65). Expression of a SpoOA-independent gene remains high late during sporulation in sigK mutant cells. We next examined whether O'K negatively regulates expression of a SpoOA-independent gene. For this purpose, we chose a lacZ fusion created by insertion of Tn 91 7lac into the ald gene (128), which encodes alanine dehydrogenase. Like expression of sigE-lacZ, expression of aldzzTn91 7Iac increases at the onset of sporulation and transcription is thought to be directed by GA RNAP (128). 49 Induction of ald transcription has been postulated to involve an unidentified regulatory factor(s), but does not require SpoOA~P (128). The old locus of wild-type cells and spoIVCBZ3 and spoIVCBA19 mutant cells was replaced with aldzzTn91 7lac by transformation with chromosomal DNA from K1220 (128). The strain containing ald::Tn9I 7lac in an otherwise wild-type background sporulated poorly in D8 medium, consistent with the previous report (128). However, the sporulation efficiency of this strain was comparable to that of the Wild-type strain in SM resuspension medium (data not shown). Apparently, the ald locus is dispensable for sporulation in SM medium. Therefore, the strains containing ald::Tn91 7lac were sporulated in SM medium and expression of aid-lacZ was measured by determining the B-galactosidase activity of samples collected at hourly intervals. As shown in Figure 3, there was no difference in the level of aid—lacZ expression in the strains early during sporulation, but expression remained high late during sporulation of spoIVCBZ3 mutant cells that fail to make OK. Thus, UK inhibits expression of the SpoOA-independent aid-lacZ fusion in wild-type cells late during sporulation. O’K also inhibits sigE-lacZ expression in late sporulating cells (157), and this effect does not require SinR (Figure 1) and is not completely bypassed by the rvtAlI mutation in SpoOA (Figure 2). Since aid is thought to be transcribed by GA RNAP and sigE is known to be, we propose that UK inhibits the expression of both genes late during sporulation by inhibiting the activity of GA RNAP. Perhaps O‘K competes directly with OK for binding to core RNAP. The finding that expression of the SpoOA-independent ald—lacZ fusion was not reduced in spoIVCBAI9 cells compared to wild-type cells (Figure 3) is in striking contrast to the result with the SpoOA-dependent sigE-lacZ fusion (157) (Figure 1). Taken together, these results suggest that the inhibition of sigE-lacZ expression observed when 6K is produced earlier than normal (157) (Figure 1) is due to an effect on SpoOA~P activity. 50 Relative Activity G I l T I I l I O 1 2 3 4 5 6 7 Hours After Resuspension Figure 3. The effects of altered OK production on aid-lacZ expression. The ald locus of wild-type (PY79; 0) cells, sigK (spoI VCB23) mutant cells (BK556, O), and spoIVCBA19 (V048, A) cells that produce O'K earlier than normal was replaced with aldzzTn917lac. The resulting strains were induced to sporulate in SM media. Samples were collected at the indicated times following the initiation of sporulation and assayed for B-galactosidase activity. Experiments were performed at least twice for each strain. For each experiment, the specific activities were normalized to the maximum specific activity in wild-type cells (typically 300 units). Points on the graph are averages of the normalized values and error bars show one standard deviation of the data. 51 This hypothesis is also supported by the finding that the rvtAII mutation in spoOA diminished the effect of earlier 0K production (Figure 2). UK has little effect on expression of (SH-dependent genes. Does GK also H is an alternate sigma factor exert a negative effect on (SH-dependent gene expression? 0 active during growth and with increased activity during sporulation (44). At the onset of sporulation, 0'” transcribes the spoIIA operon , which encodes O'F, the early-acting, forespore-specific sigma factor (23, 42, 148, 150). Since O'F does not become active in the mother cell, it may not be important for UK to negatively regulate the transcription of the spoIIA operon. 0n the other hand, some degree of inhibition was expected because expression of spoIIA, like expression of sigE, is positively regulated by SpoOA-P (15, 108, 150). For comparison, we also measured expression of the spa VG42 promoter. This promoter, like that of the spoIIA operon, is (SH-dependent, but the spa VG42 mutation renders the promoter independent of the activity of the Aer repressor (160), and therefore independent of SpoOA~P activity (SpoOA~P is a repressor of aer transcription). Figure 4A shows that expression of a spoIIA-lacZ fusion was slightly higher in the spoIVCBZ3 mutant that fails to make 6K, and slightly lower in the spoIVCBAI9 mutant that makes OK earlier, compared to expression in wild-type cells. The differences in expression of spoIIA-lacZ in mutants with altered O’K production were smaller than the differences in expression of sigE-lacZ (157) (Figure 1). These small differences may be due to the negative effect of OK on the phosphorelay that activates SpoOA, and/or on activity of 6H RNAP. As shown in Figure 4B, the SpoOA-independent spo VG42-lacZ fusion exhibited very small difference in expression in mutants with altered OK production. We conclude that 0’" has less of a negative effect on expression of (SH-dependent genes during 52 Figure 4. The effects of altered GK production on spoIIA-lacZ and spoVG42-lacZ expression. Wild-type (PY79; 0) cells, sigK (spoIVCBZ3) mutant cells (BK556, O), and spoIVCBAI9 (V048, A) cells that produce O'K earlier than normal, were lysogenized with phage SPfl::spolIA-lacZ and SP fl: :spo VG42-lacZ, respectively. The resulting strains were induced to sporulate in SM media. Samples were collected at the indicated times following the initiation of sporulation and assayed for B-galactosidase activity. Experiments were performed at least twice for each strain. Specific activities for SPfl::spoIIA-lacZ (A) or SPfl::spo VG42-lacZ (B) were normalized to the maximum specific activity in wild-type cells (typically 200 units for the spoIIA-lacZ fusion and 350 units for the spa VG42-lacZ fusion. Points on the graph are averages of the normalized values and error bars show one standard deviation of the data. 53 W Activi Relative O _ _ _ _ A _ 1 c A n u a m m :23. >22. mmmcwuoamaz 1V Activi Relative ..NO- # - - u _ _ _ A ~ 9 a m m q :87». >22 mmmcmuoammo: 54 sporulation than it has on expression of the (SA-dependent sigE gene (157) (Figure l), or the aId gene (Figure 3), which is thought to be (SA-dependent. Effect of inducing 0K during vegetative growth on expression of ald and spa VG42. Expression of the spa VG42-lacZ fusion was relatively insensitive to the appearance of OK in late sporulating cells (Figure 4B), compared with expression of ald- lacZ (Figure 3). Since both these fusions are expressed during vegetative growth, we examined the effect on expression of making different amounts of 6" using P P spac- sigK- sigKAI9 mutant cells (99, 157). The P -P . -sigKAI9 mutant contains in its .9ng spac chromosome the IPTG-inducible promoter, spac, fused to a copy of the sigK gene (sigKA19) that permits production of active OK without the need for the site-specific recombination event that normally joins the two parts (i.e., spoIVCB and spoIIIC) of the sigK gene and without the need for processing (99). Thus, Pspac-PsigK-sigKA19 cells produce O’K upon the addition of IPTG (40, 99). We induced different levels of OK production in Pspac-PsigK-sigKAN cells during vegetative growth by adding to cultures different concentrations of IPTG, and monitored the expression of ald-lacZ (aldzzTn917lac) and spa VG42-lacZ (carried on an SPB phage) fusions. At a low concentration of IPTG (25 11M ), ald—IacZ expression was reduced afier 3 h of induction (Fig. 5A), compared to the parallel culture without IPTG addition. At higher concentrations of IPTG (50 or 100 pM ), the ald-lacZ expression was reduced after 1.5 h of induction (Figure 5A). As noted above, the ald gene is thought to be transcribed by GA RNAP. Therefore, we examined the expression of several genes that are known to be transcribed by GA RNAP during vegetative growth, including spoOK, spoOA, and 55 Figure 5. The effect of making 6K during vegetative growth on aid-lacZ and spa VG42- lacZ expression. The ald::Tn9I 7 and SPfl::spo VG42-lacZ fusions were introduced into Pspac-PsigK-sigKA19 (BZ536) cells by transformation and transduction, respectively. Cells were grown in LB medium to early exponential phase. The cultures were divided into aliquots of equal volume, and IPTG was added to some of the aliquots to induce 0'" production. Incubation was continued and samples were taken every half hour for ,8- galactosidase assay. (A) Expression of ald-lacZ. IPTG concentrations were: 0 uM (O), 25 ”M (A), 50 uM (O), and 100 11M (Cl). (B) Expression of spa VG42-lacZ. IPTG concentrations were: 0 uM (O), 50 uM (O), 100 uM (Cl), and 200 uM (A). Experiments were performed at least twice for each strain, and data from a representative experiment are shown. 56 Specific Activities (Miller Units) ace: 3 b) mm t n C Aw ch. mm e( p S O . _ q _ _ a _ O. O.o o.m ._.O ...m ”.6 ~.m 6.6 w.m o.c °.m ._.O .Tm N.O M.m 0.0 “.0 12.7». >12. 532.0... 12.2.. >12 Sacnzo: 57 spoOH. Expression of lacZ fusions to these genes was reduced within 1.5 h after adding 1 mM IPTG to Pspac'Psng'SigKAl 9 cells, compared with the corresponding strains without IPTG addition (data not shown). Thus, GK production in vegetative cells appears to exert a general negative effect on (IA-dependent gene expression. Expression of the oH-dependent spo VG42-lacZ fusion was also reduced when GK production was induced in Pspac'Psng'SigKAI 9 cells; however, a higher concentration of IPTG (100 uM IPTG) and a longer induction time (approximately 3 h) were required (Figure 4B). Thus, it appears that a higher level of OK is required to inhibit transcription of the spa VG42 promoter than the aid promoter. Perhaps the level of OK produced late in sporulation is sufficient to inhibit ald expression (Figure 3) but insufficient to inhibit spo VG42 expression (Figure 43). It is possible that the inhibition of spa VG42 expression observed in vegetative cells (Figure 4B) is due to the negative effect of OK on (SA-dependent gene expression, because the gene encoding 6H, spoOH, is (SA-dependent, and its expression was reduced afier GK production was induced (data not shown). We propose that GK inhibits (SA-dependent gene expression in growing and sporulating B. subtilis cells by competing with GA for binding to core RNAP. This hypothesis assumes that the amount of core RNAP available for binding to sigma factors is limiting. Core RNAP does appear to be limiting in growing cells because competition has been demonstrated between a” and (SA (45). In sporulating cells, the level of 6A remains high, but its ability to associate with core RNAP is diminished early during sporulation (81, 143), (data not shown). Perhaps the mother-cell-specific sigma factors, 6E and UK, have a higher affinity than GA for binding to core RNAP. This can be tested by 58 competition experiments in vitro. Alternatively, the products of genes under OE and 6K control might prevent CA from associating with core RNAP. In summary, we have presented evidence that O'K directly or indirectly inhibits the activity of both positive regulators of sigE transcription, SpoOA~P and 6A RNAP. Hence, the appearance of OK both turns on late mother cell gene expression and turns off early O’E- directed gene expression. ACKNOWLEDGMENTS We are very grateful to A. Grossman, I. Smith, P. Zuber, P. Youngman, for providing B. subtilis strains. This research was supported by the Michigan Agricultural Experiment Station, by grant GM43585 from the National Institutes of Health. Chapter IV The Pro-sequence of Pro-0‘K Promotes Membrane Association and Inhibits RNA Polymerase Core Binding 59 60 ABSTRACT K is the inactive precursor of UK, a mother-cell-specific sigma factor Pro-O responsible for the transcription of late sporulation genes of Bacillus subtilis. Upon subcellular fractionation, the majority of the pro-(SK was present in the membrane fraction. K The rest of the pro-o was in a large complex that did not contain RNA polymerase core subunits. In contrast, the majority of the UK was associated with core RNA polymerase. Virtually identical fractionation properties were observed when pro- (SE was analyzed. Pro- UK was completely solubilized fiom the membrane fiaction and the large complex by Triton X-100, and was partially solubilized from the membrane fraction by NaCl and KSCN. The membrane-association of pro- 0‘" did not require spol VF gene products, which appear to be located in the mother cell membrane that surrounds the forespore, and govern pro-GK K processing in the mother cell. Furthermore, pro-o associated with the membrane when overproduced in vegetative cells. Overproduction of pro-6K in sporulating cells resulted in more pro-<5K in the membrane fraction. In agreement with the cell fractionation studies, K was localized to the mother cell immunofluorescence microscopy showed that pro-0' membranes that surround the mother cell and the forespore in sporulating wild-type cells and mutant cells that do not process pro-GK. Treatment of extracts with 0.6 M KCl appeared to free most of the pro-(SK and GK from other cell constituents. After salt K, reassociated with exogenous core RNA polymerase to form removal, OK, but not pro-0' holoenzyme. These results suggest that the pro-sequence inhibits RNA polymerase core binding and targets pro- OK to the membrane, where it may interact with the processing machinery. 