UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Branched-chain nucleosides : synthesis of structural analogs of the polyoxin complex and of puromycin Richards, Colin Maxwell 1973

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

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata

Download

Media
831-UBC_1973_A1 R51_4.pdf [ 8.98MB ]
Metadata
JSON: 831-1.0060150.json
JSON-LD: 831-1.0060150-ld.json
RDF/XML (Pretty): 831-1.0060150-rdf.xml
RDF/JSON: 831-1.0060150-rdf.json
Turtle: 831-1.0060150-turtle.txt
N-Triples: 831-1.0060150-rdf-ntriples.txt
Original Record: 831-1.0060150-source.json
Full Text
831-1.0060150-fulltext.txt
Citation
831-1.0060150.ris

Full Text

BRANCHED-CHAIN NUCLEOSIDES: SYNTHESIS OF. STRUCTURAL ANALOGS OF THE POLYOXIN COMPLEX AND OF PUROMYCIN BY COLIN MAXWELL RICHARDS B.Sc. (Honours), University of Melbourne, 1969 M.Sc, University of Melbourne, 1971 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in the Department of CHEMISTRY We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA July, 1973 In presenting t h i s thesis i n p a r t i a l f u l f i l m e n t of the requirements for an advanced degree at the University of B r i t i s h Columbia, I agree that the Library s h a l l make i t f r e e l y available for reference and study. I further agree that permission fo r extensive copying of t h i s thesis for scholarly purposes may be granted by the Head of my Department or by h i s representatives. It i s understood that copying or publication of t h i s thesis for f i n a n c i a l gain s h a l l not be allowed without my written permission. Department The University of B r i t i s h Columbia Vancouver 8, Canada Date ""o - i i -ABSTRACT The synthesis of a number of 3-C-cyanomethyl-2,3-dideoxy-hexo-pyranosides, pentofuranosides and pentopyranosides and the conversion of these compounds to their respective purine nucleosides are reported. Methyl 4,6-0-benzylidene-2-deoxy-a-g-erythro-hexopyranosid-3^ ulose (127) was condensed with the carbanion formed from diethylcyano^-methylphosphonate and sodium hydride to afford the unsaturated compounds methyl 4,6-£-benzylidene-E_-(and Z)-3-C-cyanomethylene-2, 3-dideoxy-ct-D-erythro-hexopyranoside (128 and 129). Acid hydrolysis of 128 gave the unblocked a,3-unsaturated n i t r i l e 141. Hydrogenation of 128 afforded methyl 4,6-0-benzylidene-3-C-cyanomethyl-2, 3-dideoxy-a-rj-ribo-hexo-pyranoside (140) which was subsequently debenzylidenated, p_-rchlorobenzoylated, and fused directly with 2,6-dichloropurine to afford 4,6-di-0-p_-chlorobenzoyl-3-C-cyanomethyl-2,3-dideoxy-g-ribo-hex-l-eno-pyranose (149) and 2,6-dichloro-9-(41,6'-di-0-p_-chlorobenzoyl-3'-C-cyanomethyl-2',3'-dideoxy-a- and g-Tj-ribo-hexopyranosyl)purine (150 and 151). Debenzyli-denation of 128 also yielded methyl 1,6-anhydro-3-£-cyanomethyl-2,3-dideoxy-q-D-ribo-hexopyranoside which was characterized as i t s j>-chloro-benzoyl derivative 148. Treatment of the blocked a-nucleoside 150 with aqueous dimethylamine in methanol resulted in replacement of the 6-chloro group by N,N-dimethylamine and hydrolysis of the cyanomethyl group to yield 2-chloro-6-N,N-dimethylamino-9-(3'-£-N,N-dimethylamino-carbamoylmethyl-2' , 3'-dideoxy-a-rj-ribo-hexopyranosyl)purine (152) . However identical treatment of the blocked 6-nucleoside gave 2-chloro-6-N,N^-dimethylamino-9-(3'-C-cyanomethyl-2',3'-dideoxy-B-g-ribo-hexopyranosyl)-purine (153). Reaction of 150 with anhydrous dimethylamine at -5° afforded, - i i i -2,6-di-N,N-dimethylamino-9- (3'-C-cyanomethyl-2' , 3' -dideoxy-ct-g-ribo-hexopyranosyl)purine (158). Hydrogenation of 158 followed by acetylation yielded 2,6-di-N,N-dimethylamino-9-(3'-C-[2"-acetamidoethyl]-21,31-dideoxy-q-D-ribo-hexopyranosyl)purine (170). 2-Deoxy-g-erythro-pentose (171) was treated with 0.05% methanolic hydrogen chloride followed by reaction with chlorotriphenylmethane in pyridine to afford methyl 2-deoxy-5-0-trityl-a-(and g)-D-erythro-pentofuranosides (175) and (176) respectively. Ruthenium tetroxide oxidation, followed by condensation with the carbanion formed from diethylcyanomethylphosphonate and sodium hydride and subsequent hydrogenation gave methyl 3-jC-cyanomethyl-2,3-dideoxy-5-0-trityl-a-g-erythro-(and threo)-pentofuranoside (186 and 187), and methyl 3-C-cyanomethyl-2,3-dideoxy-5-0-trityl-g-g-threo-pentofuranoside (188). Attempted selective removal of the 5-0-trityl group was unsuccessful. Compounds 187 and 188 were converted to the same pentopyranoside to aid in structural proof. Methyl 4-0-p_-bromobenzoyl-3-C-cyanomethyl-2,3-dideoxy-q-D-threo-pentopyranoside (191), derived from 187 and 188, was utilized in nucleoside synthesis by direct fusion to yield 2,6-dichloro-9-(4' -0_-p_-bromobenzoyl-3' -C-cyanomethyl-21 ,3' -dideoxy-3-(and cQ-g-threo-pentopyranosyl)purine (195 and 196), and 4-0-p_-bromobenzoyl-3-C^-cyanomethyl-2,3-dideoxy-D-threo-pent-l-eno-pyranose (197). Treatment of 195 and 196 with aqueous dimethylamine in methanol gave 2-chloro-6-N,N-dimethylamino-9- (3'-C-cyanomethyl-2' , 31-dideoxy-|3-(and q)-g-threo-pentopyranosyl)-purine (198 and 199). Reduction of 198 over platinum oxide in acetic anhydride and ethanol gave 2-chloro-6-N^-dimethylamino-9-(3'-C-[2"-acetamidoethyl]-2',3'-dideoxy-g-D-threo-pentopyranosyl)purine (200). - iv -The ketose, 1,2:5,6-di-O-isopropylidene-a-rj-ribo-hexof uranos-3-ulose (9) was reacted with phosphonoacetic acid trimethyl ester and potassium _t-butoxide to afford E-(and Z,)-3-C-(methoxycarbonyl)methylene-3-deoxy-l,2:5,6-di-O-isopropylidene-K-rj-ribo-hexofuranose (11 and 10) and E-(and ^)-3-C-(methoxycarbonyl)methylene-3-deoxy-l,2 :5 ,6-di-0-isopropylidene-g-rj-xylo-hexof uranose (205 and 204). Stereospecific hydroxylation of 10_ and 11 afforded 3-C-[S-and R-hydroxy(methoxycarbonyl)methyl]-1,2:5,6-di-O-isopropylidene-a-B-gluco-furanose, (211) and (212), respectively in a combined yield of 43% after chromatography. Selective formation of the monomesylates, 215 and 216, followed by treatment with sodium azide and reduction afforded the methyl D-2-(and L-2)-l,2:5,6-di-O-isopropylidene-a-D-glucofuranos-3-yl)glycinate 218 and 217. Base hydrolysis of the latter compounds yielded g-2- and L-2-(1,2:5,6-di-0-isopropylidene-a-D-glucofuranos-3-yl)-glycine, 221 and 220, respectively. The structures of the glycosyl amino acids were correlated with that of L-alanine by circular dichroism. In addition the glycosyl a-amino esters were derivatized as acetamido, 222 and 223, and as benzamido derivatives 224 and 225. Similarly stereospecific hydroxylation of pure E-3-C-(methoxy-carbonyl)methylene-3-deoxy-l,2:5,6-di-Q-isopropylidene-q-rj-xylo-hexofuranose (205) afforded crystalline 3-C-[R-hydroxy(methoxycarbonyl)-methyl]-l,2:5,6-di-0_-isopropylidene-a-D-galactofuranose (226) in 55% yield. Treatment of 226 with methanesulfonyl chloride in pyridine afforded 3-C-[R-methanesulfonyloxy(methoxycarbonyl)methyl]-1,2:5,6-di-0-isopropylidene-a-g-galactofuranose (229) and the unusual 3-C-[methane-sulf onyloxy(methoxycarbonyl)methylene]-3-deoxy-l,2:5,6-di-0_-isopropylidene-a-- v -g-xylo-hexofuranose (230), in 60 and 30% yields, respectively. Treatment of 229 with sodium azide, followed by reduction, afforded the a-amino esters, methyl-L-2-(1,2:5,6-di-O-isopropylidene-ct-D-galactofuranos-3-yl)-glycinate (232) and methyl D-2-(1,2:5,6-di-0-isopropylidene-a-D-galactofuranos-3-yl)glycinate (234). Base hydrolysis of 232 and 234 yielded L-2-(and D-2)-(l,2:5,6-di-O-isopropylidene-ct-g-galactof uranos-3-yl)glycine 233 and 235, respectively. The dihydroxy compound 212 was selectively monoacetylated then stereospecif i c a l l y dehydrated to give 3-C- [Z.-1'-0-acetyl-l' -(methoxy-carbonyl) methylene ] -3-deoxy-l, 2:5,6-di-O-isopropylidene-q-g-ribo-hexofuranose (237). Hydrogenation of 237 yielded the saturated acetoxy compound 238, which was catalytically deacetylated in 100% yield to afford 3-C- [S~hydroxy (methoxycarbonyl) methyl] -3-deoxy-l, 2:5,6-di-0_-isopropylidene-q-g-allofuranose (239). Treatment of 239 with methane-sulfonyl chloride and p_-toluenesulfonyl chloride gave the methanesulfonate and p_-toluenesulfonate, 240 and 241, respectively. Treatment of either 240 or 241 with sodium azide in anhydrous dimethylformamide, followed by hydrogenation resulted in methyl D-2-(and L-2)-(3-deoxy-l,2:5,6-di-0-isopropylidene-q-D-allofuranos-3-yl)glycinate, 243 and 245, respectively. Base hydrolysis of 243 gave D-2-(3-deoxy-l,2:5,6-di-0-isopropylidene-q-D-allofuranos-3-yl)glycine. In addition to correlating the above glycosyl amino acids with L-alanine by means of circular dichroism, the glycosyl methyl glycinate 243 was converted to a compound of known absolute configuration as determined by X-ray crystallography. Treatment of a l l six glycosyl amino acids with L-amino acid oxidase from Crotalus Adamanteus failed to afford additional proof of amino acid configuration. - v i -TABLE OF CONTENTS Page I. OBJECTIVE 1 II. INTRODUCTION 4 1. Branched-Chain Sugars 4 1.2 Synthesis of Branched-Chain Sugars 5 1.2.1 Synthesis of Sugars of Type A 5 1.2.2 Synthesis of Sugars of Type B ... 7 2. Methods of Oxidation . . 1 0 2.1 Oxidation of Secondary Hydroxyl Groups 10 2.2 Oxidation of Olefins to form Glycols 1 2 3. The Wit tig Reaction 1 5 3.1 The Wittig Reaction in Carbohydrate Chemistry... 1 6 3.2 The Phosphonate Modification of the Wittig 1 Q Reactxon , • 3.3 Mechanism and Stereochemistry 21 4. Nucleosides 24 4.1 Branched-Chain Nucleosides 24 4.2 Nucleoside Synthesis 27 5. Peptidyl Nucleosides 30 5.1 The Polyoxins 3 1 6. Synthesis of Glycosyl Amino Acids 37 III. RESULTS AND DISCUSSION 4° 1. Branched-Chain Nucleosides ^1 1.1 Synthesis of Branched-Chain 2-deoxy-Hexo-pyranosyl Nucleosides 1.2 Synthesis of Branched-Chain 2-deoxy-Pento-pyranosyl Nucleosides - v i i -Page 2. Glycos-3-yl ct-Hydroxy Esters, ct-Amino Esters and a-Amino Acids: Structural Analogs of the Sugar Moiety of the Polyoxins 100 2.2 Synthesis of D-2- and L-2-(1,2:5,6-di-0-isopropylidene-a-p-glucofuranos-3-yl)glycine... 108 2.3 Synthesis of D-2- and L-2-(1,2:5,6-di-0-isopropylidene-a-D-galactofuranos-3-yl)glycine. 121 2.4 Synthesis of D-2- and L-2-(3-deoxy-l,2:5,6-di-O-isopropylidene-a-D-allof uranos-3-yl)-glycine (246) and 247) 127 2.5 Action of L-Amino Acid Oxidase on Glycosyl Amino Acids 134 IV. EXPERIMENTAL 136 BLBIOGRAPHY 190 - v i i i -LIST OF FIGURES Figure Page I Fractionation Process of the Polyoxins 34 II Nmr Spectrum of Compound 128 v 43 III Nmr Spectrum of Compound 129 44 IV Nmr Spectrum of Compound 149 58 V Nmr Spectrum of Compound 150 59 VI Nmr Spectrum of Compound 151 60 VII Nmr Spectrum of Compound 153 64 VIII Nmr Spectrum of Compound 170 73 IX Nmr Spectra of Compounds 187 and 188 84 X CD Spectra of Compounds 211 and 212 111 XI CD Spectra of Compounds 217, 218, 220, and 221 ... 120 XII CD Spectra of Compounds 226, 232, 233' 234, and 235 126 XIII CD Spectra of Compounds 239, 243, 245, 246, 247, and 248 132 - ix -LIST OF TABLES Table Page I Branched-Chain Nucleosides 25 II Polyoxins A-L 32 III Antifungal Activities of the Polyoxins 33 IV Chemical Shift Data for Compounds 128 and 129 45 V Chemical Shift Data for Compounds 130-139 46 VI Observed Coupling Constants for Compounds 128 and 129 47 - X -ACKNOWLEDGEMENTS I wish to express my sincere thanks to Dr. Alex Rosenthal for his guidance and invaluable suggestions throughout the course of this research. My thanks are extended to Dr. L.D. Hall for reading parts of this thesis. In addition, I would like to thank both Dr. G.G.S. Dutton and Dr. L.D, Hall for their helpful suggestions during my period at U.B.C, and to Dr. Richard Barton for his assistance with the enzymatic studies. Also I wish to thank the members of Dr. Rosenthal's and Dr. Dutton's research group for their many worthwhile discussions. My thanks also go to my wife for her encouragement throughout this study and for her help in the preparation of this manuscript. The typing of this thesis by Miss Diane Johnson is greatly appreciated. The financial support of the University of British Columbia (1970-1973) and the Standard O i l Company of British Columbia (1971-1972) is acknowledged. - 1 -I. OBJECTIVE Although the f i r s t nucleoside antibiotic, cordycepin, was isolated in 1951 the true structure was not completely resolved u n t i l 1964, when i t was shown to be 3'^deoxyadenosine. Since that time numerous other deoxy- and branched-chain nucleosides have been synthesized, and isolated from naturally occurring antibiotics. Spme of those synthesized, for example, 5',5'-di-(3, 3'-C- and 2'-C-methyladenosine, have shown biological activity, in these cases as measured by their a b i l i t y to inhibit the growth of KB cells in culture. The primary objective of this work was to attempt to apply known methods to the synthesis of a number of 2,3-dideoxy branched-chain hexoses and pentoses, via the addition of modified Wittig reagents to 2-deoxy ketoses, and subsequently ut i l i z a t i o n of these branched-chain carbohydrates in nucleoside synthesis. The condensation of diethyl cyanomethylphosphonate with such ketoses lends i t s e l f to the incorpora-tion of a number of branched-chains such as, cyanomethyl, aminoethyl, carbamoyl, and K,N-dimethylaminocarbamoyl methyl functionalities. It has often been found that a change of functionality or branching at the C-3 position i s observed in many naturally occurring carbohydrates and hence this study was to introduce substituents at this position. The antibiotic puromycin (1) is one such nucleoside. The biological activity - 2 -of puromycin and i t s analogs has been found to be dependent on the C-3' substituent. NMe Nil 2 1 In the second part of this work a similar end was pursued. A „ range of commercially important glycosyl amino acids have recently been the centre of much synthetic endeavor. These compounds known as Polyoxins A-L, a l l show (except C and I) considerable toxicity to sheath-blight in rice plants. A l l twelve polyoxins contain three common features: (a) an a-L-amino acid residue; (b) a 5-amino-5-deoxy-furanuronoside and (c) a pyrimidine chromophore; therefore i t seemed to us, in the light of previous findings, that incorporation of the amino acid residue at the C-3 position, rather than at the C-4 position, may result in interesting biological consequences. Thus a method was derived which resulted in the synthesis of three pairs of D and L glycos-3-yl amino acids. - 3 -In order to provide some background to subsequent discussions a brief summary w i l l be made of synthesis of branched-chain sugars, glycosyl amino acids and nucleosides. - 4 -II. INTRODUCTION Widespread interest in branched-chain carbohydrates in recent years has been initiated by numerous reports of such compounds 1 2 occurring in many natural sources. Recently branched-chain sugars ' 3—6 have been isolated in large numbers from micro-organisms and higher g plants, as glycosidic components of antibiotics and phenolic compounds, as well as in c e l l wall polysaccharides.^ ^  In addition, i t has been 12 suggested that branched-chain sugars may also occur in man. 13 Branched-chain sugars have been divided into two classes. Those in which the branching occurs by substitution by a group R of a hydrogen atom are classified as Type A whilst substitution by R of an hydroxyl group affords Type B. The latter group being members of the "deoxy" classification. These two classes are exemplified by L-mycarose^ ^ (2a), and L-cladinose"^ (2b) as Type A, and L-17 18 19 2.0 garosamine ' (3) and L-evernitrose ' ' (4) representing Type B. - 5 -1.2 Synthesis of Branched-Chain Sugars A very brief summary of reactions applied to carbohydrates which have introduced branching into the sugar skeleton shall be discussed below. For clarity these methods shall be grouped in a manner such that they afford sugars of the Type A or Type B. Many of these reactions involve the use of keto sugars as starting compounds and therefore the synthesis of these compounds w i l l be briefly discussed in section 2.1. I 1.2.1 Synthesis of Sugars of Type A , RCOH Several methods have lent themselves successfully to the formation of branched-chain sugars in which a hydroxyl group is retained at the branch-point. Reaction of diazomethane with keto sugars has frequently been used 21 22 for the synthesis of branched-chain sugars. ' This reagent attacks the carbonyl to form an epoxide (5a) which may then be opened with, LiAlH^ to give a C-methyl substituent (5b); alkal i to give a C-hydroxy-21 methyl substituent (5c) ; or ammonia to give a C-aminomethyl substituent (5d). HO CH (5b) 3 / \ C=0 NH HO CH2OH (5c) (5a) HO CH2NH2 (5d) Related to the above method is the use of the dimethylsulfoxonium 22 24 methylid to effect epoxidation of a keto sugar. ' Thus the ketose (6) afforded the L-arabino product (7) predominantly (R = CH^) after reduction. ° H°'OCH 3 Organolithium and organomagnesium compounds have been used extensively in the formation of Type A branching and have been found to afford products which differ in configuration at the branch point. 25 Thus (_6) with organolithium affords the L-ribo sugar (8) whereas 13 26 Grignard reagents give L-ar^bino derivatives (7) predominantly. ' 27 Acetonitrile in liquid ammonia has been used to introduce a cyanomethyl functionality which may subsequently be catalytically reduced to an aminoethyl branched-chain. Related to this i s the base 28-31 catalyzed condensation of nitromethane to keto sugars which may be used to synthesize sygars of types A or B. By addition of nitromethane, the keto sugar affords nitromethyl derivatives of type A, which may then be reduced to amines or transformed by elimination to cx-nitro olefins, intermediates in the synthesis of sugars of type B, afforded by catalytic reduction. _ 7 -1.2.2 Synthesis of Sugars of Type B, RCH Sugars of this constitution were not readily available synthetically until recent years. The application of the Wittig reaction to keto 32 sugars by Rosenthal and Nguyen initiated considerable activity in the use of Wittig reagents in the synthesis of branched deoxy sugars and • i •j 33-40 their nucleosides. X o _ 0 — 10 CH 9C0 2CH ? ! 1 12 41 Tronchet and coworkers have utilized the Wittig product (13) to afford sugars of type A related to "streptose" and "apiose" by oxidative cleavage of the olefin to afford (14) and (1_5) • - 8 -13 3 4 Nucleophilic addition to anhydro sugars has been of considerable 42 advantage in the synthesis of branched-chain sugars. Thus reaction of (16) with alkyllithium reagents afforded only 3-C-substituted 43 products having the xylo-configuration (17) as expected by trans-diaxial cleavage of the epoxide ring. Reaction of 2,3-anhydro sugars with alkyl Grignard reagents, and with cyanide have also been investigated and have been shown to, 44 at times, afford unexpected products, such as halogenated sugars in the f i r s t case and isomerization of the tertiary carbon atom under the 45 46 strongly basic conditions of the latter. ' Stereoselectivity has 47 been achieved by the use of a triethylaluminum-HCN complex in ether. The anhydro sugar (18) may also be attacked by the carbanion formed from diethyl malonate^'^ to afford a compound of the altro-configuration (19) which may be reduced to the hydroxyethyl branched-48 chain sugar (20). Alkylation of the methyl glycopyranosidulose (21) using either 49 * 50 methyl iodide or Stork's enamine procedure afforded predominantly the branched-chain sugar (22). M M f:i" , Another method which readily affords branched-chain sugars of type B i s , the Oxo reaction, in which CO and react under high pressure, adding CE^O to an unsaturated species (23), which then can be reduced to afford,a hydroxymethyl derivative (25) , via the aldehydo sugar (24). 23: 24 - 10 -Very recently photoamidation of unsaturated sugars (26) and (30) has shown to be a feasible route to the carbamoyl branched-chain, and chain extended sugars 5 1 (27), (28), (29), (31), and (32). Other approaches to formation of branched-chain sugars have 52 53 utilized dimerization reactions, nitroethane addition to dialdehydes. 54 55 reduction of lactones, and ring contraction. 2. Methods of Oxidation 2.1 Oxidation of Secondary Hydroxyl Groups The profusion of branched-chain sugars has, interestingly, almost coincided with the development of more efficacious methods of keto-sugar synthesis. In addition to the classical methods of oxidation 56 of secondary alcohols, using platinum oxide and oxygen, chromium 57 58 trioxide-pyridine, and lead tetraacetate-pyridine, two relatively - 11 -new reagents have, to a large extent in recent years, dominated the oxidative methods of carbohydrates. These two being dimethyl sulfoxide 59 (DMSO) and ruthenium tetroxide (RuO^). Butterworth and Hanessian have recently thoroughly reviewed the methods of oxidation in carbohydrate chemistry and hence the methods which shall be discussed here are only those which have been used in this work (DMSO and RuO^). 60 Since DMSO was f i r s t used as an oxidant many variations based on this reagent have been successfully u t i l i z e d in which an activating 61 electrophile (E) such as N,N-dicyclohexylcarbodiimide, acetic 62 63 anhydride, or phosphorus pentoxide may be used. It has been 64 indicated that most DMSO oxidations involve the formation of a dimethylalkoxysulfonium intermediate (33) with the activating reagent (E) which is then displaced by the substrate to be oxidized to form the sulfonium intermediate (34) which subsequently loses a proton to base CH, V / CH„ S = 0 + E CH, V / CH 3 — 0 33 R„ OH I CH — R, CH, I " S I CH, 0 = 35 c \ CH, V + / CH, 0 — C R 2 34 - 12 -and collapses to afford the keto sugar (35) and dimethyl sulfide (DMS). Overend and coworkers^5 f i r s t reported the application of a ruthenium tetroxide oxidation on l,2:5,6-di-0-isopropylidene-a-D-glucofuranose (36) in 1964. Since then i t has been found to be a most 66 powerful and useful oxidant. Jones and coworkers have found that, in addition to adding prepared ruthenium tetroxide to the reaction mixture, i t may be readily prepared in situ from ruthenium dioxide and sodium or potassium periodate. Oxidations are generally performed 66 67 using a catalytic amount ' of ruthenium dioxide which i s reconverted to ruthenium tetroxide by the portionwise addition of sodium periodate. 68 It has been noted that the ruthenium dioxide must be prepared by a precipitation process, resulting in the dihydrate, to f a c i l i t a t e ready oxidation to the tetroxide by aqueous periodate solutions. 2.2 Oxidation of Olefins to Form Glycols The action of osmium tetroxide and potassium permanganate on alkenes has been shown to afford glycols via a stereospecific cis addition. C r i e g e e ^ ' ^ has established the following sequence as the course of the dihydroxylation reaction of alkenes with osmium tetroxide: - 13 -\ / OsC- \ / H 0 \ / r 4 C - 0 v o 2 - C - OH , Os^ ? I + H2° S°4 C — 0 ' ^0 C — OH i i 38 39 The cyclic osmate ester (38) can be isolated in non hydroxylic solvents, and then hydrolyzed to the glycol (39) in a separate step. Permanganate oxidations are also thought to proceed via a cyclic e s t e r ^ ^ although such a species has not been isolated. It has been « 73 observed by Lemieux that at low hydroxyl ion concentrations (e.g. in the presence of magnesium sulfate) ct-hydroxy ketones may be produced, whereas in higher pH solution with low permanganate concentration, 74 75 diols predominate. Since i t is generally accepted ' that these a-hydroxy ketones are not produced by further degradation of the d i o l , Wiberg^ has proposed a common intermediate (40) in the production of diol and hydroxy ketone. The reaction scheme proposed by Wiberg i s shown below: H , \ / I I C MnO. H-C — 0. ^.0 HO H-C-OH II 4 ». I ^Mn 2 I C H-C—0-^ \ - H-C-0-MnO. / X | 0 | 2 H (40) \ OH I I I -C-OH H-C-OH H-C-OH I + ^ \ I / > - I C=0 + BH B: 1 HvCrO-MnO. H-C-OH 1 + Mn03 1 - 14 -75 In addition Wiberg has investigated the total cleavage of olefins using permanganate and favours the periodate-like intermediate (41). i . I I H-C-0 0 MnO H-C-0 0 H-C=0 | Mn^ _ . | Mn^ " **• + MnO H-C-0 ^  ^0 H-C-0 ^  ^0 H-C=0 Tronchet and coworkers have utilized the permanganate oxidation on exocyclic methylenic compounds of the types shown below to afford compounds of the "streptose" and "apiose" types. - 15 -3. The Wittig Reaction The synthesis of a number of branched-chain sugars to be described later are a l l derived from condensation of appropriate phosphonates with keto-sugars according to the method generally known as the modified Wittig reaction. Thus, as this important synthetic method has been applied extensively to the preparation of branched-chain 32-41 sugars, as indicated earlier (section 1.2.2), a discussion of this reaction would appear necessary. However, this reaction has been reviewed several times (Ref. 76 and references therein) and so this discussion shall be kept brief. In 1919, H. Standinger and J. Meyer^ performed the f i r s t condensation between a carbonyl compound, phenylisocyanate and a. phosphorus ylid,benzhydrylidenetriphenylphos phorane: C H NCO (C 6H 5) 3P=C(C 6H 5) 2 ^ C 6H 5N-OC(C 6H 5) 2 + ( C ^ P O 78 Three years later G. Luscher reacted the same y l i d with diphenylketone to form tetraphenylallene: (C 6H 5) 3P-C(C 6n 5) 2 ( C 6 H 5 ) 2 C C ° ^ (c 6H 5) 2C=C=C(C 6H 5) 2 + ( C ^ P O 79 Many years later, in 1949, Wittig and Rieber synthesized the methylenetrimethylphosphorane (46) by the action of methyllithium on tetramethylphosphonium iodide (45). The y l i d (46) was reacted with benzophenone and the resultant compound (47) was isolated. - 16 -(CH 3) 3P-CH 3I + CH 3Li (CH3)3P=CH2 + L i l + CH4 (45) (46) (CH3)3P=CH2 + (C 6H 5) 2C=0 (CH 3) 3P-CH 2-C(C 6H 5) 2 I OH (47) From 1953 the number of applications of the Wittig reaction 81-87 80 multiplied greatly. In this year Wittig and Geissler reported the condensation of methylenetriphenylphosphorane with benzophenone to afford, in high yield, diphenylethylene. In addition the mechanism and stereochemistry of the Wittig reaction has been investigated . . 76,88-93 extensively. 3.1 The Wittig Reaction in Carbohydrate Chemistry Whilst the Wittig reaction has been used extensively in other fields i t s application in the carbohydrate f i e l d has been relatively 94-96 limited. Kochetkov and coworkers reported in 1963 the reaction of carbethoxymethylene triphenylphosphorane with aldoses having the 94 95 hydroxyls either free or blocked to afford a,B-unsaturated aldonic acids. C0 2Et CHO CH 1 II (CH0Ac)o . + (C,Hc)oPCHC0„Et *-* CH 3 or 4 6 5 3 2 i i n 0OAc (CHOAc). . 2 j 3 or 4 CH20Ac - 17 -Dihydroxylation of the double bond followed by reduction of the ester afforded the expected chain-extended aldose.^ Similarly the 97 Wittig reaction has been used to prepare unsaturated higher ketoses, aldonic a c i d s , ^ deoxy sugars,^ and C-glycosides. 1^ Chain extension has also been achieved at the C-5 end of a blocked sugar. Thus the aldehydo compound (48) was condensed with the y l i d formed from n-pentadecyltriphenylphosphonium bromide or with ethoxy-carbonylmethylene triphenylphosphorane to afford (49) and (50) respectively. R =CK s 0. ° C H 3 4 8 R = 0 49 R = CH(CH 2) nCH 3 50 R = CHC02Et Similarly the 5'-aldehydo nucleoside (51), when reacted with the y l i d formed from (ethoxycarbonylmethyl)triphenylphosphonium bromide afforded a number of products including the expected unsaturated compound (52), uracil, and the unsaturated compounds (53), (54) and (55). As mentioned in section 1.2.2 this laboratory has synthesized many branched-chain sugars via the modified Wittig reaction and the standard Wittig reaction. Primarily the condensation has been used to 34 afford 3-deoxy compounds containing either exocyclic methylenic, 3 2 3 3 3 7 3 5 3 6 methoxycarbonylmethylenic, ' ' or cyanomethylenic ' functionalities attached to hexoses and pentoses. - 18 -OCH 0 ^  R=CH 0 R — CH 51 52 R = CHC02Et 54 R = CHCO Et 53 R = CHC02H 55 R = CHC02H Recently Tronchet and coworkers have also synthesized numerous exocyclic-unsaturated sugars"^ ^ and investigated their nmr spectral ,„1 3 „13. , . £ , . 102,105 properties (H and C ) m relation to conformational analysis. In addition these workers have investigated epimerization processes observed in some Wittig reactions on keto-sugars.^"^''''^4 Lance and Szarek"*"^ have also described several examples of the reaction of methylene-triphenylphosphorane with keto sugars. Instead of using a carbonyl-containing sugar Zhdanov and Polenov*^ converted the carbohydrate moiety into a Wittig reagent (56) and proceeded to condense this reagent with p_-nitrobenzaldehyde to afford the a, B-unsaturated ketone (57). 0 0 U 0. // ^0 (C H ) P = C B - C y / U \ £-N0 (C H )CH= CH-C 0CH3 56 I 57 - 19 -3.2 The Phosphonate Modification of the Wittig Reaction Various phosphorus containing compounds have been shown^ to afford phosphonium ylids i f there exists a hydrogen atom adjacent to a phosphorus atom carrying a reasonable degree of positive charge as in the case of phosphine oxides (58), phosphinates (59) and phosphonates (60). The last group of compounds, the phosphonates, have found RO 0 / { ) \ R O ' - Y -* ' V R0 / V C 58 59 60 considerable u t i l i t y in the formation of Wittig reagents, the so-called modified Wittig reagents. 108 Horner and coworkers were apparently the f i r s t to use this type of reaction in the condensation of diethylbenzylphosphonate with 109 benzaldehyde using sodium amide as base. Later Wadsworth and Emmons showed that stabilized phosphonate carbanions were more reactive than triarylphosphoranes towards some aldehydes and ketones, and in addition required milder conditions, were less expensive and could be readily worked up in water, since the resultant phosphate by-products were water soluble, and did not cause the same problems in product isolation that triphenylphosphine oxide does in the classical Wittig reaction. 32 Thus with these facts in mind Rosenthal and Nguyen reacted 1,2:5,6-di-0-isopropylidene-a-g-ribo-hexofuranos-3-ulose (9) with carbomethoxymethyldimethylphosphonate in the presence of potassium - 20 -t-butoxide, followed by hydrogenation, to yield 3-C-(carbomethoxymethyl)-3-deoxy-l,2:5,6-di-0-isopropylidene-ct-D-allofuranose (12). Similarly reaction of the carbanion formed from diethylcyanomethylphosphonate 35 36 with (9) afforded after hydrogenation the cyanomethyl sugar ' (69). Application of the same reaction conditions that produced (12) to the 2-deoxy-hexopyranos-3-ulose (70) resulted in (71) after hydrogenation. - 21 -Jones and Moffatt^"* condensed the phosphonate Wittig reagent diphenyl-triphenylphosphoronylidene-methylphosphonate (72) with adenine and uracil nucleosides (73) and (74) respectively having an aldehydo function at C-5', to afford the 5'-deoxy-5'-(dihydroxyphosphinyl-methyl) nucleosides (75) and (76). R-CH ^ 0 Ph3P=CROP(OPh)2 72 0 0 73 B = adenine, R = 0 74_ B = uracil, R = 0 75 B = adenine, R = CHPO(OPh)2 76_ B = uracil, R = CHPO(OPh)2 3.3 Mechanism and Stereochemistry Whereas the mechanism of the Wittig reaction has been extensively investigated the modified Wittig reaction remains to some extent the subject of some speculation. It has been suggested by many authors^'^^ that the mechanism is analogous to that of the. Wittig reaction i t s e l f . Thus attack of the carbanion (61) upon the carbonyl carbon of the aldehydo or ketone (62) would afford either the threo or erythro betaines, (63) or (64) respectively. Either betaine may then undergo decomposition via cis-elimination of phosphate to afford the trans or cis olefins, (65) or (66), respectively. - 22 -R ( / C =s 0 62 1 V - V R 2 PO(R 30) 2 0 0 erythro 64 ( R ^ P 4 * — CHRA 41 \ R. (R30)2PO-n threo 63 where R-j i s the larger group; R2 is the smaller group. C = C / \ H R, cis 66 R H R-\ / 1 C =C / \ 4 R2 trans 65 It has been common un t i l now to describe the products of Wittig reactions as cis and trans along with some qualifying statement as to the definition of these geometrical isomers. This being of particular importance in the case of products derived from ketones. The general procedure in such cases has been to define as trans the product in which the two most bulky groups are trans to each other. In the case of cyclic ketones, such as keto-sugars, one procedure has been to relate the configuration to the relative positions of the olefinic proton and the proton on C-2. Thus using this method (11), having H-2 and H-l' i n a cissoid relationship, would be defined as c i s , and (10) as trans. - 23 -X X MeO Y ° 1 A 0 - / ' v f 11 10 However i t can be seen that such methods are prone to confusion and thus in the work to be subsequently described the IUPAC nomenclature shall be used, 1"^ based on a priority system similar to that of Cahn, Ingold and P r e l o g . 