Chemical Synthesis Of Two Repeat Units Corresponding To The Backbone Of The Pectic Polysaccharide Rhamnogalacturonan I Dominic Reiffarth BSc., University of Northern British Columbia, 1998 Thesis Submitted In Partial Fulfillment Of The Requirements For The Degree Of Master Of Science in Mathematical, Computer, and Physical Sciences (Chemistry) The University of Northern British Columbia May 2006 © Dominic Reiffarth, 2006 R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. Library and Archives Canada Bibliotheque et Archives Canada Published Heritage Branch Direction du Patrimoine de I'edition 395 W ellington Street Ottawa ON K1A 0N4 Canada 395, rue W ellington Ottawa ON K1A 0N4 Canada Your file Votre reference ISBN: 978-0-494-28405-6 Our file Notre reference ISBN: 978-0-494-28405-6 NOTICE: The author has granted a non­ exclusive license allowing Library and Archives Canada to reproduce, publish, archive, preserve, conserve, communicate to the public by telecommunication or on the Internet, loan, distribute and sell theses worldwide, for commercial or non­ commercial purposes, in microform, paper, electronic and/or any other formats. AVIS: L'auteur a accorde une licence non exclusive permettant a la Bibliotheque et Archives Canada de reproduire, publier, archiver, sauvegarder, conserver, transmettre au public par telecommunication ou par I'lnternet, preter, distribuer et vendre des theses partout dans le monde, a des fins commerciales ou autres, sur support microforme, papier, electronique et/ou autres formats. The author retains copyright ownership and moral rights in this thesis. Neither the thesis nor substantial extracts from it may be printed or otherwise reproduced without the author's permission. L'auteur conserve la propriete du droit d'auteur et des droits moraux qui protege cette these. Ni la these ni des extraits substantiels de celle-ci ne doivent etre imprimes ou autrement reproduits sans son autorisation. In compliance with the Canadian Privacy Act some supporting forms may have been removed from this thesis. Conformement a la loi canadienne sur la protection de la vie privee, quelques formulaires secondaires ont ete enleves de cette these. While these forms may be included in the document page count, their removal does not represent any loss of content from the thesis. Bien que ces formulaires aient inclus dans la pagination, il n'y aura aucun contenu manquant. i *i Canada R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. APPROVAL Name: Dominic Reiffarth Degree: Master of Science Thesis Title: CHEMICAL SYNTHESIS OF TWO REPEAT UNITS CORRESPONDING TO THE BACKBONE OF THE PECTIC POLYSACCHARIDE RHAMNOGALACTURONAN I Examining Committee: Chair: Dr. Robert Tait Dean of Graduate Studies University of Northern British Columbia Supervisor: Dr. KerrylteimeiyAssociate Professor Mathematical, Computer, and Physical Sciences Program University of Northern British Columbia Committee Menf^er^ Dr Michael Rutherford, Associate Professor Natural Resources and Environmental Studies Program University of N ort^m British Columbia Committee MembeK^AGaylPlourde, Associate Professor Mathematical, Computer' and T h e c a l Sciences Program Universit-y of Northern British Columbia External Sxafmner: Dr. David Dick, Adjunct Professor Mathematical, Computer, and Physical Sciences Program University of Northern British Columbia Date Approved: R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. Abstract •01COOMe HO SEt HO HOCOOMe COOMe AIIO- + BzO | OAc BzIO OH OMe OMe OH A tetrasaccharide sequence of Rhamnogalacturonan I has been synthesized starting from commercially available D-galacturonic acid and L-rhamnose. This synthesis relies on only two protected monosaccharides and proceeds through a common disaccharide intermediate. Synthesis of this tetrasaccharide has been designed to allow for the addition of branching elements at the 4-positions of the rhamnosyl units, or further chain elongation at the 2-position. - IX - R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. TABLE OF CONTENTS Abstract ii Table of Contents iii List of Tables vi List of Figures vii List of Reaction Schemes viii Abbreviations ix Acknowledgement xi Introduction 1 Chapter One Carbohydrate Recognition Domains (CRDs) Lectins Carbohydrate-based Vaccines Carbohydrates as Epitopes for Monoclonal Antibodies (mABs) Synthesis of the Epitope 3 3 5 6 6 Chapter Two The Plant Cell Wall Introduction Pectins Rhamnogalacturonan II (RG-II) Rhamnogalacturonan I (RG-I) 8 8 8 9 10 Chapter Three General Synthetic Strategies in Carbohydrate Chemistry Protecting Groups The Benzyl Protecting Group The Benzoyl and Acetate Protecting Groups The Allyl Protecting Group The Isopropylidene Protecting Group The para-Methoxybenzyl Protecting Group Glycosy lation Methods Common Glycosidic Donors Thioglycosides Imidates Glycosyl Halides Late-Stage Oxidation Synthetic Approach to RG-I 13 14 17 18 19 20 20 21 24 24 25 26 28 29 - iii - R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. Chapter Four Chapter Five Results and Discussion Synthesis of the Monosaccharide Acceptor Synthesis of Disaccharides 8 and 9, and Disaccharide Acceptors 10 and 11 Synthesis of the Disaccharide Donors Synthesis of the Fully Protected Tetrasaccharide, RG-I Deprotection of RG-I 32 32 Experimental Methyl (methyl-a-D-galactopyranosid)uronate (1) 67 67 37 41 49 62 Methyl (methyl 3,4-O-isopropylidene-a-Dgalactopyranosid)uronate (2) 68 Methyl (methyl 2-0-benzyl-3,4-0-isopropylidene-a-Dgalactopyranosid)uronate (3) 69 Methyl (methyl 2-0-benzyl-a-D-galactopyranosid)uronate (4) 69 Benzyl (methyl 2,3-di-0-benzyl-a-D-galactopyranosid)uronate (5) and Methyl (methyl 2,3-di-O-benzyl-a-D70 galactopyranosid)uronate (6) Methyl (2-Oacetyl-4-0-allyl-3-0-benzoyl-a-Lrhamnopyranosyl)-(l —-4)-(methyl 2,3-di-O-benzyl-a-Dgalactopyranosid)uronate (8) 71 Benzyl (2-0-acetyl-4-0-allyl-3-0-benzoyl-a-Lrhamnopyranosyl)-(l -*4)-(methyl 2,3-di-Obenzyl-a-Dgalactopyranosid)uronate (9) 72 Methyl (4-0-allyl-3-0-benzoyl-a-L-rhamnopyranosyl)-(l —4)(methyl 2,3-di-0-benzyl-a-D-galactopyranosid)uronate (10) 73 Benzyl (4-<9-allyl-3-0-benzoyl-a-L-rhamnopyranosyl)-(l —4)74 (methyl 2,3-di-0-benzyl-a-D-galactopyranosid)uronate (11) Methyl (2-<9-acetyl-4-0-allyl-3-G-benzoyl-a-Lrhamnopyranosyl)-( 1—4)-( 1-O-acetyl-2,3 -di-O-benzyl-a/p-Dgalactopyranosid)uronate (12) 76 Methyl (2-0-acetyl-4-0-allyl-3-0-benzoyl-a-Lrhamnopyranosyl)-(1^4)-(2,3-di-0-benzyl-a/p-Dgalactopyranosid)uronate (13) 77 Methyl (2-0-acetyl-4-0-allyl-3 -O-benzoyl-a-Lrhamnopyranosyl)-(l-*4)-2,3-di-0-benzyl-a-Dgalactopyranosyluronate trichloroacetimidate (14) 77 Benzyl (2-0-acetyl-4-0-allyl-3-0-benzoyl-a-Lrhamnopyranosyl)-(l-*4)-(l-6>-acetyl-2,3-di-6>-benzyl-a/p-Dgalactopyranosid)uronate (15) 78 - iv - R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. Chapter Six Benzyl (2-0-acetyl-4-0-allyl-3-0-benzoyl-a-Lrhamnopyranosyl)-(l—4)-(2,3-di-0-benzyl-a/p-Dgalactopyranosid)uronate (16) 79 Benzyl (2-0-acetyl-4-0-allyl-3-0-benzoyl-a-Lrhamnopyranosyl)-( 1—4)-2,3 -di-0-benzyl-a/p-Dgalactopyranosyluronate trichloroacetimidate (17) 80 Methyl (2-0-acetyl-4-0-allyl-3-0-benzoyl-a-Lrhamnopyranosyl)-(l —4)-(methyl 2,3-di-0-benzyl-a-Dgalactopyranosyluronate)-( 1-~2)-(2-0-acetyl-4-0-allyl-3 -0benzoyl-a-L-rhamnopyranosyl)-(l ~*4)-(methyl 2,3-di-0benzyl-a-D-galactopyranosid)uronate (18) 81 Methyl (2-0-acetyl-4-0-allyl-3-0-benzoyl-a-Lrhamnopyranosyl)-(l —4)-(benzyl 2,3-di-0-benzyl-a-Dgalactopyranosyluronate)-(l-*2)-(2-0-acetyl-4-0-allyl-3-0benzoyl-a-L-rhamnopyranosyl)-(l ^4)-(methyl 2,3-di-0benzyl-a-D-galactopyranosid)uronate (19) 82 Methyl (a-L-rhamnopyranosyl)-(l-*4)-(methyl a-Dgalactopyranosyluronate)-( 1-*2)-(a-L-rhamnopyranosyl)(1—4)-(methyl a-D-galactopyranosid)uronate (20) 84 Conclusions and Future Work 87 References 90 Appendix 1 94 -v- R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. List of Tables Table 1 Common protecting groups used in synthetic carbohydrate chemistry, including formation and cleavage conditions - vi - R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. List of Figures Figure 1 P-l —4 linkage of two glucose monomers found in cellulose 1 Figure 2 Representation of a synthetic antigen 7 Figure 3 LM5 and LM6 antibody binding 7 Figure 4 RG-I and some of its branches 10 Figure 5 Orthogonality of common hydroxyl protecting groups used in carbohydrate synthetic strategies 15 a-D-glucose in the cis-1,2 configuration and a-D-mannose in the trans-1,2 configuration 21 A typical glycoside donor with a suitably protected alcohol acceptor 22 A possible mechanism involving participating and non-participating groups in glycosylations 23 Probable mechanism for the NIS/TfOH-promoted glycosylation of a rhamnosyl thioglycoside with an acceptor alcohol to give a trans-1,2 linkage 25 Figure 6 Figure 7 Figure 8 Figure 9 Figure 10 Formation of a glycosidic linkage via an SN2-li ke mechanism 26 Figure 11 Formation of a trans-1,2 glycosidic linkage via an orthoester intermediate using a glycosyl bromide as the donor 27 Figure 12 D-galactose and D-galacturonic acid 28 Figure 13 Tetrasaccharides 18 and 19 30 Figure 14 General scheme showing the block synthesis approach of the target tetrasaccharide 31 The rhamnosyl donor used for the synthesis of the disaccharides 37 Figure 16 Synthesis of an analogous tetrasaccharide 51 Figure 17 Minimization of dipole interactions in the endo-anomeric effect 52 Molecular orbital explanation of the axial orientation of substituents on the anomeric carbon 53 Possible hydrogen bonding interactions affecting glycosylation 54 Use of a coordinating solvent (EtiO) in product formation 57 Figure 15 Figure 18 Figure 19 Figure 20 - vii - R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. List of Reaction Schemes Scheme 1 Synthesis of the monosaccharide galacturonic acid acceptor 32 Scheme 2 Synthesis of the disaccharides and the respective acceptors 38 Scheme 3 Synthesis of the disaccharide donors 41 Scheme 4 Synthesis of the fully protected tetrasaccharides 49 Scheme 5 Deprotection of the tetrasaccharide 63 - viii - R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. Abbreviations Abbreviations (Alphabetical order) 13C 'H A calcd COSY eq. GalA HG HMQC LPLC mAB NMR RG -I RG-II Rha rt satd TLC Carbon 13 NMR Proton NMR Angstrom calculated Correlated Spectroscopy equivalents Galacturonic acid Homogalacturonan Heteronuclear Multiple-Quantum Coherence Experiment low pressure liquid chromatography monoclonal antibody nuclear magnetic resonance spectroscopy Rhamnogalacturonan I Rhamnogalacturonan II Rhamnose room temperature saturated Thin Layer Chromatography Chemical Abbreviations (Alphabetical order) [(C6H5)3P]3RhCl Ag20 AgOTf BF3-Et20 BzCl BzlBr CAN CBr4 CH2C12 CSA DABCO DBU DDQ DMF DMP Et3N Et4NBr EtOAc EtOH Rhodium(I) tris(triphenylphosphine) chloride, Wilkinson’s catalyst silver(I) oxide silver triflate (silver trifluoromethane sulfonic acid) borontrifluoride etherate benzoyl chloride benzyl bromide eerie ammonium nitrate (Ce(NH4)2(N 03)6) carbon tetrabromide dichloromethane camphor sulfonic acid 1,4-diazobicyclo[2.2.2]octane 1,8-diazabicyclo [5.4.0]undec-7-ene 2,3-dichloro-5,6-dicyano-p-benzoquinone dimethyl formamide 2,2-dimethoxy propane triethyl amine tetraethyl ammonium bromide ethyl acetate ethanol - IX - R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. HC1 Hex HgCl2 HgO KI MeOH MeOTf MS Na2S2C>3 Na2S04 NaH NaHC03 Na0Ac-3H20 NaOMe NIS P(Ph)3 Ph pyr SnCl4 TFA TfOH TMSOTf TMU hydrochloric acid hexanes mercury(II) chloride mercury(I) oxide potassium iodide methanol methyl triflate (methyl trifluoromethane sulfonic acid) molecular sieves sodium thiosulfate sodium sulfate sodium hydride sodium bicarbonate sodium acetate trihydrate sodium methoxide AModosuccinimide triphenyl phosphine phenyl pyridine tin(IV) chloride trifluoroacetic acid triflic acid (trifluoromethane sulfonic acid) trimethylsilyl trifluoromethanesulfonate tetramethyl urea Protecting group abbreviations Ac All Bz Bzl Me Ph PMB STol TBDMS TBDPS Tr acetate allyl benzoyl benzyl methyl phenyl para-methoxybenzyl para-thiocresol r-butyldimethylsilyl /-butyldiphenylsilyl trityl (triphenylmethyl) R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. Acknowledgement I wish to thank my supervisor, Dr. Kerry Reimer, for all his support and guidance throughout this research. Dr. Reimer was there to offer advice or point me in the right direction when needed, yet allowed me the freedom to truly experiment so that I could make the most of this valuable opportunity. Dr. Reimer is also responsible for introducing me to the interesting world of synthetic carbohydrate chemistry. I also wish to thank my committee members, Dr. Guy Plourde and Dr. Mike Rutherford for their support and encouragement throughout my research, which helped me stay positive during the most trying of times. I wish to show my gratitude towards Dr. David Dick, who was always willing to provide his assistance and, perhaps even more importantly, provided me with some witty anecdotes with respect to the joys of research. I wish to thank my family for their support over the past several years - their company, and the willingness to listen to me ramble on about things chemical, is truly appreciated. I must thank my wife, Kikko, above all else, for having the understanding and patience to allow me to commit to the research undertaken here (and without her, I probably would have starved to death, been wearing dirty clothing, etc.). Finally, I wish to thank NSERC for providing the funding for both this project, as well as providing the financial support which allowed me to dedicate my time to my research. xi - R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. Introduction The term “carbohydrate” often evokes thoughts of sugar and diets in many people, and rightly so, as they are an important part of the energy storage system found both in plants and animals in the form of starch and glycogen, respectively. Hydrolysis of these stored polysaccharides can provide the organism with extra energy when required. The other well-known biological role of carbohydrates is as a means of providing structure. In plants, monomers of p -1 ,4-linked Figure 1 P-1,4 linkage o f two glucose monomers found in cellulose glucose residues (Figure 1) form long chains which intertwine to form fibrils that cannot be digested by humans. In contrast, the a-l,4-linked glucose residues in starch and glycogen form helical structures with a hollow center; this small change demonstrates, at a very basic level, the effect on structure and function of simply altering one linkage. Another carbohydrate known for its structural role is chitin, which forms the exoskeleton of arthropods, such as crabs, and is similar in structure to glycogen, except for a C-2 acetylated amino group rather than a hydroxyl group. Carbohydrates are highly diverse in the type and number of linkages they can form. A simple glucose monomer, when glycosidically linked to another glucose residue, can exhibit various linkages (to C-2, C-3, etc) to form higher order structures; this has made carbohydrates prime candidates for further investigation beyond their role as storage and structural compounds. Indeed, carbohydrates exhibit a higher degree of diversity than, say, amino acids. Whereas two identical amino acids can form only one dipeptide, two identical carbohydrates can form eleven different linkages1. Similarly, four - 1 - R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. different nucleotides can form only 24 unique tetranucleotides, but four different monosaccharides can form a staggering 35, 560 unique tetrasaccharides. - 2- R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. Chapter One Carbohydrate Recognition Domains (CRDs) The diversity displayed by carbohydrates in terms of their monomeric structures and available linkages make them excellent candidates as cellular recognition points and a source of intercellular communication. Several different types of molecules may bind to various, specific, carbohydrate recognition domains (CRDs), and may include lectins, pathogens, toxins and antibodies. A few of these are discussed here to demonstrate the importance of understanding the role carbohydrate structures play and how this information can be used to learn more about cellular activities. Lectins The class of compounds defined as lectins includes any protein or glycoprotein (proteins with bound carbohydrates) that contains at least one non-catalytic mono- or oligosaccharide binding site; binding must be able to take place reversibly2. Lectins are therefore classified into three major categories, according to their binding domain: merolectins simply possess one carbohydrate binding domain; hololectins contain two or more identical binding domains, or two or more highly homologous binding domains; chimerolectins have one sugar-binding domain and contain at least one other, unrelated, binding domain, with neither binding domain exhibiting the same biological activity nor the same carbohydrate binding specificity2. Two major types of lectins were initially identified - the S-type lectins, often referred to as galectins, and the C-type lectins, often called selectins3. The designations came about due to their differing binding activities. All galectins were initially thought to exhibit thiol-dependent binding activity, as was the case for the first galectin identified -3 - R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. galectin-I; however, later discoveries demonstrated this was not the case and, as such, the term "S-type" is fading from use4. Selectins, on the other hand, show Ca2+-dependent binding activity. The importance of protein-carbohydrate binding interactions seen in lectins can be demonstrated by the galectins, animal lectins with an affinity for P-galactose-containing oligosaccharides4. The expression levels of galectins in cancerous cell lines was reviewed by Danguy et al. in 2002; expression levels in tumorous tissue such as the breast, prostate and thyroid, but to name a few, were covered5. Differences in the concentrations and types of galectins expressed varied depending on the location of the tissue and the aggressiveness of the tumor. Galectin levels in several of these cancers were clearly above or below “normal”, indicating that the role of lectins could be useful in the detection/treatment of cancers. The results noted above are only one of many in a very narrow field of lectin research. More broadly, the identification of biologically active compounds make those compounds prime targets for further investigation and manipulation - manipulation by synthetic organic chemists. These synthesized structures can then be modified to determine whether there is any effect on function. The terminal a-D-mannopyranosyl residue is a feature of glycoproteins found in bacterial pathogens, parasites, yeasts and virus envelopes6. The presence of this receptor allows for various C-type lectins, such as the mannose receptors found on macrophages, to act as scavengers by binding a variety of pathogenic microorganisms and potentially harmful glycoproteins7,8’9. The mannose binding receptors of macrophages and dendritic cells can thus be exploited to recognize and administer synthetic mannose-containing -4 - R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. neoglycoconjugates of drugs and antigens in a cell-specific manner6. The degree of binding is dependent on the carbohydrate portion that is recognized, and differing structures have differing affinities. The synthesis of the CRD-binding portion for conjugation to the glycoprotein is therefore required. Carbohydrate-based Vaccines Carbohydrate-based vaccines rely on the expression or over-expression of unique carbohydrate structures in or on cells10, n . The viability of using carbohydrate-based vaccines to fight disease was demonstrated by the presence of antibodies in human blood serum that were elicited using the group A Streptococcus (GAS) polysaccharide in vitro12. (Note: GAS bacterial infections lead to a diverse group of illnesses including cellulitis, necrotizing fasciitis and myositis, often referred to as "flesh-eating disease13.) Research into creating synthetic analogues of this polysaccharide was performed in order to elicit more discrete antibody responses, allowing for more precise targeting of the different types of infections10. Pinto et al. have focused on developing a vaccine against GAS based on the cell wall polysaccharide, which can act as an antigen; such a vaccine could offer an alternative to high doses of antibiotics and aggressive surgery10,13. Creating a vaccine based on the simple recognition by antibodies of a carbohydrate expressed on the surface of a cell raised against a glycoconjugate vaccine is inadequate in some cases. In tumor cells, for example, certain carbohydrate structures become overexpressed, such as the Lewis blood group antigens Lea and LeaLex14. Recognition of these structures by antibodies could be problematic because Lea is also highly expressed on normal, healthy cells. An immune response based on the tumor cells would lead to the indiscriminate destruction of all cells expressing both the Lea and -5 - R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. LeaLex structures15. The synthesis of analogues to the Lea portion thus makes it possible to determine whether or not creating antigens that could lead to immune responses based solely on the tumor cells is possible11. In this way, a carbohydrate-based vaccine could be used to specifically target tumor cells. Carbohydrates as Epitopes for Monoclonal Antibodies (mABs) The importance of carbohydrates on the cell surface has already been discussed. Similar to cell-specific targeting of drugs and antigens, synthetic carbohydrate epitopes can be linked to antigens, which can in turn be used to raise monoclonal antibodies. These antibodies can then be used to probe for specific carbohydrate structures in investigations of structure/function relationships, such as cell growth and cell differentiation. Synthesis o f the Epitope In order to ensure that mABs are of a high binding specificity, they need to be raised against antigens of well-defined structure, such as those provided by synthetic carbohydrate epitopes. Where synthetically-based antigens are not available, other means of generating the carbohydrates required to raise the antibodies need to be used. Isolating the required carbohydrates is often attempted through biochemical means. Once the desired polysaccharide has been isolated from the corresponding plant (plant cell wall material may be digested through the use of suitable enzymes to obtain the polysaccharides16), acid hydrolysis is often used to fragment the large polysaccharide into smaller oligosaccharide fragments17. This method of fragmentation leads to a mixture of oligosaccharides with varying reducing and non-reducing ends, thereby not allowing for completely unambiguous structural characterization. Researchers wishing to probe the structure/function of carbohydrate epitopes therefore often express a need for - 6- R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. well-defined synthetic epitopes that have been clearly characterized and can ideally be easily modified18,19. Once an epitope has been generated, it synthetic needs to be coupled to a carrier protein in linker arm epitope order to complete the antigen. Most of the Carrier Protein Figure 2 Representation o f a synthetic antigen carbohydrate epitopes are of low molecular weight, so they are coupled to a suitable protein in order to produce an effective immunogen20. Short alkyl chains or similar structures can be used16,19,21 (Figure 2) for the coupling; the carrier protein used is often commercially available. Two of the more common types of protein used are human serum albumin (HSA) and bovine serum albumin (BSA). Figure 3 demonstrates the use of fluorescently labeled monoclonal antibodies in studying the structure/function relationship of polysaccharides found in carrot root cells. Localization or non-localization of a particular structure contained within the polysaccharide to which the monoclonal antibody has bound can be easily visualized by intense fluorescing or, conversely, absence thereof. LM5: anti-(1-4)-P-gatactan LM6: anti-(1-5)-a-arabinan Figure 3 LM5 and LM6 antibody binding; shaded areas represent the respective binding o f the LM5 and LM6 fluorescently labeled monoclonal antibodies to side chains o f Rhamnogalacturonan-I in the carrot root apex22. 