A NUCLEOTEDE- PEPTEDE lSOLATED FROM BOVINE LEVER 'Fhesis fiar i'ho Dogma of Ph. D. MECHIGAN STATE UNIVERSITY David Richard Wilken 1960 This is to certify that the thesis entitled A NUCIEOTIDE-PEPTDJE ISOLATED FROM BOVINE LIVER presented by David Richard Wilken has been accepted towards fulfillment of the requirements for Doctor of Philosophz degree in Agricultural Chemistry A /\/VVOL"LK' LLMajor professor LIBRARY Michigan §i§£€ Univcsmfy A.NUCLEOTIDE-PEPTIDE ISOLATED FROM BOVINE LIVER By David Richard Wilken A THESIS Submitted to the School for Advanced Graduate Studies of Michigan State University of Agriculture and Applied Science in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Agricultural Chemistry 1960 és (‘3 7 \ \I U‘ “k L) i v \3 ("k ACKNOWLEDGEMENTS The author wishes to express his appreciation to Dr. R. G. Hansen for his helpful advice and continuous interest in this project. The author also wishes to acknowledge the counsel of Dr. N. E. Talbert concerning the identification of amino acids. The author is grateful to his wife, Jo Anne, for her help in the preparation and typing of this thesis. ii To J 0 Anne iii VITA David R. Wilken was born on February 20, 195A, at Amarillo, Texas. He later moved to Alton, Illinois, where he graduated from Alton Senior High School. From 1951 to 1955 he attended Blackburn College where he received his B. A. degree in Chemistry. He then began his advanced training at the University of Illinois in the Department of Dairy Science where he received his M; S. degree in 1958. He transferred to Michigan State University to complete his formal academic training and his research requirements for the degree of Doctor of Philosophy in the Department of Agricultural Chemistry. He has been awarded a National Foundation postdoctoral research fellowship to continue his studies at the Institute for Enzyme Research at the University of Wisconsin. iv A NUCLEOTIDE-PEPTIDE ISOLATED FROM BOVINE LIVER By David Richard Wilken AN‘ABSTRACT Submitted to the School for Advanced Graduate Studies of ‘Michigan State University of Agriculture and Applied Science in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Agricultural Chemistry 1960 David Richard Wilken ABSTRACT Numerous reports have appeared in the recent literature concern- ing the detection of peptide derivatives of nucleotides in a variety of biological materials. A new nucleotide-peptide has been isolated from.bovine liver extracts. It has been purified by resin column and paper chromatography. The purified compound has been shown to be homogeneous in several paper chromatographic and paper electro- phoretic systems. ‘Analyses of the peptide moiety of the nucleotide-peptide indi- cate that it is composed of the amino acids, glutamic acid, glycine, B-alanine, cysteic acid, and taurine. In addition, a sixth ninhy- drin reactive component has been detected but remains to be identi- fied. The N-terminal amino acid of the peptide has been identified as glutamic acid. The nucleotide moiety of the nucleotide-peptide has been identi- fied as 5',5'-adenosine diphosphate by absorption spectrum, enzymatic studies, and paper chromatography of its hydrolysis products. The nucleotide-peptide contains a third phosphate moiety which presumably is a component of the peptide portion of the molecule. Although the nature of the nucleotide-peptide linkage has not been conclusively established, the number of possibilities has been considerably re- duced. vi TABLE OF CONTENTS MODUCTION . O O O O O O O O O O O O O O O O Q 0 I O 0 . LIMA-“IRE REVEW . O O O O O O O O O O O I O O O O O O O Peptide Derivatives of UDP-Acetylmuramic acid. S-Adenosylmethionine and Related Compounds . . . . . Amino Acid Adenylates. . . . . . . . . . . . . . . . Soluble RNA-Amino Acid Complexes . . . . . . . . . Miscellaneous Amino.Acid or Peptide Derivatives of Nucleotides or Polynucleotides. . . . . . . . . m THODS . O O O Q O O O O O O O O O O O O O O I O O O O C RESULE . O O O O O O Q 0 O O O O O O O O O O O O O 0 O 0 Bovine Liver Nucleotides . . . . . . . . . . . . . Isolation and Purification of the Nucleotide-Peptide . Homogeniety of the Purified Nucleotide-Peptide . . . Components of the Nucleotide-Peptide . . . . . . . . Qualitative Amino Acid Content of the Peptide. . N-Terminal Amino Acid of the Peptide . . . . . . . . Locations of the Phosphate Groups. . . . . . . . . . Retention of the Integrity of the Nucleotide-Peptide Linkage During Treatment with 5 '-Nuc1eotidase. . . Nature of the Nucleotide-Peptide Linkage . . . . . . DISCUSSION. 0 O C O O O O O O O O . O O . C O C . 0 O O O SWY . I I O C O O O O O O O O C O O Q O O O O O O o O BIBLIW. 0 O O O O O O O O O O O O O O O O O O O O 0 vii 11 17 22 I. II. III. IV. TABLES Chemical analysis of bovine liver nucleotides. . . . . Paper chromatographic and electrophoretic mobility of the nucleotide-peptide. . . . . . . . . . . . . Molar ratios of the nucleotide-peptide . . . . . . . . Cleavage of inorganic phosphate from the nucleo- tide-peptide by 5'- and 5'-nuc1eotidases . . . . . . . viii 57 1+6 50 57 10. 