OVERDUE FINES: 25¢ per day per itau RETURNING LIBRARY MATERIALS: Place in book return to remove charge from circulation records EVIDENCE FOR A NOVEL PATHWAY FOR ISOLEUCINE BIOSYNTHESIS IN CLOSTRIDIUM SPOROGENES By Ratna Siri Hadioetomo A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Microbiology and Public Health 1980 ABSTRACT EVIDENCE FOR A NOVEL PATHWAY FOR ISOLEUCINE BIOSYNTHESIS IN CLOSTRIDIUM SPOROGENES By Ratna Siri Hadioetomo Extracts of g, sporogenes (ATCC 7955, NCA PA 3679) were shown to contain an active biodegradative threonine dehydratase. The enzyme was constitutive and was markedly stimulated by ADP at low threonine concentrations. No stimulation occured at high sub— strate concentrations. The apparent Km's for threonine in the pre- sence and absence of ADP were 2.5 and 33.33 mM, respectively. No isoleucine sensitive threonine dehydratase activity was detected over a wide pH range. Threonine aldolase and threonine dehydro- genase activities were also not detected. g, sporogenes grew well in a minimal synthetic medium containing 8 essential amino acids (L-serine, L-arginine, L-phenylalanine, L-tyrosine, L-tryptophan, L-valine, L-leucine, and L-methionine), salts and vitamins (medium B-lO). The facts that exogenously added isoleucine was not required for growth and that isoleucine sensitive threonine dehydratase could not be detected prompted an investigation of isoleucine biosynthesis in this Ratna Siri Hadioetomo organism by measuring 140 incorporation from labeled individual com- ponents of medium B-lO. Of the amino acids in medium B-lO, only 14C-serine was found to contribute a significant amount of 140 to cellular isoleucine during growth. 14C-Isoleucine was also found in hydrolysates of proteins from cells grown in the presence of [3-14C1pyruvate and 14002. Cells grown in the presence of either l4C-threonine, l4C-aspartate, or 14C-glutamate did not contain . 14 14 labeled isoleucine. The amount of C from [3- C]pyruvate and 14C02 found in isoleucine was lower than would be expected to account for all 6 carbons. In contrast, results of studies of the extent of 14002 released by decarboxylation of radio— 14C incorporation and of active alanine, aspartate, threonine, lysine, glutamic acid, serine, and glycine were consistent with established biosynthetic pathways. Growth in medium B—lO was dependent on the presence of a low contaminating level of isoleucine or on the presence of substan- tial levels of 2-methylbutyric acid. The latter is the major product of isoleucine degradation by g, sporogenes. Protein from cells incubated in the presence of fermentation products of 14C-isoleucine incorporated the label specifically into the isoleucine of cell pro- tein. However, the specific activity of the isoleucine formed indicated an average of less than 4 carbons from 14C-2-methy1butyric acid incorporated per isoleucine formed. This plus the facts that the C-3 carbon of pyruvate and the carbon from l4CO2 are incorpor- ated into carbons other than the carboxyl carbon of isoleucine indicates that at least part of the isoleucine formed is synthesized Ratna Siri Hadioetomo by reactions more complex than the reductive carboxylation of 2-methylbutyric acid. 2, sporogenes grew well in medium B—lO without either leucine or isoleucine when 8 mM 2-methy1butyrate was added. Either there was sufficient contaminating leucine in the medium or this organism has the capacity to synthesize leucine. However, no label from any of the labeled substrates tested including 2-methylbutyric acid was found in the leucine isolated from cell proteins. Failure to incor- porate significant 14 C from labeled pyruvate or CO2 may have resulted from the high levels of leucine in the growth medium used in these experiments. DEDICATION To my husband, IGNATIUS J. HADIOETOMO, for his unfaltering support, encouragement and understanding. ii TABLE OF CONTENTS LIST OF TABLES . . . . . . . . . . LIST OF FIGURES . . . . . . . . . . ACKNOWLEDGEMENTS . . . . . . . . . . INTRODUCTION . . . . . . . . . LITERATURE REVIEW . . . . . . . . . . Nutritional requirements of C, sporogenes Threonine dehydratase . . . . . . . Isoleucine biosynthesis . . . . . . MATERIALS AND METHODS . . . . . . . . Cultures and cultural methods . . . Preparation of cell extracts . . . . . Enzyme assay . . . . . . . . . . Fractionation of isotopically labeled cells . Fractionation of amino acids from protein hydrolysates Determination of intracellular amino acid pools Decarboxylation of radioactive amino acids Incorporation of degradation product of L-[U-14C]isoleucine . . . . . . . Measurement of volatile fatty acids . . Preparation of [U-14C]2-methylbutyric acid Analytical methods . . . . . . . . Chemicals . . . . . . . . . . . RESULTS . . . . . . . . . . . . . Threonine metabolism . . . . . . . Threonine dehydratase activity . . . Threonine aldolase and dehydrogenase . Amino Acid Composition of Cells and Media Composition of cell protein . . . . Free amino acids in cell . Amino acids in growth medium . . . . iii Page vii viii \OU'IUJ l7 17 19 19 21 22 26 26 27 28 29 3O 3O 32 32 32 34 37 37 37 37 Isoleucine Biosynthesis . . . . . . . . . Compounds contributing no significant carbon to isoleucine . . . . . . . . . . . Incorporation of 14C from L-[U-14C]serine into isoleucine and other amino acids . . Incorporation of radioactive carbon from 14CO2 into isoleucine and other amino acids . . . . Incorporation of 14C from [3-14C1pyruvate into isoleucine and other amino acids . . . . . . Pulse labeling with [3-14C]pyruvate . . . Effect of isoleucine in incorporation of 14C from [3-14C]pyruvate into isoleucine . . . . . . Incorporation of label from a matabolite(s) produced from L-[U-l4C]isoleucine into isoleucine . . . Metabolites of Isoleucine . . . . . . . . . . Volatile fatty acids . . . . . . . . . . Growth requirements . . . . . . . . . . . . Isoleucine in the medium . . . . . . . . . Growth response to leucine and isoleucine . . . Growth response to 2-methylbutyrate . . . . . Incorporation of 14C from [U-14C12-Methylbutyric Acid into Isoleucine . . . . . . . . . . . . . . DISCUSSION . . . . . . . . . . . . . . . . BIBLIOGRAPHY . . . . . . . . . . . . . . iv Page 40 4O 46 49 54 56 58 58 62 62 64 64 64 65 68 7O 95 .Table 7. 10. 11. 12. LIST OF TABLES Composition of various synthetic media used for the cultivation of g, sporogenes . . . . . Effect of growth medium on levels of cellular threonine dehydratase activity . . . . . . . . The effect of AMP and ADP on threonine dehydratase activity at high and low substrate concentrations . Amino acid composition of protein from cells grown in various synthetic media . . . . . . . Intracellular concentration of amino acids in 9, sporogenes during exponential growth . . . . . . . Amino acid concentrations in the supernatant liquids of cultures grown in medium B-lO at various stages of growth . . . . . . . . . . . . . . . . Recovery of radioactivity from protein hydrolysates of g, sporogenes grown in the presence of various labeled substrates . . . . . . . . . . . . . . . Incorporation of radioactivity into isoleucine of cell protein by Q. sporogenes labeled for one genera- tion with various ZFCC-amino acids . . . . . . . Distribution of radioactivity in amino acids from protein hydrolysate of C, s oro enes cells labeled for one generation with L-IU-IECIthreonine . . . . Distribution of radioactivity in amino acids from protein hydrolysate of g, s oro enes cells labeled for one generation with L-IU-IECIglutamate . . . . . Distribution of radioactivity in amino acids from protein hydrolysate of g. s oro enes cells labeled for one generation with L-lU-IECIserine . . . . Decarboxylation of amino acids from protein hydrolysate of C. sporogenes cells grown in the presence of L- [E-IQC‘ISerine o o o o o o o o o o o a o o Page 18 33 33 38 39 41 42 44 45 45 47 47 Table 13. 14. 15. 16. 17. 18. 19. 20. 21. Distribution of radioactivity in amino acids from protein hydrolysate of C, sporogenes cells labeled for one and two generations with thOZ Distribution of radioactivity in amino acids from protein hydrolysate of C. sporogeng§_cells labeled for one generation with—14002 Decarboxylation of amino acids from protein hydrolysate of Q, sporogenes cells grown in the presence of 14 CO2 . . . . . . . . Distribution of radioactivity in amino acids from protein hydrolysate of C. s oro enes cells labeled EECIpyruvate . for one generation with—[3- Decarboxylation of amino acids from protein hydrolysate of cells grown in the presence of [3-14C]pyruvate Incorporation of label into protein by g, sporogenes cells pulsed for varying periods with [3-14C]pyruvate in either fresh or spent medium B-lO . Effect of unlabeled L-isoleucine on incorporation of C]pyruvate into amino acids isotopic carbon from [3- Incorporation of the product of isoleucine degradation into cellular isoleucine . . Volatile fatty acids obtained from isoleucine fermenta- tion and from a growing culture vi Page 51 53 53 55 59 LIST OF FIGURES Figure Page 1. Various pathways of isoleucine biosynthesis by microorganisms . . . . . . . . . . . . . . . 10 2. Proposed pathways of isoleucine biosynthesis in Leptospira . . . . . . . . . . . . . . . . 15 3. Distribution of 14C in amino acids of hydrolysed protein from C. sporogenes grown in medium B-lO containing L-[U—lEC]serine . . . . . . . . . . . 25 4. Velocities of threonine dehydratase as a function of L-threonine concentration in the presence and absence of ADP . . . . . . . . . . . . . . . 36 5. The effect of leucine on growth of g, 4porogenes in medium B-lo o o o o o o o o o o o o o o o 66 6. The effect of 2-methylbutyrate on growth of g, sporogenes in medium B-lO . . . . . . . . . . 67 7. Interpretative scheme of the amino acid labeling pattern for all radioactive experiments with g, spgrogenes . . . . . . . . . . . . . . . 79 8. Glutamate synthesis from oxalacetate . . . . . . . . 83 9. Fromation of citrate by gifcitrate synthase and by Egycitrate synthase . . . . . . . . . . . . . 87 vii ACKNOWLEDGEMENTS I wish to thank my major professor, Dr. R. N. Costilow, for his guidance and assistance throughout the course of this work and during the preparation of this dissertation. I would also like to express my thanks to Dr. H. L. Sadoff and Dr. C. A. Reddy for the use of their laboratory facilities. Financial support for the program under which this study was conducted was provided by the Midwest Universities Consortium for International Activities under a contract with the Agency for International Development. viii .INTRODUCTION In Escherichia coli, Salmonella typhimurium, and some other microorganisms (66), the first enzyme in the isoleucine biosynthetic pathway is threonine dehydratase. The latter is a pivotal enzyme, since, being subject to feedback inhibition by isoleucine, it regu- lates the flow of carbon over this pathway in these organisms. In E, coli and S, typhimurium, the formation of the biosynthetic threonine dehydratase is controlled by multivalent repression by isoleucine, valine, and leucine (65). The biosynthesis of isoleucine via unusual pathways in a few bacteria is also well documented. Some rumen bacteria synthesize isoleucine by carboxylation of 2-methylbutyric acid (55). Some of these organisms (e.g. Methanobacterium ruminantium strain M-1 and Bacteroides ruminicola subsp brevis, 8, 20) require Z-methylbutyrate for growth, while some others do not. Certain species of rumen bac- teria were found to preferentially synthesize amino acids gg_novo even when they are grown in a medium containing a complete mixture of amino acids (9, 10). While representative strains fo E, ruminicola were found to utilize peptide nitrogen or ammonia nitrogen, free amino acids would not serve as the nitrogen source for growth (52). A threonine dehydratase-less mutant of E, ggli_Crookes strain uses glutamate as a precursor of a-ketobutyrate (and isoleucine) (50). In Leptospira serotypes semaranga and tarassovi, isoleucine 1 biosynthesis via threonine dehydratase reportedly operates to a very limited extent; instead, most of the cellular isoleucine is synthe- sized from a-ketobutyrate which in turn is synthesized via an unusual pathway (15). Preliminary studies in this laboratory showed that Clostridium sporogenes could utilize certain amino acids as the sole source of energy, carbon and nitrogen. The data indicated that neither iso- leucine nor branched chain volatile fatty acid(s) were required for growth. The organism was found to possess a very active threonine dehydratase, suggesting that a-ketobutyrate is produced from threo- nine. However, the failure to detect (in_gi££g) an isoleucine- sensitive threonine dehydratase and the lack of repression of the latter in cells grown in the presence of isoleucine, leucine, and valine raised the possibility that threonine might not be the pre- cursor of isoleucine. This was confirmed by the inability of the cells to incorporate label from L-[U-14C] threonine into isoleucine, suggesting that in g. sporogenes isoleucine might be synthesized via a pathway different from those described above. Accordingly, experiments were designed to investigate the pathway for isoleucine biosynthesis in g, sporqgenes by isotopic labeling with specific substrates, enzymological approaches, and other analytical methods. The results are described herein. LITERATURE REVIEW Nutritional requirements of g, sporogenes. The early litera- ture describing nutritional requirements of g, botulinum and related clostridia was summarized by Mager gt §1_(43). The majority of these studies resulted from the necessity to develop synthetic media suitable for investigating the physiology of toxin production, growth, and sporulation in g, botulinum. Most of the media so developed were chemically defined and consisted of mixtures of amino acids, salts, vitamins and, usually, glucose. In general, excellent growth was usually obtained with media containing mixtures of 15 to 19 amino acids (13, 43, 48). Mager ft 311 (43), Roessler and Brewer (56) and Campbell and Frank (13) identified the following amino acids as essential for growth of C. parabotulinum type A: tryptophan, threonine, valine, leucine, iso- leucine, methionine, arginine, phenylalanine and tyrosine. The last three amino acids were required in unusually high concentra- tions (43), which is consistent with the data of Shull gt_§l_(6l) for _g. sporogenes. The requirement for excess arginine could be relieved in part by ornithine and lysine (43); however, higher con- centrations of arginine in a complete mixture of amino acids resulted in sporulation (48) instead of lysis (43, 48) following maximal growth. The requirement for methionine could be partially relieved by cysteine but less than maximal growth was obtained. Leucine 3 and isoleucine could substitute for each other to some extent (43). Not all the strains require valine. Studies with media containing only the essential amino acids revealed that glycine would eliminate the requirement for threonine, and that serine could replace glycine. Among the various strains tested, some preferred glycine, and some others preferred serine (34). From a nutritional point of view, C, sporogenes is indistinguishable from g, parabotulinum types A and B (34, 43). Campbell and Frank (13) found that 10 strains of a putre- factive anaerobe (PA 3679), although variable in their requirements for other amino acids, shared a common requirement for arginine, phenylalanine, tyrosine, valine, isoleucine, and serine. The latter amino acids, along with proline and histidine, were reported to be essential for g, sporogenes ATCC 7955 (13), the organism used throughout this investigation. It should be noted that variation in growth conditions, reagent purity, etc. can result in erroneous conclusions about the nutrient requirements of a particular organism. It was demonstrated that decreasing the size of inoculum may necessitate the inclusion of an amino acid which is otherwise dispensible when a large inoculum is used (34). These and other inconsistencies (e.g. the level of growth that is scored as positive, etc.) have resulted in conflicting reports of the nutritional requirements of a given organism. The vitamin requirements of E, botulinum vary among the individual strains. In 9, sporogenes, biotin and p-aminobenzoate are essential, whereas nicotinic acid and thiamine are merely stimulatory (34, 60). The strain used in these experiments (ATCC 7955) requires only thiamine and biotin (13). Threonine dehygratase. Threonine dehydratase (L-threonine hydrolyase deaminating, E.C.4.2.l.16) is one of the enzymes involved in ghe degradation of threonine and catalyzes the conversion of threonine to a-ketobutyrate and ammonia (65). Two distinctly dif- ferent forms of threonine dehydratases (termed biosynthetic and biodegradative, respectively) are known to be produced by E, 2911 (64, 74). In E, ggll, the biosynthetic form of threonine dehydratase is synthesized by cells growing aerobically in a glucose minimal medium. The enzyme is inhibited by isoleucine (64) and thus con- sidered to participate in the initial step of isoleucine biosyn- thesis. Threonine dehydratase has been purified from various sources, including a yeast, E, typhimurium, E, coli, Rhodospirillum gpheroides, and Rhodospirillum rubrum [for review, see Umbarger (65)]. While differences exist among the threonine dehydratases from various sources, the enzymes share several common features. In particular, biosynthetic threonine dehydratase is subject to inhibition by isoleucine which is reversed by valine. The degree of inhibition, however, varies with the source of the enzyme. For example, in E, rubrum, the enzyme EBHXEEEQ is only slightly sensitive to iso- leucine (28); however, there has been no report on the Ep_zigg control over this enzyme in this organism. In E. coli and E, Ayphimurium, synthesis of biosynthetic threonine dehydratase is subject to multivalent repression; i.e., repression which is mediated by the simultaneous presence of excess isoleucine, valine and leucine (65). However, this mechanism of regulation is not ubiquitous, since in Corynebacterium ER (6) and Pseudomonas multivorans (39) multivalent repression of the enzyme by isoleucine, leucine and valine does not occur. Moreover, in the latter two organisms, the biosynthetic enzyme is the only form of threonine dehydratase found, it is synthesized constitutively, and is sensitive to feedback inhibition by isoleucine. In contrast to the E, multivorans enzyme, which has no direct role in threonine catabolism (39), threonine dehydratase from Corynebacterium.