Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Mechanistic study of the oxidation of pteridine derivatives Oyama, Kiyotaka 1973

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-UBC_1973_A1 O93_4.pdf [ 6.44MB ]
Metadata
JSON: 831-1.0060148.json
JSON-LD: 831-1.0060148-ld.json
RDF/XML (Pretty): 831-1.0060148-rdf.xml
RDF/JSON: 831-1.0060148-rdf.json
Turtle: 831-1.0060148-turtle.txt
N-Triples: 831-1.0060148-rdf-ntriples.txt
Original Record: 831-1.0060148-source.json
Full Text
831-1.0060148-fulltext.txt
Citation
831-1.0060148.ris

Full Text

e 1 nan-A MECHANISTIC STUDY OF THE OXIDATION OF PTERIDINE DERIVATIVES BY KIYOTAKA OYAMA B.Engineering, Kyushu University, 1967 M.Engineering, Kyushu University, 1969 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in the Department of CHEMISTRY We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA August, 1973 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make i t freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the Head of my Department or by his representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department The University of British Columbia Vancouver 8 , Canada - i i -ABSTRACT Supervisor: Professor R. Stewart The permanganate oxidation of 6,7,8-trimethyllumazine (TML) (I, R = CH^ ) was investigated and its mechanism was elucidated by means of kinetic and product studies. The oxidations of 6,7-diphenyl-8-methyllumazine (I, R = C-Hc) with permanganate and of TML with potassium trans-1,2-D D diaminocyclohexanetetraacetatomanganate(III) (KMn''"''"'''CyDTA, where CyDTA represents the ligand) were also studied, as an aid in understanding the oxidation process. (I) It was found that permanganate oxidation of TML consists of a permanganate-independent path and permanganate-dependent path. The former is subject to general base catalysis by the phosphate buffer species above pH 3.5, with the rate being almost identical with that for the hydrogen-deuterium exchange reaction of the 7-methoyl group of TML multiplied by a factor of three. A kinetic isotope effect of 6.89 at 31.4° was obtained for TML deuterated at the 7-methyl group. From these results i t was concluded that the i n i t i a l step of the reaction is rate-determining removal of a proton by attack of general bases at the 7-methyl group to produce an enolate-anion. The mechanism of the subsequent steps was considered to be as follows; the enolate-anion undergoes - i i i -an electron transfer rather than the addition reaction with permanganate, yielding a free radical intermediate, which reacts further with the permanganate to give a manganate ester intermediate. The latter decomposes via hydration at the 7-position to give formaldehyde and 6,8-dimethyl-7-oxolumazine. The permanganate-independent path below pH 3.5 is subject to general acid catalysis by hydronium ion as well as by the phosphate buffer species. The oxidation is somewhat faster than the exchange and a kinetic isotope effect of 2.46 at 31.4° was obtained for TML deuterated at the 7- methyl group. These facts were explained by assuming the presence of an hydration route as well as the enolization route. The permanganate-dependent path was found to be subject to specific acid catalysis and there was no kinetic isotope effect for the compound. Addition of permanganate to the carbon-carbon double bond at the 6,7-position of the protonated TML is suggested as the mechanism for the reaction. From the results obtained in the present study the utility of the oxidation method, which is frequently used to locate the site of hydration in heterocyclic systems, has been reexamined and additional limitations to those appearing in the literature regarding the method are suggested. Similarities between the reported enzymic conversion of 6,7-dimethyl-8- ribityllumazine to 6-methyl-7-oxo-8-ribityllumazine and the oxidation of TML with KMn^ ^^ CyDTA under aerobic conditions are pointed out. The mechanism of the latter was investigated and an autoxidation path initiated by electron transfer from the enolate-anion to KMn^^CyDTA is proposed. - i v -During attempts to synthesize a model compound f o r f l a v i n nucleotides, i t was discovered that, methylation of 3-amino-2-methylaminopyridine with methyl iodide gives a product e x c l u s i v e l y a l k y l a t e d at the 3-amino nitrogen, rather than at the rin g nitrogen as i s found to be the case with most aminopyridines. The d i v e r s i o n of methylation from the r i n g nitrogen to the 3-amino nitrogen i s a t t r i b u t e d to a combination of s t e r i c and hydrogen bonding e f f e c t s . - v -TABLE OF CONTENTS Page 1. INTRODUCTION 1 1.1 The Pteridines and Lumazines 1 1.1.1 Physical Properties 3 1.1.2 Tautomerism 4 1.1.3 Electron Distribution 5 1.1.4 Covalent Hydration 5 1.1.5 Deuterium Exchange Reaction 12 1.1.6 Metabolism of Lumazines. Biosynthesis of riboflavin and 6-methyl-7-oxo--8-ribityllumazine from 6,7-dimethyl-8-ribityllumazine 14 1.2 Oxidation of Organic Substrates with Permanganate and with Potassium trans-1,2-Diaminocyclohexanetetra-acetatomanganate (III) 22 1.2.1 Oxidation of Alkenes with Permanganate 22 1.2.2 Oxidation of Ketones with Permanganate 23 1.2.3 Oxidation of Organic Substrates with Potassium trans-1,2-Diaminocyclohexanetetra acetatomanganate (III) 24 2. SCOPE OF THE INVESTIGATION 26 3. RESULTS AND DISCUSSION 29 3.1 Permanganate Oxidation of 6,7,8-Trimethyllumazine.. 29 3.1.1 Results 29 3.1.2 Discussion 49 3.1.2.1 Permanganate-independent Path 53 3.1.2.2 Permanganate-dependent Path 74 3.1.3 Conclusion 77 - vi -Page 3.2 Permanganate Oxidation of 6,7-Diphenyl-8-methyl-lumazine 80 3.2.1 Introduction 80 3.2.2 Results 80 3.2.3 Discussion 82 3.2.4 Conclusion 96 3.3 Further Discussion 96 3.3.1 Keto-enol Tautomerism of 6,7,8-Trimethyllumazine 96 3.3.2 Oxidation as a Method of Locating the Site of the Hydration in Heterocyclic Systems: Reexamination of its Utility 99 3.3.3 Mechanism of Enzymic Conversion of 6,7-Dimethyl-8-ribityllumazine to 6-Methyl-7-oxo-8-ribityl-lumazine 102 4. EXPERIMENTAL 112 4.1 Apparatus and Compounds 112 4.2 Preparation of Buffer Solutions 115 4.3 Kinetic Procedure 116 4.4 Stoichiometry and Product Analysis 120 4.4.1 Permanganate Oxidation of 6,7,8-Trimethyl-lumazine 120 4.4.1.1 At pH 6.0 120 4.4.1.2 At pH 1.0 121 4.4.2 Permanganate Oxidation of 6,7-Diphenyl-8-methyllumazine 129 4.4.3 Permanganate Oxidation of 2,4-Dioxo-5-amino-6-methylaminopyrimidine 130 4.4.4 Oxidation of 6,7,8-Trimethyllumazine with Potassium trans-1,2-Diaminocyclohexanetetra-acetatomanganate (III) 131 - v i i -Page APPENDIX. Position of Methylation of 2,3-Diaminopyridine and 3-Amino-2-methylaminopyridine 137 1. Introduction 137 2. Results 139 3. Discussion 142 4. Experimental 144 BIBLIOGRAPHY 153 - v i i i -LIST OF TABLES Table Page 1 Rate constants for TML oxidation at various permanganate concentrations and various pH 32 2 Effect of buffer concentration on the rate constant for TML oxidation at pH 6.0 39 3 Rate constants for TML oxidation at different buffer and permanganate concentrations at pH 2.0 41 4 Rate constants for the permanganate-independent path and the permanganate-dependent path of the oxidation of TML 44 5 Rate constants for the permanganate oxidation of TML-7-d3 at pH 6.0 51 6 Rate constants for the permanganate oxidation of TML-7-d3 at pH 2.0 52 7 TML-Oxidation rates (permanganate-independent path) and TML-exchange rates 57 8 Rate constants for the permanganate oxidation of 6,7-diphenyl-8-methyllumazine 83 9 Rate constants for the permanganate-independent path and the permanganate-dependent path for the oxidation of 6,7-diphenyl-8-methyllumazine 86 10 Melting points and parent mass peaks of the compounds obtained from the oxidation of TML with KMnIi:ECyDTA under nitrogen atmosphere 132 - ix -LIST OF FIGURES Figure Page 1 Typical pseudo first-order rate plot for the permanganate oxidation of TML 30 2 Relation between rate constant and permanganate concentration in the oxidation of TML at pH 0.997 and 1.39 34 3 Relation between rate constant and permanganate concentration in the oxidation of TML at pH 5.0 and 6.0 35 4 Effect of buffer concentration on the rate of TML oxidation at pH 6.0 40 5 Effect of permanganate concentration on the rate for TML oxidation at different buffer concentrations at pH 2.0 42 6 Effect of buffer concentration on the rate for the permanganate-independent path at pH 2.0 43 7 pH-Rate profiles for TML-oxidation (permanganate-independent path) and TML-exchange 45 8 pH-Rate profile for the permanganate-dependent path of TML oxidation 46 9 Log(rate)-pH plot for the permanganate-dependent path of TML oxidation 47 10 Typical kinetic plots for the permanganate oxidation of TML-7-d3 50 11 Typical pseudo first-order rate plot for the permanganate oxidation of 6,7-diphenyl-8-methyllumazine 81 12 Relation between rate constant and permanganate concen-tration in the oxidation of 6,7-diphenyl-8-methyl-lumazine 84 13 Relation between acidity and rate for the permanganate-independent path in the oxidation of 6,7-diphenyl-8-methyllumazine 87 14 Relation between acidity and rate for the permanganate-dependent path in the oxidation of 6,7-diphenyl-8-methyllumazine 88 - x -Figure_ Page 15 Relation between the oxidation rate and the degree of ionization of 6,7-diphenyl-8-methyllumazine for the permanganate-independent path 89 16 Relation between the oxidation rate and the degree of ionization of 6,7-diphenyl-8-methyllumazine for the permanganate-dependent path 90 17 Oxidation of TML with potassium trans-1,2-diaminocyclo-hexanetetraacetatomanganate (III) (K*foIIT-CyDTA). Plots of TML absorbance at 404 nm versus time 105 18 UV-Visible spectra of 6,7,8-trimethyllumazine, 6,7-diphenyl-8-methyllumazine and permanganate ion 117 19 NMR spectrum of the product of TML oxidation 125 20 IR spectra of the product of TML oxidation and of 2,4-dioxo-6-methylaminopyrimidine 126 21 UV spectra of the product of TML oxidation and of 5,6-dihydro-6-hydroxy-l,3-dimethyl-2,4-dioxopyrimidine 127 - x i -ACKNOWLEDGEMENT I wish to express my sincere appreciation to Dr. Ross Stewart for h i s many h e l p f u l suggestions and i n s p i r a t i o n s during the course of t h i s work. I would also l i k e to thank the University of B r i t i s h Columbia and the National Research Council f or f i n a n c i a l assistance. - 1 -1. INTRODUCTION 1.1 The Pterldines and Lumazlnes Pteridine chemistry can be said to have begun with the studies of the pigments of butterfly wings carried out by Hopkins at the end of the last century.Almost thirty years later, Wieland and his co-workers began systematic, studies of the chemical structures of these pigments and soon succeeded in isolating the yellow pigments from the 2 wings of brimstone butterflys, and the white pigments from the cabbage 3 butterfly wings. However, they met with unexpected d i f f i c u l t i e s in determining the structures of these substances, and i t was 15 years 4 before Purrmann was f i n a l l y able to show in 1940 that these compounds were aminohydroxy derivatives of a bicyclic nitrogenous ring system, viz. of pyrimido[4,5-b]pyrazine, which Wieland"' designated in 1941 as pteridine (I). 5 4 N (I) - N ^ Y N H 2 NH 0 (ID - 2 -It was recognized that one of the butterfly pigments, xanthopterin(II), is of wide natural distribution and bears a close relationship to 6 7 haematopoiesis. It .was also demonstrated ' that folic acid, a member of the vitamin B complex, contains the pteridine nucleus. These observations served as powderful stimulants to investigators in subsequent years. Prior to 1956 a l l known naturally occurring pteridines were derivatives of pterin(2-amino-4-hydroxypteridine). In that year Masuda reported the isolation of two deoxyribose derivatives of lumazine (2,4-dihydroxypteridine) along with riboflavin(III),from the mycelium of »8,9 "Eremothecium ashbyii substances from the mycelium of "Ashbyagossypii" was reported Subsequently, the isolation of the same 10,11 H3C H 3 C \ n / N \ ^ N \ ^0 H 3C N NH (III) (IV) R 0 ' " 0 H„C Y NH R = CH2(CHOH)3CH2OH (V) Originally, the two ribityl derivatives of lumazine were designated by Masuda as the "G-substance", because of its green fluorescence, and the "V-substance", because of its violet fluorescence. When the - 3 -structures of the two compounds became known, the names 6,7-dimethyl-8-ribityllumazine(IV) for the G-substance and 6-methyl-7-oxo-8-12 ribityllumazine(V) for the V-substance were suggested. These compounds are usually classified as derivatives of lumazine rather than of pterin. The many studies of lumazines that have recently been carried out have opened a new chapter in the chemistry and biochemistry of the pteridines. The chemistry and biochemistry of the pteridines have been treated 7 13—19 in considerable detail i n several reviews. ' A brief summary of some of their properties that are pertinent to the present investigation w i l l be presented. 1.1.1 Physical Properties The naturally occurring pteridines, which, in general, carry one or more substituents in the form of hydroxyl or amino groups, are notoriously insoluble in almost a l l solvents and possess very high melting points (usually above 350°). The presence of an hydroxyl group normally confers increase of solubility in water on an organic compound by virtue of i t s a b i l i t y to form hydrogen bonds with the solvent. However, with the hydroxy pteridines the hydrogen-bonding apparently is to a large extent intermolecular between the oxygen atoms and the negatively charged ring nitrogens. These rather inconvenient properties, very low solubility in most solvents and high melting points, coupled with the fact that the amounts of naturally occurring pteridines commonly isolable are usually minute, have made the study of their chemistry an experimentally d i f f i c u l t f i e l d . - 4 -1.1.2 Tautomerism The hydroxy derivatives of p t e r i d i n e s are capable of e x h i b i t i n g amido-iminol tautomerism, l i k e hydroxypyridines. I t i s w e l l known that 2- and 4-hydroxypyridines e x i s t mainly i n the amido-form. This i s also e s s e n t i a l l y true f o r hydroxypteridines, since the hydroxy group i n pte r i d i n e s are located at the a - p o s i t i o n to a r i n g nitrogen, except i n the case of 4-hydroxypteridine(VI). The UV spectrum of 4-hydroxypteridine i s intermediate between that of 4-methoxypteridine and that of 3-methyl-4-pteridone. This has been interpreted as i n d i c a t i n g that 4-hydroxy-p t e r i d i n e e x i s t s as an equilibrium mixture of VI and VII or as a hydrogen-20 bonded intermediate (VIII). (IX) (X) (XI) The lumazines might also be expected to e x i s t as an equilibrium mixture of IX and X, or as XI. However, i n t h i s thesis the p r a c t i c e i s adopted of expressing these structures i n the amido forms represented by X. - 5 -1.1.3 Electron Distribution Electron distribution about the pteridine nucleus has been the subject 21-25 of several theoretical studies. Although the results differ somewhat, the distribution shown in XII can be considered representative. 0.877 0.907 1.111 1.204 881 N 1.142 0.822 (XII) 1.170 Like other heteroaromatic nitrogen compounds, the pteridine system is electron-deficient because of localization of the ten ir-electrons available for aromatic stabilization on the electron-attracting hetero atoms. Thus, the pteridines do not undergo electrophilic substitution reactions but readily participate in nucleophilic substitution and displacement reactions. 1.1.4 Covalent Hydration Reversible covalent hydration across C=N double bonds occurs in a number of nitrogen-containing heterocycles. Covalent hydration can usually be recognized in these systems by the presence of anomalous ionization constants and ultraviolet spectra. The principal effects on ionization constants caused by covalent hydration are an increase of base strength arid a decrease in acid strength, as expected for the elimination of a double bond. A large hypsochromic shift occurs in the UV spectrum and this i s also accounted for by the elimination of the double bond. - 6 -For example, quinazoline (XIII) shows a s u r p r i s i n g hypsochromic s h i f t when made a c i d i c . Also, i t s pKa i s abnormally high compared to i t s 26—28 homologues. This can be explained as follows ; quinazoline has a stable anhydrous neutral species, but the cation i s stable only as the hydrated form. Thus, as soon as the unhydrated cation i s formed i n a c i d i c media, i t i s r a p i d l y hydrated to give XIV. H+/H20 J * (XIII) (XIV) By the a p p l i c a t i o n of rapid-reaction techniques, quantitative data 29 30 have been obtained ' for the k i n e t i c s of the covalent hydration of a number of pt e r i d i n e d e r i v a t i v e s . The reaction i s catalyzed by eith e r a c i d or base and i s r e v e r s i b l e . 8 9 Ever since the time Masuda ' i s o l a t e d 6,7-dimethyl-8-ribityllumazine (IV, page 2 ) from "Eremothecium a s h b y i i " , anomalies i n pKa values and u l t r a v i o l e t spectra were noticed i n 8-substituted lumazines by many 31 33 34" i n v e s t i g a t o r s . P f l e i d e r e r , P e r r i n , and t h e i r co-workers c a r r i e d out precise studies on the hydration properties of 28 examples of the 8-alkyllumazines. They showed that, when an a l k y l group i s present i n the 8-position of lumazines, transannular hydration occurs across the 1,7 p o s i t i o n s . An example of such transannular hydration i s shown i n Scheme 1, using 3,8-dimethyllumazine (XV) as an example. I t was shown using rapid-reaction techniques that, at equilibrium, there are 5360 - 7 -molecules of the anhydrous neutral species (XV) to each one of the hydrated molecules (XVI). r r OH CH ° H ,N ' / N ^ .0" \ CH„ 0 (XVII) CH_ OH 1 3 H N ^ N V ° (XV) I (XVI) = 5360 CH, (XVI) Scheme 1. The relationship between the anhydrous neutral species (XV), the hydrated neutral species (XVI), and the hydrated anion (XVII) of 3,8-dimethyllumazine 34 The presence of a methyl group.at the site of hydration usually hinders the hydration reaction as a result of a combination of steric 35 36 and electronic effects. ' Thus, the proportion of hydrate in the neutral species of 6,7,8-trimethyllumazine (XVIII) is decreased to 1 in 15100. On the other hand in 6,7-diphenyl-8-methyllumazine (XXI, R = H) 34 the ratio is increased to 1 in 2040. The increase in the proportion - 8 -H3C. H3C-CH. I 3 OH C,H3 H CH3 H N\^ N\J^° h2° H V N Y N Y ° o h n Y^ NY° J NH ^ 3 L 1 NH i H 3 C " X . X .NH (XVIII) (XIX) (XX) ? H3 -N OH fH3 H N\^0 \ ^ N \ ^ 0 H2Q ^ V / N \ / ^ (XXI) 0 (XXII) i x v i i i i = (XX) -L^-LUU, (XXI) (XXII) - 2040 of the hydrated species in the latter compound can be explained as follows: the 6- and 7-phenyl groups must interfere with each other's rotations, so that the normally flat pyrazine ring becomes distorted. This steric strain is relieved by hydration. Two isopropyl groups in 34 the 6 and 7 positions give rise to a similar, but bigger, effect. Kinetic studies show that the neutral species of 3,8-dimethyl-6,7-diphenyllumazine (XXI, R = CH3) becomes hydrated about 200 times more slowly than the neutral species of 3,8-dimethyllumazine (XV), although the former contains a higher proportion of hydrate when the equilibrium 34 is finally attained. - 9 -It must be noted here that the hydrated species of some of the lumazines undergo a rapid and reversible ring-opening of the pyrazine 34 37 ring. ' 6,7,8-Trimethyllumazine (XVIII) is known to be one such compound, existing as an equilibrium mixture of anhydrous (XVIII), hydrated (XIX), and ring-opened species (XX). Therefore, the ratio of the unhydrated and hydrated forms given earlier for XVIII actually refers to the equilibrium constant between the neutral form (XVIII) and the ring-opened form (XX). The kinetics of the ring-opening and ring-closing reactions of XVIII has been studied and i t was found that the ring-opening 34 reaction is subject to both general acid and general base catalysis. The most reliable method for locating the site of hydration is to examine the homologue which has a methyl group attached to that carbon atom which is thought to add the hydroxyl g r o u p . I f the location is correct, the anomalies in pKa values and UV spectra will disappear or diminish. NMR spectroscopy and comparison of the UV spectra with those of known dihydro derivatives are further useful methods. Since heterocyclic compounds that have water bound covalently across a C=N double bond behave as secondary alcohols, they can be converted to the corresponding carbonyl compounds by mild oxidation. Comparison of the oxidation products with substances of known constitution makes another useful method for locating the site of hydration. This method is unsatis-factory, however, when hydration takes place at more than two sites in the molecule. Hydrogen peroxide in 1 N sulfuric acid,3** ^ L and potassium 42 43 44 permanganate and potassium ferricyanide ' in 0.1 N sodium hydroxide have been found suitable for this purpose, if used at room temperature. - 1 0 -In 1966, Jacobsen applied the oxidation method for locating the site of hydration to a system with a methyl group at the suspected site of hydration. He examined the UV spectra and pKa values of 6,7,8-trimethyl-2-methyliminopteridine (XXIII) and a series of related compounds and attributed the observed anomalies to the presence of covalent hydration. When these compounds were oxidized by potassium permanganate under conditions where hydration was suspected to occur, the corresponding 7-oxo compounds were obtained i n good yields. The oxidation occurred very rapidly at room temperature. The mechanism of this reaction was suggested by Jacobsen to be the following; the reaction involves an i n i t i a l hydration across the 1,7 positions and the resultant hydrate (XXIV) undergoes an oxidative removal of the methyl group (considered as oxidation to carboxylic acid plus decarboxylation). He sought support 46 for this in the reported oxidation of 1-methylcyclohexanol (XXV) to cyclohexanone by potassium permanganate at room temperature. H3Cv H, CH I 3 NCH 3 H„C -> 3 OH CH-\ 1 3 NHCH, > N A ^ N " H2° H 3 C ^ N (XXIII) (XXIV) HO CH, KMnO, room temp. + fragmented products (XXV) - 11 -However, oxidation of t e r t i a r y alcohols usually requires vigorous conditions or prolonged reaction times. 