7- R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. Chapter Two The Plant Cell Wall Carbohydrates play an important role in the structure and function of plant cell walls. The structures of several carbohydrate moieties have been identified and will be discussed here, including the polysaccharide of interest, Rhamnogalacturonan I. Introduction The primary cell wall in plants plays a role in cell differentiation and cell growth, and consists of up to 90% polysaccharides and 10% glycoproteins23. Cellulose accounts for only 20-30% of primary plant cell wall polysaccharides; xyloglucans are the next largest substituent, and together form a cellulose-cross-linking glycan network. The glycan network can be thought of as being dissolved in a more soluble matrix, which includes polysaccharides, glycoproteins, proteoglycans, low-molecular-weight compounds and ions24. The most abundant macromolecular component of this matrix is the pectins. Pectins are also abundant in the middle lamellae, where they play an important role in regulating intercellular adhesion. Within the primary cell wall, pectins generally account for one third of the macromolecules found there. Pectins The pectic matrix is multifunctional, providing an environment for the deposition, slippage and extension of the cellulosic-glycan network24. Pectins are characterized by their galacturonic acid (GalA)-containing backbones. Three major classes of pectins have been identified: homogalacturonan (HG), Rhamnogalacturonan-I (RG-I) and Rhamnogalacturonan-II (RG-II). HG simply consists of various lengths of repeat units of - 8 - R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. galacturonic acid; the HG, RG-I and II pectic domains are thought to be covalently linked, forming a pectic network throughout the primary cell wall matrix and the middle lamellae24. Rhamnogalacturonan II (RG-II) The RG-II polysaccharide has a highly conserved structure, showing little to no variance throughout different types of vascular plants (angiosperms and gymnosperms)25. Investigation into the role of RG-II in the primary cell wall is ongoing; evidence suggests that it is structurally important in maintaining the rigidity and growth of the wall. A deficiency in boron results in an impairment in cell elongation and normal plant growth26. Several researchers have demonstrated that boron plays a role in the cross-linking of RG-II chains27,2S. Immunocytochemical methods to determine the location of RG-II in the primary cell wall have seen limited use due to a lack of well-defined antibodies . Currently, only two ill-defined antibodies have been used - ill-defined because the structures of the epitopes they bind have not been characterized. Studies performed using these antibodies could only verify that the antibody bound to some portion of RG-II, but did not characterize the epitope to which the antibodies bound29,30. One of the studies used a recombinant antibody that was raised by immunization of a mouse with a neoglycoprotein consisting of a RG-II fragment covalently bound to BSA29. The other antibody used thus far is a polyclonal antibody from a rabbit. As a result of the need for well-characterized epitopes, synthetic organic chemists have been attempting to synthesize key structures of the RG-II domain31,32 so that well-characterized mABs can be used in more precise structure/function studies. -9 - R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. Rhamnogalacturonan I (RG-I) The RG-I polysaccharide is characterized by its alternating D-galacturonic acid and L-rhamnose residues; as many as 100 repeats of this disaccharide unit have been isolated from plant cell walls33,34. The structure of the RG-I backbone is as follows: -a-L-Rha/?-(l —4)-a-D-GalA/?-( 1—-2)-a-L-Rha£>-(l —4)-a-D-GalA/?-(l —2)RG-I is a highly variable polysaccharide with a number of different types of branches, unlike the highly-conserved RG-II 0 * structure, where the types . (I • 0 and locations of branches HP 4 linked p g a in tan Sfj Q ■■ ■........ + W m ti • ^ 0 1-$ 0 11 • i- • • * are almost identical in all m 5-linkfld a arabinan 0 » f-i 0 pi • ^ 0 plants. Some of the G al A possible branches of RG-I are shown in Figure 4. 0 Rha # G al ^ 0# - 0 3,5-linkad a arabinan Fuc GteA Afabinogalactan Between 20-80% of the rhamnose residues in RG-I m Ara Figure 4 RG-I and some o f its branches are substituted at C-4, most often with neutral residues containing 50 or more repeats of a single residue, thereby forming one branch33. These neutral residues consist mainly of galactose and arabinose35. Analysis of RG-I pectins found in sugar-beet pulp has shown that substitution can also occur on the GalA residues35. Approximately one in 72 GalA residues is 0-3-substituted by glucuronic acid (GlcA). The highly substituted nature of RG-I has lead to it being called the 'hairy' region of pectin, whereas HG domains are known as the smooth region33. 10 - R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. Many of the structural features of RG-I have been identified, but little is known about how its structure relates to its function24. One of the most common ways of probing for structure/function relationships is through the use of monoclonal antibodies. Two such antibodies have been used for probing the side chains of RG-I: LM536 and LM621. LM5, used for probing galactan-containing side chains, was prepared by coupling commercially purchased GAL4 (a galactotetraose) to bovine serum albumin (BSA), against which monoclonal antibodies were raised by injection into rats and then harvested from the spleen by subsequent screening. Binding capacities and specificities were determined through immunodot assays. LM6 antibodies were similarly generated by coupling of a commercially available (l-5)-a-L-arabinoheptose to BSA, against which monoclonal antibodies were raised and analyzed in a procedure similar to those for LM5. The antibodies have since been used in immunofluorescent labeling studies, which showed the presence and aggregation of RG-I arabinose and galactose side chains in carrot root material . In both cases, generation of LM5 and LM6 was performed by using commercially available polysaccharides and coupling them to BSA. The binding specificities of these antibodies are not highly defined, and thus chemical synthesis is necessary18 to ensure that the binding specificities of the monoclonal antibodies that are generated are unambiguous. Ensuring the specificity of the binding is accomplished by using carbohydrate-based antigens that are pure and well-defined, and then performing inhibition studies. The inhibition studies use a panel of known oligosaccharides as inhibitors to mAB/antigen binding. -11 - R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. Synthesis of the carbohydrate epitope makes chemical manipulation of the RG-I fragments possible, allowing for modifications to be made which could conceivably lead to the generation of any number of monoclonal antibodies representative of all domains of the RG-I pectin, including the backbone. Currently, both LM5 and LM6 are still the “xn 18 two main antibodies being used in composition and structure/function studies of RG-I ’ . There is currently no antibody available which binds to the backbone of RG-I. This thesis describes the synthesis of the carbohydrate portion of the epitope, corresponding to two repeats of the RG-I disaccharide backbone. Several synthetic strategies were employed to allow for future addition to the C-4 site of the rhamnosyl residue, the most highly substituted position on RG-I. The synthetic strategies employed required a variety of protecting group manipulations and glycosylation techniques. A general background on protecting group strategy, glycosylation conditions, and the strategy in relation to late-stage oxidation is provided in the next chapter, followed by a brief discussion on the approach used for the synthesis of RG-I. - 12 - R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. Chapter Three General Synthetic Strategies in Carbohydrate Chemistry Rhamnogalacturonan I is a complex carbohydrate structure that requires the use of several important aspects of synthetic carbohydrate chemistry in its synthesis. Naturally occurring carbohydrate structures in general consist of numerous hydroxyl groups, all of which can be thought of as potential reactive sites. Regioselective protection of these reactive sites (i.e. being able to selectively protect, for example, the C-l, C-2 hydroxyl, etc.) must therefore be used in order to ensure that unwanted reactions do not take place at those sites. An understanding of regioselectivity as it relates to protecting group strategy is therefore required; this topic is covered in more detail, including the strategies used in general carbohydrate chemistry, in the section entitled Protecting Groups on page 14. This section also includes a general introduction to some of the more common protecting groups used in synthetic carbohydrate chemistry. The synthesis of polymeric carbohydrate structures is usually based on commercially available monomers, such as glucose, galactose, and so on. When synthesizing higher order carbohydrate structures, a linking of the synthetic monomers is required. Generally, the anomeric position (C-l hydroxyl) is linked in either the a- or (3-configuration to any one of the free hydroxyl groups (C-2, C-3, etc.) available on a suitable acceptor. Thus, an approach must be used during the glycosylation (linking) of the two compounds that will lead to either the a or (3product. A few strategies that are often employed in synthetic carbohydrate chemistry in order to ensure the correct stereoselectivity is achieved are covered in the section entitled Glycosylation Methods, which can be found on page 21. - 13 - R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. One final, general synthetic strategy must also be addressed here. Several naturally occurring carbohydrate-based compounds contain carboxylic acids; RG-I is one such structure. RG-I, for example, contains galacturonic acid, which is structurally related to the non-carboxylic acid-containing sugar galactose. Although related, the chemistries of the two compounds can be quite different, and the ease with which they can be manipulated varies. One synthetic strategy could, for example, base the synthesis on starting with the carboxylic acid, whereas another strategy may be based on starting with the sugar (galactose, for example) and then converting the sugar at some later stage into the corresponding acid. The latter strategy is referred to as late-stage oxidation. More information on this can be found in the section entitled Late-Stage Oxidation on page 28. The topics discussed here have been selected to provide an overview and understanding of the synthetic strategy used in the synthesis of RG-I. A general discussion on the synthetic approach to RG-I follows at the end of the chapter (page 29). Protecting Groups Protecting groups comprise an important part of all synthetic organic strategies. The successful selective manipulation of reactive sites on a compound is often determined by the effectiveness of the groups present on the other reactive sites in hindering reactivity at those sites. The choice of protecting group is often determined by the need for accessibility to a site. If access to a site, for example, is not required throughout the synthesis, a very stable group, such as a benzyl group, may be chosen to protect several of such sites, whereas a more labile group or one that is cleaved under unique conditions may be chosen for a site where access is required during the synthesis and protection may be more temporary. Strategies can vary according to the compound -14- R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. and required accessibility; however, some common protecting groups are often used in synthetic carbohydrate chemistry. acetyl highly orthogonal orthogonal under certain conditions benzoyl | isopropylidene H | generally not orthogonal TBDPS/TBDMS PMB [ j methyl* I trityl *extremely stable - generally not used other than for the protection o f anomeric hydroxyls or where deprotection is not required Figure 5 Orthogonality o f common hydroxyl protecting groups used in carbohydrate synthetic strategies Many of the carbohydrate compounds investigated for synthesis have their origins in some biological system, and are therefore often complex structures requiring a variety of protecting groups. A number of orthogonal protecting groups are available; orthogonal groups generally are not cleaved under the same conditions i.e. a group cleaved under basic conditions would be orthogonal to one cleaved under acidic conditions. Orthogonality is key to selectively protecting and manipulating reactive sites. A few of the common protecting groups and their relative orthogonality are shown in Figure 5. Two groups which are considered to be “highly orthogonal” generally have significantly different cleavage conditions, and thus both are not easily removed under one set of conditions. Some groups are considered to be orthogonal under certain conditions. For example, the acetate group can be removed under mildly acidic conditions in certain cases without removing a benzoyl group, but the reverse may not be - 15- R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. true. Also, both are cleaved under basic conditions. Orthogonality is thus somewhat dependent on the strategy employed. Table 1 Common protecting groups used in synthetic carbohydrate chemistry, including formation and ________ cleavage conditions_______________________________________________________________ Protecting Group Common Formation Conditions Common Cleavage Conditions acetyl AC2O or AcCl, pyridine, 0°C to rt NaOMe; can also be selectively cleaved in the presence o f benzoyl groups using mildly acidic conditions allyl AllBr, NaH, DMF Isomerization to the vinyl ether, follow ed by cleavage: (Ph3P)3RhCl (Wilkinson's catalyst) and DABCO, followed by cleavage using HgCl2 and HgO benzyl NaH, DMF, BzlBr Hydrogenation in the presence o f a palladium catalyst benzoyl BzCl, pyridine Usually a strong base, such as NaOMe isopropylidene DMP, CSA Acidic conditions, such as H+, H20 and heat or acetic acid, heat, 90% aq. TFA TBDPS/ TBDMS Silyl chloride in pyridine Structure O O O—R O—R Acetic acid/H20 ?1 R— O — S i- Ai Ri = Ph (TBDPS) or Ri = Me (TBDM S) PMB p-OM eBzlBr, NaH, DMF DDQ or CAN (oxidizing agents) MeO - 16 - R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. Table 1 Common protecting groups used in synthetic carbohydrate chemistry, including formation and _______ cleavage conditions_______________________________________________________________ Protecting Group Common Formation Conditions Common Cleavage Conditions Structure methyl CH3I, NaH, DMF Common conditions for cleavage as per Greene: Protective Groups in Organic Synthesis39: BBr3, CH2C12, -78°C —» 12°C OMe trityl TrCl, pyridine 50% acetic acid/ H20 ; H2(g), Pd/C Ph 0 ------------Ph R Ph Table 1 provides a brief description of some of the more common groups and the conditions used, as well as common cleavage conditions. Some of these groups are discussed here. Particulars for the conditions and variations in conditions as they apply to the synthesis of RG-1 can be found in the Results and Discussion section on page 32. The Benzyl Protecting Group The benzyl group is a commonly used protecting group due to the ease with which it can be added to the reactive site and removed later on. Common conditions for protecting a hydroxyl with the benzyl group involve using benzyl bromide (BzlBr) together with a strong base, such as sodium hydride (NaH). The benzyl groups can later be removed during the deprotection stage by placing the compound under hydrogen in the presence of a catalyst such as palladium over carbon (Pd/C), or a similar catalyst. The possible undesired reduction of any alkenes present in the compound must be taken into consideration if hydrogenation to remove the benzyl groups is required; there was no -17- R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. such concern for the synthesis of RG-I, and therefore it was considered to be an ideal group. The only aspect for cleavage that had to be taken into consideration was that cleavage of the allyl group needed to occur prior to hydrogenation, otherwise the allyl group would have been reduced to an alkane, making it almost impossible to remove. Two benzyl groups were used to protect C-2 and C-3 of the galacturonic acid residue. These two sites did not need to be manipulated throughout the synthesis so using the same protecting group at both sites appeared to be convenient. The benzyl group was chosen at C-2 because it was required to be present during later glycosylation of the galacturonic acid residue with the appropriate acceptor (see Glycosylation Methods, page 21, on glycosylation strategies for more information). As previously mentioned, NaH is often the base used in conjunction with BzlBr to produce the benzyl ether. NaH was never used in the synthetic strategy here. Please see Synthesis o f the Monosaccharide Acceptor (page 33) for a discussion on this topic in relation to the synthesis of RG-I and the addition of the respective benzyl groups at C-2 and C-3. The Benzoyl and Acetate Protecting Groups Benzoyl and acetate groups are often used in conjunction with benzyl groups where the selective protection of hydroxyl groups is required. Benzoyl and acetate groups are stable to hydrogenation conditions and can often be quantitatively removed during the final deprotection stage upon addition of sodium methoxide (NaOMe). The use of NaOMe for the removal of both groups is extremely easy and fast, and did not prove to be problematic in the synthesis of RG-I. - 18- R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. A common reaction condition for protecting a hydroxyl with a benzoyl group involves the use of benzoyl chloride (BzCl) with pyridine as the solvent. Addition of the acetate group is similar, with acetyl chloride substituted for benzoyl chloride. Acetic anhydride (A0 2 O) may also be used in acetate formation in place of acetyl chloride. Although both groups are normally removed under basic conditions, the acetate group at C-2 on the rhamnosyl residue was selectively removed in the presence of the benzoyl group at C-3 under mildly acidic conditions. The rhamnosyl residue was selectively protected in this way with the strategy in mind to selectively cleave the acetate at C-2, as previous literature has shown rhamnose can be manipulated in this way (please see Synthesis o f Disaccharides 8 and 9, and Disaccharide Acceptors 10 and 11 on page 39). The Allyl Protecting Group The allyl protecting group is a highly orthogonal hydroxyl protecting group that requires unique conditions to cleave (see Figure 5, page 15 and Table 1, page 16). Cleavage is achieved by isomerization of the double bond of the allyl, thereby producing the vinyl ether. The vinyl ether can then be cleaved with a combination of HgO/HgCh. As mentioned in the discussion on benzyl protecting groups, the only consideration that needs to be made is the order in which the allyl group is removed relative to any benzyl groups, as hydrogenation will reduce the double bond of the allyl group, preventing subsequent removal. The allyl group in the synthesis of RG-I was chosen with the strategy in mind that it would be removed first during the deprotection stage. C-4 of the RG-I molecule is the mostly highly substituted site in terms of side chains (see Figure 4, page 10), making the allyl group, due to its high degree of orthogonality, ideal. The only drawback for the removal of the allyl was the conditions under which it needed to be - 1 9- R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. removed. The process is a two-step process, with a reasonably long workup for both steps, which can result in substantial reductions in yield on small quantities of compound. Please see page 62 for a discussion on the removal of the allyl group and page 84 for reaction specifics. The Isopropylidene Protecting Group Isopropylidene groups can easily and quickly be used to protect diols using 2,2dimethoxy propane and CSA. The group is also fairly stable under a variety of reaction conditions; cleavage is effected under acidic conditions. The primary drawback of using this group is that it can only be applied to 1,2-diols that are in the axial/equatorial configuration. For example, the isopropylidene was ideal for the temporary protection of C-3 and C-4 of galacturonic acid (equatorial and axial configuration, respectively), but could not be used for the protection of C-2 and C-3 of galacturonic acid (equatorial/ equatorial configuration). The para-Methoxybenzyl Protecting Group Although the para-methoxybenzyl (PMB) group was not used in the synthesis of RG-I, it could possibly be used as an alternative to the allyl group on the rhamnosyl residue. Cleavage is usually orthogonal to conditions used for the benzyl group, although cleavage is possible under acidic conditions. Common cleavage conditions use oxidizing agents such as DDQ or CAN. An alternative method for cleavage is the use of primary sulfonamide resin in the presence of a catalyst such as TfOH with dioxane as the solvent, which can result in the near-quantitative transfer of the PMB group to the resin40. Furthermore, the conditions were used on carbohydrate compounds, and no chromatography was required to obtain a pure product. Use of the PMB group could be - 20 - R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. investigated as a replacement for the allyl. This may reduce the loss in yield during workup after removal of the allyl group. The stability of the PMB group during the reaction conditions employed for the synthesis of RG-I would have to be investigated, however. Glycosylation Methods One of the most challenging problems faced in synthetic carbohydrate chemistry is the formation of glycosidic linkages i.e. the linking of one carbohydrate to another. Two types of linkages with respect to the anomeric position are generally possible: the a-configuration and the P-configuration. Forming the desired a- or ■OH ■OH OH HO HO p-configuration is usually dependent on the HO HO 3 i ^ ~ . OR Figure 6 a.D. giUcose (left) in the cis-1,2 group that is present on C -2 and its OR c o n fig u ra tio n a n d a -D -m a n n o se (rig h t) in th e trans-1,2 c o n fig u ra tio n configuration. In order to illustrate the possible variations, consider two common carbohydrate monomers, glucose and mannose (Figure 6), the C-2 epimer of glucose. In the case of these two particular pyranosides, the a- and p-configurations correspond to the linkage at the anomeric position being either axial or equatorial, respectively. For glucose, the C-2 hydroxyl in the a-configuration corresponds to a cis-1,2 linkage i.e. the anomeric linkage is axial, and the hydroxyl or protecting group on C-2 is equatorial; both non-hydrogen groups in this case are on the “same side” of the ring. For mannose in the a-configuration, the anomeric linkage is again in the axial position, but now the C-2 hydroxyl is also axial, and thus on the “opposite side” of the ring. By definition, the mannose glycosidic linkage is trans. -21 - R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. Obtaining the desired linkage in many cases - in particular the cis configuration - can prove to be more than just a trivial exercise. BzO The method for producing ■OBz BzO- cis and trans linkages can be BzO ■OBzl C-1 H O -" BzIO. BzIO •Me Dom, generalized under the strategy of using° non-participating groups and r r ° ° r Acceplor Figure 7 protected A don°r wi,h a alcohol acceptor. The donor generally consists o f a species with a good leaving group at C -l, such as an imidate, thioalkyl group, or halide, which will depart durin8 glycosylation, giving an activated intermediate to which an alcohol bonds. The acceptor can be any compound containing a free hydroxyl. participating groups, respectively. The group in question is the protectmg group present at C-2. If, for example, a glucose monomer is to act as a glycosidic donor i.e. be added to some other hydroxyl acceptor (see Figure 7), such as another carbohydrate, a protecting group strategy must be employed which will place the appropriate protecting group on C-2 so that once glycosylation is attempted, the linkage will be either cis or trans. If, for example, the trans glycosidic linkage for glucose (P-glucose linkage) was desired, a suitable strategy would be to protect the C-2 hydroxyl with a participating group. The participating group would actively take part in the glycosylation mechanism, directing the formation of the P-linkage. An example of this, using galacturonic acid, is shown in Figure 8a, page 23. The galacturonic acid in the example has been prepared with an acetate on C-2, which can participate in the glycosylation according to the mechanism shown (another common participating group is the benzoyl group). A suitable Lewis acid, such as silver triflate, can be used to catalytically aid in the removal of the leaving group at the anomeric position, resulting in the intermediate shown in a of Figure 8. The acetate has effectively “blocked” addition to the bottom face of the monomer, - 22 - R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. directing the acceptor to form a (3- or trans linkage. This strategy tends to be fairly effective. Forming the cis linkage in the example used above requires the use of a non­ participating group on C-2. A typical strategy would involve protecting C-2 with a benzyl group, thereby forming an ether which would not be able to form the intermediate shown in Figure 8a to which the alcohol is added. Addition of the alcohol to the possible intermediate that is formed, again using a suitable Lewis acid such as silver triflate, is shown in Figure 8b. Obtaining the correct addition relies on coordination of the triflate anion as shown in Figure 8b, requiring the anion to provide steric hindrance in order to block formation of the P-linkage. Furthermore, a bulky non-participating group such as the benzyl group could slow down the addition of the acceptor; the longer the reaction takes to go to completion, the greater the possibility of other side reactions becoming a factor. Cis glycosylations therefore tend to be the more challenging of the two types of linkages to form. RO RO RO OMe BzIO- BzIO- + Ag BzIO- NH ■NH C CI3 Figure 8 A possible mechanism involving participating and non-participating groups in glycosylations; route a with a participating group at C-2 leads primarily to a trans product; route b, using a non-participating benzyl group at C-2, hopefully leads to a cis product. RO •OMe O — S — CF- BzIO- H O — R-i -23- R eproduced with perm ission o f the copyright owner. Further reproduction prohibited w ith o u t perm ission. Common Glycosidic Donors A variety of glycosidic donors exist; several of these rely on the use of a suitable leaving group at the anomeric position, followed by activation with a Lewis acid. Three common types of donors shall be discussed here, and include thioglycosides, imidates, and halides. These three particular types of donors have been selectively chosen, as all were considered as possible donors in the synthesis of RG-I. Thioglycosides Thioglycosides generally consist of thioalkyl or thioaryl group at the anomeric position (“thio” referring to the fact they contain sulfur). There are several advantages to thio donors: they can be easily stored with minimal decomposition; the method of activation can be selected to give a trans or cis glycoside; the donor can act as an intermediate to be converted into another donor, such as an imidate41. The thioglycoside can also be used as a donor42 or acceptor43 in the presence of an imidate, making the donor practical for block synthetic strategies. Standard conditions for forming thioglycosides from anomerically-acetylated carbohydrate monomers include using Lewis acids such as borontrifluoride etherate (BF3-Et20), zinc chloride (ZnCh) and tin(IV) chloride (SnCl4)44 together with some thiol45. Formation proceeds via a mechanism similar to that discussed in cis and trans glycoside formation. The presence or absence of a participating group at the C-2 position can thus play a role in the determination of the configuration during glycosylation using the thioglycoside. Common glycosylation conditions using a thioglycoside with a suitable alcohol acceptor include the use of iV-iodosuccinimide (NIS) with either silver triflate (AgOTf) or -24- R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. h3c h3c Figure 9 Probable mechanism for the NIS/TfOH-promoted glycosylation o f a rhamnosyl thioglycoside with an acceptor alcohol to give a trans-1,2 linkage triflic acid (TfOH)46, methyl triflate (MeOTf)47, or dimethyl(methylthio)sulfonium triflate (CH3 SS(CH3)2+OTf (DMTST)45. These promoters can be used for generating trans glycosides; upon the addition of tetrabutylammonium bromide (BujNBr) in the presence of DMTST with a nonparticipating group at C-2, cis glycosides have also been successfully formed in reasonably high yields48. Glycosylation in this instance occurs through the in situ conversion of the thioglycoside into the bromide. A possible mechanism for the NIS/TfOH-promoted cis-1,2 glycosidic linkage between a rhamnosylthioglycoside donor and a generic alcohol acceptor (ROH) is shown in Figure 9. Imidates Imidates can be used in a similar way to thioglycosides. Formation of a cis-1,2 acetimidate is often performed by adding trichloroacetonitrile and 1,8diazabicyclo[5.4.0]undec-7-ene (DBU) to the dissolved carbohydrate hemiacetal; conversely, if the trans-1,2 acetimidate is desired, l,4-diazobicyclo[2.2.2]octane (DABCO) may be used49. An example of donor activation using AgOTf is shown in Figure 8 on page 23. Imidates are also practical in that they are fairly stable and can thus -25- R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. be easily stored. Activation conditions are similar to those mentioned for thioglycosides. Mechanistically, the formation of the glycosidic linkage is subject to the presence or absence of a participating group on the C-2 hydroxyl. Glycosyl Halides Bzio°v 0Me One of the earliest, reliable nn \7r~rN AcO I methods of glycosylation used was OR AcO !