§p_was shown to have a catabolic function which is dependent on the con- comitant catabolism of branched-chain amino acids (6). In E. coli and E. typhimurium, the biodegradative form of threonine dehydratase is formed under anaerobic conditions, is induced by threonine and serine, requires adenylic acid for optimal activity, and is sensitive to catabolite repression (41, 49, 78). More recently, Yui ££.él.(77) reported that threonine, serine, aspartic acid, and methionine collectively function as inducers of the biodegradative threonine dehydratase in the Crookes strain of E, ggli_and proposed the term "multivalent induction." Valine, leucine and arginine are amplifiers of enzyme production. In con- trast, Egan and Phillips (22) reported that omission of serine from a complete synthetic medium containing 19 amino acids resulted in only a minor reduction in the synthesis of the enzyme by the same organism. Similarly, enzyme induction was not affected by omission of aspartate and methionine from the medium. Unfortunately, the requirement for aspartate could not be objectively assessed owing to the experimental conditions employed. It had been reported that omission of arginine curtailed induction of the enzyme (77). However, this was interpreted by Egan and Phillips (22) as resulting from a lack of an energy source capable of maintaining adequate arginine levels for protein synthesis. However, in spite of these apparent contradictions, the available data (22, 77) point to the conclusion that in the Crookes strain of E, 2211) threonine, leucine, and valine are essential components for induction of the biodegradative form of threonine dehydratase. Anaerobiosis is required for optimal synthesis of biode- gradative threonine dehydratase in E, 22;; (22, 64, 74). Oxygen was found to cause a rapid inactivation of purified threonine dehy- dratase in the absence of reducing agent (71). However, oxygen did not affect the enzyme stability ignzgzg_under conditions where further protein synthesis and growth was inhibited (22). This suggests that oxygen causes a metabolic condition that prevents further enzyme formation. Shizuta and Hayaishi (59) observed that cyclic adenosine 3',5'-monophosphabe (CAMP) reverses the catabolite repression caused by glucose and promotes the appearance of biodegradative threonine dehydratase in resting cells of E, 3911, and concluded that this represents a transcriptional effect similar to that reported for the CAMP stimulation of B-galactosidase. The absolute requirement for cAMP was later confirmed by Phillips, Egan and Lewis (51) in their studies using an adenylate cyclase mutant. CAMP was also required for the synthesis of the catabolic threonine dehydratase by E. typhimurium. This enzyme is immunologically different from the biosynthetic dehydratase in this organism (41). The proposed role of biodegradative threonine dehydratase is to provide ATP formation from threonine during anaerobic growth in an amino acid rich medium when ATP cannot be produced by means of oxidative phosphorylation (22). Such a role is further implied by the dehydratase activation by AMP (74). The mechanism of activa- tion has been shown to involve both changes in quaternary structure of the enzyme and specific facilitation of an early step in the reaction mechanism (53, 71). Presumably homologous with the effect of AMP on the E:.EQll enzyme is the effect of ADP on the enzyme from E, tetanomorphum (27). ADP promotes and helps to maintain the enzyme in aggregated state; however, the aggregation and dissociation appear to be a slower process with the ADP-activated enzyme than they are with the AMP-activated enzyme. .9. tetanomorphum dehydratase was shown to»'breakdown threonine into propionate and C02 as terminal products via the intermediates a-ketobutyrate and propionyl phos— phate, and the process was linked to ATP generation (62, 63, 65). Phillips, Egan and Lewis (51) recently showed that when an exponentially growing culture of E, gglE_Crookes strain was made anaerobic, a sharp increase in internal CAMP was noted, and the synthesis of biodegradative threonine dehydratase was detected immediately after the attainment of the peak CAMP levels and continued for several generations. Pyruvate addition at the time of anaerobic shock severely affected both CAMP accumulation and threo- nine dehydratase synthesis; however, externally added CAMP could partially counter the pyruvate effect on enzyme synthesis. It was thus concluded that conditions which resulted in a temporary energy deficit brought about the major accumulation of cAMP, and this elevated level served as a signal for initiation of threonine dehydratase synthesis to supply energy by the non-oxidative degra- dation of threonine. Isoleucine biosynthesis. Several pathways for isoleucine formation are known to occur in microorganisms (Fig. 1). These have been reviewed by Umbarger (66). The biosynthesis of isoleucine in most organisms initially involves the conversion of threonine to G~ketobutyrate by threonine dehydratase. The condensation of a—ketobutyrate with a two-carbon fragment followed by a succession of four reactions results in the formation of isoleucine. The enzymes of the latter four reactions are shared in valine biosyn- thesis. Exceptions to the threonine dehydratase pathway have been reported for a number of organisms (Fig. 1). These include rumen bacteria (3, 55), Bacillus subtilis (57), E, coli grown with B-methylaspartate (1), E, coli Crookes strain (50), Acetobacter suboxydans (7), Leptospira (15), and Serratia marcescens (36). Except for the rumen bacteria, a-ketobutyrate from sources other than threonine is the precursor for isoleucine synthesis in all C3888 . 10 Phospho- Aspartate ——>Homoserine ———)—homoserine ——> Threonine Methionine B-methyl- v Glutamate ——-> aspartate —)- a-ketobutyrate Pyruvate ——-)> Citramalate V 2—methylbutyric acid )I IsoleucinEl FIG. 1. Various pathways of isoleucine biosynthesis by micro- organisms. 11 While some species of rumen bacteria use exogenous amino acids (11), amino acids in peptides are more efficiently utilized for growth than free amino acids (52). Certain species utilize ammonia in preference to preformed exogenous organic nitrogen (3). Branched-chain volatile fatty acids are nutritional requirements for several groups of important ruminal bacteria. The acids are used for synthesis of branched-chain amino acids, branched-chain fatty acids and aldehydes, and probably for other cellular constituents possessing branched-carbon chains. Some representatives of the more important rumen bacteria were reported to synthesize isoleu- cine from 2-methylbutyric acid (55), presumably via reductive car- boxylation followed by amination reactions. Similar reactions are operative in the synthesis of leucine, valine, phenylalanine, and tryptophan from isovalerate, isobutyrate, phenylacetate, and indole acetic acid, respectively (3). However, not all of these bacteria require 2-methylbutyric acid for growth. For instance, Bacteroides ruminicola strain 23 was stimulated by a mixture of volatile fatty acids including 2-methylbutyric acid but these acids were not essential (10). The organism must be able to synthesize its branched-chain carbons, including isoleucine, since it was found to grow well when acetate was the only volatile fatty acid added to the medium (55). Isoleucine biosynthesis from 2-methylbutyric acid is subject to end-product control by isoleucine from casein peptides (55). The synthesis of amino acids from volatile fatty acids by rumen bacteria is a reflection of their habitat in which the concentration of amino acids is usually relatively low (free amino 12 acids are rapidly catabolized), whereas the concentrations of straight- and branched-chain fatty acids are high, these having been generated from the catabolism of peptides and amino acids (42). An enzyme found in E, subtilis,termed phosphorine deaminase, catalyzes the dephosphorylation and deamination of phosphohomoserine to a-ketobutyrate without the intermediate formation of threonine (57). The deaminase activity was found to be associated with threonine synthetase since both activities were shown to be affected by a single mutational event and coordinately derepressed. Growth experiments‘witha.strain deficient in biosynthetic threonine dehydratase conducted in the presence of l4C-L-homoserine plus 14C-L-threonine plus increasing excess unlabeled threonine or in concentrations of unlabeled homoserine showed the synthesis of isoleucine from homoserine without the intermediate formation of threonine. Abramsky and Shemin (1) demonstrated the conversion of B-methylaspartate exclusively to isoleucine in E, gglE_W. since all the radioactivity in the isoleucine synthesized from [CH3-140] B-methylaspartate was in 0-5, they suggested that B~methylaspartate is converted to a-ketobutyrate. Such labeling pattern would be obtained upon the conversion of [4-14C]:2u 233965: .acSEIQI 00...: 08:330.: I000 +£ZI.UII IOOU/ /W\I .LI “wInfirL .llllllll 0 .03 n1u\ /uoo 0' 06000530 . N .0: 002060530 «Iv O bow 1-wu1 _ 1o-n.v.un1 :80 m doom .3000 400.. M o. w 16 +. 22553 .5 Dow 295621234 .03 16:1 - _ n;QVIWIA¥a 4N I000 16 suggested that this lack of specificity does not seem to matter in wild type organisms because either the enzyme does not encounter significant amount of secondary substrate or the Km for oI-ketoisovalerate makes it the favored substrate. Most of the isoleucine revertants accumulated large amounts of leucine in the medium due to both constitutive synthesis of isoleucine resulting from.G-aminobutyric acid resistance and desensitization of a-isopropylmalate synthase resulting from isoleucine auxotrophy (35). It was found that partial reversion of isoleucine auxotrophy in leucine-accumulating revertants did not depend on the restoration of L-threonine dehydratase activity (36). The growth of the revertants was stimulated by B-methylaspartate, D(-)-citrama1ate or citraconate but not by glutamate, L(+)-citramalate or mesaconate. Also, no glutamate mutase was found in the cell extracts. It was therefore suggested that isoleucine is formed in the revertants from pyruvate by the leucine biosynthetic enzyme via Citramalate and citraconate as intermediates of a-ketobutyrate formation. The cell extracts of the revertants were shown to catalyze the formation of Citramalate from pyruvate plus acetyl coenzyme A and the isomer- ization of D(-)-citramalate to erythro-B-methyl-malate via citraconate. The formation of a-ketobutyrate from citraconate and not from mesaconate seemed to confirm the proposed pathway. MATERIALS AND METHODS Cultures and cultural methods. E, sporogenes (ATTC 7955, National Canners Association PA 3679) was used in all experiments. The growth media used were: (1) Medium A: "standard trypticase medium" consisted of 4.0% trypticase, 2 ppm thiamine hydrochloride, and 0.05% sodium thioglycolate. (2) Medium B: a synthetic medium containing salts, vitamins, and amino acids adapted from the medium of Perkins and Tsuji (48). Modifications of medium B (designated B-l through B-lO; Table l) were prepared by varying the amino acid composition. All media were prepared with deionized distilled water and adjusted to pH 7.4 before autoclaving. The culture was maintained by refrigeration of a culture that had sporulated in medium A. Vegetative cultures for inocula- tion of experimental media were initiated by inoculating tubes of this same medium (10 mI/tube) with 0.1 ml of the sporulated culture. These were heat-shocked (60°C for 10 min) and incubated for 6-8 h. All growth experiments were performed in an anaerobic chamber (Coy Mfg., Ann Arbor, MI) at 37°C. For growth in various synthetic media, the cells were preadapted in the corresponding medium. 17 18 .oumaoo Ihawownu anfipom we 0.m van .0Hum oaouamnocaamla w: H.0 .oawsmflzu w: 0.0 .cfiuOHa w: m00.0 .00mNmM we 3 38:? ma S .82 we .86 SN: 008»: we 36 69.08:: we 35 .091 $398 we 85 .ONm 5.0owmm we No.0 "HE 0H mo oESHo> Hmafim m C“ monocomEoo wafisoanom may woafimuaoo mficma HHIA 0m.HH 0m.HH 0m.HH 0m.HH 0m.HH 0m.HH 00.mH 0m.HH 0m.HH 0m.HH mcfioamHIg 00.~ m~.0 mN.0 mm.0 m~.0 mN.0 mm.0 mN.0 mN.0 mN.0 am500umhqua 00.0H 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 oaaamamahaonaIa 00.N 00.H 00.H 00.H 00.H 00.H 00.H 00.H 00.H 00.H mcfimouhqu OHIm mIm mlm 5Im 0Im mIm 0Im MIm NIm HIm unmaomaoo mfivoa :0 5280 coaumuucmoaoo umocowduomw .m.wo cowum>wuaau onu wow wow: waves ofiuonuamm m300um> mo m:oaa«monaou .H oHan 19 Preparation of cell extracts. Exponential phase cells were harvested by centrifugation at 18,000 X g, washed once with cold 0.1 M potassium phosphate buffer (pH 7.5), and resuspended in the same buffer (0.5 g of packed cells per ml). Cell extracts were prepared by ultrasonic oscillation of 1.5 to 2.0 ml of cell suspen- sion for four 15-sec intervals in a lOO-W ultrasonic disintegrator (Measuring and Scientific Equipment, Ltd., London) while the temperature was maintained below 20°C in an ice bath. The cell debris was removed by centrifugation at 20,000 X g for 20 min at 4°C. In all cases the cell extracts were passed through a 1.2 X 7.0 cm column of Sephadex G-15 to remove low molecular weights compounds. The extracts were tested immediately for enzyme activity. Enzyme assay. Threonine dehydratase was assayed by a modification of the lactic dehydrogenase coupled assay of Dunne g£_§l_(21). The incubation mixture (1.0 m1 final volume) in a cuvette with l-cm light path contained: 5 pmoles of dithiothreitol, 5 Umoles of ADP (pH 8.0), 0.15 Umoles of NADH, 15 Hg of rabbit muscle lactic dehydrogenase (Type II), 75 Umoles of potassium phosphate buffer (pH 8.0), 20 Hmoles of L-threonine (pH 8.0) and cell extract (”0.2 mg protein). The assay mixture was preincubated at 37°C for 5 min. The reaction was initiated by the addition of substrate and the rate of oxidation of NADH was followed at 340 nm with a Gilford 2000 recording spectrophotometer with the cuvette chamber maintained at 37°C by means of a Haake constant temperature regulator. One enzyme unit deaminates 1 Hmoles of threonine per min. 20 Threonine aldolase activity was assayed spectrophotometri- cally by a modification of the method described by Dainty (19). The reaction mixture (3.0 m1 total volume) contained 100 umoles potassium phosphate buffer (pH 7.0, 7.5 or 8.0) or Tris.HCl buffer (pH 8.5 or 9.0); 20 umoles of L-threonine; 0.3 Umoles of NADH; 10 units of alcohol dehydrogenase; and cell extracts (0.2-1.0 mg protein). Following a 5 min preincubation of the assay mixture at 37°C, the reaction was initiated by addition of substrate and the rate of oxidation of NADH at 37°C was followed at 340 nm as described above. Threonine dehydrogenase activity was measured colorimetri- cally by a modification of the method described by Komatsubara 2E .gl (37). The reaction mixture (1.0 ml total volume) contained 100 Umoles of potassium phosphate (pH 7.0, 7.5 or 8.0) or Tris.HCl buffer (OH 8.5 or 9.0); 0.5 Umoles NAD+ or NADP+; 20 Umoles of L-threonine; 100, 200 or 300 Umoles KCl and cell extract (0.2-1.0 mg protein). The reaction was initiated by addition of threonine and all samples were incubated at 37°C. After 30 min, the reaction was stepped by the addition of 1.0 ml of 0.3 M trichloroacetic acid (TCA) and deproteinized by centrifugation. Amino acetone in the supernatant was assayed by the method of Urata and Granick (67). The spectrophotometric assay for threonine dehydrogenase was performed with the same reaction mixture as described above. After a 5 min preincubation at 37°C, the reaction was initiated by the addition of substrate, and the NADH/NADPH formed was measured 21 by the increase in absorbance at 340 nm with a Gilford 2000 record- ing spectrophotometer described above. Fractionation of isotopically labeled cells. Cultures used for measurement of 14C02 uptake into cell protein were grown in Hungate tubes to prevent equilibration of label with atmospheric C02. All additions to or withdrawals from the tubes were made through a rubber septum using a sterile disposable syringe. All other labeling experiments were performed in 16 X 125 mm screw capped tubes. Six identical cell cultures (without label) were grown along with each culture that contained labeled substrates. Of the former, three were used to monitor growth, and three for measure- ment of cell protein. Incorporation of label into protein was determined by quantitation of radioactivity in individual amino acid fractions isolated following protein hydrolysis. The latter was performed by a modification of the method