47 Furthermore, examination of the o r i g i n a l paper on the oxidation of 1-methylcyclohexanol reveals that the reaction a c t u a l l y took two weeks to complete and gave cyclohexanone and other fragmented products (formic a c i d , a c e t i c a c i d , glutamic acid and a d i p i c a c i d ) . These conditions are i n remarkable contrast to those of the p t e r i d i n e oxidations. Consequently, i t remains uncertain whether t h i s oxidation method i s generally v a l i d f o r l o c a t i n g the s i t e of hydration, regardless of the presence of the methyl group at the suspected p o s i t i o n . Since p t e r i d i n e s are e l e c t r o n - d e f i c i e n t (see Section 1.1.3), they undergo n u c l e o p h i l i c addition e a s i l y . Covalent hydration, which i s mentioned above, can be regarded as a form of n u c l e o p h i l i c addition. Many examples of n u c l e o p h i l i c additions to p t e r i d i n e derivatives have been reported. For instance, i n cold a l k a l i n e s o l u t i o n 6-hydroxypteridine (XXVI) undergoes Michael-type addition across the 7,8 double bond with 48 such reagents as acetone, d i e t h y l malonate and eth y l cyanoacetate. R H (XXVI) R = CH2COCH3, CHCNC0 2C 2H 5, CH(C0 2C 2H 5) - 12 -8-Substituted lumazines (XXVII) also participate in nucleophilic addition at the 7 position with water (covalent hydration), alcohols, 37 ammonia, hydroxylamine and hydrogen cyanide. In spite of steric hindrance due to the methyl group at the 7 position, 6,7-dimethyl-8-ribityllumazine (IV) also gives addition products with water, hydrogen 49 cyanide and sodium bisulfite. R I *N NH 0 (XXVII) XH x R X = OH, OR, NH NHOH, CN 2, R I H 3 C ^ NH YH H3C> Y R I N^-4 r NH Y = OH, CN, HSO, (IV) Covalent hydration in heterocyclic systems has been treated in 30,35,50 several reviews. 1.1.5 Deuterium Exchange Reaction In 1969, Paterson and Wood"'"'' reported that the methyl protons at the 7-position of 6,7-dimethyl-8-ribityllumazine (IV, page 2 ) undergo 52 53 deuterium exchange. Almost at the same time two other groups ' reported independently the occurrence of the same reaction, and a l l three groups suggested essentially the same mechanism. - 13 -Precise kinetic studies of the exchange reaction of 6,7,8-trimethyl-52 5A lumazine (XVIII) were carried out by Stewart and McAndless ' by means of NMR spectroscopy. It was found that the reaction is subject to both general acid and general base catalysis. The rate of the spontaneous reaction is extremely low and there is no evidence of a product term, [HA][A-], for the catalysis. From these results, they proposed the reaction mechanism illustrated below (Scheme 2) for the general acid and general base catalyzed routes. fH3 h 3 W N Y ° NH H 3C 0 (XVIII) HA fast H C f 3 H 3 ' V N ^ N Y o (XXVIII) + A A ,slow + HA (XXIX) X D 2 ° fast >* CH, A ,slow CH, H C 1 3 H 2 V ^ N N ^ N " Y ° H 3(T N T NH + HA Scheme 2. Mechanism of the deuterium exchange reaction in 6,7,8-trimethyl-54 lumazine (XVIII) (where HA denotes the buffer acid). - 14 -The acid- and base-catalyzed routes differ only by the presence of an extra proton i n the former. It was found that the ratio of effectiveness of an anion A" reacting with a protonated substrate (XXVIII) 2 and neutral substrate (XVIII) i s 10 . This value is unusually low among prototropic reactions, such as enolization of acetone or mutarotation of glucose. This was tentatively attributed to the fact that 6,7,8-trimethyllumazine (XVIII) has a large tendency to gain and to lose protons, which i s reflected by higher P K B H+ a n c* lower pKa values than for the other compounds. Deuterium exchange has also been observed"^ for the 8-methyl group in riboflavin-5'-phosphate (XXXI, flavin mononucleotide), which i s a close structural analogue of 6,7,8-trimethyllumazine. CH„CH0HCH0HCH0HCH„0-P-0 2 2 || (XXXI) 1.1.6 Metabolism of Lumazines. Biosynthesis of Riboflavin and  6-Methyl-7-oxo-8-ribityllumazine from 6,7-Dimethyl-8- ribityllumazine. The close similarity between 6,7-dimethyl-8-ribityllumazine (IV, page 2 ) and riboflavin (III, page 2 ) raises the question as to whether IV serves as a precursor to III. Such a postulate was 56 originally suggested by Masuda when he succeeded in synthesizing - 15 -riboflavin (III) by the reaction of IV with acetoin or biacetyl. This view has been confirmed by experiments with radioactive t r a c e r s " ^ ' a s well as by the fact that IV can be converted to riboflavin (III) by the ^ g ^ 2 action of extracts of "Eremothecium ashbyii" and other microorganisms. In the past decade the mechanism of this transformation has been the subject of intensive studies. The formation of riboflavin from IV can be achieved either enzymically or chemically. In both cases, two molecules of IV form one molecule of riboflavin and one molecule of 5-amino-6-ribitylaminouracil ( x x x i i ) . 6 3 " 6 5 H 3 C 6 R H 3 C J y NH 0 (IV) "donor" H„C H 3C s R + H 3C /^'N NH 0 "acceptor" R I NH R HN H 2N H NH 0 0 (XXXII) ,63,66 that a unit, (III) The experiments with radioactive tracer revealed derived from carbon atoms 6 and 7 and the attached methyl groups from one molecule of the lumazine (the donor), i s transferred to the second molecule of the lumazine (the acceptor) to form the ortho-xylene ring of 67-72 riboflavin. Various mechanisms have been suggested for this trans-formation, but no definitive path has yet been established. The recent 72 mechanism proposed by Paterson and Wood w i l l be shown below. - 16 -In their experiments, the ribityllumazine (IV) was refluxed in phosphate buffer (pH = 7.3) in D^O for 18 hrs under nitrogen and in the dark. It was found that the deuterium entered at the 8 position and in the methyl group at the 6 position (see Scheme 3). This result was explained by making use of known reactivities of the lumazine precursor, such as covalent hydration and the deuterium exchange reactions (Scheme 3). 0 (III) (XXXII) Scheme 3. The mechanism of the formation of riboflavin from 6,7-72 dimethy1-8-ribityllumazine "in vitro". - 17 -The i n i t i a l step i n Scheme 3 involves attack on the hydrated lumazine by the carbanion produced by removal of a proton from the activated methyl group at the 7 p o s i t i o n of a second molecule of lumazine. Protonation and ring-opening of the resultant adduct leads to the simple carbinolamine, which i s i n equilibrium with a ketone and a diaminouracil fragment. C y c l i z a t i o n of the ketone gives the f l a v i n with the correct d i s t r i b u t i o n of deuterium l a b e l . 73 A s l i g h t l y d i f f e r e n t mechanism was suggested by Beach and Plaut for the enzymic transformation of 6,7-dimethyl-8-ribityllumazine (IV) into r i b o f l a v i n ( I I I ) . Because of the close s i m i l a r i t y i n the structures of III and IV i t was suspected that the l a t t e r might be a precursor of the former. However, i t does not seem to have been noticed that IV might also be a d i r e c t precursor of 6-methyl-7-oxo-8-ribityllumazine (V), even though a l l three compounds, I I I , IV and V, were i s o l a t e d from the fermentation product of "Eremothecium a s h b y i i " , and a l l possess s i m i l a r structures. 74 It was f i r s t thought that V was produced by the reaction between pyruvic a c i d (XXXIII) and 5-amino-6-ribitylaminouracil (XXXII), a probable intermediate i n the metabolism of IV. HfJ H 3 C ^ O (XXXIII) R H HN ^-N + H2N' y NH 0 (XXXII) R H 3C N H C L / N \ / N \ ^0 Y NH (V) - 18 -58 75 Later, Masuda, Kuwada and t h e i r co-workers ' found that the action on IV of crude enzyme extracted from "Eremothecium a s h b y i i " produced V, along with I I I . Thus, they were forced to consider another mechanism for the biochemical transformation of IV into V. I t was soon proved^ 6 that V i s not an intermediate i n the biosynthesis of III from IV, and therefore the transformations of IV into V and into I I I had to be considered separately. Korte and h i s co-workers suggested that the biosynthesis of V might be effected by the oxidation of the 7-methyl group of IV, on the basis of t h e i r discovery that the prolonged a i r - o x i d a t i o n of an a l k a l i n e s o l u t i o n of IV produced V. R R I H [0] H 3C N NH i n base H 3 C ^ N NH (IV) (V) - 19 -78 On the other hand, Mitsuda and his co-workers assumed that enzymic degradation of IV gives 5-amino-6-ribitylaminouracil (XXXII, page 17 ) and a compound with four carbon units, and the dimer of the latter (an aldol compound) reacts with the former to afford the intermediate (XXXIV, below )> which is then hydrolyzed to V. 49 Meanwhile, Rowan, Wood and Hemmerich obtained V by merely refluxing IV with pyruvate ester (alkyl group of the ester not specified). From this and the fact that IV undergoes hydration at the 7 position (see section 1.1.4), they suggested that displacement of the biacetyl residue of IV by a pyruvate residue is likely to occur via the hydrated species (XXXV) in the biochemical transformation of IV into V. (XXXIV) (XXXV) Kuwada and his co-workers found that IV, when subject to enzymic action at pH > 4, gave V after 30-60 minutes, the yield improving with a rise of pH. They failed to obtain V in the absence of enzyme. When the same reaction was carried out at pH 7, formaldehyde and formic acid were detected along with IV and III. From this result, they presented the mechanism shown below, in which the enol form of IV was considered as an intermediate. - 20 -Non-enzymic transformation of 6,7-dimethyl-8-ribityllumazine (IV) into 6-methyl-7-oxo-8-ribityllumazine (V) has also been studied by 80 several workers. Mitsuda and h i s co-workers found that the reaction of IV with _p_-benzoquinone i n phosphate buffer at pH 7.0 produced V i n almost quantitative y i e l d . I t was found that the reaction was pH-dependent and the reaction was f a s t e r at higher pH. The reaction rate was not a f f e c t e d by absence of oxygen and the reaction did not occur at a l l i n ethanol. From these r e s u l t s , they suggested that a hydrated monoanion of IV reacts with p_-benzoquinone. Oxidation of 6,7,8-trimethyllumazine (XVIII, page 8 ) and analogues by potassium permanganate and xanthine under various conditions 45 was studied by Jacobsen and i t was found that the products were the demethylated 7-oxo compounds. I t was suggested that the reaction goes v i a an hydrated species. (This has already been discussed i n d e t a i l i n - 21 -section 1.1.4). 81 McAndless also studied the oxidation of 6,7,8-trimethyllumazine and i t s dihydro derivative by potassium ferricyanide i n basic media. In both cases the ultimate product was the demethylated 7-oxo lumazine (V, R = CH^). Although the study was inconclusive he suggested the hydrated species of trimethyllumazine reacts with ferricyanide via electron transfer. It has been known that pteridines can be formed biogenetically from another biologically important family of heterocycles, the purines. Thus, the pteridines can be considered as the link between the purines and the isoalloxazine system (flavins). The biosynthesis of pteridines from purines w i l l not be dealt with here, but has been discussed elsewhere.19,76,82 84 ^ g biogenetic relationship between purines, pyrimidines, lumazines and flavins i s illustrated in Scheme 4. An important property of pteridines from a biochemical standpoint i s t h e i r a b i l i t y to form dihydro and tetrahydro d e r i v a t i v e s . These reduced forms are very re a c t i v e and not only can they serve as s p e c i f i c reductants, they can also p a r t i c i p a t e i n electron transport reactions i n l i v i n g systems. Further d e t a i l s on t h i s subject are a v a i l a b l e . . 19,85,86 elsewhere. 1.2 Oxidation of Organic Substrates with Permanganate and with Potassium trans-l,2-Diaminocyclohexanetetraacetatomanganate (III) In the present i n v e s t i g a t i o n , the oxidation of lumazine derivatives i s studied using two o x i d i z i n g reagents, potassium permanganate and the manganese (III) complex with trans-l,2-diaminocyclohexanetetraacetic acid (KMnI]:iCyDTA, where CyDTA represents the ligand). Therefore i t may be useful to discuss some oxidation modes of these two oxidants which are pertinent to the present i n v e s t i g a t i o n . 1.2.1 Oxidation of Alkenes with Permanganate Under most conditions permanganate oxidizes alkenes to y i e l d glycols. However, formation of a-hydroxy ketones and cleavage of the chain to produce aldehydes, ketones, or carboxylic acids frequently occurs, the 87 composition of the products being dependent on the conditions used. MnO, RCH=CR„ RCH—CR„ 0 0 \ / Mn // \ -0 0 OH OH I I -> RCH-CR„ RCHO (or RC02H) + R2C=0 (XXXVII) 0 OH - 23 -88 Wagner first suggested that the cyclic manganese ester (XXXVII) might be formed in the reaction. This idea was confirmed by stereo-89 18 90 chemical studies and the use of KMn 0^ , which revealed that the mode of addition of the hydroxyl groups is cis and that the oxygen atoms in 91 the product diol come chiefly from the permanganate. Kinetic studies showed that the reaction is first order in both alkene and permanganate, but zero order in hydroxide ion. The reaction is accelerated by alkyl, phenyl, or carbonyl substituents but is otherwise insensitive to electronic effects. These features resemble those of 1,3-dipolar addition 92 91 reactions and Wiberg and Geer suspect that the ring bonds in the cyclic ester might be formed synchronously, as in 1,3-dipolar additions. 93 Very recently, two groups, Lee and Brownridge and Wiberg, Deutsch and ^ 94 Rocek, have provided spectrophotometric evidence for the cyclic manganese ester intermediate in the oxidation of cinnamic acid. 1.2.2 Oxidation of Ketones with Permanganate Ketones can be rapidly oxidized by permanganate only in alkali or 95 concentrated acid. The products usually are degraded fragments and there is l i t t l e doubt that the process of enolization is involved in the reaction. In acidic solution the double bonds of the enol form are oxidized in the same manner as the oxidation of alkenes mentioned above. In basic solution the reactive intermediate may be either the enol or the enolate-anion. Wiberg and Geer^ showed that the rate law for the oxidation of acetone in basic solution has the form r = kfacetone][OH ] [MnO^  ]. This rate law is best accommodated by a mechanism in which the enolate-anion is attacked by peramanganate before i t goes to enol (next scheme). - 24 -- k l acetone + OH ) enolate + HO (fast) k2 - k3 enolate + MnO^ > product (slow) The f a c t that the rate constant for the second step (k^) i s of the order of 10^ M ''"sec 1 , ^  whereas the rates of permanganate oxidation 2 - i - i 91 for a v a r i e t y of alkenes are of the order of 10 M sec , suggests the enolate-permanganate reaction goes v i a an electron transfer rather 96 than add i t i o n to the double bond. From product analysis, Wiberg and 96 Geer suggested that the r a d i c a l formed i n the second step i n the above scheme reacts r a p i d l y with permanganate to form the manganate ester (XXXVIII), which then undergoes further reactions. 0 0 0 ll _ ii r» - II i v CH0CCH • + MnO, > CHoCCH-0-Mn0o > CH_CCHO + Mn 5 1 4 3 j _/ J J H -J HO (XXXVIII) 1.2.3 Oxidation of Organic Substrates with Potassium trans-1,2-Diaminocyclohexanetetraacetatomanganate(III) (KMn^^^CyDTA) Manganese (III) i s an e x c l u s i v e l y one-equivalent oxidant of considerable strength but the value of the uncomplexed ion i s impaired 97 98 by i t s i n s t a b i l i t y i n any but the most a c i d i c systems. ' In 1967 99 III Hamm and Suwyn prepared KMn CyDTA. This c r y s t a l l i n e compound i s stable i f stored at room temperature i n the dark. KMn"''''"'''CyDTA i s also 99 stable i n aqueous s o l u t i o n over a wide pH range (up to pH 11) and has been used to oxidize oxalate**^ and hindered phenols."''^"'' I t was shown that KMn^^^CyDTA i s normally reduced by electron transfer from the anions of the s u b s t r a t e s . H o w e v e r , under a c i d i c conditions where the substrate anion's concentration i s n e g l i g i b l e i t may abstract a hydrogen atom d i r e c t l y from the ne u t r a l molecule.'^"'" Very recently, the c r y s t a l structure of KMn"'''''''"CyDTA was determined 102 by R e t t i g and Trotter. Their structure indicates a s l i g h t l y d i s t o r t e d octahedron of Mn*''"'''. - 26 -2. SCOPE OF THE INVESTIGATION Riboflavin (III,R=ribityl) and other flavin nucleotides occupy a central position i n the reaction chain that i s responsible for oxidation in l i v i n g systems. Because of the close similarity in the structure of riboflavin and 6,7,8-trimethyllumazine (TML, IV, R,R' = CH3>, the latter may serve as a model compound for these oxidative nucleotides. With this in mind, studies of 6,7,8-trimethyllumazine have been carried 81 out in this laboratory for several years. It was found that the dihydro derivative (XXXIX, X = H) of TML was oxidized by potassium ferricyanide in alkaline solution, f i r s t to TML and ultimately to 6,8-dimethyl-7-oxolumazine (V, R = CH^). Such a facile oxidative demethylation i s unique and therefore the study of i t s mechanism is attractive from the point of view of mechanism. o - 27 -(XXXIX) The r e a c t i o n i s also important for other reasons. F i r s t l y , 45 Jacobsen considered that the oxidation method commonly used as a t o o l to locate the s i t e of hydration i n p t e r i d i n e s 3 5 ' " ^ i s generally v a l i d , even i f a methyl group i s present at the suspected s i t e of hydration, making the derived alcohol t e r t i a r y i n nature, for example XXXIX (X = OH). The v a l i d i t y of t h i s hypothesis needs to be examined. Secondly, 6,7-dimethyl-8-ribityllumazine (IV, R = r i b i t y l , R' = CHg) i s known to be a precursor of 6-methyl-7-oxo-8-ribityllumazine (V, R = r i b i t y l ) i n b i o l o g i c a l systems. Although the mechanism of t h i s transformation has received considerable attention, i t remains u n c l a r i f i e d . I t was hoped that the i n v e s t i g a t i o n of the mechanism of permanganate oxidation of 6,7,8-trimethyllumazine could shed some l i g h t on the matters r e f e r r e d to above. To a i d these studies the oxidations of 6,7-diphenyl-8-methyllumazine (IV, R = CH^, R' = <f>) by permanganate and of 6,7,8-trimethyllumazine by potassium trans-1,2-diaminocyclo-hexanetetraacetatomanganate (III) (KMn^^CyDTA) were also investigated. A d d i t i o n a l model compounds for f l a v i n nucleotides are 8-azaflavin (XL) and i t s quarternary s a l t s (XLI). The synthesis of the l a t t e r compounds were attempted and new information about the p o s i t i o n of - 28 -a l k y l a t i o n of 2,3-diaminopyridines (XLII) was obtained. This work i s described i n an appendix. N ^ NH 0 R (XL) (XLI) (XLII) - 29 -3. RESULTS AND DISCUSSION 3.1 Permanganate Oxidation of 6,7,8-Trimethyllumazine 3.1.1 Results General Features of K i n e t i c Results The reaction of 6,7,8-trimethyllumazine (TML) with permanganate i n unbuffered aqueous s o l u t i o n i s extremely slow, except under strongly a c i d i c or strongly basic conditions. The rate of the oxidation i s dependent on pH and, at a given pH, i s more rapid i n buffered than i n unbuffered solutions. Outside the region pH 1 to pH 6 the reaction i n phosphate buffer (Na2HP0^ = 0.1 M) i s too rapid to be followed by the usual spectrophotometric methods. Within the pH range 1 to 6, excellent l i n e a r c o r r e l a t i o n s between logarithm of the TML absorbance at 416 nm and time were always obtained up to two h a l f - l i v e s , under pseudo f i r s t -order conditions (large excess of permanganate) (Figure 1), i n d i c a t i n g the r e a c t i o n to be f i r s t order with respect to TML. The rate law f o r pseudo f i r s t - o r d e r k i n e t i c s i s given by equation 1, Where k^' i s the pseudo f i r s t - o r d e r rate constant. Integration gives equation 2, d[TML] dt = k^[TML] (1) - 31 --ln[TML] = ]n 't + C (C = constant) (2) Hence, k^' can be obtained from the slope of plots of logarithm of the TML absorbance at 416 nm against time. Effect of Permanganate Concentration on the Rate The kinetic order with respect to permanganate is somewhat more complicated than that of TML. The effect of permanganate concentration on the rate was examined between pH 1 and 6 with the buffer concentration and the ionic strength