Advances in Carbohydrate Chemistry and Biochemistry, Volume 50

Advances in Carbohydrate Chemistry and Biochemistry

Volume 50

This Page Intentionally Left Blank

Advances in Carbohydrate Chemistry and Biochemistry Editor DEREK HORTON

Board of Advisors LAURENS ANDERSON GUYG. S. DUTTON STEPHEN J. ANGYAL STEPHEN HANESSIAN BENGTLINDBERG HANSH. BAER E. BALLOU HANSPAULSEN CLINTON JOHN S. BRIMACOMBE NATHANSHARON J. GRANTBUCHANAN ROYL. WHISTLER

Volume 50

ACADEMIC PRESS, INC. San Diego New York Boston London Sydney Tokyo Toronto

This book is printed on acid-free paper.

@

Copyright 0 1994 by ACADEMIC PRESS, INC. All Rights Reserved. No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopy, recording, or any information storage and retrieval system, without permission in writing from the publisher.

Academic Press, Inc. 525 B Street, Suite 1900, San Diego, California 92101-4495 United Kingdom Edition published by

Academic Press Limited 24-28 Oval Road, London NWl 7DX International Standard Serial Number: 0065-23 18 International Standard Book Number: 0-12-007250-5 PRINTED IN THE UNITED STATES OF AMERICA

94 9 5 9 6 91 98 99

BB

9

8

1 6 5 4 3 2

1

CONTENTS PREFACE .................................................................

ix

1991 Robert Stuart Tipson, 1906 .

D. HORTON Text ....................................................................

...

x1u

How Emil Fischer Was Led to the Lock and Key Concept for Enzyme Specificity RAYMOND U. LEMIEUX AND ULRIKESPOHR I. I1. I11. IV. V. VI.

Introduction ....................................................... Asymmetric Induction .............................................. Yeast Fermentations and Enzymes .................................... TheLockandKey Concept .......................................... Insights on Enzyme Specificity ........................................ Concluding Remarks ................................................ References ........................................................

1

2 7 9 13 15 19

Anomeric-Oxygen Activation for Glycoside Synthesis: The Tnchloroacetimidate Method

RICHARDR . SCHMIDT AND WILLYKINZY I. General Introduction to Glycoside Synthesis: Activation through Anomerio OxygenExchangeReactions .......................................... I1. Anomeric-Oxygen Activation: Anomeric 0-Alkylation .................... I11. Anomeric-Oxygen Activation: The Trichloroacetimidate Method ........... IV. Other Anomeric-Oxygen Activation Methods ........................... V. Conclusions ....................................................... References ........................................................

21 23 25 114 116 117

Synthetic Reactions of AIdonoIactones ROSAM . DE LEDERKREMER AND OSCAR VARELA I. Introduction ....................................................... I1. Acetalation of Aldonolactones ........................................

V

125 125

vi

CONTENTS

111. Acylaton and Etherification of Aldonolactones .......................... IV. Reaction of Aldonolactoneswith Hydrogen Bromide ..................... V. Chain Elongation through the Aldonolactone Carbonyl Group ............. VI. Reaction of Aldonolactones with Alcohols .............................. VII . Reaction of Aldonolactones with Ammonia and Related Nucleophiles....... VIII . Reduction of Aldonolactones......................................... IX . p-Elimination Reactions ............................................. X . Synthesis of Deoxy Sugars from Aldonolactones ......................... XI . Glycosylation of Aldonolactones ...................................... XI1. Aldonolactones as Chiral Precursors for the Synthesis of Natural Products .... References ........................................................

132 134 136 148 151 157 162 170 179 181 201

Molecular Structure of Lipid A, The Endotoxic Center of Bacterial Lipopolysaccharides

ULRICHWHRINGER. BUKOLINDNER.AND ERNSTTH. RIETSCHEL

I. Introduction ....................................................... I1. Lipid A Definition and General Properties ............................. 111. Primary Structure of Lipid A: Backbone. Polar Substituents. and Fatty Acids . IV. Synthetic Lipid A .................................................. V. Conformation of Lipid A ............................................ VI. Endotoxicity of LPS and Lipid A ...................................... VII. Serology of Lipid A ................................................. VIII . Synopsis: The Structure. Activity. and Function of Lipid A ................ References ........................................................

211 213 214 252 252 256 258 260 265

Developments in the Synthesis of Glycopeptides Containing Glycosyl L.Asparagine. L.Serine. and L-Threonine

.

HARIG GARG.KARSTENVON DEM BRUCH.AND HORSTKUNZ

I. I1. 111. IV. V. VI. VII.

Introduction ....................................................... N-Glycopeptides ................................................... 3-O-Glycopeptidesof L-Serine or L-Threonine........................... Binding of Glycopeptides to Proteins .................................. Solid-Phasesynthesis ............................................... Enzymes as Tools for Glycopeptide Synthesis ........................... Addendum ........................................................ References ........................................................

217 278 287 298 299 303 306 307

Physicochemical Analyses of Oligosaccharide Determinants of Glycoproteins

ELIZABETH F . HOUNSELL

I. Introduction ....................................................... I1. Methods of Structural Analysis .......................................

311 311

CONTENTS I11. Purification and Profiling ............................................ IV. Backbonesand Core Regions of N- and 0-Linked Chains of Secreted and Plasma Membrane Glycoproteins............................................ V. Peripheral Substitutions of N- and 0-Linked Chains of Glycoproteins ....... VI . The Conformations and Molecular Recognition of Carbohydrate Determinants Distant from the Protein Oligosaccharide Core of Glycoproteins ............ VII. The Conformations and Molecular Recognition of Carbohydrate Determinants Adjacent to the Protein Moiety of Glycoproteins......................... References ........................................................

AUTHOR INDEX FOR VOLUME 50 ........................................... SUBJECT INDEX FOR VOLUME 50 ........................................... CUMULATIVE AUTHORINDEX FOR VOLUMES 46-50 ........................... CUMULATIVE INDEXFOR VOLUMES 46.50 ...................................

vii

314 315 325

332 343 345

351 378 386 389

This Page Intentionally Left Blank

PREFACE With this fiftieth volume the Advances reaches its half century, during which time the carbohydrate field has evolved dramatically across a broad range of formal scientific disciplines, both fundamental and applied. Over the years the chapters in this series have chronicled these developments for the benefit of the general reader and have concurrently provided, for the specialist, important critical insight into gaps in our knowledge. The rich legacy of the early carbohydrate literature remains a fruitful resource in addressing new problems with today’s superior tools of research. Lemieux and Spohr (Alberta) here trace our understanding of enzyme specificity in broad perspective as they assess Emil Fischer’s “lock and key” concept advanced a century ago in relation to current ideas of molecular recognition. It may be noted that the very first article in Volume 1 of Advances,by Claude s. Hudson, was devoted to the Fischer cyanohydrin synthesis and the consequences of asymmetric induction. The task of interpreting chemical transformations and the logical planning of synthetic methods have been traditionally difficult with the carbohydrates because of their polyfunctionalityand complex stereochemicalarchitecture. The vast body of empirical literature is daunting to the newcomer to the field, and the synthesis of glycosides by endless permutations of the traditional Koenigs- Knorr synthesis presents especial difficulty. A major step forward has resulted from the insightful thinking of R. R. Schmidt (Konstanz) toward the rational design of practical and versatile methodology for glycoside synthesis. His trichloroacetimidate method, here surveyed in comprehensive detail in a chapter with his colleague W. Kinzy (Basel), constitutes one of the most imaginative approachesto an important problem in synthetic methodology. Their chapter will undoubtedly comprise a key reference source for numerous researchers for many years to come. Although the use of abundant sugars as starting materials for chiral synthesis has received considerable attention, the ready availability of many aldonolactones is less well recognized by “mainstream” synthetic organic chemists. The chapter here contributed by de Lederkremer and Varela (Buenos Aires) provides a comprehensive overview of the practical potential of these cyclic esters and complements the more specializedcontribution on gulonolactones by Crawford in Volume 38. The biomedical importance of infectionsby gram-negative pathogensand the consequences of septic shock have drawn much attention to lipid A, the toxic subcomponent of the lipopolysaccharide endotoxin of these organisms. A comprehensive account of the chemical structures and biological behavior of the lipid A structures is presented here by Zahrihnger, Lindner, and Rietschel. The chapter incorporates much of their own work from the

ix

X

PREFACE

Borstel laboratory where Westphal animated his pioneering work on bacterial lipopolysaccharides. The sugar - amino acid linkage point in glycoproteinsand proteoglycans, involving the side-chain nitrogen atom of L-asparagine or the hydroxyl group of L-serine or L-threonine,is a key structural region of these glycoconjugates. Garg, von dem Bruch, and Kunz (Boston and Maim) now survey developments in the synthesis of glycopeptides containing these linkages, updating with significant new work the earlier reports in Volume 25 by Marshall and Neuberger and in Volume 43 by Jeanloz and Garg. The final chapter, by Hounsell (London), also relates to an important aspect of glycoprotein structure, namely the structures and shapes, as determined by physicochemical methods, of oligosaccharide determinants of glycoproteins that are antigens and targets for binding of adhesion molecules. For most of its existence the Advances has been guided by the breadth of scientific insight and editorial expertise of two individuals, M. L. Wolfrom and R. S. Tipson. The obituary chapter in this volume records the life and scientific work of Robert Stuart Tipson, who contributed a chapter on the nucleic acids to Volume 1 and retired as Editor with the publication of Volume 48.

Washington, D.C. April, 1994

DEREK HORTON

This Page Intentionally Left Blank

ROBERT STUART TIPSON 1906- 1991 Bob Tipson made contributions to carbohydrate chemistry during more than six decades and exerted a far-reachinginfluence on the published literature in the field. Although he was born and educated in England, his subsequent career developed in North America. He was associated with Advances in Carbohydrate Chemistry since its inception in 1945 and served in an editorial capacity with the series until his death on July 13, 1991, near his home in Kensington, Maryland. When Robert was born on November 23,1906, his parents were living in a rural hamlet named Wadshelf in the county of Derbyshire. He was the first of their five children. His father, Herbert James Tipson, was a teacher, as was his mother, Mary Jane (nCe Stuart).Their son did not attend formal elementary school and was tutored at home by his mother until the age of ten. Four daughters were subsequently born to Herbert and Mary Tipson between 1906and 1920;the second died in early childhood, at which time the family moved to Coventry, a nearby city in the Midlands of England. Herbert Tipson was an all-round academic with a flair for mathematics and the precise use of language, and he was also a gifted artist and a lover of gardening. For many years he taught at the Coventry Technical School, and his mind retained sharp focus on mathematical concepts even into his ninth decade. These attributeswere clearly passed on to his son who, like his father, received his secondary education as a day pupil at the Bablake School in Coventry, a traditional and highly regarded English boys’ school, admission to which required early demonstration of high academic potential. At Bablake School the young Robert Tipson followed the established rigorous and structured curriculum, which emphasized academic achievement and leadership qualities. He proved to be an outstanding student and passed in sequence the School Certificateand Higher School Certificate with high marks and gained university matriculation in 1924. His keen interest in chemistry led him to enter the nearby University of Birmingham, where Professor W. N. (later Sir Norman) Haworth was shortly to bring his worldrenowned school of carbohydrate chemistry, and Tipson received the B.Sc. degree with First Class Honours in 1927. CopyriShlO 1994 by Academic Ress, Inc.

AU rights of reproductionin any form reserved.

xiii

XiV

ROBERT STUART TIPSON

As the top student in his class, Tipson was awarded the Priestley Research Scholarship and began his research work on carbohydrates under the direction of Professor Haworth. A year later, under support from a teaching scholarship, he served as a junior instructor in organic chemistry in Birmingham and associated with other future leaders in the carbohydratefield, notably Maurice Stacey, Fred Smith, and J. K. N. Jones, who were then progressing through their studies at Birmingham. Haworth was a hard and demanding taskmaster who insisted on long hours in the laboratory,but the young R. Stuart Tipson (he was proud to emphasizethe Scottish ancestry on his mother’s side) nevertheless found time to enjoy sports and especially music. The drum set and piccolo of his school days had given way to the saxophone, and with a small group he played the popular music of the twenties at dances and other university functions. He also taught evening classes in chemistry and mathematics at the City of Coventry Technical College. His research in Birmingham during 1927- 1929 focused on fundamental studies on methylated mannopyranosides and the plant-storage fructan, inulin, and also with one of the earliest applied studies, sponsored by the British Empire Cancer Campaign, on the isolation of tobacco “tar” and the study of its constituents. However, he was eager to enlarge his horizons, and with the encouragement of Professor Haworth he left Birminghamin August 1929for Montreal, Canada, to work with Haworth’s friend Professor Harold Hibbert in the Department of Industrial and Cellulose Chemistry at McGill University. He worked with Hibbert on the structure of another fructan, the bacterial polysaccharide levan, and the results led to his first publication, which appeared in 1930 in the Journal of the American Chemical Society, as well as a second report that appeared in the following year. While at McGill he also gave instruction on the technique of Pregl’s microanalysis, and for a short time he worked in Prince Rupert, British Columbia, conducting a study on fish oils for the Fisheries Experimental Station (Pacific). After only a year in Canada, Tipson moved in August 1930to New York to accept a position as a research assistant to P. A. Levene at the Rockefeller Institute for Medical Research, where he began a phase of his research career that was most productive and lasted for nine years. By 1932 his work with Levene had resulted in no fewer than eight publications, mostly in the Journal of Biological Chemistry.This series of reports established the correct ring structures of the methyl glycosides of D-ribose and demonstrated that the natural purine (and later pyrimidine) ribonucleosideshave the sugar in the furanose ring form. Tipson was able to assemble the research results from his Birmingham days, along with the work with Hibbert and that with Levene, into a prodigious Ph.D. thesis “Studies in the Carbohydrate Group,” which he suc-

ROBERT STUART TIPSON

xv

cessfully defended in June 1932 before the Birmingham examiners, Professor Haworth and Dr. E. L. (later Sir Edmund) Hirst, and the external examiner, Professor (later Sir Ian) Heilbron. During Tipson’s return to New York from a trip to England the year before, he met on board ship a young lady, Constance Goodwin, from Asbury Park, New Jersey, who became his wife some months later. Her quiet and gentle personality was in sharp contrast to his forthright and outspoken demeanor, but she was nevertheless able to assert her influence in indirect ways, and she provided an excellent complement to the sometimes domineering mannerisms and quirky humor of her husband. The couple were devoted companions until Connie’s death in 1985; they had no children. Levene, a physician by training, was by all accounts a true genius, a self-taught chemist and pioneer biochemist. The environment of the Rockefeller Institute, with its provision for research opportunities for younger scientists, was a perfect situation for Tipson to exercise his experimental research talents to the fullest while absorbingthe drive and dedication manifested by Levene in addressing immensely difficult problems of the chemistry of life processes. Tipson devoted most of his years in Levene’s laboratory accomplishing seminal work on the components of nucleic acids. To determine the ring forms of the ribose component of the ribonucleosideshe applied Haworth’s methylation technique and established the furanoid structure for the sugar in adenosine, guanosine, uridine, and thymidine. He showed that formation of a monotrityl ether is not a reliable proof for the presence of a primary alcohol group in a nucleoside, whereas a tosyl ester that is readily displacedby iodide affords clear evidence that the ester is at the 5-position ofthe pentofuranose. Acetonation of ribonucleosides was shown to give the 2’,3’-O-isopropylidene derivatives, which were to become extensively used in nucleoside and nucleotide chemistry, and were utilized by Tipson in the first chemical preparation of a ribonucleotide, inosinic acid. Structural work on the nucleic acids by the traditional techniques of the 1930s provided challenges of formidable complexity, especially as the propensity of phosphate groups to migrate was not then recognized. Nevertheless, as early as 1935 Levene and Tipson advanced the accepted backbone structure for DNA when they formulated an oligonucleotideofdeoxyribonucleic acid as having 3’ 5’ phosphate diester linkages between the furanose sugar components. Over thirty publications resulted from Tipson’s work in Levene’s laboratory. Along with the work on nucleic acid components, he also studied the structures of gum arabic and other plant gums, and conducted a range of synthetic investigationson sugars, with particular emphasis on uronic acids and 5-carbon ketoses. His 1939 observation that acetylated glycosyl halides

-

xvi

ROBERT STUART TIPSON

react with oxyanions to give products having the trans configuration of substituents at the anomeric position and the neighboringcarbon atom was popularized two decades later by B. R. Baker as the “trans rule,” and Tipson made strenuous efforts in later literature to establish his own prior development of this concept. Following the death of Levene in 1939, Tipson’s position as a research associate at the Rockefeller Institute came to an end. He had held a part-time appointment as a lecturer in advanced biochemistry at Brooklyn College during 1938 - 1939, but in August 1939 he accepted a position at the Mellon Institute, and the Tipsons moved from their home on Long Island to Pittsburgh. He was affiliated with Dr. L. H. Cretcher, who headed the Department of Research in Pure Chemistry at the institute, which was then associated with the University of Pittsburgh. The work at the Mellon Institute was largely directed toward applied problems, including quinoline derivatives as antimalarial drugs,cinchona alkaloids as antipneumococcalagents, and the chemistry of alloxan. Some 40 research articles resulted from Tipson’s 18 years at the Mellon Institute, and they demonstrate that he was able to sustain some of his interest in carbohydrate chemistry, and he continued to study the reactions of sulfonic esters with sodium iodide. In 1945 he compiled his published work into a senior thesis for the D.Sc. degree that was awarded by the University of Birmingham. However, a considerable proportion of the research at the Mellon Institute was never published because of patent restrictions. This was particularly true for his work on carbohydrates and other organic compounds conducted after July 1952, when he was assigned to the Parke, Davis and Company Fellowship in Medicinal Chemistry to synthesize potential antiviral and anticancer agents. Tipson always enjoyed and took great pride in precise scientific writing. When the Advances in Carbohydrate Chemistry series was launched in 1945 under the editorship of W. W. Pigman and M. L. Wolfrom, he wrote a comprehensive article for Volume 1 on the chemistry of the nucleic acids that is to this day a model of historical accuracy, careful and economical use of language, and clear interpretation of experimental data based on thoroughly characterized crystalline compounds. Equally precise and meticulous as an experimentalist, he devoted time during his days at Pittsburgh to write comprehensive articles on such practical techniques as crystallization, vacuum distillation, and sublimation, which were published in the Weissberger Techniques of Organic Chemistry series. His preoccupation with careful experimental techniques and their accurate recording in the literature remained with him always. He abhorred vague descriptions of procedures, speculative interpretations not based on

ROBERT STUART TIPSON

xvii

solid facts, failure to give proper credit to prior work, and any imperfections in the use of the English language. Tipson’s long-standing interest in sulfonic esters led him to contribute a landmark article on carbohydrate sulfonates for Volume 8 of Advances in Carbohydrate Chemistry, which marked the start of his editorial involvement with the series; he joined as assistant editor to M. L. Wolfrom starting with Volume 9. This was the beginning of a long and fruitful association between Wolfrom and Tipson that assured researchers in the carbohydrate field of a regular seriesof authoritativearticles on a wide range of topics, both fundamental and applied, related to carbohydrates. The Wolfrom - Tipson team demanded standards of scientific accuracy and careful writing style that characterized their own works; all authors, however eminent, were subjected equally to the analyticjudgment and revision of the two editors. While Wolfrom focused on the balance of literature citation and scientificinterpretation,Tipson took infinite pains to ensure the exact use of scientific nomenclature and clear presentation in grammatically correct English. Authors from all around the world became familiar with the Tipson treatment of their manuscripts, with line-by-line correctionsin Tipson’s fine and precise writing, rendered in inks of various colors and often with mild expletives or pungent comments in the margin. The revised manuscript would be accompanied by a long handwritten letter with pointby-point queries to be addressed by the author. The final manuscript would go to the typesetter with all of the exact copyediting directions marked by Tipson; he had no confidence in publishers’ copyeditors. Any unwarranted alterationsby the publishers or erroneous “corrections” on the proofs by the authors would promptly elicit a terse and vexatious letter from Tipson. Some authors took exception to this “interference” with their personal writing style, even when it involved the rectification of obvious solecisms, but few did not finally appreciate the significant improvement in their published articles. The readership was undoubtedly the ultimate beneficiary of Tipson’s painstaking work. Although Bob Tipson was not a gregarious person and, in fact, became quite reclusive in his later years, he had a fine appreciation of the personal qualities of individuals he respected. This facet of his character is most evident in the sensitive and insightful articles he wrote for the Advances series on the life and work of two of his mentors, P. A. Levene and Harold Hibbert. In August 1957, Dr. Tipson left Pittsburgh to accept a research appointment in Washington, D.C., at the National Bureau of Standards, where he assumed a position in the prestigious research laboratory headed by Dr. Horace S. Isbell. Here he had an opportunity to devote his efforts full-time to

xviii

ROBERT STUART TIPSON

carbohydrate chemistry. He remained at the Bureau until his formal retirement in 1972. The Tipsons’ new home in Kensington, Maryland, closeto the Capital Beltway that encircles Washington, became the office for all of Bob’s editorial work for more than 30 years. While the broad mission of the National Bureau of Standards was concerned with standard reference materials, Dr. Isbell centered the work of his laboratory on his long interest in the carbohydratesand on the use of physical methods in their characterization. Infrared spectroscopy had shown promise in providing structural and conformational information on carbohydrates and their derivatives, and Isbell invited Tipson to conduct detailed infrared studies on the extensive collection of carbohydrate samples maintained by Isbell. The series of publications that rapidly resulted furnished a basis for assigning conformationsto pyranoid sugars and their derivatives. Although this work was later to be overshadowed by application of the much more powerful technique of nuclear magnetic resonance spectroscopy,the IsbellTipson work helped to define the molecular shapes involved and the terminology required for their description. In addition to these physical studies at the Bureau, Tipson was able to return to his synthetic interests, both alone and in collaboration with other staff members. He was especially pleased to prepare D-talose in crystalline form, an accomplishment that had eluded Emil Fischer. Pursuing his longstanding interest in the reaction of sulfonic esters with iodide and following an earlier observation that the tetratosyl ester of erythritol is converted into butadiene by the action of sodium iodide and zinc, he demonstrated(with A. Cohen)that nonterminal unsaturation may be convenientlyintroduced into alditol derivatives by reaction of contiguous secondary sulfonates with sodium iodide and zinc dust in boiling N,N-dimethylformamide. This Tipson - Cohen reaction subsequentlyproved of great utility in other hands for the conversion of more complex carbohydrate structures into vicinal dideoxy derivatives. While in government service,Tipson undertook special assignmentsfrom the U.S. Congress and in 1964- 1965 spent nine months as a consultant to the Surveys and InvestigationsStaff of the House Appropriations Committee. For this work he received awardsfor outstandingservice, although he did not enjoy the extensive travel that was required nor the dislocation from the editorial work that normally occupied a major part of each of his days at home. Following the retirement of Isbell from the National Bureau of Standards in 1968, the orientation of the laboratory was directed progressively toward targets outside the carbohydrate area, and in 1972 Tipson elected to retire from the Bureau and concentrate his efforts full-time on writing and editorial work.

ROBERT STUART TIPSON

xix

Although Tipson set out with stellar credentials in his career as a chemist and sustained a remarkably productive scientific life, he was never able to establish himself as an independent academic who would lead a group of research students. He had the good fortune to associate with such towering figures in the carbohydrate field as Haworth, Levene, Wolfrom, and Isbell. However these long-standing liaisons with major leaders may at the same time have deprived him of the opportunity to exert independence and to develop as the animator of a research group of his own. Tipson was a competent lecturer when he remained close to the technical material that he was presenting,and he could carry an argument forcefullyin discussion, but he was much more comfortable with the written word when interacting with other scientists. He was a regular participant in major scientific meetings during much of his career and traveled at regular intervals to visit his native England, but he later tired of travel, especially after his consulting stint for Congress. Subsequent to that time he was rarely seen outside his Kensington home or the weekend cottage that he and Connie enjoyed and where they would happily receive friends and colleagues. He was blunt and outspoken, had little interest in political finesse, and his written comments often projected a gruffand sometimesintolerant image. It is true that he had scant tolerance for fools, frauds, or pompous individuals, but those who had known him only through their “Tipsonized” manuscripts and the accompanying multicolored, handwritten letters were often surprised upon visiting him at home to find a kind and mild-mannered host who projected a friendly and engaging informality. A man of moderate stature and trim build, he enjoyed robust health for most of his life, despite his heavy smokinghabit. He participated in a range ofindividual sportsin his younger days, and later maintained his enjoyment of swimming and golf. His tastes were modest and his frugality sometimes reached the point of obsession,but he and Connie alwaysdelighted in extendingthe hospitality of their home to guests. The visitor would certainlybe given a tour of the garden and also be shown Bob’s collection of memorabilia and possibly his extensive stamp collection. An overnight visitor might be aroused by an early serenade on the saxophone, generally as a prelude to Bob’s call to a copious English breakfast announced in his stentorian voice. Throughout his career Tipson took a deep interest in organic nomenclature and played a leading role in the development of carbohydrate nomenclature. As early as 1939 he was appointed to a standing committee of the American Chemical Society’s(ACS)Division of Sugar Chemistrythat was in charge of standardizing terminology in the carbohydrate field. The nomenclature recommendations developed by this committee were subsequently adopted by the ACS, and ongoing negotiationsled subsequentlyto a revision

xx

ROBERT STUART TIPSON

that had joint British- American approval. Tipson continued for several years as head of the ACS committee and was involved with a joint commiscommission of the International Union of Pure and Applied Chemistry anc the InternationalUnion of Biochemistry in developingthe first internationa set of recommendations for the nomenclature of carbohydrates. His enor. mous experience as a researcher and as an editor in the carbohydrate fielc provided an unmatched background of structural knowledge with which tc test proposals for new terminology. He was one of the founding editors of the international journal Curbohy drute Research when it was established in 1964,and he remained active as ar editor of the journal for a quarter of a century. In this capacity he interactec with hundreds of researchers and became a familiar friend and unprejudicec critic to numerous scientists around the world, who benefited from hi! meticulous editing of their manuscripts, which he handled with astonishing swiftness. He was particularly insistent on the traditional literary use o punctuation in clarifying the text of complex material; he deplored tht journalistic style adopted by some authors. Members of the carbohydratt community coined the term “Tipsonized” for his treatment of their manu. scripts. His incisive revisions, often bearing marginal comments inspired bj the “Peanuts” cartoon series, bestowed impartially upon the mandarins o scienceas well as on the beginning investigatorstrugglingwith a language no his own, brought clarity and impact to countlesstexts that passed through hi! hands. The busy reader was the immediate beneficiary, but there are fea authors who did not ultimately recognize, after the initial trauma of retypini a heavily revised manuscript,the excellence in the presentation oftheir worl thanks to the unstinting efforts of Bob Tipson. Tipson never lost touch with his early research on nucleosides and nucleo tides. In 1968 he joined with W. W. Zorbach in producing a manual, Syn thetic Procedures in Nucleic Acid Chemistry. This volume compiles detailec descriptions of key experimental procedures contributed by a selection o authors from around the world and includes a number of Tipson’s owl procedures. The success of this volume led 2 years later to the production of; second collection. Subsequently,with the collaboration of Leroy Townsenc as coeditor, a new and expanded seriesNucleic Acid Chemistry was launchec and extended to four volumes, the final one being published in the last yea of Tipson’s life. Throughout these series the emphasis was on practicalit and thorough directions for guiding the novice experimentalist,as stressed ii the preface to the first volume: “The Editors have zealously avoided certaii detestable directions such as ‘processed in the usual manner’ . . . A re search worker with a reasonable skill in the art . . . may successfullyrepea the preparations without specialized training.”

ROBERT STUART TIPSON

xxi

The total scope of Bob Tipson’s many editorial and nomenclature activities with house publications, professional society reports, and committee service is too extensive to list in detail. In addition to his long involvement with the journal Carbohydrate Research, his editorship of Advances in Carbohydrate Chemistry and Biochemistry extended through Volume 48, published in 1990, and this completed a remarkable and profoundly influential 36 years of editorial activity with the series. Dr. Tipson’sinvolvement in scientificsocietiesincluded life memberships in the Royal Society of Chemistry, the American Society of Biological Chemists, and the American Chemical Society. He helped to organize the first Gordon Conference on Carbohydrates.He worked actively on behalf of the Carbohydrate Division of the ACS, serving a term as an officer of the division. In addition to his key role in the division’s Nomenclature Committee, he edited for many years the abstracts of papers presented at the meetings of the division. His outstanding contributionsto the carbohydrate field were recognized by the ACS in 1971 with the conferringofthe Claude S. Hudson Award, and, in 1986, the Carbohydrate Division honored him with the Melville L. Wolfrom Award. A special issue of the journal Carbohydrate Research was dedicated to Dr. Tipson on the occasion of his retirement as an editor of that journal. Tipson’sremarkablecapacity for work continued undiminished until very late in his life, even though he seldom ventured outside his home after the death of his wife. Physical impairment during his last two years finally curtailed the extent of the editorial activities from his celebrated mailbox, and he died in his 85th year after a short illness. He is survived by his sisters Joyce Kenward of Leamington Spa, England, and Jean McIlroy of Sedona, Arizona, as well as six nephews and four nieces. Although memories of Bob Tipson will inevitably center on his insistence on perfection in the preparation of scientific manuscripts, the fundamental research contributionshe made early in his long career to our knowledge of the structure of carbohydratesand especially of the nucleic acids merit equal recognition. His brusque and sometimes irascible outward manner concealed a kindly and concerned individual, a true gentleman and scholar. DEREKHORTON The author thanks the following persons for valuable discussions and information during the preparation of this article: J. S. Coombes, Joyce Kenward, Jean McIlroy, H. S. El Khadem, R. Schaffer, M. Stacey, and J. M. Webber.

This Page Intentionally Left Blank

ADVANCES IN CARBOHYDRATE CHEMISTRY AND BIOCHEMISTRY, VOL. 50

HOW EMIL FISCHER WAS LED TO THE LOCK AND KEY CONCEPT FOR ENZYME SPECIFICITY’ BY RAYMONDU. LEMIEUX AND ULRIKE SPOHR Department of Chemistry, University of Alberta, Edmonton, Alberta T6G 2G2, Canada

I. INTRODUCTION Emil Fischer’s genius was in the identification of important areas for research in the field of organic chemistry, which, as the name implies, was concerned with compoundsderived from living organisms. Once the project was identified and engaged, he brought unsurpassed creativity for successful experimental involvement and logical interpretation of the results. He seemed to have had a global view of natural science, and his driving interest was to contribute to an understanding of the chemical processes of living organisms. He is quoted ( 1) as stating, in a letter to his mentor, Adolf Baeyer, that he wished to synthesize the first “artificial ferment” (enzyme activity) and that, with the achievement of this goal, he would consider his mission in life accomplished. Later, in the course of his research on polypeptides, he realized that he would not reach this goal. He, of course, could have no idea of what the synthesis of an enzyme would involve. Our assessment of the literaturesuggeststhat Fischer’s motivation to enter the field of carbohydrate chemistry was the realization that knowledge of the relative configurationsof the asymmetric carbon atoms of the sugars was an essential stepping stone for research in the biological systems, that area of inquiry central to his interest in a scientific career. His brilliant success in meeting this challenge has rightly earned him the title of father of carbohydrate chemistry. This chapter demonstratesthat a strong claim can be made that he is also a leading pioneer of biological chemistry. Hudson (2) published a scholarly review of the monumental contribution

I Presented at the symposium “Emil Fischer: 100 Years of Carbohydrate Chemistry,” 203rd National Meeting of the American Chemical Society,Division of Carbohydrate Chemistry, San Francisco, California, April 5 - 10, 1992.

1

Copyright 0 1994 by Academic Press, Inc.

Au nghts of reproduction in any form reserved.

2

RAYMOND U. LEMIEUX AND ULRIKE SPOHR

of the relative configurationsof the sugars- the feat for which Emil Fischer is best known. Indeed, this stellar accomplishment appears to have overshadowed other major contributions, particularly the first insights into enzyme specificity. We now attempt to present, with fidelity, how Emil Fischer was led to his “lock and key” concept for enzyme specificity. The objective is to produce a more widespread appreciation of the profound significance of this landmark contribution (3), which arose from his recognition of asymmetric induction. Although not always literal, our translations into English are expected not to be misleading. The nomenclature is retained essentially as used by Fischer. Emil Fischer’s proof of the relative configurationsof the four chiral centers in open-chain glucose (4) appeared in 1891. The followingthree years saw a brilliant series of follow-up papers, based on the knowledge that certain sugarsdiffer only in configuration.These publications mark the origin of our appreciation that molecular forces provide the stereochemical guidance necessary to living processes. Especially because of the present worldwide concern with molecular recognition, it seemed most appropriate to survey these classical contributions as a centenary commemoration to the first person in the field. One of his many students and a life-long friend, Karl Freudenberg, remarked in his excellent biography (5) of Emil Fischer: Theoreticalquestionsplayeda minor role in his thoughts. . . . Emil Fischer was theclever tactician whoproceeded on a broadfront. . . . There were many who were better read than he, but no-one who had more practical experience.

11. ASYMMETRIC INDUCTION Whereas the configurationsof the sugars are truly a lasting monument to Emil Fischer, the concept of asymmetric induction, also referred to as partial asymmetric synthesis, initiated a new era in chemical research that is still with us today. The modem concept of asymmetric induction is illustrated by the formulas in Fig. 1. As shown, the addition of hydrogen cyanide to the optically active aldehyde can lead to two diastereomers (1 and 2). If the process is under thermodynamiccontrol, the formation of the more stable isomer will be favored; that is, that isomer for which the non-bonded interactions between the newly formed cyano and the hydroxyl groups with the dissymmetric R*group are weakest. On the other hand, the difference in the yields of 1 and 2 can be the result of kinetic control arising from a difference in the energies of the transition states- that state with the lower energy will form faster and lead to the product of higher yield. It is noteworthy that the tenets

LOCK AND KEY CONCEF’T FOR ENZYME SPECIFICITY

+

R’CHO

HCN

-

CN

3

CN

I

HCOH

I

+

I

HOCH

I

R’

R’

Ill

Ill

R

R

1

2

E = Nonbonded interaction energy. When R* is dissymmetric,

El

+ E‘1 # E2 + E’2

:. 111 # 121 FIG.1. -Asymmetric induction-under thermodynamiccontrol, 1 will form in higher yield if E ,

+ E’, < E, + El2.

of conformationalanalysis first provided the theoretical base for the formulation in 1952 of an appreciation for the steric control of asymmetric induction, which has become known as Cram’s rule (6). Although organic chemistry was still at a primitive state of development,a strong foundation for Emil Fischer’swork had been laid by such great chemists as Berzelius, Wohler, Liebig, Baeyer, and Kiliani. Furthermore, Louis Pasteur had reported (7), as outlined in Fig. 2, the preferential metabolism of the dextro-enantiomer of tartaric acid about 15 years prior to Fischer’s doctoral studies with Adolf Baeyer at the University of Strasbourg. An understanding of optical isomerism had been provided in 1874 by the van’t Hoff-Le Be1 theory of the asymmetric carbon atom (8) (see Fig. 3) and ball and stick three-dimensional molecular models were in use much as they are COOH I H-C-OH I HO-C-H I COOH Natural

COOH I HO-C-H I H-C-OH I COOH

-

COOH I HO-C-H I + Fermentation products H-C-OH I COOH

-------+

dl-Tartaric acid

Penicilliwn

glaucwn

I-Tartaric acid

FIG.2.-Louis Pasteur’s preparation of D-(kvo)-tartaric acid.

4

RAYMOND U. LEMIEUX AND ULRIKE SPOHR Planeof~ymmeby

. . . O

. .

COOH

H

H-C-OH

,

- HO-C-H -

H-C-OH

HO-C-H

Diymehic

~

.

COOH

COOH

0

#

H-C-OH 0

HO-C-H

#

HO-C-H

*

H -C -OH

0

0

COOH

Meso

,

COOH

Racemate

FIG.3.-The basic assumption for Fischer’s research of the optical isomerism of sugars.

today. As Emil Fischer stated (4): Allprevious observations in thesugar group are in such complete agreement with the van’t H o f - Le Be1 theory of the asymmetric carbon that the use of this theory seemsjustifiable.

The concept of the existenceofasymmetricforces in nature was not new to Fischer. Indeed, Louis Pasteur (9) was generalizing about asymmetry in 1874 when he prophesied: I am convinced that life as we know it has arisen out of asymmetrical processes in the universe. The universe is asymmetric.

He was convinced that optical activity is a peculiarity of life and, therefore, his view was directed toward asymmetry at a cellular level rather than at the molecular level developed by Emil Fischer. Thus, the stage was well set in 1882 when Emil Fischer, at the age of 30 years, was appointed Professor and Director of the Chemical Institute at the University of Erlangen and thereby gained full independencein the direction of research. He chose to study the carbohydrates, and his first publication in the field appeared in 1884. The work was concerned with the reaction of sugarswith phenylhydrazine,a compound that he had discovered as a teaching assistant about 10 years earlier while helping one of Baeyer’s students. At that time, his father, Laurenz Fischer, who was a very successful businessman, added to his portfolio a brewery in Dortmund. His son, Emil, who was already a chemist, became involved and it is recorded (1) that he had advised his father to purchase a Linde ice-making machine to cool and store the beer. Also, Emil developed an interest in mycology, the science of lower plants, and he recommended that the brewery acquire a microscope to differentiate yeast species and to detect contaminants. He had actually become highly knowledgeable about yeasts some 6 years earlier while he was studying in Adolf Baeyer’s laboratory and, in the course of a brief stay in Strasbourg in 1876, through a viniculturist named Dr. Fritz, he became intensively involved in the study of lower plants at the Strasbourg-BotanicalInstitute.

LOCK AND KEY CONCEPT FOR ENZYME SPECIFICITY

5

Fortunately for chemistry, he did not stay in Strasbourgbecause he wrote in his memoirs (lo), I certainly would have done my research in thisfield (mycology)had I stayed longer in Strasbourg.

He was then 24 years of age. Organic chemistryin the late 19th century was focused on the chemistry of natural products. The main opportunities,in terms of the techniques of the time, were offered by the major components of living tissues, that is, the carbohydrates, nucleic acids, lipids, tannins, and proteins. The idea that organic compounds could be synthesized only within living organisms had long been dispelled. Nevertheless, as already pointed out, Louis Pasteur (9) had quite recently expressed the opinion that there existed forces that could be exerted only within living cells. This idea was strenuously opposed by Liebig ( 1I), who held that fermentationand similar processes were due to the action of chemical substances. The Pasteur - Liebig controversyended when Buchner ( 12) succeeded in extracting a cell-free fermentation system from yeast that fermented glucose. Van’t Hoff had predicted, in 1874, that there should exist 16 normal straight-chain optically isomeric aldohexoses (8,13). It was the awesome challenge to substantiate this prediction that Fischer accepted in the mid1880s, and that was met by his classical contribution in 1891 entitled, “On the Configuration of Glucose and Related Compounds” (4). He represented glucose as shown in formula 3 and indicated that it should be interpreted as seen for 4. Thus, the hydrogen atoms and the hydroxyl groups at the four asymmetric carbons are considered to project above the plane of the paper. The representation was later simplified to that shown for

.. . .

COH

HC=O

HC=O

H-C-OH

H-C-OH

HYOH

HGC-H H-C-OH 0

H-C-OH CH20H 3

-

. .

HO-C-H

I

-

HO~H

I

H-C-OH

HCPH

H-C-OH

HC~H

I

CH20H

CH20H

4

5

dextro-G lucose

5 and became known as a Fischer projection formula (14). Fischer realized that an arbitrary assignment of absolute configuration was necessary to an orderly development of organic chemistry. Fortunately, his assignment of 3 to dextro-glucose proved correct.

6

RAYMOND U. LEMIEUX AND ULRIKE SPOHR

The preparation of a-hydroxy acids by way of the cyanohydrin was established by Winckler (15) in 1832 by the synthesis of mandelic acid from benzaldehyde. As Emil Fischer emphasized, it was Kiliani (16) who first applied the well-known cyanohydrin synthesis of a-hydroxy acids from aldehydes and ketones to the building up of aldonic acids from aldoses. The reduction of these acids to fatty acids then provided Kiliani’s classical proof of the structures of glucose and fructose (17). Although he had used sodium amalgam for the reduction of sugars to alditols,it was Fischer who learned to reduce the aldonolactonesto aldoses. The overall process of building aldoses to higher sugars has become known as Fischer- Kiliani synthesis. We have no evidence that Kiliani was aware of the formation of diastereoisomeric aldonic acids in the course of his cyanohydrin reaction. Fischer knew that the readily available I-arabinose had been subjected to the cyanohydrin reaction by Kiliani and found to provide a substance he called “arabinocarbonic acid.” Indeed, Fischer offered the opinion that the reaction of sugars with hydrogen cyanide discovered by Kiliani in 1885 was the most important reaction in carbohydrate chemistry (18). Fischer characterized this product as “I-mannonic acid” and realized that, likely, the compound could be epimerized to 1-gluconic acid by heating with quinoline. However, he felt this procedure was so tedious and inefficient that the acid would not have been prepared in this way unless it had first become available by some other method in order that its physical properties be known. He reported ( 18) that, in fact, a good quantity of I-gluconic acid formed along with the I-mannonicacid on the addition of hydrogen cyanide to l-arabinose (see Fig. 4). He found this strange and commented as follows: The simultaneousformation of the two stereoisomeric products on addition of hydrogen cyanide to aldehydes, which was observed here for thefirst time, is quite remarkable in theoretical as well as in practical terms.

Also, in 1890Fischer had proven that the reduction of fructose with sodium amalgam yields a mixture of mannitol and sorbitol and pointed out that this conformed with the van’t Hoff-Le Be1 theory (19). It seems, therefore, that the idea of asymmetric induction was clearly in a state of incubation prior to his publication of the relative configurations of the sugars in 1891. Fischer’s involvement with the relative configurations of the sugars required the preparation of pure substances and he gradually accumulated experimental data (20-22), which required the formation of epimeric aldonic acids in unequal amounts (see Table I). Thus, he was able to write (23) in 1892: A second question of general importance relates to the quantities in which the two stereoisomeric products are obtained on the generation of a new asymmetric carbon atom.

Starting with nonracemic optically inactive starting materials, only racemic products are

LOCK AND KEY CONCEF'T FOR ENZYME SPECIFICITY

.

H-C=O

HGC-H

.

. ..

.

COOH

H-C-OH HCN

H20

HaC-H CH20H

COOH

H-C-OH NH3

+

H-C-OH

HGC-H H-C-OH

t

HCC-H

HGC-H

HaC-H

HGC-H

.

CH2OH I-Arabinose

CHZOH

I-mannonic acid I-gluconic acid (Kiliani's arabinocarbonic acid)

FIG.4.-The discovery of asymmetric induction. The yield of the I-mannonic acid, under the conditions then used, was about three times greater than that of the ghco isomer (see Table I). formed and that means the two stereoisomers areformed in equal quantities.In the case of the present syntheses (Kiliani cyanohydrin syntheses), where the sugars used as starting materials are already asymmetric systems, this rule does not apply.

Two years later, in 1894, he wrote as follows (3): To my knowledge, by these observations strictly experimentalproof has been providedfor thejirst time that in the case of asymmetric systems the further synthesis occurs in an asymmetric sense.

111. YEASTFERMENTATIONS AND ENZYMES

Obviously, Fischer had conceived of the phenomenon we now refer to as asymmetric induction and had become deeply interested in its relevance to biological processes. It was that year that he abandoned Wurzburg University to accept the chair of chemistry at the University of Berlin, which was regarded as the highest position in the realm that could be achieved by a professor of chemistry. He was promised a large new institute and it appears that the design and financing of this laboratory met with considerable controversy. It seems probable that Fischer took advantage of this discontinuity in his research to write up much of the work he had done in Wurzburg on the TABLE I Some Early Observations of Asymmetric Induction by Fischer

I-Arabinose (50 g) d-Mannose (2 kg) d-Xylose (40g)

-

Mannonic acid Ca-gluconate a-Mannoheptonic acid Gluconic acid lactone Idonic acid

Yield (46)

Reference

34

20

11 87 51

21 22

35

8

RAYMOND U. LEMIEUX AND ULRIKE SPOHR

fermentation of sugars and the interpretations of the results, in a series of monumental publications in 1894. The use ofthe enzyme system then known as invertin, which was extracted from beer yeast with water and precipitated from the aqueous solution, was available to Fischer when he began his classical studies of the enzymic hydrolysis of glucosides reported in 1894. The stage was also set by another enzyme known as emulsin, which Fischer purchased from E. Merck, Darmstadt, and which was known to hydrolyze several natural aromatic glucosides such as salicin, coniferin, arbutin, and the synthetic phenyl glucoside. These aryl glucosides were already known to not be cleaved by invertin. The lock and key concept for enzyme specificity appears to have gelled in Fischer’s mind in the course of using yeasts in his studies on the configurations of sugars. It is noteworthy in this connection that it was not until 1878 that the term “enzyme” was introduced by Kuhne (24). Until then, the substances responsible for these biological activities were referred to as ferments. In fact, enzyme is a Greek term that means “in yeast.” It is pertinent to note with regard to enzyme action that Pasteur’s opinion, that the fermentation process could not be separated from the living cell, did not take into account the observation, made in 1833 by Payen and Persoz (25), that starch was converted into reducing sugars by a thermolabile substance present in the precipitate that formed on adding alcohol to a cell-free aqueous extract of malt. They termed the substance a “diastase” and the “-ase” ending of this term became in time used to designate the protein catalysts that we now call enzymes. In an Emil Fischer memorial lecture, Forster (26) reported that, as early as 1837, Berzelius held the opinion “that in living plants and animals there take place thousands of catalytic processes between tissues and fluids.” It took Fischer to appreciate the significanceof the diastase activity. Fermentations using ordinary beer yeasts had played key roles in the investigations on the configurations of glucose performed in Wiirzburg, which Fischer published in 1891 (4). For example, in 1889, Fischer and Hirschberger (27) reported the fermentationof d-mannose, a sugar that they had obtained by oxidation of mannitol with nitric acid and found identical to an aldohexose of widespread occurrence in plants (28). In the following year, he reported that the fermentation of racemic mannose left the l-isomer intact (29). Similarobservationswere made with regard to the nonfermentability of l-(dextro)-fructose, I-glucose, and l-galactose (29) (see Table 11). In addition, both the optically active isomericgulosesand various heptoses and octoseswere found to resist fermentation.He saw these results as an essential extension of the older observation by Pasteur (7) that microorganisms alter only one of two enantiomers;that is, the fermentation of sugars depends on

LOCK AND KEY CONCEPT FOR ENZYME SPECIFICITY

9

TABLEI1 Experiments That Demonstrated the Chemical Basis of Biology

Racemic mannose

aI-mannose + CO, + ethanol

+ ethanol

Similarly, d-glucose, I-fructose,or d-galactose

CO,

However, I-glucose, d-fructose,or I-galactose

no fermentation

the total configurationand not only whether it is the dextro or lev0 form. He therefore concluded: Thefermentabilityofhexosesis in close relationship to thegeometricshapeofthe molecule and can even be designated as a stereochemical question.

In a landmark paper (30) that he coauthored with Hans Thierfelder, a mycologist, the behavior of different sugars toward pure yeasts was described. In this connection, he realized that the yeasts he had used in his earlier investigationswere mixtures and, therefore, the results could be misleading. For this reason, he turned his attention to the fermentation of sugars by 12 different pure yeast species. Furthermore, he realized that he was in a uniquely fortunate position to undertake these studies because the research on the relative configurationsof the sugars had left him with a fine inventory of rare carbohydrate structures. It is interesting to note how the fermentations were scaled down for the study of rare sugars as substrates (Fig. 5 ) . In this regard, he wrote: Since the preparation of the artificial sugars is in part quite tedious and the experiments had to be modified repeatedly, we used a smallfermentation tube, as shown below, in order to save material (30).

As seen in Table 111, all six yeasts rapidly fermented glucose, mannose, and fructose. However, three of the yeasts had difficulty in fermenting galactose. None could metabolize either the naturally occurring sugars I-arabinose and rhamnose or synthetic sugars including 1-glucose,sorbose, a-glucoheptose, and a-glucooctose.Thus, the data presented in Table I1 were confirmed with pure yeast cultures and Fischer proposed the generalization: The same observation is likely to befound for other microorganismsas well as for other groups of organic compounds and perhaps a very great number of chemical processes occurring within an organism are influenced by the geometry of the cell.

IV. THELOCKAND KEYCONCEPT

A number of glycosides were available to Fischer by way of the KoenigsKnorr reaction and his own glycoside synthesis, which involves treatment of

RAYMOND U. LEMIEUX AND ULRIKE SPOHR

10

Actual size FIG.5.-Emil Fischer’s fermentations on a semi-micro scale using an apparatus of the size shown. (a) -70 mg sugar, 0.35 ml water, 0.35 ml sterilized yeast extract, and 13 mg yeast species; (b) S-trap for evolved COz;(c) aqueous barium hydroxide.

the sugar in alcoholic hydrogen chloride solution. Consequently, the fermentation of a number of glycosidesby different pure yeast species could be examined (30). The results, presented in Tables I11 and IV, showed that certain yeasts that avidly fermented glucose, fructose, and mannose only reluctantly fermented galactose and that a yeast that fermented sucrose and TABLE111 The Selective Fermentation of Natural Sugars by Pure Yeasts sugar

Glucose

Mannose

Galactose

Yeast

d

1

d

I

d

I

SpastorianusZ S. pastorianusZf S.pastorianusfff Brauereihefe Brennereihefe Milchzuckerhefe

+++ +++ +++ +++ +++ +++

-

+++ +++

-

+++

-

a

b

-

-

+++ +++ +++ +++

-

-

-

++ +++ +++ + +

-

-

CFructosea

d-Sorbose*

+++ +++ +++ +++ +++ +++

-

-

Used by Fischer to designate natural ~ h c t o s e . Also negative were d-talose, I-gulose, I-arabinose,rhamnose, a-glucoheptose, and a-gIucooctose.

LOCK AND KEY CONCEPT FOR ENZYME SPECIFICITY

11

TABLEIV The Fermentation of Glycosides by Different Pure Yeasts

S. pastorianus Z Brauereihefe Brennereihefe S. productivus Milchzuckerhefe a

Sucrose

Maltose

+++ +++ +++ + +++

+++ +++ +++ +++ -

~

Lactose

Methyl a-glucoside

Glucosyl resorcinolb

+ + + +-

-

~

-

-

-

+++

-

Not tested

From Fischer synthesis. synthesis.

* From Koenigs-Knorr

lactose did not ferment maltose. These observationsled to his concern as to whether or not the different yeasts possessed differentenzymes. Experiments were designed to answer this question, and it was soon established that, in fact, yeast contains at least two different enzymes. The procedure is outlined in Fig. 6, where it is seen that, whereas an extracellular enzyme of the Frohberg yeast could hydrolyze sucrose but not maltose, the cells contained an enzyme that ferments both the disaccharides. On this basis, it was concluded: These present observations are undoubtedly in favor of the assumption that the yeast contains two direrent enzymes.

Of course, it is now established that the glycolysis of glucose to carbon dioxide and ethanol occurs by way of a complex pathway involving 10 different enzymes acting on a variety of sugar phosphate intermediates.The extracellularenzyme preparation that Fischer used was termed invertin, the origin of the term for the enzyme we now know as invertase. He termed the intracellular enzyme yeast-glucase (3 1) and this enzyme is of the type we now refer to as an amylase. Ground

Frohberg Yeast

1

Powdered glass

-

Crushed Cells

1

H2O

Cell-free Extract I

Invert sugar

No hydrolysis

H20

Cell-free Exiract I1

Invert sugar

Glucose

FIG.6. -How the presence of different enzymes in a yeast cell was established.

12

RAYMOND U. LEMIEUX AND ULRIKE SPOHR

Fischer then examined the lactose yeast in the same manner as he did the Frohberg yeast and found it to contain both an invertin-like enzyme and a lactose-cleaving enzyme, which he termed lactase. From these results he concluded that the first step in the fermentation of lactose, as for the fermentation of sucrose and maltose, is the hydrolysis of the disaccharideto monosacharide. From this observation, he drew the landmark conclusion that he considered it most unlikely that any polysaccharide (the term included disaccharides)can be fermented without first being hydrolyzed to hexose (3 1). The research with Thierfelder(30)had led to the hypothesisthat the active chemical agents of yeast cells can react only with those sugars that are configurationally related. It was this stereochemical assessment of the fermentation process that, in turn, now led to the question (32): Would similar diferences be found for the ferments that could be separated from the organism and termed “enzymes”?

To answer this question he turned to a study of the properties of two glucosidases, then known as invertin and emulsin (32). The substrateswere to be the large number of artificial glucosides that he had synthesized from different sugars and alcohols. The results ofthese studiesare presented in Table V. It is TABLE V Effects of Structure and Configurationon Enzymatic Hydrolysis Crude enzyme preparation Glycoside

a-Glucosides Methyl Ethyl Sucrose Maltose &Glucosides Methyl Phenyl Salicyl a-Galactosidec Methyl 8-Galactosides Methyl Lactose

Invertin”

Emulsinb

+ + + +

+ + +

+ +

Aqueous extract of air-dried beer yeast. Product of E. Merck, Darmstadt. Also, neither of the enzyme preparations hydrolyzed aglycosides of rhamnose or arabinose.

LOCK AND KEY CONCEPT FOR ENZYME SPECIFICITY

13

to be noted that Fischer did not know the configurations of the anomeric carbon of glycosides. Furthermore, he presented these compounds as furanoside structures in accordance with the prevailing notions on sugar structures. Fischer was intrigued by the fact that emulsin caused hydrolysis of both /--glucosides and /--galactosides but had no effect on either the a-or /--xylosides (33). Since, at the time, Fischer expected glycosides to be furanosides, he suggested that both the enzymes required the presence of a free hydroxyl group at position 5 of a hexoside. At this point Fischer concluded that the enzymes, in terms of the configurations of the substrates, are as fastidious as yeast and other organisms. He then returned to the above-mentioned hypothesis that he and Thierfelder had proposed (30) and concluded (32) that the protein substances known as invertin and emulsin, like the substrates whose hydrolyses they effected, were asymmetrically formed molecules. On the basis of this consideration, he came to the momentous lock and key concept for enzyme activity and commented as follows: The restricted effectsof the enzymes may therefore be explained by the assumption that the approach of the molecules that cause the chemical process can occur only in the case of a similar geometric shape. To use a picture, Z would like to say that enzyme andglucoside have tofit to each other like a lock and key in order to exert a chemical effect on each other.

v. INSIGHTS ON ENZYMESPECIFICITY Emil Fischer developed a strong interest in the structural requirementsfor enzyme activity as the result of effects of changes in the structures of the aand /--methyl glucosides on their properties as substrates for the enzymes invertin and emulsin, which, as we have seen, he had shown to be a-and /--glucosidases, respectively. As already mentioned, he was fascinated in 1895 by the fact that emulsin had no effect on either the a-or /--methyl xylosides (33). In a 1912 publication with Karl Zach (34), he reported that /--methyl6-deoxyglucosidewas hydrolyzed by emulsin and wrote: It appears to us very strange that the effectof the enzyme on the methoxylgroup at the other end of the carbon chain depends on the sixth carbon atom.

The following question, which appears to be the origin of the use of chemical synthesis to provide probes for the assessment of the structural requirements for complex formation, was asked: How will the enzyme behave $there is a carbon richer alkyl at the end of the chain?

Soon after his death in 1919, his colleagues, as coauthors, reported the results presented in Table VI.Since methyl 2-deoxyglucosidehad not been

14

RAYMOND U. LEMIEUX AND ULRIKE SPOHR

TABLE VI The Probing by Fischer of the Active Site of an Enzyme (Modem Formula)

Substituent

R

R’

CH,OH CH,OH H CH, CH,Br

OH H OH OH OH

Hydrolysis by emulsin

+ -

-

+ -

Reference

32 35 33 34 36

hydrolyzed either by emulsin or by enzymes of yeast extract, it was concluded (35) that the presence of an hydroxyl group at the 2-position plays an essential role in the lock and key mechanism. Other collaborators reported (36) in 1920 that, although emulsin readily cleavedp-methyl6-deoxy-d-glucoside (34), the enzyme had no effect on ~-methyl6-bromo-6-deoxy-d-glucoside. Thus, it appears that among Emil Fischer’s last thoughts was the consideration of molecular recognition and how synthetic methodologies may provide the means to a precise understandingat the molecular level of his lock and key concept for enzyme activity. This idea was examined by many others, but a proper understanding had to await the development of synthetic methodologies for the synthesis of oligosaccharidesand congeners thereof for the probing of protein combining sites. As already stated, Fischer was deeply intrigued by the phenomenon of enzyme activity. He realized that the substances were proteins and this undoubtedly was why he next undertook the study of amino acids and peptides. He fully appreciated that the specificity of enzyme catalysis depended on the occurrence of a complementarityfor interacting dissymmetric surfaces. In this regard, he wrote (3): This example (the cyanohydrin reaction) appears to me to provide a simple solutionfor the natural asymmetric synthesis. Theformation of the sugar, as the plant physiologists assume, occurs in the chlorophyll grain, which itserf is composed of optically active substances. . . . Thepreparedsugar is released and later on used by theplant, as is known,for thepreparation of other organic components. Their asymmetry is thus explainedfrom the nature of the building material. Of course, they also provide materialfor new chlorophyll

LOCK AND KEY CONCEPT FOR ENZYME SPECIFICITY

I5

grains, which again produce active sugars. In this way optical activity is passed onfrom molecule to molecule, such as Ige goes from cell to cell. Therefre it is not necessary to deduce the formation of optically active substances in the plant fromasymmetric forces that reside outside the organisms, as Pasteur presumed.

In a biography prepared for the I966 issue of Advances in Carbohydrate Chemistry, Karl Freudenberg wrote (5): Under his leadership,synthetic and theoretical chemistry was reunited with biochemistry, and a broad scientijc basis restored to organic chemistry.

Fischer never allowed synthesis to become an end in itself and thereby lose contact with general problems. In this regard, Freudenberg recorded ( 5 ) the following comment made by Fischer- probably incidentally-in 1904: Only6 ofthe32 heptosesandonly2 ofthe128nonoseshavebeenprepared.But, sincethese compounds have not yet been found in nature and are, therefre, of only minor interest, their systematic elaboration may be Idt for a later period.

Freudenberg further commented (5): He never gave unbridled rein to his synthetic efforts,nor did hefall into the temptation of purposeless synthesis. He always remained a true scientist-a student of nature. . . . This great individual was a man ofinflexible veracityand simplicity. . . .Emil Fischers life was based on responsibility: a responsibilityforthe austerityandpurity of his work and its aims, responsibilityfor the university as an important organ of our cultural and economic Ige, and responsibilityfor each of his students.

Although few are endowed with comparabletalent and energy, his career is surely a splendid example for all. Hopefully, like the passing of optical activity from molecule to molecule, this chapter will help induce some transfer of Emil Fischer’s way of thinking and actions to future generations of organic chemists. VI. CONCLUDING REMARKS It is appropriate to close this chapter with an illustration of how Emil Fischer’s lock and key concept has sincebeen found to be relevant to enzyme specificity. Thus, it will be seen that, in fact, specific structural featuresof the substrate act somewhat like the wards of a key. That is, insertion of the substrate into the enzyme’sactive site is strongly demanding in complementarity as is inserting a key into the barrel of a lock. Formation of the complex brings the structural features of the substrate and the enzyme that are to interact into close proximity and proper orientation. Thus, the organization and thermal energy required to achieve the transition state are greatly diminished and catalysis is effected. The hydrolysis of maltose and other a-linked glucosides by the commercial enzyme, which is most commonly

16

RAYMOND U. LEMIEUX AND ULRIKE SPOHR

referred to as glucoamylase,but is also known as amyloglucosidaseor simply AMG, serves well because not only is there much known about the lock and key characterization of the reaction pathway but also a consideration of this enzyme establishes a connectionwith Emil Fischer (37). It was in the course of an investigationof the effect of various enzyme preparations on cellobiose that the decision was made to examine an enzyme preparation derived from the fungus Aspergillus niger. It did not catalyze the hydrolysis of cellobiose, whereas emulsin did, and the conclusion was drawn that cellobiose must have a /%linkage. As the name implies, the enzyme is an amylase. It is an em-hydrolase that releases glucose from the nonreducing ends of starch and dextrins. Along with alpha amylases, glucoamylasesare fungal enzymes produced by a variety of Aspergillus species that have found major industrial importance for the production of high-glucose syrups and related applications (38). Their use in the brewing of beer was likely the reason that Fischer examined the glucoamylase that is produced by the fungus A. niger. The enzyme was examined by Fischer following earlier studies by Bourquelot (39) in France. Bourquelot (39) reported that he had made known in 1883 that an extract from a culture ofA. niger hydrolyzed maltose. He now provided evidence that the solution also caused hydrolysis of trehalose. However, since this activity was lost on heating to 63 “Cbut that responsible for the hydrolysis of maltose was maintained up to 75 “C,he concluded that the fungus produced two different “ferments” -one a maltase, the other a trehalase. Pazur and Ando (40) separated the glucoamylase from other carbohydrases of A . niger by ion-exchange chromatography. They later found that the enzyme also hydrolyzesisomaltosebut at a much lower rate than maltose (4 I). It was reasonable, therefore, to test whether or not the enzyme would hydrolyze a (1 + 3)-linked disaccharide of glucose. Consequently, nigerose was examined and found to hydrolyze at a rate intermediate to those of maltose and isomaltose. Thus, it became apparent that, although the enzyme has a high specificity for an a-D-glucopyranosyl disaccharide, the structure of the aglycon can be varied without total loss of activity. On the other hand, the enzyme was found to be quite ineffective for the hydrolysis of either methyl or phenyl a-D-glucopyranoside (42). The profound difference in rate was attributed to a difference in the conformations of the glucosyl units of maltose and methyl a-D-glucopyranoside. This postulation was made because it had been suggested (43) that the glucosyl unit of maltose was in a boat conformation. Because methyl a-D-glucopyranoside was believed to favor a chair form, it was considered that the difference in rate of hydrolysis was related to the ease with which the glucosyl units in the compounds can transform into the conformation preferred by the enzyme (42). However,

LOCK AND KEY CONCEPT FOR ENZYME SPECIFICITY

17

with the advent of high-resolution nuclear magnetic resonance it became evident (44)that the glucosyl units of maltose are held extensively in the same ‘C,(D)chair conformationas methyl a-D-glucopyranosideand, therefore, another explanation had to be sought. The answer was provided by Bock and Pedersen (49,who studied the effect of deoxygenation at the various hydroxyl-group positions on rate of hydrolysis. Lemieux (46) recently reviewed the studies with his coworkers on the origin of the specificity in the recognition of oligosaccharidesby proteins. In the course of these studies, it became evident that hydroxyl groups are invariably involved in providing the stereochemical complementarity required for the binding of an oligosaccharide by monoclonal antibodies and lectins. However, binding studiesusing monodeoxy derivativesrevealed that only some of these hydroxyl groups establish essential polar interactionswith the protein. These were termed the key hydroxyl groups. On applying this technique to the hydrolysis of maltose by the amyloglucosidase ofA. niger, Bock and Pedersen found that hydroxyl groups on both of the glucose units were essential for efficient catalysis. The key hydroxyl groups proved to be OH-3 of the reducing glucose unit and OH-4‘ and OH-6’ of the nonreducing glucose unit of maltose. The two key hydroxyl groups ofthe nonreducing unit must establish polar interactionsthat tend to anchor this glucose unit in the enzyme’s active site. That methyl a-D-glucopyranoside is not a good substrate was no longer surprisingbecause this compound cannot provide the key hydroxyl group of the aglycon. Since maltose, isomaltose, and nigerose are all substrates, it became apparent that the catalysis could entertain structural variations in the aglycon as long as it can project an hydroxyl group toward the active site in a manner similar to OH-3 of maltose. Indeed, as may be seen in Table VII, conformational analysis of these disaccharidesindicates that in each case a similar disposition of the three key hydroxyl groups can be achieved with relative ease. In a sense, these hydroxyl groups perform as the wards of a key and thereby provide a fine illustration of Emil Fischer’s lock and key concept for enzyme specificity. The turning of the key once the complex has formed is a separate issue. In this regard, Lemieux (47) has pointed out that rotation about the glycosidic bond must weaken the exo-anomericeffect and thereby importantly activate the anomeric carbon to nucleophilic attack. Therefore, it seems likely that the role of the key hydroxyl group of the aglycon is to accommodate the rotation prior to the attack by water to formp-D-glucopyranose,which is the first product of the reaction. It appears that the overall mechanism for the hydrolysis of maltose by glucoamylasewill soon be delineated. A brief summary of how this is being accomplished deserves comment.

18

RAYMOND U. LEMIEUX AND ULRIKE SPOHR

TABLE VII The Three Key Hydroxyl Group# (Bold Letters) Necessary for Efficient Hydrolysis of an a-D-Glucopyranosideby the Glucoamylase of AsprgZus niger Structures

HO Ho+ Maltose4’ *

O

Nigerose4’

g0-

OH

OH

HoHO %

HO Ho+

OH H o = HO

OH

a,$-TrehaloseM

OH

no Ho+

OH - 0 p-Kolibio~s~~

F-$! -$&+o H HO

OH

0

Isornaliose4’

HO

a Lemieux and Bock (44) have pointed out why the two-dimensionalstructuralformulas used to represent the various disaccharides provide a useful approximation of the conformational preference.

Svensson and coworkers (48) were able to separate commercial A. niger glucoamylase into two catalytic components, which were termed G1 and G2. It is now established (49) that G2 is a proteolytic fragment derived from GI. Meagher and Reilly (50) showed the two forms to behave similarly and this finding appears common to all major variants of amyloglucosidasethat contain the catalytic domain. The amino acid sequences of GI and G2 had been established by Boel et al. (5 1). Using glucoamylase-specific synthetic oligonucleotides and molecular cloning of the complementary DNA synthesized from A. niger, the primary structure of the mRNA for the GI

LOCK AND KEY CONCEPT FOR ENZYME SPECIFICITY

19

enzyme was established. In vitro translations of the mRNA followed by immunoprecipitations with glucoamylase-specific antisera showed that both GI and G2 were in the culture medium. Thus, a goal contemplated by Emil Fischer, that is, the synthesis of an enzyme that he had examined (37), w as accomplished. Aleshin and coworkers (49) have reported the X-ray crystal structure at 2.2-A resolution of a G2-type variant produced by Aspergillus awamori. Meanwhile, an attempt was made to determine the amino acid residues that participate in the substrate binding and catalysis provided by G2 of A. niger (52). The results of the chemical approach indicated that the Asp-176, Glu179, and Glu-180 form an acidic cluster crucial to the functioning of the enzyme. This conclusion was then tested by site-specific mutagenesis of these amino acid residues, which were replaced, one at a time, with Asn, Gln, and Gln, respectively (53). The substitution at Glu-179 provided an inactive protein. The other two substitutions affected the kinetic parameters but were not of crucial importance to the maintenance of activity. The crystal structure (49) supports the conclusionthat Glu- 179 functions as the catalytic acid but Asp- 176does not appear to be a good candidate for provision of catalytic base. Thus, there still exists considerable uncertainty as to how the disaccharide is accepted into the combiningsite for hydrolysis. Nevertheless,the kind of scheme presented by Svensson and coworkers (52)almost surely prevails. As already mentioned, the glucoamylase project was chosen to illustrate Emil Fischer’s lock and key concept for enzyme specificity. It is seen that his vision has become unequivocally established. Many other developments could have been chosen, as can be appreciated from recent reviews by Hehre (54) and by Svensson (55). Cornforth (56) provided a fine overview of asymmetry and enzyme action in his Nobel prize lecture. Noteworthy is the conclusion that “stereospecificityis something not just incidental, but essential to enzyme catalysis.’’ In other words, the key must fit the lock. REFERENCES 1. K. Hoesch, Ber., 54 (1921) 3-480.

C. S. Hudson, J. Chem. Educ., 18 (1941) 353-357. E. Fischer, Ber.. 27 (1894) 3189-3232. E. Fischer, Ber., 24 (1891) 1836-1845,2683-2687. K. Freudenberg, Adv. Carbohydr. Chem., 21 (1966) 2-38. D. J. Cram, in J. L. Seeman (Ed.), From Design to Discovery, pp. 12-13. American Chemical Society, Washington, D.C., 1990. 7. L. Pasteur, Cornpt. Rend., 51 (1860) 298-299. 8. F. G. Riddell and M. J. T. Robinson, Tetrahedron, 30 (1974) 2001 -2007. 9. (a) Renk Dubos, Pasteur andModern Science, p. 28. Springer-Verlag, New York, 1988; (b) J. W. Cornforth, InterdisciplinarySci. Rev.,9 (1984) 107- 112. 10. E. Fischer, Untersuchungen iiber Kohlenhydrate und Fermente (1884- 1908). J. Springer, Berlin, 1909.

2. 3. 4. 5. 6.

20

RAYMOND U. LEMIEUX AND ULRIKE SPOHR

1 1. J. Liebig, Ann. Chem. Phurm., 153 (1870) 1-47. 12. E. Buchner and R. Rapp, Ber., 30 (1897) 2668-2678. 13. G. M. Richardson, The Foundations of Stereo Chemistry, Scientific Memoirs Series. American Book Co., New York, 1901. 14. C. S. Hudson, Adv. Curbohydr. Chem., 3 (1948) 1-21. 15. F. W. Winckler, Ann. Phurm., 4 (1832) 242-247. 16. H. Kiliani, Ber., 19 (1886) 767-772. 17. H. Kiliani, Ber., 22 (1889) 521-524. 18. E. Fischer, Ber., 23 (1890) 2114-2141. 19. E. Fischer, Ber., 23 (1890) 3684-3687. 20. E. Fischer, Ber., 23 (1890) 261 1-2624. 21. E. Fischer and F. Passmore, Ber., 23 (1890) 2226-2239. 22. E. Fischer and R. Stahel, Ber., 24 (1891) 528-539. 23. E. Fischer, Ann. Chem., 270 (1892) 64- 107. 24. W. Kiihne, Unters. Physiol. Instifut Univ. Heidelberg, 1 (1878) 291. 25. A. Payen and J. F. Persoz, Ann. Chim. (Phys.), 53 (1833) 73-92. 26. M. 0. Forster, J. Chem. Soc., 1 17 (1920) 1 157- 1201. 27. E. Fischer and J. Hirschberger, Ber., 22 (1889) 3218-3224. 28. E. Fischer and J. Hirschberger, Ber., 22 (1889) 365-376. 29. E. Fischer, Ber., 23 (1890) 370-394. 30. E. Fischer and H. Thierfelder, Ber., 27 (1894) 2031 -2037. 31. E. Fischer, Ber., 27 (1894) 3479-3483. 32. E. Fischer, Ber., 27 (1894) 2895-2993. 33. E. Fischer, Ber., 28 (1895) 1429- 1438. 34. E. Fischer and K. Zach, Ber., 45 (1912) 3761-3773. 35. E. Fischer, M. Bergmann, and H. Schotte, Ber., 53 (1920), 509-547. 36. E. Fischer, B. Helferich, and P. Ostmann, Ber., 53 (1920) 873-886. 37. E. Fischer and G. ZemplBn, Ann. Chem., 365 (1909) 1-6. 38. B. C. Saha and J. G. Zeikus, Starch/Starke, 41 (1989) 57-64. 39. E. Bourquelot, Compt. Rend. 116 (1893) 826-828. 40. J. H. Pazur and T. Ando, J. Biol. Chem., 234 (1959) 1966- 1970. 41. J. H. Pazur and T. Ando, Arch. Biochem. Biophys., 93 (1961) 43-49. 42. J. H. Pazur and K. Kleppe, J. Biol. Chem., 237 (1962) 1002- 1006. 76 (1954) 4595-4598. 43. R. E. Reeves, J. Am. Chem. SOC., 44. R. U. Lemieux and K. Bock, Arch. Biochem. Biophys., 22 1 (1983) 125- 134. 45. K. Bock and H. Pedersen, Actu Chem. Scund., B41 (1987) 617-628. Rev., 18 (1989) 347-374. 46. R. U. Lemieux, Chem. SOC. 47. R. U. Lemieux, in J. I. Seeman (Ed.), Explorations with Sugars: How Sweet It Was,p. 100. American Chemical Society, Washington, D.C., 1990. 48. B. Svensson, T. G. Pedersen, I. Svendsen, T. Sakai, and M. Ottesen, Carlsberg Res. Commun., 47 (1982) 55-69. 49. A. Aleshin, A. Golubev, L. M. Firsov, and R. B. Honzatko, J. Biol. Chem., 267 (1992) 19291- 19298. 50. M. M. Meagher and P. J. Reilly, Biofechnol.Bioeng., 34 (1989) 689-693. 51. E. Boel, I. Hjort, B. Svensson, F. Norris, K. E. Norris, and N. P. Fiil, EMBO J.,3 (1984) 1097- 1102. 52. B. Svensson,A. J. Clarke, I. Svendsen,and H. Mraller,Eur. J. Biochem., 188( 1990)29 - 38. 53. M. R. Sierks, C. Ford, P. J. Reilly and B. Svensson, Protein Eng., 3 (1990) 193- 198. 54. E. J. Hehre, Denpun Kagaku, 36 (1989) 197-205. 55. B. Svensson, Denpun Kugaku, 38 (1991) 125-135. 56. J. W. Cornforth, Science, 193 ( 1976) 121- 125.

ADVANCES IN CARBOHYDRATE CHEMISTRY AND BIOCHEMISTRY, VOL. 50

ANOMERIC-OXYGEN ACTIVATION FOR GLYCOSIDE SYNTHESIS: THE TRICHLOROACETIMIDATE METHOD BY RICHARD R. SCHMIDT* *Fakultatfur Chemie, Universtitat Konstanz, 0-7750Konstanz, Germany

WILLYK I N Z Y ~

t Zentrale Forschungslaboratorien,CIBA- GEIGY AG. CH-4002 Basel, Switzerland I. GENERAL INTRODUCTION TO GLYCOSIDE SYNTHESIS: ACTIVATION THROUGH ANOMERIC OXYGEN-EXCHANGE REACTIONS The biological significance of glycoconjugates has stimulated much synthetic activity in glycoside synthesis in the past years (1 -7). These efforts were initially concentrated mainly on improvements of the well-known Koenigs-Knorr method (8), introduced in 1901, which requires an exchange of the anomeric hydroxylgroup by bromine or chlorineas the first step (generation o f the glycosyl-groupdonor). The second step involvesglycosylgroup transfer to the glycosyl acceptor in the presence of a heavy metal ion promoter (Scheme 1, path B). Although this is the basis of a very valuable methodology that has been reviewed extensively (4,5,7), several inherent disadvantages make the Koenigs- Knorr method often experimentally demanding and certainly not very suitable for large-scale preparations. For instance, the requirement of at least equimolar amounts ofthe heavy metal salt promoter, often incorrectly termed “catalyst,” is a limiting factor (1 - 3). Therefore, alternative methods are o f interest. Other anomeric-oxygen exchange reactions have been recently investigated quite extensively. Closely related to the Koenigs- Knorr method is the introduction offluorine as the leaving group (Scheme 1, path B) (6,9 - 13). Because of the difference in halophilicity of this element as compared with bromine and chlorine, additional promoter systems besides silver salts were found useful as activators for glycosylation reactions (14 - 16). However,

21

AU

Copyright 0 1994 by Academic preSr Inc. rights of reproduction in any form reserved.

RICHARD R. SCHMIDT AND WILLY KINZY

22

Glycosyl Donors via Anomeric-Oxygen Exchange Reactions

/

e

Activation through Retention of the Anomeric Oxygen

\

H

Hal*Ha,

@

H \

Fischer-Helferich (Acid Activation)

@ Koenigs-Knorr; Br, CI, (I) Act. F-Activation RS-Activation

@

Anomeric 0-Alkylation: Base Activation

@

Trichloroacetirnidate (Imidate Act.) S02R - Act. PO(OR)2-Act.

SCHEME 1. -Synthesis of Glycosides and Saccharides.

because of the generally lower donor properties of glycosyl fluorides (17) these intermediateshave not yet gained wide application in the synthesis of complex glycoconjugates. Thioglycosides,where the anomeric oxygen atom is replaced by an alkyl or arylthio group, have recently attracted considerable attention as glycosyl donors (Scheme 1, path B) (5,18,19). They offer sufficienttemporary protection of the anomeric center and present several alternative possibilities for regioselective activation to generate glycosyl donor properties. Earlier methods for activation include mainly mercury(II), copper(II), and lead(I1) salts (20 - 28). However, besides the requirement of generally more than equimolar amounts of heavy metal salts, relatively low glycosyl-donor properties were experienced with these systems. This problem was partly overcome by the use of heterocyclic thioglycosides (2 1,23,25- 27). In addition to metal salts, bromonium and chloronium ions are also highly thiophilic and thus provide with counter-ions of bromide and chloride, respectively, the corresponding glycosyl halides for a subsequent Koenigs-Knorr type of reaction ( 18,19,29).If the counter-ion of the halonium ion is a poor nucleophile (for instance, succinimidefrom N-bromosuccinimide),then direct reaction with alcohols as competing nucleophiles is favored and thus leads to 0-glycosides. However, low a,/3 selectivities are frequently obtained for nonneighboring

ANOMERIC-OXYGEN ACTIVATION FOR GLYCOSIDE SYNTHESIS

23

group-assisted reactions (23,30). Formation of sulfoniumions from thioglycosides by the action of methyl triflate was also successfully applied to 0-glycosidebond-formation(3 1- 34). Disadvantagesof this method include the low a,@selectivity observed for nonparticipating2-0-protective groups, the health hazard of methyl triflate, and the formation of methylation products in other side reactions. The recently introduced activator dimethyl(methy1thio)sulfoniumtriflate (DMTST) proved to be highly thiophilic and gives rise to faster glycosylation than does methyl triflate (35). However, again, with nonparticipatinggroups the a,@selectivity is usually low (32). Radical activation of thioglycosides has also been recently reported, providing similar results in terms of yield and diastereoselectivity (36). The Fischer - Helferich method, as a direct anomeric-oxygen replacement reaction (Scheme 1, path A), has been very successfullyapplied for syntheses of simple alkyl glycosides. However, because of its reversibility, it has not gained general importance in the synthesis of complex oligosaccharides and glycoconjugates (1). 11. ANOMERIC-OXYGEN ACTIVATION: ANOMERIC 0-ALKYLATION 1. Introduction

The requirementsfor glycoside syntheses, high chemical and stereochemical yield, and applicability to large-scale preparations were not effectively met by any of the methodsjust described. However, it seems that the general strategy for glycoside synthesis is reasonable:

(i) The first step should consist of a sterically uniform activation of the anomeric center with formation of a stable glycosyl donor having either the a or the p configuration; (ii) The second step should consist of a catalyzed, sterically uniform, irreversible glycosyl transfer to the acceptor, proceeding with either retention or inversion of configurationat the anomeric center in high chemical yield and without affecting other bonds. Only simple means meeting these requirements will lead to a generally acceptable and useful methodology. Therefore, besides acid activation (Scheme 1, paths A and B), the simplest form of activation would be base activation generating first an anomeric alkoxide structure of a pyranose or furanose (Scheme 1, paths C and D). This approach is especially tempting because Nature has a similar approach for generating glycosyl donors, namely glycosyl phosphate formation [see Section IV.2 and Ref. (17)J.

RICHARD R. SCHMIDT AND WILLY KINZY

24

2. Anomeric 0-Alkylation

The direct 0-alkylation of the anomeric center (Scheme 1, path C) by treatment of furanoses and pyranoses with base and then with simple alkylating agents, for instance an excess of methyl iodide or dimethyl sulfate, has long been known (1,3). Surprisingly, no studies employing this simple method for synthesesof more-complex glycosidesand saccharideshave been reported prior to our work (1,37,38). In the beginning, direct anomeric 0-alkylation seemed very unlikely to fulfill all of the requirements for glycoside and saccharide synthesis. Even when all remaining functional groups (generally hydroxyl groups) are blocked by protecting groups, the ring- chain tautomerism between the anomeric forms and the open-chain form (Scheme 2) already gives three

a RO

R

o

e

RO

,

H 7

RR O

O

ROO

I I

Fl-X

1

I

a 0OR

RO

X = mainly OTf

SCHEME 2. - 1-0-Alkylation and 1-0-Acylation (Irreversible Reactions).

possible sites for attack of the alkylating agent. In addition, base-catalyzed elimination in the open-chain form of the sugar could be a destructive side-reaction. Therefore, the yield, the regioselectivity, and the stereoselectivity of such direct anomeric 0-alkylation would not generally be expected to be outstanding. In any event, the process should be governed at least by the following factors:

(i) the stability of the deprotonated species; (ii) the ring-chain tautomeric equilibrium and its dynamics; and (iii) the relative reactivities (nucleophilicities)of the three O-deprotonated species.

ANOMERIC-OXYGEN ACTIVATION FOR GLYCOSIDE SYNTHESIS

25

Because of the irreversibility of the 0-alkylation reaction, kinetic regioand stereo-control is required for selective product-formation. Therefore, selective formation of either a or p product seemed to be unattainable. The first experiments with iodide derivatives of carbohydrates revealed that better alkylating agents are required (37). However, excellent reactivity with corresponding trifluoromethanesulfonates (triflates) was observed, providing, for instance, with 2,3-0-isopropylidene-~-ribose and derivatives, depending on the reaction conditions, very high yields of either a- and P-linked disaccharides (37). Surprisingly, even partial 0-protection or, as recently discovered, 0-nonprotection was compatible with this reaction (39 -44). The stereocontrol could be effected by intramolecular metal-ion complexation, by steric effects, and by taking advantage of the increased nucleophilicity of the equatorial anomeric oxide over the axial anomeric oxide [kinetic anomeric effect (45,46)]. This method could even be employed in selectiveformation of a-glycosidesof Kdo (47,48). Thus, the direct anomeric 0-alkylation constitutes an especiallysimple procedure for glycoside and saccharide synthesis, giving generally high yields and diastereoselectivities. The limitation to primary triflates was a major drawback for the general use of this anomeric 0-alkylation in glycoside synthesis. However, this problem was recently overcome, at least in part, by modifying the reaction conditions (49). 111. ANOMERIC-OXYGEN ACTIVATION: THETRICHLOROACETIMIDATE METHOD 1. Formation of 0-Glycosyl Trichloroacetimidates

Aside from direct anomeric 0-alkylation (Scheme I , path C), base-catalyzed transformation of the anomeric oxygen atom into a good leavinggroup (Scheme 1, path D) should be easily readily effected. Therefore, it is not surprisingthat several approaches have been directed toward this goal, as will be discussed later (Section IV). However, stable and concomitantly reactive intermediates were never obtained for the separate anomers. Obviously, for achievement of stereocontrolled activation of the anomeric oxygen atom, the anomerization of the anomeric hydroxyl group or the anomeric oxide ion, respectively, has to be considered (Scheme 2). Thus, in a reversible activation process and with the help of kinetic and thermodynamic reaction-control, possibly both activated anomers should be accessible. These considerations led us to the conclusion that suitable triple-bond systems A=B (or compounds containing cumulative double-bond systems A=B=C) might be found that add pyranoses and furanoses under base

RICHARD R. SCHMIDT AND WILLY KINZY

26

catalysis directly and, because of reversibility, in a stereocontrolled manner (Scheme 1, path D) (1 - 3); thus, both activated anomers may be obtainable at will. However, the instability of open-chain aldehydic intermediates in basic media and the insufficient or undifferentiated reactivities of the a-and p-anomeric oxides lowered the expectations for stereocontrolled anomeric 0-activations along these lines. The desired formation of stable anomeric 0-activated intermediates via base catalysis requires a different catalytic system for reactivity in the subsequent glycosylation step. Therefore, after base-promoted trapping of anomeric 0-activated intermediates (first step), mild acid treatment in the presence of acceptors, leading to the formation ofglycosides(namely, acetals and derivatives) in an irreversible manner (second step), would constitute the simple means of catalysis desired for a new and efficient glycosylation method. These demands have to be considered in the selection of A=B (or A=-). Thus, the stable intermediates obtained in the first step have to exhibit by appropriate choice of the centers A and B (or A, B, and C) good glycosyl-donor properties in the presence of strictly catalytic amounts of acid. The water liberated upon glycosideformation is then transferred in two separate steps to the activating agent A=B (or A=B=C), thus providing the driving force for the glycosylation reaction (Scheme 3). This concept fulfills

Base (cat.)

Ro& RO

OH

+

,

AGE

, RO

OR

OR

h

If

\ \ -H20 \

Acid (cat.) \ \

HoR'

a

RORO

OR'

OR

SCHEME3. -Steps in the Glycosylation Reaction.

the requirementsjust given for an efficient glycosylationmethodology: truly catalytic amounts of a simple base (first step) and of a simple acid (second step) are required for anomeric 0-activation and promotion of the glycosylation, respectively; liberated water will not compete with the glycosyl acceptor for the glycosyl donor because it is concomitantly chemically bound

ANOMERIC-OXYGEN ACTIVATION FOR GLYCOSIDE SYNTHESIS

27

to the activating species A=B (or A=B=C); and thus, reversibility in the first and irreversibility in the second step provide important means for controllingthe yield and stereochemistry of the anomeric 0-activated intermediate and of glycoside-bond formation. The trichloroacetimidate method developed by us, and recent contributions from other laboratories to this methodology, have proved the validity of this concept ( 1- 3). Electron-deficient nitriles, such as for instance trichloroacetonitrile and trifluoroacetonitrile (AEB: A = N; B = CCCl,, CCF,), are known to undergo direct and reversible,base-catalyzed addition of alcoholsproviding 0-alkyl trichloroacetimidates(1,50). This imidate synthesis has the advantage that the free imidates can be isolated as stable adducts, which are less sensitive to hydrolysis than their corresponding salts. A detailed study of the addition of trichloroacetonitrileto 2,3,4,6-tetra-Obenzyl-D-glucose (la, Scheme 4) revealed (1 - 3,45) that, from the equatorial

RO

OH

CCI3CN.

l a : R=Bn lb: R=AC

It

Base

RO RO

~ RO ~

~RO o

RooYNH

c

c

NH

CC!,

m.p. [OC]

1a-a 1b-a

y

1a-p 1b-P

1b-a

SCHEME 4.-Addition of CCl, CN at the Anomeric Position.

1-oxide ion, the P-trichloroacetimidate la$ is generated preferentially or even exclusively in a very rapid and reversible addition-reaction (Schemes 2 and 3). However, this product anomerizesin a slow, base-catalyzedreaction (via retroreaction, anomerization of the 1-oxide ion, and renewed trichloroacetonitrile addition) to the a-trichloroacetimidate la- having the electron-withdrawing 1-substituent in an axial disposition, as favored by the

1

3

RICHARD R. SCHMIDT AND WILLY KINZY

28

thermodynamically operating anomeric effect. Thus, with different bases [for instance K2C03,Cs2C03and NaH or 1,8-diazabicyclo[5,4,0]undec-7ene (DBU)] both 0-activated anomers may be isolated in pure form and in high yields via kinetic and thermodynamic reaction-control. Both anomers are commonly thermally stable and may be stored easily. A similar result was TABLEI Synthesis of Trichloroacetimidatesof DGlucose

Reaction conditions

Compound"

Anomeric config.

Yield

(a$)

(%)

Ref.

~

...-&$

CHzCl2, NaH, CCl-,CN, room temp.

1 :o

78

65,66

Bno-oycci, BnO

CH2C12, K2CO3, CC13CN, room temp.

0: 1

90

46,67,68

CH,Cl,, K,CO,, CCl,CN, 48 h, room temp.

l:o

98

66

CHZCl,, K,COp, CC13CN, 2 h room temp.

0: I

78

46

CH,Cl,, NaH, CC13CN, room temp.

l:o

90

58a,61

CH2Cl2, K2CO3, CCl,CN, 6 h, room temp.

I :3

74

58a

CH,Cl,, NaH, CC13CN, 1.5 h, room temp.

1 :o

60

69,70

En0

1a-b

En0 NH

lb-a

AcO

/OAC

1b-P

A

c

O

~

O o

y

CCI, YNH C

C

l

3

AcO /

OAC

NH

O Y N H CCl,

/OPiV

idsa

PivO

CCI,

r?

Bn, benzyl; Piv, pivaloyl; Ac, acetyl.

ANOMERIC-OXYGEN ACTIVATION FOR GLYCOSIDE SYNTHESIS

29

obtained for the less reactive 0-acetyl derivative 1b of D-glucose, providing trichloroacetimidates lb-a and lb-p, respectively (see Table I). The higher nucleophilicityof the p-oxide ion may be attributed to a steric factor in combination with a kinetically effective stereoelectroniceffect that results from repulsions of lone electron pairs, dipole effects, or both (Scheme 5 ) (45,46). This effect should be more pronounced in anomericp-oxide ions (a) Dipole-Dipole Interaction

-0

(b) Lone-Pair Orbital Interaction

SCHEME5 .-Enhanced Nucleophilicity of poxides (Kinetic Anomeric Effect).

than in p-pyranosides because of the difference in the number of oxygen lone-pair orbitals and the difference in their relative energies. In addition, this kinetic anomeric efect should be particularly efficient in the p-mannopyranosyl oxide ion, where the thermodynamic anomeric effect, favoring the a-anomer, is also stronger. This expectation could be experimentally confirmed in the irreversible anomeric 0-alkylation of mannopyranose, which leads in nonpolar solvents preferentially to p-glycoside formation (see references in Section 11.2). However, in the reversible trichloroacetimidate formation, the stronger thermodynamic anomeric effect results in much faster generation of the a-trichloroacetimidate,and therefore trapping of the p-species becomes much more difficult. Thus, a distinction between the thermodynamic and the kinetic anomeric effect could be experimentally verified.

RICHARD R. SCHMIDT AND WILLY KINZY

30

The stereoselectiveanomeric 0-activation of carbohydrates and their derivatives via 0-glycosyl trichloroacetimidate formation is capable of extension to all important hexopyranoses (Glc, Gal, Man, Fuc, Rha, Qui, GlcN, GalN), hexofuranoses, pentopyranoses, and pentofuranoses, as well as to glucuronic acid, galacturonic acid, and muramic acid; to 2-deoxy-urubinohexose derivatives; and to many di-, tri-, and oligo-saccharides(see Section 111.3). It commonly provides stable compounds in a stereocontrolled manner. Thus, the requirements put forward for the first step, namely, efficient stereocontrolled formation of stable glycosyl donors, are fulfilled (see Section 11.1). Ultimately the significance of the 0-glycosyl trichloroacetimidates must be based solely on their glycosylation potential under mild acidic catalysis. This potential has indeed been confirmed overwhelminglyin various laboratories and is presented in comprehensive detail in this article. 2. Reaction with Brsnsted Acids

The trichloroacetimidatemethod for glycoside synthesis extended its versatility right at the outset ( 5 1,52a) by exhibiting an especially smooth reaction of 0-(glycosy1)trichloroacetimidateswith Brnnsted acids. Without the addition of any catalyst, simple Brnnsted acids are able to substitute the trichloroacetimidate group at room temperature in high yields, as shown (17) for la-a in Scheme 6. Because of anomerization of possible p products x la-a

+

HX (excess)

=

CI (90 %)

CHzCIz,RT X = F (88 O h ) (HX : Py ' HF) X = Na (90%)

SCHEME6. -Substitution of the Trichloroacetimidate Group by Simple Brmsted Acids.

formed at the beginning of the reaction, only a products are finally isolated in these instances. Carboxylic acids, being weaker acids, react with la-a with inversion of configuration at the anomeric center to yield B-0-acyl compounds (1,53). This mild and convenient method for 1-0-acylationof carbohydratesis also useful for pharmacological drug modification (54) or for the resolution of carboxylic acids (53). Accordingly, phosphoric acid mono- and di-esters permit uncatalyzed glycosyl transfer from 0-(glycosy1)trichloroacetimidates (52a,55 57,58a,58b). The reaction is thus very useful in the synthesis of glycophospholipids (1,55), which are important constituents of cell membranes (1). Commonly, direct phosphorylationat the anomeric hydroxyl group leads to

%

~ RO~

RORO

~

o OR

1a.a

c

c

NH

la-p

Possible Transition States for Diastereospecific Phosphate Formation

RoO YCCI, N H

y

,, '4

CH,CI,

RT, 1 h

( 95 % )

Acid-catalyzed a-Anomerization F 0

SCHEME 7. -Reaction of a-and &Trichloroacetimidates with Dibenzyl Hydrogenphosphate.

l

;

RICHARD R. SCHMIDT AND WILLY KINZY

32

low a,p-selectivity. However, with the 0-(glycosyl)trichloroacetimidates,a high a or p selectivity is observed and anomerization proceeds only in the presence of strong acids. Therefore, the generation of cyclic transition-states was proposed (56) (Scheme 7), which results in an S N type ~ of configurational inversion. Calculations on the basic structures of the a- and p-trichloroacetimidates,respectively, exhibit ground-state conformational preference for conformers that support the intermediate generation of these cyclic transition states. Presumably, the cyclictransition-stateis not a planar ring of eight atoms, because the calculated dihedral angles of the ground states deviate considerably from such a possibility, but rather resembles a chairlike transition-state with two long bonds constituting the 0 . . C - SOandN. . H - - -0connections. Thus, all systems having related A=B-C-H geometry may react via the cyclic transition-state proposed for phosphoric acid derivatives and therefore exhibit high diastereoselectivity.Accordingly,in addition to the carboxylic acids and the phosphoric acids already mentioned, phosphonic (59)and phosphinic acids (59), monoalkylsulfuricacid (56,60), and even a-pyridone (17,56) exhibit the same reaction behavior. However, the 2-pyridyl pglycoside thus obtained from la-a is subsequentlytransformed via the same kind of pyridone attack into the corresponding 2-pyridyl a-glycoside (17,6 1). This finding may also explain anomerization to the thermodynamically more stable product in formation of glycosyl phosphates (56). Further development of this idea led to the proposal (56) that reactive B=C groups, for instance carbonyl systems, would be able to activate alcohol acceptors AH by generating a related A-B-C-H intermediate (Scheme 8, path I). It seemed that chloral might act as a catalyst along these lines. However, it turned out that the rate of decay in the transition state is too low in all systems tested thus far. Therefore, the carbonyl compound is more or less a substitute for a Lewis acid catalyst, as indicated in Scheme 8, path 11. The high reactivity and diastereoselectivity in chloral-catalyzedreactions is attributable to the nitriles used as solvents in these reactions [see Section III.3.b and Ref. (62)].

-

3. Alcohols and Sugars as 0-Nucleophiles a. Introduction. -The synthesis of oligosaccharides is characterized, because of the various connections, anomericconfiguration,and branching, by a much larger number of possibilitiesfor coupling than that of other natural biopolymers, such as peptides or proteins, and ribo- or deoxy-ribonucleotides. Comparison of the number of possible isomers with those of the correspondingpeptides and nucleotidesimpressively illustrates this point as indicated earlier (1). The wide structural variety renders sugars and, in par-

K '

33

34

RICHARD R. SCHMIDT AND WLLY KINZY

ticular oligosaccharides,ideal as carriers of biological information, encoding considerably more information per building block than proteins and nucleic acids. This great structural variety, however, complicates the specific biosynthesis of complex oligosaccharides.In general, the formation of each saccharide linkage requires specific enzymes (“one linkage -more than one enzyme”); and thus, in comparison with the enzymic synthesis of proteins and nucleic acids, much more effort is needed. The chemical synthesis of oligosaccharidesis also more complicated than the synthesis of other biopolymers, because the construction of each individual oligosaccharide poses a new challenge, requiring a knowledge of methods, together with experience and experimental skill. Thus, there are no universal methods available neither for biological in vivo nor for chemical in vitro syntheses. In synthesisof a disaccharide, two polyfunctional sugar components must be specifically linked. Therefore, the reactivity and the disastereoselectivity of the glycosyl-donorspeciesand the regioselectivity(that is, differentiation of the reactivities) at the glycosyl-acceptorspecies are important prerequisites for success. Protection strategies and suitable procedures for activation of the anomeric carbon atom are required; in addition, the coupling step must occur diastereoselectivelywith respect to formation of an a or P linkage. The high glycosylation potential of variously protected 0-glycosyl trichloroacetimidates, their excellent a/P diastereoselectivity generally found, and their high regioselectivity often observed with partially O-protected sugar acceptors will be documented here. b. 0-Glucosyl Trichloroacetimidates as Donors.-D-Glucose (63,64) plays a central role in the formation of plant polysaccharides (for instance, such homoglycans as cellulose and starch). Also, the heteroglycan repeatingunits of many bacterial, plant, and animal polysaccharides contain glucose in a- and P-glycosidiclinkage. As a constituent of the oligosaccharide moieties of glycosphingolipidsand glycoproteins, D - ~ ~ U C OisS ~less frequently encountered.Glycosphingolipidscontain D-glucosein the core region, where it is P-glycosidically linked to ceramide. In N-glycoproteins, a-linked ~ - g l u cose is a terminating signal in the biosynthesis of the complex oligosaccharide chains; fully developed glycoproteins do not contain glucose. The synthesis and application of 0-glucosyl trichloroacetimidatesis focused on 0-benzyl- and 0-acetyl protected derivatives( 1,52a)because these two protective groups have proven to be the most valuable in glycoside synthesis. Representative examples of trichloroacetimidate formation are collected in Table I (la- la). As already outlined (Section 111.I), the glucosyl trichloroacetimidates are obtained in high yields and the diastereoselectivi-

ANOMERIC-OXYGEN ACTIVATION FOR GLYCOSIDE SYNTHESIS

35

ties observed for the base-catalyzed addition of the 1-0-unprotected glucose derivatives to the electron-poor trichloroacetonitrile are remarkable, thus providing a- and p-glucosyl trichloroacetimidates, respectively, depending mainly on the reaction conditions. The reaction conditions have not yet been optimized in all examples described here and in subsequent sections; this is partly attributable to the fact that a,p-mixtures can be tolerated in glycosylation reactions when neighboring-group participation controls the diastereoselectivityin glycoside-bond formation. For reaction as 0-nucleophileswith 0-glycosyl trichloroacetimidates,alcohol components generally require the presence of an acid catalyst (1 - 3). Boron trifluoride etherate (BF3 * OEtJ at -40°C to room temperature in dichloromethane or dichloromethane- n-hexane as solvents and trimethylsilyl trifluoromethanesulfonate(Me3SiOTf)at - 80°Cto room temperature in ether or acetonitrile, respectively, as solvents have proved to be eminently suitable (52a,62). This is exemplified by the reactions of la-a and la$ with various acceptors (Tables I1 and 111). It should be noted that the results reported in the tables generally have not been optimized. Obviously, even at low temperatures, la-a exhibits high glycosyl-donor activity, thus providing generally the p-products in high yields and diastereoselectivities (Table 11, reactions with 2A-2M). The reaction of 2E exhibits that the (much less reactive) thioglycosides are not affected; therefore the products obtained may be used immediatelyfor further glycosylations. However, glycosylation of 2E with the correspondingglucosyl fluoride as donor was not successful. The low diastereoselectivityfound (71) for the reaction with acceptor 2D is rather unexpected. It may be due to the use of trifluoromethanesulfonicacid as catalyst, which as a Brnrnsted acid should interfere differently with the donor la-a. 0-Acyl-protected acceptors 25 and 2K, having 0-acetyl protection vicinal to the accepting hydroxyl group, proved to be less reactive, and lower a,p selectivitieswere found in their glycosylation with la-a. 0-Acetyl protective groups at other positions did not affect the convenientp-glycoside formation. Thus far, a-glucopyranoside formation has not been extensively investigated (67) because this connection is less frequently found in glycoconjugates. However, it was observed that, with p-trichloroacetimidate la$ as donor, stronger catalyst systems, as for instance Me3SiOTf,favor formation of the thermodynamically more-stable product, especially when the reactions are performed in ethers as solvents (Table 111; reactions with 3 4 2G, and 2F) (67). The influence of solvents in glycosylation reactions has been observed and discussed extensively already (1,4,74). For instance, the participation of ethers, when anomeric leaving-groups are removed under SN1-type conditions, results [because of the reverse anomeric effect (75,76)] in the genera-

TABLEI1 Reaction of the Benzyl-Protected Glucosyltrichloroacetimidate l a a with 0-Nucleophiles Anomeric configuration Glycosyl acceptor

2A

28

HO

&-

Bno&+ ,,BnO OCH,

Reaction conditions

(%)

Reference

CH,Cl, , BF, .OEt, , - 18"C, 2.5 h

1:13

78

51,52a

CH2C12,BF, * OEt, , -4O"C, 2 h CH,CN, Me,SiOTf, -4O"C, 20 min CH,CH,CN, Me,SiOTf, -4O"C, 20 min CH,CH,CN, Me,SiOTf, -SOT, 15 min

1:19

85

5 1,52a

1.16

89

62

1.16

83

62

1:16

74

62

1:6

70

65

1.2: 1

86

71

CH,Cl, , BF, .OEt, , - 35°C

2c

Yield

BnO

OBn /OH 2D

,,B n O ,- BnO

BnO

CHzCl,, CF3SO,H, -20°C

2E

/

2F

0: 1

80

72

CHzClz, BF3 * OEtz , - 30°C. 2.5 h CH3CH,CN, Me,SiOTf, -SOT, 10 min

1 :4

81

51,52a

1:19

81

62

CHzCl,, BF, * OEt, , -38"C, 1.5 h

1:lO

90

71

CH,Cl,, BF3. OEtz , MS 4 A,-70°C

0: 1

96

73

CH,Cl,, BF3 OEt, , MS 4 A, - 70°C

0: 1

94

71

CH,Cl,, BF3.OEt, , MS 4 A, -70°C

1 :2.5

46

73

OBn

Ho+-4?l BnO

0%

F?

2G

CH,Cl,-n-hexane, BF,*OEtZ,- IOOC, 3 h

HO

w

I .

/

OBn

2H

O\/CCI,

HO BnO

21

AcO

CCI,

HO BnO &Ov

-

AcO

/OBn

2J

HO AcO

O\/CCl, Aco

(continues)

TABLE I1 (continue4

Glycosyl acceptor

O\/CCI,

2K

Aco W

m 2L

F?

Anomeric configuration

Yield

(a:B)

(%)

Reference

CH,Cl,, BF, .OEt, , MS 4 A, -70°C

1.8: 1

45

13

CHzCl,, BF, * OEt, , -35"C, 3.5 h

0: 1

32

51,52a

CH,CH,CN, Me,SiOTf, -8O"C, 10 min

1 :24

72

62

Reaction conditions

BnO

OBn

ANOMERIC-OXYGEN ACTIVATION FOR GLYCOSIDE SYNTHESIS

39

TABLE111 Reaction of the Benzyl-Protected Glucosyl Trichloroacetimidate la+ with 0-Nucleophiles Anomeric configuration Acceptor

Reaction conditions

Yield (To)

Reference

/OH

/OH

2G

Et,O, Me,SiOTf, room temp., 2 h Et,O, Me,SiOTf, - 10°C, 1.25 h

8: 1

85

67

5: 1

89

67

CH,CH,CN, Me,SiOTf, - 8 0 T , 20 min

1 :24

74

62

Et,O, Me,SiOTf, room temp., 5 h

3: 1

95

67

Et,O, MeJiOTf, room temp., 6 h

3: 1

72

67

Rn

HO

/OB"

OCH,

tion of equatorial oxonium ions ( pconfiguration in D-glucopyranose);these favor via invertive attack of the acceptor the formation of the thermodynamically more-stable axial products (aconfiguration in D-glucopyranose) (Scheme 9). The dramatic effect of nitriles as participating solvents in glycosylation reactions was first observed in 0-glycosylationswith 0-glucosyltrichloracetimidates ( 5 1,53). This effect demonstratedthat, independent of the configuration of the glucosyl donor, in the presence of a strong catalyst and at low temperatures, p-glucopyranoside formation is favored (see Tables I1 and 111; reactions with 2B,2F,and 2M). The explanation (Scheme 9) that fast kinetic a-nitrilium - nitrile - conjugate formation providing the P-product precedes formation of the thermodynamically more-stable p-nitrilium - nitrile conjugate, which then could also furnish a products as previously observed (77), was supported by several findings. Excellent leaving-group abilities even at low temperatures are required for the application of this methodology, and therefore, aside from trichloroacetimidates, not all leaving groups can be used in this highly useful reaction (78).

R%%x

+

Catalyst: Y Acceptor:A-H

Ro

S2, or SN2-Type (Nonpolar Solvents)

J F P

0

l

\

SN2 or SN2-Type (Nonpolar Solvents)

[Polar (Donor) Solvents]

OR

4T 7. q

@

A-H;

@

=E@

" R e @ - OR

A "Intramolecular" (B=C=Solvent) i Me

XI-

:

SCHEME 9.-Glycosylation Reaction courses.