+ I Ag through the formation of halide donors. Figure 10 Formation of a glycosidic linkage via an S ^-like mechanism Most commonly, the bromide is formed, although chlorides are sometimes used; the chloride is less labile than the bromide and is therefore more stable, and not as good a leaving group. Formation of the chloride can be achieved using thionyl chloride together with the appropriate hemiacetal50. For the purposes of this discussion, the focus will be on the bromide. Formation of the bromide can occur either from the 1-0-acetate or from the hemiacetal. There are numerous methods and conditions used to form the anomeric halide. Hydrogen bromide is often added to the 1-0-acetate, with the glycoside dissolved in a suitable solvent, such as dichloromethane51. Alternatively, the bromide can also be generated from the hemiacetal by using, for example, A-bromomethylene-A,Adimethylammonium bromide52. Once the bromide has been generated, glycosylation can occur through activation by a suitable Lewis Acid, such as AgOTf with or without a base such as Ag2 C0 3 or collidine present52,53. Generally, the base is used to promote orthoester formation, which is an intermediate in some instances, but may not be necessary in glycosylations which proceed via an S^-like mechanism53 (Figure 10); however, too much base can result in -26- R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. lower glycoside formation, as the conditions must be slightly acidic for the alcohol to react with the newly formed orthoester. A probable mechanism for the formation of the orthoester intermediate and the subsequent glycosylation is shown in Figure 11. (Note: An orthoester intermediate may form where a participating group is present on C-2 and the halide is trans to the C-2 hydroxyl, as in mannose; this is shown in Figure 11. If the participating group and halide are cis as in galactose (Figure 10), no intermediate can form and the reaction follows an SN2-like mechanism.) \ H Figure 11 Formation o f a trans-1,2 glycosidic linkage via an orthoester intermediate using a glycosyl bromide as the donor As with the thioglycosides and the imidates, the same problem forming the cis linkage arises with glycosyl halides, and the same non-participating group strategy is used. When a halide salt such as tetraethyl ammonium bromide (E^NBr) is used in the glycosylation, the predominate product formed is the cis-1,2 glycoside when no participating group is present on C-248. This technique is referred to as halide ion catalysis. The labile nature of the bromide group can make isolation of the glycosyl bromide difficult; one way of circumventing this problem is to use a one-pot method in which the bromide is generated in situ from the hemiacetal and, upon generation, -27- R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. subsequently glycosylated in the same reaction flask with the acceptor alcohol54,55. Onepot conditions published by Shingu et al. have shown high selectivity for the formation of cis-1,2 glycosides with a variety of acceptors. More information and the specifics of the conditions used can be found on page 49 in conjunction with a discussion on the glycosylation methods used in the synthesis of RG-I. Late-Stage Oxidation Several naturally occurring carbohydrates of synthetic interest exist as carboxylic acids or methylated ester derivatives thereof. Two such common carbohydrates are Figure 12 D -g a la c to se (left) a n d D -g ala c tu ro n ic a c id (rig h t) glucuronic acid and galacturonic acid, the carboxylic acid-containing relatives of glucose and galactose, respectively (Figure 12). Although related, the chemistries of the sugar and its acid derivative (termed a ‘uronate’) can be considerably different. Historically, galactose is often used, for example, followed by late-stage oxidation i.e. C-6 of galactose is oxidized later in the synthesis to the carbonyl rather than starting with galacturonic acid56. Although late-stage oxidation is generally easily achieved at the monosaccharide level, the difficulty in effectively oxidizing oligosaccharides can be considerable45. One of the means of overcoming the need for late-stage oxidation is by starting directly with the acid derivative. As previously mentioned, the chemistry of uronates can be considerably different from that of the normal sugar. The carboxylic acid group results in a highly polar compound which, for practical purposes, needs to be protected as an ester to increase its solubility in organic solvents, for example. Furthermore, the acid can be subjected to undesired side reactions during protecting group manipulation that would -28- R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. otherwise not occur in the normal sugar. An example of this is 3-elimination using what would be considered standard protecting group chemistry by protection of a hydroxyl on the sugar molecule by a benzyl group using benzyl bromide in the presence of sodium hydride (see Results and Discussion for compound 3, page 34). The literature involving oligosaccharide synthesis starting from the uronate monomer is not particularly extensive, and larger oligosaccharide syntheses are extremely rare. Please see the Synthetic Approach to RG-I in the next section and the Results and Discussion section on page 32 for more information on problems encountered working with the galacturonic acid derivative for the purposes of synthesizing RG-I. Synthetic Approach to RG-I RG-I is a highly diverse structure and presents many challenges from a synthetic aspect. The goal of this research was to successfully synthesize and characterize an unprotected tetrasaccharide corresponding to two repeats of the rhamnose-galacturonic acid disaccharide found in the RG-I backbone. The rhamnosyl residue was designed for the possible addition of side chains at the C-4 position of galacturonic acid, and had been previously synthesized57. There are several different approaches that could have been taken in the synthesis of the RG-I backbone. Most of the literature on galacturonic acid uses galactose as the starting material. Galactose is generally easier to work with as it is less susceptible to 3elimination reactions (see Results and Discussion for compound 3, page 33) and does not contain a highly polar carboxylic acid functional group. Furthermore, carboxylic acids generally need to be protected as an ester in order to make working with the compound easier. Transesterification may also occur under certain acidic conditions. -29- R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. Synthesis of the tetrasaccharide from galactose would have required the oxidation of galactose to galacturonic acid at some point in the synthetic scheme. Oxidation can generally be performed at one of three points: oxidation of the monomeric acceptor, oxidation of the disaccharide, or oxidation at the tetrasaccharide level. A synthetic approach involving the oxidation of galactose at the tetrasaccharide level has already been documented57,58; Maruyama et al. published the target tetrasaccharide, with oxidation performed in the final step using 2,2,6,6tetramethylpiperidine 1-oxide (TEMPO) on 18.8 mg of starting material with a yield of 50% (10 mg). Attempts to oxidize in the final step were also made in this lab under conditions similar to those used by the Maruyama group; however, a mixture of components was obtained. Furthermore, Maruyama's group used a non-convergent strategy that does not allow for the addition of branches at the C-4 position of the rhamnose residue. As a result of the difficulties AcO encountered in the final oxidation BzIO BzIO BzO' step, the synthetic strategy had to be OAII BzIO BzIO BzO' revised, and a new approach of IMe OAII working with galacturonic acid at Kgure 13 Tetraalcchari(fcs 18 (R , Me) 19 (R , Bzl) the monosaccharide level was attempted. This new synthetic strategy began with commercially available galacturonic acid and involved a convergent block synthesis approach which eventually led to tetrasaccharide 20. This approach allowed for the maximum application of the disaccharide building blocks in the manufacture of a suitable donor and acceptor for eventual glycosylation, giving the fully protected tetrasaccharides 30- R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. 18 and 19; the need for unique synthetic approaches for building a suitable donor and acceptor was thereby avoided. Figure 14 illustrates the block synthetic approach used the synthesis of the target compound (20). AcOAIIO BzIOBzO I OAc BzIO BzO' OMe iMe HO- AcO- BzIO- BzIOBzIO BzO' OC(=NH)CCI3 BzIO BzO' (Me OAII (All •OMe AcOBzIO- BzIO BzO OAII BzIOBzIO BzO' (Me OAII HOHOHO HO' OH HOHO HO' OH Figure 14 General scheme showing the block synthesis approach o f the target tetrasaccharide -31 - R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. Chapter Four Results and Discussion Synthesis o f the Monosaccharide Acceptor Late stage oxidation of galactose to galacturonic acid at the tetrasaccharide level was avoided by purchasing commercially available D-galacturonic acid from Fluka. The carboxylic acid group and the anomeric position were conveniently protected by simple reflux with Amberlyst® 15 H+ resin beads in methanol, as prepared by Schmidt and Neukomm59. Formation of 1 was dependent on reflux times. During reflux, according to Schmidt and Neukomm, a variety of isomers are present, with approximately 41% percent of the desired a-pyranoside isolatable after 70 horns. COOMe COOH COOMe OMe D-Galacturonic Acid (GalA) 1 OMe 2 c COOMe COOMe COOMe OMe HO HO COOBzl 5 COOMe 6 Scheme 1 Synthesis of the monosaccharide galacturonic acid acceptor. (a) MeOH, H resin; (b) 2,2-dimethoxypropane, CSA, acetone; (c) BzlBr, Ag20 , CH2C12 (87.1%); (d) AcOH, 90% aq. TFA, 70°C (87.4%); (e) i. Bu2SnO, toluene; ii. Bu4NBr, BzlBr (26.0% 5 and 66.5% 6). -32- R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. Once all the precipitate had been collected (see Experimental (1)), the remaining mother liquor was conveniently redissolved in methanol, acidic resin added, and set to reflux until equilibrium had once again been established. Further precipitate was collected and the procedure repeated several times over. Although purification of the precipitate was possible by means of recrystallization, the impure product was carried forward to the next step. Analysis of the 1H NMR spectrum of the recrystallized product confirmed the presence of two methyl groups. The two signals were assumed to be due to the methyl ester (5 3.79 ppm) and the methyl on the aglycone (5 3.42 ppm). (Note: The aglycone refers to C-l of the reducing end of the sugar, in this case galacturonic acid.) Isopropylidation of 1 was readily achieved. The insoluble starting material slowly dissolved in the acetone as the reaction progressed, giving a visual indication when the reaction was complete. Crystallization yielded white crystals that showed no trace of the impurity present in the starting material, giving a very pure product (2). The ‘H NMR spectrum showed the presence of the two isopropylidene methyl groups as singlets (integration of 3 protons for each group) at 8 1.50 ppm and 8 1.36 ppm. The next step required the 2-O-benzylation of the isopropylidene-protected galacturonic acid residue. Benzylations can generally be performed under basic conditions using sodium hydride (NaH) and benzyl bromide (BzlBr). Several publications are available which have studied various 4-0 substituents that can lead to p~ elimination (the formation of 4,5 unsaturated 4-deoxy-hexuronates) at the C-4 position of esterified compounds. Kovac60 et al. studied this problem specifically using 2 as the model compound and indeed found that the isopropylidene protecting the C-4 and C-5 -33- R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. hydroxy groups was eliminated in this fashion in the presence of base. Elimination has been attributed to the acidic nature of H-5. Based on the results by Kovac, the use of NaH was not attempted. Milder conditions were used52 where 2 was dissolved in freshly distilled CH2 CI2 , followed by the addition of benzyl 2,2,2-trichloroacetimidate and a catalytic amount of triflic acid. Neutralization was either performed by the addition of triethyl amine followed by filtration through celite, or by direct filtration through basic alumina. Results proved to be inconsistent, with a maximum yield of 77% obtained and yields as low as 45%. Due to the inconsistency of the results, an alternative method was desired. A publication by Kovac using Ag20 61 (see Experimental (3)) was used to produce the identical compound desired here. Obtained yields were consistent for various attempts. Workup was generally cleaner and a much purer product was readily obtained; however, if the collected product was not 100% pure, it was still possible to take the crude product onto the next step without consequence. The primary drawback of this method was the reasonably large amount of benzyl bromide (strong lachrymator) that was required for larger scale reactions. There is a discrepancy in the literature melting point value61 (see Experimental (3)) and the value obtained for compound 3; the optical rotation information obtained also did not match the literature value. The melting point was recorded for three different attempts on one occasion, and recorded once more to confirm the values approximately two months later. All four results were identical. Optical rotation data was recorded twice on two different occasions, with both results being identical. Composition analysis was not performed to confirm purity, and no NMR data was provided in the literature for -34- R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. comparison. Investigation of the NMR spectra for 3 indicated the correct compound. A considerable upfield shift of the signal corresponding to H-2 can be seen when comparing the spectra of 2 and 3 (6 3.98 ppm to 8 3.62 ppm), reflective of a change in the electronic environment at that position. The 'H spectrum also clearly shows the presence of the signals corresponding to the isopropylidene group (two singlets at 5 1.38 ppm and 5 1.35 ppm, integration of 3) and the benzyl group (a multiplet between 8 7.42-7.24 ppm, and two doublets at 8 4.70 ppm and 4.68 ppm, corresponding to CH2 of the benzyl group). Removal of the isopropylidene group for the synthesis of 4 was initially attempted according to Nolting et a l52 as described in the Experimental for 4, but with ambient temperature being interpreted as room temperature; results were poor, so the removal of the isopropylidene group was then attempted using water and Amberlyst® 15 H+ resin beads, also at room temperature; results were also not particularly good. After a few attempts using water, Nolting's method was revisited. Reagent equivalents were kept the same as in the initial attempts, but the solution was heated in an oil bath at 70°C. On TLC, removal of the isopropylidene group appeared to proceed quantitatively and very quickly. In cases where slightly impure 3 had been carried on to this step, pure 4 was easily obtained after recrystallization. 'H NMR showed the successful removal of the isopropylidene group by the absence of the two singlets at 8 1.38 ppm and 8 1.35 ppm that were visible in the spectrum for compound 3. Selective addition of the benzyl group at C-3 was performed by stannylene acetal formation under conditions similar to those cited by Nolting el al.52 Monitoring by TLC of the initial stannylene acetal formation proved to be impossible, and the quantity of starting material somewhat affected the time required; too short a time on too large a -35- R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. scale led to poly-benzylation and a mixture of C-3- and/or C-4-benzylated products. One possibility for following the reaction would have been to remove aliquots of the reaction mixture at various times during the reaction and analyzing the samples by NMR; however, this was not performed. Once the second step (post addition of benzyl bromide - see Experimental for 5 and 6) had been initiated, 3-0-benzylation of the methyl ester product first occurred (6), with the eventual formation of some of 5. The maximum yield of 6 obtainable upon the complete disappearance of starting material as indicated by TLC was approximately 66% each time; 5 was always present. Allowing the reaction to go further than simply to the point where all starting material had been consumed resulted in greater formation of the benzyl ester 5 and subsequent lower yields of 6. The reaction conditions were never such that the majority of the product obtained was the benzyl ester, although this may have been possible by perhaps adding further equivalents of benzyl bromide. At this stage, both acceptors 5 and 6 were desired in the glycosylation to form disaccharides 8 and 9 (see Scheme 2 on page 38). The strategy was to transform both these disaccharides into donors, which could be then used in glycosylation reactions to give tetrasaccharides 18 and 19. At the same time it would be possible to investigate the potential differences in reactivity of the presence of the benzyl ester versus the methyl ester protecting the carboxylic acid. Also, by demonstrating that both 5 and 6 could be incorporated in some way into a disaccharide and ultimately be used to generate the desired tetrasaccharides, neither the methyl (6) nor the benzyl monosaccharide (5) would be considered a loss at this stage of the synthesis. -36- R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. The 'H NMR spectra of both 5 and 6 indicated successful addition of the benzyl group at C-3. An identical downfield shift of the signals corresponding to H-3 was seen for both benzylated compounds from 8 3.77 ppm (compound 4) to 8 3.95 ppm. Synthesis o f Disaccharides 8 and 9, and Disaccharide Acceptors 10 and 11 The thioglycoside donor57 (7) used for the SEt synthesis of both disaccharides 8 and 9 had been previously synthesized in our lab, and available quantities were used in the respective glycosylations. Initial glycosylations were performed at room temperatures using 7 OAc Figure 15 The rhamnosyl donor used for the synthesis o f the disaccharides A-iodosuccinimide (NIS) and catalytic amounts of triflic acid in CH2C12. Reimer et al. glycosylated 7 with a galactose acceptor similar to 5 and 6 (methyl 6-0-benzoyl-2,3-di-0-benzyl-p-D-galactopyranoside was used)57 and reported a yield of 91%. Identical conditions using the galacturonic acid residues (compounds 5 and 6) as acceptors resulted in much lower yields (approx. 65% for 8 and 63% for 9; see Scheme 2, page 32). Although the yields appeared to be somewhat satisfactory, yields in the literature for these types of reactions were often considerably higher. Magaud et al. performed some glycosylations using various galacturonic acid thio donors with galacturonic acid acceptors62. Reported yields were higher than the yields initially observed here. The reactions published by Magaud were generally performed at -60°C, which prompted an investigation into whether or not temperature would have a significant effect on the yield of the glycosylations attempted here. -37- R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. AcO BzO \ I n n Y. ^ r / Bzl°lOMe is £,AII 8, 9; R=Me, B zl respectively R=Bzl R=Me HO HO BzIO BzIO BzIO BzO BzIO BzO OMe OMe OAII OAII 10 11 Scheme 2 Synthesis o f the disaccharides and the respective acceptors. (a) /V-iodosuccinimide, triflic acid (cat) in CH2C12, -40°C (77.5% 8 and 80.1% 9); (b) 3% methanolic HC1 in CH2Cl2. (79.7% 10 and 36.6% 11) The first attempt at glycosylating donor 7 at approximately -40°C (pyr/N2 (f) bath) was made with acceptor 5. The calculated amount of triflic acid to be added was 10.8 pL (0.10 eq. to the donor); five times the calculated amount was mistakenly added (50 pL; 0.45 eq. to the donor). The resulting yield was 80.1%, significantly higher than the previous 63%. Due to a limitation in access to quantities of donor 7, subsequent glycosylation attempts were made with 6 as the acceptor, as this product was desired for conversion to the disaccharide donor (14) and disaccharide acceptor (10). An initial attempt was made by simply reducing the temperature to about -40°C. The resulting yield was approximately 66%. The reaction was performed again at -40°C, but using 0.45 eq. of triflic acid. The yield increased to 77.5%, indicating the greater equivalents of triflic acid were key to improving yield. Although the triflic acid acts as a catalyst in the glycosylation, the larger quantities may have increased the rate of product formation, -38- R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. thereby decreasing the opportunity of byproducts to form. No attempt was made at room temperature with the higher equivalents in order to determine whether or not the lower temperature is absolutely required to obtain the preferential yield. NMR spectroscopy was used to determine whether the newly formed glycosidic linkages for 8 and 9 between the rhamnosyl donor and the galacturonic acid acceptor were a or p. The C'-H1coupling for 8 at C-l (the anomeric position) of the rhamnosyl residue, as determined by analysis of the HMQC (inverse) spectrum, revealed a one bond coupling constant of 173.3 Hz, indicative of the a-linkage. Comparatively, a coupling constant around 158-160 Hz in the C'-H1correlation spectrum would indicate a P-linkage at the anomeric position of rhamnose. Anomer signals for a-linked rhamnose residues also display a small J \j tram coupling value (2.2 Hz here) in the proton spectrum, higher than observed J\.i coupling constants for rhamnose cis linkages, which have been documented in the <1 Hz range . Generation of the disaccharide acceptor (10) was primarily attempted from 8. Removal of the 2-O-acetate on the rhamnosyl portion of the disaccharide was accomplished according to methods published by Byramova et a l64 Three percent methanolic HC1 (prepared by addition of 1 mL acetyl chloride to 35 mL distilled methanol, stirred under N2(g) in an ice bath) was used to selectively remove the acetate without affecting the benzoate, which would have been removed had the removal of the acetate been attempted using a base such as sodium methoxide. The rhamnosyl donor used in the generation of the disaccharides had, in fact, been orthogonally protected with the strategy for the removal of the 2-O-acetate discussed here in mind. Removal proved to be successful, resulting in a respectable yield of 79.7%. -39- R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. Using