s. :. (u ,1 ~ . #2.”... .- P L... 1.... 333‘}; o. A \ .t J- fig: {s (- :J . , . 3.x; .3 ”.00". .. mm . a. n. A .3 C nah” ‘ F‘mmmw 4 in“... , . .. if . ..¢...,...u§..1...u3€. . .. JP . 1.. .. it... OK? n .fiéxnmgmfiw . 3.3.1 . . l‘ . Luau L. 12...... .y: ..... < Wav. .5243: . . a. a» {L H z... - r. 1.. t: i u 5..” s. :5.» as "9:9 no: .Inasil .. V . : ,1 3395.! 3...in on... 0.: .. 2‘ at?) E!.fi§~azlt .l. 1 O2 )3 I!!! 3.1 u l i." .14 . aaflvfihflivt : 1.23.31. I . - v 1.... .. E 39:3 . {its . I: z. 1...». Ball. 3. .35.:i... a! l.u[3 I. .0. it. . thud“. avflfly‘MQE bmxflkpro an. i. .1 .3 why jun-s}... trawlunnui; ‘ -l ‘ .r.? 5..va ‘ (I..\1 L5,. E mulllll’flllfllllllmlllmmum 1810 3881 This is to certify that the dissertation entitled Gene Expression in Sinorhizobium Meliloti During Nutrient Deprivation presented by Mary Ellen Davey has been accepted towards fulfillment of the requirements for Doctor of Philosophy degreein Microbiology Major professor Dr. Frans J. de Bruijn Date August 20, 1999 MS U is an Affirmative Action/Equal Opportunity Institution 0-12771 v‘ five-44 LIBRARY Michigan State University PLACE IN RETURN BOX to remove this checkout from your record. TO AVOID FINE return on or before date due. MAY BE RECALLED with earlier due date if requested. DATE DUE DATE DUE DATE DUE use chIRC/DdoOupes—p.“ GENE EXPRESSION IN SINORHIZOBII /M MEL/1,077 DURING NUTRIENT DEPRIVATION By Mary Ellen Davey A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Microbiology 1999 To swtchhg Two mayo oxygen te BCOSysier for these i Physiolog; 9'0" "th. L eri‘tifOnmE Of mleObu With 3 THE IESUiting Ir ABSTRACT GENE EXPRESSION IN SINORHIZOBIUM MEL/1,077 DURING NUTRIENT DEPRIVATION B y Mary Ellen Davey To persist, bacteria must adapt dynamically to their environment by switching on or off different suites of genes in response to their surroundings. Two major parameters that bacteria constantly monitor are nutrient status and oxygen tension. In soil, these resources are often scarce. Nutrients enter this ecosystem intermittently, however the diverse bacterial populations compete for these nutrients and they are quickly utilized. Consequently, the natural physiological state of indigenous soil bacteria is either dormancy or negligible growth. Understanding how bacteria are able to monitor and respond to their environment in this nutrient-deprived state is fundamental to our understanding of microbial biology. By mutagenizing the genome of Sinorhizobium me/i/oti with a Tn5 derivative (Tn5/uxAB), which generates transcriptional fusions resulting in bacterial bioluminescence when the gene fusion is expressed, a gene was identified in S. meIi/oti that is induced by environmental parameters which are representative of life in soil; that is, nitrogen or carbon deprivation, low oxygen tension, as well as during post-exponential stationary-phase growth, and by osmotic stress. The tagged gene was found to be part of an operon consisting of two open reading frames (ORF), which were designated not/1 and deduced 3 I databases‘ appear to Z was lSOiaiE mutant stta mutant den expressron rucrterase it contained a deg'ee of s Hoodoo ore receptor or reporter ger coding reg: experiment: whose expr well as a hr environmer ndiA and ndiB for gutrient deprivation induced genes. Comparison of the deduced amino acid sequences of both mm and ndiB to the protein databases did not reveal similarities with any known genes; therefore, they appear to be novel. In addition, a gene involved in the regulation of this operon was isolated, by carrying out a second round of mutagenesis on the primary mutant strain 022 with a Tn3 derivative, Tn1721. A library of 3000 double mutant derivatives of 022 was screened for strains with altered luciferase expression patterns. One double mutant that failed to express the 022 luciferase fusion under any of the conditions tested was identified. This mutant contained a Tn1721 insertion in a gene which encodes a protein with a high degree of similarity to the tryptophan-rich sensory protein, TspO, from Rhodobacter sphaeroides, as well as to the mitochondrial benzodiazepine receptor, pK18. Furthermore, proper environmental control of the ndi-quAB reporter gene fusion was found to be restored after introduction of the tspO coding region in trans, under all inducing conditions tested. Thus, the experiments described here showed both the presence of a novel operon whose expression is induced by multiple environmental (stress) conditions, as well as a hitherto unidentified S. meliloti sensor/regulator locus involved in environmental control of gene expression. To Tom, for his friendship and encouragement in: studies as creative to, Tom Schm; discussions and presen endless ins for her adv, Hogan, Ma. EUChre At ACKNOWLEDGMENTS I would like to thank my advisor Frans de Bruijn for supporting me in my studies as well as creating an environment in the lab which encouraged creative thinking. I would like to thank my committee members John Breznak, Tom Schmidt, and Peter Wolk for their many helpful criticisms and discussions. I would like to thank all the members of the “de Bruijn—lab” past and present, for their friendship. In particular, I thank Philipp Kapranov for the endless insightful discussions about life and science, and Jodi Trzebiatowski for her advice and support. I would also like to thank my classmates Debbie Hogan, Mark Johnson, and Dan Buckley for the amazing discussions during Euchre. Above all, I would like to thank my family for their love and support. LBTOFF CHAPTER lntroductio' The The s Starvr Sinor, The Tl TABLE OF CONTENTS LIST OF FIGURES ......................................................................................................... viii CHAPTER 1 Introduction ..................................................................................................................... 1 The importance of starvation responses in bacteria .................................... 3 Starvation defined .................................................................................... 6 Response to nutrient limitation/deprivation ........................................ 7 Structural and physiological characteristics of starved cells .......... 8 The protein expression profiles of stewed cells .............................. 11 The function of starvation-induced proteins ..................................... 13 Regulation of nutrient-deprivation/stress induced genes ......................... 18 Increased cAMP levels during carbon deprivation .......................... 21 Two-component systems .................................................................... 23 Alternative sigma factors ...................................................................... 26 The soil environment ........................................................................................ 27 Starvation studies with Pseudomonas spp. ................................................. 29 Sinorhizobium meliloti as a model organism ............................................. 30 The responses of S. meliloti to environmental conditions ........................ 32 References ......................................................................................................... 37 CHAPTER 2: Identification and Characterization of a Novel Nutrient-Deprivation-Induced S. meliloti Locus (ndi) .............................................. 45 Abstract ............................................................................................................. 46 Introduction ......................................................................................................... 47 Materials and Methods ..................................................................................... 50 vi CHAPTER ldeotrf‘roatic Involved in Nutrient-De Results ............................................................................................................. 55 Discussion .......................................................................................................... 67 References ......................................................................................................... 71 CHAPTER 3: Identification and Characterization of S. meliloti Genes Involved in Regulating Expression of the Nutrient-Deprivation-lnduced ndi Locus ................................................................... 74 Abstract ............................................................................................................. 75 Introduction ......................................................................................................... 77 Materials and Methods ..................................................................................... 79 Results ............................................................................................................. 84 Discussion .......................................................................................................... 90 References ......................................................................................................... 92 CHAPTER 4: A Homologue of the Tryptophan-Rich Sensory Protein, TspO, and FixL Regulate a Novel Nutrient-Deprivation-lnduced S. meliloti Locus ............................................................................................................ 94 Abstract ............................................................................................................. 95 Introduction ......................................................................................................... 96 Materials and Methods ..................................................................................... 99 Results ........................................................................................................... 104 Discussion ........................................................................................................ 119 References ....................................................................................................... 124 vii CHAPTEF COflClUSlC Tre See Tspi PAS Re‘e' CHAPTER 5: Conclusions and Future Directions ......................................................................... 128 Transposon mediated reporter gene fusion approach ............................ 129 Search for genes involved in regulation during stress ............................. 131 TspO as a regulator of stress-induced gene expression ........................ 132 PAS domains: Internal sensors .................................................................... 134 Summary ........................................................................................................... 135 References ....................................................................................................... 1 38 viii Figure 11 melZ Fgwe2l HWEZZ meZB waZd F'gure 2 s Figure 2 6 Home 2.7 Flgure 28 Figure 2-9- FlgUre 31 Figdfe 32 Figure 33 Figure 1.1. Figure 1.2. Figure 2.1. Figure 2.2. Figure 2.3. Figure 2.4. Figure 2.5. Figure 2.6. Figure 2.7. Figure 2.8. Figure 2.9. Figure 3.1. Figure 3.2. Figure 3.3. LIST OF FIGURES Regulatory network depicted as a stimulus Response pathway ............................................................................... 20 Two-component signal transduction Paradigm ................................................................................................. 23 Map of the Tn5/uxAB tagged locus from strain 022 ....................... 56 Plasmid p022lux expression in E. coli .............................................. 58 ndi::Tn5/uxAB expression during carbon deprivation .................................................................... 60 ndi::Tn5/uxAB expression under low oxygen tension .................................................................... 61 A comparison of ndi ::Tn5/uxAB expression profiles during nitrogen and carbon deprivation. ........................................... 61 ndizzTn5luxAB expression profile during osmotic stress ........................................................................... 62 ndi::Tn5/uxAB expression during entry into post-exponential stationary-phase growth in complex TY medium. ........................................................................ 63 Growth of wild-type strain and mutant 022 in TY medium ................................................................................. 65 Survival during carbon deprivation ..................................................... 66 Secondary mutagenesis of mutant strain 022 (ndi::Tn1721) with Transposon Tn1721. ................................. 85 Advanced BLAST results of the Tn1721 tagged gene in mutant strain 1,F-1 .................................................................. 86 Advanced BLAST results of Tn1721 tagged gene in mutant strain 10,D-2 ............................................................... 87 Fgwe3~ Fgwe41 Fgwe42 Figure 4 3 me44 Figure 4 5 Figure 45, Figure 4 7 Figure 4- 8 Figure 4 g Figure411 Figure 4 1, Figure 3.4. Figure 4.1. Figure 4.2. Figure 4.3. Figure 4.4. Figure 4.5. Figure 4.6. Figure 4.7. Figure 4.8. Figure 4.9. Figure 4.10. Figure 4.11. Survival of wild-type S. meliloti strain 1021 and double mutant strain 10,D-2 during stationary phase growth. ......................................................... 89 Map of the Tn1721 tagged locus from “DARK” double mutant strain 12,C-4 .............................................. 104 Alignment of S. meliloti TspO deduced protein sequence ................................................................................. 106 Complementation of 12,C-4 Under low oxygen tension .................................................................. 108 Complementation of 120-4 during nitrogen and carbon deprivation, as well as during osmotic stress ..................................................................................... 109 Complementation of 120-4 with pRstspO during nitrogen and carbon deprivation, as well as during osmotic stress ..................................................... 110 Plasmid pC22lux expression in wild-type and 1021 ntrC::Tn5 mutant backgrounds after exposure to nitrogen deprivation ............................................. 111 Plasmid pC22lux expression in wild-type and 1021 fixL::Tn5 mutant backgrounds after exposure to low oxygen tensions ............................................ 1 12 Chromosomal ndizzTn5luxAB expression in wild-type andflxL::Tn5-233 under low oxygen tension .................................................................. 113 Chromosomal ndi::Tn5luxAB expression In wild-type and fixl::Tn5-233 During nutrient deprivation, and osmotic stress ........................... 114 ndi::Tn5luxAB expression in wild-type strain, and tspO and fixL mutant strains ...................................................... 