RELATEONSHIP BETWEEN PROTEASE AND ANTIBIOTIC IN DIFFERENTIATING BACiLLUS UCHENEFORMIS CELLS Dissertation fat" tfie Degree of PH. D. MECHEGAN STATE UNIVERSITY UUBISA VETKOVIC 19175 5“ fir’y- r. p 1-! i 41Lfil‘fd. , t L S ‘ was ‘2‘ ~ I 75 ’ . ; This is to certify that the , thesis entitled RELATIONSHIP BETWEEN PROTEASE AND. ANTIBIOTIC IN DIFFERENTIATING BACILLUS LICHENIFORMIS CELLS presented by I LJUBI‘S’A VITKOVIC has been accepted towards fulfillment of the requirements for Ph. D. degreein BIOPHYSICS 0-7639 0mm: l ’Llanmv 51mins. ll _., k ‘ V "It'lwlll “ ' x .V ' new ABSTRACT RELATIONSHIP BETWEEN PROTEASE AND ANTIBIOTIC IN DIFFERENTIATING BACILLUS LICHENIFORMIS CELLS By Ljubiéa Vitkovié Proteases and antibiotics are produced upon the initiation of sporulation by Bacillus species. Genetic, kinetic, and physiological data indicate that these early molecular events are related to the differentiation process and also to each other. The precise nature of their relationship is unknown. The principal objective of the study was to elucidate this phenomenon. The experimental approach was based on the hypothesis that some peptide antibiotics are released by the action of sporulation-related protease(s) on vege- tative cell protein(s). The hypothesis and its implications were investigated in Bacillus licheniformis A-S, a mutant derepressed for the production of the antibiotic Bacitracin. The extracellular proteases were purified and separated into three distinct serine enzymes on carboxy methyl cellulose (CMC) columns. About 1% of the total extracellular Bacitracin was found bound to crude protease. Purified Bacitracin—free, protease was inhibited by commercial Bacitracin. Ljubisa Vitkovié An antibiotic was produced in_vi££g by proteolysis of vege- tative (denatured) cell protein with the specific purified protease, CMC fraction III. Antibiotic was shown to be identical to Bacitracin in its UV spectra, biological activity, and chromatographic properties in a multiplicity of systems. Similarities exist between ethanol extractable peptides excreted by sporulating cells and those produced in_xit£g3 in their number, their Rf values on chromatography, and in their labelling pattern with 14[C]-D,L-ornithine. Bacitracin appears to exert product inhibition-type control on the protease. The distribution of Bacitracin precursor in vegetative cells is the same as the distribution of initiation factors involved in protein synthesis. The proposal is put forward that Bacitracin activation is a consequence of a sporulation-related control exerted by the specific serine protease. RELATIONSHIP BETWEEN PROTEASE AND ANTIBIOTIC IN DIFFERENTIATING BACILLUS LICHENIFORMIS CELLS By Ljubisa Vitkovié A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Biophysics 1975 DEDICATION I am pleased to dedicate this dissertation to the American people whose hospitality and financial support made it possible. ii ACKNOWLEDGMENTS I wish to thank my mentor and friend, Harold L. Sadoff, for his warmth, his support, and his patience throughout the course of this study. I wish to express my appreciation to Patricia Murphy, my friend and lover, who shared with me anguish and joy from this endeavor. I also wish to thank the members of my guidance committee, Estelle McGroarty, Ashraf El-Bayoumi, and Alfred Haug for their sup- port, interest, and help. The following individuals had significant impact on me during the course of this study: Leaf Huang, Gabor Kemeny, William Page, Harish Pant, Barbara Shimei, my parents, and sister. The research was supported by the funds from the College of Oste0pathic Medicine of Michigan State University and National Institutes of Health (grants #GM 14971, GM 01911 and #AI 01863). My entire university education was supported by the public funds at three state universities in New York, Iowa, and Michigan. iii TABLE OF CONTENTS Page LIST OF TABLES . . . . . . . . . . . . . . . . . . . . . . . . . vii LIST OF FIGURES . . . . . . . . . . . . . . . . . . . . . . . . . viii INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . 1 CHAPTER 1. LITERATURE REVIEW . . . . . . . . . . . . . . . . . . . . 3 I. Introduction . . . . . . . . . . . . . . . . . . . 3 II. General Consideration of Sporulation Process . . . 4 A. Morphological Stages and Biochemical Events Associated with Spore Formation . . . . 4 B. Sporulation as a Model of Cell Differentiation . . . . . . . . . . . . . . . . 6 C. Cautionary Remarks on Data Interpretation . . . 7 III. Early Sporulation Events . . . . . . . . . . . . . 9 A. Proteolytic Enzymes (Proteases) . . . . . . . . 9 1. General Description of Proteases in Bacilli . . . . . . . . . . . . . . . . . . 9 2. Relationship of Protease Production to Sporulation . . . . . . . . . . . . . . 11 3. Function of Proteases in Bacterial Sporulation . . . . . . . . . . . . . . . . 13 iv CHAPTER * Page B. Peptide Antibiotics . . . . . . . . . . . . . . 18 1. Physico-Chemical Attributes . . . . . . . . 18 2. Relationship of Antibiotic Production to Sporulation . . . . . . . . . . . . . . 19 3. Mode of Action and Possible Role(s) in Differentiation Process . . . . . . . . 21 C. Relationship Between Protease(s) and Antibiotics . . . . . . . . . . . . . . . . . . 22 1. Genetic and Kinetic Evidence . . . . . . . 23 2. Statement of the Problem . . . . . . . . . 24 IV. Early Sporulation Events in Bacillus licheniformis . . . . . . . . . . . . . . . . . . . 25 A. The Organism . . . . . . . . . . . . . . . . . 25 B. Proteases . . . . . . . . . . . . . . . . . . . 26 C. Peptide Antibiotic: Bacitracin . . . . . . . . 28 V. Summary . . . . . . . . . . . . . . . . . . . . . . 32 II. EXTRACELLULAR PROTEASE 0F BACILLUS LICHENIFORMIS: PURIFICATION AND PROPERTIES . . . . . . . . . . . . . . . 34 Materials and Methods . . . . . . . . . . . . . . . . . . 3S Organism and Growth Conditions . . . . . . . . . . . 3S Enzyme Assay . . . . . . . . . . . . . . . . . . . . 35 Specific Protease Inhibitors . . . . . . . . . . . . 36 Antibiotic Bioassay . . . . . . . . . . . . . . . . . 36 Kinetics of Protease Appearance . . . . . . . . . . . 36 Protease Purification . . . . . . . . . . . . . . . . 37 Results . . . . . . . . .4. . . . . . . . . . . . . . . . 38 Discussion . . . . . . . . . . . . . . . . . . . . . . . 44 CHAPTER Page III. PRODUCTION OF BACITRACIN IN_VITRO BY PROTEOLYSIS OF VEGETATIVE BACILLUS LICHENIFORMIS CELL PROTEIN . . . . . 46 Materials and Methods . . . . . . . . . . . . . . . . . . 47 Organism and Growth . . . . . . . . . . . . . . . . . 47 Preparation of the Substrate . . . . . . . . . . . . 47 Antibiotic Production in vitro . . . . . . . . . . . 48 Paper Chromatography . . . . . . . . . . . . . . . . 48 Thin-Layer Chromatography . . . . . . . . . . . . . . 48 Scintillation Counting . . . . . . . . . . . . . . . 49 Spectroscopy . . . . . . . . . . . . . . . . . . . . 50 Results . . . . . . . . . . . . . . . . . . . . . . . . . 50 Discussion . . . . . . . . . . . . . . . . . . . . . . . 60 IV. THE ASSOCIATION OF THE BACITRACIN PRECURSOR WITH RIBOSOMES OF VEGETATIVE BACILLUS LICHENIFORMIS CELLS . . 64 Materials and Methods . . . . . . . . . . . . . . . . . . 6S Organism and Growth Conditions . . . . . . . . . . . 65 Preparation of Ribosomes, Ribosomal Subunits, and the Fractionation of the Cell Lysate . . . . . . . . 6S Fractionation of NH401 Wash . . . . . . . . . . . . . 68 Results . . . . . . . . . . . . . . . . . . . . . . . . . 68 Discussion 0 O I O O O O O O O O O O 0 I O O C O O O O O 72 BIBLIOGRAPHY . . . . . . . . . . . . . . . . . . . . . . . . . . 76 vi Table II. II. II. III. LIST OF TABLES Chapter II Protease purification protocol . The inhibition of B: licheniformis extracellular proteases . . . . . . . . . . . . . . . . . . Chapter III Antibiotic production in_vitro by different protease preparations . . . . The results of chromatographic identification of in_vitro antibiotic . . Chapter IV Bacitracin yield (mg) from various ribosomal substrates. Distribution of the Bacitracin precursor in vege- tative cell lysates Bacitracin yield upon proteolysis of NH4C1 wash fractions . . . . . . . . . vii Page 43 44 52 58 7O 71 72 LIST OF FIGURES Figure Page Chapter II 1. The range and standard deviation of the antibiotic bioassay . . . . . . . . . . . . . . . . . . . . . . . . 39 2. Growth and total extracellular (--o--) and intra- cellular (——o——) protease activities for B, licheniformis A-S . . . . . . . . . . . . . . . . . . . 40 3. The elution profile of Bacitracin-free protease on a carboxy methyl cellulose column . . . . . . . . . . 42 Chapter III 1. Growth of B. licheniformis A-5 and kinetics of intracelIular (closed symbols) and extracellular (open symbols) Bacitracin appearance . . . . . . . . . . 51 2. Bacitracin production as a function of substrate concentration . . . . . . . . . . . . . . . . . . . . . 53 3. Primary and secondary structures of Bacitracin A . . . . . 55 4. Kinetics of in vitro Bacitracin production (———0. The reactiafi was inhibited by S x 10'3 M PMSF (---) . . . . . . . . . . . . . . . . . . . . . . . 56 5. Spectra of the antibiotic produced ifl_vitro . . . . . . . 57 6. Radio-bioautogram of Bacitracin produced by sporulating B, licheniformis cells chromatographed with in vitro antibiotic in two solvent systems. Shaded—areas represent bacteriocidal activity . . . . . 59 7. The incorporation of [3]H-ornithine into Bacitracin at different growth stages of B, licheniformis (data from reference 14) . . . . . . . . . . . . . . . . 62 viii Figure Page Chapter III 1. Sucrose density-gradient sedimentation of Bacillus licheniformis ribosomes . . . . . . . . . . . . . . . . 69 ix INTRODUCTION Cell differentiation requires the selective expression of the cell genome. Bacterial Sporulation is studied as a model system for understanding molecular events that control and accompany cell differentiation. This process is a time-ordered sequence of bio- chemical and morphological events resulting in the conversion of vegetative bacterial cells into spores. The production of a class of proteolytic enzymes (proteases) and uncharacterized peptide antibiotics occurs at the time of the initiation of the differentiation process and certain circumstantial evidence indicates that these two early molecular events are not only related to Sporulation, but also to each other. However, direct proof of the existence and possible nature of this relationship are lacking. The research performed for this dissertation was predicated on the hypothesis that the elaboration of an antibiotic by sporulating cells is a result of the proteolysis of vegetative cell protein by Sporulation-related protease. The data presented indicate that a specific protease releases Bacitracin from vegetative cell proteins of Bacillus licheniformis. Preliminary reports of this work were made during the course of investigation (132, 133). The dissertation is presented in four chapters. The Litera- ture Review, Chapter 1, presents a summary of the current knowledge of bacterial differentiation with the emphasis on the early molecular events. The original results constituting the main body of the dis- sertation are presented in the subsequent three chapters. They are written in the form of scientific papers and will be submitted for publication. All the references are grouped into a bibliography. CHAPTER I LITERATURE REVIEW 1. Introduction The purpose of this chapter is to introduce the reader to bacterial sporulation and to summarize the available information about early molecular events occurring during this process. A brief description of morphological changes and biochemical events is pre- sented. Sporulation is then compared to cell differentiation of other prokaryotic and higher eukaryotic organisms in order to illustrate its utility as a model system. Difficulties in the interpretation of data concerning protease and antibiotic production are outlined in order to facilitate a better understanding of experimental results. Early sporulation events are described in the next section. The apparent consanguinity of sporulation, protease production, and anti- biotic formation leads to the statement of the hypothesis to be tested. The last section describes Bacillus licheniformis, the organism used in this study. The presentation in this chapter then proceeds from the general description of the cell differentiation in Bacilli to pro- teases and Bacitracin formation in B. licheniformis. This chapter is not intended to constitute a thorough and complete review of the various aspects of sporulation nor to 3 construct an all inclusive bibliography on this subject. Several excellent reviews are available (6, 31, 40, 46, 64, 78, 98, 103, 127). The effort is made, however, to evaluate critically often conflicting data and to propose tentative generalizations. II. General Consideration of Sporulation Process A. Morphological Stages and Biochemical Events Associated with Spore Formation Vegetative cells of the family Bacillacae undergo metamorphosis and enter a dormant state when either their carbon or nitrogen sources, or both become limiting. Spore formation begins, under laboratory conditions, at the end of exponential growth, a time con- ventionally denoted as to. The various hours after initiation are denoted t t 1, 2..., sporulate if addition of glucose no longer represses the ultimate tn. A cell is considered to be committed to sporulation of the culture. The initiation of sporulation is under the control of catabolite repression (103, 104, 126). There are seven distinct morphological stages of spore development (31, 46, 64, 78, 98). Stage I. Derepression of spore genes occurs following the metabolic shift-down due to C or N source depletion. The cells under- go a final, critical, round of DNA synthesis. The nucleoids elongate into a long axial filament. At this stage intra- and extracellular levels of antibiotic and protease increase dramatically. Stage II. A membranous structure, the sporulation septum, begins to form along with concomitant nuclear segregation. The formation of this septum is under precisely the same control as a normal division septum (51). Thus, this aspect of sporulation is modified cell division which yields sister cells of unequal size. Protein turnover and the synthesis of tricarboxylic acid (TCA) cycle enzymes begins. Stage III. Cell wall deposition between two membrane layers is blocked. Then by a process of selective membrane growth, the larger cell begins to engulf the smaller cell, the prespore. The engulfment is complete, the spore protoplast is enclosed in a double membrane and free in the cytoplasm. Glucose dehydrogenase is pro- duced. Stage IV. A cortex forms by deposition of a cell wall-like peptidoglycan between the two membranes surrounding the spore proto— plast. Dipicolinic acid (DPA), a sporulation Specific compound of unknown function, begins to be synthesized. Stage V. Multiple proteinaceus layers are deposited around the developing spore, the forespore, building up the spore coat. Calcium is incorporated. Stage VI. As the spore matures, resistance to chloroform and heat develop. The spore coat becomes denser. Stage VII. The sporangium lyses in most spores of Bacillus and the spore is released into the medium. Spore formation is susceptible to environmental factors which may influence the time pattern of development and the chemical composition of the spore (64). B. Sporulation as a Model of Cell Differentiation Spores are exceedingly heat resistant resting cell forms. The study of their development was initially motivated by the need in the canning industry for devising rational heat treatments. More recently, sporulation is studied with a view to understanding cell differentiation (31). Myxospore formation in Myxococcus xanthus and encystment in Azotobacter vinelandii are fundamentally similar to sporulation in Bacillus species (98). The sequences