kept at constant values. The variation of the pseudo first-order rate constant k^' with permanganate concentration at different pH is given in Table 1. The results at pH 2.0, 2.4, 5.0 and 6.0 are exhibited graphically in Figures 2 and 3. At lower pH a plot of k^' versus permanganate concentration gave a straight line (Figure 2). Extrapolation of the straight line to zero permanganate concentration intercepts the ordinate somewhat above the origin of the graph. Hence, the rate law can be given by equation 4. k . = L _ d[TML] = o -fcl [TML] dt k l + k 2 L M n U 4 J U ; v = - ^ML] = k;L0[TML] + k2[TML][Mn04 ] (4) Equation 4 suggests that there are two routes involved in the oxidation of TML. One is first order in TML but zero order in permangante, as shown by the first term of the equation or the intercept of the plot of k^' versus permanganate concentration. This may be called the - 32 -Table 1. Rate Constants for TML Oxidation at Various Permanganate 3. Concentrations and Various pH. PH KMnO. 4 (mole/1.) R l (sec ) 6.01 7.25 X i o - 5 1.31 X i o " 2 b I I 2.42 X lO" 4 1.63 X i o " 2 b II 7.32 X IO" 4 1.82 X i o " 2 I I 1.46 X i o ' 3 2.04 X i o " 2 II 2.19 X i o " 3 2.09 X i o " 2 II 2.93 X 10"3 2.03 X i o " 2 II 3.66 X i o " 3 2.03 X i o " 2 5.01 9.67 X i o " 5 2.08 X i o " 3 b I I 1.45 X i o " 4 2.27 X i o " 3 b II 2.42 X , i o " 4 2.66 X i o " 3 b I I 3.22 X i o " 4 2.82 X i o " 3 b II 7.40 X IO" 4 2.85 X i o " 3 II 1.47 X i o - 3 3.07 X IO" 3 I I 2.20 X i o " 3 3.26 X IO" 3 II 2.92 X i o " 3 3.43 X IO" 3 II 3.64 X 10"3 3.44 X IO" 3 3.98 7.32 X i o " 4 8.57 X IO" 4 II 1.46 X i o " 3 9.84 X IO" 4 I I 2.18 X i o " 3 1.10 X IO" 3 II 2.89 X 10"3 1.22 X IO" 3 I I 3.60 X i o " 3 1.33 X IO" 3 3.49 7.30 X i o " 4 8.22 X IO" 4 ti 1.46 X i o " 3 9.47 X IO" 4 ti 2.17 X i o " 3 1.07 X i o " 3 ii 2.87 X i o " 3 1.22 X i o " 3 ii 3.58 X i o " 3 1.35 X i o " 3 3.00 7.29 X i o " 4 1.03 X IO" 3 II 1.45 X i o " 3 1.21 X i o " 3 ti 2.17 X i o " 3 1.40 X IO" 3 II 2.88 X IO" 3 1.59 X i o " 3 II 3.56 X i o " 3 1.77 X i o " 3 - 33 -Table 1. (Continued) PH KMnO 4 (mole/1.) k* R l (sec *) 2.39 7.33 X i o ' 4 2.07 X i o " 3 II 1.46 X i o " 3 2.70 X i o " 3 ti 2.18 X i o ' 3 3.18 X i o " 3 ii 2.90 X i o " 3 3.64 X i o " 3 ii 3.61 X i o " 3 4.08 X i o " 3 2.00 7.32 X i o " 4 4.13 X i o - 3 II 1.46 X i o " 3 5.29 X i o " 3 II 2.19 X i o " 3 6.49 X i o " 3 it 2.93 X io"'3 7.37 X i o " 3 it 3.66 X i o " 3 8.41 X i o " 3 1.39 7.33 X IO" 4 9.33 X i o " 3 II 1.10 X 10"3 1.13 X i o " 2 I I 1.46 X IO" 3 1.34 X i o " 2 II 1.82 X IO' 3 1.52 X i o " 2 II 2.18 X IO" 3 1.68 X i o " 2 n 2.54 X IO" 3 1.81 X i o " 2 I I 2.90 X IO" 3 2.09 X i o " 2 0.997 7.40 X i o - 4 1.49 X i o " 2 II 1.11 X i o " 3 1.97 X i o " 2 I I 1.47 X i o " 3 2.38 X i o " 2 II 1.84 X i o " 3 2.76 X i o " 2 I I 2.20 X i o " 3 3.07 X i o " 2 n 2.57 X i o " 3 3.45 X i o " 2 II 2.92 X 10"3 3.76 X i o " 2 3 TML = 5.10 x 10 - 5 M, Na„HPO. 1 4 = 0.1 M, p = 0.42, T = 31.4°. Since the concentration of permanganate is so small that the condition for pseudo first-order kinetics is not met, the i n i t i a l rate is recorded. - 36 -"permanganate-independent path". The other route is f i r s t order in both TML and permanganate as shown in the second term of the equation and the slope of the plot. This may be called the "permanganate-dependent path". The slope of the straight line of the plot of k^' versus permanganate concentration becomes smaller as the reaction medium i s made more basic, and f i n a l l y above pH 5.0 the same plot does not give a straight line but rather a curve which eventually levels off at higher permanganate concentration (Figure 3). This kind of levelling-off effect i s not due to the presence of different routes, but can be explained by the two-step reaction scheme described below. TML X + MnO » product The rate law for the above mechanism can be given by - ^ J - = k 2[X][Mn0 4-] (5) If the steady state assumption is applied to the species X, equation 5 can be expressed by equation 6. d[TML] k l k 2 [ T M L ] [ M n ° 4 ] ( & ) dt k_x + k2[Mn04-] ^ ; Equation 6 shows that when » k2[MnO^ ] the reaction i s f i r s t order - 37 -in both TML and permanganate since equation 6 can be approximated by eqaution 7. - [IMLUMnO,-] (7) When k_^ << k2[MnO^ ] equation 6 can be approximated by equation 8 and the reaction is zero order with respect to permanganate. ^ = k^ [TML] (8) When k_^ and k^MnO^ ] are of a comparable magnitude the order in permanganate changes from first to zero as the permanganate concentration increases, and this is actually observed in the present case. The change of order in permanganate from first to zero is equivalent to the shift of the rate-determining step from the second step to the first step as the concentration of permanganate increases. The identity of species X will be disclosed in the Discussion. Effect of Buffer Concentration on the Rate The oxidation is very much more rapid in buffered solution than in unbuffered solution. The effect of the buffer concentration on the rate was examined at pH 6 and 2 with the ionic strength maintained at a constant value. At pH 6, as has already been shown in Figure 3, a plot of the pseudo first-order rate constant versus permanganate concentration levels off at high permanganate concentration. Therefore the rates were compared with the permanganate concentration at a value fixed high - 38 -enough that the rate becomes independent of the permanganate concentra-tion. The data were collected i n Table 2 and k^1 was plotted against buffer concentration (Figure .4). A good linear relationship between the rate and the buffer concentration was observed. Extrapolation of the plotted line to zero buffer concentration passed close to the origin of the graph, indicating the uncatalyzed oxidation to be extremely slow. At pH 2.0 pseudo first-order rate constants k^' were recorded for different permanganate and buffer concentrations(Table 3). When k^' was plotted against permanganate concentration at fixed buffer concentration, parallel straight lines were obtained (Figure 5). The slopes and the intercepts of these straight lines were calculated by the method of least squares, these data being also shown i n Table 3. Intercepts (k^°) were plotted against buffer concentration and a good straight line: was observed (Figure 6). Extrapolation of the line to zero buffer concentra-tion intercepted the ordinate somewhat above the origin, which i s distinctly different from the case at pH 6.0. pH-Rate Profile The rate data in Table 1 were analyzed by the method of least squares with respect to each pH; the slopes (k^) and the intercepts (k^°) of the plots of k^' versus permanganate concentration are list e d in Table 4. In Figures 7 and 8, k^° and were plotted against pH. As can be seen in Figure 7, k^° is lowest near pH 3.5 and becomes more rapid with increasing acidity or basicity of the buffer solution. The pH-rate profile is typical of reactions which are catalyzed both by acids and bases. However, the kinetic measurements were carried out - 3 9 -Table 2. Effect of Buffer Concentration on the Rate Constant for TML Oxidation at pH 6.0. NaoHP0. (mole/1.) 2 4 kj^  (sec "*") 0.125 0.0258 0.100 0.0204 0.075 0.0160 0.050 0.0105 0.025 0.00541 y = 0.42, T = 31.4°, KMnO. = 3.00 x 10 M, TML = 5.10 x 10 M - 41 -Table 3. Rate Constants for TML Oxidation at Different Buffer and Permanganate Concentrations at pH 2.00.^ KMn04 (mole/1.) Na2HP04 (mole/1.) 0.025 0.050 0.075 0.100 0.125 7.32 x 10 - 4 0.00241 0.00295 0.00339 0.00413 0.00451 1.46 x 10 - 3 0.00361 0.00417 0.00468 0.00529 0.00568 2.19 x IO - 3 0.00459 0.00520 0.00563 0.00649 0.00680 2.30 x IO - 3 0.00553 0.00617 0.00675 0.00737 0.00785 3.66 x IO - 3 0.00670 0.00730 0.00782 0.00841 0.00883 slope (k 2 > 1.44 1.46 1.49 1.46 1.48 l.mole~^sec~^) intercept 0.00154 0.00195 0.00238 0.00314 0.00349 o - i (k-j. J sec ) correl. coeff. 0.997 0.999 0.999 0.998 0.999 a -1 sec b y = 0.42, T = 31.4°, TML =5.10 x 10 5 M - 42 -6.4 4.8-Figure 5. Effect of permanganate concentration on the rate for TML oxidation at different buffer concentra-tions at pH 2.0. y = 0.42, T = 31.4°, TML = 5.10 x 10~5 M Concentration O , 0.125 0 » 0.050 TML' u CO CO M 3.2 CO o 1.6-—I 1 1 0.8 2.4 10 3 KMn04 (mole/1.) 4.0 - 44 -Table 4. Rate Constants for the Permanganate-independent Path (k°) and the Permanganate-dependent Path (k^) of the Oxidation of TML. pH k„ (l.mole "*"sec L) k° (sec ^ ) correl. coeff. v2 v ~ ° " ' ~1 6.01 — 0.0204 ' -5.01 - 0.00340* -3.98 0.166 0.000737 0.999 3.49 0.184 0.000680 0.999 3.00 0.262 0.000834 0.999 2.39 0.690 0.00163 0.998 2.00 1.456 0.00314 0.998 1.39 5.128 0.00567 0.997 0.997 10.24 0.00817 0.998 a ]i = 0.42, T = 31.4°, TML = 5.10 x 10~5 M, Na HPO = 0.100 M. These rates are obtained from the flat part of the curves in Figure 3. An improved procedure for obtaining the limiting values is to plot l/k^' versus l/tKMnO^], as shown by Figure 3' on page 44-a, since equation 6 on page 36 can be written in the form of equation 6', the intercept of the graph giving the rate for the permanganate-independent path. k1' k x k ^ [Mn04] ^ ; The rates obtained in this way at pH 5.01 are 0.00331 sec" 1 (using a l l points) and 0.00360 sec - 1 (using the five points at high permanganate concentration). At pH 6.01 the value is 0.0203 sec"1 (using a l l points). These values do not differ greatly from those recorded in the table. - 47 -Figure 9. Log(rate)-pH p l o t f o r the permanganate-dependent path of TML oxidation. (Slope of the s t r a i g h t - 48 -on the acidic side of neutrality where the concentration of hydroxide ion is negligible. Therefore the results suggest that the reaction is subject to catalysis by species other than, or as well as, hydronium ion. Since, as has already been shown in Figures 4 and 6, is dependent on the buffer concentration i t can be concluded that the reaction which is represented by (i.e., the permanganate-independent path) is subject to both general-acid and general-base catalysis by the phosphate buffer species as well as by hydronium ion. In Figure 8 i t can be seen that k^ increases with a rise of the acidity of the buffer. When the logarithm of k^ was plotted against pH a straight line with nearly unit slope (0.88) was observed between pH 1 and 2.4 (Figure 9). This result, coupled with the fact that k^ is not affected by the concentration of the buffer (see Figure 5), clearly indicates that the reaction whose rate, is dependent on permanganate concentration (i.e., the permanganate-dependent path) is catalyzed by only hydronium ion, or in other words, is subject to specific acid catalysis. Kinetic Isotope Effect Permanganate oxidation of TML deuterated at the 7-methyl group (TML-7-d„) was studied at pH 6.0 and 2.0 in aqueous buffered H„0. CH 0 - 49 -In the experiment at pH 6 . 0 the concentration of permanganate was chosen high enough so that the rate was independent of the permanganate concentration. The rates thus obtained were compared with those for the protium compound under the same conditions. At pH 2.0 pseudo f i r s t - o r d e r rate constants (k^') for the deuterated compound were determined at d i f f e r e n t permanganate concentrations. The r e s u l t s at pH 6 and 2 are c o l l e c t e d i n Tables 5 and 6 , r e s p e c t i v e l y . At each pH pl o t s of logarithm of the TML absorbance at 4 I 6 nm versus time gave excellent s t r a i g h t l i n e s up to two h a l f - l i v e s , with c o r r e l a t i o n c o e f f i c i e n t s always greater than 0.999, and there were no signs of acceleration of the reaction, which would have resulted from any appreciable replacement of deuterium by hydrogen. Ty p i c a l log(absorbance) time pl o t s at pH 6 and 2 are reproduced i n Figure 10. The average deuterium isotope e f f e c t at pH 6.0 and at 31.4° was calculated to be 6.89. The rate data at pH 2.0 were analyzed by the method of l e a s t squares and also p l o t t e d i n Figure 5. The p l o t gave a good s t r a i g h t l i n e p a r a l l e l to, but lower than, that for the protium compound. The k i n e t i c isotope e f f e c t on k^, the permanganate-dependent path was obtained from the slope and found to be 1.04; from the intercept the e f f e c t on the permanganate-independent path was found to be 2.47. 3.1.2 Discussion As has been shown i n the preceding section oxidation of TML by permanganate i n the pH region between 1 to 6 consists of two major paths which are d i f f e r e n t i a t e d by the k i n e t i c order with respect to permanganate - 51 -Table 5. Rate Constants f o r the Permanganate Oxidation of TML-7-d,j at pH 6.0.a Na.HPO. (mole/1.) k' ( s e c f 1 ) k /k._ b L H A. ti D 0.100 2.83 x 10 - 3 7.21 0.075 2.44 x I O - 3 6.56 mean 6.89 3 y = 0.42, T = 31.4°, TML = 5.10 x 10 _ 5 M, MnO^ = 3.00 x 10 _ 3 M b Data f o r the protium compound are from Table 2. - 52 -Table 6. Rate Constants for the Permanganate Oxidation of TML-7-d at pH 2.O.3 MnO. (mole/1.) k' (sec ) 7.32 X i o " 4 2.30 X i o " 3 1.46 X i o " 3 3.20 X i o " 3 slope (k 2, l.mole ''"sec intercept (k°, sec _ 1)= ' h - 1.40 2.19 X i o " 3 4.29 X i o " 3 1.27 x. 10~" 2.93 X i o " 3 5.51 X i o " 3 correl. coeff. = 0.997 3.66 X i o " 3 6.31 X i o " 3 k_(TML) b = 1 04 k2(TMU--7-d3) ±- U^ k° (TML) b = 2 47 k2(TML-7-d3) a U = 0.42, T = 31.4°, Na2HP04 = 0.100 M, TML = 5.10 x 10~5 M. b Data for the protium compound are from Table 3. - 53 -These two paths, one permanganate-independent and the other permanganate-dependent, w i l l be discussed separately. 3.1.2.1 Permanganate-independent Path The permanganate-independent path i s subject to both general-acid and general-base c a t a l y s i s with the slowest rate occurring near pH 3.5. These k i n e t i c features are remarkably s i m i l a r to those of the hydrogen-deuterium exchange reaction of the 7-methyl group of TML, which i s also subject to both general-acid and general-base c a t a l y s i s and has i t s 52 minimum rate near pH 3.5. Such s i m i l a r i t i e s suggest that the oxidation v i a the permanganate-independent path and the exchange reaction might go v i a the same intermediates. The intermediates i n the exchange re a c t i o n are the enolate-anion (XLIII) i n the general-base catalyzed route and the enol (XLIV) i n the general-acid catalyzed route, formation of them being rate-determining. (See Scheme 5 shown below and also Scheme 2 on page 13 ). If the oxidation i n f a c t proceeds v i a these intermediates, then the rates of the oxidation and the exchange should be i d e n t i c a l . However, for the comparison of two rates one has to consider the s t a t i s t i c a l e f f e c t s at the 7-methyl group, because the exchange involves removal of a l l three hydrogens whereas the oxidation may require removal of only one hydrogen atom. The exchange reaction for both general-acid and general-base catalyzed routes can be regarded as stepwise f i r s t - o r d e r reactions which may be symbolized as X X A B C D - 54 -| H3 H V N v N v ° 0 NH general base catalyzed route/slow general acid s l o w \ c a t a l y z e d route CH„ H C 1 3  H 2 C V / N ^ N w O H C/^N-. 0 (XLIII) KMnO, f a s t D 2 ° fa s t DH2C, H C I1"3 H H-C^N-0 NH CH. i 3 D„0 f a s t (XLIV) KMnO fa s t N ^ N \ ^ o J L NH J 0 I I CH * I 3 D.C, x N \ ^ N N ^ 0 H3C N TNH Scheme 5. Oxidation and exchange reactions of TML v i a the same intermediates. - 55 -The r e l a t i o n s h i p between the exchange rate obtained by NMR H—D technique ( k ^ g ) and the rate f o r removal of the f i r s t hydrogen (k^) can 103 be mathematically derived. For the study of the exchange by NMR only the concentrations of proton-bearing species, A, B, and C are of i n t e r e s t . The changes of concentrations of these species with time are given by ^ = ~ V A l (9) = k x[A] - k 2[B] (10) ^ p - = k 2[B] - k 3[C] (11) Integration of these equations give the concentrations of each species as a function of time. [A] = [ A ] Q e x p H ^ t ) (12) [B] = [ A ] Q [exp(-k 1t)-exp(-k 2t)] (13) k k [ C ] = [ A ] o k (k -k ) [ " 1 / 2 exp(-k 1t)+exp(-k 2t)- 1/2 exp(-k 3t)] (14) If we neglect the presumably small secondary isotope e f f e c t , the rate of proton removal w i l l depend upon the number of proton i n the species. Hence, k± = 3k 3 (15) - 56 -and, k 2 = 2k 3 (16) S i m i l a r l y , the contributions of species A and B to the observed NMR peak area w i l l be three times and twice, r e s p e c t i v e l y , as much as that of C. Hence, the r e l a t i o n s h i p between the peak area, H, and the concentrations of A, B and C w i l l be ctH = 3[A] + 2[B] + [C] (17) where a i s a p r o p o r t i o n a l i t y constant. A combination of the previous equations gives otH = 3 [ A ] q exp(-k 3t) (18) Taking logarithms, we obtain InH = - k 3 t + C (C = constant) (19) H—D Therefore, the observed rate constant, k I s a c t u a l l y represented by the rate constant f o r the t h i r d step of the above consecutive reaction, H—D Since k, = 3k_ or 3k , , the oxidation rate would be expected to be 1 3 obs three times as rapid as the observed exchange rate, provided the oxidation indeed goes v i a rate-determining formation of the enol or the enolate-anion of TML. Hence, the exchange rates obtained by Stewart 52 81 and McAndless ' under the same conditions as the oxidation were H—D m u l t i p l i e d by a fa c t o r of three, and these values ^ k 0 D S ^ » a s w e i L a s the oxidation rates, are presented i n Table 7. The exchange rates - 57 -Table 7. TML-Oxidation Rates(Permanganate-independent Path) and TML-exchange Rates. pH . o - l b k^ (sec ) , H-D , -l.c kobs ( s e c } „ .H-D , -1, 3 x k , (sec ) obs 6 0.0204 0.00625 0.0188 5 0.00340 0.00108 0.00324 4 0.000737 0.000290 0.000870 3.5 0.000680 0.000235 0.000705 3 0.000834 0.000340 0.00102 2 0.00314 0.000698 0.00210 1 0.00817 0.00213 0.00638 a Na2HP04 = 0.100 M, u = 0.42, T = 31.4° b Taken from Table 4. ° Exchange rates obtained by Stewart and McAndless."'2'^ In their experiments the measured pH was not corrected to give the corresponding pD values. The conversion of pH obtained by a glass-calomel pH assembly to the corresponding pD can be done by the addition of the 104 scale-factor of 0.4 to the measured pH. The change in the solvent from H20 to D20 results in changes in the concentrations of each phosphate species as well as TML. Since the reactions are subject to general catalysis by the phosphate buffer species and by hydronium ion, such change in the concentrations of the buffer species makes unequivocal comparison of H20 and D20 systems impossible. However, the change in pK values of TML and phosphoric acid in going to D20 may be largely compensated by the scale-factor of 0.4, since these changes occur in the same direction with about the same magnitude.104,105 g u ch a v i e w ± s substantiated by the fact that both reactions have minimum rates at the same pH. - 58 -H-D (3k , ) are also plotted against pH on the same graph where the obs oxidation rates are plotted (Figure 7). As can be seen in Table 7 and Figure 7, although the oxidation is somewhat faster than the exchange at lower pH, the general agreement between the two rates is very close. This fact strongly suggests that the reactive intermediates in the oxidation are the enolate-anion in pH > 3.5 (general base catalysis region) and the enol in pH < 3.5 (general acid catalysis region). The oxidation mechanisms in these two regions will be discussed in more detail in subsequent sections. Above pH 3.5 Summary of the principal results obtained for the permanganate-independent path in this pH region are: (1) The reaction can be described by the equation 4MnO CH 3 H 3CHo0 + 4Mn0o + 40H - 59 -(2) The reaction i s f i r s t order in TML and zero order in permanganate at high permanganate concentration. (3) The reaction i s subject to general base catalysis by the phosphate buffer species. (4) The rate where permanganate concentration i s high enough that the reaction is no more dependent on permanganate concentration i s almost identical with that of the exchange reaction multiplied by a factor of three. (5) A primary isotope effect, k(TML)/k(TML-7-d3) = 6.89, is observed at 31.4°. (6) Presence of oxygen does not affect the rate nor the products. These observations and other pieces of evidence are best accommodated by the mechanism shown in Scheme 6i (below). (XLVI) - 60 -Scheme 6. Mechanism of oxidation of TML v i a general base catalyzed route. (The general bases are denoted by B.) - 61 -A l l i n d i c a t i o n s are that the process of i o n i z a t i o n i s involved i n the i n i t i a l step. A large k i n e t i c isotope e f f e c t f o r the 7-methyl group indicates that the i o n i z a t i o n of the 7-methyl group i s rate-determining. Removal of a proton from the 7-methyl group i s attained by attack of general bases (phosphate anions i n phosphate b u f f e r ) . I t now becomes clear that the species merely described by X on page 36 i s the enolate-anion (XLIII i n Scheme 6). As has already been discussed on page 36, the fact that the rate l e v e l s o f f at high permanganate concentration indicates that the rate-determining step changes from the second to the f i r s t step with increasing permanganate concentration. The enolate-anion thus formed i s extensively s t a b i l i z e d by e f f e c t i v e dispersion of i t s negative charge into the rest of the molecule by resonance, as shown below. CH, H2C. I H„C V <—> etc. or H C H 3C CH, I -H 2 C V / N ^ si N w > 0 NH Information about the mechanism of the subsequent steps i s not available from the k i n e t i c r e s u l t s . Probable pathways may be attack of permanganate at (a) the exocyclic double bond of the enol (XLV), which i s i n equilibrium with the enolate-anion, or (b) the enolate-anion i t s e l f , as shown i n Scheme 6. (See below). In the l a t t e r case - 62 -there are further p o s s i b i l i t i e s , e i t h e r addition of permanganate to the exocyclic double bond, or electron transfer from enolate-anion to permanganate. Pathway (a) Addition or electron transfer? Pathway (b) As w i l l be shown l a t e r , the oxidation v i a the general acid catalyzed route occurs at the enol (XLV). But the f i n a l products i n that case are d i f f e r e n t from those obtained here and i t seems l i k e l y that the reaction goes v i a pathway (b). (Because the products of permanganate 8 7 oxidation of a double bond depend upon the rea c t i o n conditions t h i s fact may not provide d e f i n i t i v e evidence to exclude the enol mechanism.) Further support f or pathway (b) may be found by reference to the study of the permanganate oxidation of acetone c a r r i e d out by Wiberg 96 and Geer. They showed on the basis of k i n e t i c r e s u l t s that, i n basic media, the enolate-anion i s the species with which permanganate reacts. They also suggested that the mode of the reaction between the - 63 -enolate-anion and permanganate i s an electron transfer, rather than the addition to the double bond. (Electron transfer i s IO"* times f a s t e r than the a d d i t i o n . ^ See section 1.2.2 f o r a more d e t a i l e d discussion of the mechanism.) The s i m i l a r i t y between the oxidation of acetone and TML suggests that the permanganate reacts with the enolate-anion of TML v i a e l e c t r o n t r a n s f e r , as i n the acetone oxidation. One-electron transfer from the enolate-anion to the permanganate w i l l lead to a r a d i c a l intermediate (XLVI i n Scheme 6), whose odd electron i s e f f e c t i v e l y dispersed into the molecule by resonance, i n a s i m i l a r way to that of the enolate-anion (see below). Direct evidence f o r such a r a d i c a l intermediate could not be obtained. This i s not s u r p r i s i n g since permanganate reacts r a p i d l y with most free r a d i c a l s . * H 2 C s CH_ I 3 I NH* > ' < 1 1 *\rm H 3 C / ^ N ^ H 3 C ^ N ' 0 (XLVI) etc. or Potassium trans-1,2-diaminocyclohexanetetraacetatomanganate (III) (KMn^^CyDTA) i s a weaker oxidant than potassium permanganate and i s known to undergo one-electron transfer from the anions of organic substrates."