116 CV2-luxAB expression in different mutant strains under low oxygen tensions. ............................................................... 1 16 Figure 4 Figure 4 ‘ Figure 4.12. Survival of wild-type S. meliloti strain 1021, C22, 12,C-4, and R-C22 during stationary phase ...................................................................... 118 Figure 4.12. Hypothetical model of ndiAB regulation ' by TspO and FixL. ................................................................................ 123 xi CHAPTER 1 Introduction Introduc‘ in ' factors C‘ most nat‘c 91011111 or t microbial c studies on special em acquired b. 000110”an stresses w resDonse c understaac characters described plOfiies dur Introduction In nature, bacterial growth is restricted by a wide variety of environmental factors. One factor of particular interest is the lack of essential nutrients. In most natural settings, resources are scarce, therefore periods of negligible growth or dormancy are likely to be the typical physiological state of the microbial cell. The aim of this chapter is to present information pertaining to studies on the physiological adaptations of bacteria to nutrient deprivation, with special emphasis on the parallel resistance to other types of stresses that is acquired by cells when confronted with starvation. The regulatory pathways controlling gene expression during nutrient deprivation, as well as during other stresses will also be highlighted. In the first section, I will describe the general response of bacteria when deprived of nutrients, as well as the significance of understanding the starved state. Next, structural and physiological characteristics of non-sporulating bacteria during nutrient deprivation will be described. Third, information will be presented regarding protein expression profiles during nutrient deprivation, as well as the function of starvation/stress induced genes during stasis. Fourth, regulatory pathways known to control gene expression during nutrient-deprivation and under low oxygen tensions will be reviewed. Lastly, the environmental parameters that bacteria must contend with in soil, as well as the characteristics of my model soil organism, Sinorhizobium meliloti, will be discussed. The imp their em" to their so comprise: nutrient de Bacteria rr order to S; Communit. unicel'auia' them to w respond tr morphotoi Mo poor. Re deCOmpO rhetabolic SUbStrate Sys‘ “ems. ”010.0151 t Er . I“ "ESEria 82m r'l We me The importance of starvation responses in bacteria A remarkable feature of prokaryotes is their ability to adapt dynamically to their environment by switching on and off different suites of genes in response to their surroundings. In the natural setting, these surroundings are often comprised of a wide variety of physiologically stressful parameters, such as nutrient deprivation, low oxygen tensions, osmotic stress, and desiccation. Bacteria must respond to and contend with these highly variable conditions in order to survive. It is evident from the ubiquitous and diverse microbial communities that inhabit almost every niche on earth (86, 132), that these unicellular organisms have evolved distinct mechanisms which clearly allow them to withstand these perturbations. Studies have shown that bacteria respond to such environmental factors with a variety of changes in cellular morphology and physiology, allowing them to survive (15, 20, 70, 93). Most natural habitats, such as soil and aquatic systems are nutrient poor. Resources enter these ecosystems only intermittently (eg. from decomposing leaf litter or from surface run-off). Moreover, multiple metabolically complex microbial communities compete for the available substrates and they are rapidly and competitively utilized. In addition, in aquatic systems, oxygen tensions can quickly diminish below the surface layers, due, not only to utilization, but also to limited solubility of oxygen in water. In terrestrial systems, the availability of oxygen is also greatly influenced in this same manner. In essence, even though oxygen is one of the most plentiful gases in the (58). This. ll require oxyg Henc energy limit does not in exception (' existence. . function ur Genus Bar: distinct get extensive process (5 Spores. st a Stasis (E DUl’lng 1h} allowing t A : liietabohC a has",C ‘6 they havi concentf.‘ gases in the atmosphere, most of the biosphere is limited for this element (58). This, in turn, severely affects the growth of strictly aerobic bacteria, which require oxygen for energy generation. Hence, the natural setting of bacteria is characteristically nutrient and energy limited, and therefore, periods of non-growth during which cell biomass does not increase or periods of negligible growth are the rule rather than the exception (71, 100). Even though this starved state is fundamental to bacterial existence, we are only just beginning to understand how bacteria adapt to and function under such conditions. Some bacteria, such as those belonging to the genus Bacillus, respond to starvation by differentiating into morphologically distinct cell types (spores) that are extremely resistant to stresses, and an extensive amount of work has been carried out to investigate this differentiation process (57). While it is evident that the majority of bacteria do not form spores, studies have shown that “non-spore-forming” bacteria also enter into a stasis (absence of growth) survival state, in response to nutrient deprivation. During this response, major structural and physiological changes occur allowing the cells to persist without essential nutrients, and, in addition, the cells are resistant to a variety of other stresses. A state of non-growth does not necessarily reflect an absence of metabolic activity. Bacteria persist in nature only if they can continually maintain a basic level of metabolic activity, even when nutrients are scarce. Accordingly, they have evolved mechanisms that allow them to capture substrates at concentrations that provide the essential nutrients for persistence, yet may not allow for an ‘pseudoser have been features in metabolic a cells can be spore—formi states. althi determine : starvation" areas. One involve res. Viable, but Studying th other main Starved Sta 0f Cells ade condlthns‘ allow for an increase in biomass (44). A variety of terms, such as “pseudosenescence” (92) “somnicell” formation (100), or “dormancy” (39), have been used to describe this starved state. Bacterial cells in this state have features in common with spores, including the absence of “detectable” metabolic activity. Moreover, under “improved” environmental conditions, the cells can be induced to return to a physiologically active state. Thus, non- spore-forming bacteria in nature clearly exist in a variety of different stewed states, although there has been controversy in the literature about how to determine and define these different states (40). Research investigating “starvation” in non-spore forming bacteria has been focused on two major areas. One field of investigation is that of the “spore-like” stage. These studies involve research on the viable but non-culturable (VBNC) state (83, 100) or viable, but not immediately culturable (NIC) state (40), and are concerned with studying the recovery of non-spore-forming bacteria from the starved state. The other main area is concerned with the response of bacteria as they enter the starved state. This area of research investigates the changes in the physiology of cells adapting from a eutrophic environment to nutrient-deprivation conditions. This area is the focus of this thesis. sian StafVE essential nut nutrient ‘5 CC particular nut phase is a te CGHS Occurs. However- a d starved cells. cells in CUHUrE complex. of d‘ are not define other hand. in Moreover. stat wrth starved ce cellular respon the molecular l may not overla; £64,106). Anoti deprivation. Th eXhosed to. tha‘ Starvation defined Starvation most commonly refers to the final result of deprivation for an essential nutrient. This starved state can be brought about when a particular nutrient is completely utilized, or by resuspending cells in medium lacking a particular nutrient, such as phosphate, nitrogen, or carbon (104). Stationary phase is a term that refers to a state in which no net increase in numbers of cells occurs, and it is often used interchangeably with the stewed state. However, a distinction should be made between stationary-phase cells and starved cells. The distinction made here is that stationary-phase describes cells in cultures that have stopped growing following balanced growth in a complex, or defined medium. In such cultures, the conditions that limit growth are not defined and, typically, growth is not limited by a single factor. On the other hand, in the case of starved cells the limiting nutrient is defined. Moreover, stationary-phase cells usually achieve high cell densities compared with starved cells and cell density can have a significant effect on the overall cellular responses (31). Furthermore, a distinct difference can be observed at the molecular level, where genes expressed by stationary-phase cells may or may not overlap with those genes that are expressed during starvation (64,106). Another term that is used through out this thesis is nutrient- deprivation. This term ls used to refer to the conditions that the cells are being exposed to, that is, resuspension in medium lacking either carbon or nitrogen. it however 0 nutrients a number of transport p metabolic j as well as (13. 63). lr nutrients. e Synthesrze as DFOduce extraceiiuia, QVOWIh conc Response to nutrient limitation Nutrient limitation is a condition in which all nutrients are available, however one or several nutrients are at concentrations that limit growth. When nutrients are limiting, bacteria can increase their chances of survival by a number of mechanisms, including the synthesis of higher-affinity enzymes or transport proteins specific for growth-limiting nutrients, utilization of alternative metabolic pathways to avoid possible blockages due to the lack of substrates, as well as decreased uptake of specific substrates that are available in excess (13, 63). In addition, depending on the relative concentrations of available nutrients, especially the ratio of carbon, nitrogen, and phosphorus, bacteria will synthesize storage polymers, such as polyphosphates and glycogen, as well as produce and release exopolysaccharides (73). These intracellular and extracellular storage compounds can be utilized later when more favorable growth conditions exist. IA Wt exhaustec differentia to the gen number of temper for nutrient de and fruiting generated i response tc Spores. the especially (1 tempeféiture While research he Them Esche, program wh j the absence l Structural and physiological characteristics of starved cells When an exogenous supply of an essential nutrient is completely exhausted, the starved state ensues. Some bacteria contend with starvation by differentiating into a morphologically distinct resistant form. Bacteria belonging to the genus Bacillus form endospores which are extremely resistant to a number of stresses including; desiccation, UV irradiation, and high temperature (24). Myxococcus spp. display another distinctive response to nutrient deprivation. These bacteria form elaborate multi-cellular aggregates and fruiting bodies when starved (38). Another stress resistant form is generated by Azotobacter species, which differentiate into a cyst-like form in response to nutrient deprivation by synthesizing a thick outer cell wall (58). Like spores, these cysts provide protection from environmental perturbations, especially desiccation; however, they are generally not as resistant to high temperatures as are spores. While it is evident that many bacteria do not form classical spores, research has shown that under starvation conditions, such species, among them Escherichia, Salmonella, and Vibn'o spp., do enter into a specific genetic program which results in major structural and physiological changes. The ultimate purpose of this new program is the persistence of the bacterial cells in the absence of growth (52, 64, 85). Vamt (32). As E, almost sphe Vibrio sp. 5 have been savahont dwdonhm size. the c mdeases still poten' had been broth. ‘nc Alcalige, Tl dlfiETErfi mafine prOpem phage ( MDQDOh Specie Various morphological changes are apparent in nutrient deprived cells (32). As E. coli cells enter into the stationary-phase, they become smaller and almost spherical in shape (73). This phenomenon has also been observed in Vibrio sp. S14 (78), as well as in a number of other marine bacteria, which have been shown to generate ultramicrocells, as small as 0.03 mm“, during starvation (44). These markedly smaller cells are the result of continued cell division without an increase in biomass. In addition to an overall change in size, the cell cytoplasm is condensed and the volume of the periplasm increases during starvation (97). However, as indicated earlier, these cells are still potentially viable. For example, when the filtrates of water samples, that had been passed through a 0.2 pm filter, were enriched with dilute nutrient broth, “normally sized” cells of Vibrio, Pseudomonas, Aeromonas, and Alcaligenes were recovered (44). The surface characteristics and adhesion properties of starved cells are different from those of cells growing under balanced growth conditions. Many marine bacteria become more hydrophobic, resulting in increased adherence properties when starved (44), and E. coli cells form aggregates in stationary phase (104). Changes in membrane fatty acid, peptidoglycan, and lipopolysaccharide composition have also been observed in a number of species (104). As stated earlier, a bacterial cell must maintain a basal level of metabolic activity in order to persist. When the exogenous source of an essential n through enr reactions w compounds U9) Endoi stored energ sufficrently r fundamenta' transport sul Endog the form of r metabolism, higher in star lnCl'ease Of 81 hours of Siam increases. wh FurthermOre RNA are met; SmWedCGHS‘ synthesis 'mm essential nutrient has been exhausted, basic functions must be maintained through endogenous metabolism. The latter is defined as “the total metabolic reactions which occur within a living cell when it is held in the absence of compounds or elements which serve specifically as exogenous substrates” (79). Endogenous metabolism allows the cells to maintain a basal level of stored energy (ATP or other high energy phosphate compounds), as well as a sufficiently high proton motive force across the membrane. Another fundamental use of basal endogenous metabolism is to sustain the ability to transport substrates into the cell, should they become available again. Endogenous storage polymers, as well as protein and RNA, mostly in the form of ribosomes, are used to sustain a basal level of endogenous metabolism. The rate of protein degradation in E. coli is approximately five-fold higher in starved cells compared to cells during balanced growth (59). An increase of sixteen-fold is observed in Vibrio sp. strain S14 during the first few hours of starvation (77). In addition, when cells are starved RNAse activity increases, which results in a decrease of 20 to 40% of the stable cellular RNA. Furthermore, the rate at which endogenous polymers, as well as protein and RNA are metabolized is carefully regulated (79). This control is evident in starved cells, which maintain the capacity to resume high rates of protein synthesis immediately after the addition of nutrients (1.). 