the benzyl ester (9) as a disaccharide acceptor was never seriously considered because it was thought 9 would most likely lose its benzyl ester functionality due to the slightly acidic conditions required to remove the acetate. These conditions are similar to ones that would be employed to perform a transesterification; however, out of interest, the reaction with 9 was attempted once simply to verify whether or not the expected outcome would occur. The reaction was allowed to proceed for approximately 77 hours, at which point two main upper spots and two main lower spots could be seen on TLC using a combination of UV light and charring as a means of detection. Chromatography performed on the compounds corresponding to the two upper spots was used to separate the two compounds into the benzyl ester starting material 9 (5.3% yield) and the benzyl ester acceptor 11 (36.6% yield). Similarly, the compounds corresponding to the two lower spots were also chromatographed to afford the fully protected methyl ester disaccharide 8 (7.3% yield) and the methyl ester acceptor 10 (44.2% yield). Although this method would not be practical for generating large quantities of the benzyl ester disaccharide acceptor (11), the results of this reaction demonstrate that if the benzyl disaccharide was not useful in synthesizing the tetrasaccharide, it could be readily and easily converted into the fully protected methyl ester disaccharide and the methyl ester disaccharide acceptor, both of which are desired compounds. *H NMR and 13C NMR spectroscopy was used to confirm the successful removal of the acetate on the rhamnosyl residue. The 'H NMR spectrum also showed that the remaining protecting groups were still present (i.e. benzyls, benzoyl and the allyl, as well as the methyl ester for 10). The successful removal of the acetate to generate 10 was evident by the lack of the acetate signal in the 'H spectrum (5 2.0 ppm for 7) and the -40- R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. disappearance of the I3C signal (8 21.0 ppm for 7). Similarly, JH NMR and 13C NMR spectroscopy was used to verify the removal of the acetate for 11. Synthesis o f the Disaccharide Donors \ U or2 AcO BzIO BzO OAII R2=Me OR 8 “^ 1 2 13 cC=^ 14 R2=Bz1 aj=±15 bt ± 1 6 cC=^ 17 Ri CH3 Ac H C(NH)CC13 Scheme 3 Synthesis o f the disaccharide donors. (a) 1% H2S 0 4 in AcOH and Ac20 (81.1% 12 and 91.4% 15); (b) 4 A powdered molecular sieves in CH2Cl2/MeOH (89.3% 13 and 76.0% 16); (c) trichloroacetonitrile, DBU, in CH2C12 (81.3% 14 and 81.9% 17). Initial attempts at generating 12 proved to be successful using known acetolysis conditions (see Experimental for 12)51. Reactions were allowed to proceed for four to four and a half hours, with monitoring attempted using 1:1 hexanes :ethyl acetate as the developing solvent for TLCs. The progress of the reaction was almost impossible to determine but, nonetheless, yields in the 70-80% range were obtained. Attempts were also made to follow the reaction by taking an aliquot of the neat reaction mixture and monitoring it by *H NMR. This also proved to be difficult because results as to whether the anomeric methyl group had been successfully removed were ambiguous. Another methyl peak was evident at approximately the same chemical shift as where the anomeric methyl would be expected (it was later determined the methyl peak was not due to the presence of the anomeric methyl, but some other species). The same conditions used to synthesize 12 were then attempted for the synthesis of 15, the benzyl ester compound, with reaction times varying between four to five hours. The resulting yields were quite low, ranging between 58% and 63%. Consequently, 41 - R eproduced with perm ission o f the copyright owner. Further reproduction prohibited w ith o u t perm ission. several different solvent systems were explored for monitoring by TLC using known samples of the starting material and the desired product. A mixture of 2:1:1 hexanesxthyl acetate:chloroform gave enough separation on TLC so that the reaction could be easily monitored. TLC showed the reaction to be complete after approximately two hours; longer reaction times most likely resulted in loss of the ester, as charring at the baseline of the TLC plate was evident, which could possibly have been due to the very polar carboxylic acid. Using a much shorter reaction time of two hours resulted in an increase in yield of 25-30%. The reported yield of 91.4% for this reaction was achieved despite some problems and consequent losses incurred during loading of the chromatography column during purification. Acetolysis to give 12 was revisited as a result of the findings for 15, and a mixture of 1:1:1 hexanes:ethyl acetate xhloroform was used to monitor the reaction by TLC. TLC showed the reaction to be complete after approximately two hours; although yields remained about the same, the reaction could be worked up after a shorter period of time. Purification of the product for both 12 and 15 by chromatography resulted in an 1 IT inseparable (by LPLC) a/p mixture. The resulting H and C NMR spectra therefore naturally reflected the mixture, so various reporter groups were used to identify the product. Although coupling constants for the *H spectrum for 12 could not be clearly assigned, reporter signals showed all the expected groups were present (i.e. allyl, benzoyl, etc.; see Appendix 1 for assigned 'H NMR spectra relating to 12). Furthermore, examination by COSY clearly showed a major component and a minor one, with chemical shifts for the ring protons of both being quite similar, and a notable chemical shift for the galacturonic acid H-l signal from 4.71 ppm to 6.50 ppm evident. HMQC -42- R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. (inverse) analysis also confirmed the a-linkage of the rhamnosyl residue to the galacturonic acid residue; the one bond C'-H 1 coupling constant for the anomeric carbon signal of the rhamnose residue was determined to be 174.8 Hz, characteristic of an alinkage. Analysis of the 13C spectrum was performed and, as expected, revealed the absence of the methyl on the aglycone (C-l of the galacturonic acid residue in this instance), for which the signal was determined to be at 5 56.3 ppm for 8 . Two peaks were now evident in the acetyl region of the I3C spectrum. The peak at 5 21.0 ppm was assumed to be due to the acetate on rhamnose, as it corresponded to the same chemical shift for the acetate for 8 , and a significant shift was not expected. There was also evidence of a smaller peak buried under the same signal due to the lesser anomer. A new peak was now evident at 2 1 . 2 ppm, which was attributed to the 1 -0 -acetate on galacturonic acid. The *H spectrum revealed an additional major and minor peak to the major and minor peaks of the rhamnose acetates for the two anomers and showed the anomeric methyl group signal previously found at 3.39 ppm was now absent. The HMQC (inverse) spectrum aided in assignment of the respective acetate signals; the one acetate signal, with a 13C chemical shift of 8 21.0 ppm, corresponded to the major ]H signal at 8 2 .0 1 ppm, which in turn was similar to the documented chemical shift of the acetate 13 signal for rhamnose as determined for 8 ( 8 2.00 ppm). The other acetate signal in the C spectrum at 8 21.2 ppm corresponded to the major ' H signal at 8 2.12 ppm. The NMR spectra for 15 were analyzed in a similar way to 12 by looking for various characteristic signals. The 1H and COSY spectra were analyzed to ensure the necessary protecting groups were still present, as well as to assign the ring protons. As -43- R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. for 1 2 , coupling constants could not be determined due to the inseparable a/p mixture. Similar to 12, analysis of the COSY spectrum revealed a significant chemical shift for the H-l galacturonic acid signal from 8 4.72 ppm for 9 to approximately 5 6.52 ppm and the absence of the anomeric methyl signal, which was previously reported at 3.38 ppm for 9. With the aid of the HMQC (inverse) spectrum, the anomeric acetates were found to correspond to the minor and major peaks at 8 2 . 1 2 and 8 2 . 1 0 ppm in the *H spectrum, respectively. The acetates on the rhamnose residues of the minor and major anomer corresponded to the respective peaks at 8 2.05 and 8 2.01 ppm, in agreement with the previously reported chemical shifts for 9. The corresponding peaks were also evident in the 13C spectrum at 8 21.3 ppm (anomeric acetate) and 21.0 ppm (rhamnosyl acetate). The 13C spectrum was used to confirm the absence of the methyl group on the aglycone, the signal for which was previously reported at 8 56.3 ppm for 9. Removal of the acetates to generate both hemiacetals 13 and 16 for subsequent transformation to the respective imidates was first attempted using a procedure published by Mori et al.65 (hydrazine acetate in DMF). Yields for both the methyl and benzyl ester disaccharides were in the 50% range. Most of the attempts made using these conditions resulted in long smears on the TLC plate when monitoring the reaction, rather than tight spots. When attempting to chromatograph the product in what appeared to be a reasonable eluting solvent, the syrup formed large amounts of insoluble precipitate. This indicated that perhaps in both cases (i.e. attempted formation of 13 and 16) the ester was being inadvertently deprotected, resulting in the very polar, non-soluble carboxylic acid. The solution to removing the 1-O-acetate in an effective and simple manner was recently published by Kartha et al.66 Kartha used various sugars as model compounds, -44- R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. dissolved these compounds in methanol, and added 4 A powdered molecular sieves, equal in weight to the sugar being deacetylated. The concept behind this method of de-0acetylation was that, due to the makeup of the molecular sieves and the presence of methanol, small amounts of sodium methoxide were being generated - a reagent frequently used for deacetylations. The amounts of sodium methoxide produced were small enough so as not to affect the rest of the compound (i.e. removal of the rhamnosyl acetate and/or benzoate); the conditions were considerably milder than those published by Mori, and workup and removal of solvent was far simpler. The first attempts with Kartha’s method were made using equivalents of molecular sieves higher than those published. None of the model compounds used by Kartha contained the ester functionality, and it was thus far unknown how easy it would be to deacetylate either of the two disaccharides. Compound 13 was successfully formed after 11 hours with a respectable yield of 67.4% (1.5 equivalents, by weight, of molecular sieves to 12); more importantly, the product was easily isolated without the smearing that was seen when using the previous method. The 1.5 equivalents of molecular sieves used for 12 appeared to be a bit too strong based on the low yield and the slightly yellow tinge of the solution, so when attempting the same reaction with 15, only 1.3 equivalents was used. The yield was only 51.7%; however, the product was once again easily isolatable. Dichloromethane was eventually introduced as part of the reaction mixture in the hopes of increasing the solubility of the product and minimizing the amount of degradation of both the starting material and product, as well as slowing down the reaction by making the conditions milder. Although the yield reported here for 16 was -45- R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. only 76%, a yield as high as 81% had been achieved by careful monitoring and continual optimization of the reaction conditions. The primary degradation product is hypothesized to be due to a loss of the benzoyl protecting group on the rhamnosyl donor, as both reaction mixtures (for 13 and 16), when left too long or put under conditions that were too strong, turned a bright yellow; no further tests were performed, however, to verify this. The successful synthesis of hemiacetals 13 and 16 was verified by NMR spectroscopy. The removal of the acetate groups from the aglycone of 13 was confirmed by the presence of only two acetate peaks of approximately the same intensity in the *H spectrum at 5 2.06 and 6 2.01 ppm. The intensity of the peaks indicated that the a/p mixture was approximately 50:50. Similar relative intensities were seen for the easily identifiable CO2CH3 and rhamnosyl H- 6 peaks, indicating that indeed both observed acetate peaks were due to the anomers of the deacetylated compound rather than to the presence of the rhamnosyl acetate and the acetate on the aglycone. Investigation of the 13C spectrum confirmed the removal of the anomeric acetates by the absence of the 1-0acetate signal at 5 21.2 ppm. The 13C spectrum further confirmed the presence of an anomeric mixture by the presence of two relatively equal rhamnose C-l signals and two relatively equal galacturonic acid C-l signals. The investigation of hemiacetal 16 was performed in a similar manner to that of hemiacetal 13. The 'H, 13C and COSY spectra for 13 all clearly provided evidence of an a/p mixture, whereas only the COSY spectrum for 16 clearly showed a mixture due to a variance in the chemical shift of the rhamnosyl H-l and H-2 protons for the minor anomer; the other signals for the minor anomer appeared to be buried directly underneath those of the major anomer. Investigation of the 1H and 13C spectra confirmed the -46- R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. anticipated results: the disappearance of the 1-O-acetate signals. The 'H spectrum only contained the major acetate signal at 8 2.03 ppm and a minor signal at 8 2.06 ppm. The 11 previously reported signals at 8 2.10 and 8 2.12 ppm were absent. The C spectrum also confirmed the removal of the anomeric acetate by the disappearance of the previously reported acetate signal for 15 at 8 21.3 ppm; only the signal for the rhamnosyl acetate could be seen at 8 21.0 ppm. Interestingly, the intensities of the methyl ester and acetate peaks for the two anomers of 13 indicated the mixture was almost 50:50 a/P; however 1- 5 1 only very minor peaks were evident in the C spectrum for 16, and the H spectrum indicated that mixture was closer to 75:25. Whether the a- or the P-anomer was the major one present was never investigated for either of the 1 -O-acetate mixtures or for the hemiacetals. Formation of imidates 14 and 17 was easily achieved as documented in the Experimental section. The primary inconvenience in this stage was purification. Purity could only be achieved by immediately placing the reaction mixture upon completion of the reaction on a 4:1 hexanes:ethyl acetate vacuum column. If the reaction mixture was left without immediate purification, breakdown was evident. After the collection of the samples containing the product, chromatography had to be performed twice using two different eluting solvents. Despite the need for multiple purification steps, yields remained reasonably high as documented in the Experimental section. Storage of the imidate for longer periods of time, however, proved to be a problem. Products which had been stored in the freezer (-20°C) for extended periods of time showed some breakdown of the imidate to what appeared to be the hemiacetal. This was most likely due to the lack of a moisture-free environment. -47- R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. Successful conversion of hemiacetals 13 and 16 to imidates 14 and 17 was primarily determined by inspection of the 13C spectra and the XH spectra. The 13C spectra were analyzed for two signals which corresponded to the -C=NH and -CCI3 carbon signals occurring in the 8 160 ppm and 5 90 ppm regions, respectively, as documented for other literature trichloroacetimidate compounds42,67; the 'H spectra were analyzed for a corresponding -C =N // signal at approximately 8 8.60-8.70 ppm. The appropriate signals for 14 could be seen at 8 160.7 ppm and 8 91.3 ppm, and at 8 160.7 ppm and 8 91.4 ppm IT 1 for 17 in the C spectra; the H spectra contained signals at 8 8.63 ppm for 14 and 8 8.61 ppm for 17 that corresponded to the expected N // signals. Of note is that the methyl ester imidate appeared to be present almost entirely as a single anomer only (most likely the aanomer under the conditions used49) and the benzyl ester imidate was a mixture of anomers. The COSY spectrum for 17 clearly showed a second set of signals for the minor anomer, and the 13C spectrum showed a minor peak at 8 99.4 ppm in the region of C-l of the rhamnosyl residue. Normally, under the conditions used for formation of the imidates, only one anomer is expected49. Both imidates 14 and 17 were generated using identical conditions, yet a mixture of anomers for 17 was obtained, suggesting perhaps a difference in conformation or interaction with the reagents during the reaction when trying to form 17. The other possibility was that one of the anomers of 14 was removed during purification, but there was no TLC evidence to confirm this. The two imidate donors mentioned here were the only donors which gave results leading to the successful formation of the desired tetrasaccharide product. Other forms of glycosylation and donor generation were attempted; these will be discussed in the following section. -48- R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. Synthesis o f the Fully Protected Tetrasaccharide, RG-I AcO- HOBzIO BzIOBzIO BzO' OC(=NH)CCI3 BzIO OMe BzO' OAII 14 ( R - M e ) or 17 (R=Bzl) AcO BzIOBzIO BzO' OAII BzIOBzIO OMe BzO' OAII 18 (R - Me) and 19 (R=Bzl) Scheme 4 Synthesis o f the folly protected tetrasaccharides. Conditions: AgOTf, 4 A molecular sieves in CH2C12 (38.8% 18 and 31.7% 19). Glycosylation of disaccharide donors 14 and 17 with acceptor 10 to generate 18 and 19, respectively, in high yield proved to be unsuccessful. Several different glycosylation methods were investigated in order to maximize yield, but only one of these proved to result in any product formation at all, with maximum yields in the 30% range. Silver triflate (AgOTf) was the only promoter that was successful in generating the respective target tetrasaccharides when dichloromethane was used as the solvent. One of the fundamental problems in making the required glycosidic linkage is the necessity for the cis configuration (a in this instance) at the anomeric center, linking the galacturonic acid residue of the donor with the C-2 hydroxyl of the rhamnosyl residue on the acceptor. Many of the glycosylation methods currently used are based on the idea of participating and non-participating groups during glycosylation (Please see Glycosylation Methods on page 21 for a detailed discussion on glycosylation methods and the use of -49- R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. participating and non-participating groups). The benzyl protecting group was chosen as a non-participating group in this instance to protect the hydroxyl at C-2 of the galacturonic acid residue. After multiple attempts, compound 18 was successfully synthesized in 38.8% yield. A prior attempt with the same donor and acceptor under almost the same conditions using approximately the same quantities of donor and acceptor resulted in only a 30% yield. The one difference was the amount of molecular sieves used; 1.24 equivalents of molecular sieves (added by weight to the donor) had been added to the reaction mixture providing the higher yield. Comparably, 2.20 equivalents of molecular sieves were used in the lower yielding reaction, with the mixture being stirred for almost two weeks. Several portions of promoter (AgOTf) had to be added on a continual basis in order to ensure the reaction proceeded. (Note: The reaction was not performed with fewer molecular sieves for the synthesis of 19.) Molecular sieves are often incorporated into reaction mixtures to aid in keeping the solvents free of water. Moisture in the solvent and atmosphere would have a negative impact on the donor used for the glycosylation. The imidate could undergo nucleophilic attack by the water, resulting in cleavage of the imidate and formation of the hemiacetal (13) (although never isolated, TLC of the reaction mixture of any given - and successful glycosylation indicated a spot corresponding in Rf to where the hemiacetal would be expected). The increased rate of reaction observed when a lower equivalence of molecular sieves was used could be due to less neutralization of the promoter. Removal of the 1-0acetate in the generation of the hemiacetal was accomplished by using powdered -50- R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. molecular sieves in methanol, which has been shown to generate small amounts of sodium methoxide (please see Synthesis o f the Disaccharide Donors, page 44). Generation of the sodium methoxide occurred as a result of the basic nature of the molecular sieves, so the presence of a large amount of molecular sieves may actually neutralize the promoter to some extent. Furthermore, it is unknown if the molecular sieves play any other role in the reaction, possibly slowing down product formation, such as through the interaction of the molecular sieves with the ester functional group on the galacturonic acid residue. The methyl and benzyl esters in the donor - and even the acceptor - are highly suspect as to the role they play in the success of the glycosylation. Reimer et al previously published the synthesis of a similar tetrasaccharide using galactose in both the acceptor and donor (see Figure 16) with the strategy of late-stage oxidation in mind. The published results indicate a yield of 36% with TMSOTf as the promoter37, but later attempts (unpublished) using AgOTf as the promoter provided yields as high as approximately 70%. Other than the p-O-methyl at the reducing end and the benzoyl to protect C- 6 of the galactose residue, the other protecting groups are identical to those used to generate tetrasaccharides 18 and 19. Naturally, these variations cannot be ruled out as to whether or not they play a role in the formation of the tetrasaccharide. / i 9 _ / I OBz 9 ,-—OBz T etrasac c h arid e BzIO OAII OC(=NH)CCI3 B z O ''A /T BzIO OAII Figure 16 Synthesis o f an analogous tetrasaccharide. Reimer et al. successfully synthesized a similar tetrasaccharide using the glycosyl donor and acceptor shown above in approximately 70% yield with AgOTf as the catalyst. -51 - R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. Magaud et al. have studied the differences in reactivities of galacturonic acid acceptors based on the configuration at the anomeric center i.e. identically protected acceptors were used; however, one was synthesized with the a-configuration at the anomeric carbon, and one with the p-configuration . All the acceptors studied involved the free hydroxyl at C-4, and included both the benzyl and methyl protected esters of galacturonic acid. Magaud’s group learned that the acceptor in the p-configuration was much more reactive in all cases. Magaud et al. proposed that the differences in reactivities between the a-glycoside acceptor and a P-glycoside acceptor could be due to the absence of an endo-anomeric effect. The endo- Figure 17 Minimization o f dipole interactions in the endo-anomeric effect; additive dipole interactions are minimized when the chlorine is axial (left) as opposed to being equatorial (right) anomeric effect generally explains why certain substituents at the anomeric center would orient in the axial position. This type of orientation would be considered unfavorable according to steric-based predictions, which would predict a large group to be oriented equatorially. Several factors have been attributed to this observed effect. One of the factors is the minimization of additive dipole-dipole interactions (see Figure 17)69. Solvent effects must also be taken into consideration (e.g. polar solvents vs. non-polar solvents). The other factor is a molecular orbital-based explanation, which says the lone pair on the ring oxygen oriented antiperiplaner to the non-bonding orbital (represented as a*) of the substituent on the anomeric position is stabilizing; the P-configuration has no such stabilizing interaction, -52- R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. and therefore the energy of the system is lower in the a-configuration, despite steric interactions (see Figure 18)69. 