11. 12. 15. 1h. 15. FIGURES Uridine diphOSphate acetylmuramic acid . . . . . . . S-adenosylmethionine . . . . . . . . . . . . . . . . An amino acid adenylate. . . . . . . . . . . . . . . A soluble RNA-amino acid complex . . . . . . . . . . Chromatography of bovine liver acid-soluble nucleo- tides on a Dowex-l (formate) resin column. . . . . . Chromatography of the partially purified nucleotide- ' peptide on a second Dowex-l (formate) resin column . Preparative paper chromatography of the nucleotide- peptide. . . . . . . . . . . . . . . . . . . . . . Acid and alkali stability of the nucleotide- peptidelinkage...............oooo Ultraviolet absorption spectrum of the nucleotide- peptide 0 0 O O O O O O O I O O O O O O O O O O O O O Chromatography of the amino acids obtained from the nucleotide-peptide . . . . . . . . . . . . . . . Chromatography of the DNP derivatives of the amino acids from the nucleotide-peptide. . . . . . . . . . Chromatographic identification of the nucleotide moiety of the nucleotide-peptide . . . . . . . . . . . Effect of 3'-nucleotidase on the chromatographic mobility of the nucleotide-peptide . . . . . . . . Retention of organic phosphate in the product of the reaction of the nucleotide-peptide with rye grass 5'-nucleotidase. . . . . . . . . . . . . . Periodate reactivity of the nucleotide-peptide before and after treatment with 3'-nucleotidase. 13 19 2h 5h 39 #1 hh 47 51 5h 59 61 INTRODUCTION INTRODUCTION A number of reports have appeared in the recent literature con- cerning nucleotides and amino acids or peptides combined by covalent linkage. Many of the compounds reported can be placed into one of four generally recognized classes: 1) Nucleotide-peptide compounds of the type originally isolated from penicillin treated Staphylococcus gugggg. This class of com- pounds is composed of various peptide derivatives of UDP-acetylmur- amic acidl'which are precursors of UDP-acetylmuramic acid-LaAla-D- Glu-L-Lys-DeAla-DeAla. In this case the linkage of the nucleotidic and peptidic moieties is an amide involving the carboxyl group of UDP-acetylmuramic acid and the amino group of L-alanine, the N-term- inal amino acid of the peptide. A similar compound in which diamino- pimelic acid replaces L-lysine has also been reported. These com- pounds are thought to be involved in the synthesis of the bacterial cell wall. Recently their biochemical formation has been more fully elucidated. 1The following abbreviations are used: Ala, alanine; AMP, ADP, and ATP, adenosine mono-, di-, and triphosphate; ADPR, adenosine diphos- phoribose; ATPR, adenosine triphosphoribose; DNP, the 2,h-dinitro- phenyl radical; DPN, oxidized diphosphopyridine nucleotide; DEAE- cellulose, N,N-diethylaminoethyl cellulose; FDNB, 1-fluoro-2,h-di- nitrobenzene; Glu, glutamic acid; GMP, GDP, and GTP, guanosine mono-, di-, and triphosphate; IMP, inosine monophosphate; Lys, lysine; Pi, inorganic phos hate; PPi, inorganic pyrophosphate; PAPS, adenosine- 3 '-phosphate-5 -phosphosu1fate; RNase, ribonuclease; UMP, UDP, and UTP, uridine mono-, di-, and triphosphate; UDPAG, UDPAH, UDH, and UDPGal, uridine diphosphate N-acetylglucosamine, N-acetylhexosamine, hexose, and galactose; TPN, oxidized triphosphopyridine nucleotide; EDTA, ethylenediaminetetraacetate; CpCpA, cytidylyl- -(3' -5 ')-cytidyl- yl- -(3' -5 ')-adenosine. 2) Nucleotides composed of a sulfur containing amino acid and 5'- deoxyadenosine which are chemically combined as thioethers or S-alka- lated thioethers (sulfonium compounds). Representative compounds of these types are S-adenosylhomocysteine, S-adenosylethionine, and S- adenosylmethionine. The latter compound is biologically important in the formation of Spermidine and also functions as a methyl group donor. 3) Nucleotides composed of individual amino acids and adenosine- 5'-phosphate in an anhydride linkage. In this case, the amino acid and the nucleotide are combined via a mixed carboxyl-phosphate anhy- dride bond. These compounds, which normally are enzyme-bound, are believed to represent the initial product of amino acid activation required for protein synthesis. h) Nucleotide amino acid ester complexes composed of various amino acids and soluble RNA (S-RNA). The ester linkage in S-RNA- amino acid compounds involves the carboxyl group of the amino acid and either the 2' or 3' hydroxyl group of the terminal nucleotide of the S-RNA. Such compounds appear to be intermediates in the protein synthetic sequence lying between initial amino acid activation and peptide bond formation. Other reports have appeared in the literature concerning the de- tection of nucleotide-peptide complexes which probably are not inti- mate members of the above chemically defined groups. They appear to be diversified but