UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Branched-chain sugar nucleosides Nguyen, Laure Marie Kim-Khanh 1968

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

Item Metadata

Download

Media
831-UBC_1969_A1 N38.pdf [ 6.96MB ]
Metadata
JSON: 831-1.0059860.json
JSON-LD: 831-1.0059860-ld.json
RDF/XML (Pretty): 831-1.0059860-rdf.xml
RDF/JSON: 831-1.0059860-rdf.json
Turtle: 831-1.0059860-turtle.txt
N-Triples: 831-1.0059860-rdf-ntriples.txt
Original Record: 831-1.0059860-source.json
Full Text
831-1.0059860-fulltext.txt
Citation
831-1.0059860.ris

Full Text

BRANCHED-CHAIN SUGAR NUCLEOSIDES BY LAURE (BENZING) NGUYEN K.K. Math. Elem; M.P.C.; C.G., U n i v e r s i t e de Strasbourg B.Sc. (Honours), U n i v e r s i t e L a v a l , 1964 M.Sc, U n i v e r s i t e L a v a l , 1965 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY i n the Department of CHEMISTRY We accept t h i s t h e s i s as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA November, 1968 In p r e s e n t i n g t h i s t h e s i s i n p a r t i a l f u l f i l m e n t o f the r e q u i r e m e n t s f o r an advanced degree a t the U n i v e r s i t y o f B r i t i s h C olumbia, I agr e e t h a t the L i b r a r y s h a l l make i t f r e e l y a v a i l a b l e f o r r e f e r e n c e and Study. I f u r t h e r a g r e e t h a t p e r m i s s i o n f o r e x t e n s i v e c o p y i n g o f t h i s t h e s i s f o r s c h o l a r l y purposes may be g r a n t e d by the Head o f my Department or by h i s r e p r e s e n t a t i v e s . I t i s u n d e r s t o o d t h a t c o p y i n g o r p u b l i c a t i o n o f t h i s t h e s i s f o r f i n a n c i a l g a i n s h a l l not be a l l o w e d w i t h o u t my w r i t t e n p e r m i s s i o n . Department of OHF.MTSTRY  The U n i v e r s i t y o f B r i t i s h Columbia Vancouver 8, Canada D a t e 26 o f F e b r u a r y 1969 - i -ABSTRACT A new route to branched-chain sugars by a p p l i c a t i o n of a modified W i t t i g r e a c t i o n to ketoses has been developed. 3-Deoxy-3-C-(2'-hydroxyethyl)-1,2:5,6-di-O-isopropylidene-a-D-allofuranose (LXVII) was prepared from 1,2:5,6-di-O-isopropylidene - q-D-ribo-hexofuranos-3-ulose (I) and subsequently used as key intermediate i n the synthesis of two novel branched-chain sugar n u c l e o s i d e s . S e l e c t i v e h y d r o l y s i s of (LXVII) to the 1,2-0_-monoisopropylidene d e r i v a t i v e (LXXII) followed by b e n z o y l a t i o n and a c e t o l y s i s y i e l d e d 1 ,2-di-0-acetyl-2 1 ,5,6-tri-0_-benzoyl-(2'-hydroxyethyl)-3-D-allofuranose (LXXIV). Condensation of t h i s compound with chloromercuri-6-benzamidopurine i n the presence of t i t a n i u m t e t r a c h l o r i d e and subsequent deblocking with methanolic sodium methoxide afforded the branched-chain sugar nucleoside 9-[3'-deoxy-3'-C-(2"-hydroxyethyl)-3-D-aUcfu r a i i o s y l ] - a d e n i n e (LXXXIII) i n 48% y i e l d based on (LXXIV). In a separate procedure, 3-C_-(carb o methoxymethyl)-3-deoxy-l, 2 :5,6-di-O - i s o p r o p y l i d e n e - q - D - a l l o f u r a n o s e (LXI) was s e l e c t i v e l y hydrolyzed to a f f o r d 3-C-(carbomethoxymethyl)-3-deoxy-l,2-0-isopropylidene-a-D-allofuranose (LXXXIV). Sodium metaperiodate degradation of the l a t t e r f o l l owed by re d u c t i o n w i t h sodium borohydride y i e l d e d 3-deoxy-3-C-(2'-hydroxyethyl)-1,2-O-isopropylidene-a-D-ribofuranose (LXXXV). In a procedure p a r a l l e l to that used f o r the p r e p a r a t i o n of nucleoside (LXXXIII) compound (LXXXV) was converted to the branched-chain nucleoside 9-[3'-deoxy-3'-£-(2"-hydroxyethyl)-g-D-ribofuranosyl]-adenine (LXXXVIII) i n 37% yield based on the d i a c e t a t e . The oxo r e a c t i o n was a p p l i e d to the unsaturated sugar 3-deoxy-1,2:5,6-di - O -isopropy1idene- q -D-erythro-hex-3-enose (II) i n an attempt to prepare a branched-chain sugar having a hydroxymethyl group on C-3. In an endeavour t o find a better method for the synthesis of compound (II), 1,2:5,6-di-0-isopropylidene-3-0_-p_-tolylsulfonyl-a-D-glucofuranose (XXIV) was reacted with a tetramethylammonium hydroxide solution in dimethyl sulfoxide and shown to yield quantitatively 1,2:5,6-di-O-isopropylidene-a-D-glucofuranose. Under similar conditions, the 3-0-p_-nitrobenzenesulfonate ester of 1,2:5,6-di-O-isopropylidene-a-D-glucofuranose (LXXXIX) gave 3-0-[p_- (3-deoxy-l,2:5,6-di-0_-isopropylidene-a-D-glucofuranose-3-yl) oxyphenylsulf onyl]-1,2:5,6-di-0-isopropylidene-a-D-glucofuranose (XC) and l,2:5,6-di-0_-isopropylidene-3-0-p-nitrophenyl-a-D-glucofuranose (XCI). A discussion of the probable mechanism of this reaction i s based on a study of the products obtained by application of the same reaction to ethyl p-nitrobenzenesulfonate. The reaction of 3-deoxy-l,2:5,6-di-0-isopropylidene-a-D-erythro-hex-3-enose (II) with carbon monoxide and hydrogen (in the ratio of 1:3) in the presence of dicobalt octacarbonyl gave eight products. Hydrogenolysis of the 5,6-0-isopropylidene group occurred as evidenced by n.m.r. and g.l.c. of the volatile portion of the oxo mixture. When (II) was allowed t o react with carbon monoxide and hydrogen (6:1), part of the starting material rearranged to afford a substance (XCIX) in 25% yield which was presumed to be a 2,3-ene. Reduction of the remaining oxo product with sodium borohydride followed by acetylation gave after separation by g.l.c. two crystalline components in a ratio of 1:6. - i i i -TABLE OF CONTENTS ABSTRACT i TABLE OF CONTENTS i i i LIST OF FIGURES v i i i ACKNOWLEDGEMENTS ix I. OBJECTIVE . 1 II. INTRODUCTION 3 1. Branched-chain sugars 3 2. Methodsof oxidation 8 3. Wittig reaction 10 3.1 Wittig reaction in carbohydrate chemistry 11 3.2 Phosphonate modification of the Wittig reaction ... 13 4. Oxo reaction 20 4.1 Unsaturated sugars (3,4-enes) 22 4.2 Reactions of the 3,4-enes 23 4.3 Mechanism and stereochemistry 25 4.4 Effect of the carbohydrate structure 27 . 5. Nucleosides 29 5.1 Branched-chain sugar nucleosides 29 5.2 Nucleoside synthesis 31 III. RESULTS AND DISCUSSION .. 38 1. Wittig reaction 38 1.1 1,2:5,6-Di-0-isopropylidene-a-D-ribo-hexo-furanos-3-ulose ". 39 - iv -1.2 Wittig reaction of 1,2:5,6-di-0-isopropylidene-a-D-ribo-hexofuranos-3-ulose (I) to yield 3-C-(carbo-methoxymethyl)-3-deoxy-l,2 :5,6-di-0-isopropylidene-a-D-allofuranose 41 1.3 3-Deoxy-3-C-(2•-hydroxyethyl)-1,2:5,6-di-0-isopropylidene-a-Q-allofuranose. 51 2. Nucleoside synthesis 54 2.1 Conversion of 3-deoxy-3-C-(2'-hydroxyethyl)-1,2:5,6-di-O-isopropylidene-a-D-allofuranose into l,2-di-0-acetyl-2',5,6-tri-O-benzoyl-(2'-hydroxyethyl)-B-D-allofuranose T .. 54 2.2 Chloromercuri-6-benzamidopurine 58 2.3 9-3-g-Glucofuranosyladenine 59 2.4 9-[3-Deoxy-3-£- (2 M-hydroxyethyl)-g-D-allofuranosyl]-adenine 65 2.5 l,2-Di-0-acetyl-2',5-di-0-benzoyl-3-deoxy-3-£-(2' -hydroxyethyl) -g-D-ribofuranose 71 2.6 9-[3-Deoxy-3-£- (2"-hydroxyethyl)-g-D-ribofuranosyl]-adenine 76 3.. Oxo reaction 79 3.1 3-Deoxy-l,2:5,6-di-O-isopropylidene-q-D-erythro-hex-3-enose (II) 7 .. 80 3.2 Reaction of the 3-0-p_-toluenesulf onate and the 3-0-p_-nitrobenzenesulfonate of compound (XIV) with tetramethylammonium hydroxide in dimethyl sulfoxide 81 3.3 Synthesis of 3-deoxy-l,2:5,6-di-0-isopropylidene-a-D-erythro-hex-3-enose (II) 90 3.4 Reaction of (II) with carbon monoxide and hydrogen under oxo conditions 90 IV. EXPERIMENTAL 98 General Considerations 98 l,2:5,6-Di-0-isopropylidene-ct-Q-glucofuranose (XIV) 99 1,2:5,6-Di-0-isopropylidene-a-D-ribo-hexofuranos-3-ulose (I) 99 Wittig reaction of 1,2:5,6-di-0-isopropylidene-a-D-ribo-hexofuranos-3-ulose (I) to yield 3-£-(carbomethoxymethyl)-3-deoxy-l,2:5,6-di-0_-isopropylidene-a-Q-allofuranose (LXI) .... 100 - v -3-Deoxy-3-£-(2 1-hydroxyethyl)-l,2:5,6-di-0-isopropylidene-a-D-allofuranose (LXVII) 102 3-Deoxy-3-£-(2'-hydroxyethyl)-1,2:5,6-di-0-isopropylidene-2'-O-p-tolylsulfonyl-a-D-allofuranose (LXVIII) 102 2'-0-p_-Bromophenylsulfonyl-3-deoxy-3-£- (2 *-hydroxyethyl)-1,2:5,6-di-O-isopropylidene-a-D-allofuranose (LXIX) 103 3-Deoxy-3-£- ( 2 1 - iodoethyl) -1,2 :5,6-di-O-isopropylidene-a-D-allofuranose (LXXI) T... 104 2' -£-p-Bromobenzoyl-3-deoxy-3-£- (2'-hydroxyethyl) -1,2:5,6-di-£-isopropylidene-a-D-allofuranose (LXX) 104 Partial hydrolysis of compound (LXVII) to yield 3-deoxy-3-£-(2'-hydroxyethyl)-l,2-0-isopropylidene-a-D-allofuranose (LXXII) 105 2',5,6-Tri-0-benzoyl-3-deoxy-3-C-(2*-hydroxyethyl)-1,2-0-isopropylidene-ct-D-allofuranose (LXXIII) 106 Acetolysis of 2',5,6-tri-0-benzoyl-3-deoxy-3-£- (2'-hydroxy-ethyl) -1,2-0^-isopropylidene-a-D-allofuranose (LXXIII) to yield 1,2-di-0-acetyl-2»,5,6-tri-O-benzoyl-(2 1-hydroxyethyl)-B-D-allofuranose (LXXIV) T... 106 6-Benzamidopurine (LXXV) 107 Chloromercuri-6-benzamidopurine (LXXVI) 108 1,2-0_-Isopropylidene-ct-D-glucofuranose (LXXVII) 108 3,5,6-Tri-O-benzoyl-l, 2-O-isopropylidene-ct-D-giiicofuranose (LXXVI 11) 7 7 108 l,2-Di-0-acetyl-3,5,6-tri-0-benzoyl-D-glucofuranose (LXXIX).. 109 9-3-g-Glucofuranosyladenine (LXXXI) 110 9-[3-Deoxy-3-C-(2"-hydroxyethyl)-g-D-allofuranosy]]-adenine (LXXXIII) 7 HI 3-Deoxy-3-C_- (carbomethoxymethyl) -1,2-0-isopropylidene-a-D-allofuranose (LXXXIV) 7 113 Sodium metaperiodate degradation of 3-deoxy-3-£-(carbomethoxy-methyl)-!, 2rO-isopropylidene-a-D-allofuranose (LXXXIV) to yield 3-deoxy-3-£- (21-hydroxyethyl)-1,2-£-isopropylidene-a-D-ribofuranose (LXXXV) 7. 114 - v i -2',5-Di-0-benzyl-3-deoxy-3-C- (21-hydroxyethyl)-1,2-0-isopro-pylidene-o-D-ribofuranose (LXXXVI) Acetolysis of 21,5-di-0-benzoyl-3-deoxy-3-C-(2'-hydroxyethyl)-1,2-0-isopropylidene-a-D-ribofuranose (LXXXVI) to yield 1,2-di 0-acetyl-2' ,5-di-0-benzoyl-3-deoxy-3-C_- (2 ' -hydroxyethyl) -B-Q-ribofuranose (LXXXVII) Condensation of 1,2-di-0-acetyl-2' ,5-di-0-benzoyl-3-deoxy-3-C-(2»-hydroxyethyl)-3-B-ribofuranose '(LXXXVII) with chloromercuri—6-benzamidopurine to yield 9-[3'-deoxy-3'-C-(2"-hydroxyethyl) - 6-D-ribofuranosyl]-adenine (LXXXVIII).... Sodium metaperiodate oxidation and reduction of the alio nucleoside (LXXXIII) to yield the ribo nucleoside (LXXXVIII). Reaction of l,2:5,6-di-0-isopropylidene-3-0-p_-tolylsulfonyl-a-D-glucofuranose with tetramethylarnmonium hydroxide in dimethyl sulfoxide to yield l,2:5,6-di-0_-isopropylidene-a-D-glucofuranose (XIV) T.. Reaction of 1,2 :5,6-di-£-isopropylidene-3-0_-p_-nitrophenyl-sulfonyl-ot-D-glucofuranose (LXXXIX) with tetramethylarnmonium hydroxide in dimethyl sulfoxide to yield 3-0- [p_- (3-deoxy-1,2:5,6-di-£-isopropylidene-a-Q-glucofuranose-3-yl)-oxy-phenylsulfonyl] -1,2 :5,6-di-0_-isopropylidene- ot-D-glucofuranose (XC) and l,2:5,6-di-0_-isopropylidene-3-£-p_-nitrophenyl-a-D-glucofuranose (XCI) 7... Characterization of product (XCI) Reductive cleavage of (XC) using sodium amalgam Ethyl p-nitrobenzenesulfonate (XCIII) Cyclohexyl p_-nitrobenzenesulfonate (XCVI) Reaction of ethyl p_-nitrobenzenesulfonate (XCIII) with tetramethylarnmonium hydroxide in dimethyl sulfoxide to yield tetramethylarnmonium p_-nitrobenzenesulfonate (XCIV) and p-nitro phenetole (XCV) 7 Reaction of cyclohexyl p_-nitrobenzenesulfonate (XCVI) with tetramethylarnmonium hydroxide in dimethyl sulfoxide 1,2:5,6-Di-£-isopropylidene-3-£-p_-tolylsulfonyl-a-g-gluco-furanose (XXIV) 7 1,2:5,6-Di-0_-isopropylidene-3-0-p_-nitrophenylsulfonyl-a-g-glucofuranose (LXXXIX) - v i i -Reaction of l,2:5,6-di-0-isopropylidene-3-0-p-tolylsulfonyl-a-Q-glucofuranose (XXIV) with potassium t-butoxide in dimethyl sulfoxide 3-Deoxy-l,2:5,6-di-0-isopropylidene-q-D-erythro-hex-3-enose (II) =. Reaction of (II) with carbon monoxide and hydrogen under oxo conditions - v i i i -LIST OF FIGURES FIGURE 1. Proton magnetic resonance spectrum at 100 MHz/sec in deuteriochloroform. of a mixture of the two stereoisomers of 3-C-(carbomethoxymethylene)-3-deoxy-l,2:5,6-di-0-isopropylidene-a-D-allofuranose 45 2. Partial proton magnetic resonance spectrum of (LX + LIX) in deuteriochloroform (a) at 100 MHz/sec; (b) at 100 MHz/sec,H-4 proton irradiated; (c) at 100 MHz/sec, H-l' proton irradiated; (d) at 100 MHz/sec, H-2 proton irrad-iated.) 47 3. Proton magnetic resonance spectrum at 100 MHz/sec of 3-£- (carbomethoxymethyl) -3-deoxy-l,2:5,6-di-O-isopropyli-dene-a-D-allofuranose in deuteriochloroform 50 Proton magnetic resonance spectrum at 100 MHz/sec in dimethyl sulfoxide-d^ of 9-B-D-glucofuranosyladenine.... 64 5. Proton magnetic resonance spectrum at 100 MHz/sec of branched-chain sugar nucleosides in dimethyl sulfoxide (a) 9- (3' -deoxy^-31 -£- (2"-hydroxyethyl) -3-D-allofuranosyl) -adenine 68 (b) 9-(3*-deoxy-31-£-(2"-hydroxyethyl)-B-D-ribofuranosyl)-adenine 68 6. O.R.D. curves of branched-chain sugar nucleosides plus adenine , adenosine; 9-(3'-deoxy-31-£-(2"-hydroxyethyl)-B-D-ribofuranosyl)-adenine;••• 9-(31-deoxy-3'-£-(2"-hydroxyethyl)-B-D-allofuranosyi)-adenine 70 7. Proton magnetic resonance spectrum at 100 MHz/sec in deuteriochloroform of 3-deoxy-3-£-(2'-hydroxyethyl)-1,2-0-isopropylidene-a-D-ribofuranose. . . . : 75 8. Partial proton magnetic resonance spectrum of (XC) in carbon tetrachloride ((a) at 100 MHz/sec; (b) at 100 MHz/sec H-l and H-l' protons irradiated; (c) at 100 MHz/sec, H-2' proton irradiated; (d) at 100 MHz/sec, H-2 proton irradiated.) 4^ - ix ACKNOWLEDGEMENTS The author wishes to express her thanks to Professor A. Rosenthal for his continual encouragement and s k i l l f u l guidance throughout the course of this work. Sincere appreciation is due to Professors G.G.S. Dutton and J.P. Kutney who read and commented on this manuscript. Particular thanks are due to Professor L.D. Hall for stimulating discussions. I have greatly appreciated the encouragement of my husband and my son Patrick, family and friends without whose help this thesis might never have been completed. Finally, I wish to thank the National Research Council of Canada for awarding me a Bursary for 1964-1965 and Studentships for 1965-1968. I. OBJECTIVE: Medicinal chemists have been actively searching for compounds capable of interfering with nucleic acid metabolism for use in cancer chemotherapy. From this search came the f i r s t antibiotic nucleoside cordycepin, which in the form of i t s triphosphate has been shown to have biological activity. It is of interest to note, that not only was cordycepin the f i r s t antibiotic nucleoside reported in the literature (1951) but the confusion this compound created among chemists and biologists was unprecedented. The structure which was thought to be that of a branched-chain sugar nucleoside''' was in fact that of a structural analog of the natural nucleoside adenosine in 2 which one hydroxyl on the sugar moiety was replaced by hydrogen. The biochemical importance of cordycepin or 3'-deoxyadenosine, suggested that the synthesis of branched-chain nucleosides, unknown in 1965, would be of interest. This thesis describes the preparation of such compounds. Our primary objective was the synthesis of hexo- and in particular pento-furanosyl branched-chain carbohydrates in which the secondary hydroxyl group on C-3' is replaced by a hydroxyethyl or a hydromethyl group. Suitable derivatives of these modified sugars may then afford when condensed with adenine novel branched-chain sugar nucleosides, one of them being a structural analog of adenosine. To achieve this objective two different methods of synthesis of branched-chain sugars have been tried. - 2 -The f i r s t method makes use of a new approach to such synthesis, namely the application of a modified Wittig reaction to the carbohydrate ketose 1,2:5,6-di-0-isopropylidene-a-D-ribo-hexofuranos-3-ulose(I)to afford after reduction the C-3' hydroxyethyl branched-chain sugar. In the second method the oxo reaction is applied to the unsaturated carbohydrate 1,2:5,6-di-0_-isopropyli-dene-q-D-erythro-hex-3-enose Ql)yielding a sugar with a hydroxymethyl group on C-3 or C-4. During the completion of this work, the synthesis of three branched-chain sugar nucleosides, analogs of adenosine, have been reported by the chemists of Merck, Sharp and Dohme Research Laboratories. The structural modification in these nucleosides consisted in the replacement, by a methyl group of a hydrogen atom on one carbon of the sugar moiety. A l l three nucleosides, 5',5'-di-C_, 3'-C and 2'-C methyl adenosine were shown to possess biological activity as measured by their a b i l i t y to inhibit the growth of KB cells in culture. More recently Reist and coworkers at the Stanford Research Institute published their synthesis of two branched-chain nucleosides, in both anomeric forms, in which a hydroxylgroup on one carbon of the sugar moiety is replaced by a hydroxymethyl group. However, unlike the former nucleosides, the nucleosides'1 3-deoxy-3-C-hydroxymethyl-D-erythrofuranose and 2-deoxy-2-C_-hydroxymethyl-D-erythrofuranose, cannot be considered as adenosine analogs since they do not contain the ribofuranosyl skeleton in the sugar moiety. In order to provide a background to subsequent discussions a brief review w i l l be made of the synthesis of the branched-chain sugars, the Wittig reaction and i t s application in carbohydrate chemistry as well as the oxo reaction. A few interesting points concerning purine nucleosides (in particular branched-chain nucleosides) and their synthesis w i l l be discussed. II. INTRODUCTION: 1. Branched-chain sugars In recent years branched-chain sugars have been discovered as constituent 3—6 parts of a number of antibiotics . The Birkbeck group of investigators has been at the forefront of much of the work done in this f i e l d , and their research in addition to that of other investigators has been reviewed up to 1963^. The continuous discovery of new naturally occurring branched-7 chain sugars , stimulated considerable interest in devising syntheses of these compounds. For convenience, branched-chain sugars have been divided into two 8 groups since branching can occur on the carbon chain of the sugar either by substitution of a hydrogen atom (type A) or (less usual) of a hydroxyl group (type B), e.g. I I I H-C-OH —> R-C-OH or H-C-R ' (A) (B) Several syntheses of branched-chain sugars have been reported in the literature and the more important methods are outlined below. A f i r s t group of procedures u t i l i s e s derivatives of either methyl glycosiduloses or glycosuloses as intermediates. In the synthesis of (type A) sugars, suitably protected methyl glyco-sides have been oxidized to products with a carbonyl group in the sugar ring and subsequently allowed to react with Grignard 8 or organol^Min^' 1 1 I 8 reagents to afford the corresponding sugar, as shown in equation (1): ^ C H O H —»^ C = 0 —»N C R (0 H) (1) However, while giving the same type of sugars, these reactions differ in one important aspect, the configuration of the products. As an example methyl 3,4-0-isopropylidene- g-L-erythro-pentopyranosidulose(III)yields g with Grignard reagents the glycoside with the L-arabino configuration t 9 (IV)while the addition of organolithium results predominantly in the form-ation of the sugar with the L-ribo configuration(V). I l l IV V L-arabino L-ribo The condensation of a keto sugar with a Grignard reagent has been successfully applied by Dyer and coworkers 1 0 in the synthesis of streptoses Another method of synthesis of type A sugars has been used, namely the condensation of an oxo glycoside with diazomethane, followed by cleavage of the resulting epoxide, either with lithium aluminum hydride to give glycosides with a C-methyl substituent 1 1 or with a l k a l i to give a C_-hydroxy 12 13 methyl substituent ' (Equation 2). - 5 -LiAlH HO— C-CH 0 C=0 +• CH2N2 (2) The latter procedure was used successfully by two different groups of 12 13 investigators ' in the synthesis of hamamelose. Since i t was shown that the reaction of diazomethane with glycopyrano-12 siduloses lead to a mixture of isomeric epoxides and also under appropriate 14 conditions resulted in ring expansion of the glycoside, the alternative method using the dimethylsulphoxoni'um methylide^ for the epoxidation of these compounds has been examined by the same research group. The latter procedure applied to the carbohydrate derivative(III)gave after reduction of the epoxide mixture the branched-chain sugar with the L-arabino config-uration (IV) as major product. Application of this method to other keto sugars showed i t s generality. Available methods for the synthesis of type B branched-chain sugars are more limited. Very recently, Overend and coworkers^ reported on C-alkylation of methyl glycopyranosiduloses using barium oxide and methyl 17 iodide or the enamine alkylation method of Stork et a l . Both methods applied to methyl 4,6-0-benzylidene-3-deoxy-a-D-erythro-hexopyranosidulose (VI)afforded the branched-chain sugar(VII> A small amount of the second isomer (VIII)was detected in the pyrolidine enamine alkylation method. - 6 -C 6 H 5 - C H - 0 VI 0 CH iOCH5 C 6H 5-CH-0 CH5 0 0 C H 3 C6HsCH-0 O C H : VII VIII 18 A group of Russian investigators published in 1964 the synthesis of a type B sugar, having as branched-chain a carbomethoxymethyl group which sub-sequently could be reduced to the hydroxyethyl group. Their procedure consisted in the cleavage of an epoxide ring with the diethyl malonate carbanion, permitting the synthesis of methyl 4,6-0-benzylidene-2-C_-carbo-methoxymethyl-2-deoxy-u-D-altroside (X) from the epoxide (IX). o — C H C 6 H 5 - C H - 0 V - f O C H C 6H 5-CH -oV-%CH* R ~ - C H 2 C O O C H 3 ix x Specific branched-chain sugars of this type have also been obtained by ring contraction of pyranosides to furanosides. As an example the prepara-tion of 3-deoxy-3-C-substituted hexose sugars has been described by either the nitrous acid deamination of methyl 3-amino-3-deoxy-a-D-glucopyranoside 19 20 21 (XI)or the solvolysis of methyl 3-0-nitrobenzene sulfonyl -a-D-glucopyranoside ' H XI X = NH2, R=CH2OH XII X = ONs, R=CH2OH XIII X = NH2, R=H An application by Reist and coworkers of the Japanese method in the synthesis of branched-chain erythrofuranose sugars from the pentose (XIII)led to the f i n a l synthesis of branched-chain nucleosides. In view of the synthesis of the structural analog of adenosine in which the 3'-hydroxyl group is replaced by a hydroxyethyl group, a suitable derivative of the hitherto undescribed 3-deoxy-3-C-(2'-hydroxyethyl)-D-ribo-furanose was required. Branched-chain carbohydrates mentioned previously can be synthesized by several procedures. Methods available for the synthesis of type B sugars are limited and among them, we were unable to find an easy method affording the desired product. A thorough search, however, in the literature led us to the following procedure. An important method permitting the substitution of a hydroxyl group by a hydroxyethyl group in a ring system is the application of a Wittig reaction to a cyclic ketone. In our proposed synthesis the starting material for such a reaction would be the carbohydrate 1,2:5,6-di-0-- 8 -isopropylidene-g-D-ribo-hexofuranos-3-ulose ( I ) . 2. Methods of oxidation 22 A number of procedures have been reported f o r oxidation of " i s o l a t e d " secondary alcohol groups to ketones i n blocked d e r i v a t i v e s of sugars. Up to 1963, the only methods a v a i l a b l e were oxidation with oxygen and platinum 23 24 25 oxide, chromium-trioxide p y r i d i n e , and lead tetraacetate-pyridine, but y i e l d s were often low with carbohydrate d e r i v a t i v e s and i n some cases the 26 reaction f a i l e d completely. Since then, much work has been done i n t h i s f i e l d , leading to the discovery of very important methods of oxidation. The f i r s t of a serie s of 27 new methods, also r e f e r r e d to as the "P f i t z n e r - M o f f a t t " technique uses a s o l u t i o n of N_,N_-dicyclohexylcarbbdiimide (DCC) i n dimethyl sulfoxide with phosphoric acid .or pyridinium t r i f l u o r o a c e t a t e as a proton source. This 26 28 technique, while producing good y i e l d s ' with many unhindered primary 29 and secondary alcohols, f a i l e d with compound (XIV). The f i r s t announcement of the preparation of the corresponding ketose (I) of compound (XIV) was made 30 by Overend et. a l . Oxidation of l,2:5,6-di-0_-isopropylidene-a-D-gluco-furanose with ruthenium tetroxide i n carbon t e t r a c h l o r i d e gave the ketose (I) i n 75% y i e l d . 30,32 O o o CH 3 CH 3 XIV I 31a Separately, Theander reported i t s synthesis in very low yield (6%) using.the chromium trioxide pyridine.complex in acetic acid. One year 33 later, Japanese investigators found that dimethyl sulfoxide containing phosphorus pento/xide rapidly oxidizes the alcoholic groups of sugars at room temperature to give aldehydes, ketones or carboxylic acids. When this method was applied to compound (XIV) 65% of the corresponding ketose (I) was obtained. Very recently, this same group found that for most of the derivatives studied the oxidation proceeds most efficiently when N,N-dimethyl formamideis used as a solvent with dimethyl sulfoxide and phosphorus 34 pent.O'Xide. An exception was however noticed with the glucofuranose derivative (XIV) in which case a lower yield was obtained. About the same 33 35 time when Onodera's results appeared, Albright and Goldman reported that dimethyl sulfoxide acetic acid anhydride mixtures are effective for 36-38 oxidation of alcohols. Since then, several papers have been published on the application of this method to carbohydrates. In particular when l,2:5,6-di-0-isopropylidene-a-D-glucofuranose (XIV) was used as starting 36 material 60% of the ketose (I) was obtained. 