1,3 diaxial interactions Due to a lack of the endo-anomeric effect, the basicity of the pyranosyl oxygen atom in P-anomers would be greater than in a-anomers68. The increased basicity of the ring oxygen would result in stronger Figure 18 Molecular orbital explanation of the axial orientation o f substituents on the anomeric carbon. Despite steric interactions (left), the lone pairs o f the ring oxygen can delocalize into the antibonding orbital of chlorine due to the anti-periplaner arrangement o f the o* orbital. No such anti-periplaner arrangement exists when the chlorine is equatorial (right). hydrogen bonding between the axial C-4 group in galacturonic acid and the ring oxygen, thereby increasing the nucleophilicity of the C-4 hydroxyl in the P-anomer70. Magaud et al. have performed modeling studies which have indeed indicated the distance between the ring oxygen and the hydrogen on the C-4 hydroxyl is shorter in the P-anomer than in the a-anomer. The same type of argument on the relative reactivities of the a- and p-anomers of galacturonic acid as acceptors in glycosylation reactions could be applied to the disaccharide acceptor (10). The rhamnose portion of the acceptor is a similar system as the galacturonic acid acceptor in the research performed by Magaud et al. The rhamnose residue is in the a-configuration, with the C-2 hydroxyl on rhamnose axial. The same type of hydrogen bonding interaction could be occurring between the hydrogen on the C2 hydroxyl and the ring oxygen of rhamnose (see Figure 19). Magaud’s group attempted a glycosylation with a galacturonic acid disaccharide whose internal glycosidic linkage was also a, as it is for acceptor 10; the overall glycosylation reaction by Magaud et al. between the acceptor and a monosaccharide donor gave rise to a low yield of only 45% in -53 - R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. the formation of the desired trisaccharide. BzO Similar interactions may be giving rise to C-4 AIIO C-2 C -l C -l the low yields observed in the synthesis of 18 and 19. Zhao and Kong noted that the successful glycosylation between a donor Figure 19 Possible hydrogen bonding interactions affecting glycosylation Magaud proposed the interaction on the left was responsible for the low reactivity o f the galacturonic acceptor; similar interactions may be occurring in the glycosylation using acceptor 10 for the rhamnosyl residue. and acceptor cannot simply be determined by taking mto consideration the substituents on the residue undergoing glycosylation ; they hypothesized that, when a substituent on a secondary residue was altered, the acceptor they were using underwent a conformational change that suddenly allowed for successful glycosylation to take place, whereas the acceptor and donor had previously been shown to be unreactive. These results suggest that the problem of obtaining a higher yield of 18 and 19 could be partially or completely due to the conformation of the acceptor and donor as opposed to electronic considerations (electronic considerations in the sense of leaving group activation). As already mentioned, the molecular sieves and their possible basic nature could be a factor, depending on how they interact with the compound and in particular with the ester portion of galacturonic acid. The possible interactions that could lead to conformations, and subsequent steric hindrances, which could affect the glycosylation reaction may be numerous. The low yields for 18 and 19 are partially attributed to the purification of the target compounds. Although all impurities are clearly visible on TLC for both tetrasaccharide products and separation is evident, both tetrasaccharides needed to be purified multiple times in order to isolate the majority of the product. For both 18 and 19, -54- R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. the compounds corresponding to the TLC spots immediately above and below the target compound were isolated and investigated by NMR. The compounds corresponding to the upper spots in both cases appeared to be non-tetrasaccharide-related products, but the compound corresponding to the lower spot in both cases was determined to be disaccharide acceptor 10. Purification of 18 from the compound corresponding to the upper spot proved to be fairly successful; however, separation from the acceptor required the sample to be chromatographed multiple times. Synthesis of 18, which led to a yield of 38.8%, indicated a small amount of acceptor was present, according to TLC analysis. A larger equivalence of donor, or subsequent additions of donor, may lead to the complete consumption of the acceptor, thereby improving yields by decreasing the amount of purification required. The determination of whether 14 or 17 was the better donor is somewhat inconclusive. Preparation of the disaccharide benzyl ester imidate 17 required more controlled reaction times and conditions for both the acetolysis step (product 15) and the subsequent removal of the 1-O-acetate to give the hemiacetal (product 16). The glycosylations would need to be performed under more controlled conditions in order to determine if there is a difference in reactivity. Due to a lack of starting material as a result of multiple attempts to optimize reaction conditions, this was not performed. Various reaction conditions were attempted in order to improve yields for tetrasaccharide 18 or 19 (all conditions discussed here were attempted prior to the 38.8% yield that was eventually achieved for 18; all involved a larger equivalence of molecular sieves than what was eventually used in the most recent attempt). One of the most basic changes involved inverting the equivalents of donor to acceptor. 1.5 equivalents of 10 to -55- R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. 17 was used with 0.25 equivalents of AgOTf. Product formation resulted in a paltry yield of 19.1%. The reaction was stopped once TLC indicated all the donor had been consumed. One of the advantages of using the acceptor in excess would have been less of the donor would have been required. Synthesis of the imidate donor requires three steps and multiple purifications, and can thus be thought of as the “more valuable” of the two compounds. Synthesis of the acceptor requires comparatively fewer resources and much less time; however, having excess acceptor remain after the completion of the reaction would result in purification difficulties. One attempt was made to determine what would occur if copious amounts of AgOTf promoter were used. 4.0 Equivalents of promoter to donor were added to the reaction mixture of 17 (1.5 equivalents to 10) and 10. The reaction was performed in dichloromethane at room temperature, in the dark, under N 2 (g>and in the presence of molecular sieves. There was no evidence of product formation, but TLC indicated complete breakdown of the donor after one hour. An attempt to increase the yield was made by substituting AgOTf with the stronger Lewis acid, trimethylsilyl trifluoromethanesulfonate (TMSOTf). The reaction was performed under similar conditions, as with successful glycosylation attempts made using AgOTf. 0.25 equivalents of TMSOTf to the imidate donor 14 were added, and dichloromethane was used as the solvent. The reaction was performed at room temperature, and after 17.5 hours, one major product had formed; however, the product showed a higher Rf than previous successful glycosylations. NMR analysis confirmed the desired product had not formed. There was no evidence of any tetrasaccharide formation. -56- R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. The potential of using a coordinating R~° solvent in order to encourage formation of the BzIO thermodynamically more stable a-glycoside was HO— R 1 investigated71. Diethyl ether was used because the lone pairs on the oxygen would be suitable for coordinating with the positive charge on the activated donor, most likely blocking the (5-face and Figure 20 Use o f a coordinating solvent (Et20 ) resulted in no product formation enhancing a-reactivity (Figure 20). Donor 17 and acceptor 10 were dissolved in diethyl ether, kept under N2(g) and stirred in the presence of 3 A molecular sieves. Additional portions of AgOTf were added. After 22.5 hours, there was no sign of product formation or any other activity. TMSOTf was then substituted for AgOTf with the hope that this more active promoter would lead to product formation in diethyl ether. Conditions were similar to the previous attempt with AgOTf (i.e. placed under N 2 (g), in the dark, and in the presence of molecular sieves); the reaction was started in a pyr/N2({) (approximately -35°C to -40°C) bath to slow the rate of reaction72. 0.23 Equivalents of TMSOTf were initially used. No reaction occurred, so additional portions were added over time and the reaction mixture was allowed to warm to room temperature. The reaction only appeared to proceed after large quantities of TMSOTf and small portions of dichloromethane were added to the reaction mixture. The reaction was stopped after five days, when one major product appeared to have formed. Examination by NMR showed no sign of the galacturonic acid residue and thus the product was discarded. -57- R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. The formation of product upon the addition of dichloromethane indicated a mixture of dichloromethane and diethyl ether may lead to product formation. This may have been due to the diethyl ether coordinating too effectively with the donor intermediate. Steric hindrance around the reaction site of the donor may already have made access difficult, so the use of straight diethyl ether may have completely blocked access to the a-face of the donor intermediate by the acceptor 10 (see Figure 20). A mixture of dichloromethane and diethyl ether in a 1:2 ratio was then used. 1.4 equivalents of donor 14 to acceptor 10 were dissolved in the solvent, to which AgOTf was added (0.35 equivalents to the donor), followed by 0.85 equivalents (by weight to the donor) of 4 A powdered molecular sieves73. A small amount of trifluoromethane sulfonic acid (TfOH; 0.1 equivalents of a TfOH in CH2 CI2 (saturated) solution) was added. There was no sign of the donor after 15 minutes of reaction time. The reaction was worked up thereafter. Analysis by NMR indicated no sign of tetrasaccharide formation. A final attempt varying the glycosylation conditions using an imidate donor was carried out using /-BDMSOTf (fcr/-butyldimethylsilyl trifluoromethansulfonate) as the activator73. Initial conditions called for 1.10 equivalents of donor 17 to acceptor 10 and 0.77 equivalents (by weight to the donor) of 4 A powdered molecular sieves in a solution of dichloromethane and diethyl ether in a 1:2 ratio. The reaction was stirred under N2 (g> and in the dark for one hour and then placed in a pyr/N2 (f) bath, after which /-BDMSOTf was added. After twenty minutes no change was observed, so the mixture was allowed to warm to room temperature. After four days of stirring, no change was evident. AgOTf was then added over a period of several days, with still no sign of any product formation. At this point, the reaction was stopped. -58- R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. An alternate type of commonly used glycoside donor is the thioglycosides (see Thioglycosides on page 24 for a discussion on these donors). Several attempts were made to generate a suitable thioglycoside using />ara-thiocresol. The first attempt used BF3 -Et2 0 as the Lewis acid with dichloromethane as the solvent in an attempt to convert the methyl ester-containing 1-O-acetate (12) into a thioglycoside. The reaction proceeded quickly according to TLC and was worked up after ten minutes when no starting material was evident; however, after workup, there was clearly starting material left, and the reaction was redissolved. Two compounds corresponding to the major spots visible on TLC were isolated and ]H NMR showed a mixture of products for both of those spots. It cannot be conclusively stated that these conditions did not work, as the reaction would need to be repeated to verify the results; however, although only the compounds corresponding to the major spots were isolated, a multitude of other byproducts were evident. Whether or not the byproducts were carbohydrate in origin was not determined. BF3 -Et2 0 was substituted in subsequent attempts to form the thioglycoside with tin(IV) chloride (SnCL). The initial attempt showed a multitude of products formed on TLC, with results appearing similar to those using BF3 -Et2 0 . The first attempt was discarded and no compound isolated. A second attempt was made using the same conditions with SnCU as the Lewis acid. TLC taken of the reaction mixture showed various faint spots, with apparently one major product forming. After workup, however, TLC indicated once again a vast number of side products, with a fair amount of starting material still evident. Difficulties were also encountered during workup, making recovery of the products difficult. The recovered products were chromatographed and eventually the compound corresponding to the major spot visible on TLC which did not appear to be -59- R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. starting material was isolated. Extensive NMR analysis of the product was not performed, but simple 'H NMR indicated the thioglycoside was present (note: This cannot be conclusively stated due to a lack of further analysis). The isolated compound was then used in a glycosylation attempt using AgOTf and NIS (please see Thioglycosides on page 24 for a discussion on thioglycosides) dissolved in dichloromethane with molecular sieves. After approximately twenty-five minutes, TLC indicated a heavy concentration of carbohydrate material on the baseline of the TLC plate. The baseline material was not recovered or analyzed further, as the tetrasaccharide product would have been expected to show a much higher Rf when compared to previous successful glycosylations using the same TLC analysis as for the imidate donor. Further attempts were not made at this point, as the imidate technique appeared to be more promising. Once again, however, it can not be conclusively stated that formation of the tetrasaccharide through the use of a thioglycoside donor cannot be achieved. Three attempts were also made to generate a bromide donor in situ from the donor hemiacetal, followed by glycosylation with the disaccharide acceptor. (Note: The general strategy of this one-pot method is to use the appropriate conditions to generate an abromide. Upon successful generation of the donor the alcohol acceptor is added into the same reaction flask.) In the first attempt, the benzyl ester hemiacetal (16) was dissolved in dimethyl formamide (DMF) followed by the addition of triphenyl phosphine (P(Ph)3 ) and carbon tetrabromide (CB^)55. No reaction appeared to have taken place - at least none that was observable by TLC. Two further attempts were made to generate the bromide under slightly different conditions. In the first attempt, hemiacetal 16 was dissolved in dichloromethane, to which -60- R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. P(Ph) 3 and CBty were added. TLC indicated significant formation of a product over several days of reaction with an Rf noticeably higher than that of the starting material. The product was assumed to be the bromide. When donor formation had appeared to have gone mostly to completion, tetramethyl urea (TMU) was added, followed by the acceptor (10)54. Chromatography performed on the reaction mixture after workup only yielded recovery of the impure acceptor. A second attempt using the same conditions as noted above was performed. After what appeared to be the formation of the bromide from hemiacetal 16, TMU was added, followed by the acceptor alcohol (10). No reaction appeared to have occurred, so a small amount of AgOTf was added. A small amount of non-acceptor was recovered, but NMR analysis indicated the desired compound had not formed. At this point, the one-pot halide ion-catalyzed method of glycosylation was abandoned. Extensive NMR analyses were performed on both tetrasaccharide products 18 and 19. Hydrogen and carbon assignments were performed through a combined analysis of 'H NMR, 13C NMR, COSY and HMQC (inverse) spectra. All signals corresponding to the ring protons and carbons could be successfully accounted for, as well as all protecting groups expected to be present for both 18 and 19 (please see Appendix 1 for all peak assignments for 18 and 19). Four signals corresponding to C-l of each of the four monomers could be seen in the 95 ppm to 100 ppm region of the 13C spectra of both tetrasaccharides. These signals were within the expected region. HMQC (inverse) analysis was used to determine the coupling constants of C -l” to H-l ” on the donor galacturonic acid residue in order to verify whether or not the newly formed linkage was indeed in the desired a-configuration. -61 - R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. The expected one bond coupling constant for the a-linkage is approximately 170 Hz, whereas the coupling constant for the p-linkage would be about 160 Hz74. For both 18 and 19, the observed coupling constants fell within the expected range at 171.5 Hz, indicating the a-linkage had been successfully formed. Investigation of the COSY, *H NMR, and 13C NMR spectra confirmed all the expected protecting groups were present. The signals for both methyl ester groups were visible for 18 at 52.7 ppm and 52.1 ppm in the 13C spectrum. For 19, only one peak corresponding to the methyl ester was evident at 52.7 ppm. The benzyl ester peak occurred at 67.2 ppm. Please see the Experimental section for detailed information on the assignment of peaks for 18 and 19 as well as Appendix 1 for the respective compounds. Deprotection o f RG-I The deprotection of RG-I was planned so that substitution at C-4 of the rhamnosyl residue would be possible after the first deprotection step. As mentioned in the introduction to RG-I, the RG-I polysaccharide is most highly substituted at C-4, and thus it was of interest to orthogonally protect C-4, allowing for further additions to take place. The allyl group was selected for just such a purpose. The allyl group was removed according to conditions published by Takeo et al. 75 The general procedure is to first isomerize the allyl to the vinyl ether using Wilkinson’s catalyst ([(C6H5)3P]3RhCl), followed by cleavage using a combination of mercury(II) oxide and mercury(II) chloride. Approximately 53-54% yield was obtained. The workup -62- R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. AcO AcO BzIO BzlO- BzlO BzO' BzIO BzO' OAII OH BzIO- BzIO BzIO OMe BzO' BzIO OMe BzO' OAII OH HO HO BzlO HO- HO HO BzIO HO' OH OH HO HO' HO OMe BzIO HO BzIO OMe Scheme 5 Deprotection o f the tetrasaccharide. (a) i. [(C6H5)3P]3RhCl, DABCO in EtOH/Toluene/H20 ; ii. HgO, HgCl2 in acetone/ H20 ; (b)l.O M NaOMe (pH 11-12) in MeOH; (c) H2 over Pd(OAc)2 (40 psi). procedures for the isomerization and the cleavage are quite extensive, and the degree of possible loss appeared to be fairly high. Furthermore, TLC during the isomerization appeared to proceed fairly cleanly, but some byproducts, including some starting material, appeared to be present upon workup. Several of the spots, other than the one corresponding to the starting material, could be due to incomplete isomerization of the allyl on either of the two rhamnosyl residues. A number of additional spots were also visible on the TLC developed after cleavage had occurred; these were attributed to the compounds in which incomplete cleavage of the allyl groups had occurred and the compounds that resulted due to the incomplete isomerization to the vinyl ether in the first step. One of the ways of improving the yield during this step would be to look into alternatives for cleaving the allyl group. An alternative method would be to use palladium-tetrakis triphenylphosphine (Pd(Ph3 P)4 ), which was used to remove the allyl group from the anomeric position by Kim et al.16 A difficulty may arise in that the allyl -63- R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. was removed from the anomeric position, which is generally much more reactive than other hydroxyls (e.g. C-4 of rhamnose), and therefore may not be as effective in this scenario. The other option may be to exchange the allyl for another orthogonal group at C-4 of rhamnose; this, however, would have been no trivial undertaking, as the change would have had to have been made at the monosaccharide level during the synthesis of the rhamnosyl donor in the preparation of the disaccharide donor. The para-methoxy benzyl group may just be such a candidate due to its orthogonality with other groups used in the synthesis of RG-I, and the supposed ease with which it can be removed (see page 20). The initial strategy was to fully deprotect the RG-I tetrasaccharide, followed by purification. No chromatography was performed once the allyl group had been removed and, although a compound of interest, the entire crude mixture was taken on to the next step for removal of the benzoates and the acetate group (this was the case for both 18 and 19). Removal of the benzoates and acetate was performed using sodium methoxide; the removal occurred without incidence, and a significant Rf change was observed. An attempt was made, without further verification of product, to take the entire crude mixture onto the next step, which involved hydrogenation to remove all the benzyl groups using palladium acetate as the heterogeneous catalyst77. Several attempts were made using both 18 and 19, but no change in R f was observed. Eventually, the lack of reaction was attributed to the palladium acetate catalyst being poisoned by traces of mercury residue left over from cleavage of the allyl78. A subsequent attempt at deprotection (compound 18 was used) involved removing the allyl, followed by removal of the benzoates and the acetate, and then purifying the residue by chromatography. Purification by chromatography was left to after removal of the benzoates and acetate -64- R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. because the change in Rf was dramatic enough to allow for fairly simple separation of the products from the impurities. The compound was not fully characterized at this point, but investigation by ’H NMR showed no sign of the allyl, benzoate, or acetate group signals. The signals for the remaining ring protons, methyl groups and benzyl groups could all be accounted for. The 'H NMR spectrum appeared to be free of other impurities. Removal of the benzyl groups present on the partially deprotected tetrasaccharide was successfully accomplished by placing the purified compound under ffyg) (40 psi) in the presence of a palladium acetate catalyst, giving 20. The reaction was left for one hour, after which TLC indicated only one spot present, with no sign of any starting material. The reaction appeared to have proceeded quantitatively. 'if NMR, 13C NMR, HMQC (inverse) and COSY analyses were performed on compound 20. ’H NMR confirmed the absence of all aromatic signals; analysis of the 1H NMR and 13C NMR spectra was also performed to confirm the presence of the methyl ester groups on the two galacturonic acid residues, as well as the presence of the methyl group on the aglycone. COSY analysis confirmed the presence of signals corresponding to the ring protons for both galacturonic acid residues, as well as for both rhamnosyl residues. HMQC (inverse) analysis confirmed all glycosidic linkages to be a; the one bond coupling constants for all C'-H 1 correlations of the four sugar residues were between 171.0-171.5 Hz, within the expected range. Although no analytical data has been included here for the deprotection of tetrasaccharide 19, tetrasaccharide 19 was used in attempts preceding those made with tetrasaccharide 18. Prior to attempting the deprotection of 19, a trial deprotection of disaccharide 9 (the benzyl ester) was attempted in order to determine the feasibility of the -65- R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. deprotection for the two tetrasaccharides. Prior to hydrogenation, only one spot was visible on TLC. After hydrogenation, two spots of varying intensity were evident, neither of which corresponded to the starting material. The sample was chromatographed and the two spots characterized. NMR showed that neither sample contained any benzyl ethers; however, one sample showed the presence of the benzyl protecting the carboxylic acid, whereas the other showed that the benzyl had been replaced by a methyl group. The methyl product was the minor product. These results are important for future work involving the deprotection of tetrasaccharide 19. At least two products of value would be expected after the hydrogenation step - one containing two methyl esters, and one containing a benzyl ester and a methyl ester. Tetrasaccharide 20 was not deprotected further to give the free carboxylic acids. The deprotection performed on disaccharide 9 resulted in difficulty purifying the carboxylic acid, as it was highly polar. Hydrolysis of deprotected disaccharide 9 using H+ resin and water did not prove to be fruitful; however, addition of a 1.0 M solution of HC1 eventually gave the deprotected carboxylic acid. More attempts would be necessary using perfectly pure starting material. The most desirable outcome would be for hydrolysis to occur using H+ resin, as the resin could simply be removed by filtration without having to remove any impurities or salts. - 66 - R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. Chapter Five Experimental General methods.— 'H NMR (300.13 MHz) and 13C NMR (75.03 MHz) spectra were compiled with a Bruker AMX 300 spectrometer. Chemical shifts are relative to external tetramethyl silane (CDCI3 ). Optical rotations were measured on an Autopol III polarimeter. Melting points were performed on an Electrothermal Digital Melting Point Apparatus (IA9100). All reactions were monitored by thin-layer chromatography (TLC) using silica gel on an aluminum support (Silicycle 250pm, F-254 indicator), with detection by charring with 5% sulfuric acid (v/v) in EtOH. TLCs were performed with solvents of appropriately adjusted polarity consisting of A, 7:2:1 EtOAc:MeOH:H2 0 ; B, 1:1 Hex.EtOAc; C, 4:1 Hex:EtOAc; D, 3:2 Hex:EtOAc; E, 2:1 Hex:EtOAc; F, 3:1 Hex:EtOAc; G, 2:1:1 Hex:EtOAc:CHCl3; H, 3:1:1 Hex:EtOAc:CHCl3; J, 2:1:1 Hex:EtOAc:CH2 Cl2; K, 4:1:1 Hex:EtOAc:CH2 Cl2; L, 8:1 Tol:Acetone; M, 4:3:3 Hex:EtOAc:CHCl3; N, 7:1 Tol:Acetone; P, 6:6:1 Hex:EtOAc:MeOH; Q, 12:2:1 EtOAc:Hex:MeOH; R, 16:2:1 EtOAc:Hex:MeOH; S, 10:3:1 EtOAc:MeOH:AcOH. All chromatography was performed using low pressure liquid chromatography columns packed with silica gel (Silicycle 230-400 mesh (40-63 microns) 60 A). Methyl (methyl-a-D-galactopyranosid)uronate (1) Compound 1 was prepared from commercially available D-galaturonic acid dissolved in MeOH. Amberlyst® 15 H+ resin beads were added to the mixture, which was then stirred under reflux for approximately 60 hours59. The solution was allowed to cool to room temperature, the resin was removed by filtration and the filtrate was evaporated to dryness to afford a brown syrup. The syrup was taken up in acetone, to which EtOAc -67- R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. was added until a precipitate formed. The brownish precipitate was collected by vacuum filtration and the mother liquor evaporated to dryness; the resulting syrup was taken up in acetone and the procedure was repeated until no more precipitate formed; subsequent precipitates were purer than the initial batch. All the precipitates were combined and carried on to the next step without further purification. TLC (solvent A) indicated a trace of impurity at a slightly higher Rf than the desired product. Recrystallization for the purposes of melting point and optical rotation was carried out using a mixture of EtOAc and acetone. m.p. 149.2-150.0°C, lit. m.p. 7 9 147°C; [a]J2 +166.4° (c 0.125, MeOH). Methyl (methyl 3,4-0-isopropylidene-a-D-galactopyranosid)uronate (2) The typical procedure for preparing 2 involved adding acetone (insoluble, approx. 