of morphological developments in the sporulation of Bacillus and Clostridium species are identical (98). The developmental patterns in lower eukaryotic organisms, yeasts and slime moulds, are more complex but similar to sporulation in many molecular events (79). All these processes depend somewhat on the derepression of certain genes mediated by starvation. Protein turnover and changes in enzymic patterns and morphology then follow. From the experimentalist's viewpoint, bacterial sporulation is perhaps the most amenable system employed in current studies of the molecular basis of morphogenesis (31, 46). Large masses of homogeneous populations of spores, vegetative, and sporulating cells can be obtained. The process is potentially subject to genetic analysis and the gene array has been established in at least one spore-forming bacterium, B, subtilis. Furthermore detailed cyto- logical and biochemical studies have been performed and provide back- ground for_advanced analysis. C. Cautionary Remarks on Data Interpretation Molecular events and molecular species detected during sporulation can be (31): (a) absolutely necessary for spore formation (viz., septation, Stage II), (b) pleiotropic for sporulation (i.e., necessary only in certain media while not required in others), or (c) irrelevant for sporulation (e.g., production of amylase). The major problem then is to distinguish sporulation-specific from sporulation-related gene products. It is important to discern whether the relation between sporulation and the appearance of an antibiotic, for example, is causal or fortuitous. Correlation does not establish causal relationship (78, 126). However, "causality is an appealing working hypothesis since finality of the production of these biologically active substances is unknown, and some finality is teleonomically needed" (103, p. 60). It is clear that there is no a_priori answer to this dilemma: causality and fortuity might both be encountered depending on the particular product under consideration. So far, no certain correlations between morphological and biochemical events have been established although many sporulation-specific events are known to exist. The second query is rarely mentioned and refers to the limits of accuracy of the experiments under discussion. The following facts must be considered in the evaluation of experimental reports: (1) The recognition of mutant traits in many mutant clones had to rely on bacteriological plate-tests of questionable sensitivity (3. St cr bu II Th Whether th -_ and specificity (6, 103). This is the case where accuracy had to be sacrificed in favor of efficiency. (2) The presence of proteases is usually detected by their activity on casein. This is an insensitive and selective assay. A variety of substrates should be used (e.g., azoderivatives of different proteins, luciferase, synthetic esters, etc.) to increase the sensitivity of the assay which should be capable of detecting usually low intracellular levels of protease. (3) Antibiotic is detected by its activity against sensitive organisms (bioassay). This procedure requires antibiotic quantities in the order of ug/ml, a level which probably exceeds by several orders of magnitude those needed for cellular function (perhaps a few molecules/cell). Initial detection should be conducted spectroscopically whenever possible (24). This should be followed by extensive con- centration and purification of the cell extract in order to confirm spectroscopic results by bioassay and/or other means. Often no antibiotic activity is detectable in cells sporulating in liquid cultures. However, activity is clearly displayed on solid medium (107). Several media must be examined before antibiotic production can be ruled out in mutants. This is crucial in detecting mutants which sporulate reasonably well but produce exceptionally low levels of antibiotics (cf. Sec. III, B). These are some considerations which must be taken in deciding whether the occurrence of a molecule is a result of derepression of a catabolite-repressed gene or the expression of a sporulation-specific gene (29). III. Early Sporulation Events Upon the initiation of sporulation, there occurs a dramatic increase in the intra- and extracellular proteolytic activity. The elaboration of one or several peptide antibiotics follows. These phenomena are termed "early" sporulation events (as opposed to "late" events such as DPA synthesis or appearance of lytic enzymes). A. Proteolytic Enzymes (Proteases) 1. General Description of Proteases in Bacilli A single Bacillus species usually produces several distinct proteases which have been classified according to the following overlapping criteria: (a) Location; intra- and extracellular (often erroneously termed "exoproteases"). This classification is a matter of current debate (see Ch. 11, discussion); (b) pH at which the activity is maximal: acidic, neutral, and alkaline; (c) Inhibitor specificity (related to the nature of the active site): (i) serine-proteases are specifically inhibited by phenylmethanesulfonyl fluoride (PMSF), diisopropylfluoro- phosphate (DFP), or other agents coupling to the seryl hydroxyl group, and (ii) metallo—proteases are inhibited by 10 ethylenediaminotetraacetate (EDTA), o-phenanthroline, or other chelating agents; (d) The nature of the proteolytic attack: exo- and endopeptidases, That is, amino or carboxypeptidases as opposed to those whose specificity is directed toward bonds between specific amino acids within a given protein (viz., trypsin). The discussion here is restricted to the polypeptidases, i.e., proteolytic enzymes that are able to cleave peptide bonds of protein molecules. Proteases in general can activate, inactivate or modify already existing enzymes. Usually denatured proteins are more sen- sitive to proteases than native proteins. Proteases (or peptidases) effect protein maturation (or "processing") by the removal of N- formylmethionine or methionine from newly synthesized polypeptides. They can convert inactive to active forms of certain enzymes and thus perform a variety of control functions (29, 53). Protease activity can be regulated by changes of enzyme's substrates or by specific inhibitors. Specific protease inhibitors are well known in plant and animal systems (25) although none are known to occur in microorganisms (see Chapter 11). Among the Bacilli, the extracellular proteases of B, subtilis have been most thoroughly investigated. Several seryl enzymes have been identified which are active at neutral and alkaline pH, and which differ slightly in their physico-chemical properties (43, 85). Neutral metal protease requiring Zn2+ for activity and Ca2+ for stability has been particularly well characterized (81). The question is, "Are 11 all these enzymes observed during sporulation and what is their rela- tionship to it?" 2. Relationship of Protease Production to Sporulation Physiological and genetic studies indicate a relationship between sporulation and extracellular protease production: (a) proteolytic activity appears at the time of initiation and increases during the course of sporulation reaching maximum at Stage V of sporulation (10, 87, 88, 92, 108). The increase in protease activity is a result of B§_ngg_synthesis, rather than delayed excretion or activation of preformed enzymes; (b) the efficiency of spore formation correlates with normal production of extracellular protease (29); (c) many asporogenous (sp') mutants do not produce proteases (pR-) and reversion to sporogeny is accompanied by the ability to produce the enzymes (29, 80, 103). The latest studies of protease-sporulation relationship confirm these general results but present a much more complicated picture. Mutants of B, subtilis lacking intracellular neutral metallo-protease sporulated as well as the parental strain indicating that this enzyme is not related to sporulation (44). Another mutant of the same organism having a pheno- type manifesting reduced extracellular proteolytic activity had defective protease consisting of a fragment of the wild-type enzyme (109). The study of 29 pR- mutants of B, cereus indicated that extra- cellular protease and spore formation are under closely related but distinct catabolic controls (5). Temperature sensitive (Ts) mutants 12 which sporulate normally at the permissive temperature but are sp- at the restrictive temperature have been recently reported. Studies on one of such mutants indicate that initiation of sporulation requires serine protease activity (70). The level of the intracellular protease activity is about 15-fold lower in another Ts (sp') mutant grown at the restrictive temperature, than that of the parental strain grown at the same temperature (84). Szulmajster and Keyer studied still another Ts (sp-, pR-) mutant (128). Their results sug- gest a functional relationship between the lack of intracellular protease activity, lack of protein turnover and asporogeny. Clearly, not all pr- mutants have an alteration in a structural gene coding for protease(s). It appears possible that some mutants, unable to produce protease, have a lesion in a gene regulating the expression of protease (see section II C). Sporulating strains that are pre- sumably pr' are known to exist (5, 29, 32). The existence of these mutants indicates that protease production may not be an absolute requirement for sporulation (cf. section II C). The following pre- liminary conclusion can be made: Although causal relation between protease production and sporulation has so far not been rigorously established, proteases are certainly sporulation-related products which may perform important function(s) in microbial differentiation. We must address ourselves to the question: "What are the functions of these molecules in sporulation?" l3 3. Function of Proteases in Bacterial Sporulation This question is currently vigorously investigated and numbers of reviewers are either cautious (7, 46, 126), or somewhat contradictory (29). Balassa et a1. state that ”with a few exceptions, the role (if any) of these proteins in the formation of the spore is unknown or hypothetical" (7). Szulmajster claims that ". . . the real function of the proteases is still undetermined" (126) while Hanson et a1. think that ". . . the function of the individual pro- teases cannot be described at present" (46). Doi appears incon- sistent: on one hand, he says that from the available data no definitive role is ascribable to the intracellular proteases; on the other, he feels that "there are enough data to suggest that proteases play a key role in the differentiation process and [that] they should be considered as regulatory enzymes at the post-transitional level" (29). The uncertainty about the role of proteases in sporulation seems incongruent with presently existing evidence and possibilities: (i) One of the main functions of microbial proteases lies in their participation in protein turnover by the hydrolytic degradation of protein (53). In fact, protein turnover is the major source of free amino acids for protein synthesis under shift-down conditions. It appears firmly established that extensive protein turnover (up to 20%/hr) occurs during the conversion of vegetative cells into spores (S3, 64, 78). There is strong evidence suggesting the relationship of turnover, protease activity and sporulation (46, 64, 128). How- ever, a single report is persistently quoted as evidence to the con- trary. Slapikoff et al. reported a strain of B, brevis which lacks 14 detectable intra- and extracellular proteolytic activity, does not turnover protein, but sporulates normally (112). Aronson pointed out that the increase in protein turnover following starvation is partly due to derepression or activation of proteases and partly to the synthesis of polypeptides more susceptible to proteolysis. The evidence for lack of turnover in B, brevis was based primarily on the stability of protein labeled during exponential growth (4). The function of proteases in protein turnover is corollary to their possible function in regulation of levels of amino acid pools. The ultimate effect of these complicated interrelationships could be that proteases provide trigger mechanism for the regulation of syn- thesis of RNA and protein (S3). Mandelstam and Waites have presented data deserving of special attention (80). They determined that protein is degraded at a rate of 8-10%/hr during sporulation of wild-type B, subtilis. Mutants that have lost extracellular protease are asporogenous and their intracellular protein is stable for many hours. When the protease is regained by back mutation all normal functions of the cell are restored. It is clear that protein turnover is essential for sporulation of cells suspended in water or in buffered solutions ("endotrophic" condition) where net protein synthesis is impossible. This, however, does not hold for conditions existing in media where substrates such as glutamate and ammonium ions supply sufficient carbon and nitrogen for a considerable net protein synthesis. Yet pr- mutants retain sp- phenotype in such media. These observations led the authors to propose that protease is needed for destruction 15 of repressor proteins that prevent transcription of sporulation— specific cistrons. Thus, protein turnover may serve less obvious but critical roles in morphogenesis. (ii) Several enzymes appear to be selectively inactivated during initiation and first two stages of spore development (12, 125). In particular, Deutscher and Kornberg found that aspartic transcar- bamylase and IMP dehydrogenase were present during exponential phase but disappeared by the time of appearance of refractile spores (28). The pattern of development of these enzymes was different in an sp-, pRE mutant. The susceptibility of these enzymes to proteolytic activity in extracts suggested that they were selectively proteolysed during sporulation. The net synthesis of nucleic acids ceases as B, subtilis cells initiate spore formation. Germinating spores are incapable of Bg_gg!2 nucleotide biosynthesis before protein synthesis takes place. Switzer et al. recently reported that aspartate transcarbamylase the first enzyme of pyrimidine nucleotide synthesis, is inactivated in B, subtilis cells when the culture enters stationary phase. They suggest that enzyme inactivation during sporulation proceeds by different, specific mechanisms and may be followed by proteolysis of the inactivated protein (124). (iii) The most interesting and diverse possibilities of protease function in morphogenesis are offered by "limited proteo- lysis." An increasing body of recent evidence indicates that post- translational modification of enzymes has widespread occurrence and considerable biological significance. This mechanism is involved in the activation of digestive enzyme precursors, the cascade mechanisms 16 involved in blood coagulation and complement action, and viral assembly (50). (a) The occurrence of post-translational modification in sporulation has been confirmed by classical studies of Sadoff and coworkers. They purified and characterized fructose 1,6-diphosphate aldolase from spores and vegetative cells of B, cereus. The enzymatic activity resided in two distinct molecular species (100). They further demonstrated that sporulation-related protease, in a limited proteolysis, converts vegetative cell aldolase into a spore aldolase (101). (b) At about the same time Losick and Sonenshein reported that DNA-dependent RNA polymerase from sporulating B, subtilis cells failed to transcribe phage De DNA ig_zi££g_while the enzyme from vegetative cells was active with the same template (74). This report of a change in the template specificity of RNA polymerase during sporulation suggested a mechanism of selective gene expression required in differentiation. It also opened highly competitive and contro- versial realm of inquiry. They further reported structural altera- tion of the same enzyme during sporulation and suggested that pro- tease may be responsible (73). Sporulation-specific alteration in B-subunit of RNA polymerase was subsequently described in a single site Ts, pR' mutant (70). Several lines of evidence indicated that vegetative 8 subunit was converted to the lower molecular weight B subunit found in sporulating cells by limited proteolysis. Orrego et a1. (84) did not confirm these conclusions working with their Ts sp- mutant. They claim that vegetative and sporulating cell enzymes 17 are identical if purified under conditions that totally inhibit proteolysis (59). Double label mixing experiments yielded results suggesting that the alteration of 8' subunit is due to proteolysis ig_xi£:2_(72). Szulmajster has been the most outspoken critique of the work reported by Leighton et al. (126, 127).: He also, however, co- authored a report of ifl.!i££2.