10 The pi It has t energy derive macromolecul peptidoglycan that E. coli car when deprive: gene express: expressed by uses two-dime labeled polyps their molecular the second din the rates of Syi Profiles Und6r < The protein expression profiles of starved cells It has been demonstrated that when bacteria are nutrient deprived, energy derived from endogenous metabolism is used to synthesize macromolecules, including proteins (21, 78), RNA (78), lipids, and peptidoglycan (80). Specifically, Matin and his co-workers have demonstrated that E. coli cells continue to produce proteins for an extended period of time when deprived of exogenous carbon (21 ). To investigate starvation-induced gene expression at the whole cell level, “proteome” (the Min complement expressed by a gengm_e) methods have been applied (37, 90). This approach uses two-dimensional polyacrylamide gel electrophoresis (2-D-PAGE) of pulse labeled polypeptides (82) to separate proteins on the basis of their charge and their molecular mass, using isoelectric focusing in the first and SDS-PAGE in the second dimension. Radiolabeling of proteins permits a direct evaluation of the rates of synthesis of individual proteins (50), as well as their expression profiles under certain conditions or at specific times (123). Using the proteomic methods, a sequential synthesis of at least fifty proteins that are induced in response to carbon, phosphate, or nitrogen starvation has been documented in different non-sporulating bacteria (62, 78, 108). Where many of these polypeptides are induced by the deprivation of a particular nutrient, some of them have been found to be induced by several or all of these conditions. For example, Spector et al, discovered that in 11 Salmonella. responded t shown to be Similarly. stL were inducer conditions (6 presence of ‘ for carbon. n found to be ir starved for th additional 519 starved for se FOUI’ dz observed in E (43) In gene; early Stages 0 lhloUghOUt Sta after 20-50 m” lhrothOUt Stag was found to b Salmonella, twenty of the starvation-inducible polypeptides identified responded to two or more stress conditions and six of these proteins were shown to be produced in response to three or more starvation conditions (107). Similarly, studies in E. coli have shown the presence of fifteen polypeptides that were induced in response to carbon, phosphorus, or nitrogen deprivation conditions (62). Furthermore, analysis of Vibrio species has shown the presence of three polypeptides that are also induced in response to starvation for carbon, nitrogen, or phosphate (78). In addition, thirteen proteins were found to be induced by multiple starvation, that were not found in cells that were starved for the individual nutrients (78). These data indicate that alternative or additional signaling and regulatory pathways are involved when the cells are starved for several different nutrients at the same time. Four distinct temporal patterns of protein expression have been observed in E. coli (21), as well as in several other non-spore forming bacteria (43). In general, one class of proteins is synthesized transiently during the very early stages of carbon starvation. The other three classes of proteins persist throughout starvation, with maximum synthesis at different time points, i.e., after 20-50 minutes of starvation, between 3 and 4 hours of starvation, or throughout starvation (21). In addition, the temporal pattern of protein synthesis was found to be the same when the cells were starved for either glucose or succinate. The expression profiles of cells grown under both succinate and glucose starvation conditions indicated that the bulk protein synthesis rate is reduced to approximately 40% at the onset of starvation. This reduced level of 12 Protein syntl for 90 minut nondeteCtab expressed ea' Thefun NyStrOm N-terminal aimlr protein DlOflles E caben- This 9': are starved for C" coinciding decree expression Home growing CBllS anc First. since SUD“r product of aerobi aerobic respirallC which protects th respiratory ac “Vi: ended Venou 8 re 8: protein synthesis is maintained for 180 minutes during glucose starvation and for 90 minutes during succinate deprivation, at which time synthesis becomes non-detectable (21). Another common finding is that proteins that are expressed early in starvation are critical for long term stasis survival (78, 97). The functions of starvation-induced proteins Nystrom and co-workers (81) have used 2-D-PAGE in combination with N-terminal amino acid sequencing analysis to investigate the changes in protein profiles and to identify protein products in cells that are starved for carbon. This global “proteomic” approach has shown that when E. coli cells are starved for carbon, an increased synthesis of glycolysis enzymes with a coinciding decrease in the synthesis of TCA cycle enzymes occurs. This expression profile is remarkably similar to that exhibited by anaerobically growing cells and two explanations for this similarity have been proposed. First, since superoxide radicals, which are detrimental to the cell, are a by- product of aerobic respiration, the down-regulation of proteins involved in aerobic respiration during starvation-stasis may be a defense mechanism which protects the cell from oxidative stress. In addition, the reduced respiratory activity would also prevent the over-utilization of valuable endogenous resources (81). 13 ln adc studies have genes speCifi expression of of cstglacZ (l‘ S, lyphimunu has been em (130). Howe Deriplasmic c the Ciloplasr restricted to l starvationqm membranegc @901th gen phosphatase seCreted Or . CflOPlBSm‘ t Tn ( DhoA Ca ri id ~ In addition to the proteome research approach, genetic and molecular studies have been used to identify and characterize expression profiles of genes specifically induced during carbon starvation and stationary-phase. The expression of various carbon starvation (cst) genes have been studied by use of cstzlacZ (B—galactosidase) transcriptional fusions, both in E. coli (6) and in S. typhimurium (109). In addition, insertional mutagenesis with kplacMu (lacZ) has been employed to identify genes induced primarily during stationary-phase (130). However, B-galactosidase cannot be usedlto examine the expression of periplasmic or membrane localized proteins, because it cannot pass through the cytoplasmic cell wall. Therefore, studies with this reporter gene are restricted to proteins located in the cytoplasm. However, many of the starvation-induced genes that have been identified encode for periplasmic or membrane-localized proteins (2). In order to examine these proteins, the reporter gene phoA (alkaline phosphatase) has been used. Alkaline phosphatase is an excellent reporter gene for studying the expression of secreted or membrane associated proteins, because it is inactive in the cytoplasm, but can be readily assayed once it passes through the cytoplasmic membrane (60). Consequently, insertional mutagenesis with transposon TnphoA carrying the promoterless reporter gene phoA has been employed to identify such proteins (2, 3). The discovery of specific functions for many starvation or stress-induced genes can be attributed to molecular studies of cells during stationary phase. 14 Over the PE microbiolog discoveries is encoded during statii well as in n transcribed For a comp CO/l and Sa. et al, (1995 Ever Several of ( known (26) stationary; lnClude tho: Cell; SUCh a enCOding h §!YCerol-3-F i0l’ ploleins 3503mm 8‘ pi'Od'dCts in be - en 'dent Over the past 15 years, there has been intensive research by molecular microbiologists on stationary-phase gene expression (45). One of the key discoveries has been the identification of an alternative sigma factor, as, which is encoded by the rpoS or katF gene (72, 114). Many of the genes expressed during stationary phase, or under carbon, nitrogen, or phosphate starvation, as well as in response to a diverse number of stresses (see below), are transcribed by RNA polymerase containing this alternative sigma factor (27). For a comprehensive review of as - dependent genes and their function in E. coli and Salmonella typhimurium, the reader is referred to a review by Loewen et al., (1998). Even though the majority of as - controlled genes are still unknown, several of the genes controlled by as encode for proteins whose function is known (26). In the following section, a brief description of some of these stationary—phase-induced proteins will be presented. as - controlled genes include those whose protein products are involved in providing energy to the cell; such as cbdAB which encode a cytochrome bd type oxidase (4),or hya encoding hydrogenase 1 (4), as well as glpD, the structural gene for aerobic glycerol-3-phosphate dehydrogenase (130). Another category of genes encode for proteins that are involved in transport of substrates, for example, proP is a transport system for glycine betaine and proline (66, 134). In addition, protein products involved in changes in cell morphology or surface properties have been identified. Among these are boIA which encodes a regulatory protein 15 required for carboxypep formation. l lipoprotein. of cell aggri As ir other physi. phase grow protection. stress. sucl protection a DrOducts pr which aid i stabilizes t plOlEIns ex damage. 5 (47’ 127) c F the with PlOtei Synthesqs required for induced expression of PBP6, a penicillin-binding protein- carboxypeptidase (48), which, in conjunction with ftsQAZ functions in septum formation. Also, the expression of osmB, which encodes an outer membrane lipoprotein, is controlled by as, and this protein may play a role in the formation of cell aggregates (26). As indicated earlier, starved cells also become resistant to a variety of other physiological stresses. Several of the genes induced during stationary- phase growth encode protein products that are involved in general stress protection. Among these are genes which provide resistance to oxidative stress, such as katE and katG (34, 53) which encode catalases providing protection against peroxide. In addition, there are genes whose protein products provide protection during osmotic stress, such as otsA, otsB and treA which aid in the synthesis and transport of trehalose, an osmoprotectant which stabilizes both membranes and proteins (28, 112). Furthermore, there are proteins expressed during entry into starvation that function to repair DNA damage, such as aidB which is involved in DNA methylation damage repair (47, 127) or xthA an exonuclease III which is also involved in DNA repair (102). Finally, some of the genes required for stasis survival are concerned with protein repair and maintenance. Since starved cells are unable to readily synthesize new proteins in response to environmental changes, they must rely on existing proteins to carry out essential cellular functions. Existing proteins are likely to become denatured or otherwise damaged due to environmental 16 shesses preventin cessation involved i. which enc (49). This naturally ll must be re proper ass to function Wthh encr residues c.- Therefore, DTOper Drot SUCh as Gr laCtor (532) have been ‘ r.erlal'clflng C Chapergnes Strains With degradation in prOtGO'YSf stresses (126). Therefore, mechanisms for repairing spontaneous damage, preventing denaturation, as well as assisting in renaturation of proteins after cessation of the stress conditions are critical for survival (63, 126). Two genes involved in damage repair have been identified. One is the surA gene (118) which encodes a periplasmic protein with peptidyI-prolyl isomerase activity (49). This activity repairs cis-conversions of prolines. Such conversions occur naturally in proteins and interfere with proper folding, therefore this damage must be repaired to recover function. Also, this isomerase is required for proper assembly of some outer membrane proteins. Therefore, it also appears to function as a chaperone (101 ). Another damage control gene is pcm (51 ), which encodes an L-isoaspartyl protein methyl transferase. Isoaspartate residues can also be formed spontaneously interfering with proper folding. Therefore, conversion of isoaspartate to L-aspartyl residues is essential for proper protein folding and function. In addition, genes encoding chaperones, such as GroEl and DnaK, which are under control of the heat-shock sigma 32 . . . . . factor (0' ) also show increased expressron during starvation. These proteins have been shown both to prevent denaturation of proteins, as well as aid in renaturing denatured proteins (116). Furthermore, even though these chaperones are not proteolytic per se, they appear to play a role in proteolysis. Strains with mutations in the dnaK gene have been found to impaired in protein degradation (41, 111). Two hypotheses as to how these chaperones function in proteolysis have been proposed (88). It may be that these proteins function as part of a protease complex or they may bind abnormal proteins in such a 17 way as to exact furlC' Regulatior Trar it with five (7). Since effluent in. bacteria. tr synthesize not very 54,, translation C‘lOlCeS. lt transcript ll this level. due to limi t'anSCllDt e and POSHI Funhermo ChrOmOSor transgnbm This is khc way as to make them more accessible to cleavage by proteases. However, the exact function of these proteins has not, as yet, been determined. Regulation of starvation/stress induced genes Transcribing an average 1kb E. coli gene just once and then translating it with five ribosomes consumes a minimum of 7000 “high energy” phosphates (7). Since energy efficiency is of critical importance to starved bacteria, an efficient mechanism for controlling gene expression must be employed. In bacteria, translation and transcription are concurrent. As mRNA is being synthesized translation commences immediately. Moreover, most mRNAs are not very stable. Therefore, the choice of templates that are available for translation by the pool of ribosomes, in essence, mirrors the transcriptional choices. It follows that extremely efficient control can occur at the level of transcript initiation and, in fact, most of the known global