30 A modification of the ruthenium tetroxide method has also been used 38 in the synthesis of the required ketose. Parikh and Jones found that a 30 combination of the two steps required in the original method of oxidation was more advantageous and gave good yields of oxidized products, 85%-90% in the case of compound (XIV). - 10 -3. Wittig reaction 39 As mentioned earlier, the Wittig reaction is a very useful tool to organic chemists for introducing a side chain. This reaction involves a condensation elimination between a phosphonium y l i d and an aldehyde or ketone to form an olefin and a phosphine oxide (Equation 3). R . (C,H ) P+-C HY + D ^C=0 • YHC=CR,R0 + (C,Hc)_P0 (3) b o o *2. Y=H or electron withdrawing group Historically, the f i r s t condensation between a carbonyl compound and a 40 41 phosphonium y l i d was reported in 1919 by Staudinger and Meyer . This • condensation then lay dormant in the literature for a few decades unt i l 39 42 Wittig and his collaborators ' developed i t and exploited i t s usefulness. The birth of the now well known Wittig reaction, came with the discovery that methyltriphenylphosphonium iodide could be converted to methylene t r i -phenylphosphorane, ( C ^ ) 3P=CH2, with phenyllithium and that i t would react with benzophenone to afford triphenylphosphine oxide and diphenylethylene in 84% yield (Equation 3, R^R^C^) . In view of i t s synthetic importance the Wittig reaction has been reviewed 53 and ref therein several times ' and i t is not our purpose here to make another review of this widely used reaction. The following pages w i l l deal primarily with the application of the Wittig reaction and i t s phosphonate modification in carbohydrate chemistry. In the latter part a brief history w i l l be given together with a summary of what is now known concerning the phosphonate carbanion, the mechanism, and the stereochemistry of this type of reaction. - 11 -3.1 Wittig reaction in carbohydrate chemistry Although the Wittig reaction has been used extensively in organic chemistry, very few chemists applied i t in the carbohydrate f i e l d . The f i r s t use of this reaction in carbohydrate chemistry was reported in 1963 < 43 44 45 by Kochetkov and coworkers ' ' even though the reaction with 2,3-0-46 isopropylidene-D-glyceraldehyde was already known. These investigators 43 were able to obtain higher a,6-unsaturated aldonic acids from protected 44 and non-protected monosaccharides. Condensation of the stabilized y l i d , carbethoxymethylene triphenylphos-phorane with aldehydo sugar acetates, in particular ^-arabinose, D-glucose 45 and D-galactose acetates gave 80% of trans unsaturated ethylaldonates 43 (Equation 4). In the case of tetra-O-acetyl-aldehydo-L-arabinose a trace of the cis isomer was also detected. COOEt I CHO CH I II (CH0Ac)T . + (C/:Hr),P=CHC00Et • CH (4) I J3or4 ^ 6 5 3 | CH„0Ac (CHOAc) . 2 | 4 CH20Ac When unprotected monosaccharides (D-arabinose, D-glucose etc.) were 45 condensed with the Wittig reagent, unsaturated ethylaldonates were obtained in addition to a considerable amount of by product having no double bond. These unsaturated aldonic esters obtained via the Wittig reaction have 45 been used as key intermediates in the general route to higher aldoses. Hydroxylation of their double bond, followed by reduction of the ester group gave the expected aldoses lengthened by two carbons (Equation 5). - 12 -RCH=CH-COOEt *• RCHOH-CHOH-COOEt RCHOH-CHOH-CHO (5) R=carbohydrate residue The synthesis of unsaturated higher ketoses has been reported by Zhdanov 47 48 and coworkers ' , who condensed aroylmethylene triphenyl phosphorane with an aldehydo-D-hexose-penta acetate (Equation 6). R I C=0 i CHO CH I II (CHOAc). + (C^Hr),P=CH-C-R • CH + (C^Hr),PO (6) I 4 6 5 3 || I 6 5 3 CH2OAc 0 (CHOAc) ' CH2OAc This reaction enables olefinic bonds to be introduced into monosacchar-ides which may then be converted into 2,3-di-deoxy derivatives. 49 This same group also reported the synthesis of aldonic acids from acyclic aldoses. By keeping triphenylphosphine with ethyl bromoacetate in ethanol followed by addition of sodium ethoxide and penta-O-acetyl-aldehydo-D-glucose they obtained the corresponding unsaturated octanoates. This technique, however, while being simpler had the disadvantage of giving a 45 lower yield compared to the yield obtained by Kochetkov's group. The Wittig reaction has also been applied in the synthesis of (^-glyco-sides"'0 of anthrone and fluorene. As an example, an anthrone glycoside was prepared by condensing anthronylidene triphenylphosphorane with penta-O-acetyl-aldehydo-D-galactose. Condensation involving a C-5 aldehydo group instead of a C-l as seen 51 previously has also been reported. Gigg and coworkers were able to condense - 13 -the aldehyde (XV)either with the phosphorane prepared from n-pentadecyltriphenyl-phosphonium bromide and phenyllithium, or with ethoxycarbonylmethylene 52 triphenylphosphorane (Ph3P=CHC02Et) to afford the corresponding unsaturated compounds (XVI) and (XVII). v R ^ ° \ O C H 3 O N \ // Ph XVI: R=-CH=CH-(CH ) CH XVII: R=-CH=CH-C02Et 3.2 Phosphonate modification of the Wittig reaction The phosphonium ylids used in the Wittig reaction have been divided into two categories, the stabilized and the non-stabilized ylids. Examples of each kind have been used in carbohydrate chemistry. The existence of 1 these phosphonium ylids and the st a b i l i t y of some of them has been attributed to the structural and electronic factors which contribute to the stabilization of the y l i d i c carbanion. On a theoretical basis and from experimental 53 evidence available, i t has been shown that ylids can be obtained from any phosphorus system which has a hydrogen atom adjacent to a phosphorus atom carrying a reasonable degree of positive charge, as in the case of phosphine oxides (XVIII} phosphinates (XlX)and phosphonates (XX) groups : - 14 -R 0 R 0 RO 0 R C RO C R O / x C " \ " \ - \ XVIII XIX XX This last group being of great interest since i t has been shown that phosphonate carbanions are more reactive than are the corresponding carbanions 58 in the olefin forming reaction. The historical part of phosphonate carbanion dates back to 1927 when 54 Arbuzov and coworkers found that carbethoxymethyldiethyl phosphonate could be converted into i t s carbanion by treatment with metallic sodium or potassium (7): 0 0 11 Na f " (C2H50)2P-CH2COOC2H5 — • (C2H50)2P-CH-C00C2H5 (7) Like phosphonium carbanions, phosphonate carbanions can undergo two types of reactions. Those in which only the carbanion is involved mechanistically and those in which both the carbanion and the heteroatom portion are involved. A l k y l a t i o n ^ 4 ' 5 5 , Michael type addition 5^ to conjugated carbonyl systems are examples of the former type. Reaction v/ith carbonyl compounds in a manner similar to the phosphonium carbanions to afford olefins and diethyl phosphates are examples of the latter (Equation 8). 0 • 0 t - R i \ t (RO)2P-CHY + ^C=0 • YHC=CR R + (RO) P=0 (8) - 15 -These condensations have often been referred to in the literature as "phosphonate modification of the Wittig reaction". The f i r s t reported reaction of this type is due to Horner and co-57 workers who obtained triphenylethylene in quantitative yield by condensing diethylbenzylphosphonate with benzophenone in the presence of sodamide, CO Equation (8): R=Et R1=R2 Y=C6H5- Wadsworth and Emmons found that phosphonate carbanions containing electron withdrawing groups were good reagents in the conversion of aldehydes and ketones to olefins, and in comparison to triarylphosphoranes (Wittig reagents) were more reactive and had a number of special features which enhanced their u t i l i t y (less expensive, milder conditions etc.). The use of different phosphonates, like carbethoxymethyl^ (XXI), acetonyl^ (51 62 (XXII} cyanomethyl ' (XXIIl)and diethylphosphonates permitted the synthesis of a Variety of olefin derivatives (9): f n base \ (C2H50)2P-CH2Y - J • OCHY (9) 2) \ j OO XXI: Y= -C00C2H XXII: Y= -C0CH3 XXIII: Y= -CN Solvents encountered in these condensations may vary from tetrahydro-furan (THF) arid dimethylformamide (DMF) to benzene, and bases from sodium hydride to sodamide and sodium ethoxide. Like the Wittig reaction the phosphonate modification plays an important role in the synthesis of natural products. An application of this method - 16 -63 is found in the preparation of vitamin A derivatives published by Pommer in 1960. The same route was used three years later by Openshaw and 64 Whittaker in their synthesis of emetine. Reaction of carbomethoxymethyl-diethylphosphonate anion with the keto derivative gave after hydrogenation the corresponding saturated ester (10): \ \ \ C=0 • C=CH-C00CHT • CH-CHo-C00CrL (10) / / 3 / 2 3 When carbomethoxymethyltriphenylphosphorane, (C^H^) gP^H-COOCH^, was used instead, no reaction occurred, however, a temperature of 150° gave the expected unsaturated ester in 58% yield. Later, Szantay and coworkers^** carried out with success the condensation of the same ketone with the carbethoxymethyldiethylphosphonate anion. Phosphonate anion The phosphonate carbanion can be generated from different systems like sodium hydride in dimethoxyethane^8'^, sodium methoxide in dimethyl formamide^ or potassium t-butoxide in dimethylformamide.^ The difference in i t s reactivity compared to the corresponding phosphonium carbanion has been related to their difference in nucleophilicity As mentioned at the beginning the sta b i l i t y of the phosphorus y l i d + -X-CR2 is attributed to the structural and electronic factors which contri-bute to the stabilization of the y l i d i c carbanion. This stabilization can be afforded by both the hetero-atom portion (X) and the two carbanion substituents (R). Since the latter substituents (R) are the same in the phosphonate carb-anion and the corresponding phosphonium carbanion the only difference in 53 sta b i l i t y is due to the hetero-atom portion (X). It is now believed that the a b i l i t y of the phosphorus to expand i t valence shell using the vacant energy 3d-orbitals for overlap, with the f i l l e d 2p-orbitals of the carbanion in a form of TT bonding, plays an important part i n the sta b i l i t y of the phosphorus y l i d . If we consider the phosphonate carbanion as a resonance hybrid having 58 the contributing structures , a,b, and c, and the phosphonium carbanion with the contributing structure e and f, the phosphonate w i l l have a lower net positive charge and thereby afford less stabilization for the carbanion by valence shell expansion than a phosphonium carbanion. Consequently, a phosphonate carbanion would be expected to be more nucleophilic and hence 53 more reactive with carbonyl compounds than a phosphonium carbanion. RO R RO R RO R RO - 18 -Mechanism and stereochemistry The mechanism of the phosphonate modification of the Wittig reaction, 53 58 67 according to many authors, ' ' is probably analogous to that of the Wittig reaction i t s e l f , The betaine is lik e l y formed by nucleophilic attack of the carbanion on the carbonyl carbon of the aldehyde or ketone and then probably undergoes decomposition to olefin and phosphate by oxygen transfer to phosphorus via a cis elimination. The driving force is the formation of a new phosphorus-oxygen bond in the phosphate. The betaine, however, may be able to dissociate to carbanion and carbonyl compound; no evidence on the reversibility of the f i r s t step has been found so far. R erythro cis 2 • R' = large group; R = small group - 19 -The stereochemistry of o l e f i n s produced by the Wittig r e a c t i o n has , „. - i . .. j53 and r e f therein „-been extensively studied . The r e a c t i o n normally r e s u l t s i n a predominance of the trans o l e f i n i n the mixture. The stereochemistry of o l e f i n s produced by the reaction of phosphonate carbanions with aldehydes or ketones has not been adequately studied, but the observations that have been made indic a t e that t h i s reaction r e s u l t s 53 62 67 68 i n an even greater predominance of the trans isomer. ' ' ' Wadsworth 67 and coworkers i n t h e i r study of the stereochemistry of the phosphonate modification of the Wi t t i g r e a c t i o n i l l u s t r a t e d i t s s t e r e o s p e c i f i c i t y and made the following observations: "In contrast to the Wittig r e a c t i o n , the use of s t e r i c a l l y and e l e c t r o n i c a l l y d i f f e r e n t s t a r t i n g materials produces only minor changes i n the s t e r e o s p e c i f i c nature of the phosphonate carbanion r e a c t i o n . Other factors a l t e r i n g the s t e r i c nature of the products from the Wittig r e a c t i o n , such as r e a c t i o n media and Lewis base additives have been i n e f f e c t i v e i n changing the tendency of the aldehyde-phosphonate condensa-t i o n to produce trans o l e f i n s " . However, i t should be emphasized here that among t h i r t y s i x examples given only two involved ketones. Bergelson 69 and Shemyakin noted the same phenomenon when using the carbethoxymethyl-diethylphosphonate carbanion with aldehydes. However recently Jones and 62 Maisey found novel s t r u c t u r a l e f f e c t s on the stereochemistry of the Wittig r e a c t i o n with d i e t h y l cyanomethyl phosphonate and alkylphenyl ketones. In these reactions the proportion of c i s o l e f i n was dependent on the structure of the ketone. 53 C l e a r l y , as pointed out by Johnson i n h i s book on y l i d chemistry there i s a need for considerable experimental i n v e s t i g a t i o n i n the stereo-chemistry of t h i s r e a c t i o n . - 20 -The numerous advantages of the phosphonate modification of the Wittig reaction in general and the good yields obtained by condensing the carbo-methoxymethyldiethylphosphonate anion with a keto alkaloid*^ were at the origin of our decision to use the latter method in the carbohydrate f i e l d . The results and the discussion concerning the reaction of carbomethoxy-methyl trimethylphosphonate anion with 1,2:5,6-di-0-isopropylidene-ct-D-ribo-hexofuranos'- 3-ulose w i l l be given in the next chapter.. 4. Oxo reaction As mentioned earlier, a second method was tried in order to introduce a hydroxymethyl group on carbon-3 of the sugar moiety. The reaction known 70 71 as the "oxo reaction" ' consists in the treatment of an olefin with carbon monoxide and hydrogen in presence of a cobalt catalyst. The oxo reaction i s now known to proceed in two stages.. In the f i r s t 72 stage called hydroformylation, hydrogen and a formyl group add to a carbon double bond to yield an aldehyde as shown in equation (11): In the second stage, the aldehyde is further reduced to the alcohol as shown in equation (12): RCH = CHR + H 2 + CO cobalt RCH2CH R CHO (11) catalyst RCH2CH R CHO + H 2 +• RCH2CH R CH20H (12) The expression "hydro(hydroxymethyl)ation" has been suggested to describe the overall conversion of olefins to alcohols. Preformed dicobalt - 21 -71 octacarbonyl is normally used as catalyst. However, a considerable amount of evidence has been accumulated that, indicates that cobalt hydrocarbonyl HCo(CO)^, i s effectivein i n i t i a t i n g the reaction.^ In general, temperatures between 75 and 200° and pressures of synthesis gas from 100 to 300 atmospheres are employed, higher temperatures being usual when alcohols (rather than aldehydes) are the desired products. Historically, the f i r s t application of this reaction to unsaturated 75 natural products namely, carbohydrates is due to Rosenthal and coworkers, who found that the oxo reaction represented an additional method for 7 6 lengthening the carbon chain of glycals. The scope and usefulness of this . . , ., , . _ T . 75 and ref, therein „, . reaction were described in Rosenthal's papers . This reaction can now be considered (with some limitations as we w i l l see later in the discussion) as a new approach for lengthening the carbon chain, or inserting a branched-chain (hydroxymethyl group) into the carbon chain of sugars. Since the application of the oxo reaction to carbohydrate deriva-73 tives has been well reviewed very recently, particular emphasis in the following pages w i l l be given only to points related to our objective, namely, the introduction of a hydroxymethyl group on carbon 3 of 1,2:5,6-di-0_-isopropylidene-a-D-glucofuranose (XIV). To achieve this purpose, a brief expose is necessary of the synthesis of 3-deoxy-l, 2: 5,6-di-0-isopropylidene-a-D-erythro-hex-3-enose (IT), i t s precursor alcohol (XIV)and in addition the reactions of (II). H 3 C V / 0 f H 2 H j C T ^ O C H XIV o H5C C H 5 o C H 3 CH 3 - 22 -4.1 Unsaturated sugars: 3-4 enes Apart from the glycals and 2-hydroxyglycals, 2 , 3 or 3,4 unsaturated sugars 77 n a v e received scant attention in reviews. Knowledge in particular, of sugars having a double bond at C^ -C^  is s t i l l limited. Compound(15 was 78 mentioned for the f i r s t time in the literature as a byproduct in 20% yield 79 in the reaction of the 3-0-tosyl derivative of (XXIV) and hydrazine. Three 80 decades later, Weygand and Wolz found that i t could be more efficiently produced (65% yield) by heating the p_-tolylsulfonyl derivative of (XXIV) with soda lime in vacuum. 3 Since the reduction of A -glycosene derivatives in an alternative route to the synthesis of important 3-deoxyfuranoses, new methods of synthesi 81 have been tried. Zinner and coworkers reacted the tosylate (XXIV) with 8 2 sodium carbonate in vacuo at 210-220°. Prokop and Murray accomplished the synthesis of (Il)using the potassium hydroxide modification of Weygand and 80 Wolz. The high specificity of this reaction may certainly result from the trans relationship of the p_-tolylsulfonoxy grouping and the hydrogen 83 atom at C-4. Similarly, Goodman and coworker found a 3-4 ene as a by product in the reaction of a 3-0-mesylate of another furanose derivative with sodium benzoate in dimethy1formamide. In the pyranose series, 3-4 unsaturated compounds have been obtained from aliphatic and bicyclic carbonyl compounds by way of the desired p_-84 tolylsulfonyl hydrazone. This method permitted the synthesis of the hex-3,4 enose (XXVI) in 56% yield. A small amount of the 2,3-unsaturated isomer (XXVI1) was also isolated. More recently the same compound (XXVI) 8 5 i;as synthesized by another route, Demethane sulfonyloxylation of methane sulphonate ( X X V ) in the presence of potassium t>butoxide in dimethyl sulfoxid - 23 -gave after three hours at room temperature a crystalline mixture of (XXVI) and (XXVII) in a 2:1 ratio. Almost complete isomerization of (XXVII) to (XXVI) could be effected in the presence of potassium t-butoxide for two days at room temperature. O — CH / O - C H . CgHj-CH-O 1 f OCH O — C H 2 I—O OCH 5 CeH5-CH-C^C=/OCH H XXV XXVI XXVII 4.2 Reactions of the 3-4 enes A brief report here, of some of the reactions of 3-deoxy-l,2:5,6-di-0-isopropylidene-a-D-erythro-hex-3-enose(II) may be of interest for the sub-sequent discussion of the oxo reaction. 80 Weygand and Wolz have shown that on catalytic reduction compound (II) is converted stereospecifically into 3-deoxy-l,2:5,6-di-0_-isopropylidene-a-D-galactofuranose. It is evident that the severe steric hindrance by the isopropylidene group is responsible for this stereospecificity and would preclude any possibility of obtaining the other isomer by use of a reductive process with cis stereochemistry. Another cis addition to the 3,4-ene (II) was accomplished separately by 86 87 88 Paulsen and Behre and Lehmann using Brown's hydroboration. This reaction paralleled the catalytic hydrogenation since the only product 24 obtained after oxidative cleavage of the diborane adduct was a sugar with the galacto configuration. 89 Lehmann in his subsequent paper reported on the radical addition of toluene-a-thiol to compound (II) affording in equal amounts the gluco (XXVIII) and the galacto (XXIX) isomers. For the same reason as mentioned earlier, the attack of the free thio radical is only possible from the exo side since the endo side is too sterically hindered. /exo H2C~ C P 0 H 3 C C H , / CH; C H XXVIII 3 RH H3C OCM X 1 XXIX 0 CH, CH, gluco R = S-CH. galacto - 25 -4.3 Mechanism and stereochemistry Evidence for the mechanism of the oxo reaction as carried out under conditions of high temperature and pressure is largely speculative and is based on the results of reactions between olefins and cobalt hydrotetracarbonyl (Equation 13) at atmospheric pressure and room temperature: Co 2(CO) g + H 2 - 2HCo (CO) (13) The mechanism outlined here is largely based on Heck and Breslow's work The conversion of olefins to aldehydes is regarded as proceeding in three distinct stages: a) The formation of a TT complex between olefin and cobalt hydro-carbonyl, which rearranges so that a carbon-metal sigma bond is formed. 90 As suggested by Heck and Breslow, this f i r s t stage involves at least the three following steps (Equation 14 to 16): 90 HCo(C0)4 * HCo(C0)3 + CO RCH=CHR + HCo(CO), RCH=CHR HCo(CO) RCH2CHR Co(C0)3 RCH0-CHRCo(CO). + CO ^ * RCH--CHRCo(CO) (14) (15) (16) b) The insertion of carbon monoxide between metal and carbon (Equation 17). CO RCH2-CHR-Co(CO)4 * RCH2•CHR-CO-Co(CO)3 5==* RCH2'CHR-COCo(CO)4 (17) - 26 -c) Hydrogeriolysis of the resulting complex to give an aldehyde (Equation 18): RCH2-CHR-0000(00)2 + H 2 > RCKyCHR-CHO + HCo(CO)3 If. RCH2'CHR-C0Co(C0)4 HCo(CO) + CO HCo(CO) (18) 4 A scheme analogous to that already described for hydroformylation has been 91 proposed by Marko for the subsequent^ ., hydrogenation of aldehydes to alcohols under oxo conditions: RCHO + HCo(CO), H RC=0 I HCo (CO), RCH20Co(CO) (19) RCH20Co(C0)3 + H 2 RCH2OH + HCo(CO) i f RCH2OCo(CO)4 (20) Similar schemes could explain the "hydro(hydroxymethyl)ation of unsatur-73 ates c a r b o h y d r a t e s . . . 92 93 76 Many examples in the carbohydrate f i e l d ' ' and a few in the steroid 94 95 fi e l d ' , conclusively demonstrate that the oxo reaction proceeds by way of cis addition of a hydrogen atom and a hydroxymethyl group to the carbon-carbon double bond. Thus, one might expect a certain analogy in the course of the oxo reaction with 3-deoxy-l,2:5,6-di-O-isopropylidene-a-D-erythro-hex-3-enose (II) - 27 -compared to the other cis addition seen previously 80,86,87,89 4.4 Effect of the carbohydrate structure Addition of ionic reagents to double bonds of vinyl ethers has been explained by the mesomeric interaction of the ring oxygen atom, which gives rise to the following contributing structure. The ring oxygen of unsaturated sugars, namely glycals, has been seen to exert a similar directing influence in this.kind of reaction. In hydroformylation, i t has been shown that with unsymmetrical olefins, available evidence indicate that the formyl group adds to the least hindered side of the double bond under normal conditions of high temperature and 96 pressure. Of particular interest are the results obtained with cyclic vinylic ethers. A simple compound like furan reacts as a typical conjugated diene under oxo conditions; one double bond is hydrogenated while the other 97 undergoes the normal reaction with carbon monoxide and hydrogen (Equation 21): + CO + 3H, Co 2(C0) g ChLOH V (21) However, when both positions adjacent to the ring oxygen are blocked, 28 98 "hydro(hydroxymethyl)ation" occurred at the a l t e r n a t i v e s i t e (Equation 22) H 5 C A 0 A C H 5 /CM 20H (22) The same i s true i n the carbohydrate f i e l d , A p p l i c a t i o n of the oxo reac t i o n to the hex-2,3-enose (XXX) gave a mixture of anhydro deoxy branched-chain a l d i t o l s (XXXI) and the branched-chain sugar (XXXII) 93 The same re a c t i o n applied to a 2-hydroxyglycal d e r i v a t i v e (XXXIII) gave a f t e r deacetyla-99 t i o n a h e p t i t o l with the chain on carbon 1 (XXXIV). CH2OAc AcO O A c OAc XXX CH 2 OAc Os CH2OAc A c C \ l OAc XXXI + CW 2OAc — < \ C H 2OAc A cO N . / OAc OAc XXXII AcO CH 2 OAc " — Os OAc OAc XXXIII CH 2OH 0 K CH 2 OH XXXIV Although a l l simple o l e f i n s have been found to undergo the oxo rea c t i o n , the rate of the r e a c t i o n i s observed to be highly dependent on the structure - 29 -of the o l e f i n . 1 0 0 Branching of the carbon chain always results in a decrease in reaction rate. The rate observed for cyclic olefins are of particular interest, since this type of compound is directly related to our worki Traynham101 pointed out that in a l l known cases the more strained is a cyclic structure, the more reactive i t is in electron donating roles. One might therefore expect the TT electrons of cyclopentene to be more available for donation to the vacant d orbitals of cobalt in complex formation, compared to cyclohexene. Nucleosides: 5.1 Branched-chain sugar nucleosides: The term nucleoside usually encompasses compounds containing a nitrogen heterocycle (purine or pyrimidine and their close analogs) in glycosidic linkage with a carbohydrate moiety. In general, nucleosides have been divided structurally into two main groups, based on whether they contain purine or pyrimidine aglycones. Because of the biological properties of cordycepin, the former class of nucleosides, in particular adenosine analogs with a structural modification present in the carbohydrate moiety, led to our research. As seen previously the alteration consisted in introducing a branched-chain into the sugar moiety. Branched-chain sugar nucleosides are nucleosides in which such an alteration is present. The f i r s t synthetic branched-chain nucleosides have been reported only 102 two years ago. This fact is not surprising in view of the rather limited . ' , . , . 8,10,12,13,18,20 availability of branched-cham sugars. - 30 -Chemists from Merck, Sharp and Dohme Research Laboratories synthesized the f i r s t two branched-chain structural analogs of adenosine, namely the 2 ,-C 1 0 2' 1 0 3 and 3'-C 1 0 2' 1 0 4 methyl adenosines (XXXV) and (XXXVI): M 0 C R , / O (XXXV): R=R'=H, R"=CH, (XXXVI): R=R"=H, R'=CH, (XXXVII): R'=R"=H, R=CH, HO OH Both compounds were reported biologically active. More recently, Reist 139 and coworkers described the synthesis of 3'-deoxy-3'-C-hydroxymethyl-D-erythrofuranosyl adenine (XXXVIII) and 2'-deoxy-2'-C-hydroxymethyl-D-erythro-furanosyl adenine (XXXIX) in both anomeric forms. In these branched-chain nucleosides a hydroxyl group on the sugar moiety was replaced by a hydroxymethyl group. XXXVIII XXXIX - 31 -In contrast to the branched-chain nucleosides already mentioned where the branching involves one of the ring atoms, another type of nucleoside involving a branched-side-chain, namely 5',5'-di-C-methyladenosine (XXXVII) has been synthesized by Nutt and Walton 1 0 5. This nucleoside, like the other C-methyl derivatives of adenosine, (XXXV) and (XXXVI), was reported to be biologically active. 