§: subtilis RNA polymerase modification by intracellular protease from B, megaterium (83). Klier and colleagues presented 12.X119 evidence in favor of sequential modification of RNA polymerase subunits during sporo— genesis (62, 63). Their results indicate that structural modifica- tion is caused by limited proteolysis. Post-translational modification of RNA polymerase is not restricted to spore-forming organisms (49, 60). The existence of several sporulation related RNA-polymerase-binding proteins indicates that a variety of proteins control differential genetic expression at the transcriptional level (35, 39). (c) Steinberg recently reported ig_xizg_evidence for phenotypic modification of a thermosensitive lysyl—tRNA synthetase in sporulating B, subtilis cells. He suggested that mutant enzyme might be modified during sporulation (118). This is more than a possibility when one considers that both leucyl- and methionyl-tRNA synthetases are proteolytically modified in B, 2911 (20, 96). In conclusion, Bacillus species produce variety of proteo- lytic enzymes. Genetic, physiological, and kinetic studies indicate that intra- and extracellular seryl-proteases are sporulation— related molecules. They cause protein turnover, enzyme inactivation, 18 and (post-transcriptional) modification of vegetative cell enzymes. More information is desirable for the final evaluation of proteases as regulatory enzymes during developmental process. These conclusions are, however, consistent with the data, with the known functions of proteases, and with the view of biological systems as teleonomic and conserving organizations. B. Peptide Antibiotics 1. Physico-Chemical Attributes Another group (class) of molecules elaborated by spore- forming microorganisms (bacteria, fungi, and streptomycetes) are peptide antibiotics. Several hundred molecular species have been described in the literature. Physico-chemical attributes of sporulation-related peptides are summarized below (47, 86): (i) A single species usually produces a family of closely related peptides rather than a single entity. The members of the group generally differ from each other by few amino acids. (ii) They contain both L- and D-forms of amino acids. (iii) Large numbers of unusual amino acids have been found in peptide antibiotics (e.g., N-methyl derivatives of common amino acids; B-amino, imino, and dihydroxy amino acids). (iv) Some antibiotics contain fatty acids, pyrimidines, amino sugars, hydroxy acids, and amines. (v) Many peptide antibiotics possess cyclic structures. 19 (vi) Their unique structure embues them with resistance to pro- teases and peptidases. (vii) They contain 6 to 15 amino acids with molecular weights ranging between 300 and 2000. (viii) All peptide antibiotics studied are synthesized enzymatically, without involvement of ribosomes and mRNA. A number 0f dere~ pressed antibiotic producing organisms have been isolated and, upon purification, the properties of their sporulation anti- biotics have been determined. There are five classes of these peptides. These are: edeines, bacitracins, tyrocidines, gramicidins, and polymixins. Their mode of action ranges from binding to, and modifying nucleic acids' function to membrane modification. 2. Relationship of Antibiotic Production to Sporulation As in the case of proteases, two lines of investigation sug- gest a relationship between peptide antibiotic production and sporu- lation: genetic and physiological (6, 46, 103) as follows: (i) Antibiotics are usually not produced by rapidly growing - bacilli; (ii) Specific sporulation inhibitors added before antibiotics' appearance prevent their elaboration; (iii) Sp- mutants which are antibiotic—negative (ab-) are blocked at to (Spa); (iv) A few mutants that are selected for ab_ phenotype are also sp . 20 (v) Restoration of antibiotic production by reversion, transduction, or transformation restores the cells' ability to sporulate. No single antibiotic, however, has yet been unambiguously shown to be essential to bacterial sporulation. And the relationship is not a simple one. The amount of antibiotic produced and the extent of sporulation appear to be inversely proportional. The existence of ab-, sp+ mutants would suggest that the relationship between two processes might be fortuitous. Such mutants have been recently reported. Ray and Bose described streptomycin- resistant, adenine-requiring mutants of B, subtilis which sporulated normally and failed to produce mycobacillin, an antifungal polypep- tide antibiotic (89). They did not, however, demonstrate the lack of other antibiotics made by this organism. This is required in order to rule out causality between the two phenotypes. Haavik and Thomassen isolated a presumptive bacitracin-negative mutant of B, licheniformis which was able to sporulate. Their procedures suffered from all the pitfalls described in section II C (42). Kambe et a1. (57) investigated Gramicidin S- mutrants of B, brevis and noted that many mutational events block gramicidin pro- duction. At the end of the paper they state, referring to an unpub- lished observation, that one of the mutants sporulated more efficiently than the wild type (57). This observation needs to be investigated further. In conclusion, presently available evidence favors the view that sporulation and antibiotic production are related. This 21 statement, however, does not take into consideration the role that these peptides may play in sporulation. 3. Mode of Action and Possible Role(s) in Differentiation Process The most important remark in this regard is that producer organisms become resistant to antibiotics at the time of their appearance. This observation suggests that, if antibiotics function in sporulation, their effect on vegetative and Sporulating cells is different, their mode of action changes, or a combination of these two possibilities. The first evidence for a biological function of a peptide antibiotic in the life cycle of the organism producing it was pro- vided by Sarkar and Paulus (102). These workers have demonstrated that a tyrothricin from B, brevis specifically inhibits this orga- nism's RNA polymerase ig_yi!g_and ig_!i££g, This inhibition could terminate the transcription of genes which function only during vege- tative growth. It would be very difficult, however, to demonstrate such specificity considering that nobody has yet isolated sporulation- specific genes. Yet further evidence in favor of this hypothesis has been recently supplied (94). The antibiotics used by Sarkar and Paulus consisted of cyclic peptide tyrocidine and linear peptide gramicidin. Ristow and co-workers showed that tyrocidine alone inhibits RNA synthesis leading to the cessation of growth. In con- trast, gramicidin does not affect net RNA synthesis ig_!izg_but inhibits sporulation. (Interestingly, it inhibits RNA synthesis E! vitro!) VThe mixture of the two partially reduces tyrocidine effect. 22 Hodgson suggested that certain antibiotics, by modifying permeability properties of membranes, affect the accumulation of DPA and calcium in the cortex, the removal of water, the contraction of prespores, and the formation of spore coat layer (52). Sadoff made an effort to rationalize antibiotic production by sporulating bacilli (97). He states that antibiotics "produced by sporulating cells are not the products of gratuitous synthesis but are functional in the control of cellular morphogenesis." The follow- ing additional arguments are envoked to support this hypothesis: --selective pressures would have eliminated those metabolic products without function in the cellular economy. --antibiotics cause events similar to those observed during sporulation (eg., cessation of DNA and cell wall synthesis, alteration of membrane structure and function, etc.). --Mutants that overproduce antibiotics are viable. This sug- gests that compensatory mechanisms exist within the cells to offset the effects of extensive antibiotic production. This would not be the case if antibiotics were only gratuitous "secondary metabolites." Evidence presently available favors the view that peptide antibiotics are sporulation—related molecules. They probably partici- pate in the regulation of spore formation but there is insufficient data to prove it (78). C. Relationship Between Protease(s) and Antibiotics The data presented so far indicate that seryl—proteases and antibiotics are independently related to sporulation. Proteases 23 effect protein turnover and enzyme inactivation. Both of these functions play a role in spore formation. Powerful arguments intimate that peptide antibiotics also fulfill sporulation-specific functions although sufficient evidence is still lacking. There is further evidence suggesting that these two events might be mutually related with respect at least to one class of antibiotics. 1. Genetic and Kinetic Evidence Schaeffer found that all sp-ab- strains of B, subtilis (more than 50) were also pr' and blocked at stage zero of sporulation. This was one of only two constant associations found among the variety of markers tested. The other one was transformation (sexual transfer utilizing DNA and intact cells) negative and pr- (103). Freeze and Fortnagel first reported the exception to this rule (32). In 46 sp- mutants studied they observed all four com- binations of ab and pr phenotypes (ab-pr+, ab+pr+, ab-pr+, ab+pr-). It is interesting that only four of these mutants were ab+pr- strains. In light of described experimental difficulties, it is likely that these four strains have mutations in their late sporulation genes making them ab+pr+, but protease assay was not sensitive enough. Several of the pr- strains described above were shown to be also ab- (70, 80). The linkage of pr, ab, sp characters in variety of mutants can be rationalized without assuming that they are related. If sporulation consists of a dependent sequence of events in which the occurrence of any one event depends on the successful completion of the earlier ones (78), the existence of ab—pr- or pr-ab' mutants would 24 be expected depending which of the events occurs first in the sequence. Or if these are pleiotropic mutants probably derived by mutation in a regulatory catabolite repression [CR] gene (103), the whole sequence of structural genes could be turned off. Many of the sporulation- related characters would be subject to the sequential genetic control which governs sporulation genes without playing any role in actual differentiation process (6). Although the statements above add to our understanding of the genetic data, they really do not explain the kinetic results. Extracellular protease and antibiotic activity display a very similar temporal pattern during the course of sporogenesis. Anti- biotics appear almost at the same time as proteases. For example, according to Bernlohr "the time of [extracellular protease] appearance coincides with the time when glucose is exhausted from the medium and the cells begin to change their metabolism for commitment to sporu- lation. During this time, and with the same kinetics, the polypep- tide antibiotic, bacitracin, is produced and released" (10, p. 539). There appears to exist a temporal relationship between intracellular protease activity and activity of various antibiotic synthetases (compare 33, 34, 68 with Figure 1, Chapter III). 2. Statement of the Problem In the search for more complete explanation of the presented data, the following hypothesis is proposed: Elaboration of an anti- biotic by sporulatinggcells is a result of a proteolysis of vege- tative cell protein(s) by sporulation-related_protease. 25 The hypothesis has implications beyond the immediate rationalization of the data. Protease and antibiotic production are two molecular events of fundamental importance in all sporulating bacilli. Their function in the differentiation process is not well understood. Evidence of their mutual relatedness is presently unexplained. Proving the hypothesis correct would demonstrate a new function for the sporulation proteases complementary to, and con- sistent with, the ones already shown. It would confirm the existence of the relationship between antibiotics and proteases. It would further indicate a novel mechanism of antibiotic production involving proteolytic activation. Lastly, it would open a possibility for a new regulatory mechanism in sporulation. This dissertation is a product of the research designed to test the above hypothesis. IV. Early Sporulation Events in Bacillus liCheniformis Preliminary investigations related to the working hypothesis utilized the organisms commonly used in the study of bacterial sporulation: B, subtilis, B, megaterium, and B, cereus (99 and Celikkol unpublished results). The need arose for an experimental system having the following properties: well defined antibiotic in high yield and protease of known properties. A. The Organism Bacillus licheniformis was deemed well suited for the investi- gation judging by the information already in existence. Physiological studies indicated that sporulating cells produce protease and 26 antibiotic in a manner already described in other bacilli: the anti- biotics consisted of a single well characterized family called Baci- tracins. Proteases of this organism had been partially purified and some of their properties were already known. The existence of a dere- pressed strain producing very high levels of Bacitracin provided an experimental advantage. Bergey's Manual (8th ed.) describes B, licheniformis as lichen-shaped bacterium which grows in many foods and whose spores occur in soil. It ferments glucose anaerobically to 2,34butanediol and glycerol. It can grow with ammonia as a sole source of nitrogen. B, licheniformis seems to be closely related to B, subtilis. Ribosomal protein patterns and antigenic prOperties of 708 ribosomes in the two organisms appear quite similar (36). Spore antigens are, however, different according to agglutimation and precipitation tests. B. Proteases Damodaran et a1. first fUnctionally characterized extra- cellular protease(s) of B, licheniformis (26). The data indicate that the enzyme system consists of a polypeptidase and one or more pep- tidases. It acts on variety of native and denatured proteins with a liberation of high proportion of free amino acids. It appears inactive against synthetic substrates specific for pepsin, trypsin, and chymotrypsin. Proteinase action optimal at pH 7.4 seems slightly inhibited by divalent metal ions and unaffected by cysteine, cyanide, or iodoacetate. The peptidase activity hydrolyses di-, tri-, and tetra peptides starting from the amino end. 27 Bernlohr showed by immunological methods that the protease was synthesized by the cell only after the growth had been completed (10). His attempts to purify the enzyme by conventional methods yielded poor results. Partially purified preparation had high heat stability and