apparently have one general feature in common, namely some type of nucleotide-amino acid or nucleotide-peptide asso- ciation. Such compounds have been obtained from fish, chlorella, yeasts, bacteria, fungi, spores, and mammalian tissue. Although no evidence concerning the nature of the nucleotide-peptide linkage has been reported in most cases, peptide hydroxamates have been observed after reacting some nucleotide-peptide preparations with hydroxyl- amine. In addition, some nucleotide-peptide preparations have been cleaved by treatment with aqueous hydrazine forming hydrazides of peptides, indicating covalently linked nucleotide-peptides. The question of the biochemical function of the above nucleotide-peptide compounds has not yet been answered. The present report concerns the isolation and identification of a new nucleotide-peptide obtained from bovine liver. LITERATURE REVIEW LITERATURE REVIEW Peptide Derivatives of UDP-Acetylmuramic Acid The first report concerning peptide derivatives of UDP-acetyl- muramic acid (Figure 1) came from Park and Johnson (1) who were at- tempting to elucidate the mechanism of action of penicillin. It had previously been observed that Staphylococcus aureus cells increased in size in a medium containing penicillin, although they did not diVide. Therefore, it seemed of interest to ascertain whether or Inot this increased growth was a function of an increased amount of one or several biological intermediates or due to some other cause. Dry weight, nitrogen, phoephorus, nucleic acid phOSphorus, and nucleic acid increased at comparable rates during growth in the presence of penicillin, although the increases in these constituents were only about one-half of the increase observed in untreated cells. In con- trast, a rather marked change in the distribution of acid-soluble phosphate was observed. While the per cent increase in acid-stable phOSphate was about equal in penicillin treated and non-treated cells, the amount of inorganic phosphate was decreased in treated cells. The most interesting change observed, however, was that while normal cells increased in acid-labile phoSphate two fold, there was a three fold increase of acid-labile phosphate in penicillin treated cells. The acid-labile phosphate compounds were partially purified by precipita- ting the barium salts with alcohol. In addition to the presence of pentose and uracil moieties, such preparations contained one mole of Boo 280583.600 meccamozqfi 86:: _ 8:9“. I. :oowolo...:o :o IO .6 i :z I M/o III\._ :0 :o _...lm_e IO :0... on n... on n__.. o o\\_ o. m \2/ \\O O O I\ /O m\} __ _ IONIMW acid-stable phosphate and 0.7 mole of potential reducing power per mole of acid-labile phosphate. These several observations were the beginning of numerous inves- tigations which in the past eleven years have led to elucidation of the structure, almost the entire biosynthetic pathway and a proposed function of the peptide derivatives of UDP-acetylmuramic acid. Park (2) succeeded in purifying the acid-labile phosphate pre- paration described above and resolved it into three distinct compo- nents by means of chromatography. All were derivatives of uridine containing molar ratios of labile phosphate:stable phOSphate:uridine: potential reducing power of l:l:1:0.8. The major difference in the three compounds was their nitrogen content which approached molar ratios of 3, h, and 9 for compounds 1, 2, and 3 respectively. The increased amount of nitrogen in compounds two and three was due to the presence of amino acids. Compound one was studied in.more detail by Park (3). He was able to establish that it was a derivative of uridine monophosphate and contained a pyrophosphate linkage as well as a glycosidic-phos- phate linkage. The postulated structure was a uridine diphosphate- reducing substance in which the latter was combined with the terminal phosphate of UDP by an acetal-like linkage. Although the exact iden- tification of the reducing substance, acetylmuramic acid, was not ascertained at that time, it was indicated that it was an acidic amino sugar with no‘gig hydroxyl groups and that the amino group was acetylated. The identification of the reducing sugar as acetylmur- amic was reported by Strange (h) and the structure suggested by him was later confirmed by synthesis (5). Park's compounds two and three were found to contain one L-ala- nine residue and a peptide respectively (6). The glycopeptide portion of compound three was separated from.the nucleotide moiety after mild acid hydrolysis and was found to be composed of three moles of alanine, one mole of L-lysine, and one mole of D-glutamic acid. Configura- tional analysis of the alanine residues of the peptide indicated a mixture of the D and L isomers. Recently it has been established that two of the alanine residues are D-alanine and the other residue is L-alanine (7). Later experiments of Strominger (8) more clearly defined