5.2 Nucleoside synthesis Nucleosides can be obtained by different methods. Besides the older and most extensively used reaction, consisting in the treatment of an acylglycosyl halide with a heavy metal salt of a base, other methods have been tried more recently. The so called fusion technique has been used in several syntheses.. Thus, condensation of l-()-acetyl derivatives with a base in presence of zinc c h l o r i d e 1 ^ or monochloroacetic a c i d 1 0 ^ or of an 108 oxazoline derivative in presence of p_-toluenesulfonic acid yielded the corresponding nucleosides. Nucleosides were also obtained by reacting 1-O-acetyl derivatives or acyl glycosylhalides with a purine base in 109 dimethylformamide in presence of phosphorus pentaoxide . Although the new methods have become available, most of the nucleosides, in particular branched-chain nucleosides, are s t i l l synthesized through condensation of acylglycosylhalides with heavy metal salts of the bases. Two important modifications of the early Fischer-Helferich1''' 0 procedure have been used concomitantly throughout the literature. The f i r s t modifica-tion introduced by Davoll and Lowy 1 1 1 was the substitution of the chloromercuri derivative of certain purines for their silver salt. - 32 -* O C H 2 o O A c OAc OAc AcOCH, O Cl OAc OAc NHBz AcOCH, ^  O Cl + N N N NHgCI OAc OAc HO CH2 ^  O OH OH 112 The second modification due to Baker and coworkers combined the two step synthesis into one step by using the titanium tetrachloride complex procedure: NHBz N N NWBz OAc NHAcOBz NHAc OBz J N N -Ti()Ac)Cl3 _ ^ 0 £ > 0 NHAc OH - 33 -The anomeric configuration of the nucleosides obtained from these 113 condensations have been explained in relation to Baker's "trans rule". "condensation of a heavy metal salt of a purine or pyrimidine with an acylated glycosyl halide w i l l form a nucleoside with a C-l, C-2 trans configuration in the sugar moiety regardless of the original configuration of C-l, C-2". The stereochemistry underlying this rule, that i s , the control of the entrv of the purine moiety by the 2-acyloxy group of the sugar has been reviewed 114 by Fox and Wempen . Although Baker himself found exception to the C-l, 112 139 C-2 trans rule, other exceptions have been reported since, the conclusion stating 1 1** that "C-l, C-2 cis configuration are not often observed, and when they are observed they occur in minor proportions" s t i l l remains true. Since these chloromercuri condensations are dependent on the purine 112 as well as on the glycosyl halide used, we w i l l report in the next pages only examples of condensations involving a specific and easily prepared base derivative, namely chloromercuri-6-benzamidopurine. Pentofuranosyladenine nucleosides have been reported by a great number of authors, and their synthesis has been reviewed. 1 1^' 1 1 (* Hexofuranosylnucleosides however, received less attention. Since one of our objectives was the synthesis of a hexofuranosyl branched-chain nucleoside, results obtained by authors in the hexose series might be of particular interest. The f i r s t glucofuranosylnucleoside was synthesized by Baker and 117 coworkers in 1958. Condensation of a presumed anomeric mixture of the chloro sugars (XL) with chloromercuri-6-benzamidopurine gave exclusively the - 34 -9-B-D-glucofuranosyladenine (XII) in 24% yield. LX XLI 118 Similarly, Wolfrom and coworkers a few years later obtained from tetra-O-acetyl-B-D-galactofuranosyl chloride only the 9-3-D-galactofurano-syladenine in the same yield. 119 Lerner and Kohn used 2,3:5,6-di-O-isopropylidene-a-D-manno-furanosyl chloride (XLII), a compound which cannot form an ortho ester, in the condensation with chloromercuri-6-benzanidopuririe a n d thus only isolated the corresponding a-nucleoside (XLIII). X = - C ( C H 3 ) 2 - 35 -They suggested that retention of configuration must imply an SNj mechanism with the formation of a single anomer resulting from steric hindrance due 120 to the bulky isopropylidene groups.. Lee and Nolan using essentially the same route arrived at the same conclusion. The more recent results of 121 Lerner and Kohn in the synthesis of 9-B-D-glucofuranosyl adenine and 9-a-L-lyxofuranosyl adenine suggested an analogous situation. In the f i e l d 82 of 3-deoxyfuranosyl nucleosides Prokop and Murray obtained by condensing chloromercuri-6-benzamidopurine with the diacetates (XLIV) and (XLV) in presence of titanium . tetrachloride the corresponding nucleosides (XLVI) and (XLVII) in 32% yield. MCOBz CH2OBz XLIV OAc OAc M C O H I CH zOH XLVI 8 2 OCH 2 OAc OAc XLV M O C H ^ XLVII - 36 -Although condensations of acylated sugars with chloromercuri-6-benzamidopurine in presence of titanium tetrachloride have been shown in 112 some instances to give rise to a mixture of a and 3 anomers, Prokop and 82 Murray only isolated the 3 anomer; this however, does not exclude the pres-ence in small quantity of the a isomer since the f i l t r a t e s were not investi-122 gated. Goodman and coworkers using Prokop and Murray's method obtained from the diacetate (XLVIII) only 3'-0-methyladenosine (XLIX) in a rather good yield (47%). MeO OAc MeO OH XLVIII R=Ac XLIX L R=C1 They noted at the same time, that with this type of compound the two step synthesis consisting in the condensation of the previously prepared chlorosugar with chloromercuri-6-benzamidopurine gave only 14% of the desired 3 nucleoside. 139 Reist and coworkers in their study on C_-3'-deoxy branched-chain sugar nucleosides found the following results. Condensation of the 1-0-acetyl erythrofuranose derivative (LI) (a and 3 anomers) with chloromercuri-6-benzamidopurine in presence of titanium tetrachloride gave after deacylation the a and 3 nucleosides (LII) in a ratio not exceeding 10 to 1 When the 3 sugar acetate (LIII) was converted f i r s t to the sugar halide (LIV) and then subjected to condensation with the base followed by - 37 -LII L I : X = OAc (a and B) R=Bz L I I I : X = BOAc R = £ - N 0 2 C 6 H 4 C O -LIV: X = C l R = £ - N 0 2 C 6 H 4 C O -d e a c y l a t i o n , the a and 8 nucleos ides (LII) were obtained i n a r a t i o of 4 to 1. It should be noted here , that the former condensation was c a r r i e d out i n the same r e a c t i o n mixture as the condensation of the 2'-deoxy d e r i v a t i v e (LV) and s ince the i n i t i a l r a t i o of the respec t ive sugar acetates were unknown, no comparison could be made between the y i e l d s obtained by t h i s one step procedure and the second method. LV LVI r - 38 III. RESULTS AND DISCUSSION 1. Wittig reaction The f i r s t step in the synthesis of hexo and pento-furanosyl branched-chain sugars and nucleosides i s the preparation of the carbohydrate in i t s furanose form. Since pentoses can be obtained by degradation of hexoses only one starting material, e.g. glucose is necessary for both syntheses. In the pentose series a sugar may be forced into i t s furanose form by suitable blocking of the terminal position, this procedure is however, not applicable in the hexose series where special methods are required to 118 123 • obtain furanoside derivatives. Wolfrom et. a l . ' proceeded by way of dithioa'cetal formation and furanose thioglycosides to effect the synthesis 124 of gluco and galacto furanosyl nucleosides. Carbonate esters and 125 y-lactones have also been used as a means of obtaining furanose rings of six carbon sugars. Likewise isopropylidene derivatives have been found 126 127 useful as i n i t i a l source of a number of 5-deoxy and 6-deoxyhexoses. Since isopropylidene cyclic acetals (1,2 and 1,2:5,6) possessing a 117 furanose ring in the glucose series have been successfully applied by Baker in the synthesis of 9-3-D-glucofuranosyladenine and more recently by Prokop 82 and Murray in the synthesis of 9-(3-deoxyaldofuranosyl)adenine, we used these derivatives in the present work. Three methods are usually available for the preparation of the isopropyli-dene compounds of sugars or sugar derivatives, procedures using either a - 39 -mineral acid, zinc chloride or copper sulfate as catalysts. In the present 128 work we have u t i l i z e d the procedure of Glen and coworkers, which con-129 sisted in the preparation of 1,2:5,6-di-0_-isopropylidene-a-D-glucofuranose (XIV) from D-glucose in presence of zinc chloride as catalyst. Yield, melting point and optical rotation data were in agreement with the values reported in the literature. 1.1 1,2:5,6-di-O-isopropylidene-ot-D-ribo-hexofuranos- 3 - ulose (I): We have seen in the introduction that the oxidation of l,2:5,6-di-0_-31a isopropylidene-a-Q-glucofuranose (XIV) with the Oppenauer reagent, lead 3 lb 2 9 tetraacetate and the Pfitzner-Moffatt reagent were reported to be 30 ineffective, while ruthenium tetroxide , ruthenium dioxide-sodium meta 38 36 periodate ,dimethyl sulfoxide-acetic anhydride or dimethyl sulfoxide 33 34 phosphorus pentoxide ' gave good yields of the desired ketose (I) . In this work, the preparation of 1,2:5,6-di-0-isopropylidene -a-D-ribo-hexofuranos-3-ulose (I) was based on the early note published by Onodera and collaborators in 1966. Since only the molar ratios of the carbohydrate and phosphorus pento xide were given with no experimental details except that the oxidation was done in dimethyl sulfoxide and at room temperature for 24 hours, the following precautions were taken: a) The entire reaction was carried out in the dry box. b) The solution of dimethyl sulfoxide containing diisopropylidene glucose was cooled to zero degrees before addition of the phosphorus pent o aide. c) The addition of chloroform and sodium bicarbonate solution at the - 40 -end of the reaction was done while maintaining the temperature below 20°. In an attempt to duplicate Onodera and collaborators results the same quantities of sugar (1 mmole) and phosphorous pentoxide 0.6 .mmole were used. Monitoring the reaction by thin layer chromatography s t i l l showed after 24 33 hours (normal reaction time) the presence of the starting material. When the reaction was allowed to proceed 24 hours longer no change occurred. The only way to complete the oxidation was upon addition of 0.5 mmole more of pentoxide. From the yields obtained in different runs i t was noticed that the dry box was superfluous and that the oxidation proceeded most readily when a slight excess of phosphorus pentoxide was used. By this procedure 1,2:5,6-di-O-isopropylidene-a-D-ribo-hexofiiranos -3-ulose (I) was 33 34 obtained in the same yield as reported in the literature. ' It is of interest to note-here that in a very recent publication, 34 Onodera et a l . , also reported the use of an excess of phosphorus pent-oxide (2 molar equivalents instead of 0.6) in the preparation of the carbohydrate ketose (I). After work up of the reaction mixture the crude product isolated did not contain more than 5% impurities, as demonstrated by thin layer chromato-graphy and the nuclear magnetic resonance spectroscopy. The signal of the anomeric hydrogen appeared as a doublet at T 3.94 with a ^ coupling constant of 4.5 Hz. These values are in agreement with those reported by 32 Overend and coworkers. The H-2 signal was located under a 3H multiplet at x 5.7 as proved by double irradiation experiment. Irradiation at T 5.7 collapsed the doublet at T 3.94 into a singlet. The infrared spectrum showed a strong carbonyl absorption at 1770 cm Purification by several recrystallizations from light petroleum ether gave a crystalline monohydra.te, with no carbonyl stretching in the infrared but with an OH absorption at - 41 -3450 cm"1. 1.2 Wittig reaction of 1,2:5,6-di-0-isopropylidene-a-Q-ribo-hexofuranos-3-ulose (I) to yield 3-£-(carbomethoxymethyl)-3-deoxy-l,2:5,6-di-0-isopropylidene-a-D-allofuranose The importance of the phosphonate modification of the Wittig reaction in the preparation of olefin derivatives has already been shown in the introduction. Since our main objective was the synthesis of a branched-chain sugar having on carbon-3 a hydroxyethyl group, we f e l t that the condensation of the ketose (I) with an appropriate Wittig reagent like carbomethoxymethyl-dimethylphosphonate (LVIII) might yield an unsaturated carbohydrate with a carbomethoxymethylene group as a branched-chain, which could be subsequently hydrogenated and reduced to the desired hydroxyethyl group. As mentioned in the Introduction such condensations are known and have been successfully 64 employed in the alkaloid f i e l d . Openshaw and Whittaker and later Szantay and coworkers^5 condensed a keto alkaloid with a phosphonoacetic ester anion, generated in one case from sodium hydride in dimethoxyethane, 64 or sodium methoxide in methanol, and in the other case from potassium t-butoxide in dimethyl sulfoxide.^ 5 In the present work, the condensation of 1,2:5,6-di-0_-isopropylidene-q-D-ribo-hexofuranos-3<-ulose (I) was based on Szantay and coworkers^5 procedure. However, in view of the fact that to our knowledge no carbohydrate ketose was ever condensed with a Wittig reagent, the condensation of the ketose (I) in dimethyl sulfoxide with carbomethoxymethyldimethylphosphonate (LVIII) in presence of potassium t-butoxide, was followed carefully by thin layer chromatography (t.l.c.) on s i l i c a gel G using benzene and methanol 95:5 as developer. Examination of the chromatogram with a sulfuric acid spray Me 0— C H X I n Me 0 — C H / ° -c o 4 - r VC0 CH ' 2 3 Me LX Me 0 — C H X • Me 0 —C H C H 30 2 C - C I I H - Pd 2 Me 0 — C H , X l ! Me 0 — C H C H 30 2 C C H 2 0-|-Me LXI Me - 43 -revealed one major compound with an Rf of 0.84 and three less mobile components. After a period of 48 hours, only one byproduct was present (Rf=0.60). The reaction mixture was then worked up according to the method described by Szantay, that is by precipitation of the reaction products with water followed by their extraction with ether affording a yield of 60% of unsaturated sugars. This yield however, was considerably improved by f i r s t evaporating the solvents of the reaction mixture in water, in this way a yield of 80% was obtained. The crude product thus isolated showed absorption bands between 1725 cm 1 and 1650 cm \ at a lower frequency than the starting material therefore indicating that condensation had taken place. Further confirmation was given by the n.m.r. spectrum of the crude mixture which showed two unsaturated products. We have pointed out in the introduction, that the phosphonate modification of the Wittig reaction applied to aldehydes and ketones yields a mixture of cis and trans isomers with respect to the ethylenic bond,^ the major compound being the trans isomer. Consequently, the components of the mixture can only be the two geometrical isomers of the a,g-unsaturated methyl ester (LIX) and (LX). This fact was confirmed by the results of a subsequent reduction as we shall see later. The stereochemistry of the olefins produced by the reaction of phosphonate carbanions with aldehydes and in particular ketones has not been adequately studied, and since a recent report emphasized the importance of the structure in the case of ketones, we thought that i t would be of interest to know the isomeric composition of the olefins obtained from the specific carbohydrate ketose (I).• 67 Examples from the literature have shown that when ketones are concerned, considerable amounts of cis isomer can be isolated. Condensation - 44 -of the ketone (LXII) with carbethoxydiethylphosphonate anion gave 16% of the cis and 18% of the trans isomer, likewise the ketone (LXIII) gave under the same conditions 19% of the cis isomer and 40% of the trans. C 2 H 5 COOCH 2 " W H2 C » O \ o / / LXII LXIII Openshaw and Whittaker^ 4 as well as Szantay and coworkers^5 reported the isolation of both stereoisomers, however respective yields were not given with certainty. In order to determine in our work the isomeric composition of the a,3-unsaturated ester, a small quantity of the crude product was chromato-graphed and the major zone eluted as quickly as possible to avoid decomposi-tion, affording a pure mixture as evidenced by the disappearance of the extraneous peaks in the n.m.r. spectrum. Hydrogens of a signal strength equivalent to 3/4 of a proton and others to 1/4 of a proton could be different-iated very easily in the spectrum. Figure (1) on page 45 represents an n.m.r. spectrum of a mixture of two isomers in which the second isomer accounts for 1/5 of the reaction mixture. This spectrum shows more clearly the signals corresponding to each isomer. In the following discussion the derivative having the carbomethoxymethyl group trans to the C^ -C^  isopropylidene group w i l l be referred to as the trans isomer (LX) and the other (LIX) as the cis isomer. A separation of the two isomers was attempted on a small scale using the same solvent system as mentioned Figure 1 . Proton magnetic resonance spectrum at 100 MHz/sec in deuteriochloroform of a mixture of the two stereoisomers of 3-C-(carbomethoxymethylene)-3-deoxy-l,2:5,6-di-O-isopropylidene-a-Q-allofuranose. - 46 -earlier. The zone corresponding to the mixture which on detection with ' sulfuric acid (30%) gave one spot, showed under UV light two different absorbing zones. The less mobile zone B being the darkest. We therefore eluted both sections separately. Their n.m.r. spectrum, however, indicated that each was in fact contaminated by the other, and by comparison with the n.m.r. of the pure mixture, fraction A was found to be the major product of the condensation. Since in the process of separation, the unsaturated sugars were found to be unstable and since our main objective was the synthesis of a saturated branched-chain sugar, further attempts of separation in order to obtain both isomers in a pure state were abandoned. However, as signals of both isomers occur at different chemical shifts, the presence in minor quantity of the second isomer did not interfere with the hydrogen assignments. Signals of the major compound A were assigned on a f i r s t order basis in the following way: 1) irradiation of H-l 1; at T 3.76 (see Fig. 2, page 47) changed the H-2 signal into two doublets (J =4 Hz, J_ . = 1.25 Hz) and the H-4 Z , I Z , 4 signal into two " t r i p l e t s " , 2) irradiation of H-4 changed H-l' into a doublet (J^, 2 = 1.25 Hz) and the H-2 signal into two doublets ( J 2 ^  = 4 Hz and 1 ( = 1.25 Hz) 3) irradiation of H-2 changed H-l' into a doublet (J j , 4=2 Hz) and H-4 into two "doublets" (J ., = 2 Hz). The signals of isomer B were assigned as follows. The tr i p l e t a t T 3.82, is attributed to H-1I0(J.., = 1-5 Hz), while the doublet at T 4.11 is assigned l , z to H-l ( J l f 2 = 4.75 Hz). The sextuplet at T 4.94 is attributed to H-2. Double irradiation of B as indicated below gave the following results: - 47 -M e w O - C H 2 X 1 Me O-CH C\ H H H - C \ 0 O f M e C O , C H 3 M e Mev / 0 - C H 2 X 1 Me N 0 - C H O V H C H 30 2 C - C O f Me H Me Wl V . H-l' m il 3.76 3.82 '.I 4.21 42S -J J-U :4> H-4 5.36 Figure 2. P a r t i a l proton magnetic resonance spectrum of (LX + LIX) i n deuteriochloroform ((a) at 100 MHz/sec; (b) at 100 MHz/sec, H-4 proton i r r a d i a t e d ; (c) at 100 MHz/sec, H-l' proton i r r a d i a t e d ; (d) at 100 MHz/sec, H-2 proton i r r a d i a t e d ) . - 48 -1) irradiation of H-l' changed H-2 into two doublets (J„ . = 1.50 Hz; £ , 4 J ? , = 4.75 Hz),2) irradiation of H-2 changed the H-l' t r i p l e t into a doublet ( J 1 1 > 4 = 1.50 Hz). A comparison of the n.m.r spectrum of the isomers A and B is best summarized in the following table: TABLE 1 H-l' H-l H-2 J l ' , 2 H z J1',4 H Z Isomer A 3.76 4.21 4.29 1.25 2.00 Isomer B 3.82 4.11 4.94 1.50 1.50 From these data, two major facts emerged; f i r s t l y the signals of H-l' and H-2 of A occur at a lower f i e l d than the corresponding signals of isomer B, secondly the long range coupling constants J^, ^  and J^, ^  have equal values .in isomer B, which is not the case in isomer A. Since not enough work has been reported in the literature in the f i e l d of long range coupling, we have made no attempt tq assign a definite structure, cis or trans to isomer A, on the basis of the n.m.r. spectrum. From a molecular model, i t seems that steric hindrance due to the free rotating 5,6-isopropylidene group is more important than that of the fixed 1,2-isopropylidene group. If this is true, the trans isomer may be the major compound obtained from the phosphonate modification of the Wittig reaction applied to the carbohydrate ketose (I); in other words, A might be the trans isomer. In a subsequent reaction the pure isomeric mixture of unsaturated branched-chain sugars was reduced over 10% palladium on charcoal catalyst in ethanol to yield compound (LXI). The hydrogenation was followed graphi-- 49 -cally (volume of hydrogen absorbed vs. time) and the reaction mixture worked up immediately after having reached the "plateau" of the curve, correspond-ing to the absorption of one mole equivalent of hydrogen. The crude crystalline compound isolated from the reaction mixture was homogeneous on t . l . c . It had the same Rf as the starting material. The reduced product could be easily differentiated from the unsaturated ester; the latter exhibited a pink color on s i l i c a gel, when sprayed with a 30% solution of sulfuric acid. The IR spectrum showed only one band in the carbonyl region at 1750 cm 1 characteristic of a saturated ester and no absorption in the region from 1725 to 1650 cm The n.m.r. spectrum (Fig. 3 page 50) was in agreement with the above results since no extraneous peaks corresponding to a second isomer were observed. The protons signals were assigned as follows: (a) the low f i e l d one proton doublet at x 4,29 (J, 7 = 3.7 Hz) to H-l; (b) the one proton t r i p l e t at T 5.25 ( J 2 3 = 4 H z) to H-2; (c) the high f i e l d one proton multiplet at T 7.64 to H-3, were in agreement with the decoupling experiments. Irradiation of H-l changed the H-2 t r i p l e t into a doublet and irradiation of H-2 collapsed H-l into a singlet and altered the multiplet of H-3. According to these results H-2 is coupled to H-l and to H-3. The coupling constants ^ = 3.7 Hz and J 2 3 = 4 , 0 ^ z a r e °^ t^ie 130 same order of magnitude as are usually encountered for cis hydrogens• Similar coupling constants were observed with l,2:5,6-di-0_-isopropylidene-a-131 D-allofuranose (LXIV) and with 3-deoxy-3-hydrazino-l,2:5,6-di-0_-132 isopropylidene-a-D-allofuranose(LXV) . From the above mentioned similar-i t i e s in their J„ , coupling constant between the branched-chain sugar (LXI) and (LXIV) and (LXV), and from the fact that reduction of a double-bond adjacent to a bridge head usually proceeds by addition of hydrogen from the Me 0 - C H 2 X ' M e / N 0 - C H XI H H^40 CH 2 O f Me | Me C0 2 CH 3 H-l (4.29) H-2(5.25) J | 2(3.7 Hz.) I J2,i(3.7 Hz.) |i|J2t3(4.0Hz.) -COOCH, 1 H-3 O -Vrf v -8 10 Figure 3. Proton magnetic resonance spectrum at 100 MHz/sec of 3-C-(carbomethoxymethyl)-3-deoxy-1,2:5,6-di-O-isopropylidene-a-D-allofuranose in deuteriochloroform. - 51 -"top side" compound (LXI) is certainly 3-deoxy-3-C-(carbomethoxymethyl)-1,2:5,6-di-0_-isopropylidene-a-D-allofuranose. However in order to establish unequivocally the configuration at C-3 an X-ray analysis is necessary, which is presently under investigation. 1.3 3-Deoxy-3-C_- (2' -hydroxyethyl) -1,2:5,6-di-O-isopropylidene-a-D-allofuranose Reduction of the saturated ester (LXI) with lithium aluminum hydride in tetrahydrofuran at room temperature according to a method previously described for the preparation of methyl 4-£-methyl-a-Q-glucopyranoside from 133 the corresponding uronate derivative afforded even after a reaction time of one hour instead of 15 minutes, two compounds in equal ratios, with Rf of 0.39 and 0.80 as shown by t . l . c . on s i l i c a gel using benzene and methanol (92:8) as a developer. At.this stage of the reaction the mixture was divided into two portions. The f i r s t portion was worked up immediately as described in the literature affording a colorless syrup. Chromatography on preparative t . l . c , followed by the elution of the two zones (Rf:0.80,Rf:0.39) gave fractions A and B respectively which were analysed. Fraction A was found homogeneous on t . l . c and different from the starting material as evidenced by their different Rf's in the same solvent system. The IR spectrum showed an absorption at 1765 cm 1 probably a carbonyl, but no hydroxyl. The follow-ing characteristic signals were noticed in the n.m.r. spectrum. One proton doublet at T 4.28 (J _ = 3.7 Hz) H-l, a one proton t r i p l e t at T 5.36, H-2 a six proton multiplet at T 7.96 consisting of a sharp singlet most probably a methyl group overlapping a multiplet and the expected four methyl peaks of the isopropylidene group at high f i e l d . From these results, structure (LXVI) has been tentatively assigned to fraction A. Fraction B, with an Rf= 0.39 (this value was very similar to that of 1,2:5,6-di-O-isopropylidene-a-D-- 52 -g l u c o f u r a n o s e ( 0 . 4 5 ) ) was s u r m i s e d t o be t h e d e r i v e d b r a n c h e d - c h a i n s u g a r (LXVII) h a v i n g a h y d r o x y e t h y l g r o u p on £ - 3 . M i c r o a n a l y s i s , i n f r a r e d , and n . m . r . s p e c t r a were i n ag reemen t w i t h t h i s p o s t u l a t i o n . The i n f r a r e d s p e c t r u m i n d i c a t e d no c a r b o n y l a b s o r p t i o n b u t a s t r o n g h y d r o x y l band a t 3460 cm 1 .• The mass s p e c t r u m o f (LXVII) showed a p r o m i n e n t M-15 peak c h a r a c t e r i s t i c o f a c e t o n i d e s . 1 3 4 The f o l l o w i n g c h a r a c t e r i s t i c s i g n a l s were f o u n d i n t h e n . m . r . s p e c t r u m ; a one p r o t o n d o u b l e t a t T 4 . 2 7 (J^ ^ * 3 . 6 Hz) c o r r e s p o n d -i n g t o t h e a n o m e r i c h y d r o g e n , a t r i p l e t a t T 5 . 