an isoelectric point of about 8.3. Bernlohr's data indicate that the enzyme does not contain an essential sulf— hydryl group, has a broad specificity, and does not require metal cofactors. In fact EDTA seems to produce a slight (ca. 5%) enhancement of the activity. These observations agree with those of Damodaran et 31. However, Bernlohr reports broad pH optimum and no inhibitory effect of divalent metal ions (10). Lack of definite pH optimum, the absence of metal requirement, and general heat stability indicate the similarity between this enzyme and the proteases produced by B, subtilis. B. licheniformis protease was classified as subtilopeptidase A by the ratio of esterase to protease activity, amino acid composition, and serological crossbreactions (58). The report by Hall and coworker completes the entire litera- ture on B, licheniformis extracellular proteases (45). These investigators partially separated crude protease into three fractions by chromatography on Sephadex G-200 columns. The first fraction eluting immediately after the void volume has high endoprotease and low esterase activity, is relatively heat stable, and slowly inhibited by DFP. The second peak consists of heat stable amino peptidase activity stimulated by cobalt ions and unaffected by DFP. The third fraction has a low ratio of protease to esterase activity, 28 is rapidly inactivated by heat, and totally inhibited by DFP. Gross proteolysis of the unfractionated system seems to be enhanced by calcium ions and is inhibited by EDTA. This is at variance with results of both Bernlohr and Damodaran et al. Their crude protease preparation shows a broad pH optimum and appears unaffected by reducing agents. In summary, extracellular protease(s) of B, licheniformis have not been extensively purified nor characterized physically. There are significant inconsistencies in the reports of their functional characteristics, inhibitor specificity, and cofactor requirements. Intracellular protease is described in a single report, as an enzyme with a broad pH optimum and an unusually broad specificity (11). The activity resides in at least two protein fractions and remains unaffected by inhibitors. Synthesis of intra- and extra- cellular activities are under catabolite control but do not seem to be coordinated. The rate of protein turnover does not seem correlated with the level of protease. The enzyme(s) is not characterized. C. Peptide antibiotic: Bacitracin B. licheniformis is commercially utilized for the production of the antibiotic Bacitracin. Biosynthesis of this peptide antibiotic is currently vigorously investigated by several groups. Snoke showed that the formation of Bacitracin by washed cell suspensions (114) and by protoplasts (115) of B, licheniformis involves the incorporation of amino acids known to be constituents of the antibiotic. Glucose is reported to stimulate (114, 115) and inhibit (41) Bacitracin biosynthesis. Bernlohr and Novelli (13) showed 29 that Bacitracin production depends on the stage of the cells' life cycle. The antibiotic is released from cells only under cultural conditions that will support spore formation (13). Bacitracin is not elaborated if a shift to spore metabolism does not occur (14). The growth of B, licheniformis is inhibited by Bacitracin and its effect is confined to the early growth phase (116). Bacitracin biosynthesis is not affected by protein synthesis inhibitors chlorampheniol and puromycin (14, 23, 54). This is consistent with the data on biosynthesis of variety of polypeptide antibiotics (86). The role of manganese in biosynthesis of Bacitracin has been sug- gested (136). Manganese is required for induction of sporulation protease. Bacitracin is purified, characterized and commercially available. A large body of early work is summarized in several over- lapping reviews (65, 90, 135). Its physio-chemical prOperties are well known. A family of bacitracins consists of at least ten dodecapep- tides differing in one or two amino acids, ultraviolet absorption maximum or some other physical characteristic. The most abundant and best characterized is Bacitracin A showing of molecular weights of 1500. At slightly alkaline conditions A is slowly transformed into an inactive F form via oxidative deamination of the amino terminal iso- leucine. The primary structure and UV absorption spectrum of Baci- tracin A are known (Chapter III). The majority of its constituent amino acids are hydrophobic. Bacitracin is not susceptible to hydroly- sis by pepsin or trypsin probably due to its cyclic structure. Divalent metal ions, especially zinc, enhance the activity of 30 Bacitracin (2, 134). Zinc and Bacitracin A form a complex with one to one stoichiometry. Ionic zinc appears to be bound by coordinate bonds, two on the histidine residue and two on the thiazoline residue (22, 24). One third of the total amino acids in Bacitracin are in D- form. Kuramitsu and Snoke showed that D amino acids in B, licheniformis are synthesized by racemization and transamination reactions (66). Bacitracin synthesis in cell free systems is currently under investigation. Shimura and coworkers showed that the incorporation of radioactive histidine into bacitracin requires ATP and component amino acids, decreases with the addition of D-amino acids, is optimal at pH 7.2, and unaffected by protein synthesis inhibitors (54). Partial purification of a Bacitracin-synthesizing enzyme system has been recently reported by several groups (33, 53, 87). The detailed description of one of the ig_zi£32_systems is given by Rieder et a1. (93). The nonribosomal synthesis of cyclic peptide antibiotics, gramicidin S and the tyrocidines, as shown by Lipmann's group (61, 95), occurs on enzyme templates. These results have inspired the search for Bacitracin synthetase(s). The synthesis requires ATP and magnesium ions. The individual synthetic steps are not known except that synthesis proceeds via amino acyl adenylates (AMP-amino acid-enzyme complex). Bacitracin synthetase is composed of 7 subunits and has molecular weight of 431,000 (93). Formation of the antibiotic's secondary structure, and the condensation reaction of terminal isoleucine and cysteine residues, are not yet understood. 31 Investigations of the mode of action of Bacitracin parallel the studies of its biosynthesis and rely on the knowledge of its physico-chemical properties. Perturbation of liposomal and planar lipid bilayer membranes (76) as well as alteration of permeability of protoplasts of B, megaterium by Bacitracin (9) indicate its ability to modify membrane properties. Bacitracin is active against gram-positive microorganisms. It inhibits cell wall synthesis by preventing dephosphorylation of the lipid cell subunit carrier and its subsequent re-entry into the reaction cycle (110). The inhibition by Bacitracin of the enzymatic dephosphorylation of CSS-isoprenyl pyrophosphate is due to the formation of a complex between lipid and Bacitracin in the presence of divalent metal ions (119). Binding constants for this interaction correlate with bacteriocidal potency of various Bacitracin peptides. Antibiotic forms a complex with different lipid pyrophoshates (121). M, lysodeikticus is often used in Bacitracin bioassay as a test organism. Binding studies between [3H] Bacitracin and cells and protoplasts of this organism support the hypothesis that the anti- biotic activity is due to the formation of Bacitracin-lipid-P-P complex (122). The toxicity of Bacitracin for animal cells could be due in part to the same mechanism. It inhibits the biosynthesis of squalene and sterols by preparations from rat liver (120). The formation of ubiquinone precursors in rat liver mitochondria and cell-free extracts of Rhodospirillum rubrum is inhibited by Ba fo: bit 3P1 evi ent fun tra1 elal thes fern miss biot for ¢ Prote SPOru BXCep ments prOPOS Vegeta hYPou; betwee 32 Bacitracin. The inhibition appears to be due to polyprenylpyrophos- phate unavailability in synthetic pathways (105, 106). V. Summary Metamorphosis of vegetative cells of bacilli into dormant forms is accompanied by a sequence of time-ordered morphological and biochemical changes. Two early events in this sequence are the appearances of proteolytic enzymes and peptide antibiotics. Current evidence favors the view that these molecules are part of the differ- entiation process. Serine proteases perform sporulation-requiring functions of protein turnover, enzyme inactivation, and post- transcriptional modification. Various peptide antibiotics are elaborated by sporulating cells. Genetic and kinetic data show that these molecules are intimately related to the process of spore formation itself, although evidence of a particular function is still missing. The majority of protease negative mutants are also anti- biotic negative and are asporogenous. Similarly, mutants selected for deficiency in antibiotic production are also asporogenous and proteaseless. The two molecules appear at the same time in the sporulation sequence and with similar kinetics. The existing exceptions to these general statements do not invalidate these state- ments' usefulness as methodological tools. Based on these data, it is proposed that Bacitracin appearance is a result of proteolysis of vegetative cell protein(s) by sporulation-related protease. This hypothesis is tested by investigation of a possible relationship between the extracellular serine protease and the peptide antibiotic Bacit prese 33 Bacitracin in sporulating B, licheniformis cells. This dissertation presents the results of the investigation. CHAPTER II EXTRACELLULAR PROTEASE OF BACILLUS LICHENIFORMIS: PURIFICATION AND PROPERTIES Bacillus species synthesize and excrete several proteolytic enzymes. Genetic and physiological data indicate that seryl- protease(s) are related to sporulation (70, 103, 128). The precise function or functions of this enzyme(s) are unknown and have been the subject of widespread inquiry (29, 53). Our investigation of extracellular protease(s) function in sporulating Bacillus licheni— formis was directed towards its possible role in antibiotic formation and was initially hampered by difficulties in protease purification. Functional characterization of the enzyme(s) performed on crude or partially purified preparations has in the past led to discrepancies between observations by various investigators (10, 26, 45). Therefore, the effort was made to obtain electrophoretically and chromato- graphically pure enzyme(s). In B, licheniformis the peptide anti- biotic is firmly bound to crude protease(s). Upon dissociation of this complex the purification of extracellular protease(s) was effected and certain properties of three enzymes obtained were examined. 34 it 1i (N volL incu was 1 the a react Exten teaSe 340 nm 35 Materials and Methods Organism and Growth Conditions B. licheniformis, strain A-S from the Oxford University col- lection, was obtained from R. W. Bernlohr. Cells were grown in a peptone-yeast extract sporulation medium described by Bernlohr and Clark (11) at 37C with vigorous aeration. The medium was made up in 10'2 M phosphate buffer pH 7.4 termed "standard buffer." Three liter cultures for protease production were grown in a fermentor (New Brunswick Scientific Co.) for 27 hours to an optical density of 4.5 at 620 nm. Enzyme Assay The protease assay employed was a measure of the enzyme's ability to render azoalbumin (Sigma Chemical Co.) acid soluble. A 1.5 ml reaction mixture in a 12 ml conical centrifuge tube contained: 1 ml of 5% (w/v) azoalbumin solution in standard buffer, an appropriate volume of protease solution, and buffer to 1.5 ml. The reaction was incubated at 37 C for 20 min and terminated by the addition of 2 ml of 10% (w/v) trichloroacetic acid (TCA). The precipitated protein was removed by centrifugation for 15 min in a clinical centrifuge, and the absorbance of the supernatant fraction was read at 340 nm. The reaction conditions were such, enzyme level below 50 ug/ml, that the extent of the reaction was linearly dependent on the amount of pro- tease added (10). Activities are expressed in absorbance units at 340 nm of the soluble peptides. 36 Specific Protease Inhibitors 10.1 M stock solutions of ethylenediaminetetraacetic acid (EDTA, Eastman Organic Chemicals) in standard buffer and pheny- methanesulfonyl fluoride (PMSF, Sigma Chemical Co.) in 100% ethanol were stored at 4°C. Inhibitors were added to the assay mixture at the time of enzyme addition to a final concentration of 5 x 10-3 M. Protease and commercial Bacitracin (Mann Research Laboratories) were incubated for 30 min at 37 C prior to the addition of substrate. The extent of inhibition is expressed as a percent of total activity observed in the absence of inhibitors. Antibiotic Bioassay The routine Bacitracin assay utilized paper discs impregnated with the antibiotic and overlayed on a lawn of Micrococcus lysodeikticus (8). The standard curve was established utilizing known amounts of commercial Bacitracin. The activity of the commercial product was stated to be 60 units/ml equivalent to 90% purity. Bacitracin was extracted from various preparations with 75% ethanol at room tempera- ture for 30 min. Precipitated protein was removed by centrifugation and the supernatant concentrated by drying in a vacuum evaporator (Buchler Instruments). The material was resuspended in distilled water and redryed to remove ethanol, resuspended again in an appropriate volume of distilled water, and 0.1 ml loaded on a test disc. Kinetics of Protease Appearance 100 m1 samples of growing cultures were removed at hourly intervals and subjected to 12000 xg for 20 min in a centrifuge (5 er! bu: inc int Pro pre Cip: penc buff at- tain with the ; acid Separ Chlor t0 0C1 tease, Chemic Those dialyz. (WhamE the Sta 37 (Sorvall Model RC2-B). The culture supernatant was a source of crude extracellular protease. The cells were washed once with standard buffer, resuspended in the same buffer containing 1 mg/ml lysozyme, and incubated at 30° for 30 min. The cell lysate was the source of crude intracellular protease. Protease Purification Supernatant media from 27 h cultures were concentrated by precipitation with ammonium sulfate at 90% of saturation. The pre— cipitate was collected by centrifugation at 23000 xg for 30 min, resus— pended in 10.1 M phosphate buffer pH 7.4, and dialyzed against standard buffer overnight. This fraction, termed "crude protease," was stored at -20C and used for further purification. This preparation con- tained measurable levels of Bacitracin which could have interfered with studies of the protease(s). Unsuccessful efforts made to remove the antibiotic included precipitation by ZnCl NaCl (3) salicilic 2, acid (111) and ammonium sulfate (65), and extraction with butanol (3). Separation was achieved by extracting concentrated protease(s) with chloroform in the proportion 1:1 (v/v) and permitting phase separation to occur. The top aqueous layer containing the CHCl -treated pro- 3 tease, was loaded onto a coarse grade Sephadex G-25 (Pharmacia Fine Chemicals) column, 2.5 x 40 cm, and eluted with the standard buffer. Those fractions containing active enzyme(s) were then concentrated, dialyzed, and placed on a diethylaminoethyl (DEAE) cellulose (Whatman DE—52) column, 1.5 x 18 cm, which had been equilibrated with the standard buffer. A gradient of 0 - 0.5 NaCl was utilized to remove all the protein from the column. The protein was put onto 38 columns of carboxymethyl cellulose (CMC, Whatman CM-52) with bed dimensions 1.6 x 27 cm. A linear gradient of phosphate buffer pH 6.0, 5 x 10-3 M to 10.1 M was employed to elute the protease(s) at a flow rate of 50 ml/hr. 5 ml fractions were collected and their absorbance at 280 nm was monitored and recorded (ISCO absorbance monitor model DA 2). The enzyme was concentrated at different stages of purifi- cation by lyophilization, Lyphogel (Gelman Instrument Co.) or dialysis against Ficoll (Pharmacia Fine Chemicals) solutions. All manipulations were performed at 4°C. Protein concentrations were determined by the method of Lowry et al. (75). Results The range and standard deviation of Bacitracin bioassay are presented in Figure l. The standard curve resulting from five separate experiments, was used for quantitative estimation of the antibiotic concentration. The growth response, total intracellular and extracellular protease(s) activities of B, licheniformis A-S are depicted in Figure 2. Intracellular protease(s) purification was not undertaken due to the low levels of this enzyme(s). The total proteolytic activity of crude enzyme preparations eluted from Sephadex G-25 or DEAE cellulose columns appeared in a single peak which coincided with the major protein peak. The protease(s) was not adsorbed by DEAE cellulose under a variety of conditions of pH and ionic strength. However, chromatography of the same preparation on CMC columns yielded three peaks of activity. Two eluted immediately at low ionic strength and the third at the end of the gradient after extensive washing of the column with 0.1 M phosphate. All fractions 39 IOOT 3 9.