4.25 mL/mmol) to a slightly impure mixture of 1 at rt; dimethoxy propane (10 eq.) and 10-camphor sulfonic acid ( 8 eq.) were then added80. The mixture was stirred for 30 min. (TLC solvent A), after which the mixture was neutralized with Et3N and evaporated to dryness. The resulting solid was taken up in CH2 CI2 and washed with H2 O (extracted 3x with CH2 CI2 ). The CH2 CI2 fraction was dried over Na2 S0 4, filtered, and evaporated to dryness. The resulting solid was recrystallized from solvent B to yield pure crystals of 2. m.p. 113.8-114.2°C, lit. m.p. 81 113-114°C; [a] 2 2 +83.4° (c 0.302, CHCI3 ). NMR (CDCI3 ): 5 4.92 (d, 1 H, A ,2 3.9 Hz, H-l), 5 4.67 (d, 1 H, J5A 2.5 Hz, H-5), 5 4.56 (dd, 1 H, J 4 ,5 2.5 Hz, H-4), 5 4.37 (dd, 1 H, J 4;3 6.4 Hz, H-3), 8 3.98 (dd, 1 H, J 2 ,3 5.7 Hz, H-2), 5 3.85 (s, 3 H, C 0 2 CH3), 5 3.53 (s, 3 H, OCH3), 8 1.50, 1.36 (2 s, 6 H, (CH3)2C). - 68 - R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. Methyl (methyl 2-0-benzyl-3,4-0-isopropylidene-a-D-galactopyranosid)uronate (3) Compound 2 (1.717 g, 6.547 mmol) was dissolved in freshly distilled CH2 CI2 , to which benzyl bromide (BzlBr; 5.6 mL) and silver(I) oxide (Ag2 0 ; 2.06 g, 8.89 mmol) were added (TLC solvent B)61. The reaction was stirred at rt in the dark under N 2 (g>and left overnight (-19 hours). Additional portions of Ag2 < 3 (2.06 g, 8.89 mmol) and BzlBr (5.6 mL) were then added. The reaction was stopped after approximately 43 h. The solids in the reaction mixture were removed by filtration and the filtrate was evaporated to dryness. The resulting liquid was put on a vacuum column (eluting solvent Q ; pure fractions containing 3 were collected and evaporated to dryness to afford a syrup (2.01 g, 87.1%); while stored in the freezer (-20°C), the product eventually crystallized, and melting point data was obtained: m.p. 60.2-61.6°C, lit. m.p. 61 80.5-81.5°C; [a] ^ 2 +57.7° (c 0.648, CHCI3 ), lit.61 [a] 2 7 +70.2°. 'H NMR (300.13 MHz, CDC13): 6 7.42-7.24 (m, 5 H, aromatic), 5 4.79 (d, 1 H, J ia 3.5 Hz, H-l), 5 4.79,4.74 (2 d, 2H, OCftPh), 5 4.61 (d, 1 H, J 5 ,4 2.8 Hz, H-5), 5 4.50 (dd, 1 H, J 4 ,5 2.8 Hz, H-4), 5 4.40 (dd, 1 H, JXA 5.6 Hz, H-3), 8 3.83 (s, 3 H, C 0 2 CH3), 8 3.60 (dd, 1 H, J 2 ,3 7.7 Hz, H-2), 8 3.42 (s, 3 H, OCH3), 8 1.38, I.35 (2 s, 6 H, (CHshC). Methyl (methyl 2-0-benzyl-a-D-galactopyranosid)uronate (4) A sample of compound 3 (2.01 g, 5.70 mmol) was dissolved in acetic acid (25.5 mL), to which 90% aq. trifluoroacetic acid (TFA, 4.5 mL) was added52 (TLC solvent B). The mixture was heated in an oil bath and stirred at 70°C for 30 min. The reaction was stopped and 45 mL of toluene was added; the mixture was then evaporated to dryness. Traces of acetic acid and TFA were removed by evaporation of repeating additions of toluene-hexanes-ethanol (5:1:1, v/v, 4 x 42 mL). The resulting solid was then -69- R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. recrystallized from Hex:EtOAc, yielding 4 (1.550g, 87.4%): m.p. 130.2-131.4°C, lit. m.p.61 133-134°C; [a] 2 2 +78.6° (c 0.585, CHC13), lit. 61 [a] 2 7 +73.3°. NMR (300.13 MHz, CDC13): 5 7.42-7.29 (m, 5 H, aromatic), 8 4.79 (d, 1 H, Jh2 3.4 Hz, H-l), 8 4.70, 4.68 (2 d, 2H, OCf/2 Ph), 8 4.42 (d, 1 H, J 5 ,4 1.6 Hz, H-5), 8 4.36 (dd, 1 H, J4,5 1.6 Hz, H4), 8 4.06 (dd, 1 H, J2,3 9.8 Hz, H-2), 8 3.83 (s, 3 H, C 0 2 CH3), 8 3.77 (dd, 1 H, J3A 3.4 Hz, H-3), 8 3.38 (s, 3 H, OCH3). Benzyl (methyl 2,3-di-0-benzyl-a-D-galactopyranosid)uronate (5) and Methyl (methyl 2,3-di-0-benzyl-a-D-galactopyranosid)uronate (6) Compound 4 (6.276 g, 20.08 mmol), di-n-butyltin oxide (6.000g, 24.10 mmol) and toluene (240 mL) were heated at reflux for 5 h 40 m; water was removed by 4 A molecular sieves placed in a Soxhlet extractor . The reaction mixture was then heated in a 70°C oil bath and tetra-n-butylammonium bromide (7.8 g, 24 mmol) was added, followed by BzlBr (5.9 mL). The reaction was stopped after 6.5 h and allowed to cool to rt; MeOH was added and the mixture was evaporated to dryness (TLC solvent B). The resulting syrup was chromatographed (eluting solvent D) to give 5 (2.499 g, 26.0%) and 6 (5.376 g, 66.5%). Anal, for 5: m.p. 101.2-102.0°C; [a] 7 2 +24.4° (c 0.583, CHCI3 ). lU NMR (300.13 MHz, CDC13): 8 7.45-7.26 (m, 15 H, aromatic), 8 5.27 (m, 2 H, C 0 2 Ctf2 Ph), 8 4.83,4.66 (2 d, 2 H, OC//2 Ph), 8 4.82,4.72 (2 d, 2 H, OC//2 Ph), 8 4.77 (d, 1 H, J 1;2 3.2 Hz, H-l), 8 4.41 (m, 1 H, H-5), 8 4.40-4.36 (m, 1 H, H-4), 8 3.95 (dd, 1 H, J 3 ,4 3.0 Hz, H-3), 8 3.90 (dd, 1 H, J2;3 9.8 Hz, H-2), 8 3.83 (s, 3 H, CO2CH3), 8 3.38 (s, 3 H, OCH3). Anal, calcd for C2 8 H3 o0 7: C, 70.28; H, 6.32. Found: C, 69.98, H 6.50. Anal, for 6 : [a] 2 2 +33.3° (c 0.624, CHC13), lit. 5 6 [a] 2 5 +33.3°. 'H NMR (300.13 MHz, CDC13): 8 7.41-7.26 (m, 10 H, aromatic), 8 4.83, 4.66 (2 d, 2 H, OCtf2 Ph), 8 4.81, 4.74 (2 d, 2 H, -70- R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. OC//2 Ph) , 8 4.76 (d, 1 H, J2,i 3.3 Hz, H-l), 6 4.39 (m, 2 H, H-4, H-5), 6 3.95 (dd, 1 H, J3A 3.1 Hz, H-3), 5 3.89 (dd, 1 H, J2,3 9.8 Hz, H-2), 6 3.82 (s, 3 H, C02CH3), 5 3.41 (s, 3 H, OCH3). Anal, calcd for C2 2 H2 6 0 7: C, 65.66; H, 6.51. Found: C, 65.24, H 6.40. Methyl (2-0-acetyl-4-0-allyl-3-0-benzoyl-a-L-rhamnopyranosyl)-(l -~4)-(methyl 2,3-di0-benzyl-a-D-galactopyranosid)uronate (8) A sample of compound 6 (0.356g, 0.885 mmol) was dissolved in freshly distilled CH2 C12 (15.8 mL); 3 A molecular sieves were added, and the solution was stirred under N2(g) in a pyr/N2({) bath for 10 min. (The pyr/N2({) bath (approx. -35°C to -40°C) was maintained for the duration of the reaction.) The donor57 (7) (0.489g, 1.24 mmol, 1.4 eq. to acceptor) was then added and the mixture was stirred for an additional 1 0 min., followed by addition of NIS (0.298g, 1.32 mmol, 1.06 eq. to donor) and stirring for another 10 min. 49 pL of TfOH (0.554 mmol, 0.45 eq. to donor) was added (TLC solvent E). The reaction mixture was neutralized after 15 min. with Et3N, and washed sequentially with satd NaHC0 3 and 10% (w/v) Na2 S2 0 3. The organic layer was dried over Na2 SC>4 , filtered, evaporated to dryness, and purified by column chromatography (eluting solvent E), giving 8 (0.504g, 77.5%). [a] 7 2 +3.3° (c 0.602, CHC13). 'H NMR (300.13 MHz, CDCI3): 8 8.05-7.95 (m, 2 H, aromatic), 5 7.61-7.24 (m, 13 H, aromatic), 8 5.78 (m, 1 H, OCH2 Ctf=CH2), 8 5.61 (dd, 1 H, d, 1 H, J 2 ,3 3.1 Hz, H-2 Rha), 8 5.49 (dd, 1 H, J 3>4 9.4 Hz, H-3 Rha), 8 5.20-5.04 (m, 2 H, OCH2 CH=C//2), 8 5.14 (d, J ia 2.2 Hz, H-l Rha), 8 4.90, 4.70 (2 d, 2 H, OCtf2 Ph), 8 4.84, 4.72 (2 d, 2 H, OC//2 Ph), 8 4.71 (d, 1 H, Jh2 3.5 Hz, H-l GalA), 8 4.44 (m, 1 H, H-4 GalA), 8 4.39 (m, 1-H, H-5 GalA), 8 4.12 (m, 2 H, 0C // 2 CH=CH2), 8 4.09 (dd, 1 H, . / 2;3 10.0 Hz, H-2 GalA), 8 3.95 (dd, 1 H, J 3 ,4 2.7 Hz, H-3 GalA), 8 3.83 (s, 3 H, C 0 2 CH3\ 8 3.79 (dd, 1 H, J5fi 6.2 Hz, H-5 Rha), 8 -71 - R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. 3.48 (dd, 1 H, «/4j5 9.5 Hz, H-4 Rha), 5 3.39 (s, 3 H, OCH3), 5 2.00 (s, 3 H, OCOCH3), 5 1.35 (d, 3 H, H- 6 Rha); 13C NMR (75.03 MHz, CDC13): 5 169.5,169.1 (O O ), 5 165.2 (C=O, benzoate), 5 138.4,133.1,130.4,129.7,128.8,128.6, 128.5,128.1,127.8,127.7 (C aromatic), 5 134.9 (OCH2 CH=CH2), 5 116.8 (OCH2 CH=CH2), 5 99.6 (C-l GalA, Jc,u 170.8 Hz), 8 99.4 (C-l Rha, Jc,n 173.7 Hz), 5 78.5 (C-4 Rha), 8 77.5 (C-3 GalA), 8 76.9 (C-4 GalA), 8 75.8 (C-2 GalA), 8 74.5 (CH2 Ph), 8 73.6 (OCH2 CH=CH2, CH2 Ph), 8 72.0 (C-3 Rha), 8 70.5 (C-2 Rha), 8 70.3 (C-5, GalA), 8 68.4 (C-5 Rha), 8 56.3 (OCH3), 6 52.7 (C0 2 CH3), 8 21.0 (OCOCH3), 8 18.2 (C- 6 Rha); mass spectrum (ESI) m/z 757.2839 (M+Na). Benzyl (2-0-acetyl-4-0-allyl-3-0-benzoyl-a-L-rhamnopyranosyl)-(l —4)-(melhyl 2,3-di0-benzyl-a~D-galactopyranosid)uronate (9) A sample of compound 5 (0.432g, 0.903 mmol) was dissolved in freshly distilled CH2 C12 (17.0 mL); 3 A molecular sieves were added, and the solution was stirred under N2(g) in a pyr/N2(Qbath for 10 min. (The pyr/N2(f) bath (approx. -35°C to -40°C) was maintained for the duration of the reaction.) The donor5 7 (7) (0.499g, 1.26 mmol, 1.4 eq. to acceptor) was then added and the mixture was stirred for an additional 1 0 min., followed by the addition of NIS (0.304g, 1.35 mmol, 1.07 eq. to donor) and stirring for another 10 min. 50 pL of TfOH (0.565 mmol, 0.45 eq. to donor) was added (TLC solvent E). The reaction mixture turned a deep purple almost immediately and was neutralized thereafter with Et3N, washed sequentially with satd NaHC03, and 10% (w/v) Na2 S2 0 3. The organic layer was dried over Na2 S04, filtered, evaporated to dryness, and purified by column chromatography (eluting solvent F), giving 9 (0.586g, 80.1%). [aj ^2 +0.4° (c 0.543, CHC13). lH NMR (300.13 MHz, CDC13): 8 8.06-7.97 (m, 2 H, aromatic), 8 7.63- -72- R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. 7.19 (m, 18 H, aromatic), 5 5.80 (m, 1 H, OCH2 C//=CH2), 5 5.65 (dd, 1 H, J 2 ,3 3.3 Hz, H2 Rha), 5 5.54 (dd, 1 H, J 3 ,4 9.5 Hz, H-3 Rha), 8 5.32, 5.14 (2 d, 2 H, C 0 2 Ctf2 Ph), 8 5.13 (m, 2 H, OCH2 CH=€f/2), 8 5.10 (d, 1 H, J L2 2.1 Hz, H-l Rha), 8 4.91,4.74 (2 d, 2 H, OCtf2 Ph), 8 4.84, 4.68 (2 d, 2 H, OC/72 Ph), 8 4.72 (d, 1 H, J u 3.5 Hz, H-l GalA), 8 4.45-4.39 (m, 2 H, H-4, H-5, GalA), 8 4.23-4.06 (m, 2 H, OCH2CU=CU2), 8 4.23-4.06 (m, 1 H, H-2, GalA), 8 3.98-3.88 (m, 1 H, J 3 ,4 2.8 Hz, H-3 GalA), 8 3.98-3.88 (m, 1 H, J6,s 6.2 Hz, H-5 Rha), 8 3.50 (dd, 1 H, J 4>5 9.5 Hz, H-4 Rha), 8 3.38 (s, 3 H, OCH3), 8 2.01 (s, 3 H, OCOCH3), 8 1.35 (d, 3 H, H- 6 Rha); 13C NMR (75.03 MHz, CDC13): 8 169.5,168.3 (C=0), 8 165.2 (C=0, benzoate), 8 138.5, 135.2, 133.1, 130.5,129.7, 128.8,128.7, 128.6, 128.5,128.1,127.9,127.7 (C aromatic), 8 134.9 (OCH2 CH=CH2), 8 116.8 (OCH2 CH=CH2), 8 99.7 (C-l Rha, Jc,u 173.3 Hz), 8 99.6 (C-l GalA, Jc,u 172.5 Hz), 8 78.6 (C-4 Rha), 8 77.7 (C-4 GalA), 8 77.5 (C-3 GalA), 8 75.9 (C-2 GalA), 8 74.6, 73.7 (CH2 Ph), 8 73.6 (OCH2 CH=CH2), 8 72.1 (C-3 Rha), 8 70.6 (C-2 Rha), 8 70.2 (C-5, GalA), 8 6 8 . 6 (C-5 Rha), 8 67.5 (C0 2 CH2 Ph), 8 56.3 (OCH3), 8 21.0 (OCOCH3), 8 18.4 (C- 6 Rha); mass spectrum (ESI) m/z 833.3148 (M+Na). Methyl (4-0-allyl-3-0-benzoyl-a-L-rhamnopyranosyl)-(l -^4)-(methyl 2,3-di-O-benzyl-aD-galactopyranosid)uronate (10) A sample of compound 8 (0.3 82g, 0.520 mmol) was dissolved in freshly distilled CH2 C12 (8.0 mL), to which 3% methanolic HC1 (8.0 mL + 1.5 mL after approx. 49 h) was added64. The reaction mixture was allowed to sit at room temperature for approx. four days, after which the reaction mixture was diluted with CH2 C12, washed with satd NaHC03, filtered, dried over Na2 S04, and evaporated to dryness (TLC solvent B). The resulting syrup was chromatographed (eluting solvent E) to give pure 10 (0.287g, 79.7%). -73 - R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. [a]“ +13° (c 0.519, CHCb). *HNMR (300.13 MHz, CDC13): 5 8.13-8.06 (m, 2 H, aromatic), 8 7.64-7.21 (m, 13 H, aromatic), 8 5.80 (m, 1 H, OCH2 C//=CH2), 8 5.49 (dd, 1 H, J 3 ,4 8.2 Hz, H-3 Rha), 8 5.13 (m, 2 H, OCH2 CH=C7/2), 8 5.13 (d, 1 H, J h2 2.8 Hz, H-l Rha), 8 4.86, 4.69 (2 d, 2 H, OC/fcFh), 8 4.84, 4.71 (2 d, 2 H, OCi/2 Ph), 8 4.75 (d, 1 H, «/u 3.4 Hz, H-l GalA), 8 4.48-4.44 (m, 1 H, H-4, GalA), 8 4.43-4.39 (m, 1 H, H-5, GalA), 8 4.32 (dd, 1 H, d, 1 H, J 3 ,2 3.1 Hz, H-2 Rha), 8 4.24-4.05 (m, 2 H, OCH2 CH=CH2), 8 4.05 (dd, 1 H, J 2 ,3 10.1 Hz, H-2 GalA), 8 3.97 (dd, 1 H, J 3 ,4 2.7 Hz, H-3 GalA), 8 3.84 (s, 3 H, CO2CH3), 8 3.77 (dd, 1 H, J5>66.2 Hz, H-5 Rha), 8 3.51 (dd, 1 H, . / 4;5 9.3 Hz, H-4 Rha), 8 3.40 (s, 3 H, OCH3), 8 1.33 (d, 3 H, H- 6 Rha); 13C NMR (75.03 MHz, CDC13): 8 168.9 (C0 2 CH3), 8 165.5 (C=0, benzoate), 8 138.4,138.2,133.3,130.4,129.9,128.7, 128.1,128.0,127.9 (C aromatic), 8 134.8 (OCH2 CH=CH2), 8 117.0 (OCH2 CH=CH2), 8 102.0 (C-l Rha, Jc,h 170.6 Hz), 8 99.4 (C-l GalA, Jc,h 170.6 Hz), 8 78.9 (C-4 Rha), 8 77.7 (C-3 GalA), 8 77.1 (C-4 GalA), 8 75.9 (C-2 GalA), 8 74.3 (CH2 Ph), 8 74.1 (C-3 Rha), 8 73.9 (CH2 Ph), 8 73.1 (OCH2 CH=CH2), 8 70.4 (C-5, GalA), 8 69.9 (C-2 Rha), 8 68.4 (C-5 Rha), 8 56.3 (OCH3), 8 52.7 (C0 2 CH3), 8 18.5 (C- 6 Rha). Anal, calcd for C38 H4 4 0 i2: C, 65.88; H, 6.40. Found: C, 65.52, H 6.32. Benzyl (4-0-allyl-3-0-benzoyl-a-L-rhamnopyranosyl)-(l -^4)-(methyl 2,3-di-O-benzyl-aD-galactopyranosid)uronate (11) A solution of 9 (0.500g, 0.617 mmol) in CH2 C12 (11.7 mL) was treated with 3% methanolic HC1 (11.3 mL)64. The reaction was stirred under N2(g) at rt for approx. 77.5 hours, after which time the mixture was worked up by diluting with CH2 C12, washing with satd NaHC03, drying over Na2 S04, filtering, and evaporating to dryness (TLC solvent D). The resulting syrup was chromatographed (eluting solvent D); two samples -74- R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. were collected, one of which consisted of the compounds corresponding to the two upper spots that appeared on TLC, and a second sample that consisted of the compounds corresponding to the two lower spots. Both of these samples were then individually chromatographed again (eluting solvent E) and yielded the following: (9, 0.027g, 5.3%; 11, 0.174g, 36.6%; 8 , 0.033g, 7.3%; 10,0.189g, 44.2%). [a] 2 2 +6.9° (c 0.404, CHC13). fH NMR data for 11 (300.13 MHz, CDC13): 5 8.14-8.07 (m, 2 H, aromatic), 5 7.65-7.22 (m, 18 H, aromatic), 5 5.81 (m, 1 H, OCH2 C//-CH2), 8 5.44 (dd, 1 H, JiA 8.2 Hz, H-3 Rha), 8 5.30, 5.15 (2 d, 2 H, C 0 2 C// 2 Ph), 8 5.15 (m, 2 H, OCH2 CH=CH2), 8 5.13 (d, 1 H ./u 2.9 Hz, H-l Rha), 8 4.87,4.70 (2 d, 2 H, OCtf2 Ph), 8 4.85,4.68 (2 d, 2 H, OCH2?h), 8 4.75 (d, 1 H, Ji ,2 3.4 Hz, H-l GalA), 8 4.45-4.40 (m, 2 H, H-4, H-5, GalA), 8 4.38 (m, 1 H, d, 1 H, Ji ,2 3.0 Hz, H- 2 Rha), 8 4.28-4.03 (m, 2 H, OCH2CU=CU2), 8 4.07 (dd, 1 H, J2A 10.1 Hz, H-2 GalA), 8 3.95 (dd, 1 H, J3A 2.7 Hz, H-3 GalA), 8 3.90 (dd, 1 H, J5,6 6.2 Hz, H-5 Rha), 8 3.54 (dd, 1 H, J5A 9.3 Hz, H-4 Rha), 8 3.39 (s, 3 H, OCH3), 8 1.34 (d, 3 H, H- 6 Rha); 13C NMR (75.03 MHz, CDC13): 8 168.2 (C0 2 CH3), 8 165.5 (C=0, benzoate), 8 138.4,138.3,135.3, 133.3,129.9,129.0 128.9,128.8,128.7,128.1,128.0, 128.0,127.8,127.2 (C aromatic), 8 134.8 (OCH2 CH=CH2), 8 117.0 (OCH2 CH=CH2), 8 102.5 (C-l Rha, Jc,h 170.4 Hz), 8 99.4 (C-l GalA, Jc,u 170.9 Hz), 8 79.5 (C-4 Rha), 8 78.2 (C-4 GalA), 8 77.5 (C-3 GalA), 8 76.0 (C-2 GalA), 8 74.0 (CH2 Ph), 8 74.2 (C-3 Rha), 8 74.0 (CH2 Ph), 8 73.2 (OCH2 CH=CH2), 8 70.3 (C-5, GalA), 8 69.8 (C-2 Rha), 8 6 8 .6 (C-5 Rha), 8 67.6 (C0 2 CH2 Ph), 8 56.3 (OCH3), 8 18.6 (C- 6 Rha); mass spectrum (ESI) m/z 791.3041 (M+Na). -75- R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. Methyl (2-0-acetyl-4-0-allyl-3-0-benzoyl-a-L-rhamnopyranosyl)-(l -*4)-(l-0-acetyl2,3-di-0-benzyl-a/fi-D-galactopyranosid)uronate (12) A sample of compound 8 (1.860g, 2.514 mmol) was dissolved in acetic acid (AcOH; 41.7 mL) and acetic anhydride (AC2 O; 41.7 mL)51. The reaction mixture was then cooled in an ice bath for 10 min, after which H2 SO4 (417 pL) was slowly added. The reaction was followed by TLC (TLC solvent G) and showed completion after approx. 2 h 15 m. The reaction mixture was allowed to warm to rt during the progress of the reaction. Sodium acetate trihydrate (NaOAc-SELO; 4.38g) was then added and the mixture was stirred until all Na0 Ac-3 H2 0 had dissolved. The mixture was then evaporated to dryness, taken up in EtOAc, washed with H2 O (extracted 4x with EtOAc), dried over Na2 S0 4 , filtered, evaporated to dryness, and finally co-evaporated three times with EtOH. The resulting syrup was chromatographed (eluting solvent B) to obtain pure 12 (1.555g, 81.1%). 'H NMR data for 12 (300.13 MHz, CDCI3): 5 2.12 (s, 3 H, GalA -OCOCH3, major anomer), 5 2.01 (s, 3 H, Rha -OCOCH3, major anomer); 13C NMR (75.03 MHz, CDCfi, major anomer): 5 169.7,169.5,169.1,168.8 (C O , Rha OC(0)CH3, GalA OC(0)CH 3 and C 0 2 CH3), 5 165.2 (C O , benzoate), 5 138.2-127.8 (C aromatic), 6 134.8 (OCH2 C H O H 2 ), 5 116.9 (OCH2 CH=CH2), 5 99.3 (C-l Rha, Jc,h 174.8 Hz), 6 90.6 (C-l GalA, JCM 177.9 Hz), 6 78.5 (C-4 Rha), 5 77.2 (C-3 GalA), 8 75.8 (C-4 GalA), 5 74.9 (C2 GalA), 5 74.1 (CH2 Ph), 8 73.7 (OCH2 C H O H 2), 8 73.4 (CH2 Ph), 8 72.6 (C-5 GalA), 8 72.0 (C-3 Rha), 8 70.5 (C-2 Rha), 8 68.5 (C-5 Rha), 8 52.9 (C0 2 CH3), 6 21.2 (GalA OCOCH3), 8 21.0 (Rha OCOCH3 ), 8 18.2 (C- 6 Rha). Anal, calcd for C4 1 H4 6 O 1 4 : C, 64.56; H, 6.08. Found: C, 64.70, H 5.79. -76- R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. Methyl (2-0-acetyl-4-0-allyl-3-0-benzoyl-a-L-rhamnopyranosyl)-(l ~'4)-(2,3-di-Obenzyl-a/f}-D-galactopyranosid)uronate (13) A sample of compound 12 (0.794g, 1.04 mmol) was dissolved in distilled MeOH (25.8 mL) and CH2 CI2 (14.7 mL)66. Powdered 4 A molecular sieves (0.529g, 0.667 eq. by weight to 12) were then added. The reaction was stirred at rt for approx. four days, after which the solids were removed by filtration (TLC solvent B). The filtrate was evaporated to dryness and the resulting syrup was chromatographed (eluting solvent B) to give pure 13 (0.670g, 89.3%). 13C NMR for 13 (75.03 MHz, CDCI3 ): 8 134.8 (OCH2 CH-CH2), 8 116.9 (OCH2 CH=CH2), 8 99.9, 99.3 (C-l 0/(3 Rha, Jc,u 173.4 Hz), 8 92.4,92.3 (C-l 0/13 GalA, J C,H 171.5 Hz, 172.3 Hz), 8 52.7 (C0 2 CH3), 8 21.0 (Rha OCOCH3 ), 8 18.4,18.2 ( C ^ Rha); mass spectrum (ESI) m/z 743.2676 (M+Na). Methyl (2-0-acetyl-4-0-allyl-3-0-benzoyl-a-L-rhamnopyranosyl)-(1 ~*4)-2,3-di-0benzyl-a-D-galaclopyranosyluronate trichloroacetimidate (14) A sample of hemiacetal 13 (0.592g, 0.821 mmol) was dissolved in freshly distilled CH2 CI2 (13.5 mL) and stirred under N2 (g) in the presence of 4 A molecular sieves (0.590g) at rt for approx. 10 min., after which the mixture was placed in an ice bath and stirred for an additional 15 min41. Tricholoroacetonitrile (837 pL, 8.35 mmol, 10.2 eq.) was then added, followed by l,8-diazabicyclo[5.4.0]undec-7-ene (DBU; 147 pL, 0.983 mmol, 1.20 eq.). The reaction was followed by TLC (TLC solvent B) and stopped after 10 min. The mixture was immediately placed on a vacuum column (eluting solvent Q ; fractions containing 14 were combined and evaporated to dryness. Impure 14 was then chromatographed twice; 14 was first chromatographed using eluting solvent E (0.498g pure 14); impure fractions were collected and chromatographed again using eluting solvent L (0.079g). The overall yield of 14 was 0.577g (81.3%). 'H NMR (300.13 MHz, -77- R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. CDC13): 8 8.63 (s, 1 H, C=NH), 5 8.05-7.97 (m, 2 H, aromatic), 8 7.63-7.20 (m, 13 H, aromatic), 8 6.67 (d, 1 H, J 1;2 3.3 Hz, H-l GalA), 8 5.79 (m, 1 H, OCH2 Gtf=CH2), 8 5.61 (m, 1 H, J 3 ,2 3.3 Hz, H-2 Rha), 8 5.48 (dd, 1 H, J 3 ,4 9.4 Hz, H-3 Rha), 8 5.25-5.03 (m, 3 H, H-l Rha, OCH2 CH=Ci72), 8 4.85-4.70 (m, 4 H, 2 OC//2 Ph), 8 4.65-4.61 (m, 1 H, H-5, GalA), 8 4.57-4.52 (m, 1 H, H-4, GalA), 8 4.30 (dd, 1 H, J 2 ,3 10.0 Hz, H-2 GalA), 8 4.224.05 (m, 2 H, OC7/2 CH=CH2), 8 4.05 (dd, 1 H, J 3 ,4 2.7 Hz, H-3 GalA), 8 3.85 (s, 3 H, C 02CH3), 8 3.78 (dd, 1 H, J5fi 6.2 Hz, H-5 Rha), 8 3.50 (dd, 1 H, Jy5 9.5 Hz, H-4 Rha), 8 2.04 (s, 3 H, OCOCH3), 8 1.37 (d, 3 H, H- 6 Rha); 13C NMR (75.03 MHz, CDC13): 8 169.6, 168.4 (0=0), 8 165.3 (C=0, benzoate), 8 160.7 (OC(NH)CCl3), 8 138.4, 138.0, 133.2,130.3, 129.7, 128.6, 128.5, 128.5, 128.1, 127.9, 127.8,127.7 (C aromatic), 8 134.8 (OCH2 CH=CH2), 8 116.9 (OCH2 CH=CH2), 8 99.3 (C-l Rha, Jc,h 174.7 Hz), 8 94.9 (C-l GalA, JCM 180.4 Hz), 8 78.5 (C-4 Rha), 8 76.3 (C-3 GalA), 8 75.7 (C-4 GalA), 8 75.5 (C2 GalA), 8 73.7 (OCH2 CH=CH2), 8 73.7, 73.3 (CH2 Ph), 8 72.8 (C-5, GalA), 8 72.0 (C-3 Rha), 8 70.5 (C-2 Rha), 8 6 8 . 6 (C-5 Rha), 8 52.9 (C0 2 CH3), 8 21.0 (OCOCH3), 8 18.2 (C- 6 Rha); mass spectrum (ESI) m/z 866.1777 (M+Na). Benzyl (2-0-acetyl-4-0-allyl-3-0-benzoyl-a-L-rhamnopyranosyl)-(l ~*4)-(l-O-acetyl-2,3di-0-benzyl-a/[l-D-galactopyranosid)uronate (15) A sample of 15 was prepared by dissolving dissacharide 9 (1.830g, 2.257 mmol) in AcOH (37.3 mL) and Ac20 (37.3 mL)51. The reaction mixture was then cooled in an ice bath for 10 min, after which H2 S0 4 (370 pL) was slowly added. The reaction was followed by TLC (TLC solvent H) and showed completion after approx. 2 h 05 m. The reaction mixture was allowed to warm to rt during the progress of the reaction. Sodium acetate trihydrate (Na0Ac-3H2 0 ; 3.91 lg) was then added and the mixture was stirred until all Na0Ac-3H20 had dissolved. The mixture was then evaporated to dryness, taken -78- R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. up in EtOAc, washed with H2 O (extracted 4x with EtOAc), dried over Na2 S0 4 , filtered, evaporated to dryness, and finally co-evaporated three times with EtOH. The resulting syrup was chromatographed (eluting solvent E) to obtain pure 15 (1.730g, 91.4%). 'H NMR data for 15 (300.13 MHz, CDC13): 5 2.12 (s, 3 H, GalA -OCOCH3, major anomer), 5 2.01 (s, 3 H, Rha -OCOCH3, major anomer); 13C NMR (75.03 MHz, CDCI3 , major anomer): 5 169.7,169.5,169.0,167.6 (C=0, Rha OC(0)CH3, GalA OC(0)CH 3 and C 0 2 CH2 Bzl), 5 165.2 (C=0, benzoate), 5 138.2-127.8 (C aromatic), 5 134.8 (OCH2 C H C H 2 ), 5 116.9 (OCH2 CH=CH2), 5 99.6 (C-l Rha, J C,H 173.9 Hz), 5 90.6 (C-l GalA, Jc,h 177.5 Hz), 5 78.5 (C-4 Rha), 5 77.2 (C-3 GalA), 5 76.6 (C-4 GalA), 5 75.0 (C2 GalA), 5 73.8 (OCH2 CH=CH2), 5 73.5, 73.3 (CH2 Ph), 5 72.5 (C-5 GalA), 5 72.1 (C-3 Rha), 5 70.5 (C-2 Rha), 8 68.7 (C-5 Rha), 8 67.8 (C0 2 Bzl), 8 21.3 (GalA OCOCH3), 8 21.0 (Rha OCOCH3), 8 18.4 (C- 6 Rha). Anal, calcd for C4 7 H5 0 O1 4 : C, 67.29; H, 6.01. Found: C, 67.08, H 6.05. Benzyl (2-0-acetyl-4-0-allyl-3-0-benzoyl-a-L-rhamnopyranosyl)-(l ~*4)-(2,3-di-Obenzyl-oJp~D-galactopyranosid)uronate (16) A sample of compound 15 (0.277g, 0.330 mmol) was dissolved in distilled MeOH (8.2 mL) and CH2 C12 (4.7 mL)66. Powdered 4 A molecular sieves (0.185g, 0.667 eq. by weight to 15) were then added. An additional 0.055g of 4 A molecular sieves was added on day 3. The reaction was stirred at rt for approx. four days total, after which the solids were removed by filtration (TLC solvent J). The filtrate was evaporated to dryness and the resulting syrup was chromatographed (eluting solvent J) to give pure 16 (0 .2 0 0 g, 76.0%). 13C NMR for 13 (75.03 MHz, CDC13): 8 134.9 (OCH2 CH=CH2), 5 116.9 (OCH2 CH=CH2), 8 99.5 (C-l ^ Rha, Jc,H 172.6 Hz), 8 92.4, (C-la/p GalA, Jc,h 173.2 Hz), -79- R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. 5 21.0 (Rha OCOCH3), 5 18.3 (C-6 a/p Rha). Anal, calcd for C4 5 H4 8 O 1 3 : C, 67.83; H, 6.07. Found: C, 67.96, H 6.16. Benzyl (2-0-acetyl-4-0-allyl-3-0-benzoyl-a-L-rhamnopyranosyl)-(l ~*4)-2,3-di-Obenzyl-oJfi-D-galactopyranosyluronate trichloroacetimidate (17) A sample of hemiacetal 16 (0.480g, 0.602 mmol) was dissolved in freshly distilled CH2 CI2 (12.9 mL) and stirred under N2(g) in the presence of 4 A molecular sieves (0.476g) at rt for approx. 10 min41. Trichloroacetonitrile (605 pL, 6.03 mmol, 10.0 eq.) was then added, followed by DBU (108 pL, 0.725 mmol, 1.20 eq.) The reaction was followed by TLC (TLC solvent J) and stopped after 20 min. The mixture was immediately placed on a vacuum column (eluting solvent C); fractions containing 17 were combined and evaporated to dryness. Impure 17 was then chromatographed twice (eluting solvent K, followed by chromatography with eluting solvent L), giving pure 17 (0.464g, 81.9%). ’HNMR (300.13 MHz, CDC13): 5 8.61 (s, C=N/7); 13C NMR (75.03 MHz, CDCI3 , major anomer): 5 169.5,167.5 (C-O), 8 165.2 (C=0, benzoate), 8 160.7 (OC(NH)CCl3), 8 138.4-127.7 (C aromatic), 8 134.8 (OCH2 CH=CH2), 8 116.9 (OCH2 CH=CH2), 8 99.6 (C-l Rha, J c ,h 173.2 Hz), 8 95.0 (C-l GalA, JC,H 180.1 Hz), 8 78.5 (C-4 Rha), 8 76.7 (C-4 GalA), 8 76.3 (C-3 GalA), 8 75.6 (C-2 GalA), 8 73.7 (OCH2 CH=CH2, CH2 Ph), 8 73.4 (CH2 Ph), 8 72.8 (C-5, GalA), 8 72.1 (C-3 Rha), 8 70.6 (C-2 Rha), 8 68.7 (C-5 Rha), 8 67.8 (C0 2 CH2 Bzl), 8 21.0 (OCOCH3), 8 18.4 (C- 6 Rha); mass spectrum (ESI) m/z 962.2087 (M+Na). -80- R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. Methyl (2-0-acetyl-4-0-allyl-3-0-benzoyl-a-L-rhamnopyranosyl)-(l ^4 )-(me thy12,3-di0-benzyl-a-D-galactopyranosyluronate)-(l -*2)-(2-0-acetyl-4-0-allyl-3-0-benzoyl-a-Lrhamnopyranosyl)-(l ~-4)-(methyl 2,3-di-0-benzyl-a-D-galactopyranosid)uronate (18) The acceptor (10) (0.243g, 0.351 mmol, 1.00 eq.) and donor (14) (0.426g, 0.492 mmol, 1.40 eq.) were placed together in a flask along with 4 A molecular sieves (~0.530g). Freshly distilled CH2 CI2 (5.7 mL) was added; the flask was then purged with N2(g) and allowed to stir at rt and in the dark for 1 h, after which AgOTf was added (0.0483g, 0.382 eq. to donor). The reaction was followed by TLC (TLC solvent M) and worked up after 17 h 20 m. Workup was performed by diluting the reaction mixture with CH2 CI2 , removing the solids by filtration, washing the filtrate sequentially with H2 O and satd NaHC0 3 , and drying the organic layer over Na2 SC>4 . The drying agent was removed by filtration and the filtrate evaporated to dryness to afford 18 as a syrup. The crude syrup was chromatographed using eluting solvent M; impure fractions were combined and again chromatographed with the same solvent. The crude mixture was chromatographed a total of four times to give pure 18 (0.190g, 38.8%). [a] ^ 2 +63.3° (c 0.196, CHCI3 ); 'H NMR (300.13 MHz, CDC13): 5 8.03-7.94 (m, 4 H, aromatic), 5 7.647.21 (m, 26 H, aromatic), 5 5.86-5.68 (m, 2 H, OCH2 C77=CH2), S 5.55 (dd, 1 H, J 3,i 3.3 Hz, H-2’” ), 5 5.44 (dd, 1 H, H-3” ’), 5 5.41 (dd, 1 H, H-3’), 5 5.31 (d, 1 H, H -l’), 5 5.19-5.00 (m, 4 H, OCH2 CH=CH2), 5 5.11 (m, 1 H, H -l’” ), 5 4.94,4.75 (2 d, 2 H, OC772 aPh), 5 4.85, 4.61 (2 d, 2 H, O C /^Ph), 8 4.78 (m, 1 H, H -l” ), 5 4.76 (m, 1 H, H-l), 8 4.72, 4.66 (2 d, 2 H, OC///Ph), 8 4.58, 4.42 (2 d, 2 H, OC/7/Ph), 8 4.47 (m, 2 H, H-4, H-5” ), 8 4.40 (m, 1 H, H-2’), 8 4.31 (m, 1 H, H-4” ), 8 4.20-4.10 (m, 4 H, OC/7 2 CH-CH 2 ), 8 4.14 (m, 1 H, H-2), 8 3.94 (m, 1 H, H-3), 8 3.94 (m, 1 H, H-3” ), 8 3.85 (s, 3 H, C 0 2 CH3), 8 3.84 (m, 1 H, H-2” ), 8 3.71 (m, 1 H, H-5’), 8 3.70 (m, 1 H, H- -81 - R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. 