several of the parameters concerning the accumulation of the UDP-acetylmur- amic acid derivatives in the presence of penicillin. Time-course studies indicated that the accumulation commenced immediately upon the addition of penicillin and that the half-time for maximum.accumu- lation was about fifteen minutes. Several other antibiotics inclu- ing aureomycin, terramycin, chloromycetin, and streptomycin at concen- trations up to 100 times the amount of penicillin required to obtain accumulation of nucleotide derivatives caused little or no accumula- tion of these compounds. Streptomycin could cause some accumulation of the compounds, however 1000 times as much of it was required com- pared to penicillin. These two findings supported the idea that penicillin was acting in a relatively specific rather than a broad capacity. It was also observed that the accumulation could be in- duced in resting cultures supplied only inorganic salts, glucose, and either the amino acids L-lysine, L-alanine, and L-glutamic acid or larger amounts of glutamic acid alone. This latter finding in con- junction with better isolation procedures has greatly aided more recent studies of these compounds. The presence of the UDP-acetylmuramic acid derivatives in normal as well as in penicillin treated Staphylococcus aureus (8,9), £2239- bacillus helveticus 335 (8) and in normal Streptococcus hemolyticus (10) indicated that these compounds might be of general importance in microorganisms. Indeed, the similarity of the molar ratios of the cell wall components of some bacteria to the amino acid and muramic acid content of UDP-acetylmuramic acid-peptide led Park and Strominger (11) to propose that this compound was a precursor of the bacterial cell wall. They found that hydrolysates of.§..§g£gg§ cell wall con- tained molar ratios of D-glutamic acid:alanine:lysine:muramic acid of 1:3:l:l, the same as is present in the UDP-acetylmuramic acid- peptide. The first estimates of the relative amounts of D and L isomers of alanine in the nucleotide-peptide (6) and in the cell wall hydrol- ysates (11) indicated approximately equal amounts of these components. In order for this finding to be consistent with the known number of alanine residues in UDP-acetylmuramic acid-peptide, namely three, it was necessary to consider the possibility that there were two nucleo- tide-peptides with.similar properties. One such compound would have two L-alanine residues and one D-alanine residue, the other would have two D-alanine residues and one L-alanine residue. Such an in- terpretation would require the further postulate that the latter two hypothetical compounds be isolated in equimolar amounts and that they be incorporated into cell wall in equal proportions. The above possi- bility seemed rather remote. Strominger and Threnn (7) therefore re- investigated the D-alanine and L-alanine content of.§..§g£gg§ cell wall and also the nucleotide-peptide. Indeed, they found that in both cases 67 per cent of the alanine was the D isomer and 33 per cent was the L isomer. This finding put the postulate that a single carbohydrate-containing peptide is incorporated into cell wall mate- rial on a more firm basis. Although the function of the nucleotide-peptide had been pro- posed, the sequence of the amino acids in the peptide was not known. It was inferred that the N-terminal amino acid was L-alanine because it was assumed that UDP-acetylmuramic acid-L-alanine was an inter- mediate in the formation of the complete nucleotide-peptide. To establish the amino acid sequence in the peptide, Strominger (l2) hydrolyzed the intact nucleotide-peptide with various concentrations of acid and succeeded in isolating several of the partial hydrolysis products. From the determination of the amino acid content of the various fragments, the number of possibilities for the amino acid sequence of the peptide could be reduced to four, namely, 1. Ala-Glu-Lys-Ala-Ala 2. Ala-Lys-Glu-AlaeAla 3. Ala-Ala-Glu-Lys-Ala h. Ala-Ala-Lys-Glu-Ala Sequence 1 was determined to be the proper sequence on the basis of the isolation of a new UDP-acetylmuramic acid-peptide (12,13). This nucleotide-peptide accumulated when a resting culture of §. aureus 10 was placed in a lysine deficient medium. It contained one residue each of L-alanine and D-glutamic acid. The new compound was con- verted to the complete nucleotide-peptide when cells in which it had accumulated were transferred to a medium containing lysine. This in- dicated that the new nucleotide-peptide was an intermediate in the formation of the complete nucleotide-peptide. Furthermore, this series of findings were consistent only with the possibility that the intact nucleotide-peptide had the sequence: UDP-acetylmuramic acid-Lqua-D-Glu-L-Lys-D-Ala-D-Ala. Subsequent isolation of a UDP- acetyLmuramic acid-peptide (1h) containing one mole each of L-alanine, glutamic acid, and lysine is consistent with