ANOMERIC-OXYGEN ACTIVATION FOR GLYCOSIDE SYNTHESIS

41

From these results there emerges a general picture of the reaction of trichloroacetimidate donors that is summarized in Scheme 9. In nonpolar solvents and with BF, - OEt, as catalyst at temperatures as low as possible S N (or ~ presumably it is better to say S~2-type) reactions (via a tight ion-pair) take place. With stronger catalysts, as for instance Me,SiOTf, a highly reactive carbenium-ionintermediatethat favors kinetic attack from the a face is generated. However, with ethers and the result of reverse anomeric attack, fast transformationinto a p-face shielded intermediatetakes place, leadingto formation of the a product; whereas with nitriles, on account of conjugate formation, a-face shielding remains efficient. A cyclic eight-membered transition state, leading to intramolecular glycoside-bond formation as shown in Scheme 9 may be hypothesized to explain the high reactivity and selectivity. Oligosacharidesas donors bearing glucose at the reducing end and having at least nonparticipating 2-0-protection provided essentially the same results (Table IV) (66,79). For instance, trichloroacetimidateformation with NaH as base gave the donors 4a,b in high yields and with high a-selectivity (65). Their reaction with acceptor 2G furnished, under BF, * OEt, catalysis at low temperatures, exclusively /3 products (65). Consideration of recent findings (see foregoing) should lead to improved yields in these reactions (62). The reaction of glucosyl trichloroacetimidates permitting neighboringgroup participation through 2-0-acyl protection [see for instance, the trichloroacetimidates 1b-a$ (Table I)] exhibits generally clean p-product formation regardlessof the configurationof the starting material (1) (Table V). However, the examples clearly show that the donor reactivity is lowered by acyl protection. Therefore, good yields are still attainable with reactive acceptors, but not as readily for acceptors of low reactivity. However, with Me,SiOTf as catalyst, very promising results even for less reactive acceptors were obtained (see Table V). In the Koenigs-Knorr reaction, orthoester formation was found to be a major drawback in these kinds of reactions (4). The mildly acidic nature of the trichloroacetimidate method decreases the problem of orthoester formation, thus leading to greatly improved glycosidation yields. Because of the presence of a C = C double bond in the sphingosinemoiety, 0-acyl protected glucosyl donors received general attention in the synthesis of glycosphingolipids (GSL). As a consequence of the many problems encountered with direct glucosylation of ceramides, employing all known glycosylation procedures, the introduction of the “azidosphingosineglycosylation” methodology, namely, glucosylation of azidosphingosine(70,84 - 95) (for instance, compounds 6A-6D, Table VI) and then attachment of the fatty acyl group to the amino group liberated from the azido function, led to

TABLE IV Reaction of Glucosyl Trichloroacetimidates of Oligosaccharideswith the Nucleophile 2G

Glycosyl donor 4a-a,p

NH

Bn0 BnO

&

oJL3

Trichloroacetimidate formation

Reaction conditions

CH,Cl, ,NaH, CCl,CN, room temp., 93%, a:/911: 1

CH,Cl,, BF, .OEt, , -4O"C, 2 h

CH,Cl, , NaH, CCl,CN, room temp., 5 h; 96%

CH,Cl, , BF, OEt, , -35"C, 5 h

Anomeric configuration

(a:B

Yield (Yo)

Reference

0: 1

51

65

0: 1

40

65

OBn

4b-a

OBn

OAC

I CCI,

-

TABLEV Glycosides and Saccharides from Acetylated Glucosyl Trichloroacetimidates

Glycosyl donor

Glycosyl acceptor

Anomeric configuration

Yield

b:B)

(Yo)

Reference

CHZCl,, BF3 * OEtz , 30°C

0: 1

85

80

Reaction conditions

OAc

-

1b-P

CCI,

BnO

w P

CHZCl,, BF3 * OEt2,

MS 4 A,- 2 0 ° C

0: I

25

81

58

CHZClz, BF3 * OEt, , room temp., 2 h

0: 1

67

51,52a

5C

CHZCl2, BF3. OEtz , room temp., 2 h

0: 1

14

51,52a

CH2Cl,, BF3* OEt, , room temp., 45 min

0: 1

64

51,52a

OCH,

H

O

e

N

O

t

lb-P

H

O

e

N

O

I

5c

(continues)

TABLEV (continued) Anomeric configuration Glycosyl donor

Glycosyl acceptop

Reaction conditions

(&:PI

Yield

(YO) Reference

YOAC

CH2C12,BF, .OEt, , room temp.

0: 1

78

51

CH2C12,Me3SiOTf, - 20"C,45 min

0: 1

81

82

CH,C12, Me,SiOTf, - 15"C, 30 min CH,Cl,, Me,SiOTf, - W C , 30 min CH,C12, Me,SiOTf, - 1 5 T , 30 min CH,Cl,, Me,SiOTf, - 1 5 T , 30 min CH2C1,, Me,SiOTf, - 1 5 T , 30 min

0: 1

65

83

0: 1

13

83

0: 1

72

83

0: 1

71

83

0: 1

67

83

5a-a

O Y N H CCI,

'OAc 5b-u

ACO

A d

OAC 5c-a

HOCH2COOCH3

(5F)

HO(CHz),COOCH3

(5G)

AdoYNH HO(CH2),COOCH,

(5H)

CCI,

HOCHzCOOCH2Ph (51) HO(CH2),COOCH2Ph (5J)

a

Cbz,Z, benzyloxycarbonyl.

TABLE VI Glycosylation of Azidosphingosine Derivatives with Trichloroacetimidates Clycosyl acceptor' OTBDMS

OBZ

6A

OBZ

68

Clycosyl dwor

R d o n conditionsb

i k

Tricbloroacetimidate: Glycosylation:

See Table I aC,CHzCIz, BFa'OEtz;

1b-a

Tricbloroacetimidate: Glycosylation: Trichloroacetimidate: Glycosylation:

See Table I 6D

87

See Table I 6B,CHzClz, BF3.0Et2, room temp.; 94% /3

84

Trichloroacetimidaw

C H z C l z , CCI,CN, DBU, 88% a 6 4 C H p , , Me3SiOTf;

91

Ida NH

u c c ,

Glycosylation:

Refereace 86

80% B

51968 6a-a

OAc

OAC

0

L3

AcO

ACO

Trichloroaoetimidate:

CHzCIz,CCI,CN, DBU,

Glycosylation:

97% (I 6.4. C H z C I z , BF3.0Et2;

91

32%/3 AcO AcO

6b-a OAc

(continues)

TABLE VI (continued)

Tririchloroacetimidate:

A.0- AcO

Glycosyhtion:

(CH2C12h,CQCN, DBU, -20°C; 88% (I 6A, CH2Cl,, Me,SiOTf; 64% B 6Ll, CH2Cl,, Me,SiOTf;

91

68% fl P

m

* : cAAcO

Trichloroacetimidate:

Glywsylation:

CH2Cl,, NaH,CC13CN, mom temp.; 52% (I 68, CH2Cl,,BFl.OEtl; 87%)

84.85

CH,C12,NaH, CC13cN,

85

6d-a

Trichloroacetimidate:

66%(I

Glywsylation:

Ca,

68, C H 2 C l 2 , BFl.OEh; 78%)

Trichloroacehmidate: Glycosylation:

v

CHzaz NaH, , m3m, 92%,a:84:1 6B, CHzCIz-n-hexane, BF30Etz;71%8

92

CHZa C2 CI& ,!N, O'C, 2 h; 94%a 6B, C H z C I z , BFs'OEt,,

93

AcO

Trichloroacetimidate: COOMe

Glycosylation:

93

0°C.4 h; 92%8

AcO OAC OAc

5 0 6h-a

OAC AcO

Trichlodmidate:

OAc

Glycosybn:

AcO

0

61-a

CHzClz,ccl,CN, DBU, 87% (2 68, CHKlzCl,, BF,.OEt,; 42% 8

94 94

CCI,

(continues)

TABLE VI (continued) Reference

Reaction conditionsb

Clyeosyl dowr

OAc

Trichloroacetimidate: Glywsylation: P

ca

AcO

CH2CIz,CCllCN, DBU; 87% a 68, CH2Clz,MeaiOTf (0.01eq.),room temp.; 75% p

96

96

61-u

I

CCI, 6k-u

TBDMS, tert-butyldimethylsilyl. DBU, 1,8-diazabicyclo[5,4,0]undec-7ene.

TrichloroaCetimidate:

57

Glywsylation:

97

ANOMERIC-OXYGEN ACTIVATION FOR GLYCOSIDE SYNTHESIS

49

a major breakthrough in GSL synthesis. This is demonstrated in Table VI, which includes representativeexamplesof this development. Thus, not only were simple glycosyl and lactosyl derivativesmade readily accessible,but all major glycosphingolipidseries were successfully synthesized,includinggangliosidesthat contain neuraminic acid. The importance of tumor-associated antigens of glycosphingolipid nature also created interest in several L-fucose-containing glycosphingolipids, for instance, the Lewis X (Lex) and Lewis Y (Ley) antigens. The acid sensitivity of the fucosyl anomeric bond generally requires special attention in glycoside-bond formation. However, it turned out that these 0-acetyl protected donors did not cause any problems under the required reaction conditions. On account of the high acceptor reactivity of the azidosphingosines, orthoester formation as a side reaction was encountered for the first time (85). In many instances a slightly higher catalyst concentration readily overcomes this problem. The problem may also be solved with the help of 2-0-acyl-protective groups, which for steric (2-0-pivaloyl) or electronic (2-0-benzoyl) reasons do not undergo orthoester formation as readily as do 2-0-acetyl groups (70,85). As just discussed, “ceramide glycosylation” seemed to cause major problems when the Koenigs-Knorr method was used (84). However, 0-acetyl protected glycosyl trichloroacetimidate donors also provided acceptable yields with glucosyl or the lactosyl trichloroacetimidates (Table VII) (84,98- 108). For higher oligosaccharidedonors the results were unsatisfactory. However, again by attaching the bulky 2-0-pivaloyl group to the glucose moiety, this drawback may be overcome (84,101), a result which reinforces the search for further improvements in this most active field. c. 0-Galactosyl Trichloroacetimidatesas Donors.-D-Galactose (63,64) is a constituent of complex glycosphingolipids and glycoproteins, where it plays an important role. It is found as a terminal or subterminal building block in a variety of differentconnections. In glycosphingolipids,galactoseis part of the lactosyl ceramide core-structure. Terminal and subterminal p( 1 + 4)-connection to 2-acetamido-2-deoxy-~-glucose(N-acetylglucosamine) leads to the N-acetyllactosamine moiety, which is preferentially represented in the lacto and the neolacto series. The a-(I + 4) and a-(1 --* 3) connectiondetermines the gala, globo, and isoglobo series. In glycoproteins, terminal galactoseis a signal of the Ashwell receptor, whose function consists in the binding of galactosylated glycoproteinsin the liver. In the asparagheconnected glycan residues of N-glycoproteins, galactose is mainly found in N-acetyllactosamine, whereas in the serine- or threonine-connected 0-glycoproteins, galactose is preferentially p-( 1 --* 3)-linked to 2-acetamido-2deoxy-D-galactose (N-acetylgalactosamine).This connection is also met in the ganglio and the isoganglio series of glycosphingolipids.

HNq

TABLE VII Glycosyhtion of Ceramides by Trichloroacetimidates

(CH2)13CH3

0 HNK

0

HNK

(CHz)tPh

5

=

OAc \

\

(CH2)1ZCH3

HOW

(W)zzCH3

(CH~)~ZCH~ HOW

(CHz)izCH3

(CHz)&H,

OBZ

OBZ

OBI

OTBDMS

7c

78

7A

7D

Ttkhloro8ceimid.te formatkllP

Glymyl dwor

Glyeasyhtion eoDditioas

Yield

7A, C H z C l z , BF,.OEtz 7B, CH2Clz,BFl.OEt2, mom temp.

57

70

86 98

7 8 4 R = AC; C C I 3 C N , DBU, 79% ma: R = Piv; 52% 7c-a R = TMP

7c

13

100

7C, Me,SiOTf

66

7D, Me,SiOTf

27

101 102

7 d a . 83% OL

7C, Me,SOTf

See Table 1

1b.iY l k

See Table I

L3 AcO

HNA

(CHz)zzW

7How

HO

0

(%)

Ref-

ACO OAC OAc

NH

31

101

ACO

COOMe

7 ~R =: CH,;CCIlCN, DBU, 92% 7fa:R = CH,OAc; CCIICN, DBU, 16%

AcO

0

7D, BFi.OEt2,

52

103

58

103

1

104

(~P,), 7 4 BF,.OEt,, (

~

Z

~

Z

L

OAC

NH

nr, "nr

/

I

OAc

0 AcO

1

7C, BF,.OEh

Io k

ACO

RICHARD R. SCHMIDT AND WILLY KINZY

52

TABLE VIII Synthesis of Trichloroacetimidates of D-Galactose" Trichloroacetimidate

Reaction conditions

Yield (To)

Reference

CCl,CN, CH,CI, ,NaH, room temp.

83

46

CC13CN, CH2C12,K2C03, room temp.

84

67,68

39 a -t45 /3

66,104,109

CCl,CN, CH,CI, , NaH, room temp., 1.5 h

60

85

CCl,CN, CH,C12, DBU

80

I10

CCI,CN, CH2C12, DBU, -5°C

I1

111

En? ,OBn

En0

I

CCI,

aa-p

En0

OBn

BnO ~

o

BnO y

c

c

l

,

NH

AcO

OAc

CCI,CN, CH,Cl, , Na, room temp.

AcO

CCI,

LCI,

BnO

OBn

BnO

CCI, ACO

oac

&-a En0

CCI,

(continues)

ANOMERIC-OXYGEN ACTIVATION FOR GLYCOSIDE SYNTHESIS

53

TABLE VIII (continued) Trichloroacetimidate

Reaction conditions

Yield (%)

Reference

CCI,CN, CH,CI, , NaH, MS 4 A,O"C, 2 h

76

112

CCI,CN, CH,CI, , NaH, room temp.

77

113

To10 OTol

8f-a ~

To'o

O

' C To10

C

I

,

8g-a

CCI,

Tol, Toluyl; DBU, 1,8-diazabicyclo[5,4,O]undec-7-ene.

1-0-Unprotected galactose derivatives may be readily transformed into the trichloroacetimidates8a-8g, as shown in Table VIII. Again, as demonstrated for the 0-benzyl-protectedcompounds 8%either the a-trichloroacetimidate 8a-a or the P-trichloroacetimidate 8a-/l may be obtained highly selectively,depending on the base used for the catalysis of the addition to the trichloroacetonitrile. Galactosylationwith the 0-benzyl-protecteddonors 8a-a and 8a-P( Table IX) shows that conditions can be found for invertive product-formation in high yields. Thus, from 8a-Pin ether, preferentially a products were formed, and from 8a-jl in the rather nonpolar solvent-mixture dichloromethane- nhexane, mainly the /3 product was obtained. The higher tendency of galactosyl donors to effect a-glycoside bond-formation compared with the corresponding glucosyl donors is well established in the literature (4)and is also observed here. This may be attributed to the generally higher reactivity ofthe galactosyl donor and to the axial 4-substituent. 2-0-Acyl-protected galactosyl donors readily provide j? products. The reactivity may be increased by having partial 0-benzyl protection, as exhibited (1 10,111)with donors 8d-a and 8e-a (Table X). The examples permit very successful 2-0-, 3-0-, and 4-0-connections, respectively. The highyielding synthesis of the /%Gal-(1 + 3)-GalNAc building-blocks [8f-a 10H, Table X (1 12), and 8a-a 9C, Table IX (1 15)] furnishes a convenient access to 0-glycoprotein moieties; for instance, selective removal of the l-O-ButMe,Si protective group in the 8a-9C P-product (Table IX) and subsequent P-trichloroacetimidate formation leads to the desired P-Gal-

+

+

TABLEIX Glycosidation with Benzylated Galactosyl Trichloroacetimidates"

Trichloroacetimidate

(W

Anomeric configuration (a:B)

Reference

(C,H,),O, TBDMSOTf, room temp., 0.75 h

75

5:1

67

2F

(C2H,),0, Me,SiOTf, room temp., 5 h

65

8: 1

67

2G

(C,H5),0, Me,SiOTf, room temp., 1 h

66

36: 1

67

(C,H5),0, Me,SiOTf, room temp., 3.5 h

77

8: 1

67

(C,H5),0, Me,SiOTf, -20°C

15

1 :o

114

Glycosyl acceptor

Reaction conditions

Yield

OH

OBn

6

HoBnO 111

P

8

OH

OCH,

OBn

,OBn

L$gyornDMs

CH,Cl, - n-hexane, BF, .OEt,

84

1 :7

115

CH,Cl,- n-hexane, Me,SiOTf, -25°C

80

1:4

116

8a-a

CH,CH,CN, Me,SiOTf, -40°C

75

0: 1

72

8a-a

CH,Cl,- n-hexane, BF3*OEt2,-25°C

83

1:3

12

8a-a

CH,Cl,-n-hexane, Me,SiOTf, - 3 0 T , 2 h

49

1 :o

117

8a-a

8a-a

HO

9c

BnoG 9D

HO

HO

OTBDMS

OH

TBDMS, lei?-butyldimethylsilyl.

(2-0)

TABLE X Glycosylation with Acetylated GalactopyranosylTrichloroacetimidatese Trichloroacetimidate

8b-a

Glycosyl acceptor

'o>P ( HO ! @ Y O T E D M S

HO

9c

Reaction conditions

Yield

(%I

Anomeric configuration (a:j?)

Reference

CHCl,, BF,-OEt,, MS3A

33

0: 1

118

CHCl,, BF, .OEt,, MS4A

31

0: 1

11s

CH,CI, - n-hexane, BF, * OEt,, room temp., 1 h

67

0: 1

117

CH,Cl,-n-hexane, BF3*OEt2,-2O"C, 2h

69

0: 1 (3-0)

117

OEzl

N3

1oc

En? ,OBn

N3

CH,Cl,- n-hexane, BF3*OEtz,-2O"C, 2h

65

0: 1

117

CH,CI,, BF3.OEt,, MS4A

73

0:1

I10

CHzCIz,Me,SiOTf, MS 4 A,- 3 0 T , 20 min

81

0: I

111

CH,CI,, Me,SiOTf, MS 4 A, - 30°C, 10 min

87

0:1

111

CH2C12,Me,SiOTf, l O T , 15 min

93

0: 1

112

CH,Cl, - n-hexane, BF3 * OEt,, room temp., 1 h

75

0: 1

113

1OD

?CH,CH,

6en

a

TBDMS, terf-butyldimethylsilyl.

-

RICHARD R. SCHMIDT AND WILLY KINZY

58

(1 43)-a-GalNAc-(1 0)-Ser glycopeptidemoiety (1 15). The reaction of 8e-a with thioglycoside (1 11) 10F demonstrates again that a thio group at the anomeric position is compatiblewith application ofthe trichloroacetimidate method. Among the gangliosides, GM4 [a-NeuAc-(2 --* 3)-&Gal-( 1 0)-Cer] has a relativelysimple chemical structure. It has been detected in human and chicken brain and also (1 19) as a major ganglioside of mouse erythrocytes, chicken-embryonic liver, and egg yolk. With the help of the azidosphingosine glycosylation it has been synthesized very efficiently from the neuraminic acid-containinggalactosyl donor l la$ (Table XI) ( 120- 122). Similarly the thio isomer was obtained from 11b-8and ( 120,123) the positional isomer from 1lc-a.

-

d. 0-Mannopyranosyl Trichloroacetimidates as Donors.-D-Mannose (63,64) is less frequently encountered in glycosphingolipids (only in the arthro series); however, it is generally a constituent of N-glycoproteins: a central a-Man-( 1 ---* 6) [a-Man-( 1 -,3)]Man trisaccharidemoiety, which is p-( 1 + 4)-connected to a chitobiose unit, is part of the core structure. Mannopyranosyl trichloroacetimidates that have been synthesized are compiled in Table XII. Because of the stronger anomeric effect (1,45,75), a-trichloroacetimidate formation is much faster than observed for corresponding glucose and galactose derivatives, and therefore the a-trichloroacetimidates were generally isolated thus far. This was not regarded as a disadvantage because a-mannopyranoside formation, for instance from 12a-q should be readily achieved because of the stronger anomeric effect under thermodynamic reaction control. The examples in Table XIII show that this is indeed the case; selective a-product formation was observed even with BF, OEt, as catalyst. However, clean p-product formation from the a anomer 12a-a under invertive conditions has not yet been achieved (5 1). Even the nitrile effect, as found recently, led to only partial success in this endeavor (62) (Table XIII). The ready formation of the a-mannopyranosyl linkage is also true for the 2-0-glycosylated 0-mannopyranosyl trichloroacetimidates 14a-a-14e-a (Table XIV). Use of Me,SiOTf as catalyst would be presumably superior in these reactions. 2-0-Acyl protection should lead, as a consequence of neighboring-group participation and the anomeric effect, exclusively to a products. This has been proved in many experiments (Table XV); with Me,SiOTf as catalyst excellent yields could be obtained in cases where all other methods essentially failed (129). It could be shown that at least some of the reactions proceed via rapid orthoester formation (129), and this intermediate then rearranges under Me,SiOTf catalysis to the desired reaction product.

-

TABLEXI Glycosylation of Trichloroacetimidatesof D-Galactose with Sphmgosine Derivative 6B ~~

Trichloroacetimidate formation

Glycosyl donor &-a

Reaction conditions

Yield (%)

Reference

See Table VIII

CHzClz, BFjOEtz, room temp.

96 B

85

CH,ClZ, CCl,CN, DBU, fOT, f : p 1:8 2 h; 79-93%,

CH2CI2, BFjOEt2, 0°C

82P

120,121,122

CH2C12,CCI,CN, NaH, 0°C;a : P O : 1,85%

CHzCl2, BFjOEtz, 0°C

70 P

123

CHzClz,CC1,CN a:p11:1

CH&

78 P

120

COOMe

AcO

NH

~ o @ o ) , c c , , 0

ACO

ACO

092

1la-B

COOMe

0

Acb 'OBz

NH

llb-P

Ca,

BFjOEtz

RICHARD R. SCHMIDT AND WILLY KINZY

60

TABLEXI1 Trichloroacetimidatesof D-Mannose

Trichloroacetimidate

Anomeric config.

Yield

(a#

(%)

Ref.

CH2Cl,, CCl,CN, NaH, room temp., 0.5 h

1 :o

99

46,124

CC13CN,NaH

1 :o

46

125

CH,Cl,, CCl,CN, NaH, 0°C-room temp., 20 min

1 :o

n.n.

91,126

(ClCH,),, CCl,CN, DBU - 5°C

1 :o

98

127,128

CHzClz, CC&CN, K2C03, la

1 :o

86

129

Reaction conditions

BnO

I

CCI,

CH30 I

ACO I

BnO

I CCI,

CCI,

e. Trichloroacetimidates of Glucosamine Derivatives as Glycosyl Donors.-2-Acetamido-2-deoxy-~-glucose (N-acetylglucosamine) (63,64, 1 19) is an important constituent of all glycoconjugates.In the glycan chains of N-glycoproteins it is part of the core and of the glycan side-chains. In glycosphingolipidsof the lucto and the lactoneo series, it is the main constituent. In proteoglycans, in bacterial lipopolysaccharides,in the murein of

ANOMERIC-OXYGEN ACTIVATION FOR GLYCOSIDE SYNTHESIS

61

TABLE XI11 Reactions with Mannosyl Trichloroacetimidate12a-a Reaction conditions

Acceptor

YJ'

HO

Anomeric config. Yield @:/ (%) I) Ref.

CH2CI2, BF3.OEt2, -WC, 2 h

5: 1

83

51

CH2Cl2, BF3. OEt2, 20°C. 5.5 h

l:o

73

51

CH& BF,. OEtz, 20°C, 20 h

1 :o

68

51

1:l

77

62

28

CH3CH2CN,Me3SiOTf, -8 0 T , 20 min CH3CH2CN-n-hexane (1 :4), Me3SiOTf, - 8 0 T , 20 min

3:2

70

62

2G

CHZCl2, BF3. OEtz, 20°C 4 h

1 :o

66

51

CH$H,CN, Me,SiOTf, -8O"C, 10 min

1:l

71

62

J3P

13A

+

BnOBnO

OMe

,OBn

2M

bacterial cell-walls, and as a polycondensation product in chitin it has wide distribution. In N-deacetylated form as free glucosamineit was identified as a constituent of glycosylphosphatidylinositols,which are membrane anchors for cell-surface glycoproteins (1 36). This wide distribution is accompaniedby a variety of different linkages, as compiled in Table XVI. Obviously, p-connection is generally favored. (i) Glucosamine Donors. -The great number of trichloroacetimidates synthesized thus far underlines the fact that compounds displaying high reactivity and high diastereocontrol are required for the great variety of

TABLE XIV Glycosides and Saccharides from Mannosy1 Tnchloroacetimidates Trichloroacetimidatd AcO

Trichloroacetimidate formation

Reaction conditions

Reference

OAc

CH2C12,CCl,CN, NaH; 95% a

14A, BF,-OEt,, MS 4 A,45% a

130,131

CH2C12,CCl,CN, DBU, O T , 30 min; 87% a!

14B; 10% a 14C, BFjOEt,, CH2Cl,; 60% a 14D, BF, .OEt,; 53% a-(l+ 6) 14G, 58% a

133,134

O Y N H CCl, AcO

.OAc

I O Y N H CCI,

S P O OP

)-$

z I

%To' 0

20

TABLE XIV (continued)

14A

14D

,OBn

..^. .. tlnu

140

14E

-

NPMh

OBn

0 'OBn 14C

14F

TABLEXV Glycosylation of Acetylated Trichloroacetimidatesof DMannose Trichloroacetimidate

Glycosyl acceptor

15A AcO

Yield

Reaction conditions CH,Cl,, Me3SiOTf, MS 3 A, - 10°C 10 min

(%)

59 a 4 1 -P 6)

135

60

135

OH

12c-a

CH,Cl,, Me,SiOTf, MS 3 A, -20°C 20 min

HO

AcF* ACO

Reference

CX-( 1

+

6)

158 I

12c-a

CH2C12,Me3SiOTf, MS 4 A, - 30"C, 10 min

88 a

127

CH,Cl,, Me,SiOTf, -3O"C, 10 min

92 a

126

15D (3%

ACO

HO

12c-a

A

c

t

d

o

12c-a

CH2C12,Me,SiOTf, MS 4 A,-3O"C, 10 min

94 a

126

CH,Cl,, Me,SiOTf, MS 3 A, room temp., 30 min

15 a

135

86 a-(1 + 3)

135

ACO

HO 15F

12c-a /-J

,o

15G

CH,Cl,, Me,SiOTf, 12c-a (2.3 q.), MS 3 A,room temp.

m

I .

n.n.

12c-a BnO L

n

a-(1 46) trisaccharide

90 a 3-012-0 1:3

91

81 a

135

82

135

151

CHzCIZ,Me,SiOTf, MS 3 A, room temp., 10 min

OH

CH,Cl,, Me,SiOTf, MS 3 A,- 1O"C, 10 min

CY-(1 + 6 )

RICHARD R. SCHMIDT AND WILLY KINZY

68

TABLEXVI Naturally Occurring Glycosidic Linkages of N-Acetylglucosamine Glycosidic linkage

p-( 1

Acceptor

Occurrence

Chitobiose core structure of N-glycoproteins Part of murein of Gram-negative bacteria Disaccharide unit of lipid A (as in Salmonella minnesota)

+

4) 4)

+

6)

GlcNAc MurAc GlcNAc

p-(l 6, p-( 1 43)

GalNAc GalNAc

Part of core structures of the 0-glycoproteins

3, 8-(1+ 6)

Gal Gal Man GlcA Man

lacto- and neolacto-series of glycosphingolipids

8-(1

p-( 1 '-(I

+

-

+

p-(1

3) 6) p-(1 -2)

a-(1

+

artho-series of glycosphingolipids Phosphoglycosphingolipids of tobacco leaves Phosphoglycosphingolipids

glycoside bond-formation processes encountered. Compounds capable of neighboring-group participation through N-acyl or N-phthaloyl groups (Table XVII) are readily obtained from glucosamine. Presumably on account of the size of the N-phthaloyl group, only P-trichloroacetimidates (17c-P- 17f-P)were obtained (139 - 144).However, for the N-acyl-protected compounds 17a-a and 17b-a it could be shown that the glycosylation reaction proceeds via intermediate oxazolines (1 37); therefore, an advantage for application of the trichloroacetimidate procedure could not be establishedin these cases. The N-phthaloyl-protectedtrichloroacetimidates permitted an enormous improvementin terms of yield and diastereoselectivity. However, the removal of the N-substituent from the glycosides sometimes caused problems. Therefore, 2-azido-2-deoxyglucosederivatives, readily obtained from glucals via the azidonitration methodology of Lemieux and Ratcliffe (1 52), seemed to be ideal; various trichloroacetimidates were accordingly prepared (Table XVII). Careful investigation of trichloroacetimidate(17g) formation (145) led again to conditions for the selective formation of both anomers. Also noteworthy is the selective formation (149) of the 4-0unprotected trichloroacetimidate 17i-P. Because the azido group is considered a nonparticipatinggroup, it remained to be shown that the a-trichloroacetimidates can be transformed cleanly into P-glycosides under S~2-type conditions. Experiments (1 37,137a) with the N-phthaloyl-protected donor 17c-/l showed excellent glycosyl-donor properties, as indicated in Table XVIII. Various galactose-and galactosamine-derived acceptors underwent successful reaction. The 0-benzyl-N-phthaloyl-protecteddonors 17d,e, and f showed comparable properties ( 141,143).

-

m rn

c€*

13 m

4

0

69

I-

m

-..

0

hl 4

9 9 hl

m

..

4

0

TABLE XVII (continued) Reaction conditions

Tnchloroacetimidate

171-p

ACO BnO &oycc,

(a:8)

Yield (%)

NPhth

/ 17e-P

Anomeric configuration

Reference

141,142 NH

OBn

BnOBnO -OuNH N3

CH2Cl2, CCl,CN, K2C0,, 20°C, 4 h

0: 1

90

145

DME, CC13CN,NaH, 0°C

4: 1

98

53

CHZClZ, CQCN, NaH, room temp.

1 :o

15

5435

CH2Cl2, CCl3CN, NaH, O'C, 12 h

1 :o

15

148

CH2C12,CC13CN,NaH, room temp.

1 :o

98

149

CCI,

4

0

17g-a

CCI, (OB" 17h-0

&a, YOTBDMS 171-a

HO

CCI,

/

OTBDMS

17j-p N3

CH,Cl,, CC13CN, Kzco3, room temp., 4 h

0:1

66b

149

CH,Cl,, CC13CN, KZCO,-NaH, room temp., 4.5 h

1

:o

66b

150

CH,Cl,, CC13CN, NaH, room temp., 1 h

1

:o

44b

137,147

CH,Cl,, CC13CN, Kzco3, room temp., 6 h

1:2

7 56

137

CHzCl,, CC13CN,

0: 1

n.n.

151

NH OTBDMS

17k-a

I

CCI,

/OAC 171-a

A C -ko

O YCCI, N H /OAC

NH /OAC

K2c03

NH

a

pMP, pmethoxyphenyl; pMBn, pmethoxybenzyl; DME,1,2-dimethoxyethane; TBDMS, tea-butyldimethylsilyl. From an epimeric mixture.

TABLEXVIII Reaction of 2-N-Phthaloyl Trichloroacetimidate 17c-B with Nucleophiles Reaction conditions

Glycosyl acceptor CHjOH

(1BA)

OTBDMS

HO

(am

Yield (To)

Reference

BF, eOEt2, - 30°C

0: 1

65

140

BF3.0Et2,-20°C

0: 1

70

140

Me,SiOTf, -7O"C, 5 min

0: 1

93

138, 153

BF3*OEt2,-20°C

0:1

71

139

Me,SiOTf, O T , 15 min

0:1

68

139

Me,SiOTf, 5°C

0: 1 3,6-di-0glycosylation

75

139

10H

N3

CH,O

HO

Anom eric configuration

Q o +

OCH,

OBn

AcO A

c

F

OAC

~

~

o

HO

B

&

OH

O

C

OH

3

1BD

BzO

HO

H

1BE

ANOMERIC-OXYGEN ACTIVATION FOR GLYCOSIDE SYNTHESIS

73

The corresponding 2-azido derivatives revealed surprisingly similar results: high reactivity was combined with extraordinaryp selectivity. Table XIX lists many important glycoside-bond formation reactions ( 145) with the 0-benzyl-protected glycosyl donor 17g-a. Only the reaction with the sterically hindered muramic acid acceptor 19E led, in the presence of Me,SiOTf, to partial a-product formation (57); however, this problem could be readily overcome by replacing the bulky ButMe,Si protectivegroup by the benzyl group, as shown (148) for 19F. Remarkable also are the reactions (149) of acceptor 17i-pwith the partially protected acceptors 19B and 19G: regioselective reaction at the 6-position and clean P-product formation was observed. The high tendency for p-product formation with the a-trichloroacetimidatederivativesof azidoglucose as donors is also because of the fact that the solubilityof the compounds permits the use of the rather nonpolar solvent-mixture dichlormethane- n-hexane, which favors S N ~ type reactions in the presence of BF, * OEt, as catalyst and low temperatures. An interesting example showingthat a-product formation may be readily achieved by the use of a p-trichloroacetimidate and Me,SiOTf as catalyst is shown in Table XX: compounds 17n-j3 20A furnish exclusively the a-glycoside in high yield (15 1).

+

(C)Lactosamine Donors. -The general importance of lactosamine in glycosphingolipid and glycopeptide synthesis is because of the frequent occurrence of this building block (1 19). For instance, the branching of the pentasaccharide core of N-glycoproteins is determined by the connection with N-acetyllactosamine, which may occur in p-( 1 --* 3) linkage in long chains. In 0-glycoproteins, N-acetyllactosamine is part of the core of mucin-type oligosaccharides.Likewise, the core structure of the glycosphingolipids of the luctoneo series is determined by N-acetyllactosamine. The connectionof these naturally occurring lactosamine units determines the protective-group pattern of the required building blocks. The general occurrence of the p linkage permits again the use of N-phthaloyl protection; however, azidolactose (15 3 , readily obtained from lactal, also should be very useful as a consequence of the advantages of this group already discussed. Both of these types of trichloroacetimidates,having different protective groups, have been very successfullyprepared, as indicated in Table XXI. Again, asjust discussed,with N-phthaloyl protection exclusivelyp-trichloroacetimidates 21a-/I-21e-/l were obtained (130,139,140,143,144,156- 159). Both isomers may be selectivelygenerated from the azidolactosederivatives, as shown (137, 160) for 21j. With sodium hydride as the base, a-trichloroacetimidates are obtained in very high yields. Some of these com-

TABLE XIX Reaction of 2-Azido-2-deoxyglucopyranosylTrichloroacetimidates with Nucleophiles Trichloraacetimidate

Glycosyl acceptor.

Reaction conditions

Anomeric configuration

Yield

(a$)

(%)

Reference

17g-c~

CH,CI, - n-hexane, Me,SiOTf, - 5 0 T , 10 min

0:1

60

146

17g-a

CH,C12- n-hexane, BF,*OEt,, -3O"C, 20 h

0:I

80

146

CH,Cl,- n-hexane, BF,*OEtz, -15°C

0:1

80

I50

lgB

0:I

92

145

19c

CH,Cl,- n-hexane, BF3*OEt2,-2O"C, 3h CH,Cl,- n-hexane, BF,.OEtz, - W C , 2h

0:1

90

154

4 P

HO

OH

17g-c~

X H 3

17g-c~ N3

N3

9c

17g-a

,,XH & o HO & O

OTBDMS N3

17g-a

0: 1

60

154

CH2Cl,- n-hexane, BF, OEt,, - 15 "C, 6h

0: 1

70

154

CH2CI,- n-hexane, Me3SiOTf,- 1 5 "C, 5h

1 : 1.6

90

57

CH2C12-n-hexane, BF3OEt2, -2O"C, 8h

0: 1

78

148

10H

pMpXH & o HO & O

CH2Clz- n-hexane, BF3 * OEt2, - 18"C, 3h

OTBDMS N3

19D

OTBDMS

17g-a Ho&OIBDMS I

N3

19E

17g-a

A -.as"

CH3

N3

H

19F

COOMe

(continues)

TABLEXIX (continued) Trichloroacetimidate

Glycosyl acceptor' HO

Reaction

Anomeric confignration

Yield

conditions

(a$)

(%)

Reference

0: 1

85

149

50

149

OH

CH2C12-n-hexane, BF30Et2

17i-/3 198

OCH,

-

CH2C12 n-hexane, BF3 * OEt2 N3

19G

pMP, pmethoxyphenyl; TBDMS, ?err-butyldimethylsilyl.

0: 1

6)

TABLE XX a-Selective Glycosidation of a j?-Trichloroacetirnidate" Reactants

B"O

111.1"'

20A

1711-p

BnO a

pMBn, pmethoxybenzyl.

Products

TABLE XXI Synthesis of Trichloroacetimidates of N-Acetyllactosamine Anomeric configuration

Yield

(%I

Reference

0: 1

12

130,139,140,156

CH,CI,, CC13CN, DBU

0: 1

92

151

CH,Cl,, CCl,CN, DBU, O”C, 3.5 h

0: 1

87

143,114

CH& CC13CN, DBU

0: 1

12

158

CH,Cl,, CC13CN, KZCO,

0: 1

66

159

Reaction conditions

Trichloroacetimidate

(a:B

.OAc A

C

O

p

o

& ‘ y

2la-p

‘“3

NPhth AcO

NH

OAc

&y OBn

A

C

O

e0

0

21b-p

NPhth

BnO

‘“3

NH

OAc -0MP

.OBn 21d-p

AcO#o&O

CCI,

NPhth NH

Aco

06” OAc

L U r N

PI 00

-..

d

0

&

& F;

-f ..

d

N

$

u

0

..

3

3

0

3qo 79

TABLE XXI (continued) Reaction conditions

Trichloroacetimidate

Anomeric configuration

Yield

(a:8)

(%I

Reference

9: 1

53"

137,160

1:5

64"

137

.OAc 21j-b

A

c I

AcO

\

o

OAC

From epimeric mixture.

~ ACV- o

\-

K

ANOMERIC-OXYGEN ACTIVATION FOR GLYCOSIDE SYNTHESIS

81

pounds are already suitable for further connections in the 3-,3’-, 4’-, and 6’-positions (1 14,115,137,155,160). Several glycosylation reactions with N-phthaloyl-protected donor ( 140) 21a-j?have been very successfully performed, as indicated in Table XXII. The j?-selectivitiesand the yields are generally very good. The reaction (130) with 22C demonstrates that dilactosaminylation can also be successfully achieved. The reaction of the 3’,4-0-unprotected 0-benzyl-lactosederivative (157) 22D led, contrary to the generally observed higher reactivity of the 3’-position, to reaction at both positions. This problem could be overcome by employing the 0-acyl protected lactosamine derivative 23A (Table XXIII). With donor 21e-j?, Veyrikres et al. (159) obtained tetrasaccharide 23B in high yield. This reaction could be repeated with the derived tetrasaccharides 23c as donor and 23D as acceptor, thus leadingto the corresponding octasaccharide in good yield. As already observed for azidoglucose-deriveddonors, glycosylations with azidolactose-derived donors (21f-a- 21j-a, Table XXIV) also exhibited high reactivityand~selectivity(92,114,115,154,161,162). Withtheseresults in hand, excellent preconditions for successful syntheses of the Le”and L e y antigens have been presented (164,165). Representative examples for the decisive glycoside-bond formations are compiled in Table XXV. Comparison of the results of N-phthaloyl protection and of the azido group does not exhibit advantages for the use of N-phthaloyl derivatives.

(iii) Chitobiose Donors. -Only a few trichloroacetimidate-basedchitobiose donors have been synthesized thus far, as indicated in Table XXVI. Their reaction with benzyl alcohol as acceptor demonstrates the potential usefulness of these donors in glycosylation reactions.

-

(iv) Muramic Acid as Donor. -The cell-wall peptidoglycan of bacteria has a p-( 1 4)-linked glycan chain, consisting of alternating 2-acetamido2-deoxy-~-glucoseand N-acylmuramicacid residues that are cross-linked by a peptide chain. The resulting peptidoglycan network (murein) and its fragments exhibit marked immunostimulatory and antitumor properties. The minimal structure for activity, the so-called Freund‘s complete adjuvant, is a “muramoyl dipeptide” (MDP). Many investigations have been directed toward the synthesis of derivatives of MDP, including glycosides and oligosaccharides; the attachment of lipophilic groups is of special interest because of their potential in combined chemotherapy and immunotherapy ( 166,167). The transformation of azidoglucose derivatives into muramic acid precursors enabled the formation of trichloroacetimidates as muramic acid donors that could be very successfully employed in glycoside bond-forma-

TABLEXXII Reaction of 2-N-Phthaloyl-2deoxytric~oroaceti1nidate 2la-p with Nucleophiles.

Glycosyl acceptor

Reaction conditions

Anomeric configuration

(a$)

Yield

(W

Reference ~~

CH2C12,BF, *OEt2, - 8 T , 3.5 h

0: 1

69

156

(CH2Cl)2,BF3.OEt2

0: 1

73

132

(CH2C1),, MS 4 A, BF3*OEtz,- 15°C

0:1

CHZCI,, BF, * OEtz, -2O'C

0: 1

67

140

BF,*OEt,, -20°C

0: 1

75

139

0

m N

22c

BnO

22D

Hoe

22E

22F

NPMh

n.n.

0: 1 /3-( 1 :3’) pq1 :4’) 2: 1

88

157

(CH,Cl),, BF3*OEt, (2.0 eq.), room temp.

0: 1

87

157

BF,.OEt,, -20°C

0: 1

71

140

22

102

80

139

N3

m W

22G

‘OBn

22H

Me3SiOTf,-20°C

ACO

6) OAC

0

0: 1

Z, benzyloxywbonyl;pMBn, pmethoxybenzyk TBDMS, tm-butyldimethylsilyl;TMB,2,4,6-trimethylbenzoyl.

RICHARD R. SCHMIDT AND WILLY KINZY

84

TABLEXXIII Synthesis of Oligomers of Lactosamine (Ref. 159)

Reactions and products

&omy&o+o&x

RO

R'O NPMh

OAc OAC

238 23C 23D

RO

OAC

R = H. X = OBn. A'. R"= -C(CH$, R = Ac. X = OC(-NH)CCi,, R'. R"- -C(CH& R = Ac, X - OBn. A', R"- H

tion. Table XXVII shows that a-and /?-trichloroacetimidates27a-a and -/? may be obtained directly and also that disaccharidedonors were successfully prepared. These compounds can be used for selective /?- and a-glycoside bond-formations(57,97).

f. Trichloroacetimidates of Galactosamine Derivatives as Glycosyl (N-acetylgalactosamine) (63, Donors.-2-Acetamido-2-deoxy-~-galactose 64) is a constituent of the core structure of mucin-type oligosaccharides;it is a-U-connected to serine and threonine. The derived U-glycoproteins constitute, along with the N-glycoproteins, a major class of glycoconjugates. In glycosphingolipids, N-acetylgalactosamine is mainly encountered in the globo, isoglobo, and ganglio series. Representative examples of these connections are compiled in Table XXVIII. Obviously, /?-(1 + 3)-, /?-(1 + 4)-, and a-(1 + 3)-connections are most important and therefore N-phthaloyl protection is not appropriate for the production of versatile donors. Table XXIX demonstrates that various protective-group patterns are compatible with trichloroacetimidate formation, and not only a but also /? derivatives may be generated highly selectively, as for instance 29b-/?(149)

TABLE XXIV Reaction of 2-Azido-2-deoxytrichloroacetimidatesof Lactosamine with Nucleophiles Trichloroacetimidate

Glycosyl acceptor.

Reaction conditions

Yield (%)

Reference

21f-a

CH,Cl, - n-hexane, BF, * OEt,, - 15°C

115

2lf-a

CH,Cl,- n-hexane, BF, * OEt,

154

2 1g-a

CH,Cl, - n-hexane, BF, * OEt,, MS 4 A, -25°C

92

21h-a

CH2C12- n-hexane, Me,SiOTf, -20 "C, 1 h

114

21i-a

CH2Cl, - n-hexane, BF3 * OEt,, -15°C

162

a

Z, benzyloxycarbonyl; TBDMS, tea-butyldimethylsilyl.

TABLE XXV Synthesis of Oligosaccharides with W Determinants Glycosyl donor

Glycosyl acceptor

"&OY

cc13

NPMh

& O B n.H o O +

NH

HO

% ' $OBn

OBn

OBn

22D

25a-P Trichbmacetimidate Formation: CHzC12. CC13CN, DBU; 67 % Glycosylatbn Condnbns: (CHzC1)z,BF30Elz;67 % B (1-3)'04)

BnO

OBn OBn

Tkhbmacetimidate Formation: n. n. Glycosylatbn Condaions:(CHZCI)~. BF30Etz: 52 % B'")

I

den BnO

OBn BnO

Tkhbmacetimidate Formation: CHzCIz,CC13CN, DBU, 5 h, m m temp.: 94 %, a : j3 5 : 1 MS 4 A. -25'C, Glycosylatbn Conditions: CH,Cl.$bhexane. 1 l), BF30Et2:81 %p3Li

OH

OBn

/OBn

25e-a

(continues)

TABLE XXV (continued) Glycosyl donor

Glycosyl acceptor

ox

eel,

H

o

e 0+oBn

25F

m m

OBn

OBn

GlycosylationCondiions:CH3CN, TMSOTI 0 02 eq.), -40°C; 80 % pl'\

&

HO@

OBn

0

OPiv BnO

OBn

25G

GlycosylationCondliins: CH CN TMSOTI 0 02 eq ), & ~ ;74% ' '

&4

TABLEXXVI Glycosylation of Chitobiose Derivatives Trichloroacetimidate formation

Glycosyl donor.

Glycosyl acceptor

Clycosylation conditions

References

NH

BnO NPhth

NPhth

OAc

OMP

CCl,CN, DBU (CICH,),, -5"C, Ar; 96%

m W

H

O 26A

B

(CICH,),, BF, . OEt,, -2O"C, Ar; 87% (two steps)

127

(C1CH2),,BF, * OEt,, -23"C, Ar; 84%

127,142

NH

BnO NPMh

NPMh OH

OMP

CCl,CN, DBU -55"C, 30 min; 78.6% MP,pmethoxyphenyl.

26A

TABLE XXVII Synthesis of Glycosides of Muramic Acid

CH,Cl,, CCl,CN, NaH, 40°C; 90% (Ref 57)

**

+cia

0 NH

CH,Cl,, CH,Cl,, K2C03,room temp.; 86% (Refs. 57,97)

na-p

A C W H e

CH,Cl,, K,CO,/NaH, CCl,CN, room temp., 8 h 75%,a : p6: 1 (Ref. 148) ccb

m G1ycosy1 acceptor

G1ycosy1 donor 27a-a

0

HO-!'-(OBn)

(27Al

27a-o

Ho&mn

0

Reaction conditions

Yield (%a)

Anomeric configuration (a:j?)

Reference

CH,Cl,, room temp., 3 h

60

0: 1

57

CHzC12, -2O"C, Me,SiOTf, 4 h

70

1 :o

57

CH,Cl,, -2O"C, Me,SiOTf

71

3: 1

57

CH,Cl, - n-hexane -lOT, BF3 * OEtz, 3 h

92

0: 1

57

CH,Cl, - n-hexane - lOT, BF, . OEt,, 30 min

80

0: 1

51

CH2C1,- n-hexane -S T , BF, * OEtZ, 6 h

80

1:4

57

EhO, MS 4 A,N,, -2O"C, 2 h, Me3SiOTf

91

1 :o

91

27a-B

CHzCl,, MS 4 A, 3 h, BF3 * OEt,

85

0: 1

97

27a-a

CHZCl,, BF3 * OEh, room temp., 14 h

38

0: 1

97

27a-a

27a-a

27a-P

HNZ

Ho-lfoBn 0

BtlO "3

TABLE XXVIII Structures of N-Acetylgalactosamine-ContainingGlycosphingolipids

-

Gala-Series a-GalNAc-( 1 + 3)-/3-GalNAc-(1 3)-cu-Gal-( 1 + 4)-/3-Gal-(1 + 0)-Cer Globo-Series &GalNAc-( 1 + 3)a-Gal-( 1 + 4)-/?-Gal-( 1 + 4)-/?-Glc-(1 + O)-Cer a-GalNAc-( 1 + 3)-/%GalNAc-(1 3)-cuGal-( 1 + 4)-/?-Gal-(1 4)-/%GlO( 1 + 0)-Cer D-Gal-( 1 3)-/?-GalNAc-(1 + 3)-cr-Gal-( 1 4)-/?-Gal-( 1 4)-/?-Glc-(1 -+ 0)Cer Isoglobo-Series BGalNAc-( 1 + 3)a-Gal-( 1 + 3)-/?-Gal-(1 +4)-/?-Glo(1 0)-Cer Ganglio-Series F a l N A c - ( 1 + 4)-&Gal-( 1 4)-/?-Glc-(1 + 0)-Cer /3-Gal-(1 3)-/?-GalNAc-(1 + 4)-/?-Gal-(1 + 4)-/?-Glc-(1 + 0)-Cer Lacto-Series a-GalNAc-( 1 3)-/?-Gal-(1 + 3)-/?-GlcNAc-(1 3)-/?-Glc-(1 4)-/?-Glc-(1 0)-Cer 2

-

+

-

-

t

a 1Fuc

-

--

-- -

-

Arthro-Series /?-GalNAc-(1 4)-/%GlcNAc-(1 3)-/?.-Man-( 1 + 4)-/?-Glc-(1 + 0)-Cer a-GalNAc-( 1 + 4)-/?-GalNAc-(1 4)-/3-GlcNAc-(1 + 3)-&Man-( 1 + 4)-/?-Glc-(1 + 0)-Cer Phosphoglycosphingolipids 4-OMe-/?-Gal-(1 + 3)-/%GalNAc-(1 + 3)-a-Fuc-( 1 4)-pGlcNAc-( 1 + 2)-Man

-

Globotetranosylceramide Forssman antigen Globopentaosylceramide Isoglobotetraosylceramide Gangliotriaosylceramide Gangliotetraosylceramide

(Fragment)

TABLEXXIX Synthesis of Trichloroacetimidates of N-Acetylgalactosamine Yield Trichloroacetimidate

29a-a

BnO

Reaction conditions

(W

Anomeric configuration (a:B)

Reference

CH,Cl,, CCl,CN, NaH, lh

68

1

:o

137

CH,Cl,, CCl,CN, K,C03

88

0: 1

149

3: 1

169

O Y N H

BnO

BnO

,OMEM

%

29du

(CH,Cl),, CCl,CN, DBU room temp., 3 h

(CHZCI),, CCI,CN, DBU, room temp., 2 h

81

1 :o

169

(CHZCl),, CCl,CN, DBU, room temp.

79

1 :o

169

I

CCI,

29e-0

BnO

kCI3

(continues)

TABLEXXIX (continued) Yield Trichloroacetimidate

W P

Reaction conditions

Anomeric configuration

(%I

Reference

(CHKl),, CCI,CN, DBU, room temp., 2.5 h

81

1 :o

169

DME, CCI,CN, NaH, 2h

64

1 :o

137

CH,Cl,, CCl,CN, K,CO,, 6h

5 1"

0: 14

137,116,170

CH,CI,, CCl,CN, NaH, Ih

63"

1 :o"

137

CCl, AcO

29h-8

O k

AcO $&oyccl, N3

NH

4

FI

00 d

c

?

w

d

FI

w

.-?

95

4

5

d

0

I

)-$

z

?

N

6

TABLEXXIX (continued)

Trichloroacetimidate

Reaction conditions

Yield (No)

Anomeric configuration (a:B)

Reference

116

CH2C12,CCl,CN, K,CO,/NaH, room temp., 6 h

95

1 :o

150

CH2Cl2, CCl,CN, K2CO3

76

0: I

150

CH2C12,CCl,CN, NaH, room temp.

75

1:o

150

CCl,

CCI, NH

0

4

5:

.$ 9" m

2

0

v) W

0

0

P

d

91

5:

0

98

RICHARD R. SCHMIDT AND WILLY KINZY

and 29i-p (1 13). Useful building blocks for efficient oligosaccharide syntheses are thus readily accessible. Trichloroacetimidates 29b-i ( 1 16, 137,149,169,170)are versatile building blocks for 3,6-branched core-structures of mucin-type oligosaccharides. The selective formation of compound (137) 29g-a from the corresponding 1,3-0-unprotectedazido-galactosederivative demonstrates again that only partial 0-protection may be required, because the anomeric hydroxylic group is more reactive toward trichloroacetonitde under basic conditions than the other hydroxyl groups. This aspect, which could decrease the number of protection and deprotection steps, has not yet been fully considered in the planning of complex of oligosaccharide syntheses. The first glycosylation experiments were carried out with donor (149) 29b-p, which with Me,SiOTf as catalyst exhibited high a selectivities;with the a-trichloroacetimidates (169) 29d-f and BF, * OEt, as catalyst in the nonpolar solvent toluene, excellent p selectivities were observed. In more recent glycosylations the a-connection to serine played a prominent role. Typical results with monosaccharide and oligosaccharide donors having azidogalactose at the reducing end vary (Table XXX). As expected, reactions with a-trichloroacetimidates, employing BF, OEt, as catalyst, are not a selective. Obviously, p-trichloroacetimidates and Me,SiOTf at low temperatures are of advantage for attaininghigh a selectivity,as indicated in the reaction of donors 29b, g, h, m, and p, and serine and threonine acceptors 30A - E.

-

g. Trichloroacetimidates of Mannosamine Derivatives as Glycosyl Donors. -The relatively rare occurrence of 2-acetamido-2-deoxy-~-mannose in Nature has consequently drawn little attention to its glycosylation reactions. The azido derivatives 31a-a and 3b-a (Table XXXI) have been successfullyprepared. Reaction of 31b-a has been successfullyemployed for phosphonate formation. h. Trichloroacetimidates of 6-Deoxyhexoses: Fucose, Rhamnose, and Quinovose. -(9 0-Fucopyranosyl trichloracetimidates:Inverse Procedure for Glycosylation.-L-FUCOS~ is an important constituent of glycosphingolipids. Because most of the tumor-associated blood-group glycosphingolipids have been found to contain a-connected ~-fucose,for instance Le" and Ley, a-fucosylation constitutes an important task in glycosphingolipidsynthesis ( 174).To this aim, the tri-0-benzylfucosyldonor (175,176)32a (Table XXXII) has been prepared in high yield. Reaction with galactose acceptors led, with Me,SiOTf as catalyst in ether, to high yields of H-disaccharide ( 174),the determinantofblood group 0.With the (lessreactive) lactosamine derivatives as acceptors, lower yields were observed mainly because of de-

TABLEXXX Glycosylation with Galactosamine Trichlomcetimidates

Trichloroacetimidate

Reaction conditions

Glycosyl acceptor HNZ

Yield

(%I

Anomeric con6guration

(a:n

Reference

30A

CH,Cl,, Me,SiOTf, -2 0 T , 30 min

81

4:1

116,150

29g-a

30A

CH,Cl,, Me,SiOTf, -20°C

43

1 :o

137

29h-a

30A

CH,Cl,, Me,SiOTf, - 15°C

81

5: 1

171

30A

CH,Cl,-n-hexane, - 30°C, Me,SiOTf

86

1 :o

116

30B

CH,Cl,-n-hexane, - 30°C, Me,SiOTf

55

1 :o

I16

3OC

CH,Cl,- n-hexane, - 3 0 T , Me,SiOTf

60

1 :o

116

30D

CH,Cl,-n-hexane, - 3 0 T , Me,SiOTf

80

1 :o

116

30E

CH,Cl,-n-hexane, -3 0 T , Me,SiOTf

78

1 :o

116

29b-/?

HO-lfOBn 0

W W

29h-8 HNBoc

29h-8 0 HNBoc

2911-8 HNZ

2911-8

HNFMOC

2911-8 0

(continues)

TABLE XXX (continued)

Trichloroacetimidate

Reaction conditions

Glycosyl acceptor

Yield (YO)

Anomeric configuration @:B)

Reference

29m-p

30A

CH,Cl,-n-hexane, -2O"C, Me,SiOTf

85

l:o

116

29n-a

30A

CH,Cl,, Me,SiOTf, - 30°C

86

2: 1

115

29p-a

30A

CH,Cl,, Me,SiOTf, -20°C

88

1 :o

154

30F

CH,Cl,, n-hexane, ZnCI,. OEt,, room temp., 15 h

81

29i-a

Y

z

HOo

O &

o OBn ~

O

B

n

1 : 1,2

113

OBn

(oho@oB

29j-a

HO

OBn

30G

CH,CN, Me,SiOTf, -40T, 15 min

46

0: 1

117

30H

CH,CN, Me,SiOTf, -40"C, 15 min

38

0: 1

117

OBn

OBn

29j-a

Po Ho&oy

COOMe

,% ,

OAc AcO

OBn

OBn

OBn

-.. 0

rn v)

s m

m'

c

0

O & : m"0

2

6

0

P

R

c.(

r"

m

m N

2 0

0

0

L(

101

m

5

W

d

2

TABLE XXX (continued)

Trichloroacetiddate

Glycosyl acceptor

Anomeric coafiguration

Reaction

Yield

conditions

(04

(a:8)

(CH,Cl)t, BF, OEt2,

n.n.

1 :2.7

Reference

L

8

COOMe

CCI,

30A

3OC-ff OAc

MS 4 A, - 15°C 30 min

121

TABLE XXXI Glycosylation of Trichloroacetimidatesof 2 - A z i d o - 2 d e o x y - ~ - m eDerivatives ~~

Tricbloroacetimidate

&

AcO AcO

Glycosyl acceptor

Reaction conditions

Reference

P(OCH,),

Trichloroacetimidateformation: CHzClZ, CCl3CN, NaH, room temp. Glycosylation: CHzClz,Me,SiOTf; 58%, a:/? 6: 1

147

31A

c

O Y N H

8

31a-a

“‘3

pivop /.Q$ 0

II

HO-P(OBn),

31B

OPiVo

PiVO

opiv

Trichloroacetimidateformation: CHzCl2, CCl3CN, NaH; 60% aP Glycosylation: CHzC12,BF, * OEt,, - 10°C;61%

O YCCI, N H

31b-a ~

From epimenc mixture.