3 2 a s s i g n e d t o H - 2 , a t h r e e p r o t o n m u l t i p l e t a t T 7 . 9 4 and t h e f o u r m e t h y l p e a k s a t h i g h f i e l d . The above a s s i g n m e n t s have b e e n c o n f i r m e d b y t h e u s e o f s p i n - s p i n d e c o u p l i n g e x p e r i m e n t s on B . B e c a u s e i r r a d i a t i o n a t x 7 . 9 4 changed t h e t r i p l e t a t T 5 . 3 2 i n t o a d o u b l e t and i r r a d i a t i o n a t x 5 . 3 2 changed t h e d o u b l e t a t x 4 . 2 7 i n t o a s i n g l e t , t h e r e f o r e , f r a c t i o n B i s 3 - d e o x y - 3 - £ - ( 2 ' - h y d r o x y e t h y l ) -1 , 2 : 5 , 6 - d i - 0 _ - i s o p r o p y l i d e n e - a - D - a l l o f u r a n o s e (LXVII) . H H LXV/ LXVII R=H LXVIII LXIX R = B r C , H . S 0 _ D 4 Z R = B r C , H , C 0 6 4 R=I LXX LXXI - 53 -The second portion of the reaction mixture was further refluxed for a period of one hour and then worked up in a similar way. The clear syrupy product was this time homogeneous on t . l . c , no trace of the more mobile compound A could be detected, indicating therefore that a higher temperature is needed to bring about the reaction in a minimum of time. Consequently, subsequent), 'reactions of this type were carried out at reflux temperature, affording the desired branched-chain sugar (LXVII)in quantitative yield. This compound, however., failed to crystallize and was characterized as the p_-toluenesulfonate (LXVIII), the p_-bromobenzeneSulfonate (LXIX) and the p_-bromobenzoyl derivative (LXX) . The latter were selected to provide a sample which would be amenable to X-ray crystallographic analysis. The preparation of the p_-toluenesulfonate and the p_-bromobenzenesulfonate were 135 done according to a method previously described in the literature. The reactions were complete after 30 minutes as shown by t . l . c . and their work up afforded both derivatives (LXVIII) and (LXIX) in high yield and good purity as indicated by microanalysis and n.m.r. spectroscopy. When the reaction mixtures were allowed to stand at room temperature for a few hours, the products were contaminated with impurities and failed to crystallize thus indicating that immediate work up was essential in the preparation of these derivatives. Unfortunately these compounds decomposed slowly at room temperature, and the idea of using the p_-bromobenzenesulfonate (LXIX) in X-ray crystallography was abandoned. Among the other derivatives suitable for X-ray analysis were iodo- and p_-bromobenzoyl derivatives, the former being the preferred one. Since replacement of a tosyl group by iodine requires vigorous condition (high pressure and high temperature) a sealed tube containing a solution of - 54 -compound (LXVIII) and sodium iodide in acetone was prepared and left for one day at room temperature before heating. During that time an unexpected precipitation in crystalline form of the presumed sodium p_-toluenesulfonate occurred. The tube was then kept at room temperature unt i l no more precipi-tate appeared and then heated at 100° in order to complete the reaction i f necessary. The solution darkened during heating and was therefore worked up after 30 minutes in the usual way. The quantity of the crystalline mass characterized as being the sodium p_-toluenesulfonate was very close to that of the theoretical amount, indicating that replacement of the tosyl group by iodine had occurred even at such low temperature. The syrupy product purified by column chromatography gave a pure compound (LXXI) that nonethe-less failed to crystallize. Compound (LXXI) gave a positive test with silver nitrate and the;n.m.r. spectrum was consistent with the structure 3-deoxy-3-C- (2' - iodoethyl) -1,2:5,v6-di-0_-isopropylidene-a-D-allofuranose. Preparation of i t s pyridinium salt in order to obtain a crystalline product gave negative results. Consequently we prepared the more stable p_-bromobenzoyl derivative (LXX) by reaction of the branched-chain sugar (LXVII) with p_-bromobenzoyl chloride in pyridine. The structure of the derivative thus obtained in quantitative yield was compatible with (LXX) as evidenced by spectroscopic data and microanalysis. Slow crystallization from aqueous methanol gave crystals suitable for X-ray crystallography. 2. Nucleoside synthesis 2.1 Conversion of 3-deoxy-3-C_- (2 '-hydroxyethyl) -1,2: 5,6-di-0-isopropylidene-a-D-allofuranose into 1, 2-di-0_-acetyl-2 ', 5,6-tri-0-benzoyl-(2 1 -hydroxyethyl)-B~D-allofuranose - 55 -Conversion of compound (LXVII) into a derivative suitable for the condensation with the chloromercuri salt of adenine in order to prepare a branched-chain sugar nucleoside required three steps. These steps 82 paralleled those used in the procedure of Prokop and Murray in their synthesis of 1,2-di-0-acetyl-5,6-di-0-benzoyl-3-deoxy-D-galactofuranose from the corresponding 3-deoxy-diisopropylidene galactose derivative. The steps were as follows: 1) selective hydrolysis of the 5,6-0-isopropylidene group; 2) blocking of the free hydroxyl groups as the benzoate ester; 3) acetolysis of the 1, 2-0_-isopropylidene group. The reaction scheme representing these steps is given on page 56. Partial hydrolysis of diisopropylidene sugars has been carried out under different conditions. The 5,6-isopropylidene group has been hydrolysed 119a with.a 70% solution of acetic acid at 50° , with a methanol solution 82 of hydrochloric acid at 40° or a methanol solution of sulfuric acid at 136 room temperature. In the present work the selective hydrolysis of the 5,6-0-isopropylidene group from the branched-chain sugar (LXVII) was accom-plished with aqueous methanol containing sulfuric acid as in the preparation 136 of 1,2-0_-isopropylidene glucose . The reaction was conducted at room temperature (about 20° ) for three hours^ Affording the partially hydrolysed derivative (LXXII) as a syrup in quantitative yield. Monitoring the reaction by t.1.c,indicated that simultaneous hydrolysis of 1,2-0-isopropyli-dene group occurred before a l l the starting material was consumed. Since the diisopropylidene derivative could be easier separated from the reaction product than the fu l l y hydrolyzed sugar, the reaction was stopped when the hydrolysis of the 1,2-0-isopropylidene group started as shown by the appearance on t . l . c . of a less mobil e compound. 56 X i C O C H ^ O C H 2 O H H 30+ HOCH C H 2 O CH^OH C H 3 •CH. C H 2 O- •CH, CH 2 OH C H 3 LXVII LXXI I B z C I p y r i d i n e CH 2 OBz BzOCH ^ O OAc A c 2 0 C H , OAc CH OBz 2 CH 2 OBz B z O C H Ac OH H 2 S 0 4 CH. CH 2 OBz C H 3 LXXIV LXXIII - 57 -The remaining unreacted starting material could be separated from the monoisopropylidene derivative by partition between chloroform and water. Extraction from the water phase with boiling chloroform gave the desired compound (LXXII) in good purity as indicated by t . l . c . and n.m.r. 3-Deoxy-3-C-(2'-hydroxyethyl)-1,2-0-isopropylidene-a-D-allofuranose 8 2 (LXXII) was then treated with benzoyl chloride in pyridine at room tempera-ture affording in high yield the syrupy 2',5,6-tri-0-benzoate (LXXIII) contaminated with traces of benzoic anhydride, which could only be removed by chromatography (t.l.c.) on s i l i c a gel using benzene methanol 95:5 as developer. Microanalysis and n.m.r. spectrum were consistent with the structure 2',5,6-tri-0-benzoyl-3-deoxy-3~C-(2'-hydroxyethyl)-1,2-0-isopropylidene-a-D-al lofuranose (LXXIII). 117 Acetolysis of (LXXIII) by the normal procedure using acetic acid, acetic anhydride and sulfuric acid for at least 3 days gave the B-anomer of (LXXIV), namely, 1,2-di-0-acetyl-2',5,6-tri-0-benzoyl-3-C-(2'-hydroxy-ethyl) -3-D-allofuranose as a crystalline solid in a yield of 75%. The assignment of the 3 configuration was based on an analysis of the n.m.r. spectrum. The anomeric hydrogen appears at x 3.92 as a singlet in agreement with the values predicted from the application of the Karplus curve to 137 neighboring trans hydrogens in a five membered ring. Since acetolysis of 1,2-0-isopropylidene groups has been reported to yield both 3 and a 117 122 anomers ' , the crude product of the reaction mixture was carefully analyzed. Results showed that no second isomer was present in sufficient amount to be detectable by t . l . c . or n.m.r. spectroscopy. If the a isomer was present, a doublet corresponding to i t s anomeric hydrogen would be visible in the n.m.r spectrum in the region below x3.92. This result does - 58 -130 not agree with Stevens and Fletcher who reported that for an anomeric pair of pentofuranose derivatives H-l is at a lower f i e l d when the substituents at C-l and C-2 are cis than when they are trans. Another rather unusual fact encountered in this reaction was the three day period compared to the 24 hours necessary to complete the acetolysis. Work up of the reaction mixture after 24 hours gave a mixture of two major products in almost equal quantities as deduced from the t . l . c . (2 spots Rf 0.80 and 0.45) on s i l i c a gel with benzene methanol 98:2 as developer, and from the n.mr. spectrum. When this mixture was treated a second time with acetic anhydride, acetic acid and sulfuric acid over a period of 2 days the only product isolated was the Banomer. An analysis of the n.m.r. spectrum of the mixture compared to that of the pure B anomer revealed that the latter was one of the products, the second being neither the starting material (LXXIII) nor the a anomer for the same reason as given above. 2.2 Chloromer.curi-6-benzamidopurine The synthesis of the chloromercuri salt derivative of adenine required in the- condensation with diacetates was carried out according to the 82 procedure described by Prokop and Murray. Fusion of adenine with benzoic anhydride gave 6-benzamidopurine in 68% yield. The latter compound was fu l l y 138 characterized because of i t s high melting point 249° compared to 242° and 8 2 239° reported in the literature. Despite this difference, other physical data: infrared, nuclear magnetic resonance spectra and the molecular weight obtained by mass spectrometry were in agreement with the structure (LXXV). Subsequent treatment with mercurichloride in basic medium gave the chloro-mercuri-6-benzamidopurine (LXXVI) in excellent yield. - 59 -NH; N I* M O (PhC) 20 NRBz NHBz N N H N HgCI; N fN H NaOH N IS HgCI LXXV LXXVI 2.3 9-3-D-Glucofuranosyladenine We have seen in the introduction that among the different methods of synthesis of nucleosides, the most widely used procedures were the conden-sation methods involving either the sugar acetate (via the titanium tetra-chloride complex) or the sugar halide with the chloromercuri salt of a base. The former method which has the advantage of being a one step procedure has been successfully used recently in the synthesis of 3-deoxypento- and hexo-82 122 furanosyl adenines, in the preparation of 3'-0-methyl purineribonucleoside 139 and even in the synthesis of branched-chain nucleosides. By this procedure 9-(3'-0-methyl-g-D-ribofuranosyl)adenine was prepared in a yield of 47% from 122 the diacetate compared to 14% from the sugar halide. Although the second procedure (Davoll and Lowy method1'''1) requires an extra step, i t has been used more extensively, in particular, in the synthesis of hexofuranosyl adenine - 60 , . " 117-120 derivatives. In order to find the best method of preparing the blocked branched-chain sugar nucleoside studies were carried out on a model sugar. When a hexofuranosyl acetate rather than a hexofuranosyl chloride was used, in the synthesis of 9-(B-D-glucofuranosyl)-adenine the yield was almost 117 identical to that obtained by Baker and coworkers in their synthesis of the same compound by Davoll and Lowy's method.'''11 The 1,2-0-isopropylidene-a-D-glucofuranose (LXXVII) which was easily prepared from the diisopropylidene derivative (XIV) by selective hydrolysis with a 0.8% solution of sulfuric acid in methanol, was benzoylated in the 82 usual way affording a colourless syrupy product, which failed to crystallize. Its infrared and nuclear magnetic resonance spectrum were compatible with the structure 3,5,6-tri-0-benzoyl-l,2-0-isopropylidene-a-rj-glucofuranose (LXXVIII). Acetolysis of this tribenzoate by the procedure described in the synthesis of branched-chain sugar diacetate (LXXIV) gave on work up after 24 hours the diacetate (LXXIX). Thin layer chromatography of this product on s i l i c a gel G with benzene-methanol (98:2) as developer indicated f i r s t that the reaction product was a mixture of two compounds ( 3 and a anomers) with Rf of 0.70 and 0.30 respectively, theB anomer being the major product. This was further confirmed by analysis of the n.m.r. spectrum which showed in the anomeric region at x 3.77 a singlet corresponding to 2/3 of a proton and at x 3.45 (J^ ^  ~ 4-5 Hz) a doublet of 1/3 of a proton. 130 These signals were assigned respectively to the 3 and a diacetates. The anomeric mixture of l,2-di-0-acetyl-3,5,6-tri-0_-benzoyl-D-gluco-furanose (LXXIX) was condensed with chloromercuri-6-benzomidopurine (LXXVI) by the titanium tetrachloride procedure. The intermediate coupling product was obtained as an amorphous solid, which was freed of mercuric salt by - 61 -Me Me XIV . LXXVII LXXVIII Ac20 Ac OH H30 + B z O — C H 2 LXX IX NHBz LXXXI LXXX - 62 -extraction of a chloroform solution with.30% aqueous potassium iodide. Thre reactions were set up successively and worked up after a period of 6, 22 and 33 hours affording three products A, B, and C. A comparison by t,I.e. (on s i l i c a gel G with benzene-methanol 98:2 as developer and a sulfuric acid spray as detector) of compound A with the starting material showed that the condensation had taken place only partial1 as evidenced by the presence in addition to the starting material of a compound with very low mobility just above the starting line. The latter compound, detectable by U.V., was attributed to a blocked nucleoside. Surprisingly, the zone corresponding to the a diacetate was not vis i b l e . The n.m.r. spectrum confirmed these results, since i t s t i l l showed the H-l singlet of the 3 diacetate at x 3.77 and in addition showed the following characteristic signals assigned to the blocked nucleoside (LXXX); a barely detectable doublet at x 3.70 (J = 1.75 Hz) and two singlets at x 1.38 (H-2) and 1.73 (H-8). No change could be noticed in the doublet at x 3.45. When the reaction was allowed to proceed for 22 hours the blocked nucleoside was obtained in higher yield (deduced from t . l . c . and n.m.r. data). On s i l i c a gel plates with ethyl acetate as a developer two zones with an Rf of 0.9 and 0.6 (major zone) were detected with a 30% sulfuric acid spray. In U.V. only the compound with Rf 0,6 was detectable, in addi-tion to one at an Rf of. 0.3 corresponding to adenine benzoate. The n.m.r. spectrum indicated the absence of the H-l singlet of the 3 diacetate, which was replaced by the doublet at x 3.70 mentioned earlier. S t i l l no change was observed for the doublet at x 3.45. The anomeric configuration of the blocked nucleoside (LXXX) thus obtained was deduced from the coupling constant observed for the anomeric hydrogen in the n.m.r. spectrum. The - 63 -value of J p 21 = 1.75 Hz < 2 Hz in this case indicates a 8 nucleoside through the application of the Karplus relationship. Since as shown from the previous n.m.r. data, no detectable change occurred in the doublet at T 3.45, i t was f e l t that a longer reaction time might perhaps complete the condensation. However,, this was not the case as indicated from the results obtained after 33 hours. Examination of the product clearly showed instead that some decomposition had taken place as evidenced by the appearance of extraneous peaks at T 0.8. It was therefore concluded that a reaction time of 22 hours as reported in the literature afforded the best yield of blocked nucleoside. With these conditions, 0.300 g of the diacetate (LXXIX) gave after work up 0.34 8 g of yellowish glass. An attempted quantitative separation of 0.10 g of this glass by thin layer chromatography on s i l i c a gel G containing electronic phosphor as indicator (in order to determine the percentage of nucleoside) gave no conclusive results due to poor recovery. The reason for this was attributed to the presence of electronic phosphor. In an attempt to improve the yield of the condensation by raising the pressure, the same condensation was carried out in a sealed Carius tube, but even under such conditions, no difference in yield could be observed (about 20%). Deacylation of the blocked nucleoside (LXXX) with sodium methoxide in methanol afforded a crystalline compound which precipitated out from the reaction mixture. The melting point and optical rotation were in agreement 117 with the values reported for 9-B-D-glucofuranosyladenine and i t s n.m.r. spectrum in deuteriodimethyl sulfoxide was likewise compatible with the structure (LXXXI). The n.m.r. spectrum was not taken in D~0 due to the low 1 0 0 0 Jo I. 250 I 100 I 50 H-2 (1.74) J H—8(1.84) NH2 300 N H , HO<J;H2 HOCH _o .OH H HN Y H H OH H-l'(4.l2) I K / U - J 1 4 > " H > 0 CPS 7 8 IO T Figure 4. Proton magnetic resonance spectrum at 100 MHz/sec in dimethyl sulfoxide-d 6 of 9-B-D-glucofuranosyladenine.• - 65 -solubility of (LXXXI) in this solvent. The anomeric hydrogen which appears as a singlet in Figure 4 page 64 is in fact a doublet with a spacing of 1 Hz as shown in an expanded spectrum. This low coupling constant confirms the 3 configuration of (LXXXI). An analysis of a second crop of crystalline material which precipita-ted in the mother-liquor after a period of two hours indicated no trace of nucleoside. In conclusion, since the yield obtained in this condensation was 117 almost identical to that reported by Reist, Spencer and Baker in their synthesis of 9-3-D-glucofuranosyl adenine via the glucofuranosyl chloride procedure, i t was f e l t that the titanium tetrachloride complex method involving one step less would be more advantageous in the preparation of branched-chain sugar nucleosides. 2.4 9-[3'-Deoxy-3'-C_- (2"-hydroxyethyl)-3-D-allofuranosyl)-adenine* In the light of the results obtained with the simple nucleoside (LXXXI), the condensation of the 3 diacetate of (LXXIV) with chloromercuri-6-benzamidopurine in ethylene dichloride containing titanium tetrachloride was carried out. The reaction afforded after 22 hours the blocked nucleo-side (LXXXII) which upon deacylation with methanolic sodium methoxide gave the crude 3 nucleoside (LXXXIII) in 48% yield (based on the diacetate). The latter product likewise precipitated upon neutralization of the solution. Paper chromatography of the:. crude product using water as solvent revealed, in addition to the nucleoside, the presence of adenine. The fact is not * two referees of the manuscript outlining this work suggested that in the nucleoside nomenclature the substituents on the sugar that are within the parentheses do not require "primes". However, referees of this thesis have suggested that the prime and double prime be retained to avoid ambiguity in the nomenclature. CH OBz C H 2 OAc CH 2 OBz LXXIV CH 2 OH CH 2OH LXXXVIII - 66 - NHBz LXXXI I Na OCH C H 3 O H NH 2 CH 2OH LXXXIII - 67 -surprising since we have shown previously that adenine benzoate constituted a by product of the blocked nucleoside (LXXX). Recrystallization of the nucleoside from aqueous methanol gave the branched-chain sugar nucleoside (LXXXIII) in good purity as indicated by paper chromatography, microanalysis and spectroscopic data. The n.m.r. spectrum of (LXXXIII) (Figure 5a, page 68) taken in deuteriodimethyl sulfoxide due to the low solubility of the nucleoside i n water, showed the following characteristic signals: two singlets at x 1.69 and 1.87 attributed to H-2 and H-8, respectively; a two proton signal at x 2.75 assigned to the NH2 hydrogens. The anomeric hydrogen appears as a doublet at x 4.18 with a rather large coupling constant J 1 , „,= 3.5 Hz compared to that of nucleoside (LXXXI). H-2' and H-3' were located respectively, at x 5.44 and 7.6. These assignments were confirmed by a spin-spin decoupling experiment. Double irradiation at x 5.44 collapsed the doublets at 4.18 and 4.44 into two singlets. The two doublets at x 4.44 and 4.54 were attributed to the protons of two hydroxyl groups on the basis that addition of some D20 to the deuterio-dimethyl sulfoxide resulted in their disappearance as well as that of the signal at x 2.75. It should be noted that these well resolved doublets could only be obtained with very dilute solutions. The assignment of the structure and anomeric configuration were based on the following: 1) ultraviolet-absorption data (208 and 260 ny ) similar to those 102-105 122 .139 reported for 9-furanosyl adenines ' ' , substantiate the site of glycosidation at N-9.1^0 113 2) as applied by Baker to reactions of 2-0-acylhalogenoses with 141 heavy metal derivatives of purines the "trans" rule requires that the - 68 -N N a Hoa-u HOCH -O. H - 2 (l.69) NH. H - 8 0 . S 7 ) 1" C H , . OH 2" CH 2 OH i H - 3 2 H - l " H - 3 ' 2H - I 8 T 9 Figure 5. Proton magnetic resonance spectrum at 100 MHz/sec of branched-chain sugar nucleosides in dimethyl sulfoxide-d • (a) 9- (3* -deoxy-3' -C_- (2"-hydroxyethyl) -3-B-allofuranosyl) -adenine (b) 9-(3'-deoxy-3'-C-(2"-hydroxyethyl)-3-D-ribofuranosyl)-adenine. - 69 -product in this case be trans 3) both negative optical rotation at the sodium D line as well as the negative Cotton effect shown by ORD measurements (Figure 6 page 70) are 142 143 consistent with the proposals advanced ' for purine B-D-nucleosides• The n.m.r. spectrum of the alio nucleoside (LXXXIII) was of very l i t t l e direct value toward establishing the configuration at C-l'. The resonance for the anomeric proton at T 4.18 appears as a doublet with a J j , 2i coupling constant of 3.5 Hz too large to permit an unambiguous assignment of the 8 trans anomeric configuration through application of. the 137 Karplus relationship. The magnitude of J 1 , «, of (LXXXIII) does however, i , z have implications concerning i t s conformation in solution. Table 2 l i s t s the J 1 , „, values of.adenosine and 31-deoxyadenosine J. , z 144 reported to be slightly concentration dependent and of 9-(3'-deoxy-3'-C-(2"-hydroxyethyl)-8-D-allofuranosyl)-adenine (LXXXIII). TABLE 2 Nucleosides J l ' , 2 ' H Z 144 Adenosine 6.1-6.4 144 3'-deoxyadenosine 2.1-2.3 LXXXIII 3.5 From this table one can see that the J coupling constant of the j — branched-chain sugar nucleoside (LXXXIII) is closer to that of 3'-deoxy-adenosine than to that of adenosine. One might therefore infer that the conformation of (LXXXIII) resembles that of the former nucleoside. If we postulate that vicinal coupling constants in (LXXXIII) depend only on the 145 dihedral angle (thus minimizing the other factors ), then the small value 2 0 0 2 5 0 3 0 0 3 5 0 A ( m fJL ) Figure 6. O.R.D. curves of branched-chain sugar nucleosides plus adenine: , adenosine; 9-(3'-deoxy-3'-C-(2M-hydroxyethyl)-g-D-ribofuranosyl)• adenine; ....9-(3'-deoxy-3'-C-(2"-hydroxyethyl)~£-D-allofuranosyl)-adenine. - 71 -J 1 t = 3.5 Hz arises in the same way as for J., of 3'-deoxyadenosine 1,2 > ^  from a smaller dihedral angle of the trans protons on C-l' and C-2' than that of adenosine. A consequence would be that C-2' is only slightly puckered out of the plane of the furanose ring in an endo fashion. However, inspection of a model of (LXXXIII) suggested that the potential steric interaction of the C-5 group with the purine moiety would be relieved to the greatest extent i f C-3' were endo in which case, C-5 would be in the quasi-equatorial conformation causing therefore less steric interaction with the purine base. Unfortunately the coupling constant involving protons at C-3' and C-4' were not obtainable to help establish this point. 146 Hence, we propose using the notational system of Jardetsky that the alio nucleoside exists in a conformation wherein C-3' is endo. 2.5 1,2-Di-0-acetyl-2',5-di-0-benzoyl-3-deoxy-3-C_- (2'-hydroxyethyl)-8-D-ribofuranose In order to prepare the ribo- nucleoside (LXXXVIII) (structural analog of adenosine) the alio branched-chain sugar (LXI) was f i r s t degraded to a ribo branched-chain sugar (LXXXV) in the following way. Partial hydrolysis of (LXI) with a methanolic solution of sulfuric acid, under careful surveill-ance by t . l . c . gave in 88% yield, 3-deoxy-3-C-(carbomethbxymethyl)-1,2-0-isopropylidene-a-D-allofuranose (LXXXIV) as an o i l which crystallized upon standing. Sublimation afforded (LXXXIV) in good purity. Its identity was substantiated by microanalysis, n.m.r. spectroscopy and t . l . c , which showed one spot with the same characteristic pink color of the ester (LXI) at a lower Rf than the starting material. Sodium metaperiodate oxidation of the latter, afforded after extraction with chloroform the aldehydo - 72 -LXXXV LXXXIV(a) derivative (LXXXlVa) in a maximum yield of 60%. The most favorable condition for such a yield was determined from the following experiments. Three parallel reactions involving the same amount of starting material were worked up respectively after 6, 30 and 45 minutes, a period of 6 minutes corresponding to the completion of the reaction as evidenced by the dis-appearance of the last traces of the starting material. It should be noted that the oxidation can be easily monitored by t . l . c . due to the blue color developed by the aldehydo sugar with the sulfuric acid spray. An analysis of the n.m.r. spectra of A, B, and C (the products obtained respectively from the three reactions) gave the following results: 1) i n a l l three products the aldehydo sugar was the major component, the by product accounting for 10, 20 and 40%, respectively. 