the above peptide se- quence. The several nucleotide-peptides discussed above all appeared to be intermediates in the synthesis of the final nucleotide-peptide compound. Thus, it seemed probable that the amino acids were being incorporated singly and not as a preformed peptide unit. Recent ex- periments of Ito and Strominger (15) have elucidated this problem. These workers have separated the enzymes responsible for the incor- poration of L-alanine, D-glutamic acid, and L-lysine singly and in that order into their respective substrates. In addition, an enzyme forming the dipeptide D-alanyl-D-alanine was also found and this di- peptide is incorporated intact into the complete nucleotide-peptide. Each reaction was found to require ATP as an energy source and a di- valent metal such as Mn++. Although the entire sequence of the components of the complete UDP-acetylmuramic acid-peptide, as well as most of the structure and 11 most of the biosynthetic pathway is known, several problems remain to be answered. Insofar as structure is concerned the problems yet to be solved are: l) the configuration of the lactic acid moiety of the molecule, and 2) the nature of the bonding of the glutamyl and ly- sinyl residues. It has not yet been established whether the linkages of these two amino acids are the same as normally found, that is that they involve the ahcarboxyl andObamino groups, or whether the amide linkages involve the yacarboxyl group of glutamic acid and/or the e-amino group of lysine. In addition, there appears to remain at present at least one missing step in the biosynthesis of the nucleo- tide-peptide, namely the synthesis of UDP-acetylmuramic acid. Strom- inger (16) has reported that cell-free extracts of.§.‘§g£gg§ and E, 321; are able to phosphorylitically condense phosphoenolpyruvate and UDPAC to form UDPAG-enol-pyruvate. This compound is a likely inter- mediate which would require the addition of a molecule of hydrogen to form UDP-acetylmuramic acid. Preliminary results indicate the pre- sence of a UDP-acetylmuramic acid-peptide containing diaminopimelic acid in place of lysine (17). Further work will be required to elu- cidate its structure and biosynthesis. S-Adenosylmethionine and Related Compounds Although methylation of nicotinamide employing methionine as the methyl donor had previously been demonstrated under aerobic conditions, Cantoni (18) was the first to obtain a cell-free preparation in which the methylation could be carried out under anaerobic conditions. In this case, an ATP generating system or ATP itself could replace the previously required oxygen dependent systems. Later experiments by l2 Cantoni (19) demonstrated that an unidentified intermediate required for the transmethylation reaction was formed by incubating ATP, Mg++, and methionine with a pig liver protein fraction. This finding was soon followed by isolation of the intermediate and presentation of a proposed structure (20,21) (Figure 2). The proposal of this struc- ture was based to a large degree on the following findings: 1) The compound exhibited an absorption spectrum identical with adenosine and had a ratio of adenine:labile methyl groupszpentose of l:l:l. When it was enzymatically prepared from C14 or 835 labeled methionine, it contained radioactivity. 2) Acid hydrolysis liberated three de— tectable compounds: adenine, homoserine, and an unidentified sulfur containing compound. 3) Evidence in favor of the postulated sulfon- ium ion included electrophoretic migration as a cation at pH 7.8 and failure to react with L-amino acid oxidase. This enzyme appears to attack only amino acids whose dipolar ions have their charges close together (22,23). On this basis, S-adenosylmethionine, like the methyl sulfonium derivative of methionine (2h), would not be expected to be attacked by L-amino acid oxidase. Additional support for the proposed structure was gained when it was found that during heating at 1000 at pH 7 S-adenosylmethionine was cleaved to homoserine and adenine thiomethylribose, and that the untreated compound reacted with periodate which indicated no 2' or 3' hydroxyl substitution (25). In addition, treatment with 0.1 N alkali for ten minutes at 250 yielded almost entirely adenine and S-ribosylmethionine (26). The final support for the structure has been its complete chemical syn- thesis (27). 