61 INTRODUCTION Endospore formation in the Gram-positive bacterium Bacillus subtilis involves the formation of two cellular compartments of unequal size. The two compartments, namely, the mother cell and the forespore, are generated by the asymmetric positioning of a septum. The smaller forespore compartment is later engulfed inside the mother cell through a phagocytic-like process. The mother cell nurtures the forespore during sporulation and is discarded by lysis upon maturation of the endospore. Gene expression in the two compartments is driven by a cascade of 6 factors, namely, OF, CF“, CG , and 6K, in order of their activity (27, 41, 68, 86). The forespore-specific program of gene expression is controlled by CF and 00, while the mother-cell program is controlled by GE and 6K. Each sigma factor is initially inactive. CF is the first to become active and this occurs only in the forespore. Activation of subsequent sigma factors in the cascade is triggered by signal transduction between the two compartments. The inactive forms of the mother-cell-specific sigma factors are precursor proteins called pro-(3'E and pro- OK. Each is synthesized about one hour before it is activated by proteolysis (l 8, 74, 89). The processing of mother-cell-specific 0' factors is controlled by signals fi'om the forespore. The putative processing enzyme for the conversion of pro-(3'E to CE is SpoIIGA (59, 135), which receives a signal from a protein, SpoIIR, generated in the forespore under the control of CF (48, 61, 83). Conversion of pro-CSK to OK requires SpoIVFB (l8, 19, 89), which is either the processing enzyme or its regulator, and is negatively regulated by SpoIVFA and BofA (18, 19, 51, 116). SpoIVFA, SpoIVFB, and BofA appear to be integral membrane proteins (19, 115, 116), and SpoIVFA and SpoIVFB have been shown to be localized at the boundary between the mother cell and the forespore (115). Activation of SpoIVFB for pro-6K processing requires the production of SpoIVB under the control of 62 0'6 in the forespore (17, 32). SpoIVB is inferred to be a secreted protein and is presumed to overcome the inhibitory effect of SpoIVFA and BofA (1 7, 145). ‘ Pro-(SK has 20 amino acid residues at its N-terminus which must be removed to generate active OK (69, 89, 136). Two lines of evidence indicate that pro-(SK is transcriptionally inactive (89). First, expression of (SK—dependent gene fusions does not begin until processing occurs. Second, when added to core RNA polymerase (RNAP), pro-(rK fails to direct transcription from (SK-dependent promoters in vitro. The role of the pro-sequence in preventing transcription is not clear. 0ne function of the pro-sequence may be to mask the DNA-binding activity of OK , since the affinity binding constant of purified pro- OK for promoter DNA is one order of magnitude lower than that of OK (21). The results presented here suggest additional functions of the pro-sequence. We show that K the majority of pro-o is membrane-associated in cell extracts, and is not associated with the core subunits of RNAP. In agreement with this observation, we find that pro-0'K immunolocalizes to the mother cell membranes that surround the mother cell and the forespore in sporulating cells. Moreover, pro- (5" fails to bind to core RNAP in vitro under conditions that permit (5" binding. These results suggest that two more functions of the K pro-sequence of pro-0' are to inhibit RNAP core binding and to promote association with the membrane, where processing may occur. 63 MATERIALS AND METHODS General methods. Sporulation was induced by resuspending growing cells in SM medium as described previously (43). The onset of sporulation (To) is defined as the time of resuspension. The B. subtilis strains used in this study are: PY79 (wild-type) (154), PY79/pSLl (89), BSLSl (spoIVFAABsscat) (90), BK183 (spoIVA67 trpCZ) (118), RL831 (spoIIlGA::neo) (121), RL87 (spoIVFBl52) (l8) and RL136 (spoIVFBI52 spoIVCBAI9) (18). Western blot analysis. Samples of different fractions equivalent to the same original volume of culture, or containing the same amount of protein, as determined by the Bradford method (13), were separated on SDS-12% Prosieve polyacrylamide gels (FMC) using Tris/Tricine electrode buffer (0.1 M Tris, 0.1 M Tricine, 0.1% SDS) and electroblotted to Immobilon-P membrane (Millipore). The membrane was probed with either polyclonal anti-pro- 0K antiserum (89), anti-FtsH antiserum (a gift from S. Cutting and T. Ogura), anti-E.coli core RNAP antiserum (a gift from M. Chamberlin and C. Kane), or monoclonal anti-0'E antibody (a gift from W. Haldenwang). In some experiments, the membrane was stripped and reprobed with a different antibody. Horseradish peroxidase conjugated secondary antibody was either goat-anti-rabbit IgG or goat-anti-mouse IgG (Bio-Rad). Chemiluminescence detection was performed following the manufacturer’s instructions (Amersham, ECL). Column chromatography. Mini-columns (5.5x70 mm) were made from pasteur pipettes with the narrow end cut off and sealed with glass fiber and beads. Three types of gel filtration media were used: Sephacryl S-300, Sephadex G-200, and Sephadex G-100 (Pharmacia). The flow rate was controlled by gravity and ranged from 50 to 80 til/min. The void volume and fractionation range were determined by passing various combinations of Dextran Blue, alcohol dehydrogenase (150 kDa), BSA (66 kDa), carbonic 64 anhydrase (29 kDa) and cytochrome C (12.4 kDa) through the columns. Usually a 100 [.11 sample was loaded, eluted with the same buffer, and 120 u] fractions were collected. Salt or detergent treated fractions were eluted with buffer adjusted to contain the same concentration of salt or detergent. If necessary, the column fractions were concentrated by 10% trichloroacetic acid (TCA) precipitation. Subcellular fractionation. Figure l is a diagram showing the fractionation scheme used in our experiments. Cells were collected by centrifugation (5,000g), washed with l M NaCl, and stored at -80°C. The cell pellet was resuspended in 7.5% of the original volume in lysis buffer [25 mM Hepes-KOH (pH 7.5), 50 mM NaCl, 10 mM MgC12, 1 mM EDTA, 0.5 mM DTT, 10% glycerol, 1 mg/ml lysozyme, 0.1 mg/ml DNase I, 20 ug/ml RNase A, 1 mM PMSF] and incubated for 10 min at 37°C. Cells were then chilled and lysed by passage through a French Pressure cell twice at 1,800 psi. The crude lysate was incubated at 30°C for 10 min. Cell debris was removed by centrifugation at, 12,000g for 10 min. No nucleic acids were detected when the supernatant was analyzed by 2% agarose gel electrophoresis. The supernatant was then subjected to high speed centrifugation (200,000g) for 1.5 hour at 4°C. The pellet was homogenized in 1/5 of the lysate volume in sucrose gradient buffer [25 mM Hepes-KOH (pH 7.5), 50 mM NaCl, 10 mM MgC12, 1 mM PMSF] plus 5% sucrose and loaded on top of a sucrose density gradient made with 2 ml of 55% (wt/vol) and 2 ml of 25% (wt/vol) sucrose in buffer in a 5-ml ultracentrifuge tube. After centrifugation at 200,000g for 4 hr at 4°C, the membrane fraction was recovered at the interface between 25% and 55% sucrose. The supernatant (cytoplasmic fiaction) after the initial high speed centrifugation (100 1.11) was loaded onto a gel filtration column and eluted with lysis buffer omitting the lysozyme, DNase I and RNase A. Fractions of 120 111 were collected and analyzed by Western blotting. In the experiments testing the effect of salt and detergent, the supernatant afier low speed centrifugation was divided into six aliquots. Salt or detergent was added to different 65 Cells collected at 3.5 h. Into sporulation l Lysozyme, DNase, RNase treatment French pressure cell Iysls 1 Crude cell lysate 12,0009, 10 min Pellet Supernatant (dlscard) 200,0009, 90 min 1 l Pellet Supernatant Sucrose gradlent Gel-filtration column centrlfugatlon chromatography 1 Membrane fraction Figure 1. Diagram of subcellular fractionation of sporulating B. subtilis cells. See Materials and Methods for details. 66 final concentrations and one fraction was left untreated. All aliquots were kept for 20 min at 4°C and then subjected to high speed centrifugation as noted above. The supernatant and pellet fractions were analyzed by Western blotting. Immunofluorescence microscopy and image processing. The affinity purified rabbit polyclonal anti-<3K antibodies (115) were a gift of 0. Resnekov, and were used at a 1:500 dilution. The secondary antibodies (Jackson Immunolabs) were affinity-purified donkey anti-rabbit antibodies conjugated to fluorescein isothiocyanite (FITC), and were used at a 1:100 dilution. DNA was stained with propidium iodide (PI; Molecular Probes) at a final concentration of 10 ug/ml. Cells were harvested 2.5 and 3.5 h after the onset of sporulation. Immunofluorescence experiments were performed as described by Pogliano et al. (112). Fluorescence micrographs were recorded using a cooled CCD camera (Princeton Instruments) and a PC with the MetaMorph imaging system (version 3.0; Universal Imaging Corp.). PI images were assigned to the red channel and FITC images to the green channel. Adobe Photoshop (version 3.0.5) was used to overlay FITC images on P1 images. In vitro reconstitution of RNA polymerase holoenzyme. B. subtilis core RNAP was partially purified as described previously (69). To isolate pro-(SK and OK, PY79/pSLl cells were collected at 4.5 hours into sporulation without IPTG induction. Cells from 3 ml of culture were pelleted by centrifugation at 5,000g for 5 min, resuspended in 100 pl lysis buffer substituting KCl for NaCl, and incubated at 37°C for 10 min. The KCl concentration was adjusted to 0.6 M and the lysate sonicated. Afier 10 min at 30°C, the lysate was cleared of unlysed cell debris by centrifugation for 10 min in a microfuge. The supernatant (120 1.11) was loaded onto a Sephadex G-100 column and eluted with lysis buffer containing 0.6 M KCl without enzymes. Fractions in the molecular weight range of K monomeric pro-0' were pooled and dialyzed against lysis buffer to remove the salt. The 67 dialyzed sample was divided into two 100 pl aliquots, each containing approximately 5 pmole of pro-(rK and 1 pmole of OK, as determined by Western blotting. Ten microliters of core RNAP in storage buffer, containing approximately 15 pmole of core subunits, as determined by SDS-PAGE and Coomassie blue staining, was added to one aliquot, and 10 u] of storage buffer was added to the second aliquot. The two aliquots were incubated on ice for 1 hr. Each aliquot was then fractionated on the same Sephadex G-100 column. Column fractions were precipitated with TCA and analyzed by Western blotting. 