- , / so 4// 50 4o —- =~ BACI TRACIN (F9) 20— ‘ ,0! lllllllllllllll ' 1213141516171819 20 '21 22 23 24 25 26 27 28 29 CNAMETER mmfl Fig. 1.--The range and standard deviation of the antibiotic bioassay. 0D620 4O ,-¢%———4 I J l J L I I :2 I E::_W g 2 H 5 8 10 12 24 TIME (HRS) 1.5 Fig. 2.--Growth and total extracellular (--o—-) and intracellular (——o——) protease activities for B. licheniformis A-S. PROTEASE ACTIVITY (UNITS/ml) 41 containing protease were also positive in the Bacitracin bioassay. In all cases the amount of Bacitracin was directly proportional to the measured proteolytic activity. Under no circumstances did any chromotographic steps yield Bacitracin-free protease. Approximately 1-10% of the total Bacitracin produced by the cells remained "bound" to crude protease. Precipitation of the antibiotic by various reagents or its extraction by butanol were equally unsuccessful. Bacitracin is known to form complexes with isoprenyl pyrophosphates involving divalent metal ions (121). The addition of EDTA to growing cultures, crude protease(s) or concentrated enzyme preparations did not signifi- cantly affect the amount of Bacitracin associated with protease. However, after extraction of crude protease(s) three times with chloroform, the antibiotic and protease(s) were easily separated by filtration on Sephadex G-25. Chloroform treatment produced an increase in the specific activity of the antibiotic (pg of Bacitracin/mg protein) and a decrease in total protease(s) activity. The specific activity of the crude protease was unaffected. Bacitracin, which is insoluble in chloroform, remained with protease in the aqueous layer but was physically separated from it by gel filtration. This Bacitracin-free protease(s) was then purified on CMC columns which resolved three distinct activities (Figure 3). The protocol for the purification of the three proteases presented in Table I indicates that complete recovery of protease was obtained and with an overall 44-fold purification of fraction III. Peak activity fractions of CMC I, II, and 111 had specific activities of 1.30, 2.74, and 62.0 units/mg respectively. Polyacrylamide slab gel electrophoresis in buffers 42 .cssHoo omoHSHHoo Hacuoa xxonueo a :o ommououm ooHMucfiomeflomm mo ofifimonm :ofiusHo ocbuu.m .wfim """" BSVBIOHd 3 9 1 3.2V. m2340> 20;: Am on" com amp 00— on o D _ H (Mllllllllll IIHIIl I I . ’o x I .oo \ x \o 6.. H: ll no.0 43 Table I. Protease purification protocol. Step Specific Activity Purification Recovery (units/mg) Ratio (%) Cell supernatant 0.89 1.00 100 Crude protease 1.65 1.86 33.4 CHCl3 treatment 1.80 1.09 22.0 Sephadex G—25 filtration 2.10 1.17 31.2 CMC fractions 1 a II 2.87 1.37 12.5 CMC fraction III 38.9 44.2 159 containing sodium dodecyl sulfate indicated that each of the three proteases consists of one major species with an occasional faint slowly migrating band. The molecular weight of the major band was estimated to be 30,000 Daltons in each case. Bovine serum albumin (68,000), catalase (60,000), gamma-globulin heavy (50,000) and light (25,000) chains, lactate dehydrogenase (36,000), and ovalbumin (43,000) were used as standards. 5 x 10-0 M PMSF caused 80% inhibition of proteases in cell lysate and culture supernatant (crude enzymes) isolated during the growth of B, licheniformis A-5. 5 x 10"3 M EDTA consistently enhanced the activities in all these fractions by 5-10%. ZnCl2 inhibited extracellular proteases when incubated with them for a 16 h period. The extent of inhibition was proportional to the concentration of zinc. The results of inhibition studies on purified extracellular proteases are presented in Table II. Bacitracin-free protease was 44 Table II.--The inhibition of B, licheniformis extracellular proteases. Percent Inhibition Protease Preparation PMSF EDTA Commerc1al 5 x 10-3 M 5 x 10-3 M Bac1trac1n 5 x 10'6 M Bacitracin-free protease 93 52 25 CMC fractions I and II 89 48 0 CMC fraction III 100 52 6 inhibited significantly by commercial antibiotic. PMSF inhibited 90 to 100 percent activity of all proteases. Discussion The extracellular proteases of B, licheniformis A-S appear to be seryl enzymes by virtue of their inhibition by PMSF. These results differ from those obtained by Hall et al. who reported that only two of the three extracellular proteases of B. licheniformis 11560 were inhibited by organofluorides (45). EDTA is reported to enhance (10) and inhibit (45) various preparations of extracellular protease(s) of B, licheniformis. Further data is required to estab- lish the exact nature of this relationship. Extracellular proteolytic activity seems to reside in three molecular species of similar size, different charge at pH 6, and different specific activities. The isoelectric point of crude pro- tease, pH 8.3, reported by Bernlohr (10) is consistent with the results reported in this paper. .13” ‘ 1 L, 45 Bacitracin and B. licheniformis extracellular protease(s) form a relatively tight complex which is resistant to various treat- ments employed in antibiotic extraction and purification. The exact nature of the complex is unknown but the fact that it can be dis- sociated with CHCl suggests at least three possibilities. The 3 complex may actually involve a lipid which is extracted by the CHCl3 treatment. Alternatively the complex may involve hydrophobic and polar interactions between protease(s) and Bacitracin which are modified when the aqueous phase is saturated with CHCl The com- 3. plexation seems to stabilize the active conformation of the peptide antibiotic. The Bacitracin-protease complex could involve a metal ion, such as zinc, acting as a bridge between the two components similar to its complex with C -isoprenoid pyrophosphate (119). 55 Bacitracin's binding of zinc (22, 24) enhances its activity (134). The preincubation of Bacitracin-free protease with commercial Bacitracin had an inhibitory effect on proteolytic activity. It has been reported that Bacitracin inhibited subtilisin, papain, and leucine aminopeptidase noncompetitively (77). The highly purified pro- teases obtained from CMC chromatography appear to be much less subject to inhibition than the Bacitracin-free (cruder) enzyme. This anomaly might be due to the lack in the test system of divalent metal ions which would be required for Bacitracin-protease complex formation and the resulting enzyme inhibition. Specific protease inhibitors are known to occur in plant and animal systems (25). This is a first report of such an inhibitor in a Bacillus species. CHAPTER III PRODUCTION OF BACITRACIN lN_VITRO BY PROTEOLYSIS OF VEGETATIVE BACILLUS LICHENIFORMIS CELL PROTEIN Upon the initiation of sporulation in Bacillus species, cells elaborate several proteolytic enzymes and peptide antibiotics. Certain temporal and genetic relationships exist between the sporulation protease and peptide antibiotic production by sporulating cells. The biochemical nature of the pleiotropic relationships in the expression of these functions during sporulation is unexplained (29, 53, 103). This prompted our investigation of the possibility that certain anti- biotics might be products of proteolysis of normal vegetative cell constituents. Sadoff and Celikkol found in Bacillus cereus that a Bacitracin-like peptide could be derived from the proteolysis of exponential cell extracts with sporulation protease (99). We have undertaken a study of these phenomena in Bacillus licheniformis A-S, the organism used in commercial production of Bacitracin. The data reported here support the above finding and suggest that Bacitracin elaboration by sporulating cells may be due to the activation of an antibiotic precursor, a constituent of vegetative cell proteins, by a specific, sporulation-related seryl—protease. 46 47 Materials and Methods Organism and Growth Bacillus licheniformis A-S used in this study was obtained from R. W. Bernlohr. The growth of the organism, protease purifi- cation and assay, and antibiotic bioassay have been described (Chapter II). The minimal medium of Bernlohr and Novelli (14) with ammonium lactate replaced by ammonium sulfate, was used for the pro- duction of radioactive Bacitracin. The medium was formulated in 10’2 M phosphate buffer pH 7.4, termed "standard buffer." 100 ml of culture containing l4[C]-D,L-ornithine-2-l4[C]HC1 (21.7 uC/mg, New England Nuclear), was grown for 52 hrs at 37 C with vigorous aeration. The cells were spun at 12000 x g for 20 min and the supernatant medium diluted four times with nonradioactive cell supernatant of a parallel culture. This preparation constituted the source of radioactive Bacitracin. Radioactive protein was obtained from vegetative cells grown in modified G medium (48). After 4 hrs of growth, the 300 ml culture containing 7.5 DC of l4[C]-ornithine was diluted 10 times with a nonradioactive parallel culture. Cells were harvested, washed twice in 10'1 M standard buffer and used as a source of pro- tein. Nonradioactive substrate was obtained from cells grown in nutrient broth for 3 hrs at usual conditions (Chapter 11). Preparation of the Substrate Protein was extracted by a modification of a procedure for cell wall preparations. Cells were suspended in 10% trichloroacetic acid (TCA) at 90°C for 30 min to hydrolyze nucleic acids, washed 48 twice to remove TCA, and the protein solubilized at pH 9.5. The solution was then adjusted to pH 7.4 and the insoluble materials removed by centrifugation. Antibiotic Production in vitro Antibiotic was routinely produced by the addition of protease to vegetative cell protein suspension in standard buffer at 37°C for 40 min. Putative antibiotic was immediately extracted and assayed as described previously (Chapter II). Phenylmethanesulfonyl fluoride (PMSF) (Sigma Co.) and ethylenediamintetraacetate (EDTA) (Eastman Organic Chemicals) at 5 x 10-3 M concentration were used as protease inhibitors. Paper Chromatography The identification of Bacitracin and putative antibiotic was carried out by paper chromatography according to the procedure of Snell et al. (113). The following solvent systems were used: (a) t—butanol-HZO (80:20) with ammonia in the atmosphere, (b) t-butanol acetic acid-H20 (60:6:34), and (c) acetone-H20 (70:30) with ammonia in the atmosphere. Thin-Layer Chromatography Antibiotics were applied in 5-30 ul samples onto thin-layer plates (Silica Gel, 250 um; Brinkman Instruments, Inc.) and developed with a solvent mixtures (A) t-butanol-HZO (80:20), (B) ethanol-H20 (80:20), or (C) n-butanol-acetic acid-H20 (4:1:2). Commercial Bacitracin radioactive Bacitracin, and samples of the putative antibiotic were always chromatographed together when separation was performed in one 49 dimension. Two dimensional chromatograms, containing one sample per plate, were developed with solvent A followed by solvent C. Lyo- philized culture supernatant containing radioactive Bacitracin, was treated three times with chloroform (1:1, v/v) to release antibiotic from protein complexes (Chapter II). The water insoluble material generated by this treatment was removed by centrifugation. Crude and commercial Bacitracin, and putative antibiotic were further extracted with ethanol (Chapter 11), before being chromatographed. The developed chromatograms were sprayed with 0.1% ninhydrin reagent in acetone to visualize resolved ethanol-soluble peptides. Bio- autograms were made utilizing 20 x 28 cm Pyrex glass pans overlayed with 100 ml nutrient agar inoculated with 10 m1 exponential nutrient broth culture of Mycrococcus lysodeikticus. It was important to remove the solvents completely by evaporation to dryness, because they also inhibited bacterial growth. The chromatograms were layed on solidified agar for 15 min. Zones of inhibition were easily visualized after overnight incubation of the plates at 30°. Nin- hydrin did not inactivate Bacitracin so that the same chromatograms were used for peptide and antibiotic detection. Radioautograms were obtained by measuring the radioactivity retained by silica gel comprising various peptide spots. Duplicate plates were prepared in every experiment. Scintillation Counting 10 ml of Handiflour scintillation fluid (Merck Co.) were added to glass vials containing samples and radioactivity was assayed with a model 3320 Packard Tri-Carb scintillation spectrometer. 50 Samples were either suspended in 100 pl of standard buffer or con- sisted of silica gel powder carefully scraped from chromatograms. Spectroscopy» Observations were made of the UV absorption spectra of commercial and radioactive Bacitracin and antibiotic produced ip_ zippp, The samples were prepared as described above and resuspended in 2 x 10.1 M phosphate buffer pH 4.0. The spectra were monitored at room temperature and 37°C by double beam scanning spectrophotometer (Perkin Elmer Model 124) and recorded by Sargent recorder (Model SR). The second spectrum of each sample was taken after the pH was adjusted to 10. The kinetics of ip_xi££p Bacitracin production were followed by monitoring the increase in the absorbance at 253 nm during pro- teolysis of cell protein. The reaction mixture contained 4.5 mg of protein and 0.27 units of protease CMC fraction III in 0.33 ml of standard buffer. The reaction was carried out at 37 C in 1 mm light path cuvettes. The blank consisted of substrate only. Protein concentration was determined either by Folin phenol reagent (75) or by Biuret reaction (56). Results Growth of B, licheniformis A-5 and the kinetics of intra- and extracellular Bacitracin appearance are presented in Figure l. Ip_vitro antibiotic release was initially demonstrated by use of unpurified intra- and extracellular proteases containing high levels of Bacitracin. The incubation of crude protease at 37 C for 52 h ‘1 :d 0 D620 2:» oo 51 TIME (HRS) Fig. 1.—-Growth of B. licheniformis A-5 and kinetics of intracellular (closed symbols) and extracellular (open symbols) Bacitracin appearance. BACITRACIN (us/m1) 52 did not affect its antibiotic content. In our efforts to demonstrate that an antibiotic could be obtained by cleavage of B, licheniformis protein, it was essential that neither the protein extract nor the protease contain Bacitracin. Extracellular protease was purified (Chapter II) and tested for ability to release antibiotic. The results are presented in Table I. It is significant that of the three extracellular proteases only CMC fraction III effected antibiotic release. In another experiment 2.7 units of protease (CMC fraction III) were incubated with increasing amounts of cell protein for 60 min at 37°C in a volume of 1.8 m1. Products of proteolysis were extracted and assayed for antibiotic. The results are depicted in Figure 2. The quantity of Bacitracin released is a linear function of the amount of substrate present. In this particular study, the equivalent of 100 pg of antibiotic was produced from 25 mg of cell protein. This value, 4 ug of antibiotic/mg of protein, is considerably less than the maximal yield of approximately 75 ug of Bacitracin/mg Table I.