2) the products show a characteristic signal, at very low f i e l d (doublet at x 0.40 ( J 5 4 = 2.25 Hz) attributed to the aldehydo proton). In addition a doublet at x 4.04 (J = 3 Hz) was assigned to H-l and a sharp methyl i > ^  signal at T 6.27 was attributed to the methyl ester. A multiplet at T 4.2 (attributed to a by product) which was barely detectable in spectrum A could be seen easily in spectrum B and even better in C. A concomitant appearance of a third peak in the high f i e l d region was also noticed. The experiments clearly show that in order to obtain the highest yield of aldehydo sugar, the reaction should be worked up immediately after the completion of the oxidation. Because sugars containing a free aldehydo group on C-4 and a free hydroxyl on C-3 are known to undergo complex re a c t i o n , 1 4 7 whereas those containing a blocked hydroxyl on C-3 behave normally, 1 4 8' 1 4 9 no attempt was made to degrade (LXVII) which contains a branched-chain on C-3- having a free hydroxyl group. - 74 -The aldehydo intermediate was immediately reduced with lithium aluminium hydride in tetrahydrofuran in the same way as described for the preparation of (LXVII) to yield 3-deoxy-3-C_-(21-hydroxyethyl)-1,2-0-isopropylidene-a-D-ribofuranose (LXXXV) in 72% yield. The reduced branched-chain sugar thus isolated, migrated as one spot on t . l . c . and no by product could be detected in the n.m.r. spectrum. An M-15 ion at m/e = 203 characteristic of acetonides 1 3 4 substantiated i t s structure. Since (LXXXV) is the f i r s t ribo branched-chain sugar reported having a hydroxyethyl group on C-3, and since i t is a key intermediate in the synthesis of a branched-chain ribo nucleoside i t s n.m.r. spectrum is of particular interest. Fortunately, unlike the n.m.r. spectrum of the corres-ponding C-6 branched-chain sugar (LXXII) the spectrum of (LXXXV) (Figure 7 page 75) permits a clearer assignment of the different signals. The anomeric hydrogen appears as a doublet at T 4.20 and H-2 as a tr i p l e t with unequal spacings 2 = 3.5 Hz and 3 = 4.5 Hz. The H-3 octet is well separated from the two methylene hydrogens which are overlapping with the two methyl signals of the 1,2-0-isopropylidene group. Double irradiation experiments were compatible with the above assignments. Irradiation of H-2 led to a collapse of the H-l doublet to a singlet. Irradiation of H-l changed the H-2 t r i p l e t into a doublet ( J 2 3 = 4.5 Hz) and irradiation of H-3 multiplet likewise collapsed the H-2 tr i p l e t into a doublet (J, ? = 3.5 Hz) . The reaction scheme.followed here, page 72 , in the synthesis of a pentofuranose from the corresponding hexofuranose has been reported by other groups of investigators particularly in the synthesis of 3-deoxy-D-• K .c A ' * 82,150,132 ribofuranose derivatives. j H-l (4.20) H-2 (5.31) H0H2C O H CH20-)-Me I Me CH20H H Ji,2(3.5Hz.) ft J2l3(4.5Hz.) | U,l2(3.5Hz.) H-3 (7.79) 8 10 Figure 7. Proton magnetic resonance spectrum at 100 MHz/sec in deuteriochloroform of 3-deoxy-3-C-(2'-hydroxyethyl)-1,2-0-isopropylidene-a-D-ribofuranose. - 76 -2.6 9-[3'-Deoxy-3'-£- (2"-hydroxyethyl)-B-D-ribofuranosyl]-adenine The conversion of the ribo branched-chain sugar (LXXXV) into the ribo nucleoside (LXXXVIII) was carried out by a sequence of steps that paralleled those used for the conversion of the alio sugar (LXXII) into the alio nucleoside (LXXXIII) (page 66). Benzbylation of (LXXXV) followed by acetolysis of the dibenzoate yielded 1,2-di-0-acetyl-2',5-di-0-benzoyl-3-deoxy-3-£-(2'-hydroxyethyl)-B-D-ribofuranose (LXXXVII) in 78% yield after 3 days. Thin layer chromatography revealed the presence of a trace of by product, which could not be identified. The only signal in the anomeric region was a singlet at x 3.97 attributed to H-l of the B anomer of (LXXXVII). The doublet at x 4.69 was assigned to H-2 because irradiation at x 3.97 led to a collapse of the former doublet into a singlet. Condensation of the B diacetate (LXXXVII) with chloromercuri-6-benzamido-purine by the titanium tetrachloride method, followed by deacylation of the blocked nucleoside with methanolic sodium methoxide afforded crystalline 9-(3'-deoxy-3'-C-(2"-hydroxyethyl)-B-D-ribofuranosyl)-adenine (LXXXVIII) in 37% yield (based on the condensation of the diacetate). Unlike the alio nucleoside (LXXXIII), recrystallization of (LXXXVIII) did not afford a pure compound. This d i f f i c u l t y , however, was easily overcome using paper chromatography with water as solvent. The trace of impurity (fluoresc-ing under U.V. light) which contaminated the product moved to the solvent front and could therefore be easily separated. The pure branched-chain sugar nucleoside thus isolated was a hydrate as determined by microanalysis and n.m.r. spectroscopy. Its physical properties were quite different from those of the alio nucleoside. Nucleoside (LXXXVIII) had a lower melting point, was unstable at 100°C and was much more soluble in water. The n.m.r. - 7 7 -H 0 C H 2 8zOCH 2 Bz Cl pyridine C H , CHOH C H 3 O CH, O CH, CH^OBz C H S LXXXV LXXXVI NH; N N NHBz 'IN N C H * OH N N HgCl 2 N a O C H 3 A c 2 0 Ac OH K , SO* 820CH 2 OAc CH 2 OH CH, OAc I CH OBz 2 LXXXVIII LXXXVII - 78 -spectrum (Fig. 5b ) taken in deuteriodimethyl sulfoxide, page 68 showed the following characteristic signals. Two one proton signals at x 1.66 and 1.94 attributed respectively, to H-2 and H-8; one two proton signal at x 2.8 (NH2) and a singlet at x 4.13 corresponding to the anomeric hydrogen. Signals of H-21 and H-3' were located at x 5.68 (triplet) and 7.62 (multiplet), respectively. The doublet at x 4.36 was attributed to the proton of the C-2' hydroxyl group. Results of spin-spin decoupling experiments just i f i e d these assignments. Irradiation of H-3' changed the H-2' tr i p l e t into a doublet ( J 2 , 2'bH = 5 ' Irradiation of H-2' collapsed the doublet at T 4.36 (2'OH) into a singlet and irradiation of the latter signal changed H-2' t r i p l e t into a doublet ( J 2 , 3, = 4.5 Hz). In D20 the signal at x 4.36 disappeared. It should be mentioned here that the chemical shifts reported for H-2, H-8 and H-l' are slightly concentration dependent, as observed by other 144 investigators. The structural and configurational assignments of (LXXXVIII) were based as for the alio nucleoside on U.V. data (absorption at 208 and 260 cm - 1), 113 on Baker's "trans rule", on negative optical rotation ([a]^2= -7°) and on the negative Cotton e f f e c t 1 4 2 ' 1 4 3 (Figure 6 p. 70) . However, in this case the n.m.r. spectrum, unlike for the alio nucleoside, provided a strong indication that nucleoside (LXXXVIII) had a B trans configuration. Since the anomeric hydrogen appeared as a singlet with no measurable coupling constant, therefore, through application of the Karplus relationsip, H-l and H-2 have the trans configuration with a dihedral angle in the vicinity of 90°. As a consequence, C-2* w i l l exist in the exo conformation. 1 4 6 It is interest-ing to note that the n.m.r. anomeric signal was spl i t into a doublet ( J ! i 2' = 1 H z) w hen the solution was heated to 70°. The diminution of - 79 -J , of (LXXXVIII) with respect to that of (LXXXIII) can be rationalized by the fact that there would be less repulsion between the C-4' (^ -CH2OH) and the purine of (LXXXVIII) than between the C-4' (CHOH-CH2OH) and the purine moeity of (LXXXIII). Sodium metaperiodate oxidation of the alio nucleoside (LXXXIII) yielded an aldehydo nucleoside that was immediately reduced with sodium borohydride to yield a nucleoside identical to the ribo nucleoside (LXXXVIII). Based on the fact that the alio nucleoside (LXXXIII) was degraded to the ribo nucleoside (LXXXVIII), by steps which are known not to alter the configura-tion of sugars, the alio nucleoside with a rather large coupling constant (J , ,= 3.5 Hz) could be assigned the structure 9-(3'-deoxy-31-C-(2"-hydroxyethyl)-8-D-ribofuranosyl)-adenine (LXXXIII), thus affording confirma-tion of i t s 8 configuration. In view of the fact that the titanium tetrachloride method of prepar-139 ing nucleosides is reported to give both anomers, the mother liquors from the recrystallization of the nucleosides (LXXXIII) and (LXXXVIII) were evaporated to dryness and the residues then separated by paper chromatography. Nucleosides identical to (LXXXIII) and (LXXXVIII) were recovered. There was no evidence for a nucleosides. These results therefore are in agreement with the results postulated by the "trans rule". The presence of a hydroxy-ethyl group at position 3' in addition to the C-2' acyloxy group might be expected to enhance the steric hindrance of the a side, making, the approach of the base only possible from the 3 side. 3. Oxo reaction As already discussed, the Wittig reaction, provided a method for inserting - 80 -a hydroxyethyl side chain on C-3 of a sugar. At the beginning of the investi-gation i t was envisaged that application of the oxo reaction to a 3,4-unsaturated sugar might provide an alternative method for inserting a side chain. Thus, application of the oxo reaction to 3-deoxy-l,2:5,6-di-0_-isopropylidene-a-D-erythro-hex-3-enose (II) might be expected to afford a branched-chain sugar with a hydroxymethyl on C-3 or C-4. 3.1 3-Deoxy-l,2:5,6-di-0-isopropylidene-a-D-erythro-hex-3-enose The synthesis of compound (II), starting material in the oxo reaction, posed a problem of considerable d i f f i c u l t y . Employment of potassium hydroxide as base to effect elimination of p_-toluenesulfonic acid from 1,2 :5,6-di-0_-isopropylidene-3-0-£-tolylsulfonyl-a-D-glucofuranose led to low, variable 80 82 yields of unsaturated sugar. ' Some success toward finding a better method of elimination of p_-toluenesulfonic acid from sulfonate esters of model alcohols was attained by using potassium t-butoxide in dimethyl s u l f o x i d e . ^ 2 Because the £-nitrobenzenesulfonate is known to be a 152 153 better leaving group than the p_-toluenesulfonate, other workers have used the former sulfonate ester in their studies but have found that ethers, in addition to olefins, are produced. From studies on the sulfonate esters of natural products, i t has been shown that the product is also dependent on the nature of the sulfonate and on i t s conformation. Thus, cholestan-38-ol-3-mesylate afforded cholestan-36-ol and an olefin in 86% and 4% yields, respectively, whereas cholestan-38-ol-3-£-toluenesulfonate gave mainly an olefin. When the epimeric form of the latter (namely, the 3ct-£-toluene-sulfonate) was used then the yield of olefin was reduced to 68%. 1 5 4 Utilization of potassium hydrogen carbonate or of collidine as base has resulted in the conversion of sulfonate esters of steroidal compounds - 81 -, i j * 155 «156 T into a mixture of alcohols, ketones and unsaturate d. compounds. In order to find a better method of synthesizing 1,2:5,6-di-0-isopropylidene-a-D-erythro-hex-3-enose (II), i t was f e l t that using a more powerful base, for example, tetramethylammonium hydroxide (TMAH) in dimethyl sulfoxide i cry (DMSO), might allow the elimination reaction to be carried out at room temperature. It was.envisaged that the unsaturated sugar subsequently could be extracted with petroleum ether or chloroform from the reaction mixture. 3.2 Reaction of the 3-0_-p_-toluene sulfonate and the 3-0-p_-nitrobenzene-sulfonate of 1,2 :5,6-di-(3-isopropylidene-a-p-glucofuranose with tetra-methylammonium hydroxide in dimethyl sulfoxide. When 1,2:5,6-di-0-isopropylidene-3-0-p_-tolylsulfonyl-a-D-glucofuranose (XXIV), prepared by reaction of p_-toluenesulfonyl chloride in pyridine with 79 (XIV), was allowed to react with a 25% solution of TMAH in DMSO at room temperature for 3 days, a chloroform extraction afforded a crystalline material in 89% yield. Characterization of the latter showed that i t was l,2:5,6-di-0-isopropylidene-a-Q-glucofuranose (XIV), which resulted from a hydrolysis D f the starting material. Consequently, we prepared by a method similar to that used for the synthesis of (XXIV) , the p_-nitrobenzenesulfonate of (XIV) . The presence of a better leaving group might be expected to yield the corresponding 3,4-unsaturated sugar (II). When 1,2 :5,6-di-£-isopropylidene-3-0-p_-nitrophenyl sulfonyl-a -D-gluco-furanose (LXXXIX) was treated with a tetramethylammonium hydroxide solution in dimethyl sulfoxide at room temperature for two days, a crystalline precipitate separated from the reaction mixture.' This product (XC) isolated 0 - C M e 2 XCII - 83 -in 54% yield showed absorption in the infrared characteristic of the sulfonyl (1190 and 1380 cm"1) and no absorption for a nitro group. Its mass spectrum gave a peak at m/e 643. Because isopropylidene derivatives of carbohydrates are known to lose a methyl group during their i n i t i a l mass breakdown,134'15** compound (XC) must have a molecular weight of 658, which strongly suggested that i t was a dimer of (LXXXIX) in which a p_-.nitropheny 1 sulfonyl and a nitro group were eliminated from 2-molecules of (LXXXIX). The elemental analysis of (XC) agreed with the molecular constitution C^H^O^S. The 159 dimeric structure was further confirmed by reductive cleavage with sodium amalgam which gave two fragments; the f i r s t was identified as 1,2:5,6-di-O-isopropylidene-a-D-glucofuranose (identical to an authentic sample) obtained in almost quantitative yield; the second fragment (XCII) was a water-soluble crystalline substance. Its nuclear magnetic resonance (n.m.r.) showed two doublets at T 2.43 and 2.91 (relative area 4), typical of the A^B^ pattern of aromatic protons. A doublet at x 3.96 (relative area 1) was assigned to the anomeric hydrogen. Signals for H-2, 3, 4, and 5 could not be assigned because of the interference with the lock signal of the HOD peak. Compound (XCII) was assumed tentatively to be the p_- (3-deoxy-l, 2 :5,6-di-£-isopropylidene-a-D-glucofuranose-3-yl)oxy-benzenesulfinic acid, sodium salt. The partial nuclear magnetic resonance spectrum of (XC) (shown in Figure 8 ) corroborated i t s dimeric structure and also permitted an easy assignment of the configuration of C-3 of each of the furanose moieties. Eight clearly defined peaks occurred at about x 8.6 (equal to 24 protons) that are undoubtedly due to the 8 methyl groups. Two doublets at x 1.98 and 2.80 were assigned to the 4 aromatic protons which gave the typical A B pattern of aromatic protons. The two pairs of doublets at x 4.19 - 84 -o •4. I9| |4,2I 5. 2 8 , ,5.36 | 5 , 5 6  I i i i i i i i i i j • i i i i i i • i I i i i i l l l i l | i i l I i i I l I | i i i i > i I i i | 4 5 6 T Figure 8. Partial proton magnetic re son apce, spectrum of (XC) in carbon tetrachloride ((a) at 100 MHz/sec; (b> at 100 MHz/sec, H-l and H-i' protons irradiated; (c) at 100 MHz/sec, H-2', proton irradiated; (d) at 100 MHz/sec, H-2 proton irradiated.) .- 85 -( J 1 2 = 3.5 Hz) and 4.21 ( J ^ 2, = 3.7 Hz) (see Figure 8a) were assigned to H-l and H-l' on the basis of the fact that the doublet of 1,2:5,6-di-0-isopropylidene-ot-D-glucofuranose (XIV) at x 4.22 has a ^ = 3.5 Hz. Irradiation at x 5.56 and 5.28 led to a collapse of the two pairs of doublets at x 4.2 (Figure 8c and d). When (XC) was irradiated at x 4.2, doublets at x 5.28 and 5.56 collapsed to singlets (Figure 8b). Since the H-2 signals of 1,2:5,6-di-0-isopropylidene-ct-D-glucofuranose occurs at x 5.6, whereas the H-2 signal of the 3-0-tosylate derivative occurs at x 5.28, therefore, the doublets at x 5.28 and 5.56 are those of H-2 and H-2', respectively. The configuration of C-3 of each furanose moiety of the dimer was deduced from the following evidence. H-2 and H-2' signals occurring as two doublets show no measurable coupling between H-2 and H-3, H-2' and H-3'. Since H-2 of l,2:5,6-di-0_-isopropylidene-a-D-glucofuranose gives a doublet (no measurable coupling between H-2 and H-3) whereas H-2 of l,2:5,6-di-0_-isopropylidene-ot-D-allofuranose gives two pairs of doublets therefore each furanose moiety of (XC) must have the D-gluco configuration. From this evidence, i t must follow that compound (XC) is 3-0-(p_-(3-deoxy-1,2:5,6-di-0-isopropylidene-a-D-glucofuranose-3-yl)oxyphenylsulfonyl)-1,2:5,6-di-O-isopropylidene-ct-D-glucofuranose. Work up of the remaining mother liquor gave a mixture of two other crystalline compounds, one of which (XCI), obtained in 19% yield, showed absorption in i t s infrared spectrum characteristic of a nitro group but none characteristic of a sulfonyl group. Its mass spectrum showed a peak at 366, thus establishing that (XCI) ^  a molecular w e i g h t of .381 (15 added to base peak to account for loss of a methyl group). The elemental analysis agreed with the molecular constitution C10H 0 N. The n.m.r. spectrum of lis z 3 is (XCI) exhibited two doublets at x 1.79 and 2.98 (relative area 4) which is - 86 -typical of an A 2B 2 pattern of aromatic protons, and a doublet atx 4.23 (relative area 1) which was assigned to the anomeric hydrogen. The H-2 signal appeared as a doublet at x 5.57. C-3 was assigned the D-gluco conf uration on the same basis as used for the assignment of the C-3 of compound (XC). The signals for H-4, H-5 and H-6 are overlapping (relative area 4) and occur at x 5.92. The four signals at x 8.5, 8.65, 8.73 and 8.78 are due to the four isopropylidene methyl groups. On the basis of this data, compound (XCI) was characterized as 1,2:5,6-di-£-isopropylidene-3-0-p_-nitrophenyl-a-D-glucofuranose. The remaining product (XIV), obtained in 5% yield, was readily charac-terized as 1,2:5,6-di-0-isopropylidene-a-D-glucofuranose by direct compar-ison with an authentic sample. In order to establish a mechanism for the reaction of p_-nitrobenzene-sulfonate ester of 1,2:5,6-di-0-isopropylidene-a-D-glucofuranose with tetrame'thylammonium hydroxide in dimethyl sulfoxide, the sulfonates of model alcohols were prepared and subjected to similar treatment. The p-nitrobenzene sulfonate of ethyl alcohol using a modification of 152 the method reported by Morgan and Cretcher afforded the desired sulfonat in 70% yield. Microanalysis, infrared and n.m.r. spectra were compatible with structure (XCIII). The n.m.r. spectrum exhibited the following characteristic signals; two doublets at x 1.59 and 1.82 (4 aromatic protons), a two proton quartet and a three proton tri p l e t atx 5.72 and 8.62 characteristic signals of an ethyl group. Treatment of (XCIII) with a 25% solution of TMAH in DMSO at room temperature resulted in the formation of a yellow crystalline mass which separated from the reaction mixture. Recrystallization from petroleum ethe gave a white substance, which was not a dimer as would be expected by the - 87 -analogy with the carbohydrate (XXIV), but presumably a salt as indicated by i t s high melting point 283-284°. The n.m.r. spectrum exhibited an A2^2 °t u a r t e t characteristic of 4 aromatic protons at T 1.26 and 1.62 suggest-ing the presence of a nitro as well as a sulfonate group, and a 12 proton t r i p l e t at T 6.46 attributed to the protons of four methyl groups. The presence of NO^  and SO^  absorption bands in the infrared spectrum confirmed the n.m.r. data. Consequently, the salt is the tetramethylammonium p_-nitrobenzenesulfonate (XCIV). Results obtained by microanalysis were in agreement with structure (XCIV) . The tetramethylammonium salt of p_-nitrobenzenesulfonic acid was recovered in an overall yield of 90%. In addition, extraction of the mother liquor with chloroform afforded a mixture of a least three components. The major product (XCV) isolated in 5% yield by preparative t . l . c . on alumina using benzene-petroleum ether as developer was a slightly yellow crystalline compound. Its n.m.r. showed two (two proton)doublets at T 1.85 and 3.13 characteristic of four aromatic hydrogens, a two proton quartet and a three proton t r i p l e t at x 5.92 and 8.55 which are undoubtedly those of an ethyl group. From these data alone the following conclusions can be drawn: Compound (XCV) is composed of a 1,4-disubstituted phenyl ring, and one substituent bearing an ethyl group. In addition, information concerning the nature of the substituents can be obtained from the chemical shifts of the two sets of aromatic protons. A x value of 3.13 is usually encountered for protons ortho to an ether group, while a x value of 1.85 results from the presence of an adjacent electro-negative group, e.g. (N02). An N02 absorption band in the infrared spectrum agreed with the latter postulation. These spectroscopic data therefore suggest p_-nitrophenetole as structure for (XCV). This was confirmed by direct comparison of the physical constants of (XCV) with an authentic - 88 -sample. Therefore, under the same conditions as used in the reaction of the carbohydrate p-nitrobenzenesulfonate (LXXXIX) , the ethyl p_-nitrobenzene-sulfonate (XCIII) yields 90% of tetramethylarnmonium p_-nitrobenzenesulfonate (XCIV) and p-nitrophenetole (XCV). Presumably, the hydroxide ion attacked the sulfur atom of the sulfonate liberating ethoxide ion. The latter ion probably then attacked the highly electron deficient carbon on the phenyl ring with expulsion of the ethoxy sulfonyl group and consequent formation of p-nitrophenetole as shown below. 0 0 xcrn 02N 0 0 0 XCIII The presence of a yellow substance (not charactertized) in the reaction mixture probably indicated that a Meisenheimer type complex might have been formed during the reaction. This reaction thus closely parallels the well known reaction of 2,4,6-trinitroanisole with potassium ethoxide to give a yellow salt (Meisenheimer complex) which on acidification affords 2,4,6-tri-nitrophenetole. Following these results, a second model compound, with a structure - 89 -closer to the carbohydrate (LXXXIX), namely, cyclohexyl p_-nitrobenzene-sulfonate (XCVI) was synthesized using the procedure of Streitwieser a n c j S c h a e f f e r . T h e physical constants of the product isolated in 70% yield were in agreement with those reported in the literature. The n.m.r. spectrum showed two two proton doublets at x 1.66 and 1.92 (4 aromatic hydrogens) one broad signal at x 5.35 and a 10 proton multiplet at x 8.5. When (XCVI) was treated under the same conditions as described earlier for the ethyl ester with tetramethylammonium hydroxide in DMSO, tetramethylammonium p_-nitrobenzene-sulfonate was also obtained in almost quantitative yield. A similar mechanism can be invoked to explain the formation of products (XC), (XCI), and (XIV) from the 3-0-p_-nitrobenzenesulfonate of l,2:5,6-di-0-isopropylidene-ct-D-glucofuranose. Obviously a simple hydrolysis of (LXXXIX) afforded l,2:5,6-di-0-isopropylidene-a-D-glucofuranose (XIV). In the presence of dimethyl sulfoxide the l,2:5,6-di-0_-isopropylidene-o-D-glucofuranose 3-oxide ion of (XIV) might be expected to exhibit enhanced basicity, and, presumably, might be expected to be able to attack compound (LXXXIX) at the carbon carrying the nitro group. This carbon would be positive in character due to the strong electron withdrawing effect of the sulfonyloxy group. The green to red-orange color of the solution during the reaction strongly indicates that a type of Meisenheimer complex was formed. Expulsion of the nitro group by the attacking 3-oxide ion of 1,2:5,6-di-0-isopropylidene-a-D-glucofuranose would then lead to the production of the dimer (XC). On the other hand, attack on the 3-oxide ion at the p-carbon on the aromatic ring (relative to the nitro group) of another molecule of (LXXXIX) might also be expected to lead to expulsion of the 1,2:5,6-di-0-isopropylidene-a-D-glucofuranose-3-0-sulfonyl group with the consequent formation of (XCI). - 90 -3.3 Synthesis of 3-deoxy-1,2:5,6-di-0-isopropylidene- (vD-erythro-hex-3-enose Following the results obtained in the reaction of the p_-nitrobenzene-sulfonate derivative (LXXXIX) in DMSO with TMA1I in an attempt to prepare the unsaturated carbohydrate (II), a second method was tried consisting in 162 the use of potassium t-butoxide in DMSO. Under these conditions, however, the major product isolated was l,2:5,6-di-0-isopropylidene-a-D-glucofuranose (XIV). Consequently, compound (XXIV) was treated with potassium hydroxide 82 according to the method previously described by Prokop and Murray afford-ing in 65% yield the desired 3,4-ene (II). Melting point, infrared and n.m.r. spectra were compatible with the structure 3-deoxy-1,2:5,6-di-0-isopropylidene-a-D-erythro-hex-3-enose (II). Under the same conditions, reaction of the p-nitrobenzenesulfonate (LXXXIX) instead of the tosylate (XXIV), led mainly to the formation of the diisopropylidene compound (XIV). A small amount of the 3,4-unsaturated sugar (15%) could only be obtained, using a constant temperature of 120° during a period of 1 hr. 3.4 Reaction of 3-deoxy-1,2:5^-di-O-isopropylidene-q-D-erythro-hex-S-enose with carbon monoxide and hydrogen under oxo conditions. Previous work has demonstrated that the oxo reaction can be applied successfully to g l y c a l s , 7 6 > 9 2 > 1 6 3 > 1 6 4 2-hydroxyglycals" and 2,3-enesf3the major products of the reactions being acetylated alcohols having one more carbon atom than the starting material. Thus, the reactions appeared to follow the expected course and a hydroxymethyl group was added to one side of the double bond. It could be anticipated that application of the same reaction to the unsaturated 3,4-ene (II) under hydro(hydroxymethyl)ation - 91 -condi t ions would therefore r e s u l t i n the formation of a d i - O - i s o p r o p y l i d e n e d e r i v a t i v e with a hydroxymethyl group as a branched-chain on C-3 or on C-4 . Under hydroformylat ion c o n d i t i o n s , 5 , 6 - d i - 0 - d e o x y - l , 2 - 0 - i s o p r o p y l i d e n e -a-D-xylo-hex-5-enofuranose (XCVII) gave 5 ,6 -d i -deoxy-1 ,2 -0 - i sopropy l idene -a-Q-xylo-heptodia ldo- l ,4 - furanose-d,e -D-7 ,3-pyranose ( X C V I I I ) 1 6 5 i n 40% y i e l d . C H 2 ii * CH C H , CH, XCVII H C s O o 0-)-CH, CH. i J HOHC H,C o f CH3 C H , X C V I I I As seen i n the i n t r o d u c t i o n , a v a i l a b l e evidence ind ica tes that the formyl group adds to the l eas t hindered s ide of the double bond under 96 normal condi t ions of high temperature and pressure . One can therefore expect that i n the case of compound (II) the formyl group w i l l add at p o s i t i o n 3. Since the oxo r e a c t i o n proceeds by way of c i s add i t i on of a hydrogen atom and a hydroxymethyl group to the carbon-carbon double bond, the r e s u l t i n g branched-chain sugar w i l l most l i k e l y be 3-deoxy-3-C-(2 ' -hydroxymethy l ) -1 ,2 :5 ,6 -d i -O- i sopropy l idene-a -D-ga lac to furanose . Reaction cond i t i ons : Su i tab le condi t ions f o r the r e a c t i o n of acety lated g l y c a l s with carbon monoxide, hydrogen and d i c o b a l t octacarbonyl have been wel l e s tab l i shed . ' At a temperature of 1 3 0 ° , g l y c a l s were l a r g e l y converted to a lcohols by 166 hydroxymethylation. With the major i ty of o l e f i n s the most r a p i d r e a c t i o n - 92 -rate was obtained when the ratio of hydrogen to carbon monoxide was high. With glycals the reaction time reported was usually two hours. Conditions for the reaction of di-O-isopropylidene compounds have not been established previously, therefore reaction of 3-deoxy-1,2:5,6-di-0-isopropylidene-a-D-erythro-hex-3-enose (II) with carbon monoxide, hydrogen and dicobalt octacarbonyl was carried out under four different reactions conditions. Table III l i s t s the i n i t i a l partial pressures of carbon monoxide and hydrogen, the solvent, temperature and reaction time for each experiment. TABLE III Experiment I II III IV CO . psi 800 2000 700 1100 H. . 