*"^' 1^''" Oxidation of TML was c a r r i e d out by using KMnIIICy.DTA at pH 6.0 i n phosphate buffer (0.1 M Na^WO^) under a nitrogen atmosphere. It was found that the products of the oxidation were XLVII when equimolar quantities of KMn I I ICyDTA and TML were used, and XLVIII when two equivalents of oxidant were used. (Molar r a t i o 2KMn^ 'I"''"CyDTA: 1 TML). - 64 -(XLVII) (XLVIII) The formation of XLVII must have been brought about by a coupling of the r a d i c a l intermediate (XLVI) which i s produced by an electron transfer from the enolate-anion to KMn I I ICyDTA. The formation of XLVIII when two equivalents of the oxidant are used may be explained by further i o n i z a t i o n and ele c t r o n transfer at the bridging e t h y l group of the i n i t i a l l y produced compound (XLVII). (A more d e t a i l e d discussion about the oxidation of TML by KMn^^^CyDTA w i l l appear i n section 3.3.3.) Formation of the coupling products (XLVII and XLVIII) i n the oxida-t i o n of KMn^^CyDTA shows that r a d i c a l s can be formed by oxidation of TML. A f a i l u r e to detect such coupling products i n the permanganate case i s probably due to further rapid reaction of the r a d i c a l intermediate with permanganate. The r a d i c a l can e i t h e r (a) undergo further one-electron transfer to the permanganate or (b) form a manganate (VI) ester. - 65 -The product of the electron transfer would be the carbonium ion (XLIX). However, XLIX i s a primary carbonium ion, and moreover, i t is connected with a strongly electron-withdrawing conjugated system, rendering i t a highly unlikely intermediate species. On the other hand, the combination of the radical with the paramagnetic permanganate ion to form the manganate ester (L) should go without any d i f f i c u l t y . Such radical-oxidant reactions-have been previously suggested for the permanganate oxidation of acetone^ 6 and formic acid,*^ 6 and for the chromic acid oxidation of some hydrocarbons. ' Normally inorganic manganate (VI) and manganate (V) ions rapidly 109 undergo disproportionation in neutral and acidic soltuions. The former is stable only above pH 12, and the latter only in concentrated potassium hydroxide. Very recently, however, Wiberg, Deutsch and 94 Rocek showed that esters of these manganate species, including LI and LII shown below, were formed as intermediate s in the course of the permanganate oxidation of olefins and that they are relatively stable even under neutral conditions. - 66 -— C C — ( >- CH CHCOOH 0 OH 0 0 \ \ v / Mn </ V (LI) (LII) 93 Lee and Brownridge have also provided spectrophotometric evidence for a c y c l i c manganate (V) ester intermediate (LII) i n the permanganate oxidation of cinnamic acid i n 0.99 M p e r c h l o r i c acid and recorded the rate constants for the formation and the decomposition of LII by stopped-flow techniques. LI i s a s t r u c t u r a l analog to the ester intermediate presumed here (L) . The recent works mentioned above strongly suggest that L may also be a r e l a t i v e l y stable intermediate. The manganate ester (L) can decompose i n e i t h e r of the following ways: (1) by proton loss from the a - p o s i t i o n to give an aldehyde (L I I I ) , (2) by hydrolysis to give a primary alcohol (LIV), or (3) by hydration at the 7 p o s i t i o n to give the 7-oxo compound (V), which i s the more stable isomer of the 7-hydroxy compound (LV), the d i r e c t product of the hydration. Ester decomposition v i a proton loss (route 1) and hydrolysis (route 2) are commonly presumed reactions i n permanganate and chromic acid oxidations. "'"'^  However, these pathways do not account for the formation of formaldehyde and the 7-oxolumazine as f i n a l products. Route 3, hydration at the 7-position, can accommodate a l l the facts known about the reaction. As has been discussed i n d e t a i l i n the Introduction, many 8-substituted p t e r i d i n e s , including TML, undergo - 67 -(1) Proton loss H - O I O . M n — 0 — C 3 ^| H B: H 3 C CH 3 N ^ 1 • N y NH H 0=C. H3C-? 3 N ^ 1 N NH + Mn IV (L) (LIII) (2) Hydrolysis 0 3Mn-0-CH 2-H20:^ H3C A V ^ » Y H 0 C H 2 N ^ Y N H H C C H 3 Y Y JL NH 0 (LIV) + Mn VI (3) Hydration H 2 0 s "uJtfn^O-CH 73 3 OfN~Y'',Y H 3C ^N NH 0 HC-H3C -f-3 NH 0 + CH 2 0 + Mn IV CH <f 0 3 H (LV) ° v N v Y° H 3 C ^ N NH (V) - 68 -n u c l e o p h i l i c addition at the 7-position with such reagents as water, 34 35 49 hydrogen cyanide, sodium b i s u l f i t e ' ' etc. The 7-position of the manganate ester (L) may be even more activated than that of TML i t s e l f by the presence of the manganate ester group. It i s known that there i s a rapid permanganate-manganate exchange reaction i n s o l u t i o n . L ± ± *MnO ~ + MnO 2 4 4 *MnO 2 + MnO." 4 4 The manganate ester (L) may thus be expected to e x i s t i n equilibrium with the permanganate ester (LVI), whose hydration and decomposition might be expected to be even more f a c i l e . P H3 H„C 0 3 M n V - H _ 0 - C H 2 ^ ^ N ^ 0 + MnO •N NH 4 <-H 3 C ^ N NH (L) + MnO 2-(LVI) MnO„ + MnO, 2 4 In conclusion, the re s u l t s that have been obtained are we l l accommodated by the mechanism outlined i n Scheme 6, and each step can be s a t i s f a c t o r i l y interpreted i n terms of the known chemistry of TML and permanganate. - 69 -Below pH 3.5 The results obtained for the permanganate-independent path in this pH region can be summarized as follows: (1) The reaction i s f i r s t order i n TML and zero order in permanganate. (2) The reaction i s subject to general acid catalysis. (3) The oxidation i s somewhat faster than the exchange rate multiplied by a factor of three. (4) A kinetic isotope effect of 2.46 at 31.4° was obtained for the oxidation of TML deuterated at the 7-methyl group. (5) The isolated products are hydroxybiacetyl (LVIII), ammonia, and compound LVII, which has been identified as either 2,4~dioxo-5,5,6-trihydroxy-6-methylaminopyridine or an isomer thereof. (See pp 123 f f for a description of the structural studies made on this compound. Throughout the thesis the structure for LVII i s that shown below.) These results suggest the mechanism outlined i n Scheme 7. It seems certain that the oxidation of TML is preceded by enolization. Thus, as in the deuterium exchange reaction, the i n i t i a l step involves a reversible formation of the protonated species, which is then deprotonated at the 7-methyl group by attack of conjugate bases of the buffer acids to give the enol form of TML. A l l the facts, except for the small isotope effect, summarized at the beginning of this section are consistent with the second step being rate-determining. The value of 0 0 II D HOCH C-CCH (LVIII) (LVII) - 70 -(LVII) Scheme 7. Mechanism of TML oxidation v i a e n o l i z a t i o n route (where HA denotes the buffer a c i d s ) . - 71 -2.46 for the deuterium isotope effect is rather small for a primary isotope effect. This, and the fact that the oxidation i s somewhat faster than the exchange, are to be discussed later. The enol thus formed should be readily oxidized by permanganate at i t s exocyclic double bond, leading to the dihydroxy compound LIX, a carbinol amine, which may be in equilibrium with i t s ring-opened form (LX), as expected from the fact that the hydrated species of TML undergoes reversible ring-opening (see below). 3 4 The double bond in the pyrimidine moiety of LIX regardless of whether i t is in the cyclic or ring-opened form w i l l be readily oxidized by the second molecule of permanganate to give another dihydroxy compound (LXI), which i s eventually converted to the f i n a l products, namely, ammonia, hydroxybiacetyl (LVIII) and 2,4-dioxo-5,5,6-trihydroxy-6-methylaminopyrimidine (LVII). 2,4-Dioxo-5-amino-6-methylaminopyrimidine (LXII) may be regarded as the model for the suggested reaction intermediates, LIX and LX. It was found that the oxidation of the b i s u l f i t e salt of LXII by permanganate at pH 1.0 occurred very rapidly, yielding 2,4-dioxo-5,5,6-trihydroxy-6-methylaminopyrimidine (LVII). This fact provides strong support for the proposed mechanism for the TML oxidation. - 72 -CH 3 H N HN .0 HN y Y MnO HO NH pH 1.0 HO NH OH n 0 0 (LXII) (LVII) As mentioned beofre, d i f f i c u l t i e s were met i n i n t e r p r e t i n g the small k i n e t i c isotope e f f e c t and the f a s t e r rate of the oxidation than the exchange. These facts can be explained by assuming the presence of another permanganate-independent path, i n a d d i t i o n to the e n o l i z a t i o n route. I t i s known that 8-substituted lumazine d e r i v a t i v e s , i n c l u d i n g TML, undergo a c i d - and base-catalyzed hydration. I t would be expected that the double bond of the pyrimidine moiety of the hydrated species, which i s e s s e n t i a l l y the same as the intermediate i n the e n o l i z a t i o n route (LIX i n Scheme 7), i s oxidized r e a d i l y by permanganate. Therefore, i f there were acid catalyzed hydration, as well as e n o l i z a t i o n , an oxidation mechanism v i a hydration might occur, as shown i n Scheme 8. The f i n a l products for the hydration route would be ammonia, b i a c e t y l and 2,4-dioxo-5,5,6-trihydroxy-6-methylaminopyrimidine (LVII) as expected from the mechanism suggested for the hydration route. Ammonia and the pyrimidine fragment are the common products for the e n o l i z a t i o n and hydration routes. Attempts to separate b i a c e t y l from the oxidation products were unsuccessful. In order to obtain support for the occurrence of the hydration route, the permanganate oxidation of 6,7-diphenyl-8-methyllumazine (LXIII) was investigated. (The r e s u l t s and discussion of t h i s r e a c t i o n - 73 -B . C . f I J 5 0 + MnO CH HN-3" / = N 0 CH„ OH H 4 f a s t CH HN 3 H H 3C Mn NH H.C CH, OH H 0 0 H0CHN-V^Nvy° NH NH 3 + CH3C-CCH3 + H 0 . 0 (LVII) Scheme 8. Oxidation mechanism of TML v i a hydration route. - 74 -will appear in Section 3.2). It was found that the oxidation of LXIII via a permanganate-independent path was acid catalyzed and its rate was almost identical with the reported rate for the hydration. C6 H5 C6 H5 Since the steric requirement for the hydration is smaller in TML than in LXIII, i t seems reasonable to assume that the hydration route, as well as the enolization route, are included in the permanganate-independent path of TML oxidation. The oxidation via rate-determining hydration is not associated with a kinetic isotope effect of the 7-methyl group. If the discrepancy between the oxidation rate and the exchange rate was the result of hydration occurring, the deuterium isotope effect for the oxidation via enolization can be calculated using the data in Tables 6 and 7. k(H) = k° b s(TML)-k h y d 0.00314-(0.00314-0.00210) = k(D) k°bs(TML-7-d )-k h y d 0.00127-(0.00314-0.00210) A value of 9.13 at 31.4° is a normal-to-large primary deuterium isotope effect. 3.1.2.2 Permanganate-dependent Path The results concerning the reaction via the permanganate-dependent path may be summarized as follows: - 75 -(1) The rate of the reac t i o n i s f i r s t order i n TML, permanganate and hydronium ion. (2) The rate i s independent of the buffer concentration. (3) Substitution of the hydrogens of the 7-methyl group by deuterium does not a f f e c t the rate of the reaction. These k i n e t i c r e s u l t s suggest the presence of one of the following routes, which are k i n e t i c a l l y i n d i s t i n g u i s h a b l e . (1) v i a protonation of TML KRH + + TML + H + , > TMLH (fast) + - k 3 TMLH + MnO^ > product (slow) (2) v i a protonation of the permanganate + K a MnO, + H „ > HMnO, (fast) 4 < 4 k 3 HMnO^ + TML — — > product (slow) The rate laws f o r the above mechanisms can be given by (1) V = k 3 [TMLH+] [Mn04 ] (20) k [H +] = [TML] [MnO. ] = k , [TML] [MnO. ] (21) hn+ 4 ° b S 4 where , kofH ] - 76 r (2) V = k 3[TML][HMn0 4] (23) k [H +] = -§ [TML] [MnO ] = k , [TML][MnO, ] (24) K 4 obs 4 where k„[H +] k . = -Az (25) obs K a The values of k^ for both routes are c a l c u l a b l e from equations (22) and (25) by using the observed rate constants, pH, and the reported 3 4 i o n i z a t i o n constants of TML (pK + = 0.85) and permanganic acid (pK = cn a - 4 . 6 ) . 1 1 7 From the rate data at pH 0.997, 1.39, 2.00 and 2.39 i n Table 4, the values of k^ were calcualted to be 19.1 + 4.0 l.mole "''sec "*" for the route (1) and (5.40 + 1.15) x 10 6 l . m o l e _ 1 s e c _ 1 f o r the route (2) re s p e c t i v e l y . Reactions whose rates exceed the order of 10 5 l.mole "'"sec * are rare except f o r proton and electron transfer reactions in v o l v i n g anions, neither of which i s l i k e l y i n the present case. Therefore, i t i s reasonable to assume the oxidation goes v i a protonation of TML (route 1). For both routes 1 and 2 i t would be expected that a p l o t of log(rate) 34 versus pH would tend to curve near the pK_TJ+ of TML (0.85) and f i n a l l y on l e v e l o f f when the a c i d i t y of the s o l u t i o n i s high enough to protonate most of the neutral TML. Such a trend could not be detected above pH 1.0 (see Figure 9). Unfortunately i t was not possible to study the pH region below 1.0, since the reaction i n such highly a c i d i c s o l u t i o n i s too f a s t to be followed. The permanganate oxidation of 6,7-diphenyl-8-methyllumazine (LXIII on page 74 ) i s slower than that of TML and i t was thus possible to study t h i s compound i n the highly a c i d i c region. - l i -lt was indeed found that the rate of the permanganate-dependent path levels off when protonation i s complete. (To be discussed i n more detail in Section 3.2). Cleavage of the C-H bond in the 7-methyl group i s not involved in the reaction between protonated TML and permanganate as there i s no kinetic isotope effect observed for TML-7-dg. The reaction l i k e l y occurs via addition of the permanganate to the carbon-carbon double bond at the 6,7 position of protonated TML, which may be somewhat more reactive than the unprotonated species because of the interruption in conjugation. Permanganate addition to the double bond w i l l lead to the dihydroxy compound (LXIV), which w i l l then s p l i t into biacetyl, ammonia and 2,4-dioxo-5,5,6-trihydroxy-6-methylaminopyrimidine (LVII). The mechanism for the oxidation via the permanganate-dependent path may be depicted in Scheme 9. Attempted isolation of biacetyl failed. This i s probably because biacetyl i s oxidized further by the large excess of permanganate used in the product analysis experiments (see Experimental section). The same mechanism i s proposed for the permanganate-dependent path of the oxidation of 6,7-diphenyl-8-methyllumazine. The products in that instance are benzil, ammonia and 2,4-dioxo-5,5,6-trihydroxy-6-methylamino-pyrimidine (LVII) (see Section 3.2). 3.1.3 Conclusions Regarding TML Oxidation Mechanism It was found that the permanganate oxidation of TML in the basic region (pH > 3.5) goes via i n i t i a l formation of the enolate-anion by removal of a proton from the 7-methyl group. The oxidation is very - 78 -(LVII) Scheme 9. Mechanism of o x i d a t i o n of TML v i a the permanganate-dependent path. - 79 -f a c i l e and produces unusual products, formaldehyde and the demethylated 7-oxo compound. This c h a r a c t e r i s t i c feature of the oxidation can be at t r i b u t e d to the unique properties of TML, namely (1) the strongly a c i d i c character of the 7-methyl protons, and (2) i t s tendency to undergo hydration at the 7-position. The f i r s t property i s responsible f o r the f a c i l e reaction and the second f o r the unusual products. In the a c i d i c region (pH < 3.5), there seem to be three routes involved i n the reaction. Each of the routes involves i n i t i a l protonation of TML, followed by competitive and rate-determining e n o l i z a t i o n , hydration or oxidation as depicted below. - 80 -3.2 Permanganate Oxidation of 6,7-Diphenyl-8-methyllumazine 3.2.1 Introduction The permanganate oxidation of TML (6,7,8-trimethyllumazine) in acidic media seems to consist of three paths, two of which are independent of the permanganate concentration and proceed via enolization and hydration, and the third which is a permanganate-dependent path. In highly acidic solution the oxidation is too fast to obtain good kinetic results, making i t difficult to elucidate the complete mechanism. 6,7-Diphenyl-8-methyllumazine (LXIII) does not have enolizable protons and hence its oxidation mechanism would be expected to be simpler than that of TML. Moreover, kinetic data for the hydration of 3,8-dimethyl-6,7-diphenyllumazine (LXV), a structural analog of LXIII, are available, whereas those of TML are not. In order to gain a better understanding of the acidic oxidation of TML, the study of the permanganate oxidation of LXIII was carried out. (LXIII) R = H (LXV) R = CH3 CH3 C 6 V N Y N Y 0 •N R 0 3.2.2 Results Under the pseudo first-order conditions (large excess of permanganate) excellent linear correlations between the logarithm of the lumazine absorbance at 416 nm and time were always observed up to two half-lives of the reaction. A typical plot is reproduced in Figure 11. The kinetic Figure 11. Ty p i c a l pseudo f i r s t - o r d e r rate p l o t for the oxidation of 6,7-diphenyl-8-methyllumazine. pH = 1.50, y = 0.5 (KH 2P0 4 = 0.02 M), T = 25°, KMnO^ = 7.32 x 10" 4 M, L.= 5.00 x 10~ 5 M. 1000 Time (sec) - 82 -data obtained at various pH and at various concentrations of permanganate and buffer are tabulated i n Table 8. In the highly a c i d i c region (pH < 0.5) manganese dioxide, the reduction product of permanganate, p r e c i p i t a t e s rather slowly, but i n pH > 0.5 the p r e c i p i t a t i o n i s rapid, making analysis of the k i n e t i c s d i f f i c u l t . This d i f f i c u l t y was overcome by the addition of a small amount of potassium phosphate (usually enough to make the s o l u t i o n 0.02 M). The e f f e c t of added phosphate on the rate was n e g l i g i b l e as shown i n Table 8 by the runs at pH 1.5, provided that the i o n i c strength of the s o l u t i o n i s maintained at a constant value. 