~

~

~~~~

150

RICHARD R. SCHMIDT AND WILLY KINZY

104

TABLE XXXII Synthesis of Trichloroacetimidatesof Fucose Yield Compound

Reaction conditions

(%)

NH

o ~ c c 1 3

Anomeric configuration

(a:n

Reference

CH2Cl2, CCl,CN, K2C03,room temp.

50

1 :o

175,58b

CH2Cl2, CCI,CN, DBU, room temp.

65

1 :o

162

CH2C12,CCl,CN, DBU. room temp.

79

1 :o

162

CH,C12, CC13CN,

71

1 :o

58b

CH,CN, CCl,CN, K2C03,room temp.

76

2:3

58b

CC13CN, K2C03

90

1:1

178

I e

o

B

n

32a

En0

0I

c a 3

I

NH

o ~ c c 1 3 NaH, room temp. CH3 AcO OAc

32C

BZO

BZO

OBZ 32d

composition of the highly reactive fucosyl donor 32a under the reaction conditions. Therefore, an alternative reaction procedure is required. Glycosylations and also fucosylations are generally carried out as a formally termolecular reaction of donor (D), acceptor (A), and promotor or catalyst (C) (depending on the amount required) (1,4). Because of differences in the affinities, the reaction course is expected to be first DC interaction, followedby interaction of the DC complex with A (Scheme 10,reaction course I). Obviously, for this sequence of interactions, donors and acceptors with matching reactivities are required. Therefore, acceptor and donor reactivities are often vaned by changing the protective-group pattern and, in addition, the donor reactivity is varied by the selection of leaving groups and

ANOMERIC-OXYGEN ACTIVATION FOR GLYCOSIDE SYNTHESIS

105

D+C+A

SCHEME 10.- Postulated Reaction Courses.

catalysts (1,4). However, this strategy is less successful for very reactive glycosyl donors, which may decompose in the presence of the catalyst while awaiting reaction with the acceptor. Therefore, complexation of acceptor A with the catalyst C prior to interaction with the donor D (Scheme 10, reaction course 11) should overcome this problem. The efficiency of this approach could be demonstrated in a-fucosylation with donor 32a and the acceptors 19C and 33A-D (Table XXXIII). Thus, with the help of this inverse procedure, the versatile building blocks for syntheses of the Lea,Le", Ley, and H antigen determinants are readily accessible (176). Presumably, this procedure may become of general importance when reactive glycosylating agents are employed. Alternatively, the reactivity of the fucosyl donor could be decreased, as has been recently proven very successfully (177). Acyl-protected fucosyl donors have also been generated very successfully (Table XXXII) (58b). Their reaction with acceptors led via neighboringgroup participation to /3 products (58b,178).

(Z)0-Rhamnopyranosyl Trichloroacetimidates.-Rhamnosides are mainly found in plant heteroglycans (63,64). Some rather preliminary investigations have been carried out with rhamnose derivatives. The trichloroacetimidates obtained as rhamnopyranosyl donors are listed (124,178 - 18 1) in Table XXXIV. Their structural similarity to mannose explains the ready formation of a-glycosidic bonds. (iii) 0-Quinovopyranosyl Trichloroacetimidates.-Quinovosides (6deoxyglucosides)are found, for instance, as constituents of many saponins, which are composed of a carbohydrate portion attached to an aglycon that is a complex steroid in asterosaponins (182). Their dramatic biological effects have provided a motivation for structure elucidation and also for synthesis (183). The trichloroacetimidate donors 35ad prepared are listed in Table XXXV. They have been successfully used in oligosaccharide synthesis. Likewise, a 6-sulfoquinovosyl trichloroacetimidate has been successfully prepared (58a).

TABLE XXXIII Reaction of Trichloroacetimidates of L-Fncose with U-Nucleophiles

Glycosyl donor

Reaction conditions

Glycosyl acceptor'

Yield (%)

Reference

(C2H5),0,Me,SiOTf (0.01 eq.), room temp., IP

85

165

(C,H,),O, Me,SiOTf (0.01 eq.), room temp., IP

78

165

72

177

NH

B"b

19c

32a

L

o 0 \

32a 33A

CH,CI,, Me,SiOTf (0.02 eq.), room temp., IP, 1.5 eq. 32a 32a

CH,CI,, Me,OTf (0.02 eq.), room temp., IP, 4 eq. 32a

N3

338

(3-0) 71

(3,2' di-0)

177

(0

I

N

2

g3

F.? e

9 O+

"a

N

2

?

e

N"

-3 -.

107

n

( 0 0

zCI)

m

5

ri 0 0

TABLEXXXIV Glycosylation of L-Rhamnopyrmosyl Derivatives Glycosyl donor

Trichloroacetimidate formation

Glycosyl acceptop

Reaction conditions

Reference

NH

I1 O A m ,

E":S+

CCl,CN, NaH, CH,Cl,, room temp., 30 min; 85%

&&

BnOBnO

OBn

OBn

OBn

34a-a

CHzCl,, ~ T s O H ; 8 6 % ~

124

CH,Cl,, pTsOH; 9 6 % ~ ~

124

CH,Cl,, pTsOH; 10% a

124

34A

''&

xo

34B

OBn

OBn En0

d d

2

I "

0

x

o\

I .

3

I

gq= 0

3

%

109

8

0

2

RICHARD R. SCHMIDT AND WILLY KINZY

110

TABLEXXXV Synthesis of Trichloroacetimidatesof D-Quinovose ~~

Glycosyl donoP

* :A

AcO

35a

Yield

Anomeric configuration

Reaction conditions

(%)

(a:m

Reference

CHzCl,, CCl,CN, DBU, room temp.

92

1 :o

183

CH,Cl,, CCl,CN, DBU, room temp.

90

1 :o

183

CH,Cl,, CCl,CN, DBU, room temp.

84

6: 1

183

CH,Cl,, CCl,CN, DBU, room temp.

88

1:o

183

KCCb NH

* :A

oKcc13 NH

35b

n

NH

Me

Bno,z95+) BnO

3%

oKcc13 NH

* All, ally].

i. Trichloroacetimidates of 2-Deoxyhexoses: 2-Deoxy-~-arabino-hexose.-The presence of the 2-deoxy-~-~-arubino-hexopyranoside (“2deoxy-/3-D-glucopyranoside”) moiety in natural products has stimulated various approaches for the selective synthesis of this glycosidic bond (184189). A temporary 2-phenylthio group as a neighboring group, generating an episulfonium-ion intermediate during glycoside-bond formation, seems to be advantageous because it is also readily removable by hydrogenation, affording the desired 2-deoxy sugar (188,189). Successful application of the trichloroacetimidatemethod to this problem required (i) a convenient synthesis of a 2-S-phenyl-2-thio-~-glucose derivative, subsequently(iz) a stable a-trichloroacetimidate,and finally (iii) high diastereoselection in the glyco-

ANOMERIC-OXYGEN ACTIVATION FOR GLYCOSIDE SYNTHESIS

11 1

syl transfer. This could be accomplished starting from tri-0-benzyl-D-glucal, as shown in Table XXXVI (190). The 2-phenylthio-substituted trichloroacetimidate 36-a! was readily obtained and it exhibited extraordinarily high reactivity; reactions with different acceptors were fast even at temperatures as low as - 95 "C,affording preferentially P-glycosides36a - d in high yields. Transformation into the desired 2-deoxy derivatives 36A-D was readily achieved by treatment with Raney nickel (190). Obviously,extension of this methodology to other 2-deoxy sugars and also to selective formation of a-glycoside bonds with 2-deoxy sugars should be feasible. The extension of this methodology to 3-deoxy-2-glyculosonates(for instance, Neu5Ac) is under investigation. j. Trichloroacetimidatesof Glucuronic Acid. -D-Glucosiduronate (glucuronide) formation is an important means for detoxification in mammals and leads to soluble conjugates that can be excreted via the urine. Glycosides of D-glucuronic acid occur also in many microbial, plant, and animal polysaccharides (for instance, in heparin) (1 9 1). Glycoside-bond formation with the help of trichloroacetimidateshas been accomplished quite successfully (169,192,193). To this aim, donors 37a-c have been synthesized from the 1-0-unprotected derivativesin high yields (Table XXXVII). Representative examples of their reaction with various acceptors are compiled in Table XXXVIII. k. Trichloroacetimidates of Pentoses. -Thus far, there has been relatively little activity in the application of the trichloroacetimidate method to formation of pentopyranosides and pentofuranosides (46,183,194- 199). The reported examples exhibit results similar to those already discussed, and thus special limitations are not expected. 1. Reactions of 0-Glycosyl Trichloroacetimidates with N-,S-, C-,and P-Acceptors. -Only a few studies with N-nucleophiles have been performed. Hydrazoic acid, as a strong acid, reacts with 0-glycosyl trichloroacetimidates and readily gives the thermodynamically most stable glycosyl azide without any additional catalyst (53)(Scheme 6). Nitrogen heterocycles require an acid catalyst for reaction; thus, bis-(trimethylsilylated)uracil and thymine gave, with trichloroacetimidate la-a, exclusively the P-linked nucleosides at room temperature with boron trifluoride etherate as catalyst (1,53,200). Reactions in nitriles as solvent lead during workup to trapping of nitrilium adducts (53,78). The strong interest in 1-thioaldoses and 1-thioglycosides (66,201) as a consequence of their recent use as anomeric protecting-groups, and concomitantly for glycosyl transfer with the help of thiophilic activators, led to

TABLE XXXVI Synthesis of 2-Deoxy-~~ar~bin~-hexopyranosides 1. PhSCI. room temp.

Na,C03, THF

+

EF3OEt2, Et,O/

En0

CH2C12 HOR

2. CCI,CN. NaH, room temp. (70%)

36-a

# Glycosyl acceptor

: -

Bno& BnO

36a-d (X = SPh) 36A-D (X = H)

cCi3

Reaction conditions

OR

X

Yield

Anomenc configuration

(To)

(a:B)

Reference

36a

90

8: 1

190

-4O"C, 15min 36b

90

1:o

190

-6O"C, 15min 36c

85

3: 1

190

1 :o

190

20"C,lh

: .

H

H

HO H

2F OMe

6

26

HO

OBn

-95"C, 15min 36d

Bno%2J3 BnO OMe

83

ANOMERIC-OXYGEN ACTIVATION FOR GLYCOSIDE SYNTHESIS

1 13

TABLEXXXVII Synthesis of Trichloroacetimidatesof D-Glucuronate Compound

Reaction conditions

COOMe

RoRXS+ Ro

373 (R = Bn)

NaH, CC13CN,CH2C12,15 min (98%) (Ref. 192)

37b (R = Ac)

DBU, CCl,CN, (CH2CI),, I h (92%) (Refs. 169,193)

37c

K2C03,CCl,CN, CH2C12,8 h (73%) (Ref. 193)

'YNH CCI,

COOMe

-&+

BnO

OBn

oKccI3

NH

TABLEXXXVIII Synthesis of /?-D-Ghcosiduronates ~

Reaction conditions

(%)

Anomeric configuration (a!:B,

CH2CI2, BF3 . OEt2,

88

0: 1

I92

Yield Donor

Acceptor

37a

~~

Reference

-25"C, 2 h

37a

CH2C12,BF3 * OEt,, - 30°C, 2 h

88

0: 1

192

37a

CHZCl2, BF3 . OEt2, - 30°C 2 h

82

0: 1

192

37b

Toluene, Me,SiOTf, -20°C

75

0: 1

169

Toluene, Me,SiOTf - 20°C

72

0: 1

169

37b HO

OMe

114

RICHARD R. SCHMIDT AND WILLY KINZY

the study of the reactivity of 0-glycosyl trichloroacetimidatesin the glycosylation of S-nucleophiles (66). In the examples investigated employing 0acyl- and 0-benzyl protected donors, generally high reactivity was observed. Surprisingly, with the 0-benzyl protected trichloroacetimidate la-a in the presence of boron trifluoride etherate as catalyst, 1-thioglycosidesof the a configurationare obtained exclusively. Because the anomeric effect in alkyl 1-thioglycosides supposedly corresponds approximatelyto that in alkyl glycosides (202,203), under the reaction conditions kinetically controlled /%product formation was expected; under thermodynamic control, both anomers should be formed. Obviously, glycosyl transfer to the S-nucleophiles in these cases occurs by a different mechanism. It was assumed, that, as in SNi reactions, the configuration is retained by intramolecularreaction via a tight ion-pair (66). Thiocarboxylic acids react again without the addition of any acidic catalyst to provide 1-S-acetyl-1-thiosugars (66,68). The great interest in C-glycosyl compounds is reflected in the extensive research in this field (204). Successful investigations with 0-glycosyl trichloroacetimidatesas glycosyl donors and phenol ethers (199,207,208), silyl enol ethers (205,206),trimethylsilyl cyanide (205,206), and allyltrimethylsilane (206) as C-acceptors underline the wide scope of these highly reactive glycosyl donors. The biological importance ofglycosyl phosphatesprompted interest in the synthesisof glycosyl phosphates as structural analogues (209,210).Excellent examples for their synthesis were contributed by the reaction of trichloroacetimidates 171-a, 29a-a, and 39a-a with trimethyl phosphite in presence of Me,SiOTf as catalyst (21 1) (Table XXXIX).Attack at phosphorus and subsequent 0-demethylation led in a Michaelis- Arbuzov type of reaction to the desired products. Obviously, various other elements or their derivatives are conceivableas glycosyl acceptors. These may react either directly as strong acids (as for instance hydrogen halides, see Scheme 6) or as good nucleophiles react in the presence of a catalyst with the highly reactive 0-glycosyl trichloroacetimidatesas donors. IV. OTHERANOMERIC-OXYGEN ACTIVATION METHODS

1. Other Glycosyl Imidates, Glycosyl Carboxylates, and Glycosyl Sulfonates Base-catalyzed addition of glycosyl oxides for anomeric 0-activation has been extended meanwhile to trifluoroacetonitrile(see Scheme 9), to dichloroacetonitrile, to 1-aryl-1,l -dichloroacetonitriles, and to ketene imines (46,5 1,52). Also 24 glycosy1oxy)-pyridineand -pyrimidine derivatives were readily prepared from the corresponding 2-halo precursors (78). However,

TABLE XXXIX Reaction of O-Glycosyl Trichloroacetimidates with P(OMe),

Donor

Reaction conditions

+& - * A AcO

Product

CH,Cl,, room temp. 1 h, Me,SiOTf

A&% AcO

Yield

Anomeric configuration

(W

(a:B)

Reference

16

l:o

21 1

N3 P(OMe),

II

0

O Y N H CCl,

171-a

Bny

211

.OAc

CCI,

CH,Cl,, room temp. 1 h, Me,SiOTf AcO

39a-a

A

c ACO

OYNH CCI3

O

59

G

;;(OMei2

0

6: 1

21 1

116

RICHARD R. SCHMIDT AND WILLY KINZY

none of the imidate donors thus obtained seems to exceed the 0-glycosyl trichloroacetimidatesin terms of ease of formation, stability, and reactivity. Acetimidate formation with N-methylacetamide and acylated glycosyl halides according to Sinay et a1 (2 12,213), using three equivalents of silver oxide as an activator, leads neither to particularly stable nor to reactive donors. Any other developments along these lines have already been summarized in previous reviews (1,3). The same is mainly true for anomeric 0-activations via 1-0-acylation (1,214), including orthoester (1) formation, 1-0-alkylation ( 1,215) and -silylation ( l), and 1-0-sulfonylation (1). 2. Glycosyl Phosphates and Related Systems

One of the most important direct nucleophilic substitutions at activated carbon atoms carried out in nature is enzymic 0-and N-glycosyl bond-formation at the anomeric carbon atom (2 16). At this activated position, the leaving groups are phosphates, pyrophosphates, and their nucleoside and lipid ester derivatives, which are biosynthesized via anomeric O-phosphorylation reactions. In vitro anomeric 0-phosphorylation readily furnishes dialkyl or diary1 glycosyl phosphates (2 17). These also exhibit, in inert solvents in the presence of boron trifluoride etherate or Me,SiOTf as catalysts, good glycosyl donor properties comparable to those of glycosyl fluoridesand sulfides, respectively, as reported elsewhere (17). However, with A=B-C-H systems as acceptors(see Section 111.2), where a catalyst is not required, their reactivity is similar to that of the very reactive trichloroacetimidate donors, as indicated by competition experiments ( 17). Thus, contrary to a recent statement (2 18), not only in vivo but also in vitro nucleophilic substitution at activated carbon atoms, as exemplified by the anomericcenter, can be efficientlyperformed with glycosyl phosphates. This was recently demonstratednot only for glycosyl phosphates but also for such derivatives related to imidates as 0-P (=X)Y2, where X = 0 and Y = NMe,, X = 0 and Y = Ph, X = S and Y = OMe, and X = NTs and Y = Ph (219-223). V. CONCLUSIONS

The requirements for new glycosylation methods outlined at the beginning of this chapter, namely convenient diastereocontrolledanomeric 0-activation (first step) and subsequent efficient diasterecontrolledglycosylation promoted by genuinely catalytic amounts of a catalyst (second step), are essentially completely fulfilled by the trichloroacetimidate method. This is clearly shown by the many examples and references given in this article. In terms of stability,reactivity, and applicabilitytoward different acceptors,the

ANOMERIC-OXYGEN ACTIVATION FOR GLYCOSIDE SYNTHESIS

117

0-glycosyl trichloroacetimidates have generally proven to be outstanding glycosyl donors, which resemble in various respects the natural nucleoside diphosphate sugar derivativesas glycosyl donors. Thus, base-catalyzed generation of 0-glycosyl trichloroacetimidates and ensuing acid-catalyzed glycosylation have become a very competitive alternative to direct, often uncontrolled acid-catalyzed transformation of sugars into glycosides(FischerHelferich method) or to glycosyl halide and glycosylsulfide formation for the activation step, which requires at least equimolar amounts of a promoter system for the glycosylation step (Koenigs- Knorr method and variations). In addition, the trichloroacetimidate method may be readily adapted for large-scale preparations. ACKNOWLEDGMENTS Thanks are expressed to the Deutsche Forschungsgemeinschaft and to the Fonds der Chemischen Industrie for financial support of our work reported in this chapter.

REFERENCES 1. R.R. Schmidt,Angew. Chem.,98 ( 1986)2 13- 236;Angew. Chem. Znt. Ed. Engl. 25 ( 1986) 212-235. 2. R.R. Schmidt in W. Bartmann and K. B. Sharpless(Eds.), StereochemistryofOrganicand Bioorganic Transformations, Workshop Conferences Hoechst, Vol. 17, pp. 169- 189. VCH Verlagsgesellschaft GmbH, Weinheim, 1987. 3. R.R.Schmidt, PureAppl. Chem., 61 (1989) 1257- 1270;B. M. Trost, I. Fleming, and E. Winterfeldt (Eds.), Comprehensive Organic Synthesis, Vol. 6, pp. 33 -64. Pergamon Press, Oxford, 1991. 4. H. Paulsen, Angew. Chem., 94 (1982) 184-201;Angew. Chem. Znt. Ed. Engl., 21 (1982) 155- 173. 5 . H. Paulsen, Angew. Chem., 102 (1990) 851 -867; Angew. Chem. Znt. Ed. Engl., 29 (1990) 823-839. 6. H. Kunz, Angew. Chem., 99 (1987) 297-311; Angew. Chem. Znt. Ed. Engl., 26 (1987) 294-308. 7. K. Krohn, Nachr. Chem. Technol.Lab., 36 (1988) 930-935. 8. W. Koenigs and E. Knorr, Ber., 34 (1901) 957-982. 9. T. Mukaiyama, Y. Murai, and S. Shoda, Chem. Lett. (1981) 431 -432. 10. K. C. Nicolaou, A. Chucholowski, R.E. Dolle, and J. L. Randall, J. Chem. Sot. Chem. Commun. (1984) 1155-1156. I 1. S. Hashimoto, M. Hayashi, and R.Noyori, Tetrahedron Lett., 25 (1984) 1379- 1382. 12. H. Kunz and W. Sager, Helv. Chim. Acta, 68 (1985) 283-287. 13. M. Kreuzer and J. Thiem, Curbohydr. Res., 149 (1986) 347-361. 14. T. Matsumoto, H. Maeta, K. Suzuki, and G.4.Tsuchihashi, TetrahedronLett., 29 (1988) 3567- 3570. 15. K. Suzuki,H. Maeta, T. Matsumoto, and (3.4. Tsuchihashi, Tetrahedron Lett. 29 (1988) 3571-3574. 16. K. Suzuki, H. Maeta, and T. Matsumoto, Tetrahedron Lett., 30 (1989) 4853-4856.

118

RICHARD R. SCHMIDT AND WILLY KINZY

17. R. R. Schmidt in H. Ogura, A. Hasegawa, and T. Suami (Eds.), Carbohydrates-Synthetic Methods and Application in Medicinal Chemistry, pp. 66 - 88. VCH Verlagsgesellschafl mbH, Weinheim, 1992. 18. P. Fugedi, P. J. Garegg, H. LMnn, and T. Norberg, GIycoconjugateJ., 4 (1987) 97- 108. 19. T. Norberg and M. Walding, GlycoconjugateJ., 5 (1988) 137- 143. 20. R. J. Femer, R. W. Hay, and N. Vethaviyasar, Carbohydr. Res., 27 (1973) 55-61. 2 1. T. Mukaiyama, T. Nakatsuka, and S. Shoda, Chem. Lett. (1979) 487 -490. 22. J. W. van Cleve, Carbohydr. Res., 70 (1979) 161- 164. 23. S. Hanessian, C. Bacquet, and N. Lehong, Carbohydr. Res., 80 (1980) C17-C20. 24. P. J. Garegg, C. Henrichson, and T. Norberg, Carbohydr. Res., 116 (1983) 162- 165. 25. T. Y. R. Tsai, H. Jin, andK. Wiesner, Can. J. Chem., 62 (1984) 1403-1405. 26. K. Wiesner, T. Y. R. Tsai, and H. Jin, Helv. Chim. Acta, 68 (1985) 300-314. 27. R. B. Woodward [and 48 collaborators], J. Am. Chem. SOC.,103 (1981) 3215-3217. 28. P. G. M. Wuts and S. S. Bigelow, J. Org. Chem., 43 (1983) 3489-3493. 29. S. Koto, T. Uchida, and S. Zen, Chem. Lett. (1972) 1049- 1052. 30. K. C. Nicolaou, S. P. Seitz, and D. P. Papahatijs, J. Am. Chem. SOC.,105 (1983) 24302434. 31. H. Lijnn, Carbohydr. Rex, 139 (1985) 105-113. 32. H. Ldnn, J. Curbohydr. Chem., 6 (1987) 301-306. 33. H. Liinn, Carbohydr. Res., 139(1985) 115- 121. 34. R. Eby and C. Schuerch, Carbohydr. Res., 39 (1975) 33-38. 35. P. Fugedi and P. J. Gareg, Carbohydr. Res., 149 (1986) C9-Cl2. 36. A. Marra, J.-M. Mallet, C. Amatore, and P. Sinay, Synlett (1990) 572-574. 37. R. R. Schmidt and M. Reichrath, Angew. Chem., 91 (1979)497-499; Angew. Chem. Int. Ed. Engl., 18 (1979) 466-468. 38. H. Bredereck,G. Hagelloch, and E. Hambsch, Chem. Ber., 87 (1954)3<-37: thetrehalose formation described in this paper may be interpreted accordingly. 39. R. R. Schmidt, M. Reichrath, and U. Moering, TetrahedronLett., 2 1 (:, (30)3561 -3564. ~ 356<-3568. 40. R. R. Schmidt, U. Moering, and M. Reichrath, Tetrahedron Lett., 21 ( 1 'Q, 41. R. R. Schmidt, M. Reichrath, and U. Moering, Chem. Ber., 115 (1987 1'9. 42. R. R. Schmidt, M. Reichrath, and U. Moering, J. Curbohydr. Chem., 3 (i2d.t) 67-84. 43. J. J. Oltvort, M. Klosterman, C. A. A. van Boeckel, and J. van Boom, Carbohydr. Res., 130 (1984) 147-163. 44. R. R. Schmidt and W. Klotz, Synlett (1991) 168- 170. 45. R. R. Schmidt and J. Michel, Tetrahedron Lett., 25 (1984) 821-824. 46. R. R. Schmidt, J. Michel, and M. Roos, Liebigs Ann. Chem. (1984) 1343- 1357. 47. R. R. Schmidtand A. Esswein, Angew.Chem., 100(1988) 1234- 1236;Angew. Chem. Int. Ed. Engl., 27 (1988) 1178-1180. 48. A. Esswein, H. Rembold, and R. R. Schmidt, Carbohydr. Res., 200 (1990) 287-305. 49. J. Tsvetkov, W. Klotz, and R. R. Schmidt, Liebigs Ann. Chem. (1992) 371 -375. 50. J. U. Nef, Liebigs Ann. Chem., 287 (1895) 265-359. 5 1. J. Michel, Dissertation, University of Constance, 1983. 52. (a) R. R. Schmidt and J. Michel, Angew. Chem., 92 (1980) 783-784; Angew. Chem. Int. Ed. Engl., 19 (1980)73 1 - 732; (b) R. R. Schmidt, M. Stumpp, and J. Michel, Tetrahedron Lett., 23 (1982) 405-408. 53. R. R. Schmidt and J. Michel, J. Curbohydr. Chem., 4 (1985) 141- 169. 54. J. E. Truelove, A. A. Hussain, and H. B. Kostenbander, J.Pharm. Sci., 69 (1980) 231 232. 55. R. R. Schmidt and M. Stumpp, Liebigs Ann. Chem. (1984) 680-691. 56. R. R. Schmidt, H. Gaden, and H. Jatzke, Tetrahedron Lett., 31 (1990) 327-330. 57. W. Kinzy and R. R. Schmidt, Liebigs Ann. Chem. (1987) 407-415.

ANOMERIC-OXYGEN ACTIVATION FOR GLYCOSIDE SYNTHESIS

119

58. (a) M. Hoch, E. Heinz, and R. R. Schmidt, Curbohydr. Res. (1989) 21 -28; (b) R. R. Schmidt, B. Wegmann, and K.-H. Jung, Liebigs Ann. Chem. (1991) 121- 124. 59. A. Esswein and R. R. Schmidt, Liebigs Ann. Chem. (1988) 675-678. 60. R. R. Schmidt and C. Regele, unpublished results. 6 1. H. Gaden, Dissertation, University of Constance, 1990. 62. R. R. Schmidt, M. Behrendt, and A. Toepfer, Synlett (1990) 694-696. 63. R. Schaffer in W. Pigman and D. Horton (Eds.), The Carbohydrates: Chemistry and Biochemistry, 2nd ed., Vol. IA, p. 69. Academic Press, New York, 1972. 64. J. F. Kennedy, Curbohydrute Chemistry, Oxford Science Publications. Clarendon Press, Oxford, 1988. 65. R. R. Schmidt and J. Michel, Angew. Chem., 94 (1982) 77-78; Angew. Chem. Int. Ed. Engl., 21 (1982) 77-78; Angew. Chem. Suppl. (1982) 78-84. 66. R. R. Schmidt and M. Stumpp, Liebigs Ann. Chem. (1983) 1249- 1256. 67. B. Wegmann and R. R. Schmidt, J. Curbohydr. Chem., 6 (1987) 357-375. 68. M. Schuhmacher, Dissertation, University of Constance, 1984. 69. R. Klaer, Dissertation, University of Constance, 1985. 70. P. Zimmermann, R. Bommer, T. Biir, and R. R. Schmidt, J. Curbohydr. Chem., 7 (1988) 435-452. 71. P. Fugedi, P. Nanasi, and J. Szejtli, Curbohydr. Res., 175 (1988) 173- 181. 72. R. Schaubach and W. Kinzy, unpublished results. 73. T. Takano, F. Nakatsubo, and K. Murakami, Cell. Chem. Technol., 22 (1988) 135- 145. 74. G. WulffandG. Rohle, Angew. Chem., 86 (1974) 173- 187;Angew. Chem. Int. Ed. Engl., 13 (1974) 157- 172. 75. R. U. Lemieux, Pure Appl. Chem., 25 (1971) 527-548. 76. I. TvaroSka and T. Bleha, Adv. Curbohydr. Chem. Biochem., 47 (1989) 45- 123. 77. R. R. Schmidt and E. Rocker, Tetrahedron Lett., 21 (1980) 1421- 1424. 78. Y. D. Vankar, P. S. Vankar, M. Behrendt, and R. R.Schmidt, Tetrahedron, 47 (1991) 9985 -9992. 79. G. 0.Aspinall, P. Capek, R. C. Carpenter, and J. Szafranek, Carbohydr.Res., 214 (1991) 95- 105. 80. P. B. Anzeveno, L. J. Creemer, J. K. Daniel, C. H. King, and P. S.Liu, J. Org. Chem., 54 (1989) 2539-2542. 81. N. A. Kraaijeveld and C. A. A. van Boeckel, Red. Truv. Chim. Puys-Bus., 108 (1989) 39-50. 82. H. Paulsen, B. Helpap, and J.-P. Lorentzen, Curbohydr. Res., 179 (1988) 173-197. 83. T. Toyokuni, personal communication. 84. R. R. Schmidt and P. Zimmermann, Angew. Chem., 98 (1986) 722-723; Angew. Chem. Int. Ed. Engl. (1986) 725. 85. P. Zimmermann, Dissertation, University of Constance, 1988. 86. N. P. Singh and R. R. Schmidt, J. Carbohydr. Chem., 8 (1989) 199-216. 87. T. Biir and R. R. Schmidt, Liebigs Ann. Chem. (1988) 699-674. 89. R. R. Schmidt and T. Maier, Curbohydr. Res., 174 (1988) 169- 179. Chem., 99(1987)813-814;Angew. Chem. 90. R.R.Schmidt,T.Biir,andH.-J.Apell,Angew. Int. Ed. Engl., 26 (1987) 793-794. 91. M. Masato, Y. Ito, and T. Ogawa, Curbohydr. Res., 195 (1990) 199-224. 92. R. Bommer and R. R. Schmidt, Liebigs Ann. Chem. ( 1989) 1 107- 11 1 1. 93. T. Murase, H. Ishida, M. Kiso, and A. Hasegawa, Curbohydr. Res., 188 (1989) 71 -80. 94. A. Kameyama, H. Ishida, M. Kiso, and A. Hasegawa, Curbohydr. Res., 200 (1990) 269-285. 95. H. Prabhanjan, H. Ishida, M. Kiso, and A. Hasegawa, Curbohydr. Res., 211 (1991) C1 -C5.

120

RICHARD R. SCHMIDT AND WILLY KINZY

96. A. Toepfer, Dissertation, University of Constance, 1992. 97. A. Toepfer and R. R. Schmidt, Curbohydr. Res., 202 (1990) 193-205. 98. R. R. Schmidt and R. Klager, Angew. Chem., 97 (1985) 60-61; Angew. Chem. Znt. Ed. Engl. (1985) 65. 99. R. Klager, Dissertation, University of Constance, 1985. 100. K. Koike, M. Sugimoto,S. Sato, Y. Ito, Y. Nakahara, andT. Ogawa, Curbohydr.Res., 163 (1987) 189-208. 101. S. Sato, S. Nunomura, T. Nakano, Y. Ito, and T. Ogawa, Tetrahedron Lett., 29 (1988) 4097 -4100. 102. S. Nunomura and T. Ogawa, TetruhedronLett., 29 (1988) 5681 -5684. 103. M. Numata, M. Sugimoto, S. Shibayama, and T. Ogawa, Curbohydr. Res., 174 (1988) 73-85. 104. S. Sato, Y. Ito, T. Nukuda, Y. Nakahara, and T. Ogawa, Curbohydr. Rex, 167 (1987) 197-210. 105. K. Mori and Y. Funaki, Tetrahedron, 41 (1985) 2379-2386. 106. K. Mori and T. Kinsho, Liebigs Ann. Chem. (1988) 807-814. 107. K. Mori and H. Nishio, Liebigs Ann. Chem. (1991) 253-257. 108. K. Ohashi, S. Kosai, M. Arizuka, T. Watanabe, Y. Yamagiwa, T. Kamikawa, and M. Kates, Tetrahedron, 45 (1989) 2557-2570. 109. P. H. Amvam-Zollo and P. Sinay, Curbohydr. Res., 150 (1986) 199-212. 110. K. Sakai, Y. Nakahara, and T. Ogawa, Tetrahedron Lett., 31 (1990) 3035-3038. 111. A. M. P. van Steijn, M. Jetten, J. P. Kamerling, and J. F. G. Vliegenthart, Red. Truv. Chim. Puys-Bus, 108 (1989) 374-383. 112. H. Paulsen, W. Rauwald, and U. Weichert, Liebigs Ann. Chem. (1988) 75-86. 113. P. Zimmermann, U. Greilich, and R. R. Schmidt, Tetrahedron Lett., 31 (1990) 18491852. 114. R. Schaubach, J. Hemberger, and W. Kinzy, Liebigs Ann. Chem. (1991) 607-614. 115. W. Kinzy and R. R. Schmidt, Curbohydr.Rex, 166 (1987) 265-276. 116. W. Kinzy and R. Schaubach, J. Curbohydr. Chem., accepted for publication. 117. U. Greilich, Dissertation, University of Constance, 1992. 118. K. Koike, M. Sugimoto, Y. Nakahara, and T. Ogawa, Curbohydr. Res., 162 (1987) 237 -246. 119. H. Wiegandt, Glycolipids,New ComprehensiveBiochemistry, Vol. 10. Elsevier, Amsterdam, 1985. 120. T. Murase, A. Kameyama, K. P. R. Katha, H. Ishida, M. Kiso, and A. Hasegawa, J. Curbohydr. Chem., 8 (1989) 265-283. 121. H. Iijima and T. Ogawa, Curbohydr. Res., 186 (1988) 107- 118. 122. M. Numata, M. Sugimoto, K. Koike, and T. Ogawa, Curbohydr.Res., 163 (1987) 209225. 123. Y. Ito, M. Kiso, and A. Hasegawa, J. Curbohydr. Chem., 8 (1989) 285-294. 124. P. Fugedi, A. Liptik, P. NBnhi, and A. Neszmelyi, Curbohydr.Res., 107 (1982) C5-C8. 125. V. A. Olah, J. Harangi, and A. Liptik, Curbohydr.Res., 174 (1988) 113- 120. 126. J. Kerekgyarto, J. P. Kamerling, J. B. Bouwstra, J. F. G. Vliegenthart, and A. Liptik, Curbohydr. Rex, 186 (1989) 51-62. 127. F. Yamazaki, S. Sato, T. Nukuda, Y. Ito, and T. Ogawa, Curbohydr. Rex, 201 (1990) 31-50. 128. H. Paulsen and B. Helpap, Curbohydr. Res., 216 (1991) 289-313. 129. H. Paulsen, F. Reck, and I. Brockhausen, Curbohydr.Rex, 236 (1992) 39-71. 130. K. K. Sadozai, T. Nukada, Y. Ito, Y. Nakahara, T. Ogawa, and A. Kobata, Curbohydr. Rex, 157 (1986) 101-123.

ANOMERIC-OXYGEN ACTIVATION FOR GLYCOSIDE SYNTHESIS

121

131. K. K. Sadozai, T. Kitajima, Y. Nakahara, T. Ogawa, and A. Kobata, Carbohydr.Res.. 152 (1986) 173-182. 132. T. Ogawa, M. Sugimoto, T. Kitajima, K. K. Sadozai, and T. Nukada,Tetrahedron Lett., 27 (1986) 5739-5742. 133. H. Paulsen, M. Heume, and H. Niirnberger, Carbohydr. Rex, 200 (1 990) I27 - 166. 134. J. J. Krepinsky, personal communication. 135. H. Paulsen, F. Reck, and I. Brockhausen, personal communication. 136. T. W. Rademacher, R. B. Parekh, and R. A. Dwek, Annu. Rev. Biochem., 57 (1988) 785-838. 137. G. Grundler and R. R. Schmidt, Liebigs Ann. Chem. (1984) 1826- 1847. 137a. R. R. Schmidt and G. Grundler, Angew. Chem., 95 (1983) 805-806; Angew. Chem. Znt. Ed. Engl., 22 (1983) 776-777. 138. P. Tavecchia, M. Trumtel, A. Veyrikres, and P. Sinay, Tetrahedron Lett., 30 (1989) 2533-2536. 139. A. Maranduba and A. Veyrikes, Curbohydr. Res., 151 (1986) 105-119. 140. G. Grundler and R. R. Schmidt, Carbohydr. Res., 135 (1985) 203-218. 141. F.Yamazaki,T. Kitajima,T.Nukada,Y.Ito, andT. Ogawa, Carbohydr.Res., 201 (1990) 15-30. 142. F. Yamazaka, T. Nukada, Y.Ito, S. Sato, and T. Ogawa, Tetrahedron Lett., 30 (1989) 4417-4420. 143. M. Kobayashi, F. Yamazaki, Y. Ito, and T. Ogawa, Curbohydr.Res., 201 (1990) 51 -67. 144. M. Kobayashi, F. Yamazaki, Y. Ito, and T. Ogawa, Tetrahedron Lett. 30 (1989) 45474550. 145. W. Kinzy and R. R. Schmidt, Liebigs Ann. Chem. (1985) 1537-1545. 146. W. Kinzy, Diploma thesis, University of Constance, 1983. 147. K. Briner and A. Vasella, Helv. Chim. Actu, 70 (1987) 1341- 1356. 148. A. Termin and R. R. Schmidt, Liebigs Ann. Chem. (1989) 789-795. 149. W. Kinzy and R. R. Schmidt, Tetrahedron Lett., 28 (1987) 1981- 1984. 150. W. Kinzy, Dissertation, University of Constance, 1986. 151. G. J. P. H. Boons, M. Overhand, G. A. van der Marel, and J.-H. van Boom,Angew. Chem., 101 (1989) 1538- 1539; Angew. Chem. Znt. Ed. Engl. (1989) 1504. 152. R. U. Lemieux, and R. M. Ratcliffe, Can. J. Chem., 57 (1979) 1244-1251. 153. M. Trumtel, P. Tavecchia, A. Veyrieres, and P. Sinay, Carbohydr. Res.. 191 (1989) 29-52. 154. W. Kinzy and R. R. Schmidt, Carbohydr. Res., 193 (1989) 33-47. 155. K.-H. Jung, M. Hoch, and R. R. Schmidt, Liebigs Ann. Chem. (1989) 1099- 1106. 156. H. Ammann and G. Dupuis, Can. J. Chem.. 66 (1988) 1651-1655. 157. Y. Ito and T. Ogawa, Agric. Biol. Chem., 50 (1986) 3231-3233. 158. S. Sato, Y. Ito, and T. Ogawa, Tetrahedron Lett., 29 (1988) 4759-4762. 159. J. Alais and A. Veyrieres, Tetrahedron Lett., 28 (1987) 3345-3348. 160. W. Kinzy, Proceedings of the 13th Symposium on Carbohydrate Chemistry, Stockholm, 1988. [Abstr.] 161. R. Windmuller, Dissertation, University of Constance, submitted. 162. W. Kinzy and A. Low, Curbohydr. Res., 245 (1993) 193-218. 163. S. Sato, Y. Ito, and T. Ogawa, Tetrahedron Lett., 29 (1988) 5267-5270. 164. R. Bommer, W. Kinzy, and R. R. Schmidt, Liebigs Ann. Chem. (199 I) 425 -433. 165. A. Toepfer and R. R. Schmidt, Tetrahedron Lett., 33 (1992) 5 161-5 164. 166. P. Lefrancier and E. Lederer, Fortschr. Chem. Org. Naturst., 40 (1981) 1-47. 167. P. Lefrancier and E. Lederer, Pure Appl. Chem., 59 (1987) 449-454. 168. A. Termin, Dissertation, University of Constance, 1991.

122

RICHARD R. SCHMIDT AND WILLY KINZY

169. J.-C. Jacquinet, Carbohydr. Res., 199 (1990)153- 181. 170. R.R. Schmidt and G. Grundler, Angew. Chem., 94 (1982)790-79 1; Angew. Chem. Znt. Ed. Engl. (1982);Angew. Chem. Suppl. (1982) 1707-1714. 171. T. Toyokuni, B. Dean, and S . 4 . Hakomori, Tetrahedron Lett., 31 (1990)2673-2676. 172. H. Iijima and T. Ogawa, Carbohydr. Rex, 172 (1988) 183- 193. 173. H. Iijima and T. Ogawa, Carbohydr. Res., 186 (1989)95- 106. 174. S.4. Hakomori, Annu. Rev. Biochem., 50 (1981)733-764. 175. B. Wegmann and R. R. Schmidt, Carbohydr. Res., 184 (1988)254-261. 176. R. R.Schmidt and A. Toepfer, Tetrahedron Lett., 32 (1991)3353-3356. 177. R.Windmiiller, Dissertation, University of Constance, 1991. 178. I. Kitagawa, N. I. Baek, K. Ohashi, M. Sakagami, M. Yoshikawa, and H. Shibuya, Chem. Pharm. Bull, 37(1989) 1131-1133. 179. T. M. Slaghek,A. H. van Oijen, A. A. M. Maas, J. P. Kamerling, and J. F. G. Vliegenthart, Carbohydr.Res., 207 (1990)237-248. 180. T. M. Slaghek, A. A. M. Maas, J. P. Kamerling,and J. F. G. Vliegenthart,Carbohydr.Rex, 211 (1991)25-39. 181. G . 0.Aspinall, N. K. Khare, R. K. Sood, D. Chattejee, B. avoire, and P. J. Brennan, Carbohydr. Rex, 216 (1991)357-373. 182. L. Minale, C. Pizza,R. Riccio, and F. Zollo, Pure Appl. Chem., 54 (1982)1935- 1950. 183. X.-B. Han, Dissertation, University of Constance, 1992;X.-B. Han and R. R. Schmidt, Liebigs Ann. Chem. (1992)817-823. 184. K. Tatsuta, K.Fujimoto, M. Kinoshita, and S. Umezawa, Carbohydr. Rex, 54 (1977) 85-104. 185. J. Thiem, H. Karl, and J. Schwentner, Synthesis (1978)696-698. 186. G. Jaurand, J.-M. Beau, andP. Sinay, J. Chem. Soc. Chem. Commun. (1981)572-573. 187. K. Bock, I. Lundt, and C. Pedersen, Carbohydr. Res., 130 (1984) 125-134. 188. K. C. Nicolaou, T. Ladduwahetty, J. L. Randall,and A. Chucholowski, J. Am. Chem. SOC.,108 (1986)2466-2467. 189. Y. It0 and T. Ogawa, Tetrahedron Lett., 28 (1987)2723-2726. 190. R.Preuss and R. R. Schmidt, Synthesis (1988)694-697. 191. G. J. Dutton, Glucuronic Acid: Free and Combined. Academic Press, New York, 1966. 192. R. R. Schmidt and G. Grundler, Synthesis(1981) 885-887. 193. G. Grundler, Dissertation, University of Constance, 1983. 194. S. Friedrich-Bochnitschek, H. Waldmann, and H. Kunz, J. Org. Chem., 54 (1989)75 1 756. 195. M. Masato, Y. Ito, and T. Ogawa, Carbohydr. Res., 195 (1990) 199-224. 196. H. Kunz and H. Waldmann, Angew. Chem., 90 (1984)49-50; Angew. Chem. Znt. Ed. Engl. (1984),71. 197. R.R.Schmidt, M. Faas, and K.-H. Jung, Liebigs Ann. Chem. (1985) 1546- 1556. 198. R.R. Schmidt, W. Guilliard, D. Heermann, and M. H o h a n n , J. Heterocycl. Chem., 20 (1983)1447- 1451. 199. R. R. Schmidt and M. H o h a n n , Tetrahedron Lett., 23 (1982)409-412. 200. M. Hoffmann, Diploma thesis, University of Constance, 1981. 201. D.Horton and D. Hutson, Adv. Carbohydr. Chem. Biochem., 18 (1 963) 123- 199. 202. E. L. Eliel and E. H. Juaristi, ACS Symp. Ser., 87 (1979)95- 106. 203. A. J. Kirby, The Anomeric Effect and Related Stereoelectronic Efects at Oxygen, p. 10. Springer, Berlin, 1983. 204. W. Frick and R. R. Schmidt, Liebigs Ann. Chem. (1989)565-570, and references cited therein.

ANOMERIC-OXYGEN ACTIVATION FOR GLYCOSIDE SYNTHESIS

123

205. R. R. Schmidt andM. Hoffmann,Angew. Chem.,95 (1983)417-418;Angew. Chem.Znt. Ed. Engl., 22 (1983) 406-407; Angew. Chem. Suppl. (1983) 543-544. 206. M. Hoffmann and R. R. Schmidt, Liebigs Ann. Chem. (1985) 2403-2419. 207. R. R. Schmidt and G. Effenberger, Curbohydr.Rex, 171 (1987) 59-79. 208. R. R. Schmidt and G. Effenberger, Liebigs Ann. Chem. (1987) 825-831. 209. H. Paulsen and J. Thiem, Chem. Ber., 106 (1973) 3850-3876. 210. R. Meuwly and A. Vasella, Helv. Chim. Actu, 68 (1985) 997- 1009. 2 1 1. K. Briner and A. Vasella, Helv. Chim. Actu, 70 ( 1987) 1341- 1356. 212. J. R. Pougny and P. Sinay, Tetrahedron Lett. (1976) 4073-4076. 2 13. J. R. Pougny, J. C. Jacquinet, M. Nassr, M. L. Milat, and P. Sinay, J. Am. Chem. Soc., 99 (1977) 6762-6763. 214. M. J. Ford and S. V. Ley, Synlett (1990) 255-256, and references cited therein. 215. B. Fraser-Reid, P. Konradsson, D. R. Mootoo, and R. E. Devitt, J. Chem. Soc., Chem. Commun. (1990) 270-272. 216. M. L. Sinnott in M. I. Page and A. Williams (Eds.),EnzymeMechanisms,pp. 259-297. The Royal Society of Chemistry, London, 1987. 217. M. Inage, H. Chaki, S. Kusumoto, and T. Shiba, Chem. Lett. (1982) 1281-1284, and references therein. 218. F. L. Westheimer, Science, 235 (1987) 1173-1178. 219. S. Hashimoto, T. Honda, and S. Ikegami, J. Chem. SOC.Chem. Commun. (1989) 685687. 220. S. Hashimoto, T. Honda, and S. Ikegami, Chem. Phurm. Bull., 38 (1990) 2323-2325. 221. S. Hashimoto, T. Honda, and S. Ikegami, TefrehedronLett., 31 (1990) 4769-5772. 222. S. Hashimoto, T. Honda, and S. Ikegami, Heterocycles, 30 (1990) 775-778. 223. S. Hashimoto, T. Honda, and S. Ikegami, Tetrahedron Lett., 32 (1991) 1653- 1654.

This Page Intentionally Left Blank

ADVANCES IN CARBOHYDRATE CHEMISTRY AND BIOCHEMISTRY, VOL. 50

SYNTHETIC REACTIONS OF ALDONOLAmONES BY

ROSAM. DE LEDERKREMER AND OSCAR VARELA

Departamento de Quimica Orgcinica, Facultad de Ciencias Exactas y Naturales, Universidad de Buenos Aires, 1428 Buenos Aires, Argentina

I. INTRODUCTION

Aldonolactones are useful starting materials for the synthesis of modified sugars. They have also been used as chiral templates in synthesis of natural products. Some of them are inexpensive, commerciallyavailable products or they may be obtained readily from the respective monosaccharides. The purpose of this chapter is to survey the main reactions of aldonolactones. Previous reviews on the subject include articles on gulono-1,Clactones (1) and D-ribonolactone (2). Methods of synthesis, conformational analysis, and biological properties are not discussed in this chapter. 11. ACETALATION AND ALDONOLACTONES

1. Reaction with Aldehydes

The hydroxyl groups of aldonolactones react with a variety of aldehydes and ketones to give the corresponding acetal derivatives. Treatment of the salts of aldonic acids with benzaldehyde and hydrochloric acid or zinc chloride as catalysts give benzylidene derivatives of aldonic acids or aldonolactones (3). The structures originally proposed (3) for the benzylidene derivatives of D-ribonolactone (1) have been revised (4). Reexamination of the 'H- and I3C-nuclearmagnetic resonance (n.m.r.) data for the product of benzylidenation of 1 with benzaldehyde and hydrochloric acid, described (3,5,6) as 3,5-O-benzylidene-~-ribonoI ,4-lactone (2), and its acetylated and benzoylated derivatives, led to the conclusion (4) that these compounds had a 1,3-dioxolanering (3a- c) rather than a 1,3-dioxanering. To corroboratethe assignments, the structure of 3c was determined by X-ray crystallography, which showed a central 1,5-1actonering, in a boat conformation, cis-fusedto a 1,3-dioxolane ring, having the R-configuration at the acetal carbon atom. 125

Copyright 0 1994 by Academic Press, Inc.

AU rights of reproductionin any form memed.

126

ROSA M. DE LEDERKREMER AND OSCAR VARELA

Reaction of 1 with benzaldehyde and zinc chloride gave a diastereomeric mixture (6) of R- and 92,3-0-benzylidene derivatives (4a and 4b). The former (4a) would be identical to the acetal described by Zinner et a!, (3) as “2,4-0-benzylidene-~-ribono-1,5-lactone.” This structure was further established as 4a, by chemical and physical studies of the product obtained on reaction of 1 with benzaldehyde dimethyl acetal(7). The 2,3-0-benzylidene derivative, obtained by Garegg et al. (8) on reaction of 1 with &,adichlorotoluenein pyridine, has the same propertiesas compound 4a, which would indicate the R-configuration for the acetal carbon.

4 a R1 = H, R2 = Ph b R1 = Ph. R 2 = H

3 a R=H b R=Ac c R=Bz

1 R=H

2 R=PhCH

The physical constants of the reassigned structures of benzylidene derivatives of D-ribonolactone are shown in Table I. Selective acetalation of gulono, galactono, idono, or talono- 1,Qlactones gave intermediateshaving a free hydroxylgroup at either the 2- or 3-position, which were used in synthesesof ascorbic acid (9). Reaction of L-gulono-1,4lactone with benzaldehyde, catalyzed by concentrated hydrochloric acid (10) or hydrogen chloride (1 1), gave 3,5-0-benzylidene-~-gulono-1,4-lactone (5). Analogous derivatives of 5 with aromatic or aliphatic aldehydes have been reported (9). Treatment of L-gulono- 1,Clactone with benzaldehyde diethyl acetal and concentrated hydrochloric acid gave ethyl 3,5 :4,6di-0-benzylidene-L-gulonate(6)in 84%yield (1 I). Ph

CHzOH

I

HCOCHPh

Ph 5

6

SYNTHETIC REACTIONS OF ALDONOLACTONES

127

TABLEI Physical Constants for the Reassigned Structures of 0-Benzylidene Derivatives of D-Ribonolactone (1)

Compound

Melting point ("C)

[aIn(solvent)

Reference

3a

233-235.5 230-231.5 163-164 164- 166 87 - 88

- 174.1" (HCONMe,) - 177" (HCONMe,)

3 6 7 8 7

4a 4b

-70"

-85" -40"

(HCONH,) (CHCl,) (CHCl, )

2. Reaction with Ketones The hydroxyl groups of aldonolactonesreact with ketones to afford cyclic acetal derivatives. Thus, treatment of D-gulono- 1,4-1actone with acetone and such acid catalysts (12,13) as hydrochloric acid, sulfuric acid or zinc chloride gave 2,3 :5,6-di-O-isopropylidene-~-gulono1,Clactone (7a). Selective hydrolysis of the di-U-isopropylidenederivative 7a gave the 2,3-0isopropylidene derivative (7b) in satisfactory yields ( 12,13). Using copper sulfate as catalyst (12), the diacetal 7a (25% yield) was obtained, together with 5,6-U-isopropylidene-~-gulono1,Clactone (8a, 10%yield). Selective isopropylidenation of aldonolactones has been achieved. For example, the 5,6-U-isopropylidene derivative 8a was the only product formed (95%yield) on treatment of D-gulono-1,Clactone with 2-methoxypropene ( 14). Similarly, D-galactono-1,Clactone gave, under the same conditions, the expected monoisopropylidenederivative (8b) in 95% yield. Reaction of L-gulono-1,Clactone with 2,2-dimethoxypropane- acetone, in the presence of tin(11) chloride, produced the 5,6-O-isopropylidene derivative 9a in 60% yield. Similarly, ~-mannono-l,4-lactonegave 10a in 77% yield ( 15). The di-U-isopropylidene derivatives of ~-gulono-( 16) (9b, wrongly depicted in Ref. 16) and D-mannono-1,Clactones (17,18) (lob) have been reported. Acetonation of D-g2ycero-D-gulo-heptOnO-1,Clactone with acetone-sulfuric acid yields 80% of the 3,5 :6,7-di-O-isopropylidene derivative (11) with a small proportion of the 2,3 :5,6-diacetal(l9). Likewise 2,3 :5,6-di-U-isopropylidene-~-g~ycero-~-gu~o-heptono1,Clactone (12) has been prepared (20). Reaction of D-glucono-1,4-lactone with 2,2-dimethoxypropane- tin(11) chloride yields the 5,6-U-isopropylidene derivative 13, which on periodate oxidation afforded 2,3-U-isopropylidene-~-glyceraldehyde (2 1). However, the acid-catalyzed isopropylidenationof D-glucono- 1,5-lactonewith 2,2-dimethoxypropane afforded methyl 3,4 :5,6-di-U-isopropylidene-~-gluconate (14) as the main product (22). Reduction of the ester function gave

-

ROSA M. DE LEDERKREMER AND OSCAR VARELA

128

HCOR

I

CH,OR

7 a R=CMe,

9 a R=H b R=CMe,

R’ =OH, R2 =H b R1 = H. R2 = OH

8 a

b R=H

f?

HCO, HCO-CMe, I

HiO/CMe2 I

~H,OH 12

11

10a R=H b R=CMe,

3,4 :5,6-di-O-isopropylidene-~-glucitol, which was further oxidized to 2,3 :4,5-di-O-isopropylidene-aldehydo-~-arabinose. C0,Me

I I

HCOH OCH, Me2.<

I

OCH W/CMe2

OCH

HCO Q0

OH

13

14

D-Ghcono- I ,5-lactonehas been transformed (23) into trihydroxycarboxylates bearing a long-chain alkyl acetal group (15a,b). The synthesis involves the acetalation of HO-4 and HO-6 with a long-chain alkyl carbonyl com-

SYNTHETIC REACTIONS OF ALDONOLACTONES

129

pound, followed by alkaline hydrolysis of the lactone group. These compounds may be utilized as a new type of cleavable surfactants. C0,Na

I I HOCH I

HCOH

HCO

I \ I /

HCOH CR1R2 H,CO

15 a R1 = Me, R2 = (CHJn CH, [n = 7, 8. l o ] b R1 = H, R2 = (CH&

CH,

D-Ribono-1,Clactone (1) readily condenses with acetone, under acidic catalysis with mineral acids or anhydrous copper sulfate, to give 2,3-0isopropylidene-D-ribono-1,Clactone (16a), which was employed for the synthesis of 5-deoxy and 5-0-substituted derivatives of D-ribono-1+lactone and D-nbitol(24). Acid removal of the 1,3-dioxolane protecting group gave products having probable inhibitory activity of arabinose 5-phosphate isomerase (25). Other applications of 16a for the synthesis of natural products will be discussed later. Treatment of 1 with p-methoxyacetophenone dimethyl acetal or a,p-dimethoxystyrene, in the presence of pyridinium p-toluenesulfonate, gave the 2,3-acetal derivative. Although a new chiral center is generated in the product, a preference for a major isomer was observed (26). Reaction of D-ribono-I ,blactone with cyclohexanone catalyzed by Amberlite IR- 120 (H+) in refluxing benzene led to two products identified by their spectral data and chemical transformationsas 2,3-0-cyclohexylideneD-ribono-1,4-lactone (16b) and 3,4-O-cyclohexylidene-~-ribono1$lactone (17), obtained from the reaction mixture in 60 and 20% yields, respectively (27). 0

HOCH,

Q RO

16 a

OR

R=>CMe2

b R = >C6H,,

17

ROSA M. DE LEDERKREMER AND OSCAR VARELA

130

The chirality of compound 16b was completely inverted, to give 2,3-0cyclohexylidene-L-ribonolactone (19), by means of a procedure (28) that involves oxidation of the hydroxymethyl group of 16b by ruthenium tetraoxide, followed by reduction of the lactone group. The resulting intermediate 1$lactone (18) underwent isomerization (with cyclohexylidene migration) upon refluxing in xylene, in the presence of a catalytic amount of D-camphorsulfonic acid [in Ref. (28)the formulas for the serieswere erroneously depicted].

16 b

OQ -

HO

O/C6H10

ia

A

19

3. Use of Aldonolactones Acetals for the Synthesis of Sugar Derivatives Acetal derivativesof aldonolactonesare useful starting compoundsfor the synthesisof lower aldoses and different sugar-derived molecules. Many syntheses of D- and L-erythrose from acetone derivatives of ribono- and gulonolactones have been reported. For example (27), borohydride reduction of 16b followedby periodate degradation of the glycol and subsequent benzoylation gave l-0-benzoyl-2,3-di-O-cyclohexylidene-~-~-erythrofuranose. The same sequence of reactions applied to compound 17 gave the corresponding derivative of L-erythrose. Lerner (29)reported a simple synthesisof L-erythrosethat involves 2,3-di0-isopropylidene-D-gulono-1,Clactone (7b) as a key intermediate. Reduction of the lactone group of 7b with sodium borohydride, followed by periodate oxidation of the L-glucitol derivative, afforded 2,3-0-isopropylidene-L-erythrose.The free sugar may be readily obtained by acidic hydrolysis of the latter. The borohydride reduction- periodate cleavage applied to 2,3-0-isopropylidene-D-ribono-1,Clactone (16a) led to L-erythrose (30). The method was also employed (31) for the synthesis of D-erythrose, starting from an 0-benzylidene-D-ribonolactone. However, in this case, the structural assignments for the intermediate compounds must be revised, as the starting material formulated as “3,5-0-benzylidene-~-ribono1,Clactone (2)” was, as discussed previously in this section, the 3,4-0-benzylidene-~-ribono13lactone (3a). Therefore, the correct structure for the product described as 3,5-0-benzylidene-~-nbitol(20, not isolated) would be 3,4-0-benzylidene-

SYNTHETIC REACTIONS OF ALDONOLACTONES

131

D-ribitol (21), and the structure formulated as 2,4-O-benzylidene-~-erythrose (22), would be 2,3-O-benzylidene-~-erythrose (23). These reassignments are supported by comparison of the properties of the product described as 22 with data from the literature. Thus, an authentic sample of 22, obtained by a different route (32), had an optical rotation value of -20°, which greatly differs from that found for the product formulated (31)as 22 (-65.2" + - 62.6'). The fact that mutarotation is observed, as well as the correspondence with the [a],value (-62") for 23, would indicate that the latter is the correct structure for the product described as 22. In any event, hydrolysis of the acetal function of both (22 and 23), leads to D-erythrose. CH,OH

CHzOH

I I

I I

HCOH

HCOH

HCO

HCO,

I \ CHPh HCOH

HAO/CHPh

I

I /

H,CO

CH,OH

21

20

HCO

I I \ HCOH I / HCO

CHPh

H,CO

22

23

Sodium 2,3-O-isopropylidene-~-nbonate (24), obtained from 16a, was selectively cleaved between C-4 and C-5 by sodium periodate and the resulting aldehyde group was reduced by sodium borohydride to give D-erythrono- 1,Clactone (25a) in 76% yield, with loss of the 0-isopropylidene group occurring during lactonization. New derivatives of 25a, including the 2,3-O-isopropylidenederivative (25b), were reported (33). C0,Na

24

25 a R = H b R=CMe,

Periodate oxidation of 5,6-O-isopropylidene-~-gulono1,Clactone (9a) gave 2,3-O-isopropylidene-~-glyceraldehyde in 69% yield. This compound was used to prepare 2,3-O-isopropylidene-~-glycerol and it was also condensed with amines and Wittig reagents (34).

132

ROSA M. DE LEDERKREMER AND OSCAR VARELA

The synthesis of L-ribofuranose derivatives from 16a has been carried out by Walker and Hogenkamp (35). The procedure involves oxidation of the hydroxymethyl group of 16a with dimethyl sulfoxide-N,N’dicyclohexylcarbodiimideand acid hydrolysis of the protectinggroup to give L-riburonic acid, which was converted into methyl (methyl a,P-ribofuranosid)uronates (26). Reduction of 26 with sodium bis(2-methoxyethoxy)aluminum hydride gave the chromatographicallyseparable anomers of methyl L-ribofuranoside (27).

C0,Me

CH,OH

26

27

Reactions of acetal derivatives of aldonolactones involving the lactone carbonyl group or used as chiral precursors in the synthesis of noncarbohydrate natural products are discussed in later sections. OF ALDONOLACTONES 111. ACYLATIONAND ETHERIFICATION

As acylation of aldonolactones is related to /%eliminationreactions, the fully acylated aldonolactone derivatives are mentioned in Section IX (see also Sections VI and VIII). Only a few studies on selective acylation of aldonolactoneshave been described. The partially acylated derivatives have been employed as glycosylating agents or for the preparation of monomethylated sugars. Partial benzoylation (36) of D-mannono or L-rhamnono-1,4-1actone under controlled conditions gave the 3-unsubstituted derivatives 28a and 29a. Attempted methylation (37) of HO-3 with diazomethane- boron trifluorideetherate gave a very low yield of the 3-O-methyl derivatives (28b and 29b). Methylation by Purdie’s method caused benzoyl migration, and the 2-O-methyl fully benzoylated derivatives (28c and 29c) were obtained in good yield. Reduction of the lactone carbonyl group, followed by debenzoylation, afforded 2-O-methyl-~-mannoseand 2-O-rnethyl-~-rhamnose.Benzoylation of D-galactono- 1,Clactone with 2.2 mol of benzoyl chloride at -23 ’ gave crystalline, 2,6-di-O-benzoyl-~-galactono-1,Clactone (30a) in 62% yield (38). A partially acetylated galactonolactone,namely, 2,3,6-tri-O-acetyl-~-galactono-1,Clactone (30b), was obtained by an indirect method. Tritylation and further acetylation of D-galactono-1,4-1actone gave 2,3,5-tri-O-acetyl-

SYNTHETIC REACTIONS OF ALDONOLACTONES

133

6-0-trityl-~-gdactono-1,Clactone (30c). Detritylation of 30c with F,B OEt, caused 0-5 + 0 - 6 acetyl migration, affording 30b, which was used for the preparation of 5-O-methyl-~-galactofuranose(39).

-

I

I

CH20R4

CH3 28 a R1 = Bz, R2=

H

b R1 = Bz, R2 = Me c R1 = Me, R2 = Bz

29 a R’ = Bz, R2 = H b R1 = Bz, R2 = Me c R1 = Me, R2 = Bz

30 a Rf = R4 = Bz, R2 = R3 = H b R’ = R2 = R4 = Ac, R3 E H c RT = R2 = R3 = Ac, R4 I Tr

Aldonolactones having their hydroxyl groups protected as ethers (mainly benzylated and silylated derivatives) have been employed for chain elongation through the carbonyl group, and therefore these type of derivatives are included in Section V. Selective silylation of aldohexono-1,Clactones with tert-butylchlorodimethylsilane gave the 2,6-di-O-silylated derivatives as major products (40). However, D-glucono-1,5-lactoneunderwent ring-contraction during silylation to give 2,6-di-O-tert-butyldimethylsilyl-~-glucono-l,4-lactone (31a) together with its 5,6-di-O-silyl isomer (31b), in 65 and 18%yields, respectively. Acylation and substitution reactions on the selectively silylated aldonolactone derivatives have been performed. FH20SiMe2But

I

OR’ 31 a R1 = SiMe,But, R2 = H b R’ =

H. R2 = SiMe2But

The 7-0-silyl and 2,7-di-O-silyl derivatives (55 and 18% yields, respectively) were obtained on treatment Of D-glycero-D-gulo-heptOnO-1 ,blactone with 1.1 eq. of tert-butylchlorodiphenylsilane(20). Sugar lactone derivatives selectively substituted with different protecting groups, which were em-

ROSA M. DE LEDERKREMER AND OSCAR VARELA

134

ployed as starting compounds for the synthesis of carbohydrate and noncarbohydrate natural products, are mentioned in later sections.

Iv. REACTION OF ALDONOLACTONES WITH HYDROGEN BROMIDE The reaction of aldonolactoneswith hydrogen bromide in acetic acid has been extensively studied by Pedersen and coworkers. Treatment of D-galactono-1,blactone with HBr-AcOH for a few hours at room temperature followed by acetylation gave the acetylated 6-bromo-6-deoxy derivative 32 in good yield (41). On the other hand, reaction of D-mannono- 1,Clactone with the same reagents for 2 - 3 h gave a high yield of the dibromolactone 33a. D-Glucono-1,4-or 1$lactones both yielded 2,6-dibromo-2,6-dideoxy~-mannono-l,4-lactone (33b) when treated for 18 h with hydrogen bromide - acetic acid, followed by acetylation. Substitution by bromide at C-2 takes place with inversion of configuration. Thus, 2-bromo-2,6-dideoxy-L-glucono-1,Clactone (34) was obtained from L-rhamnono-1,4-lactone (42).

fq HCOAc

OAc

I

CH2Br

I

Acociq R’

p OH HOCH

I

CH,Br

CH,

32

33 a R’ = Br, R2 = H b R’ = H, R2 = Br

34

The authorspostulated that substitution of the primary hydroxyl group by .bromide takes place through a 5,6-acetoxonium ion. Consequently, the reaction of pentono- 1,Clactones with HBr - AcOH for a few hours yielded the 2-bromo-2-deoxypentono-1,4 lactones, whereas the 2,5-dibromo derivatives were formed only on prolonged treatment (43). In the case of ~ - r i bono- I ,Clactone, a mixture of 2,5-dibromolactoneswas obtained. Substitution by bromine at C-5 could take place via the 4,5-acetoxonium ion derived from the aldonic acid formed on opening of the lactone ring, thus explaining the formation of 2,5-dibromo-2,5-dideoxy-~-xylono1,Clactone together with the analogous D-arabinono- and D-ribono-dibromolactones. In the heptonolactone series, D-g/ycero-D-gu/o-heptono-1,4-lactone reacted with hydrogen bromide -acetic acid to afford (44) 2,7-dibromo-2,7dideoxy-D-glycero-D-ido-heptono-1,Clactone (35), together with two anhy-

SYNTHETIC REACTIONS OF ALDONOLACTONES

135

drolactones. The latter, assumed to be 3,6 anhydro-L-glycero-D-gulo-heptono- 1 ,4-lactone (36) and 3,7-anhydro-~-glycero-~-gdo-heptonoI ,4-lactone (37),were not formed via the bromolactones; instead they were obtained on treatment of D-gl~Cero-D-gulo-heptOnOlaCtOnewith one equivalent of formic or acetic acid in anhydrous hydrogen fluoride, with subsequent deacylation. When an aqueous solution of the dibromolactone 35 was boiled, it gave 2,5 :4,7-dianhydro-~-glycero-~-gub-heptonic acid.

po poH 0

0

It

II

HO OH HCOH

OH HOCH,

I H~OH

OH

I

CH,Bi

35

37

36

The bromodeoxyaldonolactones have been used for the preparation of aminodeoxy aldonic acids and aminodeoxy sugars via azido derivatives (45,46).Likewise, a-and P-aminopolyhydroxy acids have been prepared by treatment of the bromodeoxyaldonolactones with liquid ammonia (47). Thus, 3-amino-3-deoxy-~-threonicacid and 3-amino-3-deoxy-~-arabinonic acid (40b)were obtained from 2-bromo-2-deoxy-~-threono-or D-XYlono- 1,4-lactone (38). It was shown that 2,3-epoxy carboxamides (namely, 39) are intermediates of the reaction. Heating at 90" for long periods led to the 3-amino-3-deoxyaldonamides, which upon acid hydrolysis yielded the corresponding aldonic acids. COR

CONH, HOCH,

I

0

Q-Br

38

I I HCNH, I HCOH I

HOCH

------

O/T C 'H

I HCOH I

CH,OH

39

CH,OH

40 a

R = NH2

b R=OH

In the case of the reaction of 2,5-dibromo-2,5-dideoxy-~-xylono(41a) and D-lyxono-1,4-lactones (41b)with liquid ammonia, the main product

ROSA M. DE LEDERKREMER AND OSCAR VARELA

136

was the 2,5-diamino-2,5-dideoxy-~-xylono1,5-lactarn 43 formed via a cyclic 2,3-epoxy-~-lyxonolactam (42),which opens exclusively at C-2. Similarly, 2-bromo-2-deoxylactonesreact with fluoride ions to afford the 2fluoro derivatives, through a 2,3-epoxylactone intermediate (48).

I

R‘ 41 a

NH2

R ’ = Br R2 = H

42

43

b R =H R’=B~

V, CHAINELONGATION THROUGH THE ALDONOLACTONE CARBONYL GROUP

The addition of organometallic reagents to the carbonyl group of conveniently substituted aldonolactones constitutes a viable chain-extension method. The reaction leads to the formation of hemiacetals of glyculoses, 1-methylene sugars, and C-glycosyl compounds, which are precursors of, or occur as subunits of, a variety of natural products. 1. Reformatsky-Type Reactions

Carbohydrate lactones have been used as the carbonyl reagent in the Reformatsky reaction. Thus, 2,3 :5,6-di-O-cyclohexylidene-~-mannono1,Clactone [44,obtained by oxidation of the mannofuranose derivative (49)] reacted with ethyl bromoacetate and zinc to give the protected 2deoxy-3-octulosonic acid ethyl ester (454in 69% yield (50). “Ketonic hydrolysis” with potassium hydroxide in aqueous methanol, followed by acidification and heating, afforded the 1-deoxyheptulose derivative 45b. Similarly, starting from compound 44,the 1-C-substituted allyl and propargyl lactols were prepared on reaction with allyl or propargyl bromides in the presence of zinc (5 1).

44

45 a R = C0,Et b R=H

Condensation of 2,3 :5,6di-O-isopropylidene-~-gulono1,Clactone (7a) with ethyl bromoacetate in the presence of zinc also gave the expected

SYNTHETIC REACTIONS OF ALDONOLACTONES

137

Reformatsky product (46). Treatment of 46 with methanol and an acid ion-exchange resin afforded the methyl glycoside 47a, the product of partial hydrolysis (47b), and the 3,8-anhydro sugar 48. Synthetic transformations were performed on these products (52).

7a-v +

;eC02Me

M C 20eB ;

~

CH,CO,Me

HCO I 'CMe,

-

OH

\o / o

HCOR I H,COR

H,CO'

HO

47 a R = CMe,

46

HO

48

b R=H

Improvement to the Reformatsky reactions was achieved (53) by the use of a highly activated zinc- silver couple dispersed on the surface of graphite. Treatment ofprotected aldono- 1,Clactones10b or 25b and 1,5-lactones51a or 51b with a Reformatsky reagent prepared from a-haloesters or alkyl 2-(bromomethy1)acrylates resulted in the formation of the corresponding 3or 4-glyculofuranos (or pyranos)onates 49a,b-50a,b, or 52a,b, respectively, under mild conditions (- 40" to 0") and in very good yields. Ethyl 2-deoxy2-fluoro (and 2-bromo)-a-~-manno-3,6-furanos-3-octulosonate derivatives were also obtained.

50 a

49 a R = CH2C02Et b R = CH,-C=CH,

b R = CH2-C=CH2

I

I

C02But

BnOFH,

51 a R1 = H, R2 = 0Bn b R' = OBn. R2 = H

R = CH2C02Et

C02Et BnOCH, I

52 a R' = H, R2 = OBn b R1 = OBn. R2 = H

138

ROSA M. DE LEDERKREMER AND OSCAR VARELA

Tetra-n-butylammonium ester enolates, assumed to be formed in the reaction of alkyl trimethylsilyl esters (or nitriles) with fluoride anion, show the same reactivity as the classical Reformatsky reagents. Protected aldonolactones react with ethyl trimethylsilylacetate, 2-trimethylsilyloxypropanoate, trimethylsilylacetate, methyl 2-trimethylsilylpropanoate,trimethylsilylacetonitrile, or alkyl2-(trimethylsilylmethyl)acrylates,in the presence of catalytic amounts of tetra-n-butylammonium fluoride to afford the chainelongated alkyl glyc-3-ulosonates or -3-ulononitriles,in high yields (54). For example, reaction of 10b with Me,SiCH,CO,Et gave 49a, in 92% yield. When the reaction was conducted with anhydrous solution of Bu4NF, an a$-mixture of the 3-0-trimethylsilyl derivatives of the glyculosonate was obtained. 2. Reaction with Organomagnesium and Organolithium Reagents

Reactions of sugar lactones with organomagnesium and organolithium reagents have been described. For example, the Grignard reaction (55) of 2,3 :5,6-di-O-isopropylidene-~-mannono1,Clactone (lob) with benzyloxymethylmagnesium chloride afforded 1-0-benzyl-3,4:6,7-di-O-isopropylidene-D-manno-heptofuranose,which on hydrogenolysis followed by acidcatalyzed hydrolysis gave D-manno-heptulose. Addition of lithiated heterocycles to aldonolactonesyields carbon-linked nucleosides (56). Thus, the reaction of 2,3 :5,6-di-O-isopropylidene-~-gulono- 1,blactone (9b) or 2,3-0-isopropylidene-~-ribono1 ,blactone (16a) with various lithiated heterocycles gave gulofuranosyl derivatives53a -g or ribofuranosyl derivatives 54b,c. Gulonolactols 53a - g and ribonolactols 54b,c were acetylated with acetic anhydride in pyridine to yield their acetyl derivatives. The stereochemistry of compounds 53a-g and 54b,c was discussed in terms of the Cotton effect of circular-dichroism curves of the ring-opened alcohols formed upon reduction by sodium borohydride. The configuration at C-1 of 53g was proved by means of X-ray analysis (57,58).

53 a - g

54 b - c a R = pyrid-2-yl b R = benzothiazol-2-yl c R = 1-benzylbenzimidazol-2-yl d R = benzirnidazol-2-yl e R = 1-benzylimidazol-2-yI f R = 1-benzylimidazol-5-yI g R = 1, 3-dithian-2-yl

SYNTHETIC REACTIONS OF ALDONOLACTONES

139

Carbon-linked sugar- heterocycles were also obtained by reaction of the lithiated derivatives obtained from 2-bromopyridine, a-picoline, and benzothiazole with 4-O-benzyl-2,3 :6,7-di-O-cyclohexylidene-~-g~ycero-~gufo-heptono-1,5-1actone(55). The corresponding D-gfpru-D-gufo-heptOpyranose-substituted compounds 56a-c were isolated in 35 -43.5% yields. With other heterocycles (for example furan), 1-disubstituted guloheptitols were obtained (59).

55

56 a R = 2-pyridyl b R = 2-pyridylmethyl c R = 2-benzothiazolyl

The hemiacetals obtained by nucleophilic addition of organometallic reagents to the carbonyl group of aldonolactones may be reduced to the corresponding C-glycosyl compounds. For example, treatment of 2,3,4,6-tetra0-benzyl-D-glucono-1,5-1actone (51a) with allylmagnesium bromide or with lithium ethyl acetate gave the corresponding hemiacetals, which were reduced with triethylsilane and boron trifluoride etherate to furnish stereoselectively the /K'-glycosyl compounds 57a and 57b, respectively (60).Very good yields in the reaction were also obtained with benzyl derivatives of aldonolactones having the gulucto and malzno (51b) configurations. The compatibility of the alcohol protecting-groups to the addition of organometallic reagents and to reduction with Et,SiH BF, has been studied (61). It was found that neither the methoxymethyl (MOM) groups, nor the 0-isopropylidene group was stable to the reduction conditions. However, the benzylated derivative 51a afforded stereoselectivelythe B-C-glycosyl compounds57c - h in good overall yields. The same sequenceof reactions was performed on 0-benzyl-protected arabinono-1,Clactone, providing a route to C-nucleosides of pentofuranosyl sugars (62). Ethynyl compounds react with sugar lactones to give acetylenic lactols (16,63). Reaction of 2,3-O-isopropylidene-~-~bonolactone (Ma)with lithium acetylenic derivatives gave 1-(2-substitutedethynyl)-2,3-0-isopropylidene-D-ribofuranoses.Similarly, treatment of 2,3 :5,6-di-O-isopropylidene~-gulono-1,4-lactone (9b) with various lithium acetylenic reagents gave 1-alkynyl-2,3 :5,6-di-O-isopropylidene-~-gulofuranoses. Reduction of ethynyl lactols obtained from 51a with Et,SiH-Lewis acid

-

140

ROSA M. DE LEDERKREMER AND OSCAR VARELA BnOCH,

1) RMgX or RLi 51 a

_____c

2) Et,SiH-F,B

57 a - h

a R = CH,CH=CH, b R=CH,CO,Et

c R=Ph d R = 2-fury1 e R = 2-pyridyl f R = ethenyl g R = p-MeOC,H, h R=mMeOC,H,

gave 1-(P-D-glucopyranosy1)alkynes (64). These compounds are immediate precursors of 1-(/?-D-glucopyranosyl)alkanes, of potential utility as new forms of nonionic detergents. Furthermore, the ethynyl derivatives could be selectively converted into the (E)and (Z)-1-(&D-glucosyl)ethenes. The sequence ethylynation-reduction has been employed for the synthesis of a model fragment (58) of ambruticin.

58

Applying the procedure just described, Rouzaud and Sinay (65)described a stereospecific synthesis of a “C-disaccharide” in which a methylene group takes the place of the interunit oxygen atom. Starting from 51a and the dibromoalkene 59, the hemiacetal 60 was obtained. Reduction of 60, followed by catalytic hydrogenolysis, gave the /?-(1 6)-“C-disaccharide7’61. Likewise, the lactol62, obtained by addition ofmethyllithiumto 2,3,5-triO-benzyl-D-arabinono-1,Clactone, underwent a self-condensation reaction in the presence of BF3 OEt, -MeCN, to afford an anomeric mixture of two C-disaccharides63 and 64 in 93% yield (66). The stereocontrolled synthesis of a chiral, polyhydroxy 177-dioxaspiro[5.5lundecene from 2,3,4,6-tetra-O-benzyl-~-glucono1,5-lactone

-

-

SYNTHETIC REACTIONS OF ALDONOLACTONES

141 OBn

I

HC=CBr,

51 a

CH,OBn

Q

+

BnO

$

OMe

Qf

BnO

OBn

OBn

59

60

J

1) E1,SiH

2) H,/ Pd

CH,OH

CH,-CH,

I

I

I

,

I

OH

OH 61

rn Q BnO

BnOCH,

y x

BnOCH,

0

BnOCH,

__

BnO

y

CH2

2

I

BnO

I

I

BnO 62

BnO 63

64

(51a) has been described (67). This bicyclic system having a different pattern of substitution, can be found in a number of biologically important natural products, such as the avermectins and the milbemycins. Reaction of 51a with lithium 1-trimethylsilyloxy-3-butyne, followed by treatment under mildly acidic conditions, afforded a mixture of anomers of the acetylenic lactol derivative65, which was readily converted into the spiroacetals66 and 67.

ROSA M. DE LEDERKREMER AND OSCAR VARELA

142

1) LiC=C(CH2),0SiMe, 51 a 2) H+ OBn 65

CH,OBn

CH20Bn

I

I

67

66

The per(trimethylsily1) ether of D-glucono- 1,5-lactone (68) reacted (68) with 2-lithio-l,3-dithiane to give, after removal of the protecting groups, 1-C-( 1,3-dithian-2-yl)-a-~-glucopyranose (69a) as a single isomer. Depending on the conditions of acetylation of 69a, the tetraacetyl69bor pentaacetyl 69c derivatives were obtained. Desulfurization of 69b with Raney nickel gave the tetraacetate of 1-deoxy-D-gluco-heptulose (70a), whereas similar desulhration of 69c was accompanied by removal of the tertiary acetoxyl group, providing stereospecific access to the C-p-D-glucosylcompound 70b.

LI

2) H,O, MeOH

R20

OSiMe, 68

OR’ OR2

R’ = R2 = H b Rf = H, R2= Ac c R’ = R2 = AC

69 a

AcO OAc

70 a R’ = O H b R’=H

SYNTHETIC REACTIONS OF ALDONOLACTONES

143

Similar addition of other lithiated thio derivatives to substituted aldonolactones has been performed (69). Thus, reaction of tris(methy1thio)methyllithium to benzylidenated or isopropylidenatedaldonolactones, followed by mercury(11)-catalyzed desulfuration, led to derivatives of glyc-2-ulosonic acids.

3. Methylenationof Aldonolactones Carbene-mediated methylenation of aldonolactones provides a direct route to 1-methylene sugars, which may be used as intermediates for the preparation of furanoid or pyranoid C-glycosyl compounds, or chiral precursors for the synthesis of natural products. The Tebbe reagent [p-chloro-bis(cyclopentadienyl)(dimethylaluminium)-p-methylenetitanium] in its pure, crystalline (70,7 1) or crude (72) forms has been employed for the methylenation of aldonolactones. Thus, D-ribono-1,44actone derivatives 71a and 71b reacted with Tebbe’s reagent to give (70,7 1) the exemethylene compounds 72a and 72b. ButR$30CH2

R’O

Q=

OR’

R’O

71 a R’ = CMe,, R2 = Me

b R’ = SiMe,,

Cp = cyclopentadienyl

72 a

R2 = Ph

OR’

R1 = CMe,,

b R‘ = SiMe,.

R2 = Me

R2 = Ph

Similarly, compound 73 was prepared from a persilylated D-galactono1,4-1actone precursor in 60 - 80% yield (7 1). However, when unpurified Tebbe’s reagent was employed, 1,4-lactones (for example 71a) gave a mixture of 72a and the product (74) of hydration of the double bond. The

iG)%

we

ButMe,SiOCH,

0

OH

HCOSiMe, OSiMe,

I

CH,OSiMe,

73

o\ /o

@H2

RO OR

CMe,

74

75a R=Bn b R - SiMe,

c R=Me

144

ROSA M. DE LEDERKREMER AND OSCAR VARELA

proportion of the lactol 74 greatly increases during the chromatographic purification, even when slightly basic eluants were used, and the alkene 72a was isolated in only 6% yield (72). Reaction of Tebbe's reagent with fully benzylated (7 I ,72), silylated (7I), and methylated (72) D-glucono-1$lactone gave the corresponding 2,6-anhydroheptenitol derivatives 75a - c in high yields. Another organotitanium reagent employed for the direct methylenation of aldonolactones was dicyclopentadienyldimethyltitanium(73). Reaction of the diacetone derivative of D-mannono-1,blactone (lob) with three equivalentsof the reagent afforded 85% of the D-manno-hept- 1-enitoldiacetal. Analogous products were also obtained from 0-isopropylidene or 0-benzylidene derivatives of aldonolactones (such as, D-erythrono-1,4lactone, D-arabinono-1,Clactone, D-glucono- 1,5-1actone,and D-mannono1,Slactone)in yields higher than 85% (except for the erythronolactone-derived 1-enitol, obtained in 64% yield). The lactols were by-products of the reaction. The application of lithiumtrimethylsilylacetate for the C- 1 elongation of aldonolactones has been examined (73). Although the reagent had been successfully used for the alkenation of lactone carbonyl groups (74), in the case of aldonolactones10b or 25b only insignificantyields ofthe alkenes, but 30 - 40%of the lactols 49a or 50a, were obtained (73). However, these lactols, alternatively prepared in good yields by a Reformatsky-type reaction (53,54),were readily eliminated to the desired alkenes by simple treatment with methanesulfonyl chloride-triethylamine at 0". Thus, from 49a or 50a separable E,Z mixtures (76a and 76b, or 77a and 77b, respectively) were obtained in good yields (73).

I

o\ 76 a R1 = H, R2 = C02E1 b R1 = CO,Et, R2 = H

I

/o

CMe,

77 a R1 = H, R2 = C0,Et b R1 = CO,Et, R2 = H

I-Methylene sugars are versatile starting compounds for the synthesis of aromatic and hydroxymethyl C-glycosyl compounds (7 1) and of doubleended C-glycosyl compounds (70). The double bond undergoes 1,3-dipolar cycloadditions to give isoxazoline derivatives (71), such as that used as a model for the synthesis of tunicamycin. Also, reaction with iodonium sym-

SYNTHETIC REACTIONS OF ALDONOLACTONES

145

dicollidine perchlorate gave an easy access to 1-iodoheptuloses or 1-iodomethylene derivatives (75). Dichloromethylenation and difluoromethylenation of aldonolactones have been achieved by concise sequences. Thus, aldonolactone derivatives react with tris(dimethy1amino)phosphine (hexamethylphosphorus triamide) and tetrachloromethane [(Me,N),P- CCl,] to give l-dichloromethylene sugars, which may be reduced to C-glycosyl compounds and chiral tetrahydrofurans (76). For example, dichloromethylenation of compound 71a gave the dichloroalkene78, which was cleanly reduced by Raney nickel to the glycosylmethane79. The same sequence, when applied to the 2,3 :5,6di-0-isopropylidene D-allono-1,Clactone (80) led to the C-methyl compound 81 in 83% yield.

Y)

ButMezSiOCHz 0 (Me,N),P-CCI, 71 a

+

CCI2

Raney Ni

I I o\ / o CMe,

78

80

79

81

The presence of a dichloromethylenegroup at the anomeric center of 82 facilitates proton abstraction at C-3 by a strong base (77), affording the 4-deoxyglycos-3-ulosederivative 83. Reduction of the dichloromethylene group by Raney nickel gave a 1-C-methylderivative with high stereospecificity, which opens the way to a series of 2,5-anhydro-1-deoxyalditols.Compound 83 was the key intermediate for the synthesis (78) of tosyl L-(+)-epimuscanne (84a) and tosyl L-(+)muscarine (84b). Difluoromethylenation (79,80) of aldonolactones may be readily accomplished by treatment of the lactone derivative with tris(dimethy1amino)phosphine, dibromodifluoromethane,and zinc in refluxing tetrahy-

146

ROSA M. DE LEDERKREMER AND OSCAR VARELA

83

82

84 a R1 = OH, R2 = H b R’ = H, R2 = OH

drofuran. The reaction was applied to 1,Clactones having the rib0 (71a), manno (lob), d o (80), and D-glycero-D-gulo configurations, and also to per(trimethylsily1)ated ~-glucono-I, 5-lactone (68), affording moderate yields (56 - 67%)of 1-difluoromethylenederivatives (for example,85). Catalytic hydrogenation of these compounds gave the corresponding difluoromethyl C-glycosyl derivatives (such as, 86) with excellent diastereofacial selectivity.

pF WF2 -

ButMe,SiOCH2

71 a

ButMe2SiOCH2

H2/ Pd

(Me,N),P A

CF2Br, I Zn

o\

/o

CMe,

85

o\

/o

CMe,

86

4. Formylaminomethylenationof Aldonolactones Sugars linked to amino acids via a carbon-carbon linkage are of interest as this type of structures occur in polyoxins. The formation of such a carbon - carbon bond was achieved by formylaminomethylenation of 2,3 :5,6-di-U-isopropylidene-~-mannono1,Clactone (lob) with ethyl isocyanoacetate in the presence of potassium hydride (8 1). The E-isomer 87a was the major product of the condensation ( 5 1% yield); the 2-isomer 87b was isolated in 4% yield. Catalytic hydrogenation (Raney nickel) of 87a and 87b gave the N-formyl-mannofuranosylglycinates88a and 88b, which on hydrolysis of the N-formyl group and deprotection of the sugar moiety led to the mannofuranosyl amino acids 89a and 89b. Selective hydrolysis of the 7,8-U-isopropylidene group of 88a, followed by periodate cleavage of the diol and reduction, gave the D-lyxofuranosylamino acid 89c. Similar treatment on 88b gave 89d.

147

SYNTHETIC REACTIONS OF ALDONOLACTONES

R1 = CO,Et, R2 = NHCHO b R1 = NHCHO. R2 = C0,Et

87 a

R3

89 a

Me

R’ = NH,, R2 = H, R3 = CH20H

b R1 = I-, R2 = NH,. R3 = CH,OH c R? = N H R 2 = R 3 = H

88

a R1 = NHCHO, R 2 = H

b R’ = H. R2 = NHCHO

d R1 = H, R2 = NH,. R3 = H

Under basic conditions the D-erythro-L-gluco-oconate 88a underwent racemization at C-2 and “anomerization” at C-3 to give a mixture of four isomeric products. The configurationsat C-2 and C-3 of these isomers were established by chemical and physical methods (82). Furthermore, alcohols and thiols, in the presence of a base, attack the double bond of the oct-2-enonates (87b) affording isomeric oct-3-ulofuranosidonates. Formylaminomethylenation of 2 :3,5 :6-di-O-isopropylidene-~-allono1,Clactone (80) gave the (E)alkene 90 in 59% yield. However, when the reaction was applied to the benzyl derivativeof D-glucono-1,5-lactone(51a),

CO,Et C0,Et

NHCHO

o\ CMe, /o

HLOB~

I I HCOBn I HCOH I

BnOCH

CH20Bn

90

91

ROSA M. DE LEDERKREMER AND OSCAR VARELA

148

the oxazole 91 (50% yield) was obtained (83). Use of potassium hydride instead of 1,5-diazabicyclo[4.3.O]non-5-ene(DBN) in the reaction of 10b or 80 with ethyl isocyanoacetategave the corresponding sugar oxazole derivatives. VI. REACTION OF ALDONOLACTONES WITH ALCOHOLS

1. Formation of Esters Aldono- 1,5-1actonesand free aldonic acids react with alcoholsin the presence of hydrogen chloride to give the corresponding alkyl aldonates (84). The reaction is slower with 1,4-1actones.Because esterification takes place very slowly in the absence of an acidic catalyst, aldonic acids and their lactones may be recrystallized from boiling alcohols without appreciable esterification (85). However, in some instances, alkyl esters are formed under these conditions. For example, essentially pure ethyl L-mannonate was isolated (6.490 yield) from the mother liquors of crystallization L-mannono- 1,Clactone, obtained by Kiliani synthesis from L-arabinose(86). Similarly, repeated recrystallization from ethanol of crude 2,3,4,6-tetra-Oacetyl-D-glucono-1,5-lactone afforded the corresponding ethyl gluconate derivative (87). The tetra-0-benzoyl derivatives of ~-glucono-1,5- and 1,Clactones underwent ring opening upon heating with methanol or ethanol at the reflux temperature, or under acid catalysis, to give methyl or ethyl tetra-0benzoyl-D-ghconates having an unesterified hydroxyl group (88). Ethanolysis of 2,4-di-0-benzoyl-3,6-dideoxy-~-arabino-hexono1,5-1actone (92) afforded ethyl 2,4-di-O-benzoyl-3,6-dideoxy-~-arabino-hexonate (93a). On being kept, compound 93a was slowly converted into 2,5-di-0benzoyl-3,6-dideoxy-~-arabino-hexono1,4-lactone (94). This transformation involves 0-4 + 0-5 benzoyl migration, with formation of ethyl 2,5-diO-benzoyl-3,6-dideoxy-~-arabino-hexonate (93b) as an intermediate (89). CO, Et

I I

HCOBz y

OBz

BzoQ 92

-

2

R’ OCH,

-

RdH2

I

CH3

93 a Rf = Bz, R2 = H b R1 = H. R2 = Bz

BzOCH

OBz

I

CH3 94

SYNTHETIC REACTIONS OF ALDONOLACTONES

149

The acid-catalyzed acetalation of aldonolactones with alkyl acetals of aldehydes or ketones takes place, in some instances, with esterificationof the lactone group to give acetal derivatives of alkyl aldonates ( 1 1,22). 2. Formation of Orthoester Derivatives

Addition of alcohols to lactones results in the formation of orthoacid or orthoester derivatives. Thus, reaction of lactone 95a with potassium cyanide in ethanol led to displacement of the tosyl group by cyanide and addition of ethanol to the lactone carbonyl group, to give the orthoacid derivative 95b, which was isolated as its acetate 95c. Mild deacylation of95c led back to 95b, but under more vigorous reaction conditions the open-chain methyl aldonate was obtained (90). CH, R4

R 3 0I L

95 a R’ = RZ = 0 , R3 = H, R4 = Ts b R’ = OEt, R2 = OH, R3 = H. R4 = CN c R’ = OEt, RZ = OAc, R3 = Ac. R4 = CN

Addition of diols to the carbonyl lactone group leads to the formation of cyclic orthoesters. Compounds containing a spiro, cyclic orthoester interlinkage at the anomericcarbon atom are ofinterest, as this type of structure is found in the oligosaccharide antibiotics orthosomycins (9 I ) (such as everninomicin, flambamycin, and avilamycin). Reaction of 2,3,4,6-tetra-0-benzyl-~-glucono-1,5-lactone (51a) with epoxides or with diols afforded the corresponding orthoester derivatives (92,93). When two secondary hydroxyl groups were involved, low yields were obtained. The reaction conditions were optimized, and it was shown that condensation of lactone derivativeswith bis-0-(trimethylsily1)-1,2-diols in the presence of trimethylsilyl trifluoromethanesulfonate (Me,SiOTf) as catalyst gave good yields of the spiro, cyclic orthoesters (94). Thus, reaction of 51a with the trimethylsilyl derivatives of ethanediol, and cis- and transI ,2-cyclohexanediol gave the corresponding orthoesters in 9 1.7, 89.7, and 63.9% overall yields (95). Furthermore, condensation of 51a with conve-

150

ROSA M. DE LEDERKREMER AND OSCAR VARELA

niently substituted derivatives of methyl a-D-glucopyranoside and methyl a-D-mannopyranosidegave the glucopyranosylidenederivatives 96 and 97. Compound 96 was isolated as a single isomer in 72% yield, and compound 97 was actually a mixture of isomers, obtained in 44.8 and 26.6% yields.

BnO Q Q O M e

BnO OBn 96

CH,OBn

I

97

Many other 2,3-O-~-glycopyranosylidene-a-~-mannopyranosides were obtained by the Me,SiOTf-promoted condensation (96) of D-glucono-, D-galactono-,and L-glycevo-D-gluco-heptono-1,5-lactones,with methyl 2,3di-O-trimethylsilyl-cu-D mannopyranosides having various substituents at C-4 and C-6. The absolute configuration of the orthoester carbon atom for one of the isomers of 2,3-~-~-glucopyranosylidene-cw-~-mannopyranosides (97),as the fully acetylated derivative, was established by X-ray analysis. The 'H and 13Cn.m.r. data for the isomers were also reported (97). The diastereoisomer having the higher values for optical rotation and chemical shift for the orthoester carbon had the (S)-absolute configuration.On the basis of this observation, the configurations for various glycopyranosylidenederivatives were tentatively assigned. Hydrogenation of the benzyl derivatives afforded the free glycopyranosylydene,which showed antihelminticactivity against Ascaridia galli (98).

SYNTHETIC REACTIONS OF ALDONOLACTONES

151

VII. REACTION OF ALDONOLACTONES WITH AMMONIA AND RELATED NUCLEOPHILES 1. Reaction of the Lactone Group with Ammonia and Amines

Aldonamides are readily prepared by reaction of lactones with liquid ammonia (86,99,loo), with ammonium hydroxide (101,102), or by bubbling ammonia gas into alcoholic solutions of the sugar lactones (103 - 104). Aldonamides of the tetronic acids are stable in aqueous solution (lOS), but penton- or hexon-amides are hydrolyzed, as shown by the change of the optical rotation of the amide solutions (106). The hydrolysis is catalyzed by acids and bases, and the product was the ammonium salt of the aldonic acid. Aldonolactones also react with a variety of amines, producing substituted amides. Thus, heating L-gulonolactone with pphenetidines led to N-substituted gulonamides, products having antipyretic and analgesic activities ( 107). Condensation of aldonolactones with alkeneamines (such as, allylamine) or unsaturated urea derivativesafforded the alkeneamidesand ureides of polyhydroxy aliphatic acids, suitable intermediates for the synthesis of mercurated diuretic compounds ( 108).Gluconolactone and methyl arabinonate have been condensed with n-decylamine and higher homologues in a quest for synthetic emulsifyingagents (109). The arabinonamideswere only slightly soluble in water and had no emulsifying properties. The gluconamides 98 were more soluble in water and were weak emulsifying agents.

98 (n = 11, 15, 17)

The aggregation behavior of the eight diastereomericN-octylaldohexonamides of the D series, three of the L-series, and the correspondingracemates has been investigated, mainly by electron microscopy (1 10). The amides were obtained by electrolytic oxidation of the corresponding hexoses to the lactones and amidation with octylamine. Depending on the disposition of the hydroxyl groups and on the chain conformation, different types of aggregates were observed. Lactonic disaccharides (such as lactobiono-1,5-1actone) reacted with long-chain primary amines to yield model glycolipids of a new type ( 11 1,112). For the formation of N-substituted aldonamides from aldonic acids and amines, N-N’-dicylohexylcarbodiimidewas employed as the condensing agent. However, no catalyst was needed in the case of the lactones.

152

ROSA M. DE LEDERKREMER AND OSCAR VARELA

The synthetic glycolipids (for example 99) were obtained crystalline from ethanol. CH,OH

~0QocH~

(CHOJL

(cH2),w,

OH

( n = 7,9, 11, 13, 15)

99

Malto-oligosaccharidealdonolactones react with ethylenediamine to give N-(2-aminoethyl)aldonamides ( 1 13- 1 15), which have been successfully grafted onto carriers via amide linkages. The malto-oligosaccharides were produced by degradation of amylose with alpha amylase. After purification of the oligosaccharides,they were converted into the lactones by hypoiodite or electrolytic oxidation. The high-pressure aminolysis of a variety of lactones, including uronic acid lactones, has been described (1 16). However, the procedure was not applied to aldonolactones. Aromatic polyamines react with sugar lactones to give heterocyclic compounds having an attached open-chain polyhydroxyalkyl substituent.Thus, treatment of aldonolactones with o-phenylenediamine afforded ( 1 17) 2-polyhydroxyalkylimidazoles(100).

Condensation of ~-glucono1$lactone with 4 : 5-diamino-6-diethylamino-2-methylpyrimidineafforded the 8-polyhydroxyalkylpurine101, a Cnucleoside of an open-chain sugar ( 1 18). Some other pyrimidines gave analogous products from either D-glucono- or D-nbono-lactones. NEt,

I

HCOH

I I

HCOH CH,OH 101

SYNTHETIC REACTIONS OF ALDONOLACTONES

153

Similarly, open chain C-nucleosides of adenine have been prepared by condensing 4,5,6-triaminopyrimidinewith aldonic acids (or aldonolactones) of various chain-lengths. The amides (102) first formed were thermally cyclized, affording the 8-(hydroxyalky1)adenines (103) in rather low yields ( 1 19).

102

103

N-Acyl indoles derived from amides have been employed for the conversion of lactones into protected hydroxyacids. Thus, (chloromethy1)aluminum 2-(2-propenyl)anilide reacts (120) with 1,4- and 1,5-lactones, as for example per-0-tert-butyldimethylsilyl-D-ribono1,4-1actone(104),to afford hydroxyamides. After protection of the free hydroxyl group, these amides were converted by ozonolysis into N-acyl indoles, 105, which were readily saponified to the acid 106.

-

R1O

OR'

R20

OR2 I

0

dR'

I

OSi Me2 But

But Me2Si0

104

105 R1 = SiMe2But, R2 = CH,O(CW,OMe

106 R1 = SiMe2But. Rz = CH,O (CH3,O Me

The procedure was proved to be general for the preparation of protected hydroxy acids from lactones ( 121). This apparently trivial process is often difficult to carry out, as the attempted derivatization of y or 8-hydroxyacids frequently results in relactonization rather than hydroxyl protection. The method was applied to several aldonolactones to produce the corresponding intermediate hydroxyamides. Protection using [(2-trirnethylsilyl)ethoxylmethyl chloride, methoxymethyl chloride, tert-butylchlorodimethylsilane, or tert-butylchlorodiphenylsilanefollowed by ozonolysis gave the protected N-(y- or Ghydroxyacy1)indole derivatives. Mild saponification gave indole and the acetal- or silyl-protected hydroxy acids.

154

ROSA M. DE LEDERKREMER AND OSCAR VARELA

2. Intramolecular Reaction of Aminolactones to Yield Lactams

Replacement of an hydroxyl group at C-4 or C-5 of an aldonolactone by amino leads to the formation of aldonolactams. For example, Fleet and coworkers (1 22) reported the synthesis of D- and L-mannonolactam from Land D-gulonolactones,respectively. The synthesis involves the selectiveprotection of the primary hydroxyl group of the isopropylidene derivative 7b and sulfonylation of HO-5 to give 107a. Displacement of the triflate by sodium azide gave the corresponding derivative 107b, with inversion of configuration at C-5. Hydrogenation of the azide function produced an amine, which rearranged spontaneously to the Glactam 108. An identical sequence on L-gulonolactone led to D-mannonolactam (122,123).

+

CHzOR

I CH,OSi Me,But

107 a

R’ = H, RZ = OSO,CF,

108 R = SiMe,Bu‘

b R’ = N,. R 2 = H

In an allegedly “facile synthesis of nojirimycin,” the intermediate nojirimycin d-lactam derivatives 109a and 109b were reported (124) to be obtained on treatment of 2,3,4,6-tetra-O-benzyl-~-glucono1$lactone (51a) with aqueous ammonia at room temperature, or with benzylamine in refluxing toluene, and with catalytic amounts of ion-exchange resin. Hydrogenolysis of 109b with 5% Pd-C in acetic acid gave nojirimycin (110). CH,OBn

CHzOH I

I

I

I

OBn

I09 a

R= H

OH

110

b R=Bn

Difficulties were encountered ( 125)in repeating the previously described synthesis of nojirimycin. Thus, when the lactone 51a was treated with benzylamine under a number of conditions, the only product isolated was, as

SYNTHETIC REACTIONS OF ALDONOLACTONES

155

expected, the amide 111.An authentic sample of 109b was obtained (126) by synthesisfrom the idofuranosederivative 112. Hydrogenationof 109b under the conditions reported (124) led to partial or complete debenzylation, but there was no evidence for the formation of nojirimycin (110)in the reaction mixture. The synthesis of p-lactams from aldonolactonesor their 2,3-unsaturated derivatives is discussed in Section XII. CONHBn

I I BnOCH I HCOBn I HCOH I

HCOBn

CH,OBn

111

112

3. Reaction of the Lactone Group with Hydrazine and Derivatives Sugar lactones react readily with hydrazine to give crystalline derivatives useful for isolation and identification (127). Thus, addition of hydrazine to a reaction mixture containing an aldonolactone facilitates isolation of the product. The lactone may be regenerated from the hydrazide by treatment with nitrous acid (128). The phenylhydrazides obtained on treatment of aldonolactones with phenylhydrazine are also useful for characterization (129,130). Pentono- and hexono- 1,Clactones yielded substituted arylhydrazideson treatment with m-and p-tolyl-, m-and pmethoxyphenyl, p-bromophenyl, and p-ethoxycarbonylphenyl-hydrazines(1 3 1). It was found that the rate of hydrazinolysisdepends on the configuration of the aldonolactone, as well as on the aryl substituent on the hydrazine. A series of aryl- and aroyl-hydrazide derivatives was obtained from D-glycero-D-gubheptono-1,Clactone (132). The reaction was faster with arylhydrazines than with aroylhydrazines. The aroylhydrazides may be decomposed by copper(I1) sulfate or nitrous acid to regenerate the precursor lactone. D-Glucono-1,5-1actone reacted with aroylhydrazines (133) to give the corresponding 1-aroyl-2-~-gluconylhydrazides (113), which underwent cyclization upon heating with triethyl orthoformate, affording 5-aryl-2-ethoxy-3-~-gluconyl-2,3-dihydro-1,3,4-oxadiazoles (114). Other heterocyclic compounds, precursors of C-nucleosides, were prepared by condensing D-glucono-1,5-lactone with 1-hydrazino-4-phen-

ROSA M. DE LEDERKREMER AND OSCAR VARELA

156

CONHNHCOR

I I HOCH I

HCOH HC(OEt), +

o\N/N

I I HCOH I

HCOH

c=o

I I

HCOH CH20H

HOqH

HCOH

I I

HCOH CHzOH 113 R = A r y l

114 R = A r y l

ylphthalazine ( 134). Cyclization of the intermediate hydrazide 115 by autodehydration, through the enolic form, gave 3-(~-gluc~-pentitol1-yl)5-phenyl-1,2,4-triazolo[3,4-a]phthalazine(116). N-

NH

\

/

HCOH

HCOH HCOH

Ph

Ph

I

CH20H

CH20H

115

I I

HCOH

116

Aldonhydroximo-lactones(aldonolactoneoximes) have been prepared by oxidation of monosaccharide or disaccharide oximes with activated MnO,, Hg(OAc),, or oxygen in the presence of Cu,Cl,-pyridine. The structure of the hydroximo-lactones depends both on the starting sugar and on the method of oxidation. Thus, D-glucose oxime gave the hydroximo-1,WacCH20H

I

FH20H

OH 117

OH

I18

SYNTHETIC REACTIONS OF ALDONOLACTONES

157

tone 117 by oxidation with MnO,, the 1,4-analog 118 upon oxidation with O2-Cu2Cl2-pyridine, and a mixture of 117 and 118 by treatment with Hg(OAc), . Hydroxyl-free and protected derivatives were prepared, and other approaches for their synthesis have been examined (135). Compound 117 is a competitive inhibitor of p-D-glucosidases (136). VIII. REDUCTION OF ALDONOLACTONES 1. Reduction to Aldoses and Alditols A common reagent for the reduction of lactones to aldoses is sodium amalgam, first described by Emil Fischer ( 137),in the course ofhis studies on the structureproof of aldoses.Fischer applied the cyanohydrin reaction to an aldose, obtaining the next higher pair of epimeric acids, which as lactones were reduced to aldoses with sodium amalgam. [For a current appreciation of the significanceof this work. See Ref. (137a)l. This reagent was also used for the reduction of the monolactones of aldaric acids to afford the corresponding uronic acids (1 38 - 140),which may be further reduced to aldonic acids (141). The reduction by sodium amalgam is strongly influencedby the pH of the reaction (142).For the optimum pH value (3-4), the reaction is completein 10 min. The pH must be carefully adjusted during the first 3 min of reaction, as rapid liberation of base and formation of the sugar take place during this period. At pH 2 - 3, the pH control is difficult because of the high rate of reaction, and at pH 4 - 5 the yields of sugar were 10- 15% lower than those obtained at the optimum pH. Sulfuricacid, benzoic acid, sodium hydrogenoxalate, and ion-exchange resins have been employed for maintaining the desired pH of the reaction (143). Catalytic hydrogenation of 1 3 - and 1,4-aldonolactonesin the presence of platinic oxide led, depending upon the reaction conditions, to aldoses or alditols (144). The yields are also dependent upon the configuration of the starting lactones. Thus, both lactones of D-mannonic acid were readily reduced to D-mannose, whereas D-glucono-1,Clactone gave 23.4% of the sugar and 24%of the alditol. Under the same conditions, D-glucono-I $lactone gave an 80% yield of D-glucose, with no alditol formation. Various hydride reagents reduce aldonic acid lactones to aldoses or alditols. For example, sodium borohydride has been widely used. D-glycero-Dgulo-Heptono-1,4-1actone was reduced to the corresponding heptose by maintaining the pH (3-4)through addition of sulfuric acid or to the alditol under alkaline conditions ( 145). Several aldonolactones have been reduced to the corresponding alditols with sodium borohydride by keeping the pH acidic during 1 h and then making it alkaline through addition of sodium

ROSA M. DE LEDERKREMER AND OSCAR VARELA

158

hydroxide (146).The yield of alditol was alwayshigher than 95%.Reduction of 1,4-or 1,5-1actonesdoes not proceed readily in alkaline solutionsbecause the formation of salts of aldonic acids prevent reaction with the reducing agent. Yields were improved by maintaining a low pH through buffering with boric acid, ion-exchange resins, or carbon dioxide (147). For example, under acidic conditions, D-gulono- 1+lactone was reduced to D-gulose ( 148). Also, Pedersen and coworkers (42,45) have described high-yielding reductions of bromodeoxyaldonolactoneswith sodium borohydride in the presence of an acidic ion-exchange resin. The reduction of 4,6-dideoxy-~-ribo-hexono1,5-lactone to the corresponding aldose with various reagents has been studied (149). For the reductions in aqueous medium, ammonia-borane (NH, - BH,) and sodium borohydride were used, and for tetrahydrofuran solutions, sodium and lithium aluminum hydride were employed. The latter selectivelygave the a-anomer of the hexose, in 90%yield. The mixed hydrides AlC1,H and AICIH,, generated from LiAlH, and AlCl,, in molar ratios of 1 :3 and 1 : 1, respectively, proved to be effective for the preparation of D-glucose or D-mannose from their aldonolactones ( 150).For these reductions, the O-tetrahydropyrany1 derivatives were used, as the lactones themselves were unreactive. In some instances, relatively low yields of aldoses were obtained, as they were rapidly reduced to the alditols. For the tota! synthesis of certain trideoxy sugars of biological significance [rhodinose (1 5 1) (2,3,6-trideoxy-~-threo-hexose, 120a) and DL-amicetose (152) (2,3,6-trideoxy-~~-erythro-hexose, 120b], diisobutylaluminum hydride was used in the stage of reduction of the corresponding rhodinonolactone (119a) and amicetonolactone (119b). The free trideoxy sugars were obtained as an anomeric mixture of furanoses and pyranoses, in 66 and 67% yields, respectively. R3 FH3

CH3

I

Diisobutylaluminum hydride was also employed (153) for the reduction of 2,6-dideoxy-3-C-methyl-~-arubino-hexono1,Clactone (121) to an antibiotic component, the branched-chain sugar evermicose (2,6-dideoxy-3-Cmethyl-D-arubino-hexose, 122). The L-enantiomer of 122 (olivomycose, 125a), L-amicetose 125b, and 2,6-dideoxy-3-C-methyl-~-ribo-hexose (L-

SYNTHETIC REACTIONS OF ALDONOLACTONES

159

mycarose, 12%)were also synthesized by a route that involvesthe diisobutylaluminum hydride reduction of the correspondinglactones 123a, 123b, and 124, obtained from noncarbohydrate precursors (154).

121

122

@= HOCH

125 a R1 =OH, R2 = CH, b RI=RZ=H c R1 = CH, R2 = OH 124

Borane has been used for the reduction of aldonolactones (150). The conversion of a lactone into the aldose seems to be rapid, and subsequent reduction to the alditol is slower. Employing a 10-fold excess of M diborane in tetrahydrofuran afforded isolated yields of D - ~ ~ U C Oand S ~ D-mannose of 80 - 85 and 60 - 65%, respectively, from the corresponding lactones. Brown and Bigley (1 55) described the use of borane derivatives, namely diakylboranes, for the reduction of lactones to lactols. These reagents have proved highly effective for converting 0-acylated aldonolactones into the corresponding 0-acyl aldoses having HO- 1 free. Thus, 2,3,5,6-tetra-0-benzoylD-gulono-1,Clactone was reduced with bis(3-methyl-2-butyl)borane (diisoamylborane,disiamylborane) to 2,3,5,6-tetra-0-benzoyl-~-gulofuranose

160

ROSA M. DE LEDERKREMER AND OSCAR VARELA