13 I mchEEESmocmoqnm N 8:9... Inw IO olo/ I _\_ _/ /IO \._ .12 ~Iloml NIo! NIio wl QIOI Io IO 1h The chemically synthesized S-adenosylmethionine is not completely enzymatically active because presumably all four of the possible ster- ioisomers (D-L methionine, i sulfonium ion) are formed in the synthe- sis employed and the various methylpherase enzymes are relatively Specific for one of the isomers (28). A newer synthesis of S-adeno- sylmethionine combining enzymatic and chemical methods has reduced the number of sterioisomers obtained to two (28). In this synthesis, S- adenosyl-L-homocysteine was first prepared enzymatically from adeno- sine and L-homocysteine employing a rat liver condensing enzyme which is specific for L-homocysteine (29). Methylation of the en- zymatic product yielded a preparation of (i) S-adenosyl—L-methionine which was 50% enzymatically active. Using S-adenosylmethionine pre- pared in this manner, it was demonstrated that the (-) S-adenosyl- methionine is the substrate for guanidoacetic acid and catechol methylpherases and also for the yeast enzyme which cleaves S-adeno- sylmethionine to form 5'-methylthioadenosine and 2-amino-h-butyro- lactose. This appears to be the first biological resolution of a sulfonium.ion. The major function of S-adenosylmethionine appears to be its ability to serve as a source of methyl groups in transmethylation reactions, although it also undergoes other interesting biological reactions. Evidence has been obtained for its role as a methyl donor to form the following specific compounds or general types of com- pounds: a) Nemethylnicotinamide (18), b) creatine (30), c) 0~methyl- catechols (31), d) ergosterol (32), e) N-methylamino purines (33), f) methionine (via transmethylation with homocysteine) (3n), and g) 15 choline (35). S-adenosylhomocysteine has been identified as the re- maining product in creatine synthesis (36), and it presumably is one of the products in the other transmethylation reactions described above. Its structure has been demonstrated by synthesis (37). Transethylation involving S-adenosylethionine has also been ob- served (38). This compound is found in Torulopsis utilis grown in the presence of ethionine. Formation of the S-adenosylethionine and subsequent transethylation may be due to incomplete specificity of the enzymes responsible for the analogous reactions involving meth- ionine. In addition to its transmethylation function, S-adenosyl- methionine is involved in spermidine synthesis (39). In this case it is the 05 B, and 7 carbons of the methionine moiety which are transferred to putrescine after an initial decarboxylation of the methionine moiety. S-adenosylmethionine also undergoes a biochemi- cal cleavage (h0,hl) similar to one of its known chemical cleavage reactions (#2). In both cases, the products are 5'-methylthio- adenosine and 2-amino-h-butyrolactone. The mechanism of the enzy- matic cleavage has been studied (R3). Although no mechanism has been proved, an initial elimination of 2-amino-3-butenoic acid as an intermediate in the reaction has been excluded since no tritium from the aqueous medium.was incorporated into the 2-amino-h-butyrolactone formed. A second possible mechanism.which has not been tested is a nucleophylic attack on the carbon adjacent to the sulfur atom by a carboxyl oxygen with the elimination of 2-amino-h-butyrolactone. One of the most interesting aspects of S-adenosylmethionine metabolism is its biological formation. The enzyme(s) catalyzing 16 its formation from.ATP and methionine and the proper metals have been partially purified from rabbit liver (AL) and yeast (#5) and their properties have been compared. Enzymes from both sources require high concentrations of M'g++ for optimal activity and both require monovalent cations such as N114+ or K+ for activity. Some differences between the enzymes have been found, such as the observation that the liver enzyme specifically requires Mg**'while the yeast enzyme re- aponds, although less well, to other divalent cations. In addition, while fluoride inhibits the rabbit liver enzyme, it appears to have no effect on the yeast enzyme even at higher concentrations than required to demonstrate inhibition of the liver enzyme. Although differences do exist between the enzymes, the overall reaction ap- pears to be the same. It has been demonstrated that the products of the reaction are inorganic phosphate and pyrophosphate as well as S-adenosylmethionine. The inorganic phosphate arises from the 7 or terminal phosphate of ATP and the pyrophOSphate arises from the a . and B phosphates of ATP. Although inorganic phosphate is liberated from the terminal phosphate of ATP during the reaction, it has not been possible to demonstrate any participation of free ADP in the reaction. Furthermore, neither has it been possible to separate the overall reaction into two or more fractions nor to demonstrate an exchange of Pi32 into ATP or PPi32 into ATP either with or without methionine in the presence of the purified enzyme. Incorporation of 014-1abeled methionine into S-adenosylmethionine in the presence of the purified enzyme also could not be demonstrated. Observations reported thus far indicate that the entire reaction may be catalyzed 17 by a single enzyme, however, confirmation of this possibility must await absolute purification of the enzyme(s). Aside from the inter- esting elimination of Pi and PPi, the reaction represents a rather unique utilization of ATP, namely