68 RESULTS The majority of pro-(rK is membrane-associated. To investigate whether pro-(rK is associated with core RNAP, we fractionated crude lysates of sporulating wild- type B. subtilis. To facilitate the comparison of pro-(SK and OK, cells were collected at 3.5 h after the onset of sporulation (T35), a time at which approximately equal amounts of pro- 0'" and UK are present in cells. Cells were treated with lysozyme and lysed by passage through a French Pressure cell. The crude lysate was cleared of cell debris by low speed centrifugation (12,000g) and the supernatant was then subjected to high speed centrifugation (200,000g). The resulting pellet was further fractionated on a sucrose density gradient. Samples of different fractions were analyzed by Western blotting using anti-pro-O'K antibodies (89). As shown in Figure 2A, the majority of pro-(3'K was detected in the high-speed pellet (lane 3), while UK was predominantly present in the high-speed supernatant (cytoplasmic fraction) (lane 2). After further fractionation of the high speed pellet on a sucrose density gradient, pro-Cl“K remained in the membrane fraction, whereas the small amount of OK in the sample formed a pellet at the bottom of the sucrose gradient tube (data not shown), suggesting that it was present in residual cell debris or in a large aggregate of proteins. The cytoplasmic fraction was apparently depleted of membrane vesicles as FstH, an integral membrane protein, was not detected (lane 2). All the FtsH was found in the initial high speed pellet (lane 3) and was recovered in the purified membrane fraction (lane 4). The purified membrane fraction was essentially free of core RNAP, as little B and 3' subunits were detected (lane 4). These results show that the majority of the pro-<3K in the crude lysate, unlike OK, is not associated with core RNAP, but is membrane-associated. To ask whether pro-<5K in the cytoplasmic fraction was associated with core RNAP, 69 Figure 2. Subcellular fractionation of extracts of sporulating wild-type cells. Cell extracts were fractionated as diagramed in Figure 1. Proteins in different fractions were subjected to SDS-polyacrylamide gel electrophoresis and analyzed by Western blotting. (A) Samples of different fiactions equivalent to the same original volume of wild-type T 35 culture were analyzed for pro-CfK and OK, as well as FtsH and the B and 13' subunits of RNAP by Western blotting. Lane 1, supernatant after 12,000g centrifugation. Lane 2, supernatant after 200,000g centrifugation. Lane 3, pellet afier 200,000g centrifugation. Lane 4, membrane fraction purified by sucrose density gradient. (B) The supernatant after 200,000g centrifugation was subjected to size-fractionation by passage through a Sephacryl S-300 column. Equal volumes of the column fractions were analyzed for pro-(5’K and OK and the RNAP B and 13' subunits. Fraction numbers are indicated. (C) Samples of different fi'actions equivalent to the same original volume of wild-type TI 7 culture were analyzed for pro-(5E and GE. Lane contents are the same as for panel A. (D) The supernatant after 200,000g centrifugation was subjected to size-fractionation by passage through a Sephacryl S-300 column. Equal volumes of the column fractions were analyzed for pro-(SE and 6E. pro-cK OK 1234 :3-.- FtsH -- -" [5+3 70 B 12 3 4 5 6 pro-0K,- .- cK’ [3+B' 1..- D 1234567 pro-GE- .- O‘E/ -. fi —. 71 the supernatant after high speed centrifugation was size-fractionated by passage through a Sephacryl S-300 column, which has a fractionation range of 10 kDa to 1,500 kDa. The molecular weight of CK RNAP holoenzyme is about 370 kDa, which should render it readily separated from very high molecular weight complexes and from free pro-oK (29 kDa) in this column. The column fractions were analyzed by Western blotting using anti- pro-C)J< antibodies or antibodies against E. coli core RNAP. As shown in Fig. ZB, most of the pro-(1’K was eluted at or near the void volume of the column (lane 1), suggesting that it is part of a very large complex (>l,500 kDa). UK was eluted later than pro-GK and was co- eluted with the B and 13' subunits of RNAP (lanes 2 and 3), indicating that 6K was present K in the holoenzyme form. Taken together, these results show that pro-0’ is not associated with core RNAP in the crude extract of sporulating B. subtilis; rather, most of it is associated with membrane and the rest is present in a large complex of unknown composition. We next asked whether pro-<3E fractionates in the same way as pro-6K. Pro-(SE is the inactive precursor of GE. Since pro- GE is synthesized earlier than pro- GK, wild-type cells were collected at 1 h and 40 minutes after the onset of sporulation, a time at which approximately equal amounts of pro-(rE and GE are present. Cell extracts were prepared and fractionated as described above. Pro-(1'E and 0E were analyzed by Western blotting using monoclonal anti-(rE antibody. As shown in Figure 2C, pro-(SE fractionated in a pattern similar to that of pro- OK. The majority of pro-GE was detected in the high speed pellet (lane 3) and was recovered in the purified membrane fraction (lane 4). The small amount of pro-(IE in the cytoplasmic fraction (lane 2) was eluted in the void volume of a Sephacryl S-300 column (Figure 2D, lane 1), suggesting it is part of a very large complex 72 (>1,500 kDa) of unknown composition. 0‘5 was found almost exclusively in the cytoplasmic fiaction (lane 2), and co-eluted with core RNAP (Fig. 2D, lanes 2 and 3) or as free OE (Figure 2D, lanes 5 and 6) from the sizing column. Effect of detergent and salt treatment on the membrane association of pro-6K. After the lysate of wild-type cells collected at T 3. 5 was cleared of cell debris by low speed centrifugation (12,000g), the supernatant was treated with detergent or salt, then subjected to high speed centrifugation (200,000g). The resulting supernatant and pellet fractions were analyzed by Western blotting to further characterize the membrane- association of pro- UK. As expected for a protein interacting with membranes, pro- OK was solubilized by 1% Triton X-100 treatment (Figure 3A). In contrast, the small amount of OK found in the high speed pellet remained in the pellet upon detergent treatment (Figure 3A). This result is consistent with the finding that 6K in the pellet did not fractionate with membrane in a sucrose gradient and supports the idea that O’K is not membrane-associated. Instead, we speculate that it may be associated with residual cell debris or a large aggregate of proteins. To determine the size of pro-0’K and 6K in the Triton X-lOO-treated supernatant, instead of subjecting it to high speed centrifugation it was size-fractionated by passage through a Sephadex G-200 column, which has a fractionation range of 5 kDa to 600 kDa. Figure 3B shows that the majority of UK was eluted near the void volume, suggesting that most of the UK was not dissociated from core RNAP by 1% Triton X-100. Pro-(5K was eluted in the included volume, indicating that pro-(rK was dissociated from membranes. In addition, the large complex that had remained in the supernatant of extracts not treated with detergent (Figure 28, lane I) appeared to be dissociated by Triton X-100, suggesting that 73 Figure 3. Effects of detergent and salt treatment on fractionation of pro-6K and OK. (A) Crude cell extract was cleared of cell debris by 12,000 g centrifugation. The supernatant (S 12,000g) was divided into six aliquots and treated with either 1% Triton X-100, 0.5 M NaCl, 1 M NaCl, 0.25 M KSCN, 0.5 M KSCN, or left untreated. These aliquots were then subjected to high speed centrifugation (90 min, 200,000g). Samples of the supernatant (S) and pellet (P) fractions equivalent to the same original volume of wild-type T3.5 culture were analyzed by Western blotting using anti-pro-(SK antibodies. (B) The supernatant after 12,000g centrifugation was treated with 1% Triton X-100 (lane L) and size-fractionated by passage through a Sephadex G-200 column (lanes 1 through 11). Equal volumes of the column fractions were analyzed for pro-(SK and UK by Western blotting. Numbers indicate the column fractions. Fractions 1 and 2 contained materials eluted in the void volume of the column. 74 209. 2m... _ zomx Smud— _oez s:— _oaz 2m...— co:..... 3.. — UQaOOnuC—a — 38.3 A SS PSPSPSPS PSP ,pro-cK -n--... -—-—--—---\GK 1234567891011L ~— ,pro-c" l _ l 75 the interactions of pro-(3’K in the large complex are primarily hydrophobic in nature. Both a nonchaotropic salt (N aCl) and a chaotropic salt (KCN S) partially solubilized pro-(5K from the membrane (Figure 3A), suggesting that both ionic and hydrophobic interactions are likely to be involved in the binding of pro-(3'K to the membrane. The pro-(3'K remaining in the pellet after salt treatment may be present inside vesicles and therefore incapable of release by salt. In contrast, both 0.5 M NaCl and 0.25 M KCNS completely solubilized the residual UK from the pellet. Membrane association of pro-0'K does not depend upon sporulation- specific gene products. The products of the mother cell-expressed spoI VF operon are thought to be intimately involved in the processing of pro-6K. SpoIVFB is either the processing enzyme or a regulator of the processing enzyme (18, 19, 88, 89). SpoIVFA negatively regulates the activity of SpoIVFB and these proteins are thought to form a complex in the mother cell membrane that surrounds the forespore (18, 19, 115). To investigate whether spoI VF gene products are required for the membrane association of pro-6K, a lysate from spoI VF null mutant cells was fractionated and analyzed by Western blotting. As shown in Figure 4A (lanes 1-3), the majority of pro-6K was present in the pellet after high speed centrifugation, just as in wild-type cells (Figure 2A, lanes 1-3). K Since spol VF null mutant cells are processing deficient, only pro-0’ is present. We conclude that the membrane association of pro-(SK does not depend upon spoI VF gene products. Another mother cell-specific protein, SpoIVA, is located at the forespore surface and controls the assembly of the spore cortex and coat (24). Pro-6K processing is impaired K in spa! VA mutant cells (89). We found that pro-G associates with the membrane in spa] VA mutant (spol VA 6 7) cells (data not shown), suggesting that the spa] VA mutation 76 A 1234567 ‘- ~ -- .---—pro-c" wu- Figure 4. Specificity of the membrane-association of pro-6K. (A) Sporulating (T 3. S) spol VF null mutant (BSLS 1) cells and vegetative wild-type cells expressing pro-(5'K from a plasmid (pSLl) were fractionated. Proteins equivalent to the same original volume of cells were analyzed by Western blotting. Lanes 1-3, 12,000 g supernatant, 200,000 g supernatant and 200,000 g pellet, respectively, of sporulating BSLSI cells. Lanes 4-7, 12,000 g supernatant, 200,000 g supernatant, 200,000 g pellet and gradient-purified membrane, respectively, of vegetative PY79/pSLl cells. (B) Western blot analysis of 2 p. g of protein from sucrose gradient-purified membrane of sporulating (T35) wild-type (PY79) cells (lane 1) and sporulating (T 3. 5) wild-type cells containing plasmid pSLl afier being induced to make pro-(5'K for 10 min (lane 2), 30 min (lane 3) or 3.5 h (lanes 4). 