of Roberts SE 3; (54). Cultures were harvested by centrifugation at 18,000 X g for 15 min. The cell pellets were then extracted twice with the same volume (2.5 m1) of 5% TCA. The first extraction was performed at 0°C for 15 min and the second at 90°C for 30 min. The pellets were washed and resuspended in 1.5 m1 of 6 N HCl and transferred to ampoules. The ampoules were flushed with argon, sealed and incubated at 110°C for 24 h. The hydrolysates were dried under a stream of nitrogen at 80°C, dissolved in distilled water and adsorbed onto a cation exchange resin (AG 50W-X4, hydrogen form; Bio-Rad) packed in a 22 disposable pipette (1.6 ml resin bed). The resin was washed with 10 ml water, and the amino acids eluted with 1 M ammonium hydroxide. The eluates were dried as described above, dissolved in 200-500 pl water and stored at -10°C for further analysis. Fractionation of amino acids for protein hydrolysates. Three different methods were used to separate the amino acids in the radioactive hydrolysates, i.e. (a) thin layer chromatography (TLC); , (b) paper electrophoresis; and (c) amino acid analysis using an automatic amino acid analyzer. Amino acids were resolved by ascending chromatography on Whatman LK-S-D linear K preadsorbent thin layer plates (Whatman, Inc., Clifton, N.J.). The solvent system used was methyl ethyl ketone:pyridine:water:acetic acid (70:15:15:2, v/v). Although it did not separate leucine from isoleucine, this solvent system did resolve these two amino acids from all others in the protein hydrolysates. Unlabeled amino acid standards were spotted along with samples on each plate. To correct for losses in recovery of label, a non-radioactive protein hydrolysate to which known amounts of 1('C-isoleucine were added was applied to each plate along with the sample being analyzed. The plates were developed 3-4 times at room temperature for 1.5 h each with adequate drying in between developments. A guide strip (containing amino acid standards) on the chromatograms was sprayed with a solution of 0.1% ninhydrin in acetone and heated for 15-30 min in a drying oven. Areas of the unstained portion of the chromatograms corresponding to the leucine 23 and isoleucine spots on the guide strip were scraped with a spatula and the loosened adsorbent was collected in scintillation vials. The sample was extracted with 1 m1 of distilled water, then diluted with 10 m1 of aqueous scintillation fluid (Formula 963) and the radioactivity was counted. Electrophoresis of amino acids in protein hydrolysates was performed on Whatman no. 1 paper. Buffer system A (0.25 M sodium acetate, pH 4.3) was used to separate acidic from neutral and basic amino acids; buffer system B (0.05 M potassium phosphate, pH 7.5) was used to resolve lysine from all other amino acids except arginine. However, since E, sporogenes cannot synthesize arginine, unless radioactive arginine is added to the culture the arginine in the protein hydrolysate is not labeled. Protein hydrolysates were spotted on paper, along with amino acid standards, and electrophoresed at 35 volts/cm. The paper was then dried and a guide-strip containing the amino acid standards was stained with ninhydrin as described above. Areas of the unstained portion of the electrophoretogram corresponding to standards in the guide strip were cut out, folded, immersed in toluene-based scintillation fulid in glass vials and counted for radioactivity. Recovery of label in samples resolved by electro- phoresis was only 63-65% as determined with labeled amino acid standards separated by the same procedure. Amino acids from protein hydrolysates were quantitated on an analytical scale using a Beckman Model 121 programmable amino acid analyzer (Beckman Instrument, Inc., Palo Alto, CA). Program I, 24 with 0.2 M sodium citrate buffer, pH 2.99, containing 3% by volume of n-propanol, was used to separate aspartate, threonine, serine, glutamate, glycine, and alanine. Program II with 0.2 M sodium citrate buffer, pH 3.39, containing 3% by volume of methanol was used to separate valine, methionine, isoleucine, and leucine. All separations were performed at 52°C on a 0.9 X 65 cm column of spherical sulphonated styrene copolymer resin (Type AA-lS, 8% crosslinkage; Beckman Instruments, Inc.). The flow rates for Program I and II were 28 ml/h and 70 ml/h respectively. Before each run, the column was flushed with 0.2 N NaOH for 8 min (at 70 ml/h), then equilibrated for 35 min (at 70 ml/h) with the appropriate program buffer. After injection of the sample (600 Ml), the flow rate was adjusted to that required for each program. For analysis of radioactive samples, the elution stream from the analyzer was interrupted before reaction with ninhydrin and diverted to a Gilson Microfractionator Type FC-80K fraction collec- tor (Gilson Medical Electronics, Inc., Middleton, WI). The point at which the eluant stream was diverted was determined by reference to a calibration mixture elution profile which was used as a guide. Fractions were collected in the time mode at one-minute intervals. Every second or third fraction was quantitatively transferred to a glass vial, mixed with aqueous scintillation fluid (Formula 963) and the radioactivity was counted. In all cases, plots of radioactivity vs elution time gave discrete peaks which coincided to those obtained with calibration standards (Fig. 3). Fractions of individual peaks were pooled and the total label was 25 .ocwummm00HIDHIA wcwcfimucoo 0HIm Bsfivme ca CBOHw mmcuwouoam .m.Eoum cwmuoua wmuhaouwha mo wvwom ocwfim :« 00H mo cowuanwuumfin .m .on mwFDZE’. 00 08 05 00.4. co. co. 2.. co. 8. c! .0». pm. 0.. oo. oo 1 J I .I I 4 II .I I4 ItaduJIlJ so... .0) Icon £40 100.. Icon 0 Eccoen. o I can _H 1000. J oou. Ohm own on» O?» on» ONflKx OQN OnN CNN \ OO— 00. ON. 00. On. O! _ _ _ \ J OON I. 00¢ .l 000 l 000 A 000. l CON. 1 00! . l 000. NM] / SlNflOD 0.4 l OOON 3.0 L coon and 26 quantitated by counting duplicate aliquots. The average value was used to calculate the total radioactivity in each amino acid pool. All values were corrected for quenching caused by the elution buffer; the counting efficiency was 80.0-83.5% as measured against an internal standard. The recovery of label by this method based on analysis of a mixture of radioactive amino acid standards was 98-100%. Determination of intracellular amino acidgpools. Cells grown in 50 ml medium B-lO were harvested at a culture density of 0.3 O.D. units by vacuum filtration through Metricel membrane filters (25 mm diameter and 0.2 pm pore size; Gelman Instrument Co., Ann Arbor, MI), and each filter washed with 5% TCA in the cold as described above. The solution was extracted five times with equal volumes of ether to remove TCA, and adsorbed on a cation exchange resin packed in a disposable pipette. After washing, the amino acids were eluted with l M ammonium hydroxide as described above. The eluate was dried as described previously and the crystalline amino acids were dissolved in a small volume of water. Amino acid concentrations were determined with an amino acid analyzer. Decarboxylation of radioactive amino acids. Monocarboxylic amino acids and glutamic acid were decarboxylated with chloramine-T according to the procedure of Kemble and McPherson (33). The radioactive sample was placed in a Warburg flask with 0.5 m1 of 12% chloramine-T solution in the side arm and 0.4 ml of hydroxide 14 of hyamine in the center well to trap the C02. The reaction was initiated by tipping the chloramine-T solution into the sample. The 27 reaction was allowed to proceed for 2 h at 30°C. At the end of incubation, the contents of the center well were transferred to a glass vial; the well was rinsed with 1 m1 of methanol and the wash- ings added to the same vial. Ten ml of toluene—based scintillation fluid was added to the vial and the radioactivity was counted. A vial containing 0.4 m1 of hydroxide of hyamine and 1 m1 of methanol in 10 ml of the same scintillation fluid served as a reagent blank. Aspartic acid was decarboxylated with ninhydrin by a modification of procedure of Greenberg and Rothstein (26). The l°C02 produced in the reaction vessel was driven by a stream of nitrogen into two receiving vessels containing barium hydroxide solution. The precipitated barium carbonate was collected and transferred to a Warburg flask containing 0.5 ml of 2 N HCl in the side arm and 0.4 m1 of hydroxide of hyamine in the center well. The flask was sealed, and the acid tipped into the sample. The flask was held at room temperature for 1-2 h to allow for complete absorption of the liberated 1°C02 by the alkali. The rest of the procedure was as described above. In both procedures, the buffer was that in which the indi- vidual radioactive amino acids were eluted from the amino acid analyzer since the molarities and the pH of the buffers were suitable for this purpose. Incorporation of the degradation product(s) of L-[U-1°C] isoleucine into isoleucine. Ten HCi of L-[u—140]isoleucine was added to a 10 ml culture of E, sporogenes as soon as growth was observed, and the culture incubated until it reached a turbidity of 28 0.3 O.D. units. The cells were harvested by centrifugation and discarded. The supernatant was acidified with HCl to pH"¢3.8-4.0, and shaken vigorously to remove all the C02. The solution was passed through 6 columns, one after the other, of cation exchange resin (AG 50W-X4, hydrogen form, 9.6 m1 total volume) packed in disposable pipettes to remove all amino acids and other amines. A medium was prepared with the resulting solution by adding a mixture of amino acids, salts, vitamins, and sodium thioglycollate in the amounts and proportions corresponding to medium B-lO (Table 1). After the pH was adjusted to 7.4, the medium was flushed with argon for 10 min, stOppered, placed in the anaerobic chamber and sterilized by filtration. This medium was used for culturing cells which were harvested at a culture turbidity of 0.6 O.D. units. Cell protein was isolated and hydrolyzed as described above and the hydrolysate analyzed by means of an amino acid analyzer. Measurement of volatile fatty acids. Volatile fatty acids were extracted from acidified supernatant solutions (from either a growing culture or from cells incubated in a solution containing 100 mM L-isoleucine and L-proline in 0.2 M potassium phosphate buffer; pH 7.5), and were qualitatively identified and quantitated by gas chromatography according to the procedure as described in the Anaerobe Laboratory Manual (29). A varian model 1420 gas chromatograph, equipped with a thermal conductivity detector, was used. The column was stainless steel (0.125 inch by 6 feet [ca. 0.32 by 182.9 cm]) and contained 15% SP 1220-l% H PO on Chromosorb 3 4 W AW (100/120 mesh; Supelco, Inc., Bellefonte, PA). Helium was the 29 carrier gas (25 ml/min), and temperatures were: column, 135°C; injector and detector, 165°C each. Volatile fatty acids were applied to the column as ether solutions. Preparation of [U-1°C]2‘m€thy1bUtYriC aCid- [U-1°C] 2—Methylbutyric acid was prepared by fermenting L-[U—1°C]isoleucine in the presence of L-proline using E. sporogenes cells. The cells were grown in 100 ml medium-A, harvested in the exponential phase of growth and washed twice with 0.1 M potassium phosphate buffer, pH 7.4. The cells were then suspended (1.35 g of packed cells/ml) in a 2.0 ml solution containing 10 mM L-isoleucine, 20 uCi L- [U-1°C]isoleucine (sp.act. 360 UCi/ mole), and 20 mM L-proline. The whole mixture was allowed to incubate in the anaerobic chamber at 37°C for 4 h. Following centrifugation, the clear supernatant fluid was passed through a column of cation exchange resin (AG 50W-X4, hydrogen form) packed in a disposable pipette (1.6 ml resin bed) to remove the remaining amino acids. The fluid that came through (containing the volatile fatty acids) was collected and the total radioactivity was determined. Although the latter fraction may have contained traces of other volatile fatty acids, practically all of the radioactivity was expected to reside in 2-methylbutyric acid since this is the major volatile fatty acid obtained from the fermentation system employed (see Table 21). The yield of the latter was 28% based on radioactivity counting corre- sponding to a 2.8 mM concentration in 2.25 ml volume with a specific activity of 0.967 uCi/umole. 30 Analytical methods. Growth of E, sporogenes was estimated by measuring optical density (O.D.) at 600 nm using a Mini Spec 20 spectrophotometer (Bausch and Lomb Optical Co., Rochester, N. Y.). The protein content of cell extracts and cell pellets obtained after cold TCA extraction were determined according to the methods of Kalb and Bernlohr (32) and Lowry 23 El (40), respectively. All radioactive counting was performed on a Packard Tri-Carb liquid scintillation spectrometer, model 3320 (Packard Instrument Co., Inc., Downers Grove, IL). Chemicals. Amino acids of the highest purity available were purchased from several different companies; those that were used in medium B-10 were all purchased from Sigma Chemical Co., St. Louis, MO. Dithiothreitol was obtained from Aldrich Chemical Co., Milwaukee, WI; ADP was from General Biochemicals; NAD+, NADP+, NADH, rabbit muscle lactic dehydrogenase enzyme (Type II), alcohol dehy- drogenase enzyme, and chloramine-T were from Sigma Chemical Co., St. Louis, MO. Volatile fatty acids were purchased from Eastman Kodak Co., Rochester, N.Y. and Fisher Scientific Co., Fairlawn, N.J. The following radioactive compounds were purchased from New England Nuclear, Boston, MA: [U-14C1acetate, L-[U-1°C]methionine, L-[U-14C]arginine, L-[U-l°C]orthinine, L-[U—l°c]histidine, L-[U-1°C]valine, L-[U-l4c]aspartate. L-[U-14c1threonine. L—[U-1°C] glutamate, L-[U-14C]isoleucine, [3—14c1pyruvate. 1°C—NaHC03, and ll'C-toluene. L-[u-14C]phenylalanine, and L_[U-l°c]tyrosine were purchased from International Chemical and Nuclear Corp., Irvine, 14 CA. [U-14C]glucose, L—[U—l°c]proline, and L-[U- C]leucine were purchased from Amersham Corp., Arlington Heights, IL. 31 CA. [U—1°C]glucose, L—[U-14C1proline, and L-[U-14C]leucine were purchased from Amersham Corp., Arlington Heights, IL. Hydroxide of hyamine was purchased from Packard Instrument Co., Inc., Downers Grove, IL. The aqueous scintillation fluid (Formula 963) was obtained from New England Nuclear. Toluene— based scintillation fluid was prepared by mixing 6.00 g of 2,5—diphenyloxasole and 0.01 g l,4-bis-2-(5—phenyloxazoy1)-benzene in one liter of toluene; both chemicals were obtained from Research Product International Corp., Elk Grove Village, IL. RESULTS Threonine Metabolism Threonine dehydratase activity. If both biodegradative and biosynthetic threonine dehydratases are present in E, sporogenes, it is reasonable to expect differences to occur in the specific activity of the enzyme from cells grown in media of varying composi- tion. Table 2 shows that none of the growth media tested produced significant variation in the threonine dehydratase activity as compared to that in cells grown in complete synthetic medium (con- trol). Omission of isoleucine failed to increase the activity as one might expect if a derepression of biosynthetic enzyme had occurred. Furthermore, the presence of excess isoleucine along with high concentrations of valine and leucine (medium B-l contains 17 mM L—valine and 11.5 mM L-leucine) did not reduce the enzyme activity as would be expected for a system in which multi- valent repression of enzyme synthesis was operational. Moreover, the enzyme is apparently not subject to catabolite repression, since no variation in activity was caused by glucose. No difference in specific activity was observed in cells grown in the presence of excess threonine or in the absence of threonine and/or serine, in contrast to the result expected if the catabolic enzyme were inducible by eithercnrboth amino acids (medium B-l contains 8.4 mM 32 33 Table 2. Effect of growth medium on threonine dehydratase activitya Specific Activity Growth Medium (units/mg protein) Standard trypticase (medium A) 0.725 Synthetic: complete (medium B-l) 0.951 no isoleucine 0.721 + 30 mM L-isoleucine 0.879 + 30 mM glucose 0.856 no threonine 1.146 no threonine and no serine 0.962 + 22 mM L-threonine 1.009 + 42 mM L-threonine 0.988 aAssay procedure is as described under Materials and Methods. Table 3. The effect of AMP and ADP on threonine dehydratase activity at high and low substrate concentrations.8 Specific Activity (units/mg protein) Nucleotide 20 mM L-thr 4 mM L-thr None 1,050 0.431 + 5 mM AMP 1.050 0-431 + 5 mM ADP 1.050 0-812 aAssay mixture was as described in Materials and Methods. The cells were grown in synthetic medium (B-1) and the crude extract was desalted as described. 34 L-threonine and 9.5 mM L-serine). The role of leucine and/or valine in the induction of the enzyme could not be assessed since both amino acids were essential for growth in the medium used. Table 3 shows that threonine dehydratase is activated by adenosine diphosphate (ADP) but not by adenosine monophosphate (AMP); however, the activation by ADP was only observed at low substrate concentration. From the Lineweaver-Burk plot shown in Fig. 4, the Km values for the enzyme in the presence and absence of ADP were calculated to be 2.50 mM and 33.33 mM, respectively; under these conditions the Vmax values were 0.033 and 0.045, respectively. Thus, at least one effect of ADP on the enzyme is a l3-fold reduc- tion in its Km for threonine. Since threonine dehydratase has been studied extensively in other organisms including Clostridium tetanomorphum (65), the investigation of this enzyme in E, sporogenes was not pursued any further. However, the failure to detect an isoleucine-sensitive enzyme (addition of L-isoleucine up to 5 mM to the assay mixture at pH 7.0, 8.0, and 9.5 did not affect the specific activity) prompted the investigation of the isoleucine biosynthetic pathway in this organism. Threonine aldolase and dehydrogenase. No threonine aldolase or threonine dehydrogenase activities were observed in E, gporogenes cell extract (cells were grown in medium A) under the assay condi- tions employed. Addition of 5 mM dithiothreitol to the assay mixture did not result in the detection of threonine aldolase 35 FIG. 4. Velocities of threonine dehydratase as a function of L-threonine concentration in the presence and absence of ADP. 36 90589.5 h; .I—Z . _ 000. 