3.2.3 Discussion The good s t r a i g h t l i n e p l o t obtained i n Figure 11, where ln(absorbance| i s p l o t t e d against time, indicates that the re a c t i o n i s f i r s t order i n the lumazine. The order with respect to permanganate was then obtained by maintaining the a c i d i t y constant and varying the permanganate concentration A p l o t of the pseudo f i r s t - o r d e r rate constant against permanganate concentration gave a s t r a i g h t l i n e ; extrapolation of the l i n e to zero permanganate concentration intercepted the ordinate somewhat above the o r i g i n of the graph. This was observed over the whole pH range studied. Two t y p i c a l p l o t s of k ^ g versus permanganate concentration are shown i n Figure 12. Such pl o t s can be represented by the equation k 1_ d[L] _ o + k 2[Mn0 4 ] (26) obs [L] dt v = - = k ^ t L ] + k 2[L][Mn0 4 _] (27) Table 8. Rate Constants (sec ) for the Permanganate Oxidation of 6,7-Diphenyl-8-methyllumazine (T = 25°). pH,Ho KMnQ4 (mole/1.) Buffer o 7.32 X i o " 4 1.46 X i o " 3 2.19 H X i o " 3 2.93 X i o " 3 3.66 X i o " 3 (mole/1.) V pH, 1.75 7.72 X i o " 4 1.38 X i o " 3 1.93 X i o " 3 2.45 X i o " 3 3.03 X i o " 3 0.02 a 0.5 1.50 1.36 X i o " 3 2.28 X i o " 3 3.32 X i o " 3 4.27 X i o " 3 5.08 X i o " 3 0.02 a it - 2.30 X i o " 3 - - 5.03 X i o " 3 o.io a tt 1.2 2.28 X i o " 3 3.95 X i o " 3 5.55 X i o " 3 7.13 X i o " 3 8.77 X i o " 3 0.02 a II 1! 2.25 X IO" 3 3.98 X i o " 3 5.58 X i o " 3 7.18 X i o " 3 8.70 X i o " 3 o . i b n II 2.26 X i o " 3 3.92 X i o " 3 5.55 X i o " 3 7.18 X i o " 3 8.63 X i o " 3 0.3 b ti II 2.30 X i o " 3 3.98 X i o " 3 5.57 X i o " 3 7.16 X i o " 3 8.75 X i o " 3 0.5 b it 0.93 3.78 X i o " 3 6.45 X i o " 3 9.37 X i o " 3 1.18 X i o " 2 1.42 X i o " 2 0.02 3 II 0.83 4.55 X IO" 3 7.93 X i o " 3 1.14 X i o " 2 1.47 X i o " 2 1.78 X i o " 2 0.02 a it H , 0.48 o 6.10 X IO"3 1.14 X i o " 2 1.59 X i o " 2 2.03 X i o " 2 2.55 X i o " 2 - -0.46 6.70 X i o " 3 1.17 X i o " 2 1.70 X i o " 2 2.20 X i o " 2 2.70 X i o " 2 - -0.39 6.72 X IO" 3 1.20 X i o " 2 1.72 X i o " 2 2.20 X i o " 2 2.75 X i o " 2 - -0.28 7.43 X IO" 3 1.31 X i o " 2 1.92 X i o " 2 2.42 X i o " 2 3.03 X i o " 2 - --0.07 9.25 X i o " 3 1.64 X i o " 2 2.37 X i o " 2 3.10 X i o " 2 3.77 X i o " 2 - --0.40 9.27 X i o " 3 1.78 X i o " 2 2.53 X i o " 2 3.37 X io" 2 4.08 X io" 2 - -3 KH-PO.; b CCl„C0 0H-CCl oC0„Na. - 84 -— i 1 1 1 0.8 2.4 4.0 10 3 KMn04 (mole/1.) - 85 -The equation suggests that there are two reaction paths, one permanganate-independent and the other permanganate-dependent, involved, as i n the oxidation of TML. In order to see the pH dependence of these two paths, the data i n Table 8 were analyzed by the method of l e a s t squares with respect to each pH. The intercept (k^°) and slopes (k^) of the p l o t s of k Q ^ g versus permanganate concentration are shown i n Table 9. When the logarithms of k^° and k^ are plotted against a c i d i t y of the s o l u t i o n a l e v e l l i n g - o f f of the rates i s observed under highly a c i d i c conditions (Figures 13 and 14). These plo t s show t y p i c a l i o n i z a t i o n curves. In the permanganate-independent path, i t i s obvious that the i o n i z a t i o n i s of the substrate. On the other hand i n the permanganate-dependent path, the protonation occurs e i t h e r at the substrate or at the oxidant, these routes being k i n e t i c a l l y i n d i s t i n g u i s h a b l e (see equations ( 3 0 ) and ( 3 ' 5) on page 94). In the permanganate-independent path and i n both cases of the permanganate dependent path i t would be expected that a log-log p l o t of the rate against the degree of i o n i z a t i o n of the substrate would give a s t r a i g h t l i n e of u n i t slope. The degree of i o n i z a t i o n of the lumazine at the a c i d i t i e s where the rates were obtained was determined by the usual 118 spectrophotometric method ; these r e s u l t s are included i n Table 9. Log-log p l o t s of both rates (k^° and k^) against the degree of i o n i z a t i o n of the lumazine indeed give good s t r a i g h t l i n e s , with slopes of 0.72 for k^° and 0.95 f o r k^ (Figure 15 and Figure 16). In order to see whether the r e a c t i o n i s subject to s p e c i f i c or general a c i d c a t a l y s i s the e f f e c t on!the rate of varying the buffer concentration was examined (CCl«C00H-CCl„C00Na buffer system at pH 1.20). - 86 -Table 9. Rate Constants f o r the Permanganate-Independent Path (k^) and the Permanganate-Dependent Path (k 2) of the Oxidation of 6,7-Diphenyl-8-methyllumazine (T = 25°) pH,H (l.mole Isec L ) k° (sec Co r r e l . coeff. 1.75a 0.764 2.36 X io" 4 .999 1.50a 0.0758 1.29 4.37 X io" 4 .999 1.20a 0.140 2.21 6.96 X io" 4 .998 „ b 0.137 2.20 7.16 X io" 4 .999 II b 0.140 2.18 7.16 X io" 4 .999 „ b 0.144 2.19 7.36 X io" 4 .998 0.93a 0.240 3.59 1.26 X IO" 3 .999 0.833 0.322 4.54 1.31 X IO" 3 .999 0.48 0.433 6.51 1.55 X io" 3 .997 0.46 0.440 6.94 1.63 X 10"3 .998 0.39 0.488 7.03 1.65 X io" 3 .998 0.28 0.524 7.76 1.82 X io" 3 .999 -0.07 0.644 9.76 2.19 X io" 3 .999 -0.40 0.737 10.78 1.73 X io" 3 .999 a,b have the same meaning as those i n Table 8. degree of i o n i z a t i o n of 6,7-diphenyl-8-methyllumazine. 1.2 - 91 -As can be seen i n Table 9, the rates of both paths are independent of the buffer concentration, suggesting both paths are subject to s p e c i f i c a c i d c a t a l y s i s . The mechanism of each path w i l l be discussed i n d i v i d u a l l y below. Permanganate-independent Path for 6,7-Diphenyl-8-methyllumazine The fa c t s that the reaction i s subject to s p e c i f i c a c i d c a t a l y s i s and the rate i s independent of permanganate concentration suggest that there might be an acid-catalyzed hydration involved. I f t h i s i s so, the oxidation path v i a acid-catalyzed hydration might be described as follows. OH C H 3 slow H -N NH - 92 -OH f H3 H N \ ^ 0 NH + M n ° 4 f a s t CH, H OH | 3QH •N I + Mn NH 0 (LXVI) 0 0 II II <j>-C-C-cf) + NH + H„CHN H H O — K . NH 0 (LVII) The rate of the oxidation with the above mechanism should be i d e n t i c a l with that of hydration. P f l e i d e r e r , P e r r i n and t h e i r co-34 workers showed that the hydration i s a c i d - and base-catalyzed, and that the hydration rate f o r 3,8-dimethyl-6,7-diphenyllumazine (LXV) i s -4.867,, as l o g ( r a t e ) , at pH 2.96 at 20°. Since the values of pK B H+ of 3,8-dimethyl-6,7-diphenyllumazine (LXV) and of 6,7-diphenyl-8-methyllumazine are very close (pK + of the dimethyl and the monomethyl 34 compounds are 0.20 and 0.36 r e s p e c t i v e l y ), the hydration rates of these two compounds should not d i f f e r greatly. The oxidation rate of 6,7-diphenyl-8-methyllumazine at pH = 2.96 i s extremely slow so that the rate at t h i s pH was obtained g r a p h i c a l l y . Extrapolation of the s t r a i g h t l i n e of the log k^°-pH p l o t i n Figure 13 to pH 2.96 gave a value of -4.580 to log This agrees w e l l with the reported hydration rate and hence i t i s reasonable to conclude that the oxidation i s preceded by rate-determining hydration! - 93 -The steps after the hydration will be rapid attack of permanganate at the carbon-carbon double bond of the hydrate to give a dihydroxy compound LXVI which is eventually converted to the final products, namely, benzil, ammonia and 2,4-dioxo-5,5,6-trihydroxy-6-methylamino-pyrimidine (LVII), as has been observed. Permanganate-dependent Path for 6,7-Diphenyl-8-methyllumazine Since the reaction is subject to specific acid catalysis, the reaction can be represented by either of the following schemes, as in the case of the permanganate-dependent path for the TML oxidation. (1) via protonation of the lumazine L + H + LH+ « - k 3 LH + MnO. — — > product (fast) (slow) The rate laws for the above schemes can be given by (fast) (slow) (2) via protonation of permanganate + K MnO." + H — — * HMnO 4 < 4 k 3 L + HMnO, — p r o d u c t - 94 -(1) v = k3[LH+][Mn04~] (28) = k3a[L]T[Mn04"] (29) where [L]^ = [L] + [LH+] and a is degree of ionization of the lumazine. kobs = k 3 a ( 3 0 ) or log k o b g = log a + log k 3 (31) .+ [ M n ° 4 _ ] (2) v = k3[L][HMn04] = k3[L][H'] — ( 3 2 ) since thus = [L][H +] BH+ [ L ] T a KBH+ (33) or v = k 3 [L]T[Mn04 ] (34) kobs = k3 Hr- ( 3 5 ) k3 KBH + log k Q b g = log a + log — (36) Equations (31) and (36) indicate that values of k 3 for each route can be obtained from the intercept of the plot of log versus log a (Figure 16) and the reported ionization constants of 6,7-diphenyl-8-methyllumazine (pK + = 0.36 3 4) and permanganic acid (pK = -4.6 ^ ^ ) . Da a This gives values of k 3 of 14.49 l.mole 1sec 1 for route 1 and 1.323 x - 95 -6 —1 — 1 10 l.mole sec for route 2. As mentioned in the discussion of the permanganate-dependent path for the TML oxidation, reactions whose rates 5 -1 -1 exceed the order of 10 l.mole sec are rare. Accordingly, route 1 is preferred. The value of 14.49 l.mole ''"sec "*" is slightly less than the value of 19.12 1. mole "''sec ^  obtained for the permangante-dependent path of the TML oxidation. This suggests the same reaction, namely permanganate addition to the carbon-carbon double bond of the protonated lumazine, takes place in both compounds. The slightly slower rate for the 6,7-diphenyllumazine than for TML is probably due to the larger steric requirement of the phenyl substituents than of the methyl substituents. The mechanism of the permanganate oxidation of 6,7-diphenyl-8-methyllumazine via the permanganate-dependent path may be outlined as below, which is essentially the same as that of TML CH. I 3 .NH + H KBH+. "N fast 0 CH, H N Vfe^Nv^O ? 3 + H • N v V N Y \ MnO NH 4 slow OH f H 3 H + Mn V 0 0 II II .<))-C-C-<j) + NH, H CHN H -N H0-H0-Y NH 0 - 96 -3.2.4 Conclusions Regarding the Oxidation Mechanism of 6,7-Diphenyl- 8-methyllumazine It was found that the permanganate oxidation of 6,7-diphenyl-8-methyllumazine consists of two paths, one permanganate-independent and the other permanganate-dependent. The former proceeds via rate-determining hydration, while, the latter goes via addition of permanganate to the carbon-carbon double bond at the 6,7 position of the protonated lumazine. 3.3 Further Discussion 3.3.1 Keto-enol Tautomerism of 6,7,8-Trimethyllumazine In preceding sections terms such as "enol" or "enolate-anion" have been frequently used with reference to TML. Like acetone and other carbonyl compounds, TML can undergo a sort of "keto-enol" tautomerism, as shown below. 0 0 'keto" "enol" The NMR spectrum of TML in D^O shows no peak due to the exocyclic methylene protons of the enol form, indicating the equilibrium lies well toward the keto form. The equilibrium constant for the tautomerism (K ) may be calculated by assuming the following consecutive ionization processes. - 97 -CH H C I K 2 ^ / N v / J . v . . ^ •4 + H H 3C NH 0 "enolate" H„C CH, 2\ 1 V N rN^,,,0 H 3C X ^ N + H K2 H 2 C V - N > ^ CH. I 3 NH H 3 C / ^ N NH 0 "enolate" "enol" Ionization constants and are expressed by K, [enolate][H ] [keto] (37) thus [enolate][H ] 2 [enol] K taut [enol] = \ [keto] K„ (38) (39) Unfortunately, cannot be determined experimentally because i n basic s o l u t i o n TML ex i s t s as an equilibrium mixture of the hydrate (LXVI), the hydrate anion (LXVII) and the ring-opened form (LXVIII) 34 - 98 -H„C 0 H I 3 I n H„C. 3 \p N \^ N -Y 0 +3 H 3 C ^ N -OH C H 3 '-N^ .^ O H3CHN N ^ .0 i NH /L J\Z . N H < = = = ! H C - A A N H H 3C (LXVI) 0 (LXVII) (LXVIII) 54 Stewart and McAndless estimated the value of pK^ to be about 11, on the basis of s t r u c t u r a l considerations. They assumed that the a c i d i t i e s of the hydrate and of the 7-methyl group of the unhydrated compound would not d i f f e r greatly. (The pK of the hydrate was reported £L to be 10.34. 3 4) S i m i l a r l y , the b a s i c i t i e s of the enolate-anion and the hydrated lumazines (below) should not d i f f e r greatly. OH f R N R' NH , K + H — t OH R j / N R N ^NH 0 R = H, CH 3, CH(CH 3) 2, C 6H 5; R' = CH 3, r i b i t y l The values of pK & f o r most of the hydrates shown above f a l l i n the 34 range between 6.5 and 7.0. If pK^ i s approximated by the value of 7.0, K can be calculated from equation 39. taut K. taut K„ 10 10 -11 = 10 -4 This value indicates the enol content of TML i s about 40 times more than - 99 -that of acetone, whose K i s reported to be 2.5 x 10 ' taut r -6 119 3.3.2 Oxidation as a Method of Locating the S i t e of the Hydration  i n Heterocyclic Systems: Reexamination of i t s U t i l i t y One of the frequently used methods to locate the s i t e of the hydrate i n h e t e r o c y c l i c systems i s the oxidation method developed by Albert and others"?^ (see Section 1.1.4). This method i s based on the idea that h e t e r o c y c l i c compounds that have water bound covalently across a -C=N bond are by nature secondary alcohols and can thus be oxidized * 45 to the corresponding oxo compounds. In 1966, Jacobsen showed that 6,7,8-trimethyl derivatives of some amino- and hydroxy-pteridines (for instance, LXIX) underwent rapid oxidative demethylation with permanganate i n basic s o l u t i o n to produce the corresponding 7-oxo compounds i n very good y i e l d . He assumed that the reaction takes place by i n i t i a l hydration at the 7-position, followed by oxidative demethylation (considered as oxidation plus decarboxylation (see below)). On t h i s basis he considered such oxidative demethylations could be used as an a d d i t i o n a l a i d to those already appearing i n the l i t e r a t u r e to locate the p o s i t i o n of hydration. C H3 H3 V v N OH ?H3 NHCH„ (LXIX) - 100 -This hypothesis seems untenable i n the present case, however, since the hydrate i s a t e r t i a r y alcohol by nature; oxidation and subsequent decarboxylation of such compounds usually requires vigorous conditions or prolonged reaction times. He sought support for h i s hypothesis by c i t i n g the oxidation of 1-methylcyclohexanol to cyclohexanone by potassium permanganate at room temperature. However, examination or the o r i g i n a l paper on the oxidation of 1-methylcyclohexanol reveals that the r e a c t i o n a c t u a l l y took two weeks to complete and gave cyclohexanone, plus other fragmented products. These conditions are i n remarkable contrast to those of the p t e r i d i n e oxidations. In spite of such uncertainty i n the oxidation mechanism, t h i s method has been applied to locate the s i t e of hydration of several p t e r i d i n e derivatives which have 45 71 a methyl group at the suspected s i t e of the hydration. * It has become apparent from the present study that such f a c i l e oxidative demethylation occurs v i a i n i t i a l i o n i z a t i o n of the 7-methyl group rather than v i a i n i t i a l hydration at the 7-position. Thus, Jacobsen's conclusions are unsound. At t h i s point i t may be of value to consider further the l i m i t a t i o n of the oxidation method. Albert and Armarego"^ pointed out that the oxidation method i s unsatisfactory i n the following cases: (1) when hydration takes place at more than two s i t e s i n the molecule. (2) when the i n i t i a l l y formed oxo compound forms a hydrate which i s further oxidized to a dihydroxy compound. (3) when the hydrated species i s i n equilibrium with a ring-opened structure ( i . e . , a c y c l i c amino-aldehyde), which may be oxidized considerably f a s t e r than the ring-closed structure. This could lead to the i n c o r r e c t conclusion that the o r i g i n a l material was not c y c l i c . - 101 -120 Recently, Paterson and Wood applied the oxidation method to 8-substituted pyrido[2,3-d]pyrimidines (LXX) and (LXXI). They noticed that these compounds showed a large hypsochromic s h i f t i n t h e i r UV spectra i n a c i d i c s o l u t i o n . On the assumption that t h i s i s due to covalent hydration occurring at the 7-position, they attempted to convert these compounds to the corresponding 7-oxo compounds by using permanganate i n a c i d i c s o l u t i o n . The attempts f a i l e d , however, and they were forced CHoCH_0H I 2 2 CH CH OH OH !. H |/NYNy v/£ HV"N yN Y0Mn° L NH L I NH CH.CH„0H - 0 C V / N \ I / N \ ^ C NH (LXX) R = CH, (LXXI) R = H to leave t h e i r hypothesis that the hydration i s taking place i n these systems unconfirmed, even though the UV spectra provided strong support f o r t h i s . The present study of the permanganate oxidations of TML and 6,7-diphenyl-8-methyllumazine show that the hydrated forms of these molecules are oxidized r a p i d l y at the double bond i n the pyrimidine moiety. From t h i s and the general tendency of permanganate to react f a s t e r with 121 double bonds than with alcohols, i t seems very l i k e l y that the hydrated species of LXX and LXXI are oxidized at the double bond of the pyrimidine moiety, rather than at the hydrate s i t e . These considerations lead to the conclusion that the oxidation method with permanganate i s unsatisfactory i n the case when the hydrate - 102 -contains a reactive double bond. In order to avoid the difficulties experienced by Paterson and Wood, i t seems worthwhile to add this latter criterion to the limitations already noted by Albert and Armarego"^ with regard to the oxidation-method. 3.3.3 Mechanism of the Enzymic Conversion of 6,7-Dimethyl-8-ribityllumazine to 6-Methyl-7-oxo-8-ribityllumazine In 1956 Masuda reported isolation of three compounds, riboflavin (III), 6,7-dimethyl-8-ribityllumazine (IV), and 6-methyl-7-oxo-8-ribityllumazine (V) from the fermentation product of "Eremothecium 8 9 ashbyii". ' These compounds bear a close structural resemblance to one another and i t soon became clear that IV is the precursor of both TTT10,57-62 ...58,75. I l l and V in biological systems. CH, H3C N ^ O NH H 3 C \ / N \ ^ N \ ^ 0 0 (III) H3C^ ^N NH R = ribityl (IV) |H3 H / S t H3C^ N Y NH (V) These conversions can be achieved both enzymically and chemically. A number of mechanisms for the chemical and enzymic conversions of these - 103 -systems have been proposed, but a c l e a r e l u c i d a t i o n of the mechanisms has not been reached yet i n e i t h e r case (see the Introduction). The mechanism of the enzymic conversion of IV to V w i l l now be discussed with the help of newly a v a i l a b l e knowledge, obtained from the present study of the permanganate oxidation of TML. 79 Kuwada and h i s co-workers found that treatment of IV with the enzyme s o l u t i o n prepared from "Eremothecium a s h b y i i " produced V, formaldehyde and formic a c i d . They found that the conversion occurred only above pH 4, the y i e l d s of the products improving with a r i s e of pH. They considered the enol form of TML as the intermediate, but a d e t a i l e d mechanism of the reaction has not yet been proposed (see the Introduction), Their r e s u l t s resemble those found i n the present study of the permanganate-TML reaction. The fa c t that the y i e l d s of the products are higher at higher pH suggests that i o n i z a t i o n of the 7-methyl group i s also involved i n the enzymic conversion. The i o n i z a t i o n may have been achieved by the a c t i o n e i t h e r of the enzume or of general bases present i n the l i v i n g systems. In fa c t a l i v i n g system can be regarded as a pool of b u f f e r i n g systems. Although the d e f i n i t i v e mechanism of the enzymic reaction can not be put forward at the present time, because of lack of information regarding the nature of the enzyme, i t may well be that the electron tr a n s f e r takes place from the enolate-anion to the enzyme as i n the permanganate oxidation of TML, since many of the 122 123 oxidative-reductive enzymes can undergo one-equivalent oxidations. ' Potassium trans-1,2-diaminocyclohexanetetraacetatomanganate (III) (KMn^^CyDTA) i s an e x c l u s i v e l y one-equivalent oxidant and i s known to undergo an el e c t r o n - t r a n s f e r from anions of organic substrates. - 104 -These facts make this reagent attractive to use as the model for the enzyme action supposed above, and the investigation of the oxidation III mechanism of TML with KMn CyDTA was carried out. The products of the oxidation under an oxygen atmosphere were 6,8-dimethyl-7-oxolumazine and formaldehyde; the yield of the former being greater than 81%. In Figure 17 the change in TML absorbance at 404 nm with respect to time was recorded for reactions at different pH and at different phosphate buffer concentrations, a l l reactions being conducted under an oxygen atmosphere. As can be seen in Figure 17, the reaction is faster in higher buffer concentration at the same pH, and at higher pH in the same buffer concentration. This fact indicates that ionization at carbon (presumably the 7-methyl group) is involved in the oxidation. The enolate-anion thus formed will undergo an electron-transfer with KMn'^ '''CyDTA to produce a free-radical intermediate, as in the case of the permanganate oxidation. The formation of a free-radical was demonstrated by the experiment carried out under a nitrogen atmosphere, as described below. When the reaction was carried out under nitrogen a large quantity of yellow precipitate appeared (approximately 90% of the weight of starting material). It was found that the reaction produced two kinds of product depending upon the amount of KMn^^CyDTA used. The product obtained when equimolar quantities of oxidant and TML were used was assigned the structure XLVII, and XLVIII when two equivalents of III oxidant (molar ratio 2KMn CyDTA:1 TML) were used, on the basis of mass spectra (see the Experimental). - 106 -CH, f 3 CH, CH, 3 I ^ NJV / N \ CH=CHV / N \ ^ N ^ 0 N CH 3 H 3C NH HN 0 0 N Y " V H 2 _ C H 2 Y N N Y ° ^ ' I J J L A H £ - \ . A - N ^ ^ V ' V ^ N "3 N T NH (XLVII) (XLVIII) There Is l i t t l e doubt that the formation of XLVII i s brought about by a coupling of the r a d i c a l intermediate produced by transfer of an electron from enolate-anion to oxidant, i n the way depicted below. CH H C 1 3 L\^-N \ ^ N ^ . 