77 does not impair the processing of pro-0'K by interfering with its membrane association. K To ask whether membrane association of pro-0' occurs in the absence of any sporulation-specific gene products, we induced production of pro-(1'K during vegetative growth from a multicopy plasmid, pSLl, which has the intact sigK gene fused to an IPTG- inducible promoter (Pspac) (89). A lysate was prepared from IPTG-induced cells containing pSLl, fiactionated, and analyzed by Western blotting. About half of the pro- OK was pelleted by high speed centrifugation and it remained in the membrane fraction afier sucrose gradient purification (Figure 4A, lanes 47). Hence, membrane association of pro-GK does not require expression of any sporulation-specific genes. In this experiment, about half of the pro-6K remained in the supernatant afier high speed centrifugation (Figure K 4A, lane 5), whereas in sporulating cells only a small amount of pro-6 was found in the cytoplasmic fraction (Figure 2A, lane 2). The difference may be due to overproduction of pro-(3’K in IPTG-induced vegetative cells containing pSLl. Like the pro-0'K in the cytoplasmic fraction of sporulating cells (Figure 2B, lane 1), the pro-(1'K in the cytoplasmic fraction of the vegetative cells appeared to be present in a large complex (>1,500 kDa) of unknown composition (data not shown). Pro-0' K binding sites are not saturated on the membranes of sporulating cells. To test whether membranes in sporulating cells have the ability to bind more pro- UK, we induced the production of pro-(5'K from pSLl during sporulation. Wild-type cells bearing pSLl were induced with IPTG for 10 min, 30 min, or 3.5 h before being harvested at T35. Membrane fractions from these cells were purified by sucrose gradients. Two micrograms of protein from each membrane preparation was analyzed by K Western blotting. As shown in Figure 4B, more pro-0' was detected in membranes 78 K prepared from cells overproducing pro-c , as compared to membranes prepared from wild-type cells. These results indicate that pro-0'K binding sites are not saturated on the membrane of sporulating cells. Pro-c!K localizes to the mother cell membranes that surround the forespore and the mother cell of the postengulfment sporangium. Pro-(3‘K and UK were immunolocalized in sporulating cells using affinity purified anti-(5’K antibodies (115), secondary antibodies coupled to FITC, and fluorescence microscopy. The rabbit K polyclonal anti-(SK antibodies bind pro-o as well as UK. Therefore, we were able to distinguish both forms of the transcription factor only by co-staining the nucleoids to determine the stage of sporulation and by analyzing mutants that are either deficient in pror-d'K processing or are known to synthesize mature 0K in the absence of processing. Postengulfinent sporangia at stages III and IV in sporulation can be readily identified by their DNA staining pattern. Whereas the forespore chromosome of stage III sporangia (Figure 5A], red, arrow) more closely resembles the mother cell chromosome, albeit slightly more condensed, the forespore nucleoid of stage IV sporangia assumes a characteristic toroidal structure (Figure 53,, red, arrow) upon association with the CUB-type SASP (111). Wild-type postengulfment sporangia, which could be assigned to stage III in sporulation by virtue of their DNA staining pattern (Figure 5A], red), displayed pro- (SK/GK immunostaining (Figure 5A2, green) in the periphery of most of the mother cell and on one side of the engulfed forespore. The peripheral forespore staining was often evident as a crescent at the interface between the forespore and the larger volume of the mother cell. There was very little overlap between the green immunostaining of pro-(l‘K/c'K and the red 79 Figure 5. Immunolocalization of pro-0K and 0K in sporulating cells. The sporangia were harvested at T?“5 in (A) and at T35 in (B) through (E) and prepared for immunofluorescence microscopy as described in Materials and Methods. Arrows point to the engulfed forespore compartment and are oriented perpendicularly to the long axis of the sporangia. DNA was stained with P1 (red) (A l, 8,, C1, D1, El). Immunostaining of pro— OK and OK is shown in green (A 2, 32’ C 2, D 2, E2). Where the red and green fluophores overlap, as in the doubly exposed images shown in B3 and E3, a yellow to orange color is visible. (A) Wild-type sporangia with almost equally as condensed mother cell and forespore nucleoids, which is characteristic of cells at stage III in sporulation before pro-c7K is processed to 0'". (B) Wild-type sporangia at stage IV in sporulation, when the forespore nucleoid has assumed its toroidal shape and pro-0’K has been processed to 6K in the mother cell. (C) Pro- (SK processing deficient spoIIIGA::neo mutant sporangia of strain RL831. (D) Processing deficient spoIVFBISZ mutant sporangia of strain RL87. (E) spoIVFBISZ spoIVCBA19 doubly mutant sporangia of strain RL136, which synthesize mature 6K in the absence of a filnctional protease for pro-(3'K processing. DNA and 0K II I WT IV spoIIIGA: :neo spoIVFBISZ spoIVFBISZ spoIVCBA19 81 nucleoid staining (Figure 5A 3), indicating that pro- (SK/GK is associated with the mother cell membranes that surround the mother cell and the forespore. After sporulation proceeded to stage IV, as indicated by the toroidal forespore nucleoid (Figure SB 1, red, arrow), when pro-(5’K is known to be converted to OK in the mother cell (89), the pattern of pro-ch/O’K immunostaining changed to include the cytoplasm of the mother cell (Figure SBZ, green). This staining pattern was consistent with the previously reported even distribution of CK throughout the mother cell (1 15). In spoIIlG and spoIVFB mutant sporangia, which are deficient in pro-(SK processing and do not proceed in development beyond stages III and IV, respectively, pro- GK immunostaining was detected in the periphery of the mother cell and forespore (Figure 5C and 5D, green). Because this staining pattern was similar to the one observed in wild- type stage III sporangia (Figure 5A, green), we conclude that pro-0K is associated with the mother cell membranes that surround the mother cell and the forespore. In sporangia of a spoIVFB mutant that produces mature 0'" without processing due to a deletion (spoIVCBA19) in the pro-sequence-encoding portion of sigK, immunostaining of UK was detected throughout the mother cell (Figure 5E, green). As this staining pattern is reminiscent of the one observed in wild-type stage IV sporangia (Figure SB, green), we infer that after proteolytic activation O'K is released from the membrane and becomes soluble in the mother cell cytoplasm. Therefore, the change in pro-(SK/trK immunostaining from stage III to stage IV in sporulation resulted from the conversion of membrane associated pro-(3'K to soluble O’K, consistent with our subcellular fractionation results (Figure 2). Pro-t1K does not bind to exogenous core RNAP in vitro. Very little, if 82 any, of the pro-<3K in lysates is associated with core RNAP (Figure 2). Is this because pro- UK is unable to bind to core RNAP (due either to intrinsic inability to bind or to associate with other cellular components like membranes) or because core RNAP is not available for binding? To address this question, our strategy was to dissociate both pro-(SK and UK from other components in the cell lysates and incubate them with exogenous core RNAP. To increase the production of pro-(1'K and UK, we used wild-type cells containing pSLl. In the absence of IPTG induction, leaky expression from the Pspac promoter in pSLl allows accumulation of pro-0'K during sporulation so that when cells are harvested at T a time 4.5’ at which more 6K has accumulated, both pro-GK and UK are present at a higher level than in wild-type cells at T35. A crude lysate was prepared from cells harvested at T 4' 5 and KCl was added to a final concentration of 0.6 M. The salt-treated lysate was then size-fractionated on a Sephadex G-100 column, which has a fractionation range of 4-150 kDa. Both pro- 6K and 6K in untreated crude lysate were excluded from this column (data not shown). Afier salt treatment, a portion of the pro-0'K and UK was retained in the column (Figure 6A), indicating that UK was partially dissociated from core RNAP and pro-0'K was partially dissociated from the membrane and/or the large complex that remained in the supernatant after high speed centrifugation (Figure ZB, lane 1). Fractions 5-7 containing dissociated pro-(SK and UK were pooled and dialyzed to remove the salt. The dialyzed sample was incubated with either partially purified core RNAP or with the core RNAP storage buffer, and then fractionated in separate experiments on the same Sephadex G-100 column. Upon incubation with core RNAP (Figure 6B), UK was eluted in the void volume, suggesting 83 K Figure 6. OK, but not pro-6 , reassociates with core RNAP after being dissociated by salt treatment. The 12,000g supernatant was prepared in the presence of 0.6 M KCl and separated by a Sephadex G-100 column (A). The void volume of this column was fractions 1-2, wherein pro-(1'K and OK would be eluted if not treated with salt. Fractions 5- 7 containing dissociated pro- GK and OK in approximately the monomeric size range were pooled and dialyzed. The dialyzed fractions were incubated with (B) and without (C) exogenous core RNAP. Proteins were then separated by the same Sephadex G-100 column and analyzed by Western blotting with anti-pro-(l’K antibodies. Only UK shifted back to the void volume upon incubation with core RNAP (panel B, lanes 1 and 2), indicating formation of the holoenzyme. 84 1234567891011 V/pro-c" .-.- l K ,pro-c Woo-'- l\al< ,/pro-cK kw,“ ‘0“ 85 that it had reassociated with core RNAP. Pro-6K was eluted in the included volume after incubation with either core RNAP (Figure 6B) or storage buffer (Figure 6C). The same results were obtained when the experiment was repeated with a lysate made from wild-type cells harvested at T3 5 (data not shown). These results indicate that pro- 0K does not bind to core RNAP, even after it has been dissociated from other cellular components, whereas GK readily reassociates with core RNAP under these conditions. 86 DISCUSSION K We have demonstrated that the majority of pro-6 in cell lysates is membrane- associated and is not bound to core RNAP. In contrast, nearly all of the 6K in lysates of sporulating cells is present in the cytoplasmic fraction and appears to be bound to core RNAP. In sporulating cells, pro- O'K appears to associate with the mother cell membranes that surround the mother cell and the forespore, as visualized by immunofluorescence microscopy. Processing releases 6K into the mother cell cytoplasm. Most of the pro-(5K and UK can be dissociated from large components in the cell extract by 0.6 M KCl. After removal of the salt, 6", but not pro-6K, could bind to added core RNAP. These results indicate that the pro-sequence of pro-GK promotes membrane association and inhibits RNAP core binding. The ability of pro- OK to associate with a membrane may facilitate its proteolytic processing to active CK. SpoIVFB has been proposed to be either the protease that processes pro-OK or a regulator of the protease (18, 19, 88, 89). Encoded in the same operon as SpoIVFB is SpoIVFA, which appears to inhibit SpoIVFB activity until a signal is received from the forespore (17-19). SpoIVF B and SpoIVF A have been shown to be localized at the boundary between the mother cell and the forespore (115). As depicted in Figure 7, these proteins presumably insert into the mother cell membrane that surrounds the forespore since the spa] VF operon is expressed in the mother cell (19). Likewise, bofA is thought to be expressed in the mother cell (1 16). Although BofA has not yet been shown to be localized to the mother cell membrane that surrounds the forespore, it has three putative transmembrane segments and, like SpoIVFA, it appears to inhibit SpoIVFB activity (18, 116). The signal that relieves this inhibition and leads to pro-(1'K processing is 87 mother cell forespo re Figure 7. Model depicting association of pro-(5K with the mother cell membrane that surrounds the forespore and signal transduction between the forespore and the mother cell K leading to the processing of pro-G . The shaded oval represents a possible abundant membrane protein that interacts with pro-6K. See the text for details. 