000 000 00? CON 0 CON On. 00. On 0 _ _ a _ d _ _ ON] 0.? 00 5,. _ 00 00_ EC..— W.N R EV. SE mm.mn R EV— :zEnv nod :13 and 50:13 0N— r _ _ F _ _ _ 0N 0¢ 00 ._..|> 8 00. ON. 37 activity. Cell extracts obtained from E, sporogenes grown in the synthetic media (medium B-l with or without the addition of 28.5 mM L-leucine) did not show any threonine dehydrogenase activity. In E, coli K-12, threonine dehydrogenase is induced by L-leucine and not by its substrate, L-threonine (47). Amino Acid Composition of Cells and Media Composition of cell protein. The amino acid compositions of proteins from cells grown in either B-8 or B-lO media were essen- tially the same (Table 4). The concentration of aspartate consti- tutes that of both aspartate and asparagine; similarly, the concen- tration of glutamate constitutes that of glutamate and glutamine. It is probable that both glutamate and alanine from peptidoglycan in the cell wall are included in these values. Calculation of specific activities of radioactivity incorporated into the indivi- dual amino acids are based on the data obtained with cells grown in medium B—lO. Free amino acids in cell. During the exponential phase, most amino acids in the intracellular pool were present in rela- tively low concentrations except for proline, glutamate, and valine (Table 5). The unusually high concentration of the latter three amino acids might be due to the necessity of the cells to balance the osmotic pressure exerted by the growth conditions in medium B-lO (45). Amino acids in growth medium. The concentrations of some amino acids in the growth medium at various stages of growth in 38 Table 4. Amino acid composition of protein from cells grown in various synthetic mediaa Concentration (nmoles/mg protein) Amino Acid Medium B-8 Medium B-lO Aspartate 771 752 Threonine 365 352 Serine 314 381 Glutamate 930 853 Proline 139 244 Glycine 533 522 Alanine 942 939 Valine 629 615 Methionine 346 unresolved Isoleucine 593 518 Leucine 628 568 Tyrosine 182 291 Phenylalanine 255 337 Lysine 625 586 Histidine 140 157 Arginine 265 218 aCells grown in medium B-8 and B-10 were harvested at the culture density of 0.7 and 0.6 O.D. units respectively. Proteins were isolated and hydrolyzed as described in Materials and Methods. 39 Table 5. Intracellular concentrations of amino acids in E. sporogenes during exponential growth3 Amino Acid Concentration. . (nmoles/mg protein) Proline 67 Glutamate 52 Valine 38 Serine 13 Alanine 12 Tyrosine 9 Leucine 6 Aspartate 6 Glycine 5 Lysine 3 Threonine l Methionine 1 Histidineb 1 Isoleucine l Arginine undetectable Phenylalanine unresolvedc aCells were harvested when the turbidity of the culture was 0.3 O.D. units. Amino acids were resolved and quantitated by means of an amino acid analyzer. bEstimated by comparing the area under their peaks (by weighing method) to those of known concentrations in the same chromatogram. cDue to the presence of an unidentified ninhydrin-positive substance that overlapped with phenylalanine in the chromatogram. 40 medium B-lO from three separate experiments are presented in Table 6. Although there is discrepancy on the rate of serine degradation among the different determinations, most of serine was consumed before the culture reached a density of 0.6 O.D. units. Both serine and arginine seem to be limiting at higher O.D. values. A significant amount but not all of leucine was consumed during the growth period. Of particular interest is the accumulation of alanine in the growth medium accompanying the rapid rate of serine utilization during the exponential phase (notice that alanine was initially absent). Isoleucine Biosynthesis Compounds contributing no significant carbon to isoleucine. The first attempt to determine the precursor of isoleucine was made by growing E, sporggenes in the presence of L-[U-l401threonine, and then measuring the amount of radioactivity in the isoleucine fraction isolated from whole-cell protein hydrolysates. No 1°C was found in isoleucine which indicated that this amino acid is not synthesized via the usual pathway. Therefore, other labeled components were added to the medium one at a time to find the pre- cursor of isoleucine. Table 7 shows that there was substantial incorporation of radioactivity into cell protein from all 1"C- compounds tested, but negligible incorporation of label into isoleucine and/or leucine. The low level of radioactivity recovered in the isoleucine + leucine fraction from a few substrates may have come from 1°C02 released on degradation of the labeled amino acids. 41 mums/H0005" HOG fl M20 50.0 55.~ N0.0 00.m 0m.0 mN.N N0.~ 00.0 ~0.0 0m.0 0.0 0:0:m0< 00.0 00.0 m5.m 00.0 00.n0 00.0 00.0 00.00A 0N.m 0N.0 0.0m 0:0:0wu< mz 00.m 00.0 mN.0 00.0 50.m N0.0 05.0 002 mmz 0.00 m:0:m0m05:onm 5m.0 00.0 50.0 0m.0 00.0 mw.0 mm.0 00.0 0~.0 00.0 0.N 0:0m005H 00.~ 00.0 00.0 00.0 50.0 00.m 00.0 00.00 00.5 00.0 n.00 0:00:00 05.m mm.0 00.m no.0 mm.w 05.m 0m.m 00.0 m0.5 0m.w 0.00 0:0:O0suoz 05.m 50.0 0m.0 mm.0 50.m0 0N.0 00.00 0N.m0 -.00 N0.m0 m.50 0:00m> mm.0 mm.0 mm.0 00.N 50.00 00.0 00.00A 00.n0A 00.0 m0.0 0.m~ 0:0uom m:o0umuu:oo:oo XE m:00umuu:oo:oo ZS m:00umuu:oo:ou 28 55.0 mwm.0 Nm0.0 0w~.0 00N.0 0N0.0 50m.0 500.0 000.0 00m.0 :8 "mo 000.0.0 um m .nxm "we 000.0.0 u: N .axm "mo 000.0.0 um 0 .mxm 0:000:H nuaoum mo mowmum m500um> um 00Im 650008 :0 :Bouw mousu0ao mo mv0=c00 u:mum:uma:m map :0 m:00umuu:ou:oo 00am o:0E< .0 o0nMH 42 .a0:o 0:00:00000 :0 0: 00\Eavw 0005002 0:0 00000000: ”AnamuwoumEouso 00500 :02”. .3 00000 o:080 00:00 800.0 000000000 003 0:00:00+0:00:00 IowH .0:00:00 8000 0:00:00000 00000000 00 >0000000: 00: 003 00 .Ahcmmquumeouzu 00:00 00000000 50 0:00:00 8000 000000000 003 0:00:00000 50053 0000 0:00:00m000I0HI0 £003 0000000 0000000 000 uawuxmo .0muum was 00; m :0 Hmuou «p .000000 susouw 0u0u:0 0:0 usonwsounu 0050000 050 :0 0:00000 003 0000000 .u:080u0ax0 £000 :00 000v oom.m 050.5m 000.0 o.~ 00000000 00:0 OOHV Noo.n 00m.0a ooH.o o.o0 ma0=m0masamnm oHIm 000v 000.0 000.00 000.0 0.00 0:0:o0500z 00Im co m00.m mma.m00 II II amumumu< OHIm 000v m5n.0 5N5.m0 000.0 m.0 0:000:00: wIm 000v mm0.00 0mo.00m emo.o o.o~ mmo0:00 mum m0 m5m 005.00 mmo.o o.m0 0:00000 cum mm m0~ omo.5 mmo.o o.m0 000300000 5Im mHN 0m0.m mam.wm omo.o o.o~ me0c0wu< mIm 00o: m~0.0 a:~.w0 mmo.o o.mH mc0u=ma 0Im 05 m0m.m oom.0m mmo.o o.m0 0:00m> mum m0 0mo.5 00m.05 mmo.o o.m0 0:0:owuse «:0 0000+000 000oH :000000 A00oE:\0020 000000000 000000000 80000: mummzwouvm: mo 0100\200 we\a:0 500>0uu< :5 Ho IDHIA £03000 500>0uomo0wmm 00000000 00 00000000000 0000000 m:00um> mo 00:00000 0:0 :0 :3ouw 00:0wduomm .0 mo m0ummm00005: :000000 Eoum 500>000000000 mo hu0>oo0m .5 0000a 43 Some of these labeling experiments were conducted in media containing a number of amino acids not required for growth of g, sporogenes (media B-2 through B-7, see Table 1). This could reduce the potential for incorporation due to dilution of some common intermediate. In addition, there could be a reduction in the 14C found in isoleucine from rapidly metabolized substrates such as threonine, arginine, valine and leucine due to protein turnover in stationary phase cells. This is particularly likely of a spore 14 forming bacterium such as g, sporogenes. Therefore, further C incorporation experiments were performed with some of the amino acids in the minimal medium (B-10) and the cells incubated with labeled substrate for only one generation. Table 8 shows that under these conditions, uptake of label into protein was very high, but there was negligible incorporation into the isoleucine/leucine fraction except when leucine was the labeled substrate. Further analysis of the hydrolysate from cells labeled with L-[U-14C]leucine showed that at least 98% of the label in the isoleucine + leucine fraction was in leucine. Both threonine and glutamate provide a-ketobutyrate for isoleucine biosynthesis in some bacteria. Therefore, the protein hydrolysates from cells labeled with these amino acids were examined in more detail. Practically all of the label from L-[U-14 C]threonine was in the threonine fraction (Table 9). However, a small but sig- nificant amount was present in glycine. All of the label from L—[U-14C]glutamate was found in the glutamic acid present in the protein hydrolysate (Table 10). These data support the conclusion 44 Table 8. Incorporation of radioactivity into isoleucine of cell protein by g, sporogenes labeled for one generation with various 1[*C-amino acidsa Specific Radioactivity Incorporated L-[U-MC]b mM Activity dpm/mg dpm/10p1.hydrolysate Substrate Substrate (pCi/umole) Protein Total ile+leuC Arginine 25.0 0.1 76,195 3,200 <100 Valine 17.0 0.1 59.079 2,580 <100 Leucine 11.5 0.1 95,524 4,000 3,800 Aspartate -— -- 887,550 29,370 <100 Threonine -- -- 3,406,100 114,010 <100 Glutamate -- -- 402 ,650 13 ,480 <1oo aIn each experiment the label was added to 10 ml culture grown in medium B—lO at a culture density of 0.3 O.D. units. bA total of 25 uCi of each of the followinglgere added: L-[U-IAC] aspartate (spec. act. 221 uCiigmole), L-[U- C]threonine (Spec. act. 210 uCi/Umole) and L-[U- C]glutamate (spec. act. 266 pCi/ pmole). cIsoleucine+leucine was resolved by thin layer chromatography (Materials and Methods). 45 Table 9. Distribution of radioactivity in amino acids from protein hydrolysate of g, s oro enes cells labeled for I §5CI one generation with L- U- threoninea b Total Radioactivity Specific Activity Amino Acid nmoles (dpm) (dpm/nmole) Glycine 17 4,818 283 Aspartate 25 346 14 Threonine 12 121,505 10,125 Lysine 23 211 9 Glutamate 28 314 ll Isoleucine 17 40 8 Leucine 19 128 7 Total recovered 127,362 (>100%)C a25 uCi L-[U-14C]threonine (specific activity 21011Ci/umole) was added to 10 m1 culture in medium B-10 at cell density of 0.3 O.D. units. bLysine was separated on paper electrophoresis; all other amino acids were separated and quantitated with an amino acid analyzer (Materials and Methods). CTotal radioactivity in the amount of hydrolysate analyzed was 123,537 dpm. Table 10. Distribution of radioactivity in amino acids from protein hydrolysate of g, s oro enes cells labeled for one generation with L-lU-IECIglutamatea b Total Radioactivity Specific Activity Amino Acid nmoles (dpm) (dpm/nmole) Glutamate 34 16,335 480 Aspartate 30 8 <1 Lysine 23 51 2 Neutrals -- 6O --- a25 UCi L-[U-14c]glutamate (spec. act. 266IJCi/Umole) was added to 10 ml culture in medium B-lO at cell density of 0.3 O.D. units. bThe amino acids were separated by paper electrophoresis (Materials and Methods). 46 that isoleucine is not synthesized by g, sporogenes via a classical pathway. The specific activity of the 14C-threonine incorporated into protein was about 20 times that of 1l‘C-glutamate even though the specific activities of both amino acids added to the medium (see Tables 9 and 10) were essentially the same. This may have resulted from a large dilution of the added labeled glutamic acid by pool levels (see Table 5) or by continuous rapid synthesis of glutamic acid during the one generation of growth. Incorporation of 14C from L-[U-14C]Serine into isoleucine and other amino acids. All the previous experiments designed to incorporate label into isoleucine from various radioactive substrates were without success. Since serine was the principal (if not the sole) source of pyruvate in the minimal growth medium (B-lO) employed, one would expect the incorporation of label from L—[U—lhc]serine into most amino acids which the organism can synthesize. This proved to be true (Table 11). The relative specific activities of the individual amino acids per mole or per carbon atom are within reason. Serine had the highest specific activity as might be expected since it would not be diluted by C02 exchange or uptake reactions or by pyruvate from other sources. Thus, the lower specific activity of alanine compared to serine in the cell protein could result from dilution by alanine in the pool formed prior to addition of labeled serine to the culture (see Table 5). The specific activity of glycine is less than two thirds that of serine, suggesting that glycine is not synthesized exclusively from serine. The specific activity 47 Table 11. Distribution of radioactivity in amino acids from protein hydrolysates of 9, one generation with L- U- oro enes cells labeled for s -1L1fj;p¢-T . a C serine b Total Radioactivity Specific Activity Amino Acid nmoles (dpm) dpm/nmole dpm/C atom Alanine 82 11,631 142 47 Serine 33 7,827 237 79 Glycine 46 4,753 103 51 Aspartate 66 12,127 184 46 Threonine 31 4,053 131 33 Lysine 51 11,775 231 39 Glutamate 75 18,359 245 49 Isoleucine 45 5,411 120 20 Valine 54 442 8 —- Total recovered 76,337 (87%)C a25 uCi L-[U-14C]serine (spec. act. 162 UCi/Umole) was added to a 10 ml culture in medium B-lO at a culture density of 0.3 O.D. units. bLysine was separated on paper electrophoresis; all other amino acids were resolved by means of an amino acid analyzer (Materials and Methods). CThe total radioactivity in the sample of hydrolysate analyzed was 88,051 dpm. Table 12. protein hydrolysate of C. Decarboxylation of amino acids from oro enes cells 3’57ng grown in the presence 3f L- U- C serine Radioactivity (dpm) Percent Amino Acid Initial 14002 Label Lost Alanine 1,468 508 34.6 Serine 1,111 281 25.3 Glycine 728 415 57.0 Aspartate 2,160 1,073 49.7 Threonine 822 204 24.8 Lysine 2,599 333 12.8 Glutamate 2,310 418 18.1 Isoleucine 250 66 26.4 48 of aspartate is consistent with its synthesis from oxalacetate. Furthermore, the approximately equal distribution of specific activity among aspartate, threonine, and lysine suggests that the latter two amino acids are synthesized from aspartate via the usual pathway as described in E. coli (66). The specific activity of glutamate suggests its synthesis from a-ketoglutarate rather than from proline. Isoleucine contained significant levels of label derived from the l4C-serine. Its specific activity was only slightly less than that of alanine and was approximately equal to that of threonine. No significant amount of radioactivity was found in valine and the same was true for leucine although the latter amino acid was not completely resolved. Small amounts of label might be incorporated into valine and leucine by C02 exchange reactions. These data pro- vide good evidence that there is some synthesis of isoleucine but no significant synthesis of the other two branched chain amino acids under the conditions employed. Decarboxylations of the individual radioactive amino acids from cell protein was performed to help determine which carbon atom(s) had been labeled. With the exception of isoleucine, the results (Table 12) are not different from what would be expected if all of the amino acids examined were uniformly labeled. This would be predicted if all were formed from labeled pyruvate, acetate, and CO2 from 14C-serine. The amount of radioactivity lost during iso- leucine decarboxylation (26.4%) suggests that there are 2 or 3 carbons which are not labeled. 49 Incorporation of radioactive carbon from 14C02 into isoleu- cine and other amino acids. To determine the incorporation pattern of CO2 into cellular protein, two cultures of g, sporogenes were labeled for one and two generations respectively, in medium B-lO containing 14C-NaHCO3. In both cases, incorporation of total label per mg protein and the distribution of label among the amino acids was essentially the same (Table 13). In both experiments, the starting l4C02 specific activity was subject to dilution by the pre-existing non-labeled CO2 resulting from catabolism prior to add- ing the label, and would be further diluted during growth. However, the observation that the amino acid labeling pattern was unchanged by doubling the incubation period to two generations indicates that the amount of dilution during growth had little effect. The overall distribution of radioactivity among the indi— vidual amino acids in the protein hydrolysate obtained from the culture labeled for one generation was determined with the aid of an amino acid analyzer (Table 14). The nearly equal specific activities in alanine, serine, and glycine suggest that the label originated from an exchange reaction of pyruvate with 14CO probably 2 via pyruvate synthase. However, the reaction is probably rather. limited in this system, as evidenced by the low incorporation of label into these amino acids. The abundance of radioactivity in aspartate suggests that the latter is synthesized from oxalacetate which obtained its label from the carboxylation of pyruvate. The equal distribution of label in aspartate, threonine, and lysine supports the earlier suggestion that the latter two amino acids 50 Table 13. Distribution of radioactivity in amino acids from protein hydrolysates of _C_. gorogenes cells labeled for one and two generations with 140028 Radioactivity (dpm/mg protein) b Labeled for Labeled for Amino Acid one generation two generations Aspartate 162,036 153,969 Glutamate 176,438 141,221 Lysine 104,656 134,529 Neutral 265,751 294,682 Basic 126,514 159,924 a50 uCi 14C-NaHCO3 (spec. act. 6.2 uCi/Dmole) was added to 20 ml culture in medium B-lO at the cell density of 0.15 O.D. units. Ten ml of the culture was harvested at culture density of 0.3 O.D. units and the remaining ten ml was harvested at cell density of 0.6 O.D. units. bFive- and 10 pl aliquots of protein hydrolysate were separated by paper electrophoresis and the values listed were calculated based on 65.0% recovery of total radioactivity in the samples. Direct counting of the hydrolysate resulted in total radioactivity of 730,146 and 684,355 dpm/mg protein for cultures labeled for one and two generations, respectively. 