in 97% yield (156). The reaction was also applied to the tetra-0-benzoyl-aldono- 1,Clactones of the ~-gulo,D-&, D - ~ u ~ oand , ~ - ~ l tconfigurations ro and to D-galactono-1,4-1actone tetraacetate, furnishing the corresponding acylated tetraacylhexofuranosesin very high yields ( 157). Hydroboration reactions proceed slowly in most organic solvents, but they are catalyzed by ethers. The reaction may therefore be performed in ethyl ether, tetrahydrofuran, 2-methoxyethyl ether, and the like. Tetrahydrofuran is usually the solvent of choice because it readily dissolves the diakylborane and the per0-acyllactone. The reduction is complete after 14- 20 h of stirring at room temperature and constitutes a general procedure for the preparation of furanose derivatives. Acylated aldono- l $lactones were also effectively reduced by diisoamylborane (1 58). This reagent has been widely used in our laboratory for the synthesis of deoxy sugars from deoxyaldonolactones, as described in Section X. Dijong and coworkers (159,160) reported acetyl migration and ring expansion during the diisoamylborane reduction of acetylated aldono- 1,blactones. According to these authors, ring enlargement takes place only when the aglycon and the side chain are cis oriented in the intermediate 0-dialkylboryl furanoside and when the acetoxy groups at C-4 and C-5 have a threo-configuration. Thus, 5-0-acetyl-~-rhodinono-1,4-lactone underwent ring expansion during reduction, whereas 5-0-acetyl-~amicetono- 1,4-1actonedid not. Furthermore, these authors could not reproduce the yield reported (1 57) (83Yo) for the diisoamylborane reduction of 2,3,5,6-tetra-O-acetyl-~-galactono1,Clactone, and the highest yield obtained for the galactofuranose derivative was 14%, even under optimized conditions. Reduction of tetra-0-acetyl-D-glucono-1,4-lactone led to decomposition products. However, in our hands, both the acetylated(16 1)and the benzoylated ( 162) derivatives of D-galactono-1,Clactone gave high yields of furanoses, comparable to those reported by Kohn et al. (157). Diisoamylborane has been employed for the reduction of free aldonolactones. Because diisoamylborane reacts with free hydroxyl groups, an excess of the reagent is required (163). By this means, D-galactono-1,Clactone, D-glucono-1$lactone, and D-erythrono-1,4-lactonecould be reduced to the correspondingaldoses in 60 - 70%yield. In the synthesis of 2-deoxy-~~and L-riboses ( 164)through a Reformatsky reaction from noncarbohydrate precursors, the intermediate2-deoxy-~~and L-ribonolactoneswere reduced to the deoxy sugars by diisoamylborane. Other unprotected 2-deoxylactones have been successfully reduced by diisoamylborane (42,165). Although the reduction of acylated or free lactones to aldoses with diisoamylborane is eminently satisfactory, the preparation of the reagent is tedious, and solutions of diisoamylborane are difficult to manipulate. In order to find a reagent having more desirable handling properties, other alkylboranes, including bis(2,4,4-trimethylpentyl)borane,bis(2,3-dimethyl-2-butyl)borane,

SYNTHETIC REACTIONS OF ALDONOLACTONES

161

and dicyclohexylborane, have been studied (166). It was found that, when the size of the alkyl group is increased, the extent of reduction decreases, and thus diisoamylborane is the most satisfactory reagent. On the other hand, diisocamphenylborane, which is more readily prepared and handled than diisoamylborane ( 167), was successfully employed for the reduction of 2,3 :5,6-di-O-isopropylidene-~-mannonolactone (lob) to the corresponding mannofuranose derivative. Polarographic (1 68) and electrochemical (169) procedures for the reduction of D-ribono-1,4-1actone have been developed, and the latter has been applied on a pilot-plant scale.

2. Isotopic Labeling and Substitution at the Anomeric Center of Aldoses by Reduction of Aldonolactones Reduction of aldonolactones and their derivativeswith isotopicallymodified reducing agents leads to sugars labeled at the anomeric center. Glycosidessubstituted with deuterium or labeled with tritium are widely employed for kinetic isotope-effect measurements, mechanistic studies, isotope-tracing experiments, and so on. D-( 1-2H)Glucopyranosewas prepared (170) by direct reduction of D-glucono- 1,5-1actone,in deuterium oxide solution, with sodium amalgam in the presence of phosphoric acid-d,,. or, alternatively,by reduction of the lactone tetrahydropyranyl derivativemth sodium borodeuteride in tetrahydrofuran (171). A one-pot, high-yielding synthesis of 1,2,3,4,6-penta-O-acetyI-j?-~-( 12H)glucopyranose (127) from tetra-0-acetyl-D-glucono-1,5-1actone (126) has been reported (172). Sodium borodeuteride reduction of 126, followed by in situ acetylation, gave the readily isolated and crystalline 127. The crystalline 2,4-dinitrophenyl 2,3,4,6-tetra-O-acetyl-j?-~-( 1-2H)glucopyranoside (128) was subsequently obtained from 127. CH,OAc

CH,OAc

-

Qo

AcO

AcOQH3(N02)2

AcOQ

C

OAc

126

CH,OAc

OAc 127

OAc 128

The cyanohydrin synthesis of higher sugars, which involves intermediate aldonolactones, allows the introduction of a I4C label in the sugar chain. Thus, for example, ~-[5-'~C]arabinose was synthesized (12) from D-xylose, which was first converted, by addition of K 14CNand hydrolysis, into D-[ 1-

162

ROSA

M. DE LEDERKREMER AND OSCAR VARELA

14C]gulono-1,Clactone. The hydroxyl groups at C-2 and -3were protected by isopropylidenation,and the 5,6-glycol was oxidized by sodium periodate. Treatment of the resulting syrupy product with methanolic hydrogen chloride, followed by borohydride reduction and hydrolysis, afforded L-[~-'~ Jarabinose. C IX. /%ELIMINATION REACTIONS Because of the activating effect of the carbonyl group, aldonolactone derivativesreadily undergo /I-elimination reactions to yield unsaturated lactones. 1. /%Eliminationin Aldono-1,5-lactones

a. Enono-1,5-lactones.-One of the first reports on p-elimination in aldonolactones was the observation by Haworth and Long ( 173) of the unexpected formation of furan-Zcarboxylic acid upon heating an aqueous solution of 2,3,4-tri-0-methyl-~-xylono1,5-lactonein the presence of pyridine. In 1944, Isbell (1 74) advanced a mechanism based on enolization and p-elimination of the starting lactone. In the course of investigations ( 175) on the structure of evernitrose, the corresponding 1,5-lactone 129, obtained upon oxidation of the sugar, yielded the a,j-unsaturated derivative 130 when refluxed with methanolic potassium acetate. The same enonolactone 130was obtained from L-mycarose (125c).

Meov Meo MeOH KOAc

0

__t

I

/

Me 129

Me 130

Oxidation of evermicose (122) with bromine yielded a mixture of y- and 8-lactones, which was directly acetylated. Refluxing the acetate in benzene solution in the presence ofp-toluenesulfonicacid gave ( 176)a mixture of the unsaturated lactones 131 and 132. In related work, Ganguly and Saksena (1 77) obtained an enonolactone by oxidation of D-nogalose with Jones' reagent, followedby @-eliminationpromoted by piperidine. Similarly, L-nogalose gave the enantiomeric lactone. a,fl-Unsaturatedderivatives are readily formed on acylation of aldonolactones under alkaline conditions. Thus, Dijong and Wittkotter (178) found that acetylation of 4-0-benzyl-D-g/~cero-D-~/~heptonoI ,5-lactone(133a)

SYNTHETIC REACTIONS OF ALDONOLACTONES

/

163

Me

Me

131

132

or its 7-0-trityl derivative (133b) with pyridine-acetic anhydride led to formation of the unsaturated 1$lactone derivatives 134a and 134b, respectively. CH,OR

CH,OR

I

I

HOCH

-

13nOQo

HO

OH

133 a R = H b R=Tr

BlrQ

0

OAc 134 a R = A c b R=Tr

In the hexosamine series, Kuzuhara and Emoto (179) obtained 2-benzamido-4,ii-O-benzylidene-2,3-dideoxy-~-erythro-hex-2-enono1,5-lactone (136a) by refluxing the hexosaminic acid derivative 135a with acetic anhydride-pyridine. The 2-benzyloxycarbonyl derivative (135b) also underwent&elimination (180)when acetylated in basic media, to give 136b.A similar result was obtained on acetylation of 2-amino-4,6-0-benzylidene-2deoxy-D-gluconic acid (135c) with acetic anhydride and sodium acetate, which afforded ( 181) the 2-enono- 1,5-1actone 136c.

135 a R = B z b R=CO,Bn c R=H

136 a R = B z b R = C0,Bn c R=Ac

164

ROSA M. DE LEDERKREMER AND OSCAR VARELA

Benzoylation in an excess of pyridine also favors p-elimination. Thus, treatment of D-glucono-1$lactone with an excess of benzoyl chloride and pyridine for 16 h at room temperature gave crystalline 2,4,6-tri-O-benzoyl3-deoxy-~-eryrhro-hex-2-enono1,5-lactone (137a) in 97% yield ( 182). Under similar conditions L-rhamnono-1,5-lactoneafforded ( 158) 2,4-di-0benzoyl-3,6-dideoxy-~-erythro-hex-2-enono1$lactone (138). p-Elimination is also the main reaction occurring when acylated sugars having HO- 1 free are oxidized in basic media. When 2,3,4,6-tetra-0-acetyl-a-~-glucopyranose was treated with methyl sulfoxide- pyridine -triethylamine- sulfur trioxide, oxidation followed by eliminationtook place readily to afford ( 183) 2,4,6-tri-0-acetyl-3-deoxy-~-erythro-hex-2-enono1$lactone (137b). The same product was obtained starting from p-D-glucopyranose tetraacetate and also from the mannose analog (184). On the other hand, the per-0benzyl derivativeofglucose afforded the saturated 1,5-lactone 51a under the same conditions; that is, the benzyloxy group was not eliminated. However, Tsuda er al. (185) reported the formation of 2,4,6-tri-O-benzy1-3-deoxy-~erythro-hex-2-enono-1,5-lactone (137c) upon treatment of 2,3,4,6-tetra-0benzyl-D-glucono-1,5-lactone (51a) with sodium methoxide.

137 a R = Bz b R=Ac c R=Bn

138

In the aldopentose series, 2,3,4-tri-O-acetyl-a-~-xylopyranose afforded the correspondingunsaturated 1,5-1actoneby oxidation - elimination. Likewise, hepta-0-acetylcellobiose gave, upon oxidation ( 184), the product of p-elimination at the reducing end. 2,3,4,6-Tetra-0-benzoyl-~-glucoand D-mannopyranoses were not oxidized by the same reagents. Studies on 2-acetamido-2-deoxyaldohexoses, fully acetylated except at 0-1, have been reported (186). Oxidation of the gluco isomer 139 with methyl sulfoxide-acetic anhydride afforded the unsaturated lactone 141 as the major product, along with the fully acetylated lactone 140. The manno isomer also gave 141, but in much lower yield. The 0-benzyl analog of 139 did not undergo p-elimination when oxidized with the same reagent. Oxidation of 4,6-isopropylidene or benzylidene acetals of 2-acetamido-2-deoxyhexoses of the D-gluco and ma man no configurations by the Fktizon reagent

SYNTHETIC REACTIONS OF ALDONOLACTONES

I65

(187) led (188) to the a,p-unsaturated lactones (142 or 136c), together with the saturated lactones. Treatment of the latter with ptoluenesulfonyl chloride in pyridine afforded the p-elimination products (142 or 136c). The free unsaturated lactone, 2-acetamido-2,3-dideoxy-~-erythro-hex-2-enono1,5lactone (143), was prepared from 142 by hydrolysis of the isopropylidene group. In a related paper, Pokorny et al. (189) reported the preparation of 2-acetamido-2,3-dideoxy-~-threo-hex-2-enono1,5-1actone. Me,SOAc,O

AcO

+

Qo

~

0

AcO

AcO

NHAc

NU4c

139

NHAc

140

141

0-CH,

CH,OH

-

Me2(oQo

Qo

HO

NHAc 142

NHAc

i 43

b. a-Pyrones. -Acylated aldono- 1,5-1actonesyield pyran-2-one derivatives by a double elimination, but more vigorous conditionsthan those used for the formation of 2-enono-lactonesare required. Nelson and Gratzl(l90) described the production of 3-acetoxy-6-acetoxymethylpyran-2-one (144a), in 92% yield, upon heating D-glucono-1,5-lactone with acetic anhydridepyridine for 1 h at 80°C. When the reaction was conducted at room temperature for 30 h, a mixture of the monounsaturated lactone (137b) and the a-pyrone 144ain 5.5 :26 ratio was obtained. Treatment of L-rhamnono-l,5lactone with an excess of benzoyl chloride and pyridine for 20 h, with subsequent sublimation of benzoic acid from the mixture at 120°C in vucuo, afforded 3-benzoyloxy-6-methylpyran-2-one(145) as the main product (191). The conversion ofthe 2-enono-lactones137bor 138into the 2-pyrone derivatives145 or 144a,respectively,required both acid and base, suggesting a concerted process in which the 4-acyloxy group is removed by acid, with the formation of an incipient cation stabilized by allylic resonance. Simultaneous abstraction of H-5 by a base results in the stable a-pyrone. In fact, the 3-benzoyloxy-2-pyrones144b and 145, and the 3-acylamido-2-pyrone146,

166

ROSA M. DE LEDERKREMER AND OSCAR VARELA

have been obtained by reaction of the unsaturated sugar-lactonederivatives 137a, 138, and 141, respectively,with tin(1V) chloride (192). On prolonged reaction, the 2-enonolactone 137a afforded the halogenated pyrone 144c. The 3-acetamido 2-pyrone 146 was obtained on treatment of 136c with concentrated hydrochloric acid followed by acetylation. ~o

~o

~

0

-

-

__ OBz

OR1

$44 a R1 = Ac, Rz = OAc b R1 = Bz, R2 = OBz c R1 = Bz. R2 = CI

145

NHAc

146

2. &Elimination in Aldono-1,4-lactones

2(5H)-Furanones (A%utenolides). -Aldono- 1,44actones readily undergo P-elimination on acylation to yield furan-2-one derivatives. Lederkremer and Litter (1 93) reported formation of the furanone derivative 147 on benzoylation of D-galactono-1,Clactone with an excess of benzoyl chloride and pyridine for 16 h at room temperature. The configurationof the exocyclic double bond was identified (194) as 2. P-Elimination also could be accomplished by treatment of the acylated lactone, in chloroform solution, with triethylamine, which gave 147 as an ( E , Z )diastereomeric mixture ( 194). Similarly, 2,3,5-tri-O-benzoyl-6-O-trityl-~-galactono1,Clactone afforded the furan-2-one derivative 148 in 78% yield (195). The precursor enonolactones, resulting from a single elimination, could not be isolated, probably because of a fast second elimination yielding the very stable furanone. However, prolonged acetylation (48 h) of 6-O-trityl-~-galactono-1,4lactone gave the monounsaturated lactone 149, but only in 9% yield, with

147

148

149

2,3,5-tri-O-acetyl-6-O-trityl-~-galactonoI ,4-lactone being the major product ( 196). In contrast, 5,6-O-isopropylidene-~-galactono-, L-gulono-, and

SYNTHETIC REACTIONS OF ALDONOLA~ONES

167

D-mannono 1,4-lactones(8b, 9a, and 10a) react readily with mesyl chloride in pyridine at 0 "C to produce (197) the corresponding hex-2-enono-1,4-lactone 2-mesylates 150- 152.

150

152

I51

Benzoylation of D-glycero-D-gulo-heptono-1,Clactone with an excess of benzoyl chloride and pyridine afforded the hept-2-enono-1,Clactone as the main product (198). The di- and triunsaturated compounds were isolated in very low yield from the mother liquors (199). Higher yields of the di- and triunsaturated derivatives 153 and 154 were obtained when the p-elimination reaction was performed with triethylamine on the previously synthesized per-0-benzoyl D-glycero-D-@lo-heptono-1,4-1actone.Employing 10% triethylamine in chloroform, the lactone 153 was obtained as an E, Z diastereomeric mixture in 9: 11 ratio as determined by 'H n.m.r. When 20% triethylamine was used, the furanone 154 was obtained in 59% yield (200). Its structure was assigned, on the basis of 'H and I3Cn.m.r. spectra, as 3 benzoyloxy-(5Z)- [ ( Z ) -3-benzoyloxy-2-propenyliden] -2(5H)-furanone. The stereochemistryof the exocyclic double bonds was established (20 1) by nuclear Overhauser effect spectroscopy (NOESY).

,

\

OBZ

OBZ

153

154

The elimination of benzoic acid from benzoylated aldonolactonestakes place in successive steps. The relative extent of breaking the C-H bond and the C - OBz bond depends on the ease of removal of the hydrogen as a proton and the fact that the benzoate is a poor leaving-group. The H-2 proton in aldonolactones is acidic and it is likely that an E cb mechanism, through an

,

168

ROSA M. DE LEDERKREMER AND OSCAR VARELA

intermediate stabilized carbanion, operates in the first elimination step. This hypothesis is supported by the fact that the same furanone (147) was obtained from D-galactono-, D-glucono-, or D-mannono-1,Clactones ( 194), having a cis-orientation of leaving groups in the first two and trans-leaving groups in the last one. A similar mechanism could explain the elimination of the second molecule of benzoic acid, but formation of the triunsaturated lactone 154, obtained from the D-glyCero-D-gulo-heptOnOlaCtOne perbenzoate, would require a different intermediate. The driving force in this case is the stability of the resulting furanone. An acid-base catalyzed process, such as that proposed for the formation ofthe a-pyrone 144a,could explain the formation of 154. Treatment of 2,3,5-tri-O-benzoyl-~-rhamnono-1,4-lactone with triethylamine as just described afforded 3-benzoyloxy-5-ethylidene-2( 5H)-furanone (155) in 75% yield (202). 1,8-Diazabicyclo[5.4.0]undec-7-ene(DBU) effectively causes p-elimination in aldonolactone derivatives. Thus, 2,3,5tri-0-benzyl-D-ribono-1,Clactone afforded 2,5-di-O-benzyl-~-glyceropent-2-enono-l,4-lactone(156a) in 9 1% yield when treated with DBU in hexamethylphosphoramide (203). Similarly, Barrett and Sheth (204,205) treated 2,3,5-tri-O-acetyl-~-ribono1,Clactone with DBU in tetrahydrofuran for 3 h at - 20°C and obtained 3-acetoxy-5-methylen-2(5H)-furanone (157) in 94%yield. Other bases were less efficient.The monounsaturated compound 156b was formed, together with the furanone 157, in 2: 1 ratio when the elimination reaction was conducted with DBU for 4 h at - 78°C. Similarly, the corresponding enono-lactones 156c and 156d were obtained from the 5-0-trityl and the 5-0-tert-butyldiphenylsilyl derivatives, respectively. In a related reaction, Komura et al. (206) obtained the furanone 156e by mesylation-p-elimination of 2,5-di-O-tert-butyldimethylsilyl-~-ribono1,44actone.

OBz 155

Rzocu H2&Q

-

OR1 156 a R1 = R2= Bn

OAc 157

b R1 = R2 = Ac c R1 = Ac. R2 = Tr

d R1 = Ac. R2 = SiPh,But e R1 = R2 = Si Me,But

The lactone derivative 158 obtained from D-ribono-1,Clactone afforded (-)-(R)-angelica lactone (159a) upon treatment with methanolic ammonia solution (6). In a similar way, Font and coworkers (207)synthesized (+)-(S)-

SYNTHETIC REACTIONS OF ALDONOLACTONES

169

angelica lactone (159b),and Danilova and coworkers (208) obtained (S)-hydroxymethyl-2(5H)-furanone (159c). The lactone 160 synthesized from 2deoxy-D-erythro-pentose afforded the furanone 161 by mesylationelimination (209).

BzO

159 a

I58

R1 = CH,

R2 = H

b R1 = H, R2 = CH, c R1 = CH,OH, R2 = H

I CH20SiMe,But

AH,OSiMe,But

160

161

A trans-acetoxyelimination was described (2 10)for the 2-bromodeoxyaldono-l,4-lactone 33b, induced by NaHSO,, to afford the butenolide 162a. However, the C-2 epimeric bromolactone 33a, undergoes a trans-P-bromoacetoxy elimination to produce (211) butenolide 162b. CH, Br

u I

AcOCH

0

R

162 a R = Br b R=H

2-Acylamido-hex-2-enono-1,4- (164aJ64b) and 1,5-lactone derivatives (141,165)are obtained ( 181) by acylation of 2-amino-2-deoxy-~-gluconic acid (163)with the acyl chloride in pyridine,whereas reaction of 163 with hot acetic anhydride-sodium acetate afforded (18 1,212)the butenolides 166-E and 166-2, earlier wrongly formulated (2 13)as 146. The isomers 166-2and 167-2 were obtained in high yield on treatment of furanones 164a or 164b

ROSA M. DE LEDERKREMER AND OSCAR VARELA

170

CYOR

CH,OR

I

I

co; I+ HCNH, I

NHR 141 R = A c 165 R = Bz

964 a R = A c b R=Bz

HOCH

I HCOH I HCOH I

CH,OH

+

163

NHAc

NHR

166 E

166-Z R = A c 167-2 R = B z

~

with DBU ( 1 8 1). The P-elimination reaction was extended to a lactonic disaccharide.Thus, treatment of 168 with triethylamine afforded the diunsaturated derivative 169 as an E, 2 mixture ( I 94).

HCOBz

I

HdOBz

I

bBz

CH,OBz

OBz

uocH H Q

HCOBz

I

OBz

OBz

CH,OBz 168

169

X. SYNTHESIS OF DEOXY SUGARSFROM ALDONOLACTONES 1. Catalytic Hydrogenation of Enonolactones

Catalytic hydrogenation of benzoylated 3-deoxy-2-enono-1 $lactones is stereoselective and affords the 3-deoxyaldonolactonederivatives, useful intermediates for the synthesis of deoxy sugars. Thus, 2,4,6-tri-O-benzoy1-3-

SYNTHETIC REACTIONS OF ALDONOLACTONES

171

deoxy-~-erythro-hex-2-enono1,5-lactone137a, obtained in 97% yield from the commercially available D-glucono-1,5-lactone, was selectively hydrogenated ( 182)over palladium - charcoal to give the 3-deoxy-~-arabino-hexono-l,5-lactone derivative 170. Reduction and debenzoylation of 170 afforded (2 14) 3-deoxy-~-arabino-hexose (171).On the other hand, oxidative degradation of 3-deoxy-~-arabino-hexono1,5-lactonewith ceric sulfate in M sulfuric acid gave 2-deoxy-~-erythro-pentose (2 15). Similarly, 2-deoxyD-fyxo-hexose and 2-deoxy-~-arabino-hexosewere synthesized from 3deoxy-D-gafacto-heptono-1 ,44actone and 3-deoxy-~-gfuco-heptono-1,4lactone, respectively (216). The reaction may be controlled to yield from common aldonolactones the aldose having one fewer carbon atom (2 17,218). FHzOBz

CH20H

170

171

The sequence /%elimination-catalytic hydrogenation- diisoamylborane reduction was applied (156) for the synthesis of ascarylose (174a). Starting from L-rhamnono-1,5-lactone, compound 174a was obtained via the dideoxylactone 172s and the sugar derivative 173a. When catalytic hydrogenation of the enonolactone138 was performed with deuterium, stereospecific labeling took place (2 19)at C-3and at the 2R position of ascarylose (174b). 138

-

-

BZOQ

OH

BZOQ

OBZ

OBZ

172a R = H b R=D

173 a R = H b R=D

-

H

o

p

OH

174 a R = H b R=D

When the benzoylated lactone 172a was debenzoylated with sodium methoxide and the product boiled in 1,Cdioxane, the more-stable 1,4-lactone 175 was isolated upon rebenzoylation. Reduction of the lactone carbonyl group afforded (220) the crystallinefuranose derivative (176) of ascarylose. A similar procedure, performed on the benzoylated 3-deoxy-~arabino-hexono-1,5-lactone 170, gave (22 1) the 3-deoxy-~-arabino-hexofuranose derivative.

172

ROSA M. DE LEDERKREMER AND OSCAR VARELA

172a

-

R - Icy OBz

BzOCH

OBz

BzOCH

I

I

CH3

CH3 175

176

Catalytic hydrogenation (194) of butenolide 147, obtained from D-galactono-, D-glucono-, or D-mannono-1,Clactones, afforded stereoselectively the 2,6-di-0-benzoyl-3,5-dideoxy-~,~-threo-hexono1,Clactone 177.

i77

From the glycosylfuranone 169, obtained via p-elimination, a 1 : 1 diastereoisomeric mixture of the 2(S),4(R)- and 2(R),4(S)-3,5-dideoxylactones 178-(S,R) and 178-(R,S) was obtained (194). Therefore, a glycosylation reaction was employed for the resolution of the racemic dideoxylactone derivative 177.

178 ( S , R)

178 (R, S )

Trideoxy sugars have also been prepared from aldono- 1,Clactones. Thus, 2-0-benzoyl-3,5,6-t~deoxy-cu-~,~-thveo-hexofuranose was obtained (202) from L-rhamnono-1,Clactone via the furanone 155. L-Rhamnono- 1,5-lac-

SYNTHETIC REACTIONS OF ALDONOLACTONES

173

tone yielded (222), through the intermediate 3-benzoyloxy-6-methylpyran2-one (145), 2-0-benzoyl-3,4,6-trideoxy-~,~-threo-hexopyranose (179), which on debenzoylation gave the free sugar 180. Stereospecific catalytic hydrogenation of furanone 154 afforded (200) 2,7-di-O-benzoyl-3,5,6-trideoxy-D,L-threo-heptono1,Clactone.

145

-Q

OH

---+ NaMeO

Q

OH

OBz

OH

179

180

A simple procedure based on the sequence /3-elimination - catalytic hydrogenation of acylated aldonolactones was described by Pedersen et al. (223). Acetylated pentono- and hexono- 1,Clactones, when hydrogenated under pressure in the presence of triethylamine, afforded the acetylated 3-deoxy-aldono-1,Clactones in high yield. The 3-deoxygenationof peracylated D-gzyCero-D-gulo-heptOnO-1,Clactone was also described (44,224). The perbenzoate of 3-deoxy-~-gluco-heptono-1,4-lactonewas used (224) for the preparation of crystalline 3-deoxy-~-gluco-heptose. Hydrogenation of the lactonic disaccharide 168 in the presence of triethylamine and palladium afforded (194) the 3-deoxylactone derivative 181.

HC!OBz

I

HCOBZ

I

I

OBz

OBZ

CH,OBz 181

2. Hydrogenolysis of Bromodeoxyaldono-l,4-lactones Deoxyaldono-1,Clactones have also been obtained from the respective bromo derivatives. Thus, abequose (3,6-dideoxy-~-xyZo-hexose, 182)was

174

ROSA M. DE LEDERKREMER AND OSCAR VARELA

obtained (225) from D-galactono-1,44actone(Scheme I), and 2,5-dideoxyD-erythro-pentono-1,Clactone from the 5-bromo percursor (6). The starting material for the preparation of the latter was compound 3 (erroneously formulated as 2) (see Section 11).

l a

CH3

-

Qo-

HCOBz

I

HCOBz

OBZ

I

CH,R

"

G

O

OBz

H

OH

CH3

R = OCPh3

182

R = Br

SCHEME 1

The bromine atom of 38, 41a, and 183 may be replaced by hydrogen through catalytic hydrogenolysisin the presence of triethylamine(42,43), by treatment with iodide (41), or by treatment with hydrazine (226). When no acid acceptor is present, hydrogenolysisof 2-bromo-2-deoxyaldono-1,4-lactones yields the corresponding 2,3-dideoxyaldonolactonewith removal of the bromine atom and the C-3 hydroxyl group (227). Thus the 2-bromo-aldonolactone 183 gave the 2,3-dideoxyaldonolactone184. The two isomeric dibromolactones 185a and 185b both yielded 186a, which on hydrogenolysis in the presence of triethylamine was converted into the trideoxylactone 186b. The enantiomer was obtained by hydrogenolysis of 34. 0

HOCH,

PHO

183

185 a

R1 = Br, R2 z H

b R1 = H, R 2 = Br

184

186 a

R' = Br

b R1=H

SYNTHETIC REACTIONS OF ALDONOLACTONES

175

3. Base-Catalyzed Rearrangement of Bromodeoxy Aldonolactones Treatment of 6-bromoaldohexono-or 7-bromoaldoheptono-1,Clactones with strong base leads to rearrangement via epoxide migration. The reaction has been applied to the synthesis of deoxylactones (226). For example, 6bromo-2,6-dideoxy-~-arabino-hexono1,Clactone (187) yields 2-deoxy-~ribo-hexono-1,Clactone (190) through formation of the 5,6-epoxide 188, which, in strong base, rearranges to the 4,5-epoxide 189. The conversion of 189 into 191 may take place by intramolecular attack of the carboxylate anion on C-4 with inversion to yield the 1,Clactone 190, which opens with the base to give 191.

CyBr I y 2

HOCH

I

CH,OH 187

188

189

J

F" y 2

HOCH

I I HOCH I

HOCH

CH,OH

H&H

I

CHzOH 191

190

The 2-deoxylactone 190 was used for a synthesis of L-digitoxose (193). Treatment of 190 with hydrogen bromide in acetic acid gave the 6-bromolactone 192a,which on hydrogenolysis yielded the 2,6-dideoxylactone 192b. Reduction of 192b with diisoamylboraneafforded 2,6-dideoxy-~-ribo-hexose (193).When 192awas treated with aqueous potassium hydroxide, it gave the salt of 2-deoxy-~-arabino-hexonic acid, which, after acidification and acetylation, yielded the tnacetate of 2-deoxy-~-arabino-hexono1,4-1actone (194).This base-catalyzed rearrangement was also applied to 6-bromo-3,6dideoxy-aldohexono-1,Clactones (228).

ROSA M. DE LEDERKREMER AND OSCAR VARELA

176

190

-

0-0 HO~H

I

CH, R 192 a R = Br b R=H

HO~H

I

CH3

193

CHzOAc

194

The 6-bromo-3,6-didoxy-~-urubino- and D-xyb-hexono-1,6lactones 195 and 200 were formed on treatment of the corresponding 3-deoxylactones with hydrogen bromide in acetic acid. Upon treatment of 195 or 200 with anhydrous potassium carbonate in acetone, the two bromodeoxylactones were converted into the 5,6-epoxides 196 and 201.Treatment of the epoxide 196 or the 6-bromolactone 195 with excess of aqueous potassium hydroxide afforded the 3-deoxy-~-ribo-hexonate198 as the main product, which was characterized as 2,5,6-tri-0-acetyl-3-deoxy-~-ribo-hexono1,4lactone (199).The rearrangement occurs, as monitored by 13C-n.m.r.spectroscopy, through the intermediate 197.On the other hand, the 5,6-epoxide 201,having the xylo configuration, gave the 2,6- and 2,5-anhydrides as the major reaction products as determined from the l3C-n.m.r. spectra of the methyl esters 202 and 203.3-Deoxy-~-lyxo-hexonic acid was isolated as the acetylated 3-deoxylactone 204 in 15% yield. The 6-bromolactone 195 was converted by hydrogenolysis, followed by carbonyl reduction, into 3,6-dideoxy-D-arubino-hexose(tyvelose). The behavior of 7-bromo-2,3,7-t~deoxy-~-arab~no-heptono 1 ,Clactone (205)and 7-bromo-2,7-dideoxy- (208)and 7-bromo-3,7-dideoxy-~-glucoheptono- 1,Clactone(211)toward aqueous base has also been studied (229). The 6,7-epoxide is formed by potassium carbonate treatment of each bromodeoxy heptonolactone. With potassium hydroxide, epoxide migration occurs giving mixtures of epoxidesthat undergo intramolecular attack by the

SYNTHETIC REACTIONS OF ALDONOLACTONES

177

I

CH20H

195

196

197

J F”; HOCH

I

y

2

HOCH

I

HOCH

I

ACOCH

I

CH20H

CH20Ac

I99

HCOH

OH

CH2Br I

200

198

OH

HC, H2k0 201

CH20Ac

I

CH20Ac

202

203

OAc

204

ROSA M. DE LEDERKREMER AND OSCAR VARELA

178

carboxylate or by an alkoxide. Thus 205 is converted into 2,3-dideoxy-~-arabino-heptonicacid (206),which was isolated as the acetylated lactone 207b. Treatment of crude 207a with hydrogen bromide in acetic acid gave the 7-bromolactone(207c) enantiomer of 205. The 7-bromo-2,7-dideoxy (208) and the 7-bromo-3,7-dideoxy-~-gluco-heptono1,Clactone (211) gave anhydro heptonic acids on treatment with base. Thus, reaction of 208 with potassium hydroxide led to the formation of 3,6-anhydro-2-deoxy-~-idoheptonic acid 209 as the main product, as determined by I3C-n.m.r. spectroscopy. Compound 209 was isolated as the isopropylidenederivative 210. On the other hand, compound 211 gave the 2,5-anhydride212a as the major product, accompanied by a small proportion of the 2,6-anhydride 213a. These products were isolated by column chromatographyas the benzoylated methyl esters 212b and 213b, respectively.

CH,R~

F" y 2 y 2

HCOH

H~OH

I I

HCOH

I HOCH I HOCH I

Hr,,

- Yt HCOR' 0

CH,OH

CH,Br

206

205

0

II

OH,C

\

H~OH

I HCOH I

I

Me2C"

HOLH,

CH,Br

208

209

210

Ici

0

SYNTHETIC REACTIONS OF ALDONOLACTONES o

l

CH,OR~

+

OR2

L

179

OR2 R20

R~OCH

OH

HCOH

I HCOH I

I

CH,0R2

CH2Br 21I

XI. GLYCOSYLATION OF ALDONOLACTONES Selectively substituted aldonolactones have been used as glycosylating agents for monosaccharides. Glycosylaldono-1,Clactonesare useful precursors of disaccharides having the reducing end in a furanoid-ring structure. Condensation of 2,3,5-tri-O-benzoyl-6-O-trityl-~-galactono1,Clactone with either tetra-0-acetyl-a-D-glucopyranosyl bromide or D-glucopyranose pentaacetate, in the presence of silver trifluoromethanesulfonate or stannic chloride, respectively, afforded the 6-~-~-~-ghcopyranosy~-~-galactono1,Clactone derivative 214 in good yield (230). Diisoamylborane reduction of 214 gave the P-~-Glcp-(1 + 6)-P-~-Galfderivative215.

.;jl CH,OAc

HCOBz

AcO

-

OBZ

.;; CH,OAc

HCOBz

OBZ

AcO OAc 214

OAc 215

Galactofuranose disaccharides have been obtained by condensation of D-galactofuranose pentabenzoate (216) with a conveniently substituted Dgalactono-1,Clactone. Thus the stannic chloride-catalyzed glycosylation of 216 with 2,3,5-tri-O-benzoyl-~-galactono1,4-lactone (217) or its 6-0-trityl derivative gave the benzoylated P-D-galactofuranosyl-(1 + 6)-~-galactono-

ROSA M. DE LEDERKREMER AND OSCAR VARELA

180

-

174-lactone(218). Reduction of 218 by diisoamylboraneled to the disaccharide derivative, from which the methyl glycosidewas prepared. Debenzoylation gave methyl D-galactofuranosyl-(1 6)-P-~-galactofuranoside(219). The free disaccharide P-~-Galf(1 46 ) - ~ - G a lwas p also prepared (23 1).

I

CH20H 216

I

CH,OBz

217

218

I HLOH I

AH

CH,OH 219

-

A similar strategy was used for the preparation of the P-D-Galf( 1 5 ) - ~ Galf( 221) immunogenic unit of extracellular polysaccharides of Penicillum and Aspergillus species (232). 2,6-Di-O-benzoyl-~-galactono1,44actone (220), readily obtained by partial benzoylation of D-galactono-174-lactone, was used to glycosylate the galactofuranose benzoate 216. The HO-5 group of 220 reacted preferentially to give the P-glycosyllactone 221 in 70%yield; the trisaccharide derivative 222 was isolated as a by-product. Reduction of

SYNTHETIC REACTIONS OF ALDONOLACTONES

181

-

the lactones 221 and 222 with diisoamylboranefollowed by debenzoylation gave P-D-Gal$( I + 5)-~-Galjandj?-~-Gal$(I 43)-v-~-Gd$(1 5)]-DGalf (38).

+ I

I

HCOH

I

HCOBz

OBz

I

I

I OBz

CH,OBz

CH,OBz 220

221

OBZ

41l CH,OBz

CHzOBz I OBz HCOBz

222

XII. ALDONOLACTONES AS CHIRAL PRECURSORS FOR OF NATURAL PRODUCTS

THE

SYNTHESIS

The regio- and stereo-selectivefunctionalization of aldonolactones yields optically active lactones,which are important precursors in natural product synthesis. Concepts such as “chiral templates” and “chirons,” derived from carbohydrates, have been ingeniously and widely applied in synthesis (233). Among the commercially available aldonolactones,D-ribono- 1,Clactone is

ROSA M. DE LEDERKREMER AND OSCAR VARELA

182

probably the most intensively used in organic synthesisto date. The chemistry and utility of D-ribono-1,Clactone have been reviewed (2). Butenolides derived from D-ribonolactone,having a single chiral carbon atom, have been employed for inducing asymmetry. 1. Synthetic Uses of y-Butenolides Derived from Aldonolactones

Numerous procedures for the preparation of butenolides have been developed. Font and coworkers (234- 236) prepared the 5-0-substituted derivatives 223a-c of D-ribono-1,Clactone. The &glycol system of 223a reacted with N,N-dimethylformamide dimethyl acetal and then with iodomethane to give the trimethylammonium methylidene intermediate224. Pyrolysis of 224 gave the butenolide 225.

223 a

224

R=CH,

225

b R = PhCH,

c R = Ph3C

Butenolides have also been prepared from aldohexono-1,Clactones via trimethylammonium methylidene derivatives ( 15). 5,6-0-IsopropylideneL-gulono- (9a) and D-mannono-1,44actone(10a) were converted into 2-(dimethylamino)-1,3-dioxolane derivatives, which on treatment with iodomethane followed by thermal decomposition yielded compounds 226 and 227 respectively.

226

227

Cyclic orthoformates are useful intermediates for the synthesis of butenolides (235).Treatment of D-nbonolactone,or its 5-0-substituted derivatives, with one molar equivalent of ethyl orthoformate gave diastereoisomeric

SYNTHETIC REACTIONS OF ALDONOLACTONES

183

mixtures of orthoesters(228a- c) in quantitativeyield. Pyrolysisof the orthoformates 228a-c produced the correspondingbutenolides (159c, 229a, and 2298).

228 a R =

n

159 c R = X 229 a

b R=CH, c R-PhCH,

R=CH,

b R=PhCH,

Butenolide 231 was obtained by a procedure that did not involve a pyrolytic step (237). Thus, the 5-trityl ether 223c was converted into the thioxocarbonate 230 upon treatment with N,N'-thiocarbonylbis(imidazo1e). Raney nickel effected the conversion of 230 into 231.

223 c

-

Ph,COCH,

0

Ph,COCH,

0

'cj- y, o\c/o II S

230

231

Regiospecific trans-/?-bromo- acetoxy elimination of readily accessible 2-bromodeoxyaldono-1,ClactonesWith NaHSO, afforded (2 1 1)good yields of such butenolides as 159c and 162b. Enantiomerically pure 5,6-epoxides and 5,6-diols were obtained from 162b. Butenolides have been employed as convenient chiral precursors of such molecules as the antileukemic lignan lactones (+)-trans-burseran, (-)-isostegane, and (+)-steganacin (2). Butenolide 159c was identical to the product obtained by enzymic hydrolysis of ranunculin, a glycoside present in Ranunculaceae. A short and efficient synthesis of (-)-ranunculin, via 159c, has been described (236). The reactivity of butenolides with different nucleophiles has been studied (238).A general methodology for the preparation of y-alkyl-a,/?-butenolides or y-alkylbutanolactones from ribonolactone has been developed (239). Some of these products have pheromonal activity in

184

ROSA M. DE LEDERKREMER AND OSCAR VARELA

insects and are also used as fruit fragrances. Among the y-alkyl-2,3-butenolides, the readily available (6,207,238,240)angelicalactones(159a and 159b) have served as chiral starting compounds for the synthesis of asymmetric molecules. For example, angelica lactone was used for the synthesis of (+)and (-)-blastmycinone (240,24 1). Although a$-butenolides are inert to epoxidation by peroxyacids, epoxidation may be accomplished by using sodium hypochlorite in pyridine. Thus, (+)-angelica lactone (159b) gave stereoselectivelythe epoxy derivative 232. Oxirane ring-opening by iodide produced the diastereoisomericmixture of iodides 233, which was converted into the 2-deoxy derivative 234 by hydrogenolysis. The lithium enolate of 234 was alkylated at C-2 and then esterified with isovaleryl chloride, affording (+)-blastmycinone (235).

233

232

1 1 I

I

CH3

Bu 235

I CH3 234

The double bond of butenolides undergoes stereoselective Michael addition of organometallic reagents, affording useful synthetic intermediates. Thus 1,Caddition of lithium dimethylcuprate to 231 gave 236 as a single isomer, which was employed (237) for the synthesis of the bromopentene derivative 237. Similarly, 1,Cconjugated addition (242) of lithium divinylcyanocuprate to 238 gave the adduct 239, which on treatment with potassium carbonate in methanol produced the epoxide 240, a key intermediate for the synthesis of the carbocycline 241 through highly stereoselective processes.

SYNTHETIC REACTIONS OF ALDONOLACTONES

Ph,COCH,

0

+ + -

231-

185

1 ; I .

Me 236

238

237

239

240

241

A total synthesis of (3S,4R)-(+)-eldanolide (246), a sex attractant pheromone, has been reported (243). Compound 246 was synthesized by two differentroutes, both involving the butenolide 245 as the key precursor. The higher-yielding sequence is described here. Treatment of the tosylate acetal 242 with methanolic sodium methoxide led, as previously described by Hoffman and Ladner (244), to the epoxide 243. Addition of lithium diisobutenylcuprate to 243 afforded 244, which after successivehydrolysis of the isopropylidene group, treatment with triethyl orthoformate, and pyrolysis,

186

ROSA M. DE LEDERKREMER AND OSCAR VARELA

gave 245. The Michael addition of Me,CuLi to the latter takes place with high stereocontrol, the chirality at C-4 being responsible for the asymmetry induced at C-3, and gives the aldanolide 246. The reaction of differently substituted butenolides with alkyl cuprates has also been investigated, in order to obtain molecules having the (4S,5R)-4,5-dialkyldihydro-2-(3H)furanone type of constitution. Natural products possessing such a structure, as for example (+)-eldanolide (246) and (+)-trans-cognac lactone (247), have been synthesized (245). TsOCH, ~

0

~

L

i

i

e (Me,CCH),CuLi

o\

/o

CMe2

/o

0,

242

o\

CMe, 243

/o

CMe, 244

FH2CH=CMe,

CH,CH=CMe,

bo Me2CuLi

I

Me

246

245

Many strategies for stereochemical control on butenolides have been explored, and also reviewed (246), by Hanessian. For example, starting from butenolides having the 44s) or 44R) configuration (respectively obtained from L-glutamic acid or D-ribonolactone) the C-2 -C- 10 and C- 1 1- C- 16 subunits, 252 and 253, respectively, of the ionophore antibiotic ionomycin, have been synthesized (247). Conjugate addition of lithiumtris(trimethy1thio)methane to 248, followed by C-2 hydroxylation and desulfurization led to a 3-C-methyl lactone, which was readily converted into the epoxide 249. The butenolide 250 was prepared from 249 through a four-step synthesis that involves a two-carbon homologation, a lactone replication process, and oxidative elimination. Stereocontrolled addition of LiMe,Cu to 250,

SYNTHETIC REACTIONS OF ALDONOLACTONES

187

followed by selective silylation and deoxygenation, led to 251, which served as precursor for the synthesisof 252 and 253. A similar methodology was also employed for the synthesis of the C-1 -C-13 polyol system (248), and the C-14-C-20 and C-32 -C-38 segments (249) of the antibiotic amphotericin B, from readily available, single-chiral progenitors.

250

249

J

J

251

J

J

\ \ Ph02S

252

253

The double bond of butenolides reacts under Diels - Alder conditionsand the resulting chiral bicycles have served as precursors of prostacycline analogs and chrysanthemic acids (250,25 l). The butenolide 248 was obtained by theproceduredescribedbyIrelandeta1.(237).Abicyclo[4.3.0]ringsystem (254) was prepared by Diels-Alder reaction of 248 with butadiene in the presence of aluminum trichloride. Reduction of 254 (LiBH,) yielded the

188

ROSA M. DE LEDERKREMER AND OSCAR VARELA

diol255, which was converted into the substituted tetrahydrofuran 256 via the tosyl derivative. Cleavage of the double bond of 256, followed by esterification, provided the diester 257. A bicyclo[3.3.0] ring system (258) was obtained by Dieckmann cyclization and subsequent demethoxycarbonylation. On the other hand, compound 248 reacted with diazopropane to produce a mixture of diastereoisomeric pyrazolines, which was converted into the same bicyclo[3.1.O] system (259) by irradiation.

LJ 254 R = Bu'PhzSi

255 R

Bu'Ph,Si

256 R = Bu'Ph,Si

I

259 R = ButPhzSi

258 R = ButPh,Si

257 R = ButPhzSi

The pyrazoline derivative 260 was also the precursor for the synthesis (252) of the naturally occumng umbelactone. Reaction of butenolide 159c with diazomethane gave the pyrazoline 260, which was subjected to pyrolysis to give (-)-(S)-umbelactone (261). As the natural umbelactone was described as being dextrorotatory, the synthesis of (+)-(R)-umbelactonefrom 159c was also performed. Cycloadditions (253) of butenolides with isoprene afforded a 1 : 1 mixture of Diels- Alder regioisomers. The selectivity is increased by the use of aluminum trichloride as catalyst. Although the butenolides studied did not react with furan, even in the presence of catalysts,they reacted smoothly with cyclopentadiene. For example, reaction of (-)-angelica lactone (159a) with

SYNTHETIC REACTIONS OF ALDONOLACTONES

189

CH3

159 c

260

261

isoprene for 20 h at - 200°C afforded an equimolecular mixture of the regioisomers262 and 263. However, in the presence of A1C1, (6 days, SOOC), the ratio of 262 and 263 was 85 : 15. Reaction of 159a with cyclopentadiene at 70- 100°Cgave the tricyclic endo (264)and ex0 (265) adducts. The endoex0 selectivity has been shown to be temperature-dependent.

159 a

264

265

2-Acyloxybutenolidesobtained by p-elimination from ribonolactone (see Section IX.2) have served as appropriate chiral intermediates in several synthesis of antibiotics. Barrett and Sheth (205) reported a seven-step synthesis of racemic tert-butyl-8-O-tert-butyldimethylsilylnonactate, a monomeric moiety of the antibiotic nonactin, from 157. Also, (4S,6S)-4dimethyl-tert-butylsilyloxy- 6 - [(dimethyl-tert-butylsilyloxy)methyl]- tetra-

ROSA M. DE LEDERKREMER AND OSCAR VARELA

190

hydro-2H-pyran-2-one(266), a suitable precursor for the spiro-acetal unit that occurs in milbemycinsand overmectins,was synthesized (254) from the 2-acetoxybutenolides 156b and 156d. CH,OR

266 R = ButMezSi

2. Synthetic Uses of SButenolides 2,3-Dideoxyhex-2-enono-1,5-lactone derivatives (penten-5-olides) have been prepared (255- 258) and employed as startingcompounds in synthesis. Thus, Michael addition of benzylhydroxylamine to racemic 6-0-acetyl2,3,4-trideoxy-~,~-g~ycero-hex-2-enono1,5-1actone (267) took place stereoselectively to give the unstable benzyloxyamino-2-pyne 268, which was readily converted into the j3-lactam derivative 269, a precursor of thienamycin (259). j3-Lactams were also obtained (260) by 1,3-dipolarcycloaddition of nitrone 270 to the unsaturated 1,5-1actone267, followed by hydrogenolysis and subsequent cyclization to the j3-lactam 271, having a polyol sidechain at the C-3 position.

bo Do CH,OAc

CH,OAc

f

l

R

NH 0

NHOCH,Ph

267

268

,C,H40Me

0,

H

269 R = SiMe,But

O

Y

~,

H

,,pc6H40Me

N=C Ph'

H ' Ph

0

270

27i

OR

SYNTHETIC REACTIONS OF ALDONOLACTONES

191

Racemic (267)and chiral(272a and 272b) lactoneshave been employedas dipolarophilesin cycloadditions with differently substituted nitrones (26 1). The reaction generally furnished two diastereomeric compounds, such as 273 and 274, from 272b. The configuration of the pyranoid ring was established on the basis of ‘H-n.m.r. data. The transformation of cycloadduct273 into the p-lactam 276 was achieved (262) through the intermediatehemiacetal 275. However, other adducts having methyl or benzyl substituents on nitrogen did not yield p-lactams, and deamination was observed instead. Reaction of such unsaturated 1,5-lactones as 272b with a mixture of hydroxylamine and formaldehyde (formaldoxime), in the nitrone form, gave an 8-aza-3,7-dioxa-2-oxo-bicyclo[4.3.0]nonane derivative via a 1,3-dipolar cycloaddition. Alternatively, a stepwise process yielded the 1-aza-3,9-dioxa-8oxo-bicyclo[4.3.0]nonane derivative (263). Dipolar cycloaddition of benzonitrile oxide to 272a gave the bicyclic products 277 and 278 in 58 and 7% yields respectively (264). The structure of 278 was determined by X-ray analysis. Deacetylation of 277 took place with contraction of the lactone ring. CH,OAc I

CH,OAc I

. R1 = OAc, RZ = H b R1 = H. R2 = OAc

272 a

273 R1 = H, R2 = C,H,OMe

Roq 274 R1 = C6H,0Me.

R2 = H

~, ,.pc6H40

0

Ph

I

Ph 275

276 R = SiMe,Eut

CH3

ROSA M. DE LEDERKREMER AND OSCAR VARELA

192

272 a

-

0

277

278

3. Other Syntheses from ~-Ribono-l&lactone D-Ribonolactoneis a convenient source of chiral cyclopentenones,acyclic structures, and oxacyclic systems, useful intermediates for the synthesis of biologically important molecules. Cyclopentenones derived from ribonolactone have been employed for the synthesis of prostanoidsand carbocyclic nucleosides. The cyclopentenone 280 was synthesized (265) from 2,3-0cyclohexylidene-D-ribono-1,4-1actone (16b) by a threestep synthesis that involves successive periodate oxidation, glycosylation of the lactol with 2propanol to give 279, and treatment of 279 with lithium dimethyl methylphosphonate. The enantiomer of 280 was prepared from D-mannose by convertingit to the correspondinglactone, which was selectivelyprotected at HO-2, HO-3 by acetalization. Likewise, the isopropylidene derivative 282 was obtained (266) via the intermediate unsaturated lactone 281, prepared from 16a. Reduction of 281 with di-tert-butoxylithium aluminum hydride, followed by mesylation, gave 282.

-Q LiCH,PO(OMe),

16 b

Me2c"oQo

0 ,

/o C6H10

0,

/o

279

280

o\ /o

o\ /o

CMe,

281

CMe,

282

SYNTHETIC REACTIONS OF ALDONOLACTONES

193

The 2-trifluoromethanesulfonatesof the four diastereomeric 3,5-di-Obenzyl-pentono-1,Clactones(such as D-ribono-1,6lactone, 283)gave, upon treatment with potassium carbonate in methanol, the ring-contraction product, methyl oxetane-2-carboxylate(284). The stereochemistryat C-2 of the resulting oxetanes is determined by the configuration at C-3, rather than C-2, of the starting lactones (267).

I

I

I

BnO

OBn

OSOzCF3

283

284

Kandil and Slessor (268) conducted the synthesis of (+)-lineatin (287), the aggregation pheromone of Trypodendron lineaturn, starting from 2,3-0isopropylidene-D-ribono-1,Clactone (16a). Grignard reaction of 16a with an excess of methylmagnesium iodide, produced a triol, which was selectively protected by acetylation, and then treatment with [Z(trimethylsilyl)ethoxylmethyl chloride (SEM chloride) and deacetylation gave the glycol 285. Periodate oxidation of 285 and condensation of the resulting aldehyde function with diethyl (3-cyano-1,l-dimethoxypropyl)phosphonate provided the a,P-unsaturated nitrile 286, from which (+)-lineatin (287) was obtained through a 10-step synthetic route (2.7% overall yield from 16a).

w::-

HOCH,

OH

o\ / o CMe,

285

~e

OSEM

o\

Me

/o

CMe,

286

287

The mycotoxin (-)-citreoviridin and the fungal metabolite (+)-citreoviral (291) have been synthesized (269) from D-ribonolactone, via the intermediate 288. Addition of methyllithium to the lactone 288 gave the corresponding lactol289a, which was converted into the methyl acetal289b. This compound was treated with trimethylallylsilane-ZnBr, to give the doubly alkylated C-glycosyl compound 290, a key intermediate for synthesis of the natural products.

ROSA M. DE LEDERKREMER AND OSCAR VARELA

194

Me

I

288

289 a

R=H

I

I

o\

/o

CMe, 290

b R=Me

J

Me

OH

291

A number of enantiomerically pure chiral building-blocks, such as 292 294, have been prepared (270,271) by zinc-copper cleavage of 5-bromo-5deoxy-2,3-O-isopropylidene-~-ribono1 ,Clactone, followed by reduction. Similarly, from the 5-iodo lactone analogue the enoic acid 295 was obtained by reaction with zinc/silver-graphite (272).

292

293

294

295

SYNTHETIC REACTIONS OF ALDONOLACTONES

195

Asymmetric synthesis of the C- 1- C-5 segment 298 of a bis-normaytasinoid was developed by Barton and coworkers (273) from the ditosyl derivative (296) of ribonolactone. Compound 296 was transformed into the dibromide, benzylated, and reduced to the 5-bromoaldose 297. Compound 298 was obtained in five steps from the intermediate 297.

297

296

298

4. Synthesis of Nucleosides, Amino Acids, and Other N-Containing Products from Aldonolactones

Aldonolactones have been employed as starting compounds for the synthesis of nucleosides. The first approaches to such syntheses were based on the reduction of aldono- 1,Clactone derivatives to the corresponding furanoses, which were then coupled to purines or pyrimidines. Thus, 2-C-methylD-ribono-I ,Clactone (274)was reduced with diisoamylborane,the resulting HO-1 group was benzoylated, converted into the glycosylchlorideby HC1in ether, and condensed with 9-chloromercuri-6-benzamidopunne. Removal of the benzoyl blocking groups gave 2’-C-methyladenosine (299). A similar procedure was employed (275) for the preparation of 9-fi-~-gulofuranosyladenine (300) from tetra-0-benzoyl-D-gulono-1,Clactone. Compound 300 and 9-a-~-lyxofuranosyladenine (301) were also synthesized (276) from 2,3 :5,6-di-O-isopropylidene-~-gulono1,Clactone (7a).

HO

OH

OH

OH

OH

HO

HCOH

I

CH,OH 299

300

301

196

ROSA M. DE LEDERKREMER AND OSCAR VARELA

Borchardt and coworkers (265) have employed the chiral cyclopentenones derived from aldonolactones for the synthesis of the analogue 302a of neplanocin A. Neplanocin A (302b) and aristeromycin (303), carbocyclic analogs of adenosine having antiviral and antitumor activities, have also been synthesized (277,278).

302 a R = H b R = CH,OH

303

Methyl oxetane-2-carboxylate derivatives (e.g., 284), obtained by ring contractionof aldonolactones,have been employed for the synthesis (279) of the nucleoside P-noroxetanocin [9-(P-~-erythro-oxetanosyl)adenine, 3041 and its a-anomer via an a-chloride obtained by a modified Hunsdiecker reaction. Displacement of chloride by adenine and debenzylation gave 304. The threo isomer of 304, P-epinoroxetanocin(305),was likewise synthesized from D-lyxono-1,4-1actone.The oxetane nucleosides display potent antiviral activity against the human immunodeficiency virus (HIV).

OH

304

Ad = Adenine

305

Such dideoxynucleosidesas CNT (306) and the potent anti-HIV drug ddC (307) have been obtained (280), respectively, from the butenolide 248 and from its saturated analogue. Thus, conjugate addition of cyanide to 248, followed by reduction of the lactone group, acetylation of HO- 1, and coupling with silylated thymine, afforded, after deprotection, compound 306.

SYNTHETIC REACTIONS OF ALDONOLACTONES

197

0

Me+-L

u I

I

k 2 0

HOCH,

0

HOCH,

CLo

0

CN

307

306

A general approach to the synthesis of polyfunctionalized amino acids from sugar lactones is exemplified by the synthesis(28 1,282) of the D-amino acid (2RY3S,4R)-3,4-dihydroxyproline (310) from 3,4-0-benzylidene-~-ribono-l,5-lactone(3). The C-2 hydroxyl group of 3 was derivatized as the trifluoromethanesulfonate to afford 308a. Nucleophilic displacement of the triflate by sodium azide took place, unexpectedly, with retention of configuration at C-2 to give the azide derivative 308b. Debenzylidenationfollowed by sulfonylation of the primary hydroxyl group gave the mesylate 309a. Hydrogenation of the azide group and treatment of the resulting amino mesyl lactone 309b with sodium hydroxide gave the proline derivative 310, whose structure was determined by X-ray crystallographyto ensure that no epimerization had occurred at C-2.

308 a

R = OSO,CF,

b R=N,

309 a

R’ = N, R2 = OS0,Me

b R1 = NH,

310

R2 = OS0,Me

Other polyhydroxylated prolines have been also synthesized (283) from 0-isopropylidene derivatives of D-gulonolactone (7a) and D-galactonolactone (8b). Thus, starting from 7a, the pyrrolidine derivatives 311a and 311b were prepared by successive reduction of the lactone, mesylation of the resulting diol, cyclization with benzylamine, and removal of one or both 0-isopropylidene groups and the N-benzyl substituent. Oxidation of the diol 31l a (after prior protection of the amino group) afforded the dihydroxypro-

ROSA M. DE LEDERKREMER AND OSCAR VARELA

198

line 312, the enantiomer of 310. 1,4-Dideoxy-l,4-imino-~-glucitol, an analogue of 31 l b, was likewise synthesized from 8b.

5 steps

7 a

311 a

312

R = CMe,

b R=H

The pyrrolidine derivative 314, a skeletal analog of the antitumor antibiotic anisomycin, was synthesized from the acetal derivative 16b. The 5-OH group of 16b was tosylated and then substituted with sodium azide. Reduction (sodium borohydride) of the lactone group afforded an open-chain derivative, which was selectively protected to give 313. Hydrogenation of the wide function, followed by ptoluenesulfonylation, led to 314 by an intramolecular nucleophilic displacement (284). TS

I

CH20CH2Ph

I

HCOTs

0,

/o

-

HCOH HCOCH,Ph I

I CH2N3

C .C .-H ";P ,h PhCH,O

OH

C6HVl

16 b

313

314

The d-lactam 108, obtained from D-gulonolactone (122), was employed for the synthesis of L-6-epicastanospermine(318a) and L- 1,6-diepicastanospermine (318b).The natural enantiomer of 318a was similarly synthesized from L-gulonolactone(285). Compound 108 was benzylated and the lactam function reduced (LiA1H4-A1Cl3)to the correspondingtertiary amine 315. Removal of the silyl protecting group, oxidation of the alcohol to the aldehyde, and addition of vinylmagnesium bromide gave the epimeric alcohols, which were protected by silylation (316). After hydroboration of the alkene, the diastereomers were separated, and on treatment with mesyl chlorideEt3N, the quaternary ammonium salts 317a and 317b were obtained. Hydrogenolysis and then acid hydrolysis of the protecting groups led to 318a and 318b, respectively.

SYNTHETIC REACTIONS OF ALDONOLACTONES

315 R'

316

= Bu'Me,Si

317 a

199

R2 = H, R 3 = OR'

b R2 = OR'. R3 = H

318 a

R2= H, R 3 = O H

b R2 = OH, R3 = H

Fleet and coworkers (20) also reported two different routes for the synthesis of the tetrahydropyrrolizidine321, starting from D-glycero-D-gulo-heptono- 1,Clactone derivative 12. One approach involves the introduction of an azido group at C-7 via cleavage of the silyl ether, sulfonylation of HO-7, and substitution by azide. Reduction of the lactone (NaBH,) gave the azidodiol319a, which was mesylated. Hydrogenation of the azidomesylate 319b,

4 steps

12

2 steps OS02Me c-- 12

6 steps

c-

CH2N3

319

a R=H b R = S0,Me

321

320

200

ROSA M. DE LEDERKREMER AND OSCAR VARELA

followed by heating in ethanol in the presence of sodium acetate, led directly to the di-0-isopropylidene derivative of the pyrrolizidine. Removal of the acetals by acid hydrolysis gave compound 321. The alternative synthesis of 321 starts by reduction of the lactone group of 12 and then mesylation to afford 320. Nitrogen was introduced by treatment of 320 with benzylamine to give a monocyclic pyrrolidine. The formation of the second ring was achievedby removal of the silyl group, followed by mesylation.The resulting mesylate was unstable and spontaneously closed to give the N-benzyl pyrrolizidinium salt, which was readily converted into 321. Similar strategies have been used for the synthesis (286)of the tetrahydropyrrolizidine 323 from 2,3 : 5,6-di-O-isopropylidene-~-glycero-~-taloheptono- 1,Clactone (322). These polyhydroxylated pyrrolidines and pyrrolizidines are potential specificinhibitors of glycosidases. The stereochemistryof the hydroxyl groups have a profound effect on the selectivity of the inhibition (286-288).

322

Enantiomerically pure 4,5,6-trihydroxy-norleucins(for instance 325) were obtained ( 197) from the hex-2-enono- 1,4-1actone-2-mesylates(such as 152).These butenolides were stereoselectivelyhydrogenated to afford, upon treatment with sodium azide, the C-2-inverted derivatives, such as 324. Reduction of the azide function and hydrolysis of the acetal group gave the amino acids (namely 325), which were converted into lactones in acid media.

N3

324

325

SYNTHETIC REACTIONS OF ALDONOLACTONES

20 1

2,3 :5,6-Di-O-cyclohexylidene-~-gulonoI ,Clactone (326) was the starting compound for a synthesis (289) of (+)-negamycin (329). Reduction of 326 with DIBAL gave the gulofuranose derivative, which was converted into the corresponding oxime. The nitrone 327, generated in situ by reaction of the oxime with methyl glyoxylate, reacted in a highly enantioselective 1,3dipolar cycloaddition with the benzyloxycarbonyl derivative of allylamine, to produce 3R, 5R-trans(328a)and 3S, 5R-cis (328b)adducts. The two new asymmetric centers created in 328a were elaborated to the 3R, 5R stereochemistry of 329.

HCO,

HCO,

H,~O/C6Hi0

H2~o/c6H10

326

327

I H, N H

-

I 4-

HCO I ‘CH

HC , O’

329

wNHCO,CH,Ph

‘

’O

328 a R’ = H, Rz = C0,Me b R’ = C0,Me.

R2 = H

REFERENCES 1. T. C. Crawford, Adv. Carbohydr. Chem. Biochem., 38 (1981) 287-321. 2. K. L. Bhat, S.-Y. Chen, and M. M. Joullit, Heterocycles, 23 (1985) 691-729. 3. A. H. Zinner, H. Voigt, and J. Voigt, Carbohydr. Res., 7 (1968) 38-55.

202

ROSA M. DE LEDERKREMER AND OSCAR VARELA

4. N. Baggett, J. G. Buchanan, M. Y. Fatah, C. H. Lachut, K. J. McCullough, and J. M. Webber, J. Chem. SOC.Chem. Commun.,(1985) 1826- 1827. 5. S. Y. Chen and M. M. Joullib, Tetrahedron Lett., 24 (1983) 5027-5030. 6. S. Y. Chen and M. M. Joullib, J. Org. Chem., 49 (1984) 2168-2174. 7. G. J. F. Chittenden and H. Regeling, Recl. Truv.Chim. Pays-Bus, 105 (1986) 186- 187. 8. P. J. Garegg, L. Maron, and C. G. Swahn, Actu Chem. Scund., 26 (1972) 5 18-522. 9. T. C. Crawford, U.S. Patent No. 4,111,958 (1978); Chem. Abstr., 90 (1979) 138150~. 10. M. Matsui, M. Saito, M. Okada, and M. Ishidate, Chem. Phurm. Bull., 16 (1968) 12941299. 11. T. C. Crawford and R. Breitenbach, J. Chem. SOC.Chem. Commun. (1979) 388-389. 12. R. K. Hulyalkarand J. K. N. Jones, Can. J. Chem., 41 (1963) 1898-1904. 13. L. M. Lemer, B. D. Kohn, andP. Kohn, J. Org. Chem., 33 (1968) 1780-1783. 14. C. Copeland and R. V. Stick, Austr. J. Chem., 31 (1978) 1371-1374. 15. J. A. J. M. Vekemans, J. Boerekamp, E. F. Godefroi, and G. Chittenden, Red. Truv.Chim Pu~s-Bus.,104 (1985) 266-272. 16. H. Ogura, H. Takahashi, and T. Itoh, J. Org. Chem., 37 (1972) 72-75. 17. K. Ohle and G. Berend, Ber., 58 (1925) 2590-2592. 18. 0. T. Schmidt, C. C. Weber-Molster, and H. Hauss, Ber., 76 (1943) 339-344. 19. J. S. Brimacombe and L. C. N. Tucker, Curbohydr.Res., 2 (1966) 341 -348. 20. W. F. Collin, G. W. J. Fleet, M. H d d s s o n , I. C. di Bello, and B. Winchester, Curbohydr. Res., 202 (1990) 105- 116. 21. G. J. F. Chittenden, Recl. Truv.Chim. Pays-Bus., 107 (1988) 455-458. 22. H. Regeling, E. De Rouville, and G. J. F. Chittenden, Recl. Truv.Chim. Puys-Bus., 106 (1987) 461 -464. 23. T. Kida, A. Masuyama, and M. Okahara, Tetrahedron Lett., 31 (1990) 5939-5942. 24. L. Hough, J. K. N. Jones and D. L. Mitchell, Can. J. Chem., 36 (1958) 1720- 1728. 25. E. C. Bigham, C. E. Gragg, W. R. Hall,J. E. Kelsey, W. R. Mallory, D. C. Richardson, C. Benedict, and P. H. Ray, J. Med. Chem., 27 (1984) 717-726. 26. B. H. Lipshutz and M. C. Morey, J. Org. Chem., 46 (1981) 2419-2423. 27. A. M. Sepulchre, A. Gateau, and S. Gero, C.R. Acud. Sci. Paris. C, 269 (1969) 13121314. 28. S. Takano, K. Inomata, and K. Ogasawara, Heterocycles, 27 (1988) 2413-2415. 29. L. M. Lerner, Curbohydr. Rex, 9 (1969) 1-4. 30. R. H. Shah, Carbohydr. Res., 155 (1986) 212-216. 31. R. H. Shah, J. Curbohydr. Chem., 5 (1986) 139-146. 32. N. Baggett, K. W. Buck, A. B. Foster, B. H. Rees, and J. M. Webber, J. Chem. SOC.C. (1966) 212-215. 33. D. L. Mitchell, Can. J. Chem., 41 (1963) 214-217. 34. C. Hubschwerlen, Synthesis (1986) 962-964. 35. T. E. Walker and H. P. C. Hogenkamp, Carbohydr. Res., 32 (1974) 413-417. 36. A. Fernindez Cirelli, M. Sznaidman, L. Jeroncic, and R. M. de Lederkremer, J. Carbohydr. Chem., 2 (1983) 167- 176. 37. L. 0. Jeroncic, M. L. Sznaidman, A. Fernandez Cirelli, and R. M. de Lederkremer, Curbohydr. Res., 191 (1989) 130-137. 38. R. M. de Lederkremer,C. Marino, and 0.Varela, Curbohydr.Res., 200 (1990) 227-235. 39. M. L. Sznaidman, A. FernPndez Cirelli, and R. M. de Lederkremer, Curbohydr.Res., 146 (1986) 233-340. 40. E. Mark and E. Zbiral, Monutsh. Chem., 112 (1981) 215-239. 4 1. K. Bock, I. Lundt, and C. Pedersen, Curbohydr. Res., 68 (1979) 3 13- 3 19. 42. K. Bock, I. Lundt, and C. Pedersen, Carbohydr. Res., 90 (1981) 7 - 16.

SYNTHETIC REACTIONS OF ALDONOLACTONES

203

43. K. Bock, I. Lundt, and C. Pedersen, Curbohydr.Res., 90 (1981) 17-26. 44. K. Bock, 1. Lundt, C. Pedersen, and R. Sonnichsen, Curbohydr.Res., 174 (1988) 331 340. 45. K. Bock, I. Lundt, and C. Pedersen, Actu Chem. Scund. B, 41 (1987) 435-441. 46. M. Bols and I. Lundt, Actu Chem. Scund B, 42 (1988) 67-74. 47. M. Bols and I. Lundt, Acut Chem. Scund., 45 (1991) 280-284. 48. M. Bols and I. Lundt, Actu Chem. Scund., 44 (1990) 252-256. 49. R. D. Guthrie and J. Honeyman, J. Chem. SOC.,(1959) 853-854. 50. Y. A. Zhdanov, Y. E. Alexeev, and C. A. Khourdanov, Curbohydr. Res., 14 (1970) 422 -423. 5 1. Y. A. Zhdanov, Y. E. Alexeev, and V. A. Tyumenev, Dokl.AkadNuuk. SSSR.,226 (1976) 1334-1336; Chem. Abstr., 85 (1976) 33285e. 52. V. H. Shrivastava and L. M. Lemer, J. Org. Chem., 44 (1979) 3368-3373. 53. R. Csuk and B. I. GEnzer, J. Curbohydr. Chem., 9 (1990) 797-807. 54. R. Csuk and B. I. Glanzer, J. Curbohydr. Chem., 9 (1990) 809-822. 55. A. Kampf and E. Dimant, Curbohydr. Res., 32 (1974) 380-382. 56. H. Ogura and H. Takahashi, J. Org. Chem., 39 (1974) 1374- 1379. 57. H. Ogura, K. Furuhata, and H. Takahashi, Nucleic Acids Res., Spec. Publ. 3 (1977) 23-26; Chem.Abstr., 89 (1978) 75429t. 58. H. Ogura, K. Furuhata, H. Takahashi, and Y. Iitaka, Chem. Phurm. Bull., 26 (1978) 2782-2787. 59. Y. A. Zhdanov, V. G. Alekseeva, E. L. Korol, and V. A. Polenov, Dokl. Akad Nuuk. SSSR., 230 (1976) 856-859; Chem.Abstr., 86 (1977) 90157j. 60. M. D. Lewis, J. K. Cha, andY. Kishi, J. Am. Chem. Soc., 104(1982) 4976-4978. 61. G. A. Krauss and M. T. Molina, J. Org. Chem., 53 (1988) 752-753. 62. S. Czemecki and G. Ville, J. Org. Chem., 54 (1989) 610-612. 63. H. Ogura, H. Takahashi, Synth. Commun., 3 (1973) 135- 143. 64. J. M. Lancelin, P. H. A. Zollo, and P. Sinay, Tetrahedron Lett., 24 (1983) 48334836. 65. D. Rouzaud and P. Sinay, J. Chem. SOC.Chem. Commun. (1983) 1353- 1354. 66. A. Boschetti, F. Nicotra, L. P a m , G. R u m , and L. Zucchelli, J. Chem. SOC.Chem. Commun. (1989) 1085-1086. 67. S. Hanessian and A. Ugolini, Curbohydr.Rex, 130 (1984) 261-269. 68. D. Horton and W. Priebe, Curbohydr.Rex, 94 (1981) 27-41. 69. J. E. Hengeveld, V. Grief, J. Tadanier, C. M. Lee, D. Riley, and P. A. Lartey, Tetrahedron Lett., 25 (1984) 4075-4078. 70. C. S. Wilcox, G. W. Long and H. Sugh, Tetrahedron Lett., 25 (1984) 395-398. 71. T. V. RajanBabu and G. S. Reddy, J. Org. Chem., 5 1 (1986) 5458-5461. 72. M. H. Ah, P. M. Collins, and W. G. Overend, Curbohydr. Res., 205 (1990) 428-434. 73. R. Csuk, and B. I. Glanzer, Tetrahedron, 47 (1991) 1655- 1664. 74. A. Takahashi, Y. Kirio, M. Sodeoka,H. Sasai, and M. Shibasaki, J. Am. Chem.Soc., 1 I 1 (1989) 643-647. 75. D. Noort, G.H. Veeneman, G-J. P. H. Boons, G. A. van der Marel, G. J. Mulder, and J. H. van Boom, Synlett (1990) 205-206. 76. Y. Chapleur, Chem. Commun. (1984) 449-450. 77. A. Bandzouzi and Y. Chapleur, Curbohydr.Res., 171 (1987) 13-24. 78. A. Bandzouzi and Y. Chapleur, J. Chem. SOC.Perkin Trans., I (1987) 661-664. 79. W. B. Motherwell, M. J. Tozer, and B. C. Ross, J. Chem. SOC.Chem. Commun. (1989) 1437-1439. 80. W. B. Motherwell, B. C. Ross, and M. J. Tozer, Synlett. (1989) 68-69.

204

ROSA M. DE LEDERKREMER AND OSCAR VARELA

81. K. Bischofberger, R. H. Hall, and A. Jordaan, Chem. Commun. (1975) 806-807. Perkin Trans., 82. R. H. Hall, K. Bischofberger, S. J. Eitelman, and A. Jordaan, J. Chem. SOC. I(1977) 743-753. 83. S. J. Eitelman, R. H. Hall, and A. Jordaan, Chem. Commun. (1976) 923-924. 84. 0. F. Hedenburg, J. Am. Chem. SOC..37 (1915) 345-372. 85. K. Rehorst, Ber., 63 (1930) 2279-2292. 86. M. L. Wolfrom and H. B. Wood, J. Am. Chem. Soc., 73 (1951) 730-733. 87. A. M. Dempsey and L. Hough, Curbohydr. Res., 16 (1971) 449-451. 88. R. M. de Lederkremer, A. Femindez Cirelli, and J. 0. Deferrari, Carbohydr. Res., 13 (1970) 9-13. 89. 0. Varela, A. Fernindez Cirelli, and R. M. de Lederkremer, Anales Asoc. Quim. Argentina, 69 (1981) 161-170. 90. I. Dijong and U. Wittkhtter, Chem. Ber., 104 (1971) 2806-2812. 91. D. E. Wright, Tetrahedron, 35 (1979) 1207-1237. 92. J. Yoshimura and M. Tamaru, Carbohydr. Res., 72 (1979) C9-Cll. 93. M. Tamaru, S. Horito, and J. Yoshimura, Bull. Chem. SOC.Jpn., 53 (1980) 3687-3690. 94. J. Yoshimura, S. Horito, and H. Hashimoto, Chem. Lett. (1982) 701 -703. 95. S. Horito, K. Asano, K. Umemura, H. Hashimoto, and J. Yoshimura, Carbohydr. Res., 121 (1983) 175-185. 96. J. Yoshimura, K. Asano, K. Umemura, S. Horito, and H. Hashimoto, Carbohydr. Res., 121 (1983) 187-204. 97. K. Asano, S. Horito, A. Saito, J. Yoshimura, and K. Ueno, Carbohydr. Rex, 136 (1985) 1-11. 98. K. Asano, S. Horito, J. Yoshimura, T. Nakazawa, Z.-I. Ohya, and T. Watanabe, Curbohydr. Res., 138 (1985) 325-328. 99. J. W. E. Glattfeld and D. MacMillan, J. Am. Chem. Soc., 56 (1934) 2481 -2482. 100. G. B. Robbins and F. W. Upson, J. Am. Chem. Soc., 60 (1938) 1788- 1789. 101. C.S.HudsonandS.Komatzu,J.Am.Chem.Soc.,41 (1919) 1141-1147. 102. J. W. W. Morgan and M. L. Wolfrom, J. Am. Chem. Soc., 78 (1956) 1897- 1899. 103. R. A. Weerman, Rec. Trav. Chim. Pays-Bas, 37 (1917) 52-66. 104. A. G. Renfrew and L. H. Cretcher, J. Am. Chem. Soc., 54 (1932) 4402-4404. 105. M. L. Wolfrom and R. B. Bennett, J. Org. Chem., 30 (1965) 458-462. 106. M. L. Wolfrom, R. B. Bennett, and J. D. Crum, J.Am. Chem. Soc., 80 (1958) 944-946. 107. Y. Nakajima and Y. Hoshi, Japanese Patent No. 2 1,240 (1964); Chem. Abstr., 62 (1965) 11,900d. 108. D. L. Tabem, U.S. Patent No. 2,084,626 (1937); Chem. Abstr., 31 (1937) 5949*. 109. M. Fieser, L. F. Fieser, E. Toromanoff, Y. Hirata, H. Heymann, M. T e a , and S. Bhattachanya, J. Am. Chem. Soc., 78 (1956) 2825-2832. 1 10. J.-H. Fuhrhop, P. Schneider,E. Boekema, and W. Helfrich,J. Am. Chem. Soc., 1 10( 1988) 2861 -2867. 111. T. J. Williams, N. R. Plessas, and I. J. Goldstein, Carbohydr. Res., 67 (1978) C1 -C3. 112. T. J. Williams, N. R. Plessas, and I. J. Goldstein, Arch. Biochem. Biophys., 195 (1979) 145- 151. 113. W. N. Emmerling and B. Pfannemiiller, Carbohydr. Res., 86 (1980) 321 -324. 114. V. Zabel, A. Miiller-Fahmow, R. Hilgenfeld, W. Saenger, B. Pfannemiiller, V. Enkelmann, and W. Welte, Chem. Phys. Lipids., 39 (1986) 313-327. 115. G. Ziegast and B. Pfannemuller, Carbohydr. Rex, 160 (1987) 185-204. 116. K. Matsumoto, S. Hashimoto, T. Uchida, T. Okamoto, and S. Otani, Bull. Chem. SOC. Jpn., 62 (1989) 3138-3142. 117. N. K. Richtmyer, Adv. Carbohydr. Chem., 6 (1951) 175-203.

SYNTHETIC REACTIONS OF ALDONOLACTONES

205

118. R. Hull, J. Chem. SOC.(1958) 4069-4073. 119. H. S. El Khadem and R. Sindric, Curbohydr. Res., 34 (1974) 203-207. 120. A. C. M. Barrett and D. Dhanak, Tetrahedron Lett., 28 (1987) 3327-3330. 121. A. G. M. Barrett, B. C. B. Bezuidenhoudt, D. Dhanak, A. F. Gasiecki, A. R. Howell, A. C. Lee, and M. A. Russell, J. Org. Chem., 54 (1989) 3321 -3324. 122. G. W. J. Fleet, N. G. Ramsden, and R. Witty, Tetrahedron Lett., 29 (1988) 2871 -2874. 123. T. K. M. Shing, J. Chem. SOC.Chem. Commun. (1988) 1221. 124. B. Rajanikanth and R. Seshadri, Tetrahedron Lett., 30 (1989) 755-758. 125. G. W. J. Fleet, N. G. Ramsden, N. M. Carpenter, S. Petursson, and R. T. Aplin, Tetruhedron Lett., 3 1 ( 1990) 405 -408. 126. G. W. J. Fleet, N. M. Carpenter, S. Petursson, and N. G. Ramsden, Tetrahedron Lett., 3 1 (1990) 409-412. 127. T. W. J. van Marle, Rec. Truv. Chim. Puys-Bus., 39 (1920) 562-565. 128. A. Thompson and M. L. Wolfrom, J. Am. Chem. SOC.,68 (1946) 1509- 1510. 129. L. Maquenne, Bull. SOC.Chim. Paris, 48 (1887) 719-723. 130. E. Fischer and F. Passmore, Ber., 22 (1889) 2728-2736. 131. H. H. Stroh and D. Henning, Chem. Ber., 100 (1967) 388-394. Res., 91 (1981)39-48. 132. S.M. Sharaf,M.A.E.Sallam,andH.A.ElShenawy,Curbohydr. 133. M. A. M. Nassr, M. A. M. Taha, and M. A. E. Shaban, Curbohydr. Res., 121 (1983) 119-124. 134. M. A. E. Shaban, M. A. M. Nassr, and M. A. M. Taha, Curbohydr. Res., 113 (1983) C16-C17. 135. D. Beer and A. Vasella, Helv. Chim. Actu, 68 (1985) 2254-2274. 136. D. Beer and A. Vasella, Helv. Chim. Actu, 69 (1986) 267-270. 137. E. Fischer, Ber., 22 (1889) 2204-2205; 23 (1890) 930-938. 137a. R. U. Lemieux and U. Spohr, Adv. Curbohydr. Chem. Biochem., 50 (1994) 1-20. 138. E. Fischer and 0. Piloty, Ber., 24 (1891) 521-528. 139. C. Niemann, S. Kajala, and K. P. Link, J. Biol. Chem., 104 (1934) 189- 194. 140. M. Sutter and T. Reichstein, Helv. Chim. Actu., 21 (1938) 1210- 1218. 141. L. A. Mai, U.S.S.R. Patent No. 102,627 (1956); Chem. Abstr., 52 (1958) 4682i. 142. N. Sperber, H. E. Zaugg, and W. M. Sandstrom, J.Am. Chem. Soc., 69 (1947) 9 15-920. 143. H. S. Isbell, J. V. Karabinos, H. L. Frush, N. B. Holt, A. Schwebel, andT. T. Galkowski,J. Res. Nutl. Bur. Stand., 48 (1952) 163-171. 144. J. W. E. GlattfeldandG. W. Schimpff,J. Am. Chem. Soc., 57 (1935) 2204-2208. 145. M. L. Wolfrom and H. B. Wood, J. Am. Chem. SOC.,73 (1951) 2933-2934. 146. H. L. Frush and H. S. Isbell, J. Am. Chem. Soc., 78 (1956) 2844-2846. 147. M. L. Wolfrom and A. Thompson, Methods Curbohydr. Chem., 2 (1963) 65-68. 148. L. M. Lerner, Curbohydr. Res., 44 (1975) 116- 120. 149. J. Nemek, and J. Jary, Collect. Czech. Chem. Commun., 33 (1968) 843-848. 150. S. S. Bhattacharjee,J. A. Schwarcz,and A. S. Perlin, Curbohydr.Hes., 42 (1975)259-266. 151. R. Knollmann and I. Dyong, Chem. Ber., 108 (1975) 2021-2029. 152. I. Dijong and N. Jersch, Chem. Ber., 109 (1976) 896-900. 153. I. Dijong and D. Glittenberg, Chem. Ber., 110 (1977) 2721 -2728. 154. C. Fuganti and P. Grasselli, J. Chem. SOC.Chem. Commun., 7 (1978) 299-300. 155. H. C. Brown and D. B. Bigley, J. Am. Chem. SOC.,83 (1961) 486. 156. P. Kohn, R.H. Samaritano, andL. M. Lerner,J. Am. Chem. SOC.,86 (1964) 1457- 1458. 157. P. Kohn, R.H. Samaritano, andL. M. Lerner, J.Am. Chem. SOC.,87 (1965)5475-5480. 158. 0. Varela, A. Fernandez Cirelli, and R. M. de Lederkremer, Curbohydr.Rex, 70 (1979) 27-35. 159. R.Knollmann, N. Jersch, and I. Dijong. Chem. Ber., 110 (1977) 2729-2743.

206

ROSA M. DE LEDERKREMER AND OSCAR VARELA

160. I. Dijong, L. Baumeister, and H. Bendlin, Chem. Ber., 112 (1979) 161- 174. 161. 0. Varela and R. M. de Lederkremer, unpublished results. 162. 0.Varela, C. Marino, and R. M. de Lederkremer, Curbohydr.Res., 155(1986)247-251. 163. T. A. Giudici and A. L. Fluharty, J. Org. Chem., 32 (1967) 2043-2045. 164. G . Nakaminami, M. Nakagawa, S. Shioi, Y. Sugiyama, S. Isemura, and M. Shibuya, Tetrahedron.Lett. (1967) 3983-3987. 165. G. Nakaminami, S. Shioi, Y. Sugiyama, S. Isemura, M. Shibuya, and M. Nakagawa, BUN. Chem. SOC.Jpn., 45 (1972) 2624-2634. 166. P. Kohn, L. M. Lemer, A. Chan, Jr., S. D. Ginocchio, andC. A. Zitnn, Curbohydr.Res., 7 (1968) 21-26. 167. J. B. Lee and T. J. Nolan, Tetrahedron, 23 (1967) 2789-2794. 168. V. G. Mairanovskii, J. Swiderski, and A. M. Ponomarev, Kim.-Furm. Zh., 13 (1979) 62-65; Chem. Abstr., 91 (1979) 5425f. 169. A. Korczynski, L. Piszczek, J. Swiderski, and A. Doniec, Zesz. Nuuk. Politech. Slusk. Chem., 631 (1979) 85-96; Chem. Abstr. 93 (1980) 114855. 170. M. L. Sinnott and I. J. L. Souchard, Biochem. J., 133 (1973) 89-98. 171. L. Hosie and M. L. Sinnott, Biochem. J., 226 (1985) 437-446. 172. L. A. Berven and S. G. Wethers, Curbohydr. Res., 156 (1986) 282-285. 173. W. N. Haworth and C. W. Long, J. Chem. SOC.(1929) 345-350. 174. H. S. Isbell, J. Res. Nutl. Bur. Stand., 32 (1944) 45-59. 175. A. K. Ganguly, 0. Z. Sarre, and H. Reimann, J. Am. Chem. Soc., 90(1968) 7129-7130. 176. A. K. Ganguly and 0. Z. Sarre, J. Chem. Soc. Chem. Commun. (1969) 1149- 1150. Chem. Commun. (1973) 531-532. 177. A. K. Ganguly and A. K. Saksena, J. Chem. SOC. 178. I. Dijong and U. Wittkoetter, Tetrahedron Lett. (1969) 3793-3795. 179. H. Kuzuhara and S. Emoto, Agric. Biol. Chem. (Tokyo), 26 (1962) 334-340; Chem. Abstr., 58 (1963) 4638h. 180. G. Deak, K. Gall-Istok, and P. Sohar, Acfu Chim. Acud. Sci. Hung., 75 (1973) 189- 191; Chem. Abstr.. 78 (1973) 97934. 181. D. Horton, J. K. Thomson, 0. Varela, A. Nin, and R. M. de Lederkremer, Curbohydr. Res., 193 (1989) 49-60. 182. R. M. de Lederkremer, M. I. Litter, and L. F. Sala, Curbohydr.Res., 36 (1974) 185- 187. 183. G.M.Cree,D.M.Mackie,andA.S.Perlin,Cun. J. Chem.,47(1969)511-512. 184. D. M. Mackie and A. S. Perlin, Curbohydr. Res., 24 (1972) 67-85. 185. Y. Tsuda, T. Nunozawa, and K. Yoshimoto, Chem. Phurm. Bull (Tokyo), 28 (1980) 3223-3231; Chem. Abstr., 94 (1981) 19259511. 186. N. PravdiE and H. G. Fletcher, Jr., Curbohydr. Res., 19 (1971) 353-364. 187. M. Fetizon and M. Golfier, Compt. Rend. C, 267 (1968) 900-903. 188. N. PravdiE, B. Danilov, andH. G. Fletcher, Jr., Curbohydr. Res., 36 (1974) 167-180. 189. M. Pokomy, E. Zissis, H. G. Fletcher, Jr., and N. PravdiE, Curbohydr. Res., 43 (1975) 345 - 354. 190. C. R. Nelson and J. S. Gratzl, Curbohydr.Res., 60 (1978) 267-273. 191. 0. Varela, A. Fernandez Cirelli, and R. M. de Lederkremer, Curbohydr. Res., 79 (1980) 2 19-224. 192. A. Nin, 0. Varela and R. M. de Lederkremer, Synthesis (1991) 73-74. 193. R. M. de Lederkremer and M. I. Litter, Curbohydr. Res., 20 (1971) 442-444. 194. C. Marino, 0.Varela, and R. M. de Lederkremer, Curbohydr.Res., 220 (1991) 145- 153. 195. R. M. de Lederkremer, C. Du Mortier, and 0. Varela, Rev.Latinoam. Quim., 17 (1986) 16- 19. 196. M. L. Sznaidman, A. Femandez Cirelli, and R. M. de Lederkremer, Curbohydr.Res., 146 (1986) 233-240.

SYNTHETIC REACTIONS OF ALDONOLACT'ONES

207

197. J.A. J. M. Vekemans,R.G.M.deBruyn,R.C.H.M.Caris,A. J.P.M.Kokx,J. J.H.G. Konings, E. F. Godefroi, and G. J. F. Chittenden, J. Org. Chem., 52 (1987) 1093- 1099. 198. M. I. Litter and R. M. de Lederkremer, Curbohydr. Rex, 26 (1973) 431 -434. 199. M. I. Litter and R. M. de Lederkremer, Andes Asoc. Quim.Argentina, 62 (1974) 147150. 200. L. 0. Jeroncic, 0. J. Varela, A. Fernandez Cirelli, and R. M. de Lederkremer, Tetruhedron, 40 (1984) 1425- 1430. 20 1. L. 0. Jeroncic, Doctorate Thesis, University of Buenos Aires, 1988. 202. 0.Varela, A. Fernandez Cirelli, and R. M. de Lederkremer, Curbohydr. Res., 100 (1982) 424-430. 203. W. Timpe, K. Dax,N. Wolf, and H. Weidmann, Curbohydr. Res., 39 (1975) 53-60. 204. A. G. M. Barrett and H. G. Sheth. J. Org. Chem., 48 (1983) 5017-5022. 205. A. G. M. Barrett and H. G. Sheth, Chem. Commun. (1982) 170- 171. 206. H. Komura, T. Iwashita, H. Naoki, and K. Nakanishi, J. Am. Chem. Soc., 105 (1983) 5 164- 5 165. 207. R. M. Ortuiio, D. Alonso, and 5. Font, Tetrahedron Lett., 27 (1986) 1079- 1082. 208. G. A. Danilova, V. I. Mel'nikova, and K. K. Pivnitsky, Tetrahedron Lett., 27 (1986) 2489-2492. 209. K. Tadano, Y. Iimura, T. Ohmori, Y. Ueno, and T. Suami,J. Curbohydr. Chem., 5 (1986) 423 -435. 210. C. Pedersen, K. Bock, and I. Lundt, PureAppl. Chem., 50 (1978) 1385-1400. 2 1 1. J. A. J. M. Vekemans, G. A. M. Franken, C. W. M. Dapperens, E. F. Godefroi, and G. J. F. Chittenden, J. Org. Chem., 53 (1988) 627-633. 212. C. T. Clarke, J. H. Jones, and R. Walker. J. Chem. SOC.Perkin Trans., Z (1976) 10011003. 213. M. Bergmann, L. Zervas, and E. Silberkweit, Ber., 64 (1931) 2428-2436. 214. C. Du Mortier and R. M. de Lederkremer, J. Carbohydr. Chem., 3 (1984) 219-228. 215. R. M. de Lederkremer and L. F. Sala, Curbohydr. Res., 40 (1975) 385-386. 216. L. F. Sala, A. Fernandez Cielli, and R. M. de Lederkremer, Curbohydr. Res., 78 (1980) 61 -66. 2 17. L. F. m a , A. Fernlndez Cirelli, and R. M. de Lederkremer,J. Chem. Soc. Perkin Trans., ZI(1977) 685-688. 2 18. L. F. Sala, A. Fernindez Cirelli, and R. M. de Lederkremer, Andes Asoc. Quim.Argentina,66 (1978) 57-63. 2 19. R. N. Russell, T. W. Weigel, 0. Han, and H. Liu, Curbohydr. Res., 201 ( 1990) 95 - 114. 220. 0. Varela, A. F e d n d e z Cirelli, and R. M. de Lederkremer, Curbohydr. Res., 85 (1980) 130-135. 22 1. C. Du Mortier and R. M. de Lederkremer, Curbohydr. Rex, 139 ( 1985) 47- 54. 222. M. Sznaidman, A. Fernindez Cirelli, and R. M. de Lederkremer, Andes Asoc. Quim. Argent., 70 (1982) 341-348. 223. K. Bock, I. Lundt, and C. Pedersen, Actu Chem. Scund. B, 35 (1981) 155- 162. 224. L. 0. Jeroncic, A. Fernlndez Cirelli, and R. M. de Lederkremer. Curbohydr. Rex, 167 (1987) 175-186. 225. 0.Moradei, C. Du Mortier, 0.Varela, and R. M. de Lederkremer,J. Curbohydr. Chem., 10 (1991) 469-479. 226. K. Bock, I. Lundt, and C. Pedersen, Actu. Chem. Scund. B, 38 (1984) 555-561. 227. I. Lundt and C. Pedersen, Synthesis (1986) 1052- 1054. 228. K. Bock, I. Lundt, andC. Pedersen, Actu. Chem. ScundB, 40 (1986) 163-171. 229. K. Bock, I. Lundt, and C. Pedersen, Curbohydr. Rex, 179 (1988) 87-96. 230. C. Du Mortier, 0.Varela, and R. M. de Lederkremer, Curbohydr.Res., 189(1989)79- 86.

208

ROSA M. DE LEDERKREMER AND OSCAR VARELA

231. C. Marino, 0. Varela, and R. M. de Lederkremer, Carbohydr. Res., 190 (1989) 65-76. 232. S. Notermans, G. H. Veeneman, C. W. E. M. van Zuylen, P. Hoogerhout, and J. H. van Boom, Mol. Imrnunol., 25 (1988) 975-979. 233. S. Hanessian Total synthesis of natural products: The chiron approach. Pergamon, 1983. 234. P. Camps, J. Font, and 0. Ponsati, Tetrahedron Lett., 22 (1981) 1471-1472. 235. J. Cardellach, C. Estopa, J. Font, M. Moreno-Mafias,R. M. Ortuiio, F. Sinchez-Ferrando, S. Valle, and L. Vilamajo. Tetrahedron, 38 (1982) 2377-2394. 236. P. Camps, J. Cardellach,J. Font, R. M. Ortuiio, and 0. Ponsati, Tetrahedron, 38 (1982) 2395-2402. 237. R. E. Ireland, R. G. Anderson, R. Badoud, B. J. Fitzsimmons, G. J. McGarvey, S. Thaisrivongs, and C. S. Wilcox, J. Am. Chem. SOC.,105 (1983) 1988-2006. 238. P. Camps, J. Cardellach, J. Corbera, J. Font, R. M. Ortufio, and 0.Ponsati, Tetrahedron Lett., 39 (1983) 395-400. 239. J. Cardelach, J. Font, and R. M. Ortuiio, J. Heterocyclic Chem., 21 (1984) 327-331. 240. R. M. Ortuiio, D. Alonso, J. Cardellach,and J. Font, Tetrahedron,43 (1987)2192-2198. 241. J. Cardellach, J. Font, and R. M. Ortu60, Tetrahedron Lett., 26 (1985) 2815-2819. 242. P. Magnus and D. P. Becker, J. Am Chem. Soc., 109 (1987) 7495-7498. 243. R. M. Ortuiio, R. Merck and J. Font, Tetrahedron Lett., 27 (1986) 2519-2522. 244. R. W. Hoffmann and W. Ladner, Chem. Ber., 116 (1983) 1631- 1642. 245. R. M. Ortuiio, R. Merck, and J. Font, Tetrahedron, 43 (1987) 4497-4506. 246. S. Hanessian, Aldrichim Acta, 22 ( 1989) 3 - 15. 247. S. Hanessian and P. Murray, Can. J. Chem., 64 (1986) 2231 -2234. 248. S. Hanessian, S. P. Sahoo, and M. Botta, Tetrahedron Lett., 28 (1987) 1147- 1150. 249. S. Hanessian and M. Botta, Tetrahedron Lett., 28 (1987) 1 151 - 1 154. 250. J. Mann and A. Thomas, Chem. Commun. (1985) 737-738. 251. M. G. B. Drew, J. Mann, and A. Thomas, J. Chem. SOC.Perkin Trans., I(1986) 22792285. 252. R. M. Ortuiio, J. Bigorra, and J. Font, Tetrahedron, 43 (1987) 2199-2202. 253. R. M. Ortufio, R. Batllori, M. Ballesteros, M. Monsalvatje, J. Corbera, F. SinchezFerrando, and J. Font, Tetrahderon Lett., 28 (1987) 3405-3408. 254. S. V. Attwood and A. G. M. Barren, J. Chem. SOC.Perkin Trans., I (1984) 1315- 1322. 255. J. Mieczkowski, J. Jurczak, M. Chmielewski, and A. Zamojski, Carbohydr. Res., 56 (1977) 180-182. 256. P. Rollin and P. Sinay, Carbohydr. Res., 98 (1981) 139- 142. 257. P. J. Card, Carbohydr. Rex, 104 (1982) 338-340. 258. P. Jarglis and F. W. Lichtenthaler, Tetrahedron Lett., 23 (1982) 3781 -3784. 259. M. Chmielewski and S. Maciejewski, Carbohydr. Res., 157 (1986) Cl -C3. 260. I. Panfil, M. Chmielewski, and C. Belzecki, Heterocycles, 24 (1986) 1609- 1617. 261. I. Panil and M. Chmielewski, Tetrahedron, 41 (1985) 4713-4716. 262. I. Panfil, C. Belzecki, and M. Chmielewski, J. Carbohydr. Chem., 6 (1987) 463-470. 263. I. Panfil, C. Belzecki, M. Chmielewski, and K. Suwinska, Tetrahedron, 45 (1989) 233238. 264. H. Gnichtel, L. Autenrieth-Ansoge, and J. Dachman, J. Carbohydr. Chem., 6 (1987) 673-683. 265. D. R. Borcherding, S. A. Scholtz, and R. T. Borchardt, J. Org. Chem., 52 (1987) 5457546 1. 266. P. Belanger and P. Prasit, Tetrahedron Lett., 29 (1988) 5521 -5524. 267. D. R. Witty, G. W. J. Fleet, K. Vogt, F. X.Wilson,Y. Wang, R. Storer, P. L. Myers, andC. J. Wallis, Tetrahedron Lett.. 3 1 (1990) 4787-4790. 268. A. A. Kandil and K. N. Slessor, J. Org. Chem., 50 (1985) 5649-5655.

SYNTHETIC REACTIONS OF ALDONOLACTONES

209

H. Suh and C. Wilcox, J. Am. Chem. SOC.,110 (1988) 470-481. B. Hafele, D. Schroter, and V. JQer, Angew. Chem. Znt. Ed. Engl., 25 (1986) 87-89. V. Jager and B. HSele, Synthesis(1987) 801-806. A. Furstner and H. Weidmann, J. Org. Chem., 54 (1989) 2307-231 1. D. H. R. Barton, M. Btnkchie, F. Khuong-Huu, P. Potier, andV. Reyna-Pinedo, Tetrahedron Lett., 23 (1982) 651 -654. 274. E. Walton, S. R. Jenkins, R. F. Nutt, M. Zimmerman, and F. W. Holly, J. Am. Chem. SOC.,88 (1966) 4524-4525. 275. P. Kohn, R. H. Samaritano, and L. M. Lerner, J. Org. Chem., 31 (1966) 1503- 1506. 276. L. M. Lerner, B. D. Kohn, and P. Kohn, J. Org. Chem., 33 (1968) 1780-1783. 277. M. S. Wolfe, D. R. Borcherding, and R. T. Borchardt, Tetrahedron Lett., 30 (1989) 1453- 1456. 278. M. S. Wolfe, B. L. Anderson, D. R. Borcherding, and R. T. Borchardt, J. Org. Chem., 55 (1990) 4712-4717. 279. F. X. Wilson, G. W. J. Fleet, D. R. Witty, K. Vogt, Y. Wang, R. Storer, P. L. Myers, and C. J. Wallis, Tetrahedron: Asymmetry, 1 (1990) 525-526. 280. M. Okabe, R-C Sun,S. Y. -K. Tam, L. J. Todaro, and D. L. Coffen. J. Org. Chem., 53 (1988) 4780-4786. 28 1. J. C. Dho, G. W. J. Fleet, J. M. Peach, K. Prout, and P. W. Smith, Tetrahedron Lett., 27 (1986) 5203-5206. 282. P. D. Baird, J. C. Dho, G. W. J. Fleet, J. M. Peach, K. Prout, and P. W. Smith, J. Chem. SOC.Perkin, Trans. Z(1987) 1785-1791. 283. G. W. J. Fleet and J. C. Son, Tetrahedron, 44 (1988) 2637-2647. 284. A. M. Sepulchre, A. Gateau, A. Gaudemer, and S. D. Gero, J. Chem. SOC.Chem. Commum. (1970) 759-760. 285. G. W. J. Fleet, N. G. Ramsden, R. J. Molyneux, and G. S. Jacob, Tetrahedron Lett., 29 (1988) 3603-3606. 286. A. J. Fairbanks, G. W. J. Fleet, A, H. Jones, I. Bruce, S.A. Dahler, I. Cenci di Bello, and B. Winchester, Tetrahedron, 47 ( 1991) 131- 138. 287. S.A. Dahler, G. W. J. Fleet, S. K. Namgoong, and B. Winchester, Biochem. J., 258 (1989) 6 13-6 15. 288. I. Cenci di Bello, G. W. J. Fleet, S. K. Namgoong, K. I. Tadano, and B. Winchester, Biochem. J., 259 (1989) 855-861. 289. H. Iida, K. Kasahara,and C. Kibayashi, J. Am. Chem. Soc., 108 (1986) 4647-4648.

269. 270. 271. 272. 273.

This Page Intentionally Left Blank

ADVANCES IN CARBOHYDRATE CHEMISTRY AND BIOCHEMISTRY, VOL. 50

MOLECULAR STRUCTURE OF LIPID A, THE ENDOTOXIC CENTER OF BACTERIAL LIPOPOLYSACCHARIDES' BY ULRICH ZAHRINGER, BUKOLINDNER, AND ERNSTTH. RIETSCHEL Department of Immunochemistry and Biochemical Microbiology, Forschungsinstitut Borstel, Institut fur Experimentelle Biologie und Medizin, 0-23845 Borstel, Germany

I. INTRODUCTION Gram-negative bacteria such as the Enterobacteriaceae, Neisseriaceae, and Pseudomonadaceaeexpress at their outer membrane amphiphilic macromolecules termed lipopolysaccharides (LPS, endotoxin), which are of considerable microbiological, genetical, chemical, pharmacological, and, in particular, biomedical interest ( 1 - 9b). LPS participatein various physiological membrane functionsessential for growth and survival of Gram-negative bacteria. LPS also constitute the heat-stable 0-antigens of the bacteria and, thus, serve to identify the multiplicity of 0-serotypes. Finally, LPS are endowed with an overwhelming spectrum of biological activities, expressed either in vitro or after injection into experimental animals. LPS constitute potent immunostimulators and adjuvants (lo), but have also been recognized as playing a key role in the pathogenesis and manifestationsof Gramnegative infection in general and of septic shock in particular. Thus, LPS are among the most potent agents capable of inducing local or generalized inflammatory reactions in both humans and experimental animals. In the past, much emphasis has been placed on identifying structural principles of LPS and on elucidatingtheir chemical structureswith the aim, among other reasons, of defining the substructures that determine the various biological effects of LPS. This objective has been achieved for a number of such activities. These results, in turn, have stimulated the chemical synthesis of biologically active LPS substructures and chemical analogues (1 l - 19). At the same time, the mechanisms of biological action of endotoxin are being studied at the molecular level, providing a rational basis I This article is dedicated to Professor Dr. Dr.med. h.c. Otto Westphal on the occasion of his 80th birthday (February lst, 1993).

21 1

Copyright 0 1994 by Academic press Inc. AU rightsof reproduction in any form reserved.

ULRICH ZAHRINGER et al.

212

for strategies aimed at the immunological or pharmacological prevention and treatment of endotoxicosis (20,2 1). Structurally, LPS is comprised of three genetically, chemically and antigenically distinct regions, namely the 0-specificchain, the core oligosaccharide, and lipid A (Fig. 1). The latter serves to anchor LPS to the bacterial membrane (3). The 0-specijk chain is a heteropolymeric polysaccharide and responsible for the 0-antigenic properties of LPS, which denote the induction and binding of species-specificantibodies. It is composed of up tg 60 repeating units that, themselves, are made up of two to six glycose residues. The 0-specific chain is characteristic for a bacterial species and shows, on comparison of different Gram-negativebacteria, an enormous structural variability. In a given LPS preparation it exhibits heterogeneity because of the presence of various molecules differingin the number of repeating units. In the LPS of some nonenterobacterial pathogens, such as Neisseria and Haemophilus, the 0-specific chain is absent (22). The core oligosaccharide (23) may be divided into two regions, namely the outer core and the inner core. The outer core region of enterobacterial LPS, contains such neutral sugars as D-galactose (Galp), D-glucose (Glcp), 2-acetamido-2-deoxy-~-glucose (GlcpNAc), and 2-acetamido-2-deoxy-~-galactose(GalpNAc) and is therefore also called the hexose region. The inner core is mainly composed of the rare glycose residues L-glycero-D-manno-heptose (L,D-Hep) and 3deoxy-D-manno-octulosonic acid (“2-keto-3-deoxy-~-manno-octonic acid,” Kdo), which often carry phosphate (6,23) or 2-aminoethyl(pyro)phosphate [Etn-(P)P] (24). The structure of the core-oligosaccharide of some enterobacterial LPS [Salmonella minnesota, Escherichia coli, Proteus mirabilis, Citrobacterfreundii,and other bacterial genera (23)] has been elucidated and shown to be of moderate interbacterial variability. Lipid A constitutes the covalently bound lipid component and the least variable component of LPS (25). It anchors LPS to the bacterial cell by hydrophobic and electrostaticforces and mediates or contributesto many of the functions and activities that LPS exerts in prokaryotic and eukaryotic organisms. In the following sections, the primary structure of lipid A of different Gram-negative bacteria is described, together with some of its characteristicbiological properties. Furthermore, this article describes some of the principal methods that have been used for the structural analysis of lipid A and discusses their merits and limitations.

0-Speci f ic Chain

Outer Core

Inner Core

- Lipid

A

FIG. 1. -Schematic architecture of an enterobacterial lipopolysaccharide.

MOLECULAR STRUCTURE OF LIPID A

213

11. LIPIDA: DEFINITION AND GENERAL PROPERTIES Lipid A may be obtained by subjecting an LPS preparation to mild acid hydrolysis, by which treatment the ketosidic linkage between the inner core oligosaccharide and lipid A, namely, between Kdo and the distal D - ~ ~ U C O S ~ mine residue [GlcN (11)] of the lipid A backbone, is cleaved. Acid hydrolysis must be used, because enzymes that split the ketosidic bond, or mutants that synthesize oligosaccharide-freelipid A are known only for a few cases (26). The acid lability of the linkage between lipid A and the polysaccharide portion was first detected by Boivin and coworkers (27) in 1933, but the term lipid A was introduced only in 1954 by Westphal and Luderitz (28) to describe the hydrophobic and chloroform-solubleprecipitate obtained after acid treatment of LPS and to distinguish it from another lipid (lipid B), investigated earlier by Morgan and Partridge (29). This lipid B turned out to be a noncovalently bound phospholipid that could be extracted into the organic layer from the LPS complex using ether or petroleum ether (28,30). Later, the term lipid A was also used for the designation of the bound lipid component as it is present in LPS. In order to differentiate between bound and isolated lipid A, it was proposed to restrict the term lipid A to the lipid component as it is present in LPS, and to use the term free lipid A for polysaccharide-deprived (or isolated) lipid A (2,4). Neither term, however, describesa homogeneous material having a chemically defined structure. As it is now known, lipid A harbors a certain structural heterogeneity that is either due to incomplete biosynthesis or generated by chemical degradation during the preparation of free lipid A by acid hydrolysis (3 1). Therefore, the term lipid A designates, rather, a class of structurally related molecules that are constructed according to a common architectural principle. Despite the enormous progress made in the analytical and synthetic chemistry of lipid A, the complete structural elucidation of a lipid A preparation still represents a challenge for analytical chemists. The reason for the difficulties encountered in analysis of lipid A are due, in part, to its amphiphilic nature, i.e., the presence of hydrophilic and hydrophobic regions favoring the formation of aggregates in either organic nonpolar or aqueous solution. This factor has, in the past, greatly limited the application of such advanced chromatographic purification procedures as adsorption-, ion-exchange-, or reverse-phase high-performance liquid chromatography (h.p.1.c.). Moreover, molecular aggregates of lipid A and LPS preparations also limit the application of modem proton (lH), carbon ( 13C), and phosphorus (31P)nuclear magnetic resonance (n.m.r.) spectroscopy. Further, the amphoteric character of lipid A, resulting from the simultaneouspresence of negatively and positively charged groups as well as its compositional and structural microheterogeneity,has added to the problems associated with the chemical and physical analysis of lipid A.

214

ULRICH ZAHRINGER et al.

Nevertheless, in the past decade, considerable progress has been made in this field. The most successful approach toward the isolation of homogeneous lipid A and derivatives was the application of controlled degradation reactions and the introduction of specific protecting groups in combination with the development of various h.p.1.c. purification procedures (32 - 35). In addition, improvements in soft desorption ionization techniques, such as laser-desorption mass spectrometry (1.d. -m.s.) (36- 40), fast-atom bombardment mass spectrometry(f.a.b. - m.s.) (32,33,4l), and ZszCf-plasmadesorption mass spectrometry(42 -44b), which have been found suitable also for nonderivatized macromolecules, have deepened our knowledge of the lipid A structure. Moreover, nondestructive IH-, I3C-,and 31P-n.m.r.spectroscopy contributed tremendously to the progress made in the structural elucidation of various types of lipid A. Modifications of derivatization reactions for methylation analysis, in combination with coupled gas - liquid chromatography- mass spectrometry (g.1.c. - m.s.) or h.p.1.c. - m.s. have been of great help. Finally, the recent availability of various synthetic lipid A reference compounds and partial structures thereof has allowed, in most cases, an unequivocal interpretation of the results obtained with these empirical analytical procedures. OF LIPIDA : BACKBONE, POLAR 111. PRIMARY STRUCTURE SUBSTITUENTS, AND FATTYACIDS

Comparison of endotoxic lipid A of different Gram-negative genera and families reveals that they not only share chemical constituentsbut that they are also made up according to a similar architectural principle, which is exemplified by the well-studied lipid A of the Escherichia coli Re mutant strain F5 15 (Fig. 2, for literature compare Ref. 25). This E. coli lipid A is composed of a 2-amino-2-deoxy-6-0-(2-amino-2-deoxy-~-~-glucopyranosy1)-a-D-glucopyranose disaccharide [p-~-GlcpN-(1 + 6)-a-~-GlcpN], which carries two phosphate groups: one in position 4’ [of the distal GlcpN residue, GlcN(II)] and one in the a-glycosylic position 1 [of the reducing GlcpN residue, GlcN(I)]. Of the hydroxyl groups present in this hydrophilic lipid A backbone, those in positions 4 and 6’ are not substituted. The latter primary hydroxyl group is only unsubstituted in free lipid A, because in LPS it represents the attachment site of Kdo, that is, the polysaccharide component (compare Fig. 2). Positions 3’, 2’, 3, and 2 of the backbone carry a total of four (R)-3-hydroxytetradecanoicacid residues [ 14:0(3-OH)].As these are directly linked to the GlcpN disaccharide, they are termed primary acyl groups (45). The hydroxyl groups of the two 14:O(3-OH) residues bound to GlcN(I1) are acylated by secondary fatty acids, that at position 2 by dodecanoic acid ( 12 :0) and that at position 3’ by tetradecanoic acid ( 14:0). The

MOLECULAR STRUCTURE OF LIPID A

215

0 0, II o,

’

P

HO

FIG.2.-Chemical structure of lipid A of the Escherichiu coli Re mutant strain F5 15. The hydroxyl group at position 6’ constitutes the attachment site of Kdo. The numbers in circles indicate the number of carbon atoms present in the fatty acyl chains. The 14:0(3-OH)residues possess the (R)-configuration.The glycosylic phosphate group may be substituted by a phosphate group (see Table I) (46,65,69).

glycosylicphosphate at C- 1 may carry a further phosphate group in nonstoichiometric amounts, as is the case in E. coli K-12 (46). In addition to the main structure shown in Fig. 2, molecules having a smaller number of acyl groups are also present (47,48). The same structure as that of E. coli lipid A has been identified in S. typhimurium (32). The lipid A of E. coli was the first of a series of lipid A preparations that have also been obtained by total chemical synthesis (1 4,15). The structural components encountered in E. coli are also present in lipid A of other bacterial sources. Thus, a survey of the structures analyzed shows that lipid A, in general, contain two gluco-configured and pyranosidic Dhexosamine residues (2-amino-2-deoxy-~-~ucose, GlcpN, or 2,3-diamino2,3-dideoxy-~-ghcose,GlcpN3N, also termed “DAG’ (49,50)], which are present as ap-( 1 + 6)-linkeddisaccharide (monosaccharidebackboneshave also been identified, but the respective lipid A lack endotoxic activity). The disaccharide is phosphorylated by one or, in most cases, two phosphate

216

ULRICH ZAHRINGER et al.

residues, one in a-glycosyliclinkage at position 1 and a second ester-bound to position 4’of HexN(11). Both phosphate groups may carry substituents that often contain (positively charged) amino groups. Primary fatty acids attached to the lipid A backbone are amide- and ester-linked and comprise, in general, medium- to long-chain (R)-3-hydroxyfatty acids (C Both ester- and amide-linked (R)-3-hydroxyfatty acids may be acylated at their 3-hydroxyl group to constitute (R)-3-acyloxyacyl residues. As these results and Fig. 2 show, three structural components may be defined in lipid A: (i) the lipid A backbone consisting of a pyranosidic HexN disaccharide and phosphate groups, (ii) substituents of the backbone phosphate residues (polar head groups), and (iii) fatty acids. Therefore, lipid A of different bacteria may be classified according to the nature of the backbone constituents (GlcpN or GlcpN3N), the type and nature of the polar head groups, and features of the acylation pattern. In a few instances, other backbone substituents have been encountered. These will be described later in conjunction with individual lipid A forms. In the following paragraphs, our present knowledge of these structural components and their significance for the lipid A architecture will be discussed. With few exceptions, the chemical structure of these lipid A forms was determined by analyzing LPS derived from rough R-mutant bacteria. In contrast to wild-type (S) bacteria, these mutants form, because of defects in the synthesis, transfer, epimerization, or phosphorylation of glycosyl residues, incomplete LPS (R-form LPS), consisting of lipid A and a truncated polysaccharide region. 1. The Lipid A Backbone