88 generated in the forespore by (JG-dependent expression of spa] VB (17, 32) (Figure 7). SpoIVB appears to have a signal sequence, so it may cross the forespore membrane in order to accomplish its signaling function (17, 145). If processing of pro-(rK requires it to directly interact with SpoIVFB, then the ability of pro-(1'K to associate with the mother cell membrane that surrounds the forespore may facilitate processing by promoting this interaction. Our immunolocalization studies showed that pro-<3K interacts not only with the mother cell membrane that surrounds the forespore but also with the membrane that surrounds the mother cell in sporulating wild-type cells, as well as in spoIIIG and spa] VF B mutant cells (Figure 5). Does the pro-(5'K associated with the membrane that surrounds the mother cell get processed? It appears that most, if not all, of the pro-(1'K produced in sporulating cells is processed to UK. First, very little pro-0'K was detected late during sporulation (89). Second, a pulse-chase experiment demonstrated that the half-life of pro-0K is about 30 min. The majority of the 35S-label in pro-(1‘K at T 3 was found in 0'" by T4 (data not shown). Therefore, it seems likely that pro-<3K associated with the membrane that surrounds the mother cell is either processed there or it dissociates and is processed elsewhere (e.g., at the mother cell membrane that surrounds the forespore). However, we cannot rule out the possibility that some of the pro-0K that associates with the membrane that surrounds the mother cell is degraded. The pro-sequences of both mother cell specific 0' factors appear to promote membrane association. We found that pro-(SE in lysates of sporulating cells had very similar fractionation properties as pro-(SK (Figure 2). The majority of pro-(SE was membrane-associated and not bound to core RNAP. O'E, like 6K, appeared to be associated 89 with core RNAP in the cytoplasmic fraction. The pro-sequence of pro-(SE has been proposed to form an amphipathic a helix with a highly charged face (109), which could presumably interact with negatively charged head groups of membrane lipids, but this would not explain the preferential localization of pro- 0’5 to the sporulation septum (47, 60). Genetic suppression (110) and chemical cross-linking studies (47) suggest that pro-(SE E interacts with its putative processing enzyme, SpoIIGA. However, pro-0' may also interact with another protein in the septal membrane since localization of pro-'0'E (47) and a pro-GEzzGFP fusion protein (60) to the septal membrane occurs in SpoIIGA mutant cells. K The 20 amino acid pro-sequence of pro-c is very hydrophobic, except for two charged residues at positions 13 and 14 from the N-terminus (136). The charged residues might prevent the pro-sequence from inserting into the membrane like a transmembrane domain of a typical integral membrane protein. In support of this prediction, virtually all K the pro-0' was found in the aqueous phase of a Triton X-114 fractionation experiment (data not shown). We speculate that pro- 6K is peripherally associated with the membrane, perhaps via binding of the pro-sequence to an abundant integral membrane protein (Figure 7), since the membranes in sporulating cells have the capacity to bind much more pro-(3'K when it is overproduced (Figure 48). Alternatively, it is possible that the pro-sequence of pro-(3K does not interact directly with a membrane component. Removal of the pro-sequence could induce a conformational change that prevents membrane association and/or uncovers a site that gives O'K a higher affinity for core RNAP than for the membrane. The interaction of pro-oK with membranes does not require spoI VF gene products (Figures 4A and 5D), or the products of spa] VA (data not shown) or spoIIIG (Figure 5C). Indeed, the interaction does not require any sporulation-specific gene 90 products since pro-(3’K produced in vegetative cells was membrane-associated (Figure 4A). A small portion of the pro-(SK and pro-(3'E in cell lysates was not membrane- associated (Figures 2A and 2C). Rather, the pro-G factors appeared to be present in large complexes (>1,500 kDa) (Figures 28 and 2D) of unknown composition. The large complexes could be aggregates of the pro-0' factors alone or in combination with other proteins. Different methods of cell breakage had little effect on the proportion of pro- 0'" that was membrane-associated versus present in a large complex. We tested sonication and osmotic shock lysis procedures (data not shown) in addition to the French pressure cell lysis method reported here. In addition to promoting the membrane association of pro-6K, the pro-sequence also appears to inhibit RNAP core binding. The B and 13' subunits of core RNAP were barely detectable in a membrane fraction that contained abundant pro-(SK (Figures 2A, lane 4). Also, the pro-(1'K that was not membrane-associated appeared to be present in a large complex (>l,500 kDa) containing very little [3 and 13' (Figure 2B, lane 1). Moreover, much less pro- OK than O’K bound to core RNAP after both had been released from large cellular components by salt treatment and the salt was removed by dialysis (Figure 63). We cannot rule out the possibility that pro-(SK remained in small complexes with itself or another protein(s) upon treatment with 0.6 M KCl. However, pro-(3'K showed a similar elution profile from a sizing column as UK both in the presence of 0.6 M KC1 (Figure 6A) and after salt removal when core RNAP was not added (Figure 6C). It seems unlikely that pro-<3K was irreversibly denatured by 0.6 M KCl since GK readily associated with core RNAP upon its addition (Figure 6B). Under these conditions, the pro-sequence greatly hindered RNAP core binding. The pro-sequence may be close to the core-binding domain in the 91 three-dimensional structure of pro-(SK, directly blocking core binding. Alternatively, cleavage of the pro-sequence may result in a conformational change which activates core binding. In agreement with our findings, Johnson and Dombroski (58) recently demonstrated that purified OK competes much more efficiently than purified pro-0'K for binding to E. coli core RNAP. Removal of only 6 amino acid residues from the N- terminus of pro-6K restored core binding and the holoenzyme was transcriptionally active. 0'70 of E. coli does not bind to promoter DNA unless its amino-terminal region 1.1 K is removed (22). Removal of the pro-sequence from pro-o results in a ten-fold increase in DNA-binding activity (21). Our results suggest that in addition to modulating DNA-binding activity, the pro-sequence of pro-(SK promotes its membrane association, perhaps facilitating processing to OK. Removal of the pro-sequence releases OK from the membrane and appears to unmask its RNAP core-binding activity, allowing the fimctional holoenzyme to form. ACKNOWLEDGMENTS We are very grateful to S. Lu, 0. Resnekov, W. Haldenwang, M. Chamberlin, C. Kane, S. Cutting, and T. Ogura for providing antibodies. We thank B. Johnson and A. Dombroski for communicating their results prior to publication. This research was supported by the Michigan Agricultural Experiment Station, by grant GM43585 from the National Institutes of Health to L. K., and by NIH grant GM18568 to R. Losick. A. Hofineister was a postdoctoral fellow of the Alexander von Humboldt Foundation and performed the immunofiuorescence microscopic studies described in this chapter. Chapter V Summary and Perspective 92 93 Mother-cell gene expression during B. subtilis sporulation is governed by the concerted action of transcription factors OE, SpoIIID, GK, and GerE. While the timely appearance of these transcription factors is critical to the temporal regulation of mother-cell gene expression, the timely disappearance of the early transcription factors is important as well. The replacement of GE and SpoIIID with OK and GerE completes a switch in the mother cell pattern of gene expression. Experiments described in this dissertation demonstrate that the appearance of CK initiates a negative feedback loop to facilitate the disappearance of GE and SpoIIID. 0K appears to affect the activity of the two positive regulators of sigE transcription, 0'" and SpoOA~P. O’K could inhibit 6A activity by competing with GA for binding to core RNAP. The negative effect on SpoOA~P is likely to be mediated by a (SK-dependent gene product(s). The biological significance of the negative feedback regulation of OK on the levels of SpoIIID and GE needs to be addressed. The negative effect of UK on sigE and spoIIID transcription could be bypassed by fusing these genes to a (SK-dependent promoter (e.g. the gerE promoter). For example, in an otherwise wild-type background, a single copy of PgerE-PspolllD-spoIIID in the chromosome should result in a higher level of SpoIIID expression late during sporulation. This is predicted to inhibit expression of certain cot genes (e.g. cotC, cotD, and cotX) that are repressed by SpoIIID, based on in vitro studies (39, 40, 50, 69). It would be interesting to determine whether the resistance or structural characteristics of spores is altered when the negative feedback regulation of OK on SpoIIID or CE is bypassed. The negative effect of OK on SpoOA~P may be difficult to investigate because there are many potential regulatory sites in the phosphorelay (34). It is possible to investigate the 94 negative effect of OK on 0" activity. To test the model that O'K has higher affinity than 0’" for core RNAP, in vitro assays could be developed to directly compare the core-binding activity of various sigma factors. If these studies support the idea that UK outcompetes 6A for binding to core RNAP, it may be possible to swap the core-binding domain of GA with that of OK so as to increase the core-binding affinity of GA. The strain containing 6A with the swapped core-binding domain would then be analyzed for sporulation phenotype and the production of GE and SpoIIID. The appearance of (3" not only turns on expression of genes in the 0K regulon, but also turns off genes in the OE regulon. Earlier production of UK by deleting the part of the sigK gene encoding the pro-sequence results in lower expression of sigE and the whole 6E regulon, including spoIIID. Thus, one function of making OK first as an inactive precursor is to avoid initiating the negative feedback loop prematurely. Pro-0‘K appears to be kept inactive by at least two different mechanisms. First, the pro-sequence inhibits the RNAP core binding activity of CK. This is supported by the K observation that pro-6 in extracts of sporulating cells is not associated with core RNAP and the observation that pro-(5K fails to bind to core RNAP in vitro under conditions that permit OK binding. A simple model is that the pro-sequence interacts with and masks the core-binding domain of OK. If this is true, using the pro-(5K with mutations in the pro- sequence that abolish membrane-association but retain inhibition of RNAP core-binding (discussed below), intragenic suppressor mutations in the core-binding region of OK could be identified that relieve the inhibitory effect of the pro-sequence. A second mechanism of inactivation involves the tethering of pro-(3'K to the membrane surrounding the mother cell 95 and the membrane surrounding the forespore, physically sequestering it from core RNAP in the mother cell cytoplasm. Membrane-association is apparently mediated by the pro- sequence since processing releases 0'" to the cytoplasm of mother cell where it associates with core RNAP. It is not clear how pro-0’K interacts with the membrane. Residues in the pro-sequence critical for membrane-association and core-binding could be identified by a systematic mutagenesis analysis of the 20 amino acid residues in the pro-sequence, and/or by making a set of nested deletions of the pro-sequence. Testing these mutants for core- binding and membrane association could yield information about the mechanism by which the pro-sequence promotes membrane-association and inhibits core-binding. Since the pro-(3K processing machinery appears to be localized to the membrane surrounding the forespore (l 15), association of pro-(SK with this membrane may facilitate processing by bringing pro-(rK and the processing protease in proximity. During the course of subcellular fractionation studies, various fractions, including the membrane fraction, were incubated under different conditions to test for processing activity. No in K vitro processing of pro-o was observed. There is evidence that SpoIVFB is a labile protein, which could explain the failure to observe in vitro processing. In wild type cells, SpoIVF B is stabilized and inhibited by SpoIVFA. Relieve of the inhibitory effect of SpoIVF A depends on a forespore signal protein SpoIVB. Certain mutations in SpoIVFA (bofB) render SpoIVF B active independent of SpoIVB. In spoIIID mutant cells, 0'" is not produced but SpoIVFB overaccumulates IO-fold (115). Further attempts to develop an in vitro processing system may involve mixing the membrane fraction of cells overexpressing spoIVFB (e. g. bofB spoIIID double mutant cells) with pro-(rK (purified or in a membrane fraction fi'om processing deficient cells) and monitoring the production of OK. APPENDIX Appendix Bacillus subtilis SpoIIID protein binds to two sites in the spa VD promoter and represses transcription by GE RNA polymerase Reprinted from the Journal of Bacteriology 96 JOURNAL or BACTERIOLOGY, Feb. 1997. p. 972—975 0021 -9193/97/504.00 +0 Copyright 0 1997, American Society for Microbiology 97 Vol. 179, N0. 