1 1 1 Thus the rules for specifying the configuration about the double bond are: "(I) for each double bond to be described configurationally, determine which of the two groups attached to each of the doubly bound atoms has the higher priority according to the sequence rules of Cahn, Ingold, and Prelog; (II) that configuration in which the two groups of higher priority are on the same side of the reference plane is assigned the stereo-chemical descriptor (from the German zusammen); that configuration in which these groups are on opposite sides i s assigned the descriptor. E (from the German entgegen)" [Ref. 112]. Thus, (11) becomes I! and (10) becomes on application of these principles. Recently two groups have investigated the stereochemical course of the reaction of the carbanion from diethylcyanomethylphosphonate 113,114 with various aldehydes and ketones. In the study of Jones and 113 Maisey i t was shown that the ratio of stereoisomers was dependent on - 24 -steric factors of the ketone. Thus when It, was increased in size the cis/trans ratio increased as expected [where cis and trans are as defined in (66) and ( 6 5 ) ] . 114 More recently Lefebvre and Seyden-Penne prepared the erythro-and threo-diethyl-l-cyano-2-hydroxy-2-phenylethylphosphonates > (67) and (68) respectively, the supposed prdtonated intermediates of a modified 67 o 68 0 Wittig reaction. These workers suggest that the cis/trans ratio was largely dependent on the relative rates of phosphate elimination from „ the oxyanions formed from (67) and (68) in basic media. They have found that (64) and (63) are interconvertible both via (61) and via some other mechanism not involving (61). This process was indicated by the fact that both (67) and (68) gave very similar ratios of cis and trans products, as did simple reaction of (61) with benzaldehyde. 4. Nucleosides 4 . 1 Branched-Chain Nucleosides It has become general, through common usage, to refer to compounds in which a nitrogen heterocycle, usually a purine or pyrimidine, i s linked to the anomeric position, through nitrogen, of a carbohydrate which contains branching in i t s carbon skeleton, as a branched-chain nucleoside. - 25 -Table I ; Branched-Chain Nucleosides Ref. 118 HOCH2 o B Ref. 123 NH2 OH 75 B = Ad,DMP,G,C,U HOR CH3 \ 119 OCH HO M / Cl-purine purine CH30 OCH3 71 CH2OH 0 Ad 121 CH2OH 132 73 R1=DMP;R2=H;R3=NHC02Et 74 R1=H;R2=DMP;R3=NHC02Et 2. 0 R 1 R2" HO OH 124-126 76 R=R1=H;R2=CH3;B=Ad 77. R=R2=H;R1=CH3;B-Ad 78 R1=R2=H;R=CH3;B=Ad 79. R=R1=H;R2=CH3;B=pyrimidines 80 R=R2=H;R1=CH3;B=pyrimidines 81 R^Ad; R2=H 82 R^H; R2=Ad 127, 128 83 R^Ad; R2=H 84 RX=H; R2=Ad Cl-Ad 122 - 26 -Table I: (Continued) HOCH2 0 Ad HOCH2CH2 OH 86 HOCH2-CH„ 0 Ad HOCH2CH2 OH HOCH2 o OH CH, 88 B = adenine 88 B = thymine HOCH2 0 DMP Ref. 37 37 B 131 30,133,134 RCH2 OH 89 R=N02; 90 R=CN; 91 R=CH NH 92 R=C0NH2; 93_ R=C0N(Me)2 H0CH„ 0 DMP Ref. 133 CH 0 II 0 95 H0_ HO-Ad 134 AcHNCH9CH2 OH 96 H0_ HO. 0 DMP 134 (Me)„NCCH„ OH 2 tl 2 0 94 Abbreviations used: Ad = adenine, C G = guanine. = cytosine, DMP = 6-dimethylpurine, - 27 -Since the f i r s t synthesis of cytotoxic nucleosides of branched-chain sugars by Walton and coworkers11** an increasing number of synthetic studies of nucleosides of this class of sugars have been reported; some nucleosides exhibiting biological a c t i v i t y . 1 1 ' ' A representative sample of these compounds are shown in Table I. A number of these branched-chain nucleosides have been investigated for biological activity. In particular the series of 2'-C- and 3'-C-methyl nucleosides [(76)-(80)] produced by Walton and coworkers have 125 shown inhibition of KB cells in culture and to be effective anti-126 vaccinia agents in mice, with the nucleosides in which branching occurs at the 3'-position of greatest activity. Similarly the 3'-amino-3'-hydroxymethyl derivative of adenosine exhibited weak inhibition 123 against Vaccinia Dairen. 4.2 Nucleoside Synthesis The various methods of nucleoside synthesis have expanded rapidly in the last few years. Since there have been reviews in the development 136—138 of nucleoside synthesis a discussion of the numerous methods would be unnecessary and beyond the scope of this thesis. Thus only the procedures used in synthesis of compounds discussed in the experimental section and the background to these procedures shall be dealt with here. Fusion of blocked carbohydrate, having either a halogen, or acyl group at the anomeric position, directly with a purine under diminished 135 pressure has frequently been utilized in nucleoside synthesis. This method i s applicable to substituted purines that are specifically substituted for later conversion into other derivatives. It is not necessary to further substitute such functions as amino, oxo, or thio - 28 -that might already be present as part of the structure. If the carbo-hydrate i s substituted at C-2 with an acyl group the configuration of 140 141 the resultant nucleoside may be predicted by Baker's trans rule, ' which suggests that the predominant, i f not unique, nucleoside formed w i l l have a trans arrangement of the C - l 1 and C-2' substituents. [Whilst this rule was originally enunciated for glycosyl halides reacting with heavy metal salts of bases, i t empirically appears true for simple fusion reactions.] However, i f C-2 has a nonparticipating substituent (ether) or no substituent (2-deoxy) then both anomeric 1 A A i _ 1 3 6 nucleosides may result. ROCH 2^ 0 R0CH„ o b a s e ROCH 2 0 100 R0CH2 0 \3 base OR 101 Most reactions in the literature to date have used almost exclusively acetates and other acyl groups at the anomeric position in direct fusion reactions. However i t has been observed that methyl glycosides may also be used directly in nucleoside synthesis. Furukawa 1A 2 1A 3 and coworkers found that application of Bonner's method of 0_-glucosylation for the synthesis of purine nucleosides by the use of the boron trichloride complex of methyl D-ribo-furanoside (102) afforded the required nucleosides; only the a-nucleoside (104) was observed. Fusion of a substituted purine with a sugar has been known to give rise to two basic substitution patterns on the purine heterocycle; glycosylation through either N-9 or, less frequently, through N-7. Fortunately N-9 purine nucleosides have a u.v. maxima at 275 nm whilst the N-7 analogs absorb at 295 nm thus permitting ready differentiation. Determination of the configuration of purine nucleosides has been aided by the use of optical rotatory dispersion (ORD) and circular 145-147 dichroism (CD) spectra. It has been empirically found that purine nucleosides of the B-g-configuration afford negative Cotton effects, whilst cx-D-purine nucleosides give positive Cotton effects. 118 Hata and coworkers have shown that a number of purine nucleosides of L-mycarose and L-cladinose, i.e., branched-chain L-nucleosides, afford positive Cotton effects when in the g-L-configuration. These results were confirmed by nuclear magnetic resonance studies of the hexopyranosyl compounds. - 30 -5. Peptidyl Nucleosides Branched-chain carbohydrates have e l i c i t e d interest for many years as seen in section 1 . There has been particular interest, however, in amino sugars, partly because they occur as basic constituents in 148 many antibiotics. In addition many nucleoside antibiotics contain peptide linkages and amino acid functionalities attached to the carbohydrate moiety. Examples of such compounds are: blas t i c i d i n S (105), gougerotin (106), and puromycin (1). Gougerotin (106) Blasticidin S (105) Puromycin (1) Umezawa and coworkers"'4>149,150 ^ a v e Sy nthesized a number of nucleosides which contain a free a-amino acid residue in the carbohydrate moiety. Both furanosyl and pyranosyl nucleoside derivatives, (108) and (109) respectively, have been synthesized. Unfortunately . biological testing on these compounds has not been reported. - a l -ios 109 5.1 The Polyoxlns The polyoxlns represent a new group Of peptidyl nucleoside antibiotics that are antifungal in their action. To date twelve polyoxins (A-L) have been isolated from Streptomyces cacaoi var. asoensis and characterized. A l l twelve polyoxins contain 3 common structural features: (i) an a-L-amino acid, ( i i ) a 5-aminofuranuronoside, and ( i i i ) a pyrimidine chromophore. The constitution of a l l twelve polyoxins, as elucidated by Isono and coworkers1"^ since 1966, are depicted i n Table II. Polyoxins C and I are the only members of the series which do not exhibit extreme toxicity towards the photopathogenic fungus, Pel l i c u l a r i a filamentosa f. sasakii, which causes sheathblight i n rice plants. The biological activities of the polyoxins are very characteristic since they act specifically against photopathogenic fungi and lack activity against bacteria, and in addition have l i t t l e or no toxicity towards mice, fish or plants. Polyoxin D was found to be the most effective in controlling this disease, although many other members also - 32 -Table II: Polyoxins A-L. CH.OCONH, 2. i H 2NCH 0 Polyoxin R l R 2 R3 (110a) A CH2OH A OH (b) B CH2OH OH OH (c) D C02H OH OH (d) E C02H OH H CO H I (e) F C02'H OH (f) G CH20H OH OH C02H (g) H CH3 OH (h) J CH3 OH OH CO H (i) K H OH (j) L. K OK OH . R (k) C OH COOH % (1) I OH OH - 33 -exhibited activity as shown in Table I I I . Table I I I : Antifungal Activities of Polyoxins MIC", mcg/ml Test-organism A B C D E F G H I Piricularia oryza: 3.12 6. 25 >100 3.12 12. 5 25 6.25 3.12 >100 Cochliobolus miyabeanus 3.12 3. 12 >100 6.25 12. 5 6.25 3.12 25 >100 Ptllicutaria sasakii 12.5 1. 56 >100 >1.56 1. 56 50 1.56 50 >100 Alternaria kikuchiana 50 12. 5 >100 50 50 >100 6.25 12.5 >100 Clomtrelln cingulata >100 >100 >100 >100 >100 >100 >100 >100 >100 Physalospora laricina 25 3. 12 >100 100 50 >100 6.25 12.5 >100 Cladosporium Julvum 3.12 1. 56 >100 100 25 25 3.12 6.25 >100 Fusarium oxysporurn >100 >100 > 100 >100 >100 >100 >100 >100 >100 • Minimal inhibitory concentration. [Ref. 151]. Numerous^2 ^ 4 workers have examined the mode of action of the polyoxins and have found that they inhibit glucosamine uptake which suggests'^"' that the site of action may be related to cell-wall chitin biosynthesis. A recent study''"^^ on the effect of polyoxin A on plant viruses has shown that this compound was a more successful inhibitor of tobacco mosaic virus (TMV) than blasticidin S. Since the polyoxins are now in practical use in Japan as agricultural fungicides and since the fractionation process of the various polyoxins from natural sources i s long and tedious the chemical synthesis of the components has engaged considerable study. Naka and 158 coworkers have reported the f i r s t chemical synthesis of a sugar component of the polyoxins, namely methyl(5-benzamido-3-0-benzoy1-5-deoxy-l,2-0-isopropylidene-°-g-allo-furan)uronate (111), as the f u l l y - 34 -Figure I: Fractionation Process of Polyoxins. r ~ H Culture broth | pH 2.2 Celite Filtrate Do-Lex J0W-X8 (—H) | O.JN NH<OH Eluate cooc., ipray^lry Crude polyoxin complex Amberlite IR-4B (—Cl) _ _ J Effluent Reiidue I Dowcx 50W-X8 i M/10 KH.PO,—HC1 Active fraction! Carbon J 60% acetone Eluate J cone., precipitation Mixture of polyoxin 1 A , ^B, C , G, H . I Cellulose column BuOH-AcOH-HiO (4:1:2) I C T G 1 Ad tor bate J 6#NaCI Eluate Carbon MeOH-CiH.N.HiO (3:1:4) Eluate J cone, precipitation Powder 0.2 N HCI Dowex S0W-X16 I Effluent I Carbon MeOH-C,U,N-H<0 (5:1:4) Eluate J cone, precipitation Mixture of polyoxin D, E, F Cellulose column BuOH-AcOH-H/> (3:1:4) r T E [Ref. 151]. blocked derivative. More recently Moffatt and coworkers have described the synthesis of the basic nucleoside skeleton (112a) of the polyoxins as well as i t s q-L-talo-isomer (112b). In addition Emoto et a l . 1 " 5 ^ have described the preparation of "thymine polyoxin C" (113). COOMe BzHN _ OBz 111 COOH NH„ Ur OH OH 112b COOH H2N_4- Ur OH OH 112a COOH H2N—-H o Th OH OH 113 Rosenthal and Shudo, i n this laboratory have synthesized stereospecifically an analog of the sugar moiety of the polyoxins in which the a-L-amino acid moiety was attached to C-3 of a hexofuranose to afford L-2-(3_deoxy-l,2-0-isopropylidene-a-D-allofuranos-3-yl)glycine (114) from the a-hydroxy ester (115). Much to the consternation of - 36 -this group Jordaan et a l . ^ ^ recently published the X-ray crystal structure of a similar compound, 5-brosyl-3-deoxy-3-C-(R)-(ethoxy-carbonylformamido)methyl-l,2-isopropylidene-ct-g-ribofuranose (116), which i s analogous to the D-amino acid analog described later i n this thesis. Br - A \V-S — 0 —CH 2 ^ 0 \ = / II 0 1 y y v v OHC-HN-116 -H C02Et '4 The experimental procedure for the synthesis of this compound has 162 just been published and u t i l i z e s the method of Schollkopf to be described in the next section. Various workers have synthesized carbohydrate derivatives of amino acids, however these are almost invariably linked through the nitrogen of the amino acid. A number of methods for the synthesis of these derivatives have been reported and include, condensation of a suitably protected amino acid derivative with the appropriate 163 halo sugar, at moderately elevated temperatures, direct condensation of the aldehyde group of the free sugar with the amino group of the amino acid,"'"^4'''"^ condensation of amino acid amides with glycosyl halides,"^"' and acylation of the amino group of amino sugars with 166 167 amino acid derivatives. ' A recent novel approach makes use of the activating effect of the nitro group in an ct-nitroolefinic sugar 117 , which permits addition, via the amino group, of amino acid esters. The a-nitroolefinic sugar can be generated in situ from i t s 168 B-nitroacetoxy precursor. R2 0 nrH 1 1 H 3 H 2N— CH—COOR HCl 117 6. Synthesis of Glycosyl Amino Acids Of the small number of glycosyl amino acids synthesized and reported to date the most popular procedure has been displacement of either a secondary methanesulf o n y l o x y ^ ^ ' o r toluenesulfonyloxy^"^ group with sodium azide followed by reduction of the azide to an amine. 159 The procedure of Moffatt and coworkers util i z e s the reaction of the 5'-aldehydo nucleoside (119) with sodium cyanide in aqueous methanolic potassium carbonate, and hydrogen peroxide to form the epimeric hydroxyamides 120a and 120b . Reaction of 120a with methane-sulfonyl chloride in pyridine followed by, azide displacement of the sulfonate, acid hydrolysis and hydrogenation afforded the basic polyoxin skeleton 121a . Similarly 120b. afforded 121b . Deamination of 121a and 121b with nitrous acid, which is known to proceed with retention of configuration in a-amino acids, due to participation of the carboxylic acid function, 169,170 a f f o r c j e d the two ct-hydroxy derivatives obtained by hydrolysis of 120a, and 120b . Used in an extended sense. - 38 -H — Uracil NaCN-K2C03 H 1) MsCl-pyr 2) NaN3-DMF 0 3) H 4) H2-Pd/C 119 120 a_, Series as above; b_, Series as above but inverted configuration at C-5', Similarly the methods of Naka 1 5^ and Emoto1"^*3 u t i l i z e d the azide displacement of a 5-sulfonyloxy group of a suitably blocked hexo-furanose, followed by oxidation of the 6-hydroxyl and subsequent reduction of the azide to afford the 5-amino-5-deoxy-allofuranuronic acid. The configuration of the so formed amino acids were assumed to be those arising from simple SN2 displacement of the methanesulfonate or toluenesulfonate group by azide although Emoto 1 5^ 3 and coworkers correlated "thymine polyoxin C" with a sample from natural sources. 149,150 Umezawa and coworkers,' in synthesizing the novel 3-amino-3-C-carboxy-3-deoxy derivatives 108 and 109 , used the Bucherer 171 hydantoin synthesis to prepare the hydantoin 122 . Hydrolysis with bariumhydroxide then gave the partially blocked compound 123 - 39 -BzOCH2 0 HOCH 122 123 Compound 116 was synthesized by Jordaan et a l . ^ ^ from the keto-sugar 9_ with ethyl isocyanate in the presence of sodium hydride to afford after hydrogenation the blocked glycosyl-D-amino acid (124). + N= C CH2C02Et Et0 2C 124 - 40 -III. RESULTS AND DISCUSSION The work to be described on the following pages f a l l s into two basic units, (1) chemical studies on the synthesis of analogs of the nucleoside antibiotic puromycin; (2) synthesis of analogs of the sugar moiety of the antifungal nucleosides, polyoxins A-L. The order that shall be followed in discussing this work i s summarized under the following headings: 1. Branched-Chain Nucleosides: Structural Analogs of Puromycin 1.1 Synthesis of Branched-Chain 2-Deoxy-Hexopyranosyl Nucleosides. 1.2 Synthesis of Branched-Chain 2-Deoxy-pentopyranosyl Nucleosides. 2. Glycos-3-yl q-Hydroxy Esters, q-Amino Esters and q-Amino  Acids: Structural Analogs of the Sugar Moiety of the  Polyoxins 2.1 Synthesis of D-2- and L-2-(l,2:5,6-Di-0-isopropylidene-q-D-glucofuranos-3-yl)glycine. 2.2 Synthesis of D-2- and L-2-(l,2:5,6-Di-0-isopropylidene-q-D-galactofuranos-3-yl)glycine. 2.3 Synthesis of p-2-and L-2-(3-Deoxy-l,2:5,6-di-0-isopropylidene-q-D-allofuranos-3-yl)glycine. - 41 -1.1 Synthesis of Branched-Chain 2-Deoxy-hexopyranosyl Nucleosides  Related to Nucleoside Moiety of Puromycin Previously i t was mentioned that a number of deoxy-hexopyranose 172 173 sugars and nucleosides ' have shown biological activity and are an essential component of various antibiotics such as chalcomycin, - , . 174,175 - , , . 1 7 6 lankamycin and chromomycin A^. With these facts in mind the synthesis of methyl 3-C-cyanomethyl-2,3-dideoxy-q-D-ribo-hexopyranoside (145) and a number of nucleosides derived from i t , was initiated. 1.1.1 Methyl 4,6-0- benzy lidene-2-deoxy-q-rj-erythro-hexopyranosid-3-ulose (127) To obtain f i r s t l y methyl 4,6-0-benzylidene-2-deoxy-q-D-ribo-hexopyranoside (126), from which the ketose 127 i s derived, i t was necessary to prepare 126 using known procedures'*"^ beginning with methyl 2,3-anhydro-4,6-0-benzylidene-q-p-allopyranoside (125) . Reaction of 125 with lithium aluminum hydride in tetrahydrofuran yielded 126 quantitatively, resulting from axial attack at C-2, essentially by a hydride ion, to afford the 2-deoxy compound 126 . Subsequent oxidation of 126 using either ruthenium t e t r o x i d e ^ ' ^ ^ or dimethyl sulfoxide-acetic anhydride afforded the ketose 127 in 70-80% yield. Even though the ruthenium dioxide-sodium metaperiodate reaction afforded 127 in slightly lower yield (68% as compared to 78%) than the DMSO-acetic anhydride reaction i t was the method of choice, since the product was very pure and required only several hours to complete. On the other hand the DMSO-acetic anhydride oxidation took - 42 -3-4 days and afforded a product which always contained dimethylsulfide even after recrystallization and chromatography. 0 OH u 125 126 127 1.1.2 Methyl 4,6-0-Benzylidene-E-(and Z)-3-C-cyanomethylene-2,3-dideoxy-q-D-erythro-hexopyranoside (128 and 129) Reaction of the 3-ketose (126) with the carbanion formed from diethyl cyanomethylphosphonate and sodium hydride in dimethyoxyethane 133 according to the method of Rosenthal and Baker afforded on work up two products. These compounds were separated by low pressure column 180 chromatography and recrystallized from ethanol. Both compounds showed the presence of an unsaturated n i t r i l e function by their infrared spectrum and were assumed to be the expected E_ and £ isomers about the double bond. The major component (128)(69%) was assigned the structure of methyl 4,6-0_-benzylidene-Ji-3-C-cyanomethylene-2,3-dideoxy-q-p-erythro-hexopyranoside whilst 129 (7%) was assigned the structure of the Z-isomer. The stereochemical assignment of compounds 128 and 129 was based primarily upon their nmr spectra. The shielding and deshielding influences associated with the anisotropy associated with the cyano group have a significant effect on the chemical shift of protons on _ Prepared in conjunction with Mr. B.L. C l i f f . T 4 5 6 7 8 Figure II: Partial 100 MHz NMR Spectrum of Methyl 4,6-Benzylidene-E-3-C-cyanomethylene 2,3-dideoxy-ct-g-erythro-hexopyranoside (128) in CDC13. ( a) I r r a d i a t i o n o f H - l ; (b) H - l ' ; (c) H-4 7 1 1 1 I I I I ' L x 4 5 6 7 Figure III: 100 MHz NMR Spectrum of Methyl 4.6-0-benzylidene-Z-3-C-cyanomethylene-2,3-dideoxy-a-p-erythro hexopyranoside 129 in CDC1. I r r a d i a t i o n of (a) H - l ; (b) H - l ' . - 45 -carbons two and four. When the n i t r i l e group i s cis with respect to C-2 a deshielding of the equatorial proton i s observed. Similarly when the n i t r i l e group i s trans to carbon two (and cis with respect to carbon four) the proton on C-4 i s observed at lower f i e l d . This i s summarized in Table IV. Compound T - H 2 e T _ H 2 a T _ H 4 T - H ^ » 128 6.93 7.43 5.91 . 4.50 129 7.40 7.32 5.72 4.70 Table IV. Chemical Shift Data for Compounds 128 and 129-- 46 -Tronchet and coworkers have recently investigated many such isomeric pairs of unsaturated cyanomethylene sugars and have in a l l instances observed the deshielding effect of the n i t r i l e group upon the proton in which i t i s in a cisoid relationship. Table V summarizes their relevant data. Table V. No. Configuration^ at H-l' R l R 2 x-H2 T ~ H 4 T " H 1 130 cis (a) H 5 . 0 0 4 . 9 5 4 . 2 1 131 trans (a) H 4 . 8 0 5 . 3 2 4 . 0 1 132 cis H (a) 5 . 0 5 5 . 2 6 4 . 2 5 133 trans H (a) 4 . 8 5 5 . 2 7 4 . 3 5 134 cis H H 5 . 0 8 5 . 1 7 4 . 3 8 135 trans H H 4 . 8 7 5 . 2 7 4 . 5 2 136 cis CH3 H 5 . 0 3 4 . 9 1 4 . 3 8 137 trans C H 3 H 4 . 8 6 5 . 1 6 4 . 7 1 138 cis H CH3 5 . 0 9 5 . 0 4 4 . 3 8 139 trans H CH3 4 . 8 6 5 . 3 4 4 . 6 4 Table V. Chemical Shift Data for Compounds 1 3 0 - 1 3 9 . (a) 0-Isopropylidene-l ,2-glycolyl. (b) Cis and trans are defined as in Reference 1 0 2 , i.e. as shown in the diagram above. * There are numerous misprints in this reference. The correct data was obtained from the Ph.D. thesis of J.M. Bougeois. - 47 -Similarly, H-l' was observed at lower f i e l d in 128 than in 129 implying a possible deshielding interaction of the benzylidene ring oxygen on this proton. The nmr spectra of both 128 and 129 were amenable to extensive analysis which was confirmed by decoupling experiments. These results are tabulated below. The protons on carbon two were assigned to be axial or equatorial based on the observed coupling constant with the anomeric proton, which was assumed to be equatorial. The smaller values, 0.5 Hz for 128 and 1.5 Hz for 129 imply an equatorial-equatorial"''^ 3 relationship, whilst the values of 4.0 Hz and 3.5 Hz suggest an equatorial-axial arrangement between the anomeric proton and the respective H-2. Compound 128 Compound 129 H-l H-2 e H-2 H-l' a H-4 H-l H-2 e H-2 a H-l' H-4 H-l - 0.5 4.0 - H-l - 1.5 3.5 -H-2 e 0.5 - ^15 - H-2 e 1.5 - M.5 1.0 H-2 a 4.0 ^15 2 1 H-2 a 3.5 vL5 - 2 1 H-l' - - 2 2 H-l' - 1.0 2 2 H-4 - - 1 2 - H-4 - - 1 2 Table VI. Observed Coupling Constants in Compounds 128 and 129 (in Hz). - 48 -1.1.3 Methyl 4,6-0-benzylidene-3-C-cyanomethyl-2,3-dldeoxy-a-D-. j. , ribo-hexopyranoside (140) Hydrogenatlon of the a,3-unsaturated n i t r i l e proceeded smoothly from either 128 or 129 to afford a single crystalline cyano-sugar when performed over 5% palladium-on-charcoal at atmospheric pressure; the uptake of hydrogen being slightly more than a molar equivalent. 33 Although benzylidene blocking groups have been known to hydrogenolyze no trace of unblocking was observed in this study. Infrared (ir) and thin layer chromatography (tic) revealed that the n i t r i l e group was not reduced under these conditions, whilst nmr indicated the presence of a new methylene group observed as an ABM multiplet centered at T 7.12. The configuration at carbon three of the cyanomethyl branched-chain sugar 140 was determined by nmr. A fact previously mentioned, that in general equatorial-equatorial coupling constants are of the order 0-1 Hz, whilst hydrogens having an axial-equatorial arrangement usually show a mutual coupling of 3-5 Hz, and diaxially orientated protons afford much larger couplings of 7-10 Hz. Coupling constants between H-3 and H-2e, and H-3 and H-2a were observed to be 2 Hz and 4 Hz respectively. Additionally, the coupling between H-3 and H-4 was observed to be 6 Hz. Thus i t can be concluded that H-3 must be equatorial, thereby implying that the cyanomethyl branched-chain i s "down" and 140 i s of the ribo configuration. This result was not unexpected since hydrogenation would proceed by addition to the olefinic bond from the least hindered side, that being from above the plane of the ring. As shall be seen later (section 1.2.5) the configuration of the anomeric substituent has a pronounced effect on the way in which reagents may approach the ring. Prepared in conjunction with Mr. B.L. C l i f f . - 49 -Thus compound 128 i s no exception to this. In the cx-anomer the methoxyl group at C-l prevents attack from below the ring even at the relatively remote exocyclic double bond to afford only the ribo product. 1.1.4 Methyl E-3-C-Cyanomethylene-2,3-dideoxy-a-g-erythro-hexopyranoside (141) and i t s p_-Chlorobenzoyl Derivative (142) The a,B-unsaturated n i t r i l e 128 , when treated with Dowex 50W-X8 + 28 (H ) i n anhydrous methanol readily underwent loss of the 4,6-fJ-benzylidene group within 2.25 h at reflux temperature, to afford a single crystalline product. Nmr and infrared spectroscopy revealed 22 that no anomerization had occurred (since the coupling constants between the anomeric proton and H-2 and H-2 were 4 Hz and 2 Hz r a e respectively) and that the unsaturated n i t r i l e was s t i l l present (shown - 50 -by i r maxima at 2260 cm ^ and 1645 cm ^"). In addition, the infrared revealed the presence of hydroxyl groups at 3400 cm ^. Thus based on this evidence, coupled with an elemental analysis, i t was concluded that the benzylidene group of 128 could be readily removed without yielding anomerization or altering the unsaturated n i t r i l e . This i s an interesting finding when compared to the same reaction as applied to the saturated derivative 140 to be discussed i n the next section. Compound 141 was further characterized as i t s p_-chlorobenzoate which again exhibited data consistent with the a,B-unsaturated n i t r i l e 142. . In the rinr spectrum, protons on C-4 and C-6 were observed at considerably lower f i e l d due to the deshielding effect of the p_-chlorobenzoyl groups and reveal clearly the coupling data should i t be required. 128 142 R = p_-chloro-benzoyl 1.1.5 Debenzylldenation of Methyl 4,6-0-Benzylidene-3-C-cyano-methyl-2,3-dideoxy-a-D-ribo-hexopyranoside Various methods have claimed success in removal of benzylidene 182b blocking groups. However i t does appear that in systems as highly functionalized as many carbohydrates are, that any given method may or may not be successful. Hydrogenation over palladium on carbon has been used successfully in the debenzylldenation of methyl 4,6-0-- 51 -benzylidene-3-C-(carbomethoxymethyl)-2,3-dideoxy-a -D-ribo-hexo-33 pyranoside (143) and methyl 4,6-0i-benzylidene-2-deoxy-3-£-28 nitromethyl-q-D-ribo-hexopyranoside (144). However, when the blocked CH2C02CH3 OH 1 4 3 144 cyanomethyl sugar 140 was subjected to hydrogenation over palladium on carbon at atmospheric pressure debenzylidenation was not observed. Since cyanomethyl groups may be readily reduced when hydrogenated in 13 3 X 3A the presence of platinum oxide ' the conditions for deblocking were not made more vigorous. Even though ultimately i t was desired to convert the cyanomethyl group to an aminoethyl function i t would not 183 be wise to do so at this stage. It has been observed that N-acetyl groups may participate to form nitrogen heterocycles in nucleoside synthesis. Therefore in the light of the favourable results afforded by the acid hydrolysis of the a,g-unsaturated n i t r i l e 128 using Dowex 50W-X8 (H+) resin and on other reports of successful debenzylidena-tionsusing various resins in their acid forms methyl 4,6-0-benzylidene-3-C-cyanomethyl-2,3-dideoxy-a-D-ribo-hexopyranoside (140) was treated with Dowex resin in anhydrous methanol. When the reaction mixture was stirred at room temperature a relatively clean reaction ensued, affording two products, shown by nmr, having similar R 's on t i c . - 52 -These compounds were believed to be due to anomerization during the unblocking reaction. These two compounds were separable as their £-chlorobenzoyl derivative. Nmr spectrometry revealed a narrow t r i p l e t for the anomeric proton of the suggested a-anomer (146) and a doublet of doublets for the B-anomer (147). The latter compound could not be Induced to crystallize. Towards the end of the unblocking a third component was apparent by t i c . A satisfactory analysis for the simple unblocked cyanomethyl sugar 145 could not be obtained. However, the inclusion of 0.5 mole of methanol in the calculation rendered a close approximation to the required values. In addition, a small amount of the 1,6-anhydro compound 148 was isolated on p_-chlorobenzoylation. If the cyanomethyl sugar 140 was refluxed with Dowex resin for a short time a complex mixture of products resulted, as evidenced by t i c , of which 145 comprised only a small proportion. Hence i t would appear that by simply saturating the a,B-unsaturated n i t r i l e 128 , to afford the cyanomethyl sugar 140 , a significant effect on the reactivity under acid conditions results. One of the outstanding changes that has occurred is the considerably increased a b i l i t y of 184 140 to anomerize as compared to 128 . It i s well known that 2-deoxy sugars have significantly increased rates of reaction during acid hydrolysis when compared to their "2-oxy" analogs, often showing rate 184 increases of 100 fold. Similarly 6-deoxy sugars also exhibit an increased rate of anomerization. - 53 -146 R1=p_-chlorobenzoyl;R2=H;-R3=OCH 148 R=p_-chlorobenzoyl 147 R =p_-chlorobenzoyl;R0=OCH0;R =H An additional factor to be considered in explaining the increased rate of anomerization during the unblocking of 140 must no doubt be 185 attributed to the configuration of the branched-chain. Bishop has examined extensively the relative rates of anomerization and furanoside-pyranoside interconversions and successfully correlated these results with steric interactions based on established 1,3 and 186—189 1,2 interactions. Thus on these grounds i t would be expected that 140 would ancmerize to relieve the 1,3 interaction between the a-methoxyl and the cyanomethyl function at C-3 whereas in the unblocking of 128 there is no direct steric interaction between the anomeric substituent and the substituent at C-3, thus no enhancement of the rate of anomerization would be expected. - 54 -OCH If the pyranoside anomerization proceeds through a protonated 185 238 species or acarbonium ion as postulated by Bishop and Capon an intermediate as shown below may be involved. H I It may be envisaged that subsequent attack of the C-6 primary hydroxyl on the anomeric center would result in the 1,6-anhydro compound which was isolated in 8% yield as i t s p_-chlorobenzoyl derivative. That the ring was in the "'"C, conformation was shown by - 55 -the observed coupling constants of H-4. The large trans diaxial coupling previously present between H-4 and H-5 was no longer observed in i t s nmr spectrum and a value of 0.5 Hz was revealed, indicating the diequatorial nature of these two protons. Similarly, ^ was approximately 1-2 Hz, confirming the above conformational assignment. 