a. General Principles and Structures. -The lipid A backbone constitutes the hydrophilic portion of the amphiphilic lipid A. Structurally, the lipid A backbone is a bisphosphorylated p-( 1 6)-interlinked HexN disaccharide, the HexN residues being always present in the D-glucopyranosidic form. Two types of 2-amino-2-deoxy-~-glucopyranoses have thus far been encountered as backbone constituents, namely GlcpN and GlcpN3N. Two phosphate groups are directly attached to the lipid A backbone, one in position 4’[ofHexN(II)] and one in the a-glycosidicposition 1 [of HexN(I)]. As will be shown here, various lipid A backbone structures, isolated from taxonomically related but also remote bacteria, possess this common structural principle, which comprises the most conservative region of the entire LPS molecule. The p-( 1 6)-interlinked HexpN disaccharide structure has thus far not been identified in other biomolecules such as glycoproteins, glycolipids, or glucans (5 1) and is thus unique to lipid A.

-

MOLECULAR STRUCTURE OF LIPID A

217

b. Structural Elucidation of the Lipid A Backbone.-The presence of GlcpN and GlcpN phosphate in lipid A was demonstrated in studies (1,28,30,52) on enterobacterial LPS. In 1964, Burton and Carter (53) provided evidence that free lipid A of E. coli contains a glycosidically linked GlcpN (phosphate) disaccharide. Gmeiner et al. (54) were the first to show, by using chemical degradation procedures (periodate oxidation), in combination with chemical and enzymic analysis, that the lipid A backbone of a S. minnesota Re mutant (strain R595) was a (1 + 6) interlinked GlcpN disaccharide that had been previously synthesized by Foster and Horton (55). The first methylation analysis of the lipid A backbone was carried out by Adams and Singh (56),studyinglipid A of Serratia marcescens. Also in this case, the GlcpN glycosidic linkage was found to be 1 + 6, and most probably p. A more-systematic degradation scheme for the isolation of the GlcpN disaccharide and application of methylation analysis, including g.1.c. - m.s., was first introduced and described by Hase and Rietschel(57,58).Theg.1.c. -m.s. analysis was further improved by Jensen et al. (59) with the help of synthetic reference compounds. Upon g.1.c. - m.s., the permethylated B-D-G~C~NAC(1 + 6)-~-GlcNAc-olcould be distinguished from the a-anomeric isomer a-~-GlcpNAc-( 1 + 6)-~-GlcNAc-olby its g.1.c. retention time. Moreover, the electron-impact (e.i.) fragmentation pattern of permethylated p-DGlcpNAc-(1 + 6)-~-GlcNAc-ol was characteristic for a 6-substituted GlcNAc-01. The first successful lH-n.m.r. analysis of the lipid A backbone was achieved with a purified dimethyl monophosphoryl lipid A derivative isolated from the E. coli Re mutant strain F5 15 (35). By using preparative thin-layer chromatography (t.1.c.) for the isolation of the lipid A structures, the authors also obtained evidence for heterogeneity relating to the degree of acylation of the lipid A backbone. Qureshi et al. (32) were the first to introduce the h.p.1.c. methodology for isolating homogeneous monophosphated lipid A derivatives from LPS of a S. typhimurium Re mutant strain (G30/ C2 1). H-N.m.r. analysis of a dimethyl pentatrimethylsilylmonophosphoryl lipid A derivative (60) revealed structural identity with E. coli lipid A (35). A disadvantage of the derivatization procedure used in the case of S. typhimurium (60) was that the trimethylsilylateddimethyl monophosphoryl lipid A derivative was not suitable for the characterization of the (free) hydroxyl group in position 4. This identification could be achieved, however, by the approach of Imoto et al. (33,who investigated the dimethyl monophosphoryl lipid A of E. coli and its per-0-acetylated derivative to compare the chemical shifts before and after acetylation. From the differences in the chemical shift observed in lH-n.m.r., the authors could clearly demonstrate that, in their preparation, position 4 of GlcN(1) was free. All

218

ULRICH ZAHRINGER et al.

other backbone hydroxyl groups showed no significant chemical-shift differences, indicatingthat the hydroxyl group at C-4 was the only free one in E. coli monophosphated lipid A. (The chemical shift of the primary hydroxyl group at C-6’ was not assigned and it could, therefore, not be determined as being free.) The first 13C-n.m.r. analysis of the p-(1 -6) GlcpN disaccharide was camed out on the lipid A backbone ofE. coli (6 1). However, the assignments in this study were different from those obtained by Krasikova et al. (62) who synthesized the disaccharide p-~-GlcpN-(1 6)-~-GlcpNand used it as a reference compound. I3C-N.m.r. was also used to identify the lipid A backbone of S. typhimurium (63), S. minnesota (64,64a), E. coli (34,64a, 65,66,66a), Proteus mirabilis (67), and Haemophilus injluenzae (68).

-

c. Glycosyl Region of the Backbone.-The lipid A backbone contains phosphate and glycosyl residues,the latter composed of the gluco-configured D-hexosamines GlcpN and GlcpN3N. In the following, types of lipid A backbones are described that are classified according to the nature of the constituent hexosamine are described.

(9 Lipid A Containing a GlcpN- GlcpN Disaccharide Backbone.-In addition to its presence in LPS of E. coli (57,69), S. minnesota (54,57,70), and S. typhimurium (60),thep-D-GlcpN-(1 + 6)-~-GlcpN disaccharide was also identified as the lipid A backbone structure in many other enteric and nonenteric Gram-negative bacteria. These include the Enterobacteriaceae Yersinia enterocolytica (7 I), Y.pseudotuberculosis(62), S. marcescens (56), P. mirabilis (67,72), Shigella jlexneri (57) and Enterobacter agglomerans (72a,72b); the Neisseriaceae Neisseria gonorrhoeae (33), N. meningitidis (73,74),and Acinetobacter calcoaceticus(75);the PseudomonadaceaePseudomonas aeruginosa (76,77a)and Xanthomonas sinensis (78);the Bacteroidaceae Bacteroides fragilis (79), Fusobacterium nucleatum (80) and Porphyromonas (Bacteroides) gingivalis (80a); the Rhizobiazeae Rhizobium trifolii (8 1) and R. meliloti (82), and members of families such as Haemophilus injluenzae (68), Vibrio parahaemolyticus (83), Rhodocyclus gelatinosus (84), Chromobacterium violaceum (85), Vibrio cholerae(86), Rhodopseudomonas gelatinosa,Rhodobacter (formerly Rhodopseudomonas)sphaeroides (87,88),Rhodobacter capsulatus(89),Rhodospirillum tenue (90),Rhodomicrobium vannielii (9 I), Sphaerotilus natans (92), Bordetella pertussis (93), Actinobacillus actinomycetemcomitans (94), Erwinia carotovora (95) and others ( 1,4). Moreover, the p-( 1 6) linked GlcpN disaccharide has been identified in LPS of wild-type strains (S-form LPS) of, for example, S. minnesota, E. coli 086, E. coli 0 1 1 1, and x.sinensis, indicating that this lipid A backbone structure is present (57) in both S-form and R-form LPS. In addition to a p-( 1 6) GlcpN disaccharide, an a-(1 + 6) GlcpN disac-

-

-

MOLECULAR STRUCTURE OF LIPID A

219

charide backbone has been postulated (on the basis of g.1.c. -m.s. investigations) as also being present in an R-mutant (R4) of P. mirabilis (67), although the p-( 1 + 6)-linked GlcpN backbone predominated over the a-(1 6)-linked GlcpN in the molar ratio of 8 : 1. Thus far, no other bacteria, including other P. mirabilis strains, have been identified as containingan a-(1 6) GlcpN disaccharide as a lipid A backbone constituent. It, therefore, remains to be established whether this backbone structure is unique to this particular mutant, whether it has been overlooked in other bacteria, or whether it constitutes an artifact formed during degradation or derivatization procedures.

-

(ii) Lipid A Containing a GlcpN3N-GlcpN Disaccharide Backbone. The diaminohexose2,3-diamino-2,3-dideoxy-~-glucopyranose (GlcpN3N) was not known to occur naturally until 1975, when it was identified as a constituent of the lipid A of Rhodopseudomonas viridis and Rhodopseudomonas palustris LPS (96). The GlcpN3N component replaces completely (see later) or partially the GlcpN residues of the backbone, as has been suggested by analytical data (49). Lipid A containing both GlcpN and GlcpN3N have previously been termed “mixed lipid A” (49,50). Figure 3 shows this glycose residue as a component of the backbone of Campylobacterjejuni lipid A, which thus far is the most thoroughly investigated representative of a GlcpN3N-containing,disaccharidic lipid A backbone (97). In the major lipid A component of C.jejuni, GlcpN3N positionally replaces the GlcN(11) residue of the classical p-~-GlcpN-(1 6 ) - ~ GlcpN structurediscussed in the previous section. The HexN disaccharideis p-( 1 6)-linked, and similaritiesto the E. coli lipid A structure with regard to the location of phosphate and acyl groups are also evident (Fig. 2). The lipid A backbone of C.jejuni, however, is heterogeneous with respect to the number of GlcpN and GlcpN3N residues present in the disaccharide structure. Three types out of the four theoretically possible combinations were identified: ( i ) the classical p-~-GlcpN-(1 6)-~-GlcpNlipid A backbone disaccharide, (ii)a p-~-GlcpN3N-(1 6)-~-GlcpNdisaccharide, and (iii)a p-~-GlcpN3N-(1 6)-~-GlcpN3N disaccharide. These structures are present in a molar ratio of approximately 1 :6 : 1, the GlcpN3N-GlcpN hybrid backbone, therefore, predominates(97).This type of hybrid backbone is also present in Brucella and in such phototrophic microorganisms as the Chromatiaceae(98). Thus far, a lipid A in which the backbone disaccharide contains GlcpN3N at the reducing and GlcpN at the nonreducing position has not been identified. A lipid A composed of GlcpN3Nand GlcpN was also identified in Thiobacillusferrooxidans and T. thiooxidans(99). In the case of P. carboxydovoruns (loo), which contains GlcpN3N, the lipid A structure was not elaborated [for further details, see also Ref. (lol)].

-

-

-

--

220

ULRICH ZAHRINGER et al.

FIG.3. -Chemical structure of the major component of Campylobacterjejuni lipid A. For details see the text. See also the legend to Fig. 2. For substituents of the phosphate groups see Table I. The a-anomenc phosphate has been tentatively assigned (97).

(iii) Lipid A Containing a GlcpN3N-GlcpN3N Disaccharide Backbone. -In Pseudomonas diminuta lipid A (102), a GlcpN3N disaccharide was identified as constituting the lipid A backbone (103). The nature of the glycosidic linkage of the disaccharide has been postulated to 8, but only on the basis of serological cross-reactivitywith Salmonella lipid A. It was, thus, concluded that the disaccharide wasp-( 1 + 6)-linkedand that at position 4’ of the distal GlcpN3N residue a phosphate group is present, whereas the nature of the substituent at the glycosidic position was not identified. As noted before in C.jejuni lipid A (Fig. 3), a minor fraction contains a p-( 1 6)-linked GlcpN3N disaccharide (97). In the lipid A backbone of Caulobacter crescentus GlcN could not be detected. It is possible, however, that a GlcpN3N-GlcpN3N disaccharide is present ( 103a).

-

(iv) Lipid A Containing a GlcpN3N Monosaccharide Backbone.- As a constituent of lipid A, GlcpN3N was first detected in R. viridis (96,104), where it is present as a nonphosphated bis-N,N-acylated monosaccharide. A monosaccharidic GlcpN3N residue forms also the lipid A backbone in R.

MOLECULAR STRUCTURE OF LIPID A

22 1

palustris (96), Pseudomonas diminuta (102), and Phenylobacterium immobile (105). Lipid A containing exclusively a GlcpN monosaccharide has not been identified thus far. In this context, it should be pointed out that the p-glycosidic linkage of the N-acylated lipid A backbone lacking the 4’-phosphate is very susceptible toward the acid hydrolysis that is employed to isolate lipid A (69). The possibility must, therefore, be kept in mind that lipid-A-derived monomeric HexN structures may also result from acid treatment of LPS. ( v ) Lipid A Containing Other Glycosyl Backbones.-In a recent study (106) it was claimed that, in R. trifolii, D-glucosaminuronicacid (GlcpAN), carrying amide-linked 27-hydroxyoctacosanoicacid [28 :0(27-OH)],comprises the lipid A backbone and not a p-( 1 + 6)-linked GlcpN disaccharide as had been found previously (81). Whether the reported GlcpAN residue constitutes the lipid A backbone or whether it is present in addition to the GlcpN-containingdisaccharide remains to be elaborated. Recently, Bhat et al. (107) reported the existence of three different types of lipid A backbones in the LPS of members of the family Rhizobiazeae, these being those containing only GlcpN (R. meliloti and R. fredii), a second type containing a GlcpN-GalA disaccharide (R. leguminosarum),and a third type containing GlcpN3N either in the form of a monosaccharide or as a hybrid disaccharide together with GlcpN (Bradirhizobiumspp.). The detailed structures of these lipid A backbones await further analysis.

( v f i Conclusions.-In summary, the general structure of the lipid A backbone disaccharide is represented by a bisphosphated p-( 1 + 6)interlinked HexpN disaccharide having GlcpN or GlcpN3N as glycosyl components. GlcpN disaccharides having a p-glycosidic linkage are also present in other natural molecules such as glycoproteins (5 I), the bacterial murein (peptidoglycan), and chitin (108). In these examples, however, the disaccharide linkage isp-( 1 4). To our best knowledge, ap-( 1 + 6) linked GlcpN or GlcpN3N linkage is not present in other biomolecules, and consequently, this structure appears to be unique and characteristic for one, namely, lipid A. Lipid A containing monosaccharidic GlcpN3N (but not GlcpN) has also been encountered. In these instances, however, further structural studies are indicated to establish details of these lipid A structures.

-

d. Phosphate Groups of the Backbone.-The backbone of disaccharidic lipid A contains, in general, two phosphate groups. Of these, one is a-linked to the glycosylic hydroxyl group at C- 1, and the other is ester-bound to the hydroxyl group at position C-4’ of GlcN(11).

(i) The Glycosylic Phosphate (la-P).-As

investigations of the lipid A

222

ULRICH ZAHRINGER et al.

structure were often carried out using monophosphoryllipid A, which lacks the glycosylic phosphate, the anomeric configuration of GlcN(I) remained unknown for a long time. In 1982, however, Redmond and colleagues (109) showed by 'H- and 31P-n.m.r.spectroscopythat, in the Re LPS ofS. minnesotu R595, the phosphate residue attached to GlcN(1) is a-glycosylically linked (Fig. 4). This was also shown by I3C-n.m.r. spectroscopyto be true for an Re mutant of E. coli (66), for an 0-deacylated and per-0-acetylated derivative of Re LPS of E. coli F5 15 (34) and for the E. coli mutant D3 1m4 (1 10). Indirect evidence for the a-1-phosphate in B. pertussis was obtained by Caroff et al. (1 1 l), who synthesized the 2-deoxy-2-[(R)-3-hydroxytetradecanamidol-a- and P-D-glucopyranose 1-phosphates. By comparing the (different) rates of phosphate release during acid-catalyzed hydrolysis of the synthetic a- and P-glycosylically bound phosphate on the one hand and bacterial lipid A on the other, the anomeric configuration of GlcN(1) in B. pertussis lipid A was determined to be a.In later investigations dealing with lipid A of C.jejuni (97), the 1-phosphatewas determined by a similar method also to be a.Further, in lipid A of H. influenzue, the glycosylic 1-phosphate was found to be a,as determined by 'H- and '3C-n.m.r. spectroscopy (68).

FIG.4. -Chemical structure of lipid A of the Salmonella minnesota Re mutant strain R595. For details see the text. See also the legend to Fig. 2. For substituents ofthe phosphategroups see Table I.

MOLECULAR STRUCTURE OF LIPID A

223

(ig The Non-Glycosylic Phosphate ( 4 ’ 4 . -A nonglycosylic (esterbound) phosphate, linked to GlcN(II), was detected in acid hydrolyzates of lipid A (52). From a study on deacylated S. minnesota Re LPS, it was determined (1 12)that phosphate was attached to the hydroxyl group at C-4’ of the lipid A backbone disaccharide (Fig. 4). This investigation showed that hydrazinolysis of S. rninnesota Re LPS followed by mild acid hydrolysis yielded the phosphorylated oligosaccharide P-4’-GlcpN-P-( 1 6)-GlcpN and, after N-acetylation, P-4’-GlcpNAc-P-(1 6)-GlcpNAc as well as the nonphosphated GlcpN-P-( 1 46)-GlcpN and GlcpNAc-P-( 1 + 6)GlcpNAc disaccharides, respectively, which were purified and isolated by preparative high-voltage paper electrophoresis (h.v.p.e.) ( 112). Subsequent to periodate oxidation of the lipid A backbone derivatives, the nonreducing GlcpN in P-4’-GlcpNAc-p-( 1 6)-GlcpNAc was found to be resistant to periodate oxidation, indicatingthat the phosphate group was linked to either position 3’ or position 4’. When the non-N-acetylated P-GlcpN-P-(1 6)GlcpN and GlcpN-p-( 1 6)-GlcpN disaccharides were similarly treated, GlcpN (and P-GlcpN) was quantitatively oxidized and no longer detectable, strongly indicating that position 4‘ of GlcN(I1) was the attachment site of phosphate. The first ”P-n.m.r. data (1 13) of various chemically degraded LPS from S. minnesota R595 were in accordance with these results. Periodate resistance of GlcN(11)in nonmodified and N-acetylated P-4’-GlcpNGlcpN was also shown for the lipid A disaccharideofR. sphaeroides (87) and was explained on the basis of a phosphate group being present at position 4’. An elegant chemical approach toward the determination of the position of the ester-bound phosphate in a Re mutant of E. coli K-12 (69) comprised oxidation of 0-deacylated lipid A with Me,SO/Ac,O, whereby the primary hydroxyl group at C-6’ is oxidized to the aldehyde (hydroxyl groups at C-3’ and C-3 are also oxidized to keto groups). Alkaline conditions then caused the release of the phosphate (in the form of inorganic phosphate) from position 4’ via a p-elimination reaction . The application of ‘H- and 31Pn.m.r. spectroscopy,measuring selective proton -proton couplingconstants by polarization transfer from the proton (H-4’) to phosphorus, also proved the location of phosphate at position 4’ in the lipid A of S. minnesota R595 (1 09). Another approach was devised by Helander et al. (68): LPS of adeep rough mutant of H. inJluenzae(strain 1-69 Rd-/b+)was methanolyzed (2 MHCIMeOH, 2 h,86 “C) and permethylated. The main products incorporating both GlcpN and phosphate were identified by g.1.c. - m.s. analysis as pyranosidic GlcpN 4’-phosphate carrying a 3-0-methyltetradecanoicacid residue by the characteristic fragmentation pattern in the electron impact mass spectrum (e.i. - m.s.). The same procedure was successfully applied to P. aeruginosa (mutant PAC605), which also contains a phosphate group at

-

-

-

-

-

224

ULRICH ZAHFUNGER et al.

FIG.5. -Chemical structure of the main component of lipid A of Bacteroidesfragilis NCTC 9434 LPS.For substituents ofthe phosphate groups see Table I. The anomeric configuration of the GlcN(1)has been tentatively assigned as (Y (79).

position 4’ of GlcN(11)(77). As an advantage of this latter approach,not only the position of phosphate but also the nature of fatty acid, amide-bound to the nonreducing GlcpN, may be determined in the same experiment. For the determinationof the phosphate group a new procedure was established (64a)using a two-step deacylation of the rough form LPS (E.coli F5 15 and S. minnesota R595) and h.p.1.c. purification of the intact core-backbone oligosaccharide.Data obtained by lH-, 13C-,and 31P-n.m.r.spectroscopy as well as f.a.b.-m.s. analysis unequivocally showed the presence of 4’-phosphate at GlcN (11). A similar procedure was also applied to E. coli J5 LPS (66a). Although the phosphate group bound to position 4’ of HexN(I1) is a general lipid A constituent, examples are known where it is lacking. Thus, in B. fragilis (79) (Fig. 5.), the hydroxyl group in position 4’ is free and in R. vannielii (9 l), the phosphate group in position 4’ is positionally replaced by a P-linked D-mannose residue. The ester-linked phosphate group is also lacking in other lipids A, including Chromatiurn vinosum (1 14), Selenomonas ruminantium (1 1 9 , and Thiobacillus species (99) (for further examples, compare Refs. 49 and 50).

225

MOLECULAR STRUCTURE OF LIPID A

(iii) Conclusions.-The p-( 1 + 6)-linked HexpN disaccharide of lipid A carries, in general, two phosphate groups. The glycosylicphosphate group is a-linked to C- 1 of HexN(1) and the nonglycosylicphosphate group is esterbound to position 4' of HexN(I1). All disaccharide-containinglipid A carry the a-glycoslyic phosphate residue, whereas the nonglycoslyic phosphate is lacking in lipid A of certain Gram-negative bacteria. These latter preparations are generally of low endotoxic activity (9 1,l I6 - I 18) 2. Phosphate Substitutents

Both the glycosylic and the nonglycosylic phosphate residues may be substituted, in general, by nonacylated, charged glycosyl or nonglycosyl residues. These polar head-groups are often not present in stoichiometric amounts, leading to a certain intrinsic heterogeneity of lipid A. It should be noted that, in Legionellu pneurnophilu (1 19) and H . influenzue(120), glycerol phosphate was identified as a constituent of LPS or lipid A, the location of which, however, was not studied. Some of the substituents have been structurally elucidated and their nature and location are summarized in Table I.

TABLEI Phosphate-Linked Substituentsof the Lipid A Backbone (Y-P-FD-GlcpN-(l+ 6>a-D-Gl~pN-l-P) Substituent of backbone phosphate group at Organism

C-4'

Escherichia coli Salmonella minnesota Proteus mirabilis Yersiniapestis Vibrio cholerae Neisseria meningitidis Chromobacterium violaceurn Campylobacterjejuni Pseudomonas aeruginosa Haemophilus influenzae Rhodospirillum tenue Rhodobacter capsulatus Sphaerotilus natans Moraxella catarrhalis

L-Arap4N" ~-Arap4N L-Arap4N

a

Nonstoichiometric.

P,phosphate. cEtn, 2-aminoethanol.

P-Etna ~-Arap4N P-Etn" ~-Arap4N P-Etn* P-Etn P

c-1 pa.b

P-Etna.=

-

D-Araf" P-Etn P-Etna GlcpN P-Etn D-Anf P-Etna

P-Etna

Reference 4 4 4

121 86 73 85 97 77 68 90 89 92 193b

226

ULRICH ZAHRINGER et al.

a. Substituents at the Non-Glycosylic 4’-Phosphate.-Substitution of the 4’-phosphate was, for the first time, identified in lipid A of C. violaceum (85), where 4-amino-4-deoxy-~-arabinopyranose (~-Arap4N)quantitatively substitutesthe 4’-phosphate of GlcN(11).This is also the case in lipid A of P. mirabilis (72) and Y. pestis (121). In S. minnesota Re LPS (Fig. 4), the 4’-phosphate of GlcN(11) is substituted by this aminopentose, however, in a nonstoichiometricmanner (85,113). The L-Arap4N linked to the phosphate at position 4’ is released from (hydrazine-treated)LPS under mild alkaline conditions (0.2 MNaOH, 1 h, l0O’C;) in the form of its 1-phosphate(85), and concomitantly with the release of P-~-Arap4N1-phosphate, the 4’phosphate group is removed from the lipid A backbone. In S. minnesota Re LPS, the glycoslyic linkage of ~-Arap4Nto lipid A was determined, by IH-n.m.r. (109), to be p. In Rhodobacter capsulatus (89), C.jejuni (97), S. natans (92), and N. meningitidis (73), 2-aminoethyl phosphate (Etn-P) is bound to the nonglycosylic phosphate residue (Table I). In the case of N. meningitidis, the substitution is nonstoichiometric. As Fig. 6 shows, out of four conceivable lipid A species, the major one carries 2-aminoethyl pyrophosphate at positions 4’ and 1, and a minor one contains monophosphate groups at these positions. Studyinga lipid A precursorisolated from a temperature-sensitivemutant of S. typhimurium (STiSO), Strain et al. (63), by applying ‘H- and 31P-n.m.r., found that L-Arap4N was linked to the glycosylic phosphate at C-1 and 2-aminoethyl phosphate (Etn-P) was linked to position 4’ ofGlcN(11).Thus, the polar head-groupsin the lipid A precursor produced by this mutant are at opposite locations, as previously found for a structurally similar (“neutral”) precursor (122) and for lipid A of, for instance, S. minnesota (109,113). At present it is not known whether this discrepancy is due to differences in the methodology used or whether the substitution pattern in the precursor of S. typhimurium is differentfrom that of mature lipid A of the bacterial strains investigated previously. b. Substituentsof the Glycosyl a-Phosphate.-Either the substituents of the phosphate glycosylically linked to GlcN(1) are neutral or they carry

FIG.6.-Chemical structure of the lipid A backbone of Neisseria meningitidis M986 LPS having polar substituents.The nonglycosylic and glycosylic phosphate groups are substituted by Etn-P to a similar degree (85 vs SO%, w/w). The anomeric configuration of the phosphate has been assigned tentatively as a.R’ = 12:0(3-OH), RZ= 14-0[12:0(3-OH)] (73).

MOLECULAR STRUCTURE OF LIPID A

227

positive or negative charges. A neutral substituentis present in R. tenue (90), where D-arabinofuranose (D-Araf ) is glycosylically linked to the phosphate group at C- 1. In E. coli, phosphate may be substituted the glycosylic phosphate, resulting in a pyrophosphategroup at C- 1 (69).As a positively charged group, nonacylated GlcpN was identified in C. violaceum (85) as being linked to the glycosylic phosphate. In C. jejuni, an Etn group, and in S. minnesota (Fig. 4), V. cholerae, R. capsulatus, and N. meningitidis (Fig. 6) an Etn-P group are linked to the C-1 phosphate (Table I). The location of these residues at the glycosylicphosphate has also been studied in a precursor of lipid A biosynthesis, however, as just discussed, with discrepant results. c. Conclusions.-Both phosphate groups of lipid A at C-1 and C-4’ may carry (polar) substituents,although examplesin which the phosphate groups are free are known. The substitution of phosphate groups is often nonstoichiometric, as exemplified with N. meningitidis lipid A (Fig. 6). In this interestingexamplethe glycosylicposition [GlcN(I)] and the hydroxyl group at C-4’ [GlcN(11)]both carry Etn-PP. This substitution pattern is symmetric, and careful chemical and 31P-n.m.r.analyses have shown (73) that, in the lipid A part of this LPS, both phosphate groups at positions 4’ and 1 are nonstoichiometrically substituted by Etn-P to a similar degree (85 vs 80%). In lipid A produced with sodium acetate buffer (0.1 M NaOAc, pH 4.5, 1 h, 100°C)the degree of Etn-P substitution at position 4’ (73%, w/w) is slightly lower, whereas substitution at C-1 is decreased to one half (45%,w/w) compared with native LPS (80%, w/w). This indicates that the pyrophosphate group of Etn-PP at C-4’ is resistant to mild acid hydrolysis, whereas that of the a-glycosylically linked Etn-PP group at C- 1 is susceptible to even mild acid hydrolysis. The extent of phosphate substitution is influenced by, for instance, the growth conditions ( 122- 124).It is possible that bacteria, depending on their physiological demands, are able to add or omit ionic head-groups and thereby regulate their net surface charge (1 13). It is noteworthy that most of the substituentson the phosphate, such as Etn, L-Arap4N, and GlcpN carry, at neutral pH, a positively charged amino group. Their presence in the neighborhood of phosphate residues (or of Kdo in LPS) may be considered as a regulating factor, controlling the electrostatic interaction of negatively charged residues (phosphate)with such bivalent cations as CaZ+or MgZ+of the medium. The Ca2+ and Mgz+ ions may neutralize intermolecularly negative charges of lipid A. These interactions, therefore, appear to be of great importance for the stability and function of the bacterial outer membrane. Moreover, it has been postulated that positively charged groups in lipid A, such as L-Arap4N,significantlycontribute to the resistance ofcertain Gram-negative bacteria (for instance P. mirabilis) to such antibiotics as polymyxin B ( 125). In accordance with this concept, P. mirabilis mutants lacking ~-Arap4Nare highly sensitive toward polymyxin B (126).

TABLEI1 Distribution of Acyl and 3-Acyloxyacyl Residues between Hydroxyl and Amino Groups of the Classical Lipid A Backbone I&&lcN(II)-(l6)-&lcN (I)] of Various Bacteria Nature and linkage type of acyl sobstitwots of the lipid A backbone bound to GlcN (II) at position Bacteria

3‘ (ester)

Envinia carotovora Proteus mirabilis

14:0[3-q14 :O)] 14:0[3-q 14:O)] 14 :0[3-0(14 :O)p 14:0[3-0[14:0(2-0H)]’] 14:0[3-q 14:O)] 14:0[3-q14:0)]

Haemophilus infuenzae Neisseria gonorrhmae Neisseria meningitidis Pseudomonas aeruginosa

14:013-q 14:O)] 12:0(3-OH) 12:0(3-OH) 10:0(3-0H)

Bacteroides fragilis Rhodocylclus gelatinosus

16:0(3-0H)” 15:0(3-OH)” 10 :0(3-OH)

Rhodobacter sphaeroides

10:0(3-0H)

Rhodobacter capsulaus Sphaerotilus natans Actinobacillus actinomycetemcomitans Moraxella caiarrhalis Porphyromonas gingivalis Enterobacter agglommans

Escherichia coli Salmonella typhimurium Salmonella minnesota

GIcN (I) at position 2’ (nmide)

14 :q3-q12 :O)] 14 :0[3-q 12:O)] 14 :0[3-q 12:O)]

3 (ester) 14 :0(3-0H) 14:0(3-0H) 14:0(3-OH) 14:0(3-0(16 :O)lb 14 :O(3-0H) 14:0(3-0H)

b

14: 0(3-0H) 14:0(3-0H) 14:0(3-0H)

10:0[3-0(A5-12: I)] 10:0(3-OH)

14:0[3-0(A7-14: 1)p 14:0[3-0( 14 :0)]’ 14 :o(3-0~0) 10: 0[3-0(12 :0)p

10:0(3-OH) 10:0(3-OH)

14 :0(3-OH) 14:0(3-0H) 14:0[3-0(16 :O)]’ 14:0(3-OH) 14 :0[3-q 12:O)] 14 :0[3-0(12 :O)] 12 :0[3-0( 12 :0)p 12 :0[3-0[12:0(2-0H)]p 16: 0(3-OH)” 15-Me-I6 :0(3-OH)” 10: 0[3-0( 12:0)p 10:0[3-0(14 :O)p 14 :o(3-0~0) 14: 0(3-0H) 14:0(3-0XO) 10:0[3-0(12 :0)p

14:0[3-q14 :O)] 12:0(3-0H) H 140[3-o[14:0(2-0H)]]d

14:0[3-q 14 :O)] n o [ 3 - q 100)) 15-Me-16:0[3-0(15:0)] 140[3-q120)]

14:0(3-0H) 120[3-q loo)] H 140(3-OH)

14 :0(3-0H) 12:0[3-q 12:0)] 15-Ma16 :0(3-OH) 14:O[ 3-o( 16 :O)]

14 :0[3-q12 :O)] 14 :0[3-q14 :O)] 14 :0[3-q 14 :O)] 14 :0[3-q 12 :O)] 14 :O[ 3 - q 12 :O)] 12 :0[3-0(12 :0)]’ 12:0[3-OIl2: 0(2-0H)])” 15-Msl6 :0[3-O[ I3-Me-14:0]]’ 16:0[3-0( 13-Met14 :0]’ 10:0[3-q12 :O)]

14: 0(3-0H) 12:q3-OH) 12 :0(3-OH) 10:0(3-OH)b 15:0(3-OH)” 16:0(3-0H)” 10:O(3-OH) 10:0(3-0H)

Fatty acids in one column add up to 1 mol eq of lipid A. b n t in nonstoichiomctricamounts. Intrinsinc h-neity, I2 :O may be replaced by 10:0,10: I, 12 : 1, 14:0,14: 1 ester linked to 10:0(3-0H). ”Positionof the hydroxy p u p (2-0H or 3-0H) was not daermined; intrinsic hetnogcneity. I40 (2-0H)may be nplacedby 140. a

2 (amide)

Reference

4 32,60 37,181 95 12 68 33 35 71

19 84 87,88

89 92 94 196a 80a 51a

MOLECULAR STRUCTURE OF LIPID A

229

3. Fatty Acids a. General. -As a characteristic feature, lipid A contains saturated medium- to long-chain ( 10- 22 carbon atoms) fatty acids in ester and amide linkages. These fatty acids comprise nonhydroxylated and hydroxylated ones ofwhich 60 - 75%are 3-hydroxylated. Thus, 3-hydroxylated fatty acids usually preponderate within the fatty acid profile of lipid A. The fatty acid profile of lipid A differs significantly from that of other bacterial phospholipids (127). Fatty acids may be linked directly to hydroxyl and amino groups of the lipid A backbone. These are termedprimary (45) acyl groups and they include 3-hydroxy and 3-keto fatty acids. Fatty acids may also be linked to the 3-hydroxyl groups of the ester and amide-linked 3-hydroxy fatty acids. These are termed secondary acyl residues, and they include nonhydroxylated (normal) and 2-hydroxylated fatty acids. b. Nature of Fatty Acids. -(9(R)-3-Hydroxylated Fatty Acids. 3-Hydroxylated fatty acids constitute common and obligatory constituents of LPS, namely, their lipid A component. The 3-hydroxy (or P-hydroxy) fatty acids of lipid A possess, without exception, the (R)-configuration (128,129). The (R)-3-hydroxy fatty acids of different bacteria, either amideor ester-bound, have been found generally to be saturated straight-chain and even-numbered and to have a number of carbon atoms between 10 [e.g., in P. aeruginosa lipid A (77,130,13l)] and 22 [e.g., in Chlamydia trachornatis (132) and C. psittaci (133) lipid A]. In Veillonella,however, and some other bacteria, even-numbered 3-hydroxy fatty acids are present ( 134). Of these 3-OH acyl groups, (R)-3-hydroxytetradecanoic acid [ 14 :0(3-OH)] was identified in about two thirds of all bacteria investigated so far (127). As a general rule, (R)-3-hydroxy fatty acids are directly bound to the amino or hydroxyl groups of the lipid A backbone disaccharide(positions 3’,2’,3, and 2) and are, therefore, designated as primary lipid A fatty acids (45). 3-Hydroxylated fatty acids as substituents of an amide-bound 3-hydroxy fatty acid (3-acyloxyacylresidue in amide linkage) have not been identified thus far (38). However, an ester-bound 3-acyloxyacyl residue in which both fatty acids are (R)-3-hydroxylated [ V. cholerae (86)]was identified. In general, as is the case in all Enterobacteriaceae, and in Helicobacter pylori (135,136), only one type of 3-hydroxy fatty acid is involved in amide linkage (Table 11, Figs. 2,4), but in a few instances (Bacteroides, Xanthomonas, Veillonella, and Myxobacteria) two different 3-hydroxy acids are amide-bound to the lipid A backbone. In B.fragilis, the acyl groups having the higher number of carbon atoms are linked to GlcN(I1)(79). In LPS of a number of intracellular Gram-negativebacteria such as Coxiella burnetii (137) and L. pneumophila ( 1 19),a large number ofdifferent amide-bound(R)-3-hydroxyfatty acids are present exclusively in amide-linkage, however, in unknown location.

230

ULRICH ZAHRINGER et a1

In a few cases, 3-hydroxy fatty acids that contain additional functional groups were characterized. Thus (R)-3-hydroxydodec-(5Z)-enoicacid [A512 : 1(3-OH)]was identified in ester linkage in lipid A of the chloridazon-degrading bacterium P. immobile (105). Unsaturated 3-hydroxy fatty acids are also present in R. trifolii (81) and R. meliloti (82). In an extract of L. pneumophila, 2,3-&hydroxy-12-methyltridecanoic acid [ 12-Me-13:0(2,3diOH)] and 2,3-dihydroxytetradecanoicacid [ 14 :0(2,3-diOH)]were identified in amide-linkage ( 138).It remains, however, to be established, whether these 2,3-dihydroxylatedfatty acids are lipid A constituents. Small amounts of 2-methyl-3-hydroxy-fattyacids have been detected in B. pertussis (139). The determination of (R)-3-hydroxylated fatty acids is of considerable chemotaxonomic value (30,127), and lipid A-linked 3-hydroxylated fatty acids are used as diagnostic markers for the qualitative and approximate quantitative determination of endotoxins in biological samples (140). For the analysis of 3-hydroxy fatty acids in very low concentration,they are first liberated by hydrolysis of LPS or lipid A, followed by modification of the carboxyl groups to methyl esters and of the 3-hydroxyl groups to the trimethylsilyl ethers. Using g.1.c. -m.s., this approach was found to be suitable for the determination of endotoxin concentrations in the serum with a detection limit of 200 fmol3-hydroxy fatty acid/mL (140). The concept of using 3-hydroxylated fatty acids for the detection of endotoxin in serum (140), peritoneal fluid and plasma (141), or organs of higher animals is furtherjustified by the fact that the amount and type of @)-3-hydroxylated fatty acids present in lipid A are not altered by changing the growth conditions (temperature,medium, and aeration), as has been shown for P. mirabilis, S. minnesota, E. coli (37,142,143),P. aeruginosa (144), and Y. enterolitica (145). Moreover, some bacteria, such as B. frasilis, express characteristicand diagnostic hydroxy fatty acids [ 15-Me-16:0(3-OH)]. The presence of (R)-3-hydroxylatedfatty acids in all lipid A investigated thus far shows that they constitute a highly conserved structural feature of LPS. It should be mentioned that, although very characteristic,(R)-3-hydroxylated fatty acids are not present exclusively in lipid A. They have also been identified in such other lipids as the ornithine-containing lipids of various Gram-negative bacteria. Ornithine lipids having (R)-3-hydroxylated fatty acids of a chain length similar to that of lipid A-linked (R)-3-hydroxy fatty acids (CL4-C were found, for example, in B. pertussis, Gluconobacter, Acetobacter, Brucella, and Francisella (for details see Refs. 127, 146, and 147).(R)-3-Hydroxylatedfatty acids are also present in the rhamnolipid of P. aeruginosa ( 148), in the fish poison pahutoxin ( 149), and in glycolipids of Rhodotorula (1 50).

(iq (S)-2-Hydroxylated Fatty Acids. -The 2-hydroxylated (a-hydroxylated) fatty acids encountered in lipid A possess, without exception, the

MOLECULAR STRUCTURE OF LIPID A

23 1

(S)-configuration ( 129,151). They are not as common as 3-hydroxy fatty acids and are never substituted at their 2-hydroxyl group. Among the (S)-2hydroxy fatty acids identified, (S)-2-hydroxydodecanoic[ 12 :0(2-OH)] and (S)-2-hydroxytetradecanoic acid [ 14 :0(2-OH)] predominate. Thus, 14 :O(2-OH) was found to be present in lipid A of Salmonella (Fig.4), Klebsiella, and Serratia, but to be absent from LPS of Escherichia, Proteus, Enterobacter, Shigella, and Yersinia strains (15 1). On the other hand, 12 :O(2-OH) was identified in Pseudomonas (77,152,153), Acinetobacter (1 54), Bordetella (139), Chromobacterium (85), and other genera. Unlike 3-hydroxylated acyl groups, 2-hydroxy fatty acids are not bound directly to the lipid A backbone. They are present exclusively as secondary ester-linked fatty acids, attached to the hydroxyl group of an (R)-3-hydroxylated acyl residue, thus forming an acyloxyacyl group. These 2-hydroxy fatty acid-containing 3-acyloxyacyl groups are either amide- (C. violaceurn, P. aeruginosa, X. sinensis, and A. calcoaceticus) or ester-linked [S. minnesota and A. calcoaceticus (38)J to the lipid A backbone. In several instances (such as S. minnesota and P.aeruginosa), the amount of the 2-hydroxylatedfatty acid and that of the nonhydroxylatedfatty acid of the same chain length adds up to 1 mol per molecule of lipid A. [thus, in S. minnesota 14 :O(2-OH) and 14 :0; in P. aeruginosa 12 :O(2-OH) and 12 :01. As the 2-hydroxylated fatty acid occupies the same position in lipid A as the corresponding nonhydroxylated fatty acid, it appears that a-hydroxylation takes place at the level of fully acylated lipid A. As a consequence of partial a-hydroxylation,these lipid A express heterogeneity in that they contain at least two species, one containing the 2-hydroxylated and one containing the corresponding nonhydroxylated acyl group (26,77,151) (compare Fig. 4). In P. aeruginosa PAO, the degree of hydroxylation,that is, lipid A heterogeneity with respect to hydroxylated [ 12 :0(2-OH)] and nonhydroxylated ( 12 :0) fatty acids, was found to be mainly influencedby the growth temperature. With increasing temperature, the proportion of the nonhydroxylated fatty acids ( 12 :0 and 16 :0) was decreased at the expense of such hydroxylated ones as 12 :O(2-OH)(144). A similar finding was made with Salmonella species ( 142). It has previously been reported (15 5 ) that the lipid A of Pseudomonas paucimobilis (now termed Sphingomonas paucimobilis) contains an unusual lipid A-type glycolipid having 14 :O(2-OH) as the only amide-linked fatty acid. A recent reinvestigation of this compound ( 156), however, revealed that it constitutes a glycosphingolipid containing (in addition to GlcpA, Galp, Manp, and GlcpN) 14:O(2-OH) in amide linkage to the dihydrosphingosine base. As our recent analyses show, S. paucimobilis bacteria are completely devoid of 3-hydroxylatedfatty acids, and hence we hypothesize, that, in this case, LPS is replaced by this very glycosphingolipid,which, interestingly, lacks endotoxic activity. Thus, S. paucirnobilis, together with

232

ULRICH ZAHRINGER et al.

Borrelia burgdorferi (1 57), appears to constitute a group of microorganisms, which do not, in contrast to the majority of Gram-negative bacteria, require a typical LPS for growth and multiplication.

(iii) Other Hydroxylated Fatty Acids. -P. diminuta, P. vesicularis (1 58), and L. pneumophila (138) contain, in addition to 3-hydroxylated fatty acids, 5,9-dihydroxytetradecanoicacid. It is likely, but not unequivocally proven, that this fatty acid constitutes a lipid A component. 27-Hydroxyoctacosanoic acid [28 :0(27-OH)] has been identified as a major fatty acid in LPS of R. trifolii (106), R. meliloti (82), and L. pneumophila (1 59). Using 1.d.- m.s. analysis of dephosphated lipid A, in combination with chemical analysis, 28 :O(27-OH)was determined to be ester-bound to the 3’-hydroxyl group of GlcN(11)in R. meliloti. The 28 :O(27-OH) residue was also found in R. viridis, R. palustris, Nitrobacter winogratsky, Nitrobacter hamburgensis, Pseudomonas carboxydovorans, Bradirhizobium spp., Agrobacterium spp., and Thiobacillus spp., all being members (101,160) of the 16s rRNA subgroup a-2. Moreover, two 28 :O(27-OH)homologues, namely 26 :O(25-OH) and 29 :0(28-OH), were identified in Bradirhizobium lupini (101) and R. palustris le5 (160), respectively,the latter having a branched structure. In L. pneumophila such (0-1)-0x0- and 1,o-dioic long-chain fatty acids as 27oxooctacosanoic acid [28 :O(27-0xo)l and heptacosan-1,27-&0ic acid [27 :O(dioic)] and their homologues 29-oxotriacontanoic acid [30 :O(29oxo)] and nonacosan- 1,29-dioic acid [29 :O( 1,29)dioic] were identified (1 59). In other Legionella species, such as L.jordanis, L. rneceacherinii,and L. micdadei, such a-hydroxylated long-chain fatty acids as 2-hydroxy-270x0-octacosanoic acid [28 :0(2-OH,27-oxo)], 2-hydroxy-29-0x0-triacontanoic acid [30 :0(2-OH,29-oxo)], 2-hydroxyheptacosane-1,27-dioic acid [27 :O(2-OH)dioic], and 2-hydroxynonacosan-1,29-dioic acid [29 :O(2OH)dioic] have been analyzed ( 161) in the phenol- chloroform- petroleum ether (PCP) extracts, indicating that they are constituents of lipopolysaccharide. (iv) 3-0x0 (3-Keto) Fatty Acids.-As lipid A constituents, 3-ox0 fatty acids were first encountered in Mbrio anguillarum (D. Shaw, personal communication). Later it was shown that R. capsulatus contains 3-ketotetradecanoic acid [ 14 :O(3-0xo)l in amide linkage (Fig. 7). As shown in Fig. 8, R. sphaeroides also carries at GlcN(1) a 14:o(3-0x0) group in amide linkage whereas 14:0(3-OH) is amide-bound to GlcN(I1) (50,87,88,162).As a peculiar feature, this lipid A contains the unsaturated fatty acid (Z)-A7-tetradecenoic acid (A7-14: l), which is present as a 3-(7-tetradecenoyl)oxytetradecanoic acid (14 :O[ 3-O(A7-14 : l)]) group, amide-bound to GlcN(11) (87,88,162a).Unsaturated and 3-keto fatty acids are rarely encountered in

0 -0,II

P’

0

II ,o-

0

.P

HO’

O ‘H

FIG.7. -Chemical structure of lipid A of Rhodobucter capsulatus. Dashed lines indicate nonstoichiometricsubstitution.The anomeric a! configuration of phosphates and the configuration(ZorE)ofthedoublebondinA5-12:1 areassignedonlytentatively(89).Forsubstituents of the phosphates see Table I.

0 II,op\

OH

FIG.8.-Chemical structure of lipid A of Rhodobucter sphueroides (87,88,164). For details see the text and the legend to Fig. 6.

233

234

ULRICH ZAHRINGER et al.

lipid A, and the acylation pattern of R. capsulatus and R. sphaeroides has, therefore, to be considered unusual. A 14:O(3-0x0) residue was also found in the lipid A of Paracoccus denitrificans (163). Interestingly, those lipid A containing 3-ox0 fatty acids (R. sphaeroides, R. capsulatus, and P. denitrificans) also contain unsaturated ( 12 : 1 or 14 : 1) acyl groups. Thus far, 14:O( 30x0) is the only 3-keto fatty acid identified in lipid A, where it is always present in amide linkage (87,88,164). Lipid A of R. sphaeroides exhibits, compared with enterobacteriallipid A, very low toxicity (87,162,164- 166), and it was speculated that this was related to the presence of 3-0x0 and unsaturated fatty acids. In order to evaluate the possible role of these groups in endotoxin activity, Qureshi et al. ( 162) subjected lipid A of R. sphaeroides to hydrogenation treatment in the presence of Pt(1V)oxide hydrate, whereby 14:O(3-0x0) and A'- 14 : 1 were reduced to the racemic (R,S)-3-hydroxy and the saturated fatty acid, respectively. Biological analysis of the reduced material, however, revealed that the chemical treatment had not changed, that is, augmented, the low endotoxicity of R. sphaeroides lipid A, indicating that other factors (namely the size of acyl groups) are responsible for the lack of endotoxic activity ( 162). ( v ) Nonhydroxylated, Iso-, and Ante-Iso Fatty Acids. -Nonhydroxylated fatty acids are, like 3-hydroxy fatty acids, common and obligatory constituents of lipid A. In general, they are saturated and are composed of straight-chain and even-numbered acyl residues having carbon atom numbers ranging from 12 to 18. In Veillonella, odd-numbered acyl groups are present (134). Iso- and ante-iso fatty acids were identified in lipid A of such Bacteroides strains as B. fragilis NCTC 9343 (38,79,117) (Fig. 5), B. gingivalis, B. distasonis, and B. ovatus (127). In Bacteroides strains, 13-Me-14 :0 preponderates and, because it does not occur in other lipids A, it may be considered as a diagnostic marker for Bacteroides LPS (80a,117). In LPS of the plant pathogen X. sinensis (78), 9-methyldecanoic acid (9-Me-10: 0), and its &-oxidationproduct 9-Me-10 :O(2-OH) in addition to 10 :0(3-OH), 12 :0(3-OH), and the correspondingiso-branched 1 1-Me-12 :O(3-OH) have been identified. Iso- and ante-iso branched fatty acids have also been encountered in other Gram-negative bacteria including Flexibacter,Myxococcus, and Cystobacter ( 127). Cyclopropyl fatty acids have previously been postulated to occur as ester-linked constituents of lipid A of Neisseria sicca [cyclo-13:01 and A. calcoaceticus (formerly Micrococcus aceticus). However, subsequent studies on purified lipid A preparations revealed that these are devoid of cyclic (cyclopropyl) nonhydroxylated fatty acids, and, therefore, the previously encountered lipid A-associated cyclopropyl acyl groups are likely to be derived from other, contaminating bacterial lipids.

MOLECULAR STRUCTURE OF LIPID A

235

( v i ) Unsaturated (Alkenic) Fatty Acids. -As a general rule, unsaturated fatty acids are absent from lipid A. Nevertheless, R. sphaeroides ATCC 17023(87,88,162,164)incorporatesinto its lipid A an unsaturated fatty acid (A7-14:1)in molaramounts(Fig. 8). In thisexample, 14:0[3-0[(A7-14:l)]] was unequivocally identified by g.1.c.-m.s. (87), f.a.b. -m.s. (88), and 'Hn.m.r. ( 162) analysis as the amide-linked acyloxyacyl residue at GlcN(11) (Fig. 7). In R. capsulutus lipid A, A5-12: 1 is linked to the 10:0(3-OH) residue bound to C-3 of GlcN(I1) (Fig. 7). In R. vanielii, AI4-22: 1 was identified in ester linkage; however, its position at the lipid A backbone was not determined (9 1). Also Enterobacteria are able to synthesize unsaturated fatty acids and to incorporate these into the lipid A component. Thus, when grown at low temperature (10- 15OC) E. coli (143), Salmonella spp. (142), P. mirubilis (37), and Y. enterocolitica(145)are incorporated into the lipid A component unsaturated fatty acids that are not present in LPS ofbacteria grown at 37 "C. For E. coli and Salmonella strains grown at low temperatures, it was found that (2)-A9-hexadecenoicacid (A9-16: 1)was incorporated at the expense of 12 :0 (142,143), however, not quantitatively. Further investigationsof these lipid A by 1.d.-m.s. revealed that the unsaturated fatty acid specifically replaced the 12 :0 residue in 14 :0[3-0( 12 :O)] that is bound to GlcN(11) (37). A similar effect of thermoadaptation, resulting in the formation of amidebound 14 :0[3-O(A9-16:l)], was detected in P.mirabilis and Y. enterocolitica (145). Based on the foregoing examples, it may be concluded that unsaturated fatty acids present in lipid A are always bound to the hydroxyl group of the amide-linked 3-hydroxy fatty acid residue at GlcN(I1) (position 2'). Interestingly, the substitution pattern of this particular hydroxyl group, which, in all lipid A studied, cames a fatty acid, varies with the growth temperature and within bacterial genera. This specific site of the hydrophobic region of lipid A, therefore, expresses a relatively high degree of variability.

c. Linkage of Fatty Acids to the Lipid A Backbone. -The lipid A backbone of E.coli (Fig. 2), S. minnesota (Fig. 4), and S. typhirnurium carries approximately 4 mol eq. of (R)-3-hydroxy fatty acids [ 14 :0(3-OH)], 2 of which occupy amino functions (amide linkage at positions 2 and 2') and 2 of which are linked to backbone hydroxyl groups (ester linkage at positions 3 and 3'). In C.jejuni lipid A (Fig. 3), 3 mol eq. of (R)-3-hydroxyfatty acidsare arnide-linked, and in P.diminuta the four primary fatty acids are exclusively amide-bound. Both amide- and ester-bound(R)-3-hydroxyfatty acids carry, at their 3-hydroxyl groups, secondary fatty acids in ester linkage, forming 3-acyloxyacyl groups. These 0- and N-linked 3-acyloxylacyl groups are characteristic for lipid A and are found in taxonomically remote groups of Gram-negative bacteria.

236

ULRICH ZAHRINGER et al.

( i ) Ester-Bound Fatty Acids.-It is known that the ester-bound fatty acids, that is, their number, size, and type, play an essential role in the expression of endotoxic activity both in vitro and in vivo (5,118,162). As an example, not only 0-deacylated lipid A, but also the disaccharide precursor molecule (tetraacyl precursor Ia) containingfour 14 :O( 3-OH) residues (Fig. 9A) is inactive in inducing interleukin-1 (IL- 1), interleukin-6 (IL-6), and tumor necrosis factor a (TNFa) in vitro (167 - 169) or in eliciting the local Shwartzman phenomenon (170). This is in contrast to bacterial ( 171) and synthetic (172) pentaacyl precursor Ib (Fig. 9B) containing, in addition, 16 :0 bound to 14 :O(3-OH) at GlcN(1) (173), which expresses significant endotoxicity. Ester-linked fatty acids are bound to two types of hydroxyl groups in lipid A, namely, (i) the hydroxyl groups of the backbone at position 3 and 3’, which carry either 3-hydroxyacyl or 3-acyloxyacyl groups, or (ii)the hydroxyl groups of the backbone-linked 3-hydroxy fatty acids, which carry, in general, nonhydroxylated acyl groups. In order to obtain selectively O-deacylated derivatives of lipid A of defined chemical structure and, thus, to elucidate the biological significance of ester-bound acyl groups, many attempts, including selective chemical degradation or application of specific esterases, have been made. One successful approach toward selective O-deacylation of S. minnesota Re mutant R595 (128) LPS led to the discovery of ester-bound 14 :0[3-O(14 :O)]. Under mild alkaline conditions (0.25 M NaOH, 5 min, 56“C),the 14 :0[3-O(14: O)] residue, attached to position 3’ of GlcN(I1) (Figs. 2 and 4), is released. On the other hand, treatment of 4’-monophosphated lipid A of S. minnesota with triethylamine [3% (v/v), pH 9.41 at 100°C(174) causes rapid liberation of the 14 :O(3-OH) residue at position 3 of GlcN(1) and only slow release of 14 :O[ 3-O( 14 :O)] (174). Similar observations were made with the monophosphated lipid A of S. typhimurium (32,41),S. minnesota R595, and E. coli J5 (175). The reason for the different behavior of 14:O(3-OH) at position 3 and of 14 :0[3-O(14 :O)] at position 3’ is currently not understood. It is likely that the free j3-hydroxyl group of 14 :O(3-OH) contributes to its lability toward alkaline saponification (175). Hydroxylaminolysis, treatment with stronger alkali (0.5 M NaOH, 2 h, 100°C)and alkaline methanolysis(0.25 M NaOMe, 1 h, 5OoC)lead to complete 0-deacylation of LPS and lipid A (176). Particularly in the case of alkaline methanolysis, ester-linked 3-acyloxyacyl residues undergo, in addition to transmethylation, a j3-elimination reaction, whereby the (R)-3-hydroxy fatty acid ester is first transformed into the a,P-unsaturated and then into the (S,R)-3-methoxyfatty acid methyl ester. The acyl substituent,on the other hand, is eliminated in the form of the free fatty acid (176). In fact, the presence of a 3-methoxyacyl derivative in the fatty acid spectrum of a given LPS is a strong indication for the presence of an ester-bound 3-acyloxyacyl

MOLECULAR STRUCTURE OF LIPID A

237

A 0 -0,

II ,o-

HO'

"OH

B 0 -0,I HO

/

P'

0

FIG.9.-Chemical structure of precursor Ia (A) and precursor Ib (B).

residue. This P-elimination reaction, therefore, is useful in the structural elucidation of lipid A. Furthermore, enzymes that quantitatively liberate ester-bound fatty acids of lipid A are known. Such esterases are present in the amoebae Dictyostelium discoideum (1 77 - 179) and Acanthamoeba castellanii (1 78). Selective

238

ULRICH ZAHRINGER et al.

liberation of exclusively secondary fatty acids is elegantly achieved by application of the enzyme 3-acyloxyacylhydrolase(AOAH). This enzyme, which is present in granulocytesand macrophages(45,180)and which is likely to be involved in the biodegradation of LPS, cleaves nonhydroxylated and 2-hydroxylated secondary fatty acids present in 0-and N-linked 3-acyloxyacyl groups, that is, regardless of their structural context in the lipid A molecule (181). Lipid A partial structures, exclusively lacking secondary fatty acids, have recently been found to constitute, like lipid A of R. sphueroides(1 82) and R. cupsulutus (89), strong endotoxin antagonists and to inhibit the LPS-induced production of TNFa, IL-1, and IG6 in human monocytes (167,169,183- 185a). In view of these and previous (1 86) findings, partially 0-deacylated lipid A preparations currently occupy a central place in endotoxin research. Therefore, oligoacyl lipid A structures constitute attractive targets for future chemical synthesis, and the development of new methods for selective 0-deacylation will, most likely, prove to be highly rewarding.

(ii) Amide-Bound Fatty Acids.-In contrast to the wide spectrum of different types of fatty acids involved in ester linkage, only two types of acyl groups are amide-bound, namely, (R)-3-hydroxyand 3-0x0 fatty acids. The amide linkage is relatively resistant toward alkaline or acidic hydrolysis and prolonged treatment may lead to degradation and condensation reactions of 3-hydroxy or 3-0x0 fatty acids. Therefore, other degradation reactions have been developed for the liberation and identification of amide-bound acyl groups (176). For the analysis of fatty acids amide-linked to GlcN(I), several chemical degradation procedures are available. One comprises periodate oxidation of 0-deacylatedand 2H-reducedlipid A (lipid A-OH,). Following permethylation, g.1.c. - m.s. analysis revealed a 2-deoxy-1-deuterio-1,3-di-O-methyl2-(N-methyl-3-methoxyacylamido)glycerolderivative in which the amidelinked fatty acid at GlcN(1) was identified as 14: O(3-OH) in B. pertussis (93), and as 16 :O(3-OH) and 18 :O(3-OH) in R. trifolii (8 1). A second method uses permethylation of the dephosphated (48%aqueous HF, 48 h, 4°C) and 2H-reducedlipid A. This approach allowed the assignment of amide-bound fatty acids linked to GlcN(1) and GlcN(II), as well as the identification of the backbone structure as a HexpN disaccharide (85). Mass-spectrometricanalysis of the products was performed by using either a short g.1.c. column (0.3 X 5 cm) or by direct insertion-probeanalysis (87). In the case of C.violuceum (85), the mass spectra obtained from the permethylated HexpN disaccharide bearing attached N-methylacyl residues revealed unequivocally that both amino groups camed 12 :O(3-OH). The third approach to analyze amide-linked fatty acids in lipid A (187) is

MOLECULAR STRUCTURE! OF LIPID A

239

based on selective methylation, that is, the formation of a methyl imidate in the presence of methyl iodide and silver salts as catalyst ( 188). The intermediate acyl imidate (a Schiff s base) is highly susceptible to mild acid hydrolysis and, at pH 2.0 (20 min, 20°C),the formerly N-bound acyl or acyloxyacyl group is rapidly released as the methyl ester, whereasall ester linkagesremain intact. The liberated acyl methyl esters can be analyzed directly by g.1.c. m.s.. Using this procedure, the presence of amide-bound 14 :0[3-O(12 :O)] and 14:0[3-O(16 :O)] was, for the first time, demonstrated in lipid A of S. minnesota (1 87). Later, it was shown that amide-linked 3-acyloxyacyl residues are also present in lipid A of other Gram-negative bacteria (38). A disadvantage of this procedure is, however, the fact that only the amidelinked 3-acyloxyacyl of /3-glycosides are liberated to a satisfactory extent (188). The low yields obtained in the case of a-anomeric GlcN(1) are most likely attributable to hindrance of an association with the catalyst (Ag+). Finally, amidases from the slime mould D. discoideum (179) were found to cleave the amide-bound long chain fatty acids of S. london LPS. Interestingly, 0- and N-linked acetyl groups in the oligosaccharide region of LPS were not affected. Khorana and coworkers (1 89,190) isolated and purified two different acyl amidases from D. discoideum. The two enzymesexpressed distinct substrate specificity, one liberating specifically the N-acyl residue attached to GlcN(I) and the other the amide-bound acyl group at GlcN(11). It should be pointed out that modem spectrometric techniques, such as f.a.b. - m.s., 1.d.- m.s., or even n.m.r. if applied alone, do not allow to the elucidation of the nature of amide-linked acyl groups. This problem can be solved only by using, in addition, chemical or enzymic methodologies such as the ones just described. d. Location of Fatty Acids at the Lipid A Backbone.-The chemical methods described thus far provide a great deal of information as to the type of linkage and, therefore, the possible location of acyl groups in lipid A. However, for unequivocal allocation of the fatty acidsto individual hydroxyl and amino groups of the lipid A backbone, the informationgained by chemical analysis must be supplemented by other (physical) methods such as n.m.r., f.a.b.-m.s., Z5zCf-p.d.-m.s.,and 1.d.-m.s.. The first f.a.b.-m.s. investigations were performed on highly purified mono- and bis-phosphate lipid A preparations of S. typhimurium (41). F.a.b. - m.s. permitted the determinationof the molecularweight ofboth bis([M H]+ = 1798) and monophosphated ([M H]+ = 1700) lipid A. Furthermore, cleavage at the C-1'-0-C-6 glycosidic bond gave an oxonium ion fragment (m/z = 1087)that correspondedto GlcN(11) carrying one (dimethyl) phosphate, two 14 :0(3-OH), one 14 :0, and one 12 :0 residue (Figs. 2 and 4). Fragmentsderived from the reducing end of lipid A [GlcN(I)]

+

+

240

ULRICH ZAHRINGER et al.

were not observed in the positive f.a.b. m a s spectrum but could be calculated as the difference between the molecular mass ([M H]+ = 1700)and the fragment of the oxonium ion (m/z = 1087). Thus, in this case, GlcN(1) carried two 14:O(3-OH) fatty acyl groups, which, however, could not be assigned to specific positions 2, 3, or 4 of GlcN(1) or to one of the hydroxyl groups of the two 14 :O(3-OH)residues. Nevertheless,the f.a.b. -m.s. analysis not only confirmed the presence of acyloxyacyl residues as earlier suggested by chemical analysis (187), but also allowed, for the first time, the assignment of the two acyloxyacyl residues to the distal GlcN (11) in lipid A of S. typhimurium. It was realized in consequence that, in S. typhimurium lipid A, GlcN(I1) camed four and GlcN(1) only two acyl groups; that is, the fatty acid distribution over the two GlcN residues was asymmetrical (see later). Laser-desorption mass spectrometry (1.d. - m.s.) of dephosphated and nonpurified lipid A was introduced in studies on the acylation pattern of lipid A of S, minnesota, E. coli, and P.mirabilis (37). Although these earlier studies did not establish whether some mass peaks originated from laser-induced fragmentation or from smaller, intact molecular species (namely monomeric subunits formed during acidic hydrolysis), the first 1.d. -m.s. study, taken in combination with chemical results (38,187), revealed the location of the individual fatty acids in S. minnesota, E. coli, and P.mirabilis. The success of 1.d. - m.s. for locating acyl groups was demonstrated with P.mirabilis lipid A isolated from cells grown at physiological and low temperatures (37). Cultivation of P.mirabilis at 12°Ccaused the incorporation of A9-16: 1 at the expense of 12 :0. Using 1.d. -m.s. in combination with chemical analysis, it was possible to determine precisely the position of the A9-1 6 :1. It was found (37) that the unsaturated fatty acid positionally replaced 12:O at the 3-hydroxyl group of amide-linked 14:0(3-OH) at GlcN(11).Furthermore, the study confirmed earlier analytical investigations on the fatty acid distribution in Salmonella (187). An advantage of the 1.d. - m.s. method lies in the possibility of controllingthe degree of fragmentation by the amount and kind of alkali salts (NaI, CsI) added to the sample for cationization. In this way, both the fragmentation pattern and the heterogeneity of a preparation may be studied in detail (39). It has recently been demonstrated ( 191) that the nature and location of lipid A primary fatty acids is determined by the specificity of the enzymes UDP-GlcpNAc-0-acyltransferase and UDP-3-0-[(R)-hydroxyacyl]GlcpN-N-acyltransferasefor acyl- acyl camer protein (acyl ACP). The analysis of the acyl ACP specificity of these 0- and N-acyltransferases should, therefore, constitute a biochemical approach for elucidation of the location of primary fatty acids in lipid A ( 191).

+

MOLECULAR STRUCTURE! OF LIPID A

24 1

e. Lipid A Differing in the Number and Location of Fatty Acids. -Lipid A derived from various bacterial sources may vary in the number of fatty acids present and their location. These variations have been elucidated in lipid A derived from R-form LPS. On the other hand enterobacterialS-form LPS is heterogeneous and that fraction containing a long polysaccharide component, namely, the 0-specific side chain, contains acyl-deficient lipid A, the structure of which has not been studied thus far(47). Described here are selected but representative examples of lipids A that are classified according to the number of acyl groups they carry and to their distribution over the HexpN residues of the backbone. ( i ) Hexaacyl Lipid A.-Most lipid A studied contain up to six acyl groups. Hexaacyl lipid A is present in Enterobacteriaceae (such as E. coli), but also in such taxonomically remote bacterial families as the Neisseriaceae, and the Pseudomonadaceae(for instance, C. violaceurn).These lipid A carry the four primary 3-hydroxy fatty acids and two additional acyl groups linked to hydroxyl groups of primary fatty acids. With respect to the location of the secondary acyl groups, two types of lipid A may be distinguished, namely, lipid A having an asymmetric (E. coli type) and symmetric (C. violaceurn type) acylation pattern. As noted previously (92), bacteria possessing a symmetric acylation pattern belong to the P-subclass of Protobacteria ( 192), whereas the Enterobacteriaceae, exhibiting an asymmetric acylation of lipid A, are members of the y-subclass. (a). Asymmetrical Acylation Pattern of H e x w y l Lipid A. The classical representative of an asymmetrically acylated hexaacyl lipid A is that of E. coli Re LPS (strain F5 15). As described, this lipid A carries at positions 3’,2’,3, and 2 of the backbone, four 14: O(3-OH) residues (Fig. 2) with the hydroxyl groups of the two 14 :O(3-OH) residues bound to GlcN(11) being acylated by I2 :0 and 14 :0. Therefore, a total of six fatty acids is present, four of which are associated with GlcN(I1) and two with GlcN(1). Thus, the overall acylation pattern is asymmetric (4 2). A number of other hexaacyl lipid A were identified as being asymmetrically substituted by fatty acids as in E. coli,and these include S. typhirnuriurn (4 l), the hexaacyl species of S. rninnesota (37), P. rnirabilis (37), H. injluenzae (68), Providencia rettgeri ( 193),A. actinornycetemcornitans (94) and probably Pasteurella rnultocida (1 93a). The lipid A of C. jejuni also belongs to the class of asymmetrically acylated hexaacyl species. It should be noted, however, that GlcN(I1) is replaced here by GlcpN3N (97). In asymmetrically acylated lipid A, the four primary hydroxy-fatty acids possess the same chain length of 14 carbon atoms [thus far, with one exception ( 193b), only 14 :O( 3-OH) was identified].

+

ULRICH ZAHRINGER et al.

242

- 0,

II

P

II ,o-

o,

HO'

0

FIG.10.-Chemical structure of lipid A of Chromobacterium violaceurn. The dashed line indicates nonstoichiometrica-hydroxylation(85) of 12 :0. For details, see the text, and for substitutents of phosphate residues, see Table I.

(b). Symmetrical Acylation Pattern of Hexaacyl Lipid A. In the case of C. violaceurn (Fig. lo), the lipid A backbone also carries four primary acyl groups [O-linked 10 :O(3-OH) and N-linked 12 :0(3-OH)]. The secondary fatty acids present [approximately 2 mol of 12:O plus 12:0(2-OH)] are linked to the hydroxyl groups of the two 12 :O( 3-OH)residues of which one is associated with GlcN(1) and one with GlcN(I1) (38,85). In this instance, therefore, GlcN(I) and GlcN(I1) each carry three fatty acids, resulting in a symmetric acylation pattern (3 3). Other lipid A examples possessing a symmetrically acylated backbone are N. gonorrhoeae(33), N. rneningitidis (73), the hexaacyl species of P.aeruginosa (77),R. gelatinosus (84), R. gelatinosa (194), S. natans (92), and, most likely, B. pertussis ( 139)and X . sinensis (38). In symmetrically acylated lipid A, the two ester-linked and the two amide-linked primary 3-hydroxy fatty acids possess (with the exceptions of R. gelatinosus, R. gelatinosa, and S. natans)different chain lengths, the 0-bound acyl group having 10 or 12 and the N-bound acyl group possessing 10, 12 or maximally 14 carbon atoms. It should be noted that, in C. violaceurn lipid A, the amide-linked 12 :O(3-OH) at GlcN(11) carries either 12 :0 or 12 :0(2-OH), yielding two molecular species of lipid A (Fig. 10).

+

MOLECULAR STRUCTURE OF LIPID A

243

(ii) Heptaacyl Lipid A.-Lipid A species carrying seven fatty acids have been encountered in Re LPS of S. minnesota R595 (174,187), S. typhimurium (195), E. carotovora (99, and P. mirabilis R45, (37,72) but also in a temperature-sensitivemutant of E. coli (195a). These heptaacyl forms exhibit the hexaacyl E. coli type of acylation pattern but have, in addition, one 16 :0 residue bound to the 3-hydroxyl group of 14 :0(3-OH), which is amide-linked to GlcN(1) (Fig. 4). It must be noted, however, that the heptaacyl species often represents a minor (20- 3096, w/w) lipid A component, whereas the major species is a hexaacyl form resembling, in its acylation pattern, E. coli type lipid A. Heptaacyl S. minnesota lipid A has been chemically synthesized (196). In lipid A of Moraxella catarrhalis an unusual lipid A was identified possessing an asymmetrical (3 4)acylation pattern with four 12:0(3-OH) residues as primary fatty acids, three of which are acylated to form acyloxyacyl residues [ 12:0(3-O(10:0)] at positions 2’ and 3 and [ 12:0(3-0(12:0)] at position 2 (193b).

+

(iii) Pentaacyl Lipid A. -Among the pentaacyl lipid A structuresstudied thus far are those in which only one of the four primary acyl groups is 3-0-acylated [B.fragilis(79),R. sphaeroides(87,88),R. capsulatus(89), and precursor Ib (173)l. On the other hand, lipid A are now known in which a major species lacks one of the primary fatty acids (P.aeruginosa) (77). In the major component of B. fragilis lipid A, five comparatively longchain and, in part, iso-branched fatty acids are present (79). Their location is shown in Fig. 5. It is evident that the only secondary acyl group present is bound to the 3-hydroxy fatty acid at position 2’ [GlcN(II)]. It is a peculiar feature of B.fragilis lipid A that the hydroxyl group at position 4’ ofGlcN(11) is free; that is, that the nonglycosylic phosphate group is absent. A pentaacyl lipid A is also found in R. sphaeroides (Fig. 7) (88). This lipid A has the same distribution of fatty acids as B. fragilis (79) but, as a notable feature, it contains two rare types of fatty acids, namely, A’-14 : 1 and 14 :O(~-OXO), the former being present as a 3-acyloxytetradecanoic acid group amide-linked to GlcN (11), the latter being amide-bound to GlcN(1). Furthermore in R. capsulatus (Fig. 7) a pentaacyl lipid A having two 14 :O(3-0x0) groups in amide linkage constitutes a main component (89). Two residues of 10 :0(3-OH), of which the one bound to GlcN(I1) carries dodecenoic acid (A5-12: 1) at its 3-hydroxyl group in nonstoichiometric amounts, are ester linked. The location of R. capsulatusfatty acids, however, differs significantly from that of the B. fragilis and R. sphaeroidespentaacyl lipid A in that the secondary fatty acid (12 : 1) is bound to the ester-linked 3-hydroxyfatty acid at position 3‘. Another fraction of R.capsulatuslipid A lacks 12 : 1 and, thus, exemplifies a tetraacyl lipid A species.

244

ULRICH ZAHRINGER et a1

A representative of a pentaacyl lipid A, in which the only secondary fatty acid present is associated with GlcN(I), is precursor Ib (Fig. 9 B). This compound was isolated from a temperature-sensitive mutant of S. typhimurium (mutant Ts5) (173).The only 3-acyloxyacylresidue in precursor Ib is 14 :0[3-0( 16 :O)], which is amide-linked to GlcN (I) (position 2). A pentaacyl lipid A is also present in P.aeruginosa (77) (Fig. 1 1A). Here, the main lipid A species contains a total of five fatty acids, and a minor hexaacyl species (Fig. 1 1B) corresponds structurally to lipid A of C. violuceum (Fig. 10). The prominent pentaacyl component, which makes up approximately75%(w/w) of P.aeruginosa lipid A, encompasses three structural forms that all possess the same p-( 1 + 6)-linked GlcpN backbone, but with only three (primary)3-hydroxyfatty acids attached to positions 3’, 2’, 3, and 2 (Fig. 11A). These structural forms differ from each other by the 3-0-acylation of each of the two amide-linked 12 :O(3-OH) residues by the secondary acyl groups 12 :0 or 12 :0(2-OH), as indicated by the dashed lines. Of the four conceivable structural types, the one bearing two 12 :O(2-OH) residues is not present. ( i v ) Tetraacyl Lipid A. -Tetraacyl lipid A constitute a minor species of, for example, enterobacterial lipid A preparations (32). They may be the result of incomplete biosynthesis or arise from the acid treatment applied to LPS for the preparation of free lipid A. In this sense, mature tetraacyl lipid A does not exist. However, a tetraacyl lipid A partial structure (Fig. 9 A) is produced by a S. typhimurium mutant having a conditional defect in the biosynthesis of Kdo (197). This lipid A partial structure, constituting a precursor in lipid A biosynthesis,was first termed acidic precursor ( 122,123) and (around 1982) precursor Ia. Unfortunately, the same molecule, also isolated from a S. typhimurium mutant, was later termed precursor IV or lipid IV (63,198). In order to distinguish the tetraacyl compound from an associatedand I6 :0-containingpentaacyl form (the latter is called precursor Ib or lipid IVB),the terms precursor Ia (199) and lipid IV, (63) were introduced and used synonymously from 1985 on. In view of the fact that it was the first introduced into the literature, the term precursor I (with the distinction between precursor Ia and precursor Ib) will be used in this article. Precursor Ia is composed of a p-~-GlcpN-(1 6)-a-~-GlcpN1,4’-bisphosphate carrying only primary acyl groups, namely, four 14 :O( 3-OH) residues, two in amide (2’ and 2) and two in ester linkage (3’ and 3), respectively. Lipid A precursor Ia has been chemically synthesized (compound 406 or LA-14PP) (200).

-

f. Concluding Remarks. -Fatty acids constitute important components of lipid A and confer the hydrophobic properties characteristic of the mole-

MOLECULAR STRUCTURE OF LIPID A

245

A 0 -0

II ‘ P’ /

HO

B -0,

0 IIo, P

HO’

FIG.1 1. -Chemical structure of the two preponderant lipid A forms of Psatdomonus ueruginosu. (A) Pentaacyl lipid A (major lipid A fraction; 75%,w/w). (B) Hexaacyl lipid A (minor lipid A fraction, 2596, w/w). Dashed lines indicate nonstoichiometrica-hydroxylation of 12 :0. A lipid A species having 2 mol 12 :0(2-OH)/moI lipid A was not detected (77).

246

ULRICH ZAHRINGER et al.

cule. With regard to the nature, linkage type, size, and number of the fatty acids as well as with regard to their location in lipid A, the following generalizations may be made. Lipid A contains primary fatty acids directly linked to hydroxyl and amino groups of the backbone, and secondary fatty acids bound to hydroxyl groups provided by the primary acyl residues. The number of carbon atoms of primary and secondary fatty acids is, in the majority of lipid A studied, in the range of 10 to 18. They are, in general, saturated, even-numbered, and straight-chain fatty acids and in only few cases, are unsaturated, odd-numbered, and iso- and ante-iso branched derivatives present in molar amounts. The primary fatty acids are (R)-configured 3-hydroxy fatty acids, but 3-OX0 fatty acids [thusfar only 14 :O(3-0xo)l have also been identified in lipid A. The primary 3-hydroxyacyl groups can be involved in both ester and amide linkage, whereas 14:O(3-0x0) is only amide-bound. 3-Hydroxy fatty acids are present exclusively as primary acyl groups, the only exception known so far being K cholerae (86) lipid A. Often, only one type of 3-hydroxy fatty acid is present [ 14 :O(3-OH) in Enterobacteriaceae], but examples in which lipid A contains two or more different 3-hydroxy fatty acids (C. violaceurn, and B. fragilis) are also known. If there are several homologous 3-hydroxylatedacyl groups, the one having the longest chain is amide-bound (4), the other@)being engaged in ester linkage (P. aeruginosa, and N. gonorrhoeae). In general, the chain lengths of the amide-bound 3-hydroxy fatty acids at HexN(1) and HexN(I1) are identical (E. coli), but they may also differ as in B. ji-agilis (79), X . sinensis (78), M . fulvus (201), and R. trifolii (202). If the amide-bound fatty acids associatedwith HexN(1) and HexN(I1) differ in the number of carbon atoms, the one having the longest chain is linked to HexN(1) [for example, GlcN(1) in B.fragilis (79)]. The secondaryfatty acids are particularly composed of nonhydroxylated fatty acids, but 2-hydroxylated fatty acids are also found. The 2-hydroxy fatty acids always possess the (3-configuration and, together with the nonhydroxylated fatty acids of the same chain length, they often add up to 1 mol eq. (as in S. rninnesota). Up to three secondary fatty acids may be linked to the 3-hydroxy fatty acids at positions2,2’, and 3’. With one exception (193b) the 3-hydroxy fatty acid at position 3 [HexN(I)] is never 3-0-acylated. In lipid A possessing two secondary acyl groups, these are linked either to the 3-hydroxy fatty acid at position 2’and 3’ (E. coli type asymmetric acylation) or to those at positions 2’ and 2 (C. violaceurn type symmetric acylation). If only one secondary acyl group is present, it may be linked to the 3-hydroxy fatty acid at position 3‘ (R. capsulatus), position 2’ ( B. ji-agilis and R. sphaeroides) or position 2 (precursor Ib). Secondary fatty acids are absent from the tetraacyl lipid A partial structure precursor la. Among the lipid A-associated fatty acids, 28 :O(27-OH) or 29 :O(28-OH) and 26 :O(25-OH) deserve special mention, as they deviate considerably

MOLECULAR STRUCTURE OF LIPID A

241

from the former in chain length and position of hydroxylation. These fatty acids possess twice the number of carbon atoms as the usual fatty acids present in, for example, enterobacterial lipid A. They extend completely through the outer membrane and may be important for a stable bilayer arrangement. The penultimately located hydroxyl group may be substituted by a secondary acyl group, thus contributing to the rigid organization of the outer membrane system. These long chain fatty acids have only recently been identified. Further studies are necessary to see whether they adhere to the general rules presented here, to identify their stereochemistry, and to learn more about their taxonomic, biological and physicochemical significance. Interestingly,they are associated with members ofthe a-2 subgroupof bacteria based on homologies of their 16s rRNA (203). The presence and estimation of 28 :0(27-OH), therefore, is of great taxonomical value (10 1,160). 4. Hydroxyl Group in Position 4 of the Lipid A Backbone

The presence of a free, that is, unsubstituted,hydroxyl group at the lipid A backbone was indirectly postulated after the discovery of amide-linked 14:0[3-0(12:0)] and 14:0[3-0(16:0)] in S. minnesota R595 lipid A (4,187). It was known that phosphate was bound to the hydroxyl group at position 4’, and that Kdo occupied one hydroxyl group, and because 1 mol of each 14:0[3-O( 14 :O)] and 14 :O(3-OH) was known to be ester-linked, it was obvious that one out of the three available hydroxyl groups was not substituted. However, the question as to which hydroxyl group of the lipid A backbone was free remained open for some time. The elucidation of the enterobacterial lipid A biosynthesis and the identification of lipid X “2deoxy-2-[(R)-3-hydroxytetradecanamido]-3-0-[(~)-3-hy~oxytetradecanoyll-a-D-glucopyranosylphosphate,that is a-~-GlcpN1-phosphate acylated at positions 2 and 31 as a precursor of lipid A biosynthesis rendered the hydroxyl group at position 4 of GlcN(I) a likely candidate. That indeed this hydroxyl group was free was unequivocally shown by Imoto et al. (35), who isolated from E. coli Re LPS (strain F5 15)a monophosphatedlipid A, part of which, after esterifying the phosphate groups with diazomethane, was subjected to two-dimensional ‘H-n.m.r. spectroscopy (‘H,’H-COSY). A further sample was, in addition, peracetylated (whereby five acetyl groups were introduced into the molecule) and also analyzed by H-n.m.r. spectroscopy. Comparison of the ‘H-n.m.r. data of both lipid A derivatives revealed a significant downfield shift of about 1.5 ppm for the H-4 signal (3.55 vs. 5.O 1 ppm) before and after 0-acetylation, whereas all other signalsremained unchanged. In this way, for the first time, the authors could clearly demonstrate that position 3 of GlcN(1) [and position 3’ of GlcN(II)] was acylated, whereas the hydroxyl group at C-4 of GlcN(I) was free. Unfortunately, the expected shift of the primary hydroxyl group in position 6’ (compareSection

248

ULRICH ZAHRINGER et al.

111.5) was not observed by this approach, as the signals of H-6a’ and H-6b’ could not be unequivocally assigned because of the overlapping with other signals in the region of 3.7-4.2 ppm. Takayama and coworkers (60) introduced the h.p.1.c. separation technique for such amphiphilic molecules as lipid A, and in earlier experiments they applied paired-ion reverse-phase h.p.1.c. for the preparation of homogeneous fractions deriving from 4’-monophosphated lipid A of S. typhimuriurn. The purified preparations obtained were suitable for f.a.b. - m.s. analysis. However, monophosphated lipid A isolated in this way expressed a considerable heterogeneity with respect to the number and location of 0acyl residues (60).In order to further improve the purification procedure, as well as to obtain lipid A derivativessuitable for n.m.r. spectroscopy, Qureshi et al. (174) prepared the dimethyl phosphate derivative of S. minnesota (R595)lipid A, which, after purificationby reverse-phase h.p.1.c. (C,8),could be analyzed by H-n.m.r. The n.m.r. spectrum of, for example, the heptaacyl lipid A dimethyl monophosphate fraction, unequivocally revealed 0-acyl substitution [ 14 :O(3-OH)Jat position 3 and a free hydroxyl group at position 4 of GlcN(1). The problem of determining unsubstituted free hydroxyl group(s)located at the lipid A backbone still remains a challenge in analytical chemistry. This is mainly because of the fact that this problem cannot be solved by classical methylation analysis (Hakomori procedure), because the alkaline pH of the reaction conditions causes removal of 0-acyl residues and, thereby, artificially creates free hydroxyl groups. Therefore, a new procedure to determine free hydroxyl group(s) in lipid A, using methylation analysis, has been recently introduced by our laboratory (77). The method is based on a methylation procedure originally described by Fugedi and Kovacz (204). By this method, free hydroxyl groups are transformed into their methyl ethers via electrophilic substitution with trimethyloxonium tetrafluoroborate [Me,O+ BF,-]. The reaction is performed in dichloromethane for 16 h at 20°Cin the presence of a sterically hindered base (2,5-di-tert-butyl-4-methyl-pyridine). In contrast to other methylation procedures (such as Hakomori methylation), these conditions allow free hydroxyl groups to be transformed into methyl ether derivatives without affecting or cleavingester bonds. On applying this method to monophosphated lipid A of S. rninnesota (R595) and E. coli (F5 15), followed by acetolysis, NaBH,-reduction, and per-0-acetylation, g.1.c. - m.s. analysis revealed 1,3,5,6-tetra-O-acety1-2-deoxy-4-0methyl-2-(N-methylacetamido)-~-glucitol to be the only derivative of GlcN(1). This result shows that, in the lipid A preparations studied, the hydroxyl group in position 4 of GlcN(1) was free. This procedure was also applied to lipid A of P. aeruginosa (77) and N.rneningitidis (73), in which cases the hydroxyl group at C-4 was also found to be unsubstituted. Another procedure for determining free hydroxyl groups applies diazo-

MOLECULAR STRUCTURE OF LIPID A

249

methane (CH,N,) with SiO, as a catalyst (205,206). This method was used to demonstrate the presence of free hydroxyl groups at position 4 (and 6’; compare Section 111.5) in, for instance, lipid A of R. sphaeroides ATCC 17023 (87). This convenient procedure, however, suffers from the great disadvantageof not giving quantitative yields of methyl ethers. It, therefore, does not allow the characterization of the degree of substitution of hydroxyl groups of the lipid A backbone and determination of whether the hydroxyl groups are free. One example in which the C-4 hydroxyl group of lipid A is substituted is known. Thus, in R.tenue (90), nonacylated GlcpN is glycosidicallylinked to position 4 of GlcN(I). It must be emphasized that the problem of unsubstituted hydroxyl groups is usually studied employing free lipid A prepared by treatment of LPS with acid. The demonstration of a free hydroxyl group at C-4 in monophosphated lipid A therefore does not exclude the possibility that, in LPS, a substituent, present in acid-labile linkage, could be bound to 0-4 of lipid A. This possibility has thus far been excluded for E. coli Re LPS which was analyzed by ‘H- and I3C-n.m.r.and shown to contain an unsubstitutedhydroxyl group at C-4 of GkN( I) (34,110).

5. Hydroxyl Group in Position 6’ as the Attachment Site of Kdo Based on results obtained by periodate oxidation of deacylated S. minnesota Re (R595) LPS, it was proposed (1 12)in 1971 that the hydroxyl group at position 3’ of GlcN(11) served as the attachment site of Kdo to lipid A. This structural proposal was corroborated by other groups, who investigatedthe core-backbonestructure by various degradation procedures (69) and by ‘Hand 31P-n.m.r.spectroscopy (109). However, methylation analysis of the deacylated core (Kdo)- lipid A backbone oligosaccharideobtained from a P. mirabilis Re mutant (strain R45), in combination with g.1.c. -m.s. of partially methylated glucosaminitolacetates (72), revealed l75,6-tri-0-acetyl-2deoxy-3,4,5-tri-0-methyl-2-(N-methylacetamido)-~-gluc~tol as the major product deriving from GlcN(I1). The presence of an acetyl residue at the primary hydroxyl group at position 6’ of (the reduced) GlcN(II), as well as the presence of a methyl group in position 3, clearly showed that an acid-labile substituent, namely, Kdo, had been linked to position 6’ and not to position 3’ as previously assumed. Also in 1983, ‘H- and I3C-n.m.r. spectroscopic investigationson LPS of a Re mutant of E. coli LPS (35,66) suggested that Kdo was linked to the hydroxyl group at position 6’ of GlcN(11) of the lipid A backbone. Again at approximately the same time it was discovered that lipid X represented a monomeric precursor of lipid A biosynthesis(207). The known location of lipid A acyl groups (positions 3’, 2’, 3 and 2) and phosphate residues (positions 4’ and 1) suggested that Kdo was linked to the primary hydrsxyl group at position 6’. Unger and coworkers (70) investi-

250

ULRICH ZAHRINGER et al

gated the structure of a (Kdo), - (GlcpN), tetrasaccharide derived from S. minnesota Re (R595)LPS by hydrazinolysisand h.p.1.c. purification. Based on the 13C-n.m.r. data, it was suggested that Kdo was linked to 0-6’ of GlcN(II), although the site of attachment of Kdo to the backbone could not be determined unequivocally. This is because a-ketosidically linked Kdo residues give rise to only a small and nondiagnostic glycosidic shift in the range of only 1- 2 ppm (glycosidic shifts are usually 8 - 10 ppm downfield) of the C-6’ signal (64). In addition, such model compoundsas methyl a-and P-ketosides of Kdo exhibit nondiagnostic chemical-shift differences for the determination of the anomeric linkage at (2-2, these differences being less than 1 ppm (208). In summary, these investigations strongly indicated that Kdo was linked to 0-6’ of GlcN(11)of lipid A, but final experimentalproofwas lacking. Also, the anomeric configuration of the lipid A-proximal Kdo residue was unknown. This is mainly because, in contrast to normal aldoses, the anomeric linkage of Kdo to lipid A cannot be determined by simple ‘H-n.m.r. analysis because of the lack of a proton at the anomeric center. This problem was first addressed (65) in E. coli (mutant D31m4) by applying W-n.m.r. In the proton-coupled 13C-n.m.r. spectrum of Kdo, the heteronuclear coupling constant 3JC1,H.3ax was determined to be = 5 Hz for the p-Danomer (209), whereas it appeared as a sharp singlet (3JG,,H-3an) < 1 Hz when the a-D anomer was present (64). This results from the stereochemical arrangement of the carboxyl group at C- 1 and the methylene protons at C-3 exhibiting a trans-relationship for the 8-D(3J~1,~.3ax = 5 Hz) and a gauche orientation for the a-D anomer (3JCI,H-3ax < 1 Hz). The proton-coupled 13C-n.m.r. spectrum of E. cdi Re LPS (triethylammonium salt) showed a 3JG1,H-3ax value of = 3 Hz (after subtraction of computer-generatedline broadening), which nevertheless led the authors to conclude that the Kdo linkage was a (65). Unequivocal proof for the attachment site to lipid A and the configuration of Kdo was provided in 1985 by investigations on an 0-deacylated, per-0acetylated, carboxy- and phosphate-methylated,and highly purified derivatives isolated from LPS of an Re mutant (F5 1 5) of E. cali (34) (Fig. 12). This compound was subjected to chemical analysis, 1.d.-m.s., and ‘H- and I3Cn.m.r. spectroscopy.From the chemical-shiftvalues of the H3ax and H-3eq protons in ‘H n.m.r., as well as from the observation of large differences in the chemical shift of the methylene protons (H-8a and H-8b) of the individual Kdo residues [which are known to be characteristic for a-ketosides (210)],and from the signals in the 13C-n.m.r.spectrum, the structure of the Kdo-lipid A backbone was completely elaborated as shown in Fig. 12. According to this study, Kdo is in a-D-ketosidic linkage attached to the hydroxyl group in position 6’ of GlcN(I1) of the lipid A backbone. This

25 1

MOLECULAR STRUCTURE OF LIPID A

OMe AcO

0

Me0,II Me0

/

p’

,OMe

0 AcO

AcO

I

R

0

\

I R

0

0 Me

QAc

R=

FIG.12.-Chemical structure of an 0-acetylated N,N’-bis(3-acetoxytetradecanoyl)tetrasaccharide tetramethyl phosphate isolated from the E. coli Re mutant F5 15 (34). [For details see text.]

structural arrangement was later confirmed by Qureshi et al., who purified and investigated the hexaacyl LPS species of the E. coli Re mutant D3 1m4. (1 10). An interesting case concerning the core -lipid A linkage-region was encountered in the LPS ofA. calcoaceticus(75). Here, in part of LPS, the link between the polysaccharide region and lipid A is not mediated by Kdo. Instead, the isosteric ~-glycero-~-talo-2-octulosonicacid ( ~ g ~ t - K ois) present, which, like Kdo, is a-ketosidicallylinked to C-6’ of GlcN(11)of lipid A. Interestingly, the a-ketosidic linkage of DgDt-KO to lipid A is highly resistant to acid hydrolysis, a fact most likely related to the presence (75) of the axial hydroxyl group at C-3. DgDt-KO has now been identified as being linked to the lipid A - proximal Kdo residue in Pseudomonaspseudomallei (21 1) and P. cepacia (2 12). In conclusion, only in few LPS has the Kdo-lipid A linkage been elucidated unequivocally. In these examples, Kdo is a-ketosidically linked to 0-6’ of HexN(11)of the lipid A backbone [E.coli (34,65),H. influenzae (68), an S. minnesota R595 recombinant carrying the genus-specific antigen of Chlamydia spp. (213), and S. minnesota (70)]. The same type of linkage, however, is very likely to exist also in other Gram-negativebacteria, including P. mirabilis (72), B.fragilis(79), P.aeruginosa (77), V.parahaemolyticus (83), N . meningitidis (73), Rhodobacter, Rhodocyclus, and others. It is note-

252

ULRICH ZAHRINGER et al.

worthy, that also in lipid A containing a GlcpN3N disaccharide [C. jejuni (97)] or a GlcpN3N monosaccharide backbone [P.immobile (lOS)], the hydroxyl group at position 6’ serves as the attachment site of Kdo. IV. SYNTHETIC LIPIDA Based on the results of chemical analysis, free lipid A has been chemically synthesized [compare(12 - 15,17)].The first fully synthetic lipid A molecule (preparation 506 or LA-15-PP) corresponds in structure to E. coli lipid A (Fig. 2) carrying a glycosylic (and nonglycosylic) monophosphate. Later, other lipid A and lipid A partial structures were prepared that contain a p-( 1 46)-linked GlcpN disaccharide, but that differ in the acylation and phosphorylation pattern. These preparations include the heptaacyl species of S. minnesota lipid A (compound 516 or LA-l6-PP, Ref. 196) and P. mirabilis lipid A (compound A-403, Ref. 2 14), the pentaacyl precursor Ib (LA-2O-PP,Ref. 172,Fig. 9B), and the tetraacyl precursor Ia (406 or LA- 14PP, Ref. 215, Fig. 9A). In addition, the 4’-monophosphate (for example, compound 504 or LA- 15-PH) and the 1-monophosphate (505 or LA- 15HP) partial structures of E. coli lipid A, and such acyl-deficient compounds as preparation LA-19-PP, corresponding to 0-deacylated E. coli lipid A, were prepared. Further, C. violuceum-type lipid A (LA-22-PP) containing 0-linked 14 :O( 3-OH) and N-linked 14 :O[ 3-O( 14 :O)] in a symmetrical distribution, and partial structures thereof, was synthesized (15). Also, such lipid A disaccharide analogues as the pentaacyl compound 316, having an acylation pattern different from that of lipid A, have been prepared by chemical synthesis (2 16). A, phosphonooxyethyl ( 15) compound, a 3-ether (2 17), and fluorinated (218) analogueshave also been synthesized. Finally, a great number of monosaccharide partial structures having different acylation and phosphorylation patterns have been prepared by several groups (compare see Refs. 1 1,13- 16,19). These compounds include synthetic counterparts of lipid A-related bacterial products, such as lipid X and lipid Y , as well as other partial structures and analogues, correspondingto either the reducing or the distal HexpN unit of the lipid A backbone. The 4’-monophosphate partial structure of bacterial and synthetic precursor Ia (2 15) and Ib (172), E. coli hexaacyl lipid A (33, and S. rninnesota heptaacyl lipid A (196)were compared by ‘H-n.m.r., t.l.c., and other methods (2 19),and were found to be structurally identical. OF LIPIDA V. CONFORMATION

In parallel with the increasing knowledge on the primary structure of lipid A, calculations and physical investigations have been performed in order to

MOLECULAR STRUCTURE OF LIPID A

253

gain insight into the molecular shape and the three-dimensional arrangement of lipid A and LPS. 1. Molecular Shape

Single crystals of free lipid A or LPS are as yet not available. Therefore, the most promising approach to obtain molecular models is to perform theoretical calculations.After the chemical structures of enterobacteriallipid A had been elucidated,this methodologywas successfullyapplied with heptaacyl 5'. minnesota lipid A (220)and hexaacyl E. coli Re LPS (221). As an example, Fig. 13 shows the atomic model of the E. coli lipid A molecule, as calculated by Kastowsky et al. (22 1) using energy-minimizationtechniques. The most striking property of the model is that the lipid A backbone is not orientated perpendicular to the direction of the parallel-aligned fatty acids, but that it displays an angle tilt in the range of 53 & 7 '. This tilt leads to an energeticallyfavored close packing of the acyl chainsand to the consequence that the fatty acids do not end in one plane (perpendicularto the z-axis) as is the case for phospholipids. As a further result, the fatty acids may be interdigitated, as proposed from neutron-scatteringdata (222). It should be menO

A

u

FIG.13.-Ball-and-stick (A) and space-filling(B) atomic models ofE. culi lipid A. One out of several possible conformers is shown. The ring plane (right)of GlcN(I1) almost coincides with the plane of the paper (221). Courtesy of Dr. Manfred Kastowsky, University of Berlin.

254

ULRICH ZAHRINGER et al.

tioned that a similar tilt of the lipid A backbone had also been calculated for free lipid A of S. minnesota (220).Furthermore, the primary hydroxyl group at position 6’ of GlcN(TI), which carries in LPS the polysaccharide region, is exposed on the upper surface of the molecule. The close packing of the acyl groups associated with the inclination of the lipid A backbone with respect to the fatty acid orientation seems to constitute a common and characteristic feature of the lipid A conformation. This specific(endotoxic)conformation is very likely to influence greatly the tendency of the amphiphilic lipid A to adopt peculiar supramolecular structures. 2. Three-DimensionalOrganization LPS and free lipid A are amphiphiles, and, therefore, they are not present as individual molecules in aqueous solution above the critical micellar concentration, but rather as aggregates that may adopt different three-dimensional structures. Among the supramolecular arrangements that may be assumed by amphiphiles are micellar, lamellar, hexagonal, and nonlamellar cubic, as well as inverted-hexagonal,structures (223). Factors that influence the formation of aggregates comprise intrinsic parameters such as the primary chemical structure, and extrinsic ones such as temperature, pH, and the concentrationof such bivalent cations as Mg2+and Ca2+.On elevation of the temperature, within one type of structure, transitions can take place at a phase transition temperature (T,)from the gel (p)to the liquid-crystalline(a) state of the acyl chains, reflecting an increase of the fluidity hydrophobic region of the amphiphiles. The /3 Q phase transition may be associated with a transition between different three-dimensional structures. In order to learn about the phase states adopted by LPS and lipid A, Fourier-transform infrared spectroscopy, differential scanning calorimetry, and X-ray small-angle diffraction with CuK, or synchrotron radiation have been applied. In the following section, some recent results are summarized.