3 Bacillus subtilis SpoIIID Protein Binds to Two Sites in the spoVD Promoter and Represses Transcription by oE RNA Polymerase ' BIN ZHANG,‘ RICHARD A. DANIEL,2 JEFFERY ERRINGTON,2 AND LEE KROOS" Department of Biochemistry. Michigan State University, East Lansing, Michigan 48824,‘ and Sir Mlliam Dunn School of Pathology, University of Oxford, Oxford 0X1 3R5, United Kingdom2 Received 26 September 1996/Accepted 15 November 1996 The Bacillus subtilis spoVD gene encodes a penicillin-binding protein required for spore morphogenesis. SpoIIID is a sequence-specific DNA-binding protein that activates or represses the transcription of many dilerent genes. We have defined the spoVD promoter region and demonstrated that it is recognized by 0‘ RNA polymerase in vitro and that SpoIIID represses spoVD transcription. 'IVvo strong SpoIIID-binding sites were mapped in the spoVD promoter region, one overlapping the -35 region and the other encompassing the -10 region and the transcriptional start site. In response to starvation, Bacillus subtilis initiates a devel- opmental process involving a series of morphological changes driven by a program of gene expression (6). One of the early morphological changes is the formation of an asymmetric sep- tum that divides the bacterium into two compartments, the mother cell and the forespore. Subsequent migration of the septum engulfs the forespore in a double membrane, pinching it off as a free protoplast within the mother cell. Cell wall-like material known as the cortex is then deposited between the two membranes. Finally. the mother cell synthesizes so-called coat proteins that assemble on the surface of the forespore and the mature spore is released by lysis of the mother cell. The pro- gram of gene expression driving these morphological changes involves a cascade of sigma factor activity (9, 18). In the mother cell, the products of genes transcribed by oE RNA polymerase (505) are primarily responsible for engulfment, cortex forma- tion, and production of 0", while the products of genes tran- scribed by 0" RNA polymerase (50") function mainly in for- mation of the spore coat, as well as mother cell lysis (6, 9). Another key regulator of mother cell gene expression is SpoIIID, a sequence-specific DNA-bindingfiprotein that acti- vates or represses many genes in both the a and or" regulons (8, 14). Previous genetic studies sugested that the spoVD gene is likely to be a member of the o regulon and showed that spoVD is overexpressed in spolllD mutant cells (3). The prod- uct of spoVD is a penicillin-binding protein. Penicillin-binding proteins generally cany out the final steps of the synthesis of peptidoglycan, which is the major component of the cortex. Consistent with a proposed role of SpoVD in cortex synthesis, spoVD mutant cells are defective in spore cortex development (1, 3). Here we show that Errl5 can transcribe spoVD in vitro and that SpoIIID can repress spoVD transcription by binding to the promoter region. spoVD is transcribed by Err" in vitro, and SpoIIID represses this transcription. Several lines of evidence suggest that spoVD is transcribed by 505 (3). First, expression of a spoVD-lacZ fusion was reduced or abolished in strains defective in trE production. Second, spoVD-lacZ was expressed when tran- scription of the spoIIG operon encoding pro-(7'3 and its puta- tive processing enzyme. SpoIIGA, was induced in vegetative ‘ Corresponding author. Phone: (517) 355-9726. Fax: (517) 353- 9334. E-mail: kroos@pilot.msu.edu. 972 cells. Third, the spoVD promoter region shows sequence sim- ilarity to other (re-dependent promoters. The upstream bound- ary of the spoVD promoter region was defined by deletion analysis with exonuclease III-SI nuclease digestion and the integrational plasmid vector pSGI301 (3). Plasmids with in- serts extending from a range of positions in the vicinity of the transcriptional start site (TSS) to a unique HindIII site in the middle of the spoVD coding region were integrated into the chromosome of Spo+ strain SG38 (7). Inserts with their up- stream ends located up to and including -28 bp relative to the TSS gave a Spo' phenotype upon integration, indicating that promoter function was absent, whereas inserts extending to -42 bp or farther upstream were Spo‘, indicating at least partial promoter activity. The minimal sequences needed for spoVD expression thus lie within 42 bp of the TSS. To deter- mine whether 505 could recognize this promoter in vitro. a plasmid extending well upstream of the promoter (to -211) and to position +103 downstream (pSGl362; reference 3) was used as the template in runofl' transcription assays. The plas- mid was linearized by digestion with restriction enzymes that cleave at difierent sites downstream of the spoVD promoter, generating DNA templates that would produce spoVD runofl transcripts of diflerent lengths. In vitro transcription reactions were performed as described previously (8). Runoff transcripts were electrophoresed in a 5% polyacrylamide gel containing 8 M urea and detected by autoradiography. Partially purified ErrE produced runoii’ transcripts of the expected sizes (Fig. 1, lanes 2 and 3). Some promoters are transcribed by both Eo'a and E10" (8, 16, 24), presumably because both forms of RNA polymerase recognize some similar sequences in promoters (5). To demonstrate that 0'5 can direct transcription from the spoVD promoter and to test whether o" can too, gel-purified (IE and 0'" were reconstituted with core RNA polymerase as described previously (14 and used to transcribe a spoVD tem- plate. Only EoE, not , produced a runoff transcript (Fig. 1, lanes 5 and 6). A control experiment showed that the recon- stituted Err" was active because a strong signal was produced from a template containing the oKodependent cotD promoter (lane 7). We concluded that spoVD is transcribed by Bee. Since 1305 is active only in the mother cell compartment during sporulation (4, 10. 17). expression of spoVD is expected to be mother cell specific. This is consistent with the results of a tht. 179. 1997 N I I n vrtro transcrrphon ol Iprrl’l) by 1-11' and cllccl of Spollll) {1501113 - pg). cont .IIIn mg the tpul'l) promote-1(3) wa drgext d writ IIuRl (lanes 1 and. 3. 117 base tranxcrrpt) or \Iml (lancxl .-11nd 1511 bne 11 anurlptl and Ir. Ilnrcrrhcd In a total volume 11143 1.11 IIIIth Iz‘rr' (t). ‘ 11g) alone (1 IIIc.\ ’ .IIId 3)orwh01 tag of gel- purIlIeda SpoIIID added (lane: 1 and 4) kpStiI (13 (3 111:) wan dlglest 'dwith 1er )‘RI ndtr scnbcd wrlrretun uet “(lane .5)ormrh recormrtuted Fxr' (lane (1) pLRKltll (2 pg) contarnurng the 11:1!) promote er ( 14) was digested Wilh "null” (225- hp transcrrpt) and transcrrbed wrrh reconstituted Eak (lane 7). Arrowheads dcnot )1: the posrtrons ot ru null transu lpl\ oi the expected sues. as judgedr tr om the mrgratron of end— labeled DNA fragment) of Mull-digested pBRSJZ genetic experiment showing that expression of spoVD only in the mother cell is sufficient to allow sporulation 3). The ability to transcribe spoVD in vitro with Ea allowed us to test whether SpoIIID can directly affect spoVD transcnp- tion. The twofold overexpression of spoVD-lacZ in spoIIID mutant cells suggested that SpoIIID might repress transcrip- 98 NO'l'l-ZS 071 tion of ,1/IrIl I) (.3). AddItIon of Spollll) III the Irl \‘Itr‘o tran~ scrIptron rcactlon abohshcd the April I) signal (151g. 1. lanes 1 ’1 d 4). This was not due to general inhibition of transcription by Spollll). because the same preparation of Spollll) acti- vated transcription of .YIgK by 1:111 (data not shown). as re- ported pretrorrsly (8). We concluded that SpoIIID can specif- icalb reprexs 1,1011) tranxcriptlon b_\ Ii 11" In \rtro lhe .s/IIIIIII) Irene Is transcrrbcd b_\ Fu' (13 15 33. 3.3). yet SpoIIID represses transcription of spot 'I) (Fig. 1) and some other rr'ldcpcndent genes. Thus, Spollll) limits the expres- sion of some genes In Its own regulon. Other genes In the Ir" regulon that are repressed by SpoIIID include hotel. as dem- onstrated in vivo (13) and III vitro (8). and probably the .tprII/le operon. as implied by III vivo data (11). Another possible SpoIIID-repressed operon In the 11' regulon Is slur/l 7*. based on the observation that a spoil 'F-Iur'l lusron Is overexpressed in spa/III) mutant cells (3) and the fact that three near-perfect matches to the SpoIIID binding site consensus sequenre (S) can be found In the spoIVF promoter Perhaps SpoIIID helps confine the expression of some genes in the (IL regulon to II physiologically relevant level or time period. Consistent with this notion, in spoIlID mutant cells. the products of spoIVF overaccumulatc and SpoIVFA is mislocalized (30). SpoIIID binds to two sites in the spoVD promoter region. Gel retardation assays Indicated that SpoIIID binds to DNA fragments containing the spoVD promoter region (data not shown). To localize the binding sites of SpoIIID In this pro- moter more precisely. DNase I protection experiments were performed as described prevIously (8). Radioactive DNA probes. labeled separately on the nontranscribed or tranv 1 2 3 4 .33) ‘ " -21> g -1 2) I i, l' #7) . . .. 1 FIG 3 SpoIIID lootpnnrs In the 1,141! I) promoter rcgmn Radroaetrvc DNA It. rgrnc nrs \cpar. Itclv end labeled on the nontransurhed (A) or 11. Inscrrbcd (11) strand were lnt‘ul‘lillcd In \epm ratt reactrrlms wrth a carrier pr o Itcrn (brwrnc \IIrInI albunrrn. 31 1110111111 (lane 1).»! Wllh 11025 (1. Inc 3.) lIl11< (I Inc ‘1) ti I 1.c1Ir 4) I" II ‘ (I. ..IL 5 .. In lu- ugv tar-re pIrIILIIIandIbcnI ijcttc tc)d1ol NI of protcctmn by SpoIIID and numb. Is to the left ruler to meIIonx rel. one 111 m ..l 4112 ‘llIL‘ ht tllll’l l1.ltl|l\ ”11' 155(1) Stan Indrtate enhanced thav 11,1. lul UV. IN: lIIpoIr SpoIIID brndrng 974 NOTES -35 -1o “282‘! - 1 3- 1 5 bp- CATACA-‘l e 1 r’ qaqcatcq‘l'tcflcctqtccaaatteaquA‘rAaAaf aaacaa ecu apoVD _OIOOU manna—r Consonant -20 W -20 - 5 announce +5 FIG. 3. Position of SpoIIID binding sites In the spoVD promoter region and alignment of sequences within these sites with the consensus sequence for SpoIIID binding. (A) The nucleotide sequence of the spoVD promoter region is aligned with respect to conserved nucleotides found In the - 10 and -35 regions of promoters transcribed by [311'- (21 ). shown at the top (It means G or T. and 111 means A or C). Nucleotides in the spoVD promoter that match the consensus are shown as boldface capital letters. Overlining and underlining indicate regions on the nontranscribcd and transcribed strands. respectively. protected by SpoIIID from digestion with DNase I (Fig. 2). The dashed portion of the line indicates a region of uncertain protection due to a lack of DNase I digestion in this region. (B) Nucleotide sequences within the SpoIIID-protected regions of the spoVD promoter are aligned with respect to the consensus sequence for SpoIIID bind- ing (8). shown at the top (W means A or T. R means purine. and Y means pyrimidine). Numbers refer to positions relative to the TSS. Note that the sequence shown for the binding site between —20 and —29 is from the strand opposite that shown in panel A. Nucleotides that match the consensus are shown as boldface capital letters scribed strand, were prepared as follows. pSGl362 (3) was digested with Hindi". which cleaves in the multiple cloning site downstream of the spoVD promoter. and labeled either at the 3' end by the Klenow enzyme fill-in reaction and [a-JzPldATP or at the 5' end by treatment with alkaline phoso phatase followed by T4 polynucleotide kinase and [y-“PlATP. In both cases, the labeled DNA was digested with Sacl. which cleaves upstream of the spoVD promoter, to produce a 340-bp fragment that was purified by elution from an 8% polyacryl- amide gel. The purified DNA fragments were incubated with difl'erent amounts of SpoIIID and then mildly digested with DNase 1. After DNase 1 treatment, the partially digested DNAs were electrophoresed in 6% polyacrylamide gels con. taining 8 M urea alongside a sequencing ladder generated by subjecting the apprOpriate end-labeled DNA to the chemical cleavage reactions of Maxam and Gilbert as described previ- ously (19). Figure 2 shows that SpoIIID protected two regions of the spoVD promoter from DNase I digestion. Site 1 spanned positions -33 to -15 on the