1.1.6 Nucleoside Synthesis: Formation of 4,6-pi-0-£-chlorobenzoyl-3-C-cyanomethyl-2,3-dideoxy-D-ribo-hexo-l-eno-pyranose (149), i — _ — — — — — — — — — — » — 2,6-Dichloro-9-(4' .e'-di-O-p-chlorobenzoyl-S'-^cyanomethyl-2',3'-dideoxy-a- and g-D-ribo-hexopyranosyl)purjyne (150) and (151) In order to prepare a nucleoside from the cyanomethyl branched-chain sugar 146 , containing a 6-N,N-dimethylaminopurine moiety, a number of methods were reviewed. Three principal methods have in the past been employed in nucleoside synthesis; these being (a) Condensation methods using either the sugar acetate (via ' the titanium tetrachloride complex) or the sugar halide with the chloromercuric salt of the base. This method was used by Baker and 190 coworkers in their original work on puromycin. (b) Fusion of the sugar acetate directly with an appropriately 139 substituted base, such as 6-chloropurine with or without acid catalyst, generally results in considerable yields of the required nucleoside. (c) Direct fusion of the sugar in which the C-l substituent 142 i s 0-methyl, has also recently been shown to condense in good yield with certain purine bases. - 56 -Therefore, as the cyanomethyl sugar 146 already possessed the 0-methyl at C-l the latter method was chosen as the most efficacious procedure to use, eliminating the necessity of converting the O-methyl group to either an acetoxyl or halide, as required for procedures (a) and (b). Accordingly compound 146 was fused directly with 2,6-dichloropurine at 155°, under reduced pressure, for two hours. Thin layer chromatography revealed that three products were formed which were readily separated by column chromatography using the previously described low pressure method. The three compounds were subsequently identified as the novel unsaturated branched-chain sugar, 4,6-di-0_-p_-chlorobenzoyl-3-C-cyanomethyl-2,3-dideoxy-D-ribo-hexo-l-eno-pyranose (149), 146 149 R = p_-chlorobenzoyl Cl RO Cl - 57 -resulting from loss of methanol across the 1,2 positions, 2,6-dichloro-9-(4 *,6'-di-0-p_-chlorobenzoyl-3'-C-cyanomethyl-2•,3'-dideoxy-a-D-ribo-hexopyranosyl)purine (150), and 2,6-dichloro-9-(4',6'-di-0-p_-chlorobenzoyl-3'-C-cyanomethyl-2',3'-dideoxy-g-D-ribo-hexopyranosyl) purine (151) isolated in yields of 34, 37, and 24% respectively. The unsaturated branched-chain sugar 149 was readily identified by nmr spectroscopy (Figure IV). It was immediately apparent that the 0-methyl group was no longer present. In addition the anomeric proton was now observed at considerably lower f i e l d (T 3.46), characteristic of glycals. Similarly the single C-2 proton was observed as a double doublet at T 5.15, again significantly down-field shifted from i t s position in 146 . As can be observed in Figure IV the coupling constants for this compound were readily obtainable, revealing J.. 9 to be 6 Hz and an a l l y l i c coupling of H-l to H-3 of 1 > ^ 2 Hz. H-2 was coupled to H-3 by a value of 4 Hz. Also the infrared spectrum of 149 exhibited absorbtions at 2290 cm 1 and 1660 cm 1 corresponding to the n i t r i l e and the olefinic bond further corroborating 191 the above assignment. 1,2-Glycals are common products of such fusion reactions. However, i t is not clear whether nucleoside synthesis by direct fusion goes via the intermediacy of glycals or i f glycal formation is a competing reaction. It is interesting to note, however, that nucleosides have been synthesized using acid catalyzed fusions of 192 1,2 glycals. The f i r s t blocked nucleoside to be eluted from the column was assigned the structure of 2,6-dichloro-9-(4',6'-di-0-p_-chlorobenzoyl-3-C-cyanomethyl-21,3'-dideoxy-a-D-ribo-hexopyranosyl)purine (150). H-8 ft £-ClBzOCH2 p_-ClBzO CNCH„ N Cl Cl 150 H-l* H-4' T 1 ± 10 Figure V. 100 MHz NMR Spectrum of 2,6-dichloro-9-(4' ,6 Wi-Oj-^-chlorobenzoyl-S'-^-cyano^ethyl^' ,3'-dideoxy-a-D-ribo-hexopyranosyl)purine (150) in CDC1 Cl T Figure VI. 100 MHz NMR Spectrum of 2,6-Dichloro-9-(46'-di-0-p_-chlorobenzoyl-3'-C-cyanomethyl-2',3*-dideoxy-6-D-ribo-hexopyranosyl)purine (151) in CDC1_. - 61 -s The anomeric proton in the nmr spectrum of 150 (Figure V) was observed as a doublet of doublets with coupling constants of 10 and 3 Hz. The coupling constants J. r and J„ , were readily determined as 4,5 3,4 1 and 3 Hz respectively. From this data i t must be concluded that the sugar moiety exists i n the conformation and that 150 is indeed an a-nucleoside. This assignment was later collaborated by the circular dichroism spectrum of the unblocked nucleoside. The H-8 proton was observed as a singlet at T 1.70. It i s most l i k e l y that the driving f o r c e , to achieve the conformai-tional inversion, i s the considerable 1,3 interaction that the purine base and the C-3' substituent would experience i f the sugar ring was in the ^C^ conformation. The interaction between the purine and the cyanomethyl group must then be greater than that experienced by the C-5, and C-6 substituents i n their 1,3 interaction with the hydrogen atoms on the ring. H The second nucleoside to be eluted from the column was suggested to be the g-nucleoside (151), again based on i t s nmr spectrum (Figure VI) and subsequently on the cd spectrum of the unblocked nucleoside. The anomeric proton was observed as a " t r i p l e t " at x 3.70, with coupling constants to H-2' and H-2* of 5-6 Hz each. The values are a l i t t l e e a surprising in that i f H-l' is axial as expected for the B-nucleoside then the trans-diaxial value J^, ^ should be around 8-10 Hz, whilst the axial-equatorial value J^, should be approximately 3-5 Hz. Similarly the J,, coupling constant was observed to be 6 Hz, again considerably smaller than the 8-10 Hz expected for the trans-diaxial arrangement. These facts would indicate deformation of the sugar ring towards a more planar situation thus reducing the H-4'/H-5' and H-l'/H-2^ dihedral angles. The driving force for this deformation is not readily apparent, but may be associated with the dipolar interaction of the anomeria substituent with the ring oxygen and steric interactions. H 151 1.1.7 Reaction of Blocked Hexopyranosyl Nucleosides with Dimethyl- amine The substitution of the 6-chloro atom by an N,N-dimethylaminp 193 194 functionality has been elucidated by Robins and used by Fox et a l . Accordingly, when the blocked nucleosides 150 and 151 were treated with dimethylamine in water and methanol, the 6-chloro atom was replaced in addition to the expected removal of the ester groups, to afford - 63 -,152 and 153 . The fact that only one purine chloro atom had been replaced was readily determined by mass spectrometry. However, in addition i t was observed that the molecular ion of 152 occurred at 46 mass units higher than that of 153 and contained an additional N,N-dimethyl group, evidenced by i t s nmr spectrum. The infrared spectra of 152 and 153 indicated that while a n i t r i l e was present in 153 , i t was absent in 152 , and instead, an amide carbonyl at 1660 cm 1 was observed in the latter compound. Together these facts indicate that under exactly the same reaction conditions the blocked a-nucleoside 150 afforded the N,N-dimethylcarbamoyl nucleoside 152 whilst the blocked g-nucleoside retained i t s cyanomethyl functionality to yield 153 . The cyanomethyl hydrolysis reaction had previously been observed JL33 134 in this laboratory ' i n the aqueous dimethylamine deblocking of 6-chloro-9-(2'-0-acetyl-5',6'-di-O-benzoyl-3'-C-cyanomethy1-3'-deoxy-g-D-allofuranosyl)purine (154) and 6-chloro-9-(2'-0-acetyl-5'-0-benzoyl-3'-C-cyanomethyl-3'-deoxy-g-g-ribo-furanosyl)purine (155) to afford 6-Jfl,N-dimethylamino-9-(3'-C-N,N-dimethylcarbamoylmethyl-3'-deoxy-g-D-allofuranosyl)purine (156) and 6-N,N-dimethylamino-9-(3'-jC-N,N-?dimethylcarbamoylmethyl-3'-deoxy-g-D-ribo-furanosyl)purine (157) respectively. The nmr spectra of the unblocked nucleosides 152 and 153 could not, alone, be used for anomeric or conformational assignments. The anomeric proton of compound 152 was observed as a double doublet at T 3.97, having coupling constants of 10 and 4 Hz with the C-2' protons. Similarly the anomeric proton of 153 exhibited a double doublet at NMe, H-8 HOCH C H 2 C N ON I I I' I I I I I I | I I I I ,1 I I I ! I I I I I I I t I I I I I I I I I I 1 I I I I | I I | I I I I I I I. I I I -t T 2 3 4 5 6 7 8 Figure VII: 100 MHz NMR Spectrum of 2-chloro^6-N,N-dimethylamino-9-(3'-C-cyanomethyl-2',3'-dideoxy-g-D-ribo hexopyranosyl)purine (153) in DMSO-d^ . 65 -CH2OR RO CH2CNy N 150 N Y C 1 C l C l ROCH, < RO N N CH2CN 151 R = p_-chlorobenzoyl BzO CH2CN OAc 154 R' = CH^OBz 155 R' = H CH2OH CH, 0= C \ 152 / Me V, N N—Me Cl NMe, NMe, N(Me), R" H o J „ 0. < N N CH„ OH I 2 o = c N(Me), 156 R" = CH„OH 157 R " = H - 66 -T 3.60, with coupling constants of 7 and 4 Hz. Thus again i t would appear that the a-nucleoside is in the ^C^ conformation, whilst the S-anomer i s i n a slightly deformed conformation. 190 The ultraviolet spectra of both 152 and 153 indicated the N-glycosyl bond was attached at the N-9 position of the purine moiety (A = 276 nm). A positive Cotton effect was observed for the max unblocked a-nucleoside (152) and a negative Cotton effect for the 145 B-nucleoside (153) further corroborating the anomeric assignment. When the f u l l y blocked a-nucleoside (150) was reacted with anhydrous dimethylamine at -5° for several days the cyanomethyl nucleoside (158) was recovered in 63% yield. Nuclear magnetic resonance and mass spectrometry proved that both purine chloro atoms had been replaced by N^N-dimethylamino functions and that the ester groups had been removed to afford 2,6-di-N,N-dimethylamino-9-(3'-C-cyanomethyl-2',3'-dideoxy-a-D-ribo-hexopyranosyl)purine (158). Once again the nmr spectrum indicated that the sugar ring was in the ^C^ conformation, shown by the coupling constants of 10 and 3 Hz for the anomeric proton with those protons on carbon two. - 67 -1.1.8 Hydrolysis of the Cyanomethyl Group During Unblocking Reactions In the previous section i t was observed that during identical unblocking procedures, using an aqueous solution of dimethylamine i n methanol at room temperature, that the cyanomethyl function underwent hydrolysis to a N,N-dimethylcarbamoyl derivative in the case of the ct-nucleoside (150), whilst i t remained unhydrolyzed in the case of the 133 13 A g-nucleoside (151). As mentioned previously Rosenthal and Baker ' also observed this hydrolysis with compounds 154 and 155 . However, as w i l l be seen i n section 1.2.8, compounds 159 and 160 were resistant to such hydrolysis affording only the unblocked cyanomethyl nucleosides 161 and .162 , respectively. (159) R^  = p_-bromobenzoyl; R^  = 2,6-dichloropurine; R^  = H (160) R^  = p_-bromobenzoyl; R2 = H; R^  = 2,6-dichloropurine (161) R1 = R^  = H; R-2 = 2-chloro-6-N,N-dimethylaminopurine (162) R1 = R2 = H; R^  = 2-chloro-6-N,N-dimethylaminopurine In attempting to determine the factors which increase the f a c i l i t y with which the cyanomethyl groups of 150 , 154 and 155 undergo base hydrolysis to the carbamoyl compounds other experiments were undertaken. When the a- and g-methylglycosides 146. and 147 were treated under the same conditions as above for 3.5 hours, after which time a l l starting material had been consumed, i t was found that whilst the - 6 8 -OCH 3 HO CH2CN 1 4 6 CH2CON(Me) 163 CH„OR CH OH RO OCH 3 HO c 147 164 g-glycoside afforded the unblocked cyanomethyl compound ,164 , the ct-glycoside underwent hydrolysis of the cyanomethyl group to afford the carbamoyl derivative 163 . However, when methyl 3-(>cyanomethyl-2,3-dideoxy-q-D-ribo-hexopyranoside (146a)[(146); R = K], obtained by treatment of 146 with anhydrous dimethylamine at -5°, and 3-C.-, cyanomethyl-2,3-dideoxy-B-rj-ribo-hexopyranoside (147a) [ (147) ; R = H] were treated with aqueous dimethylamine in methanol neither compound had undergone base hydrolysis of i t s cyanomethyl group after 12 hours, as evidenced by lack of change in the t i c of each reaction mixture. Thus i t would appear that the presence of an ester function f a c i l i t i t a t e s the base hydrolysis of the n i t r i l e . However this ester participation must be considerably dependent upon the relative spatial arrangement since i t has been found that while 146 and 150 were hydrolyzed, the B-anomers ,147 and 151 were not. Similarly, both - 69 -pentopyranosyl nucleosides 159 and 160 »having a threo relationship between cyanomethyl and ester functionalities, were resistant to hydrolysis. 133 13A It has also been found ' that whilst 6-chloro-9-(2'-O-acetyl-5'-0-benzoyl-3'-C-cyanomethyl-3'-deoxy-g-D-ribofuranosyl)purine (165) reacted readily with aqueous dimethylamine in methanol (3 hours 166 R=R'=H N(Me)2 required for complete reaction), the analogous compound 166 , not possessing an acetyl group at C-2', required considerably more time to afford the same product (12 hours required for complete reaction). 195 This latter reaction may be due to simple base hydrolysis of the cyanomethyl group without participation. 196 Helferich recently isolated a compound having the suggested structure 169 as an intermediate in the deacetylation of the glucose tetraacetate derivative 168 , using sodium methoxide in methanol. Thereby implying attack of the n i t r i l e nitrogen upon the carbonyl carbon of the adjacent acetate. If this type of mechanism were operative in our situation i t would involve the formation of - 70 -a six to seven cis-fused bicyclic system, which, from investigation of molecular models, does not appear to be under great conformational stress. Similarly, i f attack of an ester carbonyl on the polar n i t r i l e group occurred, a similar bicyclic system may result. Both these intermediates may afford the observed hyrolysis products in basic solution. - 71 -Anchiomeric assistance by hydroxyl groups have been suggested in 197 the rapid hydrolysis of some aldose cyanohydrins and the n i t r i l e 169a . Whilst this possibility cannot be eliminated in this work OMe M / OMe i t would appear that the presence of an adjacent ester function does enhance the rate of hydrolysis of the n i t r i l e . 1.1.9 2,6-Di-N,N-dimethylamino-9-(3'-C-[2"-acetamidoethyl]-2',3'-dideoxy-cx-D-ribo-hexopyranosyl) purine (170) Reduction of n i t r i l e s has been achieved i n numerous ways, including 198 199 201 200 catalytic hydrogenation, metal hydrides ' and diborane. 133 13 A Rosenthal and Baker ' have successfully reduced a number of cyanomethyl branched-chain sugars and nucleosides using lithium aluminum hydride, rhodium on alumina or platinum oxide, the latter two methods performed catalytically under a hydrogen atmosphere of 40-6o psi. 199 200 Since catalytic hydrogenation of n i t r i l e s has been suggested ' to proceed via an imine intermediate [equation (a)] the possibility of E2 H2 R-C3N — R - C H = N H — f L - » ~ R-CH2~NH2 (a) side reactions arises between the imine and the primary amine, to afford - 72 -secondary amines [equation (b)]. RCH=NCH2R + NH3 (b) Such reactions may be minimized by the addition of reagents which w i l l essentially prevent the primary amine from further reaction reagents. Freifelder has suggested that the presence of ammonia helps to prevent some side product formation by repression of the equilibrium formation of ammonia and the secondary imine [equation (b)]. Therefore, the cyanomethyl nucleoside (158) was hydrogenated over 5% rhodium-on-alumina in ethanol saturated with ammonia, under a pressure of 60 psi of hydrogen. After twenty-four hours at room temperature a l l starting material had been consumed as evidenced by t i c . The reaction mixture was then f i l t e r e d and evaporated to a syrup which was acetylated with acetic anhydride in pyridine. This would have the effect of ful l y acetylating both the amine and the free hydroxyl groups. The acetylation product was then treated with aqueous dimethylamine in methanol, which i s known to cleave 0-acetyl groups 205 selectively. Thus after preparative t i c the N-acetylated nucleoside (170) was recovered in 39% yield. once formed. Acetic anhydride 202 and mineral acid 203 are such 204 CHC1, H-8 HOCH AcHNCH2CH2 /N H-l' 170 N-H N>^  ^ NMe, NMe, NAc u> • • 1 • • Figure VIII. 100 MHz NMR Spectrum of 2,6-Di-N,N-dimethylamino-9-(3'-C-[2"-acetamidoe'thyl]-2',3'-di-deoxy-q-D-ribo-hexopyranosyl)purine (170) in CDC1_. - 74 -The nmr spectrum of 170 clearly showed the presence of the H-8 proton on the purine residue, and a broad signal at T 3.80 was assigned to the N-H proton of the acetamido group. A pair of doublets was observed for the anomeric proton at x 4.18 with coupling constants of 11 and 3 Hz with the protons on carbon two, thereby confirming that 170 had maintained the "^C^  conformation. Sharp singlets were observed at x 6.58 and x 6.87 corresponding to the dimethylamine groups, and a singlet at x 8.05 corresponding to the N-acetyl residue. Compound 170 exhibited a positive Cotton effect in i t s cd spectrum at 290 nm, as did i t s precursor(160). Mass spectrometry afforded the molecular ion at m/e 421 and revealed the base peak of the spectrum to be m/e 206 due to [B+H]+, (C„H.,.N,). y 14 o 1.2 Synthesis of Branched-Chain 2-Deoxy-pentopyranosyl Nucleosides It was envisaged that the previously described route for the synthesis of 2-deoxy hexopyranosyl nucleosides would be applicable for the synthesis of a number of branched-chain 2-deoxy-pehtOfuranosyl sugars and nucleosides. However, due to d i f f i c u l t i e s encountered in the reaction sequence the f i n a l objective was altered to encompass the - 75 -synthesis of 2-deoxy-pyranosyl derivatives. Compounds of this type 172 have been shown to exhibit biological activity. These compounds have already been discussed in the introduction. As mentioned previously, the introduction of the cyanomethyl functionality into a sugar moiety permits later conversion of the n i t r i l e into a number of other interesting and potentially useful functional groups, such as the amino- or amido-derivatives. 1.2.1 Methyl 2-Deoxy-q,g-D-erythro-pentofuranoside (172) The f i r s t requirement of this synthetic sequence was to obtain 2-deoxy-D-ribose (171) in i t s furanoid form. A number of pathways were possible to achieve this end; (a) degradation of an appropriate 2-deoxy hexose after the formation of the hexofuranose. 207 (b) formation of the pentose dialkyl dithioacetal which may be readily p_-nitrobenzoylated at carbon five, and then converted to 208 the corresponding 5-0-acylpentofuranose. (c) controlled reaction of 2-deoxy-D-ribose i n methanol with acid to yield directly the anomeric methyl furanosides. Fortunately, the behaviour of 2-deoxy-D-erythro-pentose (171) in anhydrous methanol, containing 0.1% hydrogen chloride had been 209 studied by Overend and coworkers. It was observed that after twelve minutes at 23° the specific rotation of the solution reached a maxima. If the reaction was quenched at this time with silver carbonate the resultant syrup did not react with lead tetraacetate, thus indicating that only methyl 2-deoxy-q,g-g-erythro-pentofuranoside (172) was present. - 76 -The presence of pyranoside would be indicated by consumption of lead 210 tetraacetate. Zorbach reports a similar finding for 2-deoxy-D- > ribo-hexose (173) which also revealed the presence of only the anomers of methyl 2-deoxy-D-ribo-hexofuranoside (174) in a one to one ratio. CHO CH20H 174 173 Thus, using the straightforward method based on that of Overend, 2-deoxy-D-ribose was treated with anhydrous methanol containing 0.05% hydrogen chloride at room temperature. After fifteen minutes the reaction was stopped by the addition of excess silver carbonate. Fi l t r a t i o n and evaporation of the f i l t r a t e afforded a clear syrup which was used directly in the next step of the reaction sequence. - 7 7 -1.2.2 Methyl 2-Deoxy-5-0-trityl-a-D-erythro-pentofuranoside (175) and Methyl 2-Deoxy-5-0-trityl-3-g-erythro-pentofuranoside (176) In order to selectively block the primary position of 172, leaving the C-3 secondary hydroxy!free, and thus available for later oxidation, treatment with chlorotriphenylmethane i n pyridine was selected. This 211 reagent i s known to react almost exclusively with primary alcohols, leaving secondary alcohols free. Thus after five days at room temperature, t i c , using benzene-ethyl acetate (9:1) as solvent, indicated that a l l 172 had been consumed to afford two components which absorbed ultraviolet and charred when sprayed with sulfuric acid and heated [R^ 0.40 and R^  0.25]. After work up the viscous gum was chromatographed on s i l i c a gel to afford \ almost equal amounts of pure 175 and 176. The total yield after chromatography was 65%. The assignment of 175 as methyl 2-deoxy-ct-D-erythro-pentofuranoside, the f i r s t component eluted from the column, was based on nmr and optical rotation data, enabling comparison with 212 reported values. The 250 Hz sweep width 100 MHz nmr spectrum revealed the anomeric proton to consist of two closely spaced doublets having couplings of 0.8 Hz and 4.5 Hz with the protons on carbon two. 172 175 176 - 78 -Similarly 176, the second component eluted, was assigned the 8-configuration. Again the anomeric signal exhibited a double doublet with couplings of 2.0 Hz and 5.0 Hz. The optical rotation in chloroform of 175 was +69° whilst that of 176 was -38°. These experimental values are i n close agreement with those 212 determined by Leonard et a l . Considerable work was done by these workers to determine unambiguously the anomeric configuration of 175 and 176 by chemical synthesis and by a complete analysis of their nmr spectra. 212 Experimental Values L i t . Values . [ot]p4 J 1 ) 2,(Hz) J 1 } 2„(Hz) [ex]*8 J 1 > 2,(Hz) J 1 > 2„(Hz) 175 +69° 4.5 0.8 +64.4° 4.8 0.7 176 -38° 2.0 5.0 -43.8° 1.9 5.5 Conversion of 176 into a compound of known anomeric configuration, methyl 2,3-di-0-p-toluenesulfonyl-5-0-triphenylmethy1-g-D-ribo-furanoside (180) was accomplished by treatment of 176 with p_-bromobenzenesulfonyl chloride in pyridine to afford the p_-bromobenzenesulfonate (177), which was converted into the olefin (178)with sodium methoxide in anhydrous DMF. Osmylation followed by alkaline hydrolysis afforded the diol (L79) which would have the ribose configuration since osmium tetroxide should attack the double bond from the less hindered side. The diol was then converted into i t s crystalline ditosylate derivative which - 79 -was identical i n a l l respect to the ditosylate (180) formed from 5-()-trityl-0-D-ribofuranoside (181). OH OH OTs OTs OH OH It i s important at this stage of the reaction scheme to have unequivocal assignment of the anomeric configuration of compounds 175 and 176. 1.2.3 Methyl 2-Deoxy-5-0-trityl-a-p-glycero-pentofuranosid-3-ulose (182) and Methyl 2-deoxy-5-0-trityl-B-p-glycero-pentofuranosid 3-ulose (183) Oxidation of 175 or 176 had not been reported in the literature and hence a satisfactory method for this reaction needed to be evolved. As indicated in the Introduction, oxidation of secondary alcohols with numerous reagents have been investigated. However, possibly one of the simplest and usually high yielding reactions i s the ruthenium tetroxide oxidation. Thus in this instance compounds 175 and 176 were - 80 -treated with ruthenium tetroxide produced i n situ by the so-called "catalytic method" previously described. The reactions i n both cases were readily monitored by t i c [ s i l i c a gel, benzene-methanol (95:5)]. The ketose in each case had a higher Rf than the starting alcohol. On completion of the reaction unreacted ruthenium tetroxide was removed by addition of a small amount of isopropanol. The reaction mixture was then worked up in the usual way to afford the t i t l e ketose in 98% crystalline yield. The infrared spectrum of 182 exhibited a strong carbonyl absorption at 1755 cm 1 in the crude product and hence does not exist as the hydrate. The nmr spectrum, analyzed on a f i r s t order basis only, afforded the coupling data (observed) tabulated below. Compound 182 H-l H-2'a H-2b H-4 H-5a H-5fa H-l - 5.2 1.0 H-2 a 5.2 - 17.6 0.8 - -H-2b 1.0 17.6 H-4 - 0.8 - - 2,6 3.4 H-5 a - - 2.6 - 10.0 H-5b - - - 3.4 10.0 0 T 4.52 7.19 7.58 5.94 6.54 6.66 - 81 -H-1 H-2 a Compound H-2b 183 H-4 H-5 a H-5b H-1 - 1.8 5.5 - -H-2 a 1.8 - 18.0 - - -H-2b 5.5 18.0 - 1.2 - -H-4 - - 1.2 3.2 6.5 H-5 a - - - 3.2 - 10.2 H-5 b - - - 6.5 10.2 -T 4.71 7.56 7.32 5.84 6.62 6.80 Similar treatment of the reaction mixture from 176 afforded 183 in 96% yield as a viscous syrup which could not be induced to cry s t a l l i z e . This compound was judged homogeneous by t i c and nmr. The infrared spectrum showed the presence of a carbonyl absorption at 1760 cm ^. The nmr data i s presented above. 1.2.4 Methyl Z_-(and E)-3-£-cyanomethylene-2,3-dideoxy-ct-rj-glycero-pentofuranoside (184) and Methyl Z-(and Jp-3-C-Cyanomethylene-2,3-dideoxy-q-rj-glycero-pentofuranoside (185) Ketoses (L82)and Q.83) were u t i l i z e d directly i n the condensation with the carbanion formed from diethylcyanomethylphosphonate and sodium hydride in dimethoxyethane under anhydrous conditions. This procedure 113 was based on that of Jones and Maisey. Ketose 182 in dimethoxyethane was slowly added to a f i l t e r e d solution of the carbanion in dimethoxyethane to afford, after work up, 184 in 94% yield which crystallized on trituration with ethanol. The infrared spectrum of 184 exhibited an absorption at 2250 cm ^ and 1660 cm - 82 -corresponding to a n i t r i l e function and an olefinic bond, respectively. The olefinic proton resonated at T 4.78 in i t s nmr spectrum, super-imposed upon the H-l multiplet. Irradiation of the anomeric hydrogen collapsed the H-2 multiplet at T 7.07 to a doublet. Similarly irradiation of H-4 at T 5.27 simplified the H-5 signal at T 6.66. The methoxyl hydrogens were observed at x 6.60. 0 1 8 3 = O C H ; \< N = H 1 8 5 _ K± = O C H 3 ; R , = H Similarly, reaction of 183 under exactly the same conditions afforded 185 in 94% yield as a clear syrup. The infrared spectrum of 185 exhibited the same information as 184. The methoxylresonated at x 6.80 in i t s nmr spectrum, 20 Hz upfield from that observed for the corresponding a-anomer. Methyl Z-(and E)-3-C~cyanomethylene-2,3-dideoxy-B-g-glycero-pentofuranoside (185) was used without further characterization. Neither 184 nor 185 could be resolved by t i c into the expected 161 IS and isomers. However i t has been previously observed that pairs of geometric isomers may move as a single spot on t i c . - 83 -1.2.5 Methyl 3-C-cyanomethyl-2,3-dideoxy-5-0-trityl-ct-D-erythro-pentofuranoside (186), Methyl 3-£-cyanomethyl-2,3-dideoxy-5-O-trityl-ct-D-threo-pentof uranoside (187), and Methyl 3-C-cyanomethyl-2,3-dideoxy-5-0-trityl-3-p-threo-pentofuranoside (188) When pure 184 was hydrogenated in ethanol, containing 10% benzene to aid solubility, over 10% palladium on carbon as catalyst, at atmospheric pressure and room temperature approximately one mole-equivalent of hydrogen was taken up. Investigation of the hydrogenation reaction mixture by t i c [ s i l i c a gel, petroleum ether (30-60°)-ethyl ether (6:4)] revealed one elongated spot. Double development of a 20 cm t i c plate in petroleum ether-ethyl ether (8:2) resolved this spot into two components [R^ 0.29 (major) and R^  0.25 (minor)] which were i n i t i a l l y separated by preparative t i c . Later, using a technique which has since found wide application in this laboratory, namely that 178 of low pressure column chromatography, on t i c grade s i l i c a gel (without binder), larger quantities of 186 and 187 were separated in a ratio of approximately 7:2. These two compounds must correspond to the two modes of addition of hydrogen to the double bond of 184. In one instance from above the plane of the ring to afford an erythro- compound and secondly from below the ring giving rise to a threo- branched-chain sugar. - 84 -OCH3 T 4 5 6 7 8 Figure IX: Partial 100 MHz NMR Spectra of Methyl 3-C-cyanomethyl-2,3-dideoxy-a and g-g-threo-pentofuranoside, (187) and (188) respectively, in CDC1„. - 85 -C H 2 ° T r n CH90Tr CHo0Tr 0 ^  1 0 2 0 H 2 - Pd/c \ y CH CN ^OCH3 CH2CN 186 187 The trans nature of the 3,4 substituents i n compound 187 was shown via transformation of 187 into a pentopyranosyl derivative (described in section 1.2.6). Compound 186 was assigned the structure of methyl 3-C-cyanomethyl-2,3-dideoxy-5-0-trityl-q-D-erythro-pento-furanoside and 187 was assigned the structure of methyl 3-C-cyanomethyl-2,3-dideoxy-5-0-trityl-a-rj-threo-pentof uranoside. It i s of interest to note the stereochemical influence of the groups directing the addition of hydrogen to the double bond. Above the ring the bulky 5-£-trityl group would hinder hydrogenation from that particular direction, whilst the methoxyl group i n the a-configura-tion would tend to hinder hydrogenation from below the ring. The fact that the erythro product dominates may indicate a greater stereo-chemical influence of the anomeric methoxyl during hydrogenation than the group at the C-5 position. A similar stereochemical effect has 213 been observed for osmylation of 2,3-didehydro glycosides. In contrast, hydrogenation of the 6-anomer (185), resulted in a single crystalline product(188^ which was homogeneous by t i c and nmr. Again the configurational proof of 188 shall be discussed in section 1.2.6. It was however based on ring expansion to show that i t i s of the threo-configuration about carbon atoms 3 and 4. Thus once again i t - 86 -would appear that the stereochemical course of hydrogenation i s significantly influenced by the C-l methoxyl group. In the case of 185 hydrogenation occurs only via an underside addition of hydrogen due no doubt to the total hindrance from above the ring by both the C-l methoxyl and the 5-0-trityl groups. CH CN 185 188 The mmr spectrum of 186 indicated the presence of a small amount of impurity which could not be removed by successive chromatographic procedures. This impurity did not however conceal any of the important features of the spectrum. The anomeric proton at T 4.95 was observed as a double doublet having coupling constants with the C-2 protons of 4.5 Hz and 1.0 Hz. H-4 exhibited a pseudo quartet at T 6.09 and H-5 exhibited a multiplet centered at T 6.76. Decoupling of H-5 from H-4, by irradiation at T 6.76 reduced the H-4 multiplet to a doublet revealing ^ as 3.0 Hz. In the 100 MHz spectrum of methyl 3-C-cyanomethyl-2,3-dideoxy-a-D-threo-pentofuranoside (187) the anomeric proton at T 4.88 was seen as a double doublet with coupling constants of 2.5 Hz and 4.5 Hz to the C-2 protons. The proton on carbon four, at x 5.70, appeared as a well defined doublet of triplets which collapsed to a doublet with a spli t t i n g of 6.5 Hz when the H-5 signal was irradiated. - 87 -Similarly, i n the nmr spectrum of methyl 3-C-cyanomethyl-2,3-dideoxy-8-D-threo-pentofuranoside (188) the anomeric proton was observed at T 5.02 as a doublet of doublets with coupling constants of 2.0 Hz and 5.0 Hz with the protons on C-2. H-4 appeared at T 5.74 as a doublet of tri p l e t s . Unfortunately, H-5' and H-5" were not well situated with respect to one another to permit irradiation of either both together or each one singly. However, the H-5 protons did form an ABM system which was amenable to approximate analysis. From this H-51 and H-5" were found to be coupled to H-4 by 7.0 Hz and 5.0 Hz. When this data was fi t t e d to the doublet of triplets for H-4, a value of 5.0 Hz could be derived for the coupling of H-3 with H-4. Configurational assignment of 186, 187 and 188 was purposely not made on nmr data due to the well documented inconsistencies observed in coupling constants for furanose rings. In fact a systematic and 214 complete survey by Hall and Slessor on l,2:5,6-di-0_-isopropylidene-D-hexoses revealed that adjacent protons having a cis orientation in a furanose ring may have coupling constants between 3.07 Hz and 8.45 Hz, whilst those having a trans