51 Table 14. Distribution of radioactivity in amino acids from protein hydrolysates of g, sporogenes cells labeled for one generation with 14002a b Total Radioactivity Specific Activity Amino Acid nmoles (dpm) (dpm/nmole) Alanine 93 5,282 57 Serine 38 1,484 39 Glycine 51 1,435 28 Aspartate 74 17,364 235 Threonine 35 7,771 222 Lysine 58 10,916 188 Glutamate 84 17,925 213 Isoleucine 51 8,404 165 Valine 61 755 12 Total recovered 71,336 (977.)c a50 uCi 14C-NaI-ICO3 (spec. act. 6.2 uCi/umole) was added to 20 m1 cul- ture in medium B-lO at a culture density of 0.15 O.D. units. bLysine was separated by paper electrophoresis; all other amino acids were resolved by means of an amino acid analyzer (Materials and Methods). CTotal radioactivity in the sample of hydrolysate analyzed was 73,848 dpm. 52 are synthesized from aspartate via the usual pathway as described in E, gpli_(66). The nearly equal distribution of label in glutamate and aspartate suggest that aKG, which is the precursor of glutamate, is synthesized largely via the TCA cycle. A very significant amount of label was incorporated into isoleucine. The specific activity of the isoleucine formed is reasonably close to that found in aspartate, lysine and threonine. This indicates that there may be one C02 incor- porated per isoleucine synthesized by g, sporogenes. The low level of radioactivity in valine is consistent with the earlier suggestion that some label in valine and leucine could originate from a 14C02 exchange reaction. The results obtained from decarboxylation of each of the amino acids labeled with 14CO2 are shown in Table 15. Except for isoleucine, the amount of label lost from each oftfluaamino acids examined was consistent with 14C02 exchange reactions and with recognized biosyn- thetic pathways. Thus, the exchange of 14CO2 with pyruvate could account for all the label being lost from alanine and a small percen- tage of it from threonine and lysine. Similarly, reversible trans- amination and C02 exchange with a-ketoisovalerate could result in the 14 small amount of C in the carboxyl group of valine. The 100% loss of label from aspartate is as expected since both carboxyls are removed from this amino acid by the ninhydrin reaction used. Failure to find 14C02 from the C-1 of glutamate indicates thattfiuasynthesis of citrate in 9. sporpgenes involves a rgfcitrate synthase as in the case of Clostridium kluyveri and few other bacteria (23, 24). This 53 Table 15. Decarboxylation of amino acids from protein hydrolysate of g, sporogenes cells grown in the presence of 14C02 Radioactivity (dpm) Percent Amino Acid Initial 1.4002 Label Lost Alanine 704 712 101 Aspartate 580 601 104 Threonine 1,460 297 20 Lysine 1,773 377 21 Glutamate 1,261 0 O Isoleucine 481 182 38 Valine 112 120 107 Serine8 N.D. N.D. --- Glycinea N.D. N.D. —-- aNot determined due to insufficient amount of radioactivity in the sample. Table 16. Distribution of radioactivity in amino acids from protein hydrolysate of g, s oro enes cells labeled for one generation with IB-EECIpyruvatea Total Radioactivity Spec. Act. Spec. Act./ Amino Acid nmoles (dpm) (dpm/nmole) Spec. Act. alanine Alanine 63 17,458 277 1.0 Serine 25 10,968 438 1.6 Glycine 35 1,305 37 0.1 Aspartate 50 16,406 328 1.2 Threonine 23 6,830 297 1.1 Lysine 39 21,603 554 2.0 Glutamate 57 35,452 622 2.3 Isoleucine 35 7,713 220 0.8 Valine 41 0 0 Total recovered 117,735 (94%)c a25 uCi [3-14C]pyruvate (spec. act. 20 uCi/umole) was added to a 10 m1 culture in medium B-lO at the culture density of 0.3 O.D. units. bLysine was separated by paper electrophoresis; all other amino acids were resolved by means of an amino acid analyzer (Materials and Methods). cThe total radioactivity in the sample of hydrolysate analyzed was 125,803 dpm. 54 results in the synthesis of glutamate withtfiuzc-l derived from the carboxyl carbon of acetate. Only 38% of the 14C in isoleucine was released by decarboxyla- tion. As in the case of valine part or all of this fraction could have been incorporated by reversible transamination and CO2 exchange reactions. However, 14C from 14CO was fixed in at least one addi- 2 tional carbon of isoleucine. Incorporation of 14C from [3—140]pyruvate into isoleucine and other amino acids. When 9, sporogenes was labeled for one generation with [3—14C]pyruvate, the observed labeling pattern (Table 16) of all of the amino acids except isoleucine was consistent with the biosynthetic pathways outlined in the two earlier sections. The incorporation of large amounts of radioactivity into serine as compared to alanine probably resulted from the relative concentra- tions of the two amino acids in the medium at the time the labeled pyruvate was added. While the fresh medium (B-lO) contained 25 mM serine and no alanine, the serine was mostly depleted and significant amount of alanine accumulated by mid-log phase of growth (Table 5). The data clearly show that the C-3 of pyruvate is incorporated into isoleucine. The specific activity of theisoleucine is close enough to that of aspartate and threonine to suggest that there is one C-3 of pyruvate incorporated per molecule. As expected, no label 14 in the form of CO was released from any of the amino acids follow- 2 ing decarboxylation (Table 17). 55 Table 17. Decarboxylation of amino acids from protein hydrolysate of cells grown in the presence of [3-14C]pyruvate Radioactivity (dpm) Percent Amino Acid Initial 1400, Label Lost Alanine 2,301 4 0 Serine 1,557 17 1 Glycine 169 0 0 Aspartate 4,264 16 0 Threonine 788 18 2 Lysine 4,094 18 0 Glutamate 5,188 0 0 Isoleucine 782 0 0 56 Pulse-labeling with [3-14C]pyruvate. The results of the foregoing experiments showed that label was incorporated into 14CD2 or 14C-pyruvate (uniformly labeled resulting from degradation of L-[U-140]serine isoleucine in cell proteins when either or labeled at C-3) were added to the growth medium. However, when the specific activities of the labeled isoleucine were com- pared to those of other l4C-amino acids from the same proteins, it was not possible to account for all the carbons in isoleucine. The data indicate that no more than three carbons were derived from pyruvate and a maximum of two carbons from C02. One possible ex- planation for this was that a high percentage of the isoleucine formed was degraded throughout the one generation of growth. This could result in a large dilution of the label incorporated. If such was the case, then one would predict an increased incorpora- tion of radioactivity into isoleucine by cells grown in the pre- sence of label for an abbreviated period of incubation. However, the results of pulse—labeling of g, sporogenes cells with [3-14C] pyruvate (Table 18) show that this was not the case. A considerable amount of radioactivity was incorporated into the cell protein even during a 5 min labeling period. However, when the hydroly- sates were chromatographed, only a very small amount of label was found to be associated with the isoleucine fractions. This observation raised the possibility that traces of isoleucine might have been supplied to the cells as contaminants present in one or more of the other amino acids used in preparing medium B-lO. Indeed, although the cells might be capable of 57 Table 18. Incorporation of label into protein by Q. sporogenes cells pulsed for varying periods with [3-14C]pyruvate in either fresh or spent medium B-lOa Labeling Time Specific Activity (dpm/mg protein) (min) Fresh Medium Spent Medium55 5 638,763 289,476 10 945,006 433,736 30 1,186,226 659,471 aCells from 160 ml culture were harvested by centrifugation when the culture turbidity was at 0.3 O.D. units. The cells were resus- pended in 10 ml of either fresh or spent medium B-lO, each contain— ing 50 uCi of [3-14C]pyruvate (spec. act. 20 uCi/umole), and re- turned to incubation at 37°C. After the indicated time intervals, aliquots were withdrawn from the culture and the cell pellets harvested and processed as described (Materials and Methods). bThe spent medium was the clear supernatant obtained when the culture was harvested at the culture turbidity of 0.3 O.D. units. 58 synthesizing isoleucine, contaminating traces of the latter could conceivably inhibit its formation. However, no significant incorporation of 14C from [3—14C]pyruvate into isoleucine was observed when the pulse experiments were repeated in "spent medium," i.e., in the supernatant medium of a culture which had been grown to mid-log phase of growth (Table 18). Effect of isoleucine on incotporation of 14C from [3-140] pyruvate into isoleucine. From the foregoing, it was not possible to conclude whether or not the incorporation of [3-1401pyruvate into isoleucine was regulated by isoleucine, and whether or not con- taminating isoleucine present in the medium was sufficient to inhi- bit biosynthesis of this amino acid. In an attempt to resolve this question, a culture of g, sporogenes was labeled with [3-14C]pyruvate in medium B-lO containing 5 mM unlabeled isoleucine and the dis- tribution of label in cellular protein determined (Table 19). The labeling pattern observed among the 6 amino acids quantified was essentially identical with that observed with cells labeled with [3-14C1pyruvate in the absence of added isoleucine (Table 19). Amino acid analysis of the culture medium from this experiment showed that 3.2 mM isoleucine was left in the medium at the time the cells were harvested. Incorporation of label from a metabolite(s)pproduced from L-[U-14Glisoleucine into isoleucine. It was subsequently demon— strated that there was a very low level of isoleucine contaminating the basal medium (B-lO) used (see below). Therefore, the possi- 59 Table 19. Effect of unlabeled L-isoleucine on incorpzration of Isotopic carbon from [3- C]pyruvate into amino acidsa b Total Radioactivity Specific Activity Amino Acid nmoles (dpm) (dpm/nmole) Alanine 63 17,471 277 Glycine 35 192 6 Aspartate 50 15,273 305 Lysine 39 22,723 583 Glutamate 56 32,747 585 Isoleucine 35 7,111 203 Total recovered 95,581 (80%)C a25 uCi [3-140]pyruvate (spec. act. 20 uCi/umole) and unlabeled L-isoleucine were added when the turbidity of the culture was 0.3 O.D. units. The initial isoleucine concentration was 5 mM. bAspartate, glutamate, and lysine were isolated by paper electro— phoresis; all other amino acids were resolved on an amino acid analyzer. cThe total radioactivity for the amount of hydrolysate analyzed is 119,180 dpm. 60 bility existed that a reaction between some product of isoleucine degradation and pyruvate and CO was responsible for the observed 2 incorporation of label into isoleucine when cells were incubated with 14 L-[U-14C]serine, [3-14C]pyruvate, or CO To test this possibility, 2. a growth medium was designed to contain not only all the ingredients of medium B-lO but also the labeled product of isoleucine degradation (the 2nd medium on Table 20). The latter was obtained from the treated supernatant medium of a culture grown in the presence of L-[U-14C] isoleucine (the lst medium on Table 20). This was collected when the culture reached a density of 0.3 O.D. unit which is the point at which label was normally added in those cultures incubated with 14 L—[U-14C]serine, [3-14C]pyruvate, or CO Following acidification 2. + and cation exchange (AG 50W-X4, H form) treatment of the supernatant from the first culture to remove residual amino acids and CO 4.6% 2. of the original label remained in solution. This constituted the total radioactivity in the second medium. Ten percent of the total 14C present in the 2nd medium was incorporated into the cell protein. Over 95% of the isotope incorporated was recovered in the isoleucine fraction (9.5% of the initial radioactivity). Upon decarboxylation of the isoleucine fraction, 36% of the label was lost as 14C02. This indicates that some metabolite(s) of isoleucine other than or in addition to 2- methylbutyric acid was incorporated into the isoleucine found in cell protein. While the specific activity of the isoleucine isolated was not high (144 dpm/ mole), it approached the levels 14 14 found in cells labeled with 002, C-serine, or 14C-pyruvate. 61 Table 20. Incorporation of the product of isoleucine degradation into cellular isoleucine Radioactivity Determination (dpm) lst Mediuma: Initial 2.05x107 Supernatant of culture at harvest 1.68x107 . b . . 5 2nd Medium : Initial 9.51x10 Supernatant of culture at harvest. 7.09x105 Protein hydrolysatec: Total 9.51x104 Isoleucine fraction 9.09xlO4 aTen UCi of L-[U-IAC]isoleucine (spec. act. 360 uCi/umole) was added to 10 m1 culture in medium B-lO as soon as growth was observed; the cells were removed at the culture density of 0.3 O.D. units and the supernatant solution was collected. This medium contained the supernatant medium from the lst culture which had been treated to remove 14C02 and amino acids (see Mate- rials and Methods). Non-labeled amino acids, salts, vitamins, and sodium thioglycollate were added to the supernatant solution at the same concentrations as in medium B-lO. CObtained from cells harvested from the second culture at an O.D. of 0.6. 62 This level of incorporation is very significant considering the low level of total radioactivity present in the 2nd medium (Table 20); less than 0.5 UCi/lO ml of medium. Metabolites of Isoleucine Volatile fatty acids. The most likely metabolites of iso- leucine produced by Q. sporogenes are volatile fatty acids, C02, and ammonia. Washed exponential phase cells suspended in a solution containing 100 mM L-isoleucine and 100 mM L-proline produced 2—methylbutyrate as the major volatile fatty acid as expected (Table 21). The low amount of acetate, propionate, and isobutyrate were probably produced from the intracellular pool of alanine, threonine, and valine respectively. In the isotope labeling experiments using L-[U-140]3erine, [3—14C]pyruvate and 14002, the label was added to the culture grown in medium B-lO at the turbidity reading of ”0.3 O.D. units. There- fore, it was of interest to find out the type and the amount of volatile fatty acids present in the culture medium at that particu- lar stage of growth. As shown in Table 21, acetate was the major acid produced. This would be expected because of the high concen- tration of L-serine in medium B-lO (Table l). Leucine and valine in the medium must have been fermented to produce the observed isovalerate and isobutyrate respectively. Trace amount of propionate were also observed. While no other volatile fatty acids were detected in these samples, the levels of metabolites necessary for the limited incorporation into isoleucine observed would be below the limits of detection by the method used. 63 Table 21. Volatile fatty acids obtained from isoleucine fermentation and from a growing culture Concentration (mM) Acids Isoleucine Fermentationa Culture Mediumb Acetate 4 11 Propionate l l Isobutyrate traces 2 Isovalerate --- 2 2-Methylbutyrate 65 —- aExponential phase cells grown in medium A were harvested, washed twice with 0.1 M potassium phosphate buffer (pH 7.5), and then resuspended in 20 ml solution containing 100 mM L-isoleucine and L-proline at the rate of 1.35 g of packed cells/ml. After 8 h incubation at 37°C in the anaerobic chamber, the cells were removed by centrifugation and the clear supernatant was assayed for volatile fatty acids. bCells were grown in 10 ml of medium B-10 and were removed by cen- trifugation when the culture density reached 0.3 O.D. units; the clear supernatant was assayed for volatile fatty acids. 