0 r + NH 0 + + BH (XLVII) - 107 -The formation of XLVIII i n the presence of two equivalents of the oxidant can be explained by further i o n i z a t i o n and e l e c t r o n - t r a n s f e r at the bridging e t h y l group of the i n i t i a l l y produced compound LXVII. (whether the dianion shown below i s involved or whether the monoanion can be oxidized to XLVIII by two equivalents of Mn*1''' i s not known.) Consequently, the r e s u l t s obtained under a nitrogen atmosphere c l e a r l y show that the reaction between the enolate-anion and KMn***CyDTA occurs v i a e l e c t r o n transfer and that further rapid reaction between the r a d i c a l and the oxidant does not take place. Under an oxygen atmosphere the r a d i c a l may quickly pick up an oxygen molecule to give a peroxy r a d i c a l . The p l o t of TML absorbance versus time (Figure 17) i s t y p i c a l of e i t h e r autocatalysis or a r a d i c a l chain reaction. The reaction seems to be the l a t t e r rather than an au t o c a t a l y s i s , since there was no i n d i c a t i o n of r a t e - a c c e l e r a t i o n by II 2-the addition of Mn CyDTA , which could be expected to accumulate during the reaction. It i s reasonable to assume that the peroxide r a d i c a l abstracts a hydrogen atom at the 7-methyl group of the s t a r t i n g TML, because the resultant r a d i c a l enjoys extensive resonance s t a b i l i z a -t i o n . From these considerations, the reaction may be described by the following r a d i c a l - c h a i n autoxidation mechanism. (XLVIII) (XLVII) - 108 -I n i t i a t i o n : - 109 -Termination: CH. H C I 3 2 % . ^ N \ ^ , N ^ „-0 H3C T NH CH, CH I 3 V N V ^ ^ Y V W N V* NN>° * HN Y ^ N ^ C H 3 H 3 C N 1 < (XLVII) NH •OOCHn H3C CH I 3 •N •N NH 0 H 9C H3C-CH, I 3 NH ^N^ 0 CH • CH -0-0-CHo HN. N ^ ^ C H 3 H 3C f 3 ' N \ ^ N \ ^ 0 NH N^ 0 (LXXIII) CH • ° 0 C H 2 ^ / N ^ N - ^ H, CH, 0. ^ -N Y ->Y HH HN f H 3 C H „ - 0 - 0 - C H 2 ^ N \ ^N\_,0 Y HoC NH (LXXIII) III Scheme 10. Mechanism of the oxidation of TML by KMn CyDTA under oxygen atmosphere. - 110 -If the r a d i c a l concentration i s very low, then most of the product would be the hydroperoxide LXXII, which may decompose to produce formaldehyde and the 7-oxolumazine v i a hydration at the 7-position, as shown below. (This i s e s s e n t i a l l y the same reaction as the decomposi-tio n of the manganate ester intermediate proposed f o r the permanganate oxidation of TML.) E2°'"\ CH, H0-0-CH_//N H 20 + CH 20 + f 3 H Y Y H 3 C ^ l 1 NH (LXXII) If the r a d i c a l concentration i s f a i r l y large, the y i e l d of the peroxide LXXIII should be increased. The peroxide LXXIII may decompose i n a s i m i l a r manner to the hydroperoxide LXXII, y i e l d i n g formaldehyde, the 7-oxolumazine and the alcohol LXXIV, as shown below. CH3 r u n 2 O ^ ^ N ^ O ^ O - ^ -H NV ^ N > " C H 3 H T H: (LXXIII) CH, i 30^ / N \ ^ ° CH I 3 N \ N \ .^0 H 3 C " N NH CH_ I 3 H0H2C. / N ^ ^ N \ ^ j O N' -CH + CH 20 + (LXXIV) - I l l -In f a c t , when a concentrated s o l u t i o n of TML was oxidized by more than equimolar quantities of KMn^^CyDTA while oxygen gas was bubbled through the s o l u t i o n the alcohol LXXIV and the 7-oxolumazine were produced, the r a t i o of the products being approximately 1:1 (see Experimental). If the enzyme functions as an electr o n - t r a n s f e r oxidant, the mechanism outlined i n Scheme 10 might also hold f o r the enzymic conversion of IV to V. (TML and KMn***CyDTA i n Scheme 10 are replaced by IV and the enzyme.) - 112 -4. EXPERIMENTAL 4.1 Apparatus and Compounds  Apparatus The pieces of apparatus used in the present study were as follows: Melting points: Buchi melting-point apparatus; temperatures are uncorrected. Ultraviolet and visible spectra: Bausch and Lomb Model 502, Cary 15 and Cary 16 spectrophotometers. A l l machines are equipped with thermostated c e l l compartments and the spectra were recorded at 25 + 0.2° unless otherwise stated. Infrared spectra; Perkin-Elmer Model 457 grating infrared spectro-photometer u t i l i z i n g potassium bromide pellets.technique. Nuclear magnetic resonance spectra: Varian A-60 60 megahertz spectrometer. Mass spectra: Atlas CH-4 and A.E.I. MS 902 mass spectrometers. Vapour-phase chromatogram: Varian Aerograph Model 90-P. pH: Radiometer Model 26 pH meter. Carbon, hydrogen and nitrogen analyses of compounds were performed by P. Borda, Microanalysis Laboratory, University of British Columbia. - 113 -Compounds The lumazine derivatives were synthesized by the five-step procedure 81 used by McAndless, as i l l u s t r a t e d below. Cl . • N — C 1 NaOH C 1 -CH, • N V - ° H C H 3 N H 2 H N < N^ OH Cl OH OH NaNO, CH 3C0 2H R . ^0 CH, CH, HN -VN. OH „ n HN «H „ /X, N a 2 S 2 ° ^ ON OH R' 0^ H 2N N OH (LXII) H O \ ^ - H 3 C ^ 0 (IV, R = CH 3) (LXIII, R = C 6H 5) (V) - 114 -Because 4-methylamino-5-amino-2,6-dihydroxypyrimidine (LXII) and 81 i t s b i s u l f i t e s a l t decompose somewhat during r e c r y s t a l l i z a t i o n , McAndless used i t without i s o l a t i o n f o r the synthesis of the lumazine d e r i v a t i v e s . The l a t t e r compounds were then p u r i f i e d by column chromatography. However, column chromatography of the lumazines i s d i f f i c u l t because of t h e i r low s o l u b i l i t y i n most solvents. In the present study, the b i s u l f i t e s a l t of LXII was su c c e s s f u l l y r e c r y s t a l l i z e d from water containing a small amount of sodium d i t h i o n i t e . The lumazines obtained by using the p u r i f i e d b i s u l f i t e s a l t were pure enough ( a f t e r the usual r e c r y s t a l l i z a t i o n ) f o r the oxidation study. Their elemental a n a l y t i c a l data and melting points are as follows. 6,7,8-Trimethyllumazine (IV, R = C E ^ ) : m.p. 305-308° (decomposition); 124 l i t . , 308-310°. Elemental analysis. Found: C, 52.56; H, 4.92; N, 27.02. Calcd. for C QH i r iN.0„: C, 52.42; H, 4.89; N, 27.17. 6.7- Diphenyl-8-methyllumazine (LXIII, R = C,H_): m.p. 299-303° D D 34 (decomposition); l i t . , 288-292°. Elemental analysis. Found: C, 68.80; H, 4.53; N, 17.01. Calcd. f o r C^H^N^Cy C, 69.08; H, 4.27; N, 16.96. 6.8- Dimethyl-7-oxolumazine (V): m.p. above 350° (decomposition); l i t . , 4 5 358°. Elemental analysis. Found: C, 46.1; H, 4.02; N, 27.15. Calcd. for CoH_N.0.: C, 46.16; H, 3.87; N, 26.9. O o 4 j Potassium trans-1,2-diaminocyclohexanetetraacetatomanganate (III) III 99 (KMn CyDTA) was obtained by the method used by Hamm and Suwyn. Melting point 199-201° (decomposition); l i t . , 1 0 1 201-203°. Elemental analysis. Found: C, 36.92; H, 4.5; N, 5.98. Calcd. for KMnC. .H J 0 'H,0: C, 37.0; H, 4.4; N, 6.17. - 115 -4.2 Preparation of Buffer Solutions Since the oxidation reaction e i t h e r generates or consumes hydroxide ion depending upon the conditions and i s , i n any case, pH dependent, the use of buffer i s necessary. In the present work phosphate buffers were used i n a l l but strongly a c i d i c (pH < 0.5) systems. Phosphate has the advantage, f i r s t , of g iving buffers with a wide pH range and 125 second, of delaying s i g n i f i c a n t l y the p r e c i p i t a t i o n of i n s o l u b l e manganese dioxide, the usual product of permanganate reduction. The second reason i s e s p e c i a l l y important f o r the present study, since a l l k i n e t i c data were obtained spectrophotometrically and interference by p r e c i p i t a t e s must be avoided. Under the conditions used the p r e c i p i t a t i o n of manganese dioxide was s u f f i c i e n t l y slow that i t did not s i g n i f i c a n t l y disturb the k i n e t i c treatment. The buffer solutions, which va r i e d i n pH and i o n i c strength, were prepared i n the following manner. An accurately weighed amount of sodium phosphate (dibasic) was taken into a 100 ml beaker, the amount being chosen to give the desired phosphate concentration when d i l u t e d to a f i n a l volume of 100 ml. 126 The i o n i c strength of the buffers i s c a l c u l a b l e i n the usual manner and the value of a given buffer s o l u t i o n f i x e d at the desired value by the addition of the appropriate amount of sodium chloride or other s a l t to the beaker. Approximately 90 ml of water was then added i n t o the beaker and the desired pH was obtained by addition of 0.1 N or 1 N hydrochloric a c i d dropwise with s t i r r i n g as indicated by the pH meter. The contents of the beaker were then transferred q u a n t i t a t i v e l y to a 100 ml volumetric f l a s k and made up to the mark with water washings of the beaker. A 5 ml a l i q u o t was removed for pH measurement and the remainder - 116 -stored i n a container under nitrogen atmosphere. With t h i s means of storage, the pH was maintained over a period of weeks. However, i n the present work fresh buffer solutions were prepared a f t e r a week of storage. The water used i n the preparation of the buffer solutions and stock sample solutions was r e d i s t i l l e d from basic permanganate through a 14" Vigreaux column. Commercial a n a l y t i c a l grades of sodium phosphate and sodium chloride were used without further p u r i f i c a t i o n . 4.3 K i n e t i c Procedure  General As i l l u s t r a t e d i n Figure 18, the wavelength maxima of neutral 6,7,8-trimethyllumazine (TML) and 6,7-diphenyl-8-methyllumazine are at 404 nm and 426 nm re s p e c t i v e l y . On the other hand, permanganate has a minimum at 416 nm and even when a 50-fold excess of the s t o i c h i o -metric concentration of permanganate i s present the absorbance at t h i s wavelength i s r e l a t i v e l y small. The k i n e t i c data were therefore obtained by observing the disappearance of the lumazine absorption at 416 nm. These spectrophotometric measurements were made by using a Bausch and Lomb 505 spectrophotometer except for some very slow reactions i n which a Cary 16 spectrophotometer was used. In a l l of the k i n e t i c experiments the pseudo f i r s t - o r d e r technique was used, the i n i t i a l oxidant concentration being i n at l e a s t 12-fold excess. Stock solutions of TML and potassium permanganate were prepared by weighing accurately the appropriate amount of material into a 10 ml volumetric f l a s k and making up to the mark with d i s t i l l e d water. The Figure 18. UV-Visible spectra of 6,7,8-trimethyllumazine, 6,7-diphenyl-8-methyllumazine and permanganate ion. — 6,7,8-trimethyllumazine (5 x 10 M); -4 , potassium permanganate (2.6 x 10 M) . -, 6,7-diphenyl-8-methyllumazine (5 x 10 5 M); 0.6 J 0.5 A ai a C ca .o u o CO n) o 0.4 H 0.2 J 0.1 J 450 wavelength (nm) 350 i - 118 -amount of TML in the stock solution was chosen to give a concentration close to 5 x 10 5 mole/1, (this gives an i n i t i a l absorbance of approximately 0.6) when 25 y l . of the stock solution was diluted into 2.5 ml of a buffer solution. On the other hand, the amount of permanganate was chosen to give a 12- or 60-fold excess of the stoichiometric concentration when 10 or 50 y l of the stock solutions were added to 2.5 ml of buffer. Since the solubility of 6,7-diphenyl-8-methyllumazine in water is very low, i t was not possible to make as concentrated a stock solution in this case as with the TML or permanganate. Therefore stock solutions of 6,7-diphenyl-8-methyllumazine were prepared by simply adding an accurately weighed sample of solid material to a known volume of buffer solution, the amount being chosen to give a concentration of the stock solution close to 5 x 10 5 mole/1, and an i n i t i a l absorbance of approximately 0.65. The stock solutions of permanganate and the lumazines were stable over several days but, in general, fesh solutions were prepared daily. A typical kinetic run was carried out as follows: buffer solution (2.5 ml) and TML stock solution (25 yl) were added to an optical c e l l by means of syringes equipped with Chaney adapters. In the case of 6,7-diphenyl-8-methyllumazine 2.5 ml of the stock solution were taken in the c e l l . To the lumazine solution thus obtained, an appropriate volume (usually 10-50 yl) of permanganate stock solution was added by means of a microsyringe. The c e l l was stoppered with a Teflon stopper, thoroughly mixed by shaking, placed in the c e l l compartment of the spectrophotometer, and the absorbance at 416 nm was followed with respect to time. The c e l l and a l l reaction components were thermostated at the - 119 -desired temperature p r i o r to use. It was found that the k i n e t i c r e s u l t s of the permanganate oxidation of TML and 6,7-diphenyl-8-methyllumazine were unaffected by conducting the re a c t i o n under a nitrogen or oxygen atmosphere instead of under a i r . Therefore, most k i n e t i c runs were c a r r i e d out under a i r . The k i n e t i c studies of the TML oxidation with KMn I I ICyDTA were ca r r i e d out e s s e n t i a l l y i n the same manner as i n the permanganate oxidation of TML. In t h i s case, the course of the reaction depends upon the atmospheric conditions. Therefore, for those runs conducted under a i r oxygen gas was bubbled through the s o l u t i o n f o r 10 minutes p r i o r to use, i n order to make sure oxygen was present i n excess. The -3 127 s o l u b i l i t y of oxygen gas i n water at 25° i s about 5 x 10 mole/1., whereas the k i n e t i c concentration of TML i s 5 x 10 5 mole/1. K i n e t i c Isotope E f f e c t The methyl group at the 7-position of 6,7,8-trimethyllumazine (TML) undergoes f a c i l e deuterium exchange. Thus, TML deuterated at the 7-methyl group was produced i n the following way. An accurately weighed amount of TML was dissolved i n a known volume pf deuterium oxide (99.5% pure) buffered by an appropriate amount of sodium phosphate. The s o l u t i o n was allowed to stand for 7 to 10 h a l f - l i v e s of the exchange reaction, as estimated from the 81 exchange rate. The deuterated compound thus formed was used without being i s o l a t e d . P r i o r to the k i n e t i c study, the NMR spectrum was run on an a l i q u o t of the s o l u t i o n to make sure that the protium compound was completely converted to the deuterium compound. The oxidation reaction was started by adding a small volume of - 120 -this solution (25 yl) to a much larger volume of the permanganate in buffered 1^0 (2.5 ml). The change of the TML absorbance at 416 nm was then recorded with respect to time. A l l the kinetic data were analyzed by the method of least squares by arranging the data in a linear form of y = ax + b. The least squares analysis and the subsequent plotting were performed by a Hewlett-Packard calculator and plotter, model 9125A. 4.4 Stoichiometry and Product Analysis 4.4.1 Permanganate Oxidation of 6,7,8-Trimethyllumazine 4.4.1.1 At pH 6.0 By means of a spectrophotometry analysis i t was found that one mole of 6,7,8-trimethyllumazine (TML) was completely consumed when 4/3 moles of permanganate were used at pH 6.0 (^£^0^ = 0.1 M). The UV spectrum of the solution after completion of the reaction was identical with that of an authentic sample of 6,8-dimethyl-7-oxolumazine. The calculation based on the i n i t i a l concentration of TML and the known extinction coefficient at 346 nm of the 7-oxolumazine revealed that TML was converted to the 7-oxolumazine in nearly quantitative yield ( 95%). Other organic product was determined in the following way. TML (20 mg, 0.1 mmoles) was dissolved in 10 ml of phosphate buffer (pH = 6.0, Na2HP04 = 0.1 M). To this was added the permanganate solution (20 mg, 0.13 mmoles in 5 ml of the same buffer). The reaction mixture was allowed to stand at room temperature for 10 minutes with sti r r i n g , and then any deposited manganese dioxide was f i l t e r e d off. - 121 -The f i l t r a t e was added to a freshly prepared solution of 1,2-dianilino-ethane (30 mg) in 10 ml of 40% aqueous acetic acid. A copious white precipitate immediately separated out. This was collected, dried in a vacuum desiccator and then weighed (16 mg). Recrystallization of the crude product from hot methanol gave white needles whose melting point is 125-126°. The melting point indicates this compound to be the condensation product of formaldehyde with 1,2-dianilinoethane (m.p. in 129 the literature is 126° ). This is confirmed by the result of the elemental analysis. Found: C, 80.59; H, 7.16; N, 12.30%. Calcd. for based on the crude condensation product was 73%. 4.4.1.2 At pH 1.0 With Stoichiometric Amount of Permanganate A solution of an accurately weighed amount of TML (500 mg, 2.43 mmole) in 30 ml of phosphate buffer (pH = 1.0, Na^PO^ = 0.1 M) was titrated by 0.05 M permanganate solution containing the same buffer. In order to avoid involvement of the permanganate-dependent path, each 1,2-Dianilinoethane is used to determine total aldehyde, since i t reacts with most aldehydes, but not with ketones, to give condensation products!28 : C 1 5H 1 6N 2: C, 80.31; H, 7.20; N, 12.49%. The yield of formaldehyde H„C-NHC,H, + OHCR The colourless condensation products are normally insoluble in water and separate in a very pure form. Their characteristic melting points can serve to identify the aldehydes. - 122 -drop was added with vigorous s t i r r i n g and only a f t e r the pink colour of the preceding drop had completely disappeared. The t i t r a t i o n was stopped when the pink colour of permanganate was retained longer than 15 seconds. C a l c u l a t i o n of the uptake of permanganate showed that approximately 4/5 of a mole of permanganate were consumed for one mole of TML. The colour of the reaction mixture at the end point was pale yellow and no deposit of manganese dioxide was observed. When the s o l u t i o n was made a l k a l i n e , a white p r e c i p i t a t e , presumably manganese hydroxide, separated out. This indicates a l l of the permanganate was reduced to the +2 stage under the present conditions. The reaction mixture was divided into two aliquots and one of them was added to a f r e s h l y prepared s o l u t i o n of 600 mg (3 mmoles) of 2,4-dinitrophenylhydrazine i n 3 ml of 50% s u l f u r i c acid. The s o l u t i o n was then heated on the steam bath for 10 minutes. A f t e r cooling the s o l u t i o n , an orange coloured p r e c i p i t a t e was c o l l e c t e d , dried i n a vacuum desiccator, and then weighed (240 mg). The crude 2,4-dinitrophenyl-hydrazone was r e c r y s t a l l i z e d from nitrobenzene to give reddish-orange coloured needles whose m.p. i s 255-265° (decomposition). Elemental analysis of t h i s compound gave C, 41.44; H, 3.05; N, 24.14%, leading to an empirical formula of C1(,H. .0-N_ (calculated values are C, 41.56; H, 3.05; N, 24.24%). Melting point and elemental analysis showed the compound to be the di-2,4-dinitrophenylhydrazone of the hydroxybiacetyl 130 (LVIII) (m.p. of t h i s hydrazone i n the l i t e r a t u r e i s 235° (decomposition)). The y i e l d of hydroxybiacetyl based on the crude phenylhydrazone was 44%. 