-

a. Three-DimensionalStructure. -For free lipid A of S. minnesota and E. coli, complete phase diagramswere established;that is, the dependence of the structuralpolymorphism on temperature, water content, and Mg2+concentrations was determined (224- 226). It was found, as depicted schematically in Fig. 14, that the a acyl chain-melting transition is in these studies,under all conditions, connected with a change in the supramolecular structure of the lipid assembly. Within the transition range, different phases usually coexist. In the gel (p) state, cubic phases (Q) preponderate at high water content (>60 %) and at high [lipid A] :[Mg2+]ratios. In the range of low water concentration,the lamellar phase (L) is the exclusive structure at all Mg2+concentrations. In the liquid-crystalline(a)state, the hexagonal H, phase is predominant. Its contribution, however, is weak at low Mg2+con-

-

MOLECULAR STRUCTURE OF LIPID A

255

I

lamellar

I I

I t

TC

Temperature

FIG.14.-Phase diagram (227) of free lipid A.The phase diagram was established using S. minnesota Re LPS-derived lipid A.

centration and high water content, but becomes prominent at higher Mg2+ concentration and lower water content. Most importantly, under near-physiological conditions of water content, divalent cation concentration, and temperature, the lipid A assemblies adopt cubic structures almost exclusively (226). A similar polymorphism was observed for Re mutant LPS (227), and for this as well as for free lipid A, the /3 a chain-melting transition and the transitionsbetween different structuresare reversible with temperature. In other studies applying small-range X-ray or neutron scattering, only lamellar structureswere found for free lipid A (2 19,228).The pentaacyl lipid A analogue 3 16,which cames two 14 :O(3-OH) residues in position 2 and 2’, and 14 :0 groups in position 3,4, and 6’, was found to have its acyl chains in an a (liquid)-type arrangement and to adopt a nonlamellar inverted (H,) structure (2 19).The monosaccharidiclipid X, on the other hand, appears to aggregate in micellar structures, with its two acyl groups [ 14 :0(3-OH)] being very mobile and in the a-phase (229). Further details of the threedimensional structure of lipid A are comprehensively reviewed by Seydel and Brandenburg (227a) and Naumann et al. (228). A clear correlation between the supramolecular structure and endotoxic activity has been found recently using synchrotron radiation X-ray diffraction and Fourier-transform infrared spectroscopy (229a). LPS or lipid A expressing lamellar structures were biologically inactive, whereas lipid A expressing a strong tendency to form nonlamellar (cubic or hexagonal) structuresexhibited strong endotoxicity. It should be noted that at present it cannot be decided whether bioactivity of endotoxin is expressed by the aggregate described (229b) or by a single endotoxin molecule having a con-

-

256

ULRICH ZAHRINGER et al.

formation which leads-at higher concentration-to the supramolecular structure observed. Recent studies appear to favor the concept that monomeric structuresare involved in the mediation of endotoxicity (229c,229d).

b. Phase States (Fluidity).-We have previously shown that, in the S. minnesota S- and R-form LPS series, the values of T, are lowest for Re mutant preparations(around 30°C)and that they increasewith the length of the polysaccharide portion, being very high for S-form LPS, which demonstrates the T, around 37 -40°C (228,230). Surprisingly, free lipid A possesses the highest T, value (approximately 45°C). The phase transition temperatures of the various mutant LPS and lipid A are significantly higher than those of the phospholipid mixture composing the inner leaflet of the outer membrane. At physiological temperature (37°C)membranes made of this mixture are exclusively in the a-state and, thus, show a high fluidity. In contrast, the phase-transition temperature of Re mutant LPS, S-form LPS, and lipid A are just below, in the range, or even above this temperature, respectively, leading to a low fluidity in membrane systems made up from these compounds. Therefore, it should be expected that the exclusive presence of LPS in the outer leaflet of the asymmetric outer membrane leads to a lower fluidity, thus contributing to its permeabilitybarrier function. Addition of Mg2+ or Ca2+and lowering of the pH cause a significant rigidification of the acyl chains of free lipid A and LPS preparationsand lead to an increase in T,. Under alkaline conditions(pH > 7), a fluidizationof the LPS and lipid A acyl chains, and a decrease in T,, takes place. Free lipid A and all R-mutant LPS preparations investigated exhibit an extremelystrong lyotropic behavior, that is a strong dependence of the p cx chain-melting transition on the water content. For example, the lipid A phase-transition is clearly observed (231) only at water concentrations higher than 50 - 60%.

-

VI. ENDOTOXOCITY OF LPS AND LIPIDA

LPS are endowed with an overwhelming spectrum of biological activities expressed either in vivo after injection into mammalians or in humoral or cellular in vitro systems ( 1,3,4,6,20,118).In fact, LPS has been postulated as playing a central role in the pathogenesis and manifestations of Gram-negative infection, in general, and of septic shock, in particular. Therefore, one of the driving forces in LPS research has been the search for that region and structure of the macromolecule that is responsible for the induction of the endotoxic reactions. 1. Lipid A, the Endotoxic Center of LPS

Almost 40 years ago, Westphal and Luderitz postulated that the lipid A component constitutes the endotoxic principle that is responsible for the

MOLECULAR STRUCTURE OF LIPID A

257

induction of the manifold pathophysiological effects of LPS (28). This hypothesis was based, in part, on the finding that all LPS, independent of their bacterial origin, showed comparable endotoxic activity and that they all, despite their different polysaccharide compositions, contained lipid A. Duridg the past 20 years a large body of data confirming this idea has been accumulated, mainly through the work of Galanos et al., who showed, for example, that water-soluble, polysaccharide-free,free lipid A exhibits potent endotoxic activity (1,3 - 6,232). The successfulchemical synthesis of lipid A and partial structuresthen offered the possibility of confirmingthis concept. By studyingthe biological activity of such synthetic preparations in comparison to natural products in various in vitro and in vivo systems, relationships between the chemical structure and biological activity could be established at the molecular level. Lethal toxicity, local Shwartzman reactivity, adjuvanticity, induction of interferon, tumor necrosis, endotoxin (cross) tolerance, and, in particular, fever-inducing capacity (pyrogenicity) were measured as typical in vivo endotoxin activities. The effectiveness of the preparations was tested in vitro in humoral systems (such as complement activation and gelation of the Limulus amebocyte lysate), and in cellular ones including murine B-lymphocyte mitogenicity and, in particular, activation of macrophages associated with the release of such bioactive mediators as prostaglandins, and leukotrienes, IL-l, IL-6, and TNFa (167,168,230,233). The results from all biological assays performed showed that chemically synthesized E. coli lipid A (compound 506 or LA-15-PP) expresses, with similar doses, the same spectrum of endotoxic effects as bacterial (E. coli) free lipid A (5,234-237). Thus, lipid A constitutes the lethal, pyrogenic, leukopenic, and mediator-inducing, that is, the endotoxically essential region of LPS, its endotoxic properties being embedded in a molecule having the structure shown in Fig. 2. The lipid A of E. coli possesses a complex structure and the question arose as to which of its constituentsis important for, or contributes to, bioactivity. To answer this question a great number of lipid A partial structures were analyzed, again in various in vivo and in vitro systems, for endotoxicactivity. The results of the in vivo studies, with emphasis on fever-producing capacity, may be summarized as follows (1 16,118,235,238). HexpN monosaccharide compounds largely lack in vivo endotoxic activity [pyrogenicity (rabbit) and lethal toxicity (mouse)]. GlcpN disaccharide structures carrying two phosphate groups and seven (LA- 16-PP),five (LA-20-PP) or four (LA- 14-PP) fatty acids exhibit pyrogenic activity that, however, is weaker than that of E. coli lipid A (containing six fatty acids in asymmetric distribution). Compounds having only two fatty acids (LA- 19-PP) are completely inactive. Glucosamine disaccharide preparations carrying only one phosphate group (at either position 1 or 4’, see Fig. 2) are also less pyrogenic than E. colilipid A (containingtwo phosphate residues).

258

ULRICH ZAHRINGER et al.

In vitro, the capacity of LPS, lipid A, and partial structures to induce the production of IL-1, IG6, and TNFa in human peripheral monocytes was analyzed. The results of these studies (167,168,183,239)were, in principle, similar to those obtained in vivo. Thus, hexosamine (GlcpN, GlcpN3N) monosaccharide structures were completely inactive. Glucosamine disaccharidecompounds carrying two phosphate groups and seven (LA- 16-PP)or five (LA-20-PP)fatty acids exhibit lower activitythan E. coli lipid A (six fatty acids), The importance of the location of fatty acids became obvious on analyzing hexaacyl compound LA-22-PP. This preparation contains, like E. coli lipid A, a p-( 1 + 6)-linked GlcpN-disaccharide and four primary 14 :O(3-OH) residues. As secondary fatty acids, 14 :0 residues are linked to the 3-hydroxyl fatty acids at C-2’ and C-2, resulting in a C.violaceurn-type symmetric acylation pattern. Compound LA-22-PP, having six fatty acids with 14 carbon atoms was less active by a factor of lo4 in inducing IL-1 in peripheral monocytes ( 167). This result shows that not only the number of acyl groups but also their location (keeping the chain length constant) is an important factor in the expression of endotoxic activity. This has also been found to be true for the anomeric configuration of GlcN(1) as observed on investigating 2-phosphonooxyethylderivativesof lipid A ( 15). The synthetic structures analyzed correspond to E. coIi lipid A except that, instead of the two 14 :O( 3-OH) residues at GlcN(I) two 14 :0 groups are present and the glycosidicallylinked phosphate is replaced by an a-orp-linked 2-phosphonooxyethyl group. As our preliminary results show (185) the a-anomeric compound was a potent inducer of TNFa, being comparable in activity to synthetic E. coli lipid A (compound 506). The p anomer, however, was less active by a factor of lo3.These data show that (i) the phosphate does not have to be directly linked to GlcN(I) for lipid A bioactivity (15 ) and, most important, (ii) that the a-anomeric configuration at GlcN(1) is essential for the expression of endotoxic activity of lipid A (1 85). The importance of the chain length of acyl groups in this respect has to be demonstrated. Most interestingly,the tetraacyl precursor Ia (LA- 14-PP), as compounds having only two fatty acids (LA-19-PP),completelylacked mediator-inducing activity. Monophosphated E. coli lipid A partial structures were also significantly less active than the corresponding bisphosphate compound LA- 15-PP. As a further point of interest, LPS, including LPS of enterobacterial Re mutants was, in general, more active (by approximately a factor of 10) than free lipid A, suggesting that Kdo contributes to the induction of mediators (1 16,118,240). VII. SEROLOGY OF LIPIDA

In the early seventies, Galanos et al. (241) showed that lipid A may be immunogenic, and acid-treated bacteria (exposing free lipid A on their sur-

MOLECULAR STRUCTURE OF LIPID A

259

face) or free lipid A complexed to suitable carriers were found to be good immunogens for the preparation of anti-lipid A antibodies. Such antisera could be raised with lipid A preparations derived from various Gram-negative bacteria, provided they contained the E. coli- or C. violaceum-type of structure. The specificityof lipid A antibodieswas, however, not clearly defined until recently when Brade et al. (6,242- 244b) could characterizeepitopes present in E. coli-type lipid A by using bacterial and synthetic lipid A antigens and polyclonal anti-lipid A antisera. Thus far, eight different antigenic determinants have been defined, all residing in the hydrophilic lipid A backbone region. One epitope is composed of the 1,4’-bisphosphorylatedp-( I 6)linked GlcpN disaccharide, a second of the 4’-phosphorylated, and a third of the 1-phosphorylated disaccharide. The other two immunoreactive determinants are located in acylated D-gluco-HexpN 1- and 4-phosphate monosaccharides, respectively. In the latter instances, the specificity is also expressed by acylated and phosphorylated preparations differing in the nature of the glycosyl residue (Glcp, GlcpN, GlcpN3N, and Glcp3N). The specificity of monosaccharide and disaccharide lipid A antibodies is dependent on the presence of phosphate groups but independent of the acylation pattern; that is, fatty acids are not part of the epitopes(243). Acyl groups, however, by influencing the three-dimensionalorganizationof lipid A, may modulate the exposure of the epitopes. In recent years, the field of research on lipid A antibodies has been rapidly expanding (6,245-249). The interest in lipid A antibodies is based on the significance of LPS and lipid A in Gram-negative septic shock and the expectation that lipid A antibodies, being directed against a structure common to all Gram-negative bacteria, which, at the same time, harbors the toxic principle, may be widely cross-reactive and possibly cross-protective. In fact, polyclonal anti-E. coli lipid A antibodies cross-react with a large variety of free lipid A of distinct bacterial origin, namely, those lipid A containing ap-( 1 46)-linked HexpN disaccharide bisphosphate. It must be emphasized, however, that these lipid A antibodies do not cross-react, at least in vitro, with LPS, that is, with the lipid A component carrying the saccharideportion. Thus, in LPS either the lipid A epitopesare not expressed (that is, free lipid A constitutes a neoantigen exposing structural or conformational determinants not present in LPS) or the epitopes are cryptic (that is, Kdo or other LPS constituents sterically hinder the binding of antibodies to lipid A). There are, however, reports in the literature of mouse and human monoclonal antibodies, in particular of the IgM class, that appear to be directed against lipid A and that are reported to cross-react with different LPS (250-25 la). Their reactivity was elucidated in the ELISA assay system, but it is known that ELISA determinations, in particular of such amphiphilic molecules as lipid A and of sticky antibodies such as those of the IgM class,

-

260

ULRICH ZAHRINGER et al.

may, unless certain precautions are taken, lead to false positive results simulating cross-reactivity (252). These IgM antibodies are reported to exhibit protective activity in septicemicstates (250,251) but one of these studieshas received critical comments (244a,253). It is evident, therefore, that further studiesare required in order to understandtheir mechanism ofaction and, in particular, to define unequivocally the epitopes that they recognize. VIII. SYNOPSIS: THESTRUCTURE, ACTIVITY, AND FUNCTION OF LIPIDA As the preceding sections show, the term lipid A does not denote a defined molecular entity but rather a family of structurally closely related but not identical phosphoglycolipids. Their structures may differ in the type and number (one or two) of HexpN residues present in the backbone; by the number, location, nature, size, and linkage of acyl residues; by the nature of phosphate substituents; and finally by the degree of phosphorylation of the backbone. Lipid A components of different bacterial origin are classified in the present article on the basis of structural variations. Despite these variations, lipid A of different origin not only share many identical constituents, but also are constructed according to a similar architectural principle (Fig. 15). Settingaside those few preparations containing a monosaccharidebackbone, lipids A contain ap-( 1 + 6)-linked disaccharide having the D-gluco configuration and pyranosidic D-hexosamine residues, which carry an a-glycosidicand, at position C-4’, a nonglycosidicphosphate group. This lipid A backbone is substituted at the amino and hydroxyl groups in positions 3’, 2’, 3, and 2 by (R)-3-hydroxyfatty acids, up to three of which are acylated at their 3-hydroxylgroup by secondary fatty acids. Lipid A expressingstrongest endotoxicitycontains six fatty acids, and, in this case, the secondary fatty acids are bound to the 3-hydroxy fatty acid either at position 2‘ and 3‘ as in E. coli, or at position 2’ and 2, as in C.violaceurn. In the latter instance, however, the chain length of the primary fatty acid is lower than that in E. coli. Figure 15 shows the structure of the two hexaacyl lipid A types that are the most abundant among lipid A preparationsstudied thus far. The two lipid A types differ only in the location of one secondary fatty acid (designated R 1or R2 in Fig. 15) and in the chain length of the primary fatty acids. Attachment of one acyl group to the 3-hydroxy fatty acid at position 2 [GlcN(I)] yields heptaacyl lipid A (see Fig. 4), whereas pentaacyl lipid A (Figs. 5,7, 8, and 9B) formally result from the omission of any one of the secondary acyl groups of the structuresshown in Figs. 2 and 10.As comparison of the structures shows, the 3-hydroxyl group of the primary fatty acid at position 3 is never acylated, whereas that of the 3-hydroxy fatty acid at position 2’ is generally acylated. Other lipid A structures are created by replacing GlcpN by GlcpN3N, or by omitting the 4’-phosphate group.

MOLECULAR STRUCTURE OF LIPID A

26 1

HO -OH

FIG.15.-Structure of the Kdo-lipid A domain. The domain is structurally highly conserved and constitutes the biologically active region of LPS. R’, RZ= secondary fatty acids: In hexaacyl E. coli-type lipid A (m = n = 14), R2= fatty acid, R‘ = H; in hexaacyl C. violaceum-type lipid A (m = 10 or 12, n = 12 or 14), RZ= H, R’=fatty acid. R3,R4= H or substituent of backbone phosphate groups (see Table I). R5 = negativelycharged groups: Kdo [in Salmonella (70) and E. coli (34)], P03Hz (in Haemophilus injluenzae (68), Bordetella pertussis, Vibrio parahaemolyticus (83), Bacteroidesgingivalis(268), Vibrio cholerae(269) GalpA [in Rhodocyclus gelatinosus Dr2 (270)], and GlcpA [in Rhizobium sphaeroides (165)l. R6= core region: L-glycerm-D-manno-Hep (Enterobacteriaceae,Neisseriaceae,and other bacteria). D-glyceroa-D-mannc-Hepp [Rhodocyclusgelatinosus Dr2(270)], a-~-Glcp4-P03H, [Acinetobactercalcoaceticus (75)], and a-D-Man [Rhizobium trifolii ANU 843 (271)].

The hydroxyl group at C-4 [GlcN(I)] is generally not substituted,whereas in almost all Gram-negative bacteria studied in detail, the primary 6’-hydroxyl group cames Kdo (or the isosteric DgDt-KO) in a-ketosidic linkage (Fig. 15). The universal presence and the structural conservation of the Kdo and lipid A components led us earlier to propose that this LPS domain should not be viewed separately but rather as one structural and functional molecular entity ( 1 18). Thus, the Kdo - lipid A domain constitutesthe minimal structure that bacteria require for growth and multiplication and that expresses full endotoxic activity. It should be realized that the separate discussion of the lipid A and Kdo regions is justified merely by one chemical feature of LPS, namely, the peculiar susceptibilityof the Kdo-lipid A link-

262

ULRICH ZAHRINGER et al.

age to acid hydrolysis and, thus, the ready dissociation ofthe Kdo-containing core region from the lipid A component, but not by genetic, biological or other properties. In Enterobacteriaceae, the Kdo-lipid A domain also forms biosynthetically one molecular entity. As elucidated in E. coli and Salmonella, UDP-N-acetyl-D-GlcpNis, after addition of two 14 :O(3-OH) residues (1 9 1,254), condensed with 2,3-bisacyl-~-GlcpN1-phosphateto form a tetraacyl disaccharide 1-phosphate(a reaction catalyzed by the enzyme lipid A synthase). This molecule is then phosphorylated at position 4’ to yield precursor Ia of lipid A biosynthesis (255). To this precursor, Kdo is transferred before the completion of the lipid A structure by addition of 12 :0 and 14 :0 [In Pseudomonas, however, mature lipid A is formed first, and Kdo is then added in a subsequent step (26)]. The lipid A synthase also accepts UDP-activated and acylated GlcpN3N. It is, therefore, understandable that backbone structures containing a &( 1 6)-linked GlcpN3N disaccharide or hybrid disaccharides such as GlcpN3N-GlcpN are likewise formed in vitro ( 13) and in vivo (see Section 111,l .c.iii). Lipid A having Kdo attached at 0-6’ comprises a domain of LPS that exhibits a remarkable structural conservatism, as far as the principal architecture is concerned (Fig. 15).This type of structure is found in most Gramnegative bacteria. It is not present in other biomolecules and is thus unique to Gram-negativebacteria and their LPS. It containsthose structures that are required for (i) bacterial viability and (ii) for optimal expression of endotoxic activity. Thus far, viable (that is growing and multiplying) Gram-negative bacteria lacking LPS have not been isolated in nature, nor have they been produced in the test tube. This observation indeed suggests that LPS is essential for the viability of this class of microbes. In the bacterial cell, LPS is located in the outer membrane and it may be assumed that the vital significance of LPS is related to the successful assembly, architecture, and function of this outer membrane. As discussed, isolated LPS in aqueous solution may self-organize to a membrane structure that, as revealed by electron microscopy (256,257), resembles fragments of the bacterial membrane. Artificially created LPS membranes may, therefore, serve as objects for studying the architecture and function of the outer membrane of Gram-negative bacteria. As X-ray diffraction studies have shown, lipid A tends to be organized in large domains comprising of up to 1200 molecules, which are partly arranged in parallel and which persist for long periods of time (220). As conformational calculations have revealed, the lipid A fatty acids are tightly packed in a dense two-dimensional hexagonal lattice, the formation of which is favored by the absence of unsaturated fatty acids (220,221,228).In this respect, LPS membranes differsignificantly from other biological membranes, which, at physiological temperature, are more fluid and in which the phospholipids

-

MOLECULAR STRUCTURE OF LIPID A

263

possess rotational freedom. The high degree of order of LPS assemblies and the nearly crystalline nature of the lipid A fatty acid arrangement yields a well-ordered and rigid membrane system that may prevent hydrophobic molecules from diffusinginto or through an LPS layer system as is present in the outer leaflet of the Gram-negative outer membrane (258). Indeed it is known that the outer membrane of enteric bacteria is relatively impermeable for certain hydrophobic antibiotics, detergents, hydrophobic dyes, and bile acids (259,260). Thus, the outer membrane serves as a permeation barrier and, thereby, protects the bacterial cell from external toxic molecules. It is probable that the highly ordered LPS molecule, and, in particular, the lipid A component, is largely responsible for the permeation-barrier properties of the outer membrane (258). The Kdo residues carrying negative charges and the phosphate groups of lipid A are of importance for the organization and function of the outer membrane in that they constitute high- and low-affinity binding-sites for Mg2+and Ca2+ions (65). The Kdolipid A domain thus facilitatesthe accumulationofbivalent ions essential for the stabilityand proper function of the membrane, as well as probably for the activity ofenzymes and integral proteins (such as porines) present in the cell or the outer membrane, respectively. On the other hand, the negatively charged Kdo -lipid A domain constitutes a target for binding of certain antibiotics, such as polymyxin B (125) and basic peptides derived from host cells such as the bactericidal/permeability-increasing protein (BPI) (261). In this respect the Kdo-lipid A domain constitutes a weak site of the outer membrane, rendering bacteria vulnerable to the action of cationic agents. Endotoxic activity (including pyrogenicity and mediator induction) is, accordingto present knowledge, optimally expressedby a molecule containing two j3-(1 + 6)-linked HexpN residues of the D-gluco configuration, two phosphate groups, and six fatty acids (saturated and, in part, 3-hydroxylated), including 3-acyloxyacyl groups with the chain length, in the linkages and the locations present in E. coli- (Fig. 2) or C. violaceurn-type lipid A (Fig. 6) and exemplified in Fig. 15. Most interestingly, none of the synthetic compounds thus far available exhibits stronger endotoxicity than natural or synthetic E. coli-type lipid A. Partial structures lacking one constituent, or molecules having a different distribution of constituents, are endotoxically less active or inactive an exception being the 2-phosphonooxyethylderivative ( 15,185)of compound 506. Thus, slight modifications at any site of the E. coli lipid A architecture result in a significant decrease of biological activity, showing that endotoxicity is not dependent on one single lipid A constituent or a toxophore group. The role of fatty acids appears to be of particular importance. As discussed, the number of acyl groups is a decisive factor for endotoxic activity, six fatty acids being optimal. However, the

264

ULRICH ZAHRINGER et al

chain length of (the primary) fatty acids also plays an important role, as indicated from biological analyses of E. coli and C. violaceurn lipid A and also of lipid A having particularly long acyl chains [such as C. psittaci (1 33) and B. fragilis (79)], and shorter ones [P. aeruginosa (77) and reduced R. sphaeroides (162)J. It appears that the expression of endotoxic activity of lipid A requires a certain equilibration of hydrophilic (bisphosphated, p( 1 ---* 6)-linked gluco-configured HexpN backbone) and hydrophobic (defined number of fatty acids having a specific chain length) regions. The peculiar arrangement of lipid A constitutents,generatinga unique chemical architecture, results in the formation of a peculiar conformation (endotoxic conformation) and a particular supramolecular organization, which allows the expression of endotoxic activity. This conformation is likely to be influencedby Kdo, as the up-regulating effect of the negatively charged inner core region on lipid A bioactivity shows. It is most remarkable that precursor Ia (Fig. 9 A) which differs from E. coli lipid A (Fig. 2) only by the absence of two secondary acyl residues ( 12 :0 and 14 :0), thus lacking 3-acyloxyacylgroups, is completelyinactive in stimulating human mononuclear cells to release bioactive mediators. This observation points to an extreme structuralspecificityofthe expression of endotoxic activity by lipid A. In turn, this factor also strongly suggests that a specific binding-protein or a specific receptor is involved in the interaction between LPS or lipid A and such responsive target cells as monocytes. The existence of lipid-A-binding and monocyte/macrophage-associatedproteins has been demonstrated (262 - 264). In accord with this concept, we and others found that the tetraacyl precursor Ia strongly competes with the specific binding of labeled LPS to monocytes and that it inhibits, as does R. sphaeroides ( 182) and R. capsulatus (265) LPS (probablyvia an antagonistic mechanism), the production of LPS or free lipid A-induced mediator in human monocytes(167,169,183,184,266). Also the presence of a Kdo-specificreceptor on LPS target cells has been postulated (240,267). Studies along these lines appear important, as they may lead to new strategies for pharmacological control of endotoxin effects and because they provide deeper insight into the mechanisms involved in the interaction of endotoxins with the host. It should be mentioned that, on biodegradation of E. coli lipid A by macrophages and granulocytes, the two secondary fatty acids 12 :0 and 14 :0 are removed by the enzyme 3-acyloxyacylhydrolase, yielding (1 8 1) a structural counterpart of precursor Ia. As discussed, this compound not only lacks endotoxic activity in the human system, but also constitutes a potent endotoxin antagonist. Thus, on enzymic degradation of lipid A, a molecule (precursorIa) is formed, which not only is detoxified, but also counteracts endotoxin bioactivity ( 1 8 1).

MOLECULAR STRUCTURE OF LIPID A

265

Bacterial endotoxins, and, in particular, their lipid A component, because of their complex structure, their significancefor bacterial viability, and their important role in Gram-negative infection and sepsis, have fascinated researchers in the field of chemistry, physics, biology, immunology, genetics, microbiology, pharmacology, and clinical medicine for several decades. It was through the efforts of these scientists that extensive molecular knowledge on various aspects on the chemistryand biology of lipid A was acquired. It is hoped that in the future advantage will be taken of this knowledge in application to the immunological and also pharmacological control of Gram-negative sepsis and associated local and generalized inflammatory reactions caused by endotoxins and their lipid A component. ACKNOWLEDGMENTS The financial support of the Deutsche Forschungs Gemeinschaft (SFB 367, project B2) and Fonds der Chemie (EThR) is greatly appreciated. We thank M. Lohs, F. Richter, B. Kohler and I. Stegelmann-Mullerfor preparing drawings and photographs. We are particularly grateful to our friends and colleaguesH. Brade, E. Coats, H.-D. Grimmecke, 0. Holst, M. Kastowsky, S. Kusumoto, H. Labischinski, 0. Liideritz, H. Mayer, G. Seltmann, and U. Seydel for valuable suggestions.

REFERENCES 1 . C. Galanos, 0. Liideritz, E. Th. Rietschel, and 0. Westphal in T. W. Goodwin. (Ed.), International Review ofBiochemistry,Vol. 14, Biochemistry ofLipids 11:Newer Aspects of the Chemistry and Biology of Bacterial Lipopolysaccharides, with Special Reference to Their Lipid A Component, p. 239. University Park Press, Baltimore, 1977. 2. 0. Westphal, 0. Liideritz, E. Th. Rietschel, and C. Galanos, Biochem. Rev., 9 (1981) 191-195. 3. 0. Liideritz, M. A. Freudenberg, C. Galanos, V. Lehmann, E. Th. Rietschel, and D. H. Shaw, Curr. Top. Membr. Tramp.. 17 (1982) 79-151. 4. E. Th. Rietschel, C. Galanos, 0. Luderitz, and 0. Westphal, in D. R. Webb. (Ed.), Immunopharmacology and the Regulation of Leukocyte Function: The Chemistry and Biology of Lipopolysaccharides and Their Lipid A Component, p. 183. Dekker, New York/Basel, 1982. 5. E. Th. Rietschel, H. Brade, L. Brade, K. Brandenburg, U. F. Schade, U. Seydel, U. Ziihringer, C. Galanos, 0. Liideritz, 0. Westphal, H. Labischinski, S. Kusumoto, and T. Shiba, Prog. Clin. Biol. Rex, 231 (1987) 25-53. 6. H. Brade, L. Brade, and E. Th. Rietschel, Zbl. Bakt. Hyg., A268 (1 988) 15 1 - 179. 7. C. R. H. Raetz, Annu. Rev.Biochem., 59 (1990) 129-170. 8. E. Th. Rietschel and H. Brade, Sci. Am., 267 (1992) 54-61. 9. E. Th. Rietschel, L. Brade, B. Lindner, and U. Zilhringer, in D. C. Morrison and J. L. Ryan. (Eds.), Bacterial Endotoxic Lipopolysaccharides:Biochemistry of Lipopolysaccharides, p. 3. CRC Press, Boca Raton, FL, 1992. 9a. E. Th. Rietschel, T. Kirikae, F. U. Schade, U. Mamat, G. Schmidt, H. Loppnow, A. J. Ulmer, U. Zahringer, U. Seydei, F. Dipadova, M. Schreier, and H. Brade, FASEB J., 8 (1994) 217-225.

266

ULRICH ZAHRINGER et al.

9b. E. Th. Rietschel, T. Kirikae, F. U. Schade, A. J. Ulmer, 0. Holst, H. Brade, G. Schmidt, U. Mamat, H.-D. Grimmecke, S. Kusumoto, and U. Zaringer, Immunobiol., 187 (1993) 169- 190. 10. C.R. Alving,J. Immunol.Meth., 140(1991) 1-13. 11. M. Kiso, S . Tanaka, M. Tanahashi, Y. Fujishima, Y. Ogawa, and A. Hasegawa, Carbohydr. Res., 148 (1986) 221-234. 12. P. Westerduin,J. H. van Boom, C. A. A. van Boeckel, and T. Beetz, Carbohydr.Rex, 137 (1985) ~ 4 - ~ 7 . 13. P. Stiitz, H. Aschauer, J. Hildebrandt, C. Lam, H. Loibner, I. Macher, D. Scholz, E. Schiitze, and H. Vyplel, in A. Nowotny (Ed.), Endotoxin Research Series, Vol. 1, Cellular and Molecular Aspects of Endotoxin Reactions: Chemical Synthesis of Endotoxin Analogs and Some Structure Activity Relationships, p. 129. Elsevier, Amsterdam, 1990. 14. S. Kusumoto, N. Kusunose, M. Imoto, T. Shimamoto, T. Kamikawa, H. Takada, S. Kotani, E. Th. Rietschel, and T. Shiba, Pure Appl. Chem., 6 1 (1989) 46 1 -464. 15. S. Kusumoto, T. Kamikawa, M. Kurosawa, K. Fukase, T. Shiba, T. Kusama, T. Soga, E. Shioya, K. Nakayama, H. Nakayima, Y. Osada, and Y. Ono, in A. Nowotny, J. J. Spitzer, and E. J. Ziegler (Eds.), Endotoxin Research Series, Vol. 1, Cellular and Molecular Aspects of Endotoxin Reactions:New Results in the Synthesis of Lipid A and Endotoxins, p. 41. Elsevier Science Publishers, Amsterdam, 1990. 16. M. Kiso, S. Tanaka, M. Fujita, Y. Fujishima, Y. Ogawa, H. Ishida, and A. Hasegawa, Carbohyr. Res., 162 (1987) 127-140. 17. D. Charon, C. Diolez, M. Mondange,S. R. Sarfati,L. Szabo, P. Szab6, and F. Trigalo,Am. Chem. SOC.Symp. Ser., 231 (1983) 301-316. 18. T. Shimizu, Y. Ohtsuka, Y. Yanagihara, I. Hajime, S.-I. Nakamoto, and K. Achiwa, Znt. J. Immunopharmacol., 13 (1991) 605-611. 19. S . Ikeda, M. Matsuura, M. Nakatsuka, J. Y. Homma, M. Kiso, A. Hasegawa, and C. Nishimura, J. Clin. Lab. Immunol., 32 (1990) 177- 181. 20. D. C. Momson and J. L. Ryan, Annu. Rev. Med., 38 (1987) 417-432. 21. M. Pollack, in G. L. Mandell, R. G. Douglas, and J. E. Bennet (Eds.), Principles and Practice of Infection Diseases, Update 8, New TherapeuticStrategies in Gram-Negative Sepsis and Septic Shock Based on Molecular Mechanism of Pathogenesis, p. 1. Churchill Livington, New York, 1990. 22. J. M. Griffiss, H. Schneider, R. E. Mandrell, R. Yamasaki, G. A. Jarvis, J. J. Kim, B. W. Gibson, R. Hamadeh, and M. A. Apicella, Rev. Inf Dis., 10 (1988) s287-s295. 23. 0. Holst and H. Brade, in D. C. Momson and J. L. Ryan (Eds.), Bacterial Endotoxic Lipopolysaccharides: Chemical Structure of the Core Region of Lipopolysaccharides. p. 135. CRC Press, Bow Raton, FL, 1992. 24. 0. Holst, E. Rohrscheidt-Andrzejewski,H. Brade, and D. Charon, Carbohydr. Res., 204 (1990) 93-102. 25. E. Th. Rietschel, H. -W. Wollenweber,H. Brade, U. Dhringer, B. Lindner, U. Seydel, H. Bradaczek, G. Barnickel, H. Labischinski, and P. Giesbrecht, in E. Th. Rietschel, (Ed.), Handbook of Endotoxin, Vol. 1, Chemistry of Endotoxin:Structure and Conformation of the Lipid A Component of Lipopolysaccharides,p. 187. Elsevier, Amsterdam, 1984. 26. R. C. Goldman, C. C. Doran, S. K. Kadam, and J. 0. Capobianco, J. Biol. Chem., 263 (1988) 5217-5223. 27. A. Boivin, J. Mesrobeanu, and L. Mesrobeanu, Compt. Rend. Sci. Biol., 114 (1933) 307- 3 10. 28. 0. Westphal and 0. Liideritz,Agnew. Chem., 66 (1954) 407-417. 29. W. T. J. Morgan and S. M. Partridge, Biochem. J.,35 (1941) 1140- 1163. 30. 0.Liideritz, K. Jann, and R. Wheat, in M. Florkin and E. H. Stotz (Eds.), Comprehensive

MOLECULAR STRUCTURE OF LIPID A

267

Biochemistry, Vol. 26A, Somatic and Capsular Antigens of Gram-Negative Bacteria, p. 105. Elsevier, Amsterdam, 1968. 31. R. S. Johnson, G.-R. Her, J. Grabarek, J. Hawiger, and V. N. Reinhold, J. Biol. Chem., 265 (1990) 8108-8116. 32. N. Qureshi, K. Takayama, and E. E. Ribi, J. Biol. Chem., 257 (1982) 1 1808- 1 1815. 33. K. Takayama, N. Qureshi, K. Hyver, J. Honovich, R. J. Cotter, P. Mascagni, and H. Schneider, J. Biol. Chem., 261 (1986) 10624-10631. 34. U. Zahringer, B. Lindner, U. Seydel, E. Th. Rietschel, H. Naoki, F. M. Unger, M. Imoto, S. Kusumoto, and T. Shiba, Tetrahedron Lett., 26 (1985) 6321-6324. 35. M. Imoto, T. Shiba, H. Naoki, T. Iwashita, E. Th. Rietschel, H. -W. Wollenweber, C. Galanos, and 0. Liideritz, Tetrahedron Lett., 24 (1983) 4017-4020. 36. R. J. Cotter, J. Honovich, N. Qureshi, and K. Takayama, Biomed. Mass Spectrom., 14 (1987) 591 -598. 37. U. Seydel, B. Lindner, H. -W. Wollenweber, and E. Th. Rietschel, Eur. J. Biochem., 145 (1984) 505-509. 38. H.-W. Wollenweber, U. Seydel, B. Lindner, 0. Luderitz, and E. Th. Rietschel, Eur. J. Biochem., 145 (1984) 265-272. 39. B. Lindner, U. ahringer, E. Th. Rietschel, and U. Seydel, in A. Fox, S. L. Morgan, L. Larsson, and G. Odham (Eds.),Analytical MicrobiologyMethods: Structural Elucidation of Lipopolysaccharides and Their Lipid A Component: Application of Soft Ionization Mass Spectrometry, p. 149. Plenum, New YorkfLondon, 1990. 40. U. Seydel, M. Haas, E. Th. Rietschel, and B. Lindner, J. Microb. Meth., 15 (1992) 167-183. 41. N. Qureshi, K. Takayama, D. Heller, and C. Fenselau, J. Biol. Chem., 258 (1983) 1294712951. 42. M. Caroff, C. Deprun, D. Karibian, and L. Szab6, J. Biol. Chem., 266 (1991) 1854318549. 43. R. Wang, L. Chen, R. J. Cotter, N. Qureshi, and K. Takayama, J. Microb. Meth., 15 (1992) 151-166. 44. D. Karibian, C. Deprun, L. Szabb, Y. Le Beyec, and M. Caroff, Int. J. Mass Spectrom. Ion Processes, 11 I (1991) 273-286. 45. A. L. Erwin and R. S. Munford, Jr., J. Biol. Chem., 265 (1990) 16444- 16449. 46. M. R. Rosner, H. G. Khorana, and A. C. Satterthwait, J. Biol. Chem., 254 (1979) 59185925. 47. B. Jiao, M. A. Freudenberg, and C. Galanos, Eur. J. Biochem.. 180 (1989) 5 15- 5 18. 48. W. Fischer, Eur. J. Biochem., 194 (1990) 655-661. 49. J. Weckesser and H. Mayer, FEMSMicrobiol. Lett., 54 (1988) 143-154. 50. H. Mayer, J. H. Krauss, A. Yokota, and J. Weckesser, in H. Friedman, T. W. Klein, M. Nakano, and A. Nowotny (Eds.), Endotoxin:Natural VariantsofLipid A, p. 45. Plenum, New York, 1990. 51. J. Montreuil, Adv. Curbohydr. Chem. Biochem., 37 (1980) 157-223. 52. A. Nowotny, J. Am. Chem. Soc., 83 (1961) 501-503. 53. A. J. Burton and H. E. Carter, Biochemistry, 3 (1964) 41 1-418. 54. J. Gmeiner, 0. Liideritz ,and 0. Westphal, Eur. J. Biochem., 7 (1969) 370-379. 55. A. B. Foster and D. Horton, J. Chem. SOC.,384 (1958) 1890- 1894. 56. G. A. Adams and P. P. Singh, Biochim. Biophys. Acta. 187 (1969) 457-459. 57. S. Hase and E. Th. Rietschel, Eur. J. Biochem., 63 (1976) 101-107. 58. S. Hase and E. Th. Rietschel, Eur. J. Biochem., 63 (1976) 93-99. 59. M. Jensen, D. Borowiak, H. Paulsen, and E. Th. Rietschel, Biomed. Mass Spectrom., 6 (1979) 559-565. 60. K. Takayama, N. Qureshi, and P. Mascagni, J. Biol. Chem., 258 (1983) 12801- 12803.

268

ULRICH ZAHRINGER et al.

61. D. Blache, M. Bruneteau, and G. Michel, Biochimie (Paris), 62 (1980) 191- 194. 62. I. N. Krasikova, V. I. Gorbach, V. V. Isakov, T. F. Solov'eva, and Y. S. Ovodov, Eur. J. Biochem., 126 (1982) 349-351. 63. S . M. Strain, I. M. Armitage, L. Anderson, K. Takayama, N. Qureshi, andC. R. H. Raetz, J. Biol. Chem., 260 (1985) 16089-16098. 64. A. Neszmelyi, P. Kosma, R. Christian, G. Schulz, and F. M. Unger, Carbohydr.Res., 139 (1985) 13-22. 64a. U. Ziihringer, V. Sinnwell, J. Peter Katalinic, E. Th. Rietschel, and C. Galanos, Tetrahedron, 49 (1993) 4193-4200. 65. S. M. Strain, S. W. Fesik,andI. M. Armitage,J. Biol. Chem., 258(1983) 13466-13477. 66. S.M. Strain, S. W. Fesik, and I. M. Armitage, J. Biol. Chem., 258 (1983) 2906-2910. 66a. 0. Holst, S. Miiller-Loennies, B. Lindner, and H. Brade, Eur. J. Biochem., 214 (1993) 695 -701. 67. J. Radziejewska-Lebrecht, U. R. Bhat, H. Brade, and H. Mayer, Eur. J. Biochem., 172 (1988) 535-541. 68. I. M. Helander, B. Lindner, H. Brade, K. Altmann, A. A. Lindberg, E. Th. Rietschel, and U. Z&ringer, Eur. J. Biochem., 177 (1988) 483-492. 69. M. R. Rosner, J. Tang, I. Barzilay, and H. G. Khorana, J. Biol. Chem., 254 (1979) 5906-5017. 70. R. Christian, G. Schulz, P. Waldstiitten, and F. M. Unger, Tetrahedron Lett., 25 (1984) 3433- 3436. 7 1. M. Jensen, Strukturanalyseder Lipoid A Komponente von Lipopolysacchariden aus Yersinia enterocolitica, Doctoral Thesis, University of Freiburg, 1980, 72. Z. Sidorczyk, U. mringer, and E. Th. Rietschel, Eur. J. Biochem., 137 (1983) 15-22. 72a. A. K. Harrata, L. N. Domelsmith, and R. B. Cole, Biological Mass Spectrometry, 22 (1993) 59-67. 72b. R. B. Cole, L. N. Domelsmith, C. M. David, R. A. Laine, and A. J. DeLucca, Rapid Communications in Mass Spectrometry, 6 (1992) 6 16-622. 73. V. A. Kulshin, U. ahringer, B. Lindner, C. E. Frasch, C. -M. Tsai, B. A. Dmitriev, and E. Th. Rietschel, J. Bacteriol., 174 (1992) 1793- 1800. 74. V. L. L'vov, I. K. Verner, L. Y. Musina, A. V. Rodionov, A. V. Ignatenko, and A. S. Shashkov, Arch. Microbiol., 157 (1992) 131 - 134. 75. K. Kawahara, H. Brade, E. Th. Rietschel, andU. ahringer, Eur. J. Biochem., 163 (1987) 489-495. 76. D. T. Drewry, J. A. Lomax, G. W. Gray, and S. G. Wilkinson, J. Biochem. (Tokyo), 133 (1973) 563-572. 77. V. A. Kulshin, U. ahringer, B. Lindner, K. -E. Jeer, B. A. Dmitriev, and E. Th. Rietschel, Eur. J. Biochem., 198 (1991) 697-704. 77a. D. N. Kurunaratne, J. C. Richards, and R. E. W. Hancock, Arch. Biochem. Biophys. 299 (1992) 368-376. 78. E. Th. Rietschel, 0. Liideritz, and W. A. Volk, J. Bacteriol., 122 (1975) 1180- 1188. 79. A. Weintraub, U. ZAhringer, H. -W. Wollenweber,U. Seydel,and E. Th. Rietschel, Eur. J. Biochem., 183 (1989) 425-431. 80. S. Haw, T. Hofstad, and E. Th. Rietschel, J. Bacteriol., 129 (1977) 9- 14. 80a. T. Ogawa, FEBS Lett., 332 (1993) 197-201. 81. R. Russa, 0. Liideritz, and E. Th. Rietschel, Arch. Microbiol., 141 (1985) 284-289. 82. T. Urbanik-Sypniewska, U. Seydel, M. Greck, J. Weckesser, and H. Mayer, Arch. Microbiol., 152 (1989) 527-532. 83. S.Kondo, U. ZAhringer, U. Seydel, K. Hisatsune, and E. Th. Rietschel, Eur. J.Biochem., 200 (1991) 689-698.

MOLECULAR STRUCTURE OF LIPID A

269

84. H. Masoud, B. Lindner, J. Weckesser, and H. Mayer, System. Appl. Microbiol., 13(1990) 227-233. 85. S. Hase and E. Th. Rietschel, Eur. J. Biochem., 75 (1977) 23-34. 86. K. W. Broady, E. Th. Rietschel, an d 0 . Liideritz, Eur. J. Biochem., 115 (1981)463-468. 87. P. V. Salimath, J. Weckesser, W. Strittmatter, andH. Mayer, Eur. J.Biochem., 136(1983) 195- 200. 88. N. Qureshi, J. Honovich, H. Hara, R. J. Cotter, and K. Takayama, J. Biol. Chem., 263 (1988) 5502-5504. 89. J. H. Krauss, U. Seydel, J. Weckesser, and H. Mayer, Eur. J. Biochem., 180 (1989) 5 19- 526. 90. R. N. Tharanathan, J. Weckesser, and H. Mayer, Eur. J. Biochem., 84 (1978) 385-394. 91. 0. Holst, D. Borowiak, J. Weckesser, and H. Mayer, Eur. J. Biochem., 137 (1983) 325 332. 92. H. Masoud, T. Urbanik-Sypniewska, B. Lindner, J. Weckesser, and H. Mayer, Arch. Microbiol., 156 (1991) 167-175. 93. M. Caroff and L. SzabB, Carbohydr. Rex, I14 (1983) 95- 102. 94. H. Masoud, S. T. Weintraub, R. Wang, R. Cotter, and S. C. Holt, Eur. J. Biochem., 200 (1991) 775-779. 95. S. Fukuoka, H. Kamishima,Y. Nagawa, H. Nakanishi,K. Ishikawa,Y. Niwa, E. Tamiya, andI. Karube, Arch. Microbiol., 157 (1992) 311-318. 96. J. Roppel, H. Mayer, and J. Weckesser, Carbohydr. Res., 40 (1975) 31 -40. 97. A. P. Moran, U. e n g e r , U. Seydel, D. Scholz, P. Stutz, and E. Th. Rietschel, Eur. J. Biochem., 198 (1991) 459-469. 98. H. Mayer, U. R. Bhat, H. Masoud, J. Radziejewska-Lebrecht,C. Widemann, and J. H. Krauss,Pure Appl. Chem., 61 (1989) 1271- 1282. 99. A. Yokota, M. Rodriguez, Y. Yamada, K. Imai, D. Borowiak, and H. Mayer, Arch. Microbiol., 149 (1987) 106-111. 100. H. Mayer, J. H. Krauss, T. Urbanik-Sypniewska, V. Puvanesarajah, G. Stacey, andG. Auling, Arch. Microbiol., 151 (1989) I 1 1 - 116. 101. H. Mayer, S. A. Campos-Portuguez, M. Busch, T. Urbanik-Sypniewska, and U. R. Bhat, in A. Nowotny, J. J. Spitzer, and E. J. Ziegler (Eds.), Endotoxin Research Series, Vol. 1, Cellular and Molecular Aspects of Endotoxin Reactions: Lipid A Variants- Or, How Constant Are the Constant Regions in Lipopolysaccharides?p. 11 I. Elsevier,Amsterdam, 1990. 102. S. G. Wilkinson and D. P. Taylor, J. Gen. Microbiol., 109 (1978) 367-370. 103. N. Kasai, S. Arata, J. -I. Mashimo, Y. Akiyama, C. Tanaka, K. Egawa, and S. Tanaka, Biochem. Biophys. Res. Commun., 142 (1987) 972-978. 103a. N. Ravenscroft, S. G. Walker, G. G. S. Dutton, and J. Smith, j . Bacteriol., 174 (1992) 7595-7605. 104. G. Keilich, J. Roppel, and H. Mayer, Carbohydr. Res., 5 1 (1976) 129- 134. 105. R. Weisshaarand F. Lingens, Eur. J. Biochem., 137 (1983) 155-161. 106. R. I. Hollingsworth and R. W. Carlson, J. Biol. Chem., 264 (1989) 9300-9303. 107. U. R. Bhat, H. Mayer, A. Yokota, R. I. Hollingsworth, and R. W. Carlson, J. Bacteriol., 173 (1991) 2155-2159. 108. Biological Properties of Peptidoglycan, P. H. Seidl and K. H. Schleifes (Eds.), Walter de Gruyter, Berlinmew York, 1986. 109. M. Batley, N. H. Packer, and J. W. Redmond, Biochemistry, 21 (1982) 6580-6586. 1 10. N. Qureshi, K. Takayama, P. Mascagni, J. Honovich, R. Wong, and R. J. Cotter, J. Biol. Chem., 263 (1988) 11971-11976. 1 1 1. M. Caroff, A. Tacken, and L. Szab6, Carbohydr. Res., 175 (1988) 273-282.

270

ULRICH ZAHRINGER et al.

112. J. Gmeiner, M. Simon, and 0. Luderitz, Eur. J. Biochem., 21 (1971) 355-356. 113. P. F. Miihlradt, V. Wray, and V. Lehmann, Eur. J. Biochem., 81 (1977) 193-203. 114. R. E. Hurlbert, J. Weckesser, H. Mayer, and I. Fromme, Eur. J. Biochem., 68 (1976) 365-371. 1 15. Y. Kamio, K. C. Kim, and H. Takahashi, J. Biochem. (Tokyo), 70 (197 1) 187 - 19 1. 116. E. Th. Rietschel, L. Brade, 0.Holst, V. A. Kulshin, B. Lindner, A. P. Moran, U. F. Schade, U. Zahringer, and H. Brade, in A. Nowotny, J. J. Spitzer, and E. J. Ziegler (Eds.), Endotoxin Research Series, Ed., Vol. I, Cellular and Molecular Aspects of Endotoxin Reactions: Molecular Structure of Bacterial Endotoxin in Relation to Bioactivity, p. 15. Elsevier, Amsterdam, 1990. 117. A. A. Lindberg, A. Weintraub, U. Zilhringer, andE. Th. Rietschel, Rev. Infect. Dis., 12 (2) (1 990) S133 - S 14 1. 1 18. E. Th. Rietschel, L. Brade, U. F. Schade, U. Seydel, U. ZAhringer, S. Kusumoto, and H. Brade, in E. Schrinner, M. Richmond, G. Seibert,and U. Schwartz (Eds.), Surface Structures of Microorganisms and Their Interaction with the Mammalian Host: Bacterial Endotoxins:Properties and StructureofBiologically Active Domains, p. 1. Verlag Chemie, Weinheim, 1988. 119. A. Sonesson, E. Jantzen, K. Bryn, L. Larsson, and J. Eng, Arch. Microbiol., 153 (1989) 72-78. 120. S. E. Zamze, M. A. J. Ferguson, E. R. Moxon, R. A. Dwek, and T. W. Rademacher, Biochem. J., 245 (1987) 583-587. 121. N. Dalla Venezia, S. Minka, M. Bruneteau, H. Mayer, and G. Michel, Eur. J. Biochem., 151 (1985) 399-404. 122. V. Lehmann and E. Rupprecht, Eur. J. Biochem., 81 (1977) 443-452. 123. V. Lehmann, J. W. Redmond, A. Egan, and I. Minner, Eur. J. Biochem., 86 (1978) 487-496. 124. S. K. B. Lakshmi, U. R. Bhat, K. Wartenberg, S. Schlecht, and H. Mayer, FEMS Microbiol. Lett., 60 (1989) 317-322. 125. M. Vaara, T. Vaara, M. Jensen, I. M. Helander, M. Nurminen, E. Th. Rietschel, and P. H. MZikelB, FEBSLett., 129 (1981) 145-149. 126. W. Kaca, J. Radziejewska-Lebrecht,and R. Bhat, Microbios, 6 I (1990) 23-32. 127. S. G. Wilkinson, in C. Ratledge and S. G. Wilkinson (Eds.,) Microbial Lipids, Vol. 1, Gram-NegativeBacteria, p. 299. Academic Press, London/San Diego, 1988. 128. E. Th. Rietschel, H. Gottert, 0.Liideritz, and 0. Westphal, Eur. J. Biochem., 28 (1972) 166-173. 129. E. Th. Rietschel, Eur. J. Biochem., 64 (1976) 423-428. 130. D. Horton, D. A. Riley, S. Samreth, and M. G. Schweitzer,Am. Chem. SOC.Symp. Ser., 231 (1983) 21-47. 131. S. G. Wilkinson, Rev. Infect. Dis., 5 (1983) 941-949. 132. M. Nurminen, E. Th. Rietschel, and H. Brade, Infect. Immun., 48 (1985) 573-575. 133. L. Brade, S. Schramek, U. Schade, and H. Brade, Infect. Immun.,54 (1986) 568-574. 134. M. J. Hewett, K. W. Knox, and D. G. Bishop, Eur. J. Biochem., 19 (1971) 169-175. 135. I. Mattsby-Baltzer, Z. Mielniczuk, L. Larsson, K. Lindgren, and St. Goodwin, Infect. Immun., 60 (1992) 4383-4387. 136. A. P. Moran, I. M. Helander, and T. U. Kosunen, J. Bacteriol., 174 (1992) 1370- 1377. 137. H.-W. Wollenweber, S. Schramek, H. Moll, and E. Th. Rietschel,Arch. Microbiol., 142 (1985) 6- 1 1. 138. W. R. Mayberry, J. Bacteriol., 147 (1981) 373-381. 139. N. Haeffner, R. Chaby, and L. Szab6, Eur. J. Biochem., 77 (1977) 535-544. 140. S. K. Maitra, M. C. Schotz, T. T. Yoshikawa, and L. B. Guze, Proc. Natl. Acad. Sci. USA, 75 (1978) 3993-3997.

MOLECULAR STRUCTURE OF LIPID A

27 1

141. A. Sonesson, L. Larsson, R. Anderson, N. Adner, and K. -G. Tranberg, J. Clin. Microbiol., 28 (1990) 1163-1168. 142. H.-W. Wollenweber,S. Schlecht,0.Liideritz,andE. Th. Rietschel, Eur. J. Biochem., 130 (1983) 167-171. 143. L. van Alphen, B. Lugtenberg, E. T. Rietschel, and C. Mombers, Eur. J. Biochem., 101 (1979) 571-579. 144. A. M. B. Kropinski, V. Lewis, and D. Berry, J. Bucteriol., 169 (1987) 1960- 1966. 145. K. Wartenberg, W. Knapp, N. M. Ahamed, C. Widemann, and H. Mayer, Zbl. Bnkt. Hyg.., 253 (1983) 523-530. 146. Y. Kawai and I. Yano, Eur. J. Biochem., 136 (1983) 531 -538. 147. 0.W. Thiele, J. Oulevey, and D. H. Hunnemann, Eur. J. Biochem., I39 (1984) 131 - 135. 148. J. R. Edwards and J. A. Hayashi, Arch. Biochem. Biophys., I I 1 (1965) 415-421. 149. D. B. Boylan and P. J. Scheuer, Science, 155 (1967) 52-56. 150. A. P. Tulloch and J. F. T. Spencer, Can. J. Chem., 42 (1964) 830-837. I5 1. K. Bryn and E. Th. Rietschel, Eur. J. Biochem., 86 (1978) 3 1 I - 3 15. 152. U. R. Bhat, A. Man, C. Galanos, and R. S. Conrad, J. Bucteriol., 172 (1990) 663 I 6636. 153. S. E. Zamze, A. R. W. Smith, and R. C. Hignett, J. Gen. Microbiol., 13I (1 985) 19411950. 154. H. Brade and C. Galanos, Eur. J. Biochem., 122 (1982) 233-237. 155. K. Kawahara, K. Uchida, and K. Aida, Biochim. Biophys. Actu, 712 (1982) 571 -575. 156. K. Kawahara, U. Seydel, M. Matsuura, H. Danbara, E. Th. Rietschel, and U. ahringer, FEBS Lett., 292 (1991) 107-110. 157. K. Takayama, R. J. Rothenberg, and A. G. Barbour, Infict. Immun., 55 (1987) 231 1 2313. 158. S . Arata, T. Hirayama, N. Kasai, T. Itoh, and A. Ohsawa, FEMS Microbiol. Lett., 60 (1989) 219-222. 159. H. Moll, A. Sonesson, E. Jantzen, R. Marre, and U. Z&ringer, FEMSMicrobiol. Lett.. 97 (1992) 1-6. 160. U. R. Bhat, R. W. Carlson, M. Busch, and H. Mayer, In?. J. Syst. BuderioZ., 41 (1991) 213-217. 161. A. Sonesson, H. Moll, E. Jantzen, and U. W n g e r , FEMS Microbiol. Lett., 106 (1993) 3 15 -320. 162. N. Qureshi, K. Takayama, K. C. Meyer, T. N. Kirkland, C. A. Bush, L. Chen, R. Wang, and R. J. Cotter, J. Biol. Chem., 266 (1991) 6532-6538. 162a. C. R. H. Raetz, J. Bucteriol., 175 (1993) 5745-5753. 163. B. J. Wilkinson, M. S. Hindahl, L. Galbraith, and S. G. Wilkinson, FEMS Microbiol. Lett., 37 (1986) 63-67. 164. W. Strittmatter, J. Weckesser, P. V. Salimath, and C. Galanos, J. Bucteriol., 155 (1983) 153-158. 165. P. V. Salimath, R. N. Tharanathan, J. Weckesser, and H. Mayer, Eur. J. Biochem., 144 (1984) 227-232. 166. N. Qureshi, K. Takayama, and R. Kurtz, Infect. Immun., 59 (1991) 441-444. 167. H. Loppnow, H. Brade, I. Diirrbaum,C. A. Dinarello, S. Kusumoto, E. Th. Rietsche1,and H.-D. Flad, J. Zmmunol., 142 (1989) 3229-3238. 168. W. Feist, A. J. Ulmer, J. Musehold, H. Brade, S. Kusumoto, and H.-D. Had, Immunobiol., 179 (1989) 293-307. 169. M. H. Wang, H.-D. Had, W. Feist, J. Musehold, S. Kusumoto, H. Brade, J. Gerdes, E. Th. Rietschel, and A. J. Ulmer, Lymphokine Cytokine Rex, 11 (1 99 1) 23 - 3 1. 170. C. Galanos, V. Lehmann, 0.Luderitz, E. Th. Rietschel, 0. Westphal, H. Brade, L. Brade, M. A. Freudenberg, T. Hansen-Hagge, T. Liideritz, G. McKenzie, U. F. Schade, W.

-

272

ULRICH ZAHNNGER et al.

Strittmatter, K. -I. Tanamoto, U. ahringer, M. Imoto, M. Yamamoto, T. Shimamoto, S. Kusumoto, and T. Shiba, Eur. J. Biochem., 140 (1984) 221 -227. 171. C. Galanos, T. Hansen-Hagge, V. Lehmann, and 0.Liideritz, Infict. Immun., 48 (1985) 335 - 358. 172. E. Th. Rietschel, L. Brade, U. F. Schade, C. Galanos, M. A. Freudenberg, 0.Liideritz, S. Kusumoto, and T. Shiba, Eur. J. Biochem., 169 ( 1987) 27 - 3 1. 173. T. Hansen-Hagge, V. Lehmann, U. Seydel, B. Lindner, and U. Zduinger, Arch. Microbiol., 141 (1985) 353-358. 174. N. Qureshi, P. Mascagni, E. E. Ribi, and K. Takayama, J. Biol. Chem., 260 (1985) 5271-5278. 175. K. R. Myers, A. T. Truchot, J. Ward, Y. Hudson, and J. T. Ulrich, in A. Nowotny, J. J. Spitzer, and E. Ziegler (Eds.),Endotoxin Research Series, Vol. 1, Cellular and Molecular Aspects of Endotoxin Reactions:A Critical Determinant ofLipid A Endotoxic Activity, p. 145. Excerpta Medica, Amsterdam/New York/ Oxford, 1990. 176. H.-W. Wollenweber and E. Th. Rietschel, J. Microb. Meth., 1 1 (1990) 195-21 1. 177. W. Drozanski, A. E. A. Moustafa, 0. Liideritz, and 0. Westphal, Carbohydr. Res., 178 (1988) 225-232. 178. W. Drozanski, C. Galanos, S. Schlecht, and 0. Liideritz, Eur. J. Biochem., 155 (1986) 433-437. 179. D. Malchow, 0.Liideritz,B. Kickhofen,0.Westphal, and G. Gerisch, Eur. J. Biochem., 7 (1969) 239-246. 180. R. S. Munford, Jr., and C. L. Hall, Infect. Immun., 48 (1985) 464-473. 181. R. S. Munford,Jr., L. Eidels, and E. J. Hansen in A. Nowotny, J. J. Spitzer, and E. J. Ziegler (Eds.), Endotoxin Research Series, Vol. I, Cellular and Molecular Aspects of Endotoxin Reactions:PurifrcationofHuman Acyloxyacyl Hydrolase UsingAfinity Chromatography, p. 305. Medica, Amsterdam/New York/Oxford, 1990. 182. K. Takayama, N. Qureshi, B. Beutler and T. N. Kirkland, Infect. Immun., 57 (1989) 1336-1338. 183. M.H. Wang, W.Feist,H. Herzbeck,H.Brade,S.Kusumoto,E.Th.Rietschel,H.-D.Flad, and A. J. Ulmer, FEMSMicrob. Immunol., 64 (1990) 179- 186. 184. D. T. Golenbock, R. Y. Hampton, N. Qureshi, K. Takayama, and C. R. H. Raetz, J. Biol. Chem., 266 (1991) 19490- 19498. 185. A. J. Ulmer, H. Heine, W. Feist, S. Kusumoto, T. Kusama, H. Brade, U. Schade, E. T. Rietschel, and H. -D. Had, Infect. Immun., 60 (1992) 3309-3314. 185a. H. Loppnow, E. Th. Rietschel, H. Brade, U. Schdnbeck, P. Libby, M.-H. Wang, H. Heine, W. Feist, 1. Diirrbaum-Landmann, M. Emst, E. Brandt, H. Grage-Griebenow, A. J. Ulmer, S. Campos Portuguez, U. Schade, R. Kirikae, S. Kusumoto, J. Krauss, H. Mayer, and H.-D. Had, Vol.2, Bacterial endotoxin:Recognition and eflectormechanisms. (J. Levin, C. R. Alving, R. S. Munford, and P. L. Stiitz, eds.)Excerpta Medica, Elsevier, Amsterdam, 1993, p. 337. 186. T.H.Pohlmann,R. S.Munford,andJ.M. Harlan,J. Exp.Med., 165(1987) 1393- 1402. I Biochem., . 187. H.-W. Wollenweber, K. W. Broady, 0.Liideritz, and E. Th. Rietschel, Eur. . 124 (1982) 191-198. 188. U. Kraska, J. R. Pougny,and P. Sinay, Curbohyd. Rex, 50 (1976) 18 1- 190. 189. R. C. Verret, M. R. Rosner,andH.G. Khorana,J. Biol. Chem., 257(1982) 10222- 10227. 190. R.C. Verret,M.R. Rosner,andH.G. Khorana,J. Biol. Chem.,257(1982)10338- 10234. 191. J. M. Williamson, M. S. Anderson, and C. R. H. Raetz, J. Bacteriol., 173 (1991) 35913596. 192. E. Stackebrandt, R. G. E. Murray, and H. G. Triiper, Znt. J. Syst. Bacteriol., 38 (1988) 321-325.

MOLECULAR STRUCTURE OF LIPID A

273

193. S.Basu, J. Radziejewska-Lebrecht,and H. Mayer, Arch. Microbiol.. 144 (1986) 213-2 18. 193a. W. Erles, H. Feist, and K. D. Feossmann, Arch. Exp. Veterinurmed., 31 (1977) 203209. 193b. H. Masoud, M. B. Perry, and J. C. Richards, Eur. J. Biochem. 220 (1994) 209-216. 194. R. N. Tharanathan, P. V. Salimath, J. Weckesser, and H. Mayer, Arch. Microbiol., 141 (1985) 279-283. 195. I. M. Helander, L. Hirvas, J. Tuominen, and M. Vaara, Eur. J. Biochem., 204 (1992) 1101-1106. 195a. I. M. Helander, B. Lindner, U. Seydel, and M. Vaara, Eur. J. Biochem., 212 (1993) 363-369. 196. C. Galanos, 0. Luderitz, M. A. Freudenberg, L. Brade, U. F. Schade, E. Th. Rietschel, S. Kusumoto, and T. Shiba, Eur. J. Biochem., 160 (1986) 55-59. 197. V. Lehmann, E. Rupprecht, and M. J. Osborn, Eur. J. Biochem., 76 (1977) 41-49. 198. C. R. H. Raetz, S.Purcell, M. V. Meyer, N. Qureshi, and K. Takayama, J. Biol. Chem., 260 (1985) 16080- 16088. 199. T. Hansen-Hagge, V. Lehmann and 0. Ludentz, Eur. J. Biochem., 148 (1985) 21 -27. 200. M. Imoto, H. Yoshimura, N. Sakaguchi, S. Kusumoto, and T. Shiba, Tetrahedron Lett., 26 (1985) 1545-1548. 201. G. Rosenfelder, 0. Luderitz, and 0. Westphal, Eur. J. Biochem., 44 (1974) 41 1-420. 202. R. Russa and Z. Lorkiewicz, J. Bucteriol.. 119 (1974) 771-775. 203. C. R. Woese, E. Stackebrandt, W. G. Weisburg, B. J. Paster, M. T. Madigan, V. J. Fowler, C. M. Hahn, P. Blanz, R. Gupta, K. H. Nealson, and G. E. Fox, Syst. Appl. Microbiol., 5 (1984) 315-326. 204. P. Fiigedi and J. Kovacs, Fifth Euro. Symp. Curbohydr.,(1989) A-69. [Abstr.] 205. K. Ohno, H. Nishiyama, and H. Nagase, Tetrahedron Lett., 45 (1979) 4405-4406. 206. Y. Inoue and K. Nagasawa, Curbohyd.Res., 168 (1987) 181-190. 207. K. Takayama, N. Qureshi, P. Mascagni, M. A. Nashed, L. Anderson, and C. R. H. Raetz, J. Biol. Chem., 258 (1983) 7379-7385. 208. A. K. Bhattacharjee,H. J. Jennings, and C. P. Kenny, Biochemistry, 17 (1978) 645 -65 1. 209. A. Neszmelyi, K. Jann, P. Messner, and F. M. Unger, J. Chem. SOC.,Chem. Commun. (1982) 1017-1019. 2 10. F. M. Unger, D. Stix, and G. Schulz, Curbohydr. Rex, 80 (1980) 19 1- 195. 21 1. K. Kawahara, H. Moll, P. Kosma, S. Dejsirilert, T. Ezaki, and U. Ziihringer, in Second Conference of thelnternutionulEndotoxin Society, Vienna, August 17- 20,1990. [Abstr.] 2 t2. U. Zannger, K. Kawahara, H. Moll, and P. Kosma, unpublished results. 213. 0. Holst, L. Brade, P. Kosma, and H. Brade, J. Bucteriol., 173 (1991) 1862- 1866. 214. K. Ikeda, T. Takahachi, H. Kondo, and K. Achiwa, Chem. Phurm. Bull., 35 (1987) 13 11- 13 14. 215. M. Imoto, H. Yoshimura, S. Kusumoto, and T. Shiba, Proc. Jupun Acud., 60 (1984) 285-288. 2 16. S. Kusumoto, M. Inage, H. Chaki, M. Imoto, T. Shimamoto, and T. Shiba, Am. Chem. SOC.Symp. Ser., 231 (1983) 237-254. 217. M. Shiozaki, Y. Kobayashi, M. Arai, N. Ishida, T. Hiraoka, M. Nishijima, S. Kuge, T. Otsuka, and Y. Akamatsu, Curbohydr. Res., 222 (1991) 69-82. 218. H. Vyplel, D. Scholz, H. Loibner, M. Kern, K. Bednarik, and H. Schaller, Tetrahedron Lett., 33 (1992) 1261-1264. 219. H. Labischinski, D. Naumann, C. Schultz, S. Kusumoto, T. Shiba, E. Th. Rietschel, and P. Giesbrecht, Eur. J. Biochem., 179 (1989) 659-665. 220. H. Labischinski, G. Barnickel, H. Bradaczek, D. Naumann, E. Th. Rietschel and P. Giesbrecht, J. Bucteriol., 162 (1985) 9-20.

274

ULRICH ZAHRINGER et al.

221. M. Kastowsky,A. Sabisch, T. Gutberlet, and H. Bradaczek, Eur. J. Biochem., 197 (1991) 707- 7 16. 222. H. Labischinski, E. Vorgel, W. Uebach, R. P. May, and H. Bradaczek, Eur. J. Biochem., 190 (1990) 359-363. 223. V. Luzzati, in D. Chapman (Ed.), Biological Membranes: Physical Fact and Function, X-Ray Diffraction Studies of Lipid- Water Systems, Academic Press, London, 1968. 224. U. Seydel, K. Brandenburg, M. H. J. Koch, and E. Th. Rietschel, Eur. J. Biochem., 186 (1989) 325-332. 225. U. Seydel and K. Brandenburg, in A. Nowotny, J. I. Spitzer, and E. J. Ziegler (Eds.), Endotoxin Research Series, Vol. 1, Cellular and Molecular Aspects of Endotoxin Reactions: Conformations of Endotoxin and Their Relationship to Biological Activity, p. 6 1. Excerpta Medica, Elsevier, Amsterdam, 1990. 226. K. Brandenburg, M. H. J. Koch, and U. Seydel, J. Struct. Biol., 105 (1991) 1 1-21. 227. U. Seydel and K. Brandenburg, in D. C. Momson and J. L. Ryan (Eds.), Bacterial Endotoxic Lipopolysaccharides: Supramolecular Structure of Lipopolysaccharides and Lipid A, p. 225. CRC Press, Boca Raton, FL, 1992. 227a. K. Brandenburg, Biophys. J.,64 (1993) 1215-1231. 228. D. Naumann, C. Schultz,A. Sabisch, M. Kastowsky, and H. Labischinski,J. Mol. Struct., 214 (1989) 213-246. 229. G . Lipka, R. A. Demel, and H. Hauser, Chem. Phys. Lipids, 48 (1988) 267-280. 229a. K. Brandenburg,H. Mayer,M. H. J. Koch, J. Weckesser,E. Th. Rietschel, andU. Seydel, Eur. J. Biochem., 218 (1993) 555-563. 229b. A. Shnyra, K. Hultenby, and A. A. Lindberg, Infect. Immun. 61 (1993) 5351-5360. 229c. K. Takayama, D. H. Mitchell, Z. Z. Din, P. Mukerjee, C. Li, and D. L. Coleman, J. Biol. Chem. 269 (1994) 2241-2244. 229d. T. Kirikae, F. U. Schade, U. Zihringer, F. Kirikae, H. Brade, S. Kusumoto, T. Kusama, and E. Th. Rietschel, FEMS Immunol. Med. Microbiol. 8 (1994) 13-26. 230. T. Liideritz,K. Brandenburg,U. Seydel,A. Roth, C. Galanos, andE. Th. Rietschel,Eur. J. Biochem., 179 (1989) 1 1 - 16. 231. K. Brandenburg and U. Seydel, Eur. J. Biochem., 191 (1990) 229-236. 232. 0.Westphal, 0.Liideritz, C. Galanos, H. Mayer, andE. Th. Rietschel, inL. Chedid, J. W. Hadden, F. Spreafico,P. Dukor, and D. WiIloughby (Eds.), Advances in Immunopharmacology: The Story of Bacterial Endotoxin, p. 13. Pergamon Press, Oxford, 1985. 233. K.4. Tanamoto, U. F. Schade, E. Th. Rietschel, S. Kusumoto, and T. Shiba, Infect. Immun., 58 (1990) 217-221. 234. C. Galanos, 0. Liideritz, E. Th. Rietschel, 0. Westphal, H. Brade, L. Brade, M. A. Freudenberg,U. F. Schade, M. Imoto, S. Yoshimura, S. Kusumoto, and T. Shiba, Eur. J. Biochem., 148 (1985) 1-5. 235. J. Y. Homma, M. Matsuura, and Y. Kumazawa, Adv. Exp. Med. Biol., 256 (1990) 101- 119. 236. S. Kotani, H. Takada, M. Tsujimoto, T. Ogawa, I. Takahashi, T. Ikeda, K. Otsuka, H. Shimauchi, N. Kasai, J.4. Mashimo, S. Nagao, A. Tanaka, S. Tanaka, K. Harada, K. Nagaki, H. Kitamura, T. Shiba, S. Kusumoto, M. Imoto, and H. Yoshimura, Infect. Immun., 49 (1985) 225-237. 237. S. Kanegasaki, K.4. Tanamoto, T. Yasuda, J. Y. Homma, M. Matsuura, M. Nakatsuka, Y. Kumazawa, A. Yamamoto, T. Shiba, S. Kusumoto, M. Imoto, H. Yoshimura, and T. Shimamoto, J. Biochem. (Tokyo), 99 (1986) 1203- 1210. 238. S. Kotani and H. Takada, Adv. Exp. Med. Biol., 256 (1990) 13-43.

MOLECULAR STRUCTURE OF LIPID A

275

239. H. Loppnow, L. Brade, H. Brade, E. Th. Rietschel, S. Kusumoto, r. Shiba, and H. -D. Flad, Eur. J. Imrnunol., 16 (1986) 1263- 1267. 240. J. M. Cavaillon and N. Haeffner-Cavaillon, Cytokine, 2 ( 1990) 3 13- 329. 24 1. C. Galanos, M. A. Freudenberg,F. Jay, D. Nerkar, K. Veleva, H. Brade, and W. Strittmatter, Rev. Infect. Dis., 6 (1984) 546-552. 242. L. Brade, 0. Holst, and H. Brade, Infect. Immun., 61 (1993) 4514-4517. 243. L. Brade, K. Brandenburg, H.-M. Kuhn, S. Kusumoto, I. Macher, E. Th. Rietschel, and H. Brade, Infect. Immun.. 55 (1987) 2636-2644. 244. H.-M. Kuhn, L. Brade, B. J. Appelmelk, S. Kusumoto, E. Th. Rietschel, and H. Brade, Infect. Immun., 60 (1992) 2201 -2210. 244a. H.-M. Kuhn, L. Brade, B. J. Appelmelk, S. Kusumoto, E. Th. Rietschel, and H. Brade, Immunbiol. 189 (1993) 457-471. 244b. H.-M. Kuhn, Infect. Immun. 61 (1993) 680-688. 245. W. C. Bogard, Jr., D. L. Dunn, K. Abernethy, C. Kilgarriff, and P. C. Kung, Infect. Immun., 55 (1987) 899-908. 246. R. Chaby, D. Charon, T. Pedron, and R. Girard, Biochim. Biophys. Res. Commun.. 143 (1987) 723-731. 247. N. Kasai, S. Arata, J.4. Mashimo, M. Ohmori, T. Mizutani, and K. Egawa, in A. Nowotny (Ed.),Endotoxin Research Series, Vol. 1, Cellular and Molecular Aspects of Endotoxin Reactions, p. 121. Elsevier, Amsterdam, 1990. 248. T. N. Kirkland, N. Qureshi, and K. Takayama, Infect. Imrnun., 59 (1991) 131 - 136. 249. M. Pollack, K. Oishi, J. Chia, M. E. Evans, G. Guelde, and N. C. Koles, Adv. Exp. Med. Biol., 256 (1990) 331-340. 250. E. J. Ziegler, C. J. Fisher, C. L. Sprung, R. C. Straube, J. C. Sadoff, G. E. Foulke, C. H. Wortel, M. P. Fink, R. P. Dellinger, N. N. H. Teng, I. E. Allen, H. J. Berger, G. L. Knatterud, A. F. LoBuglio, C. R. Smith, and the HA- 1A sepsis study group, New Engl. J. Med., 324 (1991) 429-436. 251. K. Gorelick, P. J. Scannon, J. Hannigan, N. Wedel, and S. K. Ackerman, in C. A. K. Borrebaeck and J. W. Lamck (Eds.), Therapeutic Monoclonal Antibodies: Randomized Placebo-ControlledStudy ofE.5 MonoclonalAntiendotoxinAntibody, p. 16. M. Stockton, New York, 1990. 251a. Y. Fujihara, M.-G. Lei, andD. C. Momson, Infect. Zmmun., 61 (1993)910-918. 252. M. A. Freudenberg, A. Fomsgaard, I. Mitov, and C. Galanos, Infection, 17 (1989) 322329. 253. J. -D. Baumgartner, Infect. Dis. Clin. N. Am., 5 (1991) 915-927. 253a. H. S. Warren, R. L. Damus, and R. S. Munford, New Engl. J. Med., 326 (1993) 1 1 531157. 254, C. R. H. Raetz, in Methods in Enzymology, Vol. 209, pp. 455-466. Academic Press, San Diego, 1992. 255. C. R. H. Raetz, in C. Neidhardt, J. 1. Ingraham, K. Brooks Low, B. Magasanik, M. Schaechter, and H. E. Umbarger (Eds.), Escherichia coli and Salmonella typhimurium: Cellular and Molecular Biology: Structure and Biosynthesis ofLipid A, p. 498. Amencan Society of Microbiologists, Washington DC,1987. 256. J. W. Shands, Jr., J. Infect. Dis., 128 (1973) 189- 193. 257. J. W. Shands, Jr., in S. Kadis, G. Weinbaum, and J. S. Ajl (Eds.), Microbial Toxins, Vol. IV, Bacterial Endotoxins: The Physical Structure of Bacterial Lipopolysaccharides, p. 127. Academic Press, New York, 197 1 . 258. H. Nikaido and R. E. W. Hancock, in The Bacteria, Vol. X, Outer Membrane Permeability ofpseudomonas aeruginosa, p. 146. Academic Press, New York, 1986.

276

ULRICH ZAHRINGER et al.

259. G. Seltmann, in Die bakterielle Zellwand. G. Fischer Verlag, Stuttgart, Germany, 1982. 260. R. Vuorio and M. Vaara, Antimicrob. Agents Chemother., 36 (1992) 826-829. 261. C. E. Ooi, J. Weiss, M. E. Doerfler, and P. Elsbach, J. Exp. Med., 174 (1991) 649655. 262. D. C. Momson, Microb. Pathogen., 7 (1989) 389-398. 263. S. D. Wright, R. A. Ramos, P. S. Tobias, R. J. Ulevitch, and J. C. Mathison, Science, 249 (1990) 1431-1433. 264. T. Kirikae, F. Kirikae,F. U. Schade, M. Yoshida, S. Kondo, K. Hisatsune, S. LNishikawa, and E. Th. Rietschel, FEMSMicrobiol. Immunol., 76 (1991) 327-336. 265. H. Loppnow, P. Libby, M. A. Freudenberg, J. H. Krauss, J. Weckesser, and H. Mayer, Infect. Immun.,58 (1990) 3743-3750. 266. W. Feist, A. J. Ulmer, M.-H. Wang, J. Musehold, C. Schliiter,J. Gerdes, H. Herzbeck, H. Brade, S. Kusumoto, T. Diamantstein, E. T. Rietschel, and H.-D. Had, FEMSMicrobiol. Immunol., 89 ( 1 992) 73 - 90. 266a. B. E. Henricson, P. Y. Perera, N. Qureshi, K. Takayama, and S. Vogel, Infect. Immun.,60 (1992) 4285-4290. 267. J. B. Parent, J. Biol. Chem., 265 (1990) 3455-3461. 268. H. Kumada, K. Watanabe, T. Umemoto, Y. Haishima, S. Kondo, and K. Hisatsune, FEMS Microbiol. Lett., 51 (1988) 77-80. 269. H. Brade, J. Bacteriol., 161 (1985) 795-798. 270. H. Masoud, H. Mayer, T. Kontrohr, 0.Holst, and J. Weckesser,System. Appl. Microbiol., 14 (1991) 222-227. 27 1. R. I. Hollingsworth, R. W. Carlson, F. Garcia, and D. A. Gage, J. Biol. Chem., 265 (1990) 12752.

ADVANCES IN CARBOHYLlRATE CHEMISTRY AND BIOCHEMISTRY, VOL. 50

DEVELOPMENTS IN THE SYNTHESIS OF GLYCOPEPTIDES CONTAINING GLYCOSYL L-ASPARAGINE, L-SERINE, AND L-THREONINE BY

HARIG . GARG,*KARSTEN VON

DEM

BRUCH,? AND HORSTKUNZ-~

*Pulmonary and Critical Care Unit, Department of Medicine, Massachusetts General Hospital, Harvard Medical School, Boston, Massachusetts 021 14; and flnstitutf i r Organische Chemie, Johannes Gutenberg-Universitat.0-55128 Mainz, Germany

I. INTRODUCTION In 1932, the first synthesis of a D-glucopyranosylamine derivative (1) substituted on the nitrogen atom by a glycine residue was achieved by BergH,N-CH-COOH

I I

CHZ

HO*oHHO

H

O

H

~

E

~

o

AcHN

RHN 1 R=COCHzNH2

3

2 R = COCHR'NH2

mann and Zervas (1). Twenty-one years later, Link and coworkers (2), utilizing this method, prepared a series of N-acyl derivativesof 2. Neuberger and associates (3) proposed a 2-acetamido-N-(~-aspart-4-oyl)-2-deoxy-j?-~glucopyranosyl linkage between carbohydrate and protein in egg albumin. In order to obtain a direct proof for the proposed carbohydrate-protein linkage, Marks and Neuberger (4)in 1961 reported the synthesis of derivative 3. Since that time, a number ofglycoproteinshave been found to contain this or other types of carbohydrate-protein linkages. A selection ( 5 ) is given in Table I. The afore-mentioned results have stimulated an interest in the synthesisof glycopeptides for the purpose of supplying structurallywell-defined derivatives as models for biochemical and immunochemical studies. Synthetic 277

Copyright 0 1994 by Academic Press, Inc.

AU rights of reproductionin any form reserved.

HARI G. GARG el al.

278

TABLE In The Amino Acid and Carbohydrate Residues in Carbohydrate-Protein Linkages of Mammalian Origin. ~~

Amino acid L-Asparagine

L-Serine/L-threonine

5-Hydroxy-~-lysine

Correspondingcarbohydrateresidue

Anomeric configuration

2-Acetamido-2-deoxy-~-glycose 2-Acetamido-2deoxy-~-galactose D-XylOSe D-Mannose D-Gdactose

P (Y

B (Y

P

In plants a linkage between Chydroxy-bproline and L-arabinose and in halobacteria a linkagebetween L-asparagine and DGlucose/2-acetamid~2deoxy-D-galactosehas been reported. (1

strategies for glycopeptides are more complex and difficult compared with the synthesis either of oligopeptides or of oligosaccharides. Glycopeptide synthesis requires the protection of amino and carboxyl termini of amino acids and peptides and also the protection of hydroxyl groups of carbohydrate residues. In addition, for stereoselectivesynthesis of specific anomers, participating or nonparticipating groups, respectively,at C-2of the carbohydrate moiety are needed. The respective glycosyl donors have to be coupled with amino acid derivatives to give the desired isomer. Furthermore, conditions are required throughout the synthesis that permit the deprotection of only one functional group without attacking the other protection and also without affecting the anomeric configuration or the breaking of an oligosaccharide bond. For this purpose, several new amino-protecting groups of the urethane type, new carboxyl-protecting groups, and new methodologies have been developed since the first article appeared in this series (6). This chapter focuses mainly on the advances made in glycopeptide synthesis that have been published between 1984 and 1993. Several reviews describing differentaspects of synthetic glycopeptideshave been published during this period (7 - 11). 11. N-GLYCOPEPTIDES 1. Formation of the 2-Acetarnido-2-deoxy-~-~-glucosylLAsparagine Linkage

a. Anomeric Azides as Precursors. -The reduction of a glycosyl azide derivative (4) to glycosylamine (5), followed by a coupling with the 4-carboxyl group of L-aspartic acid (6) in the presence of dicyclohexylcarbodiimide (DCC),was first employed by Marks and Neuberger ( 5 ) for synthesis of 2-acetamido-N-(~-aspart-4-oyl)-2-deoxy-~-~-glucopyranosyl (3). Since

GLYCOPEPTIDE SYNTHESIS

279

2-HN-CH-COOBzl Ad*

HZ

AcO

I I COOH

~

NH,

N3

AcHN

AcHN

4

5

11 DCc

2) depmtenon

'HZ

+

-

3

6 &I

I

PhCH2

2 .PhCHfxn

then, this method of synthesis has been extended to the preparation of a number of derivatives (6) and different condensation methods have been compared (12). In order to construct the L-asparagine derivative of per-0-acetylated 0-(2-acetamido-2-deoxy-/3-~-glucopyranosyl( 1+4)-2-acetamido-2-deoxyD-glucose (7), Spinola and Jeanloz (13) used (the sensitive) silver azide for conversion of the a-chloro anomer of the chitobiose derivative 8 into the /3-azido derivative 9. Kunz and associates ( 14,15) have accomplished the conversion of the a-chloro anomer of 8 into 9 using sodium azide in the presence of tri-n-octyl-methylammonium chloride as a phase-transfer catalyst in chloroform water. Boc-HN- CH-COO-All 10

ACHN AcHh

7 R

-

H,

k

10 R

R'

=

H,

=

I

11 Boc = (CH~JICOCO

8 R=CI. R ' = H

R'=Ns

CHZ COOH

R' = Boc-L-Asn-OAll

9 R-H,

+

All -CHZ;CHGY

NHz

Hydrogenolysis of the azide 9 furnished the amino derivative 10, which was subsequently coupled with N-tat-butyloxycarbonylaspartate( 16) (11) in the presence of ethyl 2-ethoxy-1,2-dihydroquinoline-1-carboxylate W D Q ) ( 17). Warren and associates ( I 8) have prepared a glycosyl azide derivative (15) of a heptasaccharide. This glycosyl azide was obtained from 0-a-D-mannopyranosyl-( 1-6)-O-[a-~-mannopyranosyl-( I -3)-O-a-~-mannopyrano syl - ( 1+6) - 0-a - D - mannopyranosyl- ( 1 3)] - 0-/3-D - mannopyranosyl( 1-4) - 0 - (2 - acetamido - 2 - deoxy - /3 - D - glucopyranosyl) - ( 14 4 ) - 2 acetamido-2-deoxy-~-glucopyranose (12) by treatment of its peracetylated derivative 13 with trimethylsilyl trifuoromethanesulfonate,followed by reaction of the intermediary oxazoline 14 with trimethylsilyl azide. Reduction of the glycosyl azide 15 in the presence of Lindlar catalyst gave the glycosylamine derivative 16. The condensation of 16 with I-tert-butyl N(9-fluoren-

-

280

HARI G. GARG et al.

Frnoc~HN- CH-

RO & o AcO . + l

NH,

+

AcHN

R'-HN-

COOtBu

I CH2 I

CH-COOF

I

0.O-dtelhyl phosphowlqanide

c

I

COOH

16 R - X '

17 IBu

=(CHIIIC

RO AcO * ; = O

NH AcHN

18 R = X', R = Frnlx, R" = tBu

19 R = X ' R ' = R " = H

ylmethoxycarbony1)-L-aspartate (17) in the presence of 0,O-diethylphosphoryl cyanide gave the corresponding glycosylasparagine derivative 18, which was partially deprotected to furnish derivative 19.

b. In Situ Anomerization Procedure for Attachment of CX-L-FUCOS~~ Units. -The synthesis of the 0-protected trisaccharide azide 20 containing an a-L-fucosylresidue was achieved ( 19c)by using a chitobiosylazide derivative containing a free hydroxyl group at C-6. The coupling was performed under in situ anomerization conditions (20) with benzyl-protected a-fucopyranosyl bromide ( 19b). Selective hydrogenation employing neutral (!) washed Raney nickel gave the protected trisaccharide glycosylaminederivative 21. Derivative 21, when coupled with 1-tert-butyl N-allyloxycarbonylL-aspartate (22) in the presence of EEDQ, gave (19) the protected fucosylchitobiose-L-asparagine derivative 23. Similarly, the per-0-acetylated

GLYCOPEFT'IDE SYNTHESIS

28 1

BzlO

H3C O ,Ac

Aloc-HN-

CH-COOIBu

22

20 R=N,

Ala IBu

21 R = N H 2

-

CH+H-CHzCCC

=(CH3I3C

Aloc-HN- CH- COOfBu

trisaccharide azide having the 4-0-(a-~-fucopyranosyl)lactosaminestructure was synthesized by reaction of 2,3,4-tri-0-(4-methoxybenzyl)-a-Lfucopyranosylchloride with a lactosamine azide derivative, subsequentoxidative removal of the 4-methoxybenzyl (Mpm) protecting groups, and acetylation of the liberated hydroxyl groups. Reduction of the azido function and condensation of the resulting amine with 1-ally1N-trichloroethoxycarbonyl-L-aspartate afforded the 4-0-(a-fucosyl)lactosamine glycopeptide having the Lewis" antigen structure (21). c. 1-Isocyanates as Precursors.-Per-0-acetylglycosyl isothiocyanate derivatives of monosaccharides were first coupled with L-aspartate derivative 6 by Khorlin and coworkers (22) to form l-N-(~-aspart-4-oyl)glycosyl derivatives. This procedure was modified and has successfullybeen applied for the synthesisof the P-D-mannosyl-chitobiosyl-L-asparagheglycopeptide 24 via reaction (23)of the per-0-acetyl derivative of 25 with L-aspartate derivative22. It is noteworthythat attempts to prepare the 1-azido derivative H,N-CH-COOH

I H

O

~

o

H

~

o

&

/

~ &HZ o

HO HO AcHN

AcHN

24

O H .& HO HO AcHN 25

N=C=S AcHN

HAM G. GARG et d.

282

26

I CH3

Trimsfhylsllyl ulde I SKI,

ACO

NR

AcO AcHN

AcHN

27 R = N 2 28 R = H?

(26) of this trisaccharide by treatment of oxazoline 27 with trimethylsilyl a i d e and trimethylsilyl trifluoromethanesulfonate or SnCl,, under conditions described by Warren and associates (18), with the goal of synthesizing the corresponding 1-amino trisaccharide derivative 28, were unsuccessful. d. Reactions of 1-Hydroxy-D-glycosamineswith Ammonium Hydrogen Carbonate. -2-Acetamido-2-deoxy-~-glucose (24,25)(29) and per-O-acetylated chitobiose (26,27) (30) have been reported to react directly with ammonium hydrogen carbonate (NH,HCO,) to give the 1-amino derivatives 31 and 32, respectively. These derivatives were coupled with L-aspartate derivative 33 to give the corresponding glycosyl- L-asparagine derivatives 34 and 35. Derivatives 34 and 35 are useful intermediates for solidphase synthesis of glycopeptides (described in Section V).

-../.g+ HO

AcHN

OH

or

& o O c*:A c

AcHN

29

AcO

AcHN

OH

30

31

32

GLYCOPEFTIDE SYNTHESIS Fmoc-HN-CH

283 Fmw-HN-CH

-COOtBu

Fmoc-HN -CH

&I,

C Hz

-COOtBu

I

I c=o

I

or

AcO

NH

...

, HCtlN

&OOPIp

-COOtBu

I

II

AcHN

33 34 PIP

=CeF5-

IBU

= (CH313CC

Fmac- =

35

CH,GCOI

2. Selective Deprotection and Peptide-Chain Enlongation

Glycopeptides (3) containing a 2-acetamido-N-(~-aspart-4-oyl)-2-deoxy/3-D-glucopyranosylresidue were obtained, together with their regioisomers, by using the benzyloxycarbonyl(Z)-protected L-aspartic anhydride and glycosylamine 5. After separation of the regioisomers,elongation of the peptide chain was performed in 3 in the direction of the carboxyl terminus by Garg and Jeanloz (28- 30) and in the direction of the amino terminus by Shaban and Jeanloz (31). In the past several years Kunz and associates (1416,19,21,32 - 37,39) have developed differentprotective groups and protecting-group combinations of orthogonal stability (that is, the alternative stability and cleavability of two or more protecting groups) for the amino and carboxyltermini of such glycopeptidesas 36 and have used them extensively for elongation of the peptide chains of glycopeptides. The allyl ester as a temporary protecting group (32)for protection of the L-aspartic acid residue of the chitobiose glycopeptide 36 was introduced (14) for elongation of the peptide chain. The allyl ester group may be removed either by rhodium(1)-catalyzed (16) or by palladium(0)-catalyzed (32) cleavage to give ( 14,15) the carboxyl-deprotected glycopeptide derivative 37. Chain lengthening at the C terminus with peptide allyl esters in the presence of EEDQ ( 17) gave derivatives 38 and 39, respectively. The free glycopeptide Boc-HN- CH -COOAll

Boc-HN-CH-COOH

I

36

I

37

HARI G. GARG et al.

284

Boc HN-CH-COOR

I I

CH,

+ peplide ally1 esfers (H (Xaa),-OAll:

-

EEDa

2 & c

* c :A

AcHN

38 R 39 R

AcHN L-Phe-L-Thr-OAll = L-Ala-L-Thr~L Ala~L~Ser OAll

=

amide 40 was obtained by treatment of 38 with methanolic ammonia (1 5,33). Treatment of39 with trifluoroaceticacid resulted (15,33) in removal Boc-HN-

CH -COO-L-Phe~L-Thr-NH,

I

*o&i=o

CHZ

NH

HOH O

HO AcHN AcHN 40

of the tert-butyloxycarbonyl(Boc) group to give 41. The protecting-group combination used in these syntheses offers the possibility of extension of the peptide chain in both directions. CF,COOH

I

-COO-L-Ala

L-Thr-L-Ala-L-Ser OAll

CHZ

-

I

c=o

/OAC

CF3COOH

39

H,N-CH

\

ArUhl

41

The noteworthy advantages of the allyl ester are (a)it is readily introduced into amino acids; (b) after isomerization (to 1-propenyl) by a palladium(0) catalyst it may be removed under weakly acidic or basic and neural conditions (32), even if sulfur-containingamino acids are present (34); (c)it shows orthogonal stability to the tert-butyloxycarbonyl and 9-fluorenylmethoxycarbonyl groups (lo); and (d) it is not affected by the hydrogen fluoridepyridine complex (35). The allyl protecting group has also been demonstrated to be useful for blocking of side-chain functionalities(36). Thus, the palladium(0)-catalyzed

GLYCOPEPTIDE SYNTHESIS

285

removal of the P-ally1 ester protection from asparticacid peptides, even from Asp - Gly and Asp - Ser sequences, was achieved without transpeptidation (36). The combination of the N-terminal allyloxycarbonyl (Aloc) group (34) with the tert-butyl ester proved to be a particularly efficient tool in the synthesis of complex glycopeptides,as has been demonstrated in the synthesis of sensitive fucosyl chitobiose glycopeptides (19). The Aloc group was 0

H-NH

v O C C " IIH \

OAc H & h o A c

CoOtBu

,COOtBU I

CH

I CHz ..dN

43 42

Aloc-Ala-Leu-NH

Me2N(CH2)3N=C=NE~ IDAECI

OAc

H,C-OAc

Aloc-Ala-Leu-

coot~u \

Aloc-Ala-Leu-OH

,

'CH {HZ

NH

, ,CWH \ , cn

CFiCOOH

I

,OAc

1c = o

n

44

45

/

J

46

AcHN

removed from the trisaccharide-asparagine conjugate 42 with complete chemoselectivity by palladium(0)-catalyzedally1 transfer to N,N-dimethylbarbituric acid as the allyl-trapping nucleophile, to give the N-deprotected conjugate 43 almost quantitatively. Chain extension using a water-soluble carbodiimide furnished 44.

HARI G. GARG et al.

286

Removal of the tert-butyl ester protection from 44 proceeded quantitatively to give 45 by the action of trifluoroacetic acid. Because of the indirect protection by the 0-acetyl groups, the sensitive fucoside bond remained absolutely stable. It is noteworthy, that the fucoside bond in the analogous fucosyl chitobiose derivative 23, partially protected by benzyl ether groups, was completely decomposed upon treatment with trifluoroacetic acid. Chain extension on 45 furnished the trisaccharide hexapeptide46, which is a partial sequence of an envelopeglycoprotein of a leukemia virus ( 19).Complete deprotection and the binding of the synthetic glycopeptide to bovine serum albumine has been achieved (19). The synthesis of the totally protected 2-acetamido-1-N-(~-aspart-4-oyl)-2-deoxy-~-~-ghcopyranosy~ derivative 47 containing the I ,3-dithian-2-ylmethyl (Dim) ester group has been described (37). SelectiveC-terminal deprotection to give derivative 48 was carried out by oxidation to the disulfone and its immediate /I-elimination. These cleavage conditions do not affect the Boc group, the 0-acetyl Boc-HN-

CH-

COODim

Boc-HN-CH-COOH

I

CH,

AcHN 48

groups, or the glycoside bond. In this way, the synthesisof glyco-tri-peptides and -tetrapeptides by amino- and carboxyl-terminal elongation of the peptide chain has been achieved (37). Using the 2,2,2-trichloroethoxycarbonyl(Teoc) group (38) in combination with the ally1 ester protecting group, Kunz and Marz (39) have synthesized glycosylated peptide T derivatives, which are partial structures of the -glycoproteingp 120 of the human immunodeficiency virus HIV- I. In this context, the N-Teoc group was selectively removed from the per-0acetylated Lewisa antigen trisaccharide- asparagine conjugate 49 by using zinc in acetic acid. Subsequent chain-extension furnished the trisaccharide

Hy9

R-HN-CH-COOAll

II

A A J * m c T OAc o

AcHN

CH2

AcHN

49 R = C13C-CH2-O-C0 (Tcoc)

50 R = Boc-Ala-Ser(1Bu)-Thr(lBu)-Thr(tBu)-Thr(tBu)

GLYCOPEPTIDE SYNTHESIS

287

hexapeptide allyl ester 50, which was carboxyl-deblocked via palladium(0)-catalyzedcleavage and extended to the N-glycosyl peptide T derivative. Complete deblocking of the glycosylated HIV- 1 structure has been achieved (39). Analogous glycosylated peptide T derivatives have also been synthesized by solid-phase glycopeptide syntheses employing allylic anchoring groups (40). 111. 3-0-GLYCOPEPTIDES OF L-SERINE OR L-THREONINE

1. Formation of the 3-0-Glycosidic Linkage a. Koenigs- Knorr Methods. -Jones and coworkers (4 1) prepared the 3-0-(2-acetamido-2-deoxy-/3-~-glucopyranosyl)-~-se~ne derivative 51 by treating 2-acetamido-3,4,6-tri-~-acetyl-2-deoxy-~-~-glucopyranosyl chloride with N-(benzyloxycarbony1)-L-serinemethyl ester in the presence of silver carbonate.Bencomo and Sinay (42)used the N-benzyloxycarbonyl (Z) group in combination with the tert-butyl ester for the carboxyl protection of L-serine. The tert-butyl group was removed with trifluoroacetic acid without cleaving the 0-glycosidic bond. Nevertheless, glycopeptides involving 0-glycosylated L-serine/L-threonine are more or less acid-sensitive(hydrolysis and anomerization)and alkali-labile@-elimination).This holds true, in particular, for the /3-xylose - L-serine bond.

Ace++

2-HN-CH-

I

COOMe

2-HNSilver carbonale

CH-

I

COOMe

AcO

AcHN CI

OH iHz

ACO ACO

ACHN 2 = C,H,CH,OCO.

51

To overcome these difficulties in the selective deprotection and chain extension, several carboxyl-protectinggroups, namely, allyl ( 16,32), benzyl (43,44), tert-butyl(42), 2-bromoethyl(45), 2-chloroethyl(45), heptyl(46), 4-nitrophenyl(47,48),and pentafluorophenyl(49)for L-serine/L-threonine have been introduced or applied. Similarly, amino-protecting groups for L-serine/L-threonine that have proved useful for the synthesis of glycopeptides are tert-butyloxycarbonyl (50), 9-fluorenylmethoxycarbonyl (43,44,48), 2-(2-pyridyl)ethoxycarbonyl (5 1), 2-(4-pyridyl)ethoxycarbonyl (44,52),and 2-triphenylphosphonioethoxycarbonyl(53).Some applications of these groups have been discussed in earlier reviews (7- 1 1). Schultheiss-Reimannand Kunz(43) applied silver triflate as the promoting reagent for the reaction of 2,3,4-tri-0-benzyl-c-~-glucopyranosyl bromide (52) with N-(9-fluorenylmethoxycarbonyl)-~-serine benzyl ester to

HAM G. GARG et al.

288

furnish the 3-0-(j?-glucopyranosyl)-~-serinederivative 53. The protectinggroup combination Fmoc/OBzl of 53 allows versatile and selective deprotection and chain extension for the synthesis of glycopeptides. Removal of the Fmoc group from 53 was achieved selectively and quantitativelyby using morpholine (43,44). Under these conditions, the 0-glycosyl- serine (- threonine) linkage is unaffected. Peptide condensation gave the glycotripeptide 54, which was completely deprotected by subsequent application of morpholine to remove the Fmoc group, hydrogenolysis of the benzyl ester, and methanolysis of the carbohydrate ester groups catalyzed by the weak base hydrazine, to give (43-45) 55. This methodology (4334) proved to be

Fmm-NH-CH-C00Bz1

I

1. Maphollne

2. Fmc-Asn-Leu-OH

Fmm-Asn-Leu-NH-CH

*

-COOEzl

1. MOpholine

2 H*/P&C

H-Asn-Leu-NH-CH-CWH (OH

I

FHZ

of quite general applicabilityfor glycopeptide syntheses, as in the synthesisof tumor-associated T,- and T-antigen glycopeptides (44) or in solid-phase synthesis (see Refs. 25, 26, 55, and 56; see Section V). Kenne and associates (57) have applied this procedure for the synthesisof a-D-mannopyranosylderivativeslinked to L-serine/L-threonine(59 and 60). Compounds 59 and 60 were obtained by coupling 2,3,4,6-tetra-O-acetyl-~mannopyranosyl chloride (56) with Fmoc-L-serine benzyl ester or Fmoc-Lthreonine benzyl ester in the presence of silver triflate and 4-A molecular sieves, followed by deprotection (57). Paulsen and associates (5 8) synthesized mono- and di-D-galactopyranosyl derivatives of L-serine/L-threonine. Condensation of the 2-azido-2-deoxyglycosyl halides 61 and 62 with the benzyl or tert-butyl esters of N-benzyloxycarbonyl (Z)-protectedL-serine, L-threonine,and L-leucyl-L-serine(6365) in the presence of silver carbonate, silver perchlorate, Drierite, and molecular sieves gave (59) the corresponding 0-glycopeptides67 -69. The free glycopeptides were obtained after total deprotection.

GLYCOPEPTIDE SYNTHESIS

289 Fmoc-HN-CH-

Fmoc-HN-

CH-

COOBrl

I HC-R I OH

AcO

511YBilrillae

nlol&"lor

I

*

SiBYBS

* 56

COOBzl

I HC-R

A

c 6 AcO

c

57 R = H 58 R = C H 3

O OAc e o a) Morphaline

R = H, or R = CH,

b) HydrOgBndySiSIPdC c) Hydmine hydrate

H,N-

CH-

COOH

I

HC-R

I

59 R = H 60 R = C H 3

HO

2-L-Leu-HN-

.

I

61

I

0

+

O , Ac

CH-COOBzl

R'-HN-CH-COOR

+ 2-A

w

I I

HC-R

&:+

ACO AcO

OAc

A

C

N3 62

O

F

OAc

Br

AcO

O

&

~

AcO

OAc

63 A = L Leu-L-Ser-OBzl

67 R = H, R' = 2. R"

64 A = L-Ser-OlBu

68 R =H, R'=Z-L-LeU, R " = Bzl 69 R = CH3, R' = 2, R" = IBU

65 A = L-Thr-OlBu

=

IBU

b. Trichloroacetimidate Method. -Kunz and Waldmann (32) applied the trichloroacetimidate method (60) to couple the tri-0-benzylated xylosyl residue of 70 to N-(benzyloxycarbony1)-L-serineally1 ester (71) to give the xylosyl-L-serine derivatives 72 and 73 (ratio ofp to a,6 : 1). In order to obtain the biologically interesting sialyl T, antigen glycopeptides, N-(benzyloxycarbonyl)-O-[methyl(5-acetamido-4,7,8,9-tetra-Oacetyl-3,5-d~deoxy-~-g~ycero-a-~-gu~uc~~2-nonu~opyranosy~)-onate-(~

-

290

HAM G. GARG et al. Z HN-CH-COOAll

I

I

0F3. - 3 P C

-COOAll

Z-HN- CH- COOAll

I

CHZ

I

+

OH

BZlO 70

-

2-HN-CH

71

II NH

72 73

3) - 0 - (2,4,6- tri - 0 - acetyl -p- D - galactopyranosyl)]- ( 1 3) - 0 - L - seryl[methyl(5 -acetamid0- 4,7,8,9-tetra- 0-acetyl - 3,5 -dideoxy-~-glycero-a-D gulucto-2-nonulopyranosyl)onate[(2 + 6)-0-(2-acetamido-4-0-acetyl-2deoxy-a-~-galactopyranosyl)]-(1 3)-~-serinebenzyl ester has been synthesized by employing the corresponding trichloroacetimidate derivative and N-(benzyloxycarbony1)-L-serine benzyl ester. The ratio of the a- and p-glycosides formed (6 1) was 2 :5. --j

-

c. Glycosyl Fluorides.-The synthesis of a mixture of a and p-D-xylopyranosyl-L-serinederivatives(72 and 73) and the 2,3,4,6-tetra-O-pivaloylp-D-glucopyranosyl-L-serinederivative 76 from the glycosyl fluorides74 and 75 and N-(benzyloxycarbony1)-L-serineally1 ester (71) in the presence of boron trifluoride etherate has been achieved (62). The stereoselectiveformaZ-HN-CH,OPlV

+

73

F 74

COOAll

I

72

or

0

plvo+~Hz PWO PlVO

75

tion of derivative76, the stabilityofglycosyl fluorides, and the homogeneous reaction proceeding without heavy-metal promotion make this procedure an interesting alternative of Koenig- Knorr methods. d. 1-Thioglycosides.-Thioglycosides have been introduced as efficient glycosyl donors (63). The ethyl 1-thiomono (77 and 78)-, di (79and 80)-,and tri (81)-saccharides were condensed with N-(benzyloxycarbony1)-L-serine benzyl ester (82) in the presence of dimethyl (methy1thio)sulfoniumtrifluoromethanesulfonate(DMTST) (63) as promoter to give (64) the corresponding a-0-glycopeptides (83-87). Derivative 87, on reduction with hydrogen sulfide and acetic anhydride gave the 2-acetamido-2-deoxy derivative 88, which was completely deprotected by reduction and the action of methanolic ammonia to give (64) 89. Thioacetic acid (65) has also been employed for converting the azido group of carbohydrate residues to an acetamido

GLYCOPEPTIDE SYNTHESIS

29 1

group. An alternativedirect transformationof 2-azido-2-deoxyglycosylpeptide derivatives to the corresponding 2-acetamido derivatives has been achieved by applying a Staudinger reaction with tri-ocylphosphine in dry acetic acid (67,68). CoOBzl

2-HN-CH-

I y

-

RO&O

TI R = A c

78 R

=

2

I

RO

Erl

OR

ZHN-CH-CCOBzl

I

RO

Z-HNRO

SC2Hg OR

CHH 'Z

+

I

N,

CH,

COOEzl

- +

I

RO@o&'

OH 82

OR

R'O

RO

OR

2-HN-

CH-

I

C00Erl

CH,

I

EzlO AcO

OEzl 81

Ad&AcO

OAc 87

Z-HN-CH-CCOEzI

H2N-CH-CCOH

I

I

OH

HO

89

HAM G. GARG et al.

292

a-D-Mannopyranosyl-(1 + 2)-a-~-mannopyranosylderivatives of L-serine and L-threonine glycopeptides (95 and 96) were prepared by treating ethyl 2,3,4,6-tetra-O-benzyl-1-thio-a-D-mannopyranoside (90) with bromine, followed by treatment with ethyl 3,4,6-tri-O-benzyl-1-thio-a-D-mannopyranoside (91) to give the ethyl 1-thio derivative of the disaccharide 92. Treatment of 92 with Fmoc-L-serine or L-threonine benzyl ester gave 93 and 94, respectively. On total deprotection,the freeglycopeptides95 and 96 were obtained (66).

BZlO

Fmoc-HN-CH-

I HC-R

H,N

COOBzl

-CH-

COOH

I HC-R I

I I

2 + 0 Ho

I

93 R - H 94

95 R = H 66 R=CH,

R-CH,

Using the ethyl 1-thioderivative (97)of 2,3,4-tri-O-benzoyl-~-xylose, the fully protected and free 0-glycopeptides99 and 100, having the N-terminal amino acid sequence 3 to 6 (98) of the protein core of a proteodermatan sulfate have been prepared (69).

B

z

~

z

~0 0 2S

97 Bz=C,H5C0

C

2

+ H

2-L-Ala-L-Ser-Gly-L-Ile-OBzl I 5 OH

98

DMTST

R"-L-Ala-L-Ser-Gly-L-IleOFI R O bR -

OR 99 R = Bz, R' = C6H&H2, R" = C6H,CH20CC

100 R = R ' = R " = H

e. 1-0-Acetyl Activation. -Glycopeptides 101 and 102, containing the A3- A, sequence of proteodermatan sulfate and xylose, have been synthe-

293

GLYCOPEPTIDE SYNTHESIS

sized alternativelyby activation of the 1-0-acetyl group of the xylose derivative 103 with trimethylsilyl triflate (69,70). Thus, compound 103 reacted with 104 and 105 in the presence of trimethylsilyl triflate to give the protected glycopeptides 106 and 107, respectively, which on deprotection yielded glycopeptides 101 and 102.

ii~meihyliilyl

A

~

a

~ OAc

O

R-L-Ser-Giy I c OH

+A

103

-L-lle-OBzl

rrdla,e

R-L-Ser-Gly-L

y&+b

OAC

104 R = CeH50C0

106 R

105 R = C,H,OCO-L-Ala

107 R

R -L-Ser-Gly

Ile~OR

-

-

C,H,CH,OCO C,H,OCO-L-Ala

-L-lle-OH

101 R = H

102 R = L-Ala

f. Electrophile-Induced Lactonization of Glycosyl 4-Pentenoates.Carbohydrateshaving an anomeric pent-4-enoic acyl group have been used as glycosyl donors to prepare glycopeptides (7 1). Thus, condensation of the glucopyranose derivatives 108 and 109 with pent-4-tenoic acid in the presence of N,N'-diisopropylcarbodiimide and cuprous chloride gave the protected glucopyranosyl4-pentenoates 110 and 111, respectively. Activation

+

R3 R'O

4penlenoic acid. CuCI.

R3R'O

N.N'diisoptopyl carbodiimide ~

0

OH

R2

1W R'

= Bzl. R2 = R3 = OBzl

110 R'

Frnoc-HN--CH-COOBzl

=

BzI.

OH NIS = N-lodasucontmlde

NIS

lC%CF$%H

-

R3 = OBzl

Frnoc-HN-CH-COOBzl

I

CH,

R2

BzlO

0

HARI G. GARG ef a1

294

0 Teoc- L-Ser OH I 111

-L-Ala-

Teoc-HN - CH- C

OIBu

+ 113 Tecc = CI&CH,OCO

I

A**/cH2 AcO

'%

-L-Ala-OtBu

"

114 (aP,O100)

of 110 and 111 with soft electrophiles, such as iodonium compounds, [symcollidine,I]C10, or N-iodosuccinmide (NIS), trifluoromethanesulfonic acid, or l73-dithian-2-y1iumtetrafluoroborate in the presence of 3-A molecular sieves or Drierite and treatment with Fmoc L-serine benzyl ester or the serine peptide 113 gave glycopeptides 112 and 114, respectively. Condensation of 111, having a participating group at C-2, proceeds with complete stereoselectivity.Because of the mild reaction conditions, this method is of potential interest for the glycosylation of preformed peptides. 2. Methods of Selective Deprotection, Followed by Peptide-Chain Elongation

Selectivemethods for deprotectingglycopeptideseither at the amino or at the carboxyl terminus have been developed during the past decade by introduction and application of different combinations of protecting groups (Table 11). The protected 0-xylosyl derivative (1 15) of L-serine (43) B-D-gdactopyTABLEI1 Different Amino and Carboxyl Protecting-GroupCombinations of L-Serine/L-Threonine Used for Glycopeptide Chain-Lengthening Terminus -NHz

Benzyloxycarbonyl 9-Fluorenylmethoxycarbonyl

2-(4-Pyridyl-ethoxycarbonyl Trichloroethoxycarbonyl Allyloxycarbonyl

-COOH Ally1 ferf-Butyl MY1 Benzyl 4-Nitrophenyl Pentachlorophenyl Pentafluorophenyl Phenacyl Benzyl MY1 ferf-Butyl

Reference 32,68 42 67,72 43,44,56,13 14 47,48,15 75 49 66 44,73 39 19,33,76

GLYCOPEPTIDE SYNTHESIS

295

ranosyl - (1 + 3) - 2 - acetamido - 2 - deoxy - a - D - galactopyranosyl - L serine (44,73,74) (116), and the analogous L-threonine compound (44,73,74) 117 having the Fmoc group at the amino terminus and the benzyl group at the carboxyl terminus were used successfully for the synthesis of 0-glycopeptides. As was first demonstrated for 0-glucosyl and 0-xylosyl serine derivatives (43), this protecting group combination allows selective deprotection of either the amino terminus by using morpholine (43,44) or FrnooHN-

CH-

-CH -COOBzl

COOBzl

Frnoc-HN

I

II HC-R

CHZ

I

I

0

0

LOBZ 115

116 R = H 117 R=CH,

H

B-D-Galp(l-13)-a-D-GalpNAc-(l

-to)-L-SerI

P-D-Galp(l-13)-a-D-GalpNAc-(l +O)

I

-L-Ser

I

~-D-Galp(l+3)-a-D-GalpNAc-(l-+O) -L-Ser

AH

118

the carboxyl terminus by hydrogenation (8). Beginning at the carboxyl end and in the presence of EEDQ as the coupling reagent, these derivativeswere used (74) as a step-by-step synthesis of the free glycosylated tripeptide 118. Another combination of orthogonally stable protecting groups, which proved very useful in glycopeptide synthesis, consists of the Fmoc group together with the ally1 ester (67). Removal of the Fmoc group from glycoFrnoc-HN-CH-COOAll

I

HC-R I

I

H2N- CH-COOAll

I

. HC-R

I

morphdrm

AcHN 0

12l R=H.

119 R - H

122 R = C H 3

120 R=CH,

Fmoc-HN-CH-CCOH