--Antibiotic production ip_vitro by different protease prepara— tions. Bacitracin (ug) Protease Preparation Reaction Controla crude protease 120 80 Bacitracin-free protease 29 0 CMC fractions I and II 0 0 CMC fraction III 38 0 a . . . . . . Control levels of antibiotic are those which were contained in either the protease or substrate preparations. S3 100 _. (pg) I 50.— BACITRACIN 25... I I I J I 5 IO 15 20 25' PROTEIN EXTRACT (m 9) Fig. 2.--Bacitracin production as a function of substrate concentration. S4 of cell protein produced in the course of sporulation of B, licheni- formis A-S. The peculiar feature of Bacitracin structure (Figure 3), A-Z thiazoline ring resulting from the condensation of the carboxyl group of the isoleucine with the amino and thiol groups of cysteine, is believed responsible for the antibiotic's absorption bond at 253 nm (1). The kinetics of ip_!i££p_antibiotic production were followed by monitoring the increase in the absorbance at 253 nm during pro- teolysis of cell protein (Figure 4). 5 x 10-3 M PMSF almost com: pletely inhibited the reaction while the same concentration of EDTA had no inhibitory effect. Experiments utilizing spectroscopic and chromatographic techniques resulted in the identification of putative antibiotic with Bacitracin. The UV spectra of the antibiotic pro- duced ip-zippp_taken at pH 4 and 10, are presented in Figure 5. Radioactive Bacitracin yielded the identical spectra while commercial Bacitracin's spectrum reflected the presence of large amounts of inactive Bacitracin F (24). The preliminary results of paper chroma- tography of antibiotics indicated that the three samples migrated the same distances in three different solvent systems. An improve- ment in experimental technique was needed, however, due to some dis- crepancies in Rf values caused by large zones of inhibition in bio- autograms. Thin-layer chromatography on silica plates in three solvent systems yielded similar Rf values for the three antibiotic preparations (Table II). In one such experiment 5 ul of radioactive Bacitracin (112 cpm/ug) and antibiotic produced by proteolysis of radioactive vegetative cell protein (99.0 cpm/ug) were chromato- graphed in two solvents. After development, chromatograms were 55 N'Hz CH3 D-ASP | I CH2 LeHis +L—Asp L-IIe I / \{Y- e linkage ‘3?Ef‘<3+4 [)‘FDFHZ I /L-Igs HC NHz L-IIQ, <——D-Om A V L—lle / I Cf/N\ 9 I 1" /CH-C-'>-L-Icu->D —Glu s-CH2 I L-Cgs Fig. 3.--Primary and secondary structures of Bacitracin A. 56 3nm 0,3 —- 0,2 0,1 ’ ABSORBANCE AT 25 l I I I I 10 ' 2‘0 30 40 50 60 TIM E (M IN) Fig. 4.--Kinetics of Hi vitro Bacitracin production (x). The reaction was inhibited by S x 10‘3 M PMSF (—--) S7 .60IIIIIIIII \ w __ \\~_pH IO — \ \ \ .50 — " 0 o 5 18 I40 " \ 0 V) n \ < .30 -— PH4 \ " ¥ \ - I I I I I I I I I . 230 zoo 290 320 Wave length ‘ nm Fig. S.--Spectra of the antibiotic produced _i_n_ vitro. 58 Table II.--The results of chromatographic identification of ip_vitro antibiotic. Rf Values Chro;:::g;aphy goizzgt Commercial Radioactive In Vitro o y Bacitracin Bacitracin Antibiotic a .41 --- .50 Paper ~ c .71 .76 .79 A .21 .19 .20 Thin-Layer B .39 .38 .42 assayed for peptides, bacteriocidal activity, and total radioactivity present in each spot. These data presented in Figure 6 indicate that: (i) biologically active peptides have identical Rf values, (ii) the number of peptides resolved in two systems and their respective Rf values are similar, and (iii) the distribution of l4[C]- ornithine among the peptides in each sample is very similar. Thin- layer chromatography of three antibiotic preparations in two dimen- sions yielded overlapping bioautograms. The incorporation of the label into the vegetative and sporulating cells was determined by measuring the loss of radioactivity in supernatants of cultures grown for 4 and 52 h respectively. Vegetative cells incorporated 12.5% l4[C]-ornithine after 4 hrs of growth while 16.7% of the label remained in cells after 52 hrs of incubation under the same con- ditions. 59 SOLVENT SYSTEM: ETHANOL:H20 (80:20) SOLVENT FRONT III) 0 297 O 39 O 173 @ 130 g 321 “59 3370 106 O BACITRACIN ANTIBIOTIC t-BUTANOL:H20 (80:20) 58 (3 66 () 320 O O 231 .e M 723. 0 BACITRACIN ANTIBIOTIC. Fig. 6.--Radio-bioautogram of Bacitracin produced by sporulating B. licheniformis cells cromatographed with ip_vitro antibiotic in two solvent systems. Shaded areas represent bacteriocidal activity. 60 Discussion A peptide must exist in vegetative cell proteins of B, licheniformis which can be cleaved by a unique serine protease to yield an antibiotic similar to Bacitracin in its chromatographic and spectral properties. The amount of antibiotic produced is directly proportional to the amount of protein in the reaction system. The unique UV absorbing component of Bacitracin is the thiazoline ring. The loss of spectrum accompanied by inactivation of the antibiotic occurs under alkaline condition. We note that a structure with similar spectral properties arises during the course of proteolysis and, since A-2 thiazolines would be unstable at conditions employed for preparation of the protein extract, this structure must result from the action Of protease. A reaction sequence during proteolysis which could lead to the formation of the thiazoline ring and its possible mechanism was previously described (99). Ethanol-soluble peptides produced by ip_!i££p_hydrolysis of vegetative cell proteins are similar in number, quantity, and Rf values to the same produced by sporulating cells. This is presumably due to the same mechanism Operating ip_!i!p_and ip_!i££p, Comparable amounts of l4[C]- ornithine found in these peptides are taken up by vegetative and sporulating cells. Bernlohr and Novelli investigated incorporation of radioactive amino acids into Bacitracin with respect to the age or state of differentiation of B, licheniformis in order "to prove conclusively that the antibiotic is being synthesized by the 'dif— ferentiated' cell and not simply being released at this time" (14, p. 99). The cells were exposed to 3[H]-D,L-Ornithine for one hour at ()1 different times in the life-cycle, lOO-fold excess of nonradioactive ornithine was added, and the radioactivity incorporated in Bacitracin after 24 h, was assayed. The amount of the label incorporated into Bacitracin [A] normalized by the number of cells (as reflected in Klett units; their data Fig. 3, p. 103) [B] was plotted versus the time period (Figure 7). The regression line through the data points indicates that the incorporation of 3[H]-ornithine into Bacitracin is independent of the stage of growth. The results of this analysis contradict their conclusion. The basis of their error lies in their failure to normalize 3[H]-ornithine uptake with cell concentration. The sequences of morphological developments in the sporu- lation of Bacillus and Clostridium species are almost identical (98). Sporulating Clostridia elaborate toxins in the manner very similar to antibiotic production by Bacilli (103). There are data indicating that an increase in toxicity is due to activation of toxin precursor by proteases and not to Bg_pplp_toxin synthesis (16, 27, 30, 71). Recent studies of Edeine biosynthesis lend further support to the idea that antibiotics may arise by activation of specific precursors. Edeine B was found to be bound covalently to the edeine synthetase. The edeine-polyenzymes complex associated with the DNA-membrane complex exhibited no antimicrobial activity, unless the complex was treated with alkali (67). It is possible that activation of edeine ip_xi!p. occurs by proteolytic action on the edeine-polyenzymes complex. The preincubation of the purified protease with commercial Bacitracin inhibited only the enzyme capable of ip_!i££p_Bacitracin production and left the other two unaffected. Bacitracin was found 62 4 — 3 _ o o LX103 n o o r O l _ ° 0 0 I _I I I I I I o 1 2 3 4 5 6 7 PERIOD Fig. 7.--The incorporation of [3]H-ornithine into Bacitracin at dif- ferent growth stages Of B, licheniformis (data from reference 14). 63 to form ip_xixp_tight complex with protease (Chapter 11). These results suggest that Bacitracin may exert a product inhibition-type control on the protease by actually forming a complex with it. The function Of proteases and antibiotics in the sporulation process of Bacilli and a relationship between them (if any), has been a subject of active investigation and debate in the past. However, these questions still remain unanswered. The results reported here present evidence, to our knowledge for the first time, that there exists a direct functional relationship between these two sporulation- related molecules. CHAPTER IV THE ASSOCIATION OF THE BACITRACIN PRECURSOR WITH RIBOSOMES OF VEGETATIVE BACILLUS LICHENIFORMIS CELLS Sporulating cells of Bacillus licheniformis produce the pep- tide antibiotic Bacitracin and several serine proteases. Bacitracin was found to inhibit the extracellular protease complex by binding with it. A peptide antibiotic, derived by ip_!i££p_proteolysis of exponential cell extracts with a specific protease, was shown to be identical to Bacitracin in its spectral and chromatographic properties (Chapters 11 and III). Studies of the same phenomena in Bacillus cereus yielded a Bacitracin-like peptide under similar conditions and further suggested that the protein(s) containing the antibiotic peptide was associated with the SOS ribosomal subunit of vegetative cells. The yield of antibiotic was proportional to the weight of ribosomes proteolysed (99). In B, licheniformis A-S, a derepressed antibiotic producing mutant, the data obtained support these findings and indicate that the Bacitracin precursor(s) is similar to the initiation factor proteins. 64 III I! OI I 65 Materials and Methods Orgapjim and growth Conditions B. licheniformis A—S, obtained from R. W. Bernlohr, was grown “ in modified G medium (48) or nutrient broth, as previously described (Chapter II and III). Sporulating or vegetative cells were harvested from 3 1 cultures by centrifugation at 12,000 xg for 20 min when the turbidity, measured by optical density of the medium at 620 nm, reached ca. 0.8. They were then washed with the appropriate buffer and stored at -20°C. Preparation of Ribosomes, Ribosomal Subunits, and the Fractionation of the Cell Lysate Ribosomes and ribosomal subunits were prepared by modified procedures of Bonamy et a1. (15) [Method 1] or Goldthwaite and Smith (37) [Method 11]. All Operations were performed at 0-4°C. Method 1: Frozen cells were suspended in 10'2 M tris-2 x 10"5 M magnesium acetate-5 x 10-3 M 2-mercaptoethanol buffer, pH 7.2 (TMS buffer), containing lysozyme (1 mg/ml) and DNAse (0.1 mg/ml). The mixture was incubated at 30 C for 60 min. The cells lysed and debris was removed from the extract by centrifugation at 16,000 rpm for 30 min. Supernatant liquid was spun at 41,500 rpm for 2 h in a pre- parative ultracentrifuge (Beckman, Model L3-50) and the ribosomal pellet obtained was resuspended in TMS buffer and dialyzed for 3 h against three changes of the same buffer. To complete dissociation into ribosomal subunits, the preparation was then stirred overnight in TMS buffer. The extract was layered on linear sucrose density 66 gradients (5% to 25% w/v) prepared in 10.2 M tris-10'3 M magnesium acetate buffer pH 7.2 (TM3 buffer). Preparative (38 ml) or analytical (5 ml) gradients were spun at 27,000 rpm for 4.5 h in Spinco rotors SW 27 or SW 50.1, respectively. Ribosomal particles were recovered with a density gradient fractionator (ISCO, Model 640) which measured and recorded their absorbance at 260 nm. Up to 100 ug of ribosomes in 50 ul were loaded on each analytical gradient. 0.3 m1 fractions were collected at the flow rate of 1 ml/min. Sucrose concentration was determined on control gradients, by measuring the refractive index of the fractions with a precision refractometer. Appropriate fractions were pooled, the sucrose dialyzed away, and the ribosomes reconcentrated by centrifugation at 27,000 rpm for 9 h (in a SW 27 rotor) or at 41,500 rpm for 2.5 h (in a Ti 50 rotor). Purified and separated ribosomal particles (705, 505, and 308) were resuspended in TM3 buffer and stored at -20°C. Their purity was tested in analytical sucrose density gradients under the conditions described above. Method 11: Frozen cells were thawed in three volumes of 3 x 10'2 M tris (pH 7.5), 10'2 M magnesium acetate, 5 x 10"2 M KCl, 3 x 10-3 M 2-mercoptoethanol (buffer A) and disrupted in a French pressure cell at 20,000 1b/in2. The extract was spun at 12,000 xg for 10 min and DNAse (20 ug/ml) was added to the supernatant fluid which was then centrifuged at 20,000 xg for 30 min. The pellet was discarded and the supernatant fluid spun at 40,000 rpm for 3 hrs. The top two-thirds of the supernatant ("S-100 fraction") and the bottom third ("membrane fragments fraction") were separated and 67 stored at -20°C. The ribosomal pellet was either resuspended in buffer A and stored at -20°C or subjected to a high salt (1 M NH4C1) wash. The ribosomal pellet was resuspended in a buffer containing 1.0 M NH4C1; 5 x 10.2 M tris, pH 7.5; 10.2 M magnesium acetate; and 6 x 10'3 M 2-mercoptoethanol (Buffer B), left overnight and pelleted again in the ultracentrifuge. The pellet was resuspended in buffer A, spun at 3,000 xg for 10 min, and the supernatant fluid containing "washed" ribosomal particles was stored. The supernatant fluid pro- duced by ultracentrifugation, termed "NH4C1 wash," was dialyzed against 10"2 M phosphate buffer pH 7.4 and stored at -20°C. The order of washing ribosomes, either before or after separation had no effect on the distribution of the Bacitracin precursor. The ribo- somal pellet was resuspended in a buffer containing 10.2 M tris, 3 1 M MgClz, and 10' M NH4C1 buffer, pH 7.4 (buffer C), and dialyzed against it for 40 h (123). The subunits were separated pH 7.4, 10- on sucrose density gradients (5-25% w/v) made up in buffer C as described above. Ribosomal particles were thus prepared by two pro- cedures, with and without a high salt wash. The concentration of ribosomes was expressed in absorbance units at 260 nm, were each absorbance unit was equal to 67 ug of ribosomes/ml (129). Protein concentration was measured by the method of Lowry et al. (75). Protease purification and assay, antibiotic bioassay, identification of the product, and the ip_!i££p_reaction conditions have been described (Chapters 11 and III). Purified extracellular B, licheniformis protease, CMC fraction III, was used in all experiments. 68 Fractionation of NHACl Wash _" 1.5 ml of thawed sample was mixed with 75% ethanol for 30 min at room temperature. The protein precipitate which formed was pelleted by centrifugation at 20,000 xg for 10 min and resuspended in 0.75 ml of 10"2 M phosphate buffer, pH 7.4. The supernatant was evaporated to dryness, washed with distilled water to remove alcohol, and resuspended as above. 0.25 ml of the two fractions were combined yielding 0.5 ml of "reconstituted" NH4C1 wash. The three prepara- tions, the supernatant, the precipitate, and the reconstituted NH Cl wash, were proteolysed under the standard conditions (Chapter 4 III). Results The separation of purified ribosomal particles was hampered by their resistance to dissociation. In Figure 1, the absorbance at 260 nm (on 0 to 0.5 scale) was plotted versus fraction number for five analytical gradients (A to E). Profile A describes typical distribution of 308, SOS, 708, and 1008 particles produced by Method 1. Three unresolved peaks, which were pooled and subjected again to the dissociation and separation procedures, yielded profile B indicating that the disruption of the 1005 complex has been effected. However, subunits resulting from this cumbersome procedure were only about 80% pure. The dissociation of ribosomes by the methods of Tissieres et al. (129) was equally unsuccessful. The incubation of ribosomes at 45°C for 4.5 h resulted in extensive unfolding of all, except 30$, particles [C]. 