regulators do act at this level. It should be noted, however, that our present view may be biased due to limited evidence; therefore other mechanisms, such as controlling transcript elongation and termination, RNA processing, initiation of translation, and post-translational modifications may also play a significant role (7). Furthermore, the ability to select a variety of genes at different locations on the chromosome at the same time, without utilizing precious energy for transcribing and translating non-essential operons is extremely advantageous. This is known to be accomplished by two different mechanisms. First of all, 18 histone-like ' DNA, thereb (IHF) has be expressron o Regul response pa expression c nutrient dep' sensor proti or indirectly Number of t operons. th lesponse 9 generated. trarisduce: thereby m. for SYSterT at Various histone-like proteins (IHF, H-NS, and HU) are involved in dynamically masking DNA, thereby controlling accessibility. For example, Integration Host Factor (IHF) has been found to be required for starvation survival, as well as the expression of fourteen glucose-deprivation-induced proteins (75,76). Regulatory networks, such as the one depicted in Fig. 1 as a stimulus response pathway, are used by bacteria to orchestrate the coordinated expression of a number of different genes/operons (73). A stimulus, such as nutrient deprivation or osmotic shock, is detected by a sensor protein. The sensor protein then passes this signal on to a regulatory protein, either directly or indirectly via one or more transducers. The regulatory protein then acts on a number of target operons, which, in turn, may control the expression of other operons, thus creating a regulatory cascade. When the products of the response genes have reached the proper level, a feed-back response is generated. This feedback response can act at the level of signal production, transducer activity, or even regulator activity, or some combination there-of, thereby modulating the response. In this manner, a regulatory network allows for system equilibrium, since the expression of the cascade can be controlled at various levels resulting in stabilization at the appropriate amount of expression required for persistence under the new conditions (7). 19 rn Respon Stimulus Sensor l > Sigfal \ - I Regulator I Transducer / Response / V X\ \ /T arget Operons Proteins ‘ < x \ j \ Figure 1. Regulatory network depicted as a stimulus-response pathway Secondary Operons Many operons, including those that contain genes required for the catabolism of certain sugars or amino acids, are controlled by such global regulatory networks, but are also independently regulated, creating an additional level of control (7). It is evident that bacteria have evolved multiple mechanisms to orchestrate the expression of regulatory networks (7). In some cases, the mechanism used is the same as that described for a single operon, namely, via a repressor or activator protein that recognizes a particular promoter region; however, in the case of global regulatory networks, the promoter element is 20 common to alternative si wanderem combination for example. (87). Finally. non-protein 5 (amino aCld 1 various gene guanosine te t9- 84). Sele SenmCancei sections ll'ICre; lnEc negatIVe b8: WW to C common to a set of operons. In other cases, coordinated regulation involves altemative sigma factors (see below), that direct RNA polymerase to specific promoter elements present in the target genes/operons. In addition, a combination of regulatory proteins and sigma factors can be utilized, as seen, for example, in the nitrogen utilization control pathway of a variety of bacteria (87). Finally, some regulatory networks involve other factors, including internal non-protein signaling molecules. For example, during the stringent response (amino acid starvation), as well as in response to other stresses, expression of various genes and operons appears to be regulated by the nucleotide guanosine tetraphosphate (ppGpp) in a manner that remains to be elucidated (9, 84). Selected examples of different control mechanisms and their significance during nutrient deprivation will be reviewed in the following secflons. Increased cAMP levels during general carbon deprivation In E. coli, the only known regulatory system involved in monitoring carbon availability is the cyclic adenosine monophosphate/cAMP receptor protein (CAMP/CRP) system. cAMP is a key mediator of gene expression in Gram- negative bacteria (8). This molecule is synthesized from ATP by adenylate cyclase, which is encoded by the eye gene. It acts as an effector molecule by binding to CRP which subsequently activates transcription at target promoters. 21 Studies usrn induced in re These protei' However an: (62). Moreove synthesized u proteins are 5, cultures suppl proteins regul response to s carbon-starve catabolic pote S'u'bStTates be EXDressed gr Survival lDSte their 9908 eX Studies using cya mutants of E. coli have shown that some of the genes induced in response to carbon starvation require cAMP for their induction. These proteins have been designated Cst (carbon starvation) proteins. However, another class of carbon starvation induced genes have been found to be cAMP independent; this group was named Pex (post-exponential) proteins (62). Moreover, it was discovered that the CAMP-independent (Pex) proteins are synthesized under various starvation and stress conditions, while the Cst proteins are specifically induced by the lack of carbon or in non-starving cultures supplemented with cAMP (6). Therefore, it has been proposed that the proteins regulated by cAMP levels (Cst) are specifically concerned with the response to starvation. For example, the cstA gene is specifically expressed by carbon-starved cells and functions in peptide utilization, and this increase in catabolic potential may be extremely advantageous when the more “typical” substrates become limited (103). On the other hand, the Pex proteins are expressed under various stresses and they appear to function in stress survival instead of the response to carbon starvation specifically; moreover, their gene expression appears to rely on alternative sigma factors, as well. 22 Two- A me and module called two-c comprehen: osmolarity a modifying tr (87). These units: input receiver (joy and receive deDhOSphor Of a Sensor) Two-component systems A major mechanism of signal transduction that bacteria use to sense and modulate gene expression in response to environmental stimuli is the so- called two-component system (see Blumenthal et al. [1996] for a comprehensive review). Nitrogen starvation, oxygen limitation, and changes in osmolarity are but a few of the environmental stresses that cells overcome by modifying their cellular physiology with the help of two-component systems (87). These bicomponent systems are composed of the following functional units: input sensor domains; output effector domains; and transmitter and receiver domains, as shown in Figure 2. Communication between transmitter and receiver domains is primarily accomplished by phosphorylation and dephosphorylation reactions. In essence the two-component systems consist of a sensory kinase and an associated response regulator. Environmental stimuli l l SENSOR { Phosphorylation signal >3 input domain Mtransmitter <13 f 1 ® ® receiver output domain < I REGULATOR l Figure 1. 2. Two-component signal transduction paradigm, consisting of an input sensor domain and an output effector domain with transmitter and receiver domains for protein-protein communication. 23 The t expression i availability c tension. and for respondi control the L and Ach/Ar component genes locate identified in Pseudomor, the Synthes assimilation in E. membrane - c0l‘lditions C OmPC is pr differ Signiti. Contl'Oiilng i The two-component regulatory systems most relevant here control gene expression in response to nitrogen availability, changes in osmolarity, the availability of different electron acceptors, as well as changes in oxygen tension, and include NtrB/NtrC for monitoring nitrogen availability, EnvZ/OmpR for responding to changes in the surrounding osmolarity, NarX/NarL which control the use of nitrate as a terminal electron acceptor during anaerobiosis, and Ach/ArcA which controls aerobic respiration (87). As with most two- component systems, these regulatory circuits coordinate the expression of genes located in a variety of specific operons. The NtrB/NtrC system has been identified in many different bacteria, including a variety of enteric species, Pseudomonas spp., as well as S. meliloti, and is required for the regulation of the synthesis of glutamine synthetase and other enzymes important in nitrogen assimilation in response to changes in the availability of nitrogen (74, 99). In E. coli, the EnvZ/OmpR system controls the ratio of the two outer membrane porins, OmpF and OmpC, in response to osmolarity. Under conditions of low osmolarity OmpF is favored, whereas in high osmolarity, OmpC is preferentially synthesized (120). The two porins are highly similar, but differ significantly in the diameters of the channels formed by each (94), thereby controlling passive diffusion. An example of the interconnected nature of regulatory networks is found in the pathways controlling the adaptation of E. coli to different redox environments, in this regulatory scheme three global regulatory systems interact to control the use of energy sources in this bacterium (33). A key 24 strategy of t. donors to ter allowed with available ene The N. preferentially nitrate and rr- the synthesrs addition, Nar This regulate acid sequent (CRpldescr Contains a fc that the trans redox seDSir aeromc res; anaerOhio'sis precise mec strategy of this facultative anaerobe is to channel electron transport from donors to terminal acceptors so that the drop in free energy is the maximal allowed with the available substrates. Thereby, the cells are able to exploit available energy sources to obtain the most energy. The NarX/NarL two-component system controls anaerobic respiration by preferentially activating the operon encoding the major nitrate reductase when nitrate and molybdate are available. At the same time, this system represses the synthesis of other less beneficial anaerobic terminal reductases. In addition, NarX/NarL works in concert with another global regulator, Fnr (110). This regulator can function both as an activator and a repressor and its amino acid sequence shows a high degree of similarity to the catabolite repressor (CRP) described above. Fnr, however, has an added feature. This regulator contains a four-cysteine cluster that binds Fe 2+, and there is strong evidence that the transition between the +2 and the +3 states of the iron plays a role in redox sensing (33). In addition, E. coli utilizes a system (AchlArcA) to control aerobic respiration (81 ). This control system is based on repression during anaerobiosis and relief of this repression when oxygen is available. The precise mechanism by which this regulation occurs has not been elucidated. 25 Alterr Recec transcriptiona ' environmenta systems. rely l distinct sigma 1 different enviro deprivation (r;N t‘3F 0f 0'28). iror M. In additio. only legulates i but also serves In- S described as it) for survive, um 51 -~ W'Cam Subsr Cfthe Pex Piste abiiit y lo Sun/We Alternative sigma factors Recent studies have firmly established that many mechanisms of transcriptional control of gene expression in response to different environmental stresses, including responses mediated by two-component systems, rely heavily on the utilization of distinct sigma factors. In E. coli, six distinct sigma factors have been found to be involved in the response to different environmental signals, including heat shock (OH or 032), nitrogen . . N 54 . E 24 . deprivation (G or o ), extracytoplasmic stress (6 or o ), substrate gradients F 28 . . . 19 . s 38 . . (c or o ), iron deprivation (o ), and osmotic stress (0 or o ) spec1fic factors (54). In addition, it is now evident that the Us subunit of RNA polymerase not only regulates gene expression during starvation and stationary-phase growth, but also serves as a master regulatory component of gene expression during general stress conditions, even when the cells are in exponential growth (25). In fact, as controls such a large number of genes that it has even been described as the primary sigma factor. It has also been found to be essential for survival under the physiological stressful conditions that prevail in nature (25, 26). For example, in E. coli, rpoS mutants lacking as fall to induce a significant subset of the carbon starvation induced polypeptides, including six of the Pex proteins. Moreover, these mutants are drastically impaired in the ability to survive starvation conditions, and are severely affected in their 26 response tr (61). There genetic res highly varie The soil en Althc starvation c information variables. | have evolve Constitutes fluctuating F Moreover, Ii rflatter varie recalcitrant response to other stresses, such as heat shock, oxidation, and hyperosmosis (61 ). Therefore, it is becoming clear that Cs is one of the key elements of the genetic response that enables E. coli to persist in its natural habitat under highly variable physiological conditions. The soil environment Although much is known about the changes in expression profiles under starvation conditions in a select group of enteric and marine bacteria, little information is available about the response of indigenous soil bacteria to such variables. It is likely that mechanisms for survival particular to this environment have evolved in bacteria whose normal habitat is soil. As shown in Fig. 3, soil constitutes an extremely complex heterogeneous environment with numerous fluctuating parameters that can influence microbial growth and survival (89). Moreover, like most natural habitats, soil is nutrient poor (133). Soil organic matter varies in concentration from 08-20%, with the bulk of the carbon in recalcitrant forms, such as humus, therefore bacteria indigenous to soil must constantly contend with nutrient deprivation (121 ). 27 attached microcolgny Figure 3. A soil aggregate showing bacteria localized on the surface as microcolonies, as well as the complexity and heterogeneity of bacterial niches found in a a single aggregate. (adapted from Madigan et al., 1997) 28 Starvation In th1 Pseudomor investigatec coli strains was discov multiple-nu cells showe first week ( found to be analyses w- sllnthesize: deerivatign these Drote Hovi eXpiessior three differ KT2440. c starvation Cells that V mUta tiOns Starvation studies with Pseudomonas spp. In the past few years, the response of the ubiquitous soil bacterium, Pseudomonas putida KT2442, to carbon starvation conditions has been investigated. In these studies, P. putida KT2442 was tested along with two E. coli strains and one S. typhimurium strain for their ability to survive starvation. It was discovered that P. putida was fully viable after 20 days of carbon or multiple-nutrient starvation (17). On the other hand, E. coli and S. typhimurium cells showed a decrease in viability of one to two orders of magnitude after the first week (17). A Vibrio strain was also tested in a similar manner and it was found to be fully viable after two weeks in sea water (17). In addition, 2-D-PAGE analyses were performed and the results indicate that P. putida also synthesizes an array of new proteins in a temporal fashion, in response to deprivation of an exogenous carbon source. The function and regulation of these proteins have not, as yet, been elucidated (16). However, more recently, the role of RpoS in regulating starvation gene expression, as well as in conferring stress tolerance has been examined in three different species of Pseudomonas. A rpoS mutant of P. putida strain KT 2440, C1 R1, was examined and showed reduced survival of carbon starvation, as well as reduced cross-protection to other types of stresses in cells that were carbon starved (96). In another study, it was found that rpoS mutations in P. aeruginosa resulted in a two - to three-fold decrease in resistance to a variety of stresses, although the effect was found to be less 29 pronounced system ((38 stress respc two-compor stress. In a: contained le evident the spp. Sinorhizob Bact 0f Organisn These bact in Which a , ”Odule (fOr inrmed pros atmospheric plant. hEnce W953 anc his we” est; la . tire m‘ZOSph, Tog v U] 3 aiOry mt pronounced as that observed in E. coli cells (35). In addition, a two-component system (GacS/GacA) has been found to influence rpoS accumulation and the stress response of P. fluorescens Pf-5 (131 ). Strains with mutations in this two-component system are compromised in their ability to withstand oxidative stress. In addition, during entrance into stationary-phase, these mutants contained less than 20% of the wild-type levels of 08. Therefore, it is becoming evident that as also plays an important role in stasis survival in Pseudomonas spp. Sinorhizobium meliloti as a model organism Bacteria belonging to the family Rhizobiaceae represent a unique group of organisms for research on environmental control of gene expression in soil. These bacteria form a mutualistic symbiosis with their leguminous host plant, in which a new specialized organ is formed, a nitrogen-fixing root or stem nodule (for a recent review see Spaink et al., 1998). The nodules that are formed provide the environs for the bacteria to survive and to reduce atmospheric dinitrogen to ammonia, which can then be assimilated by the plant, hence mutualism (55). The development of this symbiosis is a complex process and requires a constant exchange of signals between the symbionts. It is well established that plants influence the microbial community structure of the rhizosphere by both excreting growth promoting nutrients and releasing regulatory molecules that control bacterial gene expression (56, 91). In 30 response. t contact with soil. the rhiz in the oligot signaling in rhizosphere the host ce recluire the and coordr surroundir bacteria a GI gene 9; amenable Genetic m sequeDCe In ; Selecting eXpreSsiO wnd‘tYpe response, the rhizobia are known to produce signal molecules that alter the gene expression of their host plant (105, 122, 125). Rhizobia persist in bulk soil with a scarcity of nutrients until they come in contact with their host plant. Hence, they persist in three distinct habitats: bulk soil, the rhizosphere; and the plant nodule. In order to do so, they must survive in the oligotrophic conditions of soil, competitively sense and utilize plant signaling molecules and growth promoting nutrients excreted from the rhizosphere, as well as adapt to the very different homeostatic environs within the host cell cytoplasm. Moreover, these three distinct modes of existence require that these bacteria consistently sense the environmental parameters and coordinately regulate gene expression to adapt to changes in their surroundings, even when they are starved for essential nutrients. Hence, these bacteria afford a unique model system in which to study environmental control of gene expression in an indigenous soil bacterium. Furthermore, S. meliloti is amenable to the use of molecular genetic techniques (18), both physical and genetic maps are available, and the determination of its genomic DNA sequence is well-advanced (29) In addition, two aspects, in particular, of rhizobial biology add value to selecting these bacteria for studies on environmental control of gene expression. First, nodule occupancy competition studies between mutant and wild-type strains in order to assess the effect of mutations on symbiosis, provide a highly specific and unique assay (69). Second, during the intermediate stages of symbiotic development, the bacteria infect the plant 31 J .. .t ' "fut-r .4” ‘- nodule cells accompanie which incluo‘ adaptation to bacterords re although con terminally dif ineversrble common wrt state. Hertc rhizobia. nodule cells and differentiate into the “bacteroid state”. This differentiation is accompanied by considerable morphological and physiological changes (124), which include adjustments in both nitrogen and carbon metabolism, as well as adaptation to the micro-aerobic milieu within the plant cell (12, 36). Moreover, bacteroids represent a non-growth or even terminally differentiated state, although controversy still continues as to whether bacteroids are, in fact, terminally differentiated (95). Nevertheless, this reversible or potentially even irreversible “nitrogen-fixing organelle”- state of bacteroids has features in common with the more universal bacterial “culturable” versus “non-culturable” state. Hence, another aspect of stasis survival can be investigated using rhizobia. The responses of S. meliloti to environmental conditions The response of rhizobia to nutrient deprivation or oxygen limitation is just beginning to be investigated. Recently, evidence for a general starvation response in Rhizobium leguminosarum, similar to that found in E. coli and Vibn'o spp. has been reported (115). In addition, Uhde et al, (1997) have used transposon mutagenesis to identify S. meliloti mutants that are affected in stationary-phase survival. Some of the transposon tagged loci have been cloned and analyzed and found to be involved in amino acid metabolism and aerobic respiration. In addition, phosphate stress-induced genes in S. meliloti 32 meliloti still r this sigma fa promoter ele as cells exit functional cc experiments from E. coli lPDGpp) roii in S. melilot strain 41 an to these stri 1021 Shows starvation. E SinCe ”ltrog that the abil (113), as well as changes in chemotaxis, motility, and flagellation in response to starvation have been investigated (129). Studies on the regulatory mechanisms controlling gene expression during various stresses have, also, only just begun. A homologue of rpoS in S. meliloti still remains to be isolated, although the following data indicate that this sigma factor does exist. RpoS - dependent growth phase-regulated promoter elements from E. coli have been found to be recognized in S. meliloti as cells exit exponential growth and enter stationary-phase, indicating functional complementation (67). In addition, Southern hybridization experiments have identified DNA fragments that hybridize with the rpoS gene from E. coli (67). Accumulation of the nucleotide guanosine tetraphosphate (ppGpp) following amino acid, nitrogen, or carbon starvation has been reported in S. meliloti 1021 (Howorth, 1999). Interestingly, two other isolates (S. meliloti strain 41 and R. tropici CIAT899) were found not to produce ppGpp in response to these stresses (30). In addition, the ppGpp accumulating (stringent) strain 1021 showed much higher levels of intracellular ATP in response to nitrogen starvation, as compared to the ATP levels in the relaxed strains of rhizobia. Since nitrogen fixation requires a great deal of energy, it has been hypothesized that the ability to mount a stringent response and produce elevated levels of ATP may be directly related to the efficiency of nitrogen fixation (30). However, further studies need to be done to address the significance of these findings. Two-component signal transduction pathways that sense and induce gene expression in response to the lack of nitrogen (Ntr system) (14); the 33 presence of sensing (Fix have been 1 required for rhizobia hav and RegAB. ' Rhodobactei synthesis of signals (22). transduction low pH. whe low OXygen determine tl SYstems. it DH. Our 1 ”ment‘de; EXPleSsed ‘ AS Domed ‘ thSimOg‘C‘ Ce ”a“ Stre. recur v a tort! c presence of dicarboxylic acids (Dct BD system) (22, 128); low oxygen tension sensing (Fix LJ system) (5, 10, 19) as well as pH sensing (ActRS) (98, 117); have been identified in S. meliloti. The FixLJ is unique to rhizobia and is required for symbiotic nitrogen fixation (5). The DctBD and ActRS systems in rhizobia have strong sequence similarity to two corresponding systems (DctSR and RegAB, respectively) in Rhodobacter spp. The DctSR system in Rhodobacter has a similar function as its rhizobial counterpart, namely, synthesis of dicarboxylate transporters in response to certain environmental signals (22). However, ActRS and RegAB appear to represent distinct signal transduction pathways. The ActRS system is concerned with the response to low pH, whereas the RegAB system appears to be involved in the response to low oxygen tension (117). Although additional research is required to determine the exact mechanism and signal that is detected by these two systems, it may be that the sensor proteins in these systems respond to some environmental signal that is influenced by both low oxygen tensions and low pH. Our laboratory has focused its studies on the nature and regulation of nutrient-deprivation-induced S. meliloti genes, and a collection of 70 genes expressed during carbon or nitrogen deprivation has been described (68, 69). As pointed out above, it is evident that bacteria frequently encounter an array of physiologically stressful parameters in their natural setting. Furthermore, certain stresses, such as nutrient deprivation or osmotic stress, trigger global regulatory circuits that control the expression of genes belonging to several 34 pathways. Therefore. \ involved in , regulatory g" characterizat organism fur in nature. Ir bacteria enc re also exp genes could during Symt into bactero Towa during Nutrie tranSPOSicm . Tn5/WAS in Cal'bOn, Ditrc stationary~pf was deSlgna The chgracte round of mm regulathn Of Tn1721. and pathways, which allow these bacteria to withstand various perturbations. Therefore, we hypothesized that by searching for S. meliloti regulatory genes involved in controlling gene expression under multiple stresses, global regulatory genes could be identified, and that the identification and characterization of such global regulatory genes might shed light on how this organism functions under the different physiological conditions it encounters, in nature. In addition, since many of the environmental parameters these bacteria encounter in soil; such as low oxygen tensions and osmotic changes, are also experienced in planta, we hypothesized that these global regulatory genes could control gene expression in both the free living state, as well as during symbiosis when the bacteria become endosymbiotic and differentiate into bacteroids. Towards this goal, regulatory components for the induction of genes during nutrient deprivation were identified using successive rounds of transposon mutagenesis. A S. meliloti mutant (C22) which harbors a Tn5/uxAB insertion in a gene that is induced by a variety of stresses; including carbon, nitrogen, or oxygen deprivation, osmotic stress, as well as during stationary-phase was identified after the first round of mutagenesis. This locus was designated ndi for nutrient deprivation induced, and it appears to be novel. The characterization of this locus will be discussed in Chapter 2. A second round of mutagenesis was subsequently used to identify genes involved in the regulation of the ndi locus. Strain C22 was mutagenized with transposon Tn1721, and a library of double mutants was screened for altered patterns of 35 lux express altered ndi of two of the One c the I'ldl locus special intere Wliil a high di (65). as well a Rhodobacter mediated regi evidence (135 intermediates the maIOT pori Oherons by cc Hence, TSpO leeulate gene presented in C lux expression. Three double mutants (12,C-4, 1,F-1, 10,D-2) that display an altered ndi expression pattern were identified. The preliminary characterization of two of these loci (1 ,F-1 and 10,D-2) will be described in Chapter 3. One of the double mutants (12,C-4) that lacks transcriptional activity of the ndi locus under any of the inducing conditions tested has proven to be of special interest. The Tn1721 transposon was found to be inserted in a gene with a high degree of similarity to the mitochondrial benzodiazepine receptor (65), as well as to the outer membrane oxygen sensor protein (TspO) of Rhodobacter sphaeroides (136). Although the exact mechanism for TspO mediated regulation of gene expression is still not understood, there is evidence (135) that this outer membrane protein regulates the efflux of intermediates from the heme biosynthetic pathway (porphyrinogens) through the major porins, thereby controlling the expression of genes from certain operons by controlling the intracellular concentration of an effector molecule. Hence, TspO appears to represent a novel mechanism by which bacteria can regulate gene expression when stressed. The results of this study will be presented in Chapter 4. 36 inuneo' s ntheSis icrobiol. 2.Alexande enes in E 37335-341 3Alexander starvation- membrane 4.Atlung, T., sigma S a: Escherichi phosphate SBatut, J., Leeuwenh 6.Blum, P. I- Burgess, cyclic AMI Bacteriol. 7.8lumenth anal sis o and . Mc BlOiOgy, VI 8.Botsford, 4fl4h62o 9CaShel, Iv C-Neuha biol0gy, vc REFERENCES 1.AIbertson, N. H., T. Nystrom, and S. Kjelleberg. 1990. Macro-molecular anthesrs durin recover; of the marine Vibrio sp. S14 from starvation. J Gen icrobiol. 136: 201-220 . 2.Alexander, D. M., K. Damerau, and A. C. St. John. 1993. Carbth/c'zlrate uptake ggn3e§51n3§gchenchia coli are induced by carbon starvation. Curr icrobiol. 3.Alexander, D. M., and A. C. St John. 1994. Characterization of the carbon starvation-inducible stationa phase-inducible ene s/ encodin an outer membrane lipoprotein in Esc erichia coli. Mol icrobio. 11(6):10 9—1071. 4.A_tlung T., K. Knudsen, L. Heerfordt, and L. Brondsted. 1997. Effects of Sigma S and the transcri tional activator AppY on the induction of the Escherichia CO/I hya an cbdAB-apgA operons in response to carbon and phosphate starvation. J Bacteriol. 1 9(7):2141-2146. 5.Batut, J., and P. Boistard. 1994. Oxygen control in Rhizobium. Antonie van Leeuwenhoek. 66:129-150. 6.Blum, P. H., F. D. Jovanovich, M. P._McCann, J._ E. Schultz, S. E. Lesley, R. R. Burgess and A. Matin. 1990. Cloning and in vrvo and in vitro regulation of cyclic AMP-dependent carbon starvation genes from Escherichia coli. J Bacteriol. 172:3813-3820. 7.BIumenthaI, R. M., D. W. Borst, and R. G. Matthews. 1996. Ex erimental anal sis of Iobal gene regulation in Escherichia coli, p. 1-79. n W. E. Cohen and . Mol ave ed.), Progress in Nucleic ACId Research and Molecular Biology, vol. 55. cademic Press, New York, NY. 8.Botsford, J. L. 1981. Cyclic nucleotides in prokaryotes. Microbiol. Rev. 45(4):620-642. 9.Cashel, M., and K. E. Rudd. 1996. The stringent response, p. 1410-1431. In F. C. Neidhart (ed.), Escherichia coli and Salmonella: cellular and molecular biology, vol. 1. American Socrety for Microbiology, Washington, DC. 10.David, M. M.-L. Daveran J. Batut, A. Dedieu, O. Domergue, J. Ghai, C. Hertig, P. Soistard, and . Kahn. 1988. Cascade regulation of nif gene expression in Rhizobium meliloti. Cell. 54:671—683. 11.de Bruijn, F. J. 1987. Transposon Tn5 mutagenesis to map genes. Meth Enzymol. 154:175-196. 12.Encamacion, S., M. Dunn, K. Willms, and J. Mora. 1995. Fermentative and aerobic metabolism in Rhizobium etli. J. Bacteriol. 177(11):3058-3066. 13.Ferenci, T. 1996. Adaptation to life at micromolar nutrient levels: the re ulation of Escherichia coliglucose transport by endornduction and cAMP. F MS Microbiol Rev. 18:301- 17. 14.Fischer, H.-M. 1994. Genetic regulation of nitrogen fixation in rhizobia. Microbiol. Rev. 58(3):352-386. 15.Foster, J. W., and M. P. Spector. 1995. How Salmonella survive against the odds. Ann Rev Microbiol. 49:145-174. 16.Givskov, M., L. Eberl, and S. Molin. 1994. Reponse to nutrient starvation in Pseudomonas putida KT2442: two-dimensional electro horetic anal sis of starvation- and stress-induced proteins. J. of Bacteriol. 76:4816-48 4. 37 liGivsko to nutrieri cross-prof irelllti- ~ l8.Glazebr meliloti. Nl 19.60119 W. 1998. Stru driven Sign 20.6raham, nodulation 21 Great, R. Staravtion starvation r 22.Gu, B., J. meliloti Dc controlled receiver m 23.Hgarle , C Acrds es 24.Hecker, response 25Hengge. Osmotic cc 21(5):887. 26.Hen . stationgag e Salmone/i EShIng‘u 27.Hengge. early Stati 2‘8 Hen . TTGha Ogsee 17.Givskov, M., I. Eberl, S. Moller, L. K. Poulsen and S. Molin. 1994. Response to nutrient starvation in Pseudomonas putida KT2442: analysis of general ggoss 2 o: :l I a 0 i " . . 12,C-4 thspO 12,C-4 thspO (aerobic) (aerobic) (1% oxygen) (1% oxygen) Figure 4.3 Complementation of 12.0-4 [a double mutant S. meliloti strain carrying a Tn5-1063 and a Tn1721 insertion (ndi::Tn51063; tspO::Tn 1721) that is not responsive under all conditions tested] under low oxygen tension by thspO, after 4 and 8 hours of exposure to low oxygen tensions. Values of luminescence are defined as described in Materials and Methods. 108 120 100 a» as 80 1" § ‘9. as 6° -. g '75- 40 -~ T) :3 a: :1 20 -~ thspO thspO thspO thspO (GTS) (GTS-N) (GTS-C) (GTS+NaCI) Figure. 4.4 Complementation of 12.0-4 during nitrogen and carbon deprivation, and osmotic stress, after 1, 2, 3, and 6 hours of exposure to different stresses. 109 350 300 a 1+- I 0 § ‘79. 200 a D4 .. Q , in o‘ I 8 g; 150 - 1-1 - L: E 3 E3; 100 - 50 - pRstspO pRstspO pFlstspO pRstspO (GTS) (GTS-N) (GTS-C) (GTS+NaCI) Figure. 4.5. Complementation of 12.0-4 with pRstspO during nitrogen and carbon deprivation, and osmotic stress, after 2, 4, and 8 hours of exposure to different stresses. 110 FixL is involved in regulating expression of the ndi locus To evaluate the involvement of other signaling pathways in the control of ndl expression, plasmid p022|ux was introduced into fixL::Tn5 and ntrC::Tn5 mutant strains of S. meliloti 1021, and luciferase activity was monitored under inducing conditions overtime. A similar expression profile was found in the wild-type and ntrC mutant backgrounds during nitrogen deprivation (see Fig. 4.6). However, a much lower level of luciferase expression was found in the FixL mutant strain versus the wild-type background, under oxygen limiting conditions (Fig. 4.7). 8000 7000 . l '0 h 6000- 5000 - 4000 - Relative Light Units (IUml ‘ OD 600') 2000 ~ iooo - ,2 53 I ; 2 i wild-type ntrC::Tn5 wild-type ntrC::Tn5 (GTS) (GTS) (GTS-N) (GTS -N) Figure 4.6. Plasmid p022|ux expression in wild-type and 1021 ntrC::Tn5 mutant backgrounds after exposure to nitrogen deprivation for twelve hours. 111 3500 3000 . __ . BO . 2500- ‘ .2 f .--.-.-.-.-.--. P 8 l 5 g 2 0 0 0 ~ :Lfljfig-Ij-jfi ’J E D ::.::.....-,.-:..-1 3° 0 1500 . £3.33; 0) :1, g :E, 1000 « o ._J 04 v 500 « wild-type wild-type fixL::Tn5 fixL::Tn5 (aerobic) (1% oxygen) (aerobic) (1% oxygen) Figure 4.7 Plasmid p022|ux expression in wild-type and 1021 fixL::Tn5 mutant backgrounds after exposure to low oxygen tensions for eight hours. To further evaluate this observation, a double chromosomal (S. meliloti 1021 fixL::Tn5-233; ndi::Tn5luxAB) mutant was created by transducing the ndi::Tn5luxAB insertion into S. meliloti 1021 fixL::Tn5-233. This strain (fixL::Tn5-233; ndi::Tn5luxAB) was examined under oxygen limiting conditions. A decreased level of ndi::Tn5luxAB expression during oxygen deprivation was observed, as compared to that found in the wild-type background (see Fig. 4.8), reconfirming the data obtained with strains carrying the plasmid-borne fusion. Therefore, it appears that FixL is required for full induction of ndi::Tn5luxAB expression under low oxygen tensions. Furthermore, ndi-lux expression patterns were found to be unaltered under all other inducing conditions (carbon and nitrogen deprivation, and oSmotic stress) in the fixL mutant (see Fig. 4.9). 112 Therefore, fixL involvement appears to be specific to sensing and responding to low oxygen tensions. Relative Light Units (LUmI 1OD 600') wild-type fixL::Tn5-233 wild-type fixL::Tn5-233 (aerobic) (aerobic) (1% oxygen) (1% oxygen) Figure 4.8. Chromosomal ndi::Tn5luxAB expression in wild-type and fixL::Tn5-233 under low oxygen tension. 113 250 200 ( 150 100 Relative Light Units (LU ml I OD 600 1) 50. C22 fixL::Tn5 C22 fixL::Tn5 022 fixL::Tn5 C22 fixL::Tn5 (GTS) (GTS) (GTS—C) (GTS-C) (GTS-N) (GTS-N) (GTS+NaCI) (GTS+NaCI) Figure 4.9. Chromosomal ndi::Tn5luxAB expression in wild-type and fixL::Tn5-233 under carbon and nitrogen deprivation, and osmotic stress. 114 The specificity of TspO and FixL effects on qu expression in strain 022 Luciferase activity is very dependent on the energy status of the cell, therefore we hypothesized that the observed decrease in ndi-lux AB expression in the F ixL and TspO mutants (see Fig. 4.10) could be the result of a general effect on the physiology of the cells. In order to evaluate whether the observed low levels of luciferase activity in the TspO and F ixL mutants were due to a non- specific general physiological effect, double mutant strains CV2 tspO::Tn1721 and CV2 fixL::Tn5-233 (strain CV2 harbors a constitutive Tn5-luxAB fusion in a gene of unknown function) were created by transduction and examined under low oxygen tension (see Fig. 4.11). Relatively similar CV2-luxAB expression profiles under low oxygen tensions were observed in both mutants. Therefore, the decrease in expression of ndl locus in the FixL and TspO mutants is likely to be the direct result of these mutations. Furthermore, it should be noted that the high levels of luciferase activity from the CV2 fusion (100x greater than ndi- luxAB) indicates that these mutant strains (tspO::Tn1721 or fixL::Tn5-233) have enough reducing power to catalyze a considerable amount of luciferase activity. 115 80 701 :10 60- : '34 l ml 3% 50. .3 3 _ ‘ “Q €00. 401 55% 304 ‘65 —-D iii-3 20- 1°: - Egg; )‘ I 0, [:1- “55555 . .‘auu wild-type tspO::Tn1721 fixL::Tn5-233 wild-type tspO::Tn1721 fixL::Tn5-233 (aerobic) (aerobic) (aerobic) (1% oxygen) (1% oxygen) (1% oxygen) Figure 4.10 ndi::Tn5luxAB expression in wild-type strain, and tspO and fixL mutant strains. 1400 12001 ,. l“Oi 1000 1 n8 800 ‘ I-IlI-L-l Relative light units (LU mI‘l O.D. 600 ") 8 8 O C 8 o 0 33333§iiiiiiii 1}sz 33k CV2 tspO::Tn1721 CV2 fixL::Tn5-233 CV2 tspO::Tn1721 CV2 fixL::Tn5-233 (aerobic) (aerobic) (1% oxygen) (1% oxygen) Figure 4.11 CV2-luxAB expression in different mutant strains under low Oxygen tensions. 116 Phenotypic analysis of strain 12,0-4 Growth experiments indicated that strains 022 and 12,0-4, as well as R-022 have growth rates that are similar to that of the wild-type strain. In addition, survival experiments indicate that these mutants are not impaired in the ability to persist during starvation (Fig. 4.12). Strain 12, 0-4 was also screened for its symbiotic phenotype by inoculation on alfalfa (Medicago sativa). Six alfalfa seedlings grown on sterile Whatman filter paper, in test tubes were inoculated with wild-type or mutant strain. Six additional plants were used as uninoculated controls. The experiment was performed twice. Both wild-type and mutant strain 12,0-4 infected plants were found to be nodulated and green, while the control (uninfected) plants were not nodulated and chlorotic after seven weeks. Therefore, the mutated gene (tspO) in strain 12,0-4 does not appear to be absolutely required for the formation and function of the symbiotic association. 117 1.0E+09 1.0E+O8 1* __E E +wild-iype7f U 1.0E+07 ,_,.___022 -- ~ ————--——- : +1204 ' j --x—-Fl-C22 # 1.0E+06 T , . . , O 5 10 15 20 25 Days Figure 4.12. Survival of wild-type S. meliloti strain 1021, 022, 12,0-4, and R- 022 during stationary phase. The strains were grown in GT8 medium containing limiting amounts of carbon (0.025% glucose). A 25ml culture was grown in a 125ml flask with shaking at 28°C for 18 days. Viable cells of each culture were determined at regular intervals by plating a dilution series on TY medium. The data were recorded as colony forming units per ml (CPU/ml). The experiment was performed in triplicate. 118 DISCUSSION In order to identify regulatory components for the induction of genes during nutrient deprivation, successive rounds of transposon mutagenesis were used. A S. meliloti mutant (022) was identified after the first round of mutagenesis, which harbors a Tn5luxAB insertion in a gene that is induced by a variety of stresses; including: carbon, nitrogen, or oxygen deprivation, osmotic stress, and stationary-phase. This locus was cloned, its DNA sequence determined, and the expression was examined in various regulatory mutant backgrounds. It was found to consist of two ORFs which we have designated ndiA and ndiB for nutrient neprivation induced genes A and B. The genes of this locus are novel and their expression is, in part, controlled by F ixL, the sensor protein of a well described two-component system in S. meliloti, known to control gene expression during oxygen limitation (11, 15). A second round of mutagenesis was used to identify additional genes involved in the regulation of the ndi locus. Strain 022 was mutagenized with a second transposon (Tn1721), and a library of 3000 double mutants was screened for altered lux expression. A double mutant (12,0-4) was identified that does not express luciferase under any conditions tested. This 12,0-4 locus was then cloned, its DNA sequence determined, and the site of the Tn1721 insertion was found to be in a gene with a high degree of similarity to the mitochondrial benzodiazepine receptor (29), as well as to the outer membrane oxygen sensor protein (tspO) of Rhodobacter sphaeroides (54). 119 To our knowledge, this is the first report of TspO involvement in signal transduction, in S. meliloti. Recently, however, this same locus was identified by Oke & Long (1999) in a screen for S. meliloti genes expressed predominantly in the nodule around the time of bacA expression (during the intermediate stages of nodule development). The function and role of this locus in the development of the symbiotic association is still under invesngauon. The TspO outer membrane receptor does not appear to be ubiquitous, in nature. Homologues of the pK18 [TspO have been found in vertebrates and invertebrates; however, they have not been observed in Saccharomyces cerevisiae or in the majority of the genomes of prokaryotes whose DNA sequence has been completed. To date, orthologues of pK18 are only known to exist in purple photosynthetic bacteria and the cyanobacterium Synechocystis. Since a member of the alpha-subdivision of purple bacteria is the likely source of the endosymbiont that gave rise to the mammalian mitochondrion (51), the finding of the pK18 orthologue in R. sphaeroides, a member of this family, was of great interest. Moreover, Yeliseev et al., (1997) demonstrated that the pK18 gene from rat could complement a TspO deficient strain of R. sphaeroides, indicating functional homology. The finding of the pK18 orthologue in S. meliloti is also intriguing. On the one hand, given the fact that members of the genus Sinorhizobium are closely related to the alpha-subdivision of purple bacteria it is not surprising that similar proteins are retained in these organisms. On the other hand, since 120 only few prokaryotes contain this protein, it may be evolutionarily significant that mitochondria, members of the alpha-subdivision of purple bacteria that are the likely source of the endosymbiont that gave rise to the mammalian mitochondrion, and S. meliloti, an endosymbiont, all contain this protein. In Rhodobacter spp., TspO is located in the outer membrane and is associated with the major outer membrane porins. In the mitochondrion the data indicate that pK18 is also localized to the outer membrane and is associated with the voltage-dependent anion channel (VDAC) (29). Moreover, both of these complexes bind and transport dicarboxylic tetrapyrrole intermediates of the heme biosynthetic pathway. This has been proposed as the likely mechanism by which TspO regulates gene expression in Rhodobacter (53). Based on this information, I propose that TspO may regulate gene expression in S. meliloti by the same mechanism, that is, porphyrin transport. A hypothetical model of this regulation, including the role of FixL, is depicted in Fig. 4.12. It has been reported that, under stress conditions, bacterial cells produce and secrete porphyrins (Pronk et al., 1998 and references there in). We propose that heme itself or an intermediate of the heme biosynthetic pathway acts as an effector molecule or triggers the synthesis of another effector molecule that binds to the repressor, and then inhibits transcription of the ndl locus. On the other hand, efflux of porphyrin molecules, which appears to require TspO, is triggered and the molecules are transported out of the cell. When the concentration of porphyrins, or putative effector molecules is decreased, the repressor can no longer inhibit 121 transcription and expression of the ndi locus resumes. In the TspO mutant, however, the concentrations of porphyrins remains high under physiologically stressful conditions, therefore there is constant repression of the Ndi locus. In addition, FixL appears to be required for the full induction of the ndl locus under low oxygen tension. It is likely that FixL, either through the regulator FixJ or via another pathway, directly controls ndl expression when the oxygen tension is low. However, TspO appears to be epistatic to FixL, since no expression is observed in the TspO mutant, even though FixL is still functioning. The mechanism by which TspO controls the transport of porphyrins and the nature of the signal that TspO senses still remains to be determined. In addition, further investigations are necessary to determine if these two related TspO homologues in S. meliloti and Rhodobacter spp. actually function in a similar manner, and respond to the same signals. Nonetheless, these two outer membrane receptors are involved in regulating gene expression and are likely to provide a new and important way to think about signal transduction in prokaryotes. 122 signals: N - and C- deprivation, . high osmolarity, or low oxygen tension ‘ TspO: Controls efflux of porphyrins, hence, de-repression FixL: Controls activation (under low oxygen tension romoter O = putative effector molecule Figure 4.12. Hypothetical model of ndiAB regulation by TspO and FixL. See text for a detailed description of the above model. . = porphyrin molecule. 123 REFERENCES 1.Altschul, S. F., T. L. Madden, A. A. Schaffer, J. Zhan , Z. Zheng, W. Miller, and D. J. Li man. 1997. Gapped BLAST and PSI-B ST: a new generation of protein data ase search programs. Nucleic ACIdS Res. 25:3389-3402. 2.Armstrong, G. A., M. Alberti, F. Leach, and J. E. Hearst. 1989. Nucleotide sequence, organization, and nature of the protein products of the carotenoid biosgntheSis gene cluster of Rhodobacter capsulatus. Mol Gen Genet. 216(2- 3):2 4-268. 3.Atlas, R. M., and R. Bartha. 1998. Microbial ecolo y: fundamentals and applications, 4th ed. Benjamin/Cummings, Menlo ark, CA. 4.Beringer, J. E. 1974. R factor transfer in Rhizobium Ieguminosarum. J. Gen. Microbiol. 84:188-198. 5.Boyer, H. W., and D. Roulland-Dussoix. 1969. A complementation analysis of 2h7e2restriction and modification of DNA in Escherichia coli. J Mol Biol. 41:459- 1.- 6.Broughton, W. J., and M. J. Dilworth. 1971. Control of Ieghemoglobin synthesis in snake beans. Biochem J. 125:1075-1080. 7.de Bruijn, F. J., and S. Rossbach. 1994. Transposon mutagenesishp. 387- 405. in . Gerhardt (ed), Methods for General and Molecular Bacteriology. American SOCIety for Microbiology, Washington, DC. 8.de Brui'n F. J., S. Rossbach M. Schneider P. Ratet, S. Messmer, W. W. Szeto, ,.llil.Ausubel, and_J. Schell..1989. Rhizobium meliloti 1021 has three differentially regulated locilnvolved in glutamine biosynthesis, none of which is essentia for symbiotic nitrogen fixation. J Bacteriol. 171(3):1673—1682. 9.Ditta, G., S. Stanfield D. Corbin, and D. R. Helinski. 1980. Broad host range DNA cloning system for gram-negative bacteria: construction of a gene bank of Rhizobium meliloti. Proc. Natl. Acad. Sci. USA. 77:7347-7351. 10.Finan, T. M., E. Hartwie , L. LeMieux, K. Be man, G. C. Walker, and E. R. 18; n1e2rb1192831. General ransduction in Rhizo ium meliloti. J. Bacteriol. 11.Fischer, H.-M. 1994. Genetic regulation of nitrogen fixation in Rhizobia. Microbiol Rev. 58(3):352-386. 12.Friedman, A. M., S. R. Long, S. E. Brown, W. I. Buikema, and F. M. Ausubel. 1982. Construction of a broad host range cosmid cloning vector and its use in the genetic analysis of Rhizobium mutants. Gene. 18:289-296. 13.Gage, D. J. T. Bobo, and S. R. Long. 1996. Use of reenfluorescent protein to Visualize the earlyevents of symbIOSls between R izobium meliloti and alfalfa (Medicago saliva). J. Bacteriol. 178:7159-7166. 14.Gamier, M., A. Dimchev, N. Boujard, M. J. Price, N. A. Musto, and V. Papadopoulos. 1994. In Vitro reconstitution of a functional peri heral-type grenzzgsiigsepine receptor from mouse Leydig tumor cells. Mol armacol. 15.Gilles-Gonzalez, M. A., G. S. Ditta, and D. R. Helinski. 1991. A haem rotein With kinase actiVIt encoded by the oxygen sensor of Rhizobium meli oti. Nature. 350:170- 72. 124 16.Gray J. X., M..A. Djordjevic, and B. G. Rolfe. 1990..Two ene that regulate exopolysaccharlde production in Rhizobium sp. strain NG 2342DNA sequences and resultant phenotype. J Bacteriol. 172:193—203. 17.Gray T. R. G. 1976. Survival of ve etative microbes in soil, p. 327-364. In T. R. G, Gray and J. R. Postgate (ed) he Survival of Vegetative Microbes. ESOClety for General Microbiology Symposium 26). Cambridge University ress, Cambridge. 18.Hanahan, D. 1983. Studies on transformation of Escherichia coli with plasmids. J. Mol. Biol. 166:557-580. 19.Hastings, J. W., and G. Weber. 1963. Total quantum flux of isotropic sources. J Opt Soc Amer. 53:1410-1415. 20.Hecker, M., and W. s. a. U. Volker. 1996. Heat-shock and general stress response in Bacillus subtilis. Mol Microbiol. 19(3):417-428. 21 .Kaiser, D. 1986. Control of multicellular development: Dictyostelium and Myxococcus. Annu Rev Genet. 20:539-566. 22.Kaneko, T., S. Sato, H. Kotani, A. Tanaka, and E. Asamizu. 1996. Sequence analysis of the genome of the unicellular cyanobacterium Synechocystis sp. strain PCC6803.II. Se uence determination of the entire genome and assignment of potentia protein-coding regions. DNA Res. 3:109-136. 23.Lapidus, A., N. Galleron, A. Sorokin, and S. D. Ehrlich. 1997. S uencing and functional annotation of the Bacillus subtilus genes in the 200 b rrnB- dnaB region. Microbiology. 143:3431-3441. 24.Loewen, P. C., B. Hu, J. Strutinsky, and R. Sparlin . 1998. Regulation in the rpoS regulon of Escherichia coli. Can. J. Microbiol. :707-717. 25.Lonsq S. R. 1996. Rhizobium symbiosisznod factors in perspective. Plant Cell. (10):1885-1896. 26.Lynch J. M. 1988. The terrestrial environment, p. 224-248. In J. M. Lynch and J. . Hobbes (ed), Microorganisms in Action: Concepts and Applications. Blackwell Scientific, Oxford. 27.Matin, A. 1991. The molecular basis of carbon-starvation-induced general resistance in Escherichia coli. Mol Microbiol. 5:3—10. 28.Matin, A., E. A. Auger, P. H. Blum, and J. E. Schultz. 1989. Genetic basis of Seaervation survival in nondifferentiating bacteria. Ann Rev Microbiol. 43:293- 29. McEnery, M. W., A. M. Snowman, R. R. Trifiletti, and S. H. Snyder. 1992. Isolation of the mitochondrial benzodiazapine receptor: Assomation With the volta e-de endent anion channel and the adenine nucleotide carrier. Proc Natl cad Cl USA 89:3170-3174. 30.Meade H. M., S. R. Lon , G. B. Ruvkun, S. E. Brown, and F. M. Ausubel. 1982. Physical and gene ic characterization of symbiotic and auxotrophic mutant 0 Rhizobium meliloti induced by transposon Tn5 mutageneSis. J. Bacteriol. 149:114-122. 31.Meighen, E. A. and P._ V. Dunlap. 1993. Phxsiological, biochemical and genetic control oi bacterial bioluminescence. dv Microb Phys. 34:1 -67. 32.Metting, B. 1985. Soil microbiology and biotechnology, p. 196-214. In R. P. Ovellette and P. A. Cheremisinoff (ed), Biotechnology: Applications and Research. Technomic Publishers, Lancanster, PA. 125 33.Milcarnps,.A., and F. J. deBrui'n. 1999. Identification of a novel, nutrient deprivation induced gene (hmg ) in Sinorhizobium meliloti, involved in Tyrosine degradation. Microbiology. 145:935-947. 34.Milcamps, A., D. M. Ragatz, P. Lim, K. A. Berger, and F. J. de Bruijn. 1998. Isolation of carbon- and nitrogen-deprivation-induced loci of Sinorhizobium meliloti 1021 by Tn5-luxAB mutagenesis. Microbiology. 144:3205-3218. 35.Morita, R. Y. 1993. Bioavailability of energy and the starvation state, p. 1-23. In S. Kjelleberg (ed), Starvation in Bacteria. Plenum press, New York, NY. 36.0ke, V., and S. R. Long. 1999. Bacterial enes induced within the nodules during the Rhizobium-legume symbiosis. ol Microbiol. 32(4):837-849. 37.0stling, J., L. Holmquist, L. Flardh, B. Svenblad, A. Jou r-Jaan, and S. Kjelleberg. 1993. Starvation and recove of Vibrio, p. 10 -123. In S. Kjelleberg (ed), Starvation in Bacteria. P enum, New York. 38.Parola A. L., D. G. Stump, D. J. Pepperl, K. E. Krueger, J.. W. Regan, and H. E. Lair . 1991. Cloning and sequencmg of a pharmacologically uni ue bovine peripheral-benzodiaze ine receptor isoquinoline binding pro ein. J Biol Chem. 266(21):14082-1 087. 39.Pronk, A. F., J. Stigter, A. H. Stouthamer, F. J. de Bruin, and F. C. Boogerd. 1998. Copro orphyrin excretion by Azorhizobium caulinodans under micro- aerobic con itions. Antonie van Leeuwenhoek. 74:245-251. 40.Reese, M. G. 1994. Diploma thesis. German Cancer Research Center, Heidelberg. 41 .Reese, M. G., and F, H. Eeckman. 1995. "Novel neural network algorithms for improved eukaryotic promoter Site recognition" Presented at the Seventh International Genome Sequencmg and Analysis Conference, Hilton Head Island, South Carolina. 42.Reese, M. G., N. L. Harris, and F. H. Eeckman. 1996. "Large scale sequencing specific neural networks for promoter and s lice site recognition" Presented at the Biocomputing: Proceedings of the 199 Pacific Symposium. 43. Riond, J., M. G. Mattei, M. Ka had, X. Dumont, J. C. Guillemot, G. L. Fur, D. Caput, and P. Ferrara. 1991. olecular cloning and chromosomal localization of a human peripheral-type benzodiazepine receptor. Eur J Biochem. 195:307-312. 44.Roszak, D. B., and R. R. Colwell. 1987. Survival strategies of bacteria in the natural environment. Microbiol Rev. 51(3):365-379. 45.Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecular cloning: a laboratory manual, 2nd ed. Cold Spring Harbor Press, New York. 46.Schoffl, R., W. Arnold, A. Puhler, J. Altenbuchner, and R. Schmitt. 1981. The tetracycline resistance transposons Tn 1.721 and Tn1771 have three 38- eeseBeaigr4repeats and generate five-base pair direct repeats. Mol. Gen Genet. 47.Shields, J. A., E. A. Paul, and W. E. Lowe. 1973. Turnover of microbial tissue in soil under field conditions. Soil Biol Biochem. 5:753—764. 48.Spaink, H. P., A. Kondorosi, and P. J. J. Hooykaas. 1998. The Rhizobiaceae. Kluwer Academic Press. 498 ctor M. P. 1998.. The starvation-stress response (SSR) of Salmonella, p. 33-27 , Advances in Microbial Phy3iology, vol. 40. Academic Press. 50.Sprengel R. P. Werner, P. H. Seeburg, A. G. Mukhin, M. R. Santi, D. R. Grayson, . Guidotti, and K. E. Krueger. 1989. Molecular cloning and 126 expression of cDNA encodin a 2peripheral-type benzodiazepine receptor. J. Biol Chem. 264(34):20415-2 4 1. 51 .Woese, C. R. 1987. Bacterial evolution. Microbiol Rev. 51:221-271. 52.Wolk, C. P., Y. Cai, and J.-M. Panoff. 1991. Use of a transposon with luciferase as a re orter to identif environmentall responsive genes in a cyanobacterium. roc Natl Acad ci USA 88:535 -5359. 53.Yeliseev, A. A,, and S. Kaplan. 1999. A novel mechanism for the regulation of photosyntheSls gene expreSSion by the TspO outer membrane protein of Rhodobacter sphaeroides 2.4.1. in preparation. 54.Yeliseev, A. A., and S. Kaplan. 1995. A sensory transducer homologous to the mammalian peri heral-type benzodiazepine receptor regulates . ehotos nthetlc mem rane complex formation in Rhodobacter sphaerordes .4 1 Biol Chem. 270(36):21167-21175. 55.Yeliseev, A. A., K. E. Krueger, and S. Kaplan. 1997. A mammalian mitochondrial dru rece tor unctions as a bacterial "oxygen" sensor. Proc Natl Acad Sci US .94: 101-5106. 127 CHAPTER 5 Conclusions and Future Directions 128 CONCLUSIONS AND FUTURE DIRECTIONS Transposon mediated reporter gene fusion approach We have used the reporter gene approach to identify and characterize the expression of nutrient deprivation - induced loci by mutagenizing S. meliloti 1021 with a Tn5 derivative (Tn5-1063) containing the promoterless luciferase luxAB genes (12, 26). Tn5 has been shown to be an extremely useful tool for random mutagenesis in many Gram-negative bacterial systems, including S. meliloti (2), and luciferase has proven to be a very useful reporter of bacterial gene expression (5), as well as a useful biomarker to tag and thereby track bacteria in the soil and rhizosphere (3, 13, 21). Previously, researchers in our laboratory have used Tn5-1063 successfully in Pseudomonas fluorescens to isolate transcriptional fusions whose bioluminescence is induced under nitrogen and phosphate deprivation conditions (8), as well as in S. meliloti to isolate transcriptional fusions induced during carbon and/or nitrogen deprivation (15). There are a few limitations to this transposon/reporter gene mediated approach: (i) The insertion of a transposon into a gene generally leads to inactivation, therefore genes that are essential can not be isolated with this method, since an insertion in such a gene would be lethal. However, this caveat did not pertain to these studies, since there was no selection for growth 129 during the screening procedure; (ii) the level of expression may be too low for detection of the reporter gene; and (iii) the transposon must be inserted in the correct orientation in respect to the promoter, therefore large libraries of mutants harboring transcriptional fusion need to be created and screened, which can be labor intensive. On the other hand, creating a mutation in a gene while at the same time being able to readily examine the expression profile of that gene with a reporter is a great advantage. One potential disadvantage of using luciferase as a reporter gene for the study of genes induced by nutrient-deprivation is that luciferase activity is dependent on the energy status of the cell, due to the requirement for reducing equivalents (FMNH2) to catalyze the bioluminescent reaction. Therefore, in theory, starved cells (especially carbon-deprived) are likely to be limited in stored energy and detection of enzyme activity could be misleadingly underestimated. However, by modifying our screening protocol when testing carbon-starved cells we appear to have overcome this limitation and we were clearly able to see an increase in expression during starvation. However, this too may underestimate the amount of expression, and the use of another reporter gene, such as the Green Fluorescent Protein (GFP) (5) that is not affected by the energy status of the cell may more accurately represent the level of expression. In addition, it was observed that the levels of luciferase activity in constitutive controls began to decrease significantly after 36 hrs of carbon starvation. Therefore, the duration of expression could not be assessed over a long peroid of time. 130 Search for genes involved in regulation during stress The primary goal of this project was to identify regulatory circuits controlling stress-induced genes. Using successive rounds of transposon mutagenesis we have successfully identified such regulatory loci. The work described here in this thesis has been paralleled by an investigation by Anne Milcamps in our laboratory. For her studies, she chose to identify regulatory loci controlling the expression of a locus involved in the degradation of tyrosine whose expression is induced during both carbon and nitrogen deprivation, as well as in the presence of tyrosine (14). Using the same protocols as described here, she was able to identify a gene that encodes a protein with a high degree of similarity to a transcriptional regulator (Milcamps, in preparation), which appears to regulate gene expression in response to both carbon and nitrogen deprivation. These parallel projects clearly demonstrate the usefulness of this second site mutagenesis strategy to identify various components of regulatory pathways, including specific regulators, as well as more global “sensing” and regulatory proteins. 