0 0 II 11 HOH CC-C-CH (LVIII) - 123 -The other aliquot of the reaction mixture was made basic by the addition of sodium bicarbonate powder. After standing for one hour the precipitate of white manganese hydroxide was fil t e r e d out and the f i l t r a t e was made slightly acidic (pH 5^6) by addition of dilute sulfuric acid. The solution was then evaporated to dryness by means of a rotary * evaporator. To the residue was added methanol (30 ml) and the mixture was then boiled on a steam bath and fi l t e r e d . This was repeated two more times. The methanol extracts were combined and evaporated to dryness on a rotary evaporator. The residue was dried in a vacuum desiccator and weighed (220 mg). Two recrystallizations of the crude product from methanol-ether solution, with a charcoal treatment included in the f i r s t recrystallization, gave 30 mg of a white granular solid. M.p. 149-151° (decomposition). The observed elemental analysis of this compound i s C, 31.38; H, 4.83; N, 21.95%, leading to the empirical formula C 5H 9N 30 5, whose exact calculated values are C, 31.42; H, 4.74; N,.21.99%. The NMR spectrum of this compound in dry DMSO-dg consists of a singlet at 6 = 2.52 p.p.m. and a broad peak centered at 6 = 7.90 p.p.m. (from tetramethylsilane as an external reference. Figure 19). The ratio of the integral value of the former and the latter i s 1 to 2. This ratio corresponds to three and six protons respectively, since there are nine protons altogether, as shown by the empirical formula. When a few drops of D20 were added to the DMSO-d^  solution, the NMR peak at lower f i e l d Removal of manganese ion with the procedure described above was necessary because when the solution was evaporated to dryness without taking such action a red coloured residue, which i s very d i f f i c u l t to purify, was l e f t . This compound might be a chelate compound of the product with manganese ion, but because of d i f f i c u l t i e s with purification this point was not pursued. - 124 -disappeared completely, with the one at higher field remaining unchanged, indicating that six protons are attached to nitrogen or oxygen atoms and three protons to a carbon atom. The IR spectrum (Figure 20) shows a very wide absorption between 2500 and 3500 cm suggesting the presence of a strong hydrogen-bond. Two absorptions in the carbonyl region which are also found in 2,4-dioxo-6-methylaminopyrimidine (LXXV) indicate the uracil structure (2,4-dioxopyrimidine) of TML to be conserved in the compound. (LXXVI) The results obtained from the elemental analysis, NMR and IR spectra suggest that the compound may be 2,4-dioxo-5,5,6-trihydroxy-6-methylaminopyrimidine (LVII). The UV spectrum is also compatible with H.CHN H (LVII-a) H the proposed structure, since the spectrum does not differ greatly from L u a L o l 5 , u - u i l i ^ u t u - o - u y u r o x y - l , 3 - d i m e L u y l - 2 > H - d i o x o p y r i ' a J . d i n e (LXXVI) 131 reported by Moore and Thomson (Figure 21). Carbinolamines and gem_diols a r e usually not very stable. However in the case of LVII, the carbonyl group at the 4-position will stabilize the gem-diol and i t would also be expected that the carbinol-amine and gem-diol stabilize each other by hydrogen-bonding (as Figure 19. NMR spectrum of the product of TML oxidation ( i n DMSO-d , TMS as an external reference). <5 (p.p.m.) - 127 -- 128 -indi c a t e d by the IR spectrum). An a l t e r n a t i v e structure for the product i s LVII-a, which appears to be a more stable s t r u c t u r a l isomer of LVII and may be i n equilibrium with i t . Mass spectroscopy was not h e l p f u l i n determining the structure of the compound since reproducible values f o r the parent and fragment peaks, could not be obtained. The y i e l d s of LVII based on the crude and pure compounds were found to be 90% and 13%. I t was found that the crude compound i s contaminated with a large quantity of inorganic ma t e r i a l . Ammonia was also detected as a product by using Nessler reagent. With Excess of Permanganate On the basis of the data i n Table 4, i f one uses a 100-fold excess of permanganate over TML at pH 1.0 i n sodium phosphate b u f f e r (Na^HPO^ = 0.1 M), about 83% of the t o t a l r e a c t i o n goes v i a the permanganate-dependent path and 17% v i a the permanganate-independent path. The oxidation was c a r r i e d out under such conditions - 206 mg (1 mmole) of TML i n 50 ml of the b u f f e r plus 15.8 g (100 mmole) of permanganate i n 100 ml of the b u f f e r . A f t e r one minute (about 5 h a l f - l i v e s ) the excess of permanganate was quenched by sodium b i s u l f i t e . The subsequent work-up, conducted as before, gave (LVII), the y i e l d of which was 160 mg (84%), based on the crude compound. Attempts to i s o l a t e other l i k e l y products, uyJroxybiacctyl cr.i b i a c e t y l , by 2,4-dinii-rophenyl-hydrazine treatment f a i l e d . By using g a s - l i q u i d chromatography, a c e t i c a c i d and some u n i d e n t i f i e d compounds were detected. Ammonia was also detected by - 129 -Nessler reagent. 4.4.2 Permanganate Oxidation of 6,7-Diphenyl-8-methyllumazine  With Stoichiometric Amount of Permanganate By means of spectrophotometric analysis i t was found that one mole of the lumazine was consumed completely by approximately 2/5 moles of permanganate at pH 1.0. The products were isolated as follows. To the suspension of the lumazine (200 mg, 0.55 mmoles) in 100 ml of the buffer solution (pH = 1.0, ^ £ ^ 0 ^ = 0.1 M) was added dropwise the stoichiometric amount of permanganate in the same buffer (35 mg, 0.22 mmoles in 30 ml). In order to avoid involvement of the permanganate-dependent path, each drop was added after the colour caused by the preced-ing drop had completely consumed. After the addition was over, the colour of the solution was a clear pale yellow with no indication of the presence of a manganese dioxide precipitate. The solution was made slightly basic (pH ^ 9) by addition of sodium bicarbonate powder and the precipitated manganese hydroxide was f i l t e r e d out. The f i l t r a t e was then extracted with ether three times (3 x 50 ml). The ethereal solutions were combined, washed with water, and dried over anhydrous sodium sulfate. Ether was then removed by means of a rotary evaporator and the residue was dried in a vacuum desiccator and weighed (76 mg). Recrystallization of the crude compound from ethanol-water gave white needles. This was identified as benzil by comparison of the m.p. and the NMR spectrum with those of an authentic sample. The yield of benzil on the basis of the crude compound was 87%. The aqueous layer was worked up in the same way as the TML oxidation, giving 2,4-dioxo-5,5,6-trihydroxy-6-methylaminopyrimidine (LVII), the - 130 -yield of which based on the crude compound was 90 mg (85%). Ammonia was also detected by Nessler reagent. With Excess of Permanganate On the basis of the data in Table 9, i f one uses a 100-fold excess of permanganate at pH 1.0, 94% of the total reaction goes via the permanganate-dependent path and 6% via the permanganate-independent path. The oxidation was carried out under such conditions - 200 mg (0.56 mmoles) of the lumazine in 100 ml of buffer (pH = 1.0, Na^PO^ = 0.1 M) plus 8.8 g (56 mmoles) of permanganate in 30 ml of the same buffer. After 20 minutes sodium bi s u l f i t e was added to the reaction mixture to quench the excess permanganate. The same work-up as before gave benzil and 2,4-dioxo-5,5,6-trihydroxy-6-methylaminopyrimidine (LVII), yields of which were 75% and 79%, respectively, on the basis of crude product. 4.4.3 Permanganate Oxidation of 2>4-Dioxo-5-amino-6-methylamino- pyrimidine Permanganate in phosphate buffer (pH 1.0, ^ 2 ^ 0 ^ = 0.1 M) was added dropwise to the solution of the b i s u l f i t e salt of 2,4-dioxo-5-amino-6-methylaminopyrimidine (LXII, 500 mg) in 20 ml of the same buffer. The addition was stopped when the pink colour of the preceding drop of permanganate was retained longer than 10 seconds. Work-up of the reaction mixture i n the manner analogous to that conducted i n the case of 2,4-dioxo-5,5,6-trihydroxy-6-methylaminopyrimidine (LVII) gave a white granular (350 mg) whose melting point, NMR spectra were identical with those of LVII. - 131 -4.4.4 Oxidation of 6,7,8-Trlmethy1lumazine with potassium trans-1,2-diaminocyclohexanetetraacetatomanganate (III) (KMn***CyDTA) Under Nitrogen Atmosphere 6,7,8-Trimethyllumazine (TML) (20 mg, 0.1 mmole) in 10 ml of an aqueous buffer (pH = 6.0, Wa^RPO^ = 0.1 M) was placed in a two-neck flask. Each neck was f i t t e d with a rubber cap through which a syringe needle was inserted. High-purity nitrogen was passed through one of the needles, which was dipped below the surface of the solution. A nitrogen-flushed solution of KMn***CyDTA (45 mg, 0.1 mmole or 90 mg, 0.2 mmole) in 5 ml of the same buffer was then added in one of two ways; (1) dropwise, when equimolar quantities of KMn**'''CyDTA and TML were used, (2) a l l at once, when two equivalents of KMn***CyDTA were used. After 3 hours, a yellow precipitate was collected by f i l t r a t i o n , washed thoroughly with d i s t i l l e d water, dried i n a vacuum desiccator and weighed. In a l l cases the amount of the product was 17 mg or more. It was found from the mass spectra of the products that the reaction produced two kinds of product, depending upon the ratio of the oxidant and the substrate. Both compounds are extremely insoluble in a l l common solvents and possess high melting points (decomposition), making i t d i f f i c u l t to obtain good analytical results. Although fragment peaks of the mass spectra were not quite reproducible, the latter could s t i l l provide information regarding the structures of these compounds. The melting points and the parent mass peaks of these compounds are shown in Table 10. - 132 -Table 10. Melting Points and Parent Mass Peaks of the Compounds III Obtained from the Oxidation of TML with KMn CyDTA under Nitrogen Atmosphere 1 TML:2 KMn mCyDTA 1 TML:1 KMn mCyDTA Melting point Parent mass peak 313-315° (decomposition) 408 304-306° (decomposition) 205 A large peak at mass number 217 was also observed. The observed mass-spectrum peaks of the compound obtained when two-equivalent quantities of the oxidant are used can be assigned to the following species. m/e 408 CH„ I 3 0 CH=CH ^3 N \ ^ N CH 3 H3C Y NH (XLVIII) From these peaks i t seems reasonable to conclude that XLVIII i s produced under these conditions. This f a c t suggests that the product when equimolar quantities of the oxidant and the substrate are used might be the "dimer" of TML (XLVII), since i t i s reasonable to assume that the compound XLVIII i s produced by further oxidation of the compound XLVII with an excess of the oxidant. - 133 -m/e 410 m/e 205 0 0 (XLVII) (LXXVII) However, no peak attributable to the species XLVII was observed in the product obtained when equimolar quantities of oxidant and substrate were used. Instead, a peak of 205 which can be assigned to the species LXXVII was observed as the parent peak. There is l i t t l e doubt that the species LXXVII is produced by the splitting of the parent compound XLVII at the middle of the bridging ethyl group. Unfortuantely, satisfactory recrystallizations of XLVII and XLVIII could not be carried out because of their very low solubilities in a l l solvents that were tried. For this reason the elemental analyses that were conducted on these compounds were not reproducible. Under Oxygen Atmosphere -4 To an optical cell 2.5 ml of 1.5 x 10 mole/1, of an aqueous solution of TML (buffered at pH 6.0, Na^PO^ = 0.1 M) was added. Oxygen gas was bubbled through the solution and a small volume of aqueous KMn CyDTA (10 ul) w a s introduced, to give a final concentration of -4 about 3 x 10 mole/1. The decrease in absorbance at 404 nm of TML and the increase in the absorbance at 360 nm and 284 nm of 6,8-dimethyl-7-oxolumazine were observed. After two hours the TML peak at 404 nm had decreased to one-twentieth of its i n i t i a l value. The conversion of - 134 -TML to the 7-oxolumazine was 81%. Addition of a few drops of aqueous a c e t i c ac i d s o l u t i o n of 1,2-dianilinoethane to the reaction mixture changed the s o l u t i o n from clear to cloudy, i n d i c a t i n g the presence of aldehyde. This aldehyde can be assigned to formaldehyde, the only conceivable aldehydic product i n the present case. Next the reaction was c a r r i e d out using a more concentrated s o l u t i o n of TML and KMnI]tICyDTA. TML (50 mg, 0.25 mmoles) i n 10 ml of phosphate buffer (pH 6.0, ^ 2 ^ 0 ^ = 0.1 M) was added i n one portion to 1 ml of aqueous KMn I I ICyDTA (110 mg,0.25 mmole) with oxygen gas bubbling through the s o l u t i o n and'-with magnetic s t i r r i n g . -Soon p r e c i p i t a t i o n of: a f i n e , yellow compound was observed. The reaction was allowed to continue for two hours with s t i r r i n g and with oxygen gas bubbling through. The reaction mixture was then q u a n t i t a t i v e l y transferred to a 25 ml volumetric f l a s k and made up to 25 ml by ad d i t i o n of d i s t i l l e d water. A f t e r decanting the s o l u t i o n 25 y l of the cleaf supernatant l i q u i d was withdrawn by means of a microsynringe, d i l u t e d to 2.5 ml, and the UV spectrum was recorded. The UV spectrum showed 95% of TML had been consumed, with 45% of i t being converted to the 7-oxolumazine. The mixture i n the volumetric f l a s k was then f i l t e r e d through a sintered-glass f i l t e r and the yellow product l e f t on the f i l t e r . The f i l t r a t e was treated with a 1,2-dianilinoethane s o l u t i o n (50 mg i n 10 ml of 40% aqueous ac e t i c a c i d ) , and a white p r e c i p i t a t e i d e n t i f i e d by m.p. and elemental analysis as the condensation product of formaldehyde with 1,2-dianilinoethane was c o l l e c t e d ( y i e l d 12 mg). The yellow product (23 mg) was dissolved i n dimethyl sulfoxide, the only solvent found able to dissolve t h i s compound and f i l t e r e d . Addition of d i s t i l l e d water to the f i l t r a t e brought about the p r e c i p i t a t i o n - 135 -of f i n e yellow material which was f i l t e r e d out and thoroughly washed with d i s t i l l e d water. The a n a l y t i c a l data of the compound are as follows: Melting point, 316-320° (decomposition). Elemental analysis (found): C, 48.43; H, 4.66; N, 24.89%. Major mass spectrum peaks: 221(P), 205, 191. The observed mass peaks can be assigned to the following species. + m/e 221 CH„ H0=HCL / N ^ ^ N v . ^0 H 2C T T m/e 205 ? H3 N NH H3C y y m/e 191 CH. H 3C NH 0 These fragment-ipeaks strongly suggest the parent compound i s the alcohol LXXIV shown below. f 3 H O H ^ ^ / N x . N ^ ^0 Y (LXXIV) NH It i s not s u r p r i s i n g that the mass peak due to species LXXIV was not observed i n view of the fa c t that some alcohols, e s p e c i a l l y a l l y l i c alcohols, undergo a-cleavage to form a stable fragment which appears 132 to be a parent peak (see below). H ^ -H + R-CH=CH-C-0H — > RCH=CH-CH=OH I H - 136 -The calculated values of the elements of compound LXXIV are; C, 48.62; H, 4.55; N, 25.23%, which agree well with the observed elemental a n a l y s i s . - 137 -APPENDIX POSITION OF METHYLATION OF 2,3-DIAMINOPYRIDINE AND 3-AMINO-2-METHYLAMINOPYRIDINE 1. Introduction R i b o f l a v i n ( I I I , R = r i b i t y l ) and other f l a v i n nucleotides occupy a c e n t r a l p o s i t i o n i n the reaction chain that i s responsible for oxidation i n l i v i n g systems and model compounds f o r f l a v i n nucleotides 133 13 A have been re c e i v i n g increasing a t t e n t i o n . ' 6,7,8-Trimethyllumazine discussed i n the preceding chapters may be regarded as one such model. Add i t i o n a l models are 8-azaflavin (LXXVIII) and i t s quarternary s a l t s (LXXIX). With these i n mind the synthesis of the l a t t e r compounds were attempted. R R R' R (III) (LXXVIII) (LXXIX) A standard method to obtain r i b o f l a v i n (III) i s condensation of l-amino-2-ribitylamino-4,5-dimethylbenzen (LXXX, R = r i b i t y l ) with 135 alloxan (LXXXI). ( I l l ) (LXXX) - 138 -Accordingly, for the synthesis of LXXIX (R,R* = CU^), 3-amino-2-methylamino-l-methylpyridinium (LXXXII) was required. Methylation of 3-amino-2-methylaminopyridine (LXXXIII) appeared the obvious route to 136 LXXXII since both 3-aminopyridine (LXXXIV) and 2-methylaminopyridine 137 (LXXXV) alkylate at the ring nitrogen. It was found, however, that LXXXIII upon treatment with methyl iodide gave exclusively the 3-amino alkylation product (LXXXVI). \ N ^ N H C H 3 + I CH„ \ „<s^ NHCH-N 3 N 4? (LXXXII) (LXXXIII) (LXXXIV) N NHCH .NHCH, NHCH, \ -^^NH N INn2 (LXXXV) (LXXXVI) (LXXXVII) Other anomalies have been noted previously with regard to the 138 position of alkylation of aminopyridines. For example, although 2-aminopyridine, 2-methylaminopyridine, 3-dimethylaminopyridine and 4-dimethylaminopyridine a l l give ring alkylation, 2-dimethylaminopyridine 139 140 gives amino alkylation. ' In view of these anomalies and because l i t t l e is known about the position of alkylation of polyaminopyridines, i t was decided to investigate more fully the reaction of 2,3-diamino-pyridine (LXXXVII) and 3-amino-2-methylaminopyridine (LXXXIII) with - 139 -methyl iodide. 2. Results The product of methylation of compound LXXXIII was shown to be the hydriodide of LXXXVI by an unambiguous synthesis of the latter compound starting with 3-amino-2-chloropyridine (LXXXVTII), as described below. • N ^ C l (LXXXVIII) ClS0oC,HcCHo 2 6 5 3 NHSO_e,H(.CH„ • £ o J J (LXXXIX) <CH3>2S04 Na2C03 CH3 . N S O . C . H - C H . 2 6 5 3 (XC) NHCH„ (XCI) C H3 N H2 CuSO, NHCH, ^N^^NHCH, HI N H C H 3 Q N NHCH„ HI (LXXXVI) - 140 -The NMR spectrum of the r i n g methylated compound (LXXXII) synthesized by a d i f f e r e n t route (see below) showed the quarternary methyl group resonance at 6 = 4.0 p.p.m. No trace of such absorption was detected i n the concentrated mother l i q u o r remaining a f t e r the removal of the hydriodide of LXXXVI. Thus the amount of r i n g methylation of LXXXIII must be extremely small. 2 (CH 3CO) 20 NHCOCH NHCOCIL 3 CH 3I ^ N ^ C l + I I" CEL (LXXXVIII) (XCII) (XCIII) H CH 3NH 2 (XCIV) (LXXXII) When the reaction of 2,3-diaminopyridine (LXXXVII) with methyl iodide was c a r r i e d out without solvent and the NMR spectrum of the residue a f t e r evaporation of excess methyl iodide was determined i n D 20, two methyl peaks were observed at 6 = 3.87 and 2.85 p.p.m. These two peaks were assigned to the methyl groups of the compounds XCV and the hydriodide of XCVI by unambiguous syntheses of these compounds (see below) and subsequent determination of t h e i r NMR spectra. - 141 -. NH, _ + I CH, NH, 3 (XCIV) (XCV) NHCH, ,NHCH, .NHCH, NH_/CuSO. 3 4 HI (XCI) (XCVI) ^ N ^ N H , Product distribution between ring methylation (XCV) and 3-N-methylation (XCVI) was found to be solvent-dependent, polar aprotic solvents favouring ring methylation more than hydrogen-bonding solvents. The figures in parentheses in the following l i s t gives the ratios (x:l) of ring methylation to 3-N-methylation at room temperature: acetonitrile (7.3), acetone (3.8), tetrahydrothiophendioxide (3.0), ethanol (2.3), methanol (2.2), no solvent (1.9), 1:1 methanol-phenol (1.5), 4:1 2,2,2-trifluoroethanol-methanol (1.1). Approximate reaction times were 4 hr for the first five systems and 2-3 days for the last three. Refluxing the reaction mixtures reduced the amount of ring methylation in the case of the trifluoroethanol-methanol system. In a l l solvents studied, the total yield of monomethylation product was nearly quantitative and no absorbance at 6 = 3.13 p.p.m. due to the methyl group of the hydriodide of LXXXIII was observed. The latter indicates that l i t t l e , i f any, 2-N-methylation had occurred. - 142 -3. Discussion The r e s u l t s can be explained by assuming that the n u c l e o p h i l i c i t i e s of the various centers (as indicated by t h e i r equilibrium a c i d i t i e s ) are modified by two f a c t o r s : s t e r i c hindrance and hydrogen bonding.*4*'"''4 I f n u c l e o p h i l i c i t y were d i r e c t l y proportional to base strength, r i n g methylation would be expected to predominate over amino-methylation i n 143 a l l cases. 2-Dimethylaminopyridine, but not 2-methylaminopyridine i s methylated at the amino group rather than at the r i n g nitrogen atom 140 and Frampton, Johnson and Katr i t z k y have ascribed t h i s to a combination of s t e r i c and e l e c t r o n i c f a c t o r s . The forced co-planarity of the N-methyl groups and the r i n g that r e s u l t s from the conjugative i n t e r a c t i o n between r i n g and side-chain nitrogen atoms screens the r i n g nitrogen atom from attack by methyl iodide. The present r e s u l t s show that a 3-amino-group greatly reduces the s u s c e p t i b i l i t y of the r i n g nitrogen atom to s u b s t i t u t i o n even when the 2-amino group i s unmethylated or monomethylated. This can be explained by hydrogen-bonding i n t e r a c t i o n of the type i l l u s t r a t e d below which locks the 2-N-methyl group i n 3-amino-2-methylaminopyridine (LXXXIII) i n a p o s i t i o n where the r i n g nitrogen atom i s e f f e c t i v e l y screened. A s i m i l a r but smaller e f f e c t presumably applies to 2,3-diaminopyridine (LXXXVII). H ;.H \ / (LXXXIII, Z = CH ) (LXXXVII, Z = H) Z - 143 -Substitution occurs at the 3-position even though such hydrogen-bonding would be expected to impair the n u c l e o p h i l i c i t y of the 3-amino group. An a l t e r n a t i v e explanation i s that s u b s t i t u t i o n occurs v i a the minor imino-tautomer (XCVII), which has a f a c i l e pathway leading to a stable c a t i o n a v a i l a b l e to i t . Moreover, the 3-amino nitrogen of t h i s species i s now free from the hydrogen-bonding i l l u s t r a t e d above, and i t s n u c l e o p h i l i c i t y would be expected to be increased even more by hydrogen-bonding of the type i l l u s t r a t e d below. U C H 3 I \ H H , (XCVII) (40) An estimate of the amino-imino tautomeric constant f or 2,3-diaminopyridine (LXXXVII), K u t = [imino]/[amino], gives the value of —6 2.1 x 10 (see Experimental s e c t i o n ) , corresponding to an energy difference of 7.7 kcal.mole ^. K taut (LXXXVII) - 144 -The s t a b i l i z a t i o n of the t r a n s i t i o n state f o r reaction (40) compared to that f o r attack at r i n g nitrogen might reasonably be expected to be s u f f i c i e n t to balance t h i s energy difference. The minor solvent and temperature e f f e c t s observed i n LXXXVII might be due to changes i n the concentration of a reactive e n t i t y such as tautomer XCVII or simply to small and obscure changes i n the a c t i v i t y c o e f f i c i e n t s of substrate and activated complex. 4. Experimental A l l apparatus used i n t h i s study are the same as those used i n the pte r i d i n e s oxidation study.(page 112). Chemical s h i f t s of NMR spectra were measured from tetramethylsilane i n deuterated organic solvents or from sodium-2,2-dimethyl-2-silapentane-5-sulfonate (DSS) i n deuterium oxide as i n t e r n a l reference. Materials  Methyl iodide The compound purchased from Fisher S c i e n t i f i c Co. was washed successively with water, d i l u t e sodium carbonate, water, d i l u t e sodium t h i o s u l f a t e , and water. The s o l u t i o n was dried over calcium chl o r i d e , d i s t i l l e d , a f r a c t i o n (b.p. range between 42.7-43.3°) was c o l l e c t e d and stored over mercury. Solvents used for the methylation re a c t i o n Commercially a v a i l a b l e spectrogrades were used without further p u r i f i c a t i o n . - 145 -2,3-Diaminopyridine (LXXXVII) The compound purchased from A l d r i c h Chemical Co. was r e c r y s t a l l i z e d twice from benzene, with a charcoal treatment included i n the f i r s t r e c r y s t a l l i z a t i o n . M.p. 112-113° ( l i t . 