708 ribosomes appeared most susceptible to heat denaturation. Dialysis against 10"2 M phosphate buffer 69 303 505 705 1005 B C In ”\“J 2 sucrose 15 Fig. l.--Sucrose density—gradient sedimentation of Bacillus licheni- formis ribosomes. 70 containing 10-3 M magnesium acetate had little effect in dissociating 705 and 1005 complexes [D]. 97% pure 505 and 305 ribosomal particles were obtained by Method 11 [E]. Purified unwashed ribosomes isolated from B, licheniformis cells sporulating in G medium had Bacitracin peptides "bound" to them. Only a slight increase (if any) in the extractable antibiotic was observed upon their proteolysis. In contrast, all purified vegetative cell "Bacitracin-free" ribosomal particles, yielded high levels of the antibiotic upon incubation with sporulation protease CMC fraction III. In one experimental series, 1.8 ml containing 22 units of various ribosomal preparations and 2.8 units of protease were incubated at 37 C for 40 min. The amounts of Bacitracin pro- duced in that time are presented in Table I. The data indicate that the NH4C1 wash removed 42% and 26% of the Bacitracin precursor(s) from 708 and 505 subunits respectively. The dissociation Of 708 ribosomes into subunits did not seem to promote antibiotic activa- tion. NH Cl-washed 308 and 508 subunits yielded equal amounts Of 4 Table I.--Bacitracin yield (mg) from various ribosomal substrates. S b tr t Unwashed Washed “ 5 a 6 (mg) (mg) 705 84 49 303 + 503 -- 53 505 38 28 308 -- 28 71 antibiotic, each about half of the quantity produced from either 708 or 30S + 508 substrates. In our efforts to localize the Bacitracin precursor(s), we subjected to proteolysis fractions of a vegetative cell lysate including the S-100 fraction, membrane fragments, NH4Cl-washed ribo- somes, and the NH4C1 wash. Antibiotic produced from each of these substrates was shown to be identical to Bacitracin in its spectral and chromatographic properties. The results presented in Table 11 indicate that 65% of the total Bacitracin yield from the four functions was derived from the S-100 component, while only 4.6% from the membrane fragments. Twenty-two percent of the antibiotic pre- cursor was found firmly bound to the ribosomes whose proteins account for about 12% to 14% of the total cell protein. The fraction of the cell lysate containing the fewest protein species, the NH4C1 wash, appeared to be the most suitable material for precursor isolation. This was subject to precipitation with 75% ethanol and the precipitated protein separated from the alcohol Table II.——Distribution of the Bacitracin precursor in vegetative cell lysates. Bacitracin Produced Fraction Volume (m1) Concentration Total EErTSEEl (us/ml) (ug) S-100 24 326 7800 65 Membrane fragments 4.0 146 584 4.6 NH4C1 wash 10 105 1050 8.4 NH Cl-washed ribosomes 11 241 2650 22.0 4 72 soluble fraction. The proteolysis of these two fractions yielded antibiotic indicating that the Bacitracin precursor is principally a constituent of the alcohol precipitable protein, Table 3. However, when the two fractions were combined, they yielded Bacitracin 47% in excess of the expected value based on the yield of the individual fractions. This result is consistent with a proposed mechanism of antibiotic activation in which the A-Z thiazoline ring is formed by the transfer of a terminal isoleucine residue from one protein to another (99). Discussion Antibiotic production is a sporulation related process (Chapter 111). Our data indicate that Bacitracin is associated with sporulating B, licheniformis cell ribosomes. Proteolysis of Bacitracin- containing ribosomes does not lead to a significant increase in antibiotic activity. This is presumably due to the fact that cells were examined at a time when the Bacitracin-yielding substrate was Table III.--Bacitracin yield upon 3proteolysis of NH4C1 wash fraction. Component Observed Expected % Access Supernatant (s) (0.5 ml) 16 1° " Precipitate (p) (0.5 ml) 64 64 -- s (0.25 ml) + p (0.25 ml) 59 40 47 82.8 units of protease in 1.8 ml reaction volume. 73 depleted. Vegetative cell, Bacitracin-free ribosomes yield antibiotic upon their ip_!i££p proteolysis by a specific sporulation-related extracellular protease. The binding of Bacitracin precursor(s) may be a gratuitous phenomenon or have some function in the vegetative cell protein economy. The ribosome is polyanionic structure which binds proteins. On the other hand the distribution of the Bacitracin precursor is similar to that of initiation factors (IF), proteins functional in translational events, in Bacillus subtilis (69). The following is presented in support of likelihood that such a reaction occurs ip vivo. Under the conditions of gentle lysis 96% of the total ribosomal material of vegetative B, licheniformis cells is membrane bound (131) with a relatively large proportion of 705 particles (130). The protein composition of membrane-bound ribosomes differs from their cytoplasmic counterparts (18, 19) in that SOS subunit lacks two proteins normally found in unbound ribosomes. Fifteen percent of the total intracellular proteolytic activity in B, coli is found to be associated with membrane fragments and another 5% remains bound to ribosomes after washing with 1.5 M NH4C1 (91). The detailed studies of extracellular protease(s) from Bacillus amyloliquefaciens indicate that they are synthesized by membrane-bound ribosomes and immediately equilibrated with the entire enzyme pool. A fraction of the enzyme molecules adsorbs or complexes with cell membrane (38 and references therein). B, licheniformis produces a highly soluble extracellular penicillinase and a hydro- phobic membrane-bound form while containing only a single penicillinase structural gene. Membrane penicillinase has a molecular weight of 74 about 33,000 and contains a covalently linked phospholipid. Extra— cellular penicillinase has a molecular weight of 29,000 and is released from the membrane by tryptic cleavage. Thus, there is a precedent in B, licheniformis for a proteolytic release occurring in the mem- brane region, of an excreted biologically active polypeptide. It appears possible that an antibiotic precursor(s), presumably a con- stituent of membrane-bound vegetative ribosomes, is cleaved by a specific extracellular protease (synthesized by the same ribosomes), prior to its excretion. The reaction ip_!i!p_would take place with high probability due to the immobilization of substrate and enzyme by the membrane matrix. There are at least two possible consequences of this reaction. (1) The release of the hydrophobic peptide antibiotic could cause, or result in, modification of the membrane and/or membrane- bound ribosomes (eg., their solubilization). (2) The more specific possibility, however, is that Bacitracin release is itself a consequence of a translational control exerted by the sporulation-related serine protease. The NH4C1 wash of ribo- somes (69) and extraction of the wash with ethanol (17) yield crude and purified initiation factors (IF), respectively. Precisely the ,, same preparations yield Bacitracin upon proteolysis. Chambliss and Legault—Demare have recently shown that the IF fraction is modified upon the initiation of sporulation in B, subtilis such that the cell's ability to translate phage SPOl RNA is quite reduced or "turned off." This "turn Off" appears specific since IF from sporulating cells are 80% as active as the vegetative factors in the translation of 75 vegetative RNA and 100% as active in the translation of Q8 and T4 RNAs. Furthermore, this "turn-off" phenomenon appears to be sporulation-specific event since it does not occur in asporogenous mutants which are blocked early in sporulation (21). It seems reasonable to propose that a sporulation-specific modification of initiation factors could be effected by a specific seryl-protease. That the same action leads to an antibiotic activation is only con- sistent with the teleonomic and conservative nature of biological systems. B I BL IOGRAPHY BIBLIOGRAPHY Abraham, E. P. 1957. The bacitracins. ip_Biochemistry of some peptide and steroid antibiotics, p. 1-29. J. Wiley 8 Sons, Inc., New York. Adler, R. H., and J. E. Snoke. 1962. Requirement of divalent metal ions for bacitracin activity. J. Bacteriol. BB:1315. Anker, H. S., B. A. Johnson, J. Goldberg, and F. L. Meleney. 1949. Bacitracin: Methods of production, concentration, and partial purification, with a summary of the chemical prop- erties Of crude bacitracin. J. Bacteriol. BB,249. Aronson, A. I. 1973. The function of various proteases in bacterial sporulation. lp_Colloques Internationaux C.N.R.S. NO. 227:33. Aronson, A. I., N. Angelo, and S. C. Holt. 1971. Regulation of extracellular protease production in Bacillus cereus T: characterization Of mutants producing altered amounts of— protease. J. Bacteriol. 199,1016. Balassa, G. 1971. The genetic control of spore formation in Bacilli. Curr. Top. Microbiol. Immunol. BB,99. Balassa, G., B. Dod, and J. Zucca. 1975. Overproduction of sporulation-associated extracellular protease in Bacillus subtilis mutants, p. 279-281. In P. Gerhardt, H. L. Sadoff, and R. N. Costilow (eds.), SporES VI, American Society for Microbiology, Washington, D.C. Bauer, A. W., W. M. M. Kirby, J. C. Sheris, and M. Turck. 1966. Antibiotic susceptibility testing by a standardized single disc method. Am. J. Clin. Path. 5B3493. Beining, P. R., C. L. Pinsley, and E. D. Weinberg. 1967. Alteration Of permeability of protoplasts of Bacillus megaterium by bacitracin and selected metallic ions. Antimicrobial agents and chemotherapy 1966:308. 76 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 77 Bernlohr, R. W. 1964. Postlogarithmic phase metabolism of sporulating microorganisms. I Protease of Bacillus licheni- formis. J. Biol. Chem. 239:538. Bernlohr, R. W., and V. Clark. 1971. Characterization and regulation Of protease synthesis and activity in Bacillus licheniformis. J. Bacteriol. 105:276. Bernlohr, R. W., and B. H. Gray. 1969. Enzyme inactivation during initiation of sporogenesis, p. 186-195. lp_L. L. Campbell (ed.), Spores IV, American Society for Microbiology, Bethesda, Md. Bernlohr, R. W., and G. D. Novelli. 1960. Some character- istics of bacitracin production by Bacillus licheniformis. Arch. Biochem. Biophys. BZ,232. Bernlohr, R. W., and G. D. Novelli. 1963. Bacitracin bio- synthesis and spore formation: the physiological role of an antibiotic. Arch. Biochem. Biophys. 103:94. Bonamy, G., L. Hirschbein, and J. Szulmajster. 1973. Synthesis of ribosomal nucleic acid during sporulation of Bacillus subtilis. J. Bacteriol. 113:1296. Bonventre, P. F., and L. L. Kempe. 1960. Physiology of toxin production by Clostridium botulinum types A and B. IV. Activation of the toxin. J. Bacteriol. 22,24. Brot, N., E. Yamasaki, B. Redfield, and H. Weissbach. 1972. The properties of an Escherichia coli ribosomal protein required for the function of factor G. Arch. Biochem. Biophys. 148:148. Brouwer, J., and R. J. Planta. 1975. Protein composition of ribosomes in DNA-membrane complexes from Bacillus licheni- formis. Biochem. Biophys. Res. Comm. BB,336. Brouwer, J., and R. J. Planta. 1975. The origin of high molecular weight proteins in ribosomal preparations of Bacillus licheniformis. FEBS Lett. BB:73. Cassio, D., and J. P. Waller. 1971. Modification of methionyl- tRNA synthetase by proteolytic cleavage and properties of the trypsin-modified enzyme. Eur. J. Biochem. 39:283. Chambliss, G. H., and L. Legault-Demare. 1975. Template discrimination by the initiation factor fraction from the ribosomes Of sporulating Bacillus subtilis cells, p. 314. lp_P. Gerhardt, H. L. Sadoff, and R. N. Costilow (eds.), Spores VI, American Society for Microbiology, Washington, D.C. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 78 Cornell, N. W., and G. Guiney Jr. 1970. Binding sites for zinc (II) in bacitracin. Biochem. Biophys. Res. Comm. BB, 530. Cornell, N., and J. E. Snoke. 1964. Biosynthesis of bacitracin and protein. Biochim. Biophys. Acta 21,533. Craig, L. C., W. F. Phillips, and M. Burachik. 1969. Baci- tracin A: isolation by counter double-current distribution. Biochemistry B,2348. Creighton, T. E. 1975. Homology of protein structures: proteinase inhibitors. Nature 255:743. Damodaran, M., V. S. Govindarajan, and S. S. Subramanian. 1955. The proteolytic system of Bacillus licheniformis. Biochim. Biophys. Acta 11,99. Das Gupta, B. R., and H. Sugiyama. 1972. A common subunit structure in Clostridium botulinum type A, B and E toxins. Biochem. Biophys. Res. Comm. BB,108. Deutscher, M. P., and A. Kornberg. 1968. Biochemical studies of bacterial sporulation and germination. VIII. Patterns of enzyme development during growth and sporulation of Bacillus subtilis. J. Biol. Chem. BBB,4653. Doi, R. H. 1973. Role of proteases in sporulation, p. 1-20. Ip_B. L. Horecker and E. R. Stadtman (eds.), Current topics in cellular regulation, Vol. 7. Eklund, M. W., and F. T. Poysky. 1972. Activation of toxic component of Clostridium botulinum types C and D by trypsin. Appl. Microbiol. 23,108. Freese, E. 1972. Sporulation of Bacilli, a model of cellular differentiation. Curr. Top. Develop. Biol. 2,85. Freese, E., and P. Fortnagel. 1967. Analysis of sporulation mutants. 1. Response Of uracil incorporation to carbon sources and other mutant properties. J. Bacteriol. 23,1957. FroyshOv, 0., and S. G. Laland. 1974. On the biosynthesis of bacitracin by a soluble enzyme complex from Bacillus licheniformis. Eur. J. Biochem. 39,235. Fujikawa, K., T. Suzuki, and K. Kurahashi. 1968. Biosyn- thesis of tyrocidine by a cell free enzyme system of Bacillus brevis ATCC 8185. 1. Preparation of partially purified enzyme system and its properties. Biochim. Biophys. Acta 161:232. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 79 Fukuda, R., G. Keilman, E. McVey, and R. H. Doi. 1975. Ribonucleic acid polymerase pattern of sporulating Bacillus subtilis cells, p. 202. In P. Gerhardt, H. L. Sadoff, and R. N. Costilow (eds.), SpOTes VI, American Society for Microbiology, Washington, D.C. - Geisser, M., G. W. Tischendorf, and G. StOffler. 1973. Comparative immunological and electrophoretic studies on ribosomal proteins of Bacillaceae. Mol. Gen. Genet. 127:129. Goldthwaite, G., and 1. Smith. 1972. Physiological characteri- zation of antibiotic resistant mutants of Bacillus subtilis. Mol. Gen. Genet. 114:190. Gould, A. R., B. K. May, and W. H. Elliott. 1975. Release of extracellular enzymes from Bacillus amyloliquefaciens. J. Bacteriol. 122:34. Greenleaf, A. L., G. L. Thomas, and R. Losick. 1973. Isolation of a new RNA polymerase-binding protein from sporulating Bacillus subtilis. Proc. Natl. Acad. Sci. U.S.A. 29,490. Grelet, N. 1972. Un exemple de morphogeneSe: la sporulation des Bacillacées, p. 47. lp_Soc. bot. Fr., Mémoires (Coll. Morphologie). Haavik, H. I. 1974. Studies on the formation of bacitracin by Bacillus licheniformis: role of catabolite repression and organic acids. J. Gen. Microbiol. B3,321. Haavik, H. I., and S. Thomassen. 1973. A bacitracin-negative mutant of Bacillus licheniformis which is able to sporulate. J. Gen. Microbiol. 23,451. Hageman, J. H., and B. C. Carlton. 1970. An enzymatic and immunological comparison of two proteases from a transformable Bacillus subtilis with the subtilisins. Arch. Biochem. Biophys. 332,67. Hageman, J. H., and B. C. Carlton. 1973. Effects of mutational loss of specific intracellular proteases on the sporulation of Bacillus subtilis. J. Bacteriol. 