relationship may range between < 0.5-10.62 Hz depending upon the conformational constraints placed upon the ring. 1.2.6 Ring Expansion of 187 and 188 to Afford Methyl 3-C-cyano-methyl-2,3-dideoxy-ct,B-g-threo-pentopyranoside (189) and i t s Benzoyl Derivatives (190) and 191) To prove unambiguously that f uranosides (187) and (188) were of the threo configuration they were reacted with anhydrous methanol containing - 88 -1% hydrogen chloride for several hours at room temperature. Work up afforded clear syrups, in both cases, having identical nmr spectra, each showing identical anomeric mixtures of pyranosides. The ratio of a to g was 10:3 found by integration of the H-l proton. Anomeric assignment was based on the coupling constants with the hydrogens on carbon two. For the a-anomer couplings were 1 Hz and 3 Hz, while for the g-anomer they were 7 Hz and 2 Hz. Assignment of the configuration at carbon three could be tentatively made by analysis of the sixteen line AMBX system exhibited by the C-2 methylene protons. Irradiation of H-l collapsed this multiplet into a well-resolved eight line system which revealed the coupling constants J„ , = 12.5 Hz and J„ = 4 Hz. Thus indicating ^a,j ze,3 that H-3 must be in an axial position, with the cyanomethyl branched-chain equatorial, implying a threo-configuration for 187 and 188. The protons on carbons three and four were not amenable to analysis due to complexity in the case of the signals of H-3 and to the overlying H-5 proton signals i n the case of H-4. - 89 -To gain further evidence of ^ the £-chloro-(and j>-bromo) benzoates (190) and (191) of 189 were formed. The benzoate- ester group had the effect of shifting the H-4 signal to lower f i e l d and clear of H-5. Irradiation at T 5.22 (H-l) simplified H-2e (T 7.85) and H-2a (x 8.28) and irradiation at x 5.04 (H-4) affected the multiplet at x 6.3 (H-5). Analysis of the H-l and H-2 systems afforded the following data: J- ^ = 3.5 Hz, J. . = 1.5 Hz, J . , = 12 Hz, J„ = l,<ca 1,/e Za,i /e,J 5 Hz and J_ _ =13 Hz. The chemical shift difference between H-2a Za,Ze and H-2e was 40 Hz. From inspection of the doublet of triplets at x 5.04 (K-4) and consideration of the coupling of H-4 with the two C-5 protons, ^ = 10.5 Hz and ^ = 5.5 Hz i t was deduced that the coupling constant between H-3 and H-4 was approximately 10.5 Hz, again implying a trans orientation of the branched-chain with respect to the C-4 hydroxyl. When 186 was treated with methanol containing hydrogen chloride under similar conditions to that of 187 and 188 a considerably more complex mixture resulted. Thin layer chromatography revealed at least two components, however, the nmr spectrum showed that at least three methyl glycosides were present as evidenced by the three singlet peaks at x 6.6. The anomeric region was quite complex. Attempts to purify this mixture by treatment with p_-chlorobenzoyl chloride followed by preparative t i c were unsuccessful since each band isolated was shown to contain more than one component as evidenced by nmr. Thus i t was not possible to afford an independent proof of structure for the erythro compound (186). - 90 -1.2.7 Attempted Selective Removal of the 5-0-Triphenylmethyl Group After compounds 186, 187 and 188 were obtained methods for removing the 5-0-trityl group were investigated. Model studies were carried out on methyl 2-deoxy5-0-trityl-o; ,B-D-erythro-pentofuranosides (175) and (176). It had been noted in the literature that generally the most successful way to remove this blocking group was with 80% acetic acid, warmed on a steam bath for several minutes. Leonard and 212 workers had successfully used this procedure on methyl 3-£-bromo-benzoyl-2-deoxy-5-0_-triphenylmethyl-a-D-ribofuranoside (192). Similarly 215 Fletcher had detritylated 193 with 40% acetic acid solution in 216 dioxane in water. Shen et a l . also found that the 5-0-trityl group may be removed from certain 2',3'-didehydro-2',3'-dideoxy-uridines (193a) with warm 80% acetic acid. There are numerous other examples of acid 0 192 193 193a hydrolysis of triphenylmethyl groups exhibiting various degrees of success. Therefore i t was with some concern, that the entirely negative results of the attempted acid hydrolyses using acetic acid In concentrations ranging between 33-80%, in a number of solvents including - 91 -water, dioxane and methanol, at various temperatures and reaction times, were viewed. Tic indicated that when reaction did take place 172 was not the product, as would be expected on selective detritylation of 175 and 176, but instead a compound of lower was obtained. In the hope that 175 and 176 were not in fact sufficiently good model compounds a sample of 188 was treated with 80% acetic acid at 70° for ten minutes, then allowed to cool for thirty minutes. Nmr investigation of the sample isolated revealed the absence of both the 5-0-trityl and the methoxyl groups. The anomeric region indicated the probable presence of a- and S-pyranoses. Three other methods of acid hydrolysis were investigated without success. A solution of 175 and 176 was treated with methanol containing 0.15% hydrogen chloride un t i l a l l starting material had been consumed. After work up of the reaction mixture nmr showed the presenc e of at least three different anomeric protons. Treatment of 175 and 176 in chloroform with a trace of 90% trifluoroacetic acid cleaved both the 5-0-trityl and the methyl glycoside functionalities. Similarly treatment of 175 and 176 in acetic anhydride with BF^ in ether afforded a product which did not contain a methoxyl or t r i t y l group. 218 Related to acid hydrolysis of 5-(D-trityl groups i s the report that s i l i c a gel may catalyze detritylation of some carbohydrate derivatives. However, a l l compounds in this work were chromatographed on s i l i c a gel without detritylation. In addition to acid hydrolysis hydrogenation in ethanol in the presence of palladium on charcoal has been occasionally used to remove 219 220 t r i t y l groups. It has been noted by a number of workers, 221 222 including Todd and Khorana that this method i s , however, frequently 223 extremely slow and unsatisfactory. Nonetheless hydrogenolysis was attempted on the model substrates (175) and(176) and while no reaction was observed for reactions performed at atmospheric pressure and elevated temperature, a small amount of product (172)was isolated after several days 60 psi in a Parr hydrogenator using ethanol containing a trace of acetic acid. In addition, another product having a higher than the starting material, was formed. Based on these results and the conclusions of others, i t was decided that this method was of l i t t l e value in the solution of this problem, and hence was not further investigated. 224 Detritylation by reductive means has also been reported. 2 2 A Using lithium in liquid ammonia at -70°. Inouye et a l . found that 3-0_-benzyl-6-0-triphenylmethyl-l,2-0-isopropylidene-5-amino-5-deoxy-oi-D-r gluco-furanose was detritylated by this procedure. Using this method the model compounds (175) and (176) were readily unblocked to afford 172 as evidenced by t i c . Similarly when 188 was reacted under the same conditions t i c indicated that the t r i t y l group was removed but the formation of at least six products was observed. An infrared spectrum of the crude product showed that the n i t r i l e was no longer present and had been reduced to an amine (indicated by an absorption at 3400 cm "'"). It was expected that in addition to detritylation side 225 reactions would occur. Van Tamelen et a l . report that amongst amines and imines formed when aliphatic n i t r i l e s react with lithium in liquid ammonia alkanes may result. For example, methyl cyanide afforded methane in 98% yield. - 93 -Acetylation of the crude product was shown to afford at least seven products by t i c [ s i l i c a gel, ethyl acetate]. Preparative t i c afforded the major carbohydrate component, an acetylated methyl glycoside. The infrared spectrum of this compound revealed two carbonyl absorptions, one at 1750 cm 1 (OAc) and the other at 1670 cm 1 (NAc). Nmr indicated the presence of acetyl groups at x 7.90. The f e a s i b i l i t y of using the bis(j>-methoxyphenyl)phenylmethyl 222a group in place of the t r i t y l blocking group was investigated. When chloro-bis(p_-methoxyphenyl)phenylmethane was reacted with 172 in pyridine two blocked carbohydrates were observed to be formed by t i c . Work up of the reaction in the usual way afforded a yellow syrup which when applied to a column of s i l i c a gel became strongly orange-colored. Continued elution of the column afforded an almost quantitative amount of bis (p_-methoxyphenyl)phenyl carbinol, indicating that unblocking had occurred with great f a c i l i t y on the s i l i c a gel (cf. ref. 218). Attempts to find a chromatographic medium to achieve separation of the blocked sugars were not successful. Neutral alumina and A l u s i l [ s i l i c a gel-alumina (1:1)] were unsatisfactory, giving no separation on the former medium and continued unblocking on the latter. In a last attempt to u t i l i z e 186, 187 or 188 in their furanoid form direct fusion of 188 with 2,6-dichloropurine was attempted. The only compound isolated in amounts amenable to charazterization was triphenylmethyl-2,6-dichloropurine (194). - 94 -Cl 194 Since 2',3'-dideoxypyranosyl nucleosides are of biological 193 194 interest ' in their own right i t was judged reasonable to use 187 and 188 in the synthesis of such compounds through the now characterized compound (190 ). 1.2.8 Synthesis of 2-Chloro-6-N,N-dimethylamino-9-(3'-C-cyanomethyl-2',3'-dideoxy-g-(and a-)-g-threo-pentopyranosyl)purine (198) and (199) A number of possible routes were available for the synthesis of nucleosides from methyl 4-0^p_-bromobenzoyl-3-C-cyanomethyl-2,3-dideoxy-ot-D-threo-pentopyranoside (190). These have been discussed in the Introduction. Direct fusion of the branched-chain methyl glycoside at 145°, with 2,6-dichloropurine, in the absence of acid catalyst, ur^der reduced pressure, for three hours, resulted in the formation of three products as evidenced by t i c . Column chromatography of the resultant - 95 -glass, using t i c grade s i l i c a gel under pressure afforded the compounds 197 , 195 and 196 in order of elution. Compound 197, was obtained as a viscous syrup in 19% yield which exhibited the presence of the n i t r i l e function at 2300 cm ^ in i t s infrared spectrum. Nmr analysis did not show the presence of an H-8 proton that would be required for a nucleoside. Further nmr evidence, a positive reaction with potassium permanganate spray, chemical analysis and mass spectrometry proved 197 to be A-£-p_-bromobenzoyl-3-£-cyanomethyl-2,3-dideoxy-D-erythro-pent-l-enopyranose. Cl The f i r s t nucleoside to be eluted was shown to be 2,6-dichloro-9-(4' -0-£-bromobenzoyl-3' -C-cyanomethyl-2*, 3' -dideoxy-3-D-threo-pento-pyranosyl)purine (195) in 29% yield. A sharp singlet at T 1.66 (H-8) revealed the presence of the substituted purine moiety. The anomeric proton at x 4.05 was observed as a double doublet with coupling constants - 96 -to H-2a' and H-2e' of 10 and 2 Hz respectively thus allowing the $-anomeric configuration to be assigned. The fact was later substantiated by the circular dichroism spectrum of the unblocked nucleoside. The a-anomer (196) was isolated in 13% yield, i t s assignment also being based on the fact that the proton at C-l' showed a t r i p l e t at x 3.92 revealing an equal coupling to H-2a' and H-2e* of 4 Hz. Treatment of 195, dissolved i n methanol, with 25% aqueous dimethylamine resulted in the expected replacement of the 6-chloro group 193 by a dimethylamino functionality and removal of the hydroxyl protecting ester group. The crystalline unblocked nucleoside (198) was isolated in 76% yield after column chromatography. Similarly, reaction of the blocked a-nucleoside (196) under the same conditions afforded 2-chloro-6-N,N-dimethylamino-9-(3'-C-cyanomethyl-2',3'-dideoxy-q-D-threo-pentopyranosyl)purine (199). Investigation of these two unblocked nucleosides by uv, nmr, cd, i r , and mass spectrometry afforded the following data: (i) The ultraviolet absorption (A 274 nm) indicated that the max 144 position of attachment of the base to the sugar was at N-9, for both 198 and 199. ( i i ) A negative Cotton effect was observed in the cd spectrum of 145 147 198, indicating that i t has the 6-anomeric configuration. ' Conversely, a positive Cotton effect was exhibited by 199, showing that this compound was most probably the a-nucleoside. ( i i i ) 3-Nucleoside (198) showed the presence of a n i t r i l e absorption at 2260 cm ^ in i t s infrared spectrum. - 97 -(iv) The nmr spectra of 198 and 199 showed the presence of the H-8 proton on the purine nucleus as a singlet at T 2.08 and x 2.13 respectively. Both compounds showed a broad singlet at x 6.5 corres-ponding to the 6-N,N-dimethylamino group. The anomeric proton, in the case of 198, was observed as a double doublet at x 4.17 with coupling constants of 2.5 Hz and 10.0 Hz indicating that the anomeric proton i s trans-diaxial with one of the H-2' protons. This can be achieved i f 198 i s in the B-configuration. Similarly, the anomeric proton of 199 was observed as a t r i p l e t at x 4.06 with couplings to both H-2' protons of 4 Hz. Therefore 199 must be the a-nucleoside. The nmr spectrum of 199 was obtained on 2 mg of sample by using the Fourier Transform technique (200 transients). (v) Mass spectrometry of both 198 and 199 gave molecular ions at m/e 336 and m/e 338. High resolution mass measurement of the molecular ions of 199 showed conclusively that i t was of the molecular formula C. .H..-,JNL0„C1 as expected. 14 17 6 2 (vi) Elemental analysis of 198 was correct for the structure suggested. Compound (199) was not submitted for elemental analysis due to the small quantity obtained but was characterized by high resolution mass measurement. Thus on the data summarized above i t is safe to conclude that compounds (198) and (199) were isolated and have the suggested structures. - 98 -1.2.9 2-Chloro-6-N,N-dimethylamino-9-[3'-(2"-acetamidoethy1)-2',3'-dideoxy-g-D-threo-pentopyranosyl]purine (200) Reduction of the cyano-nucleoside (198) by catalytic hydrogenation at room temperature and 60 psi over platinum oxide in ethanol, containing 10% acetic anhydride consumed a l l the starting nucleoside in four hours. Fil t r a t i o n and evaporation afforded a syrup which showed the presence of two components [R,. 0.20 and R,. 0.15, on s i l i c a - 99 -gel using chloroform-ethanol (9:1) as developer]. The nmr spectrum of this material revealed at least three acetoxyl groups at T 8.0, hence indicating pa r t i a l acetylation of the secondary hydroxyl in addition to N-acetylation of the reduced n i t r i l e . 198 200 It i s interesting to note also that the 2-chloro group on the purine 226 moiety was not hydrogenolyzed under these conditions. Robins et a l . report that in some cases the 2-chloro functionality may be removed by hydrogenation at 60 psi over 5% palladium on carbon after several hours. Treatment of the crude syrup obtained after hydrogenation with 25% 13 3 13 A aqueous dimethylamine ' selectively unblocked the 0-acetyl groups and the product exhibited only one spot on t i c . Nuclear magnetic resonance shewed only one acetyl group at T 8.00, and a broad signal at x 4.05 which was assigned to the N-H proton. The anomeric proton was observed as a double doublet with the same values as 198. - 100 -2. Glycos-3-yl g-Hydroxy Esters T g-Amlno Esters, and q-Amino Acids:  Structural Analogs of the Sugar Moiety of the Polyoxins. In this work our prime objective was to synthesize analogs of the sugar moiety of the polyoxins. As mentioned in the Introduction the principal structural features of a l l the polyoxins include the , „ , 149-159,227,228 following: (i) possession of a 5-amino-5-deoxy-D-allo-furanosyluronic acid sugar moiety. ( i i ) a unique L-amino acid residue attached to C-4 of the ribo-furanosyl ring. In view of the fact that the introduction of branching at C-3 on the sugar moiety of naturally occurring nucleosides has been found to result in interesting changes in the biological activity of the 126 nucleosides, i t seemed of interest to synthesize structural analogs of the sugar moiety of the polyoxins in which the amino acid moiety would be attached to carbon three, rather than carbon four, of the sugar. In addition g-hydroxy-a-amino acids have been the subject of 229 considerable synthetic endeavour thus the ready synthesis to be described in the following work was an additional feature derived from this study. 2.1.1 1,2:5,6-Di-Q-isopropylidene-ot-g-ribo-hexofuranos-3-ulose (9) Condensation of acetone with D-glucose in the presence of freshly 230 prepared zinc chloride resulted in 1,2:5,6-dir-0-isopropylidene-a-D-gluco-furanose (36). Oxidation of the secondary hydroxy1 of 36 was 66 77 readily achieved via the "catalytic" ruthenium tetroxide ' method to - 101 -afford the ketose hydrate (203) which was dehydrated to 9_ by azeotroping with toluene. 2.1.2 Wittig Reaction of 1,2:5,6-Di-0-isopropylidene-q-D-ribo-hexofuranos-3-ulose (9) with Carbomethoxymethyldimethylphosphonate __ Using the procedure of Rosenthal and Nguyen ketose 9_ was condensed with the carbanion formed from carbomethoxymethyldimethyl-phosphonate and potassium t-butoxide i n dimethyl sulfoxide. After work up OH / 36 .0 two spots (A and B) were visible on t i c , however, on isolation A was shown to consist of two components, by nmr, which on hydrogenation 129 afforded a single product (12). This product was assigned the allo-configuration based on the coupling constant between H-2 and H-3. The observed value of 4.0 Hz was similar to the coupling constant between H-1 and H-2 (3.7 Hz) and is of the size often encountered for 2,3-cis 214 hydrogens in a hexofuranosyl system as shown by Hall and coworkers. 161 The second spot (B), was isolated by Rosenthal and Shudo in 6% yield. This also consisted of two components as shown by nmr which when hydrogenated afforded a single crystalline branched-chain sugar (206). The ribo-unsaturated sugars (10) and (11) were readily separated - 103 -from the epimeric mixture of unsaturated sugars (204) and (205) by column chromatography on s i l i c a gel. Fractional crystallization of the mixture of Z- and ^-unsaturated sugars (10) and (11) from n-hexane afforded pure crystalline Z-3-C-(methoxycarbonylmethylene)-3-deoxy-1,2;5,6-di-O-isopropylidene-a-D-ribo-hexofuranose (10). Similarly the E-xylo unsaturated sugar (205) was readily obtained in pure crystalline form by recrystallization of the mixture of 204 and 205 231 from n-hexane. The structure of 205 was assigned on the basis of an analysis of i t s nmr spectrum and on mechanistic considerations (discussed later). In the E-isomer (205), H-4 i s under the deshielding 232 influence of the ester function and thus exhibits a pair of doublets at lower f i e l d (x 4.60) than the Z-isomer (204) (x 5.32 having ^ = 1.8 Hz and J. _ = 6.0 Hz). Furthermore, in the Z-isomer, H-2 and the ester function are cis with respect to each other with a consequent lowering of the x value of H-2 to x 3.6 (J^ ^  ~ 3.6 Hz and J^, ^ = 1.2 Hz) in contrast to the much higher x value of H-2 in the E-isomer (x 5.15, J n „ = 4.0 Hz and J , = 0.8 Hz). 10 11 204 205 H-2 4.23 4.9 3.60 5.15 H-4 5.32 4.2 5.32 4.60 Similarly a careful analysis of 10_ and 11 revealed a similar deshielding effect by the carbomethoxy group on H-2 and H-4. - 104 -A double irradiation experiment was performed on pure 205 which showed that when H-1 was irradiated, the signal at T 5.15 collapsed to a narrow doublet indicating that i t was H-2 and was further coupled through to H-11 by a value of 0.8 Hz. When H-4 was irradiated at x 4.60, the signal assigned to H-1' collapsed to a narrow doublet, and the quartet (H-5) at x 5.42 collapsed to a t r i p l e t . The chemical shift assignments were made for the minor isomer (204) from the spectrum of a sample containing both 204 and 205. Fortunately the resonances assigned to 204 were not obscured by those of 205. Assignment was made by the usual process of f i r s t l y determining the position of H-1 and i t s coupling with H-2, then determining the position of H-2. This was done until an internally consistent set of coupling constants was reached, hence affording the coupling constants tabulated below. (204) (205) J l , 2 J2,4 J 2 , r J4.5 Xv 3.7 0 1.3 6.5 1.6 4.0 0 0.5 6.5 2.0 It i s interesting to note the relative amounts of and E stereoisomers, (10) and (11), in the case of the pair of ribo-unsaturated sugars and 204 and 205 in the case of the xylo-unsaturated sugars. 129 Rosenthal and Nguyen observed that the ribo sugar gave a value of 3:1 for the Z to E (10 and 11 respectively) ratio. Using the integral of the 100 MHz spectrum containing both xylo-unsaturated sugars (204) - 105 -and (205), a Z to JE ratio of 1:5 was determined thus indicating an inversion in the predominant stereoisomer in going from the ribo- to xylo-systems. Mechanistically this would seem reasonable since the formation of 11 (E-ribo-) would, on steric grounds, be less favoured than the formation of 10 (Z-ribo-). However i f inversion or enolization at C-4 occurred before, or concomitant with, condensation, as postulated 233 by Tronchet and Bourgeois, to afford the xylo-configuration, i n which the 5,6-"tail" i s down, an increase in the E-stereoisomer might be expected. This is observed and in fact 205 (E-xylo-) becomes the predominant product, thus indicating that the modified Wittig reaction products may be dependent in part upon the stereochemistry of the carbohydrate ketose. This fact has often been observed in non-113 carbohydrate applications of the Wittig reaction to assymetric ketones and must be simply a result of the more stable betaine (206) or (207) being formed. The structures represented below assume attack of the R, COoMe P 0 ( 0 M e ) 2 PO(OMe) 2 206 ribo: R^ =H; R2 = a 207 ribo: = H; R =• a xylo: R^  = a; R2 = H xylo: R^  = a; R2 = H a = 1,2-isopropylidene glycolyl of the phosphonate from above the plane of the ring as is observed in the hydrogenation of the subsequently formed olefins, ^ 2^» a n ( j as i s 234 believed to occur in the Grignard condensation with the cyclohexylidene analog of ketose 9_. - 106 -233 Bourgeois observed that on increasing the amount of ylide present, the extent of inversion at C-4 increased thus implying some dependence on the molar quantity of y l i d . In our work i t has also been observed that the amount of inverted product i s variable (between 0 and 12%) and may be correlated with our practice of using an 2 approximately 10% excess of Wittig reagent over ketose. A postulated mechanism for this inversion i s seen below. However, Bourgeois found that when ketose (210) was reacted with a Wittig reagent no inversion was observed at C-4. This he , suggested was due to the unfavourable requirement of refurnishing a proton from the underside of the intermediate 209. This process would be inhibited to a large extent by the steric hindrance afforded by the isopropylidene groups. In our work, however, this process was not examined. - 107 -Thus, i f the above mechanism i s accepted, the general scheme of the Wittig reaction becomes: ylide + ketose (9) — r i b o - b e t a i n e — 1 0 -I- 11 \ intermediate (209) w ylide + ketose (210) — x y l o - b e t a i n e — ^ 204 + 205 - 108 -2.2 Synthesis of D-2 and L-2-(1,2:5,6-Di-0-isopropylidene-a-D-gluco-furanos-3-yl)glycine 2.2.1 Potassium Permanganate Oxidation of the Unsaturated Esters (10) and (11) to Yield 3-C-[S- and R-Hydroxy(methoxycarbonyl)-methyl]-l,2:5,6-di-O-isopropylidene-cx-D-glucof uranose (211) and (212) respectively As mentioned previously, dihydroxylation of olefins has generally been carried out by the use of osmium tetroxide, osmic acid and hydrogen peroxide, or more economically with potassium permanganate in pyridine. These reagents are known to proceed stereospecifically by attack at the less hindered face of the olefin to afford the cis .. . 235,236 diol. Thus hydroxylation of the pure crystalline Z-unsaturated sugar (10) with either osmium tetroxide in pyridine, oxmium tetroxide in 30% hydrogen peroxide, or potassium permanganate in pyridine afforded only a single diol, 3-C-[S^hydroxy(methoxycarbDnyl)methyl]-l,2:5,6-di-0-isopropylidene-a-D-glucofuranose (211). A p r i o r i , i t could therefore be assumed that a similar reaction applied to the E-unsaturated sugar (11) would afford the diastereoisomeric diol (212). Because of the great d i f f i c u l t y of obtaining pure 1_1 i t was decided to use the mixture of unsaturated sugars (10) and (11) in the dihydroxylation step in the hope that the resultant diols could be separated. This in fact proved to be the case and oxidation of the mixture of 10_ and 1_1 afforded the mixture of diols (211) and (212) in the same ratio as that of the unsaturated sugars. The two diols were readily separated by column chromatography on t i c grade s i l i c a gel using benzene-ethyl acetate as developer under a pressure of 8 psi to afford pure 211 and 212 in a combined yield of 44%. The overall yield of this reaction was greatly - 109 -X,] C0 2Me 10 (Z) H^^co 2 Me III H--0H °~f"~ C02Me 211 ( S ) C02Me _j ft f 11 (E) KMnO^-pyr 0-% CO„Me H ' f 212 (R) influenced by temperature with a subsequent decrease in the yield of diols i f the temperature of the reaction mixture was permitted to rise over -5°. In addition, prolonged s t i r r i n g with excess potassium permanganate also afforded lower yields of diol presumably due to oxidative cleavage. Thus the reaction was always carried out in an ice-salt bath with slow, dropwise addition of permanganate. The reaction was - 110 -extremely rapid and exothermic, and could be readily followed by t i c . Compounds 211 and 212, isolated from the oxidation reaction both showed that the double bond was no longer present, evidenced by disappearance of the olefinic absorption in the infrared (1650 cm and lack of decolourization with potassium permanganate spray as expected. The infrared now exhibited a new strong absorption at 3480 cm * indicative of the presence of hydroxyls in addition to the s t i l l present carbonyl absorption at 1740 cm \ Nuclear magnetic resonance spectra of 211 and 212 were also consistent with the structures proposed revealing in each case two signals attributable to hydroxyl protons since each were readily exchanged with deuterium oxide. Molecular d i s t i l l a t i o n at 105° and 0.1 mm pressure gave analytical samples which afforded correct elemental analyses. The structures of diols 211 and 212 were assigned primarily on the mechanistic considerations described above; stereospecific c i s -dihydroxylation from the less-hindered face G f the molecule. Confirmation of the structural assignment was provided by the fact that the circular dichroism spectra (Figure X) of 211 and 212 exhibited opposite Cotton effects at approximately 210 nm. Based on the mechanistic argument above, oxidation of the jS-unsaturated ester (10) should afford an a,3-dihydroxy ester in which the configuration at the a-position would be the same as that in L-lactic acid, and so 211 should, and did, exhibit a positive Cotton effect i f the asymmetric centre adjacent to the ester chromophore determines the sign of the Cotton effect at the ultraviolet maxima of the ester function. Similarly the diastereomeric diol (212) derived from 11, exhibited a negative Cotton effect at the - I l l -same wavelength. Thus diol (211) i s suggested to be 3-C/-[S^hydroxy-(methoxycarbonyl)methyl]-l,2:5,6-dl-0-isopropylidene-a-D-glucofuranose and d i o l 212 must be 3-C-[R-hydroxy(methoxycarbonyl)methyl-l,2:5,6-di-0 isopropylidene-a-g-glucofuranose. Figure X: Circular Dichroism Spectra of Diols 211 and 212. In addition, from the products arising from the hydroxylation of the unsaturated sugars there was isolated two additional products in 19 and 3% yield, respectively. The f i r s t proved to be identical with the ketose hydrate (202) and therefore i t arose by an over oxidation of the diols. Prolonged reaction of the unsaturated sugars with per-manganate increased the yield of 202. The minor component (213) 161 i n i t i a l l y isolated by Rosenthal and Shudo, exhibited two carbonyl - 112 --1 peaks at 1720 and 1740 cm in i t s infrared spectrum, thus indicating an ct-keto grouping. The presence of the ketone group was confirmed by the fact that sodium borohydride reduction of 213 yielded the diols (211) and (212), and in addition, treatment of 213 with hydroxylamine afforded a crystalline oxime, namely 3-C-(methoxy-dicarbonyl)-1,2:5,6-dl-O-isopropylidene-a-D-glucofuranose oxime (214) 2.2.2 Selective Mono-mesylation of Diols 211 and 212 The selective formation of a sulfonate ester of a secondary 237 alcohol over a tertiary alcohol has been noted in the literature, and the formation of a secondary sulfonate being likewise slower than 237 that of a primary sulfonate. Thus when methanesulfonyl chloride in pyridine was reacted with either 211 or 212, the exocyclic secondary hydroxy1 was selectively blocked forming 3-C-[S^methane-sulfonyloxy(methoxycarbonyl)methyl]-1,2:5,6-di-O-isopropylidene-ct-D-glucofuranose (215) and 3-C-[R-methanesulfonyloxy(methoxycarbonyl)methyl]-1,2:5,6-di-0-isopropylidene-a-D-glucofuranose (216) in 74 and 80% yields respectively. C C y f e 211 R± = H; R2 = OH CO^ Me 212 Rx = OH; R2 = H - 113 -That mesylation was in fact selective to afford 216 was shown by nmr which revealed the presence of one methanesulfonyl group resona-ting at T 6.85 as a three proton singlet. Equally importantly a singlet at T 6.22 was observed which exchanged with D20 and hence must be the remaining hydroxyl. Similarly 215 exhibited a methane-sulfonyl at x 6.77 and a hydroxyl singlet at x 6.54. The presence of the tertiary hydroxyl was further confirmed by infrared (3400 cm and elemental analysis. 