I

HC-R

I A

c

AcO

O

e

' 123 R - H 124 R -CH,

HAM G. GARG et al.

296

Frnoc

121

+

123

122

+

a - D Ac,GalpNAc-[l+O) llD0

or

___)

u-D-Ac,GalpNAc-(l+O)

124

Fmoc

I -L-Ser I -L-Ser I OAll

a-D.AclGalpNAC-(l+O)

I I

-L-Thr

or

a-D-Ac~GGalpNAc-(l+O)-L-Thr

I

OAI 126

125

peptides 119 and 120 by morpholine gave the amino-deprotected derivatives 121 and 122. Selectiveremoval of the ally1ester group from 119 and 120 was performed by a pallidium (0)-catalyzed allyl-transfer reaction. However, as the Fmoc group is sensitive to morpholine, N-methylaniline a very weak base was applied as the allyl-trapping nucleophile. Under these conditions, the selectively deprotected compounds 123 and 124 were obtained in high yields and without side reactions. Coupling of 121-124 in the presence of IIDQ furnished the protected diglycosyl peptide derivatives of 125 and 126, which have the partial structure of human glycophorin(67). Similarly,other glycopeptide derivatives containing T-antigen disaccharide units (127 and 128) have been synthesized by elongation of the peptide chain (67).

p-D-Ac,Galp(l-+3)-a-D-Bz~GalpYAc.(l+O) -Xaa

I

p-o-Ac,Galp(l~3)-a-o-B~2GalpNAc-(1+0) -Xaa

12'

Xaa=L-Ser

126

L-Thr

I OAll The Fmoc/OAll protecting-groupcombination has also successfullybeen applied in the synthesis of glycopeptides constituting the linkage-region structures of proteoglycans (72). Using Fmoc amino protection and the phenacyl ester of L-serine, the glycopeptides129 and 130 have been obtained (66). H H a-D-NeupsAc-(2-6)-a-D-GaiFNAc.(l+o)

-Me,

a-D-NeugAc-(2+6)-a-~-GalFNAc-~l+O) -L.Ser

129

I

I a-D-NeuFsAc-(2+6)-a-D-G~lpNAc-(l+O) -L-Ser

I I -L-Ser I L-Val

I

o.D-NeugAc-(2+6)-a-D-Gal~Ac-(l+0) -L.Ser

I

a - D - N s u p s A c . (2 + 6 )-a -D -G ~ l ~ A c -0) -(l

L-Val I

OH

130

I

OH

Synthesisof a number of 0-glycosides (such as 131 - 134) having different chain lengths between amino acids A, and Alo, which are part of interleukin-2, was accomplished by application of the Fmoc group in combination with the tert-butyl ester (77,78).

GLYCOPEFTIDE SYNTHESIS H L-Ala-L-Pro-L-Thr

-L-Ser-L

291

H-L-Ala-L~Pra-L-Thr -L-Ser-L-Ser

Ser-OH

OH

I

I

(1~DAc,Gal@Ac

p - ~ ~ G a l --t3)-u-D~Ac3GalpNAc p(l

132

131

The synthesis were effected with the appropriate glycosylated derivatives of L-threonine (135 and 136). Selective removal of the ally1 ester derivative Z- L-Ala- L-Pro -L-Thr

Frnoc-L-Thr - 0 1 B u

I a-D-Ac,GalpNAc

- O!Bu

I u-D-AclGalpNAc

135

136

137 by using [(Ph,P),Pd(O)] catalyst to give 138, followed by coupling with dipeptide 139, gave (68) the C-terminally extended glycopeptide 140. 2-L-Ser -L-Ala-L-Ala-

OAll

2-L~Ser-L-Ala-L-Ala-

OH

llPh3Pl.Pdnl. morpholine

AcO

137

AcO

138

2-L-Ser -L-Ala-L-Ala-Gly-L-Ala-OAll AcHN

I

Tumor-associated T, and T-antigen type 0-glycopeptides (144 and 145) have been obtained by starting with the Fmoc/Bzl protecting-group combination. Continuing with condensation of the amino-deblocked component with 2-(4-pyridyl)ethoxycarbonyl(Pyoc-~-serineyielded the glycopeptide derivative 141. Hydrogenolysis to 142 and coupling with 143 promoted by EEDQ gave the TN-antigen glycopeptide 144. In the selective deprotection of the carboxyl terminus, the Pyoc group has an advantage as it is stable to hydrogenolysis. The T-antigen glycopeptide derivative 145 was obtained similarly (44,73). P y x - L-Ser -L

ACO@

AcO

Pyoc-L-Ser -L-Ser -OH

Ssi- OBd hydropnolysia

141

~

AcO@

AcO

142

HARI G.GARG et al.

298

L-Thr -0Bzl

-0B21

Pyoc-L-Ser-L-Ser-L-Thr

I Y

EEDO

A&

143

AcO

AcO

144

Y=

145 Y = AcO%

I Y

AcO OAc

AcHN

The Pyoc group is stable to acids and bases, but may be selectively removed after conversion into the N-methylpyridinium form and subsequent treatment with morpholine, to give 146. Treatment of 146 with acetic anhydride in pyridine gave 147. Hydrogenolysis of the benzyl ester group of 147 followed by methanolysis catalyzed by hydrazine afforded (44,73) the deprotected derivative 148.

-

a-D-Ac,Gal@Ac

I

1 CYl

2 mOrphOi,ne

I

H

a-D-R,GalENAc

I

i pyrrdlne AC&

H-L-Ser-L.Ser-L-Thr-oB~1

144

I

Ac-L-Sar-L.Ser-L-Thr-OR

I

R

~-D-Ac,G.~&NAc 146

1. H P d - C 2 hydrazine. MeOH

I

a-D-R,Gal@Ac

L

147 R = Ac. R = OBzl 148 R = R. =

IV. BINDINGOF GLYCOPEPTIDES TO PROTEINS As proteins are insoluble in organic solvents, their coupling with glycopeptide derivatives having a definite structure has to be effected in water solution. Thus, the carboxyl-deprotected glycopeptide derivative 148 has been coupled with bovine serum albumin (BSA) (about 25 glycopeptide molecules per protein molecule), presumably at the 6-amino group of L-lysine residues, to give (44,73) the conjugate 149. 1-~hyl-~(3dimethylimnopmyl)

o M l m d e hydmhlands

148

+

(DAEC)

BSA I-hydroxybenrbna2ole (WBI)

a-D-GalmAc

-

Ac

- L.Ser -L-SaII -L-Thr-NH-ESA

I

a-D-Gal@Ac 149

A glycopeptide- BSA conjugate (150) containing the tumor-associated T-antigen structure has been constructed (44,73). P-D-Galp(1-3)- a - 0 - G a l p A c

I

Ac-L-Sel-L-Sei-L-Thr-NH-BSA

I

P-DCalF(l--t3)-a-D-GalpNAc 150

GLYCOPEPT'IDE SYNTHESIS

299

The trisaccharide- hexapeptide partial structure (151) of a leukemia-virus envelope glycoprotein, which was obtained from 46 (see Section II,2) via saccharide and carboxyl deprotection, has also been coupled to BSA to furnish a syntheticglycoprotein 152 containing 12% of carbohydrate in.the form of the trisaccharide (1 9). HO NOC

I I

L-Na

CAla

L-La

I

"i'"

BSA

152

V. SOLID-PHASE SYNTHESIS

The solid-phase synthesis of glycopeptides was first realized by applying the polymeric benzyl ester principle of Merrifield. According to this methodology, Lavielle and associates (50) used N-(tert-butyloxycarbony1)-0glycosyl serine derivative 153 for condensation with resin-linked alanine 154. 1) couplingwith

Boc-HN-CH-COOH

I

H-Gly-L-Ser-L-Ala-OH

I H2N-CH I

154

CH3

AcO

2) N-depmleclion 31 colpling with Boc-glycine 4) deproleclion wilh dry HF

AcHN

155

153

The N-(tert-butyloxycarbonyl)protecting group was removed by trifluoroacetic acid. Subsequent coupling to N-(tert-butyloxycarbony1)glycineand final treatment with hydrogen fluoride gave the glycopeptide 155. In this instance, the successof the release of the glycopeptide from the resin through cleaving of the benzyl ester depends upon total exclusion of moisture from the HF. The glycosylated somatostatin 156 was synthesized similarly, starting from resin-linked cysteine (49).

L-Ala-Gly -L-Cys

- L-Lys -L-Asn-

L-Phe -L-Phe

- L-Trp-

L-Lys - L - l l r - L-Phe- L-Thr -L-Ser

I

p-GlcpNAc

156

I

-L-Cys

300

HAM G. GARG et al.

A new allylic anchoring technique for elongation at the N terminus of peptides linked to a polymer support has been developed (79,80). The anchoring allylic ester is formed by reactin of the cesium salt of the N-protected amino acid, for instance, tert-butyloxycarbonyl-L-alanine, with 2-(4bromo-2-butenamid0)methyl-polystyrene(157) (HYCRAM", trademark of Orpegen, Germany). The allylic ester linkage is stable to the acids and bases required for amino deprotection. Removal of the N-Boc group of 158 gave derivative 159. This product could be extended by condensation of the 0-glycosylserine derivative 160 to furnish 161.

Hp I

R- L - A h - 0 W

C

/

N

II

L

v

158 RIBOC

L

CF3CMH

159 R

=ti

161

wh*'"e

J

HycrarnTM

L 1 6 2

R=FmOC

R=H

I.Z-Ala-OHIoIc. H O B 1

Z-L-AI~-HN-CH-CO-L-AI~.OH

I Ez0-Y €310 BZO

163

Removal of the Fmoc group was achieved by treatment with morpholine- dichloromethane, leaving the allylic anchor and the O-glyco-

GLYCOPEPTIDE SYNTHESIS

30 1

sylserine linkage unaffected. The resulting resin-linked glycopeptide 162 was extended with Z-alanine. Finally, the glycotripeptide 163 was detached by palladium(0)-catalyzed allyl ester cleavage (32,79). Under the prevailing, practically neutral conditions, the other protecting groups and the O-glycosidic bond remained untouched (79). The allylic HYCRAM derivative was subsequentlymodified by insertion of a standard amino acid between the aminomethyl resin and the hydroxy butenoic acid moiety. Using this allylic anchor, the resin-linked, glycosylated HIV peptide T-derivative 164 was synthesized by application of Fmoc amino protection and sidechain protection with tert-butyl groups. The lactosamine peptide T (165) could be released from the resin by application of the palladium(0)-catalyzedallyl-transfer reaction to N-methyl aniline as the allyl acceptor.

164 (Ph3P),Pdo,cat. Ph-NHMe I DMSO

A&

165

h

166

0

302

HAlU G . GARG et al.

Similarly, the glycosylated peptide T derivative 166, carrying one N-glycosyldically and two 0-glycosylicallylinked saccharideside-chains was successfully synthesized (40,8 1). Paulsen and associates (55) described the synthesis of the glycopeptide 167, having a partial sequence of porcine glycophorine, on resins having acid-sensitivealkoxylbenzylester anchoring-groups. The application of this technique is based on use of the Fmoc group for temporary amino protection and its selective removal with morpholine. The N-terminal tevt-butyloxycarbonyl group and the alkoxylbenzylanchor were simultaneously cleaved from 167 with 95% trifluoroacetic acid, and subsequent base-catalyzed methanolysis gave 168.

a-0-Gal@Ac L-AIa-L-Thr-LL-Val-

I

L.Thr-L-Ala-GIy

-OH

I a-D-Gal@JAc 166

Luning and coworkers (56) also used the Fmoc group for temporary amino protection and employed a commercial resin (SASRIN", trademark of Bachem, Switzerland) having a dialkoxybenzyl anchoring group. The synthesized glycopeptide could be cleaved from the resin with 1% trifluoroacetic acid in dichloromethane. Meldal and Jensen (49) utilized pentafluorophenyl esters and performed solid-phaseglycopeptide synthesis, again based on temporary Fmoc protection of glycosyl- amino acid components. Per-0-acetylated derivatives, for instance 169, were used as building blocks, as in the synthesis of the N-glycopeptide (25) H - L- Thr - L - His- L -Ala - L - Ser - L - Asn(G1cNAc)- Gly L - Ser - L - Met - L - Ser - L - Gly - OH. Applying this technique, O-glycosylated derivatives of the insulin-like growth factor 1 (Ref. 82) and porcine submaxillary gland mucin (83) have been synthesized successfully, The method was shown to be also useful in multiple-column syntheses (83). Otvos et a/. (27) adopted the Fmoc-pentafluorophenyl ester methodology and have applied 0-deacetylated glycopeptide intermediates (170 and 171) for solid-phase synthesis of glycopeptides.In this way, the N-glycopeptide (26) H -Gly - L - Lys- L- Ala- L- Tyr- L- Thr- L- Ile - L- Phe- L -

GLYCOPEFITDE SYNTHESIS

303

..a'=" Fmoc-HN-

Fmoc-HN-CH-CDOPlp

CH--COOP@

I

I

Y

RO

NH AcHN

AcHN

AcHN 171

169 R = A c 170 R = H

Asn(G1cNAc)- L - Lys - L - Thr - L - Leu - L -Met - NH, and glycosylated derivatives of peptide T have been obtained (84). VI. ENZYMES AS TOOLS FOR GLYCOPEPTIDE SYNTHESIS

The high specificity and stereoselectivity of enzymes, as well as the mild conditions under which they react, make enzyme-catalyzedreactions versatile tools in the synthesis of glycoconjugates. In some instances, an enzymic one-step transformation affords higher yields then the conventional and more-complex chemical synthesis.The application of enzymes in glycopeptide synthesis is under active development for selective deprotection and glycosylation purposes. 1. Selective Enzymic Deprotection

In the synthesis of glycopeptidesof biological interest, the selectivedeprotection of various protected groups often requires neutral conditions. Enzymes operate at neutral, weakly acidic, or weakly basic conditions and in many instances combine these advantages with high selectivity. Some of them catalyze the processes with broad substrate specificity.The application of biocatalysts to effect the removal or introduction of protecting groups offers viable alternatives to chemical methods used (85 -87). Thus, the removal of the heptyl ester from the glycopeptide derivatives 172- 174 has been achieved, to give the carboxyl-deblocked glycopeptides 175 - 177, respectively, in high yield (46). During this process, the amino protection, the azido group, the base-labile acetyl groups, and glycosidic linkages remain totally unaffected. This procedure is both efficient and experimentally simple. The scope of enzymic protecting-group techniques has been demonstrated by the selective cleavage of glycopeptide tert-butyl esters by action of the enzyme termitase (87). 2. Glycosylation Reactions

Transfer of glycosyl residues to saccahride acceptors under catalysis by glycosyltransferases has been studied in recent years (85,88,89). In contrast to chemical glycosylation, enzymic extensions of the saccharide chain of

HARI G. GARG et al.

304 2-HN-CH-CCOHep

Fmoc-HN- CH-COOHep

I

I I

CH2

CH2

I

AcoE+ Aco#o

AcO

OAc

AcO

CH-

Ace# OAc

175

Acbr

ACU

CCOH

N,

OAC 174

Fmoc-HN-CH-COOH

I

I

AcO

I

173

172

2-HN-

2-HN-CH-Cmep

CHz

Ace@ AcO

OAc

177

176

glycopeptides occur not only under neutral conditions but without extensive protecting-group manipulations. As a rule, the glycosyltransferase-promoted glycosyltransfer reactions proceed with high regio- and stereoselectivity. Complex saccharide structures difficult to construct by chemical methods, such as sialoglycosides,become efficiently accessible by glycosyltransferase reactions (89). Paulson et al. (90) have reported the synthesis of the N-acetylneuraminyl -1actosamine-L-asparaginederivatives180 and 181. Starting from the 2-acetamido-2-deoxy-~-~-glucopyranosyl-~-aspara~ne derivatives 178 and 179, the (1 + 4-P-~-galactosyland (2 + 6)a-NeuSAc moieties, respectively, were coupled by two subsequent enzymic glycosylations in a "one-pot" procedure. In both glycosyltransferase reactions, alkaline phosphatase was used to decompose nucleotide diphosphate in order to prevent product-induced inhibition generated in the course of reaction. HO

I

R -HN-

R-HN-CH-COR

I

f". .NH

Ho&$o Ho 180 R 181 A

-

L-Ssr-L-Thr-L IleOH. R

-

AbC-L-Phe

GlyGlyOH. R = H-Glq-Glq

GLYCOPEFT'IDE SYNTHESIS

305

Enzymic synthesis of the trisaccharide asparagine conjugate a-DNeupAc-(2 + 6)-P-~-Galp-(1 + 4)-/3-~-GlcpNAc-( 1 + 4N)-~-Asn (186) has been reported by AugC and coworkers (9 1). R'-HN

-CH-COR I CH2 I

R'-HN-CH-COR

EC24190

HO HO

&o*i=o

=

HO HO

NH

+5J=O

NH

HO

OH

AcHN

I

CH,

AcHN

182 R = R . = H

184 R = R ' = H

183 R

185 R = M e . R = A c

Me. R' = AC

I

H,N-CH-COOH

I

AcHN

EC 2 498 1 ____L

HO

NH OH

AcHN

186

Two enzymic steps are involved in the synthesis, starting from the 2acetamido-2-deoxy-~-glucose - L-asparaghe derivative 182: (i) galactosylation catalyzed by immobilized N-acetyllactosaminesynthase (EC 2.4.1.90) to give 184 and (ii) sialylation catalyzed by soluble CMP-N-acetylneuraminate [P-D-galactopyranosyl-(1 4)-2-acetamido-2-deoxy-~-~-glucopyranose-a-(2 + 6) sialyltransferase (EC 2.4.99. l)] to give glycopeptide 186. Similarly Thiem and Wiemann (92) conducted the galactosylation of the 2-acetamido-2-deoxy-~-glucose - L-asparagine derivative 183 and the chitobiose- L-asparagine derivative 187 to furnish the di- and trisaccharide derivatives 185 and 188, respectively, by using galactosyltransferase.

-

AC-HN-CH-CQOCH,

I

I

no

CH2

187

n

o

~

no H

o AcHN

~

NH o

~

AcHN 188

Likewise, the galactosylation of 0-glycopeptide derivatives 189- 191 to give 0-lactosamine- L-serine derivatives (192- 194) using ( 1 * 4)-/?-~-

~

=

HARI G. GARG et al.

306

galactosylatransferasehas been accomplished (93). This reaction illustrates that both N- and O-glycopeptidescan be subjected to enzymic elongation of the saccharide chains.

&gNHA R-L-S~~-L-AI~-OIEU

I

R-L-Ser -L-Ala-OlBu

I

HO

OH

HO

OH 189 R C6H&HZCC0 190 R = CCI&H&O 191 R = C6HSCH$xX-L-Ala F

192 R = CsH&H2CC0

183 R = CCI,CH@O 194 R = C.H5CH2CC&L-Ala

Although the limitations of the enzymic reactions lie in the substrate selectivity and in the availability of enzymes and appropriate nucleotide sugar derivatives, further developments in this field will offer an interesting and efficient potential for the synthetic construction of glycopeptides of biological interest. VII. ADDENDUM

A number of publications on glycopeptidechemistry have appeared since this manuscript was submitted. Cohen-Anisfeldand Lansbury (94) reported on a convergent synthesisofglycopeptides. Peptidescontaining aspartic acid unblocked at the P-carboxy group were synthesized and, subsequently,condensed with glycosylamines. The optimized solid-phase synthesis of glycopeptides using Fmoc protected glycosyl amino acids activated as pentafluorophenyl esters was demonstrated by Meldal et al. in a number of examples including glycopeptide .T-cell epitopes (93, mucin glycopeptides(96), mannose-6-phosphate containing glycopeptides(97), N-glycopeptides (98), and O-glycopeptide structures of a blood clotting factor (99). Lewis” antigen glycopeptides carrying clustered saccharide side chains were synthesized according to the Boc/OAll strategy (1 00). After complete deprotection these glycopeptides were coupled to BSA and KLH (keyhole limpet hemocyanine) to give neoglycoproteins.A new glycosylationmethod consisting in the electrophile-induced activation of ally1 carbamates was applied to the synthesis of O-glycopeptides (10 1). The formation of glycopeptides was achieved by a direct glycosylation of pre-formed peptides ( 102). Lipase-catalyzed cleavage of glycopeptide n-heptyl esters was successfully used in syntheses of multiple glycosylated O-glycopeptides (103). Surpris-

GLYCOPEFTIDE SYNTHESIS

307

ingly, polar 2-(N-morpholino)ethyl esters of glycopeptides that enhance the water-solubility revealed them to be efficientlyhydrolyzed by lipases under neutral conditions (104). An enzymatic peptide bond formation catalyzed by a gene-technologically produced thiosubtilisin was used for the synthesis of glycopeptide amides (105). Oligosacharyltransferases from yeast were applied to form the N-glycosidic bonds between asparagine peptides and oligosaccharides (106,107). The saccharide portions were first transformed to their dolichyl diphosphates. Saccaride chain extension was camed out on glucosamine 0-glycopeptides by means of a galactosyltransferase-catalyzed reaction (108).

REFERENCES 1. M. Bergmann andL. Zervas, Chem. Ber., 65 (1932) 1201-1205.

D. G. Doherty, E. A. Popenoe, and K. P. Link, J. Am. Chem. Soc., 75 (1953) 3466-3468. P. G. Johansen, R. D. Marshall, and A. Neuberger, Biochem. J., 78 (1961) 518-527. G. S. Marks and A. Neuberger, J. Chem. SOC. (1961) 4872-4879. R. D. Marshall and A. Neuberger,Adv. Curbohydr.Chem. Biochem., 25 (1970)407-478. H. G. Gargand R. W. Jeanloz, Adv. Curbohydr. Chem. Biochem.. 43 (1985) 135-201. H. Kunz, Angew. Chem. Int. Ed. Engl., 26 (1981) 294-308. H. Paulsen, Angew. Chem. Int. Ed. Engl., 29 (1990) 823-839. H. Waldmann, Nuchr. Chem. Tech. Lab., 39 (1991) 615-682. H. Kunz and W. Brill, Trends Glycosci. and Glycotech. (Jupun),4 ( 1992) 7 1 - 82. H. Kunz, W. Brill, and K. von dem Bruch, in P. Pr&e (Eds.), Juhrbuch Biotechnologie, Vol. 4, pp. 103- 123. Hanser Verlag, Munich, 1992. 12. S. T. Anisfeld and P. Lansbury, Jr., J. Org. Chem., 55 (1 190) 5560-5562. 13. M. Spinola and R. W. Jeanloz, J. Biol. Chem., 245 (1970) 4158-4162. 14. H. Kunz and H. Waldmann, Angew. Chem. Znt. Ed. Engl., 24 (1985) 883-885. 15. H. Kunz, H. Waldmann, and J. Man, Liebigs Ann. Chem., (1989) 45 -49. 16. H. Waldmann and H. Kunz, Liebigs Ann. Chem. (1983) 1712- 1725. 17. B. Belleau and G. Malek, J. Am. Chem. Soc., 90 (1968) 1651 - 1652. 18. SNakabayashi, C. D. Warren,andR. W. Jeanloz, Curbohydr.Res., 174(1988)219-289. 19a. H. Kunz and C. Unverzagt, Angew. Chem. Int. Ed. Engl., 27 (1988) 1697- 1699. 19b. C. Unverzagt and H. Kunz, J. prukt. Chem., 334 (1992) 570-578. 19c. H. Kunz and C. Unverzagt, J. prukt. Chem., 334 (1992) 579-583. 19d. C. Unvenagt and H. Kunz, Bioorg. Med. Chem. 1994, in press. 20. R. U. Lemieux, K. B. Hendriks, R. V. Stick, K. James, J. Am. Chem. Soc., 91 (1975) 4056 -4062. 21. J. MIrz and H. Kunz, Synlett (1992) 589-590. 22. A. Y. Khorlin, S. E. Zurabyan, and R. G. Macharadze, Curbohydr. Rex, 85 (1980) 201-208. 23. W. Gunther and H. Kunz, Angew. Chem. Znt. Ed. Engl., 29 (199) 1050- 1051. 24. L. M. Likhoshersov, 0. S. Novikova, V. A. Dervitskaya, and N. K. Kochetkov, Curbohydr. Res., 146 (1986)Cl-C5. 25. M. Meldal and K. Bock, Tetrahedron Lett., 31 (1990) 6897-6990. 26. L. Otvos, Jr., L. Urge, M. Hollosi, K. Wroblewski, G. Graczyk, G. D. Frasman, and J. Thurin, Tetrahedron Lett., 31 (1990) 5889-5892. 2. 3. 4. 5. 6. 7. 8. 9. 10. 1 1.

308

HARI G. GARG et al.

27. L. Urge, E. Kollat, M. Hollosi, I. Lackzo, K. Wroblewski, J. Thurin, and L. Otvos, Jr., TetruhedronLett., 32 (1991) 3445-3448. 28. H. G. Garg and R. W. Jeanloz, Carbohydr. Res., 32 (1974) 37-46. 29. H. G. Garg and R. W. Jeanloz, J. Org. Chem., 41 (1976) 2480-2484. 30. H. G. Garg and R. W. Jeanloz, Carohydr. Res., 86 (1980) 69-68. 31. M.A. E. Shaban and R. W. Jeanloz, Carbohydr. Res.. 43 (1975) 281 -291. 32. H. Kunz and H. Waldmann, Agnew. Chem. Int. Ed. Engl., 23 (1984) 71 -72. 33. H. Waldmann, J. M h , and H. Kunz, Carbohydr. Res., 196 (1990) 75-93. 34. H. Kunz and C. Unverzagt, Angew. Chem. Int. Ed. Engl., 23 (1984) 436-437. 35. H. Kunz and H. Waldmann, J. Chem. SOC.Chem. Commun. (1985) 638-640. 36. H. Kunz and H. Waldmann, and C. Unvenagt, Int. J. Peptide Protein Res., 26 (1 985) 493 -497. 37. H. Waldmann and H. Kunz, J. Org. Chem., 53 (1988) 4172-4175. 38. T. B. Windholz and D. B. R. Johnston, Tetrahedron Lett (1967) 2555-2559. 39. H. Kunz and J. M&, Synlett (1992) 591-593. 40. H. Kunz, W. Kosch, and J. Man, in J. A. Smith and J. Rivier (Eds.), Peptides: Chemistry and Biology, pp. 502 - 504. ESCOM, Leiden, The Netherlands, 1992. 41. J. K. N. Jones, M.B. Perry, B. Shelton, and D. T. Walton, Can. J. Chem. 39 (1961) 1005- 1016. 42a. V. Bencomo and P. Sinay, Carbohydr. Res., 116 (1983) C9-CI2. 42b. V. Bencomo and P. Shy, GlycoconjugateJ., 1 (1984) 5-8. 43. P. Schultheiss-Reimannand H. Kunz, Angew. Chem. Znt. Ed. Engl., 22 (1983) 62-63. 44. H. Kunz and S. Birnbach, Angew. Chem. Int. Ed. Engl., 25 (1986) 360-362. 45. M.Buchholz and H. Kunz, Liebigs Ann. Chem. (1983) 1859- 1885. 45a. H. Kunz and M. Buchholz, Angew. Chem. Int. Ed. Engl., 21 (1981), 894-895. 46. P. Braun, H. Waldmann, and H. Kunz, Synlett (1992) 39-40. 47. B. Ferrari and A. A. Pavia, Int. J. Peptide Protein Rex, 22 (1983) 549-559. 48. J. M. Lacombe and A. A. Pavia, J. Org. Chem., 48 (1983) 2557-2563. 49. M. Meldal and K. J. Jensen, J. Chem. SOC.Chem. Commun. (1990) 483-485. 50. S. Lavielle, N. C. Ling, R. Saltman, and R. Guilleman, Curbohydr. Res., 89 (1981) 229 -236. 51. H. Kunz and R. Barthels, Angew. Chem. Int. Ed. Engl., 22 (1983) 783-784. 52. H. Kunz and S. Birnbach, Tetrahedron Lett.. 25 (1984) 3567-3570. 53. H. Kunz and H. Kauth, Angew. Chem. Int. Ed. Engl., 20 (1981) 895-896. 54. P. Schultheiss-Reimannand H. Kunz, Angew. Chem. Suppl. (1982) 39-46. 5 5 . H. Paulsen, G. Merz, and U. Weichert, Angew. Chem. Int. Ed. Engl., 27 (1988) 13651367. 56. B. Luning, T. Norberg, and J. Tejbrant, J. Chem. SOC.Chem. Commun. (1989) 12671268. 57. A. Helander, L. Kenne, S. Oscarson, T. Peters, and J.-R. Brisson, Carbohydr. Rex, 230 (1992) 299-318. 58. H. Paulsen, M. Schultz, J.-D. Klamann, B. Walter, and M.Paal, Liebigs Ann. Chem. (1985) 2028-2048. 59. H. Paulsen and M.Paal, Carbohydr. Rex, 135 (1984) 53-69. 60. R. R. Schmidt and J. Michel, Angew. Chem. Int. Ed. Engl., 19 (1980) 731-732; R. R. Schmidt and W. Kinzy, Adv. Curbohydr. Chem. Biochem., 50 (1994). 61. H. Ijima and T. Ogawa, Carbohydr. Res., 186 (1989) 107-118. 62. H. Kunz and W. Sager, Helv. Chim. Acta, 68 (1985) 283-285. 63a. H. Lijnn, Curbohydr. Rex, 139 (1985) 115- 121. 63b. P. Fugedi and P. J. Garegg, Carbohydr. Res., 149 (1986) C9-CI2.

GLYCOPEPTIDE SYNTHESIS

309

H. Paulsen, W. Rauwald, and U. Weichert, Liebigs Ann. Chem. (1988) 75-86. T. Rosen, I. M. Lico, andD. T. W. Chu,J. Org. Chem., 53 (1986) 1580-1582. N. Nakahara, H. Ijima, S.Siba, and T. Ogawa, TetrahedronLett., 31 (1990) 6897-6900. M. Ciommer and H. Kunz, Synlett (1991) 593-595. S. Friedrich-Bochnitschek,H. Waldmann, and H. Kunz, J. Org. Chem., 54 (1989) 75 1 756. 69. H. Paulsen and M. Brenken, Liebigs Ann. Chem. (1988) 649-654. 70. H. G. Garg, T. Hasenkamp, and H. Paulsen, Carbohydr. Res., 151 (1986) 225-232. 71. H. Kunz, P. Wernig, and M. Schultz, Synlett (1990) 631 -632. 72. S. Rio, J.-M. Beau, and J.-C. Jacquinet, Carbohydr. Rex, 219 (1991) 71-90. 73. H. Kunz, S. Birnbach, and P. Wernig, Carbohydr. Res., 202 (1990) 207-223. 74. H. Paulsen and M. Schultz, LiebigAnn. Chem. (1986) 1435-1447. 75. B. Ferrari and A. A. Pavia, Tetrahedron, 41 (1985) 1939- 1944. 76. H. Paulsen, G. Men, and I. Brockhausen,Liebigs Ann. Chem. (1990) 7 19-739 and 1272. 77. K. Paulsen and K. Aderman, Liebigs. Ann. Chem. (1989) 751 -769. 78. K. Paulsen and K. Adermann, Liebigs Ann. Chem. (1989) 771 -780. 79. H. Kunz, B. Dombo, and W. Koxh, Proc. Europ. Peptide Symp., 20 ( I 988) 154 - 156. 80. H. Kunz and B. Dombo, Angew. Chem. Int. Ed. Engl., 27 (1988) 71 1-713. 81. H. Kunz, W. Kosch, J. M h , M. Ciommer, W. Gtinther, and C. Unverzagt, in R. Epton (Ed.), Innovations and Perspectives in Solid Phase Synthesis, pp. 171- 178. Intercept, Andover, U. K., 1992. 82. A.M. Janson, M. Meldal, andK. Bock, J. Chem. SOC.Perk. Trans. Z(1992) 1699-1707. 83. S. Peters, T. Bielefeldt,M. Meldal, K. Bock, and H. Paulsen, J. Chem. SOC.Perkin TransZ (1992) 1163- 1171. 84a. I. Laczko, M. Hollosi, L. Urge, K. E. Ugen, D. B. Weiner, H. H. Mautsch, J. Thurin, and L. Otvos, Jr., Biochemistry, 21 (1990) 4282-4288. 84b. L. Urge, L. Gorbies, and L. Otvos, Jr., Biochem. Biophys. Res. Comm., 184 (1992) 1125- 1132. 85. H. Waldmann, Nachr. Chem. Tech. Lab., 40 (1992) 828-834. 86. P. Braun, H. Waldmann, W. Vogt,and H. Kunz, Synlett (1990) 105- 107. 87. M. Schultz, P. Hermann, and H. Kunz, Synlett (1992) 37-38. 88. E. J. Toone, E. S.Simon, M. D. Bednarski, and G. M. Whitesides, Tetrahedron,47 (1989) 5365-5422. 89. S. Sabesan, and J. C. Paulson, J. Am. Chem. SOC.,108 (1986) 2068-2080. 90. C. Unverzagt, H. Kunz, and J. C. Paulson, J. Am. Chem. Soc., 112 (1990) 9308-9309. 91. C. Augk, C. Gautheron, and H. Pora, Carbohydr. Res., 193 (1989) 288-293. 92. J. Thiem and T. Wiemann, Angew. Chem. Znt. Ed. Engl., 29 (1990) 80-82. 93. M. Schultz and H. Kunz, Tetrahedron Lett., 33 (1992) 5319-5322. 94. S.T.Cohen-AnisfeldandP.T.LansburyJr.,J.Am. Chem. SOC.I15(1993) 10531-10537. 95. S.Peters, T. Bielefeldt,M. Meldal, K. Bock, and H. Paulsen, Tetrahedron Lett. 33 (1992) 4126-4135. 96. M. Meldal, S. Mountsen, and K. Bock, Am. Chem. SOC.,Symp. ser. (1993) pp. 19-33. 97. M. Christensen, M. Meldal, and K. Bock, J. Chem. SOC.Perkin Trans I (1993) 14531460. 98. I. Christiansen-Brams, M. Meldal, and K. Bock, J. Chem. SOC.Perkin Trans I (1993) 146 1 - 147I. 99. K. B. Reimer, M. Meldal, S. Kusumoto, K. Fukase and K. Bock, J. Chem. SOC.Perkin Trans l(1993) 925-931. 100. K. von dem Bruch and H. Kunz, Angew. Chem. Int. Ed. Engl. 33 (1994) 101- 103. 101. H. Kunz and J. Zimmer, Tetrahedron Lett. 34 (1993) 2907-2910. 64. 65. 66. 67. 68.

310

HARI G. GARG et al.

102. C. D. Warren and H. G. Garg, Glycobiology 3 (1993) 540. 103. P. Braun, H. Waldmann, and H. Kunz, Bioorg. Med. Chem. 1 (1993) 197-207. 104. G. Braum, P. Braun, D. Kowalczyk, and H. Kunz, Tetrahedron Lett. 34 (1993) 31 1131 14. 105. C.-H. Won& M. Schuster, P. Wan& and P. Sears, J. Am. Chem. Soc. 115 (1993) 58935901. 106, J. Lee and J. K. Coward, J. Org. Chem. 57 (2992) 4126-4135. 107. J. Lee and J. K. Coward, Biochem. 32 (1993) 67694-6801. 108. M. Schultz and H. Kunz, TetrahedronAsymmetry 4 (1993) 1205- 1220.

ADVANCES IN CARBOHYDRATE CHEMISTRY AND BIOCHEMISTRY, VOL. 50

PHYSICOCHEMICAL ANALYSES OF OLIGOSACCHARIDE DETERMINANTS OF GLYCOPROTEINS BY ELIZABETH F. HOUNSELL MRC Glycoprotein Structure/Function Group, Department of Biochemistry and Molecular Biology, University College London, London, WClE 6BT, England

I. INTRODUCTION A large number of oligosaccharide structures of glycoproteins have now been unequivocally characterized by using a combination of chromatography, mass spectrometry (m.s.), and nuclear magnetic resonance (n.m.r.) spectroscopy. Information from these physicochemical techniques, together with that obtained from circular dichroism and X-ray diffraction of monoand polysaccharides [previously discussed in this series ( 1,2)], is providing data for incorporation into computer-graphics molecular models of oligosaccharides and glycopeptides in order to visualise oligosaccharide sequences important in molecular recognition. This chapter surveys the diversity of structures and shapes of oligosaccharide determinants of glycoproteins that are antigens and targets for binding of adhesion molecules. 11. METHODS OF STRUCTURAL ANALYSIS

Four major techniques (summarized in Table I) provide the basis of our current structural-analysis stratagem for oligosaccharides of glycoconjugates. Historically, the first of these was methylation analysis. The desire to delineate the linkagesthrough which monosaccharidesare coupled led to the idea of labeling the free hydroxyl groups by permethylation.The low molecular weight of the methyl group is an important factor in subsequent massspectroscopic analysis, and the hydrophobicity imparted to the oligosaccharide permits efficient purification, increased chromatographic resolution, and good m.s. ionization properties. The method of permethylation using dimethylsulfinylanion introduced by Hakomori (43) for complete derivatization of quite large oligosaccharidesbecame universally adopted (6 - 1I). A 31 1

Copyright 8 1994 by Academic Press, Inc. All rights of reproduction in any form reserved.

ELIZABETH F. HOUNSELL

312

TABLEI A General Stratagem for Oligosaccharide PhysicochemicalAnalysis Method applied

Methylation analysis

Chromatographic analysis

F.a.bm.s. or 1.s.i.-m.s

N.m.r. spectroscopy

Sequence of applied techniques

Permethylation of intact oligosaccharides Hydrolysis of glycosidic linkages Reduction to monosaccharide alditols Acetylation of free. hydroxyl groups previously involved in linkages G.1.c. separation of substituent, partially methylated alditol acetates On-line e.i.- or c.i.-m.s. analysis G.1.c. analysis after sequential hydrolysis, reduction, and acetylation G.1.c. analysis after methanolysis and trimethylsilylation H.p.a.e. chromatography after hydrolysis H.p.1.c. -h.p.a.e. chromatographic profiling of reducing oligosaccharides, alditols, or organic amine derivatives Native oligosaccharidesplaced on the m.s. probe-tip with a matrix Bombardment with atoms (f.a.b.-m.s.) or ions (1.s.i.m.s.) MA. analysis in negative-ion mode Permethylated or acetylated oligosaccharides analyzed in positive-ion mode Reducing oligosaccharidesor periodate-oxidized alditols derivatized by reductive amination and analyzed in positive-ion mode T.1.c.-m.s. analysis of oligosaccharides coupled to a lipid amine (neoglycolipids) IH n.m.r. spectrum in D20after exchange of free protons with deuterium Experiments conducted at 295 K, with acetone as the internal standard (set at 2.225 p.p.m. from 4,4-dimethyl-4-silapentane1-sulfonate) Results compared, to within +0.005 p.p.m. (laboratory-to-laboratory variation) of data in the literature Conformational studies by n.0.e. experiments Nat~ral-abundance-l~C analysis Chemical-shift assignment by 2D IH-IH and lH-l3C n.m.r. spectroscopy

Note. Abbreviations: g.l.c., gas-liquid chromatography; e.i.-m.s., electron-impact mass spectrometry; c.i.-m.s., chemical-ionizationmass spectrometry;h.p.l.c., high-performanceliquid chromatography;h.p.a.e., high pH anionexchange;f.a.b.-m.s., fast-atom-bombardment mass spectrometry; 1.s.i.-m.s.,liquid secondary-ion mass spectrometry; n.O.e., nuclear Overhauser enhancement. Details of these methodologies are given in Ref. (3).

OLIGOSACCHARIDE DETERMINANTS OF GLYCOPROTEINS

313

more facile method (12) involving the use of sodium hydroxide in dimethyl sulfoxide, based on the original Kuhn (13) procedure, is now largely used. The combination of hydrolysis, reduction, acetylation, and g.1.c. (now capillary g.1.c.) remains the best method for analysisof the large number of closely related, partially methylated monosaccharide derivatives formed, although separation by high-performance liquid chromatography (h.p.1.c.) has also been investigated ( 14). Electron-impact mass spectrometry (e.i.-m.s.) is the most generally applied robust technique for detection and characterization, although chemical-ionization(c.i.)-ms., in particular, of trifluoroacetylated, methylated alditols gives increased sensitivityand a greater abundance of the molecular ion. Gas-liquid chromatography (g.1.c.)-m.s. analysis of intact permethylated oligosaccharides ( 15) can also be achieved. As is general in biochemistry, improvements in chromatographic techniques have been a major factor in research advances. This is particularly true in the carbohydrate field, in which separation of the large number of molecules that are structurally closely related is a major challenge. Thinlayer chromatography (t.1.c.) and paper, gas-liquid, high-performance liquid, and supercriticalfluid ( 16)chromatographies,together with electrophoresis (1 7,18), have their application in oligosaccharide analysis and purification. The introduction of g.1.c. for oligosaccharide analysis constituted a major breakthrough in the field. In addition to strategies for accurate and sensitive quantitation of monosaccharide type (19,20), chiral procedures may be adopted for enantiomeric (D and L) determination (21). The sensitivity of h.p.a.e. chromatographywith pulsed amperometric detection now provides an alternative to g.1.c. for oligosaccharide compositional analysis (22). The use of m.s. in oligosaccharide analysis was advanced greatly by the advent of high-field magnets and the technique of f.a.b. -m.s. (23,24). This enabled analysis of larger oligosaccharides in their intact state, allowing for sequence determination at the nanomolar level. Analysis of native oligosaccharides in this way, and by the similar technique of 1.s.i. -m.s., gives data on acyl substitution of hydroxyl groups, substituents which may be lost during derivatization procedures. Derivatization (for example, acetylation, methylation, and reductive amination) gives further sensitivity and sequence information (23- 26). F.a.b.-m.s. of permethylated periodate-oxidized (27) oligosaccharidesand t.1.c.-1.s.i.-m.s. of periodate-oxidizedalditols coupled to lipidamines (28) provide additional strategies for gaining sequence and linkage information. Coupling to lipidamines (neoglycolipids) allows for in situ analysis of oligosaccharides on t.1.c. by m.s. and overlay with carbohydrate binding proteins in studies of receptor and antibody specificities,as introduced by Feizi and colleagues (29).

3 14

ELIZABETH F. HOUNSELL

The application of proton ('H) n.m.r. spectroscopy to oligosaccharide characterization has rapidly increased since the early days when low-field instruments were used in the characterization of synthetic products and first applied to complex oligosaccharides of glycoproteins (30 - 32); high-field n.m.r. is now a relatively routine tool for providing a first-line profile of a purified oligosaccharide (32,33). The application has been diversified into providing conformational information on oligosaccharides and glycopeptides in conjunction with 2D methods. Now available are large databanks of chemical shifts that can be readily searched to give structural intrepretation (32 - 38). The purpose of the present article is not to supply a large volume of data, but rather to categorize the structures found in mammalian glycoproteins by using 'H n.m.r. spectroscopyand mass spectrometry and to discuss the conformational information obtained. AND PROFILING 111. PURIFICATION

In addition to studies uniquely on oligosaccharidestructure, the carbohydrate moiety of glycoproteins is now being viewed in the context of the protein skeleton, in order both to characterize the complete glycoprotein molecule and to elucidate the biosynthetic controls that determine site-specific glycosylation patterns. The modern methods of purification and profiling reflect this philosophy. For example, commercial kits are available for the sensitive detection of glycosylated proteins; glycoprotein-specificdegradative enzymes are being purified and cloned; h.p.l.c., n.m.r. spectroscopy, and circular dichroism (c.d.) are being used for glycopeptide analysis; and glycoproteins produced by recombinant DNA technology are increasingly being screened for variations in glycosylation patterns (39 -42). Profiling methods originally involved the analysis of N-linked glycosylation by hydrazinolysis, followed by size-exclusionon columns of Biogel P4 before and after specific cleavage by exoglycosidases (43). Release of oligosaccharide chains by endoglycosidases (44) and h.p.1.c. analysis (reviewed in Ref. 45) have now been added to the stratagem (see Scheme 1). Both chromatography on Biogel P4 and h.p.1.c. have been perfected for purification of oligosaccharides prior to n.m.r.-spectral analysis and biological studies. Here, the h.p.a.e. technique is not generally applicable,owingto the presence of a high salt content and the lower loading and yield compared with those encountered on h.p.1.c. and Biogel P4 chromatography in volatile buffers. Initial separation by size exclusion is advantageous, because retention behavior in h.p.1.c. is affected by both composition and linkage, and, therefore, it is much easier to separate each size subset first and then to purify isomers by h.p.1.c. Reverse phase using octadecylsilyl (0.d.s.) or porous graphitized carbon (p.g.c.) columns, normal phase using amine-bonded silica (for example, a.p.s.), and anion-exchange (a.e.) chromatographies in volatile

OLIGOSACCHARIDE DETERMINANTS OF GLYCOPROTEINS

315

Glywprotein

1

Protiase digestion

Mild hydrolysis with acid for sialic acid determination

4 1.s.i.rn.s.

O(carboxymethy1)ation and protease digestion

4

h.p.1.c.

t

t

+

Peptides

GIywpeptides

e-1

Mild a

~

1.s.i.m.s. Endo H.peptide or N-terminal N-glycanase F and analysis mild alkali treatment Peptides

t

a

l

~

d

~

o

~

I

1

Oligosaccharides

~

Reductive amination and h.p.1.c. ort.1.c.-m.s. (of neoglycolipids)

h.p.a.e. chromatography

Methylation

e-7

1.s.i.rn.s. Methylation analysis

SCHEME 1. -A Flow Chart Outliningthe Methodologies for Profiling Protein Glycosylation Patterns.

buffers-organic phase are very reliable and efficient methods (reviewed in Ref. 46). It is important to integrate more than one method of purification and analysis for the unambiguous characterization of mammalian glycoconjugates. Detailed studies of glycoproteins frequently reveal new structures, in either monosaccharide type, acyl substitution,or linkage mode. In addition, the variation in glycosylation pattern from proteins translated in different cell types can be quite subtle and may be readily missed on bulk analysis. In order to alert the molecular biologist to the wide diversity of posttranslational changes available in the repertoire employed by nature, it is therefore necessary to resort to detailed studies, ideally using a combination of m.s. and n.m.r. spectroscopy and using high-resolution chromatographic techniques. IV. BACKBONES AND COREREGIONS OF N- AND 0-LINKEDCHAINS OF SECRETED A N D PLASMA MEMBRANE GLYCOPROTEINS In a series of elegant studies in the 1980s, the groups of Montreuil(30), Kobata (47), and Vliegenthart (3 1,32)began to explore the structural diver-

316

ELIZABETH F. HOUNSELL

sity of N-linked glycosylation of serum glycoproteins. This followed the extensive work of Morgan (48), Watkins (49), and &bat (50) in characterizing the blood-groupantigens and mucin structure.It became obvious that, in many instances, oligosaccharidesequences forming the periphery and backbone could be present on different protein and lipid core-regions, as described in reviews by Feizi ( 5 1) and Hakomori (52). From their studies, the concept of oligosaccharides as oncodevelopmental antigens began to be appreciated, that is, the concept that oligosaccharide structures change during development and may be reexpressed as tumor antigens in the adult. These antigens are found on glycolipids,on 0-linked chains of mucins, and on the 0-and N-linked chains of serum and cell-membrane glycoproteins. Mucins have provided a large-scale source of oligosaccharide sequences, important not only as oncodevelopmental antigens, but also as representative structures present in the peripheral and backbone regions (53) of Nlinked chains of serum glycoproteins, which are generally of more restricted availability. Several series of detailed studies of the multiple oligosaccharide chains of these specieshaving very high molecularweight have been initiated for the foregoing reasons and also because of the desire to correlate their oligosaccharide structure with (i) the known physicochemical properties important in maintaining an effective viscoelastic bamer, (ii) their role as a bacterial interface, (iii) regulation of biosynthesis, and (iv)functional molecular recognition. 1. Mucin Oligosaccharide Structure

The majority of the initial studies of mucin structure were carried out on animal gastrointestinal mucins and ovarian-cyst blood-group substances of humans. These gave a good grounding to the diversity to be expected, as described in two reviews (5334).High-field 'H-n.m.r. (270-600 MHz) and m.s. studies then began to dissect this diversity. Oligosaccharides from mucins of ovarian-cyst fluids(55 - 58) were among the first to be investigated by these new techniques, following earlier work on this material using chemical and immunochemicalmethods (50). A map of ovarian mucin structure (shown opposite)from these early studies (59) stands today as a tribute to the elegance of the early techniques and the accuracy of the understanding of oligosaccharide structure, variation, and antigenicity. In the 1980s, several structural studies of human mucin were initiated, their overall aim being to describe the oligosaccharide sequences of bronchial mucins (60-68), adult gastric (69,70)and colonic (7 1,72)mucins, and fetal gastrointestinal mucins (73- 76). The first study sought to understand the changes in oligosaccharide structure in bronchial mucins important in cystic fibrosis and bronchiectesis, in addition to the role of bacterial growth in defining much status. The second set of studies set out to establish oligosaccharide patterns in the normal and neoplastic adult mucosae. The

OLIGOSACCHARIDE DETERMINANTS OF GLYCOPROTEINS

317

New gene interaction product [now known as Y or Ley] A or B gene products

H gene

A a-L-FUC

-

a-DGalNAc-( I or a-D-Gal-(1

New gene [now known as X, SSEA-I or L e x ] h cu-L-Fuc

D-D-Gal-( 1 + 4)-p-~GlcNAc

1

1

1

.1

1

1

3)2 &D-GaI-(I 3)-

-

4 ~)-P-D-GIcNAc 1

6 a-DGalNAc 3

1

t

6 b-D-Gal-( 1 3

+

3)-/%~-GlcNAc-( 1

-

1

3)-p-~-Gal

t

a-DGalNAc-( 1 + 3)1 p-D-Gal-( 1 -,~)-/%D-GIcNAc or a-D-Gal-(1 + 3)2 3

-t

t

H gene

Leagene

I CU-L-FUC a-L-FUC 1

Lebgene interaction product

last study was begun in order to identify novel oligosaccharide sequences present in the bacteria-free environment of the fetal gut that constitute mucin oligosaccharides synthesized in early development and that might reoccur in the adult as tumor antigens (oncodevelopmentalantigens). From the extensive information on mucin structure obtained from these studies, and several additional studies on adult mucins (77 - 8 I), it is pertinent to point out the differences found. Table I1 displays the various backbone and core regions. Although it cannot be concluded that structures found in one study are not present as minor components in another, the relative amounts detected in these systematic studies demonstrated differences in each of the particular biological scenarios. The results of these studies underscore several salient points, as follows. (i) Specific carbohydrate tumor-antigens are unlikely to occur (5 I), and, therefore, quantitative differenceswill have to be exploited for any future tumor diagnosis or therapy. (ii) Biosyntheticcontrol is often reflected in the amount of an enzyme produced rather than its total absence from the gene repertoire for a particular cell. (iii) Homogeneous cell-population studies will be required in order to dissect further the diversity of oligosaccharide patterns. (iv) Outside factors will affect endogenous structure, for example, the influence of bacterial degradative enzymes. One additional aspect of studies on the core and backbone variation in

TABLE I1 The Backbone and Core Sequences Found in H u m Mucins. Core type” CY-G~~NAC &Gal( 1 -D 3)a-GalNAc [B-Gal-(1 -D 3/4)wlcNAc-( 1 -D 3)-],,, ,,-&Gal( 1 + 3)e-GalNAc

Yield from meconiume(%)

I I

13 10 4

I

-

Ib

0.5

B-Gal-(1 -D 4)-p-GlcNAc

i

6 B-Gal-( 1 + 4)-wlcNAc-( 1 -D 3)-wal( 1 -D 3)a-GalNA~ 3

T

1

W e

00

&Gal-(1 4)-wlcNAc Adult gastric mucins only +

&Gal-(1 + 4)-&GlcNAc

!

.1

6 &Gal-(1 -+3)e-GalNAc 3

t

1 &Gal-( 1 44)-&GlcNAc &Gal-(1 -D 4)-&GlcNAc 1

1

6 wf-( 1 3)a-GalNA~ +

t

1 /%Gal-(I + 3)-fl-GlcNAc Ovarian cyst only

Ib

2 /?-Gal-(1 44)-&GlcNAc

23

!

6 a-GalNAc 3

W-Gal-( 1 43/4)-&GlcNAc-( 1

-

t

1 &Gal 3)-]Gal-(1 + 4)-&GlcNAc

i

6 a-GalNAc 3

I1

t

1

W-Gal-( 1 4 3/4)-b-GlcNA0( 1 + 3)]-/?-Gal [Gal-(1 44)-/?-GlcNAc-(1 46)],-b-Gal-( 1 4 4)-&GlcNAc

2 (3 :1 ratio of branch extension on the lower branch)

L

Human skimmed milk only

6 a-GalNAc 3

I1

t

1

&Gal P-Gal-( 1 44)-&GlcNAc 1

1

6 a-GalNAc

s

IIb

i

Gal-( 1 + 4)-/?-GlcNAc-(1

6)-/?-Gal 3

T

1

F a l l - (1 4 4)-&GlcNAc] Linear GlcNAc1-6 sequence in meconium only _+

1

Continued

TABLE I1 (continued)

Core type,

Yield from meconium' (TO)

I1

-

&Gal-( 1 + 4)-&GlcNAc 1

/3-Gal-( 1 44)+GlcNAc

3

i

/3-G:-(

6

T

1

1 44)-/3-GlcNAc-(1 + 3)-&Gal

t

W h)

1

0

4)-WlcNAc Adult gastric mucin only

pGal-(I

4

&GlCNAC-( 1 + 3M-GalNAc

111

&Gal-( 1 4 3/4)-/3-GlCNAc-(1 43)-/%Gal-(1 43)-&GkNAo( 1 + 3)e-GalNAc Human colonic much is reported to have a &Gal-( 1 + 3)-j?-GlcNAc sequence linked at 0-6 of internal Gal, giving a branched structure

I11

Wal-(1 -+ 4)-WlcNAc

i

6 a-GalNAc 3

t

1 &GlCNAC Bronchial mucins only

IV

10

0.2

B-Gal-( 1 + ~)-B-GICNAC

i ?

6 a-GalNAc

/%Gal-(1

-

-

wid-(

1 4)-wlcNAc-( 1 + 6W-GalNAc Meconium and Smith degraded ovarian cyst only +

6)-~t-GalNAc

7

(2:1ratioof1-+4 and 1 + 3 on lower branch)

1 3/4)-/?-GlcNAc

1 -+ 3)-c~-GalNAc cu-GalNA~-(

a-GalNAc-(1

IV

V

4

VI

1

VII

Identified in bovine submaxillary mucin only,not in human &Gal-( -* 4)-PGlcNAc-(1 3)-,!?-Gal -+

1

5-

GlcNAc

!

6 GalNAc

2

4 I 6 1 /%Gal-(1 -* 4)-&GlcNAc-(1 -,3)-&Gal

?

1 4)-/?-GlcNAc Also,related structures reported in adult gastric much

m-( 1

+

a From references (60-8 1); a-GalNAc-(l+ 6)-GalNAc is included for completeness, although so far this has been found only in bovine submaxillary mucin (82). All monosaccharideshave the ~- ~~u l i g u r at iwhen o n ; the structure.has been identified in only one study, the source is identified. b Coreshave been named historically (53,83),but appear in the followingsemilogicalorder Gal at 0-3in the absence or presence ofGlcNAc at 0-6; GlcNAc at 0-3 in the absence or presence ofGlcNAc at 0-6;W A Cat 0-3;GlcNAc at 0-6;GalNAcat 0-6;Gal at 0-6 in the presence of Gal at 0-3. c The percentage yields for the various oligosacchatides found in the systematic study of the backbone and core moieties of fetal, gastrointestinal-tractmucins found in blocd-group H-active, meconium glycoproteins depleted of Ii antigenic activity (5% of the hexose retained on an anti-I immunoaflinitycolumn) (73- 75)are given in parenthem. The remaining 2 1% oligosaccharidesnot accounted for are a heterogeneous mixture larger than heptasaccharide..

322

ELIZABETH F. HOUNSELL

mucins is our ignorance of which factors in the protein skeleton that are distant from the glycosylation site influence biosynthesis and whether the diversity of the carbohydrate structures of mucins will also be found in the 0-linked chains of serum and cell-membrane glycoproteins. Originally, it was thought that 0-linked chains of these nonmucin and, by comparison, relatively low-molecular-weight glycoproteins were only of the following class: fa-D-NeuAc 2

1

6 a-~-GalNAc-(1

-

3

-

)Ser/Thr

1

fa-~-NeuAc-(2 ~ ) - D - D - G ~

Additional sequences have now been found, for example, those shown in Table 111. Early studies (87) of the T and B cell-membrane antigen T200 (now called CD45) also suggested the presence of 0-linked oligosaccharides larger than tetrasaccharides and provided evidence that cell-membraneglycoproteins can have clustered 0-glycosylationsites similar to those found in mucins. This has now been shown for a series of serum and cell glycoproteins, discussed in Section VII below, where multiple 0-linked chains affect protein conformation and function. The interaction of oligosaccharideswith the protein backbone can lead to joint protein-carbohydrate epitopes, and therefore at a conformational level it is important to analyze glycopeptides rather than only the released oligosaccharides. However, in order to study the diversity of oligosaccharidesequences, the 0-linked, much-type chains of glycoproteins have in general been analyzed after release from protein by treatment with mild alkali -borohydride according to the conditions originally described by Carlson (88).

2. ‘H-n.m.r. - Spectral and Mass-Spectrometric Analysis of Mucin Oligosaccharide Chains The oligosaccharide-alditolsreleased from mucins have 2-acetamido-2deoxy-D-galactit01 (GalNAc-01) at the reduced end, which may readily be detected by H-n.m.r. spectroscopy from the characteristic, multiplet pattern for protons attached at C-2and C-5 in the regions 4.40-4.24 and 4.28-3.75 ppm, respectively. The chemical shifts for these multiplets and for protons at G I , C-3,C-4, and C-6 are primarily dependent on the different substitution patterns at C-3 and C-6 as described in Table IV. However, certain substituents on the residues linked to GalNAc-ol can significantly (greater than k0.03ppm) affect the chemical shifts of GalNAc-01, in particular, the presence of CX-L-FUC at 0 - 3 or 0-4 of p-D-GlcNAclinked to Gal-

-

-

OLIGOSACCHARIDE DETERMINANTS OF GLYCOPROTEINS

323

NAc-ol and CX-L-FUC-( 1 + 2), CX-D-G~CNAC-( 1 4), or a-D-GalNAc-( 1 3) linked to the p-~-Galthat is linked to GalNAc-o1. Thus, in the database used for computer-assisted interpretation of H-n.m.r. spectra (33), the chemical shifts of the “core regions” for mucin-type alditols shown in Table IV have been defined to a tolerance of f 0 . 0 3 ppm. Unlike the situation obtaining with 13C-n.m.r.spectroscopy, no empirical rules can be deduced for predicting the effect of substitution of a hydroxyl group at a particular carbon atom on the chemical shift of the attached proton (except for sulfation, as discussed in Section Vl). Therefore, each type of linkage has to be specified in the databases used for structural interpretation. For novel oligosaccharides for which ‘H-n.m.r. data have not previously been documented in the literature, it is necessary to supplement the ‘Hn.m.r. data with composition, sequence, and linkage information provided by chromatographic and m.s. analyses. An illustrative example is the characterization of the following oligosaccharide of the glycopeptides of fetal gastrointestinal mucins of meconium (74): p-~-Gal-(1 44)-/3-~-GlcNAc 1

1

6 G~~NAC-O~

3

t

1 p-D-Gd-( 1 4~)-~-D-G~cNAc-( 1 -.* 6)-p-D-Gd

There were significant differencesin the chemical shifts of the GalNAc-ol protons in this oligosaccharide compared with the tetrasaccharide lacking the p-~-Gal-( 1 + 4)-p-~-GlcNAc-( 1 + 6) sequence on the lower branch (see Table V) and oligosaccharideshaving a P-D-Gal-( 1 4)-p-~-GlcNAc(1 3)-p-~-Galsequence linked at 0-3 of GalNAc-ol in the presence or absence of a p-D-Gal-( 1 + 4)-p-~-GlcNAc-(1 + 6) branch (see Table IV). In addition, the signals for protons at C-5 and C-6 ofp-D-Gal residues could not be readily distinguished. Thus, from ‘H n.m.r. alone, it was difficult to determine the structure of this oligosaccharide, and m.s. studieswere necessary in order to permit the interpretation that the p-D-Gal residue linked (1 3) to GalNAc-ol was itself linked at 0-6. Methylation analysis of this oligosaccharide showed the presence of terminal Gal, GlcNAclinked at 0-4, Gal linked at 0-6, and GalNAc-ol linked at 0-6 and 0-3. Sequence information from 1.s.i.-ms. of permethylated oligosaccharides showed abundant fragment-ions of m/z 464 and 432, characteristic of terminal Gal-( 1 + 4)GlcNAc sequences. Analysis of PPEADP-derivativesformed after periodate oxidation of the C-4-C-5 bond and reductive amination (28) showed the distribution of tri- and disaccharides at 0-3 and 0-6, respectively, of GalNAc-01. Together, these data gave an unambiguous assignment for the oligosaccharide structure. Thus, it was established that P-D-G~CNAC-( 1 6)-

-

-

-

-

ELIZABETH F. HOUNSELL

324

TABLE111 0-Linked Oligosaccharide Chains of Serum and Cell-Membrane Glycoproteins Glycoproteintype

Oligosaccharide structurfl

~

f [a-NeuAc-(2

Erythrocyte glycophorin (84)

-

8)I-a-NeuAc 2

.1

6 a-GalNAc 3

t

i

f [a-NeuAc-(2 + S)]-cy-NeuAc-(2+ 3)-B-Gal a-Neu Ac 2

1

6 a-GalNAc 3

t

1 &GalNAc-( 1 + 4)-PGal 3

t

2 a-NeuAc f [a-NeuAc-(2 + 3)-/?-Gal-( 1 + 4)-/.-G 1cNAc]

Human chorionic gonadotropin (85)

f

6 a-GalNAc 3

Immunoglobulin A (86)

a-Fuc-( 1

-

2)-&Gal-( 1

-

t

i

a-NeuAc-(2 -+3)-P-Gal 3)+-GlcNAc-( 1 + 3)-P-Gal-( 1 -+4)-GlcNAc 1

1

-

6 a-GalNAc 3

t

i

a-NeuAc-(2 3)-/3-Gal Also,related sequences of the mucin type a

All of the monosaccharides shown have the D configuration except for LFUC.

P - D - G can ~ occur as a linear sequence in the absence of&D-GlcNAc linked at 0-3 ofp-D-Gal [this has since also been established for oligosaccharidesof human-milk mucins (8 l)]. Similarly,it was shown by characterization ofthe trisaccharide P-D-Gal-(1 4)-p-~-GlcNAc-(1 + 6)-GalNAc-ol in the mucins of fetal gastrointestinal tract (meconium) (73,89) that the (1 + 6) linkage can occur in the absence of a (1 3) linkage in the core of mucintype oligosaccharides. This had previously been found in Smith-degraded

-

-

OLIGOSACCHARIDE DETERMINANTS OF GLYCOPROTEINS

325

mucins of human ovarian cyst (57), but is more likely in that case to be formed during the degradation procedure than by specific biosynthesis. In either event, characterization of these novel sequences has increased the diversity of the mucin-oligosaccharidestructure now expected and it underscores the necessity for stringent physiochemical parameters for structural studies. An early pointer to the potential diversity of mucin glycosylationwas the identification of core regions in addition to P-D-Gal-(1 + 3)-a-~-GalNAc and P-D-G~cNAc-( 1 6)-v-~-Gal-( 1 3)]-a-~-GalNAc found in the early classical studies. The first “new” core region was the trihexosamine core P-D-G~cNAc-( 1 6)-u-~-GlcNAc(1 + 3)]-a-~-GalNAcfound initially in sheep gastric mucins (90) and later substantiated by n.m.r. spectroscopy as present in human bronchial mucins (61), ovarian cyst fluid (57), porcine gastric mucins (9 l), and fetal gastrointestinalmucins (75,76). Additional core regions, having linear P-D-G~cNAc-( 1 3)-a-~-GalNAc,/3-DGlcNAc-(1 + 6)-a-~-GalNAc,a-D-GalNAc-(1 + 3)-a-~-GalNAc,and aD-GalNAc-(1 + 6)-a-~-GalNAc,have also been found [e.g., Refs. (73,82); Table 111. The substantiation of the core (Table 11) p-~-Gal-(1 + 6)-@-~Gal-(1 + 3)]-a-~-GalNAcawaits relevant n.m.r. data, and the other core that completes this series, namely, linearj?-D-Gal-(1 + 6)-a-~-GalNAc,has not been identified. +

+

+

-+

V. PERIPHERAL SUBSTITUTIONS OF N- AND 0-LINKED CHAINS OF GLYCOPROTEINS

Considerable structural information and n.m.r. data are available on the core and backbone regions of N- and 0-linked chains of glycoproteinsand proteoglycans as already described for mucins and well known for N-linked chains (30-32,39-47) and proteoglycans (92). The backbone regions of oligosaccharide chains of glycoproteins and proteoglycans consist of the following disaccharide units: p-~-Gal-(1 + 3/4)-p-~-GlcNAcor a-D-Man(1 2/6)-a-~-Man in glycoproteins, /~-D-G~cNAc-( 1 + 4)-p-~-GlcA(1 + 3) in hyaluronate, p-~-GalNAc-(1 + 4)-p-~-GlcA-( 1 + 3) in chondroitin sulfate (CS), and a-~-GlcNS0,-(1 4)-p-~-GlcA-(1 4) in heparan sulfate (HS). The P-D-G~cAresidues of CS and HS can then be epimerized to a-L-IdoA (CS is then called dermatan sulfate and HS, heparin). The chains are further modified by 0-sulfation and, for heparin/heparan sulfate, N-deacetylation followed by N-sulfation. Peripheral to the Gal-GlcNAc sequencesof glycoproteins, there can occur an array of blood-group-related glycosylations, anionic substitutions(sialylation, 0-sulfation, and 0-phosphorylation), or both. Chains of the highmannose type may also bear phosphate groups. Table VI reviews the pat-

-

-+

-

TABLE IV Averaged ’H n.m.r. Chemical Shifts‘ from Data in the Literature for 0-Linked Chain-Core Region Substitutions of GalNAc-ol’ GalNAc4 chemical shifts Sequence /3-Gal-( 1 + 3)-GalNAGOl a-GalNAc

H-1

H-1’

H-2

H3

H-4

H-5

Hd

Hd’

NAc

3.74

3.61

4.39

4.06

3.41

4.16

3.69

3.65

2.050

-

-

4.30

4.10

3.60

4.13

-

-

2.050

-

-

4.39

3.99

3.51

4.14

-

-

2.050

3.86

3.15

4.38

4.06

3.53

4.20

3.85

3.41

f

3 /?-Gal-( 1 + 3)-GalNAc-01 2

t

1 a-FUC /?-GlcNAc

f

6 B-Gal-( 1 + 3)GalNAc-ol 3

t i

B-GlcNAc a-NeuAc 2

1

6 G al N A c -01 3

T

1 B-Gd

2.044

B-GlcNAc 1

.1

6 GalNA001 3

-

4.39

4.06

3.46

4.28

3.93

-

2.067

-

4.40

4.08

3.57

4.26

3.95

-

2.054

-

-

4.40

4.08

3.49

4.25

-

-

2.054

-

-

4.28

4.00

3.56

4.13

3.65

-

2.034

-

4.28

3.99

3.51

4.23

3.91

-

2.044

3.72 3.67 3.66

4.39 4.25 4.24

3.89 3.84 3.84

3.68 3.38 3.41

3.75 4.03

3.65 3.93 3.84

3.67 3.71 3.53

2.049 2.044 2.056

-

t

/%GlCNAC-(1

-

i

4)-Gal WlcNAc

f

6 GalNAc-01 3

w

t:

--

t

1

(Y-Fu~-( 1 2)-Wal W l c N A o ( 1 3)-GalNAc-ol WlcNAc

t

.1

6 GalNAc-01

;i

WlcNAc 1 L-, 3)-GalNAool (Y-G~NAG( B-CICNAC-(1 6)-GalNAc-01 a-NeuAc-(2 6)-GalNAc-ol

-

0

3.8 1 3.73 3.73

-

Given to f0.03 p.p.m. (or f0.005 p.p.m. for NAc) from 4-4-dimethyl4silapentane-l-sulfonateat 295 K

* The data are those used in a computer program for spectral interpretation [asdescribed in Refs. (33)and (34)].

328

ELIZABETH F. HOUNSELL TABLE V Comparison of the 'H N.m.r. Chemical Shifts (p.p.m. from 4,4-Dimethyl-4silapentane-1-sulfonateat 295 K) of a Tetrasaccharide and Related Hexasaccharide from Meconium Glycoproteins(73-75) Proton

Tetrasaccharide /%Gal-(1 + ~)-B-GICNAC

!

Hexasaccharide &Gal-( 1 -+4)-p-GlcNAc

!

6 GalNAc-01 3

t

1 /%Gal

GalNAcol H- 1 H-1' H-2 H-3 H-4 H-5 H-6 H-6' GlcNAc H- 1 H-2 H-3 H-4 H-5 H-6 H-6' p-Gal-(1 -,4) H- 1 H-2 H-3 H-4 B-Gal-(1 3) H- 1 H-2 H-3 H-4 NAc

4.400 4.063 3.451 4.289

-

-

6 GalNAc-01 3

t

1 p-Gal-(1 + 4)+-GlcNAc-(1 + 6)-j?-Gal 3.768 3.76 4.389 4.0 13 3.416 4.235 3.946 3.10

4.558 3.171 3.725 3.692 3.614 4.000 3.840

4.599,4.556 3.76 3.13 3.71 1 3.602, 3.593 4.002, 3.998 3.858, 3.842

4.410 3.536 3.668 3.923

4.470 3.537 3.668 3.922

4.464 3.561 3.673 3.897 2.068, 2065

4.470 3.547 3.612 3.882 2.066, 2.055 2.046

+

TABLEVI Patterns of Peripheral Anionic Glycosylation Documented in 0- and N-Linked Chains of Mammalian Glycoproteins. Position found where uncommon

Sequence a-NeuAc-(2 6)-a-GalNAc-(1 +)-Ser/Thr a-NeuAc-( 2 -+ 3)-P-Gal-( 1a-NeuAc-(2 + 6)-P-Gal-( 1 +

Human colon (71), swine trachea (93), and rat sublingual mucins (54), in addition to N-linked chains Glycophorin (84), neural cell adhesion molecule (NCAM) (94) Bovine submaxillary mucins (54)

a-NeuAc-(2 + 8@-NeuAc-(2a-NeuAc-(2 + 6)$-GlcNAc-( 1a-NeuAc 2

.1

6 P-GlcNAc

Fetuin (93, rat plasma hemopexin (96) Rat serotransferrin (97)

?1 P-Gal a-NeuAc 2

.1

3 P-Gal 4

Human Tamm Horsfall glycoprotein, erythrocytes,and urinary mucins (98)

t

1

P-GalNAc a-NeuAc-(2 + 6)-/?-GalNAc-(1a-NeuGc-(2 + 3)+Gal-( 1a-NeuGc-(2 6)-(~-GalNAc-( 1 +)-Ser/Thr

Human lutropin (99) Porcine submaxillary mucins (100) Porcine ( 100)and bovine (10 1)submaxiUary mucins Keratan sulfate (102- 104), tracheobronchal mucins (64) Keratan sulfate(102- 104),hen ovomucin (105), rat gastric mucins (106), porcine thyroglobulin (107) Recombinant tissue plasminogen activator expressed in mouse epithelial cells (108) Porcine thyroglobulin (107), meconium (76) Canine submaxillary mucins (54) Glycopeptide hormones (99,109) Lysosomal hydrolases (1 10)

-

S0;-6-/l-Gal-( 1SOT-~-P-G~CNAC-( 1a-NeuAc-(2

-

3)-[HS03-6]-P-Gal-(1-

SO,-3-P-Gal-( 1SO;-3 or 4-P-GlcNAc-( 1SO;-4-P-GalNAc-( 1PO;-6-cy-Man-( 1-

a Gal, GlcNAc, and GalNAc have the D configuration. Only the broad type of sialic acid is specified as 5-N-acetyl-(NeuAc) and 5-N-glycolyl (NmGc)neuraminic acid. The large family of sialic acids varying in acylation patterns has to be reviewed as a separate topic ( I 1 I), and indeedclassicalstudiesinvolving release of chains by hydrazinolysis or alkaline-borohydride degradation have largely ignored these substitutions. In more-recent studies (1 12), in which enzymic release of N-Linked chains has been adopted, patterns of 0-acetylation and 0-sulfation are beginning to be established.

329

330

ELIZABETH F. HOUNSELL

terns of anionic substitution found for N- and 0-linked chains of glycoproteins. Short stretches of the sulfated sequencesof proteoglycanscan also be found on 0- and N-linked chains of serum and cell-membrane glycoproteins. Thus, it is pertinent to look for sulfated sequenceson all types ofglycoconjugate.Both mass-spectrometricand n.m.r.-spectral methods are ideal for their identification, and these are treated next. 1. lH-n.m.r. - spectral and Mass-Spectrometric Analysis of Sulfated Oligosaccharide Chains of Glycoproteins

The presence of sulfate groups can be readily detected by f.a.b.- or 1.s.i.m.s. because of the characteristic mass and fragmentation pattern given by successive loss of sulfate, usually as the Na+ or K+ salt. This was well exemplified by the characterization of oligosaccharides of keratan sulfate (102) carried out by negative-ion f.a.b.-m.s. of native oligosaccharides at the 5nmol level. In this study, no fragmentation of the underlying oligosaccharide sequence was seen, and, therefore, no indication was given of the position of the sulfate groups, but only the number of them present. The former information was obtained (103) by 'H-n.m.r. spectroscopy, and another study ( 104) has documented 13C-n.m.r.-spectralparameters for keratan sulfate. The characterization of monosaccharide sequence, linkage pattern, and sulfate location of di- to decasaccharides of keratan sulfate was achieved by 'H-n.m.r. spectroscopy by comparison of the chemical shifts with those of the nonsulfated disaccharide. Table VII shows this comparison for p-DGlcNAc-(1 + 3)-~-Gal and HS03(-, 6)-p-~-GlcNAc-(1 -+ 3)-~-Gal. Large, downfield-shift differences of 0.44 and 0.46 ppm were found for the protons at C-6, and progressively lesser downfield-shift differences were found for protons at C-5, C-4, C-3, and C-2. For the larger oligosaccharides of this series, 'H-n.m.r. correlated spectroscopy (COSY) gave the chemicalshift assignments of the majority of protons, which showed downfield-shift differencescompatible with 0-6 sulfation of all but the reducing-end sugar, as in the following. SOT

1

so;

SOT

1

6 6 p-~-GlcNAc-(1 -D 3)-p-D-Gd-( 1

1

. )

6 ~)-~-D-G~cNAc-( 1 + 3)-D-Gd

Although data on other sulfated oligosaccharidesare not so extensive,it has been shown, for example, that sulfation at 0-3 of galactose in 0-linked chains of meconium (76)and N-linked chains ofporcine thyroglobulin (107) increases the chemical shift of H-1 of Gal by f0.1 16 ppm (kO.00 1 ppm). Together, the published data show characteristic changes in chemical shift, caused by sulfation, which may be used as an indicator in elucidating

OLIGOSACCHARIDE DETERMINANTS OF GLYCOPROTEINS

33 1

TABLEVII Comparison of the 'H n.m.r. Chemical Shifts (p.p.m. from 4,4-Dimethyl-4-silapentane-lsulfonate at 295 K) for the Disaccharide&~-GlcNAc-(l 3)-a/&~-Galand Its Analogue Sulfated at 0-6of GlcNAc

-

Sulfated disaccharide &~-GlcNAc-(l+ 3)-cu/BD-h1 6

t

Disaccharide &~-GlcNAc-(l+ 3)u/&~-Gal Proton GlcNAc H- 1 H-2 H-3 H-4 H-5 H-6 H-6' NAc Gal H- 1

H-2 H-3 H-4 H-5 H-6 H-6'

a 4.7 17

so,

P

a

4.695

4.727

3.752

3.897 3.763 2.038 5.226 3.849 3.900 4.198 4.086

3.576 3.521 3.66 4.335 4.228

3.580 3.512 3.55 4.337 4.219 2.037

4.560 3.529 3.699 4.142 3.8-3.6 3.8-3.6

4.707 3.781

3.565 3.472 3.45

3.561 3.49 1 3.44

P

5.227 3.852 3.895 4.233 4.086

4.563 3.531 3.695 4.176 3.8-3.6 3.8-3.6

the sulfation patterns of a wide range of oligosaccharides. As shown by the results of studies (1 13) of heparin binding to antithrombin 111, sulfation patterns have significant effects on molecular recognition by virtue of their influence on charge density and oligosaccharide topography. For keratan sulfate, there has been characterized a series of monoclonal antibodies that are specific for polysulfated sequences (102,103). The n.m.r. data were consistent with the X-ray crystallographicdata showingsulfate residuespointing away from the Gal-GlcNAc backbone and defining a tetrasulfated tetrasaccharide unit involved in molecular recognition. 2. 'H-n.m.r. Analysis of Sialic Acid Substitution Patterns

The linkage of sialic acid to Gal, GlcNAc, or GalNAc may be determined on the microscale by methylation analysis before and after treatment with

332

ELIZABETH F. HOUNSELL

neuraminidase or mild acid (0.01 -0.1 M,80°C)(3). When sufficient material is available, 'H n.m.r. permits identification of the type of sialic acid present and the substitution pattern. Characteristic chemical shifts for the sialic acid of various sialylated sequencesare compared in Table VIII. It may be seen that at an accuracy of k0.03 ppm (33),it is possible to distinguish the major types of sialic acid linkages shown from the H-3 axial and equatorial signals alone. Additional differences in chemical shift are brought about by the presence of 0-acyl groups as described by Vliegenthart et al. in Ref. 11 1, in particular large downfield shifts of protons on 0-acetylated carbon atoms and on carbon atoms adjacent to those 0-acetylated. More-subtle differences in the chemical shifts of the two H-3 protons and of the H-4-H-9 of sialic acids occur in different oligosaccharide sequences, including those varying at the monosaccharide adjacent to that linked by the sialic acid; for example, the chemical shifts for NeuAc linked (2 --.* 3) to Gal are dependent on the linkage of Gal (1 + 3) to GlcNAc, (1 -+ 4)-GlcNAc, or (1 ---* 3)GalNAc-ol and on the presence of NeuAc (2 -,6), Fuc (1 ---* 3), for Fuc (1 + 4)linked to the GlcNAc. As with fucosylated sequences (1 14- 1 16), these differencesmay be interpreted in terms of conformational information additional to that obtained from nuclear Overhauser enhancement (n.0.e.) studies.

VI. THECONFORMATIONS AND MOLECULAR RECOGNITION OF CARBOHYDRATE DETERMINANTS DISTANT FROM THE PROTEIN OLIGOSACCHARIDE COREOF GLYCOPROTEINS To a certain extent, the peripheral moieties of oligosaccharidechains at a distance from the protein skeleton may be viewed as independent conformational entities. This is exemplified by the fact that the n.m.r. data from a particular structural determinant can be translated to the same determinant on a different core, as discussed previously (33). Considerable use has been made of this situation when viewing oligosaccharidesas antigens related to the blood-group structures in that glycoproteins, glycolipids, and the free oligosaccharidesof milk, urine, and feces have provided an extensive library for structural and immunochemical studies (as reviewed in Refs. 33 and 5 1 and discussed next). 'H-n.m.r.-spectroscopy of oligosaccharidesand glycolipids have provided near-complete chemical-shift assignments and some n.0.e. data on through-space interactions, that have been interpreted in terms of 3D determinants. Conformational analysis by 'H-n.m.r. spectroscopy of free oligosaccharidesis discussed next.

OLIGOSACCHARIDE DETERMINANTS OF GLYCOPROTEINS

333

TABLE VIII 'H N.m.r. Chemical Shifts" (p.p.m. from 4,4-Dimethyl-4-silapntane-lsulfonate at 295 K) of the H-3 Axial and H-3 Equatorial Protons of OligosaccharidesContaining N-Acetylneuraminic Acid and NGlycolylneuraminicAcid Resonance Structure

H-3ar

H-3eq

1.80 1.93

2.16 2.66 2.67 2.68 2.14 2.12 2.19 2.15

~~

a-NeuAc-(2 + 3)-/3-~-Gd a-NeuAc-(2 -B 3)-p-~-GalNAc-(1 -+ 4)-]B-~-Gal a-NeuAc-(2 + 6)-B-~-Galor -p~-GalNAc -P 6)-/3-~-Gal c~-NeuAc(COAc)-(2 a-NeuAc-(2 -,6)-/3-~-GlcNAc c~-NeuAc-(2-,6)-~-GalNAc-ol a-NeuGc-(2 -B 3)-/3-~-Gal a-NeuGc-(2 --* 6)-GalNAc-ol

1

I1

1.85 1.70 1.69 1.82 1.71

Chemicalshiftsareaveragedvalues(~.03p.p.m.)(33)fromourownstudiesand thosefrom Refs. 31-33,62,66,86,96-101,111,and 112.

1. Blood-Group-Related Antigens Lemieux (1 14) has discussed the recognition of the blood-group-type oligosaccharide determinants by antibodies and lectins based on (i) a series of pioneering studies on the n.m.r.-spectral analysis of chemically synthesized oligosaccharides and (iz] molecular modeling. Earlier studies reviewed by Durette and Horton ( 1 17)had established monosaccharide conformational analysis, in particular defining the ring stereochemistryand the importance of the exoanomeric effect. The use of the hard-sphere exoanomeric algorithm (HSEA) to predict qb,y angles around the glycosidic bonds, together with data from n.m.r.-spectral studies [which, for example, permits detection of strong deshielding of protons in close proximity to glycosidic-bond and ring-oxygen atoms (1 15)], was applied to the analysis of a series of oligosaccharide determinants recognized by lectins and anti-blood-group antisera ( 1 18- 121). Similar studies were performed on oligosaccharides obtained from milk (1 16,122). The extensive purification and structural studies (123- 125) of Kuhn, Kobata, Yamashita, Ginsburg, and colleagues established the diversity of the oligosaccharide molecules in milk. New structures continue to be found ( 126) and similar oligosaccharides in feces (127 - 129), and urine ( 130- 132)have provided other sources for chemicalshift assignment by COSY, relayed COSY, and long-range COSY experiments. Additional n.m.r. parameters from I3C- and 'H-n.m.r.-spectral ex-

334

wo ELIZABETH F. HOUNSELL

e

HO

H

,

o

H

OH OH

(P

4)

(3

FIG.1. -The disaccharide&D-Gal-(1 -P 4)-/.%~-GlcNAc, showing the 4 ,angles ~ defined by IUPAC a~ 0'-C'-0-C,, C'-O-C,-C~,~~~.

periments leading to conformational information on these oligosaccharide sequences [for example, analysis of spin-relaxation time, n.0.e. effects, and 13C- 'H long-range coupling constants (1 33- 136)]have provided extensive data that substantiate the modeling approach, as have molecular-dynamic simulations (1 35,137,138). In several cases, these data have been incorporatedinto molecular models built with the aid of information from molecular-mechanics algorithms (e.g., HSEA, GSEA, MM2CARB) on energetically favorable 4 , torsion ~ angles (see Fig. 1) around the glycosidic bonds. The models are made using commercial software packages, the forcefields of which (e.g., AMBER, CVFF, TRIPOS, and CHARMm) are parameterized to take account of specific oligosaccharide factors such as charge and the anomeric effect (137,138). To date, these calculations have mostly been based on simulations in a vacuum, although other studies have examined the effect of water on energy minimization and molecular dynamics or have approximatedthe

TABLEIX Calculation of the 4, yl Angles of Minimum-Energy Conformers by the HSEA Algorithm" Angles Sequence

-

a-Fuc-(1 + 2) to Galb ~-Fuc-( 1 3) to GIcNAc ~-Fuc-( 1 -+ 4) to G~cNAc /%Gal-(1 3) to GlcNAc B-Gal-( I -4) to GlCNAc p-GlcNAC-( 1 + 3) to Gal c~-Gal-( 1 3) to Gal

-

6 -70

-70 -65 - 60 -65 -66 55

Structure(Fig. 2)

135 -95 145 - 1 10 130 150 -170

a b C

d e

f 8

a Data taken from Lemieux ef al. and standardized to IUPAC format (Fig. 1). All monosaccharidesare in the D form except for L-FUC.

OLIGOSACCHANDE DETERMINANTS OF GLYCOPROTEINS

335

presence of water by use of a dielectricconstant of 1- 80 keV. The 4,vangles are shown in Table IX for the series of blood-group-relateddisaccharides at minimum-energy conformations(Fig. 2) predicted by the HSEA algorithm. These disaccharides serve as building units for oligosaccharides, which can then be refined by using computer-graphicsmolecular mechanics programs parameterized for oligosaccharides. Different methods of calculation often give similar contour patterns of energetically favorable conformations, but find different energy minima. As the probability of finding the molecule in the energetic minimum is low, conformational space around several lowenergy states should be explored ( 139).Where through-spacedistances given by n.m.r.-n.0.e. experiments are known and where interpretation of immunological- biological data is involved, models incorporating solution conformations within the low-energy wells may be of great help in visualizing oligosaccharide determinants of activity. N.m.r.-spectral data are crucial for any model building, as the relative flexibilityand heterogeneityof the oligosaccharidemoieties of glycoproteinshas thus far largely prevented their visualization by X-ray crystallography. It should be remembered, however, that the molecules could sample 3D conformational space at a higher rate than can be detected by n.m.r. relaxation and, furthermore,that, in molecular recognition of oligosaccharides, energetically unfavorable solutionconformations could be bound. With respect to the blood-group-related antigens, these generally form structures that are more stereochemically restricted and for which modeling incorporating n.0.e. data may be more appropriate than for flexible backbones. This factor may also contribute to their immunogenicity, as determined by their recognition by polyclonal serum antibodies and monoclonal antibodies raised in vitro. The first oligosaccharide determinant for a hybridoma antibody characterized in detail was the 3-fucosyl-N-acetyllactosamine(W, Table X) sequence P-D-Gal-(1 + 4)-[a-~-Fuc(1 + 3)]$-~-GlcNAc. The initial antibody was raised against a stage-specific embryonic antigen (SSEA-1) of mouse (140) and characterized by using oligosaccharides obtained from human milk ( 140,141). Other antibodiesraised against a promyelomonocytic leukemia cell line (HL60) and against the receptor for epidermal growthfactor were also found to recognizethe same or a related structure (142,143). Determination of the specificityof other monoclonal antibodiesrecognizing oligosaccharides of glycoproteins soon followed [as reviewed in Ref. (5 l)]. Studies described therein established, for example, that W, W, and combined difucosylated W/W are tumor-associated antigens of milk fatglobule membrane glycoproteins (1 16) and that the ALeb and ALeY (Table X) sequence is present on the N-linked chains of the receptor for epidermal growth factor on A43 1 cells (143). Structural studies on human, small intestinal, epithelial cells ( 144) have since revealed the presence of Ley and ALeY

ELIZABETH F. HOUNSELL

336

a

C

b e

FIG.2. -Conformations of oligosaccharides having the 4 , angles ~ shown in Table IX for minimum-energy conformationspredicted by the HSEA algorithm for use as building units in molecular modeling. The molecules are shown without the anomeric hydroxyl group.

OLIGOSACCHARIDE DETERMINANTS OF GLYCOPROTEINS

337

FIG.2. (Continued)

sequences on N-linked chains. As with the 3-fucosyl-N-acetyllactosamine and poly-N-acetyllactosamine(i antigenic structure) sequences, these were detected first on N-linked chains by antibody reactivity before they were established by structural characterization (5 1). Figure 3 shows predicted solution conformersfor oligosaccharidedeterminantsrelated to Le" and Le". These models may be used to propose molecular features recognized by the series of monoclonal antibodiesraised against these structures. They can also be used to interpretthe molecular recognition of other carbohydrate-binding proteins of nonimmune origin in plants and mammals. As discussed next, related structures have been shown to be important in the interactions of lymphocytes, platelets, and endothelial cells ( 145). 2. Sialylated Oligosaccharide Determinants

The first conformational information on the large family of sialic acids (1 1 1) of mammalian glycoproteins came from the comprehensive study ( 146) of the following glycopeptides, a-NeuAc-(2 a-NeuAe(2

-

3)-/?-~-Gd-(1 44 ) - P - ~ - G lc N A o(l~ Am)

6)-/?-DGd-( 1 + 4)-/?-~-GlcNAc-(l+ Am),

and the sialylated Le"antigen Ca 19.9, which is a tumor-associated antigen associated with several gastrointestinal malignancies ( 147): a-NeuAc-(2

-

3)-/?-~-Gal-(1

-

3)-a-[~-Fuc-(1

-

4)]-~-GlcNAc.

The data for these oligosaccharideshave been incorporated into the molecular models shown in Figs. 3 and 4. The conformation and molecular recognition of sialylated oligosaccharidesare an important area for future investi-

ELIZABETH F. HOUNSELL

338

TABLE X The Oligosaccharide Sequences of Blood-Group and Related Determinants Shown in Figs. 3 and 4

Determinant

Sequence LY-L-FUC

P-D-Gal-( 1

Lea

i

4

~)-~D-G~cNAc gal

f

4

1 3)-p-D-&NAc LY-L-FUC-( a-L-FUC

Le"

Sialyl Le"

-

-

i

4

a-~-NeuAc-(2 3)-j.-~-Gd-( 1 + 3)-/3-~-GkNAc a-~-NeuAc-(2 3)-b-~-Gal

Sialyl Le"

f

4

(Y-L-Fuc-(1 ~)-~-D-G~cNAc LY-D-G~NAc (Y-L-FUC 1

ALeb

1

3 LY-L-FUC-( 1 + 2)-p-~-Gd-( 1

-

1

1

4 3)-/3-~-GlcNAc (Y-D-G~~NAc

3 1 2)-p-~-Gd LY-L-FUC-( ALeY

i

4

LY-L-FUC-( I 3)-/3-D-GlcNAC

gation because of the perceived central importance of these prominent chain-terminating sequences in cellular interactions. There is strong evidence that the Lea and Lexdeterminants having a-NeuAc-(2 +-3)-P-~-Gal substitution are involved in the specific recognition of lymphocytes and neutrophils with endothelial cells via carbohydrate-binding proteins called selectins [reviewed in Ref. (1431. Both sialylated and sulfated oligosaccharide sequences have been implicated in trafficking of lymphocytes to high endothelial cell venules. These findings build on earlier studies on mammalian carbohydrate recognition, which showed a binding lectin of the liver ( 1 48)and a Man-6-PO,-binding lectin of lysosomal membranes ( 1 10,149), both involved in glycoprotein trafficking.

OLIGOSACCHARIDE DETERMINANTS OF GLYCOPROTEINS

339

Additional conformational studies have been carried out on the [aNeuAc-(2 s)],homopolymers that make up the type-specific antigens of meningococcus (1 50,15 1). These same sequences have been identified on the human neural cell-adhesion molecule (NCAM) as components of Nlinked glycoprotein chains (94) and are stronglyimplicated in the etiology of meningococcal disease, because they occur primarily in the young, who are susceptible, but are not expressed on adult NCAM. The importance of sialic acid in infection was first recognized for the cellular receptor of influenza virus hemagglutinin by Gottschalk (1 52). Further conformational studies ( 153) have identified the molecular features recognized by the hemagglutinin binding-site, which appears to be largely restricted to functional groups on the a-NeuAc-(2 + 3/6) moiety itself rather than internal residues of the glycoprotein or glycolipid chains that the virus binds. Sialylated sequences are also the targets for bacterial adhesion and bacterial toxins. The specificity for binding of Mycoplasma pneumoniae was shown ( 154) to involve polylactosamine sequences terminating in aNeuAc-(2 + 3)-P-~-Gal.There is a transient rise in anti-1 titer directed against the backbone [p-D-Gal-(1 4)-p-~-GlcNAc-(1 3/6)], sequence experienced after infection by M. pneumoniae. The i and I antigens were initially characterized as, respectively, linear and branched poly-N-acetylactosamine sequences recognized by naturally occurring monoclonal coldagglutinin autoantibodies(5 1). In addition, among the erythrocyte autoantibodies associated with cold-agglutinin disease of man, several are known to be monoclonal antibodies directed against sialic acid-containing determinants (1 55,156). The antibodies designated anti-Pr, anti-Gd, Fl, and Sa appear to react primarily with the sequence a-NeuAc-(2 3)-p-~-Gal(1 4)-p-GlcNAc, although there is some variability in activity with a-DNeuAc-(2 3) or c~-NeuAc-(2 6) on different backbones (156). Early data (1 57,158) on the conformations of the backbone antigens i and I are incorporated into the molecular modeling shown in Fig. 4, together with sialylated analogues. Another antigen of importance in medicine is the sequence (Fig. 4) aNeuGc-(2 3)-p-~-Gal-.This is the Hanganutziu- Diecher (H-D) antigen originally recognized as being an antigen giving rise to serum sicknesscaused by transfusion of horse serum into humans ( 159).The sequenceis present on horse erythrocyte glycoproteins, but is not biosynthesized in human cells, except as a minor component on tumor glycolipids.However, in the new era of recombinant DNA technology, this sequence has been detected on glycoproteins destined for human use that have been grown in Chinese hamster ovary (CHO) cells. This finding underlines the importance of the characterization of oligosaccharide determinants that are central factors involved in infection and immunity. +

-

- -

-

-

-

-

340

FIG.3-Possible solution conformers of oligosaccharidesequencesrelated to the Lenand Lebantigens (Lea,Lex,sialyl Lea,sialyl Lex,ALeb, ALey ). As described in the text, oligosaccharides in solution sample 3D space in local energy-minima wells, so that the conformations shown are approximate only. The models stand only to show the stereochemicalhindrance to movement in these branched oligosaccharides.The molecules are shown without the anomeric hydroxyl group.

Sialyl I

Sialyl i

-

FIG.4-A comparison of the H-D antigenic component and the receptor for influenza virus hemagglutinin, which are a-NeuGe(2 + 3)-/3-Gal and a-NeuAc-(2 3)-/3-Gal, respectively, on the backbones pD-Gal-(1 + 4)-p~-GlcNAc-(1 + 3)-&~-Gal-(1 4)-~-GlcNAc (sialyl i), /3-~-Gal-(1 +4)-&~-GlcNAc-(1 + 6)-~-Gal (sialyl I). The CH, of the acetamido group of N-acetylneuraminic acid, which is a CH,OH group in N-glycolylneuraminic acid. is shown.

-

OLIGOSACCHARIDE DETERMINANTS OF GLYCOPROTEINS

343

VII. THECONFORMATIONS AND MOLECULAR RECOGNITION PROTEIN

OF CARBOHYDRATE DETERMINANTS ADJACENT TO THE MOIETYOF GLYCOPROTEINS

The oligosaccharide sequences just discussed are linked to glycoproteins through the following(generalized)N- and 0-linked cores(to asparagineand serine or threonine, respectively).

--

-

u-D-Gd-( 1 4/3)-&~-GkNAc-(1 + 3)In-1 6/4/2-c~-~-Man-( 1 + 6/3)&&Man-( 1 4)-&~-GlcNAc-(1 + 4)-8-~-GlcNAc-(1 +Am)

-

p-D-Gd-( 1 -D 4/3)-fiD-GlCNAC-( 1 -D ~/~)-],-~Y-D-G~NAC-(I -P Ser/Thr) or p-~-Gal-( 1 + 4/3)-/3-~-GlcNAc-(1 6/3)-],-fi~-Gd-( 1 -+ 3)-a-~-GdNAc-(l-+Ser/Thr)

Numerous detailed studies have been performed on the conformations of D-Man-containing sequencesof glycoproteins( 137- 139,160- 168). The results show that there is considerable flexibility and make it difficult to summarize the solution conformations that oligosaccharide chains containing these sequences are likely to adopt. The general consensus is, as predicted by Montreuil(30), that many allowed conformations are possible. For operational reasons, many of the studies have been performed on isolated oligosaccharides and have not taken into account (i) the restrictions of degrees of freedom imposed by the hydrodynamic volume of completeoligosaccharide chains in solution and (iz] the effect ofinteraction with the protein backbone. For N-linked chains, it is pertinent to discuss the latter aspect, for which some constraintson degrees of freedom have been established.For 0-linked chains, a picture in which the chains can be said to be in intimate contact with the protein, imposingconformationalconstraintsand allowingforjoint oligosaccharide- protein determinants, is emerging. 1. N-Linked Chains

There is an extensive databank on the chemical shifts of free oligosaccharides and those linked to one or two amino acids. Several n.m.r. and modeling studies have also been performed on glycopeptides and glycoproteins ( 169- 172). The conclusion from these latter studies is that the asparagineGlcNAc bond extends away from the protein backbone, so that only small conformational changes take place upon elongation of the oligosaccharide side-chains, and attachment to protein does not affect the solution conformation of the oligosaccharide. Where X-ray crystallographic data for glycoproteinsare available, results show that the N-linked oligosaccharides either can extend away from the protein [as suggested (173), for example, for the single chain at Asn 86 in the human lymphocyte antigen HLA A2], or may interact with the rest of the molecule [as shown (42,174), for example, for the two chains of the CH,

344

ELIZABETH F. HOUNSELL

domains of the Fc region of immunoglobulin GI. In the former case, it appears that there is no significant interaction of the protein and carbohydrate, and therefore it may be envisaged that the oligosaccharide determinants at the periphery of the chains are themselvesfunctional units that have a unique accessibility at the surface of glycoproteins and cells. 2. 0-Linked Chains

Initial conformational studies suggest that, for 0-glycosylation,the glycosylated amino acid will be in a p-turn in order to accommodate the &-GalNAc residue and that there is more conformational flexibility for GalNAcSer than GalNAc-Thr, but both of these 0-a-glycosidic linkages are substantially more rigid than those in the p-GlcNAc-Asn case (170,174). An additional factor to be considered with respect to 0-glycosylation is that multiple glycosylation sites are often present, for example, in glycophorin (175- 177), the antifreeze glycoprotein of fish (178), and mucins (179 - 180). Multiple 0-glycosylation sites have also been shown on a variety of cell-membrane and serum glycoproteins (Table XI). Their effects on protein conformation are now being ascribed specific biological roles ( 181) in addition to their recognition as antigens. Because of the close interaction of 0-linked oligosaccharideswith the peptide backbone, each moiety may affect the antigenicity of the other, for example, sialic acid residues on the disaccharidep-D-Gal-( 1 +. 3)-a-~-GalNAc-( 1 -)-Ser/Thr are necessary for binding to some blood-group M and N antisera ( I 75 - 177).The ThomsenFriedenreich (T) antigen, having the unsialylated sequence p-D-Gal-( I 3)-a-~-GalNAc-(1 -+Ser/Thr, may be detected, for example, as an oncodevelopmental antigen in fetal and neoplastic colonic mucosa but not in the normal adult mucosa (175). Likewise a single-a-GalNAc linked to a Thr residue of human fibronectin forms a specific oncodevelopmental epitope

-

TABLE XI Examples of Serum and Membrane Glycoproteins Having Multiple O-Glycosylation Sites Lymphocyte antigen CD45 Human chorionic gonadotropin Lipoprotein (a) Low-density lipoprotein (LDL) receptor Human lymphocyte antigen CD8 Interleukin-2 receptor Rat renal a-glutamyltranspeptidase Sucrase and isomaltase

OLIGOSACCHARIDE DETERMINANTS OF GLYCOPROTEINS

345

recognized by a monoclonal antibody (1 82). Last, a single N-acetylglucosamine group /I-linked to Ser or Thr occurs in cytoplasmic glycoproteins ( 183,184)and appearsto be important in their intracellulartrafficking;it has been proposed ( 17)that this may be a reciprocal signal to 0-phosphorylation of proteins. The overall view that can be obtained to date is that oligosaccharide sequences have many and diverse roles that we are now beginning to dissect, starting from a base of detailed physicochemical characterization. ACKNOWLEDGMENTS The author is extremely grateful to Miss E. White for her patient typing, to Mr. D. Renouf for

his collaboration in the molecular-modelingstudies, and to Dr. T. Feizi for helpful discussions.

REFERENCES 1. D. A. Rees, Adv. Carbohydr. Chem. Biochem., 24 (1969) 267-332. 2. W. C. Johnson, Adv. Carbohydr, Chem. Biochem., 45 (1987) 73- 124. 3. E. F. Hounsell (Ed.), Methods in Molecular Biology, Vol. 14, GlycoproteinAnalysis in Biomedicine. Humana Press, 1993. 4. S . Hakomori, J. Biochem. (Tokyo), 55 (1964) 205-207. 5. K. Stellner, H. Saito, and S. Hakomori, Arch. Biochem. Biophys., 155 (1975) 464-472. 6. P. A. Sandford and H. E. Conrad, Biochemistry, 5 ( 1966) 1508- I5 12. 7. D. M. W. Anderson and G. M. Cree, Carbohydr. Res., 2 (1966) 162- 167. 8. J. S. Brimacombe, B. D. Jones, M. Stacey, and J. J. Willard, Carbohydr. Rex, 2 (1966) 167- 174. 9. H. Bjomdal, C. G . Hellerqvist, B. Lindberg, and S. Svensson, Angew. Chem. Int. Ed. Engl., 9 (1970) 610-619. 10. G. G. S. Dutton, Adv. Curbohydr. Chem. Biochem., 30 (1974) 10- 110. 11. K. A. Karlsson,I. Pascher, W. Pimlott, and B. E. Samuelsson,Biomed. Mass Spectrom., 1 (1974) 49-56. 12. I. Ciucanu and F. Kerek, Carbohydr. Res., 131 (1984) 209-217. 13. R. Kuhn and H. Egge, Chem. Ber., 96 (1963) 3338-3348. 14. S. Honda, Anal. Biochem., 140 (1984) 1-47. 15. G . C. Hansson, I. Carlstedt, and H. Karlsson, Biochemistry, 28 (1989) 6672-6678. 16. Y. Leroy, J. Lemoine, G. Ricart,J.-C. Michalski, J. Montreuil, and B. Foumet, Anal. Biochem., 184 (1990) 235-243. 17. K. Kakehi and S. Honda, J. Chromatogr., 379 (1986) 27-55. 18. P. Jackson, Biochem. SOC.Trans. 21 (1993) 121- 125. 19. C. C. Sweeley, R. Bentley, M. Makita, and W. W. Wells, J. Am. Chem. Soc., 85 (1963) 2497-2507. 20. T. Bhatti, R. E. Chambers, and J. R. Clamp, Biochim. Biophys. Acta., 222 (1970) 339347. 2 1. G. J. Genvig, J. P. Kamerling, and J. F. G. Vliegenthart, Carbohydr. Rex, 77 ( I 979) 1- 7. 22. M. R. Hardy, in Methods in Enzymolom, Vol. 179, pp. 76-82. Academic Press, San Diego, 1989. 23. A. Dell, Adv. Carbohydr. Chem. Biochem., 45 (1987) 19-72. 24. A. L. Burlingame, D. Maltby, D. H. Russell, and P. T. Holland, Anal. Chem., 60 (1988) 294R-342R.

346

ELIZABETH F. HOUNSELL

25. E. F. Hounsell, M. J. Madigan, and A. M. Lawson, Biochem. J.,219 (1984) 947-952. 26. A. M. Lawson, W. Chai, G. C. Cashmore, M. S. Stoll, E. F. Hounsell, and T. Feizi, Carbohydr. Rex, 200 (1990) 47-57. 27. A. S. Angel and B. Nilsson, in Methods in Enzymology, Vol. 193,pp. 587-607. Academic Press, San Diego, 1990. 28. M. S. Stoll, E. F. Hounsell, A. M. Lawson, W. Chai, and T. Feizi, Eur. J. Biochem., 189 (1990) 499-507. 29. T . Feizi, in Carbohydrate Recognition in Cellular Function, Ciba Foundation Symposium 145, pp. 62-79. Ciba Foundation, 1989. 30. J. Montreuil, Adv. Carbohydr. Chem. Biochem., 37 (1980) 157-223. 31. J. F. G. Vliegenthart, H. van Halbeek, and L. Dorland, Pure Appl. Chem., 53 (1981) 45-77. 32. J. F. G. Vliegenthart, L. Dorland, and H. van Halbeek,Adv. Carbohydr. Chem. Biochem., 41 (1983) 209-374. 33. E. F. Hounsell and D. J. Wright, Carbohydr. Res., 205, (1990) 19-29. 34. K. Bock, C. Pederson, and H. Pederson, Adv. Carbohydr. Chem. Biochem., 42 (1984) 193-225. 35. K. Dill, E. Berman, and A. A. Pavia, Adv. Carbohydr. Chem. Biochem., 43 (1985) 1-49. 36. J. A. van Kuik, K. Hird, and J. F. G. Vliegenthart, Carbohydr. Res. 235 (1992) 53-68. 37. K. Hermansson, P.-E. Jansson, L. Kenne, G. Widmalm, and F. Lindh, Carbohydr. Rex 235 (1992) 69-81. 38. G. M. Lipkind, A. S. Shashkov, N. E. Nifant’ev, and N. K. Kochetkov, Carbohydr. Res., 237 (1992) 11-22. 39. M. W. Spellman, Anal. Chem., 62 (1990) 1714- 1722. 40. G. Pfeiffer, H. Geyer, I. Kalsner, P. Wendorf, and R. Geyer, Biomed. Chrornatogr., 4 (1990) 193- 199. 4 1. T. Mizuochi, T. J. Matthews, M. Kato, J. Hamako, K. Titani, J. Solomon, and T. Feizi, J. Biol. Chem., 265 (1990) 8519-8524. 42. T. W. Rademacher, R. B. Parekh, and R. A. Dwek, Annu. Rev. Biochem., 57 (1988) 785-838. 43. K. Yamashita, T. Mizuochi, and A. Kobata, in Methods in Enzymolou, Vol. 83, pp. 105- 126. Academic Press, San Diego, 1982. 44. F. Maley, R. B. Trimble, A. L. Tarentino, and T. H. Plummer, Jr., Anal. Biochem., 180 (1989) 195-204. 45. K. D. Smith, A.-M. Harbin, R. A. Carruthers, A. M. Lawson, andE. F. Hounsell, Biomed. Chromatogr.,4 (1990) 26 1-266. 46. M. J. Davies, K. D. Smith, R. A. Carruthers, W. Chai, A. M. Lawson and E. F. Hounsell, J. Chromatography 646 (1993) 317-326. 47. A. Kobata, in V. Ginsburg and P. W. Robbins (Eds.), Biology of Carbohydrates,Vol. 2, p. 87. Wiley, New York, 1984. 48. W. T. G. Morgan, Proc. R. SOC. London Ser. B., 151 (1960) 308-347. 49. W. M. Watkins, Biochem. SOC.Trans., 15 (1987) 620-624. 50. E. A. Kabat, in Carbohydrates in Solution, Advances in Chemistry Series No. 117, pp. 334-361. American Chemical Society, Washington, D.C., 1972. 51. T. Feizi, Nature, 314(1985) 53-57. 52. S. Hakomori, Sci. Am., 254(5) (1986) 32-41. 53. E. F. Hounsell and T. Feizi, Med Biol., 60 (1982) 227-236. 54. A. Herp, A. M. Wu, and J. Moschera, Mol. Cell Biochem., 23 (1979) 27-44. 55. A. M. Wu, E. A. Kabat, B. Nilsson, D. A. Zopf, F. G. Gruezo, and J. Liao, J. Biol. Chem., 259 (1984) 7178-7186. 56. B. N. N. Rao, V. K. Dua, and C. A. Bush, Biopolymers, 24 (1985) 2207-2229

OLIGOSACCHARIDE DETERMINANTS OF GLYCOPROTEINS

347

57. J. H. G. M. Mutsaers, H. van Halbeek, J. F. G. Vliegenthart,A. M. Wu, and E. A. Kabat, Eur. J. Biochem., 157 (1986) 139-146. 58. V. K. Dua, B. N.N. Rao, S.-S. Wu, V. E. Dube,andC.A. Bush, J.Biol.Chem.,261 (1986) 1599- 1608. 59. T. Feizi, E. A. Kabat, G. Vicari, B. Anderson, and W. L. Marsh, J. Imrnunol., 106 (1971) 1578- 1592. 60. G. Lamblin, M. Lhermitte, A. Boersma, P. Roussel, and V. Reinhold, J. Biol. Chem.,255 (1980) 4595-4598. 6 1 . H. van Halbek, L. Dorland, J. F. G. Vliegenthart, W. E. Hull, G. Lamblin, M. Lhermitte, A. Boersma, and P. Roussel, Eur. J. Biochem., 127 (1982) 7-20. 62. G. Lamblin, A. Boersma, A. Klein, P. Roussel, H. van Halbeek, and J. F. G. Vliegenthart, J. Biol. Chem., 259 (1984) 905 1-9058. 63. G. Lamblin, A. Boersma, M. Lhermitte, P. Roussel, J. H. G. M. Mutsaers, H. van Halbeek, and J. F. G. Vliegenthart, Eur. J. Biochem., 143 (1984) 227-236. 64. T. P. Mawhinney, E. Adelstein, D. A. Moms, A. M. Mawhinney, and G. J. Barbero, J. Biol. Chem., 262 (1987) 2994-3001. 65. J. Breg, H. van Halbeek, J. F. G. Vliegenthart, G. Lamblin, M.-C. Houvenaghel, and P. Roussel, Eur. J. Biochem., 168 (1987) 57-68. 66. H. van Halbeek, J. Breg, J. F. G. Vliegenthart, A. Klein, G. Lamblin, and P. Roussel, Eur. J. Biochem., 177 (1988) 443-460. 67. J. Breg, H. van Halbeek, J. F. G. Vliegenthart, A. Klein, G. Lamblin, and P. Roussel, Eur. J. Biochem., 171 (1988) 643-654. 68. A. Klein, G. Lamblin, M. Lhermitte, P. Roussel, J. Breg, H. van Halbeek, and J. F. G. Vliegenthart, Eur. J. Biochem., 171 (1988) 631-642. 69. B. L. Slomiany, E. Zdebska, and A. Slomiany, J. Biol. Chem., 259 (1984) 2863-2869. 70. A. Slomiany,E. Zbedska,andB. L. Slomiany,J.Biol. Chem.,259(1984) 14,743- 14,749. 71. D. K. Podolsky, J. Biol.Chem., 260(1985) 15,510-15,515. 72. D. K. Podolsky, J. Biol. Chem., 260 (1985) 8262-8271. 73. E. F. Hounsell, A. M. Lawson,J. Feeney, H. C. Gooi, N. J. Pickering, M. S. Stoll, S. C. Lui, and T. Feizi, Eur. J. Biochem., 148 (1985) 367-377. 74. E. F. Hounsell, A. M. Lawson, J. Feeney, G. C. Cashmore, D. P. Kane, M. Stoll, and T. Feizi, Biochem. J.,256 (1988) 397-401. 75. E. F. Hounsell, A. M. Lawson, M. S. Stoll, D. P. Kane, G. C. Cashmore, R. A. Carruthers, J. Feeney, and T. Feizi, Eur. J. Biochem., 186 (1989) 597-610. 76. C. Capon, Y. Lertoy, J.-M. Wieruszeski, G. Ricart, G. Strecker, J. Montreuil, and B. Fournet, Eur. J. Biochem., 182 (1989) 139- 152. 77. A. Kurosaka, H. Nakajima, I. Funakoshi, M. Matsuyama, T. Nagayo, and I. Yamashina, J. Biol. Chem., 258 (1983) 11,594-11,598. 78. F.-G. Hanisch, H. Egge, J. Peter-Katalinic,G. Uhlenbruck, C. Dienst, and R. Fangmann, Eur. J. Biochem., 152 (1985) 343-351. 79. E. C. Yurewicz, F. Matsuura, and K. S. Moghissi,J. Biol.Chem.,262 (1987) 4733-4739. 80. T. Saito, T. Itoh, and S. Adachi, Biochim. Biophys. Actu, 964 (1988) 213-220. 81. F.-G. Hanisch, G. Uhlenbruck, J. Peter-Katalinic, H. Egge, J. Dabrowski, and U. Dabrowski, J. Biol. Chem., 264 (1989) 872-883. 82. W. Chai, E. F. Hounsell, G. C. Cashmore, J. R. Rosankiewicz, C. J. Bauer, J. Feeney, T. Feizi, and A. M. Lawson, Eur. J. Biochem., 203 (1992) 257-268. 83. H. Schachter, Biochem. Cell. Biol., 64(1986) 163-181. 84. M. Fukuda, M. Lauffenburger, H. Sasaki, M. E. Rogers, and A. Dell,J. Biol.Chem., 262 (1987) 11,952-11,957. 85. L. A. Cole, S. Birken, and F. Perini, Biochem. Biophys. Res. Commun., 126 (1985) 333-339.