64 Growth Requirements Isoleucine in the medium. When a low dilution of fresh B-lO medium was run through the amino acid analyzer, a very tiny peak was observed in the elution position of isoleucine. The peak was excised from the chart paper, weighed on an analytical balance and the weight so obtained was found (by comparison with those of several amino acids of known concentrations) to correspond to an isoleucine concentration of 0.16 mM. An aliquot of supernatant liquid from a culture grown in this medium until it had reached a turbidity of 0.3 O.D. units (the point of addition of radioactive compounds in the labeling experiments for one generation) was also subjected to amino acid analysis by means of amino acid analyzer. Again, a very tiny peak of isoleucine was found, and by estimation performed in the same manner as described above, the level of isoleucine was 0.08 mM which is roughly 50% of that found in the fresh medium. The protein from a 10 ml culture grown in medium B-lO that was harvested at culture turbidity of 0.6 O.D. units contained about 0.6 nmoles of isoleucine (there were about 518 nmoles isoleucine per mg pro- tein, and there were about 1.2 mg protein in the 10 ml culture). Therefore, there is a possibility that isoleucine is essential for the growth of Q. sporogenes. This amount of contaminant in the medium might be sufficient to provide all the isoleucine required for growth. Growth response to leucine and isoleucine. .Q. sporogenes exhibits a requirement for high concentrations of leucine for 65 growth in medium B-lO (Fig. 5) which does not correlate with the amount of this amino acid consumed by the cells (Table 5). Since L—leucine is the most likely source of the contaminating L-isoleucine, it was reasonable to speculate that the requirement for high concen- tration of L—leucine was only to supply enough L-isoleucine for growth. This was tested by growing 9. sporogenes in medium B-lO with 2 mM L-leucine supplemented with varying amounts of L—isoleucine. , Growth comparable to control levels (i.e. growth in medium B-lO in which 11.5 mM L-leucine was present) was observed on addition of as low as 0.1 mM L-isoleucine. Particularly surprising, however, was the fact that isoleucine could partially relieve the requirement for leucine. Addition of 10 mM of L-isoleucine to medium B-lO in the absence of leucine supported growth up to 0.42 O.D. units, which is about 50% of the normal growth obtained in this medium. Growth response to 2-methy1butyrate. Since 2-methylbutyrate is the primary catabolite of isoleucine fermentation, it was the most probable precursor of isoleucine and leucine. Growth responses to increasing concentrations of 2-methylbutyrate in the absence of both leucine and isoleucine was similar to those observed with leucine alone (Fig. 6). At the highest concentration fo 2- methylbutyrate (8 mM) the total growth was equivalent to that nor- mally obtained in medium B—lO. These observations indicated that g, sporogenes can synthesize both isoleucine and leucine when 2-methylbutyrate is present. 66 N S l o m I o m I .0 b I .0 N Maximum 0.0. at 600nm , J J I l O 2 4 6 8 IO l2 Leucine , mM FIG. 5. The effect of leucine on growth of g, spprpgenes in medium B-lO. 67 Maximum 0.0. at 600 nm l l l 0 2 4 6 8 2-Methylbu1yro’te, (m M) FIG. 6. The effect of 2-methy1butyrate on growth of 9, sporogenes in medium B-lO without leucine. 68 Incorporation of 140 from LU-IAClZ—Methylbutyric Acid into Isoleucine A total of 4.33 UCi of [U-14C12-methylbutyric acid (1.6 m1 of 0.967 uCi/Umole) was added to a growing culture in medium B-10 without added leucine but containing 4 mM unlabeled 2-methylbutyrate at a culture density of 0.3 O.D. units; the final volume was 14.2 ml. The labeling was allowed to occur for one generation. Label was found to be incorporated into the cell protein; the specific activity was 58,584 dpm/mg protein. Amino acid analysis of the protein hydrolysate showed that all of the radioactivity was in the isoleucine fraction; the calculated specific activity was 113 dpm/ nmole. Unfortunately, since the concentration of 2-methylbutyric acid in the culture at the time the label was added was not examined, the quantitative aspect of this labeling experiment may only be roughly estimated as follows: Total volume of culture = 14.2 ml. Total unlabeled 2-methylbutyrate added to the culture = 40 nmoles. Total [U-lAC]2-methylbutyrate added = 4.48 nmoles. Total initial radioactivity in the culture = 1.08 X 107 dpm. Sp. act. of 14C.-isoleucine in the cell protein = 113 dpm/ nmole. A. If all the original unlabeled 2-methylbutyrate was com— pletely consumed prior to the addition of label: sp . act. of l4C-2-methylbutyrate = 1.08 X 107 dpm/4. 48 nmoles. = 2,411 dpm/nmole. =482 dpm/nmole C. 69 This corresponds to only 25% of a single carbon of 2-methylbutyrate being incorporated into each molecule of isoleucine. B. If none of the added unlabeled 2—methylbutyrate was con- sumed prior to the addition of label: sp . act. of 14C-2-methylbutyrate = 1.08 X107 dpm/44.48pmoles. = 243 dpm/nmole . =49 dpm/nmole C. This corresponds to two carbons of 2-methylbutyrate incor- porated into each molecule of isoleucine. The above estimations suggest that no more than two carbons from 2—methylbutyric acid could have been incorporated into isoleucine. Therefore, if g, sporogenes synthesizes isoleucine via carboxylation of 2-methylbutyrate as occurs in some rumen bacteria (55), the label incorporated into isoleucine must have been diluted in the culture. However, if this was the only way in which carbons from 2- methylbutyrate were incorporated, the C-1 of isoleucine should not be labeled. One third of the total radioactivity of the isoleucine fraction was removed by decarboxylation. Therefore, the reductive carboxylation pathway must not be the only manner of isoleucine synthesis from 2-methylbutyric acid. The absence of label incorporation into leucine fraction was totally unexpected and there is no ready explanation for this. DISCUSSION The specific activity of threonine dehydratase in Q, sporogenes was found to be comparable to that in E, coli (49) but higher than that in Q. tetanomorphum cells (70). However, a biosynthetic, iso- leucine sensitive form of threonine dehydratase could not be demon- strated in g. spprogenes. Although it is possible that this form of the enzyme might have been masked by the very active biodegradative threonine dehydratase, it is unlikely, since, in all microorganisms examined thus far except for R. rubrum (28), the biosynthetic threonine dehydratase is extremely sensitive to isoleucine inhibition. For example, the biosynthetic threonine dehydratase of Cotynebacteriumlsp (6) and R. capsulata (28) is 100% inhibited by isoleucine concentra- tions of 1.0 and 0.1 mM, respectively, in the presence of 10 mM threonine; while in Pseudomonas sp, 0.15 mM L-isoleucine produces a 50% inhibition of the enzyme (39). The biosynthetic threonine dehydra- tase of R, rubrum is inhibited by isoleucine but only at low substrate concentrations (28). In E. coli (14) and.§, multivorans (39), the inhibitory effect of isoleucine on threonine dehydratase can be reversed at higher pH values. However, no inhibition by isoleucine was observed when the enzyme in extracts of g, sporogenes was assayed over a wide range of pH values and concentrations. These data and results of subsequent experiments all indicate that threonine is not a precursor of isoleucine in this organism. 70 71 The threonine dehydratase of g, sporogenes is similar to that of Q. tetanomorphum in several respects (27, 70); viz., (a) the enzyme is activated by ADP and not by AMP, (b) the activation by ADP occurs only at low threonine concentrations, (c) in the presence of ADP, the Km value of the enzyme from both sources is lowered by a factor of N13, and (d) the enzyme is constitutive. Two forms of threonine dehydratase were resolved from extracts of Q, tetanomorphum by diethylaminoethyl (DEAE) cellulose chromatography (27). However, neither form of the enzyme was subject to inhibition by isoleucine. It is probable that the function of the threonine dehydratase in both of these clostridia is catabolic. The latter has been described by Tokushige gt a1 (62) as the breakdown of threonine to propionic ' acid and ammonia accompanied by ATP production. Multivalent repression of biosynthetic threonine dehydratase is a common phenomenon in E, coli and S, typhimurium (66) and results from the simultaneous presence of excess amounts of valine, leucine, and isoleucine in the growth medium (65). Q, sporogenes required sub- stantial amounts of valine (17 mM) and leucine (11.5 mM) in medium B-lO (Table 1). This medium which supposedly "lacked" isoleucine, was subsequently found to contain traces of this amino acid as a contami- nant; however, the amount of isoleucine present Q»O.l6 mM) was far below the level required for multivalent repression as reported in other bacteria (65). Thus, although it was not possible to grow 9, pporogenes without valine and leucine, the level of isoleucine present should have allowed detection of a biosynthetic threonine dehydratase if it were truly present. Some bacteria (Cotynebacterium 72 §p_(6), and g, multivorans (39) produce biosynthetic forms of threonine dehydratase which are not sensitive to multivalent repres- sion by valine, leucine and isoleucine. However, these enzymes are sensitive to feedback inhibition by isoleucine. All the data accumulated in this study strongly suggest that g, sporogenes only contains the biodegradative form of threonine dehydratase which is constitutively synthesized. It is not repressed by glucose, and its primary role ipflxi!p_is for energy generation from threonine degrada- tion presumably at a low energy charge. Attempts to demonstrate threonine aldolase (EC 2.1.2.1) and threonine dehydrogenase (EC 1.1.1.103) activities in extracts of g, sporogenes cells were unsuccessful. Either enzyme could lead to the production of glycine from threonine. Threonine aldolase cleaves threonine to acetaldehyde and glycine in some organisms (6, 19, 31. 76); whereas threonine dehydrogenase catalyzes the oxidation of threo- nine to a-amino-B-ketobutyrate, which in turn could be degraded via two possible pathways; i.e., the aminoacetone route and the glycine route. In the first route, a—amino-B-ketobutyrate is decarboxylated non-enzymatically into aminoacetone (38). In the glycine route, G-amino-B—ketobutyrate is cleaved to acetyl coenzyme A and glycine by the action of an enzyme variously known as aminoacetone synthase (44) or glycine coenzyme A ligase (73). Threonine dehydrogenase of §, marcescens and of E, coli K-12 are controlled by catabolite repression and by leucine-mediated induction (37, 47). No threonine dehydrogenase activity was detected in extracts of g, sporogenes cells grown in the absence of glucose and in the presence of excess leucine. 73 Although neither of the two enzyme activities was detected, glycine production from threonine was found to occur to a minor extent in g, sporogenes cells (Table 9). It is not known which enzyme is responsible for that conversion. Auxotrophic mutants lacking serine transhydroxymethylase enzyme would be very useful in such investi- gation. The absence of a biosynthetic threonine dehydratase raised a question as to how isoleucine is synthesized in g, sporogenes. In preparation for isotopic labeling experiments employing 14C-substrate to find the source of carbon skeleton for isoleucine, determinations were made of the amino acid composition of cell proteins, free intra- cellular pools, and of the growth medium. Of particular interest were the high levels of glutamic acid, alanine and proline observed in the intracellular pool during the exponential phase of growth. None of these three amino acids were added to the growth medium, so they must be synthesized. g, sporogenes has been shown to accumulate proline, glutamic acid, and Y-aminobutyric acid in the amino acid pool in response to increased environmental sodium chloride (45). Measures 5(45) reported that growth of non-halophilic bacteria at low water activities (high solute concentrations) depends on the ability of the cell to balance the environmental osmotic pressure by intracellular accumulation of amino acids, and on the types of amino acid which accumulate. The synthetic medium employed to grow 9, sporogenes cells (medium B-lO) may have a sufficiently high salt content to influ- ence the pool. 74 Of the 8 amino acids that comprise medium B-lO (Table 1), serine is probably the main if not the sole source of pyruvate. E, botulinum, which is closely related to E, sporogenes (30, 75), was reported to produce alanine, acetate, CO2 and ammonia from argine, which would involve pyruvate as an intermediate. However, this fermentation occurred only to a minor extent (46). Serine appeared to be consumed very rapidly during the early exponential phase, accompanied by the concomittant accumulation of a substantial amount of alanine. Therefore, it is reasonable to assume that at a certain point during the growth the cells must have switched to using alanine as the source of pyruvate. This assumption is supported by the labeling pattern of the amino acids obtained from cells incubated in the presence of L-[U-14C]serine or [3-14C]pyruvate (Table 11 and 16)- A variety of enzymes are known that are capable of deaminating L—serine to produce pyruvate and ammonia (65); 325,, (a) dehydratases specific for L-serine, e.g. the enzyme from E, gpli_described by Alfodi gt_§E (2); (b) dehydratases that act on other substrates, most frequently L-threonine, e.g. the biosynthetic and biodegradative L-threonine dehydratases of E, coli and E, typhimurium (65); and (c) enzymes primarily catalyzing a different type of reaction, e.g. the B protein of tryptophan synthetase (EC 4.2.1.20) of E, £211,(18)- Of those enzymes capable of degrading serine as reviewed by Umbarger (65), none were activated by AMP. It is possible that there is more than one enzyme in E. §porogenes cells that could act on serine, and that such activity is not regulated, at least under the conditions of growth in medium B-lO employed. 75 The synthesis of alanine from pyruvate in E, 29;; occurs via the glutamate-alanine transaminase and valine-alanine transamin- ase (transaminase C); however, the latter serves only a minor role (66). Except for the repression control over transaminase C, alanine biosynthesis does not seem to be regulated (66). While it is not known which transaminase catalyzes the conversion between pyruvate and alanine in E, _porogenes, the assumption that the enzyme is not regulated is indicated by the accumulation of alanine both intra- and extracellularly. There was considerable incorporation of 14C into cell pro- tein when cells were grown to stationary phase in the presence of the following 14C-labeled compounds: threonine, valine, leucine, arginine, ornithine, proline, glucose, histidine, acetate, methionine, phenylalanine, and tyrosine. However, there was no significant incor- poration of label into the cellular isoleucine. Improved experimental conditions (labeling for only one generation instead of for the entire growth period, and harvesting the cells during exponential phase) gave similar results. The absence of significant incorporation of label from L-[U-lAC]aspartate or L-[U-lAC]threonine into cellular iso- leucine (Table 8) is consistent with the failure to detect the presence of biosynthetic threonine dehydratase in E, sporogenes extracts. It is very unlikely that isoleucine is synthesized from threonine in this organism. However, there were other potential sources of a-keto- butyrate in the medium used which might dilute out the a-ketobutyrate produced from threonine. E, sporogenes was found to convert L-methionine to a-ketobutyrate, ammonia and methylmercaptan (72). 76 A mutant of E, coli Crookes strain deficient in threonine dehydratase was found to synthesize isoleucine from glutamate which in turn serves as thesource of a-ketobutyrate (50). However, no labeled isoleucine was obtained when E, sporogenes cells were grown in the presence of L-[U-lAC]methionine (Table 7) or L-[U-1401glutamate (Table 8). The initial specific activities of the labeled sub— strates (aspartate, threonine, methionine or glutamate) were sufficiently high, that even if dilution of labeled<1-ketobutyrate occurred labeled isoleucine should still have been detected if the classical pathway of isoleucine biosynthesis was operative. The total amount of label incorporated into the cellular protein was much higher when the culture was labeled with L-[U-lAC] threonine than when labeled with L-[U-l401aspartate or with L—[U-lAC] glutamate even though the initial specific activities of all three substrates were nearly equal (Table 8). This suggests that threonine biosynthesis is very tightly regulated and that the biosynthesis of aspartate and glutamate are either not regulated or only weakly controlled. The small amount of threonine present in the added radio- active threonine must have been sufficient to shut off its synthesis during most of the labeling period. In E, 221$, two enzyme activities in the pathway of threonine biosynthesis, aspartokinase I and homo— serine dehydrogenase I are sensitive to inhibition by threonine (66). Aspartate is formed by transamination of oxalacetate via a glutamate- aspartate transaminase activity; in E, 2913, it is the transaminase A which is primarily responsible (66). While aspartate in the growth medium was reported to repress the bulk of glutamate-aspartate 77 transaminase activity (68), there is no evidence for a regulatory site for any effector molecule that would modulate the activity of the enzymes involved. The reaction is freely reversible (66). While further studies would be required to establish the regulation of threonine, aspartate, and glutamate biosynthesis in E, sporogenes, the lack of strict regulation of glutamate biosynthesis should be anti- cipated because of the high intracellular pool of glutamate found in this organism (Table 5). Subsequent labeling experiments showed that labeled isoleucine was obtained when cultures were incubated with L-[U-14€]serine, _14 14 [3 C]pyruvate or CO2 (Table ll, 16 and 14). The distribution of the label found in alanine, aspartate, threonine, lysine, glutamate, serine, and glycine was consistent with well established biosynthetic pathways (Fig. 7). Comparisons of the specific activities of these amino acids with that of the l4C-isoleucine formed provide a basis for estimates of the number of carbon atoms from each labeled substrate contributing to isoleucine biosynthesis. Therefore, the labeling patterns resulting from each 14C-substrate will be discussed indivi- dually. Serine degradation in a culture labeled with L-[U-l4C]serine would result in pyruvate with all of its three carbons labeled, which in turn could be directly involved in several different reactions. Those that will be specifically discussed here are: (a) the pyruvate- CO exchange reaction; (b) transamination to alanine; and (c) carboxy- 2 lation into oxalacetate. A pyruvate-CO2 exchange reaction in E. pporogenes probably occurs via pyruvate synthase, an enzyme which is 78 FIG. 7. Interpretative scheme of the amino acid labeling pattern for all radioactive experiments with E. iorogenes. 