2.2.3 Displacements of the Methanesulfonate Ester with Azide There are numerous examples in the literature of stereospecific 238 158 159 S^ 2 displacement reactions of mesyl groups by azide ' ' and so i t was with some surprise that when the displacement reaction was performed, in dimethyl formamide at 55-60° in the dark for 40 hours, ' and the resultant azido sugar (219) reduced with hydrogen, two ninhydrin positive components were observed. These two compounds were subsequently shown to be the two 8-hydroxy-cx-amino ester (218) and (217). The ratio of 218 to 217 appeared to be independent of which sulfonate ester (215 or 216) was used. The ratio of 218 to 217 was approximately of the order 1:3 and was the same for reactions times of 18 to 82 hours as determined by nmr using the distinctive methyl ester peak integrals as a measure. In a typical reaction of 40 hours duration, 218 and 217 were obtained in 17 and 51% yields, respectively, based on mesylate consumed. The reaction did not proceed to completion even after 120 hours under the above conditions. The unreacted mesylate was readily separated from the two a-amino esters by column chromatography on t i c grade s i l i c a gel under pressure. - 1 1 4 -The displacement reaction must be done under s t r i c t l y anhydrous conditions to prevent reversion of the pure sulfonate ester to a mixture of diols (211) and (212). In one attempt to isolate the intermediate azido sugar (219) by column chromatography total decomposition of 219 to 211 and 212 was observed. Both diols were isolated and proved identical with pure 211 and 212. However, samples - 115 -of the displacement reaction mixture before subsequent hydrogenation do indicate the presence of azide by a strong absorption at 2150 cm ^ in the infrared. This was presumably not due to inorganic azide as the reaction mixture had been extracted into dichloromethane and fil t e r e d to remove sodium mesylate and sodium azide. Hydrogenation of this crude azido sugar in anhydrous benzene over 5% palladium on carbon was complete in 1.25 hours and afforded the two 8-hydroxy-a-amino esters (218) and (217) as mentioned above. When the hydrogenation was performed in methanol, ethanol, or ethyl acetate a third product was observed to form having a very low in ethyl acetate and was assumed to be due to hydrolysis of the methyl ester. That 218 and 217 were methyl-D-2-(l,2:5,6-di-0-isopropylidene-a-D-glucofuranos-3-yl)glycinate and methyl-L-2-(1,2:5,6-di-O-isopropylidene-a-D-glucofuranos-3-yl)glycinate respectively, was shown by nmr, i r , cd, and elemental analysis. F i r s t l y , however, ninhydrin spray indicated the presence of amino functions in both 218 and 217. This was further confirmed by nmr which revealed that the methanesulfonyl ester was no longer present and that there were now three protons which readily exchanged with D20 in both compounds. The major isomer (217) exhibited a strong positive Cotton effect at 207 nm in i t s circular dichroism spectrum whilst the minor isomer (218) exhibited a negative Cotton effect at a similar wavelength (see Figure XI). Thus 218 and 217 are suggested to be diastereoisomers possibly having opposite chirality at the a-position to the ester function. This postulate is further confirmed by the cd spectra of the free amino acids to be discussed later. - 116 -It i s interesting to note the apparent lack of s p e c i f i c i t y i n the formation of the ct-amino esters from the sulfonates. This phenomenon was originally thought to be due to the fact that each sulfonate ester was possibly undergoing conversion to the a-azido esters by two competing reactions: (i) f i r s t l y , in which the adjacent tertiary hydroxyl on C-3 238 participates and ( i i ) secondly, a reaction involving direct displacement of the sulfonate ester by the azide ion. In the f i r s t process the hydroxyl - 117 -238 C-3 would displace, via an intramolecular mechanism, the sulfonate group, to afford an epoxide. This epoxide may subsequently be attacked 229 239 at the exocyclic position ' to afford the a-azido ester having the same stereochemistry as the starting d i o l . Thus, the a-amino ester (217) or (218) produced on reduction of the azides via the f i r s t mechanism would have the same stereochemistry as the starting diol (211) or (212). Iri the second process the sulfonate i s displaced by the usual S^ 2 mechanism and thus the amino ester might be expected to have the opposite configuration to that of the starting d i o l . As a consequence each sulfonate might be expected to afford a diastereo-isomeric mixture of products. Support for this postulation was provided by the fact that the corresponding 3-deoxy tosylate analog of 216 was converted exclusively into an a-amino ester having the opposite configura-tion to the starting compound.However, as shall be discussed i n section 2.4.1, the 3-deoxy mesylate of 215 (compound 240) was not converted exclusively into a single a-amino ester but afforded a diastereomeric mixture. Thus, i t would appear that whilst the above mechanism cannot be entirely discounted for the 3-hydroxy-a-sulfonates, there may be a further mechanism affording the diastereoisomeric mixtures. The neighboring-group participation of carboxamides in the nucleo-ph i l i c displacement of sulfonate groups bound to a v i c i n a l carbon is an , .. *. . _ 238,240a „.,. . , 240a . extensively studied reaction. Mijkovic et a l . have reported the isolation of the oxazoline derivative (b) strongly suggesting the displacement of sulfonate proceeds with participation of the carbonyl of the benzamido compound (a). Similarly a recent r e p o r t 2 4 ^ has described the participation of the N,N-dialkyl carboxamido group with - 118 -the possible formation of an imino-a-lactone (d) or an ct-lactam (e) ,. 240b intermediate. r ^ HN y OR 1 (a) R' = Ts, Ms (b) R = alkyl, aryl R = alkyl, aryl R R 0 OTs R 2 N F (c) R = alkyl (d) (e) Thus i t is possible to envisage another process whereby the carbonyl of the ester (f) may participate in a similar manner to form the epoxide derivative (g), which could subsequently be attacked by azide ion, to form the azido compound in which the sulfonate has been replaced with retention of configuration. Thus i f such a reaction was to occur, in competition with the expected S N 2 displacement reaction, hydrogenation of the resulting mixture of cx-azido esters would afford both ot-amino esters. - 119 -When the cx-amino ester (217) was carefully hydrolyzed with 1.25% aqueous methanolic sodium hydroxide then neutralized by passage through a weakly acidic anion exchange resin, followed by evaporation of the solvent, a crystalline amino acid (220) was afforded in 85% yield. The cd spectrum of 220 also exhibited a positive peak at 212 nm in 0.5 M HC1 in 95% ethanol, in agreement with the positive Cotton effects 241 of other L-amino acids and i s , therefore, suggested to be L-2-(l,2:5,6-di-0-isopropylidene-a-g-glucofuranos-3-yl)glycine. Similarly, hydrolysis of the cx-amino ester (218), yielded the crystalline cx-amino acid (221). Since both 218 and 221 exhibited strong negative Cotton effects (see Figure XI) both are suggested to possess the same stereochemistry at the chiral exocyclic carbon and 221 i s suggested to be D-2-(1,2:5,6-di-0-isopropylidene-cx-D-glucofuranos-3-yl)glycine. The cx-keto ester (213), briefly mentioned as a component of the oxidation mixture of both the oxidation of pure ^-unsaturated sugar (10) or of the oxidation of a mixture of or E-unsaturated sugars, in a potentially useful compound as an intermediate in further amino 242 243 acid syntheses. ' Attempts to obtain 213 by oxidation of the diols (211) and (212) using the potassium permanganate-pyridine method resulted in total oxidative cleavage of the diol . Similarly treatment of 211 or 212 with ruthenium tetroxide, chromic oxide-pyridine, or 244 permanganate in acetic anhydride treatment of the olefins, failed to afford the cx-keto ester (213). - 120 -3 * 221 Rj-NH ; R2=H; R3=H Figure XI: Circular Dichroism Spectra of a-Amino Esters 217 and 218, and a-Amino Acids 220 and 221. Both a-amino esters (218) and (217) were further characterized as their N-acetyl and N-benzoyl derivatives. The N-acetates (222) and (223), were readily formed by selective acetylation in acetic anhydride and pyridine for one hour. Acetylation of a tertiary hydroxyl requires relatively vigorous reaction conditions, usually using p_-toluenesulfonic acid as catalyst and elevated temperatures. Similarly, formation of the benzamido derivatives (224) and (225) was facilitated selectively by reacting the a-amino esters (218) and (217) in methanol with benzoic anhydride. Nmr showed the presence of the requisite N-blocking group and the presence of the hydroxyl function in the products isolated, as did elemental analyses. - 121 -2.3 Synthesis of D-2 and L-2-(1,2:5,6-di-O-isopropylidene-a-D-galactofuranos-3-yl)glycine 2.3.1 Potassium Permanganate Oxidation of Z-(and E)-3-C-Methoxy-carbonylmethylene-3-deoxy-l,2:5,6-di-O-isopropylidene-a-D-xylo-hexofuranose (204) and (205) When the minor unsaturated isomers (204) and (205). from the Wittig reaction on 1,2:5,6-di-O-isopropylidene-q-D-ribo-hexofuranos-3-ulose (9), were reacted with potassium permanganate in pyridine at -10° as previously described for isomers 211 and 212 (Section 2.2.1) only one spot on t i c was observed indicating that both the resultant diols (227) and (226) had similar R^'s. Thus i t was decided, that since the major unsaturated isomer (205) had been obtained as a crystalline compound, for the f i r s t time, and was completely free of 204 as evidenced by nmr, the oxidation should be performed on this compound alone. As a result the diol 3-C-[R-hydroxy(methoxycarbonyl)methyl]-l,2:5,6-di-0-isopropylidene-a-g-galactofuranose (226) was obtained in 55% crystalline yield. Diol 226 was also obtained from the mixture of diols afforded by application of the oxidation conditions to the mixture of unsaturated esters (204) and (205), by crystallization from ethanol. The diol corresponding to stereospecific oxidation of 204 could not be detected or differentiated from that of 205 by t i c and hence could not be isolated from the mixture. As previously (Section 2.2.1) the structure of diol 226 was assigned on the basis of i t s circular dichroism and optical rotatory dispersion spectra (Figure XTI) compared to that of lactic acid. It can be seen that a negative Cotton effect was observed thus implying the R-configuration. This result is consistent with the assigned stereochemistry about the double bond assigned by nmr i f dihydroxylation by the permanganate ion proceeds via attack on the - 122 -R2 = 1 ,2-0-isopropylidene glycolyl 236 less sterically hindered side of the sugar residue to afford a compound of the galacto configuration in which the exocyclic branched-241 chain has the opposite chirality to that of L-lactic acid. There-fore, since oxidation of pure 205 resulted in d i o l 226, which exhibited a strong negative Cotton effect at 210 nm, opposite in sign to that shown by L-lactic acid, diol 226 is suggested to be 3-C-[R-hydroxy(methoxycarbonyl)methyl]-l,2:5,6-di-0-isopropylidene-a-D-galactofuranose. 2.3.2 Selective Mono-mesylation of Diol 226 Treatment of 226 with methanesulfonyl chloride in pyridine at room temperature resulted in the expected mono-mesylate (229) in 60% - 123 -yield. In addition to the expected product, the unusual dehydrated mesylate (230) was obtained in 30% yield after chromatography on t i c grade s i l i c a gel using benzene-ethyl acetate (3:7) as developer. This component was, however, present before chromatography and does not appear to be an artifact of the separation process. The dehydration process could be explained by i n i t i a l formation of the OMs 230 dimesylate, the tertiary hydroxyl of the galacto diol (222) being sulfonated more rapidly than the tertiary hydroxyl in the gluco isomers (211) and (212) due to i t s prominant position above the ring. This dimesylate could then lose the elements of methanesulfonic acid to afford 230. \ ' 13 ^ Overend has found that tertiary tosylates can be quite unstable at room temperature. When 229 was treated with 65% aqueous acetic acid, several products - 124 -were formed, some of which decolourized potassium permanganate spray, thus indicating dehydration of the tertiary hydroxyl, along with the expected product in which the 5,6-O-isopropylidene group was selectively removed to yield 228. 229 228 2.3.3 Displacement of the Methanesulfonate Ester with Azide Using the same conditions as described in Section 2.2.3, treatment of 229 with sodium azide in dimethyl formamide under anhydrous conditions at 60° in the dark, displaced the mesylate as expected to afford the azido sugar (231) which was not isolated. Immediate reduction of the latter compound with hydrogen over 5% palladium on carbon in anhydrous benzene yielded both the a-L- and a-D-amino esters (232) and (234) in 38 and 9.5% yields, respectively. Reduction of the azido-ester (231) was considerably slower than in the case of the epimeric azido ester (219) having the gluco configuration, taking 7 hours in the former and 1.25 hours i n the latter case, thus possibly giving some indication of the steric environment around the azido group. - 125 -R = 1,2-0-isopropylidene glycolyl A discussion of the non-specificity in the mesylate displacement and subsequent formation of both possible amino esters has already been put forward (Section 2.2.3). Both amino esters had nmr spectra consistent with the gross structural assignments, indicating the presence of three protons which readily exchanged in D2O. Both compounds (232) and (234) afforded positive ninhydrin tests. The configuration of the resultant a-amino esters were determined by cd studies on the esters and the free amino acids obtained after 241 hydrolysis. Methyl L-2-(1,2:5,6-di-O-isopropylidene-a-g-galacto-furanos-3-yl)glycinate (232) exhibited a positive Cotton effect at 204 nm in 95% ethanol. Conversely, methyl D-2-(1,2:5,6-di-0-isopropylidene-a-D-galacto-furanos-3-yl)glycinate (234 ) exhibited a negative Cotton effect at 206 nm in contrast to the positive Cotton effects of L-amino acids. 2 ^ - 126 -Figure XII. Circular Dichroism Spectra of Diol 226, a-Amino Esters 232 and 234, and a-Amino Acids 235, and 233. - 127 -Hydrolysis of 232.and 234 in 1.25% aqueous methanolic sodium hydroxide solution followed by passage through a column of Rexyn RG:51 (H) (Carboxylic Acid Type Resin) and elution with water afforded in high yield the crystalline amino acids L-2- and D-2-(1,2:5,6-di-0-isopropylidene-a-D-galactofuranos-3-yl)glycine (233) and (235), respectively. The circular dichroism spectra of these compounds revealed strong Cotton effects of the same sign as the corresponding a-amino esters when determined in 0.5 M HC1 in 95% ethanol, at a wavelength of 210 nm, thus supporting the configurational assignment of 233 and 235 (Figure XII). 2.4 Synthesis of D-2- and L-2-(3-Deoxy-l,2:5,6-di-0-isopropylidene-a-g-allofuranos-3-yl)glycine (246) and (247) As mentioned in the Introduction Rosenthal and Shudo"'"^ 1 have prepared stereospecifically L-2-(3-deoxy-l,2:5,6-di-0-isopropylidene-a-D-allofuranos-3-yl)glycine (247) and assigned i t s configuration on the basis of i t s circular dichroism spectrum. Therefore i t was of considerable interest to synthesize the analogous D-amino acid to observe i f indeed i t had the opposite cd spectrum to that of 247. In addition, i t was hoped to increase the yields in this reaction sequence and to investigate a better method of deacetylation than that described.1^"'" The overall yield from diol (211) to the a-amino ester (245) was increased from 5% to 25% and the deacetylation of 238 to 239 was accomplished in a one-^step, 100% yield reaction as compared to the two-step, 74% yield, process for the R-analog. Thus in an attempt to stereospecifically prepare g-2-(3-deoxy-l,2:5,6-di-0-isopropylidene-a-- 128 -g-allofuranos-3-yl)glycine (246), d i o l 212 was used, since the amino ester of the D-configuration should result. Selective acetylation of 3-C-[R-hydroxy(methoxycarbonyl)methyl]-l,2:5,6-di-0-isopropylidene-a-D-glucofuranose (212) in pyridine with acetic anhydride afforded the mono-acetate (236) in 83% yield after chromatography. Nuclear magnetic resonance revealed the presence of a single acetoxyl group at T 7.81 in addition to a one-proton singlet at T 6.28 which readily exchanged with D 2 O . Therefore, i t is suggested that 236 is 3-C-[R-acetoxy(methoxycarbonyl)methyl]-1,2:5,6-di-0-isopropylidene-a-D-glucofuranose. Freshly d i s t i l l e d thionyl chloride is a well-known stereospecific dehydrating agent of alcohols, eliminating the elements of water in a 245 246 trans manner. ' Thus when 236, was treated with thionyl chloride in pyridine at 0° an unsaturated sugar (237) was obtained in 69% crystalline yield. This compound gave a positive potassium permanganate test and did not show the presence of an hydroxyl or an a-proton in i t s nmr spectrum and was therefore suggested to be Z-3-C-l'-0-acetyl-1'-methoxycarbonylmethylene-3-deoxy-l,2:5,6-di-0-isopropylidene-q-D-ribo-hexofuranose (237). Catalytic hydrogenation of 237 in ethyl acetate under a hydrogen atmosphere using 5% palladium on carbon as catalyst resulted in almost quantitative conversion to 3-C- [S-acetoxy(methoxycarbonyl)methyl]-3-deoxy-l,2:5,6-di-0_-isopropylidene-a-D-allofuranose (238). Nmr and t i c indicated that the product of hydrogenation was homogeneous and was of the alio configuration since the proton on carbon two (T 5.19) resonated as a tr i p l e t with J = 4 Hz and J = 4 Hz. The doublet - 129 -of triplets at x 7.58 was assigned to H-3 and revealed a coupling of 9.5 Hz with H-4. Similarly large values have been previously observed 214 and accurately measured by Hall and Slessor for trans orientated protons on carbons three and four of hexofuranoses. Treatment cf the acetate (238) in anhydrous methanol with a 0.05 molar equivalent of sodium methoxide for a short time resulted in catalytic deacetylation in 100% yield to afford 239 as a pure syrup. Homogeneity was shown by t i c and nmr. Nuclear magnetic resonance revealed that- the acetoxyl group had in fact been removed and replaced by an hydroxyl function. No racemization of the a-hydroxy ester was apparent by nmr which showed a single H-l' resonance at T 5.40. This - 130 -hydroxy ester exhibited a strong, positive Cotton effect at 208 nm (Figure XIII)as compared to the negative Cotton effects shown by the starting diol (212), and the Isomeric 3-C-[R-hydroxy(methoxycarbonyl)-methyl]-3-deoxy-l,2:5,6-di-O-isopropylidene-a-D-allofuranose (248). Previously the ORD spectrum of 248 was recorded,however, in Figure XIII i t s cd spectrum i s illustrated. Thus the inversion of the cd spectra confirms the stereospecific inversion of configuration at the asymmetric position a- to the ester function. 2.4.1 Preparation and Reactions of Sulfonate Esters (240) and (241) Treatment of the a-hydroxy ester (239) with methanesulfonyl chloride and toluenesulfonyl chloride in pyridine afforded the methane and toluenesulfonate esters (240) and (241), in 92 and 63% yields, respectively. Reaction of the methanesulfonate (240) with sodium azide in anhydrous dimethyl formamide at 55°, in the dark, for 40 hours, followed immediately by extraction and hydrogenation in methanol for 6 hours over 5% palladium on carbon as catalyst resulted in the formation of two ninhydrin positive components having R^  0.25 (major) and R^  0.19 (minor) [ s i l i c a gel, ethyl acetate]. Column chromatography afforded the two pure a-amino esters (243) and (245) in 34 and 22% yields, respectively. The major component 243 exhibited a strong, negative Cotton effect at 208 nm in i t s cd spectrum whereas 245 exhibited a positive Cotton effect (Figure XIII) . The a-amino ester (245) was shown to be identical to that previously synthesized 161 stereospecifically in this laboratory and therefore i s methyl L-2-(3-Deoxy-l,2:5,6-di-0-isopropylidene-a-D-allo-furanos-3-yl)glycinate, - 131 -whilst 243 must be methyl D-2-(3-deoxy-l,2:5,6-di-O-isopropylidene-a-D-allofuranos-3-yl)glycinate. Similarly, when 3-C-[S-p_-toluenesulfonyloxy(methoxycarbonyl)methyl]-1,2:5,6-di-O-isopropylidene-a-D-allofuranose (241) was reacted with sodium azide in anhydrous dimethyl sulfoxide at 55°, i n the dark, for 40 hours, followed by hydrogenation in methanol both a-amino esters (243) and (245) in 10 and 27% yields respectively were formed. That both 243 and 245 were in fact the a-amino esters was shown by nmr which revealed a braod signal centred at T 8.15 which integrated for two protons and readily exchanged in D2O. H-3 was observed as a multiplet at x 7.55. Elemental analysis confirmed the gross structure suggested. Basic hydrolysis of 243 in 1.25% aqueous methanolic sodium hydroxide solution followed by passage through a weakly acidic ion-exchange resin resulted in crystalline a-amino acid (246) in 95% yield. Similar treatment of 245 was not carried out since this has previously been reported. 1^ 1 The pure a-amino acid D-2-(3-deoxy-l,2:5,6-di-0-isopropylidene-a-D-allofuranos-3-yl)glycine (246) exhibited a strong, negative Cotton effect at 212 nm in i t s cd spectrum using 0.5 M HC1 in 95% ethanol as solvent. N-Benzoylation cf 243 in methanol with benzoic anhydride afforded the known methyl D-2-(3-deoxy-l,2:5,6-di-0_-isopropylidene-a-D-allofuranos-3-yl)-N-benzoylglycinate (244) which has been prepared by Rosenthal and 247 Dooley via an alternate route. Configurational assignments of the glycosyl a-amino acid was based on 241 the fact that many D-a-amino acids exhibit negative Cotton effects. - 132 -i Figure XIII: Circular Dichroism Spectra of a-Hydroxy Esters 239 and 248, a-Amino Esters 243 and 245, and a-Amino Acids 246 and 247. - 133 -2.4.2 Conversion of Methyl D-2-(3-deoxy-l,2:5,6-di-0-isopropylidene-a-D-allofuranos-3-yl)glycinate (243) to 3-C-t(R)-(hydroxymethyl-N-salicylideneamino)methyl]-3-deoxy-l,2:5,6-di-0-isopropylidene-a-D-allofuranose (250) Jordaan and coworkers have reported the crystal structure of 116. In addition they have converted 248 to 116 via selective unblocking of the 5,6-positions, oxidation, reduction and brosylation. Compound 248 was also converted by reduction and hydrolysis to the hydroxy amine (249) which afforded a crystalline N-salicylidene derivative (250). Therefore the unequivocal structural proof of at least one of the glycosyl amino acids (243) prepared in our work could in theory be attained. Treatment of the a-amino ester (243) with lithium aluminum - 134 -hydride In tetrahydrofuran afforded a crude product in which the ester carbonyl was no longer observable in i t s infrared spectrum. S a l i c y l -aldehyde in methanol was reacted with the crude material to afford the crystalline N-salicylidene derivative (250), having identical physical properties to those reported for the compound derived from 248. Thus we have been able to correlate via a common intermediate the compound of known structure, determined by X-ray crystallography,(116) with the compound 243, and hence the glycosyl amino acid (247), which exhibits a negative Cotton effect in i t s cd spectrum as do many D-amino acids. Therefore i t would appear reasonable to suggest that the glycosyl amino acid (246), which exhibits a positive Cotton effect, i s in fact the L-glycosyl amino acid. 2.5 Action of L-ALiinc Acid Oxidase on Glycosyl Amino Acids Since circular dichroism measurements do not afford unequivocal proof of the absolute configuration of the asymmetric centre a- to 248 the carboxyl group in multi-asymmetric centered compounds an independent alternative proof was sought. It has long been known that many amino acids may function as substrates for various enzymes which are often very specific in requirements for either D- or L-amino acids. The venoms of several poisonous snake species contain a very active L-amino acid oxidase. The venom of the eastern diamondback rattlesnake (Crotalus Adamanteus), whilst not covering the complete spectrum of L-amino acids, does react with many and is quite specific for the L-configuration. Thus, L-amino acid oxidase (Type I) from Crotalus - 135 -Adamanteus , coupled to a spectrophotometry method for product 249 detection, namely observation of the optical absorbance at 305 nm was employed. It i s known that the action of L-amino acid oxidases on a suitable substrate w i l l aford an a-keto acid. This may then be complexed with 3-hydrazinoquinoline to yield a species which absorbs very strongly at 305 nm. Whilst controls u t i l i z i n g L-alanine and L-phenylalanine exhibited increased absorption at 305 nm, the samples containing D-phenylalanine and the six glycosyl amino acids (220, 221, 233, 235, 246 and 247) failed to exhibit increased absorbtion at the same wavelength. Since there are a number of known L-amino acids which are 250 resistant to Crotalus Adamanteus, including serine, proline, lysine and hydroxyproline, no conclusion can be stated as to the configuration of the glycosyl amino acids by this assay. - 136 -IV. EXPERIMENTAL 1. General Methods Nuclear magnetic resonance spectra were determined on a Varian T-60, HA-100 or XL-100 spectrometer. Absorptions are given i n T units with tetramethylsilane as internal standard (set at 10). The following abbreviations are used in describing nmr spectra: (d) = doublet, (s) = singlet, (t) "= t r i p l e t , (q) = quartet, (d.t) = doublet of triplets etc. Mass spectra were recorded on an A.E.I. MS9 spectrometer. Optical rotations were measured at ambient temperature with a Perkin Elmer model 141 automatic polarimeter. Infrared spectra were recorded on a Perkin-Elmer model 137 spectrometer and ord and cd measurements were performed on a JASCO J-20 Automatic Recording Spectro-Polarimeter or a JASCO ORD/UV-5 spectropolarimeter. Ultraviolet spectra were recorded on a Unicam SP 800 spectrometer. Elemental analyses were performed by Mr. P. Borda, Department of Chemistry, University of British Columbia. Melting points were determined on a Leitz Microscope heating stage model 350, and are corrected. 2. Chromatography 2.1 Column S i l i c a gel column chromatography was performed using s i l i c a gel - 137 -60-200 mesh (Davidson commercial grade H) or s i l i c a gel for t i c (D-0, Mondray Ltd.) indicated as " t i c grade s i l i c a gel'.1 For t i c grade s i l i c a gel column chromatography the ratio of material to absorbent, i f not stated,was approximately 1 to 300. Columns were pressurized above the solvent reservoir to a pressure of 5-10 p s i . 2.2 Thin Layer Chromatography A l l thin layer chromatography was performed using s i l i c a gel for t i c (D-0 , Mondray Ltd.), containing 1% electronic phosphor. Compounds were detected by ultraviolet absorption, by spraying with 50% sulfuric acid followed by heating on a hot plate, or by spraying with 0.3% solution of ninhydrin in n-butanol followed by warming at 110° in an oven. Methyl 4,6-0-benzylidene-2-deoxy-a-D-ribo-hexopyranoside (126) To a solution of methyl 2,3-anhydro-4,6-0-benz ylidene-a-D-allo-pyranoside 1 7 5 (125, 10 g) in anhydrous tetrahydrofuran (280 ml) was added lithium aluminum hydride (8 g) slowly. After the reaction mixture was refluxed for 5 h, the excess lithium aluminum hydride was decomposed by dropwise addition of water. The solids were removed by f i l t r a t i o n and washed with chloroform (2 x 200 ml). The combined f i l t r a t e was washed twice with water (200 ml), dried with calcium sulfate, and evaporated under reduced pressure to give crystalline 126 . (10 g, 100%); m.p. 124-126° ( l i t . 1 7 5 m.p. 125-127°). - 138 -Methyl 4,6-0-benzylidene-2-deoxy-a-g-erythro-hexopyranosid-3-ulose (127) Method A: To a vigorously stirred solution of 126 (10 g), sodium hydrogen carbonate (2 g), ruthenium dioxide (0.17 g) i n carbon tetra-chloride (510 ml) and water (30 ml) was added portionwise a solution of 5% sodium metaperiodate. Each portion of sodium metaperiodate solution was added unt i l the characteristic yellow-green colour of ruthenium tetraoxide formed. The solution was stirred vigorously un t i l the colour was discharged to give rise to black ruthenium dioxide. At this time the next aliquot of 5% sodium metaperiodate solution was added and the process continued unt i l a l l 126 had been converted to 127 as evidenced by t i c [ s i l i c a gel, benzene-methanol (97:3);126R^ 0.30 and 127 R F0.40]. F i l t r a t i o n and extraction of the solution with chloro-form (4 x 200 ml) afforded 127 (6.7 g, 68%) after drying (calcium sulfate) and evaporation of solvent under reduced pressure; m.p. 175-177° ( l i t . 1 7 6 m.p. 176-177°). Method B; To a solution of methyl 4,6-0-benzylidene-2-deoxy-oi-D-ribo-hexopyranoside (4.6 g) in anhydrous dimethyl sulfoxide (130 ml) was added to anhydrous acetic anhydride (50 ml) and the resultant solution kept at room temperature for 70 h, after which time t i c showed the reaction was complete. The solution was then mixed with a saturated solution of sodium hydrogen carbonate (600 ml). The precipitate which formed was removed by f i l t r a t i o n , washed well with water and dried under vacuum, yield 3.51 g (78%). The product was recrystallized from ethyl acetate; m.p. 175-176.5° ( l i t . 1 7 6 m.p. 176-177°). - 139 -Methyl 4,6-0-benzylidene-E-(and Z-)-3-C-Cyanomethylene-2,3-dideoxy-oc-g-erythro-hexopyranoside (128) and (129) To a fil t e r e d solution of the carbanion of diethylcyanomethyl-phosphonate (1.5 g) and sodium hydride (0.21 g) i n dimethoxyethane (20 ml) at 0°, was added a solution of the ketose 127 (1.82 g) i n dimethoxyethane (80 ml) over a period of 1 h. The reaction mixture was stirred for 15 h at 25°. Water (80 ml) was added and the reaction mixture was extracted with diethyl ether (3 x 80 ml) and chloroform (1 x 80 ml). The combined organic phase was dried (calcium sulfate), f i l t e r e d and evapoated under diminished pressure to afford the crude product as a yellow syrup (2.34 g). A portion (1.03 g) of the reaction mixture was chromatographed on t i c grade s i l i c a gel [150 g, benzene-ethyl acetate (9:1)] to yield 128 (0.57 g, 69%) and 129 (0.062 g, 7%) having 0.70 and Rf 0.40, respectively. Compound 128 was recrystallized from ethanol; m.p. 140-141°; [ a ] * 1 +259° (c 4.2, chloroform); i r (nujol) 2250 cm"1 (C^N), 1650 cm"1 (C=C); x 3 4.39 (s, 1, H-7), 4.50 (t, 1, 4 2 Hz, 2 & 2 Hz, H-l'), 5.13 (d.d, 1, J, „ 4 Hz, J n . 0.5 Hz, H-l), 5.85 (m, 1, H-4), 1,2a l,2e 5.7-6.6 (m, 3, H-6, H-5, H-4), 6.68 (s, OMe), 6.93 (d.d, 1, J g g m 15 Hz, H-2e), 7.43 (16 line multiplet, J , 2 Hz, J_ , 1 Hz, H-2a). £.3. j 1 *CQ. y 4-Irradiation at T 4.50 simplified the multiplet atT5.85 and simplified the multiplet at T 7.43 to an 8 line multiplet. Irradiation at T 5.13 produced a doublet at T 6.93 and an 8 line multiplet at T 7.43. Irradiation at T 5.85 produced a doublet at T 4.50 and an 8 line multiplet at T 7.43. - 140 -Anal. Calc. for C.,H,,0.N: C, 66.88; H, 5.97; N, 4.88. Found: l b 1 / 4 C, 66.64; H, 5.80; N, 4.63. Compound 129 was recrystallized from ethanol; m.p. 156.0-156.5°; [ a ] 2 6 +34.5° (c 1, chloroform); i r (nujol) 2250 cm"1 (C^N), 1650 cm"1 (C=C); T 3 4.30 (s, 1, H-7), 4.70 (d.t, 1, J., „ 2 Hz, J. , 0 1.0 Hz, H-1'), 5.16 (d.d, 1, H-1, J . . 3.5 Hz, J . „ 1.5 Hz, H-1), 5.6-6.6 l,za l,^e (m, 4, H-4, H-5, H-6), 6.66 (s, 3, OMe), 7.32 (16 line multiplet, 1, J 15 Hz, J 0 , 1 Hz, H-2a), 7.50 (multiplet, 1, H-2e). Irradiation gem 2a, 4 v ' at T 4.70 collapsed the multiplet at i 7.32 to an eight line system and the multiplet at t 7.50 to a 4 line system. Irradiation at x 5.16 collapsed H-2a (x 7.32) to an 8 line system and simplified H-2e (x 7.50). Anal. Calc. for C..H.,0.N: C, 66.88; H, 5.97; N, 4.88. Found: lo 17 4 C, 66.76; H, 5.83; N, 4.96. Methyl E-3-£-cyanomethylene-2,3-dideoxy-a-g-erythro-hexopyranoside (141) Unsaturated sugar 128 (0.31 g) dissolved in methanol (25 ml) was refluxed with Dowex 50W-X8 (H+) (1.2 g, prewashed with methanol), for 2.25 h or u n t i l no starting material remained, by t i c [ s i l i c a gel, benzene-ethyl acetate (8:2)]. Removal of the resin by f i l t r a t i o n followed by evaporation of the solvent yielded a syrup which was treated with charcoal and water then f i l t e r e d . After evaporation of the water the residue was azeotroped with ethanol and benzene under diminished pressure. The clear syrup (0.20 g, 100%) crystallized on trituration with ethanol. An analytical sample was recrystallized from ethanol-petroleum ether (30-60°) and sublimed (110°/0.1 mm); m.p. 164.0-164.5°; [ a ] 2 3 +316° (c 0.14, chloroform); i r (nujol) 3400 cm"1 (OH), 2260 - 141 -cm"1 (C=N), 1645 cm'1 (C=C); T D M S 0 " d 6 4.28 (d, 1, J 7 Hz, H-4), 4.34 (m, 1, H-l'), 5.09 (d.d, 1, J , , 4 Hz, J , . 2 Hz, H-l), 6.74 (s, l,za l,ze 3, OMe), 7.31 (m, 2, J g e m 14 Hz, H-2a, H-2e). Anal. Calc. for C ^ ^ N : C, 54.26; H, 6.58; N, 7.03. Pound: C, 54.40; H, 6.72; N, 6.87. Methyl 4,6-di-0-p_-chlorobenzoyl-3-C-E-cyanomethylene-2,3-dideoxy-ct-D-erythro-hexopyranosIde (142) To the unblocked sugar 141 (0.187 g) in pyridine (10 ml) was added p_-chlorobenzoyl chloride (0.75 ml). The solution was stirred at 25° for 24 h then ice-water (200 ml) was added. The resultant solution was extracted with dichloromethane (4 x 100 ml) and the combined organic phases washed with saturated sodium bicarbonate solution (50 ml) and water (2 x 50 ml), dried (calcium sulfate) and evaporated to afford a solid which was recrystallized three times from chloroform to remove most of the p_-chlorobenzoic anhydride. Column chromatography on t i c grade s i l i c a gel [150 g, benzene-ethyl acetate (9:1)] under a pressure of 3 psi yielded 142 (0.385 g, 83%) which crystallized on trituration with ethanol. Recrystallization of 142 from ethanol 2 A afforded an analytical sample; m.p. 140.5-141.5°; [a]p +165° (c 1, - l —i r n n chloroform); i r (nujol) 2240 cm (C=N), 1740-1730 cm (C=0); T 3 4.15 (d.d, 1, J 4 5 10 Hz, l t 2 Hz, H-4), 4.58 (t, 1, 2 g 2 Hz, H-l'), 4.94 (d.d, 1, J 0 4 Hz, J „ 0.5 Hz, H-l), 5.43 (m, 2, H-6), 5.72 (m, 1, H-5), 6.55 (s, 3, OMe), 6.75 (m, 1, J 14 Hz, H-2e), gem 7.26 (m, 1, H-2a). Anal. Calc. for C^H^OgNCl^ C, 57.98; H, 4.02; N, 2.93. Found: C, 57.66; H, 3.85; N, 2.69. - 142 -Methyl 4,6-0-benzylidene-3-C-cyanoirtethyl-2,3-dideoxy-g-D-ribo-hexo-pyranoside (140) To a suspension of 5% palladium on carbon (0.50 g) in ethanol (50 ml) was added 128 (0.85 g) and stirred at room temperature under one atmosphere pressure of hydrogen u n t i l hydrogen uptake ceased. Filt r a t i o n and evporation of the mixture afforded a colourless syrup which was chromatographed on t i c grade s i l i c a gel [200 g, benzene-ethyl acetate (9:1)] under a pressure of 3 psi to yield (140) (0.57 g, 67%). 