5’” ), § 3.50 (m, 1 H, H-4’), 5 3.46 (m, 1 H, H-4’” ), 5 3.83 (s, 3 H, C 0 2 CH3”), 6 3.30 (s, 3 H, OCH3), 5 2.02 (s, 3 H, OCOC# ?’” ), 6 1.36 (d, 3 H, J 6 ,5 6.1 Hz, H-6 ’), 5 1.29 (d, 3 H, J 6 ,5 6.3 Hz, H-6 ’” ); 13C NMR (75.03 MHz, CDC13): 5 169.7, 169.2,168.4 (C=0), 6 165.3, 165.2 (O O , benzoate), 5 139.0-127.7 (C aromatic), 6 135.0,134.8 (OCH2 CH=CH2), 5 116.9,116.7 (OCH2 CH=CH2), 6 99.4 (C-l, Jc,H 173.2 Hz), 8 99.0 (C -l’” , JC,H 174.4 Hz), 8 98.4 (C-l’, J C,H 172.6 Hz), 8 96.7 (C -l” , JC,H 171.5 Hz), 8 78.5 (C-4’, C-4’” ), 8 77.6 (C-3), 8 76.7 (C-3” ), 8 76.5 (C-2), 8 76.0 (C-4, C-4” ), 8 75.5 (C-2” ), 8 75.3 (C-2’), 8 74.4 (OCH2 aPh), 8 74.1 (OCH2 bPh), 8 73.7, 73.5 (OCH2 CH=CH2), 8 73.5 (C-3’), 8 72.7 (OCH2 cPh, OCH2 dPh), 8 72.1 (C-3’” ), 8 70.6 (C-5” , C-2’” ), 8 70.5 (C-5), 8 6 8 . 6 (C-5’), 8 68.3 (C-5’” ), 8 56.3 (OCH3), 8 52.7 (C0 2 CH3” ), 8 52.1 (C0 2 CH3), 8 21.0 (OCOCH3’” ), 8 18.4 (C-6 ’) , 8 18.2 (C-6 ’” ). Anal, calcd for C77 H8 6 0 2 4 : C, 66.27; H, 6.21. Found: C, 66.60, H 6.34. Methyl (2-0-acetyl-4-0-allyl-3-0-benzoyl-a-L-rhamnopyranosyl)-(l -*4)-(benzyl 2,3-di0-benzyl~a-D-galactopyranosyluronate)-(l —2)-(2-0-acetyl-4-0-allyl-3-0-benzoyl-a-Lrhamnopyranosyl)-(l ~*4)-(methyl 2,3-di-0-benzyl-a-D-galactopyranosid)uronate (19) The acceptor 10 (0.0534g, 0.0771 mmol, 1.00 eq.) and donor 17 (0.107g, 0.114 mmol, 1.48 eq.) were placed together in a flask along with 4 A molecular sieves (-0.24lg). Freshly distilled CH2 C12 (2.0 mL) was added; the flask was then purged with N2(g) and allowed to stir at rt and in the dark for 30 min., after which AgOTf was added (0.0133g, 0.454 eq. to donor). The reaction was allowed to stir for 8 days, during which time CH2 C12 was added in order to maintain initial solvent levels. Portions of AgOTf were also added over this time period to a total of 0.956 eq. to the donor. The reaction was stopped, filtered to remove all solids, and loaded directly onto a chromatography column (eluting solvent J). Impure fractions were collected and chromatographed two -82- R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. additional times with eluting solvent N. All isolated samples were pooled to give 19 with a yield of 31.7% (0.0360g). [ a g +62° (c 0.119, CHCI3 ); 'H NMR (300.13 MHz, CDCI3 ): 8 8.04-7.90 (m, 4 H, aromatic), 8 7.74-7.14 (m, 31 H, aromatic), 8 5.84-5.68 (m, 2 H, OCH2 C//=CH2), 8 5.60 (dd, 1 H, . / 3;2 3.3 Hz, H-2” ’), 8 5.49 (dd, 1 H, J3A 9.6 Hz, H-3’” ), 8 5.42 (dd, 1 H, J3A 9.6 Hz, H-3’), 8 5.23 (d, 1 H, H -l’), 6 5.18-4.97 (m, 4 H, OCH2 CH=CH2), 8 5.06 (m, 1 H, H -l’” ), 8 4,90,4.70 (2 d, 2 H, OCtf/Ph), 8 4.85,4.56 (2 d, 2 H, OCtf/Ph), 8 4.71, 4.63 (2 d, 2 H, OC/7/Ph), 8 4.67 (m, 1 H, H-l), 8 4.61, 4.44 (2 d, 2 H, OCtf/Ph), 8 4.61 (m, 1 H, H-l), 8 4.47 (m, 1 H, H-5” ), 5 4.43 (m, 1 H, H-4), 8 4.40 (m, 1 H, H-5), 8 4.38 (m, 1 H, H-2’), 8 4.30 (m, 1 H, H-4” ), 8 4.20-4.01 (m, 4 H, OCf/2 CH=CH2), 8 4.12 (m, 1 H, H-2), 8 3.91 (m, 1 H, H-3), 8 3.90 (m, 1 H, H-3” ), 8 3.85 (s, 3 H, CO2CH3), 8 3.81 (m, 2 H, H-2” , H-5” ’), 8 3.71 (dd, 1 H, J5,6 6.2 Hz, H-5’), 8 3.46 (m, 1 H, J5A 9.3 Hz, H-4’), 8 3.46 (m, 1 H, H-4” ’), 8 3.36 (s, 3 H, OCH3), 8 2.01 (s, 3 H, OCOC/ / / ” ), 8 1.33 (d, 3 H, H-6 ’), 8 1.27 (d, 3 H, H-6 ” ’); 13C NMR (75.03 MHz, CDCI3): 8 169.6,169.2,167.9 (C=0), 8 165.2,165.1 (C=0, benzoate), 8 139.2-127.8 (C aromatic), 8 135.0,134.9 (OCH2 CH=CH2), 8 116.8,116.6 (OCH2 CH=CH2), 8 99.5 (C-l, Jc,u 173.2 Hz), 8 99.3 (C -l’” , Jc,n 174.4 Hz), 8 98.7 (C -l’, Jc,n 172.6 Hz), 8 96.4 (C -l” , JCM 171.5 Hz), 8 78.7 (C-4’), 8 78.5 (C-4’” ), 8 77.4 (C-3), 8 77.0 (C-4” ), 8 76.7 (C-3” ), 8 76.6 (C-4), 8 76.5 (C-2), 8 75.7 (C-2” ), 8 75.1 (C-2’), 8 74.5 (OCH2 aPh), 8 74.3 (OCH2 bPh), 8 73.8, 73.5 (OCH2 CH=CH2), 8 73.3 (C-3’), 8 72.9 (OCH2 cPh), 8 72.7 (OCH2 dPh), 8 72.2 (C-3’” ), 8 70.7 (C-5” ), 8 70.6 (C-2’” ), 8 70.5 (C-5), 8 68.4 (C-5’, C5’” ), 8 72.7 (C0 2 CH2 Ph), 8 56.2 (OCH3), 6 52.7 (C0 2 CH3), 6 2 1 . 1 (OCOCH3’” ), 5 18.4 (C-6 ’), 8 18.2 (C-6 ’” ). -83 - R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. Methyl (a-L-rhamnopyranosyl)-(l —4)-(methyl a-D-galactopyranosyluronate)-(l ~*2)-(aL-rhamnopyranosyl)-(l -~4)-(methyl a-D-galactopyranosid)uronate (20) A sample of 18 (0.190g, 0.136 mmol) was dissolved in EtOH/Toluene/H20 10:3:1 (28.0 mL)75 and the reaction flask was purged with N2(g). Wilkinson’s catalyst ([(C6H5)3P]3RhCl; 0.0502g, 0.0543 mmol, 0.4 eq.) was added, followed by DABCO (0.0153g, 0.136 mmol, 1.00 eq.). The reaction was set to reflux and monitored by TLC (TLC solvent M). After approx. 22 h, TLC showed the isomerization of the allyl to the vinyl ether to be incomplete so an additional 0.2 eq. of Wilkinson’s catalyst was added, followed by 0.5 eq. of DABCO. The reaction was stopped after 46 h. The mixture was evaporated to dryness, taken up in CH2C12 and sequentially washed with 0.1 M HC1, satd NaHC0 3 , and H20. The organic layer was dried over Na2 S0 4 , filtered to remove the drying agent, and evaporated to dryness once again to give the vinyl ether as the major product. The vinyl ether was cleaved by dissolving the product from the first step in 9:1 acetone/H20 (20 mL). HgCl2 and HgO were then added and the reaction mixture was allowed to stir at rt. TLC (TLC solvent M) indicated the reaction may have been incomplete, so additional portions of HgCl2 (0.0369g) and HgO (0.0290g) were added after 26 h 20 m and again after 45 h. No significant change was observed so the reaction mixture was filtered after 50 h 50 m. through celite and evaporated to dryness. The residue was taken up in EtOAc and sequentially washed with satd KI, satd Na2 S2 0 3 , and H20 . The organic layer was dried over Na2 S0 4 , filtered, and evaporated to dryness. The residue was chromatographed (TLC and eluting solvent P) and samples containing the target compound were pooled to yield a yellow-brown foam (0.090g, 54.3%, slightly crude). -84- R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. Removal of the benzoyl groups and the acetate group was accomplished by dissolving the yellow-brown foam obtained above (0.090g, 0.0683 mmol, slightly crude) in freshly distilled MeOH (12.0 mL) and adding 1.0 M NaOMe until litmus paper indicated a pH between 11-12. TLC (TLC solvent Q) taken approx. 24 h after the start of the reaction indicated the reaction had gone to completion. The reaction was neutralized by the addition of Amberlyst® 15 H+ resin beads, filtered and evaporated to dryness to afford a yellowish syrup. The syrup was chromatographed using eluting solvent R, giving a clear, colorless syrup as the product (0.048g). The pure product obtained in the previous step (0.048g, 0.045 mmol) was dissolved in EtOAc/MeOH/AcOH 5:5:1 (11.0 mL), to which palladium acetate (0.209g, 0.931 mmol) was added77. The mixture was placed under H2(g) at 40 psi for 1 h. TLC (TLC solvent R) indicated no starting material, with all material present on the baseline. The mixture was worked up by filtering through celite (rinsed with EtOAc and MeOH), followed by evaporation and drying under high vacuum; a white powder (0.032g, quantitative yield) was collected. TLC showed (TLC solvent S) one spot without any visible trace of impurities, [a]^2 +59.7° (c 0.191, MeOH); 'H NMR (300.13 MHz, CH3OD): 5 5.21 (d, 1 H, H -l’), 6 5.14 (d, 1 H, Ji>2 1.5 Hz, H -l’” ), 5 5.11 (d, 1 H, H-5), 6 5.00 (d, 1 H, Ji,2 3.7 Hz, H-l), 8 4.81 (d, 1 H, J l2 3.7 Hz, H -l” ), 8 4.52 (d, 1 H, H-5” ), 8 4.39-4.32 (m, 2 H, H-4” , H-4), 8 4.11-4.06 (m, 1 H, H-2’), 8 4.01-3.90 (m, 3 H, H-2’” , H-3, H-3” ), 8 3.83-3.69 (m, 3 H, H-2” , H-2, H-3’), 8 3.81 (s, 3 H, C 02CH3), 8 3.78 (s, 3 H, CO2 C// 3), 5 3.63 (dd, 1 H, J 3 ,2 3.2 Hz, J 3 ,4 9.4 Hz, H-3’” ), 8 3.54-3.39 (m, 2 H, H-5’, H-5’” ), 8 3.41 (s, 3 H, OCH3), 8 3.38-3.29 (m, 2 H, H-4’” , H-4’), 8 1.35-1.20 (m, 6 H, H-6’, H-6’” ); 13C NMR (75.03 MHz, CH3OD): 8 170.1,169.7 (C02CH3), 8 102.2 (C -l’” , -85- R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. Jcm 171.5 Hz), 5 100.6 (C -l” , J c ,h 171.0 Hz), 5 99.8 (C -l’, Jc,e 171.5 Hz), 698.3 (C-l, Jc,H 171.5 Hz), 6 78.4 (C-4” ), 6 77.9 (C-4), 6 76.9 (C-2’), 5 72.6 (C-4’” ), 5 72.5 (C-4’), 6 70.9 (C-2” ’), 6 70.8 (C-3), 6 70.6 (C-5), 6 70.3 (C-3’” ), 5 70.1 (C-3” ), 6 69.9 (C-5” ), 6 69.8 (C-3’), 6 69.4 (C-5’), 6 69.1 (C-5’” ), 5 6 8 . 6 (C-2 ” , C-2 ), 5 55.1 (OCH3), 6 55.1 (OCH3), 5 51.7, 51.6 (C0 2 CH3), 5 17.0 (C-6 ’), 5 17.0 (C-6 ’” ) , 6 16.9 (C-6 ’). Anal, calcd for C2 7 H4 4 O2 1 : C, 46.02; H, 6.29. Found: C, 45.88, H 6.39. - 86 - R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. Chapter Six Conclusions and Future Work The synthesis of two repeat units of the RG-I backbone proved to be challenging, particularly the glycosylation of the two disaccharides to form the fully protected tetrasaccharide. As anticipated, forming the a-linkage was no trivial task, but an increase in yield through the reduction of the quantity of molecular sieves used appears promising towards achieving yields above 40%. The possibilities of other glycosylation methods are still extensive in terms of adjusting the reaction conditions, the nature of the donor, as well as the acceptor. The reaction scheme successfully carried out in this research has demonstrated the ability to synthesize RG-I using galacturonic acid, thereby avoiding the need for latestage oxidation. Most reactions, with the exception of the glycosylation to form the tetrasaccharide and removal of the allyl groups, proceeded in relatively high yield, suggesting the synthetic method used here is extremely viable for the production of milligram to gram quantities of the tetrasaccharide. Sufficient quantities of the RG-I backbone can now be synthesized for the potential further investigation of RG-I’s role in plant physiology through the use of monoclonal antibodies. Several aspects of the work performed here can be extended or investigated further. The glycosylation of the rhamnosyl donor (compound 7) with the two variations of the galacturonic acid acceptor (compounds 5 and 6) need to be further investigated. Larger quantities of the triflic acid promoter resulted in much higher yields; however the temperature dependence for the conditions used was not fully investigated, and therefore -87- R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. no conclusion could be drawn as to the necessity of performing the reaction in relatively cool conditions (-40°C versus room temperature). The glycosylation of the disaccharide donors with the disaccharide acceptor proved to be problematic. Although several conditions were investigated here, many more variations on reaction conditions are possible, such as using other promoters, solvents, donors, etc. The most promising route appeared to be the use of an imidate donor; however, the conclusion could not be made that the use of a thioglycoside donor is not feasible. Further work would have to be done in this area in the event higher yields with the imidate cannot be achieved. Increasing the yield of the tetrasaccharide appears to be most promising by lowering the quantity of molecular sieves used, or even altogether eliminating the use of molecular sieves in the reaction. By halving the amount of molecular sieves used, the yield increased by about 10%. Also, if the efficiency of the reaction can be increased, thereby significantly reducing the amount of acceptor that remains, the difficulty in purifying the reaction mixture will be significantly reduced. Precise experiments to determine the difference in using the benzyl ester of the disaccharide donor compared to the methyl ester were not performed. Although observations were made throughout the synthesis about similarities and differences in reactivity between the two, no concrete conclusions could be drawn. The benzyl ester did appear to be more sensitive that the methyl ester disaccharide during manipulations made in donor preparation, suggesting differences do indeed exist. Removal of the allyl in the deprotection phase led to low yields for all attempts made. An improved procedure would be desirable to minimize loss. Further investigation - 88 - R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. as to how and where product loss occurs would also be valuable. Alternative methods may also exist. The possibility of replacing the allyl group with another orthogonal protecting group may be something to consider, although this would involve developing a new synthetic strategy at the level of the rhamnosyl monosaccharide; furthermore, the implications of altering the group on all other experiments involving the rhamnosyl donor is unknown. Successful removal of the allyl group also has implications for further research. The group would need to be removed in the future for addition of other glycosides to the C-4 position of the rhamnose residues. Incomplete removal of the group led to a mixture of products, which could be difficult to isolate. The pure product needs to be isolated in order to perform future experiments. The final deprotection step to the carboxylic acid was never carried out on the final compound. Difficulties in the deprotection were observed in previous attempts; however, it should be noted previous attempts did not involve a verified, pure product, and therefore other influences may have resulted in the removal of the methyl ester groups not occurring under the desired conditions. Further experimentation with the pure compound is therefore necessary. The purpose of the synthesis of RG-I is for future biological studies in the area of structure/function relationships with respect to plant physiology to be carried out. Additional research will need to be performed into what methods will work creating the necessary linker arm at the reducing end of the synthetic compound, followed by suitable attachment to a carrier protein. Ultimately, the synthesized antigen will be used to raise monoclonal antibodies, and the desired biological studies can be performed. -89- R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. References 1. Sharon, N.; Lis, Halina. Sci. Amer. 1993,268(1), 82. 2. Penmans, W. J.; Van Damme, E. J. M. Plant Physiol. 1995,109(2), 347. 3. Drickhamer, K. J. Biol. Chem. 1988,263(20), 9557. 4. Leffler, H.; Carlsson, S.; Hedlund, M.; Qian, Y.; Poirier, F. Glycoconj. J. 2004,19, 433. 5. Danguy, A.; Camby, I.; Kiss, R. Biochim Biophys Acta 2002,1572,285. 6. Becker, B.; Furneaux, R.H.; Reck, F.; Zubkov, O. A. Carbohydr.Res. 1999, 315 (1-2), 148. 7. Drickamer, K.; Taylor, M.E. Annu. Rev. Cell Biol. 1993, 9, 237. 8. Fiete, D.; Baenziger, J.U. J. Biol. Chem. 1997, 272, 14629. 9. Ezekowitz, R.A.B.; Stahl P.D. J. Cell. Sci. Suppl. 1998, 9 , 121. 10. Hoog, C.; Rotondo, A.; Joahnston, B.; Pinto, B.M. Carbohydr. Res. 2002, 337(2123), 2023. 11. Liao, L.; Auzanneau, F. Org. Lett. 2003, 5(15), 2607. 12. Blake, M.S.; Zabriskie, J.B.; Tai, J.Y.; Michon, F. US Patent 5,866,135 (1999). 13. Weiss, K.A.; Laverdiere, M. Can. J. Surg. 1997, 40, 18. 14. Hakomori, S.-I. Adv. Cancer Res. 1989, 52, 257. 15. Lemieux, R.U.; Baker, D.A.; Weinstein, W.M.; Switzer, C. M. Biochemistry 1981, 20(1), 199. 16. Jones, L.; Seymour, G.B.; Knox, J.P. Plant Physiol. 1997,113(4), 1405. 17. An, J.; O’Neill, M.A.; Albersheim, P.; Darvill, A.G. Carbohydr. Res. 1994, 252(1-2), 235. 18. Knox, J.P. International Review o f Cytology 1997,171, 79. 19. Benaissa-Trouw, B.; Lefeber, D.J.; Kamerling, J.P.; Vliegenthart, J.F.G.; Kraaijeveld, K.; Snippe, H. Infect. Immun. 2001, 69,4698. 20. Smith, D.F.; Ginsburg, V. J. Biol. Chem. 1980, 255, 55. -90- R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. 21. Willats, W.G.T.; Marcus, S.E.; Knox, J.P. Carbohydr. Res. 1998,308(1-2), 149. 22. Willats, W.G.T.; Steele-King, C.G.; Marcus, S.E.; Knox, J.P. The Plant Journal 1999,20(6), 619. 23. McNeil, M.; Darvill, A.G.; Fry, St. C.; Albersheim, P. Annu. Rev. Biochem. 1984,53, 625. 24. Willats, W.G.T.; McCartney, L.; Mackie, W.; Know, J.P. Plant Mol. Biol. 2001, 47(1), 9. 25. O’Neill, M.A.; Ishii, T.; Albersheim, P.; Darvill A.G. Annu. Rev. Plant Biol. 2004, 5 5 ,109. 26. Hu, H.; Brown, P.H. Plant Physiol. 1994,105(2), 681. 27. Kobayashi, M.; Matoh, T. Asuma, J. Plant Physiol. 1996,110(3), 1017. 28. Ishii, T.; Matsunaga, T. Carbohydr. Res. 1996, 284(1), 1. 29. Williams, M.N.V.; Freshour, G.; Darvill, A.G.; Albersheim, P.; Hahn, M.G. Plant Cell 1996,8(4), 673. 30. Matoh, T.; Takasaki, M.; Takabe, K.; Kobayashi, M. Plant Cell Physiol. 1998, 39, 483. 31. Buffet, M.A.J.; Rich, J.R.; McGavin, R.S.; Reimer, K.B. Carbohydr. Res. 2004, 339(15), 2507. 32. Chauvin, A.; Nepogodiev, S.A.; Field, R.A. J. Org. Chem. 2005, 70(3), 960. 33. Albersheim, P.; Darvill, A.G.; O'Neill, M.A.; Schols, H.A.; Voragen, A.G.J. Pectins andPectinases; Elsevier Science: Amsterdam, 1996; pp. 47-55. 34. O'Neill, M.A.; Albersheim, P.; Darvill, A. Methods in Plant Biochemistry, Academic Press: London, 1990; Vol. 2, pp. 415-441. 35. Renard, C.; Crepeau, M.J.; Thibault, J.F. Eur. J. Biochem. 1999, 266, 566. 36. Jones, L.; Seymour, G.B.; Knox, J.P. Plant Physiol. 1997,113(4), 1405. 37. Serpe, M.D.; Muir, A.J.; Ademe-Onzighi, C.; Driouich, A. Int. J. Plant Sci. 2004, 165(4), 571. 38. Obro, J.; Harhold, J.; Scheller, H.V.; Orfila, C. Phytochemistry 2004, 65(10), 1429. 39. Greene , T.W.; Wuts, P.G.M. Protective Groups in Organic Synthesis (3rd ed), John Wiley & Sons, Inc., 1999, pp 25. -91 - R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. 40. Hinklin, R.J.; Kiessling, L.L. Org. Lett. 2002, 4(7), 1131. 41. Liew, S.T.; Wei, A. Carbohydr. Res. 2002, 337(14), 1319. 42. Lefeber, D.J.; Arevalo, E.A.; Kamerling, J.P.; Vliegenthart, J.F.G. Can. J. Chem. 2002,80(1), 76. 43. Alpe, M.; Oscarson, S. Carbohydr. Res. 2002, 337(19), 1715. 44. Thijssen, M-J. L.; Halkes, K. M.; Kamerling, J. P.; Vliegenthart, J. F. G. Bioorg. Med. Chem. 1994,2(11), 1309. 45. Garegg, P. J.; Olsson, L.; Oscarson, S. J. Org. Chem. 1995, 60(7), 2200. 46. Kanie, O.; Ito, Y.; Ogawa, T. J. Am. Chem. Soc. 1994,116(26), 12073. 47. Dasgupta, F.; Garegg, P.J. Carbohydr. Res. 1988,177, cl3. 48. Andersson, F.; Fiigedi, P.; Garegg, P.J.; Nashed, M. Tetrahedron Lett. 1986,27(33), 3919. 49. Hinklin, R.J.; Kiessling, L.L. J. Am. Chem Soc. 2001,123(14), 3379. 50. Kovac, P.; Hirsch, J.; Kovacik, V.; Kocis, P. Carbohydr. Res. 1980,85(1), 41. 51. Lonn, H.; Lonngren, J. Carbohydr. Res. 1984,132(1), 39. 52. Vogel, C.; Kramer, S.; Nolting, B.; Ott, A. J. Carbohydr. Chem. 2000,19(7), 891. 53. Garegg, P.J.; Norberg, T. Acta Chem. Scand. 1979, 33(2), 116. 54. Shingu, Y.; Nishida, Y.; Dohi, H.; Kobayashi, K. Org. Biomol. Chem. 2003,1(14), 2518. 55. Shingu, Y.; Miyachi, A.; Miura, Y.; Kobayashi, K.; Nishida, Y. Carbohydr. Res. 2005,340(14), 2236. 56. Kiss, J; Burkhardt, F. Helv. Chim. Acta 1970, 53(117), 1000. 57. Rich, J.R.; McGavin, R.S.; Gardner, R.; Reimer, K.B. Tetrahedron Asym. 1999, 10(1), 17. 58. Maruyama, M.; Takeda, T.; Shimizu, N.; Hada, N.; Yamada, H. Carbohydr. Res. 2000,325(2), 83. 59. Schmidt, H.W.; Neukom, H. Helv. Chim. Acta 1966, 49(1), 510. 60. Kovac, P.; Hirsch, J.; Kovacik, V. Carbohydr. Res. 1974, 32(2), 360. -92- R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. 61. Kovac, P. Carbohydr. Res. 1972,22(2), 464. 62. Magaud, D.; Grandjean, C.; Doutheau, A.; Anker, D.; Shevchik, V.; Cotte-Pattat, C.; Robert-Baudouy, J. Carbohydr. Res. 1998, 314(3-4), 189. 63. Hodosi, G.; Kovac, P. J. Am. Chem. Soc. 1997,119(9), 2335. 64. Byramova, N.E.; Ovchinnikov, M.V.; Backinowsky, L.V.; Kochetkov, N.K. Carbohydr. Res. 1983,124(1), C8-C11. 65. Mori, M.; Ito, Y.; Ogawa, T. Carbohydr. Res. 1990,195(2), 199. 66. Kartha, K.P.R.; Mukhopadhyay, B.; Field, R.A. Carbohydr. Res. 2004,339(3), 729. 67. Zhao, W.; Kong, F. Carbohydr. Res. 2005, 340(10), 1673. 68. Magaud, D.; Domazon, R.; Anker, D.; Doutheau, A. Org. Lett. 2000,2(25), 2275. 69. Perrin, C.L.; Armstrong, K.B.; Fabian, M.A. J. Am. Chem. Soc. 1994,116(2), 715. 70. Kwon, O.; Danishefsky, S.J. J. Am. Chem. Soc. 1998,120(7), 1588. 71. Wolff, G.; Rohle, G. Angew. Chem., Int. Ed. Engl. 1974,13, 157. 72. Myers, A.G.; Liang, J.; Hammond, M.; Harrington, P.M.; Wu, Y.; Kuo, E. Y. J. Am. Chem. Soc. 1998,120(21), 5319. 73. Driguez, P-A.; Lederman, I.; Strassel, J-M.; Herbert, J-M.; Petitou, M. J. Org. Chem. 1999, 64(26), 9512. 74. Bock, K.; Pedersen, C. J. Chem Soc., Perkin Trans. 2 1974,293. 75. Takeo, K.; Murata, Y.; Kitamura, S. Carbohydr.Res. 1992,224(2), 311. 76. Kim, J-H.; Yang, H.; Park, J.; Boons, G-J. J. Am. Chem. Soc. 2005,127(34), 12090. 77. Ennis, S.C.; Cumpstey, I.; Fairbanks, A.J.; Butters, T.D.; Mackeen, M.; Wormald, M.R. Tetrahedron 2002, 58(46), 9403. 78. Nilsson, P.; Wendt, O.F. J. Organomet. Chem. 2005, 690(18), 4197. 79. Jones, J.K.N.; Stacey, M. J. Chem. Soc. 1947,1340. 80. Catelani, G.; Colonna, F.; Marra, A. Carbohydr.Res. 1988,182(2), 297. 81. Edington, R.A.; Percival, E.E. J. Chem. Soc. 1953, 2473. -93- R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. Appendix 1 - NMR Data Table of Contents Methyl (methyl 3,4-O-isopropylidene-a-D-galactopyranosid)uronate (2) 'H NMR spectra 96 Methyl (methyl 2-0-benzyl-3,4-0-isopropylidene-a-D-galactopyranosid)uronate (3) ’H NMR spectra 100 Methyl (methyl 2-0-benzyl-a-D-galactopyranosid)uronate (4) ' H NMR spectra 105 Methyl (methyl 2,3-di-0-benzyl-a-D-galactopyranosid)uronate (6) 'H NMR spectra COSY 108 111 Methyl (2-0-acetyl-4-0-allyl-3-0-benzoyl-a-L-rhamnopyranosyl)-(l-*4)-(methyl 2,3-di-0-benzyl-a-D-galactopyranosid)uronate (8) 'H NMR spectra COSY 13C NMR spectra HMQC (inverse) 112 118 120 123 Methyl (4-0-allyl-3-0-benzoyl-a-L-rhamnopyranosyl)-(l -*4)-(methyl 2,3-di-Obenzyl-a-D-galactopyranosid)uronate (10) 'H NMR spectra COSY 13C NMR spectra HMQC (inverse) 124 129 131 135 Methyl (2-0-acetyl-4-0-allyl-3-0-benzoyl-a-L-rhamnopyranosyl)-( 1—4)-( 1-0acetyl-2,3-di-0-benzyl-a/p-D-galactopyranosid)uronate (12) 'H NMR spectra COSY 13C NMR spectra HMQC (inverse) 136 143 144 149 Methyl (2-0-acetyl-4-0-allyl-3-0-benzoyl-a-L-rhamnopyranosyl)-(l—4)-(2,3-di-0benzyl-a/(3-D-galactopyranosid)uronate (13) 'H NMR spectra COSY 13C NMR spectra HMQC (inverse) 151 157 159 162 Methyl (2-0-acetyl-4-0-allyl-3-0-benzoyl-a-L-rhamnopyranosyl)-(l—4)-2,3-di-0benzyl-a-D-galactopyranosyluronate trichloroacetimidate (14) !H NMR spectra COSY 13C NMR spectra HMQC (inverse) 164 170 172 176 -94- R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. Methyl (2-0-acetyl-4-O-allyl-3-<9-benzoyl-a-L-rhamnopyranosyl)-(l —4)(methyl 2,3-di-0-benzyl-a-D-galactopyranosyluronate)-(l—2)-(2-O-acetyl-4-0allyl-3-Obenzoyl-a-L-rhamnopyranosyl)-(1^4)-(methyl 2,3-di-O-benzyl-a-Dgalactopyranosid)uronate (18) *H NMR spectra COSY 13C NMR spectra HMQC (inverse) Methyl (a-L-rhamnopyranosyl)-(l —4)-(methyl a-D-galactopyranosylnronate)-(l (a-L-rhamnopyranosyl)-(l —-4)-(methyl a-D-galactopyranosid)nronate (20) 'H NMR spectra COSY 13C NMR spectra HMQC (inverse) -95 - R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. Methyl (methyl 3,4-0-isopropylidene-a-D -galactopyranosid)uronate (2 ) X -9 6 - R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. 6690EI- Methyl (methyl 3,4-0-isopropylidene-a-D -galactopyranosid)uronate (2) P \ ElEl- 9086 0 6 0 '6 fE J- 6Z'fr9EtOE' Z9E f - 8frZ6'0 srtzer99'EZEl- 26 '6 6 £ lUr'ZOH- OOE6 0 ZE'EZft88 Z Z H - 8GE6'0 ZH tejBaiuj -9 7 - R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. ■o 0000 E CL CL 3.9 S6£6’2 (N 9 9 i6 '0 iej63}ui ZH -9 8 - R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. Methyl (methyl 3,4-0-isopropylidene-a-D -galactopyranosid)uronate (2 ) E690 E tejBaiui ZH 99 R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. ppm IZE62 £z uii ia x 3 u_ o »—sr 20) •o X ppm Methyl (methyl 2-0-benzyl-3,4-0-isopropylidene-a-D -galactopyranosid)uronate (3) < - 100 - R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. ivrnzSE’88 f286'0612£P'£6129Z.'S6t2- I0'66t2~ Z6'Q022- 2S 0 2 S 9 2 ' I022- ErZ022E6EJ22- ZH IS J 6 0 J U I 101 - R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. ppm Methyl (methyl 2-0-benzyl-3,4-0-isopropylidene-a-D -galactopyranosid)uronate (3) 02E8I2- te■teetI 6 '9 S £ l - * 8 ’9*E;89'6 t'E l6E'SSEl*S ’Q5EI - ln IE'S8EIEI ' S9E f - 0S86 0 ES’BOt'f- 6 0 ts* ;- 66'*E *;ES'6E*;- es' ss*;- ppm Methyl (methyl 2-0-benzyl-3,4-0-isopropylidene-a-D -galactopyranosid)uronate (3) srEiei8 S '6 IE ; - ZH -102 R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. Methyl (methyl 2-0-benzyl-3,4-0-isopropylidene-a-D -galactopyranosid)uronate (3) 0000 E rN ZSBO't J6E0E 103- R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. ssreopZt’tE '9 SEl'S tfr- 802'S8fr- tejSaiui ZH - 104 R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. ppm Methyl (methyl 2-0-benzyl-3,4-0-isopropylidene-a-D -galactopyranosid)uronate (3 ) X to X ■ ag aI in rvi L. < £ S Methyl (methyl 2-0-benzyl-a-D-galactopyranosid)uronate (4 ) t. X 2l i -1 0 5 - R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. 