114:612. Hall, F. F., H. O. Kunkel and J. M. Prescott. 1966. Multiple proteolytic enzymes of Bacillus licheniformis. Arch. Biochem. Biophys. 114:145. Hanson, R. S., J. A. Peterson, and A. A. Yousten. 1970. Unique biochemical events in bacterial sporulation. Ann. Rev. Microbiol. 33,53. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. S7. 58. 8O Hash, J. M. (ed.). 1975. Methods in enzymology, vol. XLIIl, Antibiotics. Academic Press Inc., New York. Hashimoto, T., S. H. Black, and P. Gerhardt. 1960. Develop- ment of fine structure, thermostability and dipicolinate during sporogenesis in Bacillus. Can. J. Microbiol. 9,203. Hermier, B., M. Pacaud, and J. Dubert. 1973. Action Of pro- tease I of Escherichia coli on RNA polymerase of same origin. Eur. J. Biochem. 3B,307. Hershko, A., and M. Fry. 1975. Post-translational cleavage of polypeptide chains: role in assembly. Ann. Rev. Biochem. 44:775. Hitchins, A. D., and H. L. Sadoff. 1972. Cell division septation and sporulation septation in Bacillus megaterium, p. 148. lp_H. O. Halverson, R. Hanson, and L. L. Campbell (eds.), Spores V, American Society for Microbiology, Washington, D.C. Hodgson, B. 1970. Possible roles for antibiotic and other biologically active peptides at specific stages during sporulation of Bacillaceae. J. Theor. Biol. 39,111. Holzer, H., H. Betz, and E. Ebner. 1975. Intracellular pro- teinases in microorganisms, p. 103-156. lp_B. L. Horecker and E. R. Stadtman (eds.), Current topics in cellular regulation, vol. 9. Ishihara, H., T. Sasaki, and K. Shimura.14l968. Biosynthesis of bacitracin. II. Incorporation of C-labelled amino acids into bacitracin by cell-free preparation from Bacillus licheniformis. Biochim. Biophys. Acta 166:496. Ishihara, H., and K. Shimura. 1974. Biosynthesis of bacitracin. III. Partial purification of a bacitracin-synthesizing enzyme system from Bacillus licheniformis. Biochim. Biophys. Acta 33B,588. Kabat, E. A., and M. M. Mayer. 1961. Estimation of protein with the biuret and ninhydrin reactions. lp_Experimental immunochemistry, 2nd ed., p. 559. Charles C. Thomas, Springfield, Illinois. Kambe, M., Y. Imae, and K. Kurahashi. 1974. Biochemical studies on gramicidin S non-producing mutants of Bacillus brevis ATCC 9999. J. Biochem. ZB,487. Keay, L., and P. Moser. 1969. Differentiation of alkaline proteases from Bacillus species. Biochem. Biophys. Res. Comm. 33,600. S9. 60. 61. 62. 63. 64. 65. 66. 67. 68. 69. 70. 81 Kerjan, P. 1973. Effect of proteases on RNA polymerase of Bacillus subtilis. Colloques Internationaux C.N.R.S. NO. 227:41. King, A. M. Q., P. A. Lowe, and B. H. Nicholson. 1974. Trypsin modification of DNA-dependent RNA polymerase of Escherichia coli B, Biochem. Biophys. Res. Comm. 99,38. Kleinkauf, H., W. Gevers, and F. Lipmann. 1969. Interrelation- ship between activation and polymerization in gramicidin 8 synthesis. Proc. Natl. Acad. Sci. U.S.A. 99,226. Klier, A. and M. M. Lecadet. 1974. Evidence in favour of the modification in vivo of RNA polymerase subunits during sporogenesis in Bacillus thuringiensis. Eur. J. Biochem. 47:111. Klier, A. F., M. M. Lecadet, and R. Dedonder. 1973. Sequential modifications of DNA-dependent RNA polymerase‘during sporogenesis in Bacillus thuringiensis. Eur. J. Biochem. 36:317. Kornberg, A., J. A. Spudich, D. L. Nelson, and P. Deutscher. 1968. Origin of proteins in sporulation. Ann. Rev. Biochem. 37:51. Korzybski, T., Z. Kowszyk-Gindifer, and W. Kurylowicz. 1967. Antibiotics origin, nature, and properties, p. 78. Pergamon Press Inc., New York. Kurimatsu, H. K., and J. E. Snoke. 1962. Biosynthesis of 0- amino acids in Bacillus licheniformis. Biochim. Biophys. Acta 99,114. Kurylo-Borowska, Z. 1975. Biosynthesis of edeine. II. Localization of edeine synthetase within Bacillus brevis Vm 4. Biochim. Biophys. Acta 399:31. Lee, S. G. and F. Lipmann. 1974. Correlation between anti- biotic synthesis, metabolite uptake and pre- spore formation. Fed. Proc. 33: 352. Legault-Demare, L., and G. H. Chambliss. 1974. Natural messenger ribonucleic acid-directed cell-free protein syn- thesizing system of Bacillus subtilis. J. Bacteriol. 120: 1300. Leighton, T. J., R. H. Doi, R. A. J. Warren, and R. A. Kelln. 1973. The relationship of serine protease activity to RNA polymerase modification and sporulation in Bacillus subtilis. J. Mol. Biol. 29,103. 71. 72. 73. 74. 75. 76. 77. 78. 79. 80. 81. 82. 82 Lettl, A., K. Nekvasilova, K. Moravec, and A. Stejskal. 1966. Etude de la toxinogénése chez Plectridium tetani. III. Synthése intracellulaire du précurseur et son activation. Ann. Inst. Pasteur 232,622. Linn, T. G., A. Greenleaf, R. Shorenstein, and R. Losick. 1973. Loss of the sigma activity of RNA polymerase of Bacillus subtilis during sporulation. Proc. Natl. Acad. Sci. U.S.A. 29,1865. Losick, R., R. G. Shorenstein, and A. L. Sonenshein. 1971. Structural alteration of RNA polymerase during sporulation. Nature (London) 227:910. Losick, R., and A. L. Sonenshein. 1969. Change in the template specificity of RNA polymerase during sporulation of Bacillus subtilis. Nature (London) 227:35. Lowry, O. H., N. J. Rosebrough, A. L. Farr, and R. J. Randall. 1951. Protein measurements with the folin phenol reagent. J. Biol. Chem. 293:265. MacDonald, R. 1., R. C. MacDonald, and N. W. Cornell. 1974. Perturbation of liposomal and planar lipid bilayer membranes by bacitracin-cation complex. Biochemistry 23,4018. Makinen, K. 1972. Inhibition by bacitracin of some hydrolytic enzymes. Int. J. Protein Res. 3,21. Mandelstam, J. 1969. Regulation of bacterial spore formation, p. 377. In Symposia of the society for general microbiology XIX, MicrOEial growth. Cambridge University Press, Cambridge, England. Mandelstam, J. 1971. Recurring patterns during development in primitive organisms, p. 1. 29_Symposia of society for experimental biology XXV, Control mechanisms Of growth and differentiation. Cambridge University Press, Cambridge, England. Mandelstam, J., and W. M. Waites. 1968. Sporulation in Bacillus subtilis: the role of exoprotease. Biochem. J. 109:793. McConn, J. D., D. Tsuru, and K. T. Yasunobu. 1964. Bacillus subtilis neutral proteinase. J. Biol. Chem. 239:3706. Millet, J. 1970. Characterization of proteinases excreted by Bacillus subtilis Marburg strain during sporulation. J. Appl. Bacteriol. 33,207. 83 83. Millet, J., P. Kerjan, J. P. Aubert, and J. Szulmajster. 1972. Proteolytic conversion ip_vitro of Bacillus subtilis vege— tative RNA polymerase into the homologous spore enzyme. FEBS Lett. 23:47. Orrego, C., P. Kerjan, M. C. Manca de Nadra, and J. Szulmajster. 84. 1973. RNA polymerase in thermosensitive sporulation mutant. J. Bacteriol. 116:636. 1970. The subtilisins, p. 199. 85. Ottesen, M., and I. Svendsen. 29_C. E. Perlman and L. Lorand (eds.), Methods in enzymology, XIX, Academic Press, London. vol. Biosynthesis of peptide 86. Perlman, D., and M. Bodanszky. 1971. antibiotics. Ann. Rev. Biochem. 39,449. 87. Pfaender, P , D. Specht, G. Heinrich, E. Schwarz, E. Kuhnle, and M. M. Simlot. 1973. Enzymes of Bacillus licheniformis FEBS Lett. 32,100. in the biosynthesis of bacitracin A. 1971. Protease 88. Prestidge, L., V. Cage, and J. Spizizen. activities during the course of sporulation of Bacillus J. Bacteriol. 107:815. Polypeptide antibiotic- J. Gen. subtilis. 89. Ray, B., and S. K. Bose. 1971. negative spore former mutants of Bacillus subtilis. 8 Appl. Microbiol. 22,491. 90. Regna, P. P. 1959. The chemistry of antibiotics. The In H. S. Goldberg (ed.), Antibiotics, bacitracins, p. 129. I__ their chemical and non-medical uses, D. Van Nostrand Inc., New York. 91. Regnier, P., and M. N. Thang. 1972. Subcellular distribution and characterization of endo- and exo-cellular proteases in Biochimie 93,1227. Escherichia coli. Characterization of an 92. Reysset, G., and J. Millet. 1972. intracellular protease in Bacillus subtilis during sporulation Biochem. Biophys. Res. Comm. 39,328. Reider, H., G. Heinrich, E. Brenker, M. M. Simlot, and P. In J. H. 93. Pfaender. 1975. Bacitracin synthetase, p. 548. ___ Hash (ed.), Methods in enzymology XLIII Antibiotics, Academic Press, New York. 1975. 94. Ristow, H., B. Schazschneider, and H. Kleinkauf. Effects of peptide antibiotics tyrocidine and the linear gramicidin on RNA synthesis and sporulation of Bacillus brevis. Biochem. Biophys. Res. Comm. 93:1085. 95. 96. 97. 98. 99. 100. 101. 102. 103. 104. 105. 106. 84 Roskoski, R., Jr., H. Kleinkauf, W. Gevers, and F. Lipmann. 1970. Isolation of enzyme bound peptide intermediates in tyrocidine biosynthesis. Biochemistry 9,4846. Rouget, P., and F. Chapeville. 1971. Leucyl-tRNA synthetase. Two forms of the enzyme: relation between structural and catalytic properties. Eur. J. Biochem. 23,459. Sadoff, H. L. 1972. The antibiotics of Bacillus species: their possible roles in sporulation. Progr. Ind. Microbiol. 11:1. Sadoff, H. L. 1973. Comparative aspects of morphogenesis in three prokaryotic genera. Ann. Rev. Microbiol. 22,133. Sadoff, H. L., and E. Celikkol. 1971. Antibiotic formation in Bacillus cereus by proteolysis of its cellular ribosomes, p. 59. In A. N. Barker, G. W. Gould, and J. Wolf (eds.), Spore reSEarch, Academic Press, London. Sadoff, H. L., A. D. Hitchins, and E. Celikkol. 1969. Properties of fructose 1,6-diphosphate aldolase from spores and vegetative cells of Bacillus cereus. J. Bacteriol. 9B, 1208. Sadoff, H. L., E. Celikkol, and H. L. Engelbrecht. 1970. Conversion of bacterial aldolase from vegetative to spore form by a sporulation-specific protease. Procl Natl. Acad. Sci. U.S.A. 99,844. Sarkar, N., and H. Paulus. 1972. Function of peptide anti- biotics in sporulation. Nature New Biology 239:228. Schaeffer, P. 1969. Sporulation and the production of anti- biotics, exoenzymes, and exotoxins. Bacteriol. Rev. 33,48. Schaeffer, P., J. Millet, and J. P. Aubert. 1965. Catabolic repression of bacterial sporulation. Proc. Natl. Acad. Sci. U.S.A. 33,704. Schechter, N., K. Mpmose, and H. Rudney. 1972. The effect of bacitracin and Ca on the formation of polyprenylpyrophos- phates and their incorporation into ubiquinone precursors in mammalian and bacterial systems. Biochem. Biophys. Res. Comm. 39,833. Schechter, N., T. Nishimo, and H. Rudney. 1973. The synthesis of 3-nonaprenyl-3;hydrpxybenzoate in rat liver mitochondria: the effect of Ca , Mg , chelators and bacitracin on the activity of p-hydroxybenzoate polyprenyl transferase. Arch. Biochem. Biophys. 293:282. 107. 108. 109. 110. 111. 112. 113. 114. 115. 116. 117. 118. 85 Schmitt, R., and E. Freese. 1968. Curing of a sporulation mutant and antibiotic activity Of Bacillus subtilis. J. Bacteriol. 99,1255. Setlow, P. 1975. Protease and peptidase activities in growing and sporulating cells and dormant spores of Bacillus mega- terium. J. Bacteriol. 122:642. Shoer, R., and H. P. Rappaport. 1972. Analysis of a Bacillus subtilis proteinase mutant. J. Bacteriol. 109:575. Siewert, G., and J. L. Strominger. 1967. Bacitracin: an inhibitor of dephosphorylation of lipid pyrophosphate, an intermediate in biosynthesis of the peptidoglycan of bacterial cell walls. Proc. Natl. Acad. Sci. U.S.A. 92,767. Simonoff, P., R. W. Price, and W. G. Bywater. 1954. Some prOperties of bacitracin methylene-disalicylate (DPS), p 395. In Antibiotic Annual, Medical Encyclopedia Inc., New York. Slopikoff, S., J. L. Spitzer, and D. Vacearo. 1971. Sporu- lation in Bacillus brevis: studies on protease and protein turnover. J. Bacteriol. 106:739. Snell, N., K. Ijichi, and J. G. Lewis. 1956. Paper chromato- graphic identification Of polypeptidic gram-positive inhibiting antibiotics. Appl. Microbiol. 3,13. Snoke, J. E. 1960. Formation of bacitracin by washed cell suspensions of Bacillus licheniformis. J. Bacteriol. B9, 552. Snoke, J. E. 1961. Formation of bacitracin by protoplasts of Bacillus licheniformis. J. Bacteriol. 92,986. Snoke, J. E., and N. Cornell. 1965. Protoplast lysis and inhibition of growth of Bacillus licheniformis by bacitracin. J. Bacteriol. 39,415. Spitnik-Elson, P., and A. Atsmon. 1969. Detachment of ribo- somal proteins by salt. I. Effect of conditions on the amount of protein detached. J. Mol. Biol. 39,113. Steinberg, W. 1975. Sporulation of Bacillus subtilis cells: in vivo evidence for phenotypic modification of a thermo— sensitive lysyl- transfer ribonucleic acid synthetase, p. 290. In P. Gerhardt, H. L. Sadoff, and R. N. Costilow (eds. ), Speres VI, American Society for Microbiology, Washington, D.C. 119. 120. 121. 122. 123. 124. 125. 126. 127. 128. 129. 86 Stone, K. J., and J. L. Strominger. 1971. Mechanism of action of bacitracin: complexation with metal ion and CSS-isoprenyl pyrophosphate. Proc. Natl. Acad. Sci. U.S.A. 99, 3223. Stone, K. J., and J. L. Strominger. 1972. Inhibition of sterol biosynthesis by bacitracin. Proc. Natl. Acad. Sci. U.S.A. 69, 1287. Storm, D. R., and J. L. Strominger. 1973. Complex formation between bacitracin peptides and isoprenyl pyrophosphates. J. Biol. Chem. 248:3940. Storm, D. R., and J. L. Strominger. 1974. Binding of baci- tracin to cells and protoplasts of M, lysodeikticus. J. Biol. Chem. 249:1823. Sun, T. T., T. A. Bickle, and R. R. Traut. 1972. Similarity in the size and number of ribosomal proteins from different prokaryotes. J. Bacteriol. 111:474. Switzer, R. L., C. L. TUrnbough, J. S. Brabson, and L. L. Waindle. 1975. Enzyme inactivation of dg_novo nucleotide biosynthesis in sporulating Bacillus subtilis cells, p. 327. ln_P. Gerhardt, H. L. Sadoff, and R. N. Costilow (eds.), Spores VI, American Society for Microbiology, Washington, D.C. Szulmajster, J. 1964. Biochimie de la sporogenése chez Bacillus subtilis. Bull. Soc. Chim. Biol. £6,443. Szulmajster, J. 1973. Initiation of bacterial sporogenesis, p. 45. lE_Symposia of the society for general microbiology XXIII, Microbial differentiation. Cambridge University Press, Cambridge, England. Szulmajster, J. 1973. Bacillus subtilis RNA polymerase from vegetative and sporulating cells and from dormant spores: hypotheses and realities. Colloques Internationaux C.N.R.S. No. 991,47. Szulmajster, J., and E. Keryer. 1975. Isolation and properties of thermosensitive sporulation mutants of Bacillus subtilis deficient in intracellular protease activity, p. 271. In P. Gerhardt, H. L. Sadoff, and R. N. Costilow (eds.), Spores VI, American Society for Microbiology, Washington, D.C. Tissieres, A., J. D. Watson, D. Schlessinger, and B. R. Hollingsworth. 1959. Ribonucleoprotein particles from Escherichia coli. J. Mol. Biol. 1:221. 130. 131. 132. 133. 134. 135. 136. 137. 87 Van Dijk-Salkinoja, M. S., and R. J. Planta. 1970. Formation and life cycle of ribosomal subunits, mono- and poly-ribosomes in Bacillus licheniformis. Arch. Biochem. Biophys. 141:477. Van Dijk-Salkinoja, M. S., T. J. Stoof, and R. J. Planta. 1970. The distribution of polysomes, ribosomes, and ribosomal subunits in exponential-phase cells of Bacillus licheniformis. Eur. J. Biochem. 19:474. Vitkovié, Lj., and H. L. Sadoff. 1974. The role of protease in antibiotic production during sporulation of Bacillus licheniformis. Fed. Proc. 99,1475. Vitkovié, Lj., and H. L. Sadoff. 1975. Relationship between sporulation protease and antibiotic in sporulating Bacillus licheniformis cells, p. 362. In P. Gerhardt, H. L. Sadoff, and R. N. Costilow (eds.), Spores VI, American Society for Microbiology, Washington, D.C. Weinberg, E. D. 1958-1959. Enhancement of bacitracin by the metallic ions of group II B, p. 924. Ifl_Antibiotics annual, Medical Encyclopedia Inc., New York. Weinberg, E. D. 1967. Bacitracin, p. 90. ln_D. Gottlieb and P. D. Shaw (eds.), Antibiotics I, mechanism of action. Springer-Verlag, New York. Weinberg, E. D., and S. M. Tonnis. 1967. Role of manganese in biosynthesis of bacitracin. Can. J. Microbiol. 19,614. Yamamoto, S., and J. 0. Lampen. 1975. Membrane penicillinase of Bacillus licheniformis 749-C, a phospholipoprotein. J. Biol. Chem. 250:3212.