nontranscribed strand (Fig. 2A) and -33 to -21 on the transcribed strand (Fig. 28). while protection in site 2 spanned positions —9 to +8 on the non- transcribed strand (Fig. 2A) and - 12 to +7 on the transcribed strand (Fig. ZB). The results of DNase I protection experi- ments are summarized in Fig. 3A. Greater than half-maximal protection of both sites 1 and 2 was achieved at a 50 nM concentration of SpoIIID (Fig. 2A and B, lanes 2). putting the two sites into the group of strong SpoIIID binding sites along with sites 1 and 2 of bofA. site 1 of spoIVCA. sites 2 and 3 of €010, and site 2 of sigK (8). Since we estimated from Western blot experiments that the SpoIIID concentration reaches the 1 11M range in sporulating cells (data not shown), it seems likely that SpoIIID occupies sites 1 and 2 in the spoVD promoter in vivo and accounts for the twofold lower expression of spoVD- lacZ in wild-type cells than in SpoIIID mutant cells (3). By aligning the nucleotide sequences of known SpoIIID binding sites. a consensus sequence. WWRRACAR-Y (where W is A or T. R is purine. and Y is pyrimidine), was found (8). Spa] 11D binding site 1 in the spoVD promoter region contains 99 .l. BM'I‘IRIOI. a perfect match to the consensus sequence. while binding site 2 contains a sequence with just one mismatch (Fig. 38). While many strong SpoIIID binding sites exhibit a second good match to the consensus sequence in inverted orientation rela- tive to the best match (8). no such second match is found in either of the two protected regions in the spoVD promoter. Of the two SpoIIID binding sites mapped in the spoVD promoter region. one overlaps the -35 region and the other encompasses the —10 region and the TSS. Hence. SpoIIID binding to either site is likely to interfere with promoter rec- ognition by EOE, explaining the strong repressive effect of SpoIIID on spoVD transcription in vitro (Fig. l). A previous study showed that SpoIIID binding in the ~35 region of the corD promoter is sufficient to mediate repression (8). Also. the SpoIIID-repressed bofA promoter has a SpoIIID-binding site encompassing the -10 region and the TSS. although It was not determined whether a second SpoIIID-binding site located at +54 to +74 is needed for repression (8). Mutational analysis of the spoVD promoter region should permit dissection of whether the two SpoIIID binding sites function in an additive. cooperative. or redundant fashion to repress transcription by Bo . We thank C. Moran for sending gartially purified Bee and R. Hal- berg for advice on purification of a , 0". and SpoIIID. This work was supported by NIH grant 43585 to LK. and by the Michigan Agricultural Experimentation Station. Work in the Er- rington laboratory was supported by the Biotechnology and Biological Sciences Research Council of the UK. was I. Coote, J. G. 1971 Sporulation in Bacillus subtilis. Characterization of oligo- sporogcnous mutants and comparison of their phenotypes with those of asporogenous mutants. J. Gen. Microbiol. 71:1—15. . Cutting. 8.. S. Rods. and R. hostel 1991. Sporulation operon spoIVF and the characterization of mutations that uncouplc mother-cell from forespore gene expression In Bacillus subri'lis. J. Mol. Biol. 211:1237-1256. . Danlel. R. A” S. Drake. C. E. Birch-an. R. Scholle. and J. Errington. 1994. The Bacillus subtilis spoVD gene encodes a mother-ccll-specific penicillin- binding protein required for spore morphogenesis. .1. Mol. Biol. 235209—220. . Drllta. A.. and R. [mic-It. 1991. Compartmentalized expression of a gene under the control of sporulation transcription factor 05 in Bacillus subtilis. Proc. Natl. Acad. Sci. USA 88:9934—9938. . Errington, J. 1991. Possible intermediate steps in the evolution of a pro- karyotic developmental system. Proc. R. Soc. Lond. B 244:117-121. . Errington. J. 1993. Bacillus subtilis spomlation: regulation of gene expression and control of morphogenesis. Microbiol. Rev. 57:1—33. . MOI, J.. and J. Mammal. 1986. Use of a lacZ gene fusion to determine the dependence pattern of sporulation operon spa!“ in spa mutants of Bacillus subtilis. J. Gen. Microbiol. [32:2967—2976. . Halberg, R. and L Kroos. 1994. Sporulation regulatory protein SpoIIID from Bacillus subtilis activates and represses transcription by both mother- cell-specific forms of RNA polymerase. J. Mol. Biol. 10:425-436. Raider-nag, W. G. 1995. The {Ulla factors of Bacillus subtilis. Microbiol. Rev. 59:1-30. . Harry. E. J., IL Pogliano. and R. latch. 1995. Use of immunofluorescencc to visualize cell-specific gene expreuion during sporulation in Bacillus sub- u‘lu. J. Bacteriol. 177:3386-3393. . [Ill-g. N.. and J. Errington. 1991. The spoIIIA operon of Bacillus subtilis defines a new temporal class of nurther-cell-specifrc sporulation genes under the control of the 05 form of RNA polymerase. Mol. Microbiol. 5:192?— 1940. . lrcton. IL. and A. Grossman. 1992. Interaction among mutations that cause altered timing of gene expression during sporulation in Bacillus sublilu. J. Bacteriol. 174:3185-3195. . Jones. C. II. K. M. Tattl. and C. P. Moran. 1992. Effects of amino acid substitutions in the -10 binding region of 17‘i from Bacillus subtilis. .l. Bac- teriol. [14:6815-6821. . Kroos, L. 8. Intel, and R. Insist. 1989. Switch protein alters specificity of RNA polymerase containing a camartmcnt-spccifrc sigma factor. Science 243:526—529. . Kunltel. IL, L Kroos. H. Poth. P. Vngman. and R. Losick. 1989. Temporal and spatial control of the mother-cell regulatory gene spoIIID of Bacillui subtilis. Genes DCV. 3:1735-1744. .Kunkel. I. K. Sandman, S. Pam. Rhiannon. and R. Losick. I988. The promoter for a sporulation gene in the spoil/C locus of Bacillus .mlmln and 100 VOL. 179, 1997 l8. 19. its use in studies of temporal and spatial control of gene expression. J. Bacteriol. l70:35I3-3522. . Lewis. P. 1., C. E. Nwogull. M. R. Ram. C. J. flamed. and J. Errington. l9“, Use of digitized video microscopy with a fluorogenic enzyme substrate to demonstrate cell- and compartment-specific gene expression in Salmo- nella cmm’udi: and Bacillus subtilis. Mol. Microbiol. [33849—3853. Losick. IL. and I’. Stragler. I992. Crisscross regulation of cell-typospccific gene expression during development in B. :ubulls. Nature 355:601—004. Manlatia. 11. E. F. Frltach, and J. Sambrook. I982. Molecular cloning: a laboratory manual. Cold Spring Harbor Laboratory. Cold Spring Harbor. NY. . Resnekov, 0.. S. Alper. and II. Loalele 1996. Subcellular localization of proteins governing the proteolytic activation of a developmental transcrip- 22. 23. 24. NOTES 975 tion {actor in Bacillus subtilis. Genes Cells 12529—542. . Roels. S.. A. Drllts. and R. Losick. I‘WI. Characterization of xpnll'rl. a sporulation gene involved in coat morphogenesis in Bacillus subtilis. J. Bac- teriol. [74:575—585. Steven, C. M., and J. Errington. 1990, Differential gene expresaion during sporulation in Bacillus subtilis: structure and regulation of the spollll) gene. Mol. Microbiol. 4543—552. Tattl. K. M.. C. ll. Jones, and C. P. Moran. 199I. Genetic evidence [or interaction of 05 with the spoIllD promoter in Bacillus subtilis. J. Bacteriol. 173:7828—7833. Zheng. L. and R. Losick. I990. Cascade regulation of spore coat gene expression in Bacillus suluilis. J. Mol. Biol. 212:645-600. BIBLIOGRAPHY Bibliography 1. Aizawa, S. 1996. Flagellar assembly in Salmonella typhimurium. Mol. Microbiol. 1921-5. 2. Alex, L. A. and M. 1. Simon. 1994. Protein histidine kinases and signal transduction in prokaryotes and eukaryotes. Trends. Genet. 10:133-138. 3. Alper, S., L. Duncan, and R. Losick. 1994. An adenosine nucleotide switch controlling the activity of a cell type-specific transcription factor in B. subtilis. Cell 77 2195-205. 4. Anderson, M., Y. T. Yu, and L. Kroos. Unpublished results. 5. Arigoni, F., K. Pogliano, C. D. Webb, P. Stragier, and R. Losick. 1995. Localization of protein implicated in establishment of cell type to sites of asymmetric division. Science 270:637-640. 6. Ala-Blane, P., F. A. Ramirez-Weber, M. P. Laget, C. Schwartz, and T. B. Kornberg. 1997. Proteolysis that is inhibited by Hedgehog targets Cubitus interruptus protein to the nucleus and convert it to a repressor. Cell 89: 1043-1053. 7. Bagyan, 1., J. Hobot, and S. Cutting. 1996. A compartmentalized regulator of developmental gene expression in Bacillus subtilis. J. Bacteriol. 178:4500-4507. 8. Bai, U., I. Mandie-Mulec, and 1. Smith. 1993. SinI modulates the activity of SinR, a developmental switch protein of Bacillus subtilis. Genes Dev. 7:139-148. 9. Baldus, J. M., C. M. Buckner, and C. P. Moran. 1995. Evidence that the transcriptional activator SpoOA interacts with two sigma factors in Bacillus subtilis. M01. Microbiol. 17:281-290. 10. Baldus, J. M., B. D. Green, P. Youngman, and C. P. J. Moran. 1994. Phosphorylation of Bacillus subtilis transcription factor SpoOA stimulates transcription from the spoIIG promoter by enhancing binding to weak 0A boxes. J. Bacteriol. 176:296-306. 1 1. Barak, 1., J. Behari, G. Olmedo, P. Guzman, D. P. Brown, and P. Youngman. 1996. Structure and function of the Bacillus subtilis SpoIIE protein and its localization at the sites of septum assembly. Mol. Microbiol. 19:1047-1060. 12. Bird, T. H., J. K. Grimsley, J. A. Hoch, and G. B. Spiegelman. 1996. The Bacillus subtilis response regulator SpoOA stimulates transcription of the spoIIG operon through modification of RNA polymerase promoter complex. J. M01. Biol. 101 102 256:436-448. 13. Bradford, M. 1976. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 72:248-254. 14. Bramucci, M. G., B. D. Green, N. Ambulos, and P. Youngman. 1995. Identification of a Bacillus subtilis spoOH allele that is necessary for suppression of the sporulation-defective phenotype of a spoOA mutation. J. Bacteriol. 177: 1 630-1633. 15. Bramueei, M. G., B. D. Green, N. Ambulos, and P. Youngman. 1995. Identification of a Bacillus subtilis allele that is necessary for suppression of the sporulation phenotype of a SpoOA mutation. J. Bacteriol. 177:1630-1633. l6. Burbulys, D., K. A. Traeh, and J. A. Hoch. 1991. Initiation of sporulation in B. subtilis is controlled by a multicomponent phosphorelay. Cell 64:545- 552. 17. Cutting, S., A. Driks, R. Schmidt, B. Kunkel, and R. Losick. 1991. Forespore-specific transcription of a gene in the signal transduction pathway that governs pro-0'K processing in Bacillus subtilis. Genes Dev. 5:456-466. 18. Cutting, S., V. Oke, A. Driks, R. Losick, S. Lu, and L. Kroos. 1990. A forespore checkpoint for mother-cell gene expression during development in Bacillus subtilis. Cell 62:239-250. 19. Cutting, S., S. Roels, and R. Losick. 1991. Sporulation operon spoI VF and the characterization of mutations that uncouplc mother-cell from forespore gene expression in Bacillus subtilis. J. M01. Biol. 221 : 1237-1256. 20. Diederich, B., J. F. Wilkinson, T. Magnin , S. M. A. Najafi, J. Errington, and M. D. Yudkin. 1994. Role of interactions between SpoIIAA and SpoIIAB in regulating cell-specific transcription factor (SF of Bacillus subtilis. Genes Dev. 82653-2663. 21. Dombroski, A. J., W. A. Walter, and C. A. Gross. 1993. Amino- terminal amino acids modulate o-factor DNA-binding activity. Genes & Dev. 7:2446- 2455. 22. Dombroski, A. J., W. A. Walter, M. T. Record, D. A. Siegele, and C. A. Gross. 1992. Polypeptides containing highly conserved regions of transcription initiation factor 0'70 exhibit specificity of binding to promoter DNA. Cell 70:501-512. 23. Driks, A. and R. Losick. 1991. Compartmentalized expression of a gene under the control of sporulation transcription factor 05 in Bacillus subtilis. Proc. Natl. Acad. Sci. USA 88:9934-9938. 24. Driks, A., S. Roels, B. Beall, C. P. Moran, and R. Losick. 1994. Subcellular localization of proteins involved in the assembly of the spore coat of Bacillus subtilis. Genes & Dev. 8:234—244. 25. Duncan, L., S. Alper, F. Arigoni, R. Losick, and P. Stragier. 1995. 103 Activation of cell-specific transcription by a serine phosphatase at the site of asymmetric division. Science 270:641-644. 26. Duncan, L. and R. Losick. 1993. SpoIIAB is an anti-0' factor that binds to and inhibits transcription by regulatory protein (SF from Bacillus subtilis. Proc. Natl. Acad. Sci. USA 90:2325-29. 27. Errington, J. 1993. Bacillus subtilis sporulation: regulation of gene expression and control of morphogenesis. Microbiol. Rev. 57:1-33. 