01 ZS60I Methyl (methyl 2-0-benzyl-a-D -galactopyranosid)uronate ( 4 ) S2'S 22l- Z L ' 90EIEEBOEt2001EJtritE I- 66Z6 0 Ofr'SSEtZ6’92Et- 68*0 J Z0'06EJ- Zl'ZOWEE'BOU­ I U EcSc c o 0 £ '9 E H ?Z'6Et't- Z8101 ZH tejSsiui - 106 - R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. ppm SE OaH- SOElOl Methyl (methyl 2-0-benzyl-a-D-galactopyranosid)uronate (4 ) M3EI 'E 0000 E lejBsuui ZH -107 - R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. U 0) OJ in O •l o o • «•I O to o (O t/) H Z □ uniMii.Q°aiiDU .(/i(/)X (n-i(9a ppm Methyl (methyl 2,3-di-0-benzyl-a-D-galactopyranosid)uronate (6) O • cn z o o o o - 108 - R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. N C O Methyl (methyl 2,3-di-0-benzyl-a-D-galactopyranosid)uronate (6) B 2'9 lE l- BIOO'B S rt6 E t- 62'EOt’l - r 9 ‘Q W Q. 86'92frt£ 9 ’6 2 K - t*92S Q- 91 '6 E H 8 0 '9 W Z9 0SH61 Y S H - lejBaiui ZH 109 R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. .o N CD Methyl (methyl 2,3-di-0-benzyl-a-D-galactopyranosid)uronate (6) 0000 E in ~to ID en ^68 2 8EBI' I I 9 H I tejBaiU! - 110 - R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. U W 0» N U> V> X 3 _ U V U « N ffl 01 III U 11 £ N € « 3 OT 3 X X ‘§!§ss is ;ks 2 ;sr ■ g ° § • § a cn o “S 4S |°8° c —* 8 tn d In 8 ii iin si D la i n i n n j n w m a j ’j o o « oi • d 5 ». 5? •‘nSr>8'ng” &d fti d oi m a jsilassa BsniaiE£!!3i! 1L L L> U. I (J U U l l U. I Methyl (methyl 2,3-di-0-benzyl-a-D-galactopyranosid)uronate (6) IDsITS - Ill - R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. (D«IO ppm - 112 - R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. 6E 2191 9Z‘229l £062 E Methyl (2-0-acetyl-4-0-allyl-3-0-benzoyl-oe-L-rhamnopyranosyl)-( I —4)(methyl 2,3-di-0-benzyl-oc-D-galactopyranosid)uronate (8) oiem i V BSSf IzO' 09S5 _c\i in \ L t'BSl S2 0f9 t BSEWt VLLO'I 89'6t?9I 20'EB9l t'O* T89T 02 £89I 9E' 9B9I | o£ o ^ mo I 6E9UI 8802ZI *B'82ZI fE' TE2.1 or/Eii I 0000’t _oo 6 0 -BEit E O 'E m 8* 8 m in 10 ten I9JB31UI ZH -113 - R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. 1 6 'IS2I6/1'8581- 1Cn) 2rct b O c x> i o € 5 i F8 08EI- £DB8'8 8B'SEEl- 1 2* sa N ro C bO -S 9 2 N =ra$• Xs) o o CL) (N 86'SOt)I OD I H tB'HH 886EE Si ZPfT S3 ESPf 8PS^t 88 £SH 8568 8 8 t'5 9 H lejBsiui -114- R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. ppm ZG' LLVX 69 9E0I Z0'9K)P BEZ2S Methyl (2-0-acetyl-4-0-allyl-3-0-benzoyl-a-L-rham nopyranosyl)-( I -~4)(methyl 2,3-di-0-benzyl-a-D-galactopyranosid)uronate (8) ES't’GOl 9fr'680t 2122H 9 2 '8 2 1 1 6E m i I E L9 v P 6 'E m 26 OSff Ll'l9l\ e y z iii 6£'8Zn triBU g|r'88H 6622 I E l '1 6 1 J t’6'8 P2t 9I222P E9'922P 1 9 ' IE21 06SE2J 9 I'iE 2 t 6 9 ' 0P2P 9Z 't* 2 I E l 9*21 161921 PES9 E 6Z'B92I tejBajui ZH -115- R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. SZJt'-E 8EZ2' I -o m s't - 116 - R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. 1929*E -o mz'z 117 R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. uidd Ibj63;ui R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. SS2 J S S S 11 £ si g I S in 3 > V u0 t) ID 1 3 IO to ’ » in o i* v opj a> S • «*T ^ Q o— tni SI 5S e m iS S S E S(VSo O " PI *HOl oSS S> fvi s m in c o in a UJ § in at oc. I a. c c s s f e i l Hs -119 - R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. ?SS5 £&£' Ll. U. U. I Methyl (2-0-acetyl-4-0-allyl-3-0-benzoyl-a-L-rham nopyranosyl)-( I —4)(methyl 2,3-di-0-benzyl-a-D-galactopyranosid)uronate (8) ItS'Bl 9 t 0 ‘ 12 LZLZS EZ2 9S 86E’89 682' Oi 2ES' OZ OOO ZL 8ZS’EZ ft-S'ft S lB 'S t 26£’9Z 9te^t VLVLL 8E9' LL 6 t S ’8 i EEI>'66 ESS'66 2EB9H9Z9'Z2lS6ZZ2I280821t'OS’82;6 8 S '82lSfZ'821W 8 '8 2 l06 9 '6 2 l96£'0E lfa t'E E l8S8*E l61^'BEf- 891'595660'69?EfS'691- uidd - 120 - R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. Methyl (2-0-acetyl-4-0-allyl-3-0-benzoyl-a-L-rham nopyranosyl)-( I -*4)(methyl 2,3-di-0-benzyl-a-D-galactopyranosid)uronate (8) geegnB^g S6Z iz\ geo 8gi ^OS'Bgl 689 851 SU 851 t'^aagt 068 651 -C O 96E OEl frgl'EEl BSB’t'El 6 I p BEI o - ID 891'991660'69TEl'S'691- uidd -121 - R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. -o Methyl (2-0-acetyl-4-0-allyl-3-0-benzoyl-a-L-rham nopyranosyl)-( I —4) (methyl 2,3-di-0-benzyl-a-D-galactopyranosid)uronate (8) U eot'S'st- moiz- I uo iiiininimiiiiiiinimim..imililllmmuimmniimiiiuniHMmN1imilii "O u>s r o« q> 91 2 6 iPSl- Methyl (4-0-allyl-3-0-benzoyl-a-L-rham nopyranosyl)-( I —4)(methyl 2,3-di-0-benzyl-a-D-galactopyranosid)uronate ( 10) 8^'6t'9l- CM S8‘ 1991W E 9 B IE I '999199St- in PB'B6Sl- > 9 'H 9 I2 Z il9 l28'2291t6 '9 2 9 l- _io in 9 6 '6 l £ tQP'GBLf6P0EZ.IES'SEZt- nzEzt- et>-' tt'Zi- 0 0 0 0 'I P92PHPVBPifEO'EQZI^ 9 ‘89ZtZH -125- R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. ------ 66'teei—— is' eebw 58 8rg£2t-v^ 02'6E2t— ts X uII X u 3f u 9S0fr2I — 06' 65' 61^21- 66'0S2t- o Methyl (4-0-allyl-3-0-benzoyl-a-L-rham nopyranosyl)-(l —4)(methyl 2,3-di-0-benzyl-a-D-galactopyranosid)uronate (10) 0*2521ZE 9521 SZZQ21 02 292I E9E921 10 6921 I* 0Z2t U 2621 t’9 5621 ZS 8621 * 6'22E l- 02'*2£[- 0Z'9EEt22'6E E l08'0*E l- 59'00*1 OZ'ZOM ZS 2 m 5261*15Z 'E 2*t- ziizn- CL Q- 52'6E *r- 20B**J 6Z 15*1 SS'6S*I EZ'E9*r -126 - R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. J902 0J- 9 6 I^ '£ 81'9201- Methyl (4-0-allyl-3-0-benzoyl-a-L-rham nopyranosyl)-(l —4)(methyl 2,3-di-0-benzyl-a-D-galactopyranosid)uronate (10) 8 9 '9fr0l- in 08 > 9 0 1 - 9 9 1 2 't 0 1 'E 9 0 t- E f9 H l- B8 22H96 9 2 U £ 0 '6 2 H £ r2 E IlE 2'9E H ilE'BEH- co pi 9*Em621911- E6Z9E E 2 /:9 H - 26>8H E 9 'Z 8 ItI0 9 6 H \L 'L 6H 9 9 E 0 2 I- Sf802lT O E B 'ttE IFS '812J- 66'1221F S '9 2 2 t16'Z221IB'EEBJS8'I'E 2I8 I'9 E 2 J02'6E 2I9 9 'O L I­ O S 'lt-2l6 5 '6fr2lZH 8696 E [S J6 3 1 U I 127 R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. ZILIBZ iejfia;ui -128 - R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. ppm -o O 6 !§52 I N t O I B I i • in m «* « in x Q. ) m o © O ID X O < r ft © V 3 « 18 - SIS: IKiSoS i (M - tn •* ! i §1 a C *-> i S S o s ssIS; sal _l I i t i » ............ _i s O 6 UU I >•**m S 3 S 3 S 3 S■?3 : a. x n p> o s S«n • at in [8*1 in in ■w ■r-. dSUS ------ *r» ^ ^ ^ °s°s CU OT ID 01 ■ OT O OT S ;*this3S S P ih E i- o o o o m O OT > a *11 Methyl (4-0-allyl-3-0-benzoyl-a-L-rham nopyranosyl)-( I —4)(methyl 2,3-di-0-benzyl-a-D-galactopyranosid)uronate ( 10) {5? ‘as!so§0g o» © "S8oot o « I »4 I! o° 4 Ol • £ i— i i i NO —cn m —v O) Q. o X -129 - R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. ppm/cm 1 1 1 1 1 4.0 1 1 1 1 I 4.5 ' I 5.0 ' ' 1 ’ ■* ■■ I ■ -130 - R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. 1 5.5 ppm 1 1 1 1 * ' 1 ' [>00r ® »d»f H2 / c » 54.6 4 8 1 2 F1HZCM ppm /cm H z /c « t 0.10200 51.60876 F2HZCM Hz i --------- — 1 . U. U. U. ll. I fipphcm j o ■“ ix w Methyl (4-0-allyl-3-0-benzoyl-a-L-rham nopyranosyl)-(l —4)(methyl 2,3-di-0-benzyl-a-D-galactopyranosid)uronate (10) ii aoo aCJ g" I a § a. X a • ; rn 3.5 to iri o § o S 5 in ■ • o j • tn tn tn ■o i n cd m 904.92 s (b a. 0.17222 >• u ln So f eof 5h S ° § ° F2PPMCM ~ 5 £ F1HI i u n n « X 3 fe *** a m«!5!!loS0m-0 - in“ ,S'-"“ -lisiSp:s » so a gnR^ S ' ' s Methyl (4-0-allyl-3-0-benzoyl-a-t-rham nopyranosyl)-( 1—4)(methyl 2,3-di-0-benzyl-a-D-galactopyranosid)uronate (10) 69 * ' 8t , 989 20 6 i2 '9 S E2*'89 £08'69 6ZE' 0Z 2EI'E£ 208 E£ 860'*£ t0 8 '9 £ E2I a £22' L L 8 *9'££ IE I L i (-06 B l I0t>'66 E /6 'tO f- tQ 6 '9 ttET6Z210 0 0 '8 2 t- 01t '82f18982P 9Z 8 '6 2 I86E'0EP162'E Et9 E 8 * E rE 1 2 8 E IESE'BEl- = 98*091J* 6 '8 9 l- uidd -131 - R eproduced with perm ission o f the copyright owner. Further reproduction prohibited w ith o u t perm ission. E^6ioi ii I uxu"" o Methyl (4-0-allyl-3-0-benzoyl-a-L-rham nopyranosyl)-(l —■4)(methyl 2,3-di-0-benzyl-a-D-galactopyranosid)uronate (10) !S 6 '9 U - E t6 ’i e i 000'82! 8i reel 189'82! 9 £ 8 '6 2 IB 6E 0E I!62'E E I9E 8> E I- X uII £ ! 2 '8 E tESE'BE!- X -uo 98*991- X !* 6 -8 9 t- 8 o " uidd - 132 - R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. eat’ 89 ZSB'69 Methyl (4-0-allyl-3-0-benzoyl-a-t-rham nopyranosyl)-(l —4)(methyl 2,3-di-0-benzyl-a-D-galactopyranosid)uronate (10) 6ZE0Z 2EI'EZ 298 EZ 960 K SEE FZ 0T6 SZ 108'9Z EZVLL EB2' ZZ 9211 LL BF9 ZZ IEZ ZZ F06 BZ u (01- 66 - < CJ E Z 6 I0 1 uidd 133- R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission U Methyl (4-0-allyl-3-0-benzoyl-a-L-rham nopyranosyl)-(l —4)(methyl 2,3-di-0-benzyl-a-D-galactopyranosid)uronate (10) 6t'9(r'9I- O (J E989'2S06Z2‘9S- o 8 2 2 fr'8 9 ► Z S B '6 9 26ZE' 0Za E l ’ EZ9 1 9 8 ' EZ6^60 O S E E't'ZB606QZH 0 B ‘9 i *E2 \ L L 9E2Z L L 1921 r u ­ ts ° o b * 9 /:z El tVLL- W6’Si­ lt OF 66uidd - 134- R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. (A fe n c w & 1 Ssl s §S|s I s I i s E B w n ii i n p» S o o ~ i | § § l # ~ 8 g o ° s a S £ ! S n O EL •» " S ’" " ! w *r u in o r«. t_ a. ^ 3 5 ru Methyl (4-0-allyl-3-0-benzoyl-a-L-rham nopyranosyl)-( I -*4)(methyl 2,3-di-0-benzyl-a-D-galactopyranosid)uronate (10) PC SSB GB LB SF NOW SI S F0 1 TO FIORE SM £™ g° •»J E 6 6 6 e iulm nm iiniiim innm nm nm nnnm nnliiiiinuiiim num m im in s - =B ISdj-o CVJ C J U U .U .U .U .U .U .U .U .U .U .I iHn n j i i u l m m o E'VE'X |8 8 S i ? S S S 2 S ! f f i S 5 g S Q. ininiiiiiiitiliiiiuiniimiiimm miinniiiiiui p | § s a s 85 a s a s a a a s a. tn NDO E u £ £ iiimmnnnuiHHtillim II U < o N X O o U rt _C a: N I vO o CQ - 135 - R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. -136 - R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. ppm Methyl (2-0-acetyl-4-0-allyl-3-0-benzoyl-a-L-rham nopyranosyl)-( I —4)( I -0-acetyl-2,3-di-0-benzyl-a/p-D -galactopyranosid)uronate ( 12) < u 137 - R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. Methyl (2-0-acetyl-4-0-allyl-3-0-benzoyl-a-L -rham nopyranosyl) (1 -0-acetyl-2,3-di-0-benzyl-a/p-D -galactopyranosid)uronate R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. ppm 5.7 5.6 5.5 5.4 5.3 5.2 Methyl (2-0-acetyl-4-0-allyl-3-0-benzoyl-a-L-rham nopyranosy!)-( I —'4)( I-0-acetyl-2,3-di-0-benzyl-a/p-D -galactopyranosid)uronate ( 12) 5 .0 ppm -139 - R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. - 140 R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. Methyl (2-0-acetyl-4-0-allyl-3-0-benzoyl-a-L-rham nopyranosyl)-( I —4)( I -0-acetyl-2,3-di-0-benzyl-a/P-D -galactopyranosid)uronate ( 12) (!) O -141 - R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. t •>. 2 £ O C rt C o E c f* -i o y -io>s jcdS OW ) cm Qi n o -r> s C v ' from EtOAc "rt *^p 60 ;H S 3 u -r •5 »g£ iSSS 01 a 'S"®' i• nn n©K J*o £U ! S S2 ■ as s a “ S S R !! ss cs KftSSSSS ttl tO 03 iSihggsi r ■^ > 01 ^ h. f t q • • • • 3 o&£s»£3dos££«aili§ sU sIiU a <4.U £U s. U .Ii.L.U.ll.ll.U.lLstL -143- R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. 90fr‘« Z33’8( 800 (3 5*3 (3 ^ 0) $ sc O g e I* E c rt _c b V S: A2 5 « S' a c d -S S- o ? I? &■ _>> § f$ -9 6 o T7 $u < n -4. 9 6s o s E I G 9 £ '£ E ! I06Z E I 9Z1BEI IZ E 'B E ! O« H 2G 9I2I8V 9IIE fr'8 9 1 8 Z 0 '6 9 !— 9 * 9 '6 9 ! - '/ O JL 8Z9 6 9 W uidd - 145 - R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. Methyl (2-0-acetyl-4-0-allyl-3-0-benzoyl-a-L-rham nopyranosyl)-( I —4)( I-0-acetyl-2,3-di-0-benzyl-a/p-D -galactopyranosid)uronate ( 12) 0501?' H 0Z 2 2 8 1 2 800'12- 55*2'J2- J■ i ! _o _o _o 17126'25- _o ID 6E09' 09ZEES '8 9 - oosroz0996T Z 17955'2Z SBTfr'EZESS9EZ52l7tt7ZS6B8'(7Z02SZ'5Z9 2 9 8 '9Z1822 ZZ6fr82ZZlOlZ'ZZBVWBL1919'BZt7t7l6' 08- — O _© _o 98E 9'06Z 25ZE6JBEE'66- uidd -146- R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. ZEES'8 9 - OOSrOZ-^. (N Methyl (2-0-acetyl-4-0-allyl-3-0-benzoyl-a-L-rham nopyranosyl)-(I -~4)( I -0-acetyl-2,3-di-0-benzyl-a/|i-D -galactopyranosid)uronate ( 12) 0 S Z S '0 Z - \ 0996' « S S 2Z1991 ELSBlfr'EZES99' EZ" F6EZEZSEU'frZ20 V S ' U 0B6Z' bLS688>Z02SZSZ9298'9Z1822 ZZ 6('82'ZZ IZIS'ZZ 10IZ'ZZ Q _. 8W8Z _o J9I9'8Z W6'0B _o O < 98E9'0 6 - Z2SZ'E6- o _o ^ 18EE'66- uidd 147 R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. ppm -148 - R eprod u ced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. ss • ? S g S ;5 S s s g s “S5og>!s-sg< S§ 2' § n ni in o u « milfl *r o Md | l 2fssg II n i as is! 2 o s u e t0) ia~ g > - .8 S * * imri S. u s . u | s s Is Is Is Is Is Is feSSRS5I8^SoSi^S™ S a iri tri ""I ® • cu ■f* • 5 p. 8 cy nj m *«s ito n x N C t. a W^ IL 4^ g ^ IV w si^sseg g iS saS sS ssisS IessI - g Sea uni iM iiiiniim iiiiim iinintiiiniiinnnininniiiiH iiliiinniniii.niinim iiiiiiuiin ausllsss nt m m l£»i§§38 Isis Hal£3SS^aSSi'8lSI aim 83 _ ... o ~ . U. U. U. U. U. I m .iiiiliiim niim m iniim iniim iim iiim iiiiiiiiiim m iiliium m iiiiitm m im iiiiim i -149- R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. 1 fe S ™3- a 2i S$ I £ i£g ^a“ss^^ss?a~s >S r a n o™ o9i n»•mfli h'T o• Si o «■* O ;s §? £ £ « Q ^WM ni a . m ; n o c !S£' a o a tt 8t£g£g£ ^ B-“ S^°S^5^$”81^SS2S *j2ScyuioSoia3cnoS^(^S i e 6 i£ 3 5! § £ LOonjio uioi'T 'tom ictnoiv al i § 5» iI hS SH e!« IR ' cs s iS s ss a S ss s & S eg g ;! ! i v a i l l ^a i iS u x *&3i - in u. to t> U U. ll. U_ L . L II k k L VL* 9- n i n i i l u m n i i j i i [ i m i n n i ( t m m n i i i i i m i i m m l i i i i m n n i i mi ii L...... nt tliiiiiniinm iinm iim iim im iniiiiiiiiiiiliiiuiiiinnninm iim iiinuiin)>nm Zl ZH -152- R eproduced with perm ission o f the copyright owner. Further reproduction prohibited w ith o u t perm ission. Methyl (2-0-acetyl-4-0-allyl-3-0-benzoyl-a-L-rham nopyranosyl)-( I —-4)(2,3-di-O-benzyl-0t/p-D-galactopyranosid)uronate (13) ZE'gEEI- ee’oset6E9E‘ t eoiot't1 9 'H t 'l EOZW 9SE2H - 2rB2HS f S E t 't - 8 f W 9692‘8 - 62 09FI- 6rom6 s ' 8 8 H ' 8 2 '2 tS f96E 191- 0Z'22SI~ 2£t'2Sf16J531UJ ZH -153- R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. erofroi ot-2zo 09 6NU 90 6901 t'6'BZOt r3 < •g^S rStfie oo rv oo ^ >CT>to 'V 8$ ro o ru n i i n n o o o i - 8' iiS' 3 i 3g 8 gf 8 •SCI e tn f • «n o s ?>f\Jash> fs. f\l S cn t »22in8«Si(>8' sag o o o o aafe8g°go cn oo *m-< o c o« ■n R 95.68081 Hz/cm S i • n 342.7B Hz 0.3207B pptn/cn 96.27525 Hz/Cffi i d SX O O OS ti!Ssix!oa!oi!i!h(AR; sM ie&irs sfe!1 is* iStelS1 J--k—L J J_ I I I .1 t I I J -* F1PPHCM O F1HI F2PPMCH F2HZCM 0.31946 ppm/cn F1HZCM fi t in i i i i ■ i i i i 1 1 J flCO fN 1 CL -I O) SB > je ro Q. •o * 1 1 1 1 ■ 1 I 1 1 1 ’ 1 ’ 1 v 5 CL ’ I 1 1 ’ ’ 1 1 1 1 1 I 1 1 r 1 1 ‘ 1 * J I' 4 23 1 ’ 1 1 ‘ 1 f 1 > r a Q- c ’ 1 ppm fs W -157 - R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. 28EH Sfr8Z i l£ 2 '9 I EJ^’91 £ Methyl (2-0-acetyl-4-0-allyl-3-0-benzoyl-a-L-rham nopyranosyl)-(I —4)(2,3-di-0-benzyl-a/p-D-galactopyranosid)uronate (13) 8t?2't2 ZZ9 2S OOZ'25 SE8'25 ES9'09 921?' 89 801? OZ 9ZS0Z WEZ 06EEZ EOZEZ EEE ’ I’L 226'9Z 190 ZZ t?l?E ’ ZZ IZZ'ZZ 6ZS8Z 52E 26 s 7 Z W '26 - f W66 6Z8' 66 - / Z 2 6 '9 tt—\ 9 8 f 'Z H - \ t?08’Z 2f-, O t6 'Z 2 t—\ B tS '8 2 t- 0 \ 265’82;-^\ 219 B2t SZ9 8c i fEZ Bet E69 621 I8 Z 6 2 ; 5B; EEl 2^8>e; 222 BE; 822 5 9 ;09E'G9l; 8 S S 9 ;E S 0'99;5t?9'89;— —= 6 2 E 6 9 \ - y /[ r 8t9’69;-77 J E 89'69; n s ;z ;- - 159 - R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. Methyl (2-0-acetyl-4-0-allyl-3-0-benzoyl-a-L-rham nopyranosyl)-( I —4)(2,3-di-0-benzyl-oc/p-D-galactopyranosid)uronate (13) tog-ftx 016' /.SI 8lS’88r 865’88t 8 f9 'B 8 t SZ9'88I t’Ei'881 E69 681 \$LSZ\ Z86'68l SBPEEl at-'Bt'Et 881'BEl 888'8EJ fSB’BEI 888’59W 0 9 E S 9 1 -A t8 5 ’S 9 t - < \ ESO'991-------- Sfr9’89t-^_ 68E’6 9 t— -^= 8 1 9 '6 9 I - / £89 691- / A ttS'Ut-' O CO ujdti - 160 - R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. Methyl (2-0-acetyl-4-0-allyl-3-0-benzoyl-a-L-rham nopyranosyl)-( I —4)(2,3-di-0-benzyl-a/p-D-galactopyranosid)uronate ( 13) -8 EZZ9 5S 9669’59 S*EB-2S t'ES9 09 E95^ B9 2251 0Z L LO t OL ZSZS 0Z EB6S IZ 8£Bt EZ 968E EZ OEOZEZ IEEE kZ 8686 SZ Z1569Z 6090 ZZ 9E7EZZ ZOZZ ZZ fr6ZS BZ 6861 08 ESBE'56 86 9158 Z6 S5E6 Z6 585 r 66 ft^E 66 l?659' 66 98ZB 66 -161 - R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. »«N 3 S i 2 in o o o t l l l i r 2i Sg II a i z a °in ^3R?S' m in • o «-• £ ) 5T O CU ii io o 5 r m o a is assess l oV oo uz c § ° g n: o(vj oo «<-< o in in ; 3s "■".1 R ■ r». in n o iM v o i 3 I^SESgSSsaSffftSSiHfe! ! S 8 i S S fc ri S 5 £ I 5 £ £ 5; 5 U U U - U . U - L i . I l U . U - U . L l U . I l U. ■o1 CO in CO X? O O Q. -162- R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. 2 :. iS 3S 1X 3 o- a 583! ij 8 S "it *Si "5 a cd i ! P s< *28:S< : « i~ K l I S g g£ S2 S£ g £ g £ i t o o ' c f f i n m o " " ID CD r a mu a o o m n o i a i ' t o ** cri in * " ’ *** *" i cmo a < S=S! °J- fi 3 ;! 1 i § 5 a> <8 B IT1 § 13-sl a o a. a a !B§SggsI §p ! cs ssiIS sss s^isiSss : isf 3 £ S3S SfcSj! h a n c i in CL HHiitniiitiHiiiliiiiininniiiiiHiiilinininniiHiiiiimniimiinn o in mim iiini nnmim mnmlmmiiiiinimm liniiHiiiiiiifuinumiiiiuiumiiiiiiHiiniiiiiiiiiininlm o ■o CO cl o (N CM -O -163 - R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. >o n o> « ■ rn r-i cr» i •0)00)' ' a o i n S J O fU I J „ l i o in m O o oi i • ■ (U I n in 0 (O C M< 1co O I rp s o o o o 01 i n • ** -i o i O) o o ss O CD o x !^5aS U. 3K [ v i in f -S P O ppm Methyl (2-0-acetyl-4-0-allyl-3-0-benzoyl-a-L-rham nopyranosyl)-( I->4)2,3-di-O-benzyl-a-D-galactopyranosyluronate trichloroacetimidate ( 14) iPa tHc ! 01 VI ID i iI w 33 2 : -164- R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. lejBaiui -165- R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. Methyl (2-0-acetyl-4-0-allyl-3-0-benzoyl-a-L-rham nopyranosyl)-(I —4)2,3-di-O-benzyl-a-D-galactopyranosyluronate trichloroacetimidate ( 14) 9 2 'S tS Ifr9'S2St- ;6'2^9r- 2ZS2E Methyl (2-0-acetyl-4-0-allyl-3-0-benzoyl-a-L-rham nopyranosyl)-( I —4)2,3-di-O-benzyl-a-D-galactopyranosyluronate trichloroacetimidate ( 14) E S 't't'S l- orogsi2Z 'l9 Q t0 [ ‘B95J16'OZSt- in 8 * '6 E 9 l- zr Est­ es'8^91t2'2S9t- 12 8 0 't fS 2 6 'f89t~ 01> 89T E 2'Z 89t- 0 9 1 0 'I 01 E2Z5S 0 8 2 Z It'9'EE Z t- 0000 t lE'OfrZt9 2 S fit69'0S Z l- [ej6a)ui ZH -166 - R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. Reproduced with permission of the copyright owner. Further reproduction prohibited without perm ission. Methyl (2-0-acetyl-4-0-allyl-3-0-benzoyl-a-L -rham nopyranosyI)-(I —4)2,3-di-O -benzyl-a-D -galactopyranosyluronate trichloroacetim idate ( 14) AcO-----f ' l bzo cn CM GO CO CO r-* nTT •n- CO in \\ 'jy \L O C (= N H )C C l3 co -tr m (MW (\l Ol GO ID CD TT CO ^ 01 cn CD CD CM r»?a o ;> 2• S>•!I 35 o ®jo ’ < A) « C Ol Cl I j° o pi v e Pv t*s ■ i-j a S ts a a = 1 iS S ““s ssa g O PI ™ o 5 .52 " S s S ni s a u 1 ^ i>lal* H i 3- s :1^5aSi!itls le ss; ui g 2 U C U E 1 c. £ s s Ol © oot e I&j5£‘ 3 « 10 01 o c. a. I UUI isgi o in LO iS£ iVkiSss in pv * n S5^3is^3£' u o u . u .u . i f e b .u . u . 1 ^ ^ 2- X- 0rt Q S '? (N (N o< 0 ^'«fn(U(p nftlM O NNW igS td u 3 < n ^ * R iK <3c3 (\i S nS S o cd i o o ; i. ?• 8 cn O H O C/1 in «n s Ife S i U O U .U .U .U .U .U .I in I • tn ~ - a? 3 S S af 3 £ ! IDiSfflSSS in O in 1 to I UD _J L_ - 171 - R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. • rs Q o r*» >C(=NH)CCI, -j ro « o in n id tn id m oj tDNCDrrNCONtDtO^ailfl^ r n c n c o a i s i n i n ’T '-'O tN rsO ! T O ' T C V J O J C O l f i C O tD cn^r^iD fn»-«oj WCDM^tflWCOM oo j n c m i d ao rr m t co ^ at cn in ’T mtn {ji v in oi 21.035 18.246 Reproduced with permission of the copyright owner. Further reproduction prohibited without perm ission. Methyl (2-0-acetyl-4-0-allyl-3-0-benzoyl-a-L -rham nopyranosyl)-( I —4)2,3-di-O -benzyl-a-D -galactopyranosyluronate trichloroacetim idate ( 14) Reproduced with permission of the copyright owner. Further reproduction Methyl (2-0-acetyl-4-0-allyl-3-0-benzoyl-a-L -rham nopyranosyl)-( I —4)2,3-di-O-benzyl-a-D -galactopyranosyluronate trichloroacetim idate (14) OC!=NH)CCl3 to to <1 u> -aromatic- prohibited without perm ission. OCH ,CH=CH OCH,CH=CH c=o c=o benzoate ppm 190 180 170 ,OC(NH)CCi 160 150 140 130 120 110 Reproduced with permission of the copyright owner. Further reproduction Methyl (2-0-acetyl-4-0-allyl-3-0-benzoyl-a-L -rham nopyranosyl)-( I —4)2,3-di-O -benzyl-oD -galactopyranosyluronate trichloroacetim idate (14) O C (= N H } C C (3 t o id cn -ri cn r to m m cn -q* prohibited without perm ission. G a iA C - 4 — r~ ppm in GO in in id Rha C -4 o m (V CD G a iA C - 3 G a iA C - 2 (D C H , Ph R ha C -2 GalAC-5 RhaC-3 R ha C -5 T 72 T 70 T n { ----,----,---- 1----(---- 1----r 68 66 Reproduced with permission of the copyright owner. Further reproduction prohibited without perm ission. Methyl (2-0-acetyl-4-0-allyl-3-0-benzoyl-a-L -rham nopyranosyl)-(I —4)2,3-di-O -benzyl-a-D -galactopyranosyluronate trichloroacetim idate ( 14) O C ( = N H ) C C l, tD S. ru B in ^4 Ol 0C (0)C H . CO, CH. Rha C-1 GalAC-l OC(NH)CCI I-------1-------1------ 1-------1-------1-------1-------'-------1---- —<------ 1------->—-----1-------1-------1-------1-------1-------r-'--- T----^ -------1 ppm 90 80 70 60 50 40 30 20 10 *=' a «cn oo o^ o• ai to «: e oi** o^ *> llsIMggs* • 8 £ a » t n t f l i N J) aOj T( n3«U< 1n 3x X W33SX « ^ * ^ ai m ,g“ !S"KS^^J8SS,,,8 -iNfvincpotntnin •g «•* s“ 58 28 ru «« o c i= is rs g ^ s rs fffe s s s ftle s s s -i £ o J g ° § • m u r . 01 z : r* in cd a ¥ 1 S g 8 2 82 2 2 22 22 22 "CuoinfflTKjipffljjn < aoo• o• ^ ru <- «** - . . in in o • »* ■ r> _■ c v hi j g 8 « o S « 5 o o ni rf S8S o Z t Eh: sk ils s s ■!fl S Ul _| (fl -176- R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. ;KS8 -U .ll.li. (D 1VNi/i 0X 3 b gnj in 3 in s *x zn ii n— rt 2 w 1 2 M = g 1 2 S :? i t o c "V e § £ 1>- 8£ O s ? nS' je "%L ^v> NQ I * S' &1 A rs 1 ? cn O Sr rt 5 H> § 23 rt § vvl S ro 4_!_ * oc n&J" I 9 6 i9 E l- v S£. 1 0 0 6 E I9 9 2 6 E I- fe 0 0n _rtu ■ 0 n.o 1 9> B 9 ^* §O JS ? * 0 fr c . S i S 82 JS y -% ■L— rt J ?> !£ 9 ^ 5 " i.o =* c^ S r t a; - i - -O i .J. o -u S'! "i "7 41 m ^ S « 21 ^ O. « « S -5 s> r^ .« o E i V EHT9 H I- W wggn- I U cT u 5^96 9 9P' 0£H61' u\\L f gznL f 09H 9S' E e n 981J- IS' 99^0 L OEU I S E 621- 9020 0 teuton *H -180 R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. >s J_ X ^ fS rare BZOOi ;ra*9 in "cn tej6a)ui -181 - R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. 680 21 i 2. f iSs1 d »- o x o ISsiSESBDiieSSS o S o SPDsCJ itisSi JitfS^igE U U U .lk lk lk lip tk U .U -l

-Q Q V B ro i 61 ftc ro a) i-1 rt m -c- -° 6 5 - M 9E5 2 9 92Z'29 0 * 2 '9S 529 09 * 9 9 '0 Z 52Z2Z 6 9 9 'E Z * 9 Z 'E Z 5 E 5 '* Z 6 Z E ’ *Z B 00 9Z 8 9 8 '9 Z 282 U 0 9 9 ZZ ZO Z'ZZ « s m 4- i -s £ (N » ■ — . ■ — ( Y 1 (II O rlP oo < iudd —Cl -186- R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. i i0 iV ^ , t/i ~~ 5 'cT &■ § £ 2 § 2 g22 = F 8 8 -E 0 V c § i v e .c >* 5 I vi- 3Q, ^N _rttj -rr ? S rt o ^ 1 *> N Q ^r oI _j_ go l i 4 1IE* ^ m >'—(N a; z: uidd -187- R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. O ,u------------ S 9 e re s LVZL'Z9- 5" O Z6E2‘99- ^ (U a* a I: S 1 2 ^ g 2 2&"0 i i gp sc "■V cS -c§ ca? & V 92 5 a ? * ?> J8 9» o S9 6I 9o n 3 ro i i . &-r0o co&• rt s ^ 4-> (U _C s: 1 (N 0129 09 tS2E 'B 9-\ 2069'8 9 - A mrOZ^NV H?990Z 9919 OZ 6690’2Z 602Z-2Z 6899 E Z ^ V 0fr9Z 'E Z -\V ~ 0 IE ffZ --O = 88ZE"VI ZZ00 9Z 6ZZP9Z KU'SZ ^ 8 9 8 '9Z £282' L L —' / I r~ 6699' LL- > I 0L 0LL L -'/ 029^'8/-y 09fr9'8Z-' t'ZSE'EB C909 96 962^'86 E6Z6 86 ZI?9E'66 uidd -1 8 8 - R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. Methyl (2-0-acetyl-4-0-allyl-3-0-benzoyl-oc-L-rhamnopyranosyl)-(I —■ (methyl 2,3-di-0-benzyl-a-D-galactopyranosyluronate)-( I —2)(2-0-acetyl-4-0-allyl-3-0-benzoyl-a-L-rham nopyranosyl)-( I —4)XI m^v*rf>fnr*.to S>OsUl s i i s l s i S £ t^t2220jS -C c30inS'^,ri00r' T r^(D lD fy in so" S 0) N (A ® ^ v r - *u r> *» < — KO ■ Z3 O l r ; S S 5 8 a 8 g ^ _ rS ~ =_ ss iri o n o S n h >0 I s ^SulSsi^SaQ ■ S! Ss ikQl OOU.U.U.U.U.U.U.U.U.U.U.U. ^ k ^ r '? X (/I rJL -.. t <0 o « £ " § 00 " S ' <2 >: Cd S: g I i 0 2 c 2. M 5o: 2« S E ° C rd ■i 2 m1nln1111111mjunmiiiiiiiiiiiiniiniininiiiiLimiiiniunimnmiiiiiinimiii......iilimmmimmnnt... 1111mimn11ninnln11111immnnnmt .liMiiiiiiliiiiuiiiiimiiiHimiHiiniiii 80: ? fc >> 2 s S 1 f. o ' *8° n 9■ Xs 9B a •*? _l ro <5> o fr n § X ro S -1 J O u N M I LA

k rS o e —o r ^ 1 - 191 - R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. o id ■'T t Q^ o -g vt O © C N ta w < eu ts: ss A c 0X ^ i>«»f* f*■* 5: - u °*■" ^-5 r~ d rt £c2 9 • > I p s 1 J3 _ O O - 192- R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. 0000 t DI e n E O- £ 0 C rr 1 § I -E 2 ^>v4A o£ ° X >L ? M Q. e t o -C 4-1 rd u> >v cd ' *-> 0 'oT iw) tE rt *^ Oi’ L SLd ®b fC ^ £ c 'T' >-k O C 0 Tt 2E __ -cL > i i *55 v— 1 0c rt O V i. fe >. I w S —% r /-» . nl rC fl; M <« «-» D« I l f ill i■£l£lw ,2 ^ 5* nCL (0 BO £ 0 9 '9 9 E ~ EQ^gZE6 0 9 ' PIZ- 829 08E- vO ± T E 6 '8 B £ - OWOO^nctov- | x \KLW- ppm V° — ZH 196- R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. OMe uidd -1 9 9 - R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. 'T t 1 a oi X v> O c re i_ E, f e O c 'T t c a . c O m 0 £ i. ??* £ o c E ? S : o u r3 & W6$\- Q 60t OZf- c 2 _J i QJ -C tt rt _£ cs^ E L3 . i _i l/> o o i a 3 X 0 1 & ">> £ _£ & r rt s_ fe O uidd - 200 - R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. 5. o S H ’SS D A, £5 ^ x1A — o o P* c fS S £ ® >L * , L. 0 X CL « ^ O ™ T > v 1 4-> £ — U J O ST ' 2 C CL &© c ^ i C +* RJ JS D t € § £5 “H 3 *>s _c 00 _ o © > s C £, ■§ c *- _>> oc E“! ■< B DC w H o J2 rt “ CM DO “ in 0SS9 9Z ^El^B ii CE6EBC uidd - 202 - R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. tEQS'El- j_ DX A >> £*• C b w a> ^ to ^ ■r O O 0 8J6’9 f9 ^ 6 '9 t - U U- E c o T IX A s -J O >-A 5 o 'r' fc r£a ^t 9o & Lu 0 V-g £E 'aT *•>o cd ^ Q 1 e * ir. si - s? 99E9'6 8 - ■is- ec ^E z*- o^ (4 oo EEgB'g? X9£SLt getg-/;^ UOB IV 8600 81' BS^E’Bt' 9SS9 8fr 33^9’19 188 r i g Mfrrgs -203- R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission. I ISSSoS^! mmmi 6 SSg£i jg S uses s Methyl (a-L-rham nopyranosyl)-(l—4)-(methyl a-D galactopyranosyluronate)-( I —2)-(a-L-rhamnopyranosyl)-( I —4)(methyl a-D-galactopyranosid)uronate ( 20 ) , U W M u (U (U N ! Sg £ £ S : « o a i o ( in « 5 — >K r iS lesas leScs mo o (V ID Q ' aoo^r fSgg s " su o s j cu o o s s ii I n |3 £ g |||i li- U. U. U- I uiiiniiiiiiiim im iim nium iim im iim nillilillllllllll o <=5 €»