1 1 3 - 1 1 4 ° 1 4 3 ) . 3-Amino-2-methylaminopyridine (LXXXIII) This compound was synthesized by the method used by Schickh, Binz 144 144 and Schulz. M.p. 99.8-101.5° ( l i t . 100-101° ). 2,3-Bismethylaminopyridine (LXXXVI) 3-Amino-2-chloropyridine (LXXXVIII) was converted to 2-chloro-3-methylaminopyridine (XCI) i n o v e r a l l y i e l d of 52% by three-step 145 procedure used by Clark-Lewis and Thompson, as i l l u s t r a t e d i n page 139. 2-Chloro-3-methylaminopyridine (XCI, 2 g) was then heated with aqueous 40% methylamine (15 ml) and copper s u l f a t e (0.5 g) i n a sealed tube at 160° for 20 hr. The reaction mixture was extracted with ether; the extract was evaporated and the r e s u l t i n g residue r e c r y s t a l l i z e d from benzene-petroleum ether (charcoal) to give white needles (1.1 g, 55%). M.p. 97.8-98.2°. Elemental analysis. Found: C, 61.1; H, 8.11; N, 30.78. Calculated f or c y : C, 61.25; H, 8.09; N, 30.65. The NMR spectrum of t h i s compound i n U^O consists of two s i n g l e t s at 6 = 2.71 (3H) and 2.87 (3H) p.p.m., and three quartets at 6 = 6.67 (IH), 6.90 (IH) and 7.5 (IH) p.p.m. The hydriodide of 2,3-bismethylaminopyridine (LXXXVI) was obtained as follows. LXXXVI was dissolved i n a d i l u t e hydrogen iodide. Evaporation and subsequent r e c r y s t a l l i z a t i o n of the r e s u l t i n g residue - 146 -from 2-propanol gave white needles. M.p. 265-270° (decomposition). Elemental analysis. Found: C, 31.66; H, 4.69; N, 16.07%. Calculated for C 7H 1 2N 3I: C, 31.71; H, 4.56; N, 15.85%. The NMR spectrum of this compound in BV^O consists of two singlets at 6 = 2.85 (3H) and 3.08 (3H) p.p.m. and three quartets at 6 = 6.83 (IH), 7.03 (IH) and 7.30 (IH) p. p. m. 3-Acetylamino-2-chloro-l-methylpyridinium iodide (XCIII) 3-Amino-2-chloropyridine (LXXXVIII) was converted to i t s 3-acetyl-144 amino derivative (XCII). A mixture of XCII (12 g) and methyl iodide (15 g) i n 20 ml of absolute methanol was then refluxed for 10 hr on a water-bath. The solution was evaporated to dryness and the resulting residue was recrystallized from 2-propanol with a small amount of ether to give white needles (12 g, 55%). M.p. 144-146° (decomposition). Elemental analysis. Found: C, 30.44; H, 3.05; N, 8.90%. Calculated for CQH..NoClI0: C, 30.72; H, 3.23; N, 8.97%. o 11) / 3-Amino-2-chloro-l-methylpyridinium iodide (XCIV) 3-Acetylamino-2-chloro-l-methylpyridinium iodide (XCIII, 12 g) in 30 ml of 1 N hydrogen chloride was warmed on a water-bath for 30 min, and then the solution was evaporated to dryness. The residue was recrystallized from ethanol to give white needles (9.8 g, 94%), m.p. 172-173° (decomposition). Elemental analysis. Found: C, 26.38; H, 2.76; N, 10.15%. Calculated for C,H0N0C1I: C, 26.62; H, 2.99; N, 10.36%. O o 2. - 147 -3-Amlno-l-methyl-2-methylaminopyridlnlum iodide (LXXXII) 3-Amino-2-chloro-l-methylpyridinium iodide (XCIV, 3 g), dissolved in water (10 ml) and aqueous 40% methylamine (15 ml), was warmed on a water-bath for 1 hr. The solution was then evaporated to dryness. Methylammonium iodide was removed by dissolving the residue i n aqueous 10% sodium carbonate, evaporating the solution to dryness, and then adding ethanol (25 ml). The suspension was shaken well and then fi l t e r e d . The f i l t r a t e was evaporated to dryness and>the residue was recrystallized twice from ethanol-2-propanol to give white needles (2.2 g, 76%), m.p. 149-149.8°. Elemental analysis. Found: C, 31.89; H, 4.57; N, 15.88%. Calculated for C ^ ^ I : C, 31.71; H, 4.56; N, 15.86%. The NMR spectrum of this compound in B^O consists of two singlets at 6 = 3.20 (3H) and 4.00 (3H) p.p.m. and three quartets at 6 = 6.93 (IH), 7.39 (IH) and 7.53 (IH) p.p.m. 2,3-Diamino-l-methylpyridinium iodide (XCV) A solution of 3-amino-2-chloro-l-methylpyridinium iodide (XCIV, 3 g) i n 40 ml of absolute methanol was bubbled through by ammonia gas for 1 hr. Ammonium iodide was removed by sodium carbonate treatment, as in the previous procedure. The product was recrystallized twice from 2-propanol (charcoal) to give white needles (1.5 g, 66%). M.p. 121-123°. Elemental analysis: Found: C, 21.81; H, 3.99; N, 16.68%. Calculated for C 6H 1 0N 3I: C, 28.69; H, 3.99; N, 16.75%. The NMR spectrum of this compound i n D^O consists of one singlet at 6 = 3.87 (3H)<p.p.m., and three quartets at 6 = 6.80 (lHj), 7.30 (IH) and 7.45 (IH) p.p.m. - 148 -2-Amino-3-methylaminopyridine (XCVI) 145 This compound and i t s hydriodide were obtained from 2-chloro-3-methylaminopyridine (XCI) in a manner analogous to that used for the preparation of 2,3-bismethylaminopyridine (LXXXVI) and i t s salt. The hydriodide of XCVI (white needles) has m.p. 149-150° (decomposition). The NMR spectrum of the hydriodide of XCVI in D^ O consists of one singlet at 6 = 2.85 (3H) p.p.m. and many peaks centered at 6 = 7.0 (3H) p.p.m. The hydriodide of 3-amino-2-methylaminopyridine (LXXXIII) was obtained by the treatment with hydrogeh-dodide as before. The NMR spectrum of this compound;in D20 consists of one singlet at 6 = 3.13 (3H) p.p.m. and three quartets at 6 = 6.85(1H), 7.35 (IH) and 7.47 (IH) p.p.m. Reaction of 3-Amino-2-methylaminopyridine (LXXXIII) with Methyl Iodide 3-Amino-2-methylaminopyridine (LXXXIII, 1.2 g) was stirred in methanol (1 ml) with excess of methyl iodide (2 g) for 15 hr at room temperature. The precipitated product was collected and recrystallized from 2-propanol with a small amount of ether to give white needles. The NMR spectrum, elemental analysis, and m.p. showed this compound to be the hydriodide of 2,2-bismethylaminopyridine (LXXXVI) (2.0 g, 85%). A solution of this compound (1 g) in water (10 ml) was made slightly basic with sodium carbonate, saturated with sodium chloride, and then extracted with ether. Evaporation, and recrystallization of the residue from benzene-petroleum ether gave white needles, identified as 2,3-bismethylaminopyridine (LXXXVI). - 149 -Reaction of 2,3-Diaminopyridine (LXXXVII) with Methyl Iodide 2,3-Diaminopyridine (LXXXVII, ca. 40 mg) and methyl iodide (200 mg) were stirred in the solvent (0.8 ml) for, in most cases, 4 hr. After the solvent and excess methyl iodide had been removed by evaporation the residue was dissolved in D20 and the NMR spectrum of the solution was recorded. In a l l cases the product consists of a mixture of the ring-methylated and 3-N-methylated products in almost quantitative yield, as shown by absorption at 6 3.87 and 2.85 p.p.m. The ratio of ring-methylation to 3-N-methylation were determined by the integral values of these two peaks. The absence of absorption at <$ 3.13 p.p.m. showed that l i t t l e , i f any, 2-N-methylation had occurred. Measurement of Amino-Imino Tautomerism of 2,3-Diaminopyridine (LXXXVII) Since both the amino- and imino-form in this system give the same mesomeric cation by protonation, the tautomeric equilibrium constant, 146 K , can be obtained by assuming the following equilibrium processes. - 150 -In t h i s scheme ^ g g + ( a m £ n o ) a n ^ ^H+Cimino) a r e t b e a c - ^ d i s s o c i a t i o n constants of the cation as the conjugate acid of the amino- and the imino-form, r e s p e c t i v e l y . Hence, the following equations can be given. K_ + = [amino] [H +] ( 4 1 ) BH +(amino) [cation] v i = [imino][H +] ( m ^BH +(imino) [cation] K } thus _ [imino] _ KBH +(imino) taut [amino] . BH^(amino) Accordingly, K can be calculated from the two d i s s o c i a t i o n taut constants. Although the i o n i z a t i o n constant of the imino-form cannot be d i r e c t l y determined, the corresponding constant of the 1-methyl de r i v a t i v e can be substituted as an approximation. Thus the d i s s o c i a t i o n constant of l-methyl-2,3-diaminopyridinium iodide (XCV) was determined by standard u l t r a v i o l e t spectroscopic methods i n phosphate buffer at 147 pH 25°. A f t e r correcting for s a l t e f f e c t s a value of 12.67 was obtained for the pK^-f. value of t h i s compound. The pK + f o r 2,3-B H J i n diaminopyridine i s 7.00. 1 4 3 By using these values, a value of 2.1 x 10 ^ was obtained f o r K from equation 43. taut Attempts to Synthesize Quarternary Salts of 9-Methyl-8-azaflavin (LXXIX) 135 A standard method to synthesize r i b o f l a v i n derivatives was applied to the synthesis of quarternary s a l t s of 9^methyl-8-azaflavin; A mixture of 3-amino-2-methylamino-l-methylpyridinium iodide (LXXXII) and alloxan was refluxed i n a c e t i c a c i d i n the presence of b o r i c acid - 151 -or zinc chloride as a c a t a l y s t under a v a r i e t y of conditions. CH, N \ ^ 0 ' (LXXXII) However, the expected reaction d i d not take place. Attempts using t r i f l u o r o b o r a t e as c a t a l y s t were also unsuccessful. Such inertness of LXXXII toward the condensation reaction may be a t t r i b u t e d to the mesomeric i n t e r a c t i o n between the quarternary r i n g nitrogen and the 2-methylamino nitrogen, which would be expected to reduce the- nucleo-p h i l i c i t y of the 2-methylamino nitrogen greatly. ? 3 N<^NHCH 3 Therefore, the synthesis of LXXIX was attempted i n d i f f e r e n t ways. 148 9-Methyl-8-azaflavin (LXXVIII), synthesized i n the standard way, was subjected to a methylation procedure using methyl iodide or dimethyl s u l f a t e under a v a r i e t y of conditions. However, these attempts were again unsuccessful. The f a i l u r e to obtain LXXIX i n t h i s case can be a t t r i b u t e d to two f a c t s ; (1) LXXVIII i s extremely inso l u b l e i n most solvents and a s u i t a b l e solvent for the reaction could not be found, and (2) the methyl group at the 9-position of LXXVIII s t e r i c a l l y prevents the methylation from occurring at the 8-position. Accordingly, attempts - 152 -to obtain LXXIX were abandoned. I t may be worthwhile to t r y to reduce 3-amino-2-methylamino-l-methylpyridinium iodide (LXXXII) to i t s dihydro compound (XCVIII) and to carry out the condensation reaction of the l a t t e r compound with alloxan i n the usual way. H H (XCVIII) - 153 -BIBLIOGRAPHY 1. F.G. Hopkins, Nature, 45, 197, 581 (1891); Philos. Trans. Roy. Soc. (London), B186, 661 (1893). 2. H. Wieland and Cl. Schopf, Ber., 58, 2178 (1925). 3. Cl. Schopf and H. Wieland, Ber., 59, 2067 (1926). 4. R. Purrman, Ann., 544, 182 (1940); 546, 98 (1940). 5. Cl. Schopf and R. Reichert, Ann., 548, 83 (1941). 6. R.B. Angier et a l . , Science, 102, 227 (1945); 103, 667 (1946). 7. M. Gates, Chem. Revs., 41, 63 (1947). 8. T. Masuda, Pharm. Bull. (Japan), 4, 71, 72, 375 (1956); 5_, 28 (1957). 9. T. Masuda, T. Kishi, and M. Asai, Pharm. Bull. (Japan), ,5, 598 (1957); Chem. Pharm. Bull. (Japan), 6_, 291 (1958). 10. G.F. Maley and G.W.E. Plaut, J. Biol. Chem., 234, 641 (1959). 11. G.F. Maley and G.W.E. Plaut, Arch. Biochem. Biophys., 80, 219 (1959). 12. T. Masuda, T. Kishi,and M. Asai, Chem. Pharm. Bull. (Japan), 6^, 447 (1958). 13. A. Albert, Quart. Revs. Chem. Soc, 6_, 197 (1952). 14. A. Albert, Fortschs. Chem. Org. Naturstoffe, 11, 350 (1954). 15. "The Chemistry and Biology of Pteridines", Ciba Foundation Symposium, L i t l e , Brown and Company, Boston, 1955. 16. W. Pfleiderer, Angew. Chem., 75, 993 (1963); Intern. Ed., 3, 114 (1964). 17. "Pteridine Chemistry", Third International Pteridine Symposium, Stuttgart, 1962, Pergamon Press, Oxford, 1964, The Macmillan Company, New York, 1964. 18. R.C. Elderfield and A.C. Mehta, "Heterocyclic Compounds", R.C. Elderfield Ed., J. Wiley and Sons. Inc., New York, 1967, Vol. 9, p. 1. - 154 -19. H. Rembold and W.L. Gyure, Angew. Chem., 84_, 1088 (1972); Intern. Ed., 11, 1061 (1972). 20. S.F. Mason, in reference 15, p. 74. 21. A. Veillard and B. Pullman, Compt. rend., 353, 2418 (1961). 22. B. Pullman and A. Pullman, Proc. Natl. Acad. Sci. (U.S.), 44, 1197 (1958). 23. T.H. Goodwin and A.L. Porte, J. Chem. Soc, 3595 (1956). 24. 0. Chalvet and C. Sandorfy, Compt. rend., 228, 566 (1949). 25. M. Kamiya and Y. Akahori, Chem. Pharm. Bull. (Japan), 20, 918 (1972). 26. A. Albert, J. Chem. Soc, 2960 (1955). 27. A. Albert, D.J. Brown, and H.C.S. Wood, J. Chem. Soc, 2066 (1956). 28. A. Albert, J. Chem. Soc, Special Publ., 3, 1 38 (1955). 29. Y. Inoue and D.D. Perrin, J. Phys. Chem., 66_, 1689 (1962). 30. D.D. Perrin, "Advances in Heterocyclic Chemistry", A.R. Katritzky Ed., Academic Press,. New York, 1965, Vol. 4, pp. 43-73. 31. W.E. Fidler and H.C.S. Wood, J. Chem. Soc, 3980 (1957). 32. W. Pfleiderer and G. Nubel, Ber., 93, 1406 (1960). 33. CH. Winestock and G.W.E. Plaut, J. Org. Chem., 26, 4456 (1961). 34. W. Pfeiderer, J. Bunting, D.D. Perrin, and G. Nubel, Ber., 99_, 3503 (1966). 35. A. Albert, Angew. Chem., 79, 913 (1967); Intern. Ed., 6_, 919 (1967). 36. A. Albert, Y. Inoue, and D.D. Perrin, J. Chem. Soc, 5151 (1963). 37. P. Hemmerich, in reference 17, p. 143. 38. A. Albert, W.L.F. Armarego, and E. Spinner, J. Chem. Soc, 2689 (1961). 39. W.L.F. Armarego, J. Chem. Soc, 5030 (1962). -155 -40. W.L.F. Armarego, J. Chem. Soc, 4094 (1962). 41. D.D. Perrin, J. Chem. Soc, 645 (1962). 42. D.J. Brown and S.F. Mason, J. Chem. Soc, 3443 (1956). 43. A. Albert and G.B. Barlin, J. Chem. Soc, 5156 (1963). 44. A. Albert and G.B. Barlin, J. Chem. Soc, 5737 (1963). 45. N.W. Jacobsen, J. Chem. Soc. (C), 1065 (1966). 46. 0. Philipow, J. Prakt. Chem., 93, 162 (1916). 47. R. Stewart, "Oxidation in Organic Chemistry", K.B. Wiberg, ed., Academic Press., New York, 1965, Part A, p. 47. 48. A. Albert and F. Reich, J. Chem. Soc, 127 (1961). 49. T. Rowan, H.C.S. Wood, and P. Hemmerich, Proc. Chem. Soc, 261 (1961). 50. A. Albert and W.L.F. Armarego, "Advances in Heterocyclic Chemistry," A.R. Katritzky, ed., Academic Press., New York, 1965, Vol. 4, pp. 1-42. 51. T. Paterson and H.C.S. Wood, Chem. Comm., 290 (1969). 52. J.M. McAndless and R. Stewart, Canad. J. Chem., 48, 263 (1970). 53. R.L. Beach and G.W.E. Plaut, Biochemistry, 9_, 760 (1970). 54. R. Stewart and J.M. McAndless, J. Chem. Soc, Perkin II, 376 (1972). 55. F.J. Bullock and 0. Jardetzky, J. Org. Chem., 30, 2056 (1965). 56. T. Masuda, Pharm. Bull. (Japan), .5, 136 (1957). 57. G.F. Maley and G.W.E. Plaut, Federation Proc, 17, 268 (1958). 58. S. Kuwada, T. Masuda, T. Kishi, and M. Asai, Chem. Pharm. Bull. (Japan), 6, 618 (1958). 59. S. Kuwada, T. Masuda, T. Kishi, and M. Asai, J. Vitaminol. (Japan), 4_, 217 (1958). 60. T. Kishi, M. Asai, T. Masuda, and S. Kuwada, Chem. Pharm. Bull. (Japan), ]_, 515 (1959). - 156 -61. H. Katagiri, I. Takeda, and K. Imai, J. Vitaminol. (Japan), 4_, 211, 278, 286 (1958); _5, 81 (1959). 62. F. Korte and H.U. Aldag, Ann., 628, 144 (1959). 63. G.W.E. Plaut, J. Biol. Chem., 235, PC 41 (1960); 28, 2225 (1963). 64. H. Wacker, R.A. Harvey, CH. Winestock, and G.W.E. Plaut, J. Biol. Chem., 239, 3493 (1964). 65. R.A. Harvey and G.W.E. Plaut, J. Biol. Chem., 241, 2120 (1966). 66. G.F. Maley and G.W.E. Plaut, J. Amer. Chem. Soc, 81, 2025 (1959). 67. R.M. Cresswell and H.C.S. Wood., Proc. Chem. Soc, 387 (1959); J. Chem. Soc, 4768 (1960). 68. A.J. Birch and C.J. Moye, J. Chem. Soc, 412 (1957); 2622 (1958); A.J. Birch, Proc. Chem. Soc, 3 (1962). 69. 0. Diels, W.M. Blanchard, and H. Heyden, Ber., 47, 2355 (1914). 70. T. Rowan and H.C.S. Wood., Proc Chem. Soc, 21 (1963). 71. T. Rowan and H.C.S. Wood., J. Chem. Soc. (C), 452 (1968). 72. T. Paterson and H.C.S. Wood., J. Chem. Soc, Perkin I, 1051 (1972). 73. R.L. Beach and G.W.E. Plaut, J. Amer. Chem. Soc, 92_, 2913 (1970). 74. T. Masuda, T. Kishi, M. Asai, and S. Kuwada, Chem. Pharm. Bull. (Japan), 6^  523 (1958). 75. M. Asai, K. Mizuno, and S. Kuwada, Chem. Pharm. Bull. (Japan), 9_, 85, 503 (1961). 76. CM. Brown, Physiol. Rev., 40, 331 (1960). 77. F. Korte, H.U. Aldag, G. Ludwig, W. Paulus, and K. Storiko, Ann., 619, 70 (1958). 78. H. Mitsuda, F. Kawai, and J. Suzuki, Vitamins (Japan), 23_, 415 (1961). 79. M. Asai, K. Mizuno, and S. Kuwada, Chem. Pharm. Bull. (Japan), 10_, 243 (1962); 11, 23 (1963). - 157 -80. H. Mitsuda, F. Kawai, J. Suzuki, and E. Satani, J. Vitaminol. (Japan), 8, 178 (1962). 81. J.M. McAndless, Ph.D. Thesis, University of British Columbia, Vancouver, 1969. 82. G.W.E. Plaut, Ann. Rev. Biochem., 30, 409 (1961); "Metabolic Pathways", D.M. Greenberg ed., Academic Press, New York, London, 1961, Vol. 2, p. 673. 83. T. Shiota, "Comprehensive Biochemistry", M. Florkin and E.H. Stotz, ed., Elsevier, Amsterdam, 1971, Vol. 21, p. 111. 84. G.M. Brown, "Metabolic Pathways", D.M. Greenberg, ed., Academic Press, New York, 1970, Vol. 4, p. 383; Advan. Enzymol., 35, 35 (1971). 85. H. Rembold and K. Buff, "Energy Transduction in Respiration and Photosynthesis", E. Quagliariello, S. Papa, and CS. Rossi, eds., Adriatica Editrice, Bari 1971, p. 467. 86. H. Rembold and K. Buff, Eur. J. Biochem., 28, 579, 586 (1972). 87. R. Stewart, i n reference 47, p. 41. 88. G. Wagner, J. Russ. Phys. Chem. Soc, 27_, 219 (1895). 89. J. Boeseken, Rec. Trav. Chim., 41, 199 (1922); 47, 683 (1928). 90. K.B. Wiberg and K.A. Saegebarth, J. Amer. Chem. Soc, 2822 (1957). 91. K.B. Wiberg and R.D. Geer, J. Amer. Chem. Soc, 88, 5827 (1966). 92. R. Huisgen, Angew. Chem. Intern. Ed., 2, 633 (1963). 93. D.G. Lee and J.R. Brownridge, J. Amer. Chem. Soc, 95_, 3033 (1973). 94. K.B. Wiberg, C.J. Deutsch, and J. Rocek, J. Amer. Chem. Soc, 95, 3034 (1973). 95. R. Stewart, in reference 47, p. 52. 96. K.B. Wiberg and R.D. Geer, J. Amer. Chem. Soc, 87, 5202 (1965). 97. W.A. Waters and J.S. L i t t l e r , in reference 47, p. 192. - 158 -98. R. Stewart, in reference 47, p. 8. 99. R.E. Hamm and M.A. Suwyn, Inorg. Chem., 6^, 139 (1967). 100. R.E. Hamm and M.A. Suwyn, Inorg. Chem., 6^  142 (1967). 101. B.L. Poh and R. Stewart, Canad. J. Chem., 50, 3432, 3437 (1972). 102. S.J. Rettig and J. Trotter, Canad. J. Chem., 51, 1303 (1973). 103. C. Rappe, Acta Chem. Scand., 20, 2236 (1966). 104. P.K. Glasoe and F.A. Long, J. Phys. Chem., 64, 188 (1960). 105. R. Garry, R.G. Bates, and R.A. Robinson, J. Phys. Chem., 68_, 3806 (1964). 106. R. Stewart, "Oxidation Mechanism", W.A. Benjamin, Inc., 1964, New York, p. 72. 107. K.B. Wiberg and G. Foster, J. Amer. Chem. Soc, 83, 423 (1961); Chem. and Ind., 1961, 108. 108. K.B. Wiberg, B. Marshall, and G. Foster, Tet. Lett., 8_, 345 (1962). 109. R. Stewart, i n reference 47, pp. 2-6. 110. R. Stewart, in reference 106, Chapters 4 and 5. 111. J.C. Sheppard and A.C. Wahl, J. Amer. Chem. Soc, 79, 1020 (1957). 112. O.E. Myers and J.C. Sheppard, J. Amer. Chem. Soc, 83, 4739 (1961). 113. A.D. Britt and W.M. Yen, J. Amer. Chem. Soc, 821, 4516 (1961). 114. A.C. Wahl, Z. Elektrochem., 64, 90 (1960). 115. L. Gjertsen and A.C. Wahl, J. Amer. Chem. Soc, 81, 1572 (1959). 116. L. Melander, "Isotope Effects on Reaction Rates", Ronald Press Co., 1960, New York, p. 22. 117. R. Stewart and M.M. Mocek, Canad. J. Chem., 41_, 1160 (1963). 118. K. Yates, Ph.D. Thesis, University of British Columbia, Vancouver (1959). 119. G. Schwarzenbach and C. Wittwer, Helv. Chim. Acta, 30, 669 (1947). - 159 -120. T. Paterson and H.C.S. Wood., J. Chem. Soc, Perkin I, 1041 (1972). 121. R. Stewart, in reference 106, p. 62. 122. Ri. Stewart, in reference 106, Chapter 11. 123. "Biological Oxidations", T.P. Singer, Ed., Interscience Publishers, New York, 1968. 124. The second one of reference 67. 125. K.B. Wiberg and R.. Stewart, J. Amer. Chem. Soc, 7_7, 1786 (1955). 126. H.A. Laitinen, "Chemical Analysis", McGraw-Hill, Inc., 1960, pp. 35-37. 127. "Handbook of Chemistry and Physics", 44th ed., Chemical Rubber Co., Cleveland, 1961, p. 1706. 128. F.Feigl,"Spot Tests in Organic Analysis",Elsevier Publ. Co., Amsterdam, 1966, p. 200. 129. H.W. Wanzlick and W. Loechel, Ber., 86, 1463 (1953). 130. G. Hesse and H. Stahl, Ber., 89, 2424 (1956). 131. A.M. Moore and CH. Thompson, Canad. J. Chem., 35, 163 (1957). 132. H. Budzikiewicz, C. Djerassi, and D.H. Williams, "Mass Spectrometry of Organic Compounds", Holden-Day, Inc., San Francisco, 1967, p. 102. 133. P. Hemmerich, "Flavins and Flavoproteins", H. Kamin, Ed., University Park Press, Baltimore, Md., 1971, p. 103. 134. CA. Hamilton, Progr. Bioorg. Chem., _1, 83 (1971). 135. R. Kahn, K. Reinemund, and F. Weygand, Ber., 67, 1460 (1934). 136. N.F. Twritsyna and A.F. Vompe, Doklady Akad. Nauk S.S.S.R., 74_, 509 (1950) (Chem. Abs., 45, 3846 (1951)). 137. A.E. Chichibabin',. R.A. Konovalova and A.A. Konovalova, Ber., 54, 814 (1921). 138. G.B. Barlin and A.C Young, J. Chem. Soc. (B), 1971, 1675, 2323. 139. A.E. Chichibabin and R.A. Konovalova, Ber., 59, 2055 (1926). - 160 -140. R. Frampton, CD. Johnson, and A.R. Katritzky, Annalen, 749, 12 (1971). 141. J. Packer, J. Vaughan and E. Wong, J. Amer. Chem. Soc, 80, 905 (1958). 142. H.C. Brown and A. Cahn, J. Amer. Chem. Soc, 77_. 1715 (1955). 143. G.B. Barlin, J. Chem. Soc, 1964, 2150. 144. 0. v. Schickh, A. Binz and A. Schulz, Ber., 69, 2593 (1936). 145. J.W. Clark-Lewis and M.J. Thompson, J. Chem. Soc, 1957, 442. 146. S.J. Angyal and CL. Angyal, J. Chem. Soc, 1952, 1461. 147. A. Albert, "Physical Methods in Heterocyclic Chemistry", A.R. Katritzky, ed., Academic Press., 1963, Volume 1, p. 12. 148. H. Rudy and 0. Majer, Ber., 71, 1243 (1938); 72, 933 (1939). 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0060148/manifest

Comment

Related Items