348

ELIZABETH F. HOUNSELL

86. A. Pierce-Crktel, J.-M. Wieruszeski, G. Strecker, J. Montreuil, and G. Spik, Eur. J. Biochem., 182 (1989) 457-476. 87. R. A. Childs, R. Dalchau, P. Scudder, E. F. Hounsell, J. W. Fabre, and T. Feizi, Biochem. Biophys. Res. Commun.. 110 (1983) 424-431. 88. R. N. Iyer and D. M. Carlson, Arch. Biochem. Biophys., 142 (1971) 101- 105. 89. J. Feeney, T. A. Frenkiel, and E. F. Hounsell, Curbohydr. Res., 152 (1986) 63-72. 90. E. F. Hounsell, M. Fukuda, M. E. Powell, T. Feizi, and S. Hakomori, Biochem. Biophys. Res. Commun., 92 (1980) 1143- 1150. 9 1. H. van Halbeek, L. Dorland, J. F. G. Vliegenthart, N. K. Kochetkov, N. P. Arbatsky, and V. A. Derevitskaya, Eur. J. Biochem., 127 (1987) 21 -29. 92. A. R. Poole, Biochem. J.,236 (1986) 1- 14. 93. S. S. Rana, E. V. Chandrasekaran, and J. Mendicino, J. Biol. Chem., 262 (1987) 36543659. 94. J. Finne, V. Finne, H. Deagostini-Bazin, and C. Goridis, Biochem. Biophys. Res. Commun., 112 (1983) 482-487. 95. E. D. Green, G. Adelt, J. U. Baenziger,S.Wilson, and H. van Halbeek, J. Biol. Chem., 263 (1988) 18,253- 18,268. 96. N. Bernard, R. Engler, G. Strecker, J. Montreuil, H. van Halbeek, and J. F. G. Vliegenthart, Glycoconjugutes, 1 (1984) 123- 140. 97. G. Spik, B. Coddeville, G. Strecker,J. Montreuil, E. Regoeczi, P. A. Chindemi, and J. R. Rudolph, Eur. J. Biochem., 195 (1991) 397-405. 98. C. P. C. Soh, A. S. R. Donald, J. Feeney, W. R. J. Morgan, and W. M. Watkins, Glycoconjugute J.,6 (1989) 319-332. 99. G. Weisshaar, J. Hiyama, G. C. Renwick, and M. Nimtz, Eur. J. Biochem., 195 (1991) 257-268. 100. A. V. Savage, C. M. Donoghue, S. M. DArcy, C. A. M. Koeleman, and D. H. van den Eijnden, Eur. J. Biochem., 192 (1990) 427-432. 101. A. V. Savage, P. L. Koppen, W. E. C. M. Schiphorst,L. A. W. Trippelvitz, H. vanHalbeek, J. F. G. Vliegenthart, and D. H. van den Eijnden, Eur. J. Biochem., 160 (1986) 123129. 102. P. Scudder, P. W. Tang, E. F. Hounsell, A. M. Lawson, H. Mehmet, and T. Feizi, Eur. J. Biochem., 157 (1986) 365-373. 103. E. F. Hounsell, J. Feeney, P. Scudder, P. W. Tang, and T. Feizi, Eur. J. Biochem., 157 (1986) 375-384. 104. G .H. Cockin, T. N. Huckerby, and I. A. Nieduszynski, Biochem. J.,236 (1986) 921 -924. 105. G. Strecker, J.-M. Wieruszeski, C. Martel, and J. Montreuil, GlycoconjuguteJ.,4 (1987) 329-337. 106. Y. Goso and K. Hotta, Biochem. J.,264 (1989) 805-812. 107. J. P. Kamerling, I. Rijkse, A. A. M. Maas, J. A. vanKuik, and J. F. G. Vliegenthart,FEES, 241 (1988) 246-250. 108. G. Pfeiffer, M. Schmidt, K.-H. Strube, and R. Geyer, Eur. J. Biochem., 186 (1989) 273-286. 109. G. Weisshaar, J. Hiyama, and A. G. C. Renwick, Eur. J. Biochem., 192 (1990) 741 -751. 110. S. Kornfeld, Biochem. SOC.Trans., 18 (1990) 367-374. 11I. R. Schauer (Ed.), The Siulic Acids. Springer-Verlag,New York, 1982. 112. J. B. L. Damm, H. Voshol, K. Hard, J. P. Kamerling, and J. F. G. Vliegenthart, Eur. J. Biochem., 180(1989) 101-110. I 13. D. H. Atha, J.-C. Lormeau, M. Petitou, R. D. Rosenberg, and J. Choay, Biochemistry, 26 (1987) 6454-6461. 1 14. R. U. Lemieux, in K. H. Laidler (Ed.),Frontiers of Chemistry, pp. 3 - 24. Pergamon, New York, 198I. 115. R. U. Lemiuux and K. Bock, Arch. Biochem. Biophys., 221 (1983) 125- 134.

OLIGOSACCHARIDE DETERMINANTS OF GLYCOPROTEINS

349

116. E. F. Hounsell, N. J. Jones, H. C. Gooi, T. Feizi, A. S. R. Donald, and J. Feeney, Curbohydr. Res., 178 (1988) 67-78. 117. P. L. Durette and D. Horton, Adv. Curbohydr. Chem. Biochem., 26 (1971) 49- 125. 118. H. Thprgersen, R. U. Lemieux, K. Bock, and B. Meyer, Can. J. Chem., 60 (1982) 44-57. 119. R. U. Lemieux, T. C. Wong, J. Liao, and E. A. Kabat, Mol. Zmmunol., 2 1 ( 1984)75 1- 759. 120. U. Spohr, 0. Hindsgaul, and R. U. Lemieux, Can. J. Chem., 63 (1985) 2644-2658. 121. U. Spohr and R. U. Lemeiux, Curbohydr. Res., 174 (1988) 21 1-237. 122. M. Biswas and V. S. R. Rao, Int. J. Quantum Chem., 20 (1981) 99- 121. 123. A. Kobata and V. Ginsburg, J. Biol. Chem., 244 (1969) 5496-5502. 124. D. F. Smith, D. A. Zopf, and V. Ginsburg, in Methods in Enzymology, vol. 50, pp. 221 -226. Academic Press, San Diego, 1978. 125. A. Kobata, K. Yamashita, and Y. Tachibana, in Methods in Enzymology, Vol. 50, pp. 2 16- 220. Academic Press, San Diego, 1978. 126. A. S. R. Donald and J. Feeney, Curbohydr. Res., 178 (1988) 79-91. 127. H. Sabharwal, B. Nilsson, M. A. Chester, S. Sjoblad, and A. Lundblad, Mol. Zmmunol., 2 I (1984) 1105-1 112. 128. H. Sabharwal, B. Nilsson, M. A. Chester, F. Lindh, G. Gronberg, S. Sjoblad, and A. Lundblad, Curbohydr. Res. 178 (1988) 145- 154. 129. H. Sabharwal, B. Nilsson, G. Gronberg, M. A. Chester, J. Dakour, S. Sjoblad, and A. Lundblad, Arch. Biochem. Biophys., 265 (1988) 390-406. 130. G. Strecker, J.-M. Wieruszeski, J.-C. Michalski, and J. Montreuil, Glycoconjugute. J.,6 (1989) 271-284. 131. J.-M. Wieruszeski, J.-C. Michalski, J. Montreuil, and G. Strecker, Glycoconjugute, J.,7 (1990) 13-26. 132. N. Platzer, D. Davoust, M. Lhermitte, C. Bauvy, D. M. Meyer, and C. Derappe, Curbohydr. Rex, 191 (1989) 191-207. 133. P. R. Rosevear, H. A. Nunez, and R. Barker, Biochemistry, 21 (1982) 1421- 1431. 134. K. Bock, PureAppl. Chem., 55 (1983) 605-622. 135. Z.-Y. Yan and C. A. Bush, Biopolymers, 29 (1990) 799-811. 136. J. Breg, D. Romijn, J. F. G. Vliegenthart, G. Strecker,and J. Montreuil, Curbohydr.Res., 183 (1988) 19-34. 137. A. D. French and J. W. Brady (Eds.), Computer Modelling of Carbohydrate Molecules, ACS Symposium Series No. 430. American Chemical Society, Washington, D.C., 1990. 138. S. W. Homans, Biochemistry, 29 (1990) 9100-91 18. 139. D. A. Cumming and J. P. Carver, Biochemistry, 26 (1987) 6664-6676. 140. H. C. Gooi, T. Feizi, A. Kapadia, B. B. Knowles, D. Solter, and M. J. Evens, Nature (London),292 (1981) 156- 158. I4 I. E. F. Hounsell, H. C. Gooi, and T. Feizi, FEBS Lett., 131 (198 1) 279 - 282. 142. H. C. Gooi, S. J. Thorpe, E. F. Hounsell, H. Rumpold, D. Kraft, 0.Forster, and T. Feizi, Eur. J. Zmmunol.. 13 (1983) 306-312. 143. H., C. Gooi, E. F. Hounsell, I. Lax, R. M. Kris, T. A. Libermann, J. J. Schlessinger,J. D. Sato, T. Kawamoto, J. Mendelsohn, and T. Feizi, Biosci. Rep., 5 (1985) 83-94. 144. J. Finne, M. E. Breimer, G. C. Hansson, L A . Karlsson, J. Leffler, J. F. G Vliegenthart, and H. van Halbeek, J. Biol. Chem., 264 (1989) 5720-5735. 145. T. Feizi, Trends Biochem. Sci., 16 (1991) 84-86. 146. J. Breg, L. M. J. Kroon-Batenburg,G. Strecker, J. Montreuil, and J. F. G. Vliegenthart, Eur. J. Biochem., 178 (1989) 727-739. 147. B. Bechtel,A. J. Wand, K. Wroblewski,H. Koprowski, and J. Thurin, J. Biol. Chem., 265 (1990) 2028-2037. 148. A. G. Morrell, G. Gregoriades, I. H. Scheinberg, J. Hickman, and G. Ashwell, J. Biol. Chem., 246 (1971) 1461-4. 149. E. F. Neufeld and M. J. Cantz, Ann. N.Y. Acud. Sci., 179 (1971) 580-587.

350

ELIZABETH F. HOUNSELL

150. F. Michon, J.-R. Brisson, and H. J. Jennings, Biochemistry, 26 1987) 8399-8405. 151. R. Yamasaki, Biochem. Biophys. Res. Commun., 154 (1988) 159-164. 152. A. Gottschalk, G. Belyavin, and F. Biddle, in A. Gottschalk (Ed.), Glycoproteins:Their Composition, Structure and Function, BBA Library Vol. 5, Part B, pp. 1082-1096. Elsevier, Amsterdam, 1972. 153. N. K. Sauter, M. D. Bednarski, B. A. Wunburg, J. E. Hansson, G. M. Whitesides, J. J. Skehel, and D. C. Wiley, Biochemistry, 28 (1989) 8388-8396. 154. L. M. Loomes, K.4. Uemura, R. A. Childs, J. C. Paulson, G. N. Rogers, P. R. Scudder, J.-C. Michalski, E. F. Hounsell, D. Taylor-Robinson, and T. Feizi, Nature, 307 (1984) 560-563. 155. D. Roelcke, R. Brossmer, and W. Riesen, Scand. J. Zmmunol., 8 (1978) 179- 185. 156. K. Uemura, D. Roelcke, Y. Nagai, and T. Feizi, Biochem. J., 219 (1984) 865-874. 157. H. Niemann, K. Watanabe, S. Hakomori, R. A. Childs, and T. Feizi, Biochem. Biophys. Res. Commun., 81 (1978) 1286-1293. 158. E. A. Kabat, J. Liao, M. H. Burzynska, C. T. Wong, H. Thngerson, and R. U. Lemieux, Mol. Zmmunol., 18 (1981) 873-881. 159. H. Higashi, M. Naiki, S. Matuo, and K. Okouchi, Eiochem. Biophys. Res. Commun., 79 (1977) 388-395. 160. J.-R. Brisson and J. P. Carver, Biochemistry, 22 (1983) 3671-3680. 161. J.-R. Brisson and J. P. Carver, Biochemistry, 22 (1983) 3680-3686. 162. D. A. Cumming, D. S. Dime, A. A. Grey, J. J. Krepinsky,and J. P. Carver, J.Biol. Chem., 261 (1986) 3208-3213. 163. D. A. Cumming, R. N. Shah, J. J. Krepinsky, A. A. Grey, and J. P. Carver, Biochemistry, 26 (1987) 6655-6663. 164. S. W.Homans,R.A.Dwek,andT. W. Rademacher,Biochemistry, 26(1987)6553-6560. 165. S. W. Homans, Biochemistry 28 (1990) 91 10. 166. A. Imberty, S. Gerber, V. Tran, and S. Perez, Glycoconjugate J., 7 (1990) 27-54. 167. C. J. Edge, U. C. Singh, R. Bazzo, G. L. Taylor, R. A. Dwek, and T. W. Rademacher, Biochemistry, 29 (1990) 1971- 1974. 168. M. Hricovini, R. N. Shah, and J. P. Carver, Biochemistry 31 (1992) 10018. 169. C. A. Bush, V. K. Dua, S. Ralapati, C. D. Warren, G. Spik, G. Strecker, and J. Montreuil, J. Biol. Chem., 257 (1982) 8199-8204. 170. A. J. Duben and C. A. Bush, Arch. Biochem. Biophys., 225 (1983) 1- 15. 171. E. Berman, U. Dabrowski, and J. Dabrowski, 176 ( 1988) 1 - 15. 172. R. L. Brockbank and H. J. Vogel, Biochemistry, 29 (1990) 5574-5583. 173. P. J. Bjorkman, M. A. Shaper, B. Samraoui, W. S. Bennett, J. L. Strominger, and D. C. Wiley, Nature, 329 (1987) 506-518. 174. H. Paulsen, Angew. Chem. Znt. Ed. Engl., 29 (1990) 823-839. 175. R Prohaska, T. A. W. Koerner, Jr., I. M. Armitage, and H. Furthmayr, J.Bid. Chem., 256 (198 1) 578 1-5791. 176. E. J. Welsh, D. Thom, E. R. Moms, and D. A. Rees, Biopolymers, 24 (1985)2301 -2332. 177. G. Pepe, Y. Oddon, D. Sin, J. P. Reboul, and A. A. Pavia, Carbohydr. Rex, 209 (1991) 67-81. 178. A. Bush and R. E. Feeney, Int. J. Peptide Protein Res. 28 (1986) 386-397. 179. N. J. Ringler, R. Selvakumar, H. D. Woodward, I. M. Simet, V. P. Bhavanandan, and E. A. Davidson, Biochemistry, 26 (1987) 5322-5328. 180. T. A. Gerken, K. J. Butenhof, and R. Shogren, Eiochemistry, 28 (1989) 5536-5543. 181. M. Kuwano, T. Seguchi, and M. Ono, J. Cell Science, 98 ( 1991) 131- 134. 182. H. Matsuura, T. Greene, and S. Hakomori, J. Biol. Chem., 264 (1989) 10,472- 10,476. 183. G. W. Hart, R. S. Haltiwanger, G. D. Holt, and W. G. Kelly, Annu. Rev. Biochem., 58 (1989) 841-74. 184. I. A. King and E. F. Hounsell, J. Biol. Chem., 264 (1989) 14022- 14028.

AUTHOR INDEX FOR VOLUME 50 Numbers in parentheses are reference citation numbers. Complete references are listed in numerical order at the end of each chapter in which they are cited. A Abernethy, K., 259(245) Achiwa, K., 21 1(18), 252(214) Ackerman, S. K., 259-260(251) Adachi, S., 317(80), 321(80) Adams, G. A., 217(56) Adelstein, E., 316(64), 321(64), 329(64) Adelt, G., 329(95) Adermann, K., 296(77-78) Adner, N., 230(141) Ahamed, N. M., 230(145), 235(145) Aida, K., 231(155) Akamatsu, Y., 252(217) Akiyama, Y., 220( 103) Alais, J., 73(159), 78(159), 81(159), 84(159) Alekseeva, V. G., 139(59) Aleshin, A., 18- 19(49) Alexeev, Y. E., 136(50-51) Ali, M. H., 143- 144(72) Men, I. E., 259-260(250) Alonso, D., 168(207), 184(207,240) Altmann, K., 218(68), 222-223(68), 225(68), 228(68), 241(68), 251(68), 261(68) Alving, C. R., 21 l(10) Amatore, C., 23(36) Ammann, H., 73( 156), 78( 156), 82( 156) Amvam-Zollo, P. H., 52(109) Anderson, B., 316(59) Anderson, B. L., 196(278) Anderson, D. M. W., 3 1 l(7) Anderson, L., 218(63), 226(63), 244(63), 249(207) Anderson, M. S., 240(191), 262(191) Anderson, R. G., 183- 184(237), 187(237) Anderson, R., 230(141) Ando, T., 16(40,41) Angel, A. S., 313(27) Anisfeld, S. T., 279( 12) Anzeveno, P. B., 43(80) ApU, H.-J., 41(90) Apicella, M. A., 212(22) Aplin, R. T., 154(125) Appelmelk, B. J., 259(244, 244a) Arai, M., 252(217)

Arata, S., 220( 103), 232( 158), 259(247) Arbatsky, N. P., 325(91) Arizuka, M., 49(108) Armitage, I. M., 215(65), 218(63, 65, 66), 222(66), 226(63), 244(63), 249(66), 250-251(65), 263(65), 344(175) Asano, K., 149(95), 150(96-98) Axhauer, H., 211(13), 252(13), 262(13) Ashwell, G., 338(148) Aspinall, G. O., 41(79), 105(181) Atha, D. H., 331(113) Attwood, S. V., 190(254) Auge, C., 305(91) Auling, G., 219(100) Autenrieth-Ansoge, L., 191(264)

B Bacquet, C., 22-23(23) Badoud, R., 183- 184(237), 187(237) Baek, N. I., 104-105(178), 109(178) Baenziger, J. U., 329(95) Baggett, N., 125(4), 13l(32) Baird, P. D., 197(282) Ballesteros, M., 188(253) Bandzouzi, A., 145(77-78) B2r, T., 28(70), 41(70,87,90), 43(87), 49(70) Barbero, G. J., 316(64), 321(64), 329(64) Barbour, A. G., 232(157) Barker, R., 334( 133) Barnickel, G., 212(25), 214(25), 253254(220), 262(220) Barrett, A. C. M., 153(120) Barrett, A. G. M., 153(121), 168(204-205), 189(205), 190(254) Barthels, R., 287(51) Barton, D. H. R., 195(273) Barzilay, I., 215(69), 218(69), 221(69), 223(69), 227(69), 249(69) Basu, S., 241(193) Batley, M., 222-223(109), 226(109) Batllori, R., 188(253) Bauer, C. J., 321(82), 325(82) Baumeister, L., 160(160) Baumgartner, J.-D., 260(253) Bauvy, C., 333( 132)

35 1

352

AUTHOR INDEX, VOLUME 50

Bazzo, R., 343( 167) Beau, J.-M., 110(186), 294(72), 296(72) Bechtel, B., 337( 149) Becker, D. P., 184(242) Bednarik, K., 252(218) Bednarski, M. D., 303(88), 339(153) Beer, D., 157(135-136) Beetz, T., 211(12), 252(12) Behrendt, M., 32(62), 35-36(62), 38(62), 39(78), 41(62), 58(62), 61(62), 11 1(78), 114(78) Belanger, P., 192(266) Belleau, B., 279(17), 283(17) Belyavin, G., 339( 152) Belzecki, C., 190(260), 191(261-263) Bencomo, V., 287(42a, 42b) Bendlin, H., 160(160) Enkchie, M., 195(273) Benedict, C., 129(25) Bennett, R. B., 151(105-106) Bennett, W. S., 343( 173) Bentley, R., 313(19) Berend, G., 127( 17) Berger, H. J., 259-260(250) Bergmann, I. M., 277( 1) Bergmann, M., 14(35), 169(213) Berman, E., 314(35), 343(171) Bernard, N., 329(96), 333(96) Berry, D., 230-231(144) Berven, L. A., 161(172) Beutler, B., 238( 182), 264(182) Bezuidenhoudt, B. C. B., 153(121) Bhat, K. L., 125(2), 182(2) Bhat, U. R., 218(67), 219(67, 98, IOl), 221(107), 227(124, 126), 231(152), 232(101, 160), 247(101, 160) Bhattachanya, S., 151(109) Bhattachajee, A. K., 250(208) Bhattacharjee, S. S., 158- 159(150) Bhatti, T., 3 13(20) Bhavanandan, V. P., 344(179) Biddle, F., 339( 152) Bielefeldt, T., 302(83), 306(95) Bigelow, S. S., 22(28) Bigham, E. C., 129(25) Bigley, D. B., 159(155) Bigorra, J., 188(252) Birken, S., 324(85) Bimbach, S., 294-295(73), 297-298(73)

Bischofberger, K., 146(81), 147(82) Bishop, D. G., 229(134), 234(134) Biswas, M., 333(122) Bjorkman, P. J., 343( 173) Bjomdal, H., 31 l(9) Blache, D., 218(61) Blanz, P., 247(203) Bleha, T., 35(76) Bock, K., 17(44,45), 18(44), 110(187), 134(41-44), 135(45), 158(42, 43), 160(42), 169(210), 173(44,223), 174(41-43,226), 175(226,228), 176(229), 282(25), 288(25), 302(25, 82-83), 306(95-99), 314(34), 327(34), 332(115), 333(115, 118), 334(134) Boekema, E., 151(110) Boel, E., 18(51) Boerekamp, J., 127(15), 182(15) Boersma, A., 316(60-63), 321(60-63), 333(62) Bogard, W. C., Jr., 259(245) Boivin, A., 213(27) Bols, M., 135(46, 47), 136(48) Bommer, R., 28(70), 41(70,92), 47(92), 49(70), 81(92, 164), 85(92) Boons,G. J. P. H., 71(151),73(151), 144(75) Borchardt, R. T., 192(265), 196(265,277, 278) Borcherding, D. R., 192(265), 196(265, 277-278) Borowiak, D., 217(59), 218(91), 219(99), 224(91,99), 225(91), 235(91) Boschetti, A., 140(66) Botta, M., 187(248-249) Bourquelot, E., 16(39) Bouwstra, J. B., 60(126), 63(126), 66,67(126) Boylan, D. B., 230(149) Bradaczek, H., 212(25), 214(25), 253(220222), 262(220-221) Brade, H., 211(5-6, 8, 9a, 9b), 212(6, 2325), 214(25), 218(66a, 67,68,75), 2 19(67), 222,223(68), 224(66a), 225(68, 1 16, 118), 228(68), 229( 132, 133), 231(154), 236(5, 118, 167-170), 238(167, 169, 183, 185), 241(68), 251(68,75,213), 256(6, l18,229d), 257(5, 116, 118, 167, 168,234,239), 258(116, 167-168, 183, 185, 185a,

AUTHOR INDEX, VOLUME 50 241), 259(6,242-244,244a), 261(68, 75, 118,269), 263(185), 264(133, 167, 169, 183,266) Brade, L., 21 l(5, 6,9), 212(6), 225(118), 229(133), 236(5, 118, 170, 172), 243(196), 251(213), 252(196), 256(6, 118), 257(5-6, 116, 118,234,239), 258(116, 118, 172), 259(6,242-244, 244a), 26 1(118), 264( 133) Brady, J. W., 334( 137), 343( 137) Brandenburg, K., 21 1(5), 236(5), 254(224226), 257(5) Brandt, E., 238( 185a) Braum, G., 307(104) Braun, P., 287(36), 303(46, 86), 307(103104) Bredereck, H., 24(38) Breg, J., 316(65-68), 321(65-68), 334(136), 337( 146) Breimer, M. E., 335( 144) Breitenbach, R., 126(1I), 149(11) Brenken, M., 292-293(69) Brennan, P. J., 105(181) Brill, W., 278(10- 1 I), 284(10), 286(10, 11) Brimacombe, J. S., 127(19),311(8) Brinbach, S., 287(44, 52), 288(44), 294, 295(44), 297-298(44) Briner, K., 70( 147), 114(21I), 115(211) Brisson, J.-R., 288(57), 339(150), 343( 160, 161) Broady, K. W., 218(86), 225(86), 228(187), 229(86), 238-240(187), 243(187), 246(86), 247( 187) Brockbank, R. L., 343(172) Brockhausen, I., 58( 129), 60( 129), 66, 67(135), 294(76) Brossmer, R., 339(155) Brown, H. C., 159(155) Bruce, I., 200(286) Bruneteau, M., 218(61), 225-226(121) Bryn, K., 225(119), 229(119), 231(151) Buchanan, J. G., 125(4) Buchholz, M., 287(45,45a), 288(45) Buchner, E., 5(12) Buck, K. W., 131(32) Bulusu, A. R. C. M. Burlingame, A. L., 3 13(24) Burton, A. J., 217(53) Burzynska, M. H., 339(158)

353

Busch, M., 219(101), 232(101, 160), 247(101, 160) Bush, A., 344(178) Bush, C. A., 232(162), 235-236(162), 264(162), 316(56, 58), 334(135), 343( 169- 170), 344(170) Butenhof, K. J., 344(180)

C

Campos-Portuguez, S. A., 219( 101), 232(101), 238(185a), 247(101) Camps, P., 182(234,236), 183(236, 238), 184(238) Cantz, M. J., 338(149) Capek, P., 41(79) Capobianco, J. O., 2 13(26), 262(26) Capon, C., 316(76), 321(76), 325(76), 329330(76) Card, P. J., 190(257) Cardellach, J., 182(235,236), 183(236, 238-239), 184(238,240,241) Cans, R.C. H. M., 167(197),200(197) Carlson, D. M., 322(88) Carlson, R. W., 221(106, 107), 232(106, 160), 247(160), 261(271) Carlstedt, I., 3 13(15) Caroff, M., 214(42,44), 218(93), 222(11I), 238(93) Carpenter, N. M., 154(125), 155(126) Carpenter, R. C., 41(79) Carruthers, R. A., 3 14(45), 3 15(46), 316(75), 321(75), 325(45,46,75), 328(75) Carter, H. E., 217(53) Carver, J. P., 335(139), 343(139, 160- 163, 168) Cashmore, G. C., 313(26), 316(74, 75), 321(74, 75, 82), 323(74), 325(75, 82), 328(74,75) Cavaillon, J. M., 258(240), 264(240) Cenci di Bello, I., 200(286, 288) Cha, J. K., 139(60) Chaby, R., 230-231(139), 242(139), 259(246) Chai, W., 313(26,28), 315(46), 321(82), 323(28), 325(46, 82) Chaki, H., 116(217),252(216)

354

AUTHOR INDEX. VOLUME 50

Chambers, R. E., 313(20) Chan, A., Jr., 161(166) Chandrasekaran, E. V., 329(93) Chapleur, Y., 145(76-78) Charon, D., 211(17), 212(24), 252(17), 259(246) Chattejee, D., 105(181) Chen, L., 214(43), 232(162), 235-236(162), 264( 162) Chen, S.-Y., 125(2, 5,6), 126-127(6), 168(6), 174(6), 182(2), 184(6) Chester, M. A., 333(127- 129) Chia, J., 259(249) Childs, R. A., 322(87), 339(154, 157) Chindemi, P. A., 329(97), 333(97) Chittenden, G. J. F., 126(7), 127(7, 15, 21, 22), 149(22), 167(197), 169(21I), 182(15), 183(21I), 200(197) Chmielewski, M., 190(255,259,260), 191(261-263) Choay, J., 33 I( 1 13) Christensen, M., 306(97) Christensen-Brams,I., 306(98) Christian, R., 218(64,70), 249(70), 250(64), 251(70), 261(70) Chu, D. T. W., 290(65) Chucholowski, A., 2 1(lo), 110(188) Ciommer, M., 291(67), 294-296(67), 302(8 1) Ciucanu, I., 313(12) Clamp, J. R., 3 13(20) Clarke, A. J., 19(52) Clarke, C. T., 169(212) Cockin, G. H., 329-330(104) Coddeville, B., 329(97), 333(97) Coffen, D. L., 196(280) Cohen-Anisfeld,S. T., 306(94) Cole, L. A., 324(85) Cole, R. B., 218(57a, 57b), 228(57a) Coleman, D. L., 256(229c) Collin, W. R., 127(20), 133(20), 199(20) Collins, P. M., 143- 144(72) Conrad, H. E., 31 l(6) Conrad, R. S.,23 I( 152) Copeland, C., 127(14) Corbera, J., 183- 184(238), 188(253) Cornforth, J. W., 4-5(9b), 19(56) Cotter, R., 218(94), 228(94), 241(94) Cotter, R. J., 214(33, 36, 43), 218(88), 222(1 lo), 228(33, 88), 232(88, 162),

234(88), 235(88, 162), 236(162), 242(33), 243(88), 249( 1 lo), 25 1(1 lo), 264( 162) Coward, J. K., 307(106, 107) Cram, D. J., 3(6) Crane, A.M., 105(181) Crawford, T. C., 125(1), 126(9, 1 l), 149(11) Cree, G. M., 164(183),311(7) Creemer, L. J., 43(80) Cretcher, L. H., 151(104) Crum, J. D., 151(106) Csuk, R., 137(53), 138(54), 144(53, 54) Cumming, D. A., 335( 139), 343( 139), 343(162- 163) Czuk, R., 144(73)

D Dabrowski, J., 317(81), 321(81), 324(81), 343(171) Dabrowski, U., 317(81), 321(81), 324(81), 343(171) Dachman, J., 19l(264) Dahler, S. A., 200(286-287) Dakour, J., 333(129) Dalchau, R., 322(87) Dalla Venezia, N., 225-226(121) Damm, J. B. L., 329(112), 333(112) Danbara, H., 231(156) Daniel, J. K., 43(80) Danilov, B., 165(188) Danilova, G. A., 169(208) Dapperens, C. W. M., 169(21I), 183(211) David, C. M., 218(57b) Davidson, D. A., 344(179) Davies, M. J., 3 15(46), 325(46) Davoust, D., 333(132) Dax, K., 168(203) de Bello, I. C., 127(20), 133(20), 199(20) de Bruyn, R. G. M., 167(197),200(197) de Lederkremer, R. M., 148(88,89), 160(158, 161, 162), 163(181), 164(182), 165(191), 166(192- 196), 167(198200), 168(194,202), 169(181), 170(181, 194), 171(182,214-218,220,221), 172(194, 202), 173(194, 200, 222, 224), 174(225), 179(230), 180(231) De Rouville, E., 127(22), 149(22)

355

AUTHOR INDEX, VOLUME 50

Deagostini-Bazin, H., 329(94) Deak, G., 163(180) Dean, B., 99(171) Deferrari, J. O., 148(88) Dejsirilert, S., 25 l(2 1 1) Dell,A., 3 13(23), 324(84), 329(84) Dellinger, R. P., 259-260(250) DeLucca, A. J., 218(57b) Demel, R. A., 255(229) Dempsey, A. M., 148(87) Deprun, C., 214(42,44) Derappe, C., 333( 132) Dervitskaya, V. A., 282(24), 325(91) Devitt, R. E., 116(215) Dhanak, D., 153(120, 121) Dho, J. C., 197(281,282) Diamantstein, T., 264(266) Dienst, C., 317(78), 321(78) Dijong, I., 149(90) Dill, K., 314(35) Dimant, E., 138(55) Dime, D. S.,343( 162) Din, Z. Z., 256(229c) Dinarello, C. A., 236(167), 238(167), 257258(167), 264(167) Diolez, C., 211(17), 252(17) Dipadova, F., 2 1l(9a) Dmitriev, B. A., 218(73, 77), 224(77), 225(73, 77), 226-227(73), 228229(77), 231(77), 242(73, 77), 243245(77), 248(73, 77), 251(73, 77), 264(77) Doerfler, M. E., 263(261) Doherty, D. G., 277(2) Dolle, R. E., 21(10) Dombo, B., 300(79-80), 301(79) Domelsmith, L. N., 218(57a, 57b), 228(57a) Donald, A. S.R., 329(98), 332( 116), 333(98, 116, 126), 335(116) Doniec, A,, 161(169) Doran, C. C., 213(26), 262(26) Dorland, L., 314-315(31, 32), 316(61), 321(61), 325(31, 32,91), 333(31, 32) Draft, D., 335( 142) Drew, M. G. B., 187(251) Drewry, D. T., 218(76) Drozanski, W., 237(177, 178) Du Mortier, C., 166(195), 171(214, 221), 174(225), 179(230) Dua, V. K., 3 16(56, 58), 343( 169)

Dube, V. E., 316(58) Duben, A. J., 343- 344( 170) Dubos, R., 4-5(9a) Dunn, D. L., 259(245) Dupuis, G., 73(156), 78(156), 82(156) Durette, P. L., 333(117) Diirrbaum, I., 236(167), 238(167), 257258(167), 264(167) Diirrbaum-Landmann, I., 238( 185a) Dutton, G. G. S.,31 I( 10) Dutton, G. J., 11l(191) Dwek, R. A., 61(136), 225(120), 314(42), 325(42), 343(42, 164, 167) Dyong, I., 158(151-153), 160(159, 160), 162(178) E

Eby, R., 23(34) Edge, C. J., 343(167) Edwards, J. R., 230(148) Effenberger, G., 114(207,208) Egan, A., 227( 123), 244( 123) Egawa, K., 220( 103), 259(247) Egge, H., 313(13), 317(78, 81), 321(78, 81), 324(81) Eidels, L., 238(181), 264(181) Eitelman, S.J., 147(82), 148(83) El Khadem, H. S., 153(119) El Shenawy, H. A., 155(132) Eliel, E. L., 114(202) Elsbach, P., 263(261) Emmerling, W. N., 152(113) Emoto, S., 163(179) Eng, J., 225(119), 229(119) Engler, R., 329(96), 333(96) Enkelmann, V., 152(114) Emst, M., 238(185a) Erwin, A. L., 214(45), 229(45), 238(45) Esswein, A., 25(47,48), 32(59) Estopa, C., 182(235) Eustache, R. J. Evans, M. E., 259(249) Evens, M. J., 335( 140) Ezaki, T., 251(211)

F Faas, M., 111(197) Fabre, J. W., 322(87) Fairbanks, A. J., 200(286)

356

AUTHOR INDEX. VOLUME 50

Fangmann, R., 317(78), 321(78) Fatah, M. Y., 125(4) Feeney, J., 316(73-75), 321(73-75,82), 323(74), 324(73, 89), 325(73, 75, 82), 328(73-75), 329(98, 103), 330331(103), 332(116), 333(98, 116, 126), 335( 1 16) Feeney, R. E., 344(178) Feist, W., 236(168, 169), 238(183, 185, 185a), 257(168), 258(168, 183, 185), 263( 185), 264( 169, 183,266) Feizi, T., 313(26, 28,29), 314(41), 316(51, 53,59, 73-75), 317(51), 321(53,7375, 82), 322(87), 323(74), 324(73), 325(41, 73,75, 90), 328(73-75), 329331(102, 103), 332(51, 116), 333(116), 335(51, 116, 140-143), 337(51, 145), 338(145), 339(51, 154, 156, 157), 343(41) Fenselau, C., 214(41), 236(41), 239(41), 241(41) Ferguson, M. A. J., 225(120) Fernhndez Cirelli, A., 132(36, 37), 133(39), 148(88, 89), 160(158), 164(158), 165(19l), 166(196), 167(200), 168(202), 17l(216 - 2 18,220), 172(202), 173(200, 222, 224) Ferrari, B., 287(47), 294(47), 294(75) Femer, R. J., 22(20) Fesik, S. W., 215(65), 218(65,66), 222(66), 249(66), 250-251(65), 263(65) Fktizon, M., 165(187) Feudenberg, M. A., 236( 170, 172) Fieser, L. F., 151(109) Fieser, M., 151(109) Fiil, N. P., 18(51) Fink, M. P., 259-260(250) Finne, J., 329(94), 335(144) Finne, V., 329(94) Firsov, L. M., 18- 19(49) Fischer, E., 2(3,4), 4(4), 5(4, lo), 6(18-23), 7(3,20-22), 8(4,27-29), 9- 10(30), 11(31), 12(30-32), 13(30, 32-34), 14(3, 32-36), 16(37), 19(37), 155(130), 157(137, 138) Fischer, W., 215(48) Fisher, C. J., 259-260(250) Fitzsimmons, B. J., 183- 184(237), 187(237) Flad, H.-D., 236(167- 169), 238(167, 169, 183, 185, 185a), 257(167, 168),

258(167, 168, 183, 185,239), 263(185), 264(167, 169,266) Fleet, G. W. J., 127(20), 133(20), 154(122, 125), 155(126), 193(267), 196(279), 197(281-283), 198(285), 198(122), 199(20), 200(286-288) Fletcher, H. G., Jr., 164(186), 165(188, 189) Fluharty, A. L., 160(163) Fomsgaard, A., 259-260(252) Font, J., 168(207), 182(234-236), 183(236, 238-239), 184(207,238,240-241), 185(243), 186(245), 188(252-253) Ford, C., 19(53) Ford, M. J., 116(214) Foridis, C., 329(94) Forster, M. O., 8(26) Forster, O., 335(142) Foster, A. B., 131(32), 217(55) Foulke, G. E., 259-260(250) Fournet, B., 313(16), 316(76), 321(76), 325(76), 329-330(76) Fowler, V. J., 247(203) Fox, G. E., 247(203) Fracsch, C. E., 218(73), 225-227(73), 242(73), 248(73), 251(73) Franken, G. A. M., 169(21 l), 183(21 1) Fraser-Reid, B., 116(215) Frasman, G.D., 282(26), 288(26), 302(26) French, A. D., 334( 137), 343( 137) Frenkiel, T. A., 324(89) Freudenberg, K., 2(5), 15(5) Freudenberg, M. A., 21 1(3), 215(47), 241(47), 243( 196), 244(3), 247(234), 252(172, 196), 256-257(3), 258(241), 259-260(252), 264(265) Frick, W., 114(204) Friedrich-Bochnitschek,S., 11I( 194), 291(68), 294(68), 297(68) Friffiss, J. M., 2 12(22) Fromme, I., 224( 114) Frush, H. L., 157(143), 158(146) Fuganti, C., 159(154) Fugedi, P., 22(18), 23(35), 35-37(71), 60( 124), 105( 124), 108(124), 248(204), 290(63b) Fuhrhop, J.-H., 151( 1 10) Fujimoto, K., 1 lO(184) Fujishima, Y., 211(11, 16), 252(11, 16) Fujita, M., 211(16), 252(16) Fukase, K., 211(15), 215(15), 252(15),

AUTHOR INDEX, VOLUME 50

258(15), 263( 15), 306(99) Fukuda, M., 324(84), 325(90), 329(84) Fukuoka, S., 218(95), 228(95), 243(95) Funaki, Y.,49(105) Funakoshi, I., 317(77), 321(77) Fiirstner, A., 194(272) Furthmayr, H., 344( 175) Furuhata, K.,138(56-57)

G Gaden, H., 28(61), 30(56), 32(56, 61) Gage, D. A., 261(271) Galanos, C., 21 l(1-5), 213(2,4),214(35), 2 15(47), 2 17( 1, 35), 218(1,4,64a), 224(64a), 225(4), 228(4, 35), 231(152), 231(154), 233(164), 234(1, 164), 235(164), 236(5, 170-172), 237(178), 241(47), 243( 196), 244(3), 246(4), 247(4, 35), 249(35), 252(35, 172, 196), 256(1, 3,4,230),257(1, 3-5,230,232, 234), 258(241), 259-260(262) Galbraith, L., 234(163) Galkowski, T. T., 157( 143) Gall-Istok, K., 163(180) Gammon, D. W., 105(181) Ganguly, A. K.,162(175-177) Garcia, F., 261(271) Garegg, P. J., 22(18, 24), 23(35), 126127(8), 290(63b) G-, H. G., 278-279(6), 283(28-30), 293(70), 306( 102) Gasiecki, A. F., 153(121) Gateau, A., 129- 130(27), 198(284) Gaudemer, A., 198(284) Gautheron, C., 305(91) Gerber, S.,343(166) Gerdes, J., 236( 169),238(169), 264(169,266) Gerisch, G., 237(179) Gerken, T. A., 344( 180) Gero, S. D.,129- 130(27), 198(284) Gerwig, G. J., 313(2 1) Geyer, H.,314(40), 325(40) Geyer, R., 314(40), 325(40), 329(108) Gibson, B. W.,212(22) Giesbrecht, P.,212(25), 214(25), 252(219), 253(220), 254(2 19,220),262(220) Ginocchio, S. D., 161(166) Ginsburg, V., 333(123- 124)

357

Girard, R., 259(246) Giudici, T. A., 160(163) GIBnzer, B. I., 137(53), 138(54), 144(53, 54, 73) Glattfeld, J. W. E., 147(144), 151(99) Glittenberg, D., 158(153) Gmeiner,J., 217-218(54), 223(112), 249( 112) Gnichtel, H., 191(264) Godefroi, E. F., 127(15), 167(197), 169(211), 182(15), 183(211), 200( 197) Goldman, R. C., 213(26), 262(26) Goldstein, I. J., 151(111, 112) Golenbock, D. T., 238( 184), 264( 184) Golfier, M., 164(185), 165(187) Golubev, A., 18- 19(49) Goodwin, S., 229(135) Gooi, H. C.,316(73), 321(73), 324325(73), 328(73), 332-333(116), 335(116, 140- 143) Gorbach, V. I., 218(62) Gorbies, L., 303(84b) Gorelick, K., 259-260(251) Goso, Y.,329(106) Gottert, H.,229(128), 236(128) Gottschalk,A., 339(152) Grabarek, J., 213(31) Graczyk, G., 282(26), 288(26), 302(26) Grage-Griebenow,H.,238(185a) Gragg, C.E., 129(25) Grasselli, P.,159( 154) Gratzl, J. S., 165(190) Gray, G. W.,218(76) Greck, M., 218(82), 230(82), 232(82) Green, E. D., 329(95) Greene, T., 345(182) Gregoriades, G., 338( 148) Greilich, U.,53(113), 55-56(117), 57(113, 117),95(113,117), lOO(113, 117), 101(117) Grey, A. A., 343(162, 163) Grief, V., 143(69) Grimmecke, H.-D., 21 l(9b) Gronberg, G., 333(128, 129) Gruezo, F. G., 316(55) Grundler, G., 68-69(137, 137a, 140), 71(137), 72(140), 73(137, 140), 78(140), 80(137), 81(137, 140), 8283( 140), 93(137), 94(137, 170),

358

AUTHOR INDEX, VOLUME 50

98(137), 99(137, 170), lll(192, 193), 113(192, 193) Guelde, G., 259(249) Guilleman, R., 287(50), 299(50) Guilliard, W., 11 l(198) Giinther, W., 281(23), 302(81) Gupta, R., 247(203) Gutberlet, T., 253(221), 262(221) Guthrie, R. D., 136(49) Guze, L. B., 230(140) Gzernecki, S., 139(62)

H Haas, M., 214(40) Haeffner, N., 230-231(139), 242(139) Haeffner-Cavaillon,N., 258(240), 264(240) Hafele, B., 194(270-271) Hagelloch, G., 24(38) Hahn, C. M., 247(203) Haishima, Y., 261(268) Hajime, I., 21 l(18) Hakomori, S., 98(174), 99(171), 31 l(4-5), 316(52), 325(90), 339(157), 345(182) Hall, C. L., 238( 180) Hall, R. H., 146(81), 147(82), 148(83) Hall, W. R., 129(25) Haltiwanger, R. S., 345( 183) Hamadeh, R., 2 12(22) Hamako, J., 314(41), 325(41) Hambsch, E., 24-25(38) Hampton, R. Y., 238(184), 264(184) Han, O., 171(219) Han, X.-B., 105(183), 110-lll(183) Hancock, R. E. W., 263(258) Hanessian, S., 22-23(23), 141(67), 181(233), 186(246, 247), 187(248,249) Hanisch, F.-G., 317(78, 81), 321(78, 81), 324(81) Hannigan, J., 259-260(251) Hansen, E. J., 238(181), 264(181) Hansen-Hagge, T., 236(170- 171, 173), 243(173), 244(173, 199) Hansson, G. C., 313(15), 335(144) Hansson, J. E., 339(153) Hara, H., 218(88), 228(88), 232-235(88), 243(88) Harada, K., 257(236) Haraldsson, M., 127(20), 133(20), 199(20)

Harangi, J., 60(125) Harbin, A.-M., 314(45), 325(45) Hard, K., 314(36), 329(112), 333(112) Hardy, M. R., 3 13(22) Harlan, J. M., 238(186) Harrata, A. K., 218(57a), 228(57a) Hart, G. W., 345(183) Hw, S., 217(57-58), 218(57, 80,85), 225227(85), 231(85) Hasegawa, A., 41(93-95), 47(93, 94), 58(120, 123),211(11, 16, 19),252(11, 16, 19) Hasenkamp, T., 293(70) Hashimoto, H., 149(94,95), 150(96) Hashimoto, S., 21(1 I), 116(220-223), 152(116) Hauser, H., 255(229) Haus, H., 127(18) Hawiger, J., 213(31) Haworth, W. N., 162(173) Hay, R. W., 22(20) Hayashi, J. A., 230(148) Hayashi, M., 21( 1I ) Hedenburg, 0. F., 148(84) Heermann, D., 11l(198) Hehre, E. J., 19(54) Heine, H., 238(185, 185a), 258(185), 263( 185) Heinz, E., 28(58), 30(58), 105(58a) Helander, A., 288(57) Helander, I. M., 2 18(68),222-223(68), 225(68), 227( 125), 228(68), 229( 136), 241(68), 243(195, 195a), 251(68), 26 l(68) Helferich, B., 14(36) Helfrich, W., 151(110) Heller, D., 214(41), 236(41),239(41), 24 l(4 1) Hellerqvist, C. G., 3 1 l(9) Helpap, B., 44(82), 60(128) Hemberger, J., 54(114), 78-79(114), 81(114), 85(114) Hendriks, K. B., 280(20) Hengeveld, J. E., 143(69) Henning, D., I55( 131) Henrichson, C., 22(24) Henricson, B. E. Her, G.-R., 213(31) Hermann, P., 303(87) Hermansson, K., 314(37)

AUTHOR INDEX, VOLUME 50 Herp, A., 316(54), 329(54) Herzbeck, H., 238(183), 258(183), 264(183, 266) Heume, M., 62(133) Hewett, M. J., 229( 134), 234( 134) Heymann, H., 151(109) Hickman, J., 338(148) Higashi, H., 339( 159) Hignett, R. C., 23 1(153) Hildebrandt, J., 211(13), 252(13), 262(13) Hilgenfeld, R., 152(114) Hindahl, M. S., 234( 163) Hindsgaul, O., 333(120) Hiraoka, T., 252(217) Hirata, Y., 151(109) Hirayama, T., 232(158) Himhberger, J., 8(27,28) Hirvas, L., 243( 195) Hisatsune, K., 218(83), 251(83), 261(83, 268), 264(264) Hiyama, J., 329(99, 109) Hjort, I., 18(51) Hoch, M., 28(58a), 30(58a), 73(155), 79( 155), 81( 155), 105(58a) Hoesch, K., 1(1),4(1) Hoffmann, M., 11I( 198-200), 114(199, 205 -206) Hoffmann, R. W., 185(244) Hofstad, T., 2 18(80) Hogenkamp, P. C., 132(35) Holland, P. T., 313(24) Hollingsworth, R. I., 221(106, 107), 232(106), 261(271) Hollosi, M., 282(26,27), 288(26), 302(26, 27), 303(84a) Holly, F. W., 195(274) Holst, O., 211(9b), 212(23, 24), 218(66a, 91), 224(66a, 91), 225(91, 116), 235(91), 251(213), 257-258(116), 261(270) Holt, G. D., 345(183) Holt, N. B., 157(143) Holt, S. C., 218(94), 228(94), 241(94) Homans, S. W., 334(138), 343(138,164,165) Homma, J. Y., 211(19), 252(19), 257(235, 237) Honda, S., 313(14, 17), 345(17) Honda, T., 116(220-223) Honeyman, J., 136(49) Honovich, J., 214(33, 36), 218(88),

359

222(1lo), 228(33, 88), 232-235(88), 242(33), 243(88), 249(1 lo), 251(110) Honzatko, R. B., 18- 19(49) Hoogerhout, P., 180(232) Horito, S., 149(94-95), 150(96-98) Horton, D., 111(201), 142(68), 163(181), 169- 170(181),217(55), 229(130), 333(117) Hoshi, Y., 151(107) Hosie, L., 161(171) Hotta, K., 329(106) Hough, L., 129(24), 148(87) Hounsell, E. F., 312(3), 313(25,26,28), 314(33, 45), 315(46), 316(53, 73-75), 321(53,73-75, 82), 322(87), 323(28, 33, 74), 324(73,89), 325(45,46,73,75, 82,90), 327(33), 328(73-75), 329331(102, 103), 332(3, 33, 116), 333(33, 116), 335(116, 141-143), 339(154), 345(184) Houvenaghel, M.-C., 316(65), 321(65) Howell, A. R., 153(121) Hricovini, M., 343(168) Hubschwerlen, C., 131(34) Huckerby, T. N., 329-330( 104) Hudson, C. S., 1(2), 5(14), 151(101) Hudson, Y., 236( 175) Hull, R., 152(1 18) Hull, W. E.,316(61), 321(61) Hultenby, K., 255(229b) Hulyalkar, R. K., 127(12), 161(12) Hunnemann, D. H., 230(147) Hurlbert, R. E., 224( 1 14) Husaain, A. A., 30(54), 70(54) Hutson, D., 11l(201) Hyver, K., 214(33), 228(33), 242(33)

P Ibrahim, I. H., 105(181) Ignatenko, A. V., 218(74) Iida, H., 201(289) Iijima, H., 58(121), lOl(172, 173), 102(121) Iimura, Y., 169(209) Iitaka, Y., 138(58) Ijima, H., 290(6 l), 292(66), 294(66), 296(66) Ikeda, K., 252(214) Ikeda, S., 211(19), 252(19) Ikeda, T., 257(236) Ikegami, S., 116(220-223)

AUTHOR INDEX, VOLUME 50

360

Ilancock, R.E. W., 218(77a) Imai, K., 219(99), 224(99) Imberty, A., 343(166) Imoto, M., 21 1(14), 214(34, 35),215(14),

217(35), 218(34), 228(35), 236(170), 244(200), 247(35), 249(34,35), 250(34), 251(34), 252(14,215,216), 257(234,236,237),261(34) Inage, M.,116(217), 252(216) Inomata, K., 130(28) Inoue, Y., 249(206) Ireland, R.E., 183-184(237), 187(237) Isakov, V. V., 218(62) Isbell, H. S., 157(143), 158(146), 162(174) Isemura, S., 160(164- 165) I~hida,H., 41(93-95), 47(93,94), 58(120), 21 1(16), 252(16) Ishida, N.,252(217) Ishidate, M., 126(10) Ishikawa, K., 218(95), 228(95), 243(95) Ito, Y.,41(91),43(91), 46(91),49(100, 101, 104),50(100, IOl), 51-52(104), 58(123), 60(91, 127), 62-63(130), 66(127), 67(9 I), 68- 69( 141- 144), 70(141-142), 73(130, 143,144,157, 158),78(130, 143,157,158), 81(130, 157), 82(130), 83(157), 89(127, 142), 110(189), 1 ll(195) Itoh, T., 127(16), 139(16), 232(158), 317(80), 321(80) Iwashita, T.,168(206), 214(35), 217(35), 228(35), 247(35), 249(35), 252(35) Iyer, R.N., 322(88)

Jantzen, E., 225(119), 229(119), 232(159,

161) Jarglis, P., 190(258) Jarvis, G. A., 212(22) Ja@,J., 158(149) Jatzke, H., 30(56), 32(56) Jaurand, G., 1 lO(186) Jay, F., 258(241) Jeanloz, R.W., 278-279(6), 279(13, 18), 282( 18), 283(28-3 1) Jenkins, S. R.,195(274) Jennings, H.J., 250(208), 339( 150) Jensen, K. J., 287(49), 294(49), 299(49),

302(49) Jensen, M., 217(59), 218(71), 227(125) Jeroncic, L.O., 132(36-37), 167(200, 201),

173(200,224) Jersch, N.,I58( 152), 160( 159) Jetten, M., 52-53( 1 I l), 57-58( 11 1) Jiao, B., 215(47), 241(47) Jin, H., 22(25,26) Johansen, P.G., 277(3) Johnson, R.S., 213(31) Johnson, W. C., 3I l(2) Johnston, D. B. R., 286(38) Jones, A. H., 200(286) Jones, B. D., 31 l(8) Jones, J. H.,169(212) Jones, J. K. N., 127(12), 129(24), 161(12),

287(41) Jones, N. J., 332-333(116), 335(116) Jordaan, A,,146(81), 147(82), 148(83) Joullit, M.M., 125(2,5 , 6), 126-127(6),

168(6), 174(6), 182(2), 184(6) J Jackson, P., 313(18) Jacob, G. S., 198(285) Jacquinet, J. C., 93- 94( 169), 98(169), 1 1 I( 169), 1 13( 169), I 16(213), 294(72),

296(72) Jager, K.-E., 218(77), 224-225(77), 228-

229(77), 231(77), 242-245(77), 248(77), 251(77), 264(77) Jager, V., 194(270,271) James, K., 280(20) Jann, K., 213(30), 217(30), 230(30), 250(209) Janson, A. M., 302(82) Jansson, P.-E., 314(37)

Juaristi, E., 114(202) Jung, K.-H., 73(155), 79(155), 81(155),

111(197) Jurczak, J., 190(255)

K Kabat, E. A.,316(37, 50,55,59), 325(57),

333(119), 339(158) Kaca, W., 227( 126) Kadam, S. K., 213(26), 262(26) Kakehi, K., 313( I7), 345( 17) Kalamnn, J.-D., 288(58) Kalsner, I., 314(40),325(40) Kamerling, J. P.,52-53(111), 57-58(111),

60( 126), 63( 126), 66-67( 126), 105( 179,

AUTHOR INDEX, VOLUME 50 180), 109(179, 180), 313(21), 329(107, 112), 330(107), 333(112) Kameyama, A., 58(120) Kamikawa, T., 49( 108), 2 1 I( 14, 15), 215(14, 15), 252(14-15), 258(15), 263( 15) Kamio, Y ., 224( 115) Kamishima, H., 218(95), 228(95), 243(95) Kampf, A., 138(55) Kandil, A. A., 193(268) Kane, D. P., 316(74-79, 321(74-75), 323(74), 325(75), 328(74-75) Kanegasaki, S., 257(237) Kapadia, A., 335(140) Karabinos, J. V., 157(143) Karibian, D., 214(42,44) Karjala, S., 157(139) Karl, H., 1 lO(185) Karlsson, H., 3 13( 15) Karlsson, K. A., 3 1I( 1 I), 335( 144) Karube, I., 218(95), 228(95), 243(95) Kasahara, K., 201(289) Kasai, N., 220(103), 232(158), 257(236), 259(247) Kastowsky, M., 253(221), 255-256(228), 262(221, 228) Kates, M., 49( 108) Katha, K. P. R., 58(120) Kato, M., 314(41), 325(41) Kauth, H., 287(53) Kawahara, K., 218(75), 231(155, 156), 251(75,211,212), 261(75) Kawai, Y., 230(146) Kawamoto, T., 333143) Keilich, G., 220( 104) Kelly, W. G., 345(183) Kelsey, J. E., 129(25) Kenne, L., 288(57), 314(37) Kennedy, J. F., 34(64), 49(64), 58(64), 60(64), 84(64), 105(64) Kenny, C. P., 250(208) Kerek, F., 313(12) Kerekgyarto, J., 60(126), 63(126), 6667(126) Kern, M., 252(218) Khare, N. K., 105(181) Khorana, H. G., 215(46,69), 2 18(69), 221(69), 223(69), 227(69), 239(189, 190), 249(69) Khorlin, A. Y., 281(22)

36 1

Khourdanov, C. A., 136(50) Khuong-Huu, F., 195(273) Kibayashi, C., 201(289) Kickhofen, B., 237( 179) Kida, T., 128(23) Kilganiff, C., 259(245) Kiliani, H., 6( 16, 17) Kim, J. J., 212(22) Kim, K. C., 224( I 15) King, C. H., 43(80) King, I. A., 345(184) Kinoshita, M., I lO(184) Kinsho, T., 49( 106) Kinzy, W., 30(57), 37(72), 48(57), 54( 114), 55(72, 115, 116), 58(115), 68(145, 149), 70(145, 149), 71(149-150), 73(145, 149, 160), 74(145, 146, 150, 154), 75(57), 76(149), 78(114), 79(114, 115), 80(160), 81(114, 115, 154, 160, 162, 164), 84(57, 149), 85(114, 115, 154, 162), 90-91(57), 93(149), 94(116), 9697(150), 98(116, 149), 99(116, 150), lOO(115, 116, 154), 103(I50), 104(162), 107(162), 288(60) Kirby, A. J., 114(203) Kirikae, F., 256(229d), 264(264) Kirikae, R., 238(185a) Kirikae, T., 21 1(9a, 9b), 256(229d), 264(264) Kirio, Y., 144(74) Kirkland, T. N., 232(162), 235-236(162), 238(182), 259(248), 264(162, 182) Kishi, Y., 139(60) =SO, M., 41(93-95), 47(93-94), 58(120, 123), 21 l(11, 16, 19), 252(11, 16, 19) Kitagawa, I., 104-105(178), 109(178) Kitajima, T., 62(131), 63(132), 68-70(141), 82( 132) Kitamura, H., 257(236) Klager, R., 28(69), 49(98-99), 50(98), 52(69), 114(69) Klein, A., 316(62, 66-68), 321(62, 66-68), 333(62) Kleppe, K., 16(42) Klosterman, M., 25(43) Klotz, W., 25(44, 49) Knapp, W., 230( 145), 235(145) Knatterud, G. L., 259-260(250) Knollmann, R., 158(15I), 160(159) Knorr, E., 21(8) Knowles, B. B., 335(140)

362

AUTHOR INDEX, VOLUME 50

Knox, K. W., 229(134), 234(134) Kobata, A., 62(130- 131), 63(130), 73(130), 78( 130), 8 1- 82( 130), 3 14(43), 3 15(47), 325(43,47), 333(123, 125) Kobayashi, M., 68-69(143-144), 73(143, 144), 78(143) Kobayashi, Y., 252(217) Koch, M. H. J., 254(224,226), 255(229a) Kochetkov, N. K., 282(24), 314(38), 325(91) Koenigs, W., 21(8) Koerner, T. A. W., Jr., 344( 175) Kohn, B. D., 127(13), 195(276) Kohn, P., 127(13), 160(156, 157), 161(166), 171(156), 195(275,276) Koike, K., 49-50(100), 56(118), 58(122) Kokx, A. J. P. M., 167(197),200(197) Koles, N. C., 259(249) Kollat, E., 282(27), 302(27) Komatzu, S., 151(101) Komura, H., 168(206) Kondo, H., 252(2 14) Kondo, S., 218(83), 251(83), 261(83,268), 264(264) Konings, J. J. H. G., 167(197), 200(197) Konradsson, P., 1 16(215) Kontrohr, T., 261(270) Koppen, P. L., 329( loo), 333( 100) Koprowski, H., 337(147) Korczynski, A., 16 1( 169) Kornfeld, S., 329( 1 lo), 338( 110) Korol, E. L., 139(59) KO&, S., 49(108) KoXh, W., 287(40), 300-301(79), 302(40, 81) Kosma, P., 218(64), 250(64), 251(211-213) Kostenbander, H. B., 30(54), 70(54) Kosunen, T. U., 229( 136) Kotani, S., 211(14), 215(14), 252(14), 257(236,238) Koto, S., 22(29) Kovacs, J., 248(204) Kowalczyk, D., 307( 104) Kraaijeveld, N. A., 43(81) Krasikova, I. N., 2 18(62) Kraska, U., 239(188) Krauss, G. A., 139(61) Krauss, J. H., 215(50), 218(89), 219(50,98, loo), 225 -226(89), 228(89), 232(50), 233(89), 238(89, 185a), 243(89), 264(265)

Krepinsky, J. J., 62(134), 343(162, 163) Kreuzer, M., 2 1( 13) Kris, R. M., 335(143) Krohn, K., 21(7) Kroon-Batenburg, L. M. J., 337( 146) Kropinski, A. M. B., 230-231(144) Kuge, S., 252(217) Kuhn, H.-M., 259(243,244,244a, 244b) Kuhn, R., 3 13(13) Kiihne, W., 8(24) Kulshin, V. A., 218(73, 77), 224(77), 225(116), 225(73, 77), 226-227(73), 228-229(77), 231(77), 242(73, 77), 243-245(77), 248(73, 77), 25 l(73, 77), 257-258(116), 264(77) Kumada, H., 261(268) Kumazawa, Y., 257(235,237) Kundt, I., 173(223) Kung, P. C., 259(245) Kunz, H.,21(6, 12), lll(194, 196),278(7, 10-11), 279(7, 14-16), 280(19a-l9c), 281(21,23), 283(14-16, 19a-l9c, 3237, 39), 284(10, 15, 32-36), 285(19a19c, 34, 36), 286(7, 10, 11, 19a-l9c, 37, 39), 287(16, 39-40,43-45,45a, 46, 51 -53), 288(43-45, 54), 289(32), 290(62), 291(67,68), 293(71), 294(19a19c, 32, 33, 39, 43, 44, 67, 68, 73), 295(67, 73), 296(67), 297(44,68, 73), 298(44, 73), 299( 19a-l9c), 300(79,80), 301(32,79), 302(40, 81), 303(46,8687), 304(90), 306(93, 100, 101), 307(103, 104, 108) Kurosaka, A., 317(77), 321(77) Kurosawa, M., 2 1 1(15), 2 15( 15), 252( 15), 258(15), 263(15) Kurtz, R., 234( 166) Kurunaraine, D. N., 218(77a) Kusama, T., 21 1(15),215(15), 238(185), 252(15), 256(229d), 258(15, 185), 263( 15), 264( 185) Kusumoto, S., 116(217),21 1(5,9b, 14, 15), 214(34), 215(14, 15), 218(34), 225(118), 236(5, 118, 167- 170, 172), 238(167, 169, 183, 185, 185a), 243( 196), 244(200), 249(34), 250(34), 251(34), 252(14, 15, 172, 196,215, 216,219), 254(219), 256(118,229d), 257(5, 118, 167-168,233,234,236, 237), 258(15, 118, 167, 168, 183, 185,

AUTHOR INDEX, VOLUME 50

239), 259(243,244,244a), 261(34, 118), 263(185), 263(15), 264(167, 169, 183,266), 306(99) Kusunose, N., 21 1(14), 215(14), 252(14) Kuwano, M., 344( 181) Kuzuhara, H., 163(179) L Labischinski, H., 21 1(5), 212(25), 214(25), 236(5), 252(219), 253(220,222), 254(219-220), 255 -256(228), 257(5), 262(220,228) Lachut, C. H., 125(4) Lackzo, I., 282(27), 302(27) Lacombe, J. M., 287(48), 294(48) Laczko, I., 303(84a) Ladduwahetty, T., 110(188) Ladner, W.,185(244) Laine, R. A., 218(57b) Lakshmi, S. K. B., 227( 124) Lam, C., 211(13), 252(13), 262(13) Lamblin, G., 316(60-63,65-68), 321(6063,65-68), 333(62) Lancelin, J. M., 139(64) Lansbury, P. T., Jr., 279( 12), 306(94) Larsson, L., 225(119), 229(119, 135), 230(141) Lartey, P. A., 143(69) Lauffenburger, M., 324(84), 329(84) Lavielle, S., 287(50), 299(50) Lawson, A. M., 313(25-26,28), 314(45), 315(46), 316(73-75), 321(73-75, 82), 323(28,74), 324(73), 325(45,46,73, 75, 82), 328(73-75), 329-331(102) Lax, I., 335(143) LeBeyec, Y., 214(44) Lederer, E., 81(166, 167) Lederkremer, R. M., 132(36-38), 133(39), 164(158), 18l(38) Lee, A. C., 153(121) Lee, C. M., 143(69) Lee, J., 307( 106, 107) Lee, J. B., 161(167) Leffler, J., 335(144) Lefrancier, P., 81(166, 167) Lehmann, V., 21 1(3), 223(113), 226(113, 122), 227(113, 122, 123), 236(170, 171, 173), 243(173), 244(3, 122, 123, 173, 197, 199), 256-257(3)

363

Lehong, N., 22-23(23) Lemieux, R. U., 17(44,46,47), 18(44), 35(75), 58(75, 152), 157(137a),280(20), 332(114-115), 333(114, 115, 118121), 339(158) Lemoine, J., 3 13(16) Lerner, L. M., 127(13), 130(29), 137(52), 158(148), 160(156, 157), 161(166), 171(156), 195(275,276) Leroy, Y., 313(16), 316(76), 321(76), 325(76), 329-330(76) Lewis, M. D., 139(60) Lewis, V., 230-231(144) Ley, S. V., 116(214) Lhermitte, M., 316(60-61, 63, 68), 321(60, 61,63,68), 333(132) Li, C., 256(229c) Liao, J., 316(55), 333(119), 339(158) Libby, P., 238(185a), 264(265) Libermann, T. A., 335( 143) Lichtenthaler, F. W., 190(258) Lico, I. M., 290(65) Liebig, J., 5( 11) Likhoshersov, L. M., 282(24) Lindberg, A. A., 218(68), 222-223(68), 225(68, 117), 228(68), 234(117), 241(68), 251(68), 255(229b), 261(68) Lindberg, B., 31 l(9) Lindgren, K., 229( 135) Lindh, F., 314(37), 333(128) Lindner, B., 21 1(9), 212(25), 214(25, 34, 37-40), 218(34, 66a, 68, 73, 77, 84, 92), 222-223(68), 224(66a, 771, 225(68, 77,92), 226(73, 92), 228(37, 68, 73, 77, 84,92), 229(38, 77), 230(37), 231(38, 77), 234(38), 235(37), 236(173), 239(38), 240(37-39), 241(37, 68,92), 242(92), 242(38,77, 84,92), 243(37, 77, 173, 195a), 244(77, 173), 245(77), 248(77), 249(34, 73), 250(34), 251(34,68,73,77), 261(34,68), 264(77) Ling, N. C., 287(50), 299(50) Lingens, F., 221( l05), 230(105), 252(105) Link, K. P., 157(139), 277(2) Lipka, G., 255(229) Lipkind, G. M., 314(38) Lipshutz, B. H., 129(26) Liptik, A., 60(124- 126), 63(126), 6667(126), 105(124), 108(124)

364

AUTHOR INDEX, VOLUME 50

Litter, M. I., 164(182), 166(193), 167(198199), 171(182) Liu, H., 171(219) Liu, P. S., 43(80) LoBuglio, A. F., 259-260(250) Loibner, H., 211(13), 252(13,218), 262(13) Lomax, J. A., 218(76) Long, C. W., 162(173) Long, G. W., 143- 144(70) Lijnn, H., 22(18), 23(31-33), 290(63a) homes, L. M., 339(154) Loppnow, H., 21 1(9a), 236(167), 238(167, 185a),257(167), 258(167,239), 264( 167,265) Loretzen, J.-P., 44(82) Lorkiewicz, Z., 246(202) Lormeau, J.-C., 331(113) Liiw, A., 81(162), 85(162), 104(162), 107(162) Liideritz,0.,211(1-5),213(2,4,28, 30), 214(35, 38), 217(1,28, 30,35, 54), 218(1,4, 54, 78, 81, 86), 221(81), 223(112), 225(4, 86), 228(4, 187), 229(38, 86, 128), 230(30,81, 142), 231(38, 142), 234(1, 38, 78), 235(142), 236(5, 128, 170- 172), 237(177- 179), 238(81, 187), 239-240(38, 187), 242(38), 243( 187, 196), 244(3, 199), 246(4, 78,86,201), 247(4, 35, 187), 249(35, 112), 252(35, 172, 196), 256(1, 3-4,230), 257(1, 3-5,28,230, 232, 234) Liideritz, T., 236(170) Lugtenberg B., 230( 143), 235( 143) Lui, S. C., 316(73), 321(73), 324-325(73), 328(73) Lundblad, A., 333( 127- 129) Lundt, I., 110(187), 134(41-44), 135(4547), 136(48), 158(42,43), 160(42), 169(210), 173(44), 174(41-43,226227), 175(226,228), 176(229) Liining, B., 288(56), 294(56) Luzzati, V., 254(223), 255(227,227a, 229a), 256(230-23 I), 257(230), 259(243) L'vov, V. L., 218(74)

M Maas, A. A.M., lOS(179, 180), 109(179, 180), 329 - 330( 107)

Macharadze, R. G., 281(22) Macher, I., 21 1(13), 252(13), 259(243), 262( 13) Maciejewski, S., 190(259) Mackie, D. M., 164(183) MacMillan, D., 151(99) Madigan, M. J., 313(25) Madigan, M. T., 247(203) Maeta, H., 21(14-16) Magnus, P., 184(242) Mai, L. A., 157(141) Maier, T., 41(89) Mairanovskii, V. G., 161(168) Maitra, S. K., 230(140) Makela, P. H., 227(125) Malcita, M., 313(19) Malchow, D., 237(179) Malek, G., 279(17), 283( 17) Maley, F., 3 14(44),325(44) Mallet, J.-M., 23(36) Mallow, W. R., 129(25) Maltby, D., 3 13(24) Mamat, U., 21 1(9a, 9b) Mandrell, R. E., 212(22) Mann, J., 187(250,251) Maquenne, L., 144(129) Marino, C., 132(38), 160(162), 166(194), 168(194), 170(194), 172-173(194), 180(23l), 18l(38) Mark, E., 133(40) Marks, G. S., 277(4) Maron, L., 126- 127(8) Mama, A., 23(36) Marre, R., 232( 159) Marsh, W. L., 316(59) Marshall, R. D., 277(3, 5), 278(5) Martel, C., 329(105) Marx, A., 23 1(152) Man, J., 279(15), 281(21), 283(15, 33, 39), 284(15, 33), 286(39), 287(39,40), 294(39), 297(33), 302(40, 8 1) Masato, M., 41(91), 43(91), 46(91), 60(91), 67(91), 11 1(195) Mascagni, P., 214(33), 217(60), 222(1 lo), 228(33,60), 236(172), 242(33), 243(172), 248(60, 172), 249(110,207), 251( 110) Mashimo, J.-L,220( 103), 257(236), 259(247) Masoud, H., 2 18(84,92,94), 2 19(98), 225 226(92), 228(84,92,94), 241(92,94),

AUTHOR INDEX, VOLUME 50 242(84,92), 26 l(270) Masuyama, A., 128(23) Mathison, J. C., 264(263) Matsui, M., 126(10) Matsumoto, K., 152(1 16) Matsumoto, T., 21(14-16) Matsuura, F., 317(79), 321(79) Matsuura, H., 345(182) Matsuura, M., 211(19), 231(156), 252(19), 257(235,237) Matsuyama, M., 317(77), 321(77) Matthews, T. J., 314(41), 325(41) Mattsby-Baltzer,I., 229( 135) Matuo, S., 339(159) Mautsch, H. H., 303(84a) Mawhinney,A. W., 316(64),321(64),329(64) Mawhinney, T. P., 316(64), 321(64), 329(64) May, R. P., 253(222) Mayberry, W. R., 230( 138), 232( 138) Mayer, H., 215(49, 50), 218(82, 84,87, 8992), 219(49-50,96,98-101), 220(96, 104), 221(96, 107), 223(87), 224(91,99, 114), 225(89-91, 121), 226(89,92, 121), 227(90, 124), 228(84, 87,89,92), 230(82, 145), 232(50, 82, 87, 101, 160), 233(87,89), 234(87, 165), 235(87,91, 145), 238(87,89, 185a), 241(92, 193), 242(84, 92, 194), 243(87,89), 247(101, 160), 249(87, 90), 255(229a), 257(232), 261(165,270), 264(265) McCullough, K. J., 125(4) McGarvey, G. J., 183-184(237), 187(237) McKenzie, G., 236( 170) Meagher, M. M., 18(50) Mehmet, H., 329-331(102) Meldal, M., 282(25), 287(49), 288(25), 294(49), 299(49), 302(25, 82,83,49), 306(95 - 99) Mel'nikova, V. I., 169(208) Mendelsohn, J., 335(143) Mendicino, J., 329(93) Meranduba, A., 68-69(139), 72-73(139), 78( 139), 82-83( 139) Merck, R., 185(243), 186(245) Men, G., 288(55), 294(76), 302(55) Mesrobeanu, J., 2 13(27) Mesrobeanu, L., 2 13(27) Messner, P., 250(209) Meuwly, R., 114(210) Meyer, B., 333( 118)

365

Meyer, D. M., 333(132) Meyer, K. C., 232(162), 235-236(162), 264( 162) Meyer, M. V., 244( 198) Michalski, J.-C., 313(16), 333(130, 131), 339(154) Michel, G., 218(61), 225-226(121) Michel, J., 25(45-46), 27(45), 28(46, 65), 29(45-46, 30(51, 52a, 53), 34-35(52a), 36(51, 52a, 65), 37-38(51, 52a), 39(51, 53), 41-42(65), 43(51,52a), 44(51), 52(46), 58(45,51), 60(46), 61(51), 70(53), 11 l(46, 53), 114(46, 51, 52a, 52b), 289(60) Michon, F., 339(150) Mieczkowski, J., 190(255) Mielniczuk, Z., 229( 135) Milat, M. L., 116(213) Minale, L., 105(182) Minka, S., 225-226(121) Minner, I., 227(123), 244(123) Mitchell, D. H., 256(229c) Mitchell, D. L., 129(24), 13l(33) Mitov, I., 259-260(252) Mizuochi, T., 314(41,43), 325(41,43) Mizutani, T., 259(247) Moenng, U., 25(39-42) Moghissi, K. S., 317(79), 321(79) Molina, M. T., 139(61) Moll, H., 229(137), 232(159, 161), 251(211-212) Meller, H., 19(52) Molyneux, R. J., 198(285) Mombers, C., 230(143), 235(143) Mondange, M., 21 1(17), 252(17) Monsalvatje, M., 188(253) Montreuil, J., 216(51), 221(51), 313(16), 3 14-3 15(30),3 16(76), 32 1(76), 324(86), 325(30, 76), 329(76,96,97, 105), 330(76), 333(86,96-97, 130, 131), 334(136), 337(146), 343(30, 169) Mootoo, D. R., 116(215) Moradei, O., 174(225) Moran, A. P., 219-220(97), 222(97), 225(979, 116), 226(97), 229(136), 241(97), 252(97), 257-258(116) Moreno-Mafias, M., 182(235) Morey, M. C., 129(26) Morgan, J. W. W., 151(102) Morgan, W. R. J., 329(98), 333(98)

366

AUTHOR INDEX, VOLUME 50

Morgan, W. T. G., 316(48) Morgan, W. T. J., 213(29) M o ~K., , 49( 105- 107) Morrell, A. G., 338( 148) Moms, D. A., 3 16(64), 32 1(64), 329(64) Moms, E. R., 344( 176) Momson, D. C., 212(20), 256(20), 264(262) Motherwell, W. B., 145(79, 80) Mountsen, S., 306(96) Moustafa, A. E. A., 237(177) Moxon, E. R., 225(120) Miihlradt, P. F., 223(113), 226-227(113) Mukaiyama, T., 21(9), 22(21) Mukerjee, P., 256(229c) Mulder, G. J., 144(75) Miiler-Fahmow, A., 152(114) Miiler-Loennies,S., 2 18(66a), 224(66a) Munford, R. S., Jr., 214(45), 229(45), 238(45, 180- 181, 186), 264(181) Murai, Y., 21(9) Murakami, K., 37-38(73) Murase, T., 41(93), 47(93), 58(120) Murray, P., 186(247) Murray, R. G. E., 241(192) Musehold, J., 236(168, 169), 257-258(168), 264( 169,266) Musina, L. Y., 218(74) Mutsaers, J. H. G. M., 316(57,63), 321(63), 325(57) Myers, K. R., 236( 175) Myers, P. L., 193(267), 196(279)

N Nagai, Y., 339(156) Nagaki, K., 257(236) Nagao, S., 257(236) Nagasawa, K., 249(206) Nagase, H., 249(205) Nagawa, Y., 218(95), 228(95), 243(95) Nagayo, T., 317(77), 321(77) Naiki, M., 339(159) Nakabayashi, S., 279( 18), 283( 18) Nakagawa, M., 160(164, 165) Nakahara, N., 292(66), 294(66), 296(66) Nakahara, Y., 49(100, 104), 50(100), 51(104), 52(104, 110), 53(110), 56(118), 57(110), 62(130, 131), 63( 130), 73( 130), 78( 130), 8 1 - 82(130)

Nakajima, H., 317(77), 321(77) Nakajima, Y., 151(107) Nakaminami, G., 160(164, 165) Nakamoto, S.-I., 21 l(18) Nakanishi, H., 218(95), 228(95), 243(95) Nakanishi, K., 168(206) Nakatsubo, F., 37-38(73) Nakatsuka, M., 21 1(19), 252(19), 257(237) Nakatsuka, T., 22(21) Nakayama, K., 211(15), 215(15), 252(15), 258(15), 263(15) Nakayima, H., 211(15), 215(15), 252(15), 258( 15),263(15) Nakazawa, T., 150(98) Namgoong, S. K., 200(287,288) Nbnhi, P., 35-37(71), 60(124), 105(124), 108(124) Naoki, H., 168(206), 214(34, 35), 217(35), 218(34), 228(35), 247(35), 249(34, 35), 250(34), 251(34), 252(35), 261(34) Nashed, M. A., 249(207) Nassr,M.A. M., 116(213), 155(133), 156(134) Naumann, D., 252(219), 253(219,220), 255-256(228), 262(220,228) Nealson, K. H., 247(203) Nef, J. U., 27(50) Nelson, C. R., 165(190) Nemek, J., 158(149) Nerkar, D., 258(241) Neszmelyi, A., 60(124), 105(124), 108(124), 2 18(64), 250(64,209) Neuberger, A., 277(3-5), 278(5) Neufeld, E. F., 338(149) Nicolaou, K. C., 21(10), 23(30), llO(188) Nicotra, F., 140(66) Nieduszynski, I. A., 329-330( 104) Niemann, C., 157(139) Niemann, H., 339(157) Nifant’ev, N. E., 314(38) Nikaido, H., 263(258) Nilsson, B., 313(27), 316(55), 333(127-129) Nimtz, M., 329(99), 333(99) Nin, A., 163(181), 166(192), 169-170(181) Nishijima, M., 252(217) Nishikawa, S. I., 264(264) Nishimura, C., 211(19), 252(19) Nishio, H., 49(107) Nishiyama, H., 249(205) Niwa, Y., 218(95), 228(95), 243(95)

A U T H O R INDEX, VOLUME 50

367

Nolan, T. J., 161(167) Noort, D., 144(75) Norberg, T.,22(18, 19,24),294(56) Noms, F.,18(51) Noms, K.E., 18(51) Notermans, S., 180(232) Novikova, 0.S.,282(24) Nowotny, A., 217(52), 223(52) Noyori, R., 21(11) Nukada, T.,68-70(141, 142), 89(142) Nukuda, T.,49(104), 51-52(104), 60(127),

Okada, M., 126(10) Okahara, M., 128(23) Okamoto, T.,152(116) Okouchi, K., 339(159) Olah, V. A., 60(125) Oltvort, J. J., 25(43) Ono, M.,344( 181) Ono, Y., 21 1(15), 215(15), 252(15), 258(15), 263( 15) Ooi, C. E., 263(261) Ortufio, R. M., 168(207), 182(235,236,

62(130), 63(130, 132), 66(127), 73(130), 78(130), 81(130), 82(130132), 89( 127) Numata, M., 49(103), 51(103), 58(122) Nunez, H.A.,334(133) Nunomura, S., 49-50(101, 102) Nunozawa, T.,164( 185) Nurminen, M., 227(125), 229(132) Niirnberger, H.,62(133) Nutt, R.F.,195(274)

238-239), 184(207,238,240,241), 185(243), 186(244), 188(252,253) Osada, Y.,211(15), 215(15), 252(15), 258(15), 263(15) Osborn, M. J., 244( 197) Oscarson, S., 288(57) Ostmann, P., 14(36) Otani, S., 152(116) Otsuka, K.,257(236) Otsuka, T.,252(217) Ottesen, M., 18(48) Otvos, L.,Jr., 282(26,27), 288(26), 302(2627), 303(84a, 84b) Oulevey, J., 230( 147) Overend, W.G.,143- 144(72) Overhand, M., 71(151), 73(151) Ovodov, Y.S., 218(62)

0

Oddon, Y., 344(177) Ogasawara, K.,130(28) Ogawa, T.,41(91), 43(91), 46(91), 49(100-

104), 50(100- 102), 51(103, 104), 52(104, IIO), 53(110), 56(118), 57(1 lo), 58(121, 122), 60(91, 127), 62( 130, 13 I), 63(130,132), 66( 127), 67(9 I), 68-69( 141 - 144), 70( 141142), 73(130, 143, 144, 157, 158), 78(130, 143), 81(130, 157), 82(130, 132), 83(157), 89(127, 142), lOl(172, 173), 102(121), 110(189), 111(195), 218(80a), 228(80a), 257(236), 290(61), 292(66), 294(66), 296(66) Ogawa,Y.,211(11, 16),252(11, 16) Ogura, H.,127( 16), 138(56,57), 139( 16,63) Ohashi, K.,49(108), 104- 105(178), 109(178) Ohle, K.,127(17) Ohmori, M.,259(247) Ohmori, T.,169(209) Ohno, K.,249(205) Ohsawa, A., 232( 158) Ohtsuka, Y.,211(18) Ohya, Z.-I., 150(98) Oishi, K., 259(249) Okabe, M., 196(280)

P Paal, M.,288(58-59) Packer, N. H., 222-223(109), 226(109) Panfil, I., 190(260), 191(261-263) Panza, L.,140(66) Papahatijs, D.P., 23(30) Parekh, R. B.,61(136), 314(42), 325(42),

343(42) Parent, J. B., 264(267) Partridge, S.M., 213(29) Pascher, I., 311(11) Passmore, F., 6-7(21), 155(130) Paster, B.J., 247(203) Pasteur, L., 3(7), 8(7) Paulsen, H.,21(4-5), 22(5), 35(4), 41(4),

44(82), 53(4, 112), 57(112), 58(129), 60(128-129), 62(133), 66-67(135), 104- 105(4), 114(209), 217(59), 278(8), 286(8), 288(55, 58-59), 290(64), 292(69), 293(69-70), 294(74, 76),

AUTHOR INDEX. VOLUME 50

368

295(8,74), 302(55,83), 306(95), 343344( 174) Paulsen, K., 296(77,78) Paulson, J. C., 303(89), 304(89-90), 339(154) Pavia, A. A., 287(47,48), 294(47,48,75), 314(35), 344(177) Payen, A.,8(25) Pazur, J. H., 16(40-42) Peach, J. M.,197(281-282) Pedersen, C.,110(187), 134(41-44), 135(45), 158(42-43), 160(42), 169(210), 173(44,223), 174(41,42, 226,227), 175(226,228), 176(229) Pedersen, H., 17(45) Pedersen, T. G.,18(48) Pederson, C.,314(34), 325’34) Pederson, H.,327(34) Pedron, T.,259(246) Pepe, G.,344( 177) Perera, P. Y. Pkrez, S., 343(166) Perini, F.,324(85) Perlin, A. S., 158-159(150), 164(184) Perry, M. B., 287(41) Persoz, J. F.,8(25) Peter-Katalinic, J., 2 18(64a), 224(64a),

317(78,81), 321(78, 81), 324(81) Peters, S., 302(83), 306(95) Peters, T.,288(57) Petitou, M., 33 1( 1 13) Petursson, S., 154(125), 155(126) Pfannemuller, B., 152(1 13, 1 15) Pfeiffer, G.,314(40), 325(40), 329(108) Pickering, N.J., 316(73), 321(73), 324,

325(73), 328(73) Pierce-Crktel,A., 324(86), 333(86) Piloty, O.,157(138) Pimlott, W.,31 l(11) Piszczek, L., 161(169) Pivnitsky, K. K.,169(208) Pizza, C., 105(182) Platzer, N., 333(132) Plessas,N.R., 151(111, 112) Plummer, T.H., Jr., 314(44), 325(44) Podolsky, D.K.,316(71, 72), 321(71, 72),

329(71) Pohlmann, T. H., 238(186) Pokorny, M., 165(189) Polenov, V. A., 139(59)

Pollack, M., 212(21), 259(249) Ponomarev, A. M., 161(168) Ponsati, O., 182(234,236), 183(236,238),

184(238) Poole, A. R., 325(92) Popenoe, E. A., 277(2) Pora, H., 305(91) Potier, P., 195(273) Pougny, J. R., 116(212-213), 239(188) Powell, M. E., 325(90) Prasit, P., 192(266) Pravdic, N., 164(186), 165(188, 189) Preuss, R., 1 1 1 - 112(190) Priebe, W.,142(68) Prohaska, R., 344( 175) Prout, K.,197(281-282) Purcell, S.,244( 198) Puvanesarajah, V., 2 19( 100)

Q Qureshi, N., 214(32, 33, 36,41,43),

215(32), 217(32,60), 218(63, 88), 222( 1 lo), 228(32, 33,60,88), 232(88, 162), 233(88), 234(88, 166), 235(88, 162), 236(32,41, 162, 174), 238(182, 184), 239(41), 241(41), 242(33), 243(63, 88, 174, 198), 244(32,63), 248(60, 174), 249(110,207), 251(1 lo), 259(248), 264(162, 182, 184)

R Rademacher, T. W., 61(136), 225(120),

314(42), 325(42), 343(42, 164, 167) Radziejewska-Lebrecht,J., 2 18(67), 2 19(67,

98),227(126), 241(193) Radzyner-Vyplel, H. Raetz, C. R. H., 21 1(7), 218(63), 226(63),

232(162a), 238(184), 240(191), 244(63, 198), 249(207), 262(191,254-255), 264( 184) RajanBabu, T. V., 143- 144(71) Rajanikanth, B., 154- 155(124) Ralapati, S.,343(169) Ramos, R. A., 264(263) Ramsden, N. G.,154(122, 125), 155(126), 198(122,285) Rana, S. S.,329(93) Randall, J. L., 21(10), llO(188) Rao, B. N.N., 316(56, 58)

AUTHOR INDEX, VOLUME 50 Rao, V. S. R., 333( 122) Rapp, R., 5 ( W Ratcliffe, R. M., 58(152) Rauwald, W., 53( 1 12), 57(112), 290(64) Ray, P. H., 129(25) Reboul, J. P., 344(177) Reck, F., 58(129), 60(129), 66-67(135) Reddy, G. S., 143-144(71) Redmond, J. W., 222-223(109), 226(109), 227(123), 244(123) Rees, B. H., 13l(32) Rees, D. A., 311(1), 344(176) Reeves, R. E., 16(43) Regele, C., 32(60) Regeling, H., 126(7), 127(7,22), 149(22) Regoeczi, E., 329(97), 333(97) Rehorst, K., 148(85) Reichrath, M., 24(37), 25(37, 39-42) Reichstein, T., 157(140) Reilly, P. J., 18(50), 19(53) Reimann, H., 162(175) Reimer, K. B., 306(99) Reinhold, V., 316(60), 321(60) Reinhold, V. N., 213(31) Rembold, H., 25(48) Renfrew, A. G., 151(104) Renwick, G. C., 329(99, 109), 333(99) Reyna-Pinedo, V., 195(273) Ribi, E. E., 214-215(32), 217(32), 228(32), 236(36, 172), 243(172), 244(32), 248( 172) Ricart, G., 313(16), 316(76), 321(76), 325(76), 329-330(76) Riccio, R., 105(182) Richards, J. C., 218(77a) Richardson, D. C., 129(25) Richardson, G. M., 5(13) Richtmyer, N. K., 152(1 17) Riddell, F. G., 3(8), 5(8) Riesen, W., 339( 155) Rietschel, E. T.,211(1-6, 8,9,9a,9b, 14), 212(25), 213(2,4), 214(25, 34-35, 3740), 215(14), 217(1,35, 57-59), 218(1, 4, 34, 57, 64a, 68, 72, 73, 75, 77-81, 83), 218(85,86), 219-220(97), 221(81), 222(68,97), 223(68), 224(64a, 77, 79), 225(4,68, 73,77, 85, 86,97, 116- 118), 226(72,73, 85,97), 227(73,85, 125), 228(4, 35, 37, 68, 72, 77, 79, 187), 229(38, 77,79,86, 128, 129, 132, 137),

369

230(37, 81, 142, 143), 231(38, 77, 85, 129, 142, 151, 156), 234(1, 38, 78-79, 117), 235(37, 142, 143), 236(5, 118, 128, 167, 169, 170, 172, 176), 238(81, 85, 167, 169, 176, 183, 185, 185% 187), 239(38, 187), 240(37-39, 187), 241(37, 68, 97), 242(38, 73,77,85), 243(37,77, 79, 187, 196), 244(3, 77), 245(77), 246(4,78, 79, 86), 247(4, 35, 187), 248(73,77), 249(34-35,72), 250(34), 251(34, 68, 72, 73, 75, 77, 79, 83)), 252(14,35,97, 172, 196,219), 253(220), 254(219,220,224), 255(229a), 256(1, 3-5,6, 118,2294 230),257(1, 3-6, 116, 118, 167,230, 232-234), 258(116, 118, 167, 183, 185, 239), 259(6,243, 244,244a), 261(34, 68,75, 83, 118), 262(220), 263(185), 264(77,79, 167, 169, 183,264,266) Rijkse, I., 329-330(107) Riley, D. A., 143(69),229( 130) angler, N. J., 344( 179) Rio, S., 294(72), 296(72) Rivoire, B., 105(181) Robbins, G. B., 151(100) Robinson, M. J. T., 3(8), 5(8) Rodionov, A. V., 2 18(74) Rodriguez, M., 219(99), 224(99) Roelcke, D., 339( 155, 156) Rogers, G. N., 339(154) Rogers, M. E., 324(84), 329(84) Rohle, G., 35(74) Rohrscheidt-Andrzejewski,E., 2 12(24) Rollin, P., 190(256) Romijn, D., 334(136) ROOS,M., 25(46), 28-29(46), 52(46), 60(46), 11 1(46), 114(46) Roppel, J., 219(96), 220(96, 104), 221(96) Rosankiewicz,J. R., 321(82), 325(82) Rosen, T., 290(65) Rosenberg, R. D., 33 1( 113) Rosenfelder, G., 246(201) Rosevear, P. R., 334(133) Rosner, M. R., 2 15(46,69), 2 18(69), 221(69), 223(69), 227(69), 239(189, 190), 249(69) Ross,B. C., 145(79, 80) Roth, A., 256-257(230) Rothenberg, R. J., 232( 157)

370

AUTHOR INDEX, VOLUME 50

Rousd, P., 316(60-63,65-68), 321(6063,65-68), 333(62) Rouzaud, D., 140(65) Riicker, E., 39(77) Rudolph, J. R., 329(97), 333(97) Rumpold, H., 335(142) Rupprecht, E., 226-227( 122), 244( 122, 197) Russa, R., 218(81), 221(81), 230(81), 238(81), 246(202) Russell, D. H., 313(24) Russell, M. A., 153(121) Russell, R. N., 171(219) Russo, G., 140(66) Ryan, J. L., 212(20), 256(20)

S Sabesan, S., 303-304(89) Sabharwal, H., 333(127- 129) Sabisch, A., 253(221), 255-256(228), 262(221,228) Sadoff, J. C., 259-260(250) Sadozai, K. K., 62(130, 131), 63(130, 132), 73(130), 78(130), 81(130), 82(130, 132) Saenger, W., 152(114) Sager, W., 21(12), 290(62) Saha, B. C., 16(38) Sahoo, S. P., 187(248) Saito, A., 150(97) Saito, H., 3 1l(5) Saito, M., 126(10) Saito, T., 317(80), 321(80) Sakagami, M., 104- 105(178), 109(178) Sakaguchi, N., 244(200) Sakai, K., 52-53(110), 57(110) Sakai, T., 18(48) Saksena, A. K., 162(177) Sala, L. F., 164(182), 171(182,215-218) Salimath, P. V., 218(87), 223(87), 228(87), 233(87, 164), 234(87, 164, 165), 235(87, 164), 238(87), 242(194), 243(87), 249(87) Sallam, M. A. E., 155(132) Saltman, R., 287(50), 299(50) Samantano, R. H., 160(156, 157), 171(156), 195(275) Samraoui, B., 343(173) Samreth, S., 229( 130) Samuelsson, B. E., 31 I( 11) Sinchez-Ferrando, F., 182(235), 188(253)

Sandford, P. A., 31 l(6) Sandstrom, W. M., 157(142) Sarfati, S. R., 21 1(17),252(17) Sane, 0. Z., 162(175-176) S a d , H., 144(74) Sash, H., 324(84), 329(84) Sato, J. D., 335( 143) Sato, S., 49(100-101, 104), SO(l00, 101), 51-52(104), 60(127), 66(127), 6870(142), 73(158), 78(158), 89(127, 142) Satterthwait, A. C., 214(46) Sauter, N. K., 339(153) Savage, A. V., 329(loo), 333( 100) Scannon, P. J., 259-260(251) Schachter, H., 321(83) Schade, F. U., 21 1(9a, 9b), 256(229d), 264(264) Schade, U., 229(133), 238(185, 185a), 258(185), 263(185), 264(133) Schade,U.F.,211(5),225(116,118), 236(5, 118, 170, 172), 243(196), 252(172, 196), 256(118), 257(5, 116, 118,233, 234), 258(116, 118), 261(118) Schaffer, R., 34(63), 49(63), 58(63), 60(63), 84(63), 105(63) Schaller, H., 252(218) Schaubach, R., 37(72), 54(114), 55(72, 116), 78-79( 1 14), 8 1(1 14), 85(114), 94( 1 16), 98- 100(1 16) Schauer, R., 329( 1 1 l), 332-333( 11l), 337(1 1 1) Scheinberg, I. H., 338(148) Scheuer, P. J., 230(149) Schimpff, G. W., 147(144) Schiphorst, E. C. M., 329(100) Schlecht, S., 227(124), 230-231(142), 235(142), 237(178) Schlessinger, J. J., 335(143) Schliiter, C., 264(266) Schmidt, G., 21 1(9a, 9b) Schmidt, M., 329(108) Schmidt, 0. T., 127(18) Schmidt, R. R., 21(1-3), 22-23(17), 24(1, 3, 37), 25(37, 39-42,44-49), 26(1-3), 27(1-3,45), 28(46, 58a, 65-67, 70), 29(45-46), 30(1, 17, 52a, 53, 55-57, 58a, 58b), 32(1, 17, 56, 59-60, 62), 34(1, 52a), 35(1-3, 52a, 62, 67), 36(52a, 62,65), 37(52a), 38(52a, 62), 39(53, 67, 77, 78), 41(1, 62, 65, 66, 70,

AUTHOR INDEX, VOLUME 50 84, 86-90, 92), 42(65), 43(52a, 84, 86, 87), 46(84), 47(92), 48(57,97), 49(70, 84,98), 50(86,98), 52(46,67), 53((113, 115), 54(67), 55(115), 57(113), 58(1, 45,62, 115), 60(46), 61(62), 68(137, 137a, 140, 145, 149), 69(137, 137a, 140), 70(53, 55, 145, 148, 149), 71(137, 149), 72(140), 73(137, 140, 145, 148, 149, 155), 74(145, 154), 75(57, 148, 154), 76(149), 78(140), 79(115, 155), 80(137), 81(92, 115, 137, 140, 154, 155, 164, 165), 82,83(140), 84(57,97, 149), 85(92, 115, 154), 90(57, 148), 91(57, 97), 93(137, 149), 94(137, 170), 95(113), 98(137, 149, 170, 175, 176), 99(137), lOO(113, 115, 154), 104(1, 58b, 175), 105(1,58a, 58b, 176), 106(165), 107(176), 111(1,46, 53,66, 78, 190, 192, 197-199), 112(190), 113(192), 114(46, 52a, 52b, 78, 199, 204- 208), 1 16(1, 3, 17), 289(60) Schneider, H., 212(22), 214(33), 228(33), 242(33) Schneider, P., 151(110) Scholtz, S. A., 192(265), 196(265) Scholz, D., 211(13), 219-220(97), 222(97), 225-226(97), 241(97), 252(13, 97, 2 18), 262( 13) Schonbeck, U., 238( 185a) Schotte, H., 14(35) Schotz, M. C., 230(140) Schramek, S., 229(133, 137), 264(133) Schreier, M., 21 l(9a) Schroter, D., 194(270) Schuerch, C., 23(34) Schuhmacher, M., 28(68), 52(68), 1 14(68) Schultheiss-Reimann, P., 287(43), 288(43, 54), 294(43) Schultz, C., 252(219), 254(219), 255256(228), 262(228) Schultz, M., 288(58), 293(71), 294-295(74), 303(87), 306(93), 307(108) Schulz, G., 218(64, 70), 249(70), 250(64, 210), 251(70), 261(70) Schuster, M., 307(105) Schiitze, E., 211(13), 252(13), 262(13) Schwarcz, J. A., 158-159(150) Schwebel, A., 157(143) Schweitzer,M. G., 229( 130) Schwentner, J., 1 lO(185)

37 1

Scudder, P., 322(87), 329-331(102, 103) Scudder, P. R., 339(154) Sears, P., 307( 105) Seguchi, T., 344(181) Seitz, S. P., 23(30) Seltmann, G., 263(259) Selvakumar, R., 344(179) Sepulchre, A. M., 129(25,27), 130(27), 198(284) Seshadri, R., 154-155(124) Seydel, U., 21 1(5,9a), 212(25), 214(25, 34, 37-40), 218(34, 79, 82,83, 89), 219(97), 220(97), 222(97), 224(79), 225(89, 97, 118), 226(89,97), 228(37, 79, 89), 229(38,79), 230(37, 82), 231(38, 156), 232(82), 233(89), 234(38, 79), 235(37), 236(5, 118, 173), 238(89), 239(38), 240(37-39), 241(37,97), 242(38), 243(89, 173, 195a), 244(173), 246(79), 249(34), 250(34), 251(34, 79, 83), 252(97), 254(224-226), 255(227, 229a), 256(118,230,231), 257(5, 118, 230), 26 1(34,83, 118), 264(79) Shaban, M. A. E., 155(133), 156(134), 283(31) Shah, R. H., 130(30-31), 131(31) Shah, R. N., 343(163, 168) Shands, J. W., Jr., 262(256,257) Shaper, M. A., 343(173) Sharaf, S. M., 155(132) Shashkov, A. S., 218(74), 314(38) Shaw, D. H., 21 1(3), 244(3), 256-257(3) Shelton, B., 287(41) Sheth, H. G., 168(204-205), 189(205) Shiba, T., 116(217), 211(5, 15), 214(34-35), 215(15), 217(35), 218(34), 228(35), 236(5, 170, 172), 243(196), 244(200), 247(35), 249(34, 35), 250(34), 251(34), 252(15,35, 172, 196,215,216,219), 254(219), 257(5, 233,234,236,237), 258(15,239), 261(34), 263(15) Shibasaki, M., 144(74) Shibayama, S., 49(103), 51(103) Shibuya, H., 104-105(178), 109(178) Shibuya, M., 160(164, 165) Shimamoto, T., 211(14), 215(14), 236(170), 252(14, 216), 257(237) Shimauchi, H., 257(236) Shimizu, T., 21 l(18) Shing, T. K. M., 154(123)

372

AUTHOR INDEX, VOLUME 50

Shioi, S., 160(164, 165) Shioya, E., 211(15), 215(15), 252(15), 258( I5), 263( 15) Shiozaki, M., 252(217) Shnyra, A., 255(229b) Shoda, S., 21(9), 22(21) Shogren, R., 344( 180) Shrivastava, V. H., 137(52) Siba, S., 292(66), 294(66), 296(66) Sidorczyk,Z., 21 8(72), 226(72), 228(72), 249(72), 251(72) Sierks, M. R., 19(53) Silberkweit, E., 169(213) Simet, I. M., 344( 179) Simon, E. S., 303(88) Simon, M., 223(112), 249( 112) Sinay, P., 23(36), 52( 109), 69(138), 72(138, I53), 1 10(186), 116(212-2 13), 140(64, 65), 190(256),239( 188), 287(42a, 42b) Sindric, R., 153(119) Singh, N. P., 41(86), 43(86), 50(86) Singh, P. P., 2 17(56) Singh, U. C., 343(167) Sinnott, M. L., 116(216), 161(170, 171) Sinnwell, V., 2 19(64a),224(64a) Sin, D., 344(177) Sjoblad, S., 333(127- 129) Skehel, J. J., 339(153) Slaghek, T. M., 105(179, 180), 109(179, 180) Slessor, K. N., 193(268) Slomiany, A., 316(69,70), 321(69,70) Slomiany, B. L., 316(69,70), 321(69,70) Smith, A. R. W., 231(153) Smith, C. R., 259-260(250) Smith, D. F., 333( 124) Smith, K. D., 314(45), 315(46), 325(45,46) Smith, P. W., 197(281,282) Sodeoka, M., 144(74) Soga, T., 211(15), 215(15), 252(15), 258(15) Soh, C. P. C., 329(98), 333(98) Sohar, P., 163(180) Solomon, J., 314(41), 325(41) Solov’eva, T. F., 218(62) Solter, D., 335(140) Son, J. C., 197(283) Sonesson, A., 225( 119), 229(119), 230(141), 232(159, 161) Sonnichsen, R., 134(44), 173(44) Souchard, I. J. L., 161(170)

Spellman, M. W., 314(39), 325(39) Spencer, J. F. T., 230( 150) Sperber, N., 147(142) Spik, G., 324(86), 329(97), 333(86,97), 343(169) Spinola, M., 279( 13) Spohr, U., 157(137a),333(120, 121) Sprung, C. L., 259-260(250) Stacey, G., 219(100) Stacey, M., 31 l(8) Stackebrandt, E., 241(192), 247(203) Stahel, R., 6-7(22) Stellner, K., 31 l(5) Stick, R. V., 127(14),280(20) Stix, D., 25q210) Stoll, M. S., 313(26,28), 316(73-75), 321(73-75), 323(28, 74), 324(73), 325(73,75), 328(73-75) Storer, R., 193(267), 196(279) Strain, S. M., 215(65), 218(63,65,66), 222(66), 226(63), 244(63), 249(66), 250-25 1(65), 263(65) Stmube, R. C., 259-260(250) Strecker, G., 316(76), 321(76), 324(86), 325(76), 329(76,96,97, 105), 330(76), 333(86,96-97, 130, 131), 334(136), 337(146), 343(169) Strittmatter, W., 2 18(87),223(87), 228(87), 233-235(87, 164), 236(170), 238(87), 243(87), 249(87), 258(241) Stroh, H. H., 155(131) Strominger, J. L., 343(173) Strube, K.-H., 329(108) Stumpp, M., 28(66), 30(55), 41(66), 52(66), 70(55), 1I1(66), 114(52b, 66) Stiitz, P., 21 1(13),219-220(97), 222(97), 225-226(97), 241(97), 252(13,97), 262( 13) Stiitz, P. L. Suami, T., 169(209) S ~ g hH., , 143- 144(70) Sugimoto, M., 49-50( loo), 49( 103), 51(103), 56(118), 58(122), 63(132), 82( 132) Sugiyama, S., 160(164, 165) Suh, H., 193(269) Sun, R.-C., 196(280) Sutter, M., 157(140) Suwinska, K., 191(263)

AUTHOR INDEX, VOLUME 50 Suzuki, K., 21(14, 15) Svendsen, I., 18(48), 19(52) Svensson, B., 18(48,51), 19(52-53,55) Svensson, S., 31 l(9) Swahn, C. G., 126-127(8) Sweeley, C. C., 3 13( 19) Swiderski,J., 161(168, 169) Szab6, L., 211(17), 214(42,44),218(?3),

222(11 l), 230(139), 232(139), 238(93), 252( 17) Szab6, P., 211(17), 252(17) Szafranek, J., 41(79) Szejtli, J., 25-27(71) Sznaidman, M., 132(36-37), 133(39), 166(196), 173(222)

T Tabern, D. L., 151(108) Tachibana, Y., 333( 125) Tacken, A., 222( 1 1 1) Tadanier, J., 143(69) Tadano, K., 169(209) Tadano, K. I., 200(288) Taha, M. A. M., 155( 133), 156(134) Takada, H., 21 1(14),215(14), 252(14), 257(236,238) Takahachi, T., 252(214) Takahashi, A., 144(74) Takahashi, H., 127(16), 138(57-58), 139( 16,63),224( 1 15) Takahashi, I., 257(236) Takano, S., 130(28) Takano, T., 37-38(73) Takayama, K., 214(33, 36,41,43), 217(60), 218(63,88),222(1 lo), 226(63), 228(33, 60,88),232(88, 157,162),233(88), 234(88, 166), 235(88, 162),236(41, 162,174),238(182, 184),239(41), 241(41), 242(33), 243(88, 174), 244(63, 198), 248(60, 174),249( 110,207), 25(I 1 lo), 256(229c), 259(248), 264(162, 182,184) Tam, S. Y.-K., 196(280) Tamaru, M., 149(92-93) Tamiya, E., 218(95), 228(95), 243(95) Tanahashi, M., 21 (1 1 I), 252( 1 1) Tanaka, A., 257(236) Tanaka, C., 220(103)

373

Tanaka, S., 211(11, 16),220(103), 252(11,

16),257(236) Tanamoto, K . 4 , 236( 170),257(233,237) Tang, J., 215(69), 218(69), 221(69), 223(69),

227(69), 249(69) Tang, P. W., 329-331(102, 103) Tarentino, A. L., 314(44),325(44) Tatsuta, K., 1 10(184) Tavecchia, P., 69(138), 72(138, 153) Taylor, D. P., 220-221(102) Taylor, G. L., 343(167) Taylor-Robinson, D., 339( 154) Tefft, M., 151(109) Tejbrant, J., 288(56), 294(56) Teng, N.N.H., 259-260(250) Termin, A., 70(148), 73(148), 75(148),

81(168), 90(148) Thaisrivongs, S., 183-184(237), 187(237) Tharanathan,R. N., 218(90), 225(90),

227(90), 234( 165),242(194), 249(90), 261( 165) Thiele, 0. W., 230(147) Thiem, J., 21(13), 110(185), 114(209), 305(92) Thierfelder, H., 9-10(30), 12-13(30) Thergersen, H., 333(118), 339(158) Thorn, D., 344(176) Thomas, A., 187(250,251) Thompson, A., 155(128), 158(147) Thomson, J. K., 163(181), 169-170(181) Thorpe, S. J., 335(142) Thurin, J., 282(26-27), 288(26), 302(26, 27), 303(84a), 337(147) Timpe, W., 168(203) Titani, K., 314(41), 325(41) Tobias, P. S., 264(263) Todaro, L. J., 196(280) Toepfer, A., 32(62), 35-36(62), 38(62), 41(62), 48(96-97), 58(62), 61(62), 81(165), 84(97), 91(97), 98(176), 105(176), 106(165), 107( 176) Toone, E. J., 303(88) Toromanoff, E.,151 ( 109) Toyokuni, T., 44(83), 99(171) Tozer, M. J., 145(79, 80) Tran, V., 343(166) Tranberg, K.-G., 230(141) Trigalo, F., 21 1(17), 252(17) Trimble, R. B., 314(44), 325(44)

374

AUTHOR INDEX, VOLUME 50

Trippelvitz, L. A. W., 329(100), 333(100) Truchot, A. T., 236( 175) Truelove, J. E., 30(54), 70(54) Trumtel, M., 69(138), 72(138, 153) Triiper, H. G., 241(192) Tsai, C.-M., 218(73), 225-227(73), 242(73), 248(73), 25 l(73) Tsai, T. Y. R., 22(25,26) Tsuchihashi, G.-I., 21(14, 15) Tsuda, Y., 164(185) Tsujimoto, M., 257(236) Tsvetkov, J., 25(49) Tucker, L. C. N., 127(19) Tulloch, A. P., 230( 150) Tuominen, J., 243( 195) Tvaroska, I., 35(76) Tyumenev, V. A., 136(51)

U Uchida, K., 231(155) Uchida, T., 22(29), 152(116) Uebach, W., 253(222) Uemura, K.-L,339( 154, 156) Ueno, K., 150(97) Ueno, Y., 169(209) Ugen, K. E., 303(84a) Ugolini, A., 141(67) Uhlenbruck, G., 317(78, 81), 321(78, 81), 324(8 1) Ulevitch, R. J., 264(263) Ulmer, A. J., 21 1(9a, 9b), 236(168, 169), 238(183, 185, 185a), 257(168), 258(168, 183, 185), 263(185), 264(169, 183,266) Ulrich, J. T., 236(175) Umemoto, T., 261(268) Umemura, K., 149(95), 150(96) Umezawa, S., 1 lO(184) Unger, F. M., 214(34), 218(34,64, 70), 249(34, 70), 250(34,64,209,210), 251(34, 70), 261(34, 70) Unverzagt, C., 280( 19a-19c), 283( 19a, 19c, 34,46), 284(34, 36), 285(19a, 19c, 34, 36), 286(19a, 19c), 294(19a, 19c), 299(19a, 19c), 302(81), 303(90) Upson, F. W., 151(100) Urbanik-Sypniewska,T., 2 18(82,92), 219(100, 101), 225-226(92), 228(84,

92), 230(82), 232(82, IOl), 241242(92), 247(101) Urge, L., 282(26-27), 288(26), 302(26,27), 303(84a, 84b)

V Vaara, M., 227(125), 243(195, 195a), 263(260) Vaara, T., 227(125) Valle, S., 182(235) van Alphen, L., 230(143), 235(143) van Boeckel, C. A. A., 25(43), 43(81), 211(12), 252(12) van Boom, J., 25(43) van Boom, J. H., 71(151), 73(151), 144(75), 180(232),211(12), 252(12) van Cleve, J. W., 22(22) van den Eijnden, D. H., 329(100), 333(100) van der Marel, G. A., 71(151), 73(151), 144(75) van Halbeek, H., 314-315(31, 32), 316(57, 61-63,65-68), 321(61-63, 65-68), 325(31-32, 57,91), 329(95, 96, 101), 333(31-32,62,96, 101), 335(144) van Kuik, J. A., 314(36), 329-330(107) van Marle, T. W. J., 155(127) van Oijen, A. H., 105(179), 109(179) van Steijn, A. M., 52-53(11 I), 57-58(111) van Zuylen, C. W. E. M., 180(232) Vankar, P. S., 39(78), 111(78), 114(78) Vankar, Y. D., 39(78), 1 I1(78), 114(78) Varela, O., 132(38), 148(89), 160(158, 161162), 163(18l), 164(158), 165(19I), 166(192, 194- 195), 167(200), 168(199?, 202), 169(181), 170(181, 194), 171(220), 172(194,202), 173(194, 200), 174(225), 179(230), 180(231), 18l(38) Vasella, A., 70(147), 114(210, 21 l), 115(211), 157(135, 136) Veeneman, G. H., 144(75), 180(232) Vekemans, J. A. J. M., 127(15), 167(197), 169(211), 182(15), 183(211),200(197) Veleva, K., 258(241) Verner, I. K., 2 18(74) Verret, R. C., 239(189, 190) Vethaviyasar, N., 22(20) Veyritres, A., 68( 139), 69( 138, 139), 72(138-139, 153), 73(139, 159),

AUTHOR INDEX, VOLUME 50 78(139, 159), 81(159), 82-83(139), 84( 159) Vicari, G., 316(59) Vilamjo, L., 182(235) Ville, G., 139(62) Vliegenthart, J. F. G., 52-53(111), 5758(1 ll), 60(126), 63(126), 66-67(126), 105(179, 180), 109(179, 180), 313(21), 314(31,32, 36), 315(31, 32), 316(57, 61-63,65-68), 321(61-63,65-68), 325(31, 32, 57,91), 329(96, 101, 107, 112), 330(107), 333(31, 32,62,96, 101, 112), 334(136), 335(144), 337(146) Vogel, H. J., 343(172) Vogel, S. Vogt, K., 193(267), 196(279) Voigt, H., 125- 127(3) Voigt, J., 125-127(3) Volk, W. A., 218(78), 234(78), 246(78) von dem Bruch, K., 278( 1 l), 286( 1 I), 306(100) Vorgel, E., 253(222) Voshol, H., 329(112), 333(112) Vuorio, R., 263(260) Vyplel, H., 211(13), 252(13, 218), 262(13) W Walding, M., 22(19) Waldmann, H., 11l(194, 196), 278(9), 279(14-16), 283(14-16, 32, 33, 3537), 284(15, 32, 33, 35, 36), 285(3536), 286(9, 37), 287(16,46), 289(32), 291(68), 294(32, 33,68), 297(68), 301(32), 303(46, 85, 86), 307(103) Waldstiitten, P., 218(70), 249(70), 251(70), 261(70) Walker, R., 169(212) Walker, T. E., 132(35) Wallis, C. J., 193(267), 196(279) Walter, B., 288(58) Walton, D. T., 287(41) Walton, E., 195(274) Wand, A. J., 337(149) Wang, M.-H., 238(183, 185a), 258(183), 264(183,266) Wang, P., 307( 105) Wang, R., 214(43), 218(94), 228(94), 232( 162), 235-236(162), 241(94), 264( 162)

375

Wang, Y., 193(267), 196(279) Ward, J., 236(175) Warren, C. D., 279(18), 283(18), 306(102), 343( 169) Wartenberg, K., 227(124), 230(145), 235( 145) Watanabe, K., 261(268), 339(157) Watanabe, T., 49(108), 150(98) Watkins, W. M., 316(49), 329(98), 333(98) Webber, J. M., 125(4), 131(32) Weber-Molster, C. C., 127(18) Weckesser, J., 215(49, 50), 218(82, 84, 87, 89-92)), 219(49, 50,96), 220-221(96), 223(87), 224(91, 114), 225(89-92), 226(89,92), 227(90), 228(84, 87, 89, 92), 230(82), 232(50,82), 233(87, 89, 164), 234(87, 164- 165), 235(87,91, 164), 238(87,89), 241(92), 242(84,92, 194), 243(87, 89), 249(87, 90), 255(229a), 261(165,270), 264(265) Wedel, N., 259-260(251) Weerman, R. A., 15I( 103) Wegmann, B., 28(67), 30(58b), 35(67), 39(67), 52(67), 54(67), 98(175), 104105(58b), 104(175) Weichert, U., 53(112), 57(112), 288(55), 290(64), 302(55) Weidmann, H., 168(203), 194(272) Weigandt, H., 58( I 19), 60( 1 19), 73( 119) Weigel, T. W., 171(219) Weiner, D. B., 303(84a) Weintraub, A., 218(79), 224(79), 225(117), 228-229(79), 234(79, 117), 243(79), 246(79), 251(79), 264(79) Weintraub, S. T., 218(94), 228(94), 241(94) Weisburg, W. G., 247(203) Weiss, J., 263(261) Weisshaar, G., 329(99, 109), 333(99) Weisshaar, R., 221( 105), 230(105), 252(105) Wells, W. W., 313(19) Welsh, E. J., 344(176) Welte, W., 152(114) Wendorf, P., 314(40), 325(40) Wernig, P., 293(71), 294-295(73), 297298(73) Westerduin, P., 2 11(12), 252( 12) Westheimer, F. L., 116(218) Westphal, O., 211(1,2,4-5), 213(2,4, 28), 217(1, 28, 54), 218(1, 4, 54), 225(4), 228(4), 229(128), 234(1), 236(5, 128,

376

AUTHOR INDEX, VOLUME 50

170),237(177, 179),246(4,201), 247(4), 256(1,4),257(1,4,5,28,232, 234) Wethers, S. G., 161(172) Wheat, R., 213(30), 217(30), 230(30) Whitesides, G.M., 303(88), 339(153) Widemann, C., 219(98), 230( 145),235(145) Widmalm, G., 314(37) Wiemann, T., 305(92) Wieruszeski, J.-M., 316(76), 321(76), 324(86), 325(76), 329(76, 105), 330(76), 333(86, 130- 131) Wiesner, K., 22(25-26) Wilcox, C., 193(269) W~~CO C.XS., , 143- 144(70), 183- 184(237), 187(237) Wiley, D. C., 339(153), 343(173) Wilkinson, B. J., 234( 163) Wilkinson, S. G., 218(76), 220-221(102), 229(127, 131), 230(127), 234(127, 163) Willard, J. J., 31 l(8) Williams,T.J., 151(111, 112) Williamson, J. M., 240(191), 262(191) Wilson, F. X., 193(267), 196(279) Wilson, S., 329(95) Winchester, B., 127(20), 133(20), 199(20), 200(286- 288) Winckler, F. W., 6(15) Windholz, T. B., 286(38) Windmuler, R., 81(161), 83(161), 105, 106(177) Wittkiietter, U., 149(90), 162( 178) Witty, D. R., 193(267), 196(279) Witty, R.,154(122), 198(122) Woese, C. R., 247(203) Wolf, N., 168(203) Wolfe, M. S., 196(277, 278) Wolfrom, M. L., 148(86), 151(86, 102, 105-106),155(128), 157(145), 158(147) Wollenweber, H.-W., 212(25), 214(25, 35, 37-38),217(35), 218(79), 224(79), 228(35, 37,79,187),229(38, 79,137), 230(37, 142),231(38), 234(38, 79), 235(37, 142),236(176), 238(176, 187), 239(38, 187),240(37, 38,187),241(37), 242(38), 243(37,79,187),246(79), 247(35, 187),249(35), 251(79), 252(35), 264(79) Wong, C.-H., 307( 105) Wong, C. T., 339(158)

Wong, R., 222(1 lo), 249(1 lo), 251(110) Wong, T. C., 333( 119) Wood, H.B.,148(86), 151(86), 157(145) Woodward, H. D., 344(179) Woodward, R. B., 22(27) Wortel, C. H., 259-260(250) Wray, V., 223(113), 226-227(113) Wright, D. E.,149(91) Wright, D. J., 314(33), 323(33), 327(33),

332-333(33) Wright, S. D., 264(263) Wroblewski, K., 282(26,27), 288(26),

302(26-27), 337(147) Wu, A. M., 316(54, 57), 325(57), 329(54) WU, S.-S., 316(58) Wulff, G., 35(74) Wurzburg, B. A., 339(153) Wuts, P. G. M., 22(28)

Y Yamada, Y., 219(99), 224(99) Yamagiwa, Y., 49( 108) Yamamoto, A., 257(237) Yamamoto, M., 236(170) Yamasaki, R., 212(22), 339(151) Yamashina, I., 317(77), 321(77) Yamashita, K., 314(43), 325(43), 333(125) Yamazaki, F., 60(127), 66(127), 6869(141- 144),70( 141- 142), 73( 143,

144), 89(127, 142-143) Yan, Z.-Y., 334( 135) Yanaghara, Y., 21 l(18) Yano, I., 230(146) Yasuda, T., 257(237) Yokota, A., 215(50), 219(50,99), 221(107),

224(99), 232(50) Yoshida, M., 264(264) Yoshikawa, M., 104-105(178), 109(178) Yoshikawa, T. T., 230(140) Yoshimoto, K.,164(185) Yoshimura, H., 244(200), 252(2 15),

257(236-237) Yoshimura, J., 149(92-95), 150(97,98) Yoshimura, S., 257(234) Yurewicz, E. C., 317(79), 321(79) Z

Zabel, V., 152( 1 14) Zach, K., 13- 14(34)

AUTHOR INDEX, VOLUME 50

Zahringer, U., 21 l(9, 9a, 9b), 212(25), 214(25, 34, 39), 218(34,64a, 68, 72, 73, 75, 77, 79, 83), 219-220(97), 222(68,97), 223(68), 224(64a, 77, 79), 224(77,79), 225(68,73, 77,97, 116118), 226(72-73,97), 227(73), 228(68, 72,77,79), 229(77,79), 231(77, 156), 232(159, 161), 234(79, 117), 236(118, 170), 240(39), 241(68, 97), 242(77), 242(73, 77), 243(77,79, 173), 244(77, 173), 245(77), 246(79), 248(73, 77), 249(34, 72), 250(34), 251(34,68, 72, 73, 75, 77, 79,83,211,212), 252(97), 256( 1 l8,229d), 257( 116, 118), 258(116, 118),261(34,68, 75, 83, 118), 264(77, 79) Zamojski, A., 190(255) Zamze, S. E., 225(120), 231(153) Zaugg, H. E., 147(142) Zbedska, E., 316(69-70), 321(69,70) Zbiral, E., 133(40)

Zeikus, J. G., 16(38) Zempkn, G., 16(37), 19(37) Zen, S., 22(29) Zervas, L., 169(213), 277(1) Zhdanov, Y. A., 136(50, 51), 139(59) Ziegast, G., 152(115) Ziegler, E. J., 259-260(250) Zimmer, J., 306(101) Zimmerman, M., 195(274) Zimmermann, P., 28(70), 41(70, 84,85), 43(84), 46(84-85), 49(70, 84, 85), 52(85), 53(113), 57(113), 95(113), 100(113) Zinner, A. H., 125-127(3) Zissis, E., 165(189) Zitrin, C. A., 161(166) Zollo, F., 105(182) Zollo, P. H. A., 139(64) Zopf, D. A., 316(55), 333(124) Zucchelli, L., 140(66) Zurabyan, S. E., 28 l(22)

377

SUBJECT INDEX FOR VOLUME 50 A Abequose, 173-174 Acetalation, 125- 126 2-Acetamido-2-deoxyaldohexoses, 164- 165 2-Acetamido-2deoxy-~~-glucosyla as paragine, linkage with N-glycopeptides, 278-283 anomeric azides as precursors, 278-280 1-hydroxy-D-glycosamines reactions with ammonium hydrogencarbonate, 282-283 in situ anomektion, 280-281 1-isocyanatesas precursors, 28 1-282 3-Acetoxy-6-acetoxymethylpyran-2-one, 165-166 3-Acetoxy-5-methylen-2(5H)-furanone,168 1-0-Acetyl, activation, 3-0-glycosidic linkage formation, 292-293 N-Acetylgalactosamine,trichloroacetimidates, synthesis, 84, 93-98 N-Acetylglucosamine,naturally occurring glycosidic linkages, 6 1,68 N-Acetyllactosamine,trichloroacetimidates, synthesis, 73, 78 - 80 2-Acylamido-hex-2-enonolactones, 169- 170 3-Acylamido-2-pyrone, 165- 166 Acylation, aldonolactones, 132- 134 Alcohols as 0-nucleophiles, see 0-Nucleophiles reaction with aldonolactones, 149- 15 1 Aldehydes, reaction with aldonolactones, 125-127 Alditols, aldonolactone reduction to, 157161 Aldohydroximo-lactones, 156- 157 Aldonamides, preparation, 15 1 Aldonolactones, 125- 201 acylation and etherification, 132- 134 benzoylated, benzoic acid elimination, 167- 168 p-elimination, 162- 166 carbonyl group, chain elongation, 136148 formylaminomethylenation of aldonolactones, 146- 148 methylenation of aldonolactones, 143146 reaction with organomagnesiumand organolithium reagents, 138- 143

Reformatsky-typereactions, 136- 138 as chiral precursors for synthesis of natural products, 18 1-20 1 y-butenolides, 182- 190 Gbutenolides, 190- 192 nucleosides, amino acids, and other Ncontaining products, 195-20 1 from ~-ribono-l,4-lactone,192- 195 glycosylation, 179- 181 intramolecular reaction yielding lactams, 154-155 lactone group, reaction with ammonia and amines, 151 - 153 hydrazine and derivatives, 155- 157 methylenation, 143- 146 reaction with alcohols, 149- 15 1 aldehydes, 125-127 hydrogen bromide, 134- 136 ketones, 127- 130 reduction to aldoses and alditols, 157- 161 by borane, 159- 160 isotopic labeling and substitution at anomeric center of aldoses, 161 - 162 by sodium amalgam, 157 synthesis of deoxy sugars, 170- 179 use of acetals for sugar derivative synthesis, 130- 132 Aldoses, aldonolactone reduction to, 157161 0-Alkylation, anomeric, 23 -25 y-Alkylbutanolactones, synthesis, 183- 184 y-Alkyl-a,pbutenolides, synthesis, 183- 184 Allyloxycarbonylgroup, N-terminal, with tert-butyl ester, 285-286 Amicetose, 158- 159 Amines, lactone group reaction with, 151153 Amino acids, synthesis from aldonolactones, 195-201 N-(2-Aminoethyl)aldonamides, 152 Aminolactones, intramolecular reaction, lactam preparation, 154- 155 Ammonia, lactone group reaction with, 151 - 153 Ammonium hydrogencarbonate, reactions with 1-hydroxy-D-glycosamines,282 283

378

SUBJECT INDEX. VOLUME 50

379

Amylase, 1 1, 16 hydrogenolysis, 173- 174 Amyloglucosidase, I6 - 17 Bromodeoxyaldonolactones,preparation of Angelica lactone aminodeoxy aldonic acids and sugars, as chiral starting compound, 184 135 reaction with isoprene, 188 189 Brernsted acids, reaction with Anisomycin, 198 0-(glycosyl)trichloroacetimidates,30Anomeric oxygen-exchange reactions, 2 1 32 1 17; see also Trichloroacetimidate Butenolide, 169 method d-Butenolides, synthetic uses, 190- I9 1 0-alkylation, 23-25 y-Butenolides, synthetic uses, 182- 190 Anomerization, in situ, attachment of a-Ldouble bond reaction, 187 188 fucosyl units, 280-28 I Antibodies, lipid A, 259-260 C Antigens, blood-grouprelated, 333- 337 Aristeromycin, 196 Carbohydrate-protein linkages, amino acid I-Aroyl-2-~-gluconylhydrazides, 155- 156 and carbohydrate residues, 277 - 278 5-Aryl-2-ethoxy-3-~-gluconyl-2,3d~hydro-Ceramides, glycosylation by trichloroaceti1,3,4-oxadizoles, 155- 156 midates, 49 - 5 1 Ascarylose, 171- 172 Chitobiose Aspergillus niger, glucoamylase, I6 - 18 derivatives, glycosylation, 8 1, 89 Azides, anomeric, as precursors, 278-280 as donor, 8 1,89 2-Azido-2-deoxyglucopyranosyltrichloroaChromatography, oligosaccharide physicocetimidates, reaction with nucleophiles, chemical analysis, 3 12- 3 13 73-76 Chromobacterium violaceum, lipid A, 242 2-Azido-2-deoxy-~-mannose derivatives, (+)-Citreoviral, 193- 194 trichloroacetimidates,glycosylation, 98, CNT, 196- 197 (+)-trans-Cognac lactone, 186 103 Azidosphingosine, derivatives, glycosylation 2,3-O-Cyclohexylidene-~-ribono1,4-lactone, 129-130 with trichloroacetimidates,41,45-48

-

-

B

D

Bacteria, gram-negative, lacking lipopolysaccharides, 262-263 Base-catalyzed rearrangement, bromodeoxy aldonolactones, 175- 179 2-Benzamido-4,6-0-benzylidene-2,3dideoxy-~-erythro-hex-2-enono- 1,5lactone, 163 Benzoylation, in excess of pyndine, 164 3-Benzoyloxy-5-ethylidene-2( 5SI)-furanone, 168 3-Benzoyloxy-6-methyIpyran-2-one, 165166 Biology, chemical basis, experiments demonstrating, 8-9 Bis-normaytasinoid, 195 (+)-Blastmycinone, 184 Bromodeoxy aldonolactones base-catalyzed rearrangement, I75 - 179

ddC, 196-197 2-Deoxy-~-~-arabino-hexopyranosides, 110- I12 3-Deoxy-~-arabino-hexose,I7 I 2-Deoxyhexoses,trichloroacetimidates, 110-112 Deoxy sugars, synthesis from aldonolactones, 170- 179 base-catalyzed rearrangement of bromodeoxy aldonolactones, 175- 179 bromodeoxyaldono-1$-lactone hydrogenolysis, 173- 174 enonolactone catalytic hydrogenation, 170- 173 Deprotection, selective enzymic, glycopeptide synthesis, 303 N-glycopeptides, 283-287 methods, 294-298

380

SUBJECT INDEX, VOLUME 50

2,4-Di-O-benzoyl-3,6-dideoxy-~-erythrohex-2-enono-1,5-1actone, 164 2,6-Di-O-benzoy1-3,5-dideoxy-~,~-threo1,4lactone, 172 Dibenzyl hydrogenphosphate,reaction with a and 8-trichloroacetimidates, 30- 32 2,5-Di-O-benzyl-~-glycero-pent-2-enono1,4-lactone, 168 Dichloromethylenation,aldonolactones, 145 2,3-Dideoxyhex-2-enono-1,5-1actone derivatives, 190 1,4-Dideoxy-1,4-imino-~-glucitol,198 Dideoxylactones, 172 L-1,6-Diepicastanospermine,198- 199 Difluoromethylenation,aldonolactones, ' 145- 146 (2R,35',4R)-3,4-Dihydroxyproline, 197 Diisoamylborane, reduction of aldonolactones, 160- 161 (4S,6S)-4-Dimethyl-tert-butylsilyloxy-6[(dimethyl-tert-butylsilyloxy)methyl]-tetrahydro-2H-pyran-2-one,190

E (+)-Eldanolide, 185- 186 8-Elimination aldono-1,Clactones, 166- 170 aldono-l,5-lactones, 162- 166 Elongation, peptide chain following selectivedeprotection, 294- 298 N-glycopeptides, 283 -287 Emulsin, 8, 12 effect on glucosides, 13 - 14 lack of effect on methyl xylosides, I3 Enonolactones /3-elimination, 162- 165 catalytic hydrogenation, 170- 173 Enzymes active site probing, 13- 14 introduction of term, 8 specificity, 13- 15 ~-6-Epicastanospermine,198- 199 /3-Epinoroxetanocin, 196 ~-Erythrono-l,4-lactone, 131 L-Erythrose, synthesis, 130- 131 Escherichiu coli, lipid A chemical structure, 214-215 Esters, formation, aldonolactones reaction with alcohols, 148- 149

Etherification,aldonolactones, 132- 134 Ethynyl compounds, reaction with sugar lactones, 139- 140

F Fatty acids, 228-247 amide-bound, 238-239 ester-bound, 236-238 hexaacyl lipid A, 24 1- 242 (S)-2-hydroxylated, 230- 232 (R)-3-hydroxylated, 229 - 230 linkage to lipid A backbone, 235-239 location at lipid A backbone, 239-240 nonhydroxylated, iso-, and anti-iso, 234 other hydroxylated, 232 3-0X0,232-234 primary and secondary, in lipid A, 246 unsaturated, 235 Fermentation glycosides, 10- 1 1 SUW, 8- 10 Fischer, Emil asymmetric induction concept, 2 - 7 probing of enzyme active site, 13- 14 yeast fermentations, 7 -9 Fischer-Helferich method, 23 Fischer- Kiliani synthesis, 6 Fluorine, as leaving group, 21 -22 Formylaminomethylenation, aldonolactones, 146- 148 0-Fucopyranosyl trichloroacetimidates, inverse procedure for glycosylation, 98, 104- 107 Fucose, trichloroacetimidates reaction with N-nucleophiles, 105- 107 synthesis, 98, 104 ~-L-Fuco units, s ~ ~attachment, in situ anomerization, 280-281 2( 5H)-Furanones, &elimination, 166- 170 G Galactosamine trichloroacetimidates glycosylation, 98- 102 as glycosyl donors, 84,92- 102 O-Galactosyl trichloroacetimidates acetylated, glycosylation with, 53, 56-57 benzylated, glycosidation, 53- 55 as donors, 49, 52 - 59

SUBJECT INDEX, VOLUME 50 glycosylation with sphingosine derivatives, 58-59 synthesis, 52-53 &~-Galf-(1 5)-D-Galf; 180- 18 I Glycosylation, 0-fucopyranosyl trichloroacetimidates, inverse procedure, 98, 104- 107 Glucoamylase, 16 Gluconamides, 151 a-D-Glucopyranoside,hydrolysis by glucoamylase, 17- 18 Glucosamine as donors, 61,68-77 trichloroacetimidates,as glucosyl donors, 60-61,68-91 chitobiose donors, 8 1, 89 glucosamine donors, 6 1,68- 77 lactusamine donors, 73,78-88 muramic acid donors, 8 1, 84,90 - 9 1 D-Glucosamine, trichloroacetimidates,synthesis, 68 - 7 1 dextro-Glucose, structure, 5 D-Glucose, trichloroacetimidates,synthesis, 27 - 29 Glucosides emulsin effect, 13- 14 synthesis, 2 1 - 1 17 anomeric 0-alkylation, 23 - 25 P-D-Glucosiduronates, synthesis, 1 1 1, 113 0-clucosyl trichloroacetimidates acetylated, glycosides and saccharides from, 4 1,43-44 benzyl-protected, 35 - 39 as donors, 34-49 acid catalyst, 35 glycosylation, 39-40 azidosphingosine derivatives, 4 1,4548 ceramides, 49 - 5 1 solvent effects, 35, 39 synthesis, 34-35 Glucuronic acid, trichloroacetimidates, 1 1 1, 113 Glycoconjugates, biological significance,2 1 Glycolipids, synthesis, 151- 152 Glycopeptide n-heptyl esters, lipase-catalyzed cleavage, 306-307 Glycopeptides, 277 - 307 binding to proteins, 298-299 synthesis, 277- 307

-

38 1

enzymes as tools, 303- 306 lactosamine importance, 73 solid-phase, 299 - 303, 306 N-Glycopeptides, 278 -287 2-acetamido-2-deoxy-~~-glucosylLasparagine linkage, 278-283 selective deprotection and peptide-chain elongation, 283-287 3-O-Glycopeptides,L-serine or L-threonine, 287-298 selective deprotection methods, 294 - 298 3-0-glycosidic linkage formation, 287 294 G1ycoproteins with multiple 0-glycosylation sites, 344 N- and 0-linked chains, peripheral substitutions, 325,239-332 oligosaccharide determinants, 3 1 1 - 345 backbones and core regions N- and 0-linked chains, secreted and plasma membrane glycoprotein, 315-325 'H-NMR-spectral and mass-spectrometric analysis, 322- 328 mucin oligosaccharide structure, 3 16-322 conformations and molecular recognition determinants adjacent to protein moiety, 343 - 345 determinants distant from core, 332342 blood-group-relatedantigens, 333337 sialylated oligosaccharide determinants, 337 - 342 purification and proliling, 3 14- 3 15 structural analysