...:3...: 2. . ... . Ix... .5 .2. _.....i.“. . .35... “Emu.“ £31123. :3. z: s» 21...}... ; I .. mm... M. was“? El? b Ii. THESIS Z 2600 Willi”!!!"lull!!!ll!MlUllllllllllllllll 3925 This is to certify that the dissertation entitled DEVELOPMENTAL REGULATION OF STREPTOMYCES COELICOLOR ANTIBIOTIC SYNTHESIS BY THE AbsA TWO—COMPONENT SYSTEM presented by Todd B. Anderson has been accepted towards fulfillment of the requirements for mm ; degree in 11121.01:inng Mééwm ajor professor / Date244?L 4.}an 0 MS U is an Affirmative Action/Equal Opportunity Institution 0-1277 1 LIBRARY Michigan State University PLACE IN RETURN BOX to remove this checkout from your record. TO AVOID FINES return on or before date due. MAY BE RECALLED with earlier due date if requested. DATE DUE DATE DUE ' ' DATE DUE 11/00 c/CIRCJDatoDuepfis-plu DEVELOP) SY} DEVELOPNIENTAL REGULATON OF Streptomyces coelicolor ANTIBIOTIC SYNTHESIS BY THE AbsA TWO-COMPONENT SYSTEM By Todd B. Anderson A DISSERTATION Submitted to Michigan State University In partial fulfillment of the requirements For the degree of DOCTOR OF PHILOSOPHY Department of Microbiology 2000 DEVELOP. Si Strepton temporally com coelz'color, the c globally regular absA produce d igmlhiotic gm! kimse gene. Ir mutations in [h to lock AbsAl Site suppressor Abs' mUlants v ABSTRACT DEVELOPMENTAL REGULATON OF Streptomyces coelicolor ANT IBIOTIC SYNTHESIS BY THE AbsA TWO-COMPONENT SYSTEM By Todd B. Anderson Streptomycetes synthesize antibiotics in a growth-phase dependent manner temporally commensurate with, but spatially independent of sporulation. In Streptomyces coelicolor, the absA locus encodes a two-component signal transduction system which globally regulates antibiotic production independent of morphogenesis. Mutations in absA produce drastically opposing phenotypes. Mutations responsible for an Abs' (antibiotic synthesis deficient) phenotype were previously localized to the absAI histidine kinase gene. In this study Abs' mutants C542 and C577 were shown to contain point mutations in the region of absAI encoding the transmitter domain; these were proposed to lock AbsAl into a kinase dominant, phosphatase deficient enzymatic state. Second- site suppressor mutations, sab (suppressor of ng), that restored antibiotic synthesis to the Abs‘ mutants were previously mapped very near to the absA locus. Sequence analysis identified several sab mutations in the absAI or absAZ genes; the latter encodes a two- component response regulator. A genetic analysis of the absA locus was performed to examine the mechanism of AbsA regulation. A disruption in absAZ caused the precocious hyperproduction of pigmented antibiotics (Pha phenotype), demonstrating that AbsA2 negatively regulates antibiotic synthesis. Gene replacements in absA, that disrupted phosphoryl- group transfer, also caused the Pha phenotype, demonstrating that the phosphorylated form of AbsA2 is the active negative regulator of antibiotic synthesis. These Pha strai: sinthesis, in ad ln-ritro phospi‘. meated AbsA biochemical ex" phosphoryl-gro ope back groun also possesses . demonstrated 1: With-phase r Pigmented anti contrast to its a activate transcr F652, and Cda} Emporally I61; phosphorl'latic high levels COi Tmscfipllm] i Pa‘allel with ll .ibsA_ The p0 SWis in th: 9 s . . OILEnaUOnO These Pha strains demonstrated temporal acceleration of calcium-dependent antibiotic synthesis, in addition to previously characterized acceleration of pigmented antibiotics. In-vitro phosphorylation experiments utilizing a maltose-binding protein fusion to truncated AbsAl ('AbsAl) and a His-tag fusion to AbsA2 provided preliminary biochemical evidence supporting AbsAl autokinase activity and AbsAl-AbsAZ phosphoryl-group signal transduction. High-copy expression of absAI alleles in a wild type background resulted in a Pha phenotype, providing genetic evidence that AbsAl also possesses AbsA2-P phosphatase activity. 81 nuclease mapping of the absA locus demonstrated leaderless cotranscription of absAI and absAZ. Transcription of absA was growth-phase regulated, experiencing a dramatic increase prior to the appearance of pigmented antibiotics. Transcription of the absA locus was also autoregulated. In contrast to its activity as a negative regulator of antibiotics, phospho-AbsA2 appeared to activate transcription of abSA. AbsA regulation of pathway-specific regulators redD, redZ, and cdaR was examined by $1 nuclease protection assays. Expression of redD was temporally retarded with respect to absA and was clearly regulated by AbsA2 in a phosphorylation-dependent manner. Conversely, redZ transcription was expressed at high levels coincident with absA expression and showed no dependence on AbsA. Transcription of the cdaR pathway-specific regulator was also temporally regulated in parallel with the absA time course. Transcription of cdaR does not appear to depend on AbsA. The possible role and mechanism of AbsA-mediated regulation of antibiotic synthesis in the S. coelicolor life cycle is discussed. Preliminary evidence supporting the conservation of absA in other species of Streptomyces is presented. ldedicate this healthy [3813]}: also dedicatec achieve this g Anderson, in ldedieate this motivated me I dedicate this work to my beautiful boys, Todd and David, who helped me keep a healthy perspective and balance in my life over the course of this project. This work is also dedicated to my wife, Marywbska, whose support and sacrifice allowed me to achieve this goal; and to my mother, Alice Anderson, and the memory of my father, Ellef Anderson, in appreciation of the values and work ethic they helped instill in me. Finally, I dedicate this work to my sister, Pamela, whose commitment to good science has always motivated me. iv lwm patience and Renqiu Kori; support. I ti also grateful knowledge : Dr. David .1 lhiern for i Plfimids U: ACKNOWLEDGEMENTS I would like to thank my mentor, Dr. Wendy Champness, for her guidance, patience and support during this period of my professional development. Thanks to ' Renqiu Kong, Marywbska Calderon, and Mark Kazmierczak for valuable technical support. I thank Paul Brian for the lab training that helped me get up and nmning. I am also grateful to N. Jamie Ryding, Gary Brown, and David Aceti for their sharing of knowledge and ideas. My appreciation goes out to the members of my thesis committee, Dr. David Amosti, Dr. Michael Bagdasarian, Dr. Michael Thomashow, and Dr. Suzanne 'I‘hiem for helpful guidance and discussion. Thank you to Mervyn Bibb for providing plasmids used in this study. LIST OF TABLE llST 0F F lGUK KEY TO AB BRI NlRODL'CTlO Regulatio CHWIER 2 SEQUENCE M MLTANT S ...... CHER 3 A GENETIC A} OF AbsA-MED] lNTROD MATERJ G A P. D C S R S RESLU N a G ll P1 1 Pi E H. S G DISCL'Sc1 TABLE OF CONTENTS LIST OF TABLES ..................................................................... ix LIST OF FIGURES .................................................................... x KEY TO ABBREVIATIONS ........................................................ xii lNTRODUCTION ..................................................................... 1 Regulation of Cellular Difi'erentiation and Antibiotic Production ...... 4 Factors triggering difl‘erentiation ................................... 5 Uncoupling of differentiation: regulation of antibiotics. . . .. 16 Two-Component Signal Transduction Systems ........................... 24 The AbsA two-component system of S. coelicolor .............. 32 CHAPTER 2 SEQUENCE ANALYSIS OF absA ALLELES OF Abs' AND sab MUTANT S .............................................................................. 36 CHAPTER 3 A GENETIC AND TRANSCRIPT ANALYSIS INTO THE MECHANISM OF AbsA-MEDIATED ANT IBIOTIC REGULATION ........................... 40 INTRODUCTION ............................................................. 41 MATERIALS AND METHODS ............................................. 45 Growth Conditions .................................................... 45 Antibiotic Assays ...................................................... 46 Plasmid and DNA Manipulations .................................... 46 Disruption of 0193/12 in C500 ......................................... 50 Construction of an In-Frame Deletion in absAIin C530 .......... 51 Site-Directed Mutagenesis ............................................ 54 RNA Isolation ........................................................... 58 $1 Nuclease Protection Assays ....................................... 59 RESULTS ........................................................................ 61 Negative Regulation of Antibiotics by the AbsA2 Regulator and AbsAl Histidine Kinase ......................................... 61 Genetic Evaluation of the Role of Phosphorylation in AbsA2- Mediated Regulation ................................................... 63 Precocious Hyperproduction of Calcium-Dependent Antibiotic, Undecylprodigiosin, and Actinorhodin in absA Mutants. . . .. 64 Precocious Hyperproduction of Antibiotics Resulting from AbsA Domain Overexpression ............................................... 66 High Resolution 81 Nuclease Mapping of the absA Transcription Start Site ................................................................. 69 Growth-Phase Dependent Expression and Autoregulation of absA. 71 DISCUSSION ..................................................................... 73 vi ClLlPIER 4 TEMPORAL EXP REGULATORS A LN'TRODL' ll-L-iTERlA RESULTS De , on De DISC L'SSl CHIPIER 5 0\EREXPRESSI OEAbsAl AND .L‘ INTRODL MATERLA Bat CHAPTER 4 - TENTPORAL EXPRESSION OF red AND cda PATHWAY-SPECIFIC REGULATORS AND THEIR DEPENDENCE ON AbsA ......................... 80 INTRODUCTION ............................................................... 81 MATERIALS AND METHODS ............................................... 83 RESULTS ......................................................................... 84 Dependence of Pathway-Specific Regulators redZ and redD on AbsA ................................................................. 84 Dependence of cdaR Expression on AbsA .......................... 87 DISCUSSION ................................................................... 90 CHAPTER 5 OVEREXPRESSION, PURIFICATION, AND PHOSPHORYLATION OF AbsAl AND AbsA2 PROTEINS .................................................. 95 INTRODUCTION ............................................................... 96 MATERIALS AND METHODS .............................................. 98 Bacterial Strains ......................................................... 98 Construction of AbsA2-His”) and AbsA2-Hi35 Expression Plasmids ................................................................. 99 Construction of MBP-‘AbsAl Expression Plasmids ............... 101 Culture Media and Growth Conditions .............................. 102 Purification of MBP-‘AbsAl Proteins ................................ 103 Purification of AbsA2-His Proteins ................................... 104 'AbsAl and AbsA2 Phosphorylation Assays ........................ 105 RESULTS ......................................................................... 106 Overexpression and Purification of AbsA2-His“) from E. coli. .. 106 In- Vitro Phosphorylation of AbsA2-His“) with Acetyl Phosphate. 108 Overexpression and Purification of AbsA2-His; from S. lividans.. 109 In- Vitro and In- Vivo Analysis of AbsA2—His6 ........................ 111 Overexpression and Purification of 'AbsAl from E. coli ............ l 13 Autophosphorylation of 'AbsAl and Phosphorylation of AbsA2-His6 ................................................................ 115 DISCUSSION ....................................................................... 1 17 CHAPTER 6 AbsA2 HOMOLOGUES IN OTHER STRAINS OF STREPTOMYCES ........... 120 INTRODUCTION .................................................................. 1 2 1 MATERIALS AND METHODS ................................................. 123 Bacterial Strains and Growth Conditions ............................... 123 PCR Amplification of Putative absA2 Homologs ..................... 124 absAZ Homolog Identification and Sequencing ........................ 125 vii CHAPTER 7 CONCLUSION 1 MOlCCUlIil for the At The Role . Model Su: AbsA2 Ta The AbsA REFERENCES. .. RESULTS ............................................................................ 127 PCR Amplification of Putative absAZ Homologs from Heterologous DNA ........................................................ 127 Identification of Putative absAZ Homologs ............................. 128 Sequence Analysis of Putative Homologs .............................. 130 DISCUSSION ....................................................................... 1 34 CHAPTER 7 CONCLUSION AND FUTURE RESEARCH ........................................... «138 Molecular Genetic Characterization of absAI Mutations Responsible for the Abs' Phenotype and Certain sab Suppressors of Abs' .................. 139 The Role of Phosphorylation in the AbsA Regulatory Mechanism ........... 141 Model Summary ..................................................................... 143 AbsA2 Targets ....................................................................... 146 The AbsA Signal and Signal-Sensing Mechanism .............................. 150 REFERENCES ................................................................................ 153 viii TABLE 1. Oligo LIST OF TABLES TABLE 1. Oligonucleotide primers used in this study ................................. 47 ix Figure I. ll Figure 2. T Figure3. P figure 4. T Figure 5.Pt Figure 6. I Figure 7. C Figure 8. I Figure 9. "i Fig; 10. Figure 11. Figure 12. Elm: 13, Figure 14, Flgurels Figure 16 Fl ’1'? re 17 FlSure 1 8 Flg‘dre 19 Figure 20 LIST OF FIGURES Figure l. Hierarchy of antibiotic regulatory loci ...................................... 15 Figure 2. Two-component signal-transduction proteins .............................. 26 Figure 3. Paradigm two-component signal-transduction system activity ........... 30 Figure 4. The AbsA two-component system ....... ................................... 33 Figure 5. Position of absA within the cda gene cluster ................................. 44 Figure 6. The absA locus and plasmid inserts based on absA ........................ 49 Figure 7. Creation of the absAZ disruption in strain C500 ............................ 51 Figure 8. The absA IA530 in-frame deletion ............................................. 52 Figure 9. The effect of an absA gene disruption and gene replacements on antibiotic production ............................................................ 62 Figure 10. Calcium-dependent antibiotic assays in Pha mutants ....................... 65 Figure 11. The effects of high-copy expression of absA alleles on antibiotic production ...................................................................... 68 Figure 12. SI nuclease protection mapping of the (16571 locus ........................ 70 Figure 13. High-resolution Sl nuclease protection analysis of the absA transcript. 72 Figure 14. Position of absA within the cda gene cluster ................................ 82 Figure 15. High-resolution S1 nuclease protection analysis of the redD and redZ transcripts ............................................................. 86 Figure 16. High resolution 81 nuclease protection analysis of the cdaR transcript. 89 Figure 17. Plasmids for 'absAI and absAZ overexpression ............................ 100 Figure 18. Purification of AbsA2 proteins overexpressed in E. coli ................. 108 Figure 19. Purification of AbsA2-Hi36 overexpressed in S. lividans ................. 110 Figure 20. In-vivo analysis of pAB8270 .................................................. 112 Figure Zl. Figure 33. Figure 33. Figure 24. Figure 25. Figure 26. ‘ Figure 21. Figure 22. Figure 23. Figure 24. Figure 25. Figure 26. Purification of 'AbsAl expressed in E. coli ................................. l 14 In-vitro phosphorylation of AbsA2-Hi56 by 'AbsAl-phosphate .......... 1 15 PCR amplification of putative absAZ homologs ..................... . ...... 128 Southern blots of putative abSAZ homologs ................................. 130 Amino acid sequence alignment of S. coelicolor AbsA2 with putative homologs from S. amobofaciens, S. griseus, S. peucetius, and S. clavuligerus .................................................................. 132 Model of AbsA-mediated antibiotic regulation in S. coelicolor .......... 144 xi Abs‘ Act CDA HI H Mary 0er Pha PSR Red sub SARP TM LIST OF ABREVIATIONS antibiotic synthesis deficient actinorhodin calcium-dependent antibiotic histidine kinase helix-turn-helix methylenomycin Oxoid nutrient agar precocious hyperproduction of antibiotics pathway-specific regulator undecylprodigiosin response regulator suppressor of abs Streptomyces antibiotic regulatory proteins transmembrane xii CHAPTER 1 INTRODUCTION The ‘ order Act/no may ofbiol oi the thous. half of the b metabolites antiparasitic herbicides, I"SEIher wit other actino if e also Und characterize the lncidem “31111111131 81 Sire’pmm'ycl DIESem We the fabUlOu lino“ acti (filling the} Sire The genus Streptomyces is composed of high-G+C, Gram-positive bacteria of the order Actinonrycetales. Streptomyces are most noted for their ability to produce a diverse array of biologically active secondary metabolites. This genus accounts for over one-half of the thousands of naturally produced antibiotics discovered and is the source of over half of the bio-active compounds currently in clinical use (165). Streptomyces metabolites important to health and agriculture include antibacterial, antifungal, antiparasitic, and antitumor drugs, immunosuppressive agents, insecticides, and even herbicides. The abundance of naturally occurring compounds produced by this genera, together with the relative ease of culturing streptomycetes (in comparison to fungi or other actinomycetes), makes them a prime target for bioexploration. Numerous efforts are also underway to produce novel metabolites through genetic engineering of well characterized Streptomyces biosynthetic pathways (82). Given the alarming increase in the incidence of antibiotic resistance in common human pathogens, coupled with the continual emergence of new human, animal and plant pathogens, it is critical that Streptomyces continue to be exploited as a source of new drugs. Unfortunately, at present we are able to culture only a handful of Streptomyces. Therefore, to fully harness the fabulous diversity of chemical compounds synthesized by this genus, and other little known actinomycetes, we must gain a better understanding of the biological processes driving their growth and development. Streptomyces are notable for their morphological similarity to filamentous fungi. Spore germination and germ tube extension is followed by vegetative growth through hyphal extension and branching to form a dense weave of substrate mycelium. Vegetative growth frequently ends in response to some nutrient limitation (90a) at which time the cell enle" the Production of? biomass EICCUmUla mycelium. the CO“ differentiation Whl physiological ditIC emerge from the Si 34; 37). Meanvvhil metabolically distii secondary metabol Streptonr_t‘c (reviewed by 150; consistencies in the evolutionary con se 0rgatiizationally, l: toEither on the chr actinOfliodin genes identified (cataIOgr meme blOSYntheri slIlllIESlS is grOBTl‘; expreSSlOn of m0 5 L’e-nsi I I. genus (30)- Likeu time the cell enters the second phase of biphasic biomass accumulation coincident with the production of aerial mycelium. DNA replication continues throughout the period of biomass accumulation (90a; 112). Prior to the appearance of antibiotics and aerial mycelium, the colony undergoes the initial stages of a complex program of cellular differentiation which has been divided into two distinct processes - morphological and physiological differentiation (31). During morphological differentiation, aerial hyphae emerge from the substrate mycelium, terminally septate, and develop spores (reviewed by 34; 37). Meanwhile, in the substrate mycelium, the temporally parallel but spatially and metabolically distinct process of physiological differentiation results in the synthesis of secondary metabolites. Streptomyces produce several chemically distinct categories of antibiotics (reviewed by 150; 173; 163). In spite of this tremendous diversity, there are certain consistencies in their genetic organization and regulation that suggest substantial evolutionary conservation in the mechanism controlling their synthesis. Organizationally, biosynthetic genes encoding a particular antibiotic are clustered together on the chromosome. Since first demonstrating the clustered organization of actinorhodin genes (109), a great many Streptomyces antibiotic gene clusters have been identified (cataloged by 119; 72). Clusters occupy approximately 15 to 100 kb and include biosynthetic genes, resistance determinants, and regulatory genes. Antibiotic synthesis is growth-phase dependent, occun'ing during the stationary phase. Temporal expression of most regulatory genes for antibiotic production are upregulated during transition phase growth. Cluster-specific regulatory features are common throughout the genus (30). Likewise, preliminary evidence using PCR and Southern hybridization techniques suggest present in numeror bidet, (99); absA 3. introduced into het suggesting some le together, these I’CSL may be conserved a review advances i 0rf-fartism S. coy/rev AbSA m'O‘COmpon antibiotic productit‘ Regulation of Cell Regulation tiers. At the earlies jointly regulated I V ‘ . egetative and/or tr unthesis from me r Sps‘tlli C r eSlllators r1 l . C. i. rereases greatly d Show - llll . (A techniques suggests that homologs of many antibiotic pleiotropic regulatory genes are present in numerous streptomycetes (absB, (136); cfs‘Q, (84); afiR, (115) ; cutRS, (3 2); bIdA, (99); absA2, see Chapter 6). Moreover, cloned antibiotic gene clusters that were introduced into heterologous host streptomycetes showed proper temporal expression, suggesting some level of regulatory conservation (109; reviewed in 119). Taken together, these results imply that regulatory mechanisms governing antibiotic production may be conserved at various stages of development. The remainder of this chapter will review advances in the field of Streptomyces antibiotic regulation with a focus on model organism S. coelicolor. Special attention will also be given to the topic of this study, the AbsA two-component signal transduction system, and its contribution to regulation of antibiotic production in S. coelicolor. Regulation of Cellular Differentiation and Antibiotic Production Regulation of antibiotic production in streptomycetes can be divided into three tiers. At the earliest stages of development, antibiotic synthesis and sporulation are jointly regulated. The precise timing of the uncoupling of joint regulation during the vegetative and/or transition phases has not been widely studied. Uncoupling of antibiotic synthesis from morphogenesis is characterized by two levels of regulation; pathway- specific regulators control the synthesis of a single antibiotic, where as pleiotropic regulators influence the production of two or more. Expression of these regulators increases greatly during the transition phase and continues well into stationary phase growth, which is characterized by the onset of antibiotic accumulation. In order to understand the role of any individual regulator on differentiation, it is valuable to have an appreciation of its Therefore, Ivvill e» the onset of dift ere only in antibiotic p Factors triggering It has been carbon, nitrogen, p metabolism (revie\ mechanism ofiniti required in second repressed by grow free assays with at presence ofglucog did not appear to 1 and references the antibiotic synthes: fiddie, ammOHIUn used to generate F Conversely, addit min. appreciation of its temporal and physiological significance in the global process. Therefore, Iwill examine in greater detail, various physiological processes involved in the onset of differentiation prior to focusing attention on specific regulators implicated only in antibiotic production. Factors triggering differentiation It has been well established that exhaustion of one or more nutrients such as carbon, nitrogen, phosphorous, or trace elements leads to the onset of secondary metabolism (reviewed in 152). Easing of catabolite repression was demonstrated as one mechanism of initiating antibiotic synthesis through derepression of specific enzymes required in secondary metabolism. Production of numerous Streptomyces antibiotics was repressed by growth on glucose, and certain other carbon sources (reviewed by 46). Cell- free assays with antibiotic synthases showed that a number of these were repressed in the presence of glucose. In contrast to the inducible enzymes of enteric bacteria, cyclic-AMP did not appear to play a role in relieving carbon source repression in Streptomyces (45 and references therein). Readily assimilable ammonium salts also negatively affected antibiotic synthesis in numerous strains (reviewed by 151; 46). In tylosin producing S. fradie, ammonium repressed several enzymes associated with catabolism of amino acids used to generate propionate and butyrate precursors of the macrolide ring (103). Conversely, addition of branched chain amino acids promoted tylosin production in this strain. In general, greater production of antibiotics was favored by growth on complex, slowly assimilable nitrogen sources. The biosynthesis of a large number of antibiotics was subject to regulation by phosphate (reviewed by 111). Those chemical classes of antibiotics Cf macrolideS a much l€5S Set biosynthetic 5 classes of anti function at the early in Colon}. may be due In l genes. or repres Given ti stationary phase secondary metal: examined whetht of nutrient depm venezue/ae ( l 72). both induced at lo granaticin was prc absence of nutrien Cases where some mduce antibiotic s Stn'ngent r response 10 low 2' IFSPOHSC IS 3550C? antibiotics especially sensitive to phosphate included aminoglycosides, tetracyclines, macrolides and polyenes, while antibiotics directly assembled from amino acids were much less sensitive. Phosphate was shown to repress phosphatases required in biosynthetic steps of aminoglycosides, and several other biosynthetic enzymes from other classes of antibiotics. However, in the synthesis of macrolides, phosphate seemed to function at the level of primary metabolism by inhibiting precursor formation. Thus, early in colony development, growth-phase dependent regulation of antibiotic synthesis may be due in part to catabolite repression of antibiotic biosynthetic and/or regulatory genes, or repression of pathways involved in precursor formation. Given that antibiotic production is associated with the low growth rates of stationary phase metabolism, it was thought that nutrient deprivation may also induce secondary metabolism through a growth-rate sensitive response. Several studies examined whether antibiotics could be induced by low growth rate alone in the absence of nutrient deprivation. In continuous culture experiments with S. cattleya (106) and S. venezuelae (172), synthesis of cephamycin C and cloramphenicol, respectively, were both induced at low grth rates independent of nutrient limitations. Likewise, granaticin was produced by S. thermocyolaceus in response to low growth rate in the absence of nutrient deficiencies (85). In contrast, Demain and Fang (1995) cite various cases where some form of nutritional stress was required along with low growth rate to induce antibiotic synthesis. Stringent response is a common mechanism involved in bacterial adaptation in response to low growth-rate or nutrient deprivation. In Enterobacteriaceae, the stringent response is associated with the accumulation of the intracellular effector molecules (:3)me in regulation 01 some cases r griserrs ( l 29, subunit that 1 synthesis of l coelicolor rel {Red} synthe: was unable tc were also imp to reduced lev Were made in PPGPP Synthe Simulation ar CDA Synthesi to make Act 0 CODdIIIOna] d6 87 156715, Si n Ce ‘ (1996) witnegS acturnulatiOn C was no obligau SIC/mm? ’rus shift—down (1‘?) (p)ppGpp (reviewed by 24). In Streptomyces, the involvement of ppGpp-dependent regulation of antibiotic synthesis and sporulation appeared to be species specific and in some cases metabolite specific. Mutations in the relC gene of S. coelicolor (128) and S. griseus (129) produced "relaxed" mutants with altered L11 protein in the SOS ribosomal subunit that were impaired in their ability to bind RelA - an enzyme that catalyzes the synthesis of pppGpp from GTP (or GDP) and ATP during the stringent response . The S. coelicolor relaxed mutants were deficient in actinorhodin (Act) and undecylprodigiosin (Red) synthesis and unable to form aerial hyphae. Similarly, the S. griseus reIC mutant was unable to produce streptomycin or form aerial hyphae. Unfortunately, these mutants were also impaired in growth so that it was uncertain whether their phenotypes were due to reduced levels of (p)ppGpp or general growth deficiencies. More recently, deletions were made in the reIA gene of S. coelicolor (113; 114; 27; 26) that completely eliminated ppGpp synthesis. Martinez-Costa, et al. (1996) found that the relA deletion reduced sporulation and severely affected Act production, but had little or no effect on Red and CDA synthesis. Chakraburtty and Bibb (1997) reported that the reIA null mutant failed to make Act or Red; however, only under nitrogen limiting conditions. A similar conditional dependence on ppGpp may also extend to streptomycin production in S. griseus, since contrary to the finding of Ochi (1990b) mentioned above, Neumann et al. (1996) witnessed substantial streptomycin production without any significant accumulation of ppGpp in S. grisezrs when grown on minimal medium. Likewise, there was no obligatory relationship between antibiotic production and ppGpp accumulation in S. clavuligerus when grown in defined or complex medium and subjected to nutritional shift-down (12). Therefore, (p)ppGpp may be required to trigger sporulation and/or synthesis Of‘ role appears 1 products of 5: Own mechanism fc effector mole levels and elic normally acyl cytoplasmic n sensing is four hanerr‘. This accumulate in the phosphOr}. reaching thresl their indepmdl 5611501' kinasesg ei‘PFeSSlon 0ft QUOI‘un reSlllallng d€\'e are rel’le‘aed bv synthesis of certain antibiotics under a particular set of culture conditions; however, its role appears to be species dependent, nutritionally conditional, and limited to certain products of secondary metabolism. Quorum sensing, cell-density-dependent gene expression, is another common mechanism for the initiation of developmental responses. In quorum-sensing systems, effector molecules accumulate in the medium until such time that they reach threshold levels and elicit a response. In Gram-negative bacteria the effector molecules are normally acyl-homoserine lactone autoinducers that are able to diffuse across the cytoplasmic membrane (58). A well characterized example of Gram-negative quorum sensing is found in autoinducer regulated gene expression of bioluminescence in Vibrio harveyi. This organism produces two signaling molecules, AL] and AI-2, that accumulate in the medium during vegetative growth (13). In the absence of autoinducer, the phosphorylated response regulator protein, LuxO, represses bioilluminescence. Upon reaching threshold concentrations, AL] and AI-2 activate the phosphatase activity of their independent cognate sensor molecules LuxN and LuxQ (hybrid two-component sensor kinases) that are able to integrate their signal through LuxO to derepress expression of bioilluminescence genes(56). Quorum sensing is also a common mechanism among Gram-positive bacteria for regulating developmental gene expression. Examples of the best characterized processes are reviewed by Kleerebezem, et al. (1997) and include genetic competence in Bacillus subtilis and Staphylococcus pneumoniae, virulence response in Staphylococcus aureus, and the production of peptide antibiotics by various species of Gram-positive bacteria. In contrast to their Gram-negative counterparts, these cell-density dependent systems incorporale a p‘ atrrusible N'aci dedicated ATP- s}.'stem5 Which i mommponenl regulélte gene ex the processed PC the peptide pher' cell-density deli“ mechanism has I hierarchical regu does appear to in below). Likewis (http://vmw. sang ABC transporters Interesting involving peptide lactone analogs, y signaling molecul species (eg, S. W Q’aneofuscams (6 differentiation is f Induction of aeria' incorporate a post-translationally processed peptide as a signaling molecule. Unlike the diffusible N-acyl homoserine lactones, these peptide pheromones are secreted by dedicated ATP-binding-cassette exporters. Another common characteristic of these systems which is not as ubiquitous in the Gram-negative mechanisms is the integration of two-component signal transduction systems as sensor and response mechanisms to regulate gene expression. The histidine kinase (HK) of these systems is thought to sense the processed peptide pheromone and activate response regulator-mediated expression of the peptide pheromone, the two-component system, and the genes responsible for the cell-density dependent phenotype (96). A peptide pheromone quorum-sensing mechanism has yet to be conclusively demonstrated in Streptomyces; however, a hierarchical regulatory cascade controlling aerial mycelium formation in S. coelicolor does appear to incorporate small peptide signaling molecules (185; 126; discussed below). Likewise, the S. coelicolor genome sequencing project (http://www.sanger.ac.uk/Projects/S_coelicolorl) has uncovered numerous homologs to ABC transporters with as yet unassigned functions. Interestingly, a variant of the Gram-positive quorum sensing systems not involving peptide pheromones is found in the genus Streptomyces. N-acyl homoserine lactone analogs, y-butyrolactones, have been implicated as cell-density-dependent Signaling molecules in morphological and/or physiological differentiation in several species (e. g., S. virginiae (189; 188); S. viridochromo (61); S. bikiniensis and S. cyaneofuscatus (60)). The best known example of butyrolactone regulation of differentiation is found in S. griseus where A-factor autoregulator is essential for the induction of aerial mycelium formation and streptomycin production (reviewed by 75; I30). A-factOI the Ciiol’lasmi spurulation in z grisetrs (92) “in regulatory funC in the cyIOPlaS’l independent lOC Subsequent binc promoter region specific activato (177; 130). A-fa vegetative grout genetically progr accumulates to m where it acts on A of S. virginiae, vvl mycelium) and is It is uncert SWP’OHU’CLJS di in lllSllIlCi frOm A‘f" t ‘ I Wing exPerime: L, MFG/0r (14) 1- 130). A-factor is described as a microbial hormone that is able to diffuse freely across the cyt0plasmic membrane. It was originally discovered by its ability to cause sporulation in a bid mutant (deficient in sporulation and antibiotic production) of S. griseus (92) when wild type and mutant were streaked side-by-side. A—factor exerts its regulatory function through an interaction with A-factor binding protein (ArpA), present in the cytoplasm of the cell. In the absence of A-factor, ArpA acts as a repressor of independent loci required for activation of aerial mycelia and streptomycin genes. Subsequent binding of A-factor to ArpA causes dissociation of the repressor from the promoter region of adpA. AdpA then activates expression of the streptomycin pathway- specific activator StrR and is also implicated in activating aerial mycelium formation (177; 130). A-factor is produced in low concentrations during the “decision phase” of vegetative growth, described by Neumann, et al. (1996), as an integral part of a genetically programmed pathway required for the onset of differentiation. Later, it accumulates to much higher concentrations during the transition and stationary phases where it acts on ArpA. This is in contrast to VB, the y-butyrolactone signaling molecule of S. virginiae, which induces production of the antibiotic virginiamycin (but not aerial mycelium) and is synthesized just prior to the antibiotic itself (94). It is uncertain how widespread y—butyrolactone signaling is in the regulation of Streptomyces differentiation. S. coelicolor produces six such molecules (8) that are distinct from A-factor, but are able to complement an A-factor deficient mutant in cross- feeding experiments (66). One of these, expressed only during transition and stationary phase growth, was purified and shown to stimulate both Act and Red production in S. coelicolor (l4). Horinouchi and Beppu (1992) predicted y-butyrolactone hormonal lO regulators I0 be regulatory “50 60% of streptor say that as mOI’t will be found IC particular regul; Wherea: induction of dif quomm-sensing through a progr cell ages. EV’lClt convincingly de for programmec The idea developmental p became commit nUtrient limitatit IIutrient to (term, Who POstulated t pIOdUCfiOn Was l exponential Qrou t2 .- auxtc growth (L regulators to be general constituents of morphological and/or secondary metabolite regulatory cascades. More conservatively, Yamada, et al. (1997) estimated that about 60% of streptomycetes produce butyrolactone signaling molecules. It is probably safe to say that as more systems are studied, this type of quorum-sensing signaling mechanism will be found to act both as global inducers of differentiation and as signals for a particular regulatory cascade of a single antibiotic. Whereas stringent response and easing of catabolite repression represent induction of differentiation by way of a particular stress, synthesis of A-factor, as a quorum-sensing signal or microbial hormone, seems to induce cellular differentiation through a programmed developmental cycle; a regulatory cascade that progresses as the cell ages. Evidence for this type of programmed developmental cycle has been convincingly demonstrated in S. griseus. Inconclusive, but growing, evidence also exists for programmed development in S. coelicolor. The idea that S. griseus differentiation was under the control of a timed developmental program was first proposed by Ensign (1988). He demonstrated that cells became committed to sporulation at 10-12 hours after spore germination in the absence of nutrient limitations and that this timing was not varied by 10X additions of any individual nutrient to defined medium. That study was further elaborated by Neumann, et al. (1996) who postulated that A-factor-induced aerial mycelium formation and streptomycin production was initiated during the first stage (the "decision phase") of diauxic exponential growth in S. griseus. The authors demonstrated, using an A-factor deficient mutant, that low concentrations of A-factor were required during the first stage of diauxic growth (up to 10 hours in the growth medium tested) for cellular differentiation 11 to take place, ever entered stationary decision phase, tli by the concentrati lag (IO-24hr), dtit not induce differe I production, L-val prior to the diauxi et al. (1996) note; perceivable nutrit doubling the conc proposed that A-f programmed difr‘t mOlilhological an exllOIIential grow A second 1 Pattern of comppe Sm’P’Omvces b] 1 v C formation; most a collaborators (1 84' hierarchical casc glotem . rEqu 1 red f on the ability ofc to take place, even though appreciable quantities were not produced until the culture entered stationary phase at 24 hours. The earlier that A-factor was supplied during the decision phase, the higher the final yield of streptomycin, although yield was not affected by the concentration of A-factor during this phase. Addition of A-factor after the diauxic lag (IO-24hr), during the second stage of exponential growth (the "execution phase"), did not induce differentiation. In the same manner, known inhibitors of streptomycin production, L-valine and staurosporin (a kinase inhibitor), were only effective if added prior to the diauxic lag (i.e., during the "decision phase"). Like Ensign (1988), Neumann, et al. (1996) noted that the onset of the diauxic lag occurred in the absence of any perceivable nutritional or physical perturbation, and its timing could not be altered by doubling the concentration of any individual nutrient or all nutrients. The authors proposed that A-factor is an inducer of cellular differentiation that forms part of a programmed differentiation cycle in which the decision by the cell to switch on morphological and physiological differentiation is made during the first phase of diauxic exponential growth. A second example of programmed development may come from the hierarchical pattern of complementation witnessed in the bid (bald) mutants of S. coelicolor (185). Streptomyces bld mutants are so called because they are deficient in aerial mycelium formation; most are also blocked in antibiotics (reviewed by 31; 35). Losick and collaborators (184; 185; 127; 126) believe that many of the bld mutants are involved in a hierarchical cascade of extracellular signaling that controls SapB synthesis - a small protein required for the initiation of aerial mycelium formation. This hypothesis is based on the ability of certain bld mutants to complement the aerial mycelium deficiency of 12 other bid strains T medium(184i 18' cultivation of am‘ ordered by mutati (and possibly cm the first signal in spanning transpor associated with h, tRVA that transla be determined xvi: identified in antib bIdG mutations a: antisigma genes (7. the hierarchy, b/u' molecule that is h formation to all t}. SapB SYnthesis (1 ability to Suppress antibiotic defects Several Ch other bld strains through "cross feeding" when grown next to one another on solid medium (184; 185), or by preconditioning agar medium with one b/d mutant prior to cultivation of another (1 85; 126). The hierarchy of steps in the regulatory cascade are ordered by mutations in bld261, -K, -A, -H, -G, -C, and -D. Briefly, bid 26] is required (and possibly encodes) for the synthesis of an extracellular factor that is believed to be the first signal in the cascade (126). The bldK gene encodes an ABC membrane- spanning transporter (127) that is possibly involved in importing the extracellular signal associated with b1d261. The most extensively studied bld gene, bldA, encodes the only tRNA that translates the rare UUA leucine codon of Streptomyces (101; 104). It is yet to be determined what function BldA has in this cascade, but UUA codons have been identified in antibiotic pathway-specific regulators redZ (182) and actII-ORF 4 (52). The bldG mutations are complemented by a locus that shows similarity to antisigma/anti- antisigma genes of Bacillus subtilis (28; 30 and references therein). The final gene of the hierarchy, bldD, is thought to code for or regulate synthesis of an extracellular molecule that is heat and protease resistant and capable of restoring aerial mycelium formation to all the other bld mutants of the cascade, presumably by its ability to restore SapB synthesis (185). Interestingly, these complementation studies demonstrated the ability to suppress morphological differentiation without pleiotrOpic suppression of antibiotic defects. Several characteristics of the bid mutants suggest their involvement in a programmed developmental cycle. First, bld mutations tend to be pleiotropic for morphological and physiological differentiation, and therefore represent early components of the developmental process where these pathways are still coupled l3 (Figure ll T that are UUCOL Spomiarion) a 134), Likewis can Suppress b differentiation Very few trans: above, b/dD tra although this Wt began to appear predominantly u second leg of the Therefore, given regulatory genes. signaling cascade that is active duri: Obviously, there Part of a developr st: ' ess. Likewise redilation of ant?‘ (Figure 1). There are many other regulators of either sporulation or antibiotic production that are uncoupled from each other (discussed below). Whi mutants (impaired in sporulation) are able to extracellularly compliment bld mutations (e. g., whiF C99, 33; 184). Likewise, overexpression of antibiotic pleiotropic or pathway specific regulators can suppress bid mutations (29; 65), demonstrating that bld genes control cellular differentiation upstream of uncoupled sporulation or antibiotic regulatory pathways. Very few transcript studies have been performed on the bld genes. Of those mentioned above, bldD transcription was most prominent at the earliest time point tested, 15 hr, although this was predicted to have already been in transition phase since aerial mycelia began to appear by 18 hr and antibiotics by 24 hr (48). Antibiotic regulatory genes are predominantly up-regulated during the transition phase, which likely corresponds to the second leg of the diauxic curve ("execution phase") described by Neumann, et al. (1996). Therefore, given that the bid genes act upstream of both whi genes and antibiotic regulatory genes, and the latter are expressed during transition and stationary phase, the signaling cascade proposed by Willey, et al. (1993) may form part of a genetic program that is active during the decision phase to commit the culture to differentiation. Obviously, there is still much work to be done to define whether the proposed cascade is part of a developmentally programmed pathway or is induced by nutritional or physical stress. Likewise, it will be exciting to see how this pathway is integrated into the global regulation of antibiotics. l4 PleiOtrO regulal’ Figure 1. Hi ofantibiotic prod .lutations in ma: leiotropic regulz pathway-specific antibiotic gene cl erar blrl genes antibiotics sporulation Pleiotropic abaA regulators “55A absB absC/mia ast/K/R2 afsQ l redZ Pathway-specific ““114 cdaR ? mmyR regulators whi genes rerID Red Act CDA Mmy Figure l. Hierarchy of antibiotic regulatory loci. Genetic loci involved in the regulation of antibiotic production in S. coelicolor can be divided into a three-tier hierarchy. Mutations in many bld genes block both antibiotic production and sporulation. Pleiotropic regulators influence the production of more than one antibiotic. Genes for the pathway-specific regulators are linked to the biosynthetic genes of their respective antibiotic gene clusters and regulate only the antibiotic that they are associated with. 15 Uncoupling or d In additio differentiation dc also subject to ur‘ detail in model 0' sporulation, will i Losick (1997). I production center lividtms, but will differences that 0 process to other 5 Champness (199‘ Streplonrt undecylpfodigios ,oi. (hit Uncoupling of differentiation: regulation of antibiotics In addition to the joint regulation of morphological and physiological differentiation demonstrated by bld mutants, sporulation and antibiotic production are also subject to uncoupled regulatory mechanisms. These have been studied in greatest detail in model organism S. coelicolor. Numerous genes uniquely involved in sporulation, whi genes, have recently been reviewed by Chater (1998) and Chater and Losick ( 1997). This section will concentrate principally on the regulation of antibiotic production centered around mechanisms discovered in S. coelicolor and closely related S. lividans, but will also incorporate examples from other species to highlight similarities or differences that will aid in conferring the current state of understanding of this complex process to other species within the genus. Reviews on this topic can be found in Champness (1999); Bibb (1996); Chater and Bibb (1997); and Hopwood, et al. (1995). Streptomyces coelicolor synthesizes four distinct antibiotics, actinorhodin (Act), undecylprodigiosin (Red), calcium-dependent antibiotic (CDA), and methylenomycin (Mmy). The gene clusters for Act and Red have been cloned (109; 110) and the biochemical pathway for Act synthesis has been extensively studied (reviewed by 72). An advantageous characteristic of Act and Red is that they are pigmented (blue and red, respectively, at alkaline pH) which facilitates phenotypic screens for factors affecting their synthesis. The sequence of cda genes has recently been revealed through the ongoing Streptomyces sequencing project (http://www.sanger.ac.uk/Projects/S_coelicolor/). Previous work had partially cloned and characterized this locus (39), but its regulatory and biochemical characteristics are still largely unknown. Methylenomycin carries the distinction of being the only 16 Seaplane?“ 3"“ plasmid (95) pre 201th region and A commr more pathway-s; that antibiotic. ( species. Most as of S. per/celius h. andrea’Z. Most l more ofthe oper. of the methyleno apparently only c while the others i requred for expp dnrl transcriptior Dnrl REIdZ and response requ lat c (65; 59). A num" pleiotropic reout ~. c Stggest they Con: my - and dttr biosynth. Streptomyces antibiotic whose sequence is carried on a plasmid, SCPl, a 350 kb linear plasmid (95) present at a copy number of four. The mmy gene cluster was localized to a 20 kb region and found to contain a regulatory gene, mmyR (3 6). A common characteristic of antibiotic gene clusters is that they encode for one or more pathway-specific regulators (PSR) that solely influence the expression of genes for that antibiotic. Champness (1999) lists PSRs for antibiotics produced in numerous species. Most antibiotic clusters encode a single PSR, although, the daunorubicin cluster of S. perrcetius has three, dnrl, dnrN, and dnrO, and the Red cluster contains two, redD and redZ. Most PSRs contain a DNA-binding motif that transcriptionally activates one or more of the operons carrying antibiotic biosynthetic genes; although at least one, MmyIL of the methylenomycin pathway is a repressor (36). Where multiple PSRs exist, apparently only one interacts with biosynthetic gene promoters (e. g., RedD and Dan) while the others regulate expression of the first. For example, Red Z and DnrN are required for expression of red and dnr biosynthetic genes because they activate redD and dnrl transcription, respectively (65; 59). They do not bypass the need for RedD and Dan. RedZ and DnrN show extensive full-length sequence similarity to two-component response regulators (RR) of the FixJ subfamily, but have lost the ability to phosphorylate (65; 59). A number of other PSRs, RedD, ActII-ORF4, Dan, SnoA, and CcaIL and one pleiotropic regulator, Ast, have recently been classified as a special group of Streptomyces antibiotic regulatory proteins (SARPs) based on amino acid similarities that suggest they contain a conserved C-terminal DNA-binding motif like the one found in the two-component response regulator OmpR (183). Similar heptameric direct repeats in act and dnr biosynthetic promoters have been predicted as binding sites for ActII-ORF4 and 17 Dnrl, resPeC‘lv' another (I66)’ 5 with. One final copy number TC suggest that thei Acting b differentiation, 3 number ofgenet affecting gromh pleiotropic regul, lirt'dans (which I perform visual sc loci which have istKRz (1 15; putative homolor: “42» each om. HERSduction path The min a cloned on a mu“; Slii'idans and S responsible fOr A Fig! Dan, respectively, and although these two SARPs can apparently substitute for one another (166), SARPs generally only regulate genes of the antibiotic they are associated with. One final observation about PSRs is that their overexpression at relatively low copy number results in dramatic increases in antibiotic production (162; 62) which may suggest that their expression is tightly controlled by the cell. Acting between the bid genes that regulated morphological and physiological differentiation, and the PSRs which influence the synthesis of individual antibiotics, are a number of genetic loci that pleiotropically regulate more than one antibiotic without affecting grth or sporulation. Remarkably, the only knowledge of antibiotic pleiotropic regulators in the genus Streptomyces comes from work in S. coelicolor and S. Iividcms (which like S. coelicolor produces Act and Red) due largely to the ability to perform visual screens for the pigmented antibiotics Act and Red in these strains. The loci which have so far been implicated in pleiotropic regulation are abaA (53), absA (2), qst/K/RZ (115; 116; 174), mia (28), and micX (143). With the exception of absB (a putative homolog of E. coli RNaseIII (136)) and micX (a possible antisense fragment (142)) each of the loci have at least a suspected involvement in phosphorylation signal- transduction pathways. The mia and abaA loci were isolated because they produced a phenotype when cloned on a multicopy plasmid (28; 53). The 2.7 kb abaA clone isolated from S. coelicolor contained five open reading frames and caused overexpression of Act in both S. Iividans and S. coelicolor when carried on a multicopy plasmid. The region responsible for Act overexpression was isolated to cotranscribed orfA and orfB (53). Fragments containing orfA and orfB, or oer alone were able to overexpress Act. 18 Furthermor e severely redl locus shows OffB did not there is signli coelicolor, m u'hiJ show sir putative OrfA In cont four antibiotic. been narrowed intergenic regi: possesses sequ. region has rece to contain the ti shows sequence however, the C. re'r‘lacement in r Promoter region Caused by the titr Containing 055C QmP‘OHEnije k Furthermore, single cross-over disruption of orfB abolished production of Act and severely reduced that of Red and CDA while having no effect on Mmy. OrfA of this locus shows similarity to the transmitter region of two-component sensor kinases (146). OrfB did not show similarity to any proteins deduced from gene databases. Interestingly, there is significant structural organization between abaA and the whiJ locus of S. coelicolor, mutants of which are impaired in sporulation. Three open reading frames of Mull show similarity to those in abaA, including Orfl which is homologous to the putative OrfA kinase (146). In contrast to abaA, the high copy clone of the mia locus abolished synthesis of all four antibiotics in S. coelicolor (28). The region responsible for the niia phenotype has been narrowed down to a 90 nt sequence of unknown fianction (30) which lies in an intergenic region, contains an open reading frame for a 20 amino acid peptide, and possesses sequence consistent with the formation of a large stem loop (29). The mia region has recently been included in the newly described absC locus (29) and is believed to contain the transcriptional start site for 01f! of absC (146). The N-terminus of ORF] shows sequence similarity to two-component histidine kinases just as in abaA ORF A; however, the C-terminal domain of ORF 1 is similar to protein phosphatases (146). Gene replacement in orfl produced an increase in Act. Since the mia fragment lies in the promoter region of absC orf], it is possible that the high-copy number mia phenotype is caused by the titration of a regulator of orfl . Therefore, both abaA and the mia- containing absC loci encode proteins that are not histidine kinases but possess two- component-like kinase domains and appear to be closely associated with the pleiotropic 19 phenotypes ob: phosphorylatio The rer associated witl threonine phos AC1 and Red or 54). In vitro, 1 product of aft} l5 33% identic; predicted to CC can-ling mum] consistent Wit} rm” e Sl'nthes not blpass the “it“ on antib devoid of Act rich medium .5 also found to r effect On Milly Through 3 (“EC Au and Red or phenotypes obtained from strains overexpressing these regions. The role of these loci in phosphorylation signal-transduction pathways is currently under investigation. The remaining pleiotropic regulatory loci all encode for proteins commonly associated with phosphorylation signal-transduction mechanisms. Ast is a serine— threonine phosphoprotein that also was isolated by its ability to cause overexpression of Act and Red when expressed on multicopy plasmids in S. lividans and S. coelicolor (78; 54). In vitro, Ast is phosphorylated by the downstream, convergently transcribed gene product of asz, a serine-threonine protein kinase (115). The N-terminal domain of Ast is 33% identical to the PSRs ActII-ORF4 and RedD (115), which have recently been predicted to contain an OmpR-like DNA-binding motif (1 83). The strain of S. coelicolor carrying multiple copies of ast produced increased levels of act/In! and redD transcripts consistent with the effect of Ast on these antibiotics. Similarly, Ast was not able to restore synthesis of Act or Red in AactII-4 or AredD mutants, demonstrating that it could not bypass the PSRs (54). A chromosomal in-frame deletion of ast had a conditional effect on antibiotic synthesis (54). On minimal medium, the ast mutant was essentially devoid of Act or Red, while on complex medium Act was only slightly delayed, and on rich medium Act and Red production were comparable to the wild-type strain. Ast was also found to reduce CDA synthesis in a similar medium-dependent nature, but had no effect on Mmy. Thus, the Ast Ser-Thr phosphoprotein appeared to regulate antibiotics through a direct or indirect interaction with the PSRs; however, it was only essential for Act and Red under certain nutritional conditions. The final set of pleiotrOpic regulatory loci, absA, afsQ, and cutRS, all encode independent two-component signal transduction systems. CutRS is a negative regulator 20 ofAct Si’mheSiS or cmR resPO”S which could be reported 8 Sim” however, recent S, coelicolor (15 production. The afSQ production of pig plasmid into S. 11 Surprisingly, no the S. coelicolor was able to suppi synthesis). Givei number, the abse mean that, like A conditions. The final regdlaior in S_ CU hum mutants that ambimics (2). L is , . We the AbsA of Act synthesis in S. Iividans (32). Gene disruptions in either cutS histidine kinase (HK) or cutR response regulator (RR) resulted in accelerated and increased production of Act, which could be suppressed by introducing the cloned culR gene. Chang, et al. (1996) reported a similar phenotype upon creating a gene replacement in cutS of S. coelicolor; however, recent evidence suggests that this locus does not regulate antibiotic synthesis in S. coelicolor (15). No information has been given as to the effect of CutRS on Red production. The afsQ locus of S. coelicolor was discovered by virtue of its ability to stimulate production of pigmented antibiotics and A-factor when introduced on a multicopy plasmid into S. Iividans (84). This locus encoded the AfsQl RR and AfsQ2 HK. Surprisingly, no phenotype was obtained when either of these genes were disrupted on the S. coelicolor chromosome. In contrast, a low copy-number plasmid carrying afsQI was able to suppress an S. coelicolor absA mutation (globally deficient in antibiotic synthesis). Given the ability of this gene to suppress the absA mutation in low copy number, the absence of a phenotype in the afsQ chromosomal disruptions may simply mean that, like Ast, this system is only essential or active under certain nutritional conditions. The final two-component system thus far identified as an antibiotic pleiotropic regulator in S. coelicolor is encoded by the absA locus. The absA locus was isolated from mutants that grew and sporulated normally but were deficient in pigmented antibiotics (2). Like the absB and mia strains, absA mutants were shown to be globally blocked in synthesis of all four S. coelicolor antibiotics. The absA locus was shown to encode the AbsAl HK and the AbsA2 RR. It was later discovered that mutations in the 21 AbsA] HK W61 (19). Gene rep expression of d overexpression antibiotics) (19 was a negative phenotypes, th uctII-ORF I V s produced (I). or indirect trar It is ap PhOSphorylatir onset of antibi sporulation 1. cascade or are repressing the independent 52 in other Gram. regulamrs, Sci: Streptomdp,Ces_ The b 8 including Spor i 50“! baCtti‘riurn AbsAl HK were responsible for the Abs‘ phenotype exhibited by the original mutants (19). Gene replacement or disruption of absA I, believed to have a polar effect on the expression of downstream absA 2, resulted in the opposite phenotype, early onset and overexpression of pigmented antibiotics (Pha phenotype; precocious hyperproduction of antibiotics) (19). Therefore, it was hypothesized that the AbsA two-component system was a negative regulator of antibiotic synthesis. Consistent with their antibiotic phenotypes, the Abs‘ and Pha absA mutants affected expression of the PSRs redD and actII-ORFIV such that their levels varied in agreement with the amount of antibiotic produced (1). Therefore, AbsA seemed to regulate antibiotic synthesis through the direct or indirect transcriptional control of the PSRs. It is apparent fi'om a review of antibiotic pleiotropic regulators that phosphorylation signal transduction plays an extremely important role in regulating the onset of antibiotic synthesis after this process has been uncoupled from that of sporulation. It is yet to be determined if these regulators comprise a linear regulatory cascade or are part of an integrated network of independent cascades either activating or repressing the production of multiple or individual antibiotics in response to numerous independent signals. Examples of two-component regulation of developmental processes in other Gram-positive organisms, in addition to data from S. coelicolor pleiotropic regulators, seems to support an integrated network of antibiotic regulation in the genus Streptomyces. The best example of regulation of growth-phase dependent differentiation including sporulation and secondary metabolite formation comes for the Gram-positive soil bacterium B. subtilis (reviewed by 192). This organism demonstrates stationary- 22 phase productiO Surfactin synthd negative control I signal transduct: various compon molecules (e. g. cell tip the whol regulatory netwt pathway are alsc competence, spc component and t (6g. surfactin s_~. signals rather th. Transcri p anAlysis of the e; phase production of a small enzymatically synthesized peptide antibiotic surfactin. Surfactin synthesis is regulated by a complex integrated network of multiple positive and negative controls involving various two-component systems and other phosphorylation signal transduction molecules (e. g., aspartyl-phosphate phosphatases). In addition, various components of this regulatory web are under the control of independent effector molecules (e.g., CodY, ComX, Cfx), whose relative concentrations inside or outside the cell tip the whole mechanism toward overall positive or negative control. The surfactin regulatory network is not physiologically isolated; instead, the molecules active in this pathway are also involved in regulation of other stationary-phase processes such as competence, sporulation, and degradative enzyme synthesis. Thus, where numerous two- component and other phosphoprotein signal transduction regulators mediate a process (e.g., surfactin synthesis), there is precedent for integration of multiple pathways and signals rather than a single linear regulatory cascade. Transcript studies suggest that pleiotropic regulators absA, afsQ, ast, and cutRS are transcribed simultaneously during transition and stationary phase growth; however, a detailed examination of their simultaneous temporal expression is lacking, as is a detailed analysis of the epistatic relationship between them. AbsA and Ast have been shown to affect the transcription of PSRs. In addition, multiple copies of ast or afsQI suppressed the absA mutation in an Abs’ strain to restore antibiotic synthesis. This would suggest that Ast and AfsQ act either downstream of AbsA in a linear cascade or in a separate cascade that integrates into a global antibiotic regulatory pathway. In support of the latter, chromosomal disruptions of afsQI and ast did not affect antibiotics under conditions in which 0123/! mutations do show a phenotype (84; 54; 2; 19). The fact that 23 45R and afsQ demonstrated 1 pathways respt deal of data is 1 suggests that tl synthesis in S. . In orde: coelicolor, it is involved. Onc each independi allow one to in identified base biochemical m tWO‘COmPOIler the Signal that “Eduction, 5 WEEKS) OfIhe Sl’stem FCgulai pathway“) m TmeOmPOi Two“ ast and afsQ are not essential for antibiotic production under conditions in which AbsA demonstrated global control suggests that they may be part of independent integrated pathways responding to different environmental signals. Therefore, even though a great deal of data is missing on the interactions of pleiotropic regulators, that which exists suggests that they do not form a single linear regulatory cascade controlling antibiotic synthesis in S. coelicolor. In order to begin constructing a unified model for antibiotic regulation in S. coelicolor, it is not sufficient to simply identify the regulatory molecules that are involved. Once identified, it is necessary to determine the biochemical mechanism of each independent system and establish the interactions that exist between systems to allow one to integrate these into the overall pathway. Of the multiple loci that have been identified based on their ability to produce a phenotype, very little is known about the biochemical mechanisms which account for their activity. With respect to the various two-component systems AbsA, AfsQ, and CutRS, questions of interest include: what is the signal that modifies HK activity; is phosphorelay the mechanism of signal transduction, and if so, what is its role in altering the activity of the RR; what is the target(s) of the RR and is its activity positive or negative; and, how is expression of this system regulated during the course of growth and in comparison to other genes of the pathway(s) they regulate? Two-Component Signal Transduction Systems Two-component signal transduction systems are ubiquitous throughout eubacteria as a means of regulating varied aspects of cell physiology in response to environmental 24 conditions, inc The paradigm ' kinase and rest component sys component prc Stock et al. (I HK prc domain and a I two to eight tr domain is hi gl specificity of a region r88pm]: understood in been best Chap transmitter do aSSOciated Wit {minimal hor mo‘compone in these Prote mimics; aUl catFried Out in axial-late phO conditions, including physical, behavioral, nutritional, and developmental responses. The paradigm two-component system (Figure 2) is composed of two proteins, a histidine kinase and response regulator. The mechanism of signal transduction in all two- component systems studied to date is phosphorelay. Extensive reviews of two- component protein structure and function are presented in Parkinson and Kafoid (1992), Stock, et al. (1995), and Volz (1995). HK proteins are generally composed of two-domains, an N-terminal sensor domain and a C-terminal transmitter domain. Sensor domains frequently contain from two to eight transmembrane helices which anchor the HK to the cell membrane. This domain is highly divergent between systems, which may in part be due to the high signal specificity of each sensor. As the name implies the sensor domain is thought to contain a region responsible for signal molecule recognition, although this mechanism is poorly understood in most two-component systems. Among those HKs where signal sensing has been best characterized are NarX (25; 102), VirA (154), and PhoR (179). The HK transmitter domain is more highly conserved and ispresponsible for the enzymatic activity associated with this molecule. Several subdomains of the transmitter share sequence and functional homology and are conserved in their organization. The H-box, present in all two-component HKs, contains a conserved His residue that is the site of phosphorylation in these proteins. The transmitter domain is associated with two independent enzyme activities; autophosphorylation at the conserved His residue which is believed to be carried out in trans between the monomers of HK homodimers (161) and phospho- aspartate phosphatase activity directed at the phosphorylated form of the R. The 25 Figure 2. Twc N’O-componer features conser consists of two transmembrane domain has als for the high SE . terminal transn “lth the amino Which is the co mains have i activity. Th e U the histidine ki rEileiver domai CC reSldLles orm the acidic e ECtor domai; regulalOrs POSS Elix‘m‘helin t{afield-pt1.01121] A. Histidine Kinase I E U |H| lNlhrl IFIIGI l Sensor Transmitter Domain Domain B. Response Regulator [lDl lDl lKlHlHTHl l Receiver Effector Domain Domain Figure 2. Two-component signal-transduction proteins. Primary structure diagrams of two-component histidine kinase and response regulator proteins demonstrate common features conserved among members of this family. (A.) The histidine kinase frequently consists of two domains. The N-terminal sensor domain contains between two and eight transmembrane helices that anchor the histidine kinase to the cell membrane. The sensor domain has also been implicated in signal sensing in some systems, which may account for the high sequence divergence of this domain among all histidine kinases. The C- terminal transmitter domain has several highly conserved subdomains indicated by boxes with the amino acid residue for which they are named. The H-box contains the histidine, which is the conserved site of phosphorylation in all histidine kinases. Other sub- domains have been implicated in nucleotide and Mg2+ binding required for enzymatic activity. The transmitter domain possesses the auto kinase and phosphatase activity of the histidine kinase. (B.) The response regulator normally consists of two domains. The receiver domain is highly conserved among all members of the two-component family. Three residues conserved in all response regulators are the Asp, Asp, and Lys, which form the acidic pocket in which the central Asp is phosphorylated. The C-terminal effector domain is responsible for the regulatory action of the protein. Most response regulators possess a DNA binding region in the effector domain, indicated here by a helix-tum-helix motif, which is consistent with the function of these proteins as transcriptional regulators. 26 phosphorylat HK to a cons description, 1 additional dc Lord and Li C-terminusc additional re accepting his accepting dc The ' terminal rec. With three if central Asp mm Ofien r trartscriptio1 Se‘ernCe Sir Wiants 0f ] SW0}: 0f 8 the receiver Ema-media: 83%“ dor abseme or. phosphorylated HK also acts as a donor for phosphoryl group transfer from the His of the HK to a conserved Asp of its cognate RR. While the majority of HKs fit the above description, there are also numerous hybrid HKs, most of which have one or more additional domains C-terminal to the transmitter domain. For example, hybrid HKs LuxN and LuxQ of Vibrio harveyi have response regulator receiver domains fused to the C-terminus of their transmitter domains (13). Likewise, BarA and Ach have an additional receiver domain, but also contain another C-terminal domain with a phospho- accepting histidine independent of the transmitter His (83). These extra phospho- accepting domains undoubtedly complicate signal transduction between the HK and R. The two-component R is usually a two-domain cytoplasmic protein. The N- terrninal receiver domain is highly conserved among virtually all members of this family, with three invariable residues, Asp, Asp, Lys, which form an acidic pocket in which the central Asp is the site of phosphorylation (Figure 2). The C-terminal effector domain most ofien contains a DNA-binding motif, such that most response regulators act in transcriptional regulatiOn. Response regulators are grouped into subfamilies based on sequence similarity within their effector domains (reviewed by 161). The most common variants of Rs are proteins which possess only the receiver or effector domain, such as SpoOF of B. subtilis (191) and GerE of E. coli (40). Those proteins which contain only the receiver domain make up part of signal-transduction cascades in which they act as intermediate signal-transduction molecules between an orthodox HK and RR. Where the effector domain exists on its own, it is able to mediate transcriptional regulation in the absence of signal. An offshoot of the signal-independent GerE protein are pseudo 27 response my domain, full-l Phosp systems (revi region of the membrane tc domain. Sui This causes : permits reco transcription suggests tha that pl’Opoge Cu (139), , bl’ Other res. activity f0” l50late the S response regulators such as RedZ and DnrN mentioned earlier (65; 59), which have dual- domain, full-length homology to RRs, but have lost the requirement for phosphorylation. Phosphorylation is the mechanism of signal transduction in two-component systems (reviewed by 161). Under the two-component paradigm (Figure 3), the sensor region of the HK senses a signal which causes a conformational change across the cell membrane to stimulate autophosphorylation at the conserved His of the transmitter domain. Subsequently, the phosphoryl group is transferred to the Asp of its cognate RR. This causes a conformational change between the receiver and effector domains that permits recognition and binding of the effector domain to its target promoter and transcriptional activation of the target gene. A review of two-component systems suggests that most HKs are phosphatase dominant in the absence of signal. Exceptions that propose kinase default HKs in the absence of signal include the EnvZ/OmpR (135), Cpx (139), and LuxN/Q/O (56) systems; however, the status of the first two are debated by other researchers (55; 132). The difficulty in unequivocally determining the default activity for the great majority of two-component systems is the inability to define and isolate the signal. Nevertheless, with the exception of Lux , in all other systems where good signal identification data exists, the signal stimulates HK kinase activity. While two-component phosphorylation-mediated signal transduction may appear at first glance to confer all or nothing regulation, a closer look at the mechanism reveals its ability to provide subtle, finely tuned responses. In part, this is due to the dual 28 Tune 3. Paradi pen-1e of signal 11; Under the: cgiator interacts aresponse regul. alts of INA ante of signa sline kinase ca train stimulates mime. The pl calator. Again fill region to re ipfled here as 1 {non in which ‘ trigalconcentra mithe histidin :i‘phanse activ fiOSphoryIated Dpnrtion of p ‘ffél‘onding our nations. Figure 3. Paradigm two-component signal-transduction system activity. (A.) In the absence of signal most histidine kinases are thought to be in a phosphatase dominant state. Under these condition, the unphosphorylated receiver domain of the response regulator interacts with the effector domain to inhibit binding to the target promoter. If the response regulator is an activator, then transcription is blocked; indicated here by the inability of RNA polymerase holoenzyme to recognize the promoter. (B.) In the presence of signal, ligand binding to a recognition site in the sensor domain of the histidine kinase causes a conformational change across the membrane. The transmitter domain stimulates autokinase activity and phosphorylation occurs at the conserved histidine. The phosphoryl group is transferred to the Asp of its cognate response regulator. Again phosphorylation causes a conformational change, now allowing the HTH region to recognize its target promoter, which effects transcriptional regulation (depicted here as activation). (C.) Module C of this diagram depicts the hypothetical situation in which the concentration of signal decreases or is lower than that in module B. A signal-concentration dependent kinase/phosphatase equilibrium is established in which part of the histidine kinase population is in the kinase state, while the remainder possess . phosphatase activity. Some phosphorylated response regulator molecules may be dephosphorylated such that the regulatory output at the target promoter reflects the proportion of phosphorylated to unphosphorylated response regulator. The corresponding output would be intermediate to signal saturation or signal deficient conditions. 29 12.3% macs— oaaEvn uo~a>=o< 22.3% .U .m “53.555 on5ai~ao5~ Enummxo connects .ft noun—=9“ coo—60¢ Gama .9.qu .0 55953... Enotmtomnauh cum—EM m8~>uu< .anwE .m noun—awe“ oz 2358.5 832323 iuwmm a: 92522 .< 30 kinase-it’ll" 51 signal Slim“I saturating le‘ eerClSe phOf signal-concer phosphatase a output interm 159). In addi characteristics more than one : HR and, therefr NarX autophosi signal associate target promoter variable not onl For example, th Entensivel y Char Sequence has be diversity With re OmpR rec0gniz Similar ity betwe- Ql require any, kinase/phosphatase activities of the HK (161). In the two-component paradigm, the signal stimulates HK kinase activity. However, unless the signal is present at sensor- saturating levels, only part of the HK population is in the kinase mode, the rest still exercise phosphatase activity toward the phosphorylated response regulator. Therefore, a signal-concentration dependent equilibrium is established between the kinase and phosphatase activities of the HK that combine to produce a corresponding regulatory output, intermediate to that of the signal saturated or signal deficient extremes (145; 159) In addition to the signal-concentration-dependent response, several other characteristics of two-component systems add to their complexity. Several systems have more than one signal that may vary in their ability to trigger autophosphorylation of the HK and, therefore, produce different degrees of response as was recently exemplified by NarX autophosphorylation in response to nitrate versus nitrite (102). In addition to the signal associated properties of the I—IK, the response of two-component regulators at their target promoters can also vary greatly. Promoter recognition sites can be extremely variable not only between different response regulators, but even for the same regulator. For example, the promoter region for at least eight operons regulated by NarL have been extensively characterized (reviewed by 159). While a weak heptameric consensus sequence has been proposed for NarL binding, heptameric sequences exhibit great diversity with respect to number, location, orientation, and spacing. Similarly, while OmpR recognizes multiple decameric sites in the ompF and ompC promoters, there is no similarity between C-box and F-box sequences (135). In addition, regulatory responses can require anywhere from a single to multiple copies of the regulator in only the 31 phosphoryle phosphoryla while some response in multiple twi mechanism: competence appear to pl outside of t. signal to a 1 The AbsA The that negatis HK is Pfed highly Con: ClOmain (a; deduced pr fideofag the 5’0me dgmatn (1s F‘gmentatil phosphorylated state (e. g., SpoOA, reviewed by 157), or a combination of the phosphorylated and unphosphorylated states (e. g., OmpR (159) and UhpA (41)). Finally, while some two-component systems seem to be the only regulator of a particular response in simple pathways (e.g., Uhp transport system (89)), other pathways integrate multiple two-component systems together with other transcription factors and feedback mechanisms to fine tune a response through a combination of interacting regulators (e. g., competence gene expression in B. subtilis (121)). Therefore, while phosphorelay does appear to play a central role in the signal transduction of all two-component systems, outside of this mechanism enormous variability exists in the process of converting a signal to a response. The AbsA two-component system of S. colelicolor The absA locus encodes a two-component signal transduction system (Figure 4) that negatively regulates production of all four S. coelicolor antibiotics (19). The AbsA] HK is predicted to be membrane bound and undergo phosphorylation at Hi5202 of the highly conserved transmitter domain. AbsAl also has a relatively large C-terminal domain (approximately 160 amino acids) that does not show sequence similarity to deduced proteins from gene databases and is of yet unknown function. A vector-bome allele of absA] used to complement the Abs' C542 and C577 mutants was truncated at the BamHI site of absA 1, which removed the last 69 amino acids of the C-terminal domain (19). Nevertheless, this truncated form of AbsAl was able to restore wild type pigmentation to the Abs‘ mutants, calling into question the requirement 32 AA P X m w X Z 2 U3 1K SCESJ 7c I I absAI m sass. 20c I B. IHI INIIDIlGI Ll El El lHTHl Sensor Transmitter C-terminal Receiver Effector Domain Domain Domain Domain Domain Histidine Kinase Response Regulator Figure 4. The AbsA two-component system. (A.) A physical map of the absA locus and surrounding genome; (B.) Primary sequence diagrams of the AbsAl histidine kinase and AbsA2 response regulator. AbsAl possesses three domains. The N-terminal sensor domain is predicted to contain four transmembratne helices. The transmitter domain contains sub-domains conserved in other histidine kinases including the H-box with the putative site of phosphorylation at His202. The unique C-terminal domain of AbsAl is of unknown function and shows no similarity to other proteins deduced from gene databases. The AbsA2 response regulator consists of two domains. The receiver domain contains the highly conserved residues of other response regulators including a putative site of phosphorylation at Asp54. The C-terminal effector domain contains a helix-tum- helix DNA-binding motif, which allows AbsA2 to effect transcriptional regulation at a target promoter(s). Lettered boxes represent highly conserved sub-domains of functional importance named after a highly conserved residues or motifs found therein. Dark boxes represent transmembrane helices. Restriction sites are A, ApaI; B, BamHI; N, NaeI; P, PstI; S, SacI; and X, )0101. 33 of the C-terminal domain in AbsAl signal sensing and transduction. The AbsA2 R is a member of the FixJ subfamily of two-component regulators. It has two domains, an N- terminal receiver domain with the putative site of phosphorylation at Asp54, and a C- terminal effector domain containing a helix-tum-helix DNA-binding motif, which is predicted to regulate target promoter transcription. The absA locus was originally identified in Abs' mutants that were globally deficient in antibiotic synthesis (2). The mutations responsible for the Abs' phenotype were localized to a 1.45 kb region of absA 1 (19) in two independent isolates, C542 and G577. Interestingly, sab (suppressors of Abs) mutants arose spontaneously at a relatively high frequency (0.1%) in C542 or C577 protoplasts (19). Also, a gene replacement and gene disruption of the absA 1 gene resulted in precocious hyperproduction of antibiotics (Pha), characterized by early onset and overproduction of antibiotics. The absAI gene knockouts were believed to have a polar effect on downstream absAZ due to the close proximity of the two genes. Given these results, it was hypothesized that absAZ encoded a negative regulator of antibiotics since its elimination resulted in the Pha phenotype (19). The Abs' and Pha mutants were found to affect the level of expression of redD and actII-ORF4 in a manner which corresponded to their phenotypes (1). In relation to parental stain 11501, PSR expression was lower in the Abs‘ mutant and higher in the Pha strain. Therefore, AbsA appeared to mediate production of antibiotics through direct or indirect transcriptional regulation of the antibiotic pathway-specific regulator genes. 34 fart] 13m: is C0 In this study I continued to examine basic characteristics of alleles and gene products of the absA locus in order to define in greater detail the mechanism of AbsA- mediated antibiotic regulation. Given previous demarcation of the mutations causing the Abs‘ phenotype to a 1.45 kb region of absA l, the C542 and C577 absA] mutant alleles were sequenced in order to characterize, at a molecular level, changes responsible for these phenotypes. Characterization of numerous sab mutants was also discussed, including the identification of various second-site suppressor mutations localized within the absA locus. The AbsAl HK and the AbsA2 RR possess conserved residues consistent with the formation of active sites involved in phosphorelay signal transduction. Thus, a genetic and biochemical approach was taken to determine whether phosphorelay was active in this system, and to define the role of phosphorylation in mediating AbsA2 activity. Because antibiotic production is growth-phase dependent, a transcript analysis of absA was performed. Identification of the absA transcription start site permitted analysis of its promoter region. In addition, temporal expression of absA was determined, which allowed for qualitative characterization of its timing with respect to that of growth and antibiotic production. It was known from previous work that AbsA influenced the expression of redD and actII-ORF 4 PSRs. In this study I examined the temporal expression of the PSRs redD, redZ, and cdaR in comparison to that of absA and further examined the influence of AbsA on expression of these regulators as putative targets of AbsA2. Finally, preliminary results were presented which suggest that AbsA2 is conserved in other species of Streptomyces. 35 CHAPTER 2 SEQUENCE ANALYSIS OF absA ALLELES OF Abs' AND sub MUTANTS 36 The absA locus of S. coelicolor was isolated from mutants that demonstrated an uncoupling of the temporally parallel processes of sporulation and antibiotic synthesis (2). Two independent isolates, C542 and C577, that mapped to the same region of the chromosome, were globally deficient in the synthesis of all four S. coelicolor antibiotics (Abs', antibiotic synthesis deficient) while unaffected in sporulation. The absA locus was later shown to encode a two-component signal transduction system comprised of the AbsAl histidine kinase and AbsA2 response regulator (19). Marker exchange and marker rescue experiments in the C542 and C577 Abs‘ mutants localized the mutations responsible for this phenotype to a 1.45kb region of absAI (19). This phenotype was dramatically opposed to the early onset and overproduction of antibiotics (Pha phenotype) obtained from an absAI gene disruption and gene replacement (19). The Pha phenotype resulting from absAI disruptions was hypothesized to result from polar effects on downstream absAz. Therefore, it was proposed that AbsA2 encoded a negative regulator of antibiotics and that the C542 and C577 Abs' strains mutationally locked the AbsA system into a negative regulatory state. Another characteristic of the C542 and C577 Abs' mutants was that they underwent apparent spontaneous reversion. Pseudorevertants of the Abs‘ phenotype, sab (auppressor of alas) mutants, which restored synthesis of all four antibiotics, spontaneously arose in the C542 and C577 absA mutant protoplasts at a frequency of 0.1% (19). The sub mutants were of considerable interest because identification of second site suppressors is a useful tool for finding additional members of a regulatory pathway. Alternatively, if localized to the absA locus, these pseudorevertants might provide insights into the mechanism of AbsAl/AbsA2 interactions. 37 In this study further characterization of the C542 and C577 Abs' mutants was undertaken. It was of considerable interest to sequence the aim“ alleles of the Abs' mutants C542 and C5 77 to confirm that mutations responsible for the Abs' phenotype were indeed present. Additionally, it was hoped that these mutations would provide evidence, when analyzed together with sab mutations (below) and site-directed mutations (Chapter 3) into the biochemical mechanism that locks AbsAl into a negatively acting state. This chapter was also concerned with the characterization of sab mutants, beginning with colony purification and phenotypic analysis on through to genetic mapping and sequence identification of various sab mutations within the absA locus. This work is presented in " Genetic suppression analysis of non-antibiotic-producing mutants of Streptomyces coelicolor absA locus" (6) . My contribution to this publication includes designing the sequencing strategy, amplification and preparation of DNA for sequencing, and analysis of raw sequencing data for the absAI and sub mutant alleles. Conclusions drawn from this work as a whole are reprinted from the text (6): (i) Non-antibiotic-producing (Abs-) mutants of the absA locus, which seem to lock the AbsA regulatory system into a negatively regulating mode, contain point mutations in conserved domains of the AbsA] histidine kinase sensor-transmitter protein. (ii) The absAI mutants spontaneously acquire suppressive mutations that restore antibiotic synthesis. (iii) Plasmid-mediated and protoplast fusion mapping techniques were useful for genetic analysis of suppressive (sab) mutations, locating some close to absA. 38 (iv) Actinophage ¢C31-derived vectors were useful for marker rescue and marker exchange experiments that verified the existence and location of sab mutations and allowed transduction of sab mutations from strain to strain. (v) Sequence analysis defined sab mutant residues in the absA two-component system. Some sab alleles (Type I) restore a wild-type phenotype to Abs" mutants, whereas some (Type 11) cause antibiotic overproduction. (vi) Antibiotic overproduction in sab strains can result from deletion of absA, consistent with absA's proposed role as a negative regulator, but the most strongly pigmented sab strain contains a point mutation in the AbsA2 response regulator, suggesting a complex role for the absA locus in production of antibiotics. 39 CHAPTER 3 GENETIC AND TRANSCRIPTIONAL ANALYSIS OF absA, AN ANTIBIOTIC GENE CLUSTER-LINKED TWO-COMPONENT SYSTEM THAT REGULATES MULTIPLE ANTIBIOTICS IN S. coelicolor 40 31 De INTRODUCTION Streptomycetes are notable among prokaryotes for their fiangal-like developmental cycles. Early in the growth of a colony, multinucleoidal vegetative hyphae extend through the growth medium, branching extensively to form a mycelia] mat. Later, in response to poorly understood signals, the vegetative hyphae initiate a program of multicellular differentiation. Morphological differentiation produces sporulating aerial hyphae on the colony surface (reviewed by 37; 34) while the temporally parallel but spatially distinct process of secondary metabolite (“antibiotic”) production occurs in the substrate mycelium (reviewed by 31; 35). Streptomycete antibiotic biosynthetic pathways involve multiple enzymes that are encoded in large clusters of genes. Each species typically contains several antibiotic gene clusters and these are subject to a complex network of regulation. Much of what is known about the regulation of antibiotic genes has come from genetic studies in Streptomyces coelicolor. One level of regulation that was discovered in S. coelicolor, and is now known to be common to most if not all streptomycetes, involves so-called “pathway-specific regulation,” a mechanism in which a cluster-linked transcriptional regulator — usually an activator -— regulates expression of numerous polycistronic transcripts in an antibiotic gene cluster. In the cases of the S. coelicolor antibiotics actinorhodin and undecylprodigiosin, which are especially well characterized, the pathway-specific activators are ActII-ORF4 and RedD, respectively (122; 167; 62; 57). Both are OmpR-like DNA binding proteins and are founding members of the SARP (for streptomycete antibiotic regulatory protein) family of regulators, which also includes many of the known cluster-linked regulators for other streptomycete antibiotics (183). 41 Studies of S. coelicolor antibiotics have been facilitated by the ease of assaying the antibiotics. Two are pigments: actinorhodin (Act) and undecylprodigiosin (Red) are blue and yellow, respectively, at alkaline pH; both are red at acidic pH. The other two S. coelicolor antibiotics, calcium-dependent antibiotic (CDA) and methylenomycin (Mmy), can be assayed in simple plate culture-inhibition assays. Production of the S. coelicolor antibiotic pigments can easily be seen to be growth-phase regulated in both plate and liquid cultures. It has been demonstrated that this temporal regulation results fi'om growth-phase regulated expression of the pathway-specific activators (167; 62; 182). It is less well understood what regulates the pathway-specific activators. However, one such control involves the absA two-component system, which was discovered in a genetic analysis of global, or coordinate, antibiotic regulation. Mutants of absA were first identified because of their actinorhodin/undecylprodigiosin-minus, sporulation-plus phenotype; subsequently, they were shown to be calcium-dependent antibiotic-minus and methylenomycin-minus, as well (2). This phenotype was named Abs’ and classical genetic mapping showed that the Abs' phenotype was attributable to mutations in the absA locus (2; 19) Further work showed a deficiency of actII-ORF 4 and redD transcription in absA mutants (1), explaining the Abs' phenotype, at least with respect to actinorhodin and undecylprodigiosin. The genetic map location of absA was far from the act and red gene clusters, but was close to the only existing cda mutant. Recent genomic sequencing of S. coelicolor has revealed that absA is associated with the cda gene cluster (http://www.sanger.ac.uk/Project/S_coelicolorl). Previous to the genomic sequencing, only a peptide synthetase-encoding segment of the cda cluster had been defined (39). 42 Now, it is apparent that absA lies in a 12 kb region between the peptide synthetase genes and a putative SARP-like regulator for the cda cluster, cdaR (Figure 5). The function of absA as a regulator of multiple antibiotic clusters, while being genetically associated with one cluster, makes absA highly unusual among antibiotic regulators. In typical two-component systems, a dimeric histidine kinase uses ATP to autophosphorylate, with one subunit transphosphorylating the other on a specific conserved histidine residue (reviewed by 161). The phosphoryl group is then transferred to an aspartate residue on a cognate response regulator, modulating its activity as a transcriptional regulator. The absA-encoded two-component system is highly “orthodox,” including the features common to many of the better-studied two-component systems. The absA] gene is predicted to encode a histidine kinase and the adjacent, downstream gene, absA2, is predicted to encode a response regulator with a C-terrninal helix-turn-helix DNA binding domain. Following the two component paradigm, sequence conservation predicts that the AbsAl protein would autOphosphorylate at His 202 and the phosphoryl group would transfer to Asp54 of AbsA2. AbsA2 is highly homologous to NarL of E. coli and the transmitter domain of AbsAl is similar to the cognate kinases, NarX (63). Closely related two-component systems from Bacillus subtilis include DegS/DegU and ComP/ComA (reviewed by 121). 43 17.5 kl: Figure 5. Position of absA with respect to the cda gene cluster. This 58.3 kb region of the cda cluster was reconstructed from sequence data made available by the Streptomyces coelicolor Sequencing Project (The Sanger Centre). Genes shown in white have been named and given putative functions based on genetic or functional analysis. cdaR is homologous to pathway-specific activators. Biosynthetic genes cdaPSI, cdaPSII and cdaPSIII encode peptide synthases which catalyze steps in the enzymatic synthesis of the lipopeptide antibiotic CDA. Shaded genes have been assigned putative functions based on sequence similarity to other proteins (annotated in http://www.sanger.ac.uk/Projects/S_coelicolor/). Marker rescue experiments (19) and subsequent sequence amlysis (6) of absA mutants located the mutations that were responsible for the Abs" phenotype to the transmitter domain of AbsA]. Below, we refer to these alleles as absA 1'. Additional genetic experiments revealed that absA could also mutate to a phenotype essentially opposite to Abs'; this phenotype was characterized by an early onset and increased level of antibiotics (l9). Antibiotic gene transcription was correspondingly increased in the 44 overproducing mutants (1). Two absA disruption mutations caused the overproduction phenotype, suggesting that the role of absA in antibiotic regulation was primarily negative (19). We have undertaken a genetic dissection of the absA locus, which we describe here. This work evaluates the role of phosphorylation in absA-mediated regulation of actinorhodin, undecylprodigiosin, and CDA and establishes the genetic basis for the two opposing phenotypes observed in absA mutants. We also describe a transcriptional analysis of the absA genes which reveals autoregulatory behavior of AbsA2. Together, the results of these experiments have implications for the mechanism by which AbsA signal transduction regulates antibiotics during the Streptomyces coelicolor life cycle. MATERIALS AND METHODS Growth Conditions Streptomyces strains were cultured in YEME broth (73) for use in plasmid and protoplast preparations. Cultures used for chromosomal DNA extraction were grown for two days either in YEME broth or on SpMR (91) plates overlaid with cellophane disks. Cultures used for RNA extraction were grown in 50 ml of SpMR broth in 300 ml baffled flasks, inoculated with 108 spores, and incubated at 30°C, 250 rpm for 18, 30 or 54 hours. Thiostrepton was added to obtain a final concentration of 10 ug/ml in liquid culture or 200 ug/ml in agar. Hygromycin (Hyg) was added to agar plates to a final concentration of 200 pig/ml. Escherichia coli was grown in L broth or L agar (147). Ampicillin was added to obtain a final concentration of 50 ug/ml in both agar and broth. 45 Antibiotics Assays Assay conditions for the calcium-dependent antibiotic were as previously described (2). Strains were grown on Oxoid nutrient agar (ONA), or R5 (73), and were placed onto plates with or without added calcium (as Ca(N03)2 to 12 mM). Soft ONA or ONA plus Ca was seeded with CDA-sensitive Staphylococcus aureus and was overlaid around the plugs. Plates were incubated overnight at 37°C. Actinorhodin and undecylprodigiosin determinations were as previously described (2). Plasmid and DNA Manipulations All Oligonucleotide primers used in this study (Table l) were prepared by the Macromolecular Structure Facility at Michigan State University (East Lansing, MI, USA). Streptomyces plasmid preparations and transformations were performed as described by Hopwood, et al. (1985). Streptomyces chromosomal DNA was isolated using the method of Pospiech and Neumann (1995). Escherichia coli plasmid preparations were done by alkaline lysis (147) or using QIAprep spin columns (QIAGEN). All replicative plasmids shown in Figure 6 were constructed by first cloning the S. coelicolor absA region of interest into pBluscriptII SK+ (Stratagene) by standard cloning techniques (147). Inserts flanked by BamHI sites were then subcloned directly into the Bng site of pl] 702 (73) as in the case of pCB220 and pTBA155. Inserts used to construct pCB520, pCBS30, pCB540, pTBA156, and pTBA175 were first subcloned into pU2925 (86) and then excised as BglII fragments from the pIJZ925 polylinker for ligation into the BgIII site of pl] 702. Replicative ligations were transformed into S. lividans 46 cocoa moan 5:235“ 05 mo 0.28205: “Em ofi an Eamon watonfiac :88 gm :ouflmqab 05 mo 02820:: “we 05 “a mfimon matching @5ch cum—Q88 u .3 659. wave Epsom—mam Ho: n .ad MQW o H .m.o N .—.5.O.U.? .3 3 8 3. ad OUOUEm? .3. v: 9 N: ohm/Rpm OUUOH? .3 mam 8 SN Boa Hm N “and UEOO? .md owl 8 Nat om ~? .3 NB 9. wwm cm Fara UOOOHOODHO» .3 oz: 8 3m dd UU‘ 33 i E H202L 3f .9. pTBA156 // 79 a: B D54E P B B pCB220 l l B H202L B pTBA155 l.___.!___l 49 1326; plasmids recovered from these transformants were then transformed into S. coelicolor J 1 501 . Disruption of absAZ in C500 The integrative plasmid pTBASOO (Figure 6) was constructed using an absAZ fragment amplified by PCR from primers WC8 and WC9, both of which contained BgIII restriction sites at their ends. The truncated region of AbsA2 encoded by the WC8/W C9 PCR product is illustrated in Figure 7. PCR amplification was carried out in a 100 pl reaction volume with 100 ng J l 501 chromosomal DNA template, under buffer and thermal-cycler conditions for absAZ amplification described in Anderson, et al. (1999). The absAZ amplification product was purified on Wizard PCR preparatory columns (Promega) prior to and following digestion with BgIII. The resulting 5’- and 3’-truncated absAZ fragment was cloned as a BglII fragment into the BamHI site of pIJ963 (86) to produce the integrative plasmid pTBASOO. Strain C500 possessing a chromosomal disruption in absA2 was created by single cross-over integration of pTBASOO. pTBASOO was passed through dam', dcm' E. coli ET12567 prior to transformation into S. coelicolor J 1501. Hygr resistance was used to select for single-crossover recombinants; these displayed the Pha phenotype. Plasmid integration was analyzed using Southern hybridization with absA- and hyg-specific probes. SO linker { helix ymmtfheux W VLLADDETIW-uI/«D-n-l/m-Kn ----NAEIAQRLHLVEGTIKT ----- WC8 WC9 Receiver Domain Effector Domain Figure 7. Creation of the absAZ disruption in strain C500. An internal region of absAZ was generated by PCR from primers WC8 and WC9. Primer WC8 recognized the region of absAZ around the highly conserved Asp13 codon. Primer WC9 annealed to the region of absA2 encoding the first helix of the helix-tum-helix DNA-binding motif. Construction of an In-Frame Deletion in absA] in C530 The integrative plasmid pTBA533 (Figure 6) was constructed by first digesting pCB4OO (pIJ2925 with a 2 kb absA l BamHI fragment) with NaeI to remove the 0.8 kb fragment internal to abaA] (Figure 8) which created pCB420. The resulting 1.2 kb BamHI absA] fragment was ligated into the BamHI site of pCB3OO (pSK+ with a 1.8 kb BamHI/X1101 absAZ-containing region) to produce pCBSOO. The 2.4 kb XhoI/PstI region of pCBSOO carrying the whole absA locus with the absA 1 in-frame deletion was subcloned from the pCBSOO polylinker as a SacI/Pstl fragment into pIJ2925 to create pCBSOl. The entire pCBSOl insert was removed from the polylinker as a BgIII fragment and ligated into BamHI-digested pIJ963 to produce the integrative plasmid pTBA533. 51 absA1A530 11.321 * IHI INHDIIGD-i 1 Sensor Transmitter C-terminal Domain Domain Domain Figure 8. The absA 1 A530 in-frame deletion (diagonal hatch) was created by the removal of a 0.8kb NaeI region internal to absA] . Horizontally hatched boxes in the sensor domain represent four transmembrane helices predicted for AbsAl. Lettered boxes of the transmitter domain symbolize highly conserved sub-domains of two-component histidine kinase transmitters. The H-box contains Hi3202 which is the putative site of phosphorylation in AbsAl. Initial attempts at gene replacement used a strain with a deletion/ermE replacement in absA (C430), but this strain transformed extremely poorly (19). Therefore, gene replacements were created in strain J 1501, as follows. Strain C530 possessing a chromosomal absA] in-frame deletion was created through double cross- over gene replacement with integrative plasmid pTBA533. pTBA533 was passed through dam', dcm' E. coli DM-l (GIBCO BRL) prior to transformation into S. coelicolor J 1501. Hygr resistance was used to select for single-crossover recombinants. Plasmid integration was analyzed using Southern hybridization with the same absA- and hyg- specific probes described above (see Disruption of absAZ). Single-crossover 52 recombinants were subjected to multiple rounds of propagation and spore isolation on solid and liquid media without Hyg to allow double-crossover curing of the plasmid. Single colonies were chosen for making spore preparations and chromosomal extractions. Initial screening for double crossovers was performed by PCR amplification from primers WC12/WC13, which were both internal to the in-frame deletion, and WC16/WC26, which produced a 1 kb product for absA] with the in—frame deletion, versus a 1.8 kb product for wild type absA 1. If no PCR products corresponding to wild type absA] were amplified, then Southern hybridization was performed on an XhoI digest of chromosomal DNAs. Colonies which had successfiilly undergone double cross-over integration of the absA] allele carrying the in-frame deletion showed no signal for the hyg probe and a single signal of 2.4 kb for the absA probe. F inal confirmation of the integrity and fidelity of the C530 absA locus was obtained by sequencing the entirety of absA] and absAZ. Procedures for the amplification and sequencing of absA] and abs/12 are described elsewhere (6). In each of the gene replacements described in this study, double-crossover integration of the mutant allele required propagation of single-crossover transformants for numerous generations under nonselective conditions. It should be noted that although single-crossover transformants demonstrated Hyg sensitivity afier only a few generations of growth in the absence of antibiotic, many of these still possessed the hyg marker as determined by Southern analysis. Similarly, PCR screening that indicated complete resolution of the double crossover was frequently contradicted by Southern analysis. Final confirmation of successful gene replacement required careful analysis by Southern hybridization and sequencing. 53 Site-Directed Mutagenesis The absAZ D54E allele was generated using PCR amplification with mutagenic primers. Separate upstream and downstream absAZ fragments, with an overlapping region centered at the site of the D54E-encoded mutation, were amplified from pCB46O (pSK+ carrying a 3.9 kb BamHI/Xhol fragment with the entire absA locus) using primer pairs WC24/W C29 and WC15/W C28 (Table 1). The resulting GAC to GAG change also introduced an XhoI site into the PCR products. Thus, the D54E-containing fi'agments were digested with XhoI and an additional restriction enzyme, the site for which was present in the upstream or downstream region surrounding absA2: BamHI for the (WC24/W C29) product upstream of absA 2, -and PM for the downstream (WCl 5/W C28) product. A three-way ligation between these fi’agments and pSK+ BamHI/Pstl produced pTBA16O containing a 1.2 kb insert with the entire absAZ D54E allele. Confirmation of the site-directed change was obtained by sequence analysis of the entire absA2 D54E allele from pTBAl60. Subsequently, a 2 kb BamHI absA] region was ligated into pTBAl6O BamHI to create pTBA162. The 3.2 kb BamHI/Pstl insert of pTBA162 containing the entire absA locus was removed as a XbaI/Kpnl fragment from the pTBA162 polylinker for ligation into pIJZ925 to produce pTBA166. The same 3.2 kb insert was excised from the pTBA166 polylinker as a Bng fragment and ligated into pU963 BamHI to generate the integrative plasmid pTBA57O (Figure 6). Strain C570 possessing a chromosomal absAZ D54E mutation was created through double-crossover gene replacement with integrative plasmid pTBA5 70. pTBA570 was demethylated as described above prior to transformation into S. coelicolor 54 J 1501. Hygr resistance was used to select for single-crossover recombinants. Plasmid integration was analyzed using Southern hybridization with the same absA- and hyg- specific probes described above (see Disruption of absA2). Single-crossover recombinants were subjected to multiple rounds of propagation on solid and liquid media without Hyg to allow double-crossover curing of the plasmid. Single colonies were chosen for making spore preparations and chromosomal extractions. Initial screening for plasmid curing was performed by PCR amplification from primers WC30/WC28, which amplified a 1.2 kb product containing the entire absA2 allele. Only the absA2 D54E allele was susceptible to digestion with X7101. Therefore, if no PCR products corresponding to wild type absA2 were amplified, then Southern hybridization was performed on XhoI digests of chromosomal DNAs. Colonies which had successfully undergone double cross-over integration of the absA2 D54E allele showed no signal for the hyg probe and signals of 1.7 and 1.5 kb for the absA probe. Final confirmation of the integrity and fidelity of the C570 absA locus was obtained by sequencing the entirety of absA] and absA2 as described (6). The D54A and D54N alleles of absA2 were created by PCR-mediated introduction of these mutations using the QuickChange Site-Directed Mutagenesis Kit (Stratagene). The D54A mutation was introduced into absA2 through a single nucleotide change (GAC to GCC) with complimentary primers WC65 and WC66. Likewise, complimentary primers WC67 and WC68 produced a single mismatch (GAC to AAC) in absA2 to generate the D54N mutation. The mutagenesis reactions were carried out on 50 ng of pTBA400 (pSK+ carrying a 1.8 kb BamHI/X1101 absA 2-containing fragment) under the manufacturer-prescribed buffer conditions with the addition of 5% glycerol and 2.5% 55 dimethyl sulfoxide (DMSO). The thermal cycler conditions were 95°C for 5 min, followed by 12 cycles of 95°C for l min, 65°C for 45 sec, and 72°C for 12 minutes. The PCR products were digested with DpnI and transformed into E. coli DHSa (GIBCO BRL) to create pTBA410 (containing absA2 [D54A]) and pTBA430 (containing absA2 [D54N]). Both the D54A and D54N mutations removed a T an restriction site at the mutagenized codon. Therefore, to screen for successful incorporation of site-directed mutations, plasmid preparations from transformant colonies were used as template in PCR reactions with primers WC35 and P11, which amplified a 360 nt region internal to absA2. Amplification products were digested with Tan and analyzed on 1.2% agarose gel. In each case, over 90% of the transformants tested screened positive for the mutation. Confirmation of the desired mutations was obtained by sequencing the absA] and absA 2 portions of the mutagenized plasmids, pTBA410 and pTBA430. Integrative plasmid pTBA532 (Figure 6) encoding the absA2 D54N mutation was made by subcloning the 1.8 kb BamHI/Kpnl fragment from the polylinker of pTBA430 into pU963. In order to construct integrative plasmid pTBA516 (Figure 6), the pTBA410 insert was increased in size to 3.8 kb by cloning in a 2 kb BamHI fragment containing the upstream portion of absA] to produce pTBA516. The 3.2 kb XhoI fragment of pTBA516 (containing absA2 [D54A]) was then cloned into the SaII site of pIJ2927 (86) to create pTBA41‘4. This same 3.2 kb insert was then removed as a 3.2 kb BgIII fragment from the polylinker of pTBA414 and cloned into pIJ963 BamHI to create the integrative plasmid pTBA5 16. Strains C516 and C532, possessing chromosomal mutations absA2 [D54A] and absAZ [D54N], respectively, were created through double-crossover gene replacements 56 with integrative plasmids pTBA516 and pTBA532, respectively. Single- and double- crossover integration procedures were the same as described for C570 gene replacement. Initial screening for plasmid curing was performed by PCR amplification of T an restriction digest analysis of the WC3 5/P11 absA2 products. Chromosomal DNA from strains that did not amplify wild type absA2 alleles were digested with XhoI and analyzed by Southern hybridization with hyg and absA probes as described above. Colonies which had successfirlly undergone gene replacement with the absA2 (D54A) and (D54N) alleles showed no signal for the hyg probe and signals of 3.2 kb for the absA probe. Final confirmation of the integrity and fidelity of the C516 and C532 absA loci was obtained by sequencing the entirety of absA] and absA 2 using methods described previously (6). The absA] H202L allele was generated using PCR overlap extension (171). Separate upstream and downstream absA 1 fragments, with an overlapping region centered at the site of the H202L mutation, were amplified from pCB460 (pSK+ carrying a 3.9 kb BamHI/Xhol fragment with the entire absA locus) using primer pairs WCl6/WC30 and WC14/WC26 (Table 1). Then, the full-length absA] H202L allele was amplified by combining 5 to 10 ng of the upstream and downstream PCR products as template together with primers WC 16 and WC26. PCR amplifications were carried out in 50 pl reactions using high-fidelity Pfu DNA polymerase (Stratagene). Bufi‘er conditions and thermal cycler settings were as previously described (6) except that the extension time was increased from 1 to 4 minutes. The expected size product was agarose-gel purified and digested with XhoI and BamHI to produce a 1.45 kb fragment that was ligated into pSK+ to create pTBA140. Confirmation of the site-directed change was obtained by sequence analysis of the absA I XhoI/BamHI fragment from pTBA140. 57 Subsequently, a 5 kb BamHI fragment (containing the 3' region of absA] and all of absA2) was ligated into pTBA14O BamHI to produce pTBAl42. The 3.2 kb XhoI region (containing most of the absA] H202L allele and all of absA 2) was excised from pTBAl42 and ligated into XhoI-digested pCB3 60 (pIJZ925 containing a 1.6 kb SacI/Xhol insert with the 5‘ region of absA] and upstream SCE8. 17c) to create pTBA144. The entire 5.4 kb insert was removed from pTBA144 as a BgIII fragment for ligation into pIJ963 BamHI to create the integrative plasmid pTBAISO (Figure 6). Strain C550 possessing a chromosomal absA 1 H202L mutation was created by gene replacement with pTBAISO. Single- and double-crossover integration procedures were the same as described for the C5 70 gene replacement. Initial screening for plasmid curing took advantage of a T an restriction site introduced by the H202L mutation. An internal region of absA] was amplified from chromosomal DNA of putative C550 strains using primers WC12 and WC13, followed by T an digest analysis. Chromosomal DNA from strains that did not amplify wild type absA] were digested with X7101 and analyzed by Southern hybridization with hyg and absA probes as described above. Colonies which had successfiilly undergone gene replacement with the absA] H202L showed no signal for the hyg probe and a signal of 3.2 kb for the absA probe. Final confirmation of the integrity and fidelity of the C550 absA locus was obtained by sequencing all of absA] and absA2 using methods described previously (6). RNA Isolation Streptomyces RNA isolation was carried out as described by Hopwood, et al. (1985) using the preparation method for dot-blotting and northern blotting . Two 58 independent isolations at 18, 30 and 54 hours of growth were performed for S. coelicolor strains J 1501 (hisAI uraAI strAl SCPl' SCP2" Pgl'), C542 (absA1-542 (6)), and C570 (absA2 [D54E] hisAI uraAI strAI SCPI" SCPZ-Pgl'). Four 50 ml cultures were pooled for each 18 hour RNA preparation, whereas two 50 ml cultures were pooled for 30 and 54 hr samples. The concentration, purity and integrity of the RNA samples was evaluated by spectrophotometry and agarose gel electrophoresis. Isolation of E. coli RNA, for use as a negative control, was performed with an RN Aeasy RNA-purification column (QIAGEN). Sl Nuclease Protection Assays All experiments were performed using 50 pg of RNA and 60,000-100,000 cpm of 32P-end-labeled double-stranded DNA probe. The absA transcript time-course analysis incorporated an absA] probe together with a glk (glucose kinase (7)) probe - which served as an internal standard for normalizing the quantity of RNA in each assay; (1) The 455 bp absA] probe was generated by PCR using the WC64 forward primer and the 5’-32P-end-labeled WC20 reverse primer. The template for absA! probe synthesis was pCB400, containing the 2 kb BamHI region of the absA locus cloned into pU2925. A 309 bp 32P-end-labeled glk probe was also generated by PCR from the primers and template described by Aceti and Champness (1998). Primers (50 pmoles) were end- labeled by the T4 polynucleotide kinase (Promega) forward reaction as described by the manufacturer with minor modifications. Prior to initiating the end-labeling reaction, each primer was incubated with spermidine (10 mM final concentration) at 70°C for 10 minutes. Likewise, ethanol precipitation of the labeled oligos was facilitated by the 59 addition of 2 ug of glycogen. The labeled oligo was divided between duplicate 50 ul PCR reactions. The reaction mix contained 20 pmoles of each primer, 100 ng template, 0.2 mM dNTPs, 1.5 mM MgC12, 5% glycerol, 2.5% DMSO, 1% formamide and 1.25 U T aq polymerase (Perkin Elmer). Thennal-cycler conditions were 95°C for 5 min, followed by 30 cycles of 95°C for 1 min, 65°C for 45 sec, and 72°C for 1 min, and a final extension at 72°C for 10 minutes. The S] nuclease protection assay was performed as previously described (1). Replicate RNA isolates were tested in independent 81 experiments. Time-course experiments included E. coli RNA as a negative control. In addition, the presence of excess probe was verified by treating an RNA sample with 2- fold concentrations of each probe and comparing their signals to the same sample treated with normal levels of probe. Results were analyzed by electrophoresis on 6% polyacrylamide sequencing gels (147) and autoradiography. Transcript sizes were estimated by running 32P-end-labeled ¢X174lHian molecular weight markers (Promega) on the same gel. To map the absA] transcription start site, a sequencing ladder was generated from primer WC20 using the fmoiD DNA Sequencing System (Promega) and compared to Sl-treated 18 hr C542 RNA hybridized to the absA] probe. The region upstream of the absA 2 translation start site was examined for promoter activity using a 504 nt probe. The absA2 probe protected the region from 330 nt upstream of the absA2 translational start to 174 nt downstream into the coding region. The absA2 probe was generated by PCR using forward primer WC24 and 5'-32P-labeled WC29 reverse primer. The PCR template was pCB46O (pSK+ carrying a 3.9 kb BamHI/Xhol insert with the entire absA locus). Primer end-labeling, PCR reaction 60 conditions, and probe purification were performed as described for the absA] probe. The absA2 probe was used in SI nuclease protection assays with 18 and 30 hr C542 RNA. RESULTS Negative Regulation of Antibiotics by the absA2-Encoded Response Regulator and absA] Histidine Kinase In previous genetic studies of the absA locus, certain mutations that disrupted the locus suggested that the absA-encoded two-component system fimctioned as a negative regulator of antibiotics (19). These mutations caused a visible phenotype of early, enhanced production of the actinorhodin and undecylprodigiosin antibiotics; we refer to this phenotype as Pha, for precocious hyperproduction of antibiotics. The Pha mutant alleles were created by insertions into the absA! gene, which is upstream in a putative absA I-absAZ operon. The phenotype in these mutants may have resulted from disruption of absA] or from polar effects on expression of absA 2. To distinguish between these possibilities, we directly tested the fiinction of absA2 by specifically disrupting the absA2 gene. A fragment internal to absA2 (Figure 6) was cloned into the nonreplicating plasmid pIJ963 to create pTBASOO, which was then integrated into the absA locus of strain J 1501. The resulting strain, C500, was absA 1+ absA2::pTBA500, with absA2 truncated upstream of the predicted helix-tum-helix domain (Figure 7). Disruption of absA2 in C500 caused a Pha phenotype (Figure 9), thereby demonstrating the involvement of absA2 in negative regulation of antibiotic production. Both repressor and activator functions are well-documented for two-component response regulators; 61 functional regions of the transmitter domain of histidine kinases, including the conserved H, N, D, and G boxes. The phenotype of the absA 121530, strain C530, was Pha (Figure 9) and, moreover, was identical to that of C500. This result implicated the protein kinase activity of AbsA] in the negative regulation effected by AbsA2. J1501 C530 Figure 9. The effect of an absA gene disruption and gene replacements on antibiotic production. Strains were grown for 4 days on SpMR agar. Strains are S. coelicolor J 1501 (wild type), C550 (absAI [H202L]), C570 (absA2 [D54E]), C500 (absA2zszBA500), C532 (absA2 [D54N]), and C530 (absA I A530). Actinorhodin and undecylprodigiosin pigments were assessed as described in Materials and Methods. Images in this dissertation are presented in color. 62 Genetic Evaluation of the Role of Phosphorylation in AbsA2-Mediated Regulation For most response regulators, phosphorylation of a conserved aspartate residue is essential for the regulatory functions of the proteins in viva. Following this precedent, the AbsA2 regulatory activity would likely require that AbsA2 be phosphorylated; AbsAl would likely be responsible for AbsA2 phosphorylation. The Pha phenotypes of C500 and C530 would be consistent with this. scenario, but it was important to consider the additional factor that many of the characterized two-component system histidine kinases are bifunctional enzymes that possess both kinase and phosphatase activities; the phosphatase activity dephosphorylates the phosphorylated response regulator. In the case of AbsAl, the in-frame deletion in C530 would remove AbsA2-specific phosphatase activity, as well as the kinase activity associated with the transmitter domain. Thus, in strain C530, phospho-AbsA2 may be present if AbsA2 can be phosphorylated by an alternative kinase or low molecular weight phosphate donors, and the C530 phenotype might be caused by a lack of the AbsA] phosphatase and a resulting overabundance of phospho-AbsAZ. In this case the negatively regulating form of AbsA2 would be the unphosphorylated form. In order to distinguish whether phospho-AbsA2 or unphosphorylated AbsA2 functions as the negative regulator, we constructed several mutants with site-directed changes to the chromosomal absA2 gene (Figure 6), altering the AbsA2 aspartate residue (D54) that is analogous to the conserved phosphorylated aspartate of response regulators. Separate gene replacements created strains C570, C516, and C532 with AbsA2 amino acid replacements, D54E, D54A, and D54N respectively. These aspartate substitutions have been shown to prevent phosphorylation of numerous response regulators (88; 42; 63 22). All three mutant strains exhibited the Pha phenotype (Figure 9; C516 not shown). Thus, these results supported the hypothesis that phospho-AbsAZ functions as the negative regulator. The histidine residue in AbsAl that corresponds to the site of phosphorylation in well-characterized members of the histidine kinase family is His 202 (Figure 6). A site- directed mutation, H202L, was made in the chromosomal absA! gene of 11501, creating strain C550. Strain C550 exhibited a Pha phenotype, a result consistent with a requirement for histidine kinase activity in negative regulation. However, the phenotype differed fi'om that of C530 (AabsA 1) in several respects. First, C550 visibly produced undecylprodigiosin earlier than actinorhodin, whereas C530 produced both antibiotics ‘ precociously. Second, hyperproduction of antibiotics, relative to strain 11501, never reached the levels seen for C530 (Figure 9). The reason for the weaker Pha phenotype of C550 is not clear at this time. We considered the possibility that AbsAl (HZOZL) contained a second site of phosphorylation; however a fiision protein containing the AbsAl (HZOZL) transmitter domain fused to maltose-binding protein did not demonstrate autokinase activity in vitro, whereas an MBP:: AbsAl+ fusion did (Chapter 5). Precocious Hyperproduction of Calcium-Dependent Antibiotic, Undecylprodigiosin and Actinorhodin in absA Mutants We sought to determine whether Pha mutations affected synthesis of calcium dependent antibiotic, in addition to actinorhodin and undecylprodigiosin. To assess CDA activity, plugs from plate grown cultures were tested for anti-Staphylococcus aureus activity. In the presence of added calcium, the lipopeptide CDA is active, damaging cell 64 de: act membranes (100). For CDA assays, culture plugs were tested on plates with and without added calcium. In a 2-day time course, shown in Figure 10, CDA activity was detected in Pha mutants at least 7 hrs earlier than in J 1501. Pha mutants C550 and C530 are shown. Similar results were obtained with C550, C530 and C570 on R5 media (data not shown). These results showed that AbsA negatively regulates CDA as well as actinorhodin and undecylprodigiosin. 41hr f5 Figure 10. Calcium-dependent antibiotic assays in Pha mutants. Growth conditions are described in Methods. Plugs were taken from ONA plates at the times indicated. CDA activity is detected in the presence of calcium, right. 65 Over the course of cultivating Pha mutants we have observed variability in how much earlier a given Pha mutant produces antibiotics compared to J1501. The amount of acceleration has ranged from at least 7 hours to several days on different media, e.g. R5, SpMR and ONA In addition, in quantitative assays of actinorhodin and undecylprodigiosin, the Pha-related overproduction has varied from 5-fold to more than 60-fold (data not shown). An exploration of this phenomenon will be reported in more detail elsewhere. Besides the effect on antibiotics, the Pha phenotype includes a defect in morphology. Pha mutants produce aerial hyphae relatively sparsely and their colony surfaces are notably crenulated. Precocious Hyperproduction of Antibiotics Resulting from AbsA Domain Overexpression In some cases, overexpression of an unphosphorylated response regulator can mimic the regulation of target promoters that normally is effected by a phosphorylated response regulator (e. g., 180). To evaluate whether overexpression of unphosphorylated AbsA2 could regulate antibiotics, we introduced a high copy clone of the absA2 (D54E) mutant allele (pTBA175; Figure 6) into J 1501 and C577SZS, a strain deleted for absA2 and most of absA] (6). The pTBAl75 plasmid included absA1+ and the absA promoter region. If unphosphorylated AbsA2 could negatively regulate antibiotics we might have observed a delay of antibiotics in the Pha C577825 strain. However, we observed no change in the Pha phenotype (data not shown) suggesting that phosphorylation of AbsA2 is required even at high protein abundance, for negative regulation. 66 When pTBA175 was introduced into J1501, the resulting phenotype was Pha, indicating an interference with normal AbsA-mediated regulation (Figure 11B). To further examine the phenomenon, we evaluated the effects of overexpressing selected domains of the AbsAl and AbsA2 proteins (Figure 6). First, we excluded an effect of the absA promoter region by introducing plasmid pCB540; this plasmid did not alter the J 1501 phenotype (Figure 11A). Second, we observed that multiple copies of the entire absA locus, in pCB520, produced no change in the Abs+ phenotype (Figure 11A). Next, we evaluated a set of high-copy plasmids that expressed the wild-type AbsA2 but carried phosphorylation-minus absA] alleles; these included pCBS30, carrying the in-frame deletion of absA 121530, and pTBA156, carrying the absA! [HZOZL] allele. These produced no change in the Abs+ phenotype (Figure 11). In contrast, a Pha phenotype resulted from plasmids that lacked absA2+, but contained absA] sequences. Two such plasmids were pCB220 and pTBA155. A pattern that emerged from these results was that an increase in gene dosage of absA 2", with or without an increase in abs/11+, did not alter antibiotic regulation. However, an increase in absA 1 sequences without an increase in absAZ+ deregulated antibiotics. One interpretation of these results is that a high absA] gene dosage causes a shift in the ratio of AbsA] kinase to phosphatase activity to favor the phosphatase activity, and relatively low expression of AbsA2+ may not allow sufficient AbsA2-P accumulation to down-regulate antibiotics. What would cause the ratio of AbsAl phosphatase to kinase activity to be higher than normal in these strains? In the cases of the pCB220 and pTBA155 plasmids, the C- terminal 69 aa of absA] are truncated. In previous complementation analyses, the 67 115011 pABSl7S 11501/ 11501/ pCBS30 pABSlS6 Figure 11. The effects of high-copy expression of absA alleles on antibiotic production. All plasmids were derivatives of pIJ702 expressed in a S. coelicolor J 1501 background. Plasmid inserts are shown in Figure 6. Strains were grown for 3 or 4 days on SpMR agar. Early onset of both Red and Act are characteristics of the Pha phenotype. Frame A demonstrates the early production of Red in the Pha phenotype, whereas frame B shows the early onset of Act synthesis. pCB220 absA] allele was capable of restoring a wild-type phenotype to absAI', but it is possible that this allele and the H202L version in pTBA155 have higher than normal phosphatase activities. However, absAI is wild-type on pTBA175. One speculation about the observed effects might be that the AbsAl-kinase activity is normally activated by the binding of a low-abundance signal molecule and the AbsAl polypeptides produced by the high-copy constructs are present in quantities sufficient to titrate the ligand; thus the population of AbsAl molecules may be predominantly in the phosphatase mode. 68 High Resolution 81 Nuclease Mapping of the absA Transcription Start Site The absA] and absA2 ORFs are separated by only 17 at and are likely cotranscribed. To define the transcription start site for absA, high resolution 81 nuclease mapping was performed. First, a PCR-generated double-stranded DNA probe specific to the predicted promoter region for absA] (Figure 12B) was used to map the transcription start site of absA] by analyzing the 81 product alongside a sequencing ladder generated from the same 32P-labeled primer used to synthesize the probe. The absA] probe protected a single product of 291 nt (Figure 12A), identifying the transcription start site for absA] as the adenosine nucleotide that is also predicted to be the putative translation start site. To evaluate cotranscription of absA 1 and absA 2, the region upstream of absA2 was probed with a 504 bp double-stranded DNA probe (Materials and Methods). The 81 product showed no indication of independent promoter activity for absA2 (data not shown). These results indicated that absA] and absA2 are transcribed as a single, leaderless transcript. Leaderless transcripts are not uncommon in actinomycetes, as documented by Strohl (1992), who reported that 11 of 139 promoters analyzed produced leaderless transcripts. Inspection of the sequence upstream of absA! revealed a —10 region with the sequence TAGCGT (Figure 12); this is similar to the consensus sequence proposed by Strohl (1992) for transcription from Streptomyces RNA polymerase that contains an E. coli-like Eo7° sigma factor, e.g. HrdB or HrdD (23). There was, however, no recognizable consensus sequence in the —35 region. 69 A T - T G T* Air G* c* -10region G* T* G C T G G A A‘M T G C H A C C R G A B. 455m DNA probe <_136%6291nt «ac BamHI XhoI BamHI Figure 12. S] nuclease protection mapping of the absA locus. (A.) High resolution S1 nuclease protection mapping on total RNA isolated from an 18 hour culture of S. coelicolor C542 grown in SpMR liquid medium. The AGCT sequencing ladder was generated from 5'-labeled Oligonucleotide WC20 (see Materials and Methods section). The transcription start site (a) and the hexameric -10 promoter region (*) are shown. (B.) The absA probe was a 455 bp PCR product amplified fi'om primer WC64 (Table 1) and primer WC20 - uniquely labeled with 32P at the 5' end. The shaded areas represent coding regions of absA] and SCE8. 1 7c contained on pCB400. The absA probe extends 291 nt downstream of the putative translation start site and 136 nt upstream. 7O Growth-Phase Dependent Expression and Autoregulation of absA To evaluate the temporal profile of absA expression, RNA was isolated over a 54 hr time course fiom cultures grown in liquid media. The media used, SpMR (see Materials and Methods), supported production of the actinorhodin and undecylprodigiosin antibiotics by strain J1501. Streptomyces coelicolor does not sporulate when grown in liquid cultures, so temporal comparisons of antibiotic production and sporulation could not be made in this experiment. However, as is generally observed, the antibiotics showed growth-phase dependent production kinetics, appearing only after a period of biomass accumulation. Sl nuclease protection assays were performed on the absA promoter region from RNA isolated from J 1501 cultures grown for 18 hours, 30 hours, and 54 hours. Antibiotics were not produced in the 18 hour culture but were visible in the 30 hr J] 501 culture. Figure 13 shows that the absA transcript was present in the 18 hr culture, and it increased significantly in abundance from 18 hours to 30 hours. The transcript then remained at a constant level through 54 hours. Comparisons of transcript abundance in different cultures were aided by the addition of a probe for the glk (glucose kinase) gene to each S1 assay. The absA signal increased about five-fold relative to the glk signal over the course of culture growth. Figure 13 also includes 81 nuclease protection assays of RNAs isolated from two strains that are mutant for the absA locus. One strain was C542, an Abs“ strain mutant in absA] (i.e., an absA1* strain), as described above. The second was C570, the Pha strain carrying the D54E mutation in absA2 that was described in Figure 6. The profile of absA expression was altered in both absA mutants. In C570, the absA transcript abundance 71 11501 (Abs‘) C542 (Abs') 0170 (Pha) 11501 (Absl) €170 (Pha) 11501 (Abs+) c542 (Abs') c170 (Pha) MW Marker (nt) ; a‘ ‘3 E '3 :3 N v V) O 30hr «absA Figure 13. High resolution 81 nuclease protection analysis of the absA transcript, using RNA isolated fi'om 18, 30 and 54 hour S. coelicolor cultures in SpMR liquid medium. S. coelicolor strains are 11501 (absA +), C542 (absA 1-542, (6)), and C570 (absA2 [D54E]). The absA probe was the 455 bp probe in Figure 12. Glucose kinase (glk) was measured to normalize the amount of RNA assayed at each time point (1). 72 was very low at all time points. In contrast, in C542, the absA transcript was several-fold more abundant than in J1501, at all time points. The 81 protection assays revealed several aspects of absA regulation. First, the effects of the absA mutations indicated that absA expression is autoregulated. Second, the mutant effects on the absA transcript were opposite to the previously-observed effects on antibiotic transcripts: whereas the Abs' and Pha phenotypes were found to correlate with decreased or increased antibiotic gene transcription, respectively, the absA transcript was decreased in the Pha strain but increased in the Abs‘ strain. These results suggest that autoregulation by absA is positive, in contrast to absA negative regulation of antibiotics. Third, the low level of absA transcript in C570, the absA2 (D54E) mutant, suggests that phospho-AbsAZ is the autoregulatory form of AbsA2, which is consistent with data from the genetic analysis that implicates phospho-AbsA2 as the antibiotic- regulatory form. Finally, the absence of any growth phase-related increase of absA transcript in C570 suggests that phospho-AbsAZ was responsible for the growth-phase regulation observed in J 1501 and C542. Thus, the growth phase regulation of absA appears to result from phosphorylation-dependent AbsA2-mediated autoregulation. We have not determined at this time whether the absA autoregulation is direct or indirect. DISCUSSION In this paper we have described a genetic and transcriptional analysis of the absA locus which firrther characterizes aspects of the mechanism of AbsA-mediated regulation of antibiotic production. Disruptions in the absA] and absA2 genes demonstrated that the AbsA two-component system is a negative regulator of the multiple antibiotics produced 73 by S. coelicolor, including calcium-dependent antibiotic, actinorhodin and undecylprodigiosin. In addition, gene replacements in the absA locus altered the putative sites of phosphorylation of AbsAl or AbsA2. As predicted from sequence conservation with other two-component systems, both the His at position 202 of AbsAl and the Asp residue at position 54 of AbsA2 were required for normal regulation of antibiotic synthesis: each of the gene replacement strains tested attained an antibiotic overproducing phenotype (Pha) consistent with a mechanism in which the phosphorylated form of AbsA2 is the active negative regulator of antibiotic synthesis. Our results did not, however, distinguish whether or not AbsA2~P is a direct repressor of the antibiotic genes or whether it is the activator of a repressor. Without absA regulation, the timing of antibiotic production is advanced but, even in Pha cultures, a period of approximately 2 days passes before antibiotics appear. One interpretation of this observation is that the appearance of antibiotics in Pha cultures indicates the time at which the culture enters an antibiotic production-competent state, but the AbsA system normally imposes a delay on production. The heterogeneities in a growing mycelia] biomass complicate distinctions of growth phases, but for the purposes of firrther discussion, we refer to the postulated “AbsA-repressed” period as the “transition stage.” We can envision several models for how AbsA, as a signal transduction system, could modulate production of antibiotics during culture growth. One model, which accommodates both the genetic and transcriptional data, supposes the following. Early in growth, a culture is not competent for antibiotic synthesis; also, the absA genes are expressed at a low level. Following a period of growth, the culture enters the “transition stage.” During this time, the signal that 74 regulates AbsA may be present at significant levels. If AbsAl is like many sensor kinase/phosphatases, it will require signal binding to activate the kinase fiinction, and exist in a phosphatase-dominant mode if signal is absent (131; 69; 186). Once the signal is present, and AbsAl is shifted to a kinase-dominant form, AbsA2-P will accumulate and negatively regulate antibiotics and also positively autoregulate, accounting for the AbsA2~P-dependent, growth-phase-related increase of absA transcript seen in 11501. Easing of AbsA-repression may require that the signal be depleted or degraded, allowing AbsAl to switch to the phosphatase form and dephosphorylate AbsA2, allowing antibiotic gene expression. At present, we have no information regarding the nature of the signal hypothesized to regulate AbsAl. Ifthe normal fimction of the AbsA system is in negative regulation of antibiotics, what explains the Abs” phenotype in the mutants that first defined the absA locus? We hypothesize that these absA 1‘ alleles lock the AbsA system into the negatively-regulating mode, i.e., in which AbsA2 is phosphorylated. In support of this notion, the Abs" phenotype requires absA 2+ (6). The mutant AbsAl' proteins might be constitutively kinase-dominant forms, either lacking phosphatase capability or functioning as signal- independent kinases. The latter possibility would be most consistent with the increased level of absA transcript observed in C542, e.g. AbsA2-P would be present even in young cultures lacking signal and would autoregulate. Another observation that could be explained by signal-independent AbsAl kinase activity is the persistence of the Abs' phenotype over the life of mutant cultures: even colonies that grow for several weeks remain unpigmented. 75 An alternative model for signal regulation in the AbsA system is that AbsAl is a kinase in the absence of signal and a phosphatase in the presence of signal, as a few sensor kinase-phosphatases are proposed to fiinction (135; 139; 138; 56). In this case, the transition stage culture lacks the signal regulating AbsAl and AbsAl -kinase activity would generate AbsA2-P. Later a signal would switch AbsA] to the AbsAl-phosphatase mode so it could dephosphorylate AbsA2-P, allowing antibiotic synthesis. We consider this model to be less compelling than the first because the AbsA2-P-dependent transcription profiles are more simply explained if the AbsAl -kinase activity is activated by a transition stage signal. What purpose does AbsA regulation of antibiotics serve in the S. coelicolor life cycle? One relevant observation is the substantial perturbation of morphogenesis observed in most Pha mutants: these mutants usually produce only sparse aerial hyphae. Conversely, antibiotic production is not altered in Abs' strains. One possibility is that precocious antibiotic synthesis per se is deleterious to normal sporulation. Calcium- dependent antibiotic may be especially inhibitory as suggested by strain C577825 (6; Champness, unpublished). This strain demonstrates a strong Pha phenotype for pigmented antibiotic production, but is blocked in synthesis of CDA due to a deletion in this gene cluster. In contrast to other Pha mutants, C5 77825 is wildtype for sporulation. Thus, it may be that S. coelicolor acquires competence for antibiotic production before the sporulation process has proceeded adequately and the function of the AbsA system is to delay antibiotic production to allow optimal sporulation. What factors establish the state that we have referred to as "antibiotic-production competent"? Likely candidates include the genes that have been identified on the basis of 76 mutant defects in antibiotic synthesis. Among these is a second gene found in Abs’ mutant hunts, absB, which encodes the S. coelicolor homolog of RNase H1 (136). Another large group of genes is known to regulate both antibiotic synthesis and the onset of sporulation. Some genes in this group are the bid genes, several of which encode regulators of gene expression (reviewed by 30). Another is reIA, which encodes pppGpp synthetase (27; 26; 113; 114). Also important are the components of y-butyrolactone signaling pathways (188; 130). Additional antibiotic regulatory genes have been isolated on the basis of multi- copy stimulation of antibiotic production. The best characterized of these are the AfsQl/Q2 two-component system (84) and the Ast/K serine-threonine phosphoprotein/kinase pair (76; 77; 71; 115; 116; 54). Mutations in the afsQl/QZ genes cause no phenotype, but disruptions to the cy'sR/K locus conditionally reduce antibiotic synthesis, especially on high phosphate media (115; 54). The multicopy effect of qst/K has been shown to correlate with increased antibiotic pathway-specific activator transcription (54). Multi-copy clones of the ast/K locus can restore antibiotic synthesis to Abs' absA]. mutants (28), and overexpression of the AfsQ response regulator does the same (84). These observations imply that these genes can compete against the postulated persistent negative regulation imposed by absA 1‘ alleles. It is widely observed that phosphorylation of response regulators modifies their activities, likely causing conformational changes that affect promoter recognition or cooperative binding at the target promoter (161). However, the extent to which phosphorylation is required for DNA binding and transcriptional regulation in viva varies for different response regulators. In the Nar system of E. coli (reviewed by 159), which 77 regulates nitrate/nitrite-responsive anaerobic respiratory pathways, phosphorylation of NarL is absolutely required for DNA binding and regulatory activity (11). Conversely, in E. coli UhpA-mediated regulation of sugar-phosphate uptake, high-copy expression of the unphosphorylated UhpA D54N protein allowed phosphorylation-independent activation of the uhpT promoter (180; 181). If unphosphorylated AbsA2 was fiinctional in viva, high-copy expression of the absA2 D54E allele, on plasmid pTBA175 (Figure 6), might have repressed antibiotic synthesis. Since it did not (Figure 11B), it appears that AbsA2 regulatory activity is strongly dependent on phosphorylation. It is noteworthy to contrast AbsA2 with several other recently discovered antibiotic cluster-linked regulators that are closely related in sequence. One such protein is RedZ, a red-cluster-linked activator of redD transcription (182). The amino acid sequence of RedZ has end-to-end similarity to AbsA2, including the putative helix-tum- helix region, with 27% identical residues overall. However RedZ lacks the conserved aspartate residue that is normally the site of phosphorylation in response regulators (65). A homolog of RedZ, DnrN, is found in the daunorubicin gene cluster of S. peucetius, where it regulates dan, which encodes a SARP pathway-specific regulator of the dnr cluster. Although the DnrN protein sequence has retained the conserved aspartate, other residues of the phosphorylation pocket are not conserved and phosphorylation appears to not be involved in DnrN function in viva (59). Thus DnrN and RedZ appear to serve as regulators in the unphosphorylated state. It is not known if a modification other than phosphorylation regulates the activity of RedZ or DnrN. Our results have shown that the AbsA two-component system is a negative regulator of the calcium-dependent antibiotic, actinorhodin, and undecylprodigiosin. For 78 the later two antibiotics, absA negatively regulates the SARP pathway-specific activator genes. It will be important to determine whether AbsA2 regulates the SARP genes directly. For the cda cluster, interesting questions are whether AbsA2 regulates the SARP homolog, cdaR, or whether it regulates another, as yet unidentified cda regulator, or directly represses cda biosynthetic genes. 79 CHAPTER 4 TEMPORAL EXPRESSION OF red AND cda PATHWAY-SPECIFIC REGULATORS AND THEIR DEPENDENCE ON AbsA 8O INTRODUCTION Global negative regulation of antibiotics by AbsA has been demonstrated to act through transcriptional control of redD and actII-4 pathway-specific regulators (PSRs) (1). While these genes constitute possible targets of AbsA2, the red gene cluster contains a second PSR redZ, which is required for expression of redD, but can not activate red biosynthetic gene expression in the absence of redD (182). The effect of AbsA on redZ expression is unknown. Another plausible target of AbsA2 is the putative PSR of the cda gene cluster, cdaR, the sequence of which was recently made available by the S. coelicolor genome sequencing project (http://www.sanger.ac.uk/Projects/S_coelicolor/). Very little is known about the temporal expression of the cda gene cluster of S. coelicolor. The deduced cdaR gene product shows sequence similarity to the N—terminal region of Streptomyces antibiotic regulatory proteins (146), including redD and actII-4, that contain an OmpR-like DNA-binding motif (183). A chromosomal disruption of cdaR was found to eliminate CDA production, suggesting it had a positive regulatory function (156). Sequencing of the cda cluster also revealed that the absA locus lies within it (Figure 14). The significance of absA association with the cda genes is not yet known. The only other Streptomyces antibiotic gene cluster known to encode a two- component signal transduction system is the rapamycin gene cluster of S. hygroscopicus, which contains genes encoding a putative two-component system of as yet unknown function (149). However, various Gram-positive bacteria which produce class I and class II antimicrobial peptide (AMP) antibiotics do contain two-component genes as part of the antibiotic gene cluster 81 17.5 kb 40 51 58.3 Figure 14. Position of absA with respect to the cda gene cluster. This 58.3 kb region of the cda cluster was reconstructed from sequence data made available by the Streptomyces coelicolor Sequencing Project (The Sanger Centre). Genes shown in white have been named and given putative fimctions based on genetic or functional analysis. cdaR is homologous to pathway-specific activators. Biosynthetic genes cdaPSI, cdaPSII and cdaPSIII encode peptide synthases which catalyze steps in the enzymatic synthesis of the lipopeptide antibiotic CDA. Shaded genes have been assigned putative functions based on sequence similarity to other proteins (annotated in http://www.sanger.ac.uk/Projects/S_coelicolor/). 82 (reviewed by 123; 96). AMPS have a conserved regulatory gene organization in which an open-reading frame encoding an autoinducing peptide pheromone precedes the genes of a two-component system. While demonstration of response regulator binding to operons within AMP clusters is still pending, disruption of either the autoinducer or two- component genes abolishes AMP production (47; 10; ‘98). It is hypothesized that two- component systems regulate most or all of the regulatory, biasynthetic and immunity/transport operons of AMP clusters (123; 96). Although the precise target(s) of AMP two-component systems remains to be elucidated, unlike AbsA, these systems have not been implicated in regulation of genes outside of the cluster in which they are found. Given the recent revelation of the association of absA with the cda cluster, and dependence of other PSRs on AbsA, it was of interest to determine whether inhibition of CDA synthesis in the Abs' mutant was correlated with transcriptional regulation of cdaR. Similarly, I examined the effect of AbsA mutants on the expression of refl and the possibility that AbsA-mediated regulation of redD was a consequence of its effect on red. Finally, growth-phase dependent expression of PSRs was compared to that ofabsA to establish temporal relationships between AbsA activity and the appearance of PSRs of the red and cda clusters. MATERIALS AND METHODS Time-course analyses of pathway specific regulators redD, refl, and cdaR transcripts were performed using high-resolution Sl nuclease protection assays. The same RNA samples used for Si analyses of the absA locus were used here, therefore, 83 growth conditions and RNA isolation procedures were as described in the Materials and Methods of Chapter 3. Double-stranded DNA probes were also labeled and synthesized under the same conditions described above (Chapter 3, Materials and Methods). A redD probe of 497 bp, with a predicted 330 nt 81 product (167; l), was generated from primers and template described in Aceti and Champness (1998). The expression of redZ was evaluated with a 405 bp probe synthesized from template pIJ4132 (White and Bibb, unpublished) with forward primer WC43 (5' AGATCTTGGAGCGGGAACTCTC CCTGC) and ”P-labeled reverse primer WC96 (5' GTCGCAGCACACACCAGGA CACG). Previous studies of this gene predict this probe would produce an 81 product of 141 nt (65). A cdaR probe of 584 bp was amplified from S. coelicolor 11501 chromosomal DNA with forward primer WC106 (5' GGCGCACTGACGAAA GCAAGGGC) and 32P-labeled reverse primer WC94 (5' CCGCCCACCGTAAGACC TCGGCC). The transcription start site for this locus had not previously been determined. 81 nuclease assay conditions were identical to those used for absA. Likewise, the same glk probe was used to normalize RNA loading, and S1 product sizes were estimated alongside 32P-labeled ¢XI74lHinfl molecular weight markers (Promega). RESULTS Dependence of Pathway-Specific Regulators rail and redD on AbsA. Red antibiotic synthesis is under the control of two pathway-specific regulators, RedD and RedZ (reviewed by 35; 31). The RedD activator is required for Red synthesis (144; 50) and overexpression of redD causes an increase in red biosynthetic gene expression and Red production (122; 167). Expression of redZ is required for both Red 84 synthesis and redD expression, suggesting that RedZ is an activator of redD transcription (182). Negative regulation of Red synthesis by AbsA was previously demonstrated to act through transcriptional regulation of redD (1). Therefore, it was of interest to see if the dependence of redD on AbsA2 was mediated through redZ, a possible target of AbsA2. Similarly, while redD (167), red (182) and absA (Chapter 3) have all been shown to be temporally regulated, I sought to gain a better understanding of the temporal relationship between the global regulator AbsA2 and expression of pathway-specific regulators which are possibly under its control. As such, S1 nuclease protection assays were performed to evaluate redD and redZ transcription with the same RNA samples used to analyze absA. These were extracted from 18, 30, and 54 hr cultures representing time points prior to and during antibiotic synthesis. AbsA-mediated regulation of red PSRs was tested in wild type (J 1501) versus absA mutant strains C542 (Abs’) and C570 (Pha). The redD transcript gave a single product of 330m (Figure 15) with a temporal pattern and AbsA2 dependence in agreement with that observed previously (1). The 18 hr cultures although turbid, still had rather sparse growth and no pigmentation. As expected, no redD transcript was discemable. redD was not present until 30 hours when it was strongest in J1501, coinciding with the Red pigment observed in this culture at that time. The level of redD in 30 hour C570 was surprisingly low since this strain reproducibly demonstrated early onset of Red on SpMR agar plates. 85 bp,. f‘r‘,‘ (‘6 ‘5 3171 m3 chm" v $88 £55. .388 12 F" v— o —-a 88$. 9.8:; 328$ E :00 :00 :00 E Fishrll 30hr1154hrl Figure 15. High-resolution Sl nuclease protection analysis of redD and redZ transcripts. Temporal regulation of redD and redZ was followed by analyzing RNA samples isolated from S. coelicolor strains grown for 18,30 and 54 hours 1n liquid SpMR. The dependence of redD and redZ expression on the AbsA two-component system was tested by examining their transcription in strains 11501 (wild type), C542 (Abs'), and C570 (Pha) at each time point. Glucose kinase (gIk) transcript was measured to normalize the quantities of RNA loaded at each time point. 86 Nevertheless, the low level of signal did coincided with the lack of pigmentation observed in this culture. By 54 hours, J1501 cultures had attained a strong red hue, while the C570 cultures were maroon, having overtaken J1501 in Red as well as visibly producing Act. At no point was there any pigmentation in the Abs' C542 cultures. As expected, based on previous observation (1), the level of redD transcript in the C542 Abs' strain was extremely low at all times tested. Contrary to the marked temporal regulation seen for redD, redZ (141 nt signal) was strongly expressed at all stages of growth examined in this culture medium. There was, however, an increase in redZ from 18 to 30 hours, indicating that this gene is temporally regulated. The fact that it was expressed so strongly hours before redD is in contrast to the almost simultaneous upregulation of redD and redZ reported by White and Bibb (1997). Equally apparent from the transcript data was the lack of dependence that refl had on AbsA2, as witnessed by the similarity in signal intensities for each of the strains tested. Dependence of cdaR Expression on AbsA No previous transcript data was available for cdaR, but its expression was hypothesized to be growth-phase dependent like other antibiotic PSRs. The intergenic region between the translation start codon of the deduced open reading frame for cdaR and the stop codon of upstream gene SCE8.09 is 744 bp. 81 nuclease protection assays of cdaR transcript employed a 584 bp double-stranded DNA probe that protected a region 439 nt upstream of the predicted translation start codon. Expression of cdaR was evaluated at 18, 30, and 54 hours in strains J1501 (wild type), C542 (Abs-) and C570 87 (Pha). S1 nuclease protection results presented in Figure 16 revealed a major product of approximately 500 nt, corresponding to a transcription start site about 380 nt upstream of the translation start codon. Although undigested probe was present in some samples, its inconsistency lead me to believe that it did not represent an additional transcription start site beyond that of 380 nt. There were some minor signals of smaller size that could represent additional promoters of this gene; however, these were not firrther analyzed in this study. Inspection of sequence upstream from the region encompassing the transcription start point identified hexamers around -10 and -35 with strong similarities to consensus sequences proposed by Strohl (1992) to be recognized by Streptomyces RNA polymerase containing and E. coli 67° -like sigma factor (data not shown). Comparison of cdaR expression in J1501 between 18 and 30 hours suggested that this gene was under growth-phase dependent regulation. It also appeared as though cdaR was expressed considerably earlier than redD under these growth conditions, but not earlier than refl. It was not clear fiom the results presented in Figure 16 whether AbsA regulated the expression of cdaR. A comparison of 18 hour J 1501 and C542 transcripts indicated that there was a strong effect. Given that C542 is an Absi strain that does not produce antibiotics and cdaR expression was dramatically increased in the Abs’ strain, the transcript results suggested that CdaR was a negative regulator. In contrast, a chromosomal disruption of cdaR reportedly blocked synthesis of CDA (156) implying that CdaR was an activator of CDA. An additional point of confirsion was that the 18 hour C570 signal was also stronger than that of J1 501. C570 is an overproducing Pha strain, but like the Abs' strain showed increased expression of cdaR in comparison to 88 :3 FAA CG... QGA 5 B 36’ 2 —° 3 a D B a 3 332-, $216: 53%. ‘23 § 8 E § 8 E 5’» 9. E 3 L: DU :4 U U 'v—i U U 2 I 18hr ll 30hr l I 54hr l 713» 553) " ’ ” ‘AbSAZ'HiSw ‘* era-u— 18.4P Figure 18. Purification of AbsA2 proteins overexpressed in E. coli. (A.) SDS-PAGE (12%) analysis oinS1o-tagged AbsA2 D54E protein purified fi'om E. coli BL21(DE3) containing pTBA240. Lanes: (1) Molecular weight markers; (2) Whole-cell lysate of BL21(DE3)/pET16b, 3 hours post-induction; (3) Whole-cell lysate of BL21(DE3)/pTBA240, 3 hours post-induction; (4) Souble phase of BL21(DE3)/pTBA240 whole-cell lysate; (5) 6M urea-solubilized cell lysate pellet fi'om BL21(DE3)/pTBA240; (6) Ni2+-column flow through of 6M urea-solubilized pellet; (7) Niz+-column wash with binding buffer; (8) Ni2+-column wash with wash buffer (60mM imidazole); (9) Dilute eluate of 6M urea-solubilized pellet from BL21(DE3)/pTBA240; (10) Molecular weight markers. (B.) SDS-PAGE (12%) analysis of Ni2+-column purified Hisro-tagged AbsA2 and AbsA2 D54E. Lanes: (1) Molecular weight markers; (2) Eluate of 6M urea-solubilized AbsA2-His“) fiom BL21(DE3)/pTBA235; (3) Eluate of 6M urea-solubilized AbsA2(D54E)-Hisro fi'om BL21(DE3)/pTBA240. In- Vitra Phosphorylation of AbsA2-His". with Acetyl Phosphate Response regulators can be readily phosphorylated by small molecular weight phosphate donors such as acetyl phosphate (107; 117). Moreover, many cases exist where His-tagged RRs retained both in-vitra phosphorylation and in-viva physiological activity without the necessity of cleaving the fusion tag (137; 132; 70; 105; 108). Thus, I 108 sought to examine the requirement of Asp54 for the phosphorylation of AbsA2 by in- vitra phosphorylation assays with AbsA2-His“, and AbsA2 (D54E)-Hisro using enzymatically synthesized 32P-acetyl phosphate as a phosphate donor. Prior to use in in- vitra phosphorylation, purified 6M urea-denatured AbsA2-His") proteins were refolded by the gradual removal of urea through dialysis. Refolding was also attempted while AbsA2-His“) was still bound to the nickel column by applying a 20 ml linear gradient of 6 M to O M urea. Quon, et al. (1996) used E. coli acetate kinase to synthesize 32P-acetyl phosphate for subsequent phosphorylation of RR CtrA-Hiss. Phosphorylation of renatured AbsA2-H1810 (after dialysis or gradient renaturation) was unsuccessful by this method. However, from the controls that were included in the assay it did appear as though the enzymatic synthesis of 32P-acetyl phosphate was proceeding to completion. Thus, it was suspected that my attempts to refold the denatured AbsA2-His“) was not producing an active conformer of AbsA2. Overexpression and Purification of AbsA2-His; from S. lividans Unsuccessful phosphorylation of heterologously produced-renatured AbsA2- Hism led me to overproduce AbsA2 in a Streptomyces expression system. It was hypothesized that weaker expression from the tipA promoter, together with production in its natural environment, might favor accumulation of AbsA2-His6 in a soluble and active form. The same allele of absA2 used to construct pTBA235 was ligated into pIJ4123 (168) to create pTBA270, which encoded an N-terminal Hiss fusion to AbsA2. Fusion protein expression was under the control of the thiostrepton-inducible tipA promoter. Given the slow growth of Streptomyces relative to E. coli, S. Iividans 1326/pTBA270 was 109 grown for 16 hours prior to induction and an additional 12 hours post induction. Comparison of the whole-cell lysates from strains 1326 and 1326/pTBA270 did not clearly indicate a band due to overproduced AbsA2-His; fusion protein (Figure 19, lanes 1 and 2). Nevertheless, there was a significant product with an Mr of ~29kDa in the whole cell lysate and the soluble and particulate fi'actions of the lysate which could be the fusion protein (Figure 19, lanes 2, 3 and 4). When the soluble phase of the lysate was purified on a nickel column, a 29 kDa protein was eluted at better that 90% purity (Figure 19, lanes 8 and 9), which was in close agreement with the predicted size of 26.2 kDa for AbsA2-Hig. 12345678910 rd), .1. “Manama“: w! .‘ a A A N 4 18.4 ‘ 14.3 ‘ 6.2 Figure 19. Purification of AbsA2-Hig overexpressed in S. Iividans. SDS-PAGE (12%) analysis of Hisa-tagged AbsA2 purified from S. Iividans 1326 containing pTBA270. Lanes: (1) Uninduced 16-hour S. Iividans 1326; (2) Whole-cell lysate of 12-hour post induction S. Iividans l326/pTBA270; (3) Soluble phase of 1326/pTBA270; (4) 6 M urea- solubilized pellet fi'om 1326/pTBA270; (5) Molecular weight mkers; (6) Ni2+-column flow through from soluble phase; (7) Ni2+-column wash with 60 mM imidazole; (8) Ni”- column soluble-phase post-wash eluate (Fraction 1); (9) Soluble-phase eluate (Fraction 2); (10) Molecular weight markers. 110 It was noticed after repeated purification of 1326/pTBA27O cultures that very low yields of AbsA2-Hig were obtained in the final eluted fraction. Examination of a low imidazole wash fiom the nickel column revealed that most of the 29 kDa AbsA2-His; was eluted in this fraction, suggesting that at least part of the reduction in yield was due to deterioration of the nickel-column affinity matrix from repeated use. Therefore, the wash sample was dialyzed and analyzed by SDS-PAGE, revealing that AbsA2-Hig represented the major single band and approximately 30% of the total protein in this sample. In- Vitra and In-viva Analysis of AbsA2-Hi“ Purified AbsA2-Hig produced in Streptomyces was tested by in-vitra phosphorylation experiments as described above. Once again I was unable to demonstrate phosphorylation of AbsA2-Hig. Therefore, it was of concern that the His; and Him fusion peptides might be rendering AbsA2 inactive. To test the activity of AbsA2-Hig in viva, pTBA270 was transformed into C570 (see Chapter 3), which carried a chromosomal allele that encoded AbsA2 with a D54E mutation at the proposed site of phosphorylation. Strain C570 caused an antibiotic overproducing phenotype (Pha) which has been associated with an inactive form of AbsA2. When pTBA270 was introduced into C570, antibiotic production was regulated at the wild type level (Figure 20). Conversely, when pIJ6021 (tipA-containing overexpression vector without insert (168) was introduced into C570, there was no effect on overproduction of antibiotics (Figure 20). This suggests that the AbsA2-His; fiision protein encoded by pTBA270 was 111 C570 C570/ pIJ6021 Figure 20. In-viva analysis of pTBA270. The biological function of AbsA2-His; encoded by pTBA270 was tested by its ability to complement the Pha phenotype of strain C570, which carries a chromosomal absA2 (D54E) mutation. Plasmid pTBA270 was transformed into strain C570. Antibiotic production of C570/pTBA270 was compared to that of C570 and wild type J1501. As a control, pIJ6021 (Hisr; overexpression plasmid identical to pTBA270 without the absA2 insert) was also transformed into C570 to test its ability to complement the Pha phenotype. Strains were analyzed after 4 days of growth on SpMR agar. functional when expressed in viva. Therefore, it was hypothesized that the inability to phosphorylate AbsA2-H156 in vitro may have been due to an inherent instability of this protein under the purification or assay conditions, or a faulty phosphorylation reaction. An alternative to the enzymatic generation of 32P-acetyl phosphate used in this study is the chemical synthesis of 32P-labeled small molecular weight phosphate donors. The latter method has been more widely used for in-vitra phosphorylation of Rs (107; 117) and will be implemented in firture trials with AbsA2. 112 Overexpression and Purification of 'AbsAl from E. coli Although small molecular weight phosphate donors have proven useful for in- vitra phosphorylation of various response regulators, there is at least one report in which it did not serve as a suitable donor (107). A preferred approach with several two- component systems has been the purification of cognate HR for use as an in-vitra phosphate donor (148; 178; 153; 42). This approach allows for more specific and efficient phosphorylation of Rs (161) as well as the opportunity to study properties of the HK and HK-RR interactions. Given difficulties phosphorylating AbsA2-His proteins with acetyl phosphate, I decided to overproduce and purify AbsAl to use as a phosphate donor in vitro. Since AbsAl is membrane bound, a truncated form, 'AbsAl, which lacks the N- terrninal transmembrane domain but possesses the entire transmitter and C-terminal domains, was expressed to favor accumulation in the cytoplasm. The truncated alleles of wild type 'absAI and 'absAI (HZOZL), which encodes an H202L change at the putative site of phosphorylation, were ligated into pMAL-C2 to produce N-terminal maltose- binding protein fusions (MBP-‘AbsA1) (Figure 17). Protein expression from pTBA350 (MBP-‘AbsAl) and pTBA360 (MBP-‘AbsAl- [1-1202L]) was IPTG inducible from a tac promoter. The MBP-‘AbsAl fiisions have a predicted Mr of 90 kDa. The whole cell lysates from these cultures demonstrated a band of minor intensity at about 90 kDa in comparison to the same strain carrying the pMal-c2 vector with no insert (data not shown). The soluble phase of the JM109/TBA350 cell lysate (Figure 21, lane 1) was purified on an amylose resin column as described in Materials and Methods. Purified MBP-‘AbsAl fusion protein with a predicted Mr of 90 kDa produced a diffuse band of 113 products ranging from about 75 to 90 kDa (Figure 21, lane 2) suggesting possible proteolysis in the host strain E. coli JM109. pMAL-c2-generated fusion proteins contain a Factor Xa recognition sequence situated between MBP and 'AbsAl. Cleavage of purified MBP-'AbsAl with Factor Xa (Figure 21, lane 4) produced a major band in the range of ~42 to 47 kDa, which was probably composed principally of the liberated MBP domain (~42 kDa) and some 'AbsAl (predicted M,of 47 kDa). Another major band represented a product of ~38 kDa and there was also one minor band at ~30 kDa, again suggesting either proteolysis of the fusion protein in the host strain or secondary cleavage sites within 'AbsAl that were recognized by Factor Xa 175 83 62 47.5 32.5 25 16.5 Figure 21. Purification of 'AbsAl expressed in E. coli. SDS-PAGE (10%) analysis of maltose binding protein fusions to truncated absA] ('absAI) purified fi'om E. coli JM109 containing pTBA350. Lanes: (1) Soluble-phase whole-cell lysate fi'om 3-hour post- induction JM109/pTBA350; (2) Post-wash eluate of soluble-phase JM109/pTBA350 purified on an amylase resin column; (3) Concentrated JM109/pTBA350 eluate; (4) Concentrated JM109/pTBA350 eluate treated with Factor Xa; (5) Molecular weight markers. 114 Autophosphorylation of 'AbsAl and Phosphorylation of AbsA2-His; Autokinase activity of MBP-'AbsAl and the 'AbsAl Factor Xa cleavage products was tested by addition of [y-3ZP] ATP. The major phosphorylation product in the untreated MBP-'AbsAl sample (Figure 22, lane 1) corresponded to the same diffuse band from ~75 to 90 kDa observed by Coomassie blue staining (Figure 21, lane 2). Afier proteolysis with Factor Xa, the major phosphorylation products corresponded to the 38 and 30 kDa proteins (Figure 22, lane 4) previously observed by Coomassie blue staining (Figure 21 , lane 4). An equivalent concentration of MBP-’AbsAl H202L was treated MBP-’AbsAl D Figure 22. In-vitro phosphorylation of AbsA2-Hig by 'AbsAl-phosphate. Radioactively labeled MBP—'AbsAl-phosphate and 'AbsAl-phosphate were prepared by incubating purified MBP-'AbsAl and 'AbsAl (cleaved fiom MBP by Factor Xa) with [y- 32P] ATP in the presence of Mg“. After incubation, semi-purified soluble-phase AbsA2- Hisa from S. lividans was added to each sample. Samples were analyzed by SDS-PAGE (10%) and autoradiography. Lanes: (1) MBP-'AbsAl + [y-3ZP] ATP; (2) MBP-'AbsAl- 32P + AbsA2-Hig; (3) Molecular weight marker; (4) Factor Xa-treated MBP-'AbsAl + [y-3’P] ATP; (5) 'AbsAl-32P + AbsA2-His6. 115 with Factor Xa and subjected to the same phosphorylation conditions, demonstrating no phosphorylation of the 'AbSAl H202L mutant (data not shown). These results indicate that phosphorylation in MBP-'AbsAl is occurring in the 'AbsAl portion of the firsion protein, and that His202 of AbsAl is required for phosphorylation. MBP-'AbsAl and the 'AbsAl cleavage product were phosphorylated with [y-32P] ATP and used as donors for the in-vitra phosphorylation of AbsA2-His; produced in 1326/pTBA270. Semi-purified AbsA2-Hig was added to each of MBP-'AbsAl- phosphate (Figure 22, lane 2) and 'AbsAl -phosphate (Figure 22, lane 5). Addition of AbsA2-Hig to 'AbsAl-phosphate produced a strong signal at ~29 kDa (Figure 22, lane 5), which coincided with the size of AbsA2-H156 in the semi-purified sample. The fact that this product did not appear when incubated with MBP-'AbsAl-phosphate suggested that its phosphorylation was dependent upon the liberated 'AbsAl domain. In contrast, heterologously produced AbsA2-His“), renatured by dialysis after 6 M urea solubilization and purification, was not phosphorylated under similar conditions. Addition of AbsA2- Hiss to MBP-'AbsAl-phosphate produced a faint signal at ~36 kDa (Figure 22, lane 2). This same signal appeared when [y-nP] ATP was added to soluble phase 1326/pTBA270 whole-cell lysate in the absence of MBP-'AbsAl or 'AbsAl (data not shown). Therefore, this signal appeared to represent a contaminating protein whose phosphorylation did not depend upon the presence of 'AbsAl. Preliminary evidence from these experiments suggests that AbsA2-Hig could be phosphorylated from 'AbsAl-phosphate after cleavage of the MBP fitsion domain. 116 DISCUSSION Purification of AbsA2 and AbsA2 D54E was pursued for use in in-vitra phosphorylation experiments to support genetic evidence for the role of Asp54 in AbsA2 phosphorylation and activity. Highly purified AbsA2-His“) proteins renatured from the insoluble fraction of E. coli cultures, and soluble phase AbsM-Hi36 from S. lividans were not phosphorylated in vitra by acetyl phosphate. Subsequent overproduction of MBP- ‘AbsAl fusions was performed to provide an alternative phosphate donor for AbsA2-His fusions. When 'AbsAl-phosphate was reacted with semi-purified AbsA2-His; from C270, preliminary results suggested cognate HK-RR phosphoryl-group transfer to form AbsA2-Hiss-phosphate. Formation of an active phosphorylated conformer by the AbsA2-Hiss fusion was fisrther supported by its ability to complement mutant AbsA2 D54E in viva. In addition, comparison of autokinase activity by 'AbsAl and 'AbsAl H202L provided evidence that His202 is the site of phosphorylation in AbsA] due to the inability of 'AbsAl H202L to become phosphorylated under identical conditions in which wild type 'AbsAl was readily phosphorylated. Numerous two-component regulators overexpressed in E. coli and renatured after 6 M urea solubilization have been successfully phosphorylated in vitra by low molecular weight phosphate donors (137; 132; 108). Initial attempts to phosphorylate AbsA2-His"), produced in E. coli and renatured after 6 M urea solubilization, were unsuccessfirl utilizing an enzymatic preparation of 32P- acetyl phosphate. Similarly, AbsA2-His; purified from the soluble phase of S. lividans was not phosphorylated in this reaction. It is uncertain at this time if these experiments were unsuccessful because the AbsA2-His fission conformers were inactive as a result of purification or reaction conditions, if there 117 was a problem with the enzymatic synthesis of acetyl phosphate, or if AbsA2 can not utilize acetyl phosphate as a phosphate donor. The affinities of response regulators to small molecular weight phosphate donors are low (161) and differences in the reactivities of Rs to acetyl phosphate have been reported (118). At least one RR, CheB, has been reported that can not use acetyl phosphate as a phosphate donor (107). An alternative method to that presented in the Materials and Methods for in-vitra phosphorylation of RRs utilizes chemically synthesized 32P--labe1ed small molecular weight phosphate donors (107; 117). Given that certain studies of two-component protein behavior are facilitated by phosphorylation with species such as acetyl phosphate (e. g., AbsA2-P DNA-binding, AbsAl phosphatase activity), we will continue to investigate expression, purification, and reaction conditions necessary to achieve small molecular weight donor in-vitra phosphorylation of AbsA2. An alternative method that offers specific and efficient in-vitra phosphorylation of Rs has been routinely demonstrated using purified truncations of cognate 1-IKs(148; 178; 153). Preliminary data presented here suggests that AbsA2-His; was successfully phosphorylated in vitra by 'AbsAl -phosphate. Nevertheless, fisrther work needs to be done to optimize overexpression and purification of 'AbsAl and AbsA2-Hig. The conditions for consistent recovery of highly-purified AbsA2-His; from S. lividans need to be reestablished. Similarly, multiple phosphorylation products due to proteolysis of MBP-'AbsAl tend to confound the results of the in-vitra phosphorylation reactions. Proteolysis of MBP-'AbsAl might be avoided by expression of pTBA350 in an alternative host, such as BL21 which has protease gene knockouts, or possibly by 118 expressing only the transmitter domain of AbsAl instead of the transmitter and C- terrninal domains. Finally, while the scope of the experiments presented in this chapter were originally oriented at analyzing the role of Asp54 in phosphorylation and activity of AbsA2, the utility of obtaining purified AbsAl and AbsA2, and establishing the conditions for in-vitra phosphorylation, project into various fisture pursuits of this project. I have already mentioned determining the phosphatase activity of AbsAl. In addition, AbsA2 in the phosphorylated and unphosphorylated states will be used in gel-shift and footprint assays to determine targets and a possible consensus binding sequence(s) of AbsA2, as well as the requirement for phosphorylation on DNA binding. Another projected use for purified AbsA2 is in the preparation of antibodies for immunoprecipitation and in-viva phosphorylation studies aimed at differentiating between the kinase and phosphatase default models of the AbsA mechanism discussed in Chapter 3. 119 CHAPTER 6 AbsA2 HOMOLOGUES IN OTHER STRAINS OF ST KEPT OMYCES 120 INTRODUCTION Streptomyces are presently the single greatest natural source of chemotherapeutic agents for health and agriculture (165). Nevertheless, a. paucity of knowledge on the genetic regulation of antibiotic synthesis in the genus as a whole has traditionally led industry to depended on expensive and laborious random mutagenesis and screening programs to achieve improvements in product yield. Currently, numerous groups are working in the model organism S. coelicolor to piece together the complex array of pathways involved in antibiotic regulation (reviewed in Chapter 1; 30; 74; 35). It is hoped that many of the regulatory processes uncovered in S. coelicolor will be conserved in other streptomycetes such that a general model for the regulation of antibiotic synthesis can be constructed. New data presented here, along with existing evidence from Streptomyces and other genera, lend support to generalized conservation of developmentally regulated processes. Precedent for the evolutionary conservation of genetic regulation of developmental processes such as sporulation and antibiotic synthesis exists in other Gram-positive bacteria. Producers of class II antimicrobial peptides (AMP) from the genera Camabacierium and Lactabacillus show remarkable similarities in the organization and function of genes within clusters encoding their function (reviewed by 96). Regulatory features conserved in these clusters include genes encoding a peptide pheromone precursor, a two-component system, and an ATP-binding cassette (ABC) exporter. Similarly, gene clusters for the regulation and synthesis of lanbiotics by species of Lactabacillus, Bacillus, and Staphylococcus show pronounced conservation of gene organization and fisnction (reviewed by 96). Once again, conserved regulatory elements 121 include signal precursors, two-component regulators, and transport systems. While these examples demonstrate conserved regulation of the products of a single gene cluster within which both regulatory and biosynthetic genes are found, more complex regulatory schemes may also be conserved. Conservation of a two-component regulator that functions pleiotropically in differentiation is exemplified by SpoOA. The phOSphorylated state of SpoOA triggers the initiation of sporulation in B. subtilis by activating and/or repressing transcription of several other regulators encoded on independent operons (reviewed by 68). Youngman and collaborators (21) presented compelling evidence for evolutionary conservation of SpoOA in phylogenetically diverse species of Bacillus and Clastridium. Based on the similarity of partial or complete DNA sequences, spOOA was proposed to be conserved in _ each of 8 Bacillus and 6 Clastridium species analyzed. Furthermore, spaOA homolog gene disruptions performed in B. anthracis and C. acetabutylicum demonstrated sporulation deficient phenotypes similar to spaOA mutants of B. subtilis. Thus, both structural and functional homology were shown to be conserved. In addition to the examples sited in other Gram-positive bacteria, several observations suggest that numerous aspects of antibiotic regulation may be conserved among streptomycetes. First, throughout the genus Streptomyces, antibiotic synthesis is growth-phase dependent, which logically suggests an evolutionarily conserved genetic basis for their temporal regulation. A possible link to temporal regulation is the cell- density dependent accumulation of structurally similar y-butyrolactones, which has been implicated in signaling the onset of morphological and physiological differentiation in several species of Streptomyces (reviewed in Chapter 1). As seen in the production of 122 peptide antibiotics in other Gram-positive bacteria, a routinely conserved feature of Streptomyces antibiotic gene clusters is the presence of one or more pathway specific regulators (reviewed by 30). On a broader scale, preliminary evidence for the conservation of many S. coelicolor pleiotropic antibiotic regulatory genes in other streptomycetes has been implied fi'om PCR and Southern hybridization data (absB (136); afsQ (84); (1st (115); thS (32); bIdA (101)). In this chapter, I examine the conservation of the global antibiotic regulator AbsA2 in the genus Streptomyces. An attractive feature of AbsA2 is that it acts as a global negative regulator of antibiotics, so that by knocking it out, antibiotic production begins earlier and reaches greater concentrations. If this mechanism were conserved, it could have important economic implications in industrial relevant strains. Directed mutagenesis of AbsA2 homologs would offer an alternative to traditional random mutagenesis and screening protocols for increasing antibiotic yields. Here, I present preliminary data that suggests the existence of AbsA2 homologs in each of ten industrial strains of Streptomyces examined. MATERIALS AND METHODS Bacterial Strains and Growth Conditions Streptomyces strains utilized in this study were S. coelicolor M600, S. albus, S. ambafaciens 2035, S. clavuligerus, S. halstedii JM8, S. halsredii 2581, S. lincalnensis, and S. peucetius. All studies were done with laboratory stocks of chromosomal DNA or samples received from other labs. 123 Escherichia coli K12 strain DHSor was used for proliferation of plasmids that carried inserts to be sequenced. E. coli was grown in culture tubes containing 10 ml of L broth supplemented with 100 pg/ml ampicillin. Cultures were incubated for 12 to 16 hours at 37°C and 250 rpm prior to plasmid extraction and purification on QIAprep spin columns (Qiagen). PCR Amplification of Putative absA2 Homologs Amplification of absA2 homologs was based on the strategy for spaOA homolog amplification (21). Members of the response regulator superfamily share extensive structural (primary, secondary and tertiary) similarity in the N-terminal receiver domain (176; 11). Sequence similarities are even stronger within subfamilies, such that a primer designed to a canserved region of the receiver domain would be expected to prime PCR synthesis from most response regulators of this subfamily. The C-terminal effector domain of Rs is more highly divergent such that primers specific to important fisnctional regions of this domain (e. g., the helix-turn-helix [HTH] DNA-binding motif) would be expected to prime more specifically for homologs of absA2. In addition, streptomycetes have a codon preference favoring a high GC content, especially at position three which has a G or C over 90% of the time (16). Consequently, where a codon allowed either G or C at position three, degeneracy was designed into the absA2 homolog primers to provide this option. Implementing these parameters, 5' forward primers were designed to conserved blocks around residues Asp9 (HO, 5' GCS GAC GAC GAG ACS ATC ATC CGS GCS; where S = G or C) and Asp54 (P9, 5' GCS CTS CTS GAC ATC CGS ATG CCS G) of the receiver domain (Figure 25). To impart 124 specificity toward absA2 homologs, a 3‘ reverse primer (P8, G TGS AGS CGC TGS GCG ATC TCS GCG) was designed around the HTH motif of the effector domain. Primer P8 is specific to the first helix of the HTH motif. While this is not the so-called 'recognition helix', it is projected to possess residues that are important in DNA recognition and binding (67; 11). Five of the eight codons recognized by P8 encode amino acids that are moderately to highly divergent with respect to a HTH motif alignment of RRs of the same subfamily (11). Of these, the most highly divergent codons lie at the 3' and 5' ends of P8. PCR amplification was carried out in a 100 pl reaction volume containing 100 ng chromosomal DNA template, 1X PCR buffer (with 1.5mM MgClz), 5% glycerol, 0.2 mM of each dNTP, 40 pmole of each primer, and 2.5U Taq polymerase. Thermal cycler conditions were: denaturation at 95°C for 5 min, followed by 30 cycles of 1 min at 94°C, 1.5 min at 60°C, and 1 min at 72°C, and a final extension at 72°C for 10 min. PCR amplification products were sized by separation on a 1.5% agarose gel with comparison to pBR322 HaeIII DNA molecular size marker (Boehringer Mannheim). absA2 Homolog Identification and Sequencing Southern hybridization of PCR amplification products was utilized to identify putative absA2 homologs. PCR amplification products were separated on a 1.5% agarose gel. DNA was transferred from the gel to a positively charged nylon membrane (Hybond-N+, Amersham) by capillary transfer (147). DNA was fixed to the membrane by UV. crosslinking. Hybridization and colorometric detection of an absA 2-dioxigenin labeled probe was performed as recommended (The Genius System User's Guide for 125 Filter Hybridization, Boehringer Mannheim) with the exception that all washes were performed at room temperature. Probes were prepared by purifying S. coelicolor absA2 PCR products with Wizard PCR preparatory columns (Promega) and using random primed DNA labeling of 100 to 300 ng of absA2 DNA overnight as recommended (The Genius System). In addition to probes prepared from the S. coelicolor P8/P9 and P8/P10 PCR products, another primer, P11 (5' SSWSAGGC ASSWSCCSCCSSWSGCSAC; S = G or C; W = A, C, G, or T), was used in combination with P9 to amplify a region of the AbsA2 receiver domain internal to the P8/P10 product to use as a probe against P8/P10 products. ' Sequencing of putative absA2 homologs was performed by cutting the P8/P10 PCR product of interest from 1% low-melting-point agarose gel and purifying with a QIAquick Gel Extraction Kit (Qiagen). The purified product (10 ng) was used as template for PCR amplification with primers WC8 (5' TTT TAG ACT TGA CGA CGA GAC SAT CAT CCG SGC SGG G) and WC9 (5' TTT TAG ATC TGT GSA GSC GCT GSG CGA TCT CSG CG), which are identical to primers P10 and P8, respectively, except that they contain BlgII restriction sites on their ends. The WC8/W C9 PCR products were digested with BgIII and purified on Wizard PCR preparatory columns (Promega). The absA2 BgIII homologs were ligated into BamHI-digested pBluescript II SK+ (Stratagene) sequencing vectors and sequenced as described by Anderson, et al. (1999). The resulting sequence was compiled, analyzed and compared to that of S. coelicolor absA2 using the Wisconsin GCG software package. 126 RESULTS PCR Amplification of Putative absA2 Homologs from Heterologous DNA Primers specific to highly conserved regions of the receiver domain and the HTH motif of the effector domain were used to amplify putative absA2 homologs from chromosomal DNA of industrially important strains of Streptomyces. Receiver domains of the same subfamily of RRs show extensive sequence similarity. Therefore, primers P10 and P9 were expected to prime numerous RRs of the AbsA2-containing subfamily since they were specific to highly conserved regions around codons for Asp9 and Asp54 of AbsA2. Conversely, primer P8, which was specific to the more highly divergent HTH encoding region of the effector domain, was predicted to permit more specific priming of absA2 homologs. Furthermore, G—C base degeneracy was designed into the third position of codons where either base encoded the same residue to allow for silent mutations in evolutionary divergence, while maintaining the high G-C codon preference of streptomycetes. The PCR products obtained using primer combinations P8/P9 and P8/P10 on chromosomal templates from S. coelicolor and eight industrial stains are shown in Figure 23. S. coelicolor DNA generated a single product of expected molecular size from each primer combination (PS/P9, 360 nt, Figure 23A lane 2; P8/P10, 522 nt, Figure 23B lane 2). Conversely, an assortment of products of various sizes and intensities were amplified from the other strains. Every strain except S. halstedii JM8 and S. albus generated a product of the same size as S. coelicolor M600 fi'om primers P8/P9, some of which were very concentrated (Figure 23A). Primer combination P8/PlO only generated a product equivalent in size to M600 from S. ambafaciens, although both S. griseus and S. albus 127 generated products only slightly smaller or larger (Figure 23B). In addition, each of these P8/P10 products was of very modest abundance in comparison to many of those amplified with P8/P9. A. B. 1234567891011 1234567891011 - - a: ..--= - .- ~ Figure 23. PCR amplification of putative absA2 homologs. Putative homologs of absA2 were amplified with primer pairs P8/P9 (A.) and P8/PlO (B.). Reactions were carried out in 100 pl volumes containing 100 ng chromosomal DNA template, 1X PCR buffer (with 1.5 mM MgClz), 5% glycerol, 0.2 mM of each dNTP, 40 pmole of each primer, and 2.5U Taq polymerase. Thermal cycler conditions were: denaturation at 95°C for 5 min, followed by 30 cycles of 1 min at 94°C, 1.5 min at 60°C, and 1 min at 72°C, and a final extension at 72°C for 10 min. PCR amplification products were sized by separation on a 1.5% agarose gel. Lanes; (1) DNA molecular weight marker pBR322 HaeIII; (2) S. coelicolor M600; (3) S. Iincalnensis; (4) S. halstedii JM8; (5) S. griseus; (6) S. albus; (7) S. avermitilis; (8) S. cinnemanium; (9) S. halstedii 2581; (10) S. ambafaciens; (l 1) DNA molecular weight marker pBR322 HaeIII. Identification of Putative abs/42 Homologs Although major amplification products from P8/P9 corresponded in size to absA 2, there were numerous other products of significant abundance generated in almost every 128 strain. Moreover, the major products amplified from P8/P10 were not of expected size. Therefore, in order to determine if any of these products represented possible absA2 homologs, they were transferred to nylon membranes and hybridized against S. coelicolor P8/P9 or P8/PlO absA2 probes. The Southern blot of P8/P9-generated products provided strong evidence that genes of similar size and sequence to that of absA2 were amplified from each of the strains tested (Figure 24A). Not only was hybridization specific to fragments of the expected size on these blots, but the absA2 probe revealed products of similar size in S. halstedii JM8 and S. albus that were not visible on the gel alone. The lack of hybridization to fi'agments of other sizes together with the appearance of signals in S. halstedii JM8 and S. albus, suggested high annealing specificity for full-length products of similar sequence to that of absA 2. In marked contrast, the Southern blot of P8/P10-generated products (Figure 24B) demonstrated a hybridization pattern virtually identical to the pattern of amplification products seen on the gel. There was no apparent specificity for products of equivalent size to P8/P10 absA2. It was suspected that hybridization conditions for this blot were not stringent enough to exclude annealing based solely on primer recognition. Therefore, a fragment internal to P8/PlO was generated fiom M600 and hybridized to the P8/P10 products. The new P9/Pll probe recognized a 210 bp region of the receiver domain of absA2. Remarkably, when this probe was hybridized to P8/P10 PCR products, essentially all signals corresponded in size to absA2 (Figure 24C). As observed in the P8/P9 blot, hybridization signals were revealed for S. Iincalnensis, S. halstedii JM8, and S. cinnemanium that were not visible on the gel. In their initial rounds of PCR 129 amplification of spoOA homologs using a similar approach, Brown et a1. (1996) obtained products of similar size to those of SpoOA from four out of eight Bacillus species. Here, Southern hybridization of PCR products suggested that homologs of similar size to absA2 existed in each of the strains tested. Indeed, sequence analysis of the amplified region of four putative homolog genes confirmed that these were of similar size to the same region of absA2 (see below). A. B. C. 123456789 123456789123456789 Figure 24. Southem blots of putative absA2 homologs. Putative absA2 homologs generated by PCR from primer pairs P8/P9 and P8/P10 were hybridized to absA2 probes generated from S. coelicolor M600. (A.) P8/P9 PCR products hybridized to a P8/P9 absA2 probe; (B.) P8/P10 PCR products hybridized to a P8/P10 absA2 probe; (C.) P8/P10 PCR products hybridized to a P9/Pll absA2 probe. Lanes: (1) S. coelicolor M600; (2) S. IincaInensis; (3) S. halstedii JM8; (4) S. griseus; (5) S. albus; (6) S. avermitilis; (7) S. cinnemanium; (8) S. halstedii 2581; (9) S. ambofaciens. Sequence Analysis of Putative Homologs Given the promising results obtained from Southern hybridizations, several PCR products were sequenced to examine their similarity to absA2. Although products fiom P8/P9 were more abundant, I chose to sequence P8/P10 products because they were 180 nt larger, encoding a region from the extreme N-terminus through to the HTH motif of AbsA2. PCR products fiom S. griseus and S. ambafaciens were gel purified and used as 130 templates for amplification from primers WC8 and WC9, which were identical to P8 and P l 0 but had BgIII restriction sites designed into their ends to facilitate cloning of the amplified genes. In addition, it was found that P8/P10 PCR products of similar size to that of absA2 could be generated from S. peucetius and S. clavuligerus (data not shown). The sequence of these products was also determined. An alignment of the translated sequences from S. ambafaciens, S. griseus, S. peucetius, and S. clawligerus with that of AbsA2 are shown in Figure 25. Inspection of the consensus sequence from this alignment revealed substantial amino acid similarity over the length of the partial gene products with the exception of the region fi'om residue 1 3 8 to 154 of AbsA2. This region of the RR encompasses a solvent accessible loop which acts as a flexible tether joining the receiver and effector domains. It is devoid of secondary structure (prediction data not shown) and is highly divergent among response regulators (131; 11). A second apparently divergent region was observed between residues 67 and 81 of AbsA2; however, closer inspection revealed that only a four amino acid stretch fi'om position 70 to 73 was highly variable. Comparison with crystolgraphic data fi'om NarL RR predicted this small region to encompass a loop between helix 01-3 and strand B-4 of the receiver domain (11; AbsA2 modeling data not shown). Although the alignment shows significant sequence similarity over most of the amplified region, receiver domains from different RRs of the same sub-family tend to be 131 P10 1 I 50 AbsA2 mirvllaDDE TIIRAGvrsI LttepgiEVV.AEAsdGreAv eLarkHRPDV Ambofaciens DDE TIIRAGaraI LsadpeiEVV.AEAstGreAv eLvrrHRPDV Griseus DDE TIIRAGvraI LardphvEVV AEAgdGheAi‘aLtraHRPDV Peucetius DDE TIIRAGvraI LsadtgiEVV AEAddGrqu eLaerRPDV Clavaligerus DDE TIIRAGvcaI LaaepgiEVV.AEAadGheAv eLterRPDV Consensus ------- ans: Tunas-"I r. ------ svv mus-41- -r.---rramv * 99 100 AbsA2 aLlDirMPem DGLtAAgemr ttnpdtavvv lTTFgeDrYi eRAquGvaG Ambofaciens aLleqMqu DGLdAAseil ksdagtavii fTTanDin aRALgeGasG Griseus vLmDirMPgl DGLtAAarlh resatvglim lTTngDeYv tRALeeGadG Peucetius aLlDirMPrl DGLaAAeelr raaprtavvm 1TTFseDeYv eRALgansG Clavaligerus leDvrMPrf DGLrAAeeiq rvapdtavvm lTTFseDeYi aRALdsGasG Consensus -L-D--MP-- DGL-AA -------------- -TTr--D-!- -RAL--G--G * P11 101 ‘__— 150 AbsA2 FLLKasDPRd LisGVnAvas GgscLSPlvA rletelrra pspRsevsge Ambofaciens FLLngDPRd LiaGVhAvad GaayLSPeaA thirglpta rmaRgsaare Griseus FLLKadDPRe LlnGV2Avga GgayLSPrvA ngiagm... rahRaahphr Peucetius FLLngDPRe LiaGVnAavq GaacLSPeiA erldrlggg rmsRaaeara Clavaligerus FLLKagDPRe LiaGVrAvad GaacLSPevA eriarlgdg rlsRawaarr Consensus ILLR--DPR- L--GVeA--- G---LSP--A -R ----------- R ------ * P8 151 4— 200 AbsA2 rttlLthEq eVlgmlgaGL SNAEIAQRLH lvegtiktyv saiftqlevr Ambofaciens rversaREr eVltllgeGL SNAEIAQRLH Griseus slarLteREr eVlaglgaGL SNAEIAQRLH Peucetius alethgREr eVValvaaGL SNAEIAQRLH Clavaligerus tlethrREr dVvalvadGL SNAEIAQRLH Consensus ----L--RE- -V------GL SNAEIAQRLH -------------------- \_J L_l \ J helix turn helix 201 223 AbsA2 nrvqaaiiay eaglvkdadl nr* Ambofaciens Griseus Peucetius Clavaligerus Consensus ----------------------- Figure 25. Amino acid sequence alignment of S. coelicolor AbsA2 with putative homologs from S. amabafaciens, S. griseus, S. peucetius, and S. clavuligerus. A consensus sequence of residues conserved in all five strains are in bold type. Primers P8 and P10 were used to amplify these genes. The orientation and location of amino acid condons recognized by primers P8, P9, P10, and P11 are indicated by arrows. Highly conserved residues within the acidic pocket are indicated C“). The predicted location of the helix-tum-helix DNA-binding region of AbsA2 is also identified. 132 the most highly conserved region of the RR. Therefore, regions of similarity were sought that might set the AbsA2 homolog apart from other RRs. Various residues were conserved among the putative AbsA2 homologs that were highly variable in Rs of the same sub-family as AbsA2. For example, amino acids H46 and R47 of AbsA2 were highly divergent not only among RRs from different genera, but also from an alignment of eleven randomly chosen S. coelicolor RRs of the same subfamily as AbsA2 (data not shown). These residues were predicted by homology modeling of AbsA2 to the NarL crystalography structureto represent the last amino acid of helix 01-2 and the first amino acid of the 01-2 to B-3 loop (1 l), which is one of the most highly solvent exposed regions of the receiver domain (176). It is tempting to speculate that these and possibly other uniquely conserved residues are important in cognate HK-RR recognition, or some other system-specific fisnction. A comparison of each homolog sequence with that of AbsA2 produced the following amino acid similarities and identities: S. griseus, 78% similarity, 60% identity; . S. amabafaciens, 78% similarity, 61% identity; S. clavuligerus, 67% similarity, 63% identity; and S. peucetius, 69% similarity, 63% identity. Excluding the highly divergent linker region, which constituted about 10% of the sequence, amino acid identities would have been around 70%. Brown, et al. (1996) did not report amino acid identities between B. subtilis SpoOA and its homologs retrieved by PCR. They did, however, mention that DNA sequence identity for spaOA homologs from different strains of C. acetaburylicum was as low as 66%. DNA sequence identity between putative absA2 homologs from different species tested here ranged fi'om 70 to 72%. Sequence identity between homologous proteins of Streptomyces has been shown to vary greatly. The sigma factor, 133 OF, required for sporulation in streptomycetes was found to be 87% identical between S. coelicolor and S. aureafaciens (134). In another study, five polyketide synthase (PKS) genes encoding the type II polyketide antibiotic frenolicin in S. raseafulvus, demonstrated approximately 40 to 70% amino acid identity with similar PKS proteins from four other species of Streptomyces (17). Finally, a BLAST search of the nearly complete Streptomyces genome sequence (httpz/lwww.sanger.ac.uk/Projects/S_coelicolor/) recognized 52 RS (in addition to absA 2) with full-length similarity to AbsA2. Amino acid identities for all but one RR (excluding AbsA2) ranged from 26 to 49%, well below the 60 to 63% identity obtained here from putative absA2 homologs fi'om other Species of Streptomyces. However, one RR found on cosmid St8D11 was 55% identical to AbsA2 and contained residues highly conserved in the putative homolog proteins. This locus will be discussed below. DISCUSSION Evidence exists for the evolutionary conservation of developmental regulation in Gram-positive bacteria. The well characterized SpoOA response regulator, which triggers sporulation in B. subtilis, was shown to be conserved in all species of highly diverse Bacillus and Clastridium tested (21). Using a strategy similar to that employed for SpoOA, degenerate PCR primers with varying specificities for absA2 successfully amplified nucleotide sequences of the same size as absA2 from ten industrial strains of Streptomyces. Southern hybridization analysis demonstrated that only those products that were the same size as absA2 annealed with absA2 probes. Yet, it was quite possible that R genes with full-length sequence similarity to absA 2, but different biological function, 134 were being primed and amplified. Therefore, the next step in confirming whether or not these were actual homologs of absA2 was to inspect their sequence. Sequence analysis of putative homologs fi'om S. griseus, S. ambafaciens, S. clawligerus, and S. peucerius revealed amino acid identities with absA2 of approximately 60%. By comparison, S. coelicolor and S. aureafaciens homologs of the highly conserved o'F sigma factor, required for transcription of late sporulation genes, shared 87% sequence identity (134). In contrast, homologous PKS genes, that synthesize different polyketide antibiotics through similar enzymatic reactions, have amino acid similarities that commonly range from just 40 to 60% (17). Moreover, only one S. , coelicolor gene encoding a R with fisll-length similarity to AbsA2 has yet demonstrated greater than 49% amino acid identity. Therefore, the fact that partial gene sequences from heterologous strains demonstrate at least 60% identity is suggestive evidence that these genes encode AbsA2 homologs. A blast search conducted on the nearly complete S. coelicolor genome identified a RR from cosmid St8D11 that is 55% identical to AbsA2. The St8Dll RR possesses the highly conserved residues of putative AbsA2 homologs that are missing from other RRs of the same sub-family. Also, the HTH region of St8Dll RR shows 54% identity and 83% similarity to that of AbsA2. Not only is the St8Dll RR highly similar to AbsA2, but the St8D11 I-IK is a remarkable 44% identical to full-length AbsAl. These HKs share significant similarity in the C-terminus of the sensor domain, which is normally highly divergent among HKs. Moreover, the St8D11 HK has a C-terminal domain that is approximately 40%.identical to that of AbsAl, making it the only protein from any database to demonstrate similarity to this region of AbsAl. 135 Streptomyces have linear chromosomes, the ends of which are genetically unstable, undergoing deletion and duplication events at a rather high frequency (reviewed by 175). As a likely consequence of this instability, no S. coelicolor housekeeping genes are found within at least several hundred kilobases of the chromosomal ends. Cosmid St8Dll lies very close to the end of the chromosome on restriction fragment AseI-A (140). According to Kieser et al. (1992), this region of the J1501 chromosome seems to have undergone several deletions and possibly duplication events in comparison to strain M145, which is the progenitor of J1501 and the strain currently being sequenced. Thus, it is uncertain whether the St8D11 two-component system is present in S. coelicolor 11501, the strain from which absA was isolated and the strain used in this study. Given its remarkable similarity to AbsA, it is possible that the St8D11 two-component system functions in antibiotic regulation, but is missing from J 1501. Studies are currently underway to determine whether the St8D11 two-component system is present in J1501 and whether it is a regulator of antibiotic synthesis or sporulation. Future amplification of absA homologs might take advantage of the unique C- terrninal domain of AbsAl. This domain is about 160 amino acids long, is of unknown function, and shows homology to only one other S. coelicolor HK. Assuming that this domain is conserved, it could serve as a criterion for amplifying absA homologs. One strategy would be to prime for regions of the RR receiver domain and the HK transmitter domain that are highly conserved and of specialized function in two-component systems. This should reduce the amplification of spurious products of varying size. At the same time, putative homologs with a C-terminal absA] domain would generate products 136 approximately 500 bp larger than those of two-component systems with 'orthodox' primary structure. The recently revealed association of absA within the cda gene cluster raises obvious questions about the possible conservation of absA throughout the genus Streptomyces. Can it be simply coincidental that a global regulator of antibiotics is located in an antibiotic gene cluster? If AbsA was originally associated only with cda regulation, how did it expand its range to regulate all S. coelicolor antibiotics? If AbsA was originally specific to cda regulation, in a manner similar to pathway-specific regulators, why don't more antibiotic gene clusters possess two-component systems? It is still not clear whether the AbsA two-component system is an idiosyncrasy of S. coelicolor or a highly conserved mechanism of antibiotic regulation throughout streptomycetes. Further analysis into the conservation of absA may help establish whether specific mechanisms of pleiotropic antibiotic regulation tied to multicellular development are highly conserved in Streptomyces or predominantly unique due to random evolutionary pressures of the organisms environment. 137 CHAPTER 7 CONCLUSIONS AND FUTURE RESEARCH 138 Prior to this study the AbsA two-component signal transduction system had been shown to globally regulate antibiotics of S. coelicolor. Interestingly, absA mutants were associated with dramatically opposing phenotypes. Mutations mapped to the absA! histidine kinase gene caused global inhibition of all four S. coelicolor antibiotics (Abs' phenotype), while gene disruptions of absA caused the early anset and overproduction of Act and Red (Pha phenotype) (19). Given the probable cotranscription of absA] and absA2, it was hypothesized that AbsA2 was a negative regulator of antibiotic production since gene disruptions in absA caused the Pha phenotype. This study sought to understand the basis for the dramatically opposed phenotypes obtained from absA mutants through further molecular genetic characterization of the absA locus and an examination of basic aspects of the AbsA two-component mechanism. Definition of AbsA2 as a positive or negative regulator was essential to more fully understanding any possible interactions with other antibiotic pathway-specific or pleiotropic regulators. Likewise, knowledge of the biochemical mechanism of signal transduction was considered prerequisite to defining more complex characteristics of the AbsA mechanism such as target binding and signal sensing. Molecular Genetic Characterization of absA] Mutations Responsible for the Ahs' Phenotype and Certain sab Suppressors of Abs' Mutations responsible for the Abs' phenotype had previously been localized to the absA] histidine kinase gene (2; 19). Sequence analysis of absA] from two independently isolate Abs' mutants, C542 and C577, identified point mutations that caused amino acid substitutions to the transmitter domain of AbsAl. The mutations to C542 (1360L and 139 R365Q) were contained within the G-box, which is proposed to play a critical role in phosphotransfer (158). The C577 mutation (L253R) was found in a region of moderate conservation among histidine kinases termed the X-box (79). X-box mutations in EnvZ (79), NtrB (9) and DegS (169) lock each of these HKs into a kinase dominant, phosphatase deficient state. Sequence characteristics of the Abs' mutations, taken together with results obtained from the genetic and transcript analyses of the absA locus (Chapter 3), lead to the hypothesis that mutations identified in C542 and C57 7 lock AbsAl into a kinase dominant, phosphatase deficient state which causes constitutive negative regulation of antibiotics. It is uncertain at this point whether the Abs’ mutants are signal-independent; however, the early accumulation of absA transcript together with the stability of the Abs’ phenotype over days and weeks suggests signal-independent behavior. Abs“ stains C577 and C542 undergo apparent pseudoreversion to attain wild type (Type I) or Pha (Type H) levels of antibiotic production. Genetic mapping of five sab (suppressor of glgs) mutants placed the mutations responsible for this phenotype very close to the absA locus (Appendix A). Marker exchange experiments demonstrated that restoration of antibiotic production was due to second-site suppressors of the Abs' mutations. The mapping data suggested that the sab mutations were located downstream of the absA! mutations. Therefore absA2 was sequenced in the five mapped sab strains plus six unmapped sab mutants. Two sab mutations were identified in absA2 while a third contained a deletion that included most of absA 1 and all of downstream absA2. The absA! gene was then sequenced in 2 of the 4 remaining mapped sab strains. Both of these were found to contain a sab mutation. Six remaining sab mutants that are wild type 140 for absA2, have not been sequenced for absA]. Second-site suppressors are a useful tool for finding additional members of a regulatory pathway. Identifying whether the remaining sab mutations lie in absA] would provide valuable insight into whether another locus is involved in the AbsA regulatory pathway. If all sab mutations were localized to absA, the likelyhood that AbsA functions through an intermediate regulator (e.g., that AbsA2 is an activator of a repressor) would be diminished. If a mutation was found to lie outside absA, then evidence would exist for the involvement of a second gene product. Four sab mutants that do not possess mutations in absA2 are Type II (Pha); three of these are unmapped. An initial experiment to test whether these mutants possess a sab mutation in absA] would be to transform them with pCBZOO (19) containing a truncated, fisnctional form of absA 1, or with pCBSZO (Chapter 3) carrying the entire absA locus. Ifa sab mutation causing the Pha phenotype lies outside of absA 1, the absA clones would not be expected to complement the mutation. Sab mutants not complemented by the absA clones could first be sequenced for absA] to assure that there is not a dominant Pha-producing mutant, and then studied through mapping and complementation to isolate the suppressor locus. The Role of Phosphorylation in the AbsA Regulatory Mechanism Disruptions in the absA] and absA2 genes demonstrated that the AbsA two- component system is a negative regulator of the multiple antibiotics produced by S. coelicolor, including calcium-dependent antibiotic, actinorhodin and undecylprodigiosin. While synthesis of Act and Red had previously been shown to be accelerated in absA disruptions (19), this is the first demonstration that absA mutants also cause early onset 141 and overproduction of CDA. The absA locus was recently found to lie within the cda gene cluster. Thus, AbsA is also the first antibiotic-gene-cluster-associated regulator that demonstrates regulatory activity outside of its cluster. Much of the attention of this study was focused on defining the role of phosphorylation in AbsA2-mediated negative regulation. Gene replacements in the absA locus altered the putative sites of phosphorylation of AbsAl or AbsA2. As predicted from sequence conservation with other two-component systems, both the His at position 202 of AbsAl and the Asp residue at position 54 of AbsA2 were required for normal regulation of antibiotic synthesis. Furthermore, each of the gene replacement strains tested attained an antibiotic overproducing phenotype (Pha) consistent with a mechanism in which AbsA2-P is the active negative regulator of antibiotic synthesis. In addition, high-copy expression of AbsA2 D54E was not able to complement the Pha phenotype of an absAA strain; moreover, it caused overexpression of antibiotics in a absA wild type background. Taken together, these results suggest that AbsA2 activity is strongly dependent upon phosphorylation. Transcription of the absA locus also demonstrated phosphorylation-dependent autoregulation of absA expression. Transcript studies of absA suggest either positive autoregulation by AbsA2-P or negative autoregulation by unphosphorylated AbsA2. Data fi'om this study coupled with precedent from other two-component systems favors a model for phospho-AbsA2-mediated positive autoregulation. First, the genetic data presented in Chapter 3 showed AbsA2-P to be the active regulator of antibiotics with a strong dependence on phosphorylation for activity. Likewise, the absence of growth- phase-related change in the absA transcript in C570 Pha strain (Chapter 3) suggests that 142 AbsA2-P was responsible for the growth-phase increase observed in J 1501 and C542. Finally, the majority of response regulators from other two-component systems are activated by phosphorylation (161). Therefore, I predict that AbsA2-P is not only the active negative regulator of antibiotic synthesis, but that it also positively autoregulates its own expression. An experiment that could resolve whether absA autoregulation is positive or negative would examine absA expression in absAA strain C577S25 (Appendix A). C577S25 possesses a deletion in all of absA2 and most of absA 1, excluding the promoter region and part of the 5' region of absA] encoding the sensor domain. IfabsA positively autoregulates its own expression, this strain would be expected to contain low levels of transcript over the course of growth, similar to C570. Conversely, if absA expression is negatively autoregulated, C577S25 absAA should produce elevated levels of transcript similar to that of C542. Model Summary A working model which accommodates both the genetic and transcriptional data summarized above is illustrated in Figure 26 and supposes the following. Early in growth, a'culture is not competent for antibiotic synthesis and the absA genes are expressed at a basal level. In the absence of signal, AbsAl is in a phosphatase dominant form. Following a period of growth, the culture enters the “transition stage.” During this time, the signal that regulates AbsA may be present at significant levels. Once the signal is present, AbsAl is shifted to a kinase-dominant form and AbsA2-P will accumulate. AbsA2-P negatively regulates antibiotics and also positively autoregulates absA expression, accounting for the AbsA2-P-dependent, growth-phase-related increase of 143 Signal Activates Absence of Signal . AbsAl Phosphatase AbSAl Kinase Dominant . . . . A1) A2 Inaztive 3’ n V” 9 9 Less Signal (stationary phase) 7 Relaxation of negative regulation EdaR _ abs cdar A ltedZ - -?redD red r ' Focal-4 act r Transcriptional regulation Figure 26. Model of AbsA-mediated regulation of antibiotic production in S. coelicolor. 144 absA transcript seen in 11501. A signal-concentration-dependent equilibrium is established between the kinase and phosphatase forms of AbsAl, such that the level of signal determines the extent to which antibiotics are repressed. Easing of AbsA- repression may require that the signal be depleted or degraded. As signal decreases, the AbsAl kinase/phosphatase equilibrium would shift toward the phosphatase form, causing dephosphorylation of AbsA2 and a decrease in the negative regulation of antibiotic gene expression. This model proposes the establishment of an AbsAl signal-dependent kinase/phosphatase equilibrium. Genetic evidence for AbsAl autokinase activity was demonstrated by loss of function phenotypes in gene replacements that targeted the transmitter domain or Hi5202 of AbsAl. Furthermore, purified MBP-'AbsAl (HZO2L) was not phosphorylated in vitro under the same conditions in which wild type MBP- ‘AbsAl was readily phosphorylated, suggesting that AbsAl is phosphorylated at this residue. Similar direct evidence for AbsAl phosphatase activity toward AbsA2-P has yet to be demonstrated. However, genetic evidence from two experiments can be interpreted to support phosphatase activity by AbsAl. First, overexpression of wild type or an H202L mutant of AbsA] in a wild type absA background caused a Pha phenotype (Chapter 3). These results are consistent with a decrease in the proportion of AbsA2-P, and thus an increase in the number of AbsAl molecules exercising phosphatase activity. Second, in a recent experiment, pTBA155 carrying absA1* (HZOZL) (Chapter 3) was transformed into C542. If AbsAl * H202L possesses phosphatase activity, I would expect some level of restoration of antibiotic production to strain C542, which is hypothesized to be locked in a kinase dominant, phosphatase deficient state. The 145 C542/pTBA155 strain attained wild type levels of antibiotic regulation, which suggests AbsAl H202L phosphatase activity. The assumption that AbsAl H202L can possesses phosphatase activity while being kinase deficient is consistent with phosphatase activity demonstrated by other HKs carrying mutations at the conserved histidine residue (80). A more direct analysis of AbsAl phosphatase activity would test purified 'AbsAl (Chapter 5) in-vitro phosphatase activity against purified AbSAZ‘HISG‘P. The principle focus for the immediate future of this project will be to refine and extend the model of AbsA-mediated regulation to include the identification of targets of AbsA2 regulation and the signal-sensing mechanism of AbsAl. A discussion of additional data generated in this study will be presented in the context of future work in these two areas. AbsA2 Targets A previous study demonstrated that mutations responsible for Abs' and Pha phenotypes correspondingly affected transcription of redD and actII-ORF 4 (1), thus establishing these pathway-specific activators as part of the AbsA regulatory pathway and possible targets of AbsA2. RedZ is a second pathway-specific regulator of the red biosynthetic pathway which appears to activate expression of redD (182). 1 found that mutations in absA had no affect on the expression of redZ, suggesting that it is not a target of AbsA2 regulation. Interestingly, RedZ is a "pseudo response regulator” which possesses full-length sequence similarity to AbsA2, but has lost the requirement for phosphorylation (65). RedZ possesses a HTH motif in its effector domain, making it possible that AbsA2 and RedZ compete for binding at the redD promoter, although 146 probably recognizing different sequences. Exclusion ofredZ from the AbsA regulatory pathway makes the redD promoter an attractive potential target of AbsA2-mediated regulation of Red synthesis. The effect of AbsA mutants was also tested on the expression of cdaR, a putative pathway-specific activator of the cda gene cluster (156). While redD and actII-4 were consistently repressed in Abs' mutants over the course of growth (1; Chapter 4), the expression of cdaR was equivalent in Abs', Abs+ and Pha cultures by the time these reached the stationary phase (Chapter 4). Only at the early 18 hr time point was there a difference in the levels of expression of cdaR; however, both the Abs‘ and Pha cultures had more abundant levels of cdaR transcript than J1501. The Abs' strain does not produce CDA, so if cdaR is an activator of CDA biosynthesis and a target of AbsA2, I would expect it to be repressed in the Abs‘ mutant. New evidence has demonstrated that Abs' mutants dramatically reduce expression of various cda biosynthetic operons over the course of grth (Ryding, unpublished), explaining the CDA‘ phenotype of Abs’ mutants. Inspection of the S. coelicolor genome sequence reveals potential additional regulatory genes adjacent to the cda cluster (146). Thus, AbsA may be acting at a target other than cdaR in the cda gene cluster. Nonetheless, CDA production is evident several hours before the appearance of Red, and cdaR expression was evident at the earliest time point in which absA was witnessed. Therefore, I can not yet exclude that AbsA may be regulating the early expression of cdaR. As such, cdaR expression will be examined at closely grouped time points between mid-exponential and stationary phase grth of Abs', Abs+, and Pha strains. Growth curves will be generated to assure that similar time points from each culture represent the same stage of development. Likewise, assays will 147 be conducted to determine the timing and abundance of CDA in the Abs+ and Pha cultures. At the same time, 81 analyses will be performed on other putative regulators of the cda gene cluster to determine their dependence on AbsA and possible involvement in the AbsA regulatory pathway. It is uncertain at this time whether the autoregulation of absA expression is direct or indirect. Nevertheless, an obvious target to begin testing for AbsA2 binding is the absA promoter region. The absA promoter has an imperfect heptameric inverted repeat, a motif frequently recognized by response regulators. The sequence and arrangement of the heptameric repeat sequence is loosely conserved in other possible targets of AbsA2 (e.g., redD and actII-4). Initial targets to be used in gel-shift AbsA2 DNA-binding assays will include the promoter regions of absA, redD, and actII-4. Successful target identification with gel-shift assays could be followed by DNaseI or hydroxy-radical footprint analyses to identify a possible AbsA2 consensus binding sequence, the identification of which could be valuable in the search for other targets of AbsA2, for example in the cda cluster. Gel-shifi assays will utilize purified AbsA2. Purification of AbsA2 and AbsA2 D54E was pursued for use in in-vitra phosphorylation experiments to support genetic evidence for the role of Asp54 in AbsA2 phosphorylation and activity. AbsA2-IEs-tag proteins purified from E. coli and S. lividans were not phosphorylated in vitro with enzymatically synthesized 32P-acetyl phosphate under the conditions tested. Genetic data strongly suggest that AbsA2-P is required for target recognition. Therefore, it may prove necessary to resolve the problems with overexpression and in-vitra phosphorylation of AbsA2 in order to perform in-vitra DNA-binding studies. The following strategy is 148 being pursued. AbsA2-His; overexpression will concentrate on the Streptomyces overexpression system since, i.) AbsA2-His; was shown to be functional in viva when expressed in S. coelicolor C570; ii.) a good yield of soluble phase AbsA2-His6 was produced in the soluble phase, thus avoiding denaturation and renaturation (i.e., excess handling); and iii.) preliminary evidence exists for the in-vitra phosphorylation of AbsA2-His; by purified phospho-‘AbsAl. A steady decrease in product yield was seen over time in the Streptomyces/Ni2+ expression/purification system. Suspected causes include plasmid instability in the S. lividans host and purification column degradation. I am currently examining the stability of the pTBA27O AbsA2 expression plasmid in strain C570 and establishing overexpression and purification conditions that will lead to consistent recovery of soluble phase AbsA2-Hig. Both AbsA2 DNA-binding studies and determination of AbsAl phosphatase activity would be facilitated by AbsA2-Hist; in-vitra phosphorylation by small molecular weight phosphate donors. Once reproducible recovery of AbsA2-His; has been achieved, in-vitra phosphorylation trials will be conducted using chemical synthesis of 32P-acetyl phosphate. This method (118) has been employed more frequently and for a greater number of response regulators than the enzymatic preparation of 32P-acetyl phosphate discussed in Chapter 5. Phosphorylation of AbsA2-Hi56 utilizing purified ‘AbsAl as a phosphate donor will be pursued simultaneously. Preliminary experiments with MBP- 'AbsA1 provided evidence that the Factor Xa-liberated phospho-‘AbsAl could serve as a phosphate donor for AbsA2-His; from S. lividans. Nonetheless, the MBP-'AbsAl fusion protein appears to undergo proteolysis in the E. coli JM109 host, leading to purification products of multiple sizes. Two alternatives are currently being pursued to eliminate 149 proteolysis of MBP-'AbsAl. First, the fusion protein will be expressed in E. coli BL21, a strain containing ampT and Ian protease knockouts. A second alternative is to overexpress only the transmitter domain of 'AbsAl versus the current truncation which includes both the transmitter domain and the C-terminal domain of unknown function. The size of the MBP-'AbsAl fusion products suggests that proteolysis is occurring in the C-terminal domain of AbsAl. Removal of the C-terminal domain should not affect the function of the transmitter domain since a 69 aa tmncation of the AbsAl C-terminus has previously been shown to complement Abs' mutants (19). The AbsA Signal and Signal-Sensing Mechanism Identification of the signal and signal-sensing mechanism has proven elusive for most two-component systems. Understanding the signal-sensing mechanism of AbsAl would help define whether AbsAl is kinase or phosphatase default in the absence of signal (i.e., whether the signal activates or relieves negative regulation of antibiotics). Likewise, identifying the signal recognized by AbsAl may provide insight into what external environmental or cell-generated factors are involved in the onset of antibiotic synthesis and how this is coordinated with other deve10pmentally regulated processes such as sporulation. Transcriptional analysis of absA expression revealed that it was growth-phase regulated, experiencing dramatic upregulation prior to the appearance of antibiotics in a wild type culture. The temporal profile of absA expression suggests that the signal recognized by AbsAl increases during transition phase growth. At this time, the signal that AbsAl senses is not known. Thus, a more immediate focus on elucidating the 150 AbsAl signal-sensing mechanism could serve to fisrther refine the AbsA mechanism and possibly provide clues into the type of signal recognized by AbsAl. Precedent from other two-component systems establishes two general trends for HK signal sensing mechanisms. The first class of HK proteins, exemplified by EnvZ (reviewed by 55) and NarX (25), have a large periplasmic loop with a proposed ligand- binding box lying between two transmembrane (TM) helices in the N-terrninal sensor domain. Ligand (signal) binding is proposed to cause movement of one transmembrane helix with respect to the other, resulting in modulation of enzymatic activity in the transmitter domain (38). Alternatively, HKs such as FixL sense the signal on a cytoplasmic domain of the HK situated between the transmembrane and transmitter domains (reviewed by (4)). Deletion of the membrane-bound sensor domain of NarX locks the HK into a signal-independent kinase dominant state (25). Conversely, deletion of the transmembrane domain of FixL does not alter its signal sensitivity (43). An initial approach to identifying the region of AbsAl involved in signal sensing could implement a series of in-frame deletions and amino acid substitutions. Unlike EnvZ and NarX, AbsAl is predicted to contain four TM helices with relatively small external loops. Deletion mutations might include the entire TM domain or the two central TM helices. If the signal sensing region is contained within the deletions, I would expect an Abs' phenotype consistent with constitutive kinase activity demonstrated in viva by truncations of HKs with the signal-sensing region contained within the transmembrane domain of the HK (170; 25). A recently identified HK from the S. coelicolor genome sequence of cosmid St8D11 is 40% identical to absA! and shows unusual similarity in the C-terminus of the normally highly divergent sensor domain. Striking within the alignment of this 151 region are 5 consecutive residues conserved in a predicted external solvent exposed loop between TM helix 3 and TM helix 4, which is also the largest of the helical loops. It is tempting to speculate that this highly conserved region represents a ligand-binding box within the sensor domain. Site-directed amino acid substitutions could provide useful in examining this region’s functional significance. This study has broadened our knowledge of the AbsA two-component system and allowed us to establish a working model which sets clear goals for future research. Whether the AbsA system is part of a genetically-programmed developmental cycle or an isolated component of an integrated network subject to external environmental stimuli will become more clear as the targets and signal-sensing mechanism of AbsA are elucidated. Finally, as approximately eighty novel two-component systems emerge from the Streptomyces Sequencing Project, knowledge of the structural features and mechanism of AbsA will surely provide insight into the function of other Streptomyces two-component systems. 152 REFRENCES 153 10. 11. Aceti, D. J., and W. C. Champness. 1998. Transcriptional regulation of Streptomyces coelicolor pathway-specific antibiotic regulators by the absA and absB loci. J Bacteriol. 180(12):3100-6. Adamidis, T., P. Riggle, and W. Champness. 1990. Mutations in a new Streptomyces coelicolor locus which globally black antibiotic biosynthesis but not sporulation. J Bacteriol. 172(6):2962-9. Agron, P. G., G. S. Ditta, and D. R. Helinski. 1993. Oxygen regulation of nifA transcription in vitro. Proc Natl Acad Sci U S A. 90(8):3506-10. Agron, P. G., and D. R. Helinski. 1995. Syrnbiotic expression of Rhizobium melilati nitrogen fixation genes is regulated by oxygen, p. 275-287. In J. A. Hoch and T. J. Silhavy (ed.), Two-Component Signal Transduction. ASM Press, Washington, DC. Aiba, H., F. Nakasai, S. Mizushima, and T. Mizuno. 1989. Evidence for the physiological importance of the phosphotransfer between the two regulatory components, EnvZ and OmpR, in osmoregulation in Escherichia coli. J Biol Chem. 264(24): 14090-4. Anderson, T., P. Brian, P. Riggle, R. Kong, and W. Champness. 1999. Genetic suppression analysis of non-antibiotic-producing mutants of the Streptomyces coelicolor absA locus [In Process Citation]. Microbiology. 145(Pt 9):2343-5 3. Angel], 8., E. Schwarz, and M. J. Bibb. 1992. The glucose kinase gene of Streptomyces coelicolor A3(2): its nucleotide sequence, transcriptional analysis and role in glucose repression. Mol Microbiol. 6(19):2833-44. Anisova, L. N ., I. N. Blinova, O. V. Efremenkova, P. Koz'min Iu, and V. V. Onoprienko. 1984. [Regulators of development in Streptomyces coelicolor A3(2)]. Izv Akad Nauk SSSR [Biol](1):98-108. Atkinson, M. R., and A. J. Ninfa. 1993. Mutational analysis of the bacterial signal-transducing protein kinase/phosphatase nitrogen regulator 11 (NRH or NtrB). J Bacteriol. l75(21):7016-23. Axelsson, L., and A. Holck. 1995. The genes involved in production of and immunity to sakacin A, a bacteriocin from Lactobacillus sake Lb706. J Bacteriol. l77(8):2125-37. Baikalov, 1., 1. Schroder, M. Kaczor-Grzeskowiak, 1c Grzeskowiak, R. P. Gunsalus, and R. E. Dickerson. 1996. Structure of the Escherichia coli response regulator NarL. Biochemistry. 35(34): 1 1053-61. 154 12. l3. 14. 15. l6. 17. 18. 19. 20. 21. 22. 23. Bascaran, V., L. Sanchez, C. Hardisson, and A. F. Brana. 1991. Stringent response and initiation of secondary metabolism in Streptomyces clavuligerus. J Gen Microbial. 137(Pt 7):]625-34. Bassler, B. L., and M. R. Silverman. 1995. Intercellular communication in marine Vibria species: density-dependent regulation of the expression of bioluminescence, p. 431-445. In J. A. Hoch and T. J. Silhavy (ed.), Two- Component Signal Transduction. ASM Press, Washington, DC. Bibb, M. 1996. 1995 Colworth Prize Lecture. The regulation of antibiotic production in Streptomyces coelicolor A3(2). Microbiology. 142(Pt 6):1335-44. Bibb, M. J. personal communication Bibb, M. J., P. R. Findlay, and M. W. Johnson. 1984. The relationship between base composition and codon usage in bacterial genes and its use for the simple and reliable identification of protein- coding sequences. Gene. 30(1-3):157-66. Bibb, M. J., D. H. Sherman, S. Omura, and D. A. Hopwood. 1994. Cloning, sequencing and deduced functions of a cluster of Streptomyces genes probably encoding biosynthesis of the polyketide antibiotic frenolicin. Gene. 142(1):31-9. Bourret, R. B., J. F. Hess, and M. 1. Simon. 1990. Conserved aspartate residues and phosphorylation in signal transduction by the chemotaxis protein CheY. Proc Natl Acad Sci U S A. 87(1):41-5. Brian, P., P. J. Riggle, R. A. Santos, and W. C. Champness. 1996. Global negative regulation of Streptomyces coelicolor antibiotic synthesis mediated by an absA-encoded putative signal transduction system. J Bacteriol. 178(1 1):3221- 31. Brissette, R. E., K. L. Tsung, and M. Inouye. 1991. Suppression of a mutation in OmpR at the putative phosphorylation center by a mutant EnvZ protein in Escherichia coli. J Bacteriol. l73(2):601-8. Brown, D. P., L. Ganova-Raeva, B. D. Green, S. R. Wilkinson, M. Young, and P. Youngman. 1994. Characterization of spoOA homologues in diverse Bacillus and Clostridium species identifies a probable DNA-binding domain. Mal IVIicrobiol. l4(3):411-26. Burbulys, D., K. A. Trach, and J. A. Hoch. 1991. Initiation of sporulation in B. subtilis is controlled by a multicomponent phosphorelay. Cell. 64(3):545-52. Buttner, M. J., K. F. Chater, and M. J. Bibb. 1990. Cloning, disruption, and transcriptional analysis of three RNA polymerase sigma factor genes of Streptomyces coelicolor A3(2). J Bacteriol. 172(6):3367-78. 155 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. Cashel, M., D. R. Gentry, V. J. Hernandez, and D. Vinella. 1996. The Stringent Response, p. 1458-1496. In F. C. Neidhardt (ed.), Escherichia coli and Salmonella Cellular and Molecular Biology, Second ed, vol. I. ASM Press, Wahington, D. C. Cavicchioli, R., R. C. Chiang, L. V. Kalman, and R. P. Gunsalus. 1996. Role of the periplasmic domain of the Escherichia coli NarX sensar- transmitter protein in nitrate-dependent signal transduction and gene regulation. Mol Microbiol. 21(5):901-11. Chakraburtty, R., and M. Bibb. 1997. The ppGpp synthetase gene (relA) of Streptomyces coelicolor A3(2) plays a conditional role in antibiotic production and morphological differentiation. J Bacteriol. l79(18):5854-61. Chakraburtty, R., J. White, E. Takano, and M. Bibb. 1996. Cloning, characterization and disruption of a (p)ppGpp synthetase gene (relA) of Streptomyces coelicolor A3(2). Mol Microbiol. l9(2):357—368. Champness, W., P. Riggle, T. Adamidis, and P. Vandervere. 1992. Identification of Streptomyces coelicolor genes involved in regulation of antibiotic synthesis. Gene. 115(1-2):55-60. Champness, W. C. personal communication. Champness, W. C. 1999. Cloning and analysis of regulatory genes involved in streptomycete secondary metabolite biosynthesis, p. 725-739. In A. L. Demain and J. E. Davies (ed.), Industrial Microbiology and Biotechnology, Second ed. ASM Press, Washington, D. C. Champness, W. C., and K. F. Chater. 1994. Regulation of antibiotic production and morphological differentiation in Streptomyces spp., p. 61-94. In P. Piggot, C. P. Moran, and P. Youngman (ed.), Regulation of Bacterial Differentiation. American Society for Microbiology, Washington, DC. Chang, H. M., M. Y. Chen, Y. T. Shieh, M. J. Bibb, and C. W. Chen. 1996. The cutRS signal transduction system of Streptomyces lividans represses the biosynthesis of the polyketide antibiotic actinorhodin. Mol Microbiol. 21(5): 1075-85. ' Chater, K. F. 1972. A morphological and genetic mapping study of white colony mutants of Streptomyces coelicolor. J Gen Microbial. 72(1):9-28. Chater, K. F. 1998. Taking a genetic scalpel to the Streptomyces colony. Microbiology. 144: 1465-1478. 156 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. Chater, K. F., and M. J. Bibb. 1997. Regulation of Bacterial Antibiotic Production, p. 57—105. In H. Kleinkauf and H. van Dohren (ed.), Products of Secondary Metabolism, vol. 6. VCH, Weinheirn, Germany. Chater, K. F., and C. J. Bruton. 1985. Resistance, regulatory and production genes for the antibiotic methylenomycin are clustered. Embo J. 4(7): 1 893-7. Chater, K. F., and R. Losick. 1997. The mycelial life-style of Streptomyces coelicolor A3(2) and its relatives, p. 149-182. In J. H. Shapiro and M. Dworkin (ed.), Bacteria as Multicellular Organisms. Oxford University Press, New York. Chervitz, S. A., and J. J. Falke. 1996. Molecular mechanism of transmembrane signaling by the aspartate receptor: a model. Proc Natl Acad Sci U S A. 93(6):2545-50. Chang, P. P., S. M. Podmore, H. M. Kieser, M. Redenbach, K. Turgay, M. Marahiel, D. A. Hopwood, and C. P. Smith. 1998. Physical identification of a chromosomal locus encoding biosynthetic genes for the lipopeptide calcium- dependent antibiotic (CDA) of Streptomyces coelicolor A3(2). Microbiology. 144(Pt 1):193-9. ' Cutting, S., and J. Mandelstam. 1986. The nucleotide sequence and the transcription during sporulation of the gerE gene of Bacillus subtilis. J Gen Microbiol. 132(Pt 11):3013-24. Dahl, J. L., B. Y. Wei, and R. J. Kadner. 1997. Protein phosphorylation afi‘ects binding of the Escherichia coli transcription activator UhpA to the-uhpT promoter. J Biol Chem. 272(3):1910-9. Dahl, M. K., T. Msadek, F. Kunst, and G. Rapoport. 1991. Mutational analysis of the Bacillus subtilis DegU regulator and its phosphorylation by the DegS protein kinase. J Bacteriol. l73(8):2539-47. de Philip, P., E. Soupene, J. Batut, and P. Boistard. 1992. Modular structure of the F ixL protein of Rhizobium melilati. Mol. Gen. Genet. 235:49-54. Delgado, J ., S. Forst, S. Harlocker, and M. Inouye. 1993. Identification of a phosphorylation site and fimctional analysis of conserved aspartic acid residues of OmpR, a transcriptional activator for ompF and ompC in Escherichia coli. Mol ‘ Microbial. 10(5):1037-47. Demain, A. L. 1989. Carbon Source Regulation of Idiolite Biosynthesis in Actinomycetes, p. 127-134. In S. Shapiro (ed.), Regulation of Secondary Metabolites in Actinomycetes. CRC Press, Boca Raton. 157 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57. Demain, A. L., and A. Fang. 1995. Emerging concepts of secondary metabolism in actinomycetes. Actinomycetologica. 9:98-117. Diep, D. B., L. S. Havarstein, and I. F. Nes. 1996. Characterization of the locus responsible for the bacteriocin production in Lactobacillus plantarum C11. J Bacteriol. 178:4472-4483. Elliot, M., F. Damji, R. Passantino, K. Chater, and B. Leskiw. 1998. The bldD gene of Streptomyces coelicolor A3(2): a regulatory gene involved in morphogenesis and antibiotic production. J Bacteriol. 180(6): 1549-55. Ensign, J. C. 1988. Physiological Regulation of Sporulation of Streptomyces griseus, p. 309-315. In Y. Okami, T. Beppu, and H. Ogawara (ed.), Biology of Actinomycetes '88. Japan Scientific Societies Press, Tokyo. Feitelson, J. S., F. Malpartida, and D. A. Hopwood. 1985. Genetic and biochemical characterization of the red gene cluster of Streptomyces coelicolor A3(2). J. Gen. Microbiol. 131:2431-2441. Feng, J., M. R. Atkinson, W. McCleary, J. B. Stock, B. L. Wanner, and A. J. N infa. 1992. Role of phosphorylated metabolic intermediates in the regulation of glutamine synthetase synthesis in Escherichia coli. J Bacterial. 174(19):6061-70. Fernandez-Moreno, M. A., J. L. Caballero, D. A. Hopwood, and F. Malpartida. 1991. The act cluster contains regulatory and antibiotic export genes, direct targets for translational control by the bldA tRNA gene of Streptomyces. Cell. 66(4):769-80. Fernandez-Moreno, M. A., A. J. Martin-Triana, E. Martinez, J. Niemi, H. M. Kieser, D. A. Hopwood, and F. Malpartida. 1992. abaA, a new pleiotropic regulatory locus for antibiotic production in Streptomyces coelicolor. J Bacteriol. 174(9):295 8-67. Floriano, B., and M. Bibb. 1996. ast is a pleiotropic but conditionally required regulatory gene for antibiotic production in Streptomyces coelicolor A3(2). Mol Microbiol. 21(2):385-96. Forst, S. A., and D. L. Roberts. 1994. Signal transduction by the EnvZ-OmpR phosphotransfer system in bacteria. Res Microbial. 145(5-6):363-73. Freeman, J. A., and B. L. Bassler. 1999. A genetic analysis of the function of LuxO, a two-component response regulator involved in quorum sensing in Vibria harveyi. Mol Microbiol. 31(2):665-77. Fujii, T., H. C. Gramajo, E. Takano, and M. J. Bibb. 1996. redD and actII- ORF4, pathway-specific regulatory genes for antibiotic production in 158 58. 59. 60. 61. 62. 63. 64. 65. 66. 67. 68. Streptomyces coelicolor A3(2), are transcribed in vitro by an RNA polymerase holoenzyme containing sigma hrdD. J Bacteriol. 178(11):3402-5. Fuqua, W. C., S. C. Winans, and E. P. Greenberg. 1994. Quorum sensing in bacteria: the LuxR-LuxI family of cell density- responsive transcriptional regulators. J Bacteriol. l76(2):269-75. Furuya, K., and C. R. Hutchinson. 1996. The DnrN protein of Streptomyces peucetius, a pseudo-response regulator, is a DNA-binding protein involved in the regulation of daunorubicin biosynthesis. J Bacteriol. l78(21):6310-8. Grafe, U., G. Reinhardt, W. Schade, I. Erritt, W. F. Fleck, and L. Radics. 1983. Interspecific inducers of cytodifferentiation and anthracycline biosynthesis from Streptomyces bikiniensis and S. cyaneafuscatus. Biotechnol Lett. 5:591-596. Grafe, U., W. Schade, I. Eritt, W. F. Fleck, and L. Radics. 1982. A new inducer of anthracycline biosynthesis from Streptomyces viridochromogenes. J Antibiot (Tokyo). 35(12):l722-3. ' Gramajo, H. C., E. Takano, and M. J. Bibb. 1993. Stationary-phase production of the antibiotic actinorhodin in Streptomyces coelicolor A3(2) is transcriptionally regulated. Mol Microbiol. 7(6):837-45. Grebe, T. W., and J. B. Stock. 1999. The histidine protein kinase superfamily. Advances in Microbial Physiology. 41:139-227. Grimsley, J. K., R. B. Tjalkens, M. A. Strauch, T. H. Bird, G. B. Spiegelman, Z. Hostomsky, J. M. Whiteley, and J. A. Hoch. 1994. Subunit composition and domain structure of the SpoOA sporulation transcription factor of Bacillus subtilis. J Biol Chem. 269(24):]6977-82. Guthrie, E. P., C. S. Flaxman, J. White, D. A. Hodgson, M. J. Bibb, and K. F. Chater. 1998. A response-regulator-like activator of antibiotic synthesis fiom Streptomyces coelicolor A3(2) with an amino-terminal domain that lacks a phosphorylation pocket [published erratum appears in Microbiology 1998 Jul;l44(Pt 7):2007]. Microbiology. 144(Pt 3):727-38. Hara, 0., S. Horinouchi, T. Uozumi, and T. Beppu. 1983. Genetic analysis of A-factor synthesis in Streptomyces coelicolor A3(2) and Streptomyces griseus. J Gen Microbiol. 129(Pt 9):2939—44. Harrison, S. C., and A. K. Aggarwal. 1990. DNA recognition by proteins with the helix-turn-helix motif. Annu Rev Biochem. 59:933-69. Hoch, J. A. 1995. Control of Cellular Development in Sporulating Bacteria by the Phosphorelay Two-Component Signal Transduction System, p. 129-144. In J. A. 159 69. 70. 71. 72. 73. 74. 75. 76. 77. 78. 79. Hoch and T. J. Silhavy (ed.), Two-Component Signal Transduction. ASM Press, Washington, D. C. Hoch, J. A., and T. J. Silhavy (ed.). 1995. Two-Component Signal Transduction. ASM Press, Washington, DC. Holman, T. R., Z. Wu, B. L. Wanner, and C. T. Walsh. 1994. Identification of the DNA-binding site for the phosphorylated VanR protein required for vancomycin resistance in Enterococcus faecium. Biochemistry. 33(15):4625-31. Hang, S. K., M. Kito, T. Beppu, and S. Horinouchi. 1991. Phosphorylation of the Ast product, a global regulatory protein for secondary-metabolite formation in Streptomyces coelicolor A3(2). J Bacteriol. l73(7):23l 1-8. Hopwood, D. A. 1997. Genetic Contributions to Understanding Polyketide Synthases. Chem. Rev. 97 :2465-2497. Hopwood, D. A., M. J. Bibb, K. F. Chater, T. Kieser, C. J. Bruton, H. M. Kieser, D. J. Lydiate, C. P. Smith, J. M. Ward, and H. Schrempf. 1985. Genetic Manipulation of Streptomyces; a Laboratory Manual. The John Innes Foundation, Norwich, United Kingdom. Hopwood, D. A., K. F. Chater, and M. J. Bibb. 1995. Genetics of antibiotic production in Streptomyces coelicolor A3(2), a model streptomycete. Biotechnology. 28:65-102. Horinouchi, S. 1996. Streptomyces genes involved in aerial mycelium formation. FEMS Microbiol Lett. 141:1-9. Horinouchi, S., O. Hara, and T. Beppu. 1983. Cloning of a pleiotropic gene that positively controls biosynthesis of A-factor, actinorhodin, and prodigiasin in Streptomyces coelicolor A3(2) and Streptomyces lividans. J Bacteriol. 155(3):1238-48. Horinouchi, S., M. Kito, M. Nishiyama, K. Furuya, S. K. Hang, K. Miyake, and T. Beppu. 1990. Primary structure of Ast, a global regulatory protein for secondary metabolite formation in Streptomyces coelicolor A3(2). Gene. 95(1):49-56. Horinouchi, S., F. Malpartida, D. A. Hopwood, and T. Beppu. 1989. afsB stimulates transcription of the actinorhodin biosynthetic pathway in Streptomyces coelicolor A3(2) and Streptomyces lividans. Mol Gen Genet. 215(2):355-7. Hsing, W., F. D. Russo, K. K. Bernd, and T. J. Silhavy. 1998. Mutations that alter the kinase and phosphatase activities of the two-component sensor EnvZ. J Bacteriol. 180(17):4538-46. 160 80. 81. 82. 83. 84. 85. 86. 87. 88. 89. 90. Hsing, W., and T. J. Silhavy. 1997. Function of conserved histidine-243 in phosphatase activity of EnvZ, the sensor for porin osmoregulation in Escherichia coli. J Bacteriol. 179(1 1):3729—35. Huang, K. J., C. Y. Lan, and M. M. Igo. 1997. Phosphorylation stimulates the cooperative DNA-binding properties of the transcription factor OmpR. Proc Natl Acad Sci U S A. 94(7):2828-32. Hutchinson, C. R. 1997. Antibiotics from Genetically Engineered Microorganisms, p. 683-702. In W. R. Strohl (ed.), Biotechnology of Antibiotics, Second ed. Marcel Dekker, Inc., New York. Ishige, K., S. N agasawa, S. Tokishita, and T. Mizuno. 1994. A novel device of bacterial signal transducers. Embo J. 13(21):5195-202. Ishizuka, H., S. Horinouchi, H. M. Kieser, D. A. Hopwood, and T. Beppu. 1992. A putative two-component regulatory system involved in secondary metabolism in Streptomyces spp. J Bacteriol. 174(23):7585-94. James, P. D. A., C. Edward, and M. Dawson. 1991. The effects of temperature, pH, and growth rate on secondary metabolism in Streptomyces thermavialaceus grown in a chemostat. J. Gen. Microbiol. 133:1715-1720. Janssen, G. R., and M. J. Bibb. 1993. Derivatives of pUCl8 that have BglII sites flanking a modified multiple cloning site and that retain the ability to identify recombinant clones by visual screening of Escherichia coli colonies. Gene. 124(1):l33-4. Jin, S., T. Roitsch, R. G. Ankenbauer, M. P. Gordon, and E. W. N ester. 1990a. The VirA protein of Agrobacterium tumefaciens is autophosphorylated and is essential for vir gene regulation. J Bacteriol. l72(2):525-30. Jin, S. G., R. K. Prusti, T. Roitsch, R. G. Ankenbauer, and E. W. N ester. 1990b. Phosphorylation of the VirG protein of Agrobacterium tumefaciens by the autophosphorylated VirA protein: essential role in biological activity of VirG. J Bacteriol. l72(9):4945-50. Kadner, R. J., M. D. Island, J. L. Dahl, and C. A. Webber. 1994. A transmembrane signalling complex controls transcription of the Uhp sugar phosphate transport system. Res Microbial. l45(5-6):381-7. ~ Kahn, D., and G. Ditta. 1991. Modular structure of FixJ: homology of the transcriptional activator domain with the -35 binding domain of sigma factors. Mol Microbiol. 5(4):987-97. 161 90a. 91. 92. 93. 94. 95. 96. 97. 98. 99. 100. 101. Karandikar, A., Sharples, G. P., and G. Hobbs. 1997. Differentiation of Streptomyces coelicolor A3(2) under nitrate-limited conditions. Microbial. 143: 3581-3590. Kendrick, K. E., and J. C. Ensign. 1983. Sporulation of Streptomyces griseus in submerged culture. J Bacteriol. 155(1):357-66. Khokhlov, A. S., I. I. Tovarova, L. N. Borisova, S. A. Pliner, L. A. Schevchenko, E. Y. Kornitskaya, N. S. Ivkina, and I. A. Rapoport. 1967. A- factor responsible for the biosynthesis of streptomycin by a mutant strain of Actinomyces streptomycini. Dokl. Akad. Nauk SSSR. 177:232-235. Kieser, H. M., T. Kieser, and D. A. Hopwood. 1992. A combined genetic and physical map of the Streptomyces coelicolor A3(2) chromosome. J Bacterial. l74( 17):5496—507. Kinoshita, H., H. Ipposhi,-S. Okamoto, H. Nakano, T. Nihira, and Y. Yamada. 1997. Butyrolactone autoregulator receptor protein (BarA) as a transcriptional regulator in Streptomyces virginiae. J Bacteriol. 179(22):6986-93. Kirby, R., and D. A. Hopwood. 1977. Genetic determination of methylenomycin synthesis by the SCPl plasmid Streptomyces coelicolor A3(2). J. Gen. Microbiol. 98:239-252. Kleerebezem, M., L. E. Quadri, O. P. Kuipers, and W. M. de Vos. 1997. Quorum sensing by peptide pheromones and two-component signal- transduction systems in Gram-positive bacteria. Mol Microbiol. 24(5):895-904. Kleman-Leyer, K., D. W. Armbruster, and C. J. Daniels. 1997. Properties of H. volcanii tRNA intron endonuclease reveal a relationship between the archaeal and eucaryal tRNA intron processing systems. Cell. 89(6):839-47. Kuipers, O. P., M. M. Beerthuyzen, P. G. de Ruyter, E. J. Luesink, and W. M. de V08. 1995. Autoregulation of nisin biosynthesis in Lactococcus lactis by signal transduction. J Biol Chem. 270(45):27299-304. Kwak, J., L. A. McCue, and K. E. Kendrick. 1996. Identification of bldA mutants of Streptomyces griseus. Gene. 17 1(1):75-8. Lakey, J. H., E. J. Lea, B. A. Rudd, H. M. Wright, and D. A. Hopwood. 1983. A new channel-forming antibiotic from Streptomyces coelicolor A3(2) which requires calcium for its activity. J Gen Microbiol. 129(Pt 12):3565-73. Lawlor, E. J., H. A. Baylis, and K. F. Chater. 1987. Pleiotropic morphological and antibiotic deficiencies result from mutations in a gene encoding a tRNA-like product in Streptomyces coelicolor A3(2). Genes Dev. 1(10): 1305-10. 162 102. 103. 104. 105. 106. 107. 108. 109. 110. 111. Lee, A. I., A. Delgado, and R. P. Gunsalus. 1999. Signal-dependent phosphorylation of the membrane-bound NarX two- component sensor- transmitter protein of Escherichia coli: nitrate elicits a superior anion ligand response compared to nitrite. J Bacteriol. l8l(17):5309-16. Lee, S. H., and K. J. Lee. 1993. Aspartate aminotransferase and tylosin biosyntheis in Streptomycesfradiae. Appl. Environ. Microbiol. 59:822-827. Leskiw, B. K., E. J. Lawlor, J. M. Fernandez-Abalos, and K. F. Chater. 1991. TTA codons in some genes prevent their expression in a class of developmental, antibiotic-negative, Streptomyces mutants. Proc Natl Acad Sci U S A. 88(6):2461-5. Li, J., S. Kustu, and V. Stewart. 1994. In vitro interaction of nitrate-responsive protein NarL with DNA target sequences in the fdnG, narG, narK and frdA operon control regions of Escherichia coli K-12. J. Mol. Biol. 241: 150-165. Lilley, G., A. E. Clark, and G. C. Lawrence. 1981. Control of the production of cephamycin C and thienamycin by Streptomyces cattleya NRRL 8057. J. Chem. Technol. Biotechnol. 31:127. Lukat, G. S., W. R. McCleary, A. M. Stock, and J. B. Stock. 1992. Phosphorylation of bacterial response regulator proteins by low molecular weight phospho-donors. Proc Natl Acad Sci U S A. 89(2):718-22. Lynch, A. S., and E. C. Lin. 1996. Transcriptional control mediated by the ArcA two-component response regulator protein of Escherichia coli: characterization of DNA binding at target promoters. J Bacteriol. l78(21):6238-49. Malpartida, F., and D. A. Hopwood. 1984. Molecular cloning of the whole biosynthetic pathway of a Streptomyces antibiotic and its expression in a heterologous host. Nature. 309(5967):462-4. Malpartida, F., J. Niemi, R. N avarrete, and D. A. Hopwood. 1990. Cloning and expression in a heterologous host of the complete set of genes for biosynthesis of the Streptomyces coelicolor antibiotic undecylprodigiosin. Gene. 93(1):91-9. Martin, J. F. 1989. Molecular Mechanisms for the Control by Phosphate of the Biosynthesis of Antibiotics and Other Secondary Metabolites, p. 213-237. In S. Shapiro (ed.), Regulation of Secondary Metabolism in Actinomycetes. CRC Press, Boca Raton. 163 112. 113. 114. 115. 116. 117.. 118. 119. 120. 121. 122. Martin, J. F., and L. E. McDaniel. 1975. Specific inhibition of candicidin biosynthesis by the lipogenic inhibitor cerulenin. Biochirn BiOphys Acta. 411(2):186-94. Martinez-Costa, O. H., P. Arias, N. M. Romero, V. Parro, R. P. Mellado, and F. Malpartida. 1996. A relA/spoT homologous gene from Streptomyces coelicolor A3(2) controls antibiotic biosynthetic genes. J Biol Chem. 271(18):]0627-34. Martinez-Costa, O. H., M. A. Fernandez-Moreno, and F. Malpartida. 1998. The relA/spoT-homologous gene in Streptomyces coelicolor encodes both ribosome-dependent (p)ppGpp-synthesizing and -degrading activities. J Bacteriol. l80(16):4123-32. Matsumoto, A., S. K. Hang, H. Ishizuka, S. Horinouchi, and T. Beppu. 1994. Phosphorylation of the Ast protein involved in secondary metabolism in Streptomyces species by a eukaryotic-type protein kinase. Gene. 146(1):47-56. Matsumoto, A., H. Ishizuka, T. Beppu, and S. Horinouchi. 1995. Involvement of a small ORF downstream of the ast gene in the regulation of secondqry metabolism in Streptomyces coelicolor A3(2). Actinomycetologica. 9:37-43. McCleary, W. R., and J. B. Stock. 1994. Acetyl phosphate and the activation of two-component response regulators. J Biol Chem. 269(50):31567-72. McCleary, W. R., J. B. Stock, and A. J. Ninfa. 1993. Is acetyl phosphate a global signal in Escherichia coli“? J. Bacteriol. 175:2793-2798. Meurer, G., and C. R. Hutchinson. 1999. Genes for the Biosynthesis of Microbial Secondary Metabolites, p. 740-758. In A. L. Demain and J. E. Davies (ed.), Manual of Industrial Microbiology and Biotechnology, Second ed. ASM Press, Washington, D. C. ' Moore, J. B., S. P. Shiau, and L. J. Reitzer. 1993. Alterations of highly conserved residues in the regulatory domain of nitrogen regulator I (NtrC) of Escherichia coli. J Bacteriol. l75(9):2692-701. Msadek, T., F. Kunst, and G. Rapoport. 1995. A signal transduction network in Bacillus subtilis includes the DegS/DegU and ComP/ComA two-component systems, p. 447-471. In J. A. Hoch and T. J. Silhavy (ed.), Two-Component Signal Transduction. ASM Press, Washington, DC. N arva, K. E., and J. S. Feitelson. 1990. Nucleotide sequence and transcriptional analysis of the redD locus of Streptomyces coelicolor A3(2). J Bacteriol. l72(1):326-33. 164 123. 124. 125. 126. 127. 128. 129. 130. 131. 132. 133. 134. N es, 1. F., D. B. Diep, L. S. Havarstein, M. B. Brurberg, V. Eijsink, and H. Halo. 1996. Biosynthesis of bacteriacins in lactic acid bacteria. Antonie Van Leeuwenhoek. 70(2-4):1 13-28. N eumann, T., W. Piepersberg, and J. Distler. 1996. Decision phase regulation of streptomycin production in Streptomyces griseus. Microbiology. 142:1953- 1963. Ninfa, E. G., M. R. Atkinson, E. S. Kamberov, and A. J. Ninfa. 1993. Mechanism of autophosphorylation of Escherichia coli nitrogen regulator H (NRH or NtrB): trans-phosphorylation between subunits. J Bacteriol. l75(21):7024-32. N odwell, J. R., and R. Losick. 1998. Purification of an extracellular signaling molecule involved in production of aerial mycelium by Streptomyces coelicolor. J Bacteriol. 180(5): 1334-7. Nodwell, J. R., K. McGovern, and R. Losick. 1996. An oligopeptide permease responsible for the import of an extracellular signal governing aerial mycelium formation in Streptomyces coelicolor. Mol Microbiol. 22(5):881-93. Ochi, K. 1990a. A relaxed (rel) mutant of Streptomyces coelicolor A3(2) with a missing ribosomal protein lacks the ability to accumulate ppGpp, A-factor and prodigiosin. J Gen Microbiol. 136(Pt 12):2405-12. Ochi, K. 1990b. Streptomyces relC mutants with an altered ribosomal protein ST- L11 and genetic analysis of a Streptomyces griseus relC mutant. J Bacteriol. 172(7):4008-16. Ohnishi, Y., S. Kameyama, H. Onaka, and S. Horinouchi. 1999. The A-factor regulatory cascade leading to streptomycin biosynthesis in Streptomyces griseus : identification of a target gene of the A-factor receptor. Mol Microbiol. 34(1):]02- 11. Parkinson, J. S., and E. C. Kafoid. 1992. Communication modules in bacterial signaling proteins. Annu Rev Genet. 26:71-112. Pogliano, J., A. S. Lynch, D. Belin, E. C. Lin, and J. Beckwith. 1997. Regulation of Escherichia coli cell envelope proteins involved in protein folding and degradation by the Cpx two-component system. Genes Dev. ll(9):1169-82. Pospiech, A., and B. Neumann. 1995. A versatile quick-prep of genomic DNA from gram-positive bacteria. Trends Genet. ll(6):217-8. Potuckova, L., G. H. Kelemen, K. C. Findlay, M. A. Lonetto, M. J. Buttner, and J. Kormanec. 1995. A new RNA polymerase sigma factor, sigma F, is 165 135. 136. 137. 138. 139. 140. 141. 142.. 143. 144. 145. required for the late stages of morphological differentiation in Streptomyces spp. Mol Microbiol. l7(1):37-48. Pratt, L. A., and T. J. Silhavy. 1995. Porin regulation of Escherichia coli, p. 105-127. In J. A. Hoch and T. J. Silhavy (ed.), Two-Component Signal Transduction. ASM Press, Washington, DC. Price, B., T. Adamidis, R. Kong, and W. Champness. 1999. A Streptomyces coelicolor antibiotic regulatory gene, absB, encodes an RNase III homolog. J Bacteriol. 181(19):6142-51. Quon, K. C., G. T. Marczynski, and L. Shapiro. 1996. Cell cycle control by an essential bacterial two-component signal transduction protein. Cell. 84(1):83-93. Raivio, T. L., D. L. Popkin, and T. J. Silhavy. 1999. The Cpx envelope stress response is controlled by amplification and feedback inhibition. J Bacteriol. 181(17):5263-72. Raivio, T. L., and T. J. Silhavy. 1997. Transduction of envelope stress in Escherichia coli by the Cpx two- component system. J Bacteriol. 179(24):7724- 33. Redenbach, M., H. M. Kieser, D. Denapaite, A. Eichner, J. Cullum, H. Kinashi, and D. A. Hopwood. 1996. A set of ordered casmids and a detailed genetic and physical map for the 8 Mb Streptomyces coelicolor A3(2) chromosome. Mol Microbiol. 21(1):77-96. Reyrat, J. M., M. David, C. Blonski, P. Boistard, and J. Batut. 1993. Oxygen- regulated in vitro transcription of Rhizobium meliloti nifA and fixK genes. J Bacteriol. l75(21):6867-72. Romero, N. M., and R. P. Mellado. 1995. Activation of the actinorhodin biosynthetic pathway in Streptomyces lividans. FEMS Microbiol Lett. 127(1- 2):79-84. Romero, N. M., V. Parro, F. Malpartida, and R. P. Mellado. 1992. Heterologous activation of the actinorhodin biosynthetic pathway in Streptomyces lividans. Nucleic Acids Res. 20(11):2767-72. Rudd, B. A., and D. A. Hopwood. 1980. A pigmented mycelial antibiotic in Streptomyces coelicolor: control by a chromosomal gene cluster. J Gen Microbiol. 119(Pt 2):333-40. Russo, F. D., J. M. Slauch, and T. J. Silhavy. 1993. Mutations that affect separate frmctions of OmpR the phosphorylated regulator of porin transcription in Escherichia coli. J Mol Biol. 231(2):261-73. 166 146. 147. 148. 149. 150. 151. 152. 153. 154. 155. 156. 157. Ryding, N. J. personal communication. Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecular Cloning: a Laboratory Manual, 2nd ed. Cold Springs Harbor Laboratory Press, Cold Springs Harbor, N. Y. - Schroder, I., C. D. Wolin, R. Cavicchioli, and R. P. Gunsalus. 1994. Phosphorylation and dephosphorylation of the NarQ, NarX, and NarL proteins of the nitrate-dependent two-component regulatory system of Escherichia coli. J Bacteriol. l76(16):4985-92. Schwecke, T., J. F. Aparicio, I. Molnar, A. Konig, L. E. Khaw, S. F. Haydock, M. Oliynyk, P. Caffrey, J. Cortes, J. B. Lester, and et al. 1995. The biosynthetic gene cluster for the polyketide irnmunasuppressant rapamycin. Proc Natl Acad Sci U S A. 92(17):7839-43. Seno, E. T., and R. H. Baltz. 1989. Structural Organization and Regulation of Antibiotic Biosynthesis and Resistance Genes in Actinomycetes. In S. Shapiro (ed.), Regulation of Secondary Metabolites in Actinomycetes. CRC Press, Boca Raton. Shapiro, S. 1989a. Nitrogen Assimilation in Actinomycetes and the Influence of Nitrogen Nutrition on Actinomycete Secondary Metabolism, p. 135-212. In S. Shapiro (ed.), Regulation of Secondary Metabolites in Actinomycetes. CRC Press, Boca Raton. Shapiro, S. (ed.). 1989b. Regulation of Secondary Metabolism in Actinomycetes. CRC Press, Boca Raton. Shi, L., and F. M. Hulett. 1999. The cytoplasmic kinase domain of PhoR is sufficient for the law phosphate-inducible expression of pho regulon genes in Bacillus subtilis. Mol Microbiol. 31(1):211-22. Shimoda, N., A. Toyoda-Yamamoto, S. Aoki, and Y. Machida. 1993. Genetic evidence for an interaction between the VirA sensor protein and the ChvE sugar- binding protein of Agrobacterium. J Biol Chem. 268(35):26552-8. Simms, S. A., M. G. Keane, and J. Stock. 1985. Multiple forms of the CheB methylesterase in bacterial chemosensing. J Biol Chem. 260(18): 10161-8. Smith, C. P. personal communication Spiegelman, G. B., T. H. Bird, and V. Voon. 1995. Transcription regulation by the Bacillus subtilis response regulator SpoOA, p. 159-179. In J. A. Hoch and T. J. 167 158. 159. 160. 161. 162. 163. 164. 165. 166. 167. Silhavy (ed.), Two-Component Signal Transduction. ASM Press, Washington, DC. Stewart, R. C., R. VanBruggen, D. D. Ellefson, and A. J. Wolfe. 1998. TNP- ATP and TNP-ADP as probes of the nucleotide binding site of CheA, the histidine protein kinase in the chemotaxis signal transduction pathway of Escherichia coli. Biochemistry. 37(35): 12269-79. Stewart, V., and R. S. Rabin. 1995. Dual sensors and dual response regulators interact to control nitrate- and nitrite-responsive gene expression in Escherichia coli, p. 233-252. In J. A. Hoch and T. J. Silhavy (ed.), Two-Component Signal Tansduction. ASM Press, Washington, DC. Stock, J. B., A. J. Ninfa, and A. M. Stock. 1989. Protein phosphorylation and regulation of adaptive responses in bacteria. Microbiol Rev. 53(4):450-90. Stock, J. B., M. G. Surette, M. Levit, and P. Park. 1995. Two-Component Signal Transduction Systems: Structure-Function Relationships and Mechanisms of Catalysis, p. 25-52. In J. A. Hoch and T. J. Silhavy (ed.), Two-Component Signal Transduction. ASM Press, Washington, D. C. Strauch, E., E. Takano, H. A. Baylis, and M. J. Bibb. 1991. The stringent response in Streptomyces coelicolor A3(2). Mol Microbiol. 5(2):289-98. Strohl, W. R. (ed.). 1997. Biotechnology of Antibiotics, Second ed, vol. 82. Marcel Dekker, New York. Strohl, W. R. 1992. Compilation and analysis of DNA sequences associated with apparent streptomycete promoters. Nucleic Acids Res. 20(5):961-74. Strohl, W. R. 1997. Industrial Antibiotics: Today and the Future, p. 1-48. In W. R. Strohl (ed.), Biotechnology of Antibiotics, Second ed, vol. 82. Marcel Dekker, Inc., New York. Stutzman-Engwall, K. J., S. L. Otten, and C. R. Hutchinson. 1992. Regulation of secondary metabolism in Streptomyces spp. and overproduction of daunorubicin in Streptomyces peucetius. J Bacteriol. 174(1): 144-54. Takano, E., H. C. Gramajo, E. Strauch, N. Andres, J. White, and M. J. Bibb. 1992. Transcriptional regulation of the redD transcriptional activator gene accounts for growth-phase-dependent production of the antibiotic undecylprodigiosin in Streptomyces coelicolor A3(2). Mal Microbial. 6(19):2797-804. 168 168. 169. 170. 171. 172. 173. 174. 175. 176. 177. 178. 179. Takano, E., J. White, C. J. Thompson, and M. J. Bibb. 1995. Construction of thiostrepton-inducible, high-copy-number expression vectors for use in Streptomyces spp. Gene. l66(1):133-7. Tanaka, T., M. Kawata, and K. Mukai. 1991. Altered phosphorylation of Bacillus subtilis DegU caused by single amino acid changes in DegS. J Bacteriol. 173(17): 5507-15. . Tokishita, S., A. Kojima, H. Aiba, and T. Mizuno. 1991 . Transmembrane signal transduction and osmoregulation in Escherichia coli. Functional importance of the periplasmic domain of the membrane-located protein kinase, EnvZ. J Biol Chem. 266(11):6780-5. Vallejo, A. N., R. J. Pogulis, and L. R. Pease. 1995. Mutagenesis and synthesis of novel recombinant genes using PCR, p. 603-612. In C. W. Dieffenbach and G. S. Dveksler (ed.), PCR Primer; A Labaraboratory Manual. Cold Spring Harbor Laboratory Press, Plainview, NY. Vining, L. C., and S. Shapiro. 1984. Chloramphenicol production in carbon- lirnited media: efi'ect of methyl alpha-glucoside. J Antibiot (Tokyo). 37(1):74-6. Vining, L. C., and C. Stuttard (ed.). 1995. Genetics and Biochemistry of Antibiotic Production, vol. 28. Butterworth-Heinemann, Boston. Vogtli, M., P. C. Chang, and S. N. Cohen. 1994. ast2: a previously tmdetected gene encoding a 63-amino-acid protein that stimulates antibiotic production in Streptomyces lividans. Mol Microbiol. 14(4):643-53. Volff, J. N., and J. Altenbuchner. 1998. Genetic instability of the Streptomyces chromosome. Mol Microbiol. 27(2):239-46. Volz, K. 1995. Structure and Functional Conservation in Response Regulators, p. 53-64. In J. A. Hoch and T. J. Silhavy (ed.), Two-Component Signal Transduction. ASM Press, Washington, D. C. Vujaklija, D., S. Horinouchi, and T. Beppu. 1993. Detection of an A-factor- responsive protein that binds to the upstream activation sequence of strR, a regulatory gene for streptomycin biosynthesis in Streptomyces griseus. J Bacteriol. 175(9):2652-6l. Walker, M. S., and J. A. DeMoss. 1993. Phosphorylation and dephosphorylation catalyzed in vitro by purified components of the nitrate sensing system, NarX and NarL. J. Biol. Chem. 268:8391-8393. Wanner, B. L. 1995. Signal Transduction and Cross Regulation inthe Escherichia coli Phosphate Regulon by PhoR, CreC, and Acetyl Phosphate, p. 169 180. 181. 182. 183. 184. 185. 186. 187. 188. 189. 203-222. In J. A. Hoch and T. J. Silhavy (ed.), Two-Component Signal Transduction. ASM Press, Washington, D. C. Webber, C. A., and R. J. Kadner. 1995. Action of receiver and activator modules of UhpA in transcriptional control of the Escherichia coli sugar phosphate transport system. Mol Microbiol. 15(5):883-93. Webber, C. A., and R. J. Kadner. 1997. Involvement of the amino-terminal phosphorylation module of UhpA in activation of uhpT transcription in Escherichia coli. Mol Microbiol. 24(5): 1039-48. White, J., and M. Bibb. 1997. bldA dependence of undecylprodigiosin production in Streptomyces coelicolor A3 (2) involves a pathway-specific regulatory cascade. J Bacteriol. l79(3):627-33. Wietzorrek, A., and M. Bibb. 1997. A novel family of proteins that regulates antibiotic production in streptomycetes appears to contain an OmpR-like DNA- binding fold [letter]. Mol Microbiol. 25(6): 1 181-4. Willey, J., R. Santamaria, J. Guijarro, M. Geistlich, and R. Losick. 1991. Extracellular complementation of a developmental mutation implicates a small sporulation protein in aerial mycelium formation by S. coelicolor. Cell. 65(4):641-50. Willey, J., J. Schwedock, and R. Losick. 1993. Multiple extracellular signals govern the production of a morpha genetic protein involved in aerial mycelium formation by Streptomyces coelicolor. Genes Dev. 7(5):895-903. Williams, S. B., and V. Stewart. 1997. Discrimination between structurally related ligands nitrate and nitrite controls autokinase activity of the NarX transmembrane signal transducer of Escherichia coli K-12. Mol Microbiol. 26(5):91 1-25. Wolfe, A. J., and R. C. Stewart. 1993. The short form of the CheA protein restores kinase activity and chemotactic ability to kinase-deficient mutants. Proc Natl Acad Sci U S A. 90(4):]518-22. Yamada, Y., T. Nihira, and S. Sakuda. 1997. Butyrolactone Autoregulators, Inducers of Virginiamycin in Streptomyces virginiae: Their Sturctures, Biosynthesis, Receptor Proteins, and Induction of Virginiamycin Biosynthesis, p. 63-80. In W. R. Strohl (ed.), Biotechnology of Antibiotics, Second ed, vol. 82. Marcel Dekker, New York. Yamada, Y., K. Sugamura, K. Kondo, M. Yanagimoto, and H. Okada. 1987. The structure of inducing factors for virginiamycin production in Streptomyces virginiae. J Antibiot (Tokyo). 40(4):496-504. 170 190. 191. 192. Yang, Y., and M. Inouye. 1992. Intermolecular compalementation between two defective mutant signal-transducing receptors. Proc. Natl. Acad. Sci. USA. 88:1 1057-11061. Yoshikawa, H., J. Kazami, S. Yamashita, T. Chibazakura, H. Sane, F. Kawamura, M. Oda, M. Isaka, Y. Kobayashi, and H. Saito. 1986. Revised assignment for the Bacillus subtilis spoOF gene and its homology with spaOA and with two Escherichia coli genes. Nucleic Acids Res. 14(2): 1063-72. Zuber, P., and M. A. Marahiel. 1997. Structure, Function, and Regulation of Genes Encoding Multidomain Peptide Synthetases, p. 187-216. In W. R Strohl (ed.), Biotechnology of Antibiotics, Second ed, vol. 82. Marcel Dekker, New York. 171 "11111111111111.1000“