as an adenosine donor. Amino Acid Adenylates Although other reports had appeared in the literature which demonstrated the incorporation of radioactive amino acids into pro- tein, Zamecnik and Keller (#6) were the first to demonstrate incor- poration under anaerobic conditions. Their system required an ATP generating system, amino acids, a soluble, heat-labile, non-dializ- able fraction, and a microsome-rich fraction into which amino acids were incorporated. Later systems were simplified in that the mem- branous material could be removed from the microsomes leaving ribo- nucleoprotein particles into which amino acids were incorporated (#7). In ascites tumor cell preparations it was observed that ATP could replace the previously required ATP generating systems (#7). As the overall protein incorporating system was refined, it became evident that GDP or GTP also played some role in the synthetic pro- cess (#8). Study of individual components required for protein synthesis has lead to the discovery of amino acid adenylates and the S-RNA-amino acid complexes. The first report of amino acid carboxyl group activation came from Hoagland and co~workers (49,50) as a result of their studies on protein synthesis. The enzymatic activation was dependent upon ATP, amino acids, and a protein fraction obtained from the supernatant (soluble fraCtion above) of rat liver homogenates by precipitation 18 at pH 5.2. Enzymatic reactions were measured either by exchange of PP132 into ATP or by the formation of amino acid hydroxamates in the presence of hydroxylamine. In the latter case the hydroxylamine acts as an acceptor of the activated amino acid. Although net accumula- tion of the activated amino acid could not be demonstrated, a con- comitant equimolar formation of PPi and amino acid hydroxamate could be observed in the presence of hydroxylamine. Failure of C14-labeled AMP to exchange with.ATP under conditions in which PPi32 rapidly ex- changed with ATP indicated that the activated amino acid and AMP were enzyme-bound. The above findings were consistent with the postulated intermediate, namely, a tightly enzyme-bound amino acid adenylate in which the mode of linkage was a carboxyl-phosphate anhydride bond (Figure 5). It was demonstrated that the pH 5 fraction supplemented with GTP could incorporate amino acids into microsomal protein in the presence of ATP. Thus the carboxyl activation represented the initial step in protein synthesis. Of several amino acids tested none showed com- petitive inhibition of activation of other amino acids. Instead, an additive PP132 exchange with ATP or hydroxamate formation was ob- served. This finding, as well as partial separation of leucine, ala- nine, and methionine-activating enzymes indicated that each.amino acid has a Specific activating enzyme. Support for this hypothesis has come in the form of purification of several of the specific amino acid-activating enzymes (51-57). Support for the carboxyl-phosphate anhydride structure of amino acid adenylates has come from several laboratories in various forms. l9 20388 28 05:6 :4 m 8:9“. IO IO H\mllm/_ / I\._ I... .e .IIIIIoomuowloum Io \__ _ 20 De Moss.gt‘§l. (57) were able to synthesize leucine adenylate using the disilver salt of AMP and the acid chloride of L-leucine. The stability of the product decreased with increasing pH between pH 5.1 and 8.0. In the presence of the synthetic compound, PPi, and a puri- fied leucine-activating enzyme fromuE. 291;, a net appearance of ATP could be observed with a concomitant equimolar decrease of the leucine adenylate. This demonstrated that the proposed amino acid adenylates could react in the reverse direction of amino acid activa- tion and thus could represent the product of amino acid activation. Reversibility of this reaction had previously been indicated by the PP132 exchange into ATP (#9). The formation of ATP from L-leucine adenylate and PPi appeared to be Specific for L-leucine adenylate Since neither D-leucine adenylate nor L-alanine adenylate showed appreciable capacity to form ATP although the latter had been active with crude extracts. These findings were consistent with the sugges- tion of specific amino acid-activating enzymes. The fact that the synthetic material reacted with hydroxylamine at lower concentrations than required in the enzymatic reaction, combined with the observa- tion that there was no net breakdown of ATP in the enzymatic reaction of ATP, leucine, and enzyme even in the presence of inorganic pyro- phosphatase, emphasized that the reaction product was tightly enzyme- bound. Additional confirmation of the proposed carboxyl-phOSphate anhy- dride structure of amino acid adenylates has come from isotope studies. Hoagland gt 31. (58) and Bernlohr and Webster (55) have studied the fate of carboxyl labeled 018 of tryptophan and alanine. 