2 psi 2400 300 2200 1100 solvent C6 H6 C6 H6 C6 H6 (CH3)2C0 Temperature °C 130° (1 hr) 140°(1.5 hr) 130° 135° 135° Reaction time hr. 2.5 1 1 1 Experiment I was carried out under the conditions normally used in oxo reactions. 1 6 6 After a period of 1 hr at 130° and 1.5 hr at 140° the amount of gas absorbed corresponded to 4 moles of synthesis gas per mole of substrate, instead of the anticipated 3 moles, as shown below. - C = C - + 2H + CO 3 moles > -CH2-CH-CH20H - 93 -When t h e r e v e r s e r a t i o s o f p a r t i a l p r e s s u r e s o f c a r b o n monox ide and h y d r o g e n were u s e d ( e x p e r i m e n t I I ) , 2 m o l e s o f s y n t h e s i s gas p e r mole o f s u b s t r a t e were a b s o r b e d w h i c h s u g g e s t e d t h a t h y d r o f o f m y l a t i o n i n s t e a d o f h y d r o ( h y d r o x y m e t h y l ) a t i o n had t a k e n p l a c e l e a d i n g t o an a l d e h y d o s u g a r . E x p e r i m e n t I I I i s a l m o s t a d u p l i c a t e o f I w i t h t h e e x c e p t i o n t h a t t h e r e a c t i o n was c a r r i e d o u t a t 135° o v e r a p e r i o d o f 1 h r o n l y , l e a d i n g l i k e -w i s e t o t h e a b s o r p t i o n o f . 2 m o l e s o f s y n t h e s i s g a s . I t i s t o be n o t e d t h a t t h i s r e a c t i o n was done u n d e r e s s e n t i a l d r y c o n d i t i o n s u s i n g r e a g e n t g r a d e c a r b o n m o n o x i d e . As a r e s u l t o f t h e s e t h r e e e x p e r i m e n t s o t h e r v a r i a t i o n s i n t h e r e a c t i o n c o n d i t i o n s were t r i e d , n a m e l y , t h e r e p l a c e m e n t o f benzene by a c e t o n e and t h e use o f e q u a l p a r t i a l p r e s s u r e s o f c a r b o n monox ide and h y d r o g e n ( e x p e r i m e n t I V ) . . H o w e v e r , i n t h i s c a s e a g a i n o n l y two m o l e s o f gas p e r mole o f s u b s t r a t e were a b s o r b e d . Remova l o f t h e c a t a l y s t V a r i o u s methods f o r t h e r e m o v a l o f t h e d i c o b a l t o c t a c a r b o n y l c a t a l y s t 168 166 f rom oxo r e a c t i o n p r o d u c t s have b e e n d e s c r i b e d . ' The two most c o n v e n i e n t methods c o n s i s t e d e i t h e r i n h e a t i n g t h e r e a c t i o n m i x t u r e on a s t eam b a t h , u n t i l c a r b o n monox ide was no l o n g e r e v o l v e d , o r i n f i l t e r i n g t h r o u g h F l o r i s i l (a s y n t h e t i c m a g n e s i a - s i l i c a g e l a b s o r b e n t ) where t h e c a t a l y s t was e l u t e d w i t h p e t r o l e u m e t h e r and t h e r e a c t i o n p r o d u c t s s u b s e -q u e n t l y e l u t e d w i t h a more p o l a r s o l v e n t . I n w o r k i n g w i t h t h e oxo p r o d u c t s o f ( I I ) b o t h methods were u s e d ; i t was f o u n d t h a t f i l t r a t i o n t h r o u g h F l o r i s i l r e s u l t e d i n t h e l o s s o f a l a r g e q u a n t i t y o f r e a c t i o n p r o d u c t ( e x p e r i m e n t . I l l ) , w h i c h c o u l d no t be e l u t e d f rom t h e c o l u m n . T h e r e f o r e , i n o r d e r t o be a b l e t o e v a l u a t e a p p r o x i m a t e l y t h e number o f p r o d u c t s o b t a i n e d , t h e c a t a l y s t was decomposed t h r o u g h h e a t i n g . - 94 -Fi l t r a t i o n , followed by treatment with Norite afforded a pale yellow syrup. However, in some cases, due to the colloidal form of the cobalt a subsequent f i l t r a t i o n through Celite was necessary. Fractionation and characterisation of reaction products Chromatographic separation of products in experiments I and IV: Thin layer chromatography of the oxo product in experiment I, showed the presence of 8 components in almost equal quantities, indicating that l i t t l e would be gained in attempting to fractionate the mixture. Thus reaction of the di-O-isopropylidene unsaturated compound (II) with carbon monoxide and hydrogen in presence of dicobalt octacarbonyl appears to be more complex than the oxo reaction for the glycals as evidenced by the number of products and the consumption of more than three moles of gas per mole of substrate. The n.m.r. spectrum of the crude oxo product showed that hydro-genolysis of the 5,6-isopropylidene group had occurred. This was further confirmed by gas-liquid chromatography of the volatile portion ( d i s t i l l e d at atmospheric pressure at 100°) of the reaction mixture; which showed the presence of benzene and isopropyl alcohol (same retention time). In experiment I I , where the reverse ratio of gases were used thin layer chromatography showed one major compound (XCIX) with an Rf of 0.7 using s i l i c a gel G and benzene-methanol 98:2 as developer. Separation of one part of the oxo product on a preparative scale gave an almost pure compound in 25% yield. The infrared spectrum, however, showed an absorp-tion at 1670 cm 1 corresponding to a carbon-carbon double bond and no hydroxyl absorption. On t . l . c . compound (XCIX) and the starting material (II) had very similar Rf 0.7 and 0.75 respectively. Despite these similarities compound C was not the 3,4-ene (II) as evidenced by the n.m.r. spectrum, which showed the following characteristic signals; a one proton mutiplet - 95 -at x 3.86, a two proton doublet at T 5.06 and three singlets at high f i e l d corresponding to 12 hydrogens. The signals in the spectrum accounted for a total of 18.protons. On the above mentioned basis compound (XCIX) might have resulted from a double bond migration from the 3,4-position to the 2,3-position. H 3 C \ < O C H 2 XCIX The separation of an unsaturated compound from the oxo product of experiment II seems to indicate that the reaction conditions were not adequate to bring about completely the hydro(hydroxymethyl)ation. However, the infrared, as well as the n.m.r. spectra of the combined slow moving components showed that addition to the double bond had occurred. The infrared spectrum indicated no absorption at 1670 cm 1. Consequently, the second proton of the oxo product was reduced with sodium borohydride and then acetylated. No indication concerning the number of components of the crude acetylated product could be obtained by t . l . c . The n.m.r. spectrum showed that i t was a mixture and in addition showed a characteristic doublet at T 2.66. Fractionation by gas-liquid chromatography (g.l.c.) was attempted on a column (Chromosorb W. carrying 10% (by weight) silicon gum rubber SE-52) - 96 -at 250° with helium as carrier gas at a flow rate of 40 ml/minute. Under these conditions the di-0-isopropylidene sugar (XIV) was stable. It is interesting to note that when nitrogen was used as carrier gas instead of helium, separation by g.l.c. led to decomposition of compound (XIV). An n.m.r. spectrum of the chromatographed product of (XIV) showed the same characteristic doublet at low f i e l d ( T 2.66) and similar peaks in the region (T 7-8) as the crude acetylated product. Analysis of the acetylated mixture by g.l.c. revealed the presence of two main components with retention times of 16 min. and 40 min,. respectively. The latter was obtained crystalline (m.p. 101-102°, but could' not be analyzed due to poor recovery. In the experiment I I I , elution of the oxo product with benzene -2% isopropyl alcohol from the F l o r i s i l column gave 4 successive fractions A, B, C, and D. Fraction A was a mixture of two major components, .The n.m.r. spectrum, however, indicated that almost complete hydrogenolysis of the isopropylidene group had occurred; consequently, the separation of the two compounds was not attempted. The other three fractions, B, C, and D indicated no major component. The last experiment, where equal partial pressures of carbon monoxide and hydrogen were used and where benzene as solvent was replaced by acetone, did not, unfortunately, give better results. T.l.c. of the crude oxo product showed no major component. Two strong absorptions at 1710 cm 1 and 1600 cm 1 were observed in the infrared spectrum. As in the three previous experiments, the n.m.r. spectrum indicated that hydrogenolysis of the isopropylidene group had occurred. - 97 -Conclusion The results obtained from the reaction of 3-deoxy-1,2:5,6-di-0_-isopropylidene-a-D-erythro-hex-3-enose with carbon monoxide and hydrogen in the presence of preformed dicobalt octacarbonyl, under four different conditions (Experiments I, I I , I I I , IV) therefore clearly indicate that the isopropylidene group is not a useful protecting group in oxo reactions. Due to the considerable number of oxo products and the d i f f i c u l t i e s encountered in their separation, no attempt was made to elucidate the structure of the oxo products. IV EXPERIMENTAL General Considerations Nuclear magnetic resonance (n.m.r.) spectra were obtained on deuterio-chloroform solutions (unless otherwise stated) with tetramethylsilane as the internal standard (set at T 10) using Jeolco 60, Varian A-60, or Varian HA 100 spectrometers. The following conventions and abbreviations are used: m = multiplet, o = octet, t =triplet, d = doublet, s = singlet. Mass spectra were obtained with an A.E.I.M.S.9 spectrometer. Infrared spectra (i.r.) were recorded on a Perkin Elmer model 137 spectrophotometer. The ORD measurements were performed on a Jasco Model 0RD/UV-5 spectropolar-meter at room temperature in aqueous solutions. High pressure reactions 9 were carried out using an Aminco 2 /16" o.d. Micro Series reaction vessel of manganese steel (American Instrument Co. Inc., Silver Spring,. Md.) Gas-liquid partition chromatography (g.l.c.) separations were effected using an Aerograph Model 1525, employing a column (12' x 3/8") of Chromosorb W carrying 10% (by weight) silicone gum rubber SE-52 operated at 250° with helium as carrier gas at a flow rate of 40 ml/minute. The ultraviolet spectral measurements were performed on a Cary 14 spectrophotometer. A l l melting points (micro hot stage) are corrected. S i l i c a gel G and aluminum oxide were used in the thin layer chromatography ( t . l . c ) . Elemental analysis were performed by the Micro-analytical Laboratory, University of - 99 -British Columbia. 128 l,2:5,6-Di-0-isopropylidene-a-D-giucofuranose (XIV) To an efficiently stirred suspension of a-D-glucose (150 g) in absolute acetone (1 l i t r e ) was added pulverized anhydrous zinc chloride (140 g) and 8% phosphoric acid (7.5 g). The mixture was allowed to shake at room temperature for two days. The unreacted sugar (54 g) was removed by f i l t r a t i o n and the f i l t r a t e was made slightly alkaline with sodium hydroxide (85 g in 85 ml of water). The insoluble inorganic material was removed by f i l t r a t i o n and washed with acetone. The f i l t r a t e and washings were concentrated under reduced pressure. The residue was dissolved in water (150 ml) and extracted with chloroform (150 ml x 3). The combined chloroform extracts were washed again with water, then dried over sodium sulfate. Evaporation in vacuo yielded a solid residue, which was recrystal-lized from petroleum ether (65-110°) affording a white crystalline compound 110 g (80%) (based on D-glucose consumed), m.p. 109°, l i t . 1 2 4 , m.p. 110-111°. [ a ] 2 2 -14° (c, 2 in chloroform), l i t . 1 2 5 [ a ]22 -13.5° in chloroform. 1,2:5,6-Di-0-isopropylidene-a-D-ribo-hexofuranos-3 -ulose.(I) To an ice-cold mixture of 1,2:5,6-di-£-isopropylidene-a-D-glucofuranose (13 g) in anhydrous dimethyl sulfoxide(l50 ml), phosphorus'pent;0;xide (8 g) was added. After being stirred at room temperature for two days, the mixture was diluted with chloroform (300 ml) and then a saturated solution of sodium bicarbonate (300 ml) was added while cooling. After vigorous shaking the chloroform phase which separated was washed successively with a solution of sodium bicarbonate (400 ml) and then twice with the same amount of water. The chloroform solution was dried over sodium sulfate and - 100 -then evaporated in vacuo to afford a syrup, 8.6 g (yield 65%), which slowly •i DG * crystallized on standing. Rf 1.35 R n r on s i l i c a gel G using benzene-32 methanol (95:5) as developer, [a]^ 2 +106° (c, 3 in chloroform), l i t . [a] +107°; l i t . 1 2 3 [a] Q +105° (c, 2 in chloroform); l i t . 3 3 , 3 4 [a] D +40° (c, 2 in chloroform), v™^ 0 1 (cm"1) 1770 (OO). n.m.r. T C D C 1 3 3.94 (d, H-l, J 1 2 = 4.5 Hz), 5.7 (H-2), 8.58 and 8.60 (2s, CH3 of isopropylidene group), 8.70 (s, 2CH3 of isopropylidene group). Irradiation of (I)at the H-2 signal collapsed H-l to a singlet. The crude product can be used as such in subsequent reactions. Several recrystallizations from light petroleum ether gave the monohydrate m.p. 115-116° [a]p 2 +43° (c, 2 in chloroform), l i t . 3 2 [a] +44.5° m.p. 113-114°, l i t . 3 1 a m.p. 108-110° [a]+40.2° (c, 0.5 in water), l i t . 3 3 ' 3 4 m.p. 118-119° [a]* 8 +110° (c, 1.0 in chloroform), v™J o 1 (cm-1) 3450 (OH), no C=0 absorption. Wittig reaction of 1,2:5,6-di-O-isopropylidene-q-D-ribo-hexofuranos-3-ulose (I) to yield 3-C- (carbomethoxymethyl)-3-deoxy-l,2:5,6-di-£-isopropylidene-a-D-allofuranose (LXI) An ice-cold solution of phosphbnoacetic acid trimethyl ester (11 ml) and potassium t-butoxide (2.5 g) in anhydrous N,N-dimethylformamide (11 ml) was added slowly to a solution (kept at 0°) of 5.5 g of l,2:5,6-di-0-isopro-33 34 pylidene-q-D-ribo-hexofuranos-3-ulose (I) ' in 33 ml of anhydrous N,N-dimethylformamide. The reaction mixture was kept at 0° for 1 hr and then at a room temperature for 48 hr (or until a l l compound (I) was consumed as evidenced by monitoring by t . l . c . on s i l i c a gel G using benzene-methanol * RD(-, refers to the Rf of 1,2 :5,6-di-O-isopropylidene-q-D-glucofuranose - 101 -(95:5) as developer). The solvent was removed under reduced pressure and the residue after addition of 150 ml of water, was extracted twice with ether (150 ml), the ether layer washed with water (30 ml), dried over magnesium sulfate, f i l t e r e d , and the f i l t r a t e evaporated under reduced pressure; yield, 5.49 (81%) of syrup. This product consisted of a mixture of cis and trans unsaturated branched-chain sugars. Preparative t . l . c . of part (0.35 g) of this product using s i l i c a gel G and benzene-methanol (95:5.) as developer gave 0.29 g of two different unsaturated sugars (in the ratio GDC1 of 1:3, each fraction having traces of contamination of the other): T 3 (of the major fraction) 3.76 (q, C-l'H, J1, 2 = 1.25 Hz and 4 = 2.0 Hz), 4.21 (d, H-l, J = 4.0 Hz), 4.29 (2t, H-2, J = 4.0 Hz and J = 1.25 Hz) 5.36 (o, H-4, J 4 5 = 6.25 Hz), 5.8.0 (m, H-5, and H-6), 6.27 (s, GH^  of ester): X C D C 1 3 ( 0f t j l e m i n o r component), 3.82 (2 overlapping d, C-l'H, J r 2 = 1.75 Hz and J p 4 = 1.50 Hz), (d, H-l, J± 2 = 4.75 Hz), 4.94 (2t, ^2,1 4.75 Hz and = 1.75 Hz). The mixture (4.7 g) of unsaturated sugars in 140 ml of ethanol was hydrogenated using 10% palladium on charcoal (2.2 g) as catalyst; 380 ml (1 mole equivalent) of gas was absorbed during a period of 3.5 hr. The catalyst was removed by f i l t r a t i o n and the f i l t r a t e then evaporated giving compound (LXI) (4 g, 85% yield, homogeneous by chromatography) that was recrystallized from petroleum ether (35-65°) or from methanol-water (4:1); m.p. 57-58°, [ a ] 2 2 +65° (c, 2 in ethanol), vmax 0 1 ( c m _ 1 ) 1 7 5 0 ( C = 0 ) ' t C D C l 3 4.29 (d, H-l, ^  2 = 3.7 Hz), 5.25 (t, H-2, J 2 3 = 4.0 Hz), 6.35 (s, methyl ester group), 7.64 (m, H-3). Irradiation of (LXI) at the H-2 signal collapsed H-l to a singlet. Anal. Calcd. for C^H^O.^ C, 56.95; H, 7.65; molecular wt., 316. Found: C, 57.14; H, 7.88; m/e 301 (the base peak in the mass spectrum is at M+-15 (loss of CH3). - 102 -3-Deoxy-3-C-(21-hydroxyethyl)-!,2:5,6-di-O-isopropylidene-ct-D-allofuranose (LXVII) To a stirred solution of 2 g of the branched-chain sugar (LXI) in 100 ml of anhydrous tetrahydrofuran was added slowly a mixture of 3 g of lithium aluminum hydride in 200 ml of anhydrous tetrahydrofuran. The reaction mixture was heated under reflux for 1 hr. and then left stand at room temperature for 1 hr. The excess lithium aluminum hydride was destroyed by dropwise addition of water. After the residue was removed by f i l t r a t i o n , the f i l t r a t e was evaporated under reduced pressure. The residue was dissolved in 8 ml of ethanol and then 16 ml of ether was added. After the precipitate was removed by f i l t r a t i o n , the f i l t r a t e was evaporated under reduced pressure. The product (LXVII)(1.73 g, 94%), resisted crystalliza-tion but was chromatographically homogeneous (Rf 0.39 on s i l i c a gel G using benzene-methanol (92:8). The analytical sample was obtained by d i s t i l l a t i o n of (L'XVII) under high vacuum: [ a]g2 +61° (c, 2 in ethanol). T C D C l 3 4.27 (d, H-l J1 2 = 3.6 Hz), 5.32 (t, H-2 J"2 3'= 4.0 Hz) 7.94 (m, H-3 and 2H-1'), 8.49, 8.56, 8.65 and 8.68 (4s, CH^ of the isopropylidene groups). Irradia-tion of the solution at x 7.94 collapsed the t r i p l e t at x 5.32 into a doublet and irradiation at x 5.32 collapsed the doublet at x 4.27 into a singlet. Anal. Calcd. for C 1 4 H 2 0 0 6 : c> 58.32; H, 8.39; mol. wt. 288. Found: C, 58.20; H, 8.35; mol. wt. 288 (15 added to m/e 273). 3-Deoxy-3-£- (21-hydroxyethyl)-1,2:5,6-di-0-isopropylidene-21-O-p-tolylsulfonyl-a-Q-allofuranose (LXVIII) To a solution of 0.058 g of compound (LXVII) in 0.2 ml of anhydrous pyridine was added 0.090 g of p_-toluenesulfonyl chloride. The reaction - 103 -mixture, protected from moisture, was agitated and kept at room temperature for 0.5 hr. A few drops of water were then added to dissolve the pyridine hydrochloride crystals and the reaction mixture left stand for a further 0.5 hr. Water was added to precipitate the p_-toluenesulfonate which was removed by f i l t r a t i o n , washed with water and then air-dried; yield 0.082 g (92%). The p-toluenesulfonate recrystallized from methanol, decomposes slowly when left at room temperature; m.p. 88-89°, [ a ] 2 2 +48° (c, 3 in chloroform); Rf 0.7 on s i l i c a gel G using benzene-methanol (94:4) as r n n developer, x 3 2.18 and 2.66 (2d, 4 aromatic H), 4.28 (d, H-l J = 3.7 Hz), 5.54 (H-2), 7.56 (s, CH3 of tosyl group). Anal. Calcd. for C^H^OgS 1/2H20: C, 55.85; H, 6.92; S, 7.10. Found: C, 55.50; H, 6.90; S, 7.20. 2'-0-p_- Bromopheny lsulfonyl-3-deoxy-3-C_- (2' -hydroxyethyl)-1,2 :5,6-di-0-isopropylidene-a-D-allofuranose (LXIX) To a solution of 0.068 g of substance (LXVII) in 0.3 ml of anhydrous pyridine was added 0.11 g of freshly recrystallized p_-bromobenzenesulfonyl chloride. The product was worked up in the same way as the p-toluene-sulfonate and recrystallized from methanol: yield, 0.100 g (84%); m:p. 110-111°, [a] 2) 2 +38° (c, 1 in chloroform); Rf 0.8 on s i l i c a gel G using benzene-methanol (96:4) ad developer. The compound slowly decomposes at CDC1 room temperature, x 3 2.25 (m, 4 aromatic H), 4.28 (d, H-l, J = J., z 3.7 Hz), 5.50 (t, H-2). Anal. Calcd. for C^H^BrOgS: C, 47.33; H, 5.36; Br, 15.74; S, 6.51. Found: C, 47.41; H, 5.41; Br, 15.55; S, 6.18. - 104 -3-Deoxy-3-C-(2'-iodoethyl)-1,2:5,6-di-O-isopropylidene-ct-D-allofuranose (LXXI) To a solution of 82 mg of tosyl product previously dried at 50° under high vacuum, in 3 ml of acetone was added 87 mg of anhydrous sodium iodide. The reaction mixture protected from moisture, was kept at roomtemperature for three days (or unt i l a l l compound was consumed as evidenced by monitoring by t . l . c . on s i l i c a gel G using benzene-methanol (95:5) as developer. The crystalline mass was then filtered and washed with acetone and the f i l t r a t e evaporated under reduced pressure affording a half oily-half crystalline residue. The crystalline material which separated after addition of methylene chloride was discarded, the organic phase was dried over sodium sulfate and evaporated under reduced pressure yielding a colorless syrup. Purification by column chromatography on s i l i c a gel using methylene chloride-ethyl acetate (94:6) as developer gave a pure product as shown by t . l . c . CDC1 (Rf: 0.8) and n.m.r., but resisted a l l attempts at crystallization T 3 4,24 (d, H-l, J1 2 = 3.7 Hz), 5.34 (t, H-2). 2' -0_-p_-Bromobenzoyl-3-deoxy-3-C- (21 ^-hydroxyethyl) -1,2:5,6-di-0_-isopropylidene-a-D-allofuranose (LXX) To a solution of 0.101 g of compound (LXVII) in 3 ml of anhydrous pyridine was added 0.152 g of p-bromobenzoyl chloride. The reaction mixture, protected from moisture, was agitated to give a homogeneous solution and then left at room temperature for 6 hr. Water was then added to the reaction mixture and the p-bromobenzoate was'extracted twice with 5 ml of methylene chloride. The combined extracts were washed with a saturated solution of sodium bicarbonate, then with water, and f i n a l l y dried over anhydrous magnesium sulfate, and fil t e r e d . The f i l t r a t e was evaporated under reduced - 105 -pressure to yield a partially crystalline product. This product was triturated with a small volume of ethanol. The ethanolic f i l t r a t e was evaporated to dryness giving an o i l . Crystallization of the o i l from 95% aqueous methanol gave crystals; m.p. 58-59°, M 2 ) 2 + 4 9 ° (c» 1 i n chloroform). Rf 0.75 on CDC1 s i l i c a gel G using benzene-methanol (95:5) as developer. x 3 2.08 and 2.45 (2d, 4 aromatic H), 4.24 (d, H-l J } 2 = 3.8 Hz), 5.30 (t, H-2). Anal. Calcd. for C 2 1H 2 yBr0 7: C, 53.50; H, 5.77; Br, 16.98. Found: C, 53.63; H, 5.96; Br, 17.00. Partial hydrolysis of compound (LXVII) to yield 3-Deoxy-3-C_-(21-hydroxy-ethyl)-!, 2-0-isopropylidene-a-D-allofuranose (LXXII) To a solution of 0.78 g (0.0027 mole) of the branched-chain sugar (LXVII) in 4.5 ml of methanol was added 4.5 ml of 0.8% sulfuric acid. The reaction mixture was left stand at room temperature; for 3 hr. then neutralized with barium carbonate, boiled a few minutes to coagulate the precipitate, and f i l t e r e d . The f i l r a t e was evaporated to dryness. To the residue was added 20 ml of water and 4 ml of chloroform and the mixture vigorously shaken. The chloroform extract was dried over anhydrous sodium sulfate, fi l t e r e d and evaporated to dryness to yield = 0.04 g of starting material: The aqueous solution was evaporated to dryness and the residue was azeotroped with ethanol. The resulting o i l was extracted with magnesium sulfate, f i l t e r e d , and evaporated to dryness under reduced pressure; yield 0.622 g of o i l (92%); the substance was homogeneous by t . l . c , Rf 0.5, benzene-methanol (3:1) as developer; [a] 2, 2 +81° (c, 1 in chloroform). Mol. wt. (by mass spectrometry), 248 (15 added to m/e). Calcd. mol. wt. 248. - 106 -2',5,6-Tri-O-benzoyl-3-deoxy-3-C-(2'-hydroxyethyl)-1,2-0-isopropylidene-a -D-allofuranose (LXXIII) To a solution of 0.600 g of compound (LXXII) in 10 ml of anhydrous pyridine was added with stirring 1.2 ml of freshly d i s t i l l e d benzoyl chloride. The reaction mixture was left stand at room temperature overnight, although monitoring by t . l . c . showed the reaction to be complete in two hr. A few drops of water were added to dissolve the pyridine hydrochloride. The solution was then poured into 100 ml of a mixture of ice and water and extracted twice with 50 ml portions of chloroform. The combined chloroform extracts were washed with water, aqueous sodium bicarbonate, water, dried over anhydrous sodium sulfate, and fi l t e r e d . The f i l t r a t e was evaporated under reduced pressure and the residue azeotroped three times with 10 ml of toluene to remove traces of pyridine. The residue, dissolved in ethanol, was treated with Norite and the solution filtered . The f i l t r a t e on evaporation gave a colorless syrup, 1.18 g (87%); Rf 0.8 on s i l i c a gel G using benzene-methanol (98:2) as developer. For analysis, a sample was purified by thin layer chromatography on s i l i c a gel G; [a] 2) 2 +21° r n n (c, 1 in chloroform; x 3 4.12 (d, H-l, J = 3.7 Hz); 7.75 (m, H-3,1'), 8.46 and 8.70 (2s, 2CH3 of isopropylidene group), 2.01 and 2.55 (2m, 15 aromatic H). Anal. Calcd. for C 3 2 H 3 2 0 9 ' C ' 6 8 - 5 5 ; H' 5- 9 3- Found: C, 68.50; H, 5.74. Acetolysis of 2',5,6-tri-0-benzoyl-3-deoxy-3-C-(2'-hydroxyethyl)-1,2-0-isopropylidene-a-D-allofuranose (LXXIII) to yield l,2-Di-0-acetyl-2',5,6-tri-0-benzoyl-(21-hydroxyethyl)-3-D-allofuranose (LXXIV) - 107 -To a well s t i r r e d s o l u t i o n (kept at 0°) of the benzoate (LXXIII) (1.18 g) i n g l a c i a l a c e t i c a c i d (8 ml) and a c e t i c anhydride (0.8 ml) was added dropwise 0.44 ml of concentrated s u l f u r i c acid. A f t e r storage at room temperature f o r 3 days the s o l u t i o n was poured into 100 ml of vigorously s t i r r e d ice-water, and the r e s u l t i n g mixture was extracted with chloroform (3x30 ml). The combined chloroform extracts were washed with water (50 ml), aqueous saturated sodium bicarbonate (50 ml), water (50 ml), dried over magnesium s u l f a t e , f i l t e r e d , and evaporated to dryness i n vacuo to give 0.80 (75%) of a syrup (LXXIV). For analysis, a sample was sublimed at 175°/50 y, r n n m.p. 38-39°, [ a ] 2 2 =0° (c, 1 i n chloroform); T 3 3.92 (s, H - l ) , 4.82 (d, H-2, J 2 3 = 4.5 Hz), 7.95 (s, 2CH 3 of A c ) ; 2.02 and 2.58 (2m, 15 aromatic H). Work up of the r e a c t i o n mixture a f t e r one day yielded a mixture i n equal quantities of the 3anomer and a second component probably an i n t e r -mediate. 