21 Compound 140 was recrystallized from ethanol; m.p. 84-86°; [a]^ - i r n n +126° (c 1.4, chloroform); i r (nujol) 2250 cm (CEN); T 3 4.40 (s, 1, H-7), 5.30 (d.d, 1, J. _ 4 Hz, J. _ 2 Hz, H-l), 5.72 (d.d, 1, J. y Z3. X ) /6 J . 8 Hz, J, , 6 Hz, H-4), 4.1-4.4v(m, 3, H-6, H-5), 6.66 (s, 3, OMe), 7.12 (8 line multiplet, 2, J 16 Hz, J,, ,10 Hz, J, , 0 4 Hz, H-l'), gem 1A,3 1 B,J 7.46 (m, 1, H-3), 7.78 (m, 1, J 15 Hz, J„ . 2 Hz, J . .2 Hz, H-2e), gem 2e,l 2e,3 8.02 (m, 1, J. ,4 Hz, J. . 4 Hz, H-2a). za,l 2a,J Anal. Calc. for C^H^O^N: C, 66.24; H, 6.62; N, 4.48. Found: C, 66.20; H, 6.56; N, 4.64. Methyl. 3-C-cyanomethyl-2,3-dideoxy-a,g-D-ribo-hexopyranoside (145), Methyl 4,6-di-0-p_-chlorobenzoyl-3-C-cyanomethy 1-2, 3-dideoxy-a (and (3 ) -D-ribo-hexopyranoside (146 and 147), and l,6-Anhydro-4-0-p_-chlorobenzoyl-3-C-cyanomethyl-2,3-dideoxy-a-D-ribo-hexopyranose (148) Compound 140 (0.57 g) was dissolved in methanol (50 ml) and stirred with Dowex 50W-X8(H^) (2.2 g) (prewashed with methanol) u n t i l a l l 140 was consumed. Fi l t r a t i o n and evaporation of the solvent afforded a clear syrup which after treatment with charcoal in water, - 143 -f i l t r a t i o n and removal of the water under diminished pressure afforded 23 145 (0.374 g, 100%) as a clear syrup; [a] D +42.5° (c 1, chloroform); i r (film) 3400 cm A (OH), 2250 cm x (C=N); x 3 4.97 (m, 1, H-1), 6.25 (m, 4, H-4, H-5, H-6), 6.65, 6.75 (2s, 3, OMe), 7.0-8.4 (m, 7, H-2, H-1', H-3, OH, two protons exchange with D^O). Anal. Calc. for CgH^NO^.O.SCH^H: C, 52.50; H, 7.86; N, 6.45. Found: C, 52.49; H, 7.00; N, 6.62. To compound 145 (0.310 g) in pyridine (20 ml) was added p_-chloro-benzoyl chloride (1.8 g) and the solution l e f t stand for 16 h at 25°. Ice-water (200 ml) was then added and the resultant solution was extracted with dichloromethane (4 x 200 ml) and the combined organic phases washed with saturated sodium bicarbonate solution (100 ml) and water (2 x 100 ml), dried (calcium sulfate) and evaporated to afford a crystalline solid (1.20 g) which was recrystallized three times from chloroform to remove £-chlorobenzoic anhydride. Column chromatography on t i c grade s i l i c a gel [200 g, toluene-ethyl acetate (9:1)] afforded 146 (0.33 g, 44%), 147. (0.29 g, 38%), and 148 (0.05 g, 8%). Compound 146 crystallized on trituration with ethanol and was recrystallized from ethanol n-hexane to yield and analytical sample, m.p. 129.5-130.5°; [ a ] 2 3 +118.5° (c 1, chloroform); i r (nujol) 2260 cm"1 (C=N), 1730 cm"1 rnn (C=0); x 2 4.77 (d.d, 1, J 5 Hz, J 10 Hz, H-4), 5.30 (t, 1, 4 , J 4, J J. „ 3 Hz, J, _ 3 Hz, H-1), 5.61 (m, 2, H-6), 5.90 (m, 1, H-5), 6.71 1,2a l,2e (s, 3, OMe), 7.1-7.3 (m, 3, H-1*, H-3), 7.9 (m, 2, H-2). Irradiation at x 7.92 produced a singlet at T 5.30. Irradiation at x 7.30 produced a doublet at x 4.77 (J 9.5 Hz) and irradiation at x 4.77 produced a tr i p l e t at x 5.90. - 144 -Anal. Calc. for C.oH..0,NClo: C, 57.73; H, 4.42; N, 2.93. Found: ZJ Z l O Z C, 57.67; H, 4.44; N, 2.72. 23 Compound 147 could not be induced to crystallized, [ a ] n -37.6°(c_ 1, - i - l rnn chloroform) ; i r (film) 2250 cm x (CEN) , 1740 cm (C=0); x 3 4.57 (d.t, 1, J_ , 8 Hz, J c , 3 Hz, J c , 6 Hz, H-5), 4.94 (d.d, 1, J. _ 5 , 4 5,63 5 , 6 b 1 » / a 5 Hz, J. „ 0.5 Hz, H-1), 5.16 (d.d, J 12 Hz, H-1'), 5.48 (d.d, l,2e £ gem 1, H-6h), 5.85 (d.d, 1, J . . 6 Hz, H-4), 6.64 (s, 3, OCH,). 3,4 J An analytical sample of 148 was obtained by recrystallization from ethanol and sublimation (13070.1 mm); m.p. 148.5-149.0°; [ a ] 2 5 -189° rnn (c 1, chloform); x 3 4.33 (s, 1, H-4), 4.93 (t, 1, ^ 2 & 3 Hz, ^ 2 & 3 Hz, H-1), 5.20 (d.d, 1, J e 4 Hz, J_ 2 Hz, H-5), 6.10 (m, 2, 5,6 a 5,6 b H-6), 7.2-8.4 (m, 5, H-3, H-1', H-2). Anal. Calc. for C^H^O^Cl: C, 58.57; H, 4.59; N, 4.55. Found: C, 58.29; H, 4.43; N, 4.34. 4,6-Di-0-p_-chlorobenzoyl-3-C-cyanomethyl-2,3-dideoxy-D-ribo-hexo-l-eno-pyranose (149) , 2,.6-Dichloro-9-(4 *,6' -di-0-p_-chlorobenzoyl-31 -C-cyanomethyl-2',3'-dideoxy-a-D-ribo-hexopyranosyl)purine (150) and 2,6-Dichloro-9- (4', 6' -di-0-p_-chlorobenzoyl-3' -C-cyanomethyl-2', 3 * -dideoxy-g-D-ribo-hexopyranosyl)purine (151) A mixture of finely powdered 2,6-dichloropurine (300 mg) and 146 (480 mg) was dried by d i s t i l l a t i o n of toluene (10 ml) from i t and evaporated to dryness. The mixture was fused at 155° and 30 mm for 1.5 h then at 155° and 0.1 mm for a further 0.5 h. Dissolution of the light brown glass in warm ethanol-ethyl acetate (1:1), f i l t r a t i o n and evaporation afforded a syrup which was column chromatographed on t i c - 145 -grade s i l i c a gel [170 g, benzene-ethyl acetate (6:4)] under a pressure of 8 psi to yield 149 (153 mg, 34%), 150 (238 mg, 37%) and 151 (159 mg, 24%). 23 Compound 149, syrup; [ a ] D +175 (c 2, chloroform); i r (film) 2280 - i - i - i rnn cm x (C=N), 1720 cm (C=0), 1600 cm A (C=C); x 3 3.46 (d.d, 1, J, „ 6 Hz, J, _ 2 Hz, H-1), 4.27 (d.t, 1, J. , 6 Hz, J. , 2 Hz, J c . 1,2 1,3 5 , b a 5,Ob 5,4 6 Hz, H-5), 5.15 (d.d, 1, J 4 Hz, H-2), 5.4-5.6 (m, 3, H-6, H-4), 2 , J 6.92 (m, 1, H-3), 7.50 (m, 2, H-1'). Anal. Calc. for C 2 2H 1 70 5NC1 2: C, 59.23; H, 3.84; N, 3.14. Found: C, 60.50; H, 3.98; N, 2.86. Compound 151 was recrystallized from ethanol, m.p. 184-185°; rnn [a]* +33.4 (c 1.6, chloroform); x 3 1.78 (s, 1, H-8), 3.70 (t, 1, J l \ 2 ' 6 H z ' H - 1 , ) ' 4 - 5 0 ( c L t ' X» J5',4' 6 H z ' J 5 \ 6 a 3 H z ' J 5 \ 6 b 6 Hz, H-5'), 5.37 (m, 3, H-6', H-4'), 6.8-7.4 (m, 5, H-2', H-3', H - f ) . Anal. Calc. for C ^ H ^ N ^ C l ^ C, 51.05; H, 3.00; N, 11.02. Found: C, 50.71; H, 2.84; N, 10.83. Compound 150 was recrystallized from ethanol, m.p. 230-231° (sinters ? 7 rnn at 197-199°); [a]^' -54.5° (c 0.7, chlof orm); x ^ x 3 1.70 (s, 1, H-8), 3.64 (d.d, 1, J x , 2 a , 10 Hz, J 1 ? 2 g , 3 Hz, H-1'), 4.62 (m, 1, H-4'),. 4.87 (d.d, 1, J,, c, 9 Hz, J,, ,, 12 Hz, H-6'^, 5.34 (m, 1, H-5'), 6 a ' 5 6 a ' 6 b 5.59 (d.d, 1, Jg, 5, 4 Hz, H-6^, 6.8-7.7 (m, 5, H-2', H-3', H i " ) . Found: C, 50.87; H, 3.07; N, 10.91 Anal. Calc. for C ^ H ^ N ^ C ^ : C, 51.05; H, 3.00; N, 11.02. - 146 -2-Chloro-6-N,N-dimethylamino-9-(3'-C-cyanomethyl-2',31-dideoxy-B-D-ribo-hexopyranosyl)purine (153) Crude 151 (130 mg) was suspended i n methanol (10 ml) and 25% aqueous dimethylamine (10 ml) and stirred at room temperature for 4 h. The mixture was evaporated to dryness and applied to a column of t i c grade s i l i c a gel (45 g) packed and eluted with chloroform-ethanol (9:1) to afford 16_ (30 mg, 36%) as a foam which crystallized on addition of a small amount of ethanol. An analytical sample was recrystallized from ethanol, m.p. 187.0-188.5°; [ a ] 2 3 +33.4° (c 1.6, chloroform); uv Xmax 217 nm (e 9400), 276 nm (e 8350, in ethanol); cd (c 0.003, rnn ethanol) [ 6 ] - - - -3000; T 3 1.51 (s, 1, H-8), 3.60 (d.d, 1, J l ' 2a' 7 , 5 H z ' J l ' 2e' 4 H z ' H ~ 1 ' ) ' 6 , 4 6 ( b r * s« N ( M e ) 2 ) . Anal. Calc. for C._H._IT.0-C1: C, 49.11; H, 5.23; N, 22.79. I D i y o J> Found: C, 49.13; H, 5.10; N, 22.45. 2-Chloro-6-N,N-dimethylamino-9-(3'-C-N,N-dimethylaminocarbamoylmethyl-2',3'-dideoxy-a-D-ribo-hexopyranosyl)purine (152) 2,6-Dichloro-9- (4', 6' -di-0-p_-chlorobenzoyl-31 -C-cyanomethyl-2 *, 3' -dideoxy-B-D-ribo-hexopyrano sy1)pur ine (150, 230 mg), in methanol (20 ml) and 25% aqueous dimethylamine(20 ml) were stirred at room temperature for 5 h (until a l l nucleoside dissolved). The mixture was evaporated to dryness and azeotroped under reduced pressure with toluene (20 ml) and methanol (20 ml). Column chromatography on t i c grade s i l i c a gel [45 g, chloroform-ethanol (9:1)] afforded an unidentified component (43 mg) and 152 (80 mg, 54%). An analytical sample was recrystallized from ethanol, m.p. 189.5-190.0°; [ a ] 2 3 +8.2° (c 0.6, chloroform); - 147 -\r (nujol) 3300 cm"1 (OH), 1640 cm"1 (CO) ; uv X 217 nm (e 16,450) 276 nm (E 15,000); cd (c 0.004, ethanol) [ 9 ] 2 7 6 +2030 ; ^"^3 2.10 (s, 1, H-8), 3.96 (d.d, 1, 2 & , 10 Hz, 2 b , 4 Hz, H-l*), 6.48 (broad s, 6, NMe2~carbamoyl), 6.96, 7.04 (2s, 6, NMe2~purine). Anal. Calc. for C.,HOCN,O.C1: C, 49.45; H, 6.10; N, 20.35. 1/ ID D 4 Found: C, 49.65; H, 6.17; N, 20.08. Molecular weight by mass spectrometry 412, 414. C^H^NgO^Cl requires 412, 414. 2,6-Di-N,N-dime thylamino-9-(3'-C-cyanomethyl-2',3'-dideoxy-<x -g-ribo-hexopyranosyl)purine (158) To the blocked nucleoside (150, 112 mg) was added anhydrous dimethyl-amine (15 ml). After the solution was l e f t stand 20 days at -5° the dimethylamine was removed by gentle warming to afford the crude unblocked nucleoside (158) as a syrup. Column chromatography on t i c grade s i l i c a gel [40 g, dichloromethane-ethanol (93:7)] afforded 158 (40 mg, 63%) as a crystalline solid. An analytical sample was 23 recrystallized from methanol, m.p. 220-221°; [a] D +46.1° (c 0 . 3 , chloform); uv X 244 nm ( e 6400), 264 nm (sh, e 3160), 292 nm ( e 3100), max (in ethanol); cd (c 0.003, ethanol) [6] 2 g 2+1070 ; T D M S 0 ~ d 6 2.09 (s, 1, H-8), 4.10 (d.d, 1, J ' , 3 Hz, J.. , 0 , 10 Hz, H-l'), 6.62 (s, 6, X y X y Z c l NMe2), 6.90 (s, 6, NMe2) Anal. Calc. for C^H^N.^: C, 54.39; H, 6.72; N, 26.12. Found: C, 54.40; H, 6.59; N, 26.51. - 148 -2,6-Di-N,N-dimethylamino-9-(31 -C- [2"-acetamldoethyl]-2', 3'-di-deoxy-cx-D-ribo-hexopyranosyl)purine (170) Hydrogenation of 158 (15 mg) in ethanol (25 ml) saturated with ammonia at 0°, using 5% rhodium on alumina (15 mg) as catalyst, at 60 psi for 24 h afforded after f i l t r a t i o n and evaporation a clear syrup which was immediately acetylated in pyridine (0.5 ml) and acetic anhydride (0.5 ml) at room temperature for 12 h. The acetylation mixture was evaporated to dryness and treated with 25% aqueous dimethyl-amine (2.5 ml) in methanol (2.5 ml) for 3 h. This solution was evaporated to dryness, taken up in chloroform-methanol (1:1) and purified by preparative t i c [ s i l i c a gel, dichloromethane-ethanol (9:1)] to afford 170 (6.5 mg, 39%) as a clear syrup which could not be induced 23 to crystallize, [a]„ +34.6 (c 0.7, chloroform); uv X 244 nm (e 7700), D — max 263 nm (e 3750 sh), 292 nm (e 3600 in ethanol); cd (c 0.005, ethanol) [ 9 ] 2 9 2 +750; T 3 2.38 (s, 1, H-8), 3.80 (m, 1, N--E), 4.18 (d.d, 1, J , , 11 Hz, J , , 3 Hz, H-l'), 6.58, 6.87 (2 x s, 12, 2 x NMe„), X y C.cL X y Z.G-8.05 (s, 3, NAc). Molecular weight required for C H Q 1N 0.: 421. Found: 421 by i y 3 i / H mass spectrometry. 209 Methyl 2-deoxy-q,g-D-erythro-pentofuranoside (172) To 2-deoxy-D-ribose (171) (5.0 g) was added 0.05% methanolic hydrogen chloride solution (175 ml). After st i r r i n g at room temperature for 15 minutes silver carbonate (10 g) was added and the suspension stirred vigorously for a further 20 minutes. The reaction mixture was then passed through charcoal and the solvent removed under reduced - 149 -pressure at a temperature of 40°, to yield 172 as a mobile, colourless, syrup (4.96 g, 90%). Methyl 2-deoxy-5-0-trityl-q-g-erythro-pentofuranoside (175) and Methyl 2-deoxy-5-0-trityl-g-D-erythro-pentofuranoside (176) Chlorotriphenylmethane (19.0 g) in pyridine (200 ml) was added to 212 172 (9.6 g) and stirred for 5 days at room temperature. The solution was then poured slowly into ice-water (1 1.) with stirring and the resultant gum washed several times with cold water. Drying was completed by azeotroping with ethanol and toluene to afford a crude mixture of 175 and 176 (28.5 g). Column chromatography of the mixture ( s i l i c a gel G, 1300 g, col. dim: 7 x 60 cm, benzene-ethyl acetate, 9:1) afforded 175 (6.7 g), 176 (6.5 g) and a mixture of 175 and 176^  (3.1 g); overall yield 65%. Methyl 2-deoxy-5-0-trityl-a-g-glycero-pentofuranosid-3-ulose (182) To a solution of 175 (2.0 g) in carbon tetrachloride (40 ml) and water (6 ml) containing sodium hydrogen carbonate (0.4 g) and ruthenium dioxide (40 mg), was added a 5% solution of sodium metaperiodate with vigorous stir r i n g . The metaperiodate solution was added slowly u n t i l the characteristic yellow-green colouration of ruthenium tetroxide was observed, no further metaperiodate was added until the solution had ful l y reverted to the black ruthenium dioxide stage. This process was repeated un t i l a l l starting material had been consumed (16 h) as evidenced by t i c ( s i l i a gel, benzene-methanol, 95:5). Unreacted oxidant was decomposed by the addition of 3-4 drops of isopropanol. The - 150 -reaction mixture was then f i l t e r e d , and the resultant aqueous layer extracted with chloroform (4 x 200 ml). The combined organic phases were washed with 5% sodium thiosulfate solution (25 ml), water (25 ml), dried over sodium sulfate, f i l t e r e d and evaporated to afford crystalline 182 (1.96 g, 98%). An analytical sample was recrystallized from 24 ethanol; m.p. 122.2-122.7°; [ a ] n +182.4° (c. 1.2, chloroform); i r (nujol) - i r nn 1755 cm x (C=0); T 3 4.52 (d.d, 1, J. . 5.2 Hz, J. . 1.0 Hz, H-1), !»-• a x'4. 5.94 (t, 1, J 4 5 a 2 . 6 Hz, J 4 5 b 3 . 4 Hz, H-4), 6.53 (s, 3, OCH.), 6.58 (ABM multiplet, 2, AAtJ 12 Hz, J A n (observed) 10 Hz, H-5), 7.37 (m, 2, AB AD A._. 34 Hz, J A D (observed) 18 Hz, H-2). AD Ao Anal. Calc. for C 25 H24°4 : C ' 7 7 - 3 ° ; H» 6-18". Found: C, 77.33; H, 6.06. Methyl 2-deoxy-5-0-trityl-B-g-glycero-pentofuranosid-3-ulose (183) To a solution of 176 (2.0 g) i n carbon tetrachloride (40 ml) water (6 ml), sodium hydrogen carbonate (0.4 g), and ruthenium dioxide (40 mg), was added a 5% solution of sodium metaperiodate in the manner described for 182. After 17 h the reaction was complete as shown by ti c ( s i l i c a gel, benzene-methanol, 95:5). After f i l t r a t i o n , the aqueous layer was extracted was chloroform (4 x 200 ml). The combined organic phases were then washed with 5% sodium thiosulfate solution (25 ml), water (25 ml), dried over sodium sulfate, f i l t e r e d and evaporated to afford 183 (1.90 g, 96%) as a mobile syrup which moved as sinyle spot on t i c . An analytical sample was prepared by molecular d i s t i l l a t i o n at 125°/0.1 mm; [ a ] 2 7 -26° (£0.5, chloform); i r (film), _ i cnci 1760 cm x (C=0); T v 3 4.71 (d.d, 1, J. . 5.5 Hz, J. „ 1.8 Hz, H-1), 1,2a l , 2 b - 151 -5.84 (m, 1, J, 6.5 Hz, J, . , 3.2 Hz, J . 1.2 Hz, H-4), 6.72 (s, *»^a 4,z a 3, OCH3), 6.70 (ABM multiplet, 2, A A f i 20 Hz, J g e m 10 Hz, H-5), 7.46 (m, 2, H-2). Anal. Calc . for C^H-.O.: C, 77.30; H, 6.18. Found: C, 77.12; 25 24 4 H, 6.00. Methyl Z,E-3-C-cyanomethylene-2,3-dideoxy-5-0-trityl-q-D-glycero-pentofuranoside (184) A solution of 182 (0.96 g) in dimethoxyethane (10 ml) was added to a fil t e r e d solution of the carbanion formed from diethylcyanomethyl-phosphonate (0.60 g) and sodium hydride (0.08 g) in dimethoxyethane (20 ml) over a period of 1.5 h at 0°, under a nitrogen atmosphere, then stirred for 1 h at room temperature. Water (40 ml) was added and the resultant aqueous layer extracted with chloroform (2 x 100 ml) and ether (2 x 100 ml). The combined organic phases were washed with water (2 x 50 ml) and dried over sodium sulfate. Evaporation of the solvent afforded 184 (0.96 g, 94%). Recrystallization of 184 from 25 ethanol yielded an analytical sample, m.p. 165.5-170.0°; [a]^ +120° (c 3.1, chloform); i r (film) 2250 cm"1 (C=N), 1660 cm"1 (C=C,olefin); T 3 4.78 (m, 2, H-l, H-l'), 5.27 (m, 1, H-4), 6.61 (m, 5, 0CH3, H-5), 7.07 (m, 2, H-2). Irradiation at T 5.27 simplified the multiplet at T 6.66. Irradiation at x 4.78 produced a doublet at T 7.07. Anal. Calc . for C^H^O^: C, 78.78; H, 6.08; N, 3.40. Found: C, 78.70; H, 5.78; N, 3.70. - 152 -Methyl Z,E-3-C-cyanomethylene-2,3-dideoxy-5-0-trityl-8-D-glycero-pentofuranoside (185) A solution of 183 (1.80 g) in dimethoxyethane (10 ml) was added to a fi l t e r e d solution of diethylcyanomethylphosphonate (1.2 g) and sodium hydride (0.16 g) in dimethoxyethane (30 ml) at 0°, under a nitrogen atmosphere. After st i r r i n g at room temperature for 1 h, water (80 m.) was added and the resulting solution extracted with chloroform (2 x 200 ml) and ether (2 x 200 ml). The combined organic phases were washed with water (2 x 100 ml) and dried over sodium sulfate. F i l t r a t i o n and evaporation of the solvent afforded 185 as a syrup (1.78 g, 94%) which was used without further purification. Methyl 3-C-cyanomethyl-2,3-dideoxy-5-0-trityl-a-g-erythro-pentofuranoside (186) and Methyl 3-C-cyanomethyl-2,3-dideoxy-5-0-trityl-q-D-threo-pentofuranoside (187) Hydrogenation of 184 (0.95 g) in ethanol-benzene (9:1) using 10% palladium on carbon (150 mg) as catalyst, at atmospheric pressure, afforded, after work up, a colourless syrup in quantitative yield. Tic of the latter syrup revealed two components 0.29 (major component) and R^  0.25 (minor) ( s i l i c a gel, petroleum ether-ethyl ether, 8:2) after double development. Column chromatography of a portion of the above material (0.55 g) on t i c grade s i l i c a gel (180 g) packed and eluted with the above solvent system, under a pressure of 8 psi afforded the two components. From compound 186 (403 mg, 79%), an analytical sample was prepared 21 by molecular d i s t i l l a t i o n at 125°/0.1 mm; [c t ] n +79° (c 0.6, dichloro-- 153 -- i r n n methane); i r (film) 2250 cm A (C=N); x 3 4.95 (d.d, 1, J , 4.5 Hz, J1 2„ 1.0 Hz, H-l), 6.09 (m, 1, J 4 5 S.5 Hz, J 4 3 3.0 Hz, H-4), 6.67 (s, 3, OCH3), 6.76 (m, 2, H-5), 7.4-8.2 (m, 5, H-2, H-3, H-l'). Irradiation at T6.76 produced a doublet at x 6.09. Anal. Calc. for C^H^C^N: C, 78.45; H, 6.54; N, 3.39. Found: C, 78.81; H, 6.60; N, 3.15. Compound 187 (55 mg, 10%) was recrystallized from ethanol, m.p. 155-* ?s - i rnn 157°; [ a ] " +67° (c 2.1, chloforom); i r (nujol) 2250 cm (C=N); x 3 4.88 (d.d, 1,J 4.5 Hz, J 2.5 Hz, H-l), 5.70 (d.t, 1, J. , 6.5 Hz, l , * a l , / b J 4 5 4.0 Hz, H-4), 6.63 (s, 3, 0CH3), 6.83 (m, 2, H-5), 7.0-7.7 (m, 1, H-3), 7.4-8.2 (m, 4, H-2, H-l'). Irradiation at x 6.83 produced a doublet at x 5.70 (J 6.5 Hz). Anal. Calc. for C H 2 0 N: C, 78.45; H, 6.54; N, 3.39. Found: C, 78.15; H, 6.35; N, 3.45. Methyl 3-C-cyanomethyl-2,3-dideoxy-5-0-trityl-g-g-threo-pentofuranoside (188) Hydrogenation of 185 (1.70 g) in ethanol (60 ml) using 10% palladium on carbon (0.20 g) as catalyst at atmospheric pressure afforded 188 (1.61 g, 94%) as a clear syrup which crystallized on trituration 23 with ethanol, m.p. 107-108°; [a] n -50° (c 0.8, chloform); i r (nujol) - i rnn 2255 cm (C^N); x 3 5.02 (d.d, 1, J 2.0 Hz, J 5.0 Hz, H-l), 1 , 1 a. 'lb 5.74 (d.t, 1, J. . 5.0 Hz, J. 5.0 Hz, J. _ 7.0 Hz, H-4), 6.71 (s, 3, 0CH3), 6.78 (ABX multiplet, 1, H-5), 7.3-8.1 (m, 5, H-2, H-3, H-l'). Irradiation at x 5.74 reduced the ABX multiplet centered atx6.78 to an AB system. - 154 -Anal. Calc. for C^H^C^N: C, 78.45; H, 6.54; N, 3.39. Found: C, 78.40; H, 6.56; N, 3.59. Methyl 3-C-cyanomethy1-2,3-dideoxy-a ,g-D-threo-pentopyranoside (189) From compound 187: In 1% methanolic hydrogen chloride solution (6 ml) 187 (123 mg) was stirred at room temperature overnight after which time silver carbonate (1.3 g) was added, and the suspension stirred for a further 2 h. F i l t r a t i o n of the mixture and evaporation of the f i l t r a t e afforded a quantitative yield of 189 as a mobile syrup (analysis of the latter by nmr revealed a ratio of a to 8 of 10:3). An analytical sample was prepared by preparative t i c ( s i l i c a gel, 22 benzene-methanol, 9:1), [a] Q +45° (c 1.2, chloroform); i r (film) 2250 cm r n r l (C=N); T 3 5.20 (d.d, 1, J 1 Hz, J 3 Hz, H-l), 6.2-6.8 (m, a  l , Z b 7, 0CH3, H-4, H-5, OH), 7.35-7.45 (m, 2, H-l'), 7.9 (m, 1, H-3), 7.9-8.5 (ABMX multiplet, 2, H-2). Irradiation a t T 5.20 simplified the ABMX system. Anal. Calc. for C0H._N0..0.5CH„OH: C, 54.55; H, 8.01; N, 7.48. o 1 3 3 J Found: C, 54.93; H, 7.43; N, 7.53. From compound 188: Compound 188 (39 mg) in 1% methanolic hydrogen chloride solution (2 ml) was stirred for 7.5 h. Silver carbonate (0.5 g) was added and the suspension stirred for a further 1 h. Fi l t r a t i o n and evaporation afforded 189 (15 mg, 93%) after preparative t i c . Ir, nmr and t i c indicated that this product was identical to that derived from 187. - 155 -Methyl 4-jO-p_-chloroben2oyl-3-C-cyanomethyl-2,3-dideoxy-q-D-threo-pentopyranoside (191) To a solution of 189 (11 mg) i n pyridine (2 ml) at 0°, was added p_-chlorobenzoyl chloride (0.16 ml) and stirred at room temperature overnight. Water (20 ml) was added and the solution was then extracted with dichloromethane (4 x 20 ml). The combined organic phases were washed with saturated sodium bicarbonate solution and water, dried over sodium sulfate and, evaporated. The residue was extracted with dichloromethane (5 ml). .Evaporation of the dichloromethane afforded the crude reaction mixture (44 mg) which was then purified by preparative t i c ( s i l i c a gel, benzene-ethyl acetate, 9:1) to yield pure 191 (17 mg, 85%), recrystallized from ethanol; m.p. 121.5-122.5°, [ a ] 2 1 -13.4° (c 4.0, chloroform); i r (KBr) 2250 cm**1 (C=N), 1730 cm*"1 rnn (C=0); x 3 5.07 (doublet of t r i p l e t s , 1, _ 10.5 Hz, J 4 _ a 10.5 Hz, J. 5.5 Hz, H-4), 5.24 (d.d, 1, J- 3.5 Hz, J. 0 1.5 Hz, H-1), 6.0-6.5 (m, 2, H-5), 6.62 (s, 3, 0CH3), 7.4 (m, 1, H-3), 7.55 (m, 2, H-1'), 7.87 (8 line multiplet, J 13 Hz, J„ , 5 Hz, H-2e), 8.28 gem 2e, 3 (6 line multiplet, J , 12 Hz, H-2a). Irradiation at x 5.07 simplified the multiplet at x 6.0-6.5 and that at x 7.4. Irradiation at x 5.24 simplified the multiplet at x 7.8-8.4. Anal. Calc. for C. _H,,0,NC1: C, 58.15; H, 5.17; N, 4.52. Found: 15 ID 4 C, 57.86; H, 5.10; N, 4.25. Methyl 4-0_-p_-bromobenzoyl-3-C-cyanomethyl-2,3-dideoxy-q-D-threo-pentopyrancside (190) To a solution of 189 (113 mg) in pyridine (20 ml) at 0° was added p_-bromobenzoyl chloride (441 mg) and the mixture stirred at room - 156 -temperature overnight. Ice-water (20 ml) was added and the mixture was then f i l t e r e d . The solid residue was dissolved in dichloromethane (25 ml), washed with saturated sodium bicarbonate solution (25 ml), water (25 ml), dried over sodium sulfate and evaporated. Preparative t i c , 4 (20 x 20) plates ( s i l i c a gel, benzene-ethyl acetate, 9:1) afforded 190 (214 mg, 85%), recrystallized from methanol, m.p. 149.5-150.0°; [a]* 1 -11° (c 1, chloroform); i r (KBr) 2250 cm"1 (C^N), 1760 - l rnn cm (C=0); T 3 5.04 (d.t, 1, J 4 3 10 Hz, J 4 5 & 10.5 Hz, J 4 5 g 5.5 Hz, H-4), 5.22 (d.d, 1, J , , , 3 Hz, J . „ Hz, H-l), 6.02-6.48 (ABX 1,1a. l,2e multiplet, 2, H-5), 6.61 (s, 3, 0CH3), 7.35 (m, 1, H-3), 7.55 (m, 2, H-l'), 7.75-8.43 (ABMX multiplet, 2, H-2). Anal. Calc. for C. _H,,0.NBr: C, 50.89; H, 4.53; N, 3.96. Found: I J I D 4 C, 50.91; H, 4.64; N, 3.95. 2,6-Dichloro-9-(4'-0-p_-bromobenzoyl-3'-C-cyanomethyl-21,3'-dideoxy-0-D-threo-pentopyranosyl)purine (195 , 2.6-Dichloro-9- (4' -0-p_-bromobenzoyl-3'-C-cyanomethyl-2', 3'-dideoxy-ct-g-threo-pentopyranosyl)purine (196), and 4-0-p_-Bromobenzoyl-3-C-cyanomethyl-2,3-dideoxy-D-threo-pent-l-eno-pyranose (197). Compound 190 (98 mg) and 2,6-dichloropurine (113 mg) were intimately mixed, and dried overnight under vacuum. The mixture was fused at , 145° for 1 h, under 30 mm pressure, then at 145-160°, under 0.1 mm pressure for 2 h. The brown glass was dissolved in benzene-ethyl acetate, 6:4 (15 ml) and f i l t e r e d . Column chromatography of the resulting product ( s i l i c a gel, 60 g, benzene-ethyl acetate, 6:4) afforded 197 (17 mg, 19%), 195 (41 mg, 29%) and 196 (19 mg, 13%). - 157 -23 Compound 197, syrup, [ a ] D -127° (£0.8, dichloromethane); i r (film) -1 - l - i r n n 2300 cm (C=N), 1730 cm A (C=0), 1650 cm (C=C); x m" L3 3.50 (d.d, 1, J 6.5 Hz, J. , 2.0 Hz, H-l), 4.95 (d.t, 1, J. . 6 Hz, J. _ -L,^ -»->J 4,J 4,5a 6.5 Hz, J 4 5 e 3.2 Hz, H-4), 5.30 (d.d, 1, J 2 3 3.0 Hz, H-2), 5.98 (ABX multiplet, 2, H-5), 7.30 (m, 1, H-3), 7.56 (m, 2, H-l'). Irradiation atx 3.50 produced a doublet at x 5.30 and simplified the multiplet at x 7.30. Irradiation at x 4.95 simplified the multiplets at x 5.98 and 7.30. Anal. Calc. for C^H^C^NBr: C, 52.19; H, 3.76; N, 4.35. Found: C, 52.48; H, 3.72; N, 4.29. Compound 195, recrystallized from ethanol, m.p. 222.5-223.5°; 24 r n n [a]p -15° (c 5, chloroform); x ^ (1.66 (s, 1, H-8), 2.0-2.6 (d.d, 4, aromatic protons), 4.05 (d.d, 1, J l t 2 & , 10 Hz, 1 & , 2 Hz, H-l'), 4.90 (m, 1, H-4'), 5.51 (d.d, 1, J , ., 5 Hz, J_ , , , 12 Hz, H-5e'), j 6 y H j 6 yDSi 6.34 (d.d, 1, J , ., 10 Hz, H~5a'), 7.3 (m, 5, H-3', H-2', H-l"). j3i y H> Anal. Calc. for C,_H1.N 0 BrCl_: C, 44.64; H, 2.76; N, 13.70. 19 14 5 3 2 Found: C, 44.74; H, 2.69; N, 13.64. Compound 196, recrystallized from ethyl acetate-ethanol (1:1), m.p. 250.5-252°; [a]^ -38° (£0.06, dichloromethane); x 3 1.65 (s, 1, H-8), 1.9-217 (d.d, 4, aromatic protons), 3.92 (t, 1, J-, . , 1. y /.Si 4 H z ' Jl',2e' 4 H z ) * Anal. Calc. for C ^ H ^ N ^ B r C ^ : C, 44.64; H, 2.76; N, 13.70. Found: C, 43.20; H, 2.31; N, 18.47. A satisfactory analysis of this compound could not be obtained. - 158 -2-Chloro-6-N,N-dimethylamino-9-(3' -C-cyanomethyl-2',31-dideoxy-8-D-threo-pentopyranosyl)purine (198) Nucleoside 195 (47 mg) in methanol (13 ml) and 25% aqueous dimethylamine (13 ml) was stirred at room temperature for 6 h and then at 5° for 12 h. After evaporation of the reaction mixture the residue was taken up in benzene-ethyl acetate (6:4) and chromatographed on t i c grade s i l i c a gel (30 g) using the above solvent system, to afford 198 (25 mg, 76%) as an amorphous solid. An analytical sample was prepared 2 A by recrystallization from ethanol-pentane, m.p. 188.5-189.5°; [ct] n +30.7 (£ 0.6, chloroform); uv max (ethanol) 217 nm (e 8050), 275 nm rnn (e 7,500); cd (£0.005, ethanol) [e]2?3 -3,000 (trough); x 3 2.12 (s, 1, H-8), 4.17 (d.d, 1, J , , 2.5 Hz, J., . , 10.0 Hz, H-1'), 5.90 (d.d, 1 , • J 5 e i 4 t 4 Hz, J 5 e , j 5 a , 11 Hz, H-5e'), 6.1-6.7 (m, 8, N(CH 3) 2, H-4', H-5a'), 7.3-8.1 (m, 5, H-2', H-1", H-3'). Irradiation at x 7.98 produced a singlet at T 4.17. Anal. Calc. for C .H N,0 Cl: C, 49.95; H, 5.09; N, 24.96. Found: W 1/ b / c, 49.98; H, 5.19; N, 24.67. 2-Chloro-6-N,N-dimethylamino-9-(3'-C-cyanomethyl-2',3'-dideoxy-a-D-threo-pentopyranosyl)purine (199); Crude 196 (6 mg) was treated with a 25% aqueous solution of dimethylamine (1 ml) and methanol (1 ml) and stirred at room temperature for 3 h. Evaporation to dryness followed by purification by preparative t i c ( s i l i c a gel, 10 x 20 cm, multiple developed in benzene-methanol, 9:1) afforded 199 (2.9 mg) which was recrystallized from methanol, 23 m.p. 213.0-214.5°; [ct]n +29.4° (£0.3, chloroform); uv max (ethanol) - 159 -217 nm (e 7940), 274 nm (e 7330); cd (£ 0.005, ethanol) [6 ]2J2 +2100 (peak); T C D C 1 3 2.08 (s, 1, H-8), 4.06 (t, 1, J 1 # ^ 4 Hz, H-1*). 35 Molecular weight by high resolution mass spectrometry: C^H^^^N^Cl 37 requires: 336.109937. Found: 336.109937; C ^ H ^ O ^ C l requires: 338.107241. Found: 338.107241. 2-Chloro-6-N,N-dimethylamino-9-(3'-C^-[2"-acetamidoethyl]-2",3'-dideoxy-B-D-threo-pentopyranosyl)purine (200) Compound 198 (7 mg), dissolved i n ethanol (10 ml) and acetic anhydride (1 ml), was hydrogenated over platinum oxide (15mg) at 60 psi for 4 h. F i l t r a t i o n and evaporation of the mixture afforded a crude syrup (10 mg) which showed the presence of two components by t i c , 0.20 and R^  0.15, in chloroform-ethanol (9:1). Treatment of the crude material wtih 25% aqueous dimethylamine (0.5 ml) in methanol (0.5 ml) for 3 h at room temperature converted one of the components (R^ 0.15) to the other (R^ 0.20). The reaction mixture was evaporated to dryness and purified by preparative t i c ( s i l i c a gel, 10 x 20 cm, chloroform-ethanol, 23 9:1) to afford 200 (6.3 mg, 79%) as a syrup [a] n -6.2° (c 0.6, chloroform); uv max (ethanol) 217 nm (e 9100), 275 nm (E 8600); cd rnn (d 0.003, ethanol) [8]__, -4450; T 3 2.13 (s, 1, H-8), 4.05 (m, 1, — 2 / 0 N-H), 4.26 (d.d, 1, 2 f l t 10 Hz, 2 g , 3 Hz, H-1'), 6.50 ( s , 6 , N(CH„) 0), 8.00 (s, 3, NAc). Molecular weight required for C_-H_-N,0,,C1: 5 2. 1 0 25 O 5 382,384. Found: 382,384 by mass spectrometry. - 160 -Methyl 3-0-acetyl-2-deoxy-5-0-trityl-q ,g-D-erythro-pentofuranoside (201) A mixture of 175 and 176 (310 mg) in pyridine ( 3 ml) and acetic anhydride (1 ml) was stirred at room temperature for 24 h. Water (10 ml) was added and the resultant solution extracted with chloroform (3 x 10 ml). Combination, drying (magnesium sulfate), and evaporation of the organic phase afforded 201 (292 mg, 85%) as a clear syrup, which was used directly without further purification. Attempted Model Nucleoside Synthesis Using 201 in the Direct Fusion Method with 2,6-Dichloropurine to Afford 2,6-Dichloro-9(?)