131 TspO as a regulator of stress-induced gene expression Many of the environmental parameters bacteria encounter in soil, such as low oxygen tensions, nitrogen and carbon limitation, as well as osmotic changes, are also experienced in planta. Consequently, a number of genes have been identified during studies on symbiosis that are involved in both physiological adaptations in the free-living state, as well as for legume infection (16). These include functions such as synthesis of cytochrome complexes, amino acids, nucleotides, chaperonins (GroEL), as well as utilization of different carbon and energy sources (16). However, the potential role of these genes in the starvation/stress response of rhizobia has not, as yet, been explored. Interestingly, we have identified a locus (TspO) in S. meliloti involved in regulating gene expression during nutrient deprivation, osmotic stress, and under low oxygen tensions (see Chapter 4). At the same time, researchers in the laboratory of Sharon Long identified the same locus as being induced within the nodule during symbiosis when the rhizobia differentiate into bacteroids (18). Consequently, it appears that TspO represents a novel regulatory protein/mechanism that plays an important role in controlling gene expression during various physiologically stressful conditions, including the development of symbiosis. As described in Chapter 4, TspO is a recently identified outer membrane receptor in Rhodobacter spp. that is associated with the major porins of the cell wall, and is involved in the cell’s ability to sense oxygen levels (28, 29). A 132 homologue exists in mitochondria that is also localized to the outer membrane and is associated with a voltage-dependent anion channel (VDAC) (11). Moreover, both of these complex channels interact with these receptors to bind and transport dicarboxylic tetrapyrrole intermediates of the heme biosynthetic pathway, and this has been proposed to be the likely mechanism by which TspO regulates gene expression in Rhodobacter (27). However, the mechanism by which TspO controls the transport of porphyrins and the nature of the environmental signal to which TspO responds still remains to be determined. Rhodobacter spp. synthesize the intracytoplasmic membrane system which contains bacteriochlorophyl and carotenoids in response to a decrease in oxygen tension (30). Therefore, it follows that the pathway of porphyrin biosynthesis would be induced in response to this environmental signal. However, one of the more curious aspects of these findings is that if the same mechanism in Rhodobacter is used in S. meliloti, then the synthesis of porphyrins would be induced in S. meliloti in response to carbon, nitrogen, or oxygen deprivation, as well as during osmotic Shock. It has been documented that bacteria synthesize and excrete porphyrin molecules under certain conditions, such as iron deficiency and oxygen limitation (7). In addition, Azorhizobium cau/inodans, a unique stem-nodulating symbiont of the tropical legume Sesbania rostrata (1) has been found to excrete coproporphyrin under micro-aerobic conditions. Moreover, the expression of the genes involved in the synthesis of this porphyrin molecule are controlled by F ixL ( 17, 20). However, 133 the precise reason for this synthesis and excretion has not been determined. Under low oxygen tension, it is likely that the porphyrins are used to synthesize alternative cytochromes, in order to adapt to the change in redox. This may also be the case in response to stress, however, we are not aware of any reports of bacteria synthesizing and excreting porphyrins in response to carbon or nitrogen deprivation, or osmotic stress. On the other hand, TspO may regulate gene expression in response to stress via another unidentified mechanism, such as the one described below. PAS domains: internal sensors PAS domains are newly described sensor modules that monitor changes in light, redox potential, oxygen tensions, small ligands, as well as the overall energy status of the cell (25). Unlike the sensor domains of two- component systems, which are usually located within the periplasmic space, these modules are located in the cytosol, and therefore represent internal sensors. More than 300 PAS domains have been identified in diverse organisms (25). They have been found in all three kingdoms of life: Archaea, Bacteria, and Eukarya, and the designation PAS is simply an acronym formed from the names of the three proteins in which this domain was first recognized (25). Most PAS domains in prokaryotes are in histidine-kinase sensors, and these domains are known to detect the “signal” via a bound cofactor, such as heme or flavin. For example, F ixL contains a PAS domain that overlaps with the 134 heme binding region of FixL. When oxygen levels are low, oxygen dissociates from the heme molecule located in the PAS domain changing the conformation of the PAS domain, which results in an altered structure and increased autophosphorylation activity of the transmitter, thereby activating signal transduction (4). Even though the finding of this internal sensing domain is a significant breakthrough, there are still many questions to address. The mechanism of signaling by PAS domains to downstream components of signal transduction pathways is not well understood, and the sensory role of the PAS domain for a variety of environmental signals remains to be determined. Since TspO is involved in “sensing” many of the environmental parameters detected by the PAS domain, it is possible that TspO may also contain a similar “sensing” domain. Even though database searches have not identified a PAS domain in TspO, it may be that another unidentified domain is contained within this protein that is also involved in sensing changes in redox or the overall energy status of the cell. Summary Prokaryotes have a remarkable capacity to adjust themselves, both structurally and functionally, to changes in their surroundings. This holds true especially in their response to one of the most common environmental constraints that they encounter, namely, nutrient deprivation. Research over the 135 past two decades has provided strong evidence that not only spore-forming, but non-spore forming bacteria as well, differentiate into a stress-resistant, stasis-survival state (6, 10, 19). Moreover, because nutrients are scarce in most natural habitats, this starved state very likely represents the more typical physiological state of many bacteria in nature (22). Since the workings of the biosphere rely heavily on the activities of diverse microbes (9), understanding how different types of bacteria are able to monitor, sense, and respond to their environment in this more typical state is fundamentally important. We postulated that bacteria that are indigenous to soil are likely to have evolved particular mechanisms for persisting in this environment. Therefore, we chose to study nutrient-deprivation-induced gene expression in a ubiquitous soil bacterium, S. meliloti, and our investigations have, indeed, discovered a mechanism by which this bacterium modulates gene expression in response to a variety of stress conditions that is distinct, and is shared with only a few closely related bacteria. Finding this TspO/pK18 orthologue in S. meliloti that appears to work in concert with the two-component oxygen-sensing system FixLJ, and is apparently involved in multiple stress signaling pathways, in addition to oxygen sensing, is intriguing, and will likely form the basis for a variety of future studies pertaining to stress-induced gene expression in this bacterium. Furthermore, although stasis-survival research in S. meliloti has only just begun, there is great potential for rapid progress in this well studied organism. This is possible because not only because numerous 136 environmentally controlled genes have been identified through studies on the development of the symbiotic association, but, in addition, the recent advances in determining the entire genomic sequence of this organism in combination with genomic arrays and DNA chip technology (23, 24) provides a significant advancement in technology that will make it quite feasible to address these issues of global regulation of gene expression during nutrient-deprivation. 137 REFERENCES 1.de Bruijn, F. J. 1989. the unusual symbiosis between the diazatrophic stem- nodulating bacterium Azorhizobium caulinodans 0R8571 and its host, the tro ical legume Sesbania rostrata, . 457-493. In E. Nester and l. Kosuge (e .), Plant-Microbe Interactions, v0. 3. McGraw Hill, New York. 2.de BruinDn, F. J., and S. Rossbach. 1994. Transposon mutagenesis, p. 387- 405. in . Gerhardt (ed), Methods for General and Molecular Bacteriology. American Society for Microbiology, Washington, DC. 3.De Weger, L. A., P. Dunbar, W. F. Mahaffee, B. J. J. Lugtenber , and G. S. Sa Ier. .1991. use of bioluminescence markers to detect Pseu omonas spp. in t e rhizosphere. Appl Environ Microbiol. 57:3641-3644. 4.Gong, W., B. Hao S. S. Mansy, G. Gonzalez, and M. A. Gilles-Gonzalez. 1998. Structure of a biological oxy en sensor: A new mechanism for heme-driven Signal transduction. Proc Na l Acad Sci USA 95(15177-15182). 5.Jansso_n, J. J., and F. J_. de Bruijn. 1999. Biomarkers and biore orters to track microbes and monitortheirgene expression, p. 651-665. n J. Davies (ed), Manual of Industrial Microbiology and Biotechnology, 2nd ed. ASM ress, Washington DC. 6.|KjeI(;eberg, S. 1993. Starvation in bacteria. Plenum Press, New York and on on. 7.Kragelund L., B. Christoffersen, O. Nybroe, and F. J. de Bruijn. 1995. Isolation ofI lux reporter gene fusions in Pseudomonas fluorescens DF57 i1n86uCible by nitrogen or phosphorous starvation. FEMS Microbiol Ecol. 17:95- 8.Madigan, M. T., J. M. Martinko, and J. Parker. 1996. Brock: Biology of Microorganisms, eigth ed. Prentice Hall, Upper Saddle River, NJ. 9.Matin, A., E. A. Auger, P. H. Blum, and J. E. Schultz. 1989. Genetic basis of et1aé'vation survival in nondifferentiating bacteria. Ann Rev Microbiol. 43:293- 10.McEnery, M. W., A. M. Snowman, R, R. Trifiletti, and S. H. Snyder. 1992. Isolation of the mitochondrial benzodiazapine receptor: ASSOClatlon With the volta e-de endent anion channel and the adenine nucleotide carrier. Proc Natl cad Cl USA 89:3170-3174. 11.Meighen, E. A., and P._ V. Dunlap. 1993. Ph siological, biochemical and genetic control of bacterial bioluminescence. dv Microb Phys. 34:1-67. 12.Meikle, A., L. A. Glover, K. Killham, and J. I. Prosser. 1994. Potential luminescence as an indicator of activation of genetically modrfled. 2Psesu8c11ogiggiznas fluorescens in liqUid culture and in sorl. Sorl Biol Biochem. 13.Mi|camps, A., and F. J. deBrui'n. 1999. Identification of a novel, nutrient deprivation induced gene (hmg ) in Sinorhizobium meliloti, involved in Tyrosine degradation. Microbiology. 145:935-947. 14.Milcamps, A., D. M. Ragatz, P. Lim, K. A. Berger, and F. J. de Bruijn. 1998. Isolation of carbon- and nitrogen-deprivation-lnduced loci of Sinorhizobium meliloti 1021 by Tn5-luxAB mutagenesis. Microbiology. 144:3205-3218. 15.Niner, B. M., and A. M. Hirsch. 1998. How many Rhizobium Genes, in addition to nod, nif/fix, and exo, are needed for nodule development and function? Symbiosis. 24:51-102. 138 16.0'Brian, M. R. .1996. Heme synthesis in the rhizobium-le ume symbiosis: a palette for bacterial and eukaryotic pigments. J Bacteriol. 1 8(9):2471-2478. 17.0ke, V., and S. R. Long. 1999. Bacterial enes induced within the nodules during the Rhizobium-legume symbiosis. o| Microbiol. 32(4):837-849. 18.0stling, J., L. Holmquist, L. Flardh, B. Svenblad, A. Jou r-Jaan and S. K'elleberg. 1993. Starvation and recove of Vibrio, p. 10 -123. In S. elleberg (ed.), Starvation in Bacteria. P enum, New York. 19.Pronk, A. F., J. Stigter, A. H. Stouthamer, F. J. de Bruijn, and F. C. Boogerd. 1998. Copro orphyrin excretion by Azorhizobium caulinodans under micro- aerobic con itions. Antonie van Leeuwenhoek. 74:245-251. 20.Prosser, J. I. 1994. Molecular marker systems for the detection. of genetically modified microorganisms in the environment. Microbiology. 140:5-17. 21 .Roszak, D. 8., and R. R. Colwell. 1987. Survival strategies of bacteria in the natural environment. Microbiol Rev. 51(3):365-379. 22.8chena, M., D. Shalon, R. W. Davis, and P. 0. Brown. 1995. Quantitative monitorin of ene expression patterns with complementary DNA microarray. Science. 70: 67-470. 23.8halon, D., S. J. Smith, and P. 0. Brown. 1996. A DNA microarray system for analyzingcomglex DNA sam les usugg two-color fluorescent probe hybridization. enome Res. :639—64 . 24.Taylor, B. L., and I. B. Zhulin. 1999. PAS domains: Internal sensors of oxygen, redox potential, and light. Microbiol Mol Biol Rev. 63(2):479-506. 25.Wolk, C. P., Y. Cai, and J.-M. Panoff. 1991. Use of a transposon with luciferase as a re orter to identif environmentall responsive genes in a cyanobacterium. roc Natl Acad Cl USA 88:535 -5359. 26.Yeliseev, A. A., and S. Kaplan. 1999. A novel mechanism for the regulation of photosynthesis gene expreSSion by the TspO outer membrane protein of Rhodobacter sphaeroides 2.4.1. in preparartion. 27.Yeliseev, A. A., and S. Kaplan. 1995. A sensory transducer homologous to the mammalian peri heral-type benzodiazepine rece tor regulates _ hotos nthetic mem rane com lex ormation in Rho obacter sphaerOIdes .4.1. Biol Chem. 270(36):21 67-21175. 28.Yeliseev, A. A., K. E. Krue er, and 8. Kaplan. 1997. A mammalian mitochondrial dru receptor unctions as a bacterial "oxygen" sensor. Proc Natl Acad Sci US 94:5101-5106. 29.2eilstra-Rgalls J., M. Gomelsky, J. M. Eraso, A. Yeliseev J. O'Gara, and S. Kaplan._19 8. Control of photos stem formation in Rhodobacter sphaerOIdes. J Bacteriol. 80(11 :2801-2809. 139 HICH Ian STATE UNIV. LIBRARIES lllllllllllllllllllllllllllllllllllllllllllllllllllllllllHI 1293018103881 13