6-Benzamidopurine (LXXV) A mixture of 5.40 g (0.040 mole) of adenine and 27.40 g (0.120 mole) of benzoic anhydride was melted and then heated at approximately 180°C fo r 15 min. The s o f t yellow cake obtained a f t e r cooling was dissolved i n 300 ml of b o i l i n g ethanol, treated with Norite. and f i l t e r e d . On cooling the product c r y s t a l l i z e d as white meedles, which were c o l l e c t e d by f i l t r a t i o n , washed with 10 ml of cold ethanol and d r i e d , y i e l d 6.50 g (68%), p. 248-249°, l i t y i e l d 65% m.p. 238-239°; l i t . y i e l d 71% m.p. 239-242°. nujol , -1-. ',, o c- r n 0-. , v , d e u t e r i o d i m e t h y l sulfoxide Vmax m (amide C=0) 1600 (phenyl); T . 1.26 (s, H-2), 1.48 (s, H-8), 1.87 and 2.50 (2m, aromatic H), 7.92 (s, m - 108 -H-9). Mol. wt. (by mass spectrometry), 239, calcd., 239. Chloromercuri-6-benzamidopurine (LXXVI) To a stirred solution of 7.8 g (0.028 mole) of mercuric chloride in 100 ml of 50% aqueous ethanol was added 6.8 g (0.028 mole) of 6-benzamido-purine. To the resulting syspension, 10.3 ml of 10% aqueous sodium hydroxide (0.028 mole) was added dropwise with s t i r r i n g . The yellow mixture was stirred 1 hr. and then allowed to stand at room temperature for a period of 20 hr. The white solid was f i l t e r e d , washed with 25 ml of cold 50% aqueous ethanol and dried in vacuo over phosphorus pentaoxide: yield, 13 g (96%); l i t . 8 2 yield 96%. 136 1,2-0-Isopropylidene-ct-D-glucofuranose (LXXVII) To a solution of 4.5 g of 1,2:5,6-di-0-isopropylidene-a-D-glucofuranose in 23.4 ml of methanol was added 23.4 ml of 0.8% sulfuric acid. The reaction mixture was left stand at room temperature for 3 hr, then neutralized with barium carbonate, boiled for a few minutes to coagulate the precipitate and fi l t e r e d . The f i l t r a t e was evaporated under reduced pressure. Colorless needles of 1 ,:2-0-isopropyiidene-a-D-glucofuranose (LXXVII) were obtained by crystallization of the residue from methanol-ether, yield 3 g (79%), m.p. 160-161°, l i t . 1 3 6 yield 95%, m.p. 160-161°. 3,5,6-Tri-0_-benzoyl-1,2-0-isopropylidene-a-g-glucofuranose (LXXVIII) To a solution of 0.600 g of 1,2-0-isopropylidene-a-D-glucofuranose 8 ml of anhydrous pyridine was added with stirring 1.7 ml of freshly d i s t i l l e d benzoyl chloride. The reaction mixture was left stand at room temperature - 109 -overnight, and then quenched with a few drops of water. The solution was poured into 100 ml of a mixture of ice and water, then extracted three times with 50 ml portions of chloroform. The combined chloroform extracts were washed with water, aqueous sodium bicarbonate, water, dried over anhydrous sodium sulfate and f i l t e r e d . The f i l t r a t e was evaporated under reduced pressure and the residue azeotroped three times with 10 ml of toluene to remove traces of pyridine. The residue, dissolved in ethanol was treated with Norite and the solution fil t e r e d yielding after evaporation 1.2 g (83%) of (LXXVIII) as a colorless syrup. Rf 0.85 on t . l . c . using s i l i c a gel G and CDC1 benzene-methanol (98:2) as developer, T 3 3.93 (d, H-l, J = 3.5 Hz), » 8.40 and 8.24 (2s, 2CH3 of the isopropylidene group), 2.02 and 2.52 (2m, 15 •P-i 1 m ' aromatic H). v 1730 (benzoate C=0). max ^ l,2-Di-0-acetyl-3,5,6-tri-0_-benzoyl-D-glucofuranose (LXXIX) To a well stirred solution (kept at 0°) of the tribenzoate (LXXVIII) (1.2 g) in 10 ml of glacial acetic acid and 1 ml of acetic anhydride was added dropwise 0.65 ml of concentrated sulfuric acid. After storage at room temperature for 24 hr., the solution was poured into 100 ml of vigorously stirred ice-water, whereupon a solid material separated. The precipitate was f i l t e r e d , washed with cold water and dried under high vacuum affording 1.0 g (80% yield) of (LXXIX). Rf 0.70 (B anomer) and Rf 0.30 (probably the a anomer) on t . l . c . using s i l i c a gel G and benzene-methanol (98:2) as developer m.p. 45°, v n U' , 0~'' (cm-1) 1730 (benzoate C=0) , TDfl 1760 (acetate C=0) ; x 3 3.77 (s, H-l, 2/3 of a proton,6-anomer),.3.43 (d, H-l, 1/3 of a proton probably of thea anomer, ^ ~ 4-5 Hz); 2.04 and 2.54 (2m, 15 aromatic H), 7.83 (s, 4 protons), 7.88 (s, 2 protons). - 110 -9-B-D-Glucofuranosyladenine (LXXXI) A mixture of 0.300 g (0.0005 mole) of.sugar (LXXIX), 0.310 g (0.0065mole) of chloromercuri-6-benzamidopurine (LXXVI), 0.310 g of Celite, and 25 ml of ethylene dichloride was d i s t i l l e d under anhydrous conditions un t i l 5 ml of d i s t i l l a t e had been collected. To the partially cooled mixture was added a solution of 0.077 ml of titanium tetrachloride in 1 ml of ethylene dichloride, and the reaction was heated under reflux for 22 hr. The cooled reaction mixture was then poured into 25 ml of saturated sodium bicarbonate and stirred vigorously for two hours, and then filt e r e d through Celite. The Celite cake was washed with chloroform (3x10 ml) and the combined chloroform extracts shaken with the aqueous extract. The chloroform extract was evaporated to near dryness in vacuo at 40°. A solution of the residue in 7. ml of chloroform was washed with 7 ml of 30% aqueous potassium iodide and 7 ml of water, dried over sodium sulfate, filtered and evaporated to dryness under reduced pressure at 50° to give 0.348 g of a light yellow glass. This product was not homogeneous by t . l . c . on s i l i c a gel G with ethyl acetate as developer. Detection with a 30% solution of sulfuric acid revealed two zones with Rf of 0.9 and 0.6 (major zone corresponding to the blocked nucleoside (LXXX). T C D C 1 3 1.38 (s, H-2), 1.73 (s, H-8), 3.70 (d, H-l", J j , 2, =1.75 Hz), 3.45 (d, J = '4.5 Hz). A mixture of 0.19 g of (LXXX) (fi r s t dried by azeotroping with anhydrous benzene) in 5 ml of 0.1 N methanolic sodium methoxide was heated under reflux for 3 hr., with solution occurring at the boiling point. The cooled solution was neutralized with glacial acetic acid and then left stand at 0° for 1 hr. The crude amorphous nucleoside (LXXXI) was removed by f i l t r a t i o n - I l l -washed with 0.5 ml of cold methanol and dried, yield 0.024 g (22% from (LXXIX). Recrystallization of the product from water gave a white crystal-line solid, m.p. 267-269° with decomposition, [a]p 2 = -58° (1% in N HC]), T - * . l l ? O ^ Q ->™° r i 2 2 roo no. • v, un^ deuteriomethyl sulfbxid L i t . m.p. 268-270 , [a]p = -58 (1% in N HC1). T / 1.66 (s, H-2), 1.84 (s, H-8), 2,71 (s, NH^ ) 4.12 (d, H-l', J p 2, = 1.0 Hz) 5.75 (s, H-2'). Irradiation of the solution at T 5.75 collapsed the doublet at T 4.12 into a singlet. 9- (3-Deoxy-3-C- (2 * -hydroxyethyl) -3-Q-allofuranosyl) -adenine (LXXXIII) A mixture of 0.600 g (0.001 mole) of sugar (LXXIV), 0.540 g (0.00012 mol of chloromercuri-6-benzamidopurine (LXXVI), 0.540 g of Celite, and 60 ml of ethylene dichloride was d i s t i l l e d under anhydrous conditions un t i l 15 ml of d i s t i l l a t e had been collected. To the partially cooled mixture was added a solution of 0.137 ml of titanium tetrachloride in 4 ml of ethylene dichloride, and the reaction was heated under reflux for 22 hr. The cooled reaction mixture was poured into 50 ml of saturated sodium bicarbonate and stirred vigorously for 2 hr., and then filtered through Celite. The Celite cake was washed with chloroform (3x20 ml), and the combined chloroform extracts shaken with the aqueous extract. The chloroform extract was evaporated to near dryness in vacuo at 40°. A solution of the residue in 15 ml of chloroform was washed with 15 ml of 30% aqueous potassium iodide and 15 ml of water, dried over sodium sulfate, f i l t e r e d , and evaporated to dryness under reduced ^pressure at 50° to give 0.647 g of the crude blocked nucleoside (LXXXII). A mixture of 0.647 g of (LXXXII) ( f i r s t dried by azeotroping with anhydrous benzene) in 16 ml of 0.1 N methanolic sodium methoxide was heated under reflux for 3 hr., with solution occurring at the boiling point. The - 112 -cooled solution was neutralized with glacial acetic acid and stirred at 0° for 1 hr., which caused the separation of the crude crystalline nucleoside (LXXXIII). The product was removed by f i l t r a t i o n , washed with 1 ml of cold methanol and dried; yield 0.156 g (48% from (LXXIV)). Two recrystalliz-ations from 80% aqueous methanol (with treatment with a small quantity of charcoal) gave a white crystalline solid, m.p. 265-266°, [a]^ 2 -44° (c, 1.5 in dimethyl sulfoxide); X p H 1 (m y) 257 (e 13,300), 208 (e 19,100); Xmax 1 3 ( m y ) 2 6 0 ( e 1 3> 5 0°)> 2 1 1 Ce 13,800); (m u) 260 (e 13,700), 208 (e 19,000); Rf (adenine) 1.5 on Whatman paper No. 1 (water as developer); deuteriodimethyl sulfoxide n , n , u n Q_ __ , x ' 1.69 (s, H-2), 1.87 (s, H-8), 2.75 (s, NH2), 4.18 (d, H-l', Jv 2, = 3.5 Hz), 4.44 and 4.54 (2d), 5.44 (m), 6.3 (m, near D20 peak), 7.6 (m, H-31),..8.2 (m, 2H-1"). Double irradiation of (LXXXIII) at x 5.44 collapsed the doublets at x 4.18 and x 4.44 into singlets. Addition of some D20 resulted in the disappearance of the doublets at x 4.44 and 4.54. The 0.R.D. curve (see Discussion), of (LXXXIII) showed a negative Cotton effect (c, 0.003, water). For analysis, compound (LXXXIII) was dried at 100° for 1 hr. under vacuum. Anal. Calcd. for C^H^N^: C, 47.96; H, 5.89; N, 21.6. Found: C, 48.22; H, 6.08; N, 21.8. The mother liquor obtained after separation of the crude nucleoside (LXXXIII) was evaporated to half i t s volume whereupon 0.070 g of crystalline material separated. A n.m.r. spectrum of this indicated i t was a mixture of sodium acetate and sodium benzoate. The remaining mother liquor was evaporated to dryness and the.residue was then partitioned between equal volumes of chloroform and water. The insoluble material was removed by f i l t r a t i o n and combined with the water fraction for preparative paper chromatography. The aqueous solution was evaporated to dryness jn vacuo. - 113 -A 0.2 ml aliquot of the combined residue, was dissolved in water, and fractionated by paper chromatography using water as developer, and detection by ultraviolet. The main zone was extracted with water; yield, about 0.0015 g, m.p. 261-263°; mixed m.p. with authentic nucleoside (LXXXIII) was 261-263°. The remaining zones were adenine and a degraded sugar. An unsuccessful attempt was made to separate the remaining portion of the residue from the mother liquor by chromatography on Dowex 1x2 (OH) (3 g, 1.0 x 10 cm) according to the procedure of Reist, Calkins and Goodman.13^ The eluted zones did not contain any nucleoside. 3-Deoxy-3-C_- (carbomethoxymethyl)-l,2-0-isopropylidene-a-D-allofuranose (LXXXIV) To a solution of 1.27 g of the branched-chain sugar (LXI) in 7 ml of methanol was added 0.7 ml of a 0.8% aqueous solution of sulfuric acid. The reaction mixture was l e f t stand at room temperature for 24 hr (the reaction mixture was monitored by t . l . c . and quenched when almost a l l of the starting material was consumed and only a trace of the completely de-O-acetonated product had appeared. The reaction mixture was neutralized with barium carbonate f i l t e r e d , and the f i l t r a t e evaporated under reduced pressure. The solid residue was extracted with 15 ml of chloroform and the mixture filtered . Evaporation of the f i l t r a t e yielded 0.972 g (88%) of an o i l which crystallized upon standing and was pure (by t.l.c.) for the next step. For analysis, a sample was sublimed at 98°/5 y to yield 3-deoxy-3-C_-(carbomethoxymethyl)-1,2-O-isopropylidene-a-D-allofuranose, m.p. 89-90°, [ a ] 2 2 +63° (c, 1 in chloroform) which migrated as one spot with an Rf of 0.6 on s i l i c a gel G CDC1 using benzene-methanol (3:1) as developer, x 3 4.17 (d, H-l, J. 9 = 3.5 Hz), 5.18 (t, H-2), 6.26 (s, due to CH3 of ester), 8.47 and 8.66 (2s, , - 114 -assigned to the CH^ of the isopropylidene group). Anal. Calcd. for C 1 2H 0 : C, 52.16; H, 7.29. Found: C, 51.88; H, 7.10. Sodium metaperiodate degradation of 3-deoxy-3-C_- (carbomethoxymethyl)-1,2-0-isopropylidene- a-D-allofuranose (LXXXIV) to yield 3-deoxy-3-C-(2'-hydroxy-ethyl) -1,2-0-isopropylidene-a-D-ribofuranose (LXXXV) To a well-stirred solution of 3-deoxy-3-C_-(carbomethoxymethyl)-1,2-0-isopropylidene-a-D-allofuranose (0.443 g, 0.0016 mole) in 22 ml of water was added sodium metaperiodate (0.342 g, 0.0016 mole). The pH of the solution was adjusted to 7 and kept at this value (pH paper) during the oxidation by the careful addition of 0.1 N sodium hydroxide. The reaction mixture, was extracted with chloroform (4x35 ml) and the combined extracts were washed with 35 ml of water, dried over anhydrous sodium sulfate, and evaporated under reduced presssure to yield 0.236 g (60%) of an o i l ; TC D C 1 3 4.04 (d, H-l, J. =3 Hz), 0.40 (d, H of CHO, J = 2.25 Hz), 3 D , 4 5.14 (t, H -2), 6.27 (s, CH3 of ester), 8.47 and 8.63 (2s, assigned to the CH^  of the isopropylidene group). To the above aldehydp sugar (0.163 g, 0.00067 mole) dissolved in tetrahydrofuran (6 ml) was added a mixture of lithium aluminum hydride (0.500 g) in tetrahydrofuran (20 ml) and the resulting mixture was heated under reflux for 1 hr. The excess lithium aluminum hydride was destroyed by dropwise addition of water. After the residue was triturated with 3 ml of ethanol, the mixture was f i l t e r e d , and the f i l t r a t e evaporated under reduced pressure. The residue was now triturated with chloroform, filtered and the f i l t r a t e evaporated under reduced pressure to give a pure sugar (LXXXV), - 115 -mri yield 0.105 g (72%), [ a ] 2 2 +72° (c, 1 in chloroform); t 3.4.20 (d, H-l, J1 2 = 3.5 Hz), 5.31 (t, H-2, J 2 z = 4.5 Hz), 7.79 (m, H-3), 8.99 and 8.65 (2s, assigned to the CH^  of the isopropylidene group). Irradiation of H-2 led to a collapse of H-l to a singlet. Irradiation of H-l led to a collapse of the H-2 t r i p l e t to a doublet, ^  g = 4.5 Hz. Irradiation of H-3 multiplet led to a collapse of the H-2 tr i p l e t to a doublet having 2 = 3.5 Hz. The molecular weight of (LXXXV) by mass spectroscopy was 218 (15 added to m/e). Calcd. 218. 2 *-5-Di-0-benzoyl-3-deoxy-3-C-(21-hydroxyethyl)-1,2-0-isopropylidene-a-D-ribofuranose (LXXXVI). 3-Deoxy-3-C_- (2'-hydroxyethyl)-1,2-0-isopropylidene-a-D-ribofuranose (LXXXV) was benzoylated using the same procedure as for compound (LXXIII) (page 106 ) to give 2',5-di-0-benzoyl-3-deoxy-3-£-(2'-hydroxyethyl)-1,2-£-isopropylidene-a-D-ribofuranose as a syrup in 81% yield; [a] 2) 2 +66° (c, CDri 1 in chloroform); T 3 4.12 (d, H-l, J± 2 = 4.12 Hz), 5.24 (t, H-2), 7.86 (m, H-3 and the two hydrogens of H-l'), 8.44 and 8.68 (2s, assigned to the CH^ of the isopropylidene group). For analysis, a sample was purified by t . l . c . ( s i l i c a gel G using benzene-methanol, 98:2, as developer). Anal. Calcd .for C 2 4 H 2 6 ° 7 : c> 67.58; H, 6.15. Found: C, 67.48; H, 6.21. Acetolysis of 2',5-di-0-benzoyl-3-deoxy-3-C-(2'-hydroxyethyl)-1,2-0-isopropylidene-a-D-ribofuranose (LXXXVI) to yield 1,2-di-0-acetyl-2',5-di-0-benzoyl-3-deoxy-3-C-(2'-hydroxyethyl)-8-D-ribofuranose (LXXXVII). An amount of 0.112 g of 2',5-di-0-benzoyl-3-deoxy-3-^C-(2*-hydroxyethyl)-1,2-0-isopropylidene-a-D-ribofuranose was allowed to react with a mixture of acetic acid, acetic anhydride and concentrated sulfuric acid and the product worked up in the same way as that already described for compound (LXXIV) (page 106 ), The diacetate (obtained in 78% yield) was almost pure 3-anomer when the acetolysis was allowed to proceed for three days. Rf of 8-anomer = 0.8 and of the non identified product = 0.6 (on s i l i c a gel G using benzene-CDC 1 methanol, 95:5, as developer) T 3 3.87 (s, H-l), 4.69 (d, H-2, J 2 3 = 4.12 Hz), 7.91 and 8.08 (2s, assigned to the CH^ of the two acetate groups). Irradiation of H-l led to a collapse of the H-2 doublet into a singlet. Condensation of l,2-di-0-acetyl-2*,5-di-0-benzoyl-3-deoxy-3-C-(2'-hydroxy-ethyl) -8-D-ribofuranose (LXXXVII) with chloromercuri-6-benzamidopurine to yield 9- [3' -deoxy-3' -C- (2''-hydroxyethyl^B-D-ribofuranosyl] -adenine (LXXXVIII) An amount of 0.404 g of 1,2-di-0-acetyl-2',5-di-0-benzoyl-3-deoxy-3-C-(2'-hydroxyethyl)-8-D-ribofuranose was condensed with chloromercuri-6-benzamidopurine (0.506 g) according to the same procedure already described (see preparation of nucleoside (LXXXIII)) to give 0.490 g of the blocked nucleoside as an amorphous glass. The blocked nucleoside (0.490 g) was deacylated with sodium methoxide as described in the preparation of compound (LXXXIII) to afford an amount of 0.090 g of 9-[3'-deoxy-3'-C-(2"-hydroxyethyl)-g-D-ribofuranosyl]-adenine (LXXXVIII). The nucleoside was recrystallized from methanol and dried at 80° for 1 hr under reduced pressure, R A d = 1.6 (paper, water as developer); m.p. 187-188°, [a] 2) 2 -7° (c, 3 in dimethyl sulfoxide). For analysis the nucleoside was purified (a trace of impurity in the crystallised product) by paper chromatography using water as developer and dried at 70° under reduced pressure. At 100° the nucleo-side decomposes. The chromatographed nucleoside was a monohydrate; m.p. 182-184°, [ a ] 2 2 -7° (c, 3 in dimethy] sulfoxide) ; A ? ^ 1 208 and 257 with e 16,800 and 11,800; A p H 7 208 and 260 with e 16,900 and 11,600; X p H 1 3 ' ' max max - 117 -on J n/n • ^  -,, nnn J 1 n rno deuteriodimethy1 sulfoxide . ' 211 and 260 with E 13,000 and 11,500; x 1.66 (s, H-2), 1.94 (s, H-8), 2.8 (s NH0), 4.13 (s, H-l»), 4.36 (d, C-2' (OH)), 4.9 (m), 5.52 (t ) , 5.68 (t, H-21), 6.2 (m), 6.5 (ra, overlaps D20 peak), 7.62 (m, H-3'), 8.40 (m, H-l"). In.D2° t h e signal at T 4.36 disappeared. Irradiation of the solution at x 4.36 collapsed the tr i p l e t a t T 5.68 to a doublet having J 2 , 3, = 4.5 Hz. Irradiation at T 5.68 collapsed the doublet at T 4.36 to a singlet and altered the H-3' signal at x 7.6. The O.R.D. spectrum of (LXXXVIII) showed a negative Cotton effect (see Discussion). [ 4 ] 2 7 2 - 2300 (t) , [ c|>] 2 5 8 0°, [ * ] 2 3 8 +4460 (p) (c, 0.004 in H,,0) . The mother liquor from the crude nucleoside (LXXXVIII) was worked up as described for the mother liquor of the nucleoside (LXXXIII) to yield an additional 0.014 g of (LXXXVIII) ( the total yield of (LXXXVIII) was 37% based on the diacetate). Anal. Calcd. for C12H N 0 1H20: C, 45.99; H, 6.06; N, 22.35. Found: C, 46.00; H, 5.96; N, 22.23. Sodium metaperiodate oxidation and reduction of the alio nucleoside (LXXXIII) to yield the ribo nucleoside (LXXXVIII) To a solution of the alio nucleoside (LXXXIII) (0.005 g) in 1.5 ml of water and 0.3 ml of ethanol was added a solution of sodium metaperiodate (0.0045 g) in 0.25 ml of water. The reaction mixture was left stand at room temperature for 0.5 hr. Sodium borohydride (0.020 g) was added to the reaction mixture which was then left stand at room temperature for 35 min. 10% Acetic acid (0.25 ml) was added to decompose excess sodium borohydride. The reaction mixture was evaporated under reduced pressure and the residue azeotroped three times with ethanol. The product was separated by paper chromatography using water as a developer to give 0.005 g of a nucleoside - 118 -having an identical n.m.r. spectrum to that of the ribo nucleoside (LXXXVIII). Reaction of 1,2:5,6-di-0_-isopropylidene-3-£-p-tolylsulfonyl-a-D-glucofuranose with tetramethylarnmonium hydroxide in dimethyl sulfoxide to yield 1,2:5,6-di-0-isopropylidene-a-D-glucofuranose (XIV) 1,2:5,6-Di-£-isopropylidene-3-0-p_-tolylsulfonyl-a-D-glucofuranose (1.3 g) was allowed to react with 1.5 ml of an aqueous solution of tetramethyl-arnmonium hydroxide i n 20 ml of dimethyl sulfoxide for 3 days at room temperature. An amount of 40 ml of water was then added to the reaction mixture. The mixture was extracted three times with 60 ml of chloroform. The combined chloroform extracts were washed with water, dried over anhydrous sodium sulfate and then evaporated under reduced pressure to yield 0.7 g (89%) of crystalline material which was recrystallized from petroleum ether (b.p. 65-110°), m.p. 110° (0.6 g), [a] 2, 2 -19° (c, 3.5 acetone). The mixed m.p. of this substance and authentic l,2:5,6-di-0_-isopropylidene-a-D-glucofuranose was 109-110°. Authentic l,2:5,6-di-0-isopropylidene-a-D-gluco-furanose has [a].2,2 -19°. Reaction of 1,2:5,6-di-0-isopropylidene-3-G)-p_-nitrophenylsulfonyl-a-D-glucofuranose (LXXXIX) with tetramethylarnmonium hydroxide in dimethyl sulfoxide to yield 3-0-[p-(3-deoxy-l,2:5,6-di-0-isopropylidene-a-D-glucofuranose-3-yl)-oxyphenylsulfonyl] -1,2:5,6-di-O-isopropylidene-a-Q-glucofuranose (XC) and 1,2:5,6-di-0-isopropylidene-3-£-p_-nitrophenyl-a-D-glucofuranose (XCI) To a solution of l,2:5,6-di-0-isopropylidene-3-0-p-nitrophenylsulfonyl-a-D-glucofuranose (10 g) in 40 ml of dimethyl sulfoxide was added 10 ml of a 25% aqueous solution of tetramethylarnmonium hydroxide. The reaction mixture, kept at room temperature for 2 days, gradually changed in color from - 119 -green to orange-red. The crystalline compound (XC) that separated was removed by f i l t r a t i o n , washed with water, dried, and then recrystallized from petroleum ether (b.p. 30-60°), yield 4 g (54%), m.p. 93-94°, [a ] 2 D 2 -78° CC1 (c, 2 in chloroform), Rf 0.3 (benzene-methanol 99:1 v/v), v 4 1470 and v J ' J ' max - l r n n 1315 cm . (O -SGy) ; ^ 3 1.98, 2.80 (4H, aromatic), 4.19 (H-l, doublet, J j 2 = 3.5 Hz), 4.21 (H-l«, doublet, J 1 t 2, =3.7 Hz), 5.28 (1 H, doublet J„ . = 3.5 Hzi due to H-2), 5.56 (1 H, doublet, J = 3.7 Hz, due to ' 2,1 ' J , \ > 2' ,1' • H-2'), 5.36 (2 H, doublet, J 3 4 and J 3 , 4, = 3.0 Hz, due to H-3 and H-3'), 8.34 (24 H, 8 peaks due to CH3). The partial n.m.r. of (XC) is shown in Figure 8 ) . Mass spectrometry: 643, 585, 527, 502 (molecular weight found by adding 15 to 643). Anal. Calcd. for C^H^O^S: C, 54.70; H, 6.43; S , 4.87; mol. wt. 658.73. Found: C, 54.42; H, 6.33; S, 5.08; N, 0.0; mol. wt. 658. The mother liquor of the above reaction mixture was poured into an equal volume of water and the solution then extracted 3 times with 100 ml aliquots of chloroform. The combined chloroform extracts were dried with anhydrous sodium sulfate, and subsequently evaporated under vacuum. The residue was extracted with 100 ml of petroleum ether (b.p. 30-60°). The extract was evaporated under reduced pressure affording 0.32 g of (XIV) as a solid which was recrystallized from petroleum ether (b.p. 30-60°), m.p. 109-110°; mixed m.p. with an authentic sample of 1,2:5,6-di-O-isopropylidene-128 a-Q-glucofuranose was 109-110°. Characterization of product (XCI) The o i l remaining after the petroleum ether extraction of the residue was a mixture of the compound (XC) and (XCI). This mixture was separated - 120 -by preparative t . l . c . Product (XCI) (1.2 g (10%)), was recrystallized from petroleum ether (b.p. 30-60°), m.p. 136-137°, [ a ] 2 2 -52° (c, 2 in chloroform), Rf 0.75.(benzene-methanol 99:1 v/v) V C C 1 4 cm"1 (N0 7), T C D C 1 3 1.79 and 2.96 (4H, 2 doublets), 4.23 (IH, ^  2 = 4.0 Hz, due to H-l), 5.35 (IH, doublet, J 3 4 = 3.0 Hz, due to H-3), 5.57 (IH, doublet, J 2 1 = 4-° H z> due to the H-2), 8.50, 8.65, 8.73, 8.78 (12H, due to 4 CH3). Mass spectro-metry: m/e 366, 351, 336, 308, 266, 248. Anal. Calcd. for C..H 0oN: C, 56.68; H, 6.07; N, 3.67; mol.wt. 381.39. 18 23 o Found: C, 56.35; H, 6.66; N, 3.90; S. 0.0; mol. wt. 381 (15 added to parent m/e peak). 159 Reductive cleavage of (XC) using sodium amalgam To a solution of 100 mg of (XC) in 6 ml of absolute methanol, was added 1.6 g of sodium amalgam. After the reaction mixture was stirred for 14 hr 159 at room temperature, i t was worked up according to a known procedure. The white residue obtained after evaporation of the fin a l ether extract showed by t . l . c . (using 100% isopropyl alcohol as developer) two major products and one minor one. Elution of each zone of the preparative t . l . c . plates with ethanol:water 1:1 v/v gave a quantitative yield of (XIV) (130 mg), and (XCII) (43 mg). Fraction (XIV) was recrystallized from petroleum ether m.p. 109-110°; mixed m.p. with authentic sample of 1,2:5,6-di-0-isopropylidene-a-D-glucofuranose, 109-110°. Product (XCII), m.p. 190-191°, Rf 0.30 (100% isopropyl alcohol) was water soluble. x D2° 2.43, 2.91 (2 doublets due to 4 aromatic protons),3.98 (IH, J =4.0 Hz, due to H-l), 8.55 (12H, t r i p l e t , due to 4CH3), [ a ] 2 2 = -19° (c, 0.4 in water). Compound (XCII) had a trace of an impurity which could not be removed and was not analysed. - 121 -T T T J 5 2 Ethyl p_-nitrobenzenesulfonate (XC i i JJ A cooled solution of sodium ethoxide prepared by dissolving 1.15 g of sodium in 50 ml of absolute ethanol was added dropwise with stirring to a solution of 11.5 g of freshly recrystallized p_-nitrobenzenesulfonyl chloride in 150 ml of ether and 20 ml of ethanol. After 24 hr, the reaction mixture was poured into 300 ml of water and then evaporated under reduced pressure to remove a l l of the ether. The aqueous layer was extracted with two portions (100 ml) of chloroform and the combined extract washed successively with 50 ml of water, 150 ml of sodium bicarbonate, and two 60 ml portions of water. The chloroform solution was dried over anydrous sodium sulfate, f i l t e r e d and evaporated to dryness at reduced pressure. The crude pale yellow ester obtained in 70% yield (8.4 g) was recrystallized from petroleum ether 152 to give practically colorless crystals m.p. 90-91°, l i t . m.p. 92-92.5°, rue i yield 81%. N.m.r. T 3 1.59, 1.82 (A 2B 2 quartet of 4 aromatic H), 5.72 ( q. CH„) 8.62 (t. CH ). v C C l 4 (cm"1) 1180 and 1340 (-SO,-), 1520 (NO.) Cyclohexyl p-nitrobenzenesulfonate (XCVI) 1^ To a mixture of 13.3 g (0.133 mole) of cyclohexanol and 150 ml of dry benzene contained in a flask protected by a drying tube and immersed in an ice-salt-bath was. added 39.2 g (0.177 mole) of freshly recrystallized p_-nitrobenzenesulfonyl chloride. After stirring magnetically for 40 min. the reaction mixture was poured into 150 ml of concentrated hydrochloric acid in 700 ml of ice and water. The precipitated solid was fi l t e r e d , dried and recrystallized from benzene-petroleum ether yielding 27 g (70% yield) of light yellow crystals, m.p. 77-78°, l i t . 