(triphenyl-methyl)purine (194) 2,6-Dichloropurine (35 mg) was intimately mixed with 201 (77 mg) and the mixture azeotroped, under diminished pressure, with toluene ( 3 x 3 ml) then evaporated to dryness. The reaction mixture was placed in an o i l bath, preheated to 150°, and fused at that temperature for 10 minutes under 30 mm pressure, then at 0.1 mm for a further 45 minutes. After cooling dichloromethane (10 ml) was added to dissolve the brown glass. Tic [ s i l i c a gel, benzene-methanol (95:5)] revealed the presence of at least seven components 0.95 (triphenylcarbinol), 0.72, 0.46, 0.43, 0.23, 0.12, 0.05 (2,6-dichloropurine) with the major component at R^  0.43. Preparative t i c using the aforementioned solvent system afforded the major component 194 (16 mg, 16%) as a crystalline solid, which was recrystallized from ethanol, m.p. 221.0-221.5°; T 3 1.83 (s, 1, H-8), 2.5-3.0 (m, 15, 3 x phenyl). Molecular weight Calc. for C^^E^^^Cl^' 430,432,434. Found by mass spectrometry: 430,432,434. - 161 -Anal. Calc: for C^H^N^Cl^ C, 66.75; H, 3.77; N, 12.98. Found: C, 66.77; H, 3.92; N, 12.74. Attempted Selective Detritylation by Acid Hydrolysis of 175, 176 and 178 (a) A solution of 175 and 176 (200 mg) in 80% acetic acid solution was warmed for 0.5 h at 60°, then cooled. Water (2 ml) was added and the solution was then filtered and evaporated to afford the product (82 mg) as a syrup which was insoluble in chloroform. The nmr of the syrup in D^O revealed the absence of the methyl and the triphenylmethyl ether blocking groups. (b) A solution of 175 and 176 (30 mg) i n 66% acetic acid solution was stirred at room temperature. Monitoring of the reaction by t i c ( s i l i c a gel, ethyl acetate) indicated that while detritylation did take place, hydrolysis of the methyl ether also proceeded since no trace of 172 was observed. (c) A solution of 175 and 176 (88 mg) in methanol (3 ml), acetic acid (3 ml), and water (3 ml) was stirred at room temperature for 24 h. No significant hydrolysis of either the triphenylmethyl group or the methyl ether could be detected by t i c . (c) To 175 and 176 (120 mg) was added 0.15% methanolic hydrogen chloride (8 ml). The reaction mixture was stirred at room temperature un t i l a l l starting material was consumed (0.4 h). Silver carbonate (420 mg) was added and the suspension stirred for a further 0.2 h. Fil t r a t i o n and evaporation of the solvent afforded a syrup which was purified by preparative t i c [ s i l i c a gel, ethyl acetate-ethanol (1:1)] to afford a clear syrup (19 mg). Nmr showed similarities to the - 162 -glycosidation product 172, however the product failed to react with triphenylmethyl chloride i n pyridine in 5 days. (e) Compound 188 (82 mg) dissolved in 80% acetic acid (3 ml) was warmed to 60° for 5 min then cooled to room temperature over a period of 0.5 h. Tic [ s i l i c a gel, benzene-methanol (99:1)] revealed that starting material s t i l l remained thus the heating and cooling process was repeated. Water (3 ml) was then added to the solution which was then fi l t e r e d to remove triphenylcarbinol. Evaporation of the solvent under diminished pressure and azeotroping with methanol afforded a syrup (26 mg). Nmr revealed the absence of the methyl ether; i r (film) 2250 cm"1 (C=N). (f) A solution of 175 and 176 (50 mg) in acetic anydride (1 ml) and BF^-etherate (0.05 ml) was stirred for 0.25 h and then poured into ice-water. The solution was extracted with dichloromethane (2 x 10 ml). The organic phase was washed with a saturated solution of sodium hydrogen carbonate, dried (magnesium sulfate) and evaporated to yield a syrup (17.5 mg) which did not exhibit a methyl ether in i t s nmr CDC1 -1 spectrum; x 3 8.00 (m, OAc); i r (film) 1750 cm (C=0). (g) Compounds 175 and 176 (50 mg) in chloroform (0.5 ml) and 90% trifluoroacetic acid (0.03 ml) were stirred at room temperature for 5 minutes after which time reaction was complete as evidenced by t i c [ s i l i c a gel, benzene-methanol (9:1)], to afford approximately six products. This reaction was not further investigated due to the large number of products. (h) A solution of 175 and 176 (50 mg) was dissolved in p_-dioxane (0.5 ml) and 80% acetic acid solution (0.5 ml) and stirred at room - 163 -temperature. After 2 h a l l starting material s t i l l remained. The solution was then warmed at 60° for a further 2 h after which time a l l 175 and 176 had been consumed to afford a major product 0.10 [ s i l i c a gel, benzene-methanol (9:1)] which did not correspond to the required hydrolysis product 172 (R^ 0.18). Attempted Hydrogenolysis of the 5-0-Triphenylmethyl Ether Group of 175, 176 and 188 (a) Compounds 175 and 176 (134 mg) in ethanol (15 ml) were hydrogenated for 4 days at 50° at atmospheric pressure using 10% palladium on calcium carbonate (120 mg) as catalyst. No significant change was observed by t i c . Similarly when 188 (44 mg) was treated under the same conditions, no change was observed. (b) To an anomeric mixture of 175 and 176 (100 mg) in ethanol (30 ml) was added 5% palladium on carbon (150 mg). The mixture was hydrogenated for 58 h under 60 psi hydrogen pressure. Tic [ s i l i c a gel, benzene-methanol (95:5)] showed no significant consumption of starting material. (c) A solution of 175 and 176 (50 mg) in ethanol (25 ml) and acetic acid (0.03 ml) was hydrogenated at room temperature over 5% palladium on carbon (50 mg) for 16 h. A small amount of detritylation was observed. CDC1 to afford 172 (9 mg) isolated by preparative t i c ; T 3 4.95 (m, 1, H-l), 6.60 (s, 3, OMe). (d) A solution of 175 and 176 (100 mg) in ethanol (50 ml) and acetic acid (1 ml) with 5% palladium on carbon (140 mg) was hydrogenated - 164 -at 60 psi for 72 h. Tic of the reaction mixture indicated a small amount (10%) of detritylation i n addition to the formation of another component 0.64 [ s i l i c a gel, benzene-ethyl acetate (8:2)]. Prolonged hydrogenation did not significantly increase the yield of 172). Attempted Reductive Removal of the 5-0-Triphenylmethyl Ether Group of 175, 176 and 188 (a) To a solution of 175 and 176 (250 mg) in tetrahydrofuran (2 ml) held at -70°, was added liquid ammonia (10 ml), then freshly cut lithium (30 mg). The blue coloured solution slowly became white on stirr i n g under anhydrous conditions. Aliquots were removed after 1 h and 2 h revealed the presence of starting material, hence a further amount (15 mg) of lithium was added. After a further 1 h a l l 175 and 176 had been converted to 172 as evidenced by t i c ( s i l i c a gel, ethyl acetate). (b) To a solution of 188 (40 mg) in tetrahydrofuran (1 ml) at -70° was added liquid ammonia (4 ml) and freshly cut lithium (20 mg). The solution was removed from the cooling mixture unt i l an intense blue colour dominated the solution (1-2 minutes). On further st i r r i n g at -70° the soluton changed to bright red then f i n a l l y became white. After this time t i c indicated that a l l 188 had been consumed. Ammonium chloride (300 mg) was added and the solution was allowed to warm to room temperature over a period of 3 h. Dichloromethane was added and the mixture was fi l t e r e d . Evaporation of the solvent afforded a syrup, which contained at least 6 components as evidenced by t i c [ s i l i c a gel, benzene-methanol (99:1)]; i r (film) 3400 cm"1 (OH, NH_). - 165 -To the above syrup in pyridine (5 ml) was added acetic anhydride (0.05 ml) and stirred for 36 h. The mixture was then poured into ice-water (25 ml) and extracted with dichloromethane (4 x 10 ml). The combined organic phases were dried (magnesium sulfate), and evaporated to afford a crude, brown o i l (48 mg). Tic ( s i l i c a gel, ethyl acetate) revealed the presence of at least 7 components. Preparative t i c afforded the major components, triphenylmethane (19 mg) and an acetylated methyl - i - i rnn glycoside (4 mg); i r (film) 1750 cm (OAc), 1670 cm (NAc); T 3 6.65 (s, OMe), 7.90-8.00 (3s, OAc). Methyl 2-deoxy-5-0-bis (p_-methoxyphenyl)phenylmethyl-a,B~erythro-pentofuranoside (202) To chloro-bis(p_-methoxyphenyl)phenylmethane (6.9 g) in anhydrous pyridine (60 ml) was added 172 {2.9 g). After 20 h t i c [ s i l i c a gel, benzene-ethyl acetate (9:1)] indicated that the reaction was complete. Ice-water (500 ml) was added, stirred, then decanted from the orange gum. Fresh water (400 ml) was added and again decanted. The remaining gum was dissolved in chloroform (300 ml), dried (magnesium sulfate), and evaporated to afford a yellow syrup (9 g). A portion (5.6 g) of this syrup was applied to a column of s i l i c a gel (500 g, Davison, 60-200 mesh) and eluted with benzene:ethyl acetate (9:1), to afford only bis-(p_-methoxyphenyl)phenyl-carbinol (4.1 g) corresponding to almost quantitative unblocking. Chromatography of 202 (620 mg) on A l u s i l [ s i l i c a gel-alumina (1:1); 35 g, benzene-ethyl acetate (8:2)] afforded bis(p_-methoxyphebyl)phenyl carbinol (301 mg) and unresolved 202 (244 mg). - 166 -Chloro-bis(p-methoxyphenyl)phenylmethane was prepared by the method c „. _ . 222a of Khorana et a l . Potassium Permanganate Oxidation of the Unsaturated Esters (10) and (11) to Yield 3-C-[S- and R-Hydroxy(methoxycarbonyl)methyl]-l,2:5,6-di-O-isopropylidene-a-D-glucofuranose (211) and (212) Respectively, and 3-C-(Methoxydicarbonyl)-l,2:5,6-di-0_-isopropylidene-a-D-glucofuranose (213) A mixture of the unsaturated esters 10 and 1J (2.6 g) in water (20 ml) and pyridine (40 ml), maintained internally at -5°, was treated dropwise with vigorous st i r r i n g , with a solution of potassium permanganate (1.4 g) in water (40 ml), added over a period of 20 min. The reaction mixture was extracted with chloroform (5 x 200 ml). The combined organic extracts were washed with water, dried over Sodium sulfate and evaporated to yield a yellow syrup (2.2 g). Column chromatography on t . l . c . grade S i l i c a Gel (20 g, column dimensions: 4 x 25 cm) packed and eluted with benzene-ethyl acetate (3:1), under a pressure of 8 psi, afforded the a-keto ester (213) (75 mg, 3%), ketose hydrate (203) (0.41 g, 19%), diol (212) (0.34 g, 12%) and diol (211) (0.89 g, 31%). An analytical sample of 213 was prepared by molecular d i s t i l l a t i o n : 99 -1 -1 [a]£ +78.2° (c 0.7, chloforom); i r (CHC13) 3400 cm 1 (OH), 1720 cm rnn (C=0) (CO.CH,); T 3 3.99 (d, 1, J. _ 4 Hz, H-1), 5.06 (d, 1, J. , 15 l,z 4,5 7 Hz, H-4), 5.45 (d, 1, J 4 Hz, H-2), 6.01 (s, 1, OH) [Ref. 161]. z, 1 - 167 -Analytical samples of 211 and 212 were prepared by molecular d i s t i l l a t i o n at 105°/0.1 mm. Diol 212: R f = 0.25 ( s i l i c a gel, benzene-ethyl acetate (3:1), [ a ] 2 2 +19° (c 1.6, chloforom); i r (film) 3480 cm'1 (OH), 1740 cm"1 (C0 2CH 3); ord (C 0.07, ethanol) [<j>]210 +2140°, t<f>]216 0°, [<j,]220 -1960°, [ $ ] 2 2 8 -3510° (trough), -3400°, [<fr]25() -1080°, [ $ ] 3 0 Q -258°; cd (c 0.13 ethanol) [e]205 -6410°, [e]^ -7640°, [e]n2 -7760° (trough), [e]22Q -5730°, [Q]23Q -2180°; T C D C 1 3 4.13 (d, 1, J1 2 4 Hz, H-l), 5.40-6.05 (m, 5), 6.20 (s, 3, C0 2CH 3), 6.30 (s, 1, OH, exchanges in D 20), 6.36 (s, 1, OH, exchanges in D 20). Anal. Calc. for C^H^Og: C, 51.72; H, 6.94. Found: C, 51.59; H, 6.99. Diol 211: Rf = 0.15, [a]* 2 +54° (c_ 1.5, chloroform); ord (c 0.07, ethanol) [<j,]210 +950°, L<j,]220 +3530°, U] 2 2 4 +3770° (peak), [ ^ 2 3 Q +3530°, U l 2 5 0 +1720°, [<f>]30 +620°; cd (c 0.10, ethanol), [e]2Q5 +2800°, [e]20g +5280° (peak), [e]22Q +3135°, [6] 2 3 0+660°; T ^ 3 4.15 (d, 1, J1 2 3.8 Hz, H-l), 5.56 (d, 1, J_2 1 3.8 Hz, H-2), 5.60-6.10 (overlapping peaks), 6.18 (s, e, C0 2CH 3), 6.54 (s, 2, two OH, exchanges in D 20). Anal. Calc. for C 1 5H 2 4O g: C, 51.72; H, 6.94. Found: C, 51.50; H, 6.93. 3-C-[R-Methanesulfonyloxy(methoxycarbonyl)methyl]-l,2:5,6-di-0-isoppopylidene-a-Q-glucofuranose (216) Methanesulfonyl chloride (0.960 g) was added dropwise to a solution of 212 (0.965 g) in pyridine (15 ml) at 0°. After the solution was stirred overnight at room temperature dichloromethane (50 ml) and ice-water (50 ml) were added and the resultant aqueous layer extracted - 168 -with dichloromethane (2 x 25 ml). The combined dichloromethane extracts were washed with saturated sodium bicarbonate solution (25 ml), water (25 ml), dried over calcium sulfate and evaporated under reduced pressure to a brown o i l which was then column chromatographed on t i c grade S i l i c a gel (50 g, column dimensions 4 x 15 cm) under a pressure of 8 psi packed and eluted with benzene-ethyl acetate (4:1) to afford 216 (0.917 g, 80%). An analytical sample was prepared by recrystalliza-tion from ether-n-hexane and sublimation at 130°/0.1 mm; m.p. 126.5-127.0°; [ a ] * J +54.8° (c 1, chloroform); T 3 4.03 (d, 1, J± 2 3.5 Hz, H-1), 4.74 (s, H-1'), 5.56 (d, 1, J 3.5 Hz, H-2), 5.65-6.05 (over-lapping peaks, 4), 6.14 (s, 3, C0 2CH 3), 6.22 (s, OH, exchanges in D 20), 6.85 (s, 3, -S0 3CH 3), 8.47-8.68 (4s, 12, 4CH3). Anal. Calc. for C. ,Ho,0.... S: C, 45.07; H, 6.15. Found: c , 44.94; lb zo 11 H, 6.23. 3-C- [^-Methanesulfonyloxy(methoxycarbonyl)methyl]-1,2:5,6-di-0-isopropylidene-a-g-glucofuranose (215) To the diol 211 (0.700 g) in pyridine (8 ml) at 0°C, was added methanesulfonyl chloride (0.700 g). After the solution was stirred for 18 h dichloromethane (25 ml) and ice-water (25 ml) were added. The resulting aqueous phase was then extracted with dichloromethane (2 x 25 ml). The combined organic extracts were subsequently washed with a saturated sodium bicarbonate solution (25 ml) followed by water (25 ml), dried over calcium sulfate, f i l t e r e d and evaporated under reduced pressure to yield an orange solid (0.806 g). Pressure chromato-graphy on t i c grade s i l i c a gel (40 g) packed and eluted with benzene-ethyl acetate (3:1) afforded 215 as a clear syrup which crystallized on - 169 -standing (0.641 g, 74%). An analytical sample was prepared by recrystallization from ether-n-hexane and sublimation at 145° and 0.1 25 mm pressure; m.p. 160.0-160.5°; [ct] D +39.2° (c 0.4, chloroform); i r (film) 3400 cm"1 (OH), 1740 cm"1 (00,0^); T C D C 1 3 4.10 (d, 1, J± 2 4 Hz, H-l), 4.44 (s, H-l'), 5.36 (d, 1, J 4 5 6 Hz, H-4), 5.56 (d, 1, J„ _ 4 Hz, H-2), 5.55 (q, 1, J c , 6 Hz, H-5), 5.86 (m, 2, H-6), 6.19 (s, 3, C0 2CH 3), 6.54 (s, 1, OH, exchanges with D 20), 6.77 (s, 3, -S0 3CH 3), 8.53-8.70 (4s, 12, 4CH3). Anal. Calc. for C,-H.,0..,S: C, 45.07; H, 6.15. Found: C, 44.95; l o Z o 1 1 H, 6.09. Methyl-D-2-(1,2:5,6-di-C^-isopropylidene-a-g-glucofuranos-3-yl)glycinate (218) and Methyl L-2-(1,2:5,6-di-0-isopropylidene-a-g-glucofuranos-3-yl)glycinate (217) Mesylate 215 (300 mg) and sodium azide (300 mg) in DMF (20 ml) were warmed for 40 h at 55-60° (not higher) in the dark after which time the reaction mixture was evaporated to dryness. The residue was suspended in dichloromethane and f i l t e r e d . Evaporation of the dichloromethane afforded a clear syrup (295 mg) which revealed two spots on t i c , 0.66 and 0.30 ( S i l i c a Gel, benzene-ethyl acetate (4:1)), the faster moving spot corresponding to the azido-sugar 219, and the slower component to unreacted mesylate. Benzene (15 ml) and 5% Pd on charcoal (150 mg) were added immediately and the mixture hydrogenated at room temperature and atmospheric pressure for 1.25 h. Fi l t r a t i o n and evaporation of the solution afforded a clear syrup (293 mg) which showed that the faster moving a-azido-ester had been reduced - 170 -to give two ninhydrin-positive components 0.58 (major) and R^  0.47 (minor) (Sili c a Gel, ethyl acetate) whilst the mesylate remained unchanged. Preparative t i c (6 plates of 20 x 20 cm) afforded the cx-amino esters (96 mg, 70% based on mesylated consumed) and unreacted mesylate (126 mg). Column chromatography of the partly purified a-amino esters on t i c grade S i l i c a Gel (60 g) packed and eluted with benzene-ethyl acetate (1:9) under 8 psi pressure, afforded two pure compounds. Compound 218 (18 mg, 17%) was twice d i s t i l l e d at 105°/0.1 mm; O O _1 _1 [a]* +33.5° (c_ 2.3, chloroform); i r (film) 3400 cm (OH, NH2), 1740 cm (C02Me); cd (c 0.19, 95% ethanol) [Z]2Q3 -4070°, [ e ] n o -5430°, [ 6 ] ^ rnn -5700°-(trough), [ 6 ] 2 2 0 -4710°, [ 6 ] ^ -1550°; T 4.13 (d, 1, J x 2 4 Hz, H-l), 5.57 (d, 1, J 2 1 4 Hz, H-2), 5.50-6.15 (overlapping multiplets), 6.21 (s, 3, C0 2CH 3), 7.45 (broad singlet, 3, OH and NH2 exchanges in D 20), 8.48 (s, CH 3), 8.56 (s, CH 3), 8.67 (s, CH3), £.69 (s, CH3). Irradiation at x 4.13 collapsed the doublet at x 5.57 to a singlet. Anal. Calc. for C._Ho-No0: C, 51.88; H, 7.25; N, 4.03. Found: I J 2.0 o C, 51.91; H, 7.29; N, 3.95. Compound 217 (59 mg, 51%) was twice d i s t i l l e d at 105°/0.1 mm; [ a ] 2 7 +53° (c 1, chloform); i r (film) 3400 cm"1 (OH, NH2), 1740 cm"1 (C02Me); cd (c 0.19, 95% ethanol) [ e ] 2 Q 3 +3620°, [ 9 ] 2 0 7 +4070° (peak), rnn [ 6 ] 2 2 0 +2260°, [ e ] 2 3 Q +450°; T L 3 4.16 (d, 1, J 3.5 Hz, H-l), 5.71 (d, 1, J 2 1 3.5 Hz, H-2), 5.45-6.15 (overlapping multiplets), 6.25 (s, 3, C0 2CH 3), 7.16 (broad s, 3, OH, NH2> exchanges in D 20), 8.56 (s, two CH 3), 8.64 (s, CH 3), 8.74 (s, CH3). Irradiation at x 4.16 collapsed - 171 -the doublet atx 5.71 to a singlet. Anal. Calc. for C 1 cH o c0 oN: C, 51.88; H, 7.25; N, 4.03. Found: C, 51.77; H, 7.26; N, 3.93. L-2-(1,2:5,6-Di-O-isopropylidene-a-g-glucofuranos-3-yl)glycine (220) A solution of the a-amino ester 217 (48 mg) in 1.25% aqueous methanolic sodium hydroxide (2 ml of 1:1 solution) was stirred for 25 min (reaction completed as indicated by t i c on S i l i c a Gel with ethyl acetate as solvent, the free acid remains at the origin) then passed through 15 ml of Rexyn RG 51 (H) (Polystyrene Carboxylic Acid Type Resin) which was prewashed with 1% acetic acid then water u n t i l the effluent was neutral. The column was eluted with water and the fractions which gave a positive ninhydrin test were combined and evaporated under reduced pressure to yield the crystalline amino acid 220 (39 mg, 85%). An analytical sample was recrystallized from ethanol; 99 m.p. 185.5-186.5° (dec); [a]jj +51.5° ( c l , water); cd (c_0.15, 0.5 N HCl in 95% ethanol), [ 6 ] ^ +3860°, [ 8 ] 2 1 2 +5110° (peak), [ 6 ] ^ +3750°, [ 6 ] 2 3 ( ) +1480°, [ 6 ] 2 4 0 +340°. The cd was taken within 10 min; x D2° (external TMS) 3.99 (d, 1, ^ 2 3.8 Hz, H-1), 5.23 (d, 1, J 2 ± 3.8 Hz, H-2), 5.59-6.00 (m, 4), 5.92 (s, 1, H-1'), 8.43 (s, CH 3), 8.49 (s, CH 3), 8.55 (s, CH3), 8.61 (s, CH3). Anal. Calc. for C,.HooN0o.l/2Ho0: C, 49.12; H, 7.06; N, 4.09. Found: C, 49.23; H, 6.84; N, 4.14. - 172 -g -2 - (1 ,2 :5 ,6-Dl-O-isopropylidene-a - g-glucofuranos -3-yl)glycine (221) A solution of 218 (17 mg) in 1.25% aqueous methanolic sodium hydroxide (0 .5 ml of 1 : 1 solution) was stirred at room temperature. Tic ( S i l i c a Gel, ethyl acetate) indicated that the reaction was complete in 15 min. Elution with water through 10 ml of Rexyn RG 51 (H) (Polystyrene CArboxylic Acid Type Resin) which was prewashed with 1% acetic acid then with water and evaporation of the fractions which afforded a ninhydrin positive test resulted in crystalline amino acid 221 (12 mg, 75%). An analytical sample was recrystallized from ethanol; m.p. 1 9 3 . 5 - 1 9 5 . 0 ° ; [ a ] 2 7 +35° (c 0 . 9 , water); cd (c 0 . 1 4 , 0 . 5 N HCl in 95% ethanol) [ 6 ] ^ - 2 8 1 0 ° , [ e ' 2 1 2 -4650° (trough), [ e ] 2 2 Q - 3 7 9 0 ° , [ e ] 2 3 0 - 1 5 9 0 ° , [ 6 ] 2 4 0 - 2 4 5 ° . The cd was determined within 10 min; D 2 ° (external TMS) 3 .92 (d, 1 , J 4 Hz, H - 1 ) , 5.37 (d, 1 , J T X j 2. f X 4 Hz, H -2 ) , 5 . 4 0 - 6 . 0 5 (overlapping peaks), 8 .40 (s, CH 3), 8 . 5 1 (s, CH3), 8 .60 (s, two CH3). Anal. Calcd. for C^H^NOg^^O: C, 4 5 . 5 1 ; H, 7 . 3 7 ; N, 3 . 8 0 . Found: C, 4 5 . 7 2 ; H, 7 . 2 7 ; N, 3 . 6 0 . Methyl L - 2 - ( 1 , 2 : 5 , 6-di - 0-isopropylidene-a - g-glucofuranos - 3-yl)-N-benzoylglycinate (225) To 217 (25 mg) in methanol ( 0 . 5 ml) was added benzoic anhydride (25 ml). The reaction mixture was stirred at room temperature for 1.5 hrthen evaporated to dryness. The resultant syrup was column chromatographed on t i c grade s i l i c a gel [4 g, benzene-ethyl acetate (1 :1 ) ] to afford 225 (31 mg, 95%) as a solid glass which was sublimed at 130 *70 . 1 mm, m.p. 63 -66°; [<x]J +12.3 (£ 1 . 2 , chloroform); T 3 I - 173 -4.08 (d, 1, J 3.5 Hz, H-l), 4.88 (d, 1, N_ f i 6 Hz, H-l'), 5.57 (d, 1, H-2), 5.55 (s, 1, OH), 6.20 (s, 3, C02Me). Anal. Calc. for C 2 2 H 2 9 N ° 9 : C ' 5 8' 5 3'» H» 6«47; N» 3.10. Found: C, 58.83; H, 6.57; N, 3.27. Methyl D-2-(1,2:5,6-di-O-isopropylidene-a-D-glucofuranos-3-yl)-N-benzoylglycinate (224) To 218 (28 mg) in methanol (0.5 ml) was added benzoic anhydride (30 mg). The reaction mixture was stirred at room temperature for 2 h then evaporated to dryness. The resultant syrup was passed through a column of t i c grade s i l i c a gel [4 g, benzene-ethyl acetate (1:1)] to yield 224 (38 mg, 100%) as a syrup which crystallized from ethanol-n-hexane overnight at 0°, m.p. 151.5-152.5°; [ a ] 2 3 +39.5° (c 2.6, rnn chloroform); T 3 4.06 (d, 1, J. _ 4 Hz, H-l), 4.55 (d, 1, J., u J - , £ 1 » « — n 9 Hz, H-l'), 5.42 (d, 1, H-2), 5.70 (s, 1, OH, exchanges in D 20), 6.20 (s,3, C02Me). Anal. Calc. for C^H^NOg: C, 58.53; H, 6.47; N, 3.10. Found: C, 58.55; H, 6.60; N, 3.07. Methyl L-2-(1,2:5,6-di-0-isopropylidene-a-g-glucofuranos-3-yl)-N-acetamidoglycinate (223) To amino ester 217 (110 mg) in methanol (3 ml) was added acetic anhydride (0.1 ml). The reaction mixture was stirred u n t i l a l l 217 had beenconsumed (1 h) as evidenced by t i c [ s i l i c a gel, dichloromethane-ethyl acetate-ethanol (5:5:1)], evaporated to dryness and column chromatographed on t i c grade s i l i c a gel (45 g, solvent as above) - 174 -under a pressure of 8 psi, to afford 223 (115 mg, 92%) as a hard 23 glass which was sublimed at 125°/0.1 mm, m.p. 62-62°; [a]^ +74° rnn (c 1.9, dichloromethane); T 3 3.08 (broad d, 1, J.T „ . , 6.5 Hz, N-H), N — n , l 4.17 (d, 1, J x 2 3.5 Hz, H-1), 5.03 (d, 1, H-1'), 5.65 (d, 1, H-2), 5.70 (s, 1, OH, exchanges in D20), 6.22 (s, 3, CO^le), 7.96 (s, 3, OAc). Anal. Calc. for C^H^OgN: C, 52.43; H, 6.99; N, 3.60. Found: C, 51.24; H, 7.27; N, 3.53. Methyl D-2-(1,2:5,6-di-O-isopropylidene-a-D-glucofuranos-3-yl)-N-acetamidoglycinate (222) To amino ester 218 (20 mg) in methanol (0.5 ml) was added acetic anhydride (0.02 ml). The solution was stirred until a l l 218 had been consumed (1 h) as evidenced by tic [silica gel, benzene-ethyl acetate (1:1)], evaporated to dryness and purified by preparative tic (one 20 x 20 cm silica-gel plate, above solvent system) to afford 222 (15 mg, 68%). Compound 222 was sublimed at 110°/0.1 mm to afford 22 a hard glass, m.p. 54-56°; [a] D +49.5° (c 0.2, dichloromethane); rnn x 3 3.10 (broad d, 1, J N _ R ^  10 Hz, N-H), 4.11 (d, 1, J 1 2 4 Hz, H-1), 4.82 (d, 1, H-1'), 5.52 (d, 1, H-2), 5.72 (s, 1, OH, exchanges in D20), 6.20 (s, 3, C02Me), 8.00 (s, 3, OAc). Anal. Calc. for C._H0_O.N: C, 52.43; H, 6.99; N, 3.60. Found: C, 52.18; H, 7.07; N, 3.39. - 175 -Wittig Reaction of 1,2:5,6-Di-0-isopropylidene-a-D-ribo-hexofuranos-3-ulose (9) to Yield IS and Z.-3-C-(Methoxycarbonyl)methylene-3-deoxy-l,2:5,6-di-0-isopropylidene D-ribo-hexofuranos (11) and (10) and E and Z-3-C - (Methoxycarbonyl)methylene-3-deoxy-l,2:5,6-di-0-isopropylidene-ot-g-xylo-hexof uranose (205) and (204) Anhydrous ketose (9) (20 g) was reacted with phosphonoacetic acid trimethyl ester (17.5 g) and potassium t-butoxide (9.5 g) in N,N-dimethyl formamide (60 ml). The product (17.9 g) was chromatographed on s i l i c a gel ("Davison", 60-200 mesh, column dimensions: 45 x 7.5 cm) using benzene-ethyl acetate 4:1 as developer. The more mobile zone consisted of 10 and 11 in a yield of 45% (11.0 g) and has been previously characterized. The less mobile zone (2.8 g, 12%) was shown to consist of two new unsaturated sugars (205) and (204) in a ratio of 5:1 as shown by nmr. The major isomer readily crystallized from the mixture in n-hexane and is tentatively assigned the structure (205): m.p. 68.0-69.0°; [ a ] 2 9 -67.0 (c 1, chloroform); T C D C 1 3 3.86 (d.d, 1, J , 0.5 Hz, J , , 2 Hz, H-l*), 4.13 (d, 1, J 4 Hz, H-l), 4.60 (d.d, 1, J. , 6.5 Hz, H-4), 5.15 (d.d, 1, H-2), 5.43 (q, 1, J c , 6.5 Hz, H-5), 6.10 (d, 2, H-6), 6.25 (s, 3, C0 2CH 3). Irradiation at T 4.13 collapsed the double doublet at T 5.15 to a doublet (J 0.5 Hz), similarly irradiation at T 4.60 collapsed the signal at T 3.86 to a narrow doublet (J 0.5 Hz). Anal. Calc. for C 1 5H 2 ?0 7: C, 57.32; H, 7.05. Found: C, 57.23; H, 6.95. - 176 -3-C- [R-Hydroxy (methoxycarbonyl)methyl ] -1,2:5,6-di-0-isopropylidene-a -D-galactofuranose (226) Oxidation of pure 205 (1.71 g) in water (20 ml) and pyridine (40 ml) was carried out according to the previously described procedure at -10° using potassium permanganate (0.95 g) in water (40 ml) added dropwise as oxidant. The reaction mixture was extracted with chloroform (6 x 150 ml). The combined organic extracts were washed with water, dried over sodium sulfate and evaporated to yield a pale yellow syrup (1.04 g, 55%) which crystallized on trituration with ethanol. An analytical sample of 226 was prepared by recrystallization from ethanol; m.p. 145.5-146.5°; [ex] 2 9 -12.0° (c 1, chloroform); cd (£0.3, ethanol) t e ] 2 Q 2 -1930, [ e l n o -2340 (trough), [ 6 ] ^ -1730, [ 6 ] ^ 0, [ 6 ] ^ r n n 780 (peak), [ e ] 2 5 0 0; T x 3 4.12 (d, 1, 3± 2 4 Hz, H-l), 5.32 (doublet of triplets, 1, J c . 8 Hz, J- . 6 Hz), 5.40 (d, 1, H-2), 5.68 5,6 5,4 (d, 1, J-, - , 9 Hz, collapses to a singlet on D-0 addition, H-l'), 1 , 1 . UH 2. 5.82-6.25 (multiplet, 2, H-6), 6.06 (s, 1, C-3 OH, exchanges in D 20), 6.14 (s, 3, C0 2CH 3), 6.33 (d, 1, C-l' OH, exchanges in D 20). Anal. Calc. for C^H^Og-. C, 51.72; H, 6.94. Found: C, 51.58; H, 7.07. 3-C-[R-Methanesulfonyloxy(methoxycarbonyl)methyl]1,2:5,6-di-0-isopropylidene-ot-g-galactofuranose (229) and 3-C-]Methanesulfonyloxy(methoxycarbonyl)-methylene]-3-deoxy-l,2:5,6-di-0-isopropylidene-a-D-xylofuranose (230). Methanesulfonyl chloride (0.1 g) was added dropwise to a solution of 226 (83 mg) in pyridine (5 ml) at 0°. After st i r r i n g the mixture overnight at room temperature chloroform (10 ml) and ice-water (10 ml) were added and the resultant aqueous phase extracted with chloroform - 177 -(4 x 10 ml). The combined organic extracts were washed with saturated sodium bicarbonate solution (10 ml), water (10 ml), dried over sodium sulfate, and evaporated under reduced pressure to afford a yellow syrup (120 mg). The product was chromatographed on t i c grade s i l i c a gel (25 g) under a pressure of 8 psi, packed and eluted with benzene-ethyl acetate (3:7), to afford two pure compounds. Compound 230 (29 mg, 30%) was recrystallized from ethanol, m.p. 154-155°; [ a ] 2 3 -8° (c 0.5, dichloromethane); T C D C 1 3 4.08 (d, 1, 2 4.0 Hz, H-1), 4.57 (d, 1, 5 6.5 Hz, H-4), 4.67 (d, 1, H-2), 5.43 (q, 1, J 5 6, 6.5 Hz, J 5 6 „ 7.2 Hz, H-5), 6.10 (s, 3, C0 2CH 3), 6.12 (multiplet, 2, H-6), 6.64 (s, 3, SO^H^). Anal. Calc. for C,,Ho,01ftS: C, 47.06; H, 5.92. Found: C, 47.24; 16 24 10 H, 5.72. Analysis was performed on 229 (61 mg, 60%) as a pure, undistilled 23 syrup which was unstable to d i s t i l l a t i o n , [ a ] n +3.4° (c 0.6, chloroform); rnn T 3 4.06 (d, 1, J± 2 4 Hz, H-1), 4.66 (s, 1, C-l'), 5.48 (d, 1, H-2), 5.4-5.6 (m, 1, H-5), 5.90 (d, 1, J 4 7 Hz, H-4), 5.94-6.35 (m, 2, H-6), 6.12 (s, 3, C0 2CH 3), 6.83 (s, 3, S0 2CH 3). Anal. Calc. for C^H^O^S: C, 45.07; H, 6.15. Found: C, 45.14; H, 6.11. 3-C-[R-Methanesulfonyloxy(methoxycarbonyl)methyl]-l,2-0-isopropylidene-a-p-galactofuranose (228) To 229 (200 mg) was added 65% aqueous acetic acid (20 ml) and the solution stirred for 5 h at room temperature, after which time a l l starting material had been consumed, to afford at least four components - 178 -as evidenced by t i c ( s i l i c a gel-ethyl acetate). The reaction mixture was then evaporated on an o i l pump and azeotroped with toluene (3 x 5 ml) to yield a mobile syrup (170 mg) which after column chromatography on t i c grade s i l i c a gel (45 g), packed and eluted with ethyl acetate under a pressure of 8 psi afforded 228 (0.040 g, 22%), recrystallized 22 from ethanol, m.p. 119.5-120.0° (decomp.); [ a ] D +25.5° (£ 0.1, dichloro-methane); x D2° 3.90 (d, 1, 4 Hz, H-1), 4.34 (s, 1, H-1*), 6.05 (s, 3, C0 2CH 3), 6.62 (s, 3, S0 2CH 3), 8.34 (s, 3, CH 3), 8.62 (s, 3, CH3). Anal. Calc. for c 1 3 H 2 2 0 i i S : C> 4 0 « 4 1 ; H» 5- 7 i*- Found: C, 40.55; H, 5.71. Methyl-L-2-(1,2:5,6-di-O-isopropylldene-a-D-galactofuranos-3-yl)-glycinate (232) and Methyl D-2-(1,2:5,6-di-0-isopropylidene-a-D-galactofuranos-3-yl)glycinate (234) Mesylate (229) (100 mg) and sodium azide (100 mg) in anhydrous DMF (7 ml) were stirred in the dark at 60° for 40 h under anhydrous conditions. The reaction mixture was then evaporated to dryness, taken up in dichloromethane, fi l t e r e d and evaporated to afford a clear syrup which was immediately hydrogenated in anhydrous benzene using 5% Pd on charcoal (50 mg) under one atmosphere of hydrogen for 7 h. F i l t r a t i o n and evaporation of the hydrogenation reaction mixture afforded a clear syrup (77 mg) which revealed two ninhydrin-positive components, 0.52 (major), and R^  0.31 (minor) (s91ica gel, ethyl acetate-dichloromethane-ethanol, 5:5:1). No unreacted mesylate was observed. Column chromato-graphy, of the above material on t i c grade s i l i c a gel (20 g), packed and eluted with the aforementioned solvent system under a pressure of 8 psi, afforded two pure components, 232 and 234. - 179 -Compound 232 (31 mg, 38%) was recrystallized from ethanol; m.p. 158.0-158.5°; [ a ] 2 3 -21° (£0.3, chloroform); cd (c 0.13, 95% ethanol) [ e ] 2 Q 4 4050 (peak), [Bl21Q 3880, [ e ] 2 2 Q 0, [9 J^'-3380 (trough), [ e ] 2 4 Q -2160, [ 9 ] 2 5 5 0; 4.09 (d, 1, J 4 Hz, H-1), 5.38 (triplet of doublets, 1, J, . 8 Hz, J r ,, 8 Hz, J c ,„ 6 Hz, H-5), 5,4 5,0 0,0 5.55 (d, 1, H-2), 5.84-6.12 (m, 2, H-6), 6.16 (d, 1, H-4), 6.21 (s, 3, C0 2CH 3), 6.32 (s, 1, C-l*), 6.9-8.5 (v. broad s, 3, NH2, OH, exchanges in D 20). Anal. Calc. for C 1 cH. c0 oN: C, 51.88; H, 7.25; N, 4.03. Found: 15 25 8 C, 51.70; H, 7.18; N, 3.87. Compound 234 (8 mg, 9.5%) was obtained in pure form by recrystalliza-tion from ethanol, then sublimed at 120° and 0.1 mm pressure, m.p. 175.5-176.5°, [ a ] 2 2 -15.4° (£0.5, chloroform); cd (£0.15, 95% ethanol) [ e ] 2 0 4 -3820, [ e ] 2 Q 6 -4550 (trough), [ e ] 2 2 0 -1150, [ e ] 2 2 3 o, [ e ] 2 3 3 rnn 3400 (peak), [ 6 " 2 5 2 0; x 3 4.05 (d, 1, 2 4.0 Hz, H-1), 5.48 (d, 1, H-2), 5.55 (m, 1, H-5), 5.91 (d, 1, J 7 Hz, H-4), 6.08 (d.d, 1, 4,0 J,,, _ 6 Hz, H-1*), 6.19 (s, 4, H-1', C0„CH o), 6.42 (d.d, 1, J,„ _ 0 , 5 2 5 0 , 0 8 Hz, H-6"), 7.1-8.2 (v. broad s, 3, NH2, OH). Anal. Calc. for C1cHot.0oN: C, 51.88; H, 7.25; N, 4.03. Found: 15 2D O C, 52.17; H, 7.15; N, 4.01. L-2-(1,2:5,6-Di-0-isopropylidene-a-D-galactofuranos-3-yl)glycine (233) The a-amino ester 232 (28 mg) i n 1.25% aqueous methanolic sodium hydroxide (2 ml of 1:1 solution) was stirred for 30 min at room temperature. Tic ( s i l i c a gel, dichloromethane-ethyl acetate-ethanol, 5:5:1) indicated that the reaction was complete after this time. The - 180 -solution was then passed through 5 ml of Rexyn RG 51(H) (Polystyrene Carboxylic Acid Type Resin) which had been prewashed with 1% acetic acid then water u n t i l the effluent was neutral. Elution of the column with water, combination and evaporation of the ninhydrin positive fractions afforded the crystalline amino acid 233 (24 mg, 89%). An analytical sample was recrystallized from ethanol; m.p. 180-181° (decomp.), [a]J -22.7° (c 1, ethanol); cd (c 0.17, 0.5 N HC1 in 95% ethanol) [ 6 ] 2 ( ) 2 +5900, [6] 2 1 0+7100 (peak), [ 6 ] ^ +4540, [ 9 ] ^ +2560, [ 6 ] ^ +400. Anal. Calc. for C^H^OgN: C, 50.46; H, 6.95; N, 4.20. Found: C, 50.37; H, 6.97; N, 4.10. D-2-(1,2:5,6-Di-O-isopropylidene-a-g-galactofuranos-3-yl)glycine (235) A solution of 234 (8 mg) i n 1.25% aqueous methanolic sodium hydroxide (1 ml of 1:1 solution) was stirred at room temperature for 30 min (complete by t i c , s i l i c a gel, dichloromethane-ethyl acetate-ethanol, 5:5:1) and then passed through 5 ml of Rexyn RG 51(H) (Polystyrene Carboxylic Acid Type Resin) which had been prewashed with 1% acetic acid then water u n t i l the effluent was neutral. Elution of the column with water, combination and evaporation under reduced pressure of the ninhydrin positive fraction afforded the crystalline amino acid 235 (7 mg, 88%). An analytical sample was recrystallized 22 from methanol; m.p. 214-215° (decomp.); [ a ] n -13.6° (£0.2, ethanol-water, (1:1)); cd (c 0.14, 0.5 N HC1 in 95% ethanol) [ 9 ] 2 0 2 "5240, [ 9 ] 2 1 0 -6420 (trough), [ e ] 2 2 Q -4780, [ 6 ] ^ -1670. Anal. Calc. for C,.Hoo0oN: C, 50.46; H, 6.95; N, 4.20. Found: 1 4 25 o C, 50.21; H, 6.85; N, 4.18. - 181 -3-C-[R-Acetoxy(methoxycarbonyl)methyl]-l,2:5,6-di-0-isopropylidene-ct-D-glucofuranose (236) Acetic anhydride (20 ml) was slowly added to an ice-cold solution of 212 (2.1 g) in pyridine (40 ml) and stirred at room temperature overnight. Evaporation of the reaction mixture afforded a pale yellow syrup (2.3 g, 100%) which was column chromatographed ( s i l i c a gel H, 60 g, benzene-ethyl acetate, 1:1) under a pressure of 8 psi to afford crystalline 236 (1.96 g, 83%). An analytical sample was recrystallized from ethanol-n-hexane; m.p. 131.0-132.0°; [ a ] n +60.4° (£1.3, chloroform); rnn T 3 4.06 (d, 1, J 3.5 Hz, H-1), 4.52 (s, 1, H-1'), 5.57 (d, 1, x., z H-2), 5.6-6.1 (m, 4, H-4, H-5, H-6), 6.15 (s, 3, CO^H^, 6.28 (s, 1, OH, exchanges with D 20), 7.81 (s, 3, OAc). Anal. Calc. for c 1 7 H 2 6 ° i o : C ' 5 2 , 3 ° 5 H» 6.71. Found: C, 52.09; H, 6.76. 3-C-[Z-l'-O-Acetyl-1'-(methoxycarbonyl)methylene]-3-deoxy-l,2:5,6-di-0-isopropylidene-a-D-ribo-hexofuranose (237) Freshly d i s t i l l e d thionyl chloride (6 ml) was added to a solution of 236 (1.68 g) in pyridine (20 ml) at 0° and stirred i n the dark for 20 h. Addition of ice-water (20 ml to the mixture followed by extraction with dichloromethane (6 x 50 ml) afforded, after drying (calcium sulfate) and evaporation gave an orange syrup (1.8 g). Column chromatography of the product ( s i l i c a gel H, 120 g, benzene-ethyl acetate, 4:1) afforded crystalline 237 (1.11 g, 69%). An analytical sample was 21 recrystallized from n-hexane; m.p. 89.0-89.5°; [ a ] n +103.7° (£1.5, CDC1 dichloromethane); T 3 4.07 (d, 1, J 4.5 Hz, H-1), 4.27 (d.d, 1, 1, 4. J 2 4 1.5 Hz, J 4 5 3.0 Hz, H-4), 4.80 (d.d, 1, H-2), 5.77 (d.q, 1, - 182 -c t 6.0 Hz, J _ ,„ 7.5 Hz, H-5), 6.04 (d.d, 1, J, , ,„ 9 Hz, H-6'), j,o 5,0 o,o 6.20 (s, 3, C0 2CH 3), 6.29 (d.d, 1, H-6"), 7.76 (s, 3, OAc). Anal. Calc. for C^H^Og! C, 54.83; H, 6.50. Found: C, 54.71; H, 6.70. 