21 In both experiments a partially purified specific amino acid-activa- ting enzyme was used. The enzyme, ATP, amino acid, and hydroxyl- amine were incubated and the inorganic pyrophosphate and phosphate from AMP which.were formed were collected for determination of 018 content. It was found that 018 was incorporated into the phOSphate oxygen of AMP in nearly the theoretical amount and that no 018 was incorporated into the oxygen of the inorganic pyrophosphate. This finding was consistent with the idea that the oxygen of the carboxyl group condensed with the stable phOSphate of ATP via a nucleophylic attack with concomitant release of inorganic pyrophOSphate. The amino acid adenylate thus formed was then cleaved at the carbon oxygen linkage of the anhydride bond to form the amino acid hydroxamr ate and AMP containing 018. The final confirmatory evidence that the product of amino acid activation was an amino acid adenylate came from Karasek _£‘§l. (59). These workers found that in the presence of substrate amounts of tryptophan-activating enzyme, ATP, tryptophan, and inorganic phos- phatase, tryptophan adenylate was formed and could be isolated from the incubation mixture. The isolated product had the electrophoretic mobility of synthetic tryptophan adenylate and reacted rapidly with hydroxylamine to form the expected tryptophan hydroxamate. This finding has been confirmed and extended by Kingdon‘gt‘gl. (60). These workers identified the enzymatic reaction product as being the same as synthetic material by electrophoresis, resin column chroma- tography, reactivity with hydroxylamine, and by its ability to under- go the reverse enzymatic reaction to form.ATP. Only one other amino 22 acid adenylate, namely serine adenylate, has been isolated thus far (56). Soluble RNAPAmino Acid Complexes The discovery of S-RNA-amino acid complexes as well as part of the structural identification Of these complexes came as the result of continued study of protein synthesis. These complexes have been found to be the product of the transfer of the amino acid of amino acid adenylates to S-RNA. This represents the second step in protein synthesis. Hoagland 35 21. (61,62) were the first to demonstrate that RNA present in the amino acid-activating enzyme of rat liver (usually termed soluble RNA, S-RNA, or transfer RNA) became labeled when it was incubated with the pH 5 enzyme (from the soluble fraction above), ATP, Mg++, and C14-amino acid. Berg (63) has indicated that the amino acid-activating enzyme may also catalyze the transfer of the activated amino acid to S-RNA. This suggestion came from the obser- vation that during approximately a hundredfold purification of methi- onine-activating enzyme the ratio of amino acid activation to incor- poration into S-RNA was constant. Incorporation of a mixture of amino acids into S-RNA was additive (62) as in the case of amino acid activation and indicated the existence of a specific S-RNA for each amino acid. Furthermore, it was found that preformed S-RNA- leucine could transfer the amino acid to microsomes in the presence of GTP and an ATP generating systemc Since GTP was not required for incorporation of amino acids into S-RNA, it appears that it functions after the formation of the S-RNA-amino acid complex. Some component 25 of the pH 5 fraction also seemed to be necessary for the transfer of the amino acid of the S-RNA-leucine to ribonucleoprotein, since care- fully washed microsomes incorporated the amino acid moiety only poor- ly unless the pH 5 fraction was also present. That the pH 5 fraction was not merely reconverting the S-RNA-leucine to leucine adenylate ‘which in turn might transfer the amino acid residue to the microsomes was indicated by the failure of Ole-leucine to lower labeling of the microsomes, and by the failure of S-RNA-leucine pretreated with mild alkali to incorporate the amino acid into microsomes. Treatment of the S-RNA-amino acid complex with mild alkali liberates the amino acid from.the S-RNA. Several other properties of the S-RNAramino acid complex were studied (62). It was noted that the compound was non-dializable and stable against water, 10 per cent NaCl, and 8 M urea. In addition, ,the amino acid was liberated by treatment with 0.01 N alkali for twenty minutes at room temperature, however, at pH h to 6 it was relatively stable. It reacted with anhydrous hydroxylamine to form the respective amino acid hydroxamate indicating carboxyl activation. Since several of these properties were similar to the amino acid adenylates, it seemed possible that a carboxyl-phosphate anhydride linkage was present although other possibilities were not excluded. Among other possibilities mentioned by Hoagland‘gg‘gl. (62) was an ester involving the carboxyl group of the amino acid and the 2' or 3' hydroxyl group of the S-RNA (Figure A). It is now known that this is the mode of linkage and that the nucleotide moiety involved is the adenosine terminal nucleotide of the S-RNA. 24 $388 Boo oc_Eo-