1 6 1 m.p. 78-79°, yield 58%, n.m.r. CDC1 T 3 1.66, 1.92 (A 2B 2 quartet of 4 aromatic H), 5.35 (s, IH), 8.5 (m, 10H), v C C l 4 (cm-1) 1540 (N0„), 1190 and 1350 (-SO.-). H13.X j - 122 -Reaction of ethyl p-nitrobenzenesulfonate (XCIII) with tetramethylarnmonium hydroxide in dimethyl sulfoxide to yield tetramethylarnmonium p_-nitro-benzenesulfonate (XCIV) and p_-nitfophenetole (XCV) To a solution of 1 g of ethyl p_-nitrobenzenesulfonate (XCIII) in 8 ml of dimethyl sulfoxide was added 1.7 g of 25% aqueous tetramethylarnmonium hydroxide. After the reaction mixture was left standing at room temperature for 5 hr, the crystals (0.50 gj were removed by filtration', and then recrystallized from petroleum ether (b.p. 60-90°), m.p. 283-284°, Rf 0.80 CDC1 on alumina (benzene:methanol, 75:25), T 3 1.44 (4H, quartet, due to 4 aromatic protons), 6.46 (12H, t r i p l e t , due to 4 CH^). The mother liquor was poured into an equal volume o f water and then extracted three times with chloroform. The remaining water-dimethyl sulfoxide mixture was evaporated under reduced pressure. The remaining residue (0.450 g), had a m.p. of 283-284° and was identical to the crystals which precipitated from the reaction mixture. The over-all yield of tetramethylarnmonium p_-nitrobenzenesulfonate (XCIV) was 90%. Anal. Calcd. for C-pH^NO S: C, 43.46; H, 5.90; N, 10.13. Found: C, 43.84; H, 5.94; N, 10.32. The chloroform extract was dried with sodium sulfate and then evapora-ted under reduced pressure. The residual o i l (0.070 g) was purified by preparative t . l . c . on alumina (benzene-petroleum ether (30-60°)(60-40 v/v)) to give a crystalline (m.p. 56-57°) material (XCV). TCDCl3 1.85, 3.13 (2 doublets due to 4 aromatic H), 5.92 (2 protons, quartet), 8.55 (3 protons t r i p l e t ) . The n.m.r. spectrum and m.p. of an authentic sample of p-nitro-phenetole were identical to those of the substance obtained from the above chromatographic separation. - 123 -Reaction of cyclohexyl p_-nitrobenzenesulfonate (XCVI) with tetramethyl-ammonium hydroxide in dimethyl sulfoxide Cyclohexyl p_-nitrobenzenesulfonate (XCVI) was allowed to react with tetramethylammonium hydroxide in dimethyl sulfoxide as described in the preceding section. Work up of the reaction mixture gave an almost quantitative yield pf tetramethylammonium p_-nitrobenzenesulfonate (XCIV). 79 1,2:5,6-Di-0_-isopropylidene-3-0_-£-tolylsulfonyl-a-D-giucofuranose (XXIV) A mixture of 1,2:5,6-di-£-isopropylidene-a-D-glucofuranose (60 g), £-toluenesulfonyl chloride (66 g) in anhydrous pyridine (120 ml) was allowed to stand at room temperature for 24 hr. After addition of enough water to solubilize the pyridine hydrochloride formed,an o i l separated which crystallized upon cooling. Filtration and recrystallization from 70 methanol gave 77 g (75% yield), m.p. 119-120°, l i t . m.p. 120-121°, n.m.r. x 3 2.22, 2.69 (A 2B 0 quartet of 4 aromatic H ), 4.18 (d, H-l, J ' • = 3.5 Hz), 7.56 (s, CH ), V C C 1 4 (cm-1) 1190 and 1380 (-SO--), j. y 2 «J in 3.x o 1,2:5,6-Di-0_-isopropylidene-3-0_-£-nitrophenylsulfonyl-a-D-glucofuranose (LXXXIX). A mixture of l,2:5,6-di-0_-isopropylidene-a-D-glucofuranose (50 g), p_-nitrobenzenesulfonyl chloride (65 g) in anhydrous pyridine (100 ml) was allowed to stand at room temperature for 48 hr. After addition of enough water to solubilize the pyridine hydrochloride formed, an o i l separated which crystallized upon cooling. The crystalline compound was recovered by f i l t r a t i o n . Recrystallization from methanol yielded 63 g (84%) m.p. rnn 108-109°. N.m.r. x 3 1.64, 1.86 ( A 2 B quartet of 4 aromatic H), 4.10 r n - i (d, H-l, J = 3.5 Hz, v 4 (cm ) 1530, 1340 (NO ), 1190 and 1380 (-SO--). JL s Z ITlcLX Z. o - 124 -Reaction of 1,2:5,6-di-0-isopropylidene-3-0-£-tolylsulfonyl-a-D-gluco-furanose (XXIV) with potassium jt-butoxide in dimethyl sulfoxide A solution of 0.5 g of (XXIV) and 0.5 g of potassium t_-butoxide in 20 ml of dimethyl sulfoxide was protected from moisture and stirred at room temeprature. After 30 min. .the reaction mixture was poured into 500 ml of ice-water and a few drops of acetic acid (30%) were added unt i l neutrality. The solution was then extracted three times with 300 ml of chloroform, the combined chloroform extracts were washed with water and dried over sodium sulfate. Evaporation under reduced pressure gave a half oily-half crystalline residue which was shown by n.m.r. and t . l . c . using s i l i c a gel G (benzene-methanol 93:7 as developer) to be mainly composed of the di-0-isopropylidene derivative (XIV) Rf = 0.5. 82 3-Deoxy-l,2:5,6-di-0-isopropylideneTq-D-erythro-hex-3-enose (II) A finely ground mixture of 10 g of 1,2: S^-di-O-isopropylidene-S-O-^tolylsulfonyl-a-Q-glucofuranose (XXIV) and 30 g of potassium hydroxide flakes was placed in a suction flask equipped with a cold finger for sublim-ation and half immersed in an o i l bath. Sublimation over a period of 4 hr. at a pressure of 50 y .and a temperature rising from 80° to 110° gave 8 2 3.3 g (56% yield) of a white crystalline product (II), m.p. 49-50°, l i t . on KBr -1 m.p. 51-52°, yield 65% , l i t . m.p. 50°. v (cm ) 1670 (-C=C-), r ' J r max v • n.m.r. T 3 4.06 (d, 2 = 4.75 Hz), 8.64 (s, 3CH3), 8.71 (s, ICHj). This material changed slowly to a syrup after a few days at room temperature but can be kept almost indefinitely at -10°. Under the same conditions, when l,2:5,6-di-£-isopropylidene-3-0-£-nitrobenzenesulfonyl-a-D-glucofuranose (LXXXIX) was used instead of (XXIV) the sublimation led mainly to the formation of the diisopropylidene compound - 125 -(XIV) and to a mixture having a m.p. 80-90°, with no formation of pure 3,4-unsaturated compound (II). When (LXXXIX) was reacted with KOH at a temperature of 120° for a short time (1 hr.), (II) was obtained in 15% yield. Reaction of (II) with carbon monoxide and hydrogen under oxo conditions Experiment I To a solution of 1.5 g of 3-deoxy-l,2:5,6-di-0-isopropylidene-a-D-erythro-hex-3-enose (II) in anhydrous benzene (20 ml), contained in the glass liner of a high pressure reaction vessel, was added dicobalt octacarbonyl (0.3 g). The stoppered liner was inserted into the autoclave, which was flushed with carbon monoxide. Carbon monoxide was then added to a pressure of 800 psi, followed by hydrogen to a total pressure of 3200 psi and the reactants were heated, with rocking at 130° for 1 hr. and at 140° for 1.5 hr. After cooling to room temperature overnight, unreacted gas was released and almost a l l of the dark colored solution was heated until the volume of the solution was reduced to half, then filtered . Evaporation of the solvent under reduced pressure gave a colored syrup which was treated with Norite in chloroform, fi l t e r e d and evaporated affording a yellow syrup 1.2 g. Thin layer chromatography on s i l i c a gel G using benzene-methanol (98:2) as developer indicated 8 spots, and the n.m.r. spectrum showed mainly 2CH3 signals in the high f i e l d region. The remaining aliquot was d i s t i l l e d and the d i s t i l l a t e analyzed by g.l.c. which showed the presence of benzene and isopropyl alcohol (same retention time as authentic samples). - 126 -Experiment II The oxo reaction was carried out and worked up as previously using 3 g of (II), 15 ml of benzene, 0.3 g of dicobalt octacarbonyl, 2000 psi of carbon monoxide and 300 psi of hydrogen, at a temperature of 130° for 1 hr. The dark coloured syrup (2.4 g) obtained after treatment with Norite, f i l t e r e d through Celite,jand evaporated afforded 2.2 g of a pale | . i • . • • • yellow syrup; major component,1 Rf 0.7, on s i l i c a gel G using benzene-methanol (99:1) as developer. J Separation of 390 mg of oxo product by t . l . c . using the latter system yielded after extraction of the major component with chloroform a pale yellow syrup'70 mg. v f l l m (cm"1) 1670, T C C l 4 3.84 (m, IH), 5.06 (d, ni3.x ' \ f t . 2H), 8.5 (s, 2CH3 of isopropylidene group), 8.50 and 8.68 (2s, 2CH3 of isopropylidene group). Extraction with ethanol of the remaining zone on t . l . c , gjave after a second extraction with boiling chloroform, 150 mg of a dark yellow syrup v f l l m (cm-1) 1600, T C D C 1 3 3.so (d), 4.25 (broad s), signals il between 7 and 8, and between 8 and 9. To a solution of 330 mg of crude oxo product in 5 ml of methanol-water (2:3),-was added with sti r r i n g a solution of 600 mg of sodium borohydride in 10 ml of water. The reaction was stirred for 4 hr, neutralized with glacial acetic acid and evaporated to dryness. The resulting residue was dried over anhydrous magnesium sulfate and extracted with hot chloroform. Evaporation afforded a syrup (300 mg) which was immediately acetylated with 0.8 ml of acetic anhydride in 6 ml of pyridine. After being kept at room temperature for one day the solution was poured into 50 ml of ice-Water whereupon a dark o i l separated. Extraction with 3x50 ml of chloroform gave 200 mg of acetylated product. G.l.c. of this product showed two - 127 -major components; retention times 16 min. and 40 min. (m.p. 101-102°) in a ratio of 1 to 6 respectively. Experiment III Experiment III was carried out using 2 g of (II) in 20 ml of benzene with 2200 psi of hydrogen and 700 psi of carbon monoxide at a temperature of 135° for 1 hr. After release of the unreacted gas the solution was poured on a F l O r i s i l column and the catalyst eluted with 500 ml of petroleum ether (b.p. 30-60°). Elution of the reaction products with a 500 ml portion of benzene-isopropyl alcohol 98:2 gave fraction A and B. Subsequent elution with a 10% and 25% solution gave fraction C and D. T . l . c , using s i l i c a gel G and benzene-methanol as developer, of fraction A eluted from the column with 100 ml of benzene-isopropyl alcohol (98:2) showed two spots with! Rf = 0.39 and 0.44. In the n.m.r. spectrum almost no signal could be detected in the T region 8 to 9 (CH^ of isopropylidene groups). T.l.c. of the remaining B, C, and D portions indicated no major component. Experiment IV Reaction of 2.7 g of (II) in 20 ml of acetone in a high pressure vessel with 1100 psi of carbon monoxide and 1100 psi of hydrogen in presence of 0.3 g of preformed dicobalt octacarbonyl at a temperature of 135° for 1 hr. afforded after the usual work up (Experiments I and I I ) , 1.2 g of a yellow syrup: v (cm" ) 1710 and 1600. T.l.c. on s i l i c a gel G max -using benzene-methanol (95:5) as developer indicated at least six components. The n.m.r. spectrum showed that partial hydrogenolysis of the isopropylidene group had taken place. -- 128 -REFERENCES 1. H.R. Bentley, K.G. Cunningham and F.S. Spring, J. Chem. Soc. 2301 (1951). 2. E.A. Kaczka, N.R. Trenner, B. Arison, R.W. Walker and F. Folkers, Biochem. Biophys. Res. Commun. 14_, 456 (1964). 3. R.U. Lemieux and M.L. Wolfrom, Advances Carbohydrate Chem. 3_, 337 (1948). 4. F. ShafizH.deh, Advances Carbohydrate Chem. 11_, 263 (1956). 5. J.S. Webb, R.W. Broschard, D.B. Casulich, J.H. Mowat, J.E. Lancaster, J. Am, Chem. Soc. 84, 3183 (1962). 6. W.G. Overend, Chem. and Ind., 342 (1963). 7. a) W. Keller-Schierlein and G. Roncari, Helv. Chem. Acta £7_, 78 (1964) . b) B.P. Vaterlaus, K. Docbel, J. Kiss, A.I. Rachlin and H. Spj-egelberg, Experientia 1£, 383 (1963). c) D.J. Cooper and M.D. Yudis, Chem. Commun., 821 (1967). 8. J.S. Burton, W.G. Overend and N.R. Williams, J. Chem. Soc, 3433 (1965). 9. A.A.J. Feast, W.G. Overend and N.R. Williams, J. Chem. Soc, 303 (1966). 10. J.R. Dyer, W.E. Mc Gonigal and K.C. Rice, J. Am. Chem. Soc. 87_, 654 (1965) . 11. B. Flaherty, W.G. Overend and N.R. Williams, J. Chem. Soc, 398 (1966). 12. W.G. Overend and N.R. Williams, J. Chem. Soc, 3446 (1965). 13. J.J.K. Novak and F. Sorm, Collection Czech. Chem. Commun. 3_0, 3303 (1965). 14. B. Flaherty, W.G. Overend and N.R. Williams, Chem. Commun., 434 (1966) 15. R.D. King, W.G. Overend, J. Wells and N.R. Williams, Chem. Commun., 726 (1967). 16. R.F. Butterworth, W.G. Overend and N.R. Williams, Tetrahedron Letters 28, 3239 (1968). 17. G. Stork, R. Terrell and J. Szmuszkovicz, J. Am. Chem. Soc. 76>, 2029 (1954). 18. L.I. Kudryashow, M.A. Chelenov and N.K. Kochetkov, Izv. Akad. Nauk. S.S.S.R. Ser. Khim. 1, 75 (1965); Chem. Abstr. 6_2, 14793 (1965). 19. S. Inoue and H. Ogawa, Chem. Pharm. Bull. (Tokyo), 8_, 79 (1960). - 129 -20. P.W. Austin, J.G. Buchanan and R.M. Saunders, Chem. Commun, 146 (1965). 21. P.W. Austin, J.G. Buchanan and R.M. Saunders, J. Chem. Soc. C 372 (1967). 22. . R.S. Tipson, Advances Carbohydrate Chem. 8^, 107 (1953). 23. K. Heyns and H. Paulsen, Advances Carbohydrate Chem. 17_, 169 (1962). 24. a) J.M. Sugihara and G.U. Yuen, J. Am. Chem. Soc. 79, 5780 (1957). b) G.U. Yuen and J.M. Sugihara, J. Org. Chem. 26, 1598 (1961). c) M.L. wolfrom and S. Hanessian, J. Org. Chem. 27_, 1800 (1962). 25.. R.E. Partch, Tetrahedron Letters, 3071 (1965); J. Org. Chem., 2498 (1965). 26. W.W. Epstein and F.W. Sweat, Chem. Rev. 67^_ 247 (1966). 27. K.E. Pfitzner andJ.G. Moffatt, J. Am. Chem. Soc. 85_, 3027 (1963). 28. J.S. Brimacombe, J.G.H. Bryan, A. Husain, M. Stacey and M.S. Tolly, Carbohyd. Res. 3_, 318 (1967). 29. B.R. Baker and D.H. Buss, J. Org. Chem. 30, 2304 (1965). 30. P.J. Beynon, P.M. Collins and W.G. Overend, Proc. Chem. Soc. 342 (1964). 31. a) 0. Theander, Acta Chem. Scand. 1_8, 2209 (1964). b)\ R.F. Nutt, B. Arison, F.W. Holly and E. Walton, J. Am. Chem. Soc. 87, 3273 (1965). 32. P.J. Beynon, P.M. Collins, P.T. Doganges and W.G. Overend, J. Chem. Soc. 1132 (1966). 33. K. Onodera, S. Hirano, and N. Kashimura, J. Am. Chem. Soc. 87, 4651 , (1965). 34. K. Onodera, S. Hirano and N. Kashimura, Carbohyd. Res., 6, 276 (1968). 35. J.D. Albright and L. Goldman, J. Am. Chem. Soc. 87, 4214 (1965). 36. W. Sowa and G.H.S. Thomas, Can. J. Chem. 44_, 836 (1966). 37. a) B. Lindberg and K.N, Slessor, Carbohyd. Res. 1_, 492 (1966). b) D. Horton and J.S. Jewell, Carbohyd. Res. 2_, 251 (1966). c) K. Bredereck, Tetrahedron Letters, j3, 695 (1967). d) A.J. Fatiadi, Chem. Commun. 441 (1967). e) A.F. Cook and J.G. Moffatt, J. Am. Chem. Soc. 89, 2697 (1967). 38. V.M. Parikh and J.K.N. Jones, Can. J. Chem. 43, 3452 (1965). 39. a) G. Wittig and H. Haib, Ann. 580, 57 (1953). b) G. Wittig and G. Geissler, Ann. 580, 44 (1953). - 130 -40. G. Wittig and G. Felletschin, Ann. 555, 133 (1944). 41. H. Staudinger and J. Meyer, Helv. Chim. Acta 2, 635 (1919). 42. a) G. Wittig and Schollkopf, Chem. Ber. 87, 1318 (1954). b) G. Wittig and W. Haag, Chem. Ber. 88_, 1654 (1955). 43. N.K. Kochetkov and B.A. Dmitriev, Chem. and Ind., 864 (1963). 44. N.K. Kochetkov and B.A. Dmitriev, Dokl. Akad. Nauk, S.S.S.R. 151 106 (1963); Chem. Abstract 59, 10215g (1963). 45. N.K. Kochetkov and B.A. Dmitriev, Tetrahedron 21, 803 (1965). 46. R. Kuhn and R. Brossmer, Angew. Chem. 14_, 252 (1952). 47. Y.Z. Zhdanov, L.A. Uzlova and G.N. , Dorofeenko, Carbohyd. Res. 3_, 69 (1966). 48. Y.Z. Zhdanov, G.N. Dorofeenko and L.A'. Uzlova, Dokl. Akad. Nauk. S.S.S.R. 160(2), 339 (1965); Chem. Abstracts 62, 11890 (1965). 49. Y.Z. Zhdanov, G.N. Dorofeenko and L.A. Uzlova, Zh. Obshch. Khim. 35_, (1), 181 (1965); Chem. Abstracts 62_, i4804b (1965). 50. Y.Z. Zhdanov, L.A. Uzlova and G.N. Dorofeenko, Zh. Vses. Khim. Obshchestva im. D.I. Mendeleeva 1_0, 600 (1965); Chem. Abstracts 64, 3671d (1966). 51. J. Gigg, R. Gigg and CD. Warren, J. Chem. Soc'. C, 1872 (1966). 52. J. Gigg, R. Gigg and CD. Warren, J. Chem. Soc. C, 1882 (1966). 53. A.W. Johnson, Ylid Chemistry, Vol. 7, Academic Press, New York 1966. 54. a) A.E. Arbuzov and A.A. Dunin, Ber. 60, 29 (1927). b) A.E. Arbuzov and A.I. Razumov, J. Russ. Chem. Soc. 61_, 623 (1929); Chem. Abstracts 23, 4444 (1929). 55. a) A.N. Pudovik and N.M. Lebedeva, Zhur. Obshchei Khim. 2_5, 1920 (1955); Chem. Abstracts 50, 8442 (1956). b) G.M. Kosolapoff, J. Am. Chem. Soc. 75, 1500 (1953). 56. A.N. Pudovik and N.M. Lebedeva, Doklady Akad. Nauk., S.S.S.R. 90, 799 (1953); Chem. Abstracts 50, 2429 (1956). 57. L. Horner, H. Hoffmann and H.G. Wippel, Chem. Ber. 91_, 61 (1958). 58. W.S. Wadsworth and W.D. Emmons Jr., J. Am. Chem. Soc. 83, 1733 (1961). - 131 -59. a) B.G. Kovalev, L.A. Yanovskaya and V.F. Kucherov, Izvest. Akad. Nauk., S.S.S.R., 1876 (1962); Bull. Acad. Sci. USSR., 1788 (1962). b) H. Takahashi, K.E. Fujiwara and Mohta, Bull. Chem. Soc. Jap. 35, 1498 (1962). c) A.K. Bose and R.T. Bahill, J. Org. Chem. 30, 505 (1965). 60. H. Normant and G.Sturtz, Compt. Rendu. C 256, 1800 (1963). 61. G. Drefahl, K.Ponsold and H. Schick, Chem. Ber. 97, 2011 (1964). 62. G. Jones and R.F. Maisey, Chem. Commun. 543 (1968). 63. H. Pommer, Angew. Chem. 72_, 811, 911 (1960). 64. H.T. Openshaw and N. Whittaker, J. Chem. Soc. 1461 (1963). 65. C. SZantay, L. Toke and P. Kolonits, J. Org. Chem. 31_, 1447 (1966). 66. L. Horner, W. Klink and H. Hoffmann, Chem. Ber. 96, 3133 (1963). 67. D.H. Wadsworth, O.E. Schupp, E.J. Seus and J.A. Ford, Jr . , J. Org. Chem. 30, 680 (1965) . 68. R.J. Sundberg, P.A. Bukowick and 0. Holcombe, J. Org. Chem. 32, 2938 (1967). 69. L.D. Bergelson and M.M. Shemyakin, Pure and Applied Chem.. £, 271 (1964). 70. I. Wender, H.W. Sternberg and M. Orchin, in Catalysis, Vol. 5,73 (1957), ed. by P.H. Emmett, New York, Reinhold Publishing Corporation. 71. 0. Roelen, Ger. Pat. 103,362 (filed 1938); U.S. Pat. 2,327,066 (1943); Chem. Abstracts 38_, 550 (1944). 72. H. Adkins and G. Krsek, J. Am. Chem. Soc. 71_, 3051 (1949). 73. A. Rosenthal, Advances Carbohydrate Chem. 23_, 59 (1968) in press. 74. M. Orchin, L. Kirch and I-Golafarb, J. Am. Chem. Soc. 78, 5450 (1956). 75. a) A. Rosenthal, D. Read and C. Cameron, Science 123, 1177 (1956). b) A. Rosenthal and D. Read, Can. J. Chem. 3_5, 788 (1957). 76. A. Rosenthal and D. Read, Methods in Carbohydrate Chem., Vol. I I , ed. by R.L. Whistler and M.L. Wolfrom, New York, Academic Press, 1963, p. 457. 77. R.J. Ferrier, Advances Carbohydrate Chem. 20_, 67 (1965). 78. K. Freudenberg and F. Brauns, Ber. 55_, 3233 (1922). 79. K. Freudenberg and 0. Ivers.-Ber. 55, 929 (1922). - 132 -80. F. Weygand and H. Wolz, Ber. 85_, 256 (1952). 81. H. Zinner, G. Wulf and R. Heitnatz, Chem. Ber. 97, 3536 (1964). 82. J. Prokop and D.H. Murray, J. Pharm. Science 54_, 359 (1965). 83. R.J. Ryan, H. Arzoumanian, E.M. Acton and L. Goodman, J. Am. Chem. Soc. 86, 2497 (1964). 84. W.R. Bam'ford and T.S. Stevens, J. Chem. Soc. 4735 (1952) . 85. a) J. Kovar, U. Dienstbierova and J. Jary, Coll. Czech. Chem. Commun. 32, 2498 (1967). b) S. Hanessian and N.R. Plessas, Chem. Commun. 706 (1968) . 86. H. Paulsen and H. Behre, Carbohyd. Res. 2_, 80 (1966). 87. a) J. Lehmann, Angew. Chem. 77, 863 (1965). b) J. Lehmann, Carbohyd. Res. 2, 1 (1966). 88. H.C. Brown, Hydroboration, Benjamin, New York, 1968. 89. J. Lehmann, Carbohyd. Res. 2_, 486 (1966). 90. D.S. Breslow and R.F. Heck, Chem. and Ind. 467 (1960). R.F. Heck and D.S. Breslow, J. Am. Chem. Soc. 83, 4023 (1961). 91. L, Marko, Proc. Chem. Soc. 67 (1962). 92. A. Rosenthal and D. Abson, Can. J. Chem. 42, 1811 (1964). 93. a) A. Rosenthal, S. Rawalay and H.J. Koch, Unpublished results. b) H.J. Koch, Ph.D. Thesis, University of British Columbia, Vancouver Canada (1967). 94. P.F. Beal, M.A. Rebenstorf and J.E. Pike, J. Am. Chem. Soc. 81_, 1231 (1959). 95. A.L. Nussbaum, T.L. Popper, E.D. Oliveto, S. Friedman and L. Wender, J. Am. Chem. Soc. 81_, 1228 (1959). 96. I. Wender, J. Feldman, S. Metlin, B.H. Gwynn and M. Orchin, J. Am. Chem. Soc. 77_, 5760 (1955) . 97. I. Wender, R. Levin and M. Orchin, J. Am. Chem. Soc. 72, 4375 (1950). 98. E.I. du Pont de Nemours and Co., Brit. Patent 614,010; Chem. Abstracts 4J5, 4685 (1949). 99. A. Rosenthal and D. Abson, Carbohyd. Res. 3, 112 (1966). - 133 -100. I. Wender, S. Metlin, S. Ergun, H.W. Sternberg and H. Greenfield, J. Am. Chem. Soc. 7J3, 5401 (1956). 101. J.G. Trayham, J. Am. Chem. Soc. 78, 402 (1956). 102. E. Walton, S.R. Jenkins, R.F. Nutt, M. Zimmerman and F.W. Holly, J. Am. Chem. Soc. 88, 4524 (1966). 103. S.R. Jenkins, B. Arison and E. Walton, J. Org. Chem. 33_, 2490 (1968). 104. R.F. Nutt, M.J. Dickinson, F.W. Holly and E. Walton, J. Org. Chem. 33, 1789 (1968). 105. R.F. Nutt and E. Walton, J. Med. Chem. 11_, 151 (1968). 106. K. Onodera, S. Hirano and H. Fukumi, Agr. Biol. Chem. (Tokyo), 28., 173 (1964). 107. CP. Whittle and R.K. Robins, J. Am. Chem. Soc. 87_, 4940 (1965). 108. M.L. Wolfrom and M.W. Winkey, J. Org. Chem. 31_, 3711 (1966). 109. K. Onodera, S. Hirano and F. Masuda, Carbohyd.Res. 4_, 263 (1967). 110. E. Fischer and B. Helferich, Ber. 47_, 210 (1914). 111. J. Davoll and B.A. Lowy, J. Am. Chem. Soc. 73, 1650 (1951). 112. B.R. Baker, R.E. Schaub and H.M. Kissman, J. Am. Chem. Soc. 77_, 5911 (1955). 113. B.R. Baker, Ciba Foundation Symposium, Chemistry and Biology of Purines, L i t t l e , Brown and Co., Boston, Mass. 1957, p. 120. 114. J.J. Fox and I. Wempen, Advances Carbohydrate Chem. 1_4, 283 (1959). . 115. J.A. Montgomery and H.J. Thomas, Advances Carbohydrate Chem. 17_ 301 (1962). 116. J.J. Fox, K.A. Watanabe and A. Bioch, Progress in Nucleic Acid Research and Molecular Biology 5_, 251 (1966) . 117. E.J. Reist, R.R. Spencer and B.R. Baker, J. Org. Chem. 23_, 1958 (1958). 118. M.L. Wolfrom, P. Mc Wain, R. Pagnucco and A. Thompson, J. Org. Chem. 2£, 454 (1964). 119. a) L.M. Lerner and P. Kohn, J. Org. Chem. 31, 339 (1966). b) P. Kohn and L.M. Lerner, J. Org. Chem. 32, 4076 (1967). 120. J.B. Lee and T.J. Nolan, Tetrahedron 23, 2789 (1967). - 134 -121. L.M. Lerner, B.D. Kohn and P. Kohn, J. Org. Chem. 33, 1780 (1968). 122. G.L. Tong, W.W. Lee and L. Goodman, J. Org. Chem. .32, 1984 (1967). 123. M.L. Wolfrom and P. Mc Wain, J. Org. Chem. 30, 1099 (1965). 124. W.N. Haworth, in The Constitution of Sugars, Longmans Green and Co. New York, N.Y. 52 (1929). 125. P. Kohn, R.H. Samaritano and L.M. Lerner, J. Am. Chem. Soc. 86, 1457 (1964) . 126. a) K.J. Ryan, H. Arzoumanian and E.M. Acton, J. Am. Chem. Soc. 86, 74, 2503 (1964). b) E.J. Reist, R.R. Spencer and B.R. Baker, J. Org. Chem. 23, 1757 (1958). 127. a) E.J. Reist, L. Goodman, R.R. Spencer and B.R. Baker, J. Am. Chem. Soc. 80, 362 (1958). b) B.R. Baker and K. Hewson, J. Org. Chem. 22, 966 (1957). c) E.J. Reist, R.R. Spencer and B.R. Baker, J. Org. Chem. 23, 1753 (1958). d) E.J. Reist, L. Goodman, R.R. Spencer and B.R. Baker, J. Am. Chem. Soc. 80, 3962 (1958). e) E.J. Reist, L. Goodman and B.R. Baker, J. Am. Chem. Soc. 80, 5575 (1958). 128. W.L. Glen, G.S. Myers and G.A. Grant, J. Chem. Soc. 2568 (1951). 129. a) E. Fischer and C. Rund, Ber. 49., 88 (1916). b) H. Ohle and L. van Vargha, Ber. 62., 2425 (1929). 130. J.D. Stevens and H.G. Fletcher, J. Org. Chem. 33, 180 (1968). 131. A. Rosenthal, personal communication, 1967. 132. D.M. Brown and G.H. Jones, J. Chem. Soc. C, 252 (1967). 133. F. Smith, Methods Carbohydrate Chemistry, Vol. I I , ed by R.L. Whistler and M.L. Wolfrom, New York, Academic Press, 1963, p. 72. 134. D.C. De Jongh and K. Biemann, J. Am. Chem. Soc. 86, 67 (1964). 135. F. Shafizadeh, Methods Carbohydrate Chemistry, Vol. I I , ed. by R.L. Whistler and M.L. Wolfrom, New York, Academic Press, 1963, p. 192. 136. O.T. Schmidt, Methods in Carbohydrate Chemistry, Vol. I I , ed. by R.L. Whistler and M.L. Wolfrom, New York, Academic Press, 1963, p. 322. 137. R.U. Lemieux and D.R. Lineback, Ann. Rev. Biochem. ,32, 155 (1963). 138. F. Weygand and H. Wolz, Chem. Ber. 85_, 256 (1952). 139. a) E.J. Reist, Chem. and Ind. 1957 (1967). b) E.J. Reist, D.F. Calkins, and L. Goodmann, J. Am. Chem. Soc. 90, 3852 (1968). 140. J.M. Gulland and E.R. Holiday, J. Chem. Soc. 765 (1936). - 135 -141. R.S. Tipson, J. Biol. Chem. 130, 55 (1939). 142. T.R. Emerson, R.J. Swan and T.L.V. Ulbricht, Biochem. Biophys. Comm. 22, 505 (1966). 143. T. Nishimura, B. Shimizu and I. Iwai, Biochim. Biophys. Acta 147, 221 (1968). 144. A.D. Broom, M.P. Schweizer and P.O.P. Tso 1, J. Am. Chem. Soc. 89, 3612 (1967). 145. a) M. Karplus, J. Am. Chem. Soc. 85_, 2870 (1963). b) R.U. Lemieux, J.D. Stevens and R.R. Eraser. Can. J. Chem. 40_, 1955 (1962). 146. CD. Jardetsky, J. Am. Chem. Soc. 84, 62 (1962). 147. R. Schaffer and H.S. Isbell, J. Am. Chem. Soc. 80, 756 (1958). 148. M.L. wolfrom and S. Hanessian, J. Org. Chem. 27_, 1800 (1961). 149. T.D. Inch, Carbohyd. Res. 5_, 45 (1967). 150. P. Szabo' and L. Szabo', J. Chem. Soc. 2944 (1956). 151. CH. Snyder, Chem. Ind. (London), 121 (1963). 152. M.S. Morgan and L.H. Cretcher, J. Am. Chem. Soc. 70, 375 (1948). 153. CH. Snyder and A.R. Soto, J. Org. Chem. 2£, 742 (1964). 154. F.C Chang, Tetrahedron Letters, 306 (1964). 155. F.X. Jarreau, B. Tchoubar and R. Goutarel, Bull. Soc. Chim., 887 (1962). 156. D.N. Jones and M.A. Saeed, J. Chem. Soc. 4657 (1963). 157. D. Dolman and R. Stewart, Can. J. Chem. 45, 911 (1967). 158. N.K. Kochetkov and O.S. Chizhov, Advances Carbohydrate Chem. 21_, 39 (1966). 159. P.A. Levene and J. Compton, J. Biol. Chem. 116, 189 (1936). 160. A. Streitwieser and W.D. Schaeffer, J. Am. Chem. Soc. £9, 6233 (1957). 161. P.D. Bartlett, J. Am. Chem. Soc. 87_, 1308 (1965). 162. J.P. Horwitz, J. Chua, M.A. Da Rooge, M. Noel and I.L. Klundt, J. Org. Chem. 31_, 209 (1966). - 136 -163. A. Rosenthal and D. Abson, Can. J. Chem. 43_, 1318 (1965); ibid. 43, 1985 (1964). 164. A. Rosenthal and H.J. Koch, Can. J. Chem. 42, 2025 (1964); ibid. 43, 1375 (1965). 165. A. Rosenthal and C. Kan, unpublished results (1967). 166. D. Abson, Ph.D. Thesis, University of British Columbia, Vancouver, Canada (1964). 167. A.R. Martin, Chem. and Ind., 1536 (1954). 168. I. Wender, Ph.D. Thesis, University of Pittsburgh (1950). 

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.831.1-0059860/manifest

Comment

Related Items