3-C-[S-Acetoxy(methoxycarbonyl)methyl]-3-deoxy-l,2:5,6-di-0-isopropylidene-ct-D-allofuranose (238) Compound 237 (1.10 g) in ethyl acetate (75 ml) was hydrogenated, using 5% palladium on carbon (500 mg) as catalyst, u n t i l no starting material remained as evidenced by t i c ( s i l i c a gel, benzene-ethyl acetate, 4:1). F i l t r a t i o n and evaporation of the solution afforded pure 238 (1.07 g, 97%) as a clear syrup. An analytical sample was 23 prepared by molecular d i s t i l l a t i o n at 105°/0.1 mm; [ a ] n +55.3 (c 1, rnn dichloromethane); T U A 3 4.22 (d, 1, J 3.9 Hz, H-1), 4.31 (d, 1, i., z J 1 ? 3 4.5 Hz, H-1'), 5.19 (t, 1, J 2 3 4 Hz, H-2), 5.55-6.05 (m, 4, H-4, H-5, H-6), 6.28 (s, 3, C0 2CH 3), 7.58 (d.t, 1, J 3 4 9.5 Hz, H-3), 7.85 (s, 3, OAc). Anal. Calc. for C^H^Og: C, 54.54; H, 7.00. Found: C, 54.38; H, 7.08. 3-C-[S-Hydroxy(methoxycarbonyl)methy1]-3-deoxy-l,2:5,6-di-0-isopropylidene-a-D-allofuranose (239) Acetate 238 (0.88 g) in anhydrous methanol (50 ml) was treated with a solution of 0.1 M sodium methoxide in methanol (1 ml) and stirred under anhydrous conditions for 4 h. To this solution IR 120 (H+) resin (0.5 ml) was added, stirred for 5 minutes, f i l t e r e d and evaporated - 183 -to afford pure 239 (0.76 g, 100%) as a clear syrup. Molecular 23 d i s t i l l a t i o n at 130°/0.1 mm afforded an analytical sample; +59.4° (c 2.4, dichloromethane); cd (jc 0.15, 95% ethanol) [ 6 ] 2 0 2 +1230, [ 9 ] 2 0 8 +1720 (peak), [ 6 ] ^ +1140, [ e ] 2 3 3 -370 (trough); T C D C 1 3 4.22 (d, 1, J 3.6 Hz, H-l), 5.18 (d.d, 1, J 4.5 Hz, H-2), 5.40 (d.d, 1, 1 , / 2., 5 J x , 3 6.5 Hz, J l t 0 R 3 Hz, H-l*), 5.65-6.10 (m, 4, H-4, H-5, H-6), 6.18 (s, 3, C0 2CH 3), 7.57 (m, 1, J g 4 9.5 Hz, H-3). Anal. Calc. for C^H^Og: C, 54.21; H, 7.28. Found: C, 54.21; H, 7.18. 3-C-[S-Methanesulfonyloxy(methoxycarbonyl)methyl]-3-deoxy-l,2:5,6-di-0-isopropylidene-a-D-allofuranose (240) To 239 (387 mg) in pyridine (10 ml) at 0°, was added methanesulfonyl chloride (0.35 ml). Tic ( s i l i c a gel, benzene-ethyl acetate, 6:4) revealed complete reaction after 5 h. Ice-water (50 ml) was added and the solution extracted with dichloromethane (4 x 50 ml) to afford an orange syrup after drying and removal of solvent. Column chromatography of the product ( s i l i c a gel H, 60 g, benzene-ethyl acetate, 6:4) afforded crystalline 240 (438 mg, 92%), m.p. 136.0-136.5°; [ a ] 2 4 +59.4° rnn (c 1, dichloromethane); T 3 4.20 (d, 1, J 4 Hz, H-l), 4.38 (d, 1 > <-1, J l f 3 4.5 Hz, H-l'), 5.19 (t, 1, J 2 3 4 Hz, H-2), 5.61 (m, 1, H-4), 5.80-6.05 (m, 3, H-5, H-6), 6.22 (s, 3, C0 2CH 3), 6.82 (s, 3, S0 2CH 3), 7.48 (d.t, 1, J. . 9.5 Hz, H-3). 3,4 Anal. Calc. for C.-H.-O.. _S: C, 46.82; H, 6.39. Found: C, 46.95; lb ZD 1U H, 6.60. - 184 -i 3-C- [ S_-p_- To luenesulf onyloxy (methoxycarbonyl)methyl ] -3-deoxy-l ,2:5,6-di-O-isopropylidene-a-D-allofuranose (241) To 239 (120 mg) in pyridine (5 ml) at 0° was added p_-toluene-sulfonyl chloride (300 mg) and the mixture stirred at room temperature for 48 h. Water (5 ml) was then added and the mixture extracted with dichloromethane (3 x 20 ml). Evaporation of solvent afforded 241 as an orange solid. Column chromatography ( s i l i c a gel, 30 g, benzene-ethyl acetate, 4:1) afforded 241 (110 mg, 63%) which was recrystallized from 24 r.nn ethanol; m.p. 92.0-92.5°; [a ]J +48.4° (c 1, dichloromethane); x 3 4.27 (d, 1, J 4 Hz, H-l), 4.57 (d, 1, J , 5 Hz, H-l'), 5.29 (t, 1, J 2 3 4 Hz, H-2), 6.40 (s, 3, C0 2CH 3), 7.60 (s, 3, CH 3), 7.80 (m, 1, H-3). Anal. Calc. for C 22 H30°10 S : C ' 5 4 , 3 2 ; H» 6 - 1 3 - Found: C, 54.32; H, 6.22. Methyl D-2-(3-deoxy-l,2:5,6-di-O-isopropylidene-a-D-allofuranos-3-yl)-glycinate (243) and Methyl L-2-(3-deoxy-l,2:5,6-di-0-isopropylidene-a-D-allofuranos-3-yl)glycinate (245) From 240: A solution of 240 (300 mg) and sodium azide (300 mg) in anhydrous dimethylformamide was maintained at 55°, for 40 h, in the dark, with stir r i n g . The solution was evaporated to dryness under vacuum, then extracted with dichloromethane to afford, after evaporation of the f i l t r a t e , a pale yellow syrup, which revealed that a l l mesylate had been consumed to afford crude azido-sugar 242, 0.45, and a small amount of another component, R^  0.27 ( s i l i c a gel, benzene-ethyl acetate, 7:3). The mixture was then immediately hydrogenated in benzene (25 ml) with 5% palladium on carbon (150 mg) at room temperature and atmospheric - 185 -pressure for 6 h. F i l t r a t i o n and evaporation of the f i l t r a t e afforded a clear syrup which showed two ninhydrin positive components by t i c , 0.25 (major component) and 0.19 (minor) [ s i l i c a gel, ethyl acetate]. Column chromatography of the product ( s i l i c a gel, 60 g, ethyl acetate) afforded the pure a-amino esters 243 and 245. Compound 243, (84 mg, 34%) was d i s t i l l e d at 105°/0.1 mm; [ a ] 2 3 -22.3° (c 5.9, dichloromethane); cd (c 0.19, 95% ethanol) [e]_ n,-1370, [6]„,c — Z U O Zlj rnri -750, [ 6 ] 2 2 Q 0, [ 6 ] 2 3 1 +1020 (peak), [ 6 ] 2 5 5 0; T ^ L 3 4.21 (d, 1, J 3.8 Hz, H-l), 5.22 (t, 1, J 4 Hz, H-2), 5.5-6.1 (m, 5, H-l', H-4, H-5, Z,5 H-6), 6.20 (s, 3, C0 2CH 3), 7.55 (m, 1, H-3), 8.15 (broad s, 2, NH2). Anal. Calc. for C^H^C^N: C, 54.37; H, 7.60; N, 4.23. Found: C, 54.52; H, 7.76; N, 4.25. Compound 245, 1 6 1 (54 mg, 22%) was d i s t i l l e d at 105°/0.1 mm; cd (c 0.19, 95% ethanol) [ 0 ] 2 Q 6 +3080, [ 9 ] ^ +1400, [ e ] 2 2 0 0, [ 6 ] m -780 (trough), [ 6 ] 2 5 6 0. From 241: A solution of the p_-toluenesulfonate ester 241 (55 mg) and sodium azide (55 mg) in anhydrous DMSO (2 ml) was stirred in the dark at 55° for 40 h. The solution was evaporated to dryness under vacuum and the residue extracted with dichloromethane. F i l t r a t i o n and evaporation of the solution afforded a light brown o i l which revealed a strong absorption at 2150 cm 1 in i t s i r spectrum. The mixture was immediately hydrogenated in methanol (10 ml), using 5% palladium on carbon (50 mg) as catalyst, at atmospheric pressure for 3 h. Tic ( s i l i c a gel, ethyl acetate) revealed that hydrogenation was complete after this time to afford two ninhydrin positive components [R^ 0.25 (minor component) and R^  0.19 (major)]. F i l t r a t i o n and evaporation of the - 186 -solvent afforded a clear syrup which was chromatographed on a column of t i c grade s i l i c a gel (15 g, ethyl acetate) to afford 243 (5 mg, 10%) and 245 (11 mg, 27%). Methyl D-2-(3-deoxy-l,2:5,6-di-O-isopropylidene-a-D-allofuranos-3-yl)-N-benzoylglycinate (244) To 243 (6 mg) in methanol (0.5 ml) was added benzoic anhydride (6 mg). The reaction mixture was stirred at room temperature for 3 h. then evaporated to dryness to afford a clear o i l which crystallized on standing. Tic revealed only one component [R^ 0.15, s i l i c a gel, benzene-ethyl acetate, 4:1], which was recrystallized from ethanol-n-hexane, m.p. 141-143.5° ( l i t . 2 4 7 m.p. 138-140°). D-2-(3-Deoxy-l,2:5,6-di-O-isopropylidene-a-D-allofuranos-3-yl)glycine (246) To a solution of 243 (22 mg) in methanol (1 ml) was added 2.5% aqueous methanolic sodium hydroxide (1 ml) and the solution stirred for 20 minutes. The solution was then passed through a short column of Rexyn RG 5 1 ( H + ) (Polystyrene Carboxylic Acid Type Resin) (5 ml) which had been prewashed with 1% acetic acid then water u n t i l neutral effluent was detected. Collection and evaporation of the ninhydrin positive fractions afforded 246 (20 mg, 95%) as a hard glass which 23 was recrystallized from ethanol-water, m.p. 198-199° (decomp.); [ o t ] n +56.7° (c 1.2, ethanol-water,1:1); cd (c 0.1, 0.5 M HCl in 95% ethanol) [ e ] 2 Q 5 -2400, [ 6 ] 2 1 2 -2860 (trough), [ e ] 2 2 Q -2160, [ 9 ] 2 3 Q -800, [ e ] 2 4 5 0; x D2° (ext. TMS) 3.48 (d, 1, ^ 2 3.5 Hz, H-1), 4.62 (t, 1, J 2 3 4 Hz, H-2), 6.90 (d.t, 1, H-3). - 187 -Anal. Calc. for C^H^C^N: C, 52.98; H, 7.30; N, 4.41. Found: C, 53.31; H, 7.30; N, 4.62. Attempted Enzymatic Oxidation of Glycosyl Amino Acids 251 To 0.13 M sodium phosphate buffer (pH 7.9) (0.8 ml) containing EDTA (0.62 mM) was added 12.50 mM amino acid solution (0.4 ml), 40 ug catalase (0.4 ml of 1 mg of 10 ml), and L-amino acid oxidase (20 u l of 5 mg in 250 y l ) . The preparation was incubated at 37° for 60 minutes. The reaction was stopped by the addition of 30% trichloroacetic acid (0.7 ml) and the resulting precipitate removed by centrifugation. One m i l l i l i t e r of the deproteinized solution was added to 1 M NaOH (0.5 ml) and adjusted to pH 2.0 by the addition of 1 M NaOH or 1 M HC1. The volume of the solution was f i n a l l y raised to 1.6 ml with water. After the addition of 0.016% 3-hydrazinoquinoline (1.4 ml) in water to the acidified solution, the mixture was incubated for 30 minutes at 37°. The absorbance of the solution was determined at 305 nm 10 minutes after the addition of 0.1 M HC1 (1.0 ml). Using these conditions the following results were obtained: Run A: Tube No. Enzyme Amino Acid Incubation Absorbance 1 present ^-phenylalanine 1 h 0.70 2 " L-phenylalanine 1 h 4.60 3 " none 1 h 0.68 4 absent L-phenylalanine 1 h 0.55 5 present L-phenylalanine 2.5 h 5.20 - 188 -Run B: Tube No. Enzyme Amino Acid Incubation Absorbance 1 present 247 1 h 0.84 2 " 2M 1 h 0.85 3 " 220 1 h 0.86 4 " 221 1 h 0.91 5 " 233 1 h 0.82 6 " 235 1 h 0.81 7 " none 1 h 0.80 8 " 220 1 h 0.85 Conversion of Methyl D-2-(3-deoxy-l,2:5,6-di-0-isopropylidene-a-D-allofuranos-3-yl)glycinate (243) to 3-C-[(R)-(hydroxymethyl)-N-salicylideneamino)methyl]-1,2:5,6-di-O-isopropylidene-a-D-allofuranose (250) To a solution of lithium aluminum hydride (30 mg) in anhydrous tetrahydrofuran (1 ml) was added the glycosyl a-amino ester (243, 25 mg) in tetrahydrofuran (2 ml). The reaction mixture was refluxed for 1.5 h, then stirred at room temperature for 2.5 h. The reaction mixture was decomposed with, 30 ul water, 90 u l 1 M aqueous sodium hydroxide, 30 u l water, stirred for 10 minutes and the granular precipitate removed by centrifugation. Decantation and two washings of the residues afforded the crude hydroxyamine (20 mg). To the above crude product in methanol (0.5 ml) was added s a l i c y l -aldehyde (30 ul). Immediately on this addition the solution became strongly yellow in colour. The solution was permitted to stand overnight - 189 -at room temperature then purified by preparative t i c to afford crude 250 as a yellow o i l (6.9 mg), which was recrystallized from ethanol (30 ul); m.p. 156.5-158.0° ( l i t . 1 6 0 b m.p. 156-157°); [ct] 2 3 +120° j (c 0.012, ethanol); ord (£0.02, ethanol) [ e ] , _ , +2765, ' e ] _ f t , -370. - 190 -REFERENCES 1. H . Grisebach and R. Schmid, Angew. Chem. Int. Ed., 11, 159 (1972). 2. J.S. Brimacombe, Angew. Chem. Int. Ed., 8, 401 (1969). 3. M.W. Miller: The Pfizer Handbook of Microbial Metabolites. McGraw-Hill, New York, 1961, p. 37,38. 4. H . Umezawa: Recent Advances in Chemistry and Biochemistry of Antiobiotics. Microbial Chemistry Research Foundation, Tokyo, 1964, p. 67. 5. W.G. Overend, Chem. and Ind., 342 (1963). 6. S. Hanessian and T.H. Haskell, in W. Pigman and D. Horton: The Carbohydrates, 2nd Edit., Academic Press, New York, 1970, Vol. IIA, p. 139. 7. W. Mayer, W. Kunz, and F. Loebich, Liebigs Ann. Chem., 688, 232 (1965). 8. (a) R.B. Duff, Biochem. J., 94, 768 (1965); (b) E. Beck and 0. Kandler, Z. Naturforsch., 20b, 62 (1965). 9. W. Mayer, and W. Kunz, Naturwissenschaften, 46_, 206 (1959). 10. J. Sellmair, E. Beck, and 0. Kandler, Z. Pflanzenphysiol. 59, 70 (1968). 11. E. Beck, Z. Pflanzenphysiol., 61, 360 (1969). 12. C.C. Kuenzle, Biochem. J. 119, 411 (1970). 13. J.S. Burton, W.G. Overend and N.R. Williams, J. Chem. Soc, 3433 (1965) 14. H. Grisebach, Helv. Chim. Acta, 51, 928 (1968). 15. M. Brufani and W. Keller-Schierlein, Helv. Chim. Acta, 49, 1962 (1966). 16. D.M. Lemal, P.D. Pacht, and R.B. Woodward, Tetrahedron, 18, 1275 (1962) '. . " . " '%.< " _ ",'. - ^ - — " ~ -' - - — — * \ l ' - y* -' w ~ ," ~» £ s w f * (C), 961 (1971). - 191 -18. W. Meyer zu Reckendorf and E, Bischof, Tetrahedron Lett., 2475 (1970). 19. A.K. Ganguly, O.Z. Sarre and H. Reimann, J. Amer. Chem. Soc, 90_, 7129 (1968). 20. A.K. Ganguly, O.Z. Sarre, and S. Samulewicz, Chem. Commun., 746 (1971). 21. W.G. Overend and N.R. Williams, J. Chem. Soc, 3446 (1965). 22. A.D. Ezekiel, W.G. Overend, and N.R. Williams, Tetrahedron Lett., 1635 (1969). 23. B. Flaherty, W.G. Overend, and N.R. Williams, J. Chem. Soc, 398 (1966). 24. R.D. King, W.G. Overend, J. Wells, and N.R. Williams, Chem. Commun., 726 (1967). 25. A.A.J. Feast, W.G. Overend, and N.R. Williams, J. Chem. Soc (C), 303 (1966). 26. J.S. Burton, W.G. Overend, and N.R. Williams, Chem. and Ind., 175 (1961), 27. A. Rosenthal and G. Schollnhammer, Carbohydr. Res., 15, 421 (1970). 28. A. Rosenthal and K.S. Ong, Can. J. Chem., 48, 3034 (1970). 29. A. Rosenthal, K.S. Ong, and D.A. Baker, Carbohydr. Res., JL3, 113 (1970). 30. H.P. Albrecht and J.G. Moffat, Tetrahedron Lett., 1063 (1970). 31. G.J. Laurens, Carbohydr. Res., _17, 35 (1971). 32. A. Rosenthal and L. Nguyen, Tetrahedron Lett., 2393 (1967). 33. A. Rosenthal and P. Catsoulacos, Can. J. Chem., 46, 2868 (1968). 34. A. Rosenthal and M. Sprinzel, Can. J. Chem., 47^ 4477 (1969); 47, 3941 (1969); 48, 3253 (1970). 35. A. Rosenthal and D.A. Baker, Tetrahedron Lett., 397 (1969). 36. A. Rosenthal, D.A. Baker, and M. Sprinzel, Tetrahedron Lett., 4233 (1970). - 192 -37. A. Rosenthal and L. Nguyen, J. Org. Chem., 34, 1029 (1969). 38. J.M.J. Tronchet, J.M. Bourgeois, R. Graf, and J. Tronchet, C.R. Acad. Sci. Paris (C), 269, 420 (1969). 39. J.M.J. Tronchet and J.M. Chalet, Helv. Chim. Acta, 53, 364 (1970). 40. J.M.J. Tronchet and J.M. Bourgeois, Helv. Chim. Acta, 53_, 1463 (1970). 41. J.M.J. Tronchet, J.M. Bourgeois, J.M. Chalet, R. Graf, R. Gurny, and J. Tronchet, Helv. Chim. Acta, 54, 687 (1971). 42. M. Sharma and R.K. Brown, Can. J. Chem., 46^ , 757 (1968). 43. A.A.J. Feast, W.G. Overend, and N.R. Williams, J. Chem. Soc. (C), 7378 (1965). 44. T.D. Inch and G.J. Lewis, Carbohydr. Res., 15, 1 (1970). 45. N.R. Williams, Chem. Commun., 1012 (1967). 46. P.W. Austin, J.G. Buchanan, and E.M. Oakes, Chem. Commun., 374 (1965). 47. B.E. Davidson, R.D. Guthrie, and A.T. McPhail, Chem. Commun., 1273 (1968). 48. L.I. Kudryashov, M.A. Chlenov, and N.K. Kochetkov, Izv. Akad. Nauk, SSSR Khim, 1, 75 (1965); Chem. Abstr., 62, 14793 h (1965). 49. S. Hanessian and P. Dextraze, Can. J. Chem., 50, 226 (1972). 50. A. Rosenthal and H.J. Koch, J. Amer. Chem. Soc, 90, 2181 (1968). 51. (a) A. Rosenthal and A. Zanlungo, Can. J. Chem., 50, 1192 (1972); (b) A. Rosenthal and K. Shudo, J. Org. Chem., 37, 1608 (1972). 52. D. Horton and E.K. Just, Carbohydr. Res., 9_, 129 (1969). 53. H.H. Baer and G.V. Rao, Ann., 686, 210 (1965). 54. H. Yanagisawa, M. Kinoshita, and S. Umezama, Bull. Chem. Soc. Jap., 44, 3399 (1971). 55. N.M.K. Ng. Ying Kin, J.M. Williams, and A. Horsington, Chem. Commun., 971 (1969). - 193 -56. K. Heyns and H. Paulsen, Advances in Carbohydr. Chem., 17, 169 (1962). 57. (a) J.M. Sugihara and G.U. Yuen, J. Amer. Chem. Soc, 7£, 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). 58. R.E. Partch, Tetrahedron Lett., 3071 (1965); J. Org. Chem., 30, 2498 (1965). 59. R.F. Butterworth and S. Hanessian, Synthesis, 2, 70 (1971). 60. N. Kornblum, J.W. Powers, G.J. Anderson, W.J. Jones, H.O. Larson, 0. Levand, and W.M. Weaver, J. Amer. Chem. Soc, _79, 6562 (1957). 61. K.E. Pfitzner and J.G. Moffatt, J. Amer. Chem. Soc, 85_, 3027 (1963). 62. J.D. Albright and L. Goldman, J. Org. Chem., 30, 1107 (1965); J. Amer. Chem. Soc, 87, 4214 (1965). 63. K. Onodera, S. Hirano and N. Kashimura, J. Amer. Chem. Soc, 87, 4651 (1965). 64. W.W. Epstein and F.W. Sweat, Chem. Rev., 67_, 247 (1967). 65. P.J. Beynon, P.M. Collins, P.T. Doganges and W.G. Overend, J. Chem. Soc, (C), 1131 (1966). 66. (a) V.M. Parikh and J.K.N. Jones, Can. J. Chem., 43, 3452 (1965); (b) B.R. Lawton, W.A. Szarek and J.K.N. Jones, Carbohydr. Res., 10, 456 (1969). 67. S. Nahar, W.G. Overend and N.R. Williams, Chem. and Ind., 2114 (1967). 68. P.J. Beynon, P.M. Collins, D. Gardiner, and W.G. Overend, Carbohydr. Res., 6, 431 (1968). 69. R. Criegee, Ann., 522, 75 (1936). 70. R. Criegee, B. Marchand, and H. Wannowius, Ann., 550, 99 (1942). 71. G. Wagner, J. Russ. Phys. Chem. Soc, 27^ 219 (1895). 72. J. Boeseken, Rec. Trav. Chim., 40, 553 (1921). - 194 -73. R.U. Lemieux and E. von Rudloff, Can. J. Chem., 33_, 1701 (1955). 74. R. Stewart, Oxidation Mechanism, Applications to Organic Chemistry, W.A. Benjamin, Inc., N.Y., 1964, p. 62. 75. K.B. Wiberg and K.A. Saegebarth, J. Amer. Chem. Soc, 79_, 2822 (1957). 76. A.W. Johnson, Ylid Chemistry, Vol. 7, Academic Press, New York, 1966. 77. H. Standinger and J. Meyer, Helv. Chim. Acta, 2, 635 (1919). 78. G. Luscher, Dissertation Eidg. Techn. Hochschule, Zurich (1922). 79. G. Wittig and M. Rieber, Ann., 562, 177 (1949). 80. G. Wittig and G. Geissler, Ann., 580, 44 (1953). 81. G. Wittig and U. Schollkopf, Chem. Ber., §]_, 1318 (1954). 82. G. Wittig and W. Haag, Chem. Ver., 88, 1654 (1955). 83. G. Wittig, Angew. Chem., 68, 505 (1956). 84. J. Levisalles, Bull. Soc. Chim. France, 1021 (1958). 85. U. Schollkopf, Angew. Chem., 71, 260 (1959). 86. A. Maercker, Organic Reactions, 14_, 270 (1965). 87. U. Schollkopf, Newer Methods of Preparative Organic Chemistry, ^3, pp. 111-150, Ed. by W. Foerst, Academic Press, N.Y. (1964). 88. M. Schlosser, Topics in Stereochemistry, 5_, pp. 1-30, Ed. by E. E l i e l and N.L. Allinger, John Wiley and Sons, Inc. (1970). 89. S. Fliszar, R.E. Hudson and G. Salvadori, Helv. Chim. Acta, 47, 159 (1964). 90. C. Ruechardt, P. Panse and S. Eichler, Chem. Ber., 100, 1144 (1967). 91. M. Schlosser and K.E. Christmann, Ann., 708, 1 (1967). 92. J. Speziale and D.E. Bissing, J. Amer. Chem. Soc, 85_, 3878 (1963). 93. M. Schlosser, Angew Chem. Internat. Ed., ]_, 650 (1968). 94. N.K. Kochetkov and B.A. Dmitriev, Chem. and Ind., 864 (1963). - 195 -95. N.K. Kochetkov and B.A. Dmitriev, Dikl. Akad. Nauk, SSSR, 151, 106 (1963); Chem. Abstracts 59, 10215g (1963). 96. N.K. Kochetkov and B.A. Dmitriev, Tetrahedron, 21, 803 (1965). 97. Y.A. Khdanov, L.A. Uzlova, and G.N. Dorofeenko, Carbohydr. Res., 3, 69 (1966). 98. Y.A. Khdanov, G.N. Dorofeenko, and L.A. Uzlova, Dokl. Akad. Nauk, SSSR, 160, 336 (1965); Chem. Abstracts, 62, 14804b (1964). 99. Y.A. Khdanov and V.G. Alexeeva, Carbohydr. Res., 10, 184 (1969). 100. Y.A. Khdanov, L.A. Uzlova, and G.N. Dorofeenko, Zh. Vses, Khim. Obsh. In. D.I. Mendeleeva, 10, 600 (1965); Chem. Abstracts, 64, 3671d (1966). 101. J. Gigg, R. Gigg, and CD. Warren, J. Chem. Soc. (C), 1872 (1966). 102. J.M.J. Tronchet, F. Barbalat-Rey, J.M. Bourgeois, R. Graf, and J. Tronchet, Helv. Chim. Acta, 55_, 803 (1972). 103. J.M.J. Tronchet and J.M. Bourgeois, Helv. Chim. Acta, 54, 1580 (1971). 104. J.M.J. Tronchet and J.M. Chalet, Carbohydr. Res., 24, 263 (1972). 105. J.M.J. Tronchet and Br. Baehler, Helv. Chim. Acta, 55, 2112 (1972). 106. D.G. Lance and W.A. Szarek, Carbohydr. Res., 10, 300 (1969). 107. Y.A. Khdanov and V.A. Polenov, Carbohydr. Res., J.6, 466 (1971). 108. L. Horner, H. Hoffmann, and H.G. Wippel, Chem. Ber., 91_, 61 (1958). 109. W.S. Wadsworth and W.D. Emmons, J. Amer. Chem. Soc, j83, 1733 (1961). 110. American Chemical Society, J. Org. Chem., 35, 2849 (1970). 111. E.S. Chan, CK. Ingold, and V. Prelog, Angew. Chem., Int. Ed., 5, 385 (1966). 112. J.E. Blackwood, C L . Gladys, K.L. Loening, A.E. Petrarca and J.E. Rush, J. Amer. Chem. Soc, 90, 509 (1968). - 196 -113. G. Jones and R.F. Maisey, Chem. Commun., 543 (1968). 114. G. Lefebvre and J. Seyden-Penne, Chem. Commun., 1308 (1970). 115. G.H. Jones and J.G. Moffatt, J. Amer. Chem. Soc, 90, 5337 (1968). 116. E. Walton, S.R. Jenkins, R.F. Nutt, M. Zimmerman, and F.W. Holly, J. Amer. Chem. Soc, 88, 4524 (1966). 117. T.Y. Shen, Angew. Chem., Int. Edit., 9, 678 (1970). 118. S. Koto, S. Zen, S. Omura, and T. Hata, Bull. Chem. Soc Jap., 45, 532 (1972). 119. G.B. Howarth, W.A. Szarek, and J.K.N. Jones, J. Org. Chem., 34, 476 (1969). 120. G.B. Howarth, W.A. Szarek, and J.K.N. Jones, Can. J. Chem., 4j5, 3691 (1968). 121. A. Rosenthal, M. Sprinzel, and H.J. Koch, Can. J. Chem., 47, 3263 (1969). 122. F. Perini, F.A. Carey, and L. Long, Carbohydr. Res., 11, 159 (1969). 123. H. Uanagisawa, M. Kinoshita, and S. Umezawa, Bull. Chem. Soc. Jap., 44, 3399 (1971). 124. S.R. Jenkins, B. Arison, and E. Walton, J. Org. Chem., 33, 249C (1968). 125. R.F. Nutt, and E. Walton, J. Med. Chem., 11, 151 (1968). 126. E. Walton, S.R. Jenkins, R.F. Nutt, F.W. Holly, and M. Nemes, J. Med. Chem., 12, 306 (1969). 127. E. Walton (Merck and Co., Inc.) Fr. 1,504,091; CA. 70, 20308 (1969). 128. E.J. Reist, D.F. Calkins, and L. Goodman, J. Amer. Chem. Soc, 90, 3852 (1968). 129. E.J. Reist, Chem. and Ind., 1957 (1967). 130. J.J. Novak, J. Smejkal and F. Sorm, Tetrahedron Lett., 1627 (1969). - 197 -131. J.J. Novak, and F. Sorm, Coll, Czech. Chem. Commun., 27_, 902 (1962). 132. G.J. Lourens and A. Jordaan, J. Het. Chem., £, 975 (1972). 133. A. Rosenthal and D.A. Baker, J. Org. Chem., 38, 198 (1973). 134. A. Rosenthal and D.A. Baker, J. Org. Chem., 38, 193 (1973). 135. T. Sato, Syn. Proc. Nucleic Acid Chem., 1, 264 (1968). 136. W.W. Zorbach, Synthesis, 7_t 329 (1970). 137. W.W. Zorbach and R.S. Tipson, Editors in "Synthetic Procedures in Nucleic Acid Chemistry", John Wiley and Sons., Inc., New York, N.Y., 1968. 138. T.L.V. Ulbricht, Angew Chem., Int. Edit., _1, 476 (1962). 139. T. Sato, T. Shimadate, and Y. Ishado, Nippon Kagaku Zasshi, 81, 1440, 1442 (1961). 140. 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. 141. J.J. Fox and I. Wempen, Adv. Carbohydr. Chem., 14, 283 (1959). 142. Y. Furukawa, K. Imai and M. Honjo, Tetrahedron Lett., 4655 (1968). 143. T.G. Bonner, E.J. Bourne, and S. McNally, J. Chem. Soc, 761 (1962). 144. B.R. Baker, J.P. Joseph, R.E. Schaub, and J.H. Williams, J. Org. Chem., 19, 1780 (1954). 145. J. Ingwall, J. Amer. Chem. Soc, 94, 5487 (1972). 146. C. Djerassi, Proc. Chem. Soc, 314 (1964). 147. G.T. Rogers and T.L.V. Ulbricht, Biochem. Biophys. Res. Commun., 39, 419 (1970). 148. J.J. Fox, K.A. Watanabe, and A. Bloch, Prog. Nucleic Acid Res. Mol. Biol., 5, 251 (1966). 149. S. Zen, Y. Takeda, A. Yashuda, and S. Umezawa, Bull. Chem. Soc, Japan, 40, 431 (1967). - 198 -150. H. Yanagisawa, M. Kinoshita, and S. Umezawa, Bull. Chem. Soc. Japan, 42, 1719 (1969). 151. R.J. Suhadolnik, in Nucleoside Antibiotics, John Wiley and Sons, Inc., 1970. 152. K. Isono and S. Suzuki, 156th Meeting, Americal Chemical Society, Atlantic City, N.J., September, 1968. 153. N. Ohata, T. Akashiba, S; Sasaki, S. Kuroda, and T. Misato, Abstracts i Annual Meeting, Agricultural Chemical Society of Japan, Tokyo, April 1, 1969, !. 16. 154. A. Endo and T. Misato, Biochem. Biophys. Res. Commun., _37, 718 (1969). 155. J. Eguchi, S. Sasaki, N. Ohta, T. Akashiba, T. Tsuchiyama, and S. Suzuki, Ann. Phytopathol. Soc, 34, 280 (1968). 156. T.G. Yun and T. Hirai, Nippon Shokubutsu Byori Gakkaiho, 34, 109 (1968). 157. K. Isono, K. Ashahi, and S. Suzuki, J. Amer. Chem. Soc, 91, 7490 (1969). 158. T. Naka, T. Hashizume, M. Nishimura, Tetrahedron Lett., 95 (1971). 159. (a) N.P. Damodaran, G.H. Jones, and J.G. Moffatt, J. Amer. Chem. Soc, 93, 3812 (1971); (b) H. Ohrui, H. Kuzuhara, and S Emoto, Tetrahedron Lett., 4267 (1971). 160. (a) J. Coetzer, A. Jordaan, G.J. Lourens, and M.J. Nolte, Acta Cryst., 28, 3537 (1972), (b) A.J. Brink, J. Coetzer, A. Jordaan, and G.J. Lourens, Tetrahedron Lett., 5353 (1972). 161. A. Rosenthal and K. Shudo, J. Org. Chem., 37_, 4391 (197 2> • 162. U. Schollkopf, F. Gerhart, and R. Schroder, Angew. Chem., Internat., Edit. , 8_, 672 (1969). 163. A. Kornhauser and D. Kegleyic, Carbohydr. Res., 11, 407 (1969). - 199 -164. J.C. Irvine and A.Hynd, J. Chem. Soc, 99, 161 (1911). 165. H. von Euler and K. Zeile, Ann., 487, 163 (1931). 166. M. Bergmann and L. Zervas, Chem. Ber., 65_, 1201 (1932). 167. D.G. Doherty, E.A. Popenoe, and K.P. Link, J. Amer. Chem. Soc, 75, 3466 (1953). ' 168. F.J.M. Rajabalee, Synthesis, 318 (1971). 169. P. Brewster, F. Hiron, E.D. Hughes, CK. Ingold, and P.A.D.S. Rao, Nature (London), 166, 179 (1950). 170. L. Benoiten, M. Winitz, M. Birnbaum, and J.P. Greenstein, J. Amer. Chem. Soc, 79, 6192 (1957). 171. H.L. Hoyer, Ber., 83, 491 (1950). 172. P. Langen and G. Etzold, Biochem. Z., 339, 190 (1963). 173. W.W. Zorbach, H.R. Munson, and K.V. Bhat, J. Org. Chem., 30, 3955 (1965) 174. P.W.K. Woo, H.W: Dion, and Q.R. Bartz, J. Amer. Chem. Soc, 83, 3352 (1961); 86, 2726 (1964). 175. P.W.K. Woo, H.W. Dion, and n.F. Johnson, J. Amer. Chem. Soc, 84, 1066 (1962). 176. M. Miyamoto, Y. Kawamatsu, M. Shinohara, K. Nakanishi, Y. Nakadira, and N.S. Bhacca, Tetrahedron Lett., 2371 (1964). 177. D.A. Prins, J. Amer. Chem. Soc, 70> 3955 (1948). 178. N.K. Richtmeyer and C.S. Hudson, J. Amer. Chem. Soc, 63, 1727 (1941). 179. L.F. Wiggins, in "Methods in Carbohydrate Chem.", 2, 188 (1963). 180. G.B. Howarth, W.A. Szarek, and J.K.N. Jones, Chem. Commun., 62 (1968). 181. B.J. Hunt and W. Rigby, Chem. and Ind., 1868 (1967). 182. S. Coffey (Edit.), "Rodd's Chemistry of Carbon Compounds1' Vol. IF, Elsevier Publ. Co., New York, 1967, (a) p. 148; (b) p. 351. 183. H. Paulsen and K. Todt, Adv. in Carbohydr. Chem., 23, 116 (1968). - 200 -184. S. Shafizadeh, Adv. in Carbohydr. Chem., 13, 9 (1958). 185. CT. Bishop and F.P. Cooper, Can. J. Chem., 41, 2743 (1963). 186. R.E. Reeves, Advan. Carbohydr. Chem., 6^, 107 (1951). 187. R.U. Lemieux, Advan. Carbohydr. Chem., 9_, 1 (1954). 188. R.J. Ferrier and W.G. Overend, Quart. Rev. (London), 13, 265 (1959). 189. B. Capon and W.G. Overend, Advan. Carbohydr. Chem., 15, 11 (1960). 190. B.R. Baker, R.E. Shaub, J.P. Joseph, and J.H. Williams, J. Amer. Chem. Soc, 77, 12 (1955). 191. D.M. Brown and T.L.V. Ulbricht in M. Florkin and E. Stolz: Comprehensive Biochemistry, Elsevier, Amsterdam, Vol. 2 (1962). 192. E.E. Leutzinger, T. Meguro, L.B. Townsend, D.A. Shuman, M.P. Schweizer, CM. Stewart and R.K. Robins, J. Org. Chem., 3_7» 3 6 9 5 (1972), 193. R.K. Robins, J. Amer. Chem. Soc, 82, 2654 (1960). 194. J.J. Fox, I. Wempen, A. Hampton, and I.L. Doer, J. Amer. Chem. Soc, 80, 1669 (1958). 195. (a) P. Sykes, In "A Guidebook to Mechanism in Organic Chemistry", 2nd Edit., Longmans, London (1965), p. 188; (b) E.N. Zilberman, Russ. Chem. Rev., 31, 615 (1962). 196. B. Helferich and K.L. Bettin, Chem. Ber., 104, 1701 (1971). 197. P.W. Austin, J.G. Buchanan, and E.M. Oakes, Chem. Commun., 374, 472 (1965). 198. P.N. Rylander, In "Catalytic Hydrogenation Over Platinum Metals", Academic Press, New York, 1967. 199. N.M. Yoon and H.C. Brown, J. Amer. Chem. Soc, 90_, 2927 (1968). 200. H.C. Brown, P. Heim, and N.M. Yoon, J. Amer. Chem. Soc, 92, 1638 (1970). - 201 -201. R.F. Nystrom, J. Amer. Chem. Soc, 77, 2544 (1955). 202. CG. Overberger and J.E. Mulvaney, J. Amer. Chem. Soc, 81, 4697 (1959). 203. Y.H. Ng and M. Freifelder, J. Pharm. Sci., 54, 1204 (1965). 204. M. Freifelder, J. Amer. Chem. Soc, 82_, 2386 (1960). 205. A. Thompson and M.L. Wolfrom, in "Methods i n Carbohydr. Chem.", 2, 215 (1963). 206. H. Zinner, H. Nimz, and H. Venner, Chem. Ber., 90, 2696 (1957). 207. H. Zinner, K. Wessely, W. Bock, K. Rieckhoff, F. Strandt, and W". Nimmich, Chem. Ber., 90, 500 (1957). 208. R.K. Ness, and E.G. Fletcher, Jr., J. Amer. Chem. Soc, 82, 3434 (1960). 209. R.E.Deriaz, W.G. Overend, M. Stacey, and L.F. Wiggins, J. Chem. Soc, 2836 (1949). 210. C.C. Bhat, K.V. Bhat, W.W. Zorbach, Chem. Commun., 808 (1968). 211. B. Helferich, Adv. Carbohydr. Chem., _3> 79 (1948). 212. N.J. Leonard, F.C Sciavolino, and V. Nair, J. Org. Chem., 33, 3169 (1968). 213. (a) CL. Stevens, J.B. F i l i p p i , and K.G. Taylor, J. Org. Chem., 31, 1292 (1966); (b) R.J. Ferrier, W.G. Overend, and G.H. Sankey, J. Chem. Soc, 2830 (1965). 214. L.D. Hall, S.A. Black, K.N. Slessor, and A.S. Tracey, Can. J. Chem., 50, 1912 (1972). 215. C.P.J. Glaudemans and H.G. Fletcher, Jr., J. Amer. Chem. Soc, 87, 4636 (1965). 216. W.V. Rugle, T.Y. Shen, and A.A. Patchett, J. Org. Chem., 30, 4353 (1965), - 202 -217. (a) L.J. Stegerhock and P.E. Verkade, Rec. Trav. Chim., 75, 143 (1956); (b) D.R. Strobach and L. Szabo, J. Chem. Soc, 3970 (1963); (c) J.G. Buchanan and R. Fletcher, J. Chem. Soc, 6316 (1965); (d) D.A. Applegarth, J.G. Buchanan, and J. Baddiley, J. Chem. Soc, 1213 (1965). 218. J. Lehrfeld, J. Org. Chem., 32, 2544 (1967). 219. P.E. Verkade, F.D. Tollenaar, and T.A.P. Posthumus, Rec. Trav. Chim, 61, 375 (1942). 220. J.R. Turvey and T.P. Williams, J. Chem. Soc, 2119 (1962). 221. W. Anderson, D.H. Hayes, A.M. Michelson, and A.R. Todd, J. Chem. Soc, 1882 (1954). 222. (a) M. Smith, D.H. Rammler, I.H. Goldberg, and H.G. Khorana, J. Amer. Chem. Soc, j?4, 430 (1962); (b) P.T. Gilham and H.G. Khorana, J. Amer. Chem. Soc, 81, 4647 (1959). 223. D.L. MacDonald and H.G. Fletcher, J. Amer. Chem. Soc, 81_, 3719 (1959). 224. S. Inouye, T. Tsuruoka, T. Ito, and T. Niida, Tetrahedron, 23_, 2125 (1968). 225. E.E. van Tamelen, H. Rudler and C. Bjorklund, J. Amer. Chem. Soc, 93, 7113 (1971). 226. M.J. Robins and R.K. Robins, J. Amer. Chem. Soc, 87, 4934 (1965). 227. (a) S. Suzuki, K. Isono, J. Nagatsu, T. Mizutani, Y. Kawashima, and T. Mizuno, J. Antibiotics, A18, 131 (1965); (b) K. Isono, J. Nagatsu, Y. Kawashima, and S. Suzuki, Agr. Biol. Chem., 29, 848 (1965); (c) K. Isono, J. Nagatsu, Y. Kawashima, K. Yamagata,. K. Sasaki, and S. Suzuki, Agri. Biol. Chem., 3J_, 190 (1967). 228. K. Isono and S. Suzuki, Tetrahedron Lett., 203, 1133 (1968). - 203 -229. Y. Liwschitz, Y. Rabinsohn, and D. Perera, J. Chem. Soc, 1116 (1962); 3589 (1962). 230. W.L. Glenn, G.S. Meyers, and G.A. Grant, J. Chem. Soc, 2568 (1951). 231. A. Rosenthal and CM. Richards, Carbohydr. Res., in press. 232. J.W. Emsley, J. Feeney, and L.H. Sutcliffe, in "High Resolution Nuclear Magnetic Resonance Spectroscopy", Vol. 2, Pergamon Press, (1966), p. 735. 233. J.M. Bourgeois, Ph.D. Thesis, Universite de Geneve, (1972). 234. R.D. Rees, K. James, A.R. Tatchell, and R.H. Williams, J. Chem. Soc, 2716 (1968). 235. M. Fieser, A. Quilco, A. Nickon, W.E. Rosen, E.J. Tarlton, and L.F. Fieser, J'. Amer. Chem. Soc, 75, 4066 (1953). 236. (a) H.O. House, in "Modern Synthetic Reactions", W.A. Benjamin, New York, N.Y., 1965, p. 142; (b) F.D. Gunstone, in R.A. Raphael, E.C Taylor, and H. Winberg, "Advances in Organic Chemistry, Methods and Results", Vol. 1, Wiley-Interscience, New York, N.Y., 1960, pp. 103-147. 237. D.H. Ball and F.W. Parrish, Adv. in Carbohydr. Chem., 23, 233 (1968). 238. B. Capon, Chemical Reviews, 69_, 407 (1969), and references therein. 239. H.E. Carter and CL. Zirkle, J. Biol. Chem., 178, 709 (1949). 240. (a) M. Miljkovic, T. Satoh, M. Konopka, E.A. Davidson, and D. Miljkovic, J. Org. Chem., 38, 716 (1973); (b) M. Miljkovic, D. Miljkovic, A . Jokic, V. Andrejevic, and E.A. Davidson, J. Org. Chem., 36, 3218 (1971). 241. J.C. Craig and S.K. Roy, Tetrahedron, 21, 391, 1847 (1965). 242. E.J. Corey, R.J. McCaully, and H.S. Sachdev, J. Amer. Chem. Soc, 92, 2476 (1970), and references therein. - 204 -243. E.J. Corey, H.S. Sachdev, J.Z. Gougoutas, and W. Saenger, J. Amer. Chem. Soc, 92, 2488 (1970). 244. K.B. Sharpless, R.F. Lauer, 0. Repic, A.Y. Teranishi, and D.R. Williams, J. Amer;. Chem. Soc, 93_, 3303 (1971). 245. T.D. Inch, G.J. Lewis, and N.E. Williams, Carbohydr. Res., 19, 17 (1971). 246. (a) D.H.R. Barton, A. DaS. Campos-Neves, and R.C. Cootson, J. Chem. Soc, 3500 (1956); (b) J.L. Beton, T.G. Halsall, E.R.H. Jones, and B.C. Ph i l l i p s , J. Chem. Soc, 753 (1957). 247. A. Rosenthal and K.C. Dooley, unpublished results. 248. B. Sjoberg, in "Optical Rotatory Dispersion and Circular Dichroism in Organic Chemistry", Ed. G. Snatzke, Heyden and Son, London, 1965, p. 173. 249. M. Nakano and T.S. Danowski, in "Methods in Enzymology", Vol. XVII, Academic Press, New York, 1971, p. 601. 250. J.P. Greenstein and M. Winitz, in "Chemistry of the Amino Acids", Vol. 1, John Wiley and Sons, Inc., New York, 1961, p. 731. 251. G. Gomori, in "Methods in Enzymology", S.P. Colowick and N.O. Kaplan (Eds.), Vol. I, Academic Press, Inc., New York, 1955, p. 138. 

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}]}"
                            data-media="{[{embed.selectedMedia}]}"
                            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:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0060150/manifest

Comment

Related Items