79 "$0-9 _. “2*?” A Raw-cu: Phocwoolycoron "0" CH2 \\ Glycine coon :‘P - ® cu, Macao! ""3 W” coon l run-4H coon tflKhA é", co to CH3 CH8 é“: Pym" H1" 4H2 1 CH'-COOH PH. N°,N'°- Methylene m. I one uo-c-coou coon c0: CHQ-COO;‘\\\\\_ do (anon! comma “2" 3°" co-oooa “Hrw'm‘ - U . ..___amcrcoon O'm.00'. coon 4%,. HO-CH Threonine raven-coon cH-coou (whecoou KOCHI’O'. 2H,C02 a-nmqlmmn coon Haw-é" :"2 CR2 coon G'UNMC'. 80 widely distributed in anaerobic organisms (12). Consequently, the label of C-1 of pyruvate would likely be diluted by any unlabeled CO2 present in the medium. However, studies with 14CO2 indicate that this exchange reaction in E, sporogenes only occurs to a very limited extent under the growth condition employed (see below). Thus, pyru— vate synthase in this organism probably operates mainly in the oxi- dation of pyruvate to acetyl coenzyme A and C02, a property reported for the same enzyme from E, acidi-urici and E, pasteurianum (12). Transamination of pyruvate derived from L-[U-14C]serine would result in alanine with all of its three carbons labeled. As pointed out earlier, alanine was excreted into the medium very actively during the early phase of growth. Although this process might continue to occur only shortly after the addition of the label, the high specific activity of the added label (sp. act. 162 pCi/ mole; 25 uCi total) would be sufficient to establish a pool of alanine in the medium with a relatively high specific activity. However, because of the relative concentrations of unlabeled alanine and serine at the time the lAGO-serine was added to the medium, there would be more dilution of the labeled alanine formed and then incorporated into protein than of the 14C-serine incorporated directly into cells. This was reflected by the specific activity per carbon atom of the serine from protein; it was about twice that of the activity per carbon in alanine. Therefore, it is more appropriate to compare the specific activities per carbon atom for other amino acids with alanine than with serine. Such comparisons indicate that alanine, glycine, aspartate, threonine, lysine, and glutamate were uniformly labeled (Table 11), although the 81 specific activity per carbon of threonine was low. Decarboxylation of each of these amino acids also indicated that they were uniformly labeled. In contrast, the radioactivity per carbon of isoleucine was only about 50% that in alanine, and decarboxylation resulted in the release of about 25% of the total label present. These data inci- cate that only 3 or 4 of the 6 carbons in isoleucine were synthesized from pyruvate. Oxalacetate is probably synthesized from pyruvate via pyru- vate carboxylase reaction which is known to occur in many organisms (58). Oxalacetate is an important intermediate; it serves as a precursor of not only the biosynthesis of several amino acids, but also of gluconeogenesis. The first reaction in gluconeogenesis is probably the conversion of oxalacetate into phosphoenolpyruvate (PEP) cata- lyzed by PEP-carboxykinase, an enzyme which is widely distributed in many organisms including a number of bacteria (69). These suggested pathways are supported by the labeling pattern of the amino acids in proteins of cells grown in the presence of labeled serine, pyruvate, or 002 (Table ll, 16 and 14), and each will be discussed in detail accordingly. Uniformly labeled oxalacetate is required for the synthesis of uniformly labeled aspartate, threonine, lysine, and glutamate, respectively. Therefore, the C02 molecule that was incorporated into oxalacetate must have been radioactive and contained approximately the same specific activity as that of pyruvate or alanine. This would be possible if a significant amount of CO2 is derived from the oxida- tion of [U-1401pyruvate via the pyruvate synthase reaction. This 82 proposed pathway is consistent with the earlier suggestion on the role of the latter enzyme in E, sporogenes. Had the C-1 of pyruvate been greatly diluted by C02 in the medium, the results obtained on decar— boxylation of the labeled amino acids (Table 12) would not have been expected. The data all indicate that E, sporogenes utilizes the same biosynthetic pathways for biosynthesis of aspartate, threonine and lysine as known to occur in other bacteria (66). Cooper and Costilow (16) suggested glutamate synthesis might occur from proline. In such a case, glutamate from cells incubated with L—[U—14C132rine should not have been so heavily labeled because proline in this organism is produced from arginine (which is essen- tial for growth) via ornithine (17, 46). The specific activity per carbon of glutamate formed from 1['C-serine was the same as that per carbon of alanine. Therefore, glutamate must be synthesized from pyruvate (Fig. 8). In most organisms,(x-ketoglutarate is synthesized from pyruvate via the forward TCA cycle (66), and these data are consistent with this pathway. However, some anaerobes use a reduc- tive pathway for glutamate synthesis (4, 5) and this is not excluded by these results. The results of decarboxylation of 14C-isoleucine formed from L-[u-IAC]serine (26.4% of label loss) suggest that a total of three to four carbons in that amino acid were labeled. Since the total specific activity of isoleucine was 120 dpm/nmole (Table 11), if only three carbons were labeled, the specific activity of each labeled carbon would be 40 dpm. This value is similar to that for alanine. However, this information is not sufficient to give any indication 83 A CoASH CH3-CO- SCOA (EHZ'COOH OH - ('3- COOH B co-ooou 9H2‘COOH 5H2” C OOH curd“ oxalacetate 2” HO-C'IH-COOH CH-COOH HO-CH-COOH CHz- coon-4 éHZ- coo... 5306i "'01. melon 2H , C02 H20 COOI-l EH -COOH HOOC- CH 7//i:2 fumoron CO H 2’2 econ 2H Cl-Iz- COOH 01- kcfoolutoroto CH2- COOH succinon CO OH "‘2"" v.5" :"2 CH | 2 COOH mm FIG. 8. Glutamate synthesis from oxalacetate. A, via the forward TCA cycle (66); B, via the reductive pathway (4, 5). 84 on which carbons of pyruvate were incorporated into isoleucine or how the process occurred. Cultures labeled with 14CO2 resulted in the incorporation of relatively high levels of radioactivity into the cellular protein (Table 13). The amount of radioactivity incorporated per mg protein in the culture labeled for two generations was only 6% lower than that labeled for one generation. This suggests taht there was little dilution of the label by unlabeled CO2 resulting from doubling the incubation period. Following is an interpretation of the labeling pattern of the amino acids obtained from cells incubated with 14CO2 for one generation (see Fig. 7). Pyruvate would acquire label at C-1 due to the pyruvate-14002 exchange reaction, thus transamination of pyruvate would result in alanine also labeled at C-1. This suggestion is supported by the complete loss of label upon decarboxylation of the alanine isolated from cell protein (Table 15). Oxalacetate synthesized by carboxylation of pyruvate would be labeled from 14CO at C—4; some label would be expected at C-1 depend- 2 ing on the extent of pyruvate-CO exchange reaction. During the forma- 2 tion of PEP from oxalacetate, C-4 of the latter would be lost and thus the PEP would be labeled only at C-1. Accordingly, serine would also acquire label at C-1 via the phosphoglycerate pathway. Glycine from serine would also be labeled at C-1. As shown in Table 14, alanine, serine, and glycine showed relatively low specific activities. This suggests that pyruvate-CO2 exchange reaction in E, Sporogenes only occurred to a limited extent. The specific activity of glycine is 85 only slightly less than that of serine, which is consistent with the earlier suggestion that the majority of glycine is synthesized from serine. Biosynthesis of aspartate, threonine, and lysine from oxa- lacetate which was labeled at C-1 and C-4 would result in all three amino acids being labeled also at C—1 and C-4. As expected, aspartate lost all its label upon decarboxylation by ninhydrin (Table 15); this treatment removes both carboxyl carbons (26). On the other hand, the chloramine-T treatment should remove only C-1 of threonine and lysine respectively (33). As shown (Table 15), both amino acids lost only W20% of their label. Since, by the classical pathways, C-l for both amino acids would be derived from C-1 of pyruvate, these results are consistent with the earlier suggestion that the pyruvate-CO2 exchange reaction occurred to a limited extent. The specific activities per nmole of alanine, serine, and glycine were only W20% of that of aspartate. The Specific activity of glutamate labeled from 14CO2 was to be nearly equal to that of aspartate. There are two ways glutamate is known to be synthesized from oxalacetate; (a) via the forward TCA cycle as in E, gpll_and other bacteria (66, Fig. 8A); (b) via the reduction of oxalacetate to succinate and the carboxylation of suc- cinate to a-ketoglutarate as found in some rumen bacteria (4, 5, Fig. 8B). During the synthesis of glutamate from C-4 labeled oxal- acetate there would be one molecule of CO2 lost in the first pathway, whereas one molecule of CO2 would be incorporated in the second one. Therefore, if glutamate in E, sporogenes was synthesized via the 86 reverse TCA cycle in the presence of 14002, one would expect the specific activity of glutamate to be about 2X that of aspartate. However, the data show that the two amino acids had about the same specific activity. Because of the high intracellular glutamate pool at the time 14CO2 was added to the culture, these data alone are not conclusive. However, failure to release any 14CO2 on decarboxylation of the labeled glutamate excludes the possibility of synthesis via the reverse TCA cycle. Glutamate in this organism must be synthesized via the forward TCA cycle. The distribution of label from 14C0 among the 5 carbons 2 of glutamate would be different depending on the stereospecificity of the citrate synthase enzyme (23, Fig. 9), Most organisms contain the giftype synthase and the glutamate synthesized from oxalacetate labeled at C-1 and C-4 is only labeled at C-1 due to the removal of one molecule of CO2 in the isocitrate dehydrogenase step. On the other hand, under the same conditions tgfcitrate synthase which is found in a small group of anaerobic bacteria, gives rise to glutamate being labeled only at C-5 derived from C-4 of oxalacetate. The C-1 of oxalacetate is lost in the isocitrate dehydrogenase step. No label was lost upon decarboxylation of glutamate (Table 15). Only C-1 of glutamate is removed by the decarboxylation method used (33). Therefore, it is likely that E, sporogenes possesses citrate synthase of the rgftype stereospecificity. The amount of radioactivity incorporated into isoleucine during one generation of growth in the presence of 14CO2 was quite significant. The specific activity of the isoleucine was not 87 .Ammv ommufiaoumlwau %n “mums mo Hm>oEoH uamaowamoumum .m. .oONH vocusu wasomaoa .< .wmmsuo%m mumuufiol now an mam wmmzucmm oumuumul Hm >3 oumuufio mo coauMEkom .m .on roowlmxw, I E ooo: N”cow :oow No To]: 1an \u§ of II lvosxuuxnuoo: . ,3 E w E 085532050 IL 0 (coo... :8 18¢ .. 4!... am: Am :08 1o xoxofiugo 01.85%. 28% 398 /u«... coo: 3 IWOOI 88 significantly less than that of lysine which should be equivalent to that of aspartate (Table 14). Upon decarboxylation, isoleucine lost 38% of its label while threonine and lysine lost 20%. If classical pathways were operative in the formation of the latter 2 amino acids, the label lost was probably derived from pyruvate-CO The 14CO2 fixed into oxalacetate via pyruvate carboxylase would be in 2 exchange reactions. the C-4 of threonine and lysine. Simple calculations indicate that for each carbon fixed, the dpm/pmole should be near 160-170. This is the level of radioactivity found in isoleucine. However, since 38% of the label was lost on decarboxylation, radioactivity from 14002 must have been incorporated into more than one carbon of isoleucine. The low specific activity of valine in cell proteins produced in the presence of 14CO2 and the complete loss of the label upon its decarboxylation is consistent with the possibility suggested above that valine could acquire label from a low level of valine-CO2 exchange reaction. Growth in the presence of [3-14C]pyruvate should give rise to alanine labeled at C-3 via transamination and acetyl coenzyme A labeled at C-2 via oxidation catalyzed by pyruvate synthase. Carboxy- 1ation of pyruvate would result in oxalacetate labeled at C—4 and upon its decarboxylation the resulting PEP would be labeled at C-3, there- fore, phosphoglycerate synthesized from PEP would also be labeled at C-3. The incorporation of significant radioactivity into serine suggests the presence of serine synthesis via the phosphoglycerate pathway in E, pporogenes as occurs in other organisms (66). This was 89 rather unexpected since the growth medium employed (medium B-lO) contained 25 mM L-serine. However, analysis of the growth medium at the time that labeled substrate was added showed that the residual serine concentrations were usually less than 1 mM. As noted above, considerable alanine had accumulated in the medium. Therefore, one would expect more dilution of the labeled pyruvate by alanine than serine. The results agreed with this. Therefore, the specific acti- vities of other amino acids labeled with [3—14C1pyruvate are most appropriately compared to that of alanine. If glycine is synthesized from serine, 0-3 of serine would be removed during the serine hydroxymethylase reaction, and glycine would not be labeled. The very low amount of 140 found in glycine could not have been due to glycine biosynthesis from threonine via established pathways. Threonine in this experiment should be labeled at C-3, and glycine synthesized from threonine should retain only the C-1 and C-2 of threonine. It is of interest,however, that results obtained earlier did show incorporation of 140 into cellular glycine when L-[U-14C]threonine was used as substrate (Table 9). Oxalacetate labeled at C-3 would give rise to aspartate and threonine labeled at their C-3 positions. However, since one molecule of pyruvate is incorporated into lysine during its synthesis from aspartate, lysine would be expected to be labeled at C-3 and C-5 (label at C-5 is derived from pyruvate incorporation). If E, gpgrgf ggpg§_contains rgfcitrate synthase as suggested earlier, the synthesized glutamate would be labeled at C-2 and C-4 (the label at C—2 is derived from C—3 of oxalacetate). The ratios of the specific activities of 90 alanine, aspartate, threonine, lysine, and glutamate synthesized from [3-14C]pyruvate were very close to values expected from these organisms (Table 16). Although the exact location of the label in each of the foregoing amino acids cannot be proved conclusively at this point, the results of their decarboxylation also supports the suggested pattern of carbon flow during their syntheses. There was no label lost upon decarboxylation of all the amino acids examined (Table 17); therefore, the label resided at a carbon atom(s) other than the carboxyl carbon. Isoleucine isolated from cell proteins produced in the presence of [3-14C1PYruvate had a specific activity not significantly differ- ent from alanine. In addition, none of this 14C was lost from the labeled isoleucine upon decarboxylation. This is very direct evidence that E. sporogenes is able to synthesize at least some isoleucine from rather simple precursors. While the data suggest that no more than one C-3 of pyruvate is incorporated per isoleucine formed, this is quite speculative because of the possible dilution by the contaminating isoleucine (see above). The fact that there was a low concentration of isoleucine con- taminating the growth medium greatly complicated time interpretation of the data. It was not possible to accurately estimate the number of labeled carbons in isoleucine by comparison of the specific activity of the isoleucine with other amino acids. However, the extent to which 14C from [3-14C]pyruvate was incorporated into isoleucine was not influenced by the addition of excess isoleucine to the growth medium. Indeed it appears probable that degradation products of 91 isoleucine are necessary for incorporation of carbons from CO2 and pyruvate. This is one possible explanation for failure to find any significant label in the isoleucine of cell protein when washed cells were incubated for very short periods of time in the presence of [3-14C]pyruvate. The accumulated evidence at this point suggested that iso- leucine or a catabolite of isoleucine was essential for growth of E, sporogenes and that the observed incorporation of label from pyruvate and CO depended on the presence of low amounts of isoleucine in the 2 medium. The No.16 mM isoleucine found contaminating medium B-lO is more than would be required for protein synthesis during the entire growth period. Approximately 2 mg cell protein were obtained from a 10 ml culture (harvested at the end of log phase) in medium B-10 and there are N518 nmoles isoleucine per mg protein (Table 4), so 0.1 mM isoleucine would be sufficient for protein synthesis. Furthermore, the requirement of a high concentration of leucine (11.5 mM) for maximal growth in medium B-lO did not correlate with leucine consumption (Table 6), suggesting that leucine might be the source of the contamination isoleucine. This was supported by the ability of E, §porogenes to grow in the presence of a low concentration of leucine provided that a small amount of isoleucine (0.1 mM) was added to the medium. A very significant amount of label was incorporated into the isoleucine of the cell protein during growth in medium B-lO containing the non-cationic degradation products of L-[U-lAC]isoleucine (Table 20). The specific activity oftfiueisoleucine fraction (144 dpm/nmole) 92 approached that obtained in cells labeled with 14002, 14C-serine, or 14C—pyruvate (Table 14, 11, 16), However, the efficiency of label incorporation from the degradation product(s) based on the ini— tial amount of label added to the culture was very high (9.5%) as compared to the efficiencies observed with labeled C02, serine or pyruvate (0.15-0.22%). The latter three substrates were subject to much more dilution in the growth medium than the catabolite(s) of labeled isoleucine. An attempt was made to identify the degradation product of isoleucine by extraction and separation of the volatile fatty acid fraction obtained from isoleucine fermentation by E, sporogenes. As expected, the major fraction was found to be 2-methylbutyric acid (Table 21). Although the latter was not detected in the mid-log phase culture grown in medium B-lO, this fact does not necessarily indicate its non-involvement in isoleucine synthesis. In fact, growth experiments showed that 2-methylbutyrate eliminated the requirement for both isoleucine and leucine during growth in medium B-lO. This finding suggested the synthesis of both isoleucine and leucine from 2-methylbutyrate, and E, sporogenes cells labeled for one generation with [U—lAC]2-methy1butyrate showed considerable incor— poration of label into the cell protein. However, all of the label was recovered in the isoleucine fraction; none in leucine. It was estimated that no more than 3 carbons from Z-methylbutyrate could have been incorporated into isoleucine, and decarboxylation of the 14C- isoleucine isolated released one third of the radioactivity. While this does not rule out isoleucine synthesis via reductive carboxylation of 93 2-methy1butyrate, obviously other reactions are involved. It is possible that some of the 2-methylbutyrate in the culture was degraded to release 14CO2 and this would have been greatly diluted by the CO in the medium. The culture was incubated in a 5% CO2 atmosphere. 2 The accumulated evidence thus far suggests that E, sporogenes is able to synthesize isoleucine via an unusual pathway and the syn- thesis is not regulated by isoleucine. Pyruvate, CO or 2-methylbuty- 2’ rate alone probably cannot serve as the sole precursor of the carbon skeleton of isoleucine. This is supported by the finding that growth in medium B-lO can only be supported at low leucine concentrations when a small amount of isoleucine or a significant level of 2-methyl- butyrate is present since both pyruvate and CO2 are abundant under the growth conditions used. The incorporation of significant levels of l4 14 14 , C from both [3— C]pyruvate and CO2 into carbons otner than the carboxyl group of isoleucine represents conclusive evidence that isoleucine synthesis in E. sporogenes can occur by mechanisms other than simple reductive carboxylation of 2-methylbutyrate. Obviously, further studies will be required to resolve this question. Studies of l4 14 14 the extent of C incorporation from C-pyruvate and CO2 in the presence and absence of 2-methylbutyrate should be helpful. Also, labeling of cellular isoleucine with specifically labeled substrates followed by stepwise degradation of the isoleucine formed should provide information regarding possible intermediates and reactions involved. 94 Growth data indicated that E, sporogenes could synthesize leucine when 2-methylbutyrate was added to the medium. Addition . of this volatile fatty acid eliminated the requirement for leucine while substitution of isoleucine for leucine resulted in only an intermediate level of growth. However, a labeling experiment with l4C—2-methylbutyrate showed that the latter volatile fatty acid is not a precursor of leucine. There is no ready explanation at this point as to how leucine is synthesized by E, sporogenes, but it would be of interest to determine the extent of 14C incorporation from both 14C-pyruvate and 14C02 into leucine in leucine-free medium supplemented with 2-methylbutyrate. 10. 11. BIBLIOGRAPHY Abramsky, T. and D. Shemin. 1965. The formation of isoleucine from B-methylaspartic acid in Escherichia coli W. J. Bacteriol. 240: 2971-2975. Alfodi, L., I. Rasko, and E. Kerekes. 1968. L-serine deaminase of Escherichia coli. J. Bacteriol. 2E; 1512-1518. Allison, M. J. 1969. Biosynthesis of amino acids by ruminal micro— organisms. J. Anim. Sci. E2; 797-807. Allison, M. J. and I. M. Robinson. 1970. Biosynthesis of a- ketoglutarate by the reductive carboxylation of succinate in Bacteroides ruminicola. J. Bacteriol. 104: 50-56. Allison, M. J., I. M. Robinson, and A. L. Baetz. 1979. Synthesis of arketoglutarate by reductive carboxylation of succinate in Veillonella, Selenomonas, and Bacteroides species. J. Bacteriol. $39: 980-986. Bell, 8. C., and J. M. Turner. 1977. Bacterial catabolism of threonine. Threonine degradation initiated by L-threonine hydro-lyase (deaminating) in a species of Corypebacterium. Biochem. J. 1E3, 579-587. Belly, R. T., S. Greenfield, and G. W. Claus. 1970. B-methylas- partate metabolism and isoleucine synthesis in Acetobacter suboxygans. Bacterial. Proc. p. 141. Bryant, M. P. 1965. Rumen methanogenic bacteria, p. 411-418. Ep_R. W. Daugherty, R. S. Allen, W. Burroughs, N. L. Jacobson, and A. D. McGilliard (ed.), Physiology of digestion in the ruminant. Butterworth, Inc., Washington, D. 0. Bryant, M. P., and I. M. Robinson. 1961. Studies on the nitrogen requirements of some ruminal cellulolytic bacteria. Appl. Microbiol. 2, 96-103. Bryant, M. P. and I. M. Robinson. 1962. Some nutritional charac- teristics of predominant culturable rumen bacteria. J. Bacteriol. E3: 605-614. Bryant, M. P. and I. M. Robinson. 1963. Apparent incorporation of ammonia and amino acid carbon during growth of selected species of ruminal bacteria. J. Dairy Sci. 4E, 150-154. 95 12. 13. 14. 15. l6. l7. 18. 19. 20. 21. 22. 23. 96 Buchanan, B. B. 1972. Ferredoxin-linked carboxylation reactions, p. 193-214. Ep_P. D. Boyer (ed.), The enzymes, vol. VI. Academic Press, New York. Campbell, L. L., Jr., and H. A. Frank. 1956. Nutritional require- ments of some putrefactive anaerobic bacteria. J. Bacteriol. 1;, 267-269. Changeux, J. P. 1961. The feedback control mechanism of biosyn- thetic L-threonine deaminase by isoleucine. Cold Spring Harbor Symp. Quant. Biol. EE: 313-318. Charon, N. W., R. C. Johnson, and D. Peterson. 1974. Amino acid biosynthesis in the spirochete Leptoppira: evidence for a novel pathway of isoleucine biosynthesis. J. Bacteriol. 117: 203-211. Costilow, R. N. and D. Cooper. 1978. Identity of proline dehy- drogenase and A'-pyrroline-5-carboxylic acid reductase in Clostridium sporoggnes. J. Bacteriol. 134: 139-146. Costilow, R. N. and L. Laycock. 1971. Ornithine cyclase (deamin- ating). Purification of a protein that converts ornithine to proline and definition of the optimal assay conditions. J. Biol. Chem. gig: 6655-6660. Crawford, 1. P. and J. Ito. 1964. Serine deamination by the B protein of Escherichia coli tryptophan synthetase. Proc. Natl. Acad. Sci. (U.S.) El, 390-397. Dainty, R. H. and J. L. Peel. 1970. Biosynthesis of amino acids in Clostridium pasteurianum. Biochem. J. 117: 573-584. Dehority, B. A. 1966. Characterization of several bovine rumen bacteria isolated with xylan medium. J. Bacteriol. 2;, 1724-1729. Dunne, C. P. gt EE, 1973. The mechanism of action of 5'-adenylic acid-activated threonine dehydrase. IV. Characterization of kinetic effect of adenosine monophosphate. J. Biol. Chem. ESE: 8189-8199. Egan, R, M. and A. T. Phillips. 1977. Requirements for induction of the biodegradative threonine dehydratase in Escherichia coli. J. Bacteriol. 132: 370-376. Gottschalk, G. and H. A. Barker. 1966. Synthesis of glutamate and citrate by Clostridium kluyveri. A new type of citrate synthase. Biochem. 23 1125-1133. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 97 Gottschalk, G. and H. A. Barker. 1967. Presence and stereo- specificity of citrate synthase in anaerobic bacteria. Biochem. g, 1072-1034. Gottschalk, G. 1979. Bacterial metabolism. Springer-Verlag, New York. Greenburg, D. M. and M. Rothstein. 1957. Methods for chemical synthesis, isolation and degradation of labeled compounds as applied in metabolic studies of amino acids and proteins, p. 652-731. Ep_S. P. Colowick and N. 0. Kaplan (ed.), Methods in enzymology, vol. IV. Academic Press, New York. Hayaishi, 0., M. Gefter, and H. Weisbach. 1963. Adenosine diphos- phate-dependent threonine dehydrase activity in extracts of Clostridium tetanomopphum. J. Biol. Chem. 238: 2040-2044. Hughes, M., C. Brenneman, and H. Gest. 1964. Feedback sensiti- vity of threonine deaminases in two species of photosynthetic bacteria. J. Bacteriol. EE: 1201-1202. Holdeman, L. V. and W. E. C. Moore. 1972. Anaerobe laboratory manual. Virginia Polytechnic Institute and State University, Blacksburg. Johnson, J. L. and B. S. Francis. 1975. Taxonomy of the clas- tridia: Ribosomal ribonucleic acid homologies among the species. J. Gen. Microbiol. EE: 229-244. Jungerman, K. A. gt_gl, 1970. Glycine formation via threonine and serine aldolase. Its interaction with the pyruvate- formate lyase pathway of one-carbon unit synthesis in Clostri- dium klpyveri. Eur. J. Biochem. g9; 424-429. Kalb, V. F. and R. W. Bernlohr. 1977. A new spectrophotometric assay for protein cell extracts. Anal. Biochem. EE, 362-371. Kemble, A. R. and H. T. McPherson. 1954. Determination of mono- amino monocarboxylic acids by quantitative paper chromato- graphy. Biochem. J. EE: 548-555. Kindler, S. H. and J. Mager. 1956. Nutritional studies with the Clostridium botulinum group. J. Gen. Microbiol. E2, 386-393. Kisumi, M., S. Komatsubara, and I. Chibata. 1973. Leucine accu- mulation by isoleucine revertants of Serratia marcescens resis- tant to a-aminobutyric acid: lack of both feedback inhibition and repression. J. Biochem. 1;: 107-115. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 98 Kisumi, M., S. Komatsubara, and I. Chibata. 1977. Pathway for isoleucine formation from pyruvate by leucine biosynthetic enzymes in leucine-accumulating isoleucine revertants of Serratia marcescens. J. Biochem. EE: 95—103. Komatsubara, S., K. Murata, M. Kisumi, and I. Chibata. 1978. Threonine degradation by Serratia marcescens. J. Bacteriol. 135: 318-323. Laver, W. G., A. Neuberger, and J. J. Scott. 1959. a-amino-B- keta acids. Part 11. Rates of decarboxylation of the free acids and the behaviour of derivatives on titration. J. Chem. Soc., p. 1483-1491. Lessie, T. G. and H. R. Whitely. 1969. Properties of threonine deaminase from a bacterium able to use threonine as sole source of carbon. J. Bacteriol. 100: 878-889. Lowry, O. H., N. J. Roseburg, A. L. Farr, and R. J. Randall. 1951. Protein measurement with the folin phenol reagent. J. Biol. Chem. 193: 265-275. Luginbuhl, G. H., J. G. Hofler, C. J. Decedue, and R. 0. Burns. 1974. Biodegradative L—threonine deaminase of Salmonella typhimurium. J. Bacteriol. 120: 559-561. Macy, J. M. 1979. The biology of gastrointestinal bacteroides. Ann. Rev. Microbiol. EE: 561-594. Mager, J., S. H. Kindler, and N. Grossowicz. 1954. Nutritional studies with Clostridium parabotulinum type A. J. Gen. Microbiol. l9: 130-141. McGilvray, D., and J. G. Morris. 1969. Utilization of L-threonine by a species of Arthrobacter. A novel catabolic role for aminoacetone synthase. Biochem. J. 112: 657-671. Measures, J. C. 1975. Role of amino acids in osmoregulation of non-halophilic bacteria. Nature 257: 398-400. Mitruka, B. M. and R. N. Costilow. 1967. Arginine and ornithine catabolism by Clostridium botulinum. J. Bacteriol. .EE: 295-301. Newman, E. B., V. Kapoor, and R. Potter. 1976. Role of L—threonine dehydrogenase in the catabolism of threonine and synthesis of glycine by Escherichia coli. J. Bacteriol. 126: 1245-1249. Perkins, W. E. and K. Tsuji. 1962. Sporulation of Clostridium botulinum. II. Effect of arginine and its degradation products on sporulation in a synthetic medium. J. Bacteriol. E3: 86-94. 49. 50. 51. 52. 53. 54. 55. 56. 57. 58. 59. 60. 61. 99 Phillips, A. T. and W. A. Wood. 1965. The mechanism of action of 5'-adenylic acid-activated threonine dehydrase. J. Biol. Chem. 240: 4703-4709. Phillips, A. T., J. I. Nuss, J. Moosic, and C. Foshay. 1972. Alternate pathway for isoleucine biosynthesis in Escherichia coli. J. Bacteriol. 109:714-719. Phillips, A. T., R. M. Egan, and B.Lewis. 1978. Control of bio- degradative threonine dehydratase inducibility by cAMP in energy restricted Escherichia coli. J. Bacterial. 135: 828-840. Pittman, K. A. and M. P. Bryant. 1964. Peptides and other nitrogen sources for growth of Bacteroides ruminicola. J. Bacteriol. EE: 401-410. Rabinowitz, K. W., R. A. Niederman, and W. A. Wood. 1973. The mechanism of action of 5'-adenylic acid-activated threonine dehydratase. VI. Identification of the partial reaction activated by adenosine monophosphate. J. Biol. Chem. 23E: 8207—8215. Roberts, R. B., P. H. Abelson, D. B. Cowie, E. T. Bolton, and R. J. Britten. 1955. Studies of biosynthesis in Escherichia coli. Carnegie Institution of Washington, Publication no. 607. Robinson, 1. M. and M. J. Allison. 1969. Isoleucine biosynthesis from 2-methylbutyric acid by anaerobic bacteria from the rumen. J. Bacteriol. 21: 1220-1226. Roessler, W. G. and C. R. Brewer. 1946. Nutritional studies with Clostridium botulinum, toxin types A and B. J. Bacteriol. El: 386-391. Schildkraut, I. and S. Greer. 1973. Threonine synthetase—catalyzed conversion of phosphohomoserine to a-ketobutyrate in Bacillus subtilis. J. Bacteriol. 115: 777-785. Scrutton, M. C. and M. R. Young. 1972. Pyruvate carboxylase, p. 1-34. Ep_P. D. Boyer (ed.), The enzymes, vol. VI. Academic Press, New York. Shizuta, Y. and O. Hayaishi. 1970. Regulation of biodegradative threonine deaminase synthesis in Escherichia coli by cyclic adenosine 3',5'-monophosphate. J. Biol. Chem. 245: 5416-5423. Shull, G. M. and W. H. Peterson. 1948. The nature of the "sporo- genes viatmin" and other factors in the nutrition of Clostri- dium sporogenes. Arch. Biochem. 18: 69-83. Shull, G. M., R. W. Thomas, and W. M. Peterson. 1949. Amino acid and unsaturated fatty acid requirements of Clostridium sporogenes. Arch. Biochem. EE: 227-241. 62. 63. 64. 65. 66. 67. 68. 69. 70. 71. 72. 73. 74. 100 Tokushige, M., H. R. Whiteley, and O. Hayaishi. 1963. Energy linked regulation of threonine metabolism by ADP. Biochem. Biophys. Res. Commun. 12: 380-385. Tokushige, M. and O. Hayaishi. 1972. Threonine metabolism and its regulation in Clostridium tetanomorphum. J. Biochem. 12, 469-477. Umbarger, H. E. and B. Brown. 1957. Threonine deamination in Escherichia coli. J. Bacteriol. 1;: 105-112. Umbarger, H. E. 1973. Threonine deaminases. Adv. Enzymol. E1: 349-395. Umbarger, H. E. 1978. Amino acid biosynthesis and its regulation. Ann. Rev. Biochem. £1: 533-606. Urata, G., and S. Granick. 1963. Biosynthesis of a-aminoketone and the metabolism of aminoacetone. J. Biol. Chem. 238: 811-820. ' Urm, E., T. Leisinger, and H. J. Vogel. 1973. Magnesium sensi- tivity of L-aspartate:2-oxoglutarate aminotransferase in Escherichia coli. Biochem. Biophys. Acta 302: 249-260. Utter, M. F. and H. M. Kolenbrander. 1972. Formation of oxal- acetate by C02 fixation on phasphoenospyruvate, p. 117-165. Ep_ P. D. Boyer (ed.), The enzymes, vol. VI. Academic Press, New York. Vanquickenborne, A. and A. T. Phillips. 1968. Purification and regulatory properties of the adenosine diphosphate-activated threonine dehydratase. J. Biochem. 243: 1312-1319. Whanger, P. D., A. T. Phillips, K. W. Rabinowitz, J. R. Piperno, J. D. Shada, and W. A. Wood. 1968. The mechanism.of action of 5'-adeny1ic acid-activated threonine dehydrase. II. Protomer-oligomer interconversions and related properties. J. Biol. Chem. gig: 167-173. Wiesendanger, S. and B. Nisman. 1953. La L-methionine demercap- todesaminase: un nouvel enzyme a pyridoxal-phosphate. Compt. rend. acad. sci. 237: 764-765. Willets, A. J. 1972. Metabolism of threonine by fusaria: growth on threonine as the sole carbon and nitrogen source. Antonie van Leeuwenhoek; J. Microbiol. Serol. EE, 591-603. Wood, W. A. and I. C. Gunsalus. 1949. Serine and threonine deaminases of Escherichia coli: activators for a cell—free enzyme. J. Biol. Chem. 181: 171-182. 75. 76. 77. 78. 101 Wu, J. I. J., H. Riemann, and W. H. Lee. 1972. Thermal stability of the deoxyribonucleic acid hybrids between the proteolytic strains of Clostridium botulinum and Clostridium sporpgenes. Can. J. Microbiol. EE} 97-99. Yamada, H., H. Kumagai, T. Nagate, and H. Yoshida. 1971. Formation of threonine aldolase by bacteria and yeasts. Agr. Biol. Chem._§§: 1340—1345. Yui, Y., Y. Watanabe, S. Ito, Y. Shizuta, and 0. Hayaishi. 1977. Multivalent induction of biodegradative threonine deaminase. J. Bacteriol. 132: 363-369. Yutaka, S. and O. Hayaishi. 1970. Regulation of biodegradative threonine deaminase synthesis in Escherichia coli by cyclic adenosine, 3',5'-monophosphate. J. Biol. Chem. 245: 5416- 5423.