28. Freese, E., J. E. Heinze, and E. M. Gelliers. 1979. Partial purine deprivation causes sporulation of Bacillus subtilis in the presence of excess ammonia, glucose, and phosphate. J. Gen. Microbiol. 115: 193-205. 29. Fuqua, C., S. C. Winan, and E. P. Greenberg. 1996. Census and consensus in bacterial ecosystem: the LuxR-Luxl family of quorum-sensing transcriptional regulators. Ann. Rev. Microbiol. 50:527-551. 30. Gaur, N. K., J. Oppenheim, and 1. Smith. 1991. The Bacillus subtilis sin gene, a regulator of alternate developmental processes, codes for a DNA-binding protein. J. Bacteriol. 173:678-686. 31. Gholamhoseinian, A. and P. J. Piggot. 1989. Timing of spa]! gene expression relative to septum formation during sporulation of Bacillus subtilis. J. Bacteriol. 171:5747-5749. 32. Gomez, M., S. Cutting, and P. Stragier. 1995. Transcription of spa] VB is the only role of 0'0 that is essential for pro-(3'K processing during spore formation in Bacillus subtilis. J. Bacteriol. 177:4825-4827. 33. Gorham, H. C., S. J. McGowan, P. R. Robson, and D. A. Hodgson. 1996. Li ght-induced carotenogenesis in Myxococcus xanthus: light-dependent membrane sequestration of ECF sigma factor CarQ by anti-sigma factor CarR. Mol. Microbiol. 19:171-186. 34. Grossman, A. 1995. Genetic networks controlling the initiation of sporulation and the development of genetic competence in Bacillus subtilis. Ann. Rev. Genetics 29:47 7-508. 35. Grossman, A. D., T. Lewis, N. Levin, and R. DeVivo. 1992. Suppressors of a spoOA missense mutation and their effects on sporulation in Bacillus subtilis. Biochimie 74:679-688. 36. Guzman, P., J. Westpheling, and P. Youngman. 1988. Characterization of the promoter region of the Bacillus subtilis spoIIE operon. J. Bacteriol. 170: 1598- 1609. 37. Halberg, R. 1994. Ph.D. Thesis. Michigan State University, East Lansing. 38. Halberg, R. and L. Kroos. 1992. Fate of the SpoIIID switch protein during Bacillus subtilis sporulation depends on the mother-cell sigma factor, 0K. J. Mol. Biol. 228:840-849. 104 39. Halberg, R. and L. Kroos. 1994. Sporulation regulatory protein SpoIIID from Bacillus subtilis activates and represses transcription by both mother-cell-specific forms of RNA polymerase. J. Mol. Biol. 243:425-436. 40. Halberg, R., V. Oke, and L. Kroos. 1995. Effects of Bacillus subtilis sporulation regulatory protein SpoIIID on transcription by GK RNA polymerase in vivo and in vitro. J. Bacteriol. 177:1888-1891. 4]. Haldenwang, W. G. 1995. The sigma factors of Bacillus subtilis. Microbiol. Rev. 5921-30. 42. Harry, E. J., K. Pogliano, and R. Losick. 1995. Use of immunofluorescence to visualize cell-specific gene expression during sporulation in Bacillus subtilis. J. Bacteriol. 177:3386-3393. 43. Harwood, C. R. and S. M. Cutting. 1990. Molecular biological methods for Bacillus. John Wiley & Sons, Chichester, England. 44. Healy, J., J. Weir, 1. Smith, and R. Losick. 1991. Post-transcriptional control of a sporulation regulatory gene encoding transcription factor 0'" in Bacillus subtilis. Mol. Microbiol. 5:477-487. 45. Hicks, K. A. and A. D. Grossman. 1996. Altering the level and regulation of the major sigma subunit of RNA polymerase affects gene expression and development in Bacillus subtilis. M01. Miocrobiol. 20:201-212. 46. Hoeh, J. 1994. The Phosphorelay Signal Transduction Pathway in the Initiation of Sporulation, p. 41-60. In Piggot, P.J., C.P. Moran and P. Youngman (ed.), Regulation of Bacterial Differentiation. American Society for Microbiology, Washington, DC. 47. Hofmeister, A. 1997. Activation of the proprotein transcription factor pro-(1'E is associated with three changes in its subcellular localization during sporulation in Bacillus subtilis. J. Bacteriol. submitted: 48. Hofmeister, A., A. Londono-Vallejo, E. Harry, P. Stragier, and R. Losick. 1995. Extracellular signal protein triggering the proteolytic activation of a developmental transcription factor in B. subtilis. Cell 83:219-226. 49. Hughes, K. T., K. L. Gillen, M. J. Semon, and J. E. Karlinsey. 1993. Sensing structural intermediates in bacterial flagellar assembly by export of a negative regulator. Science 262: 127 7- 1 280. 50. Ichikawa, H. and L. Kroos. Unpublished data. 51. Ireton, K. and A. Grossman. 1992. Interaction among mutations that cause altered timing of gene expression during sporulation in Bacillus subtilis. J. Bacteriol. 174:3185-3195. 52. Ireton, K. and A. D. Grossman. 1992. Coupling between gene expression and DNA synthesis early during development in Bacillus subtilis. Proc. Natl. Acad. Sci. USA 89:8808-8812. 105 53. Ireton, K. and A. D. Grossman. 1994. A developmental checkpoint couples the initiation of sporulation to DNA replication in Bacillus subtilis. EMBO J. 13:1566- 1573. 54. Ireton, K., N. W. Gunther, and A. D. Grossman. 1994. spoOJ is required for normal chromosome segregation as well as the initiation of sporulation in Bacillus subtilis. J. Bacteriol. 176:5320-5329. 55. Ireton, K., D. Z. Rudner, K. J. Siranosian, and A. D. Grossman. 1993. Integration of multiple developmental signals in Bacillus subtilis through the spoOA transcription factor. Genes Dev. 7 :283-294. 56. Ireton, K., D. Z. Rudner, K. J. Siranosian, and A. D. Grossman. 1995. Krebs cycle function is required for activation of the SpoOA transcription factor. Proc. Natl. Acad. Sci. USA 92:2845-2849. 57. Jin, S. and A. L. Sonenshein. 1994. Transcritional regulation of Bacillus subtilis citrate synthase genes. J. Bacteriol. 176:4680-4690. 58. Johnson, B. and A. Dombroski. 1997. The role of the pro-sequence of Bacillus subtilis UK in controlling activity in transcription initiation. J. Biol. Chem. in press: 59. Jonas, R. M., E. A. Weaver, T. J. Kenney, C. P. Moran, and W. G. Haldenwang. 1988. The Bacillus subtilis spoIIG operon encodes both OE and a gene necessary for 0’5 activation. J. Bacteriol. 1702507-511. 60. Ju, J., T. Luo, and W. Haldenwang. 1997. Bacillus subtilis pro-0'E fusion protein localizes to the forespore septum and fails to be processed when synthesized in the forespore. J. Bacteriol. 179:4888-4893. 61. Karow, M. L., P. Glaser, and P. Piggot. 1995. Identification of a gene, spoIIR, that links the activation of 0’5 to the transcriptional activity of CF during sporulation in Bacillus subtilis. Proc. Natl. Acad. Sci. USA 92:2012-2016. 62. Katsukake. 1994. Excretion of the anti-sigma factor through a flagella structure couples flagella gene expression with flagella assembly in Salmonella typhymurium. Mol. Gen. Genet. 243:605-612. 63. Kellner, E. M., A. Decatur, and C. P. Moran. 1996. Two-stage regulation of an anti-sigma factor determines developmental fate during bacterial endospore formation. Mol. Microbiol. 21:913-924. 64. Kenney, T. J. and C. P. Moran. 1987. Organization and regulation of an operon that encodes a sporulation-essential sigma factor in Bacillus subtilis. J. Bacteriol. 169:3329-3339. 65. Kenney, T. J., K. York, P. Youngman, and C. P. Moran. 1989. Genetic evidence that RNA polymerase associated with GA uses a sporulation-specific promoter in Bacillus subtilis. Proc. Natl. Acad. Sci. USA 86:9109-9113. l06 66. Kirchman, P. A., H. DeGrazia, E. M. Kellner, and J. C.P. Moran. 1993. Forespore-specific disappearance of the sigma-factor antagonist SpoIIAB: implications for its role in determination of cell fate in Bacillus subtilis. Mol. Microbiol. 8:663-671. 67. Kobayashi, K., K. Shoji, T. Shimizu, K. Nakano, T. Sato, and Y. Kobayashi. 1995. Analysis of a suppressor mutation ssb (kinC) of sur0B20 (SpoOA) mutation in Bacillus subtilis reveals that kinC encodes a histidine protein kinase. J. Bacteriol. 177:176-182. 68. Kroos, L. and S. Cutting. 1994. Intercellular and intercompartmental communication during Bacillus subtilis sporulation, p. 155-180. In Piggot, P.J., C.P. Moran and P. Youngman (ed.), Regulation of Bacterial Differentiation. Amer. Soc. for Microbiol., Washington, DC. 69. Kroos, L., B. Kunkel, and R. Losick. 1989. Switch protein alters specificity of RNA polymerase containing a compartment-specific sigma factor. Science 243:526-529. 70. Kunkel, B., L. Kroos, H. Poth, P. Youngman, and R. Losick. 1989. Temporal and spatial control of the mother-cell regulatory gene spoIlID of Bacillus subtilis. Genes Dev. 3:1735-1744. 71. Kunkel, B., R. Losick, and P. Stragier. 1990. The Bacillus subtilis gene for the developmental transcription factor 6K is generated by excision of a dispensable DNA element containing a sporulation recombinase gene. Genes Dev. 4:525-535. 72. Kunkel, B., K. Sandman, S. Panzer, P. Youngman, and R. Losick. 1988. The promoter for a sporulation gene in the spolVC locus of Bacillus subtilis and its use in studies of temporal and spatial control of gene expression. J. Bacteriol. 17023513- 3522. 73. Kutsukake, K., Y. Ohya, and T. lino. 1990. Transcriptional analysis of the flagellar regulon of Salmonella typhimurium. J. Bacteriol. 172:741-747. 74. LaBell, T. L., J. E. Trempy, and W. G. Haldenwang. 1987. Sporulation-specific sigma factor, 629, of Bacillus subtilis is synthesized from a precursor protein, P31. Proc. Natl. Acad. Sci. USA 84:1784-1788. 75. Lazazzera, B. A., J. M. Solomon, and A. D. Grossman. 1997. An Exported peptide functions intracellularly to contribute to cell density signalling in B. subtilis. Cell 89:917-925. 76. LeDeaux, J. R. and A. D. Grossman. 1995. Isolation and Characterization of kinC, a gene that encodes a sensor kinase homologous to the sporulation sensor kinases KinA and KinB in Bacillus subtilis. J. Bacteriol. 177:166-175. ‘77. LeDeaux, J. R., N. Yu, and A. D. Grossman. 1995. Different roles for KinA, KinB, and KinC in the initiation of sporulation. J. Bacteriol. 177:861-863. 78. Levin, P. A. and R. Losick. 1996. Transcription factor SpoOA switches the 1 ()calization of the cell division protein FtsZ from a medial to a bipolar pattern in Bacillus 107 subtilis. Genes Dev. 10:478-488. 79. Lewis, P. J., S. R. Partridge, and J. Errington. 1994. 0 factors, asymmetry, and the determination of cell fate in Bacillus subtilis. Proc. Natl. Acad. Sci. USA 91:3849-3853. 80. Lin, D. C. H., P. A. Levin, and A. D. Grossman. 1997. Bipolar localization of a chromosome partitioning protein in Bacillus subtilis. Proc. Natl. Acad. Sci. USA 94:4721-4726. 81. Linn, T. G., A. L. Greenleaf, R. G. Shorenstein, and R. Losick. 1973. Loss of the sigma activity of RNA polymerase of Bacillus subtilis during sporulation. Proc. Natl. Acad. Sci. USA 70:1865-1869. 82. Londono-Vallejo, J. A. 1997. Mutational analysis of the early forespore/mother-cell signalling pathway in Bacillus subtilis. Microbiology 143:2753- 2761. 83. Londono-Vallejo, J. A. and P. Stragier. 1995. Cell-cell signaling pathway activating a developmental transcription factor in Bacillus subtilis. Genes Dev. 9:503-508. 84. Lopez, J. M., A. Dromeriek, and E. Freese. 1981. Response of guanosine S'-triphosphate concentration to nutritional changes and its significance for Bacillus subtilis sporulation. J. Bacteriol. 146:605-613. 85. Lopez, J. M., C. L. marks, and E. Freese. 1979. The decrease of guanine nucleotides initiates sporulation of Bacillus subtilis. Biochem. Biophys. Acta. 587:238- 252. 86. Losick, R. and P. Stragier. 1992. Crisscross regulation of cell-type-specifrc gene expression during development in B. subtilis. Nature 355:601-604. 87. Losick, R., P. Youngman, and P. J. Piggot. 1986. Genetics of endospore formation in Bacillus subtilis. Ann. Rev. Genet. 20:625-669. 88. Lu, 8., S. Cutting, and L. Kroos. 1995. Sporulation protein SpoIVFB from Bacillus subtilis enhances processing of the sigma factor precursor pro-(1'K in the absence of other sporulation gene products. J. Bacteriol. 177 :1082-1085. 89. Lu, 8., R. Halberg, and L. Kroos. 1990. Processing of the mother-cell 0' factor, 0K, may depend on events occurring in the forespore during Bacillus subtilis development. Proc. Natl. Acad. Sci. USA 87:9722-9726. 90. Lu, S. and L. Kroos. 1994. Overproducing the Bacillus subtilis mother-cell sigma factor precursor, pro-6K, uncouples (SK-dependent gene expression fiom dependence on intercompartmental communication. J. Bacteriol. 176:3936-3943. 91. Mandic-Mulee, 1., L. Doukhan, and 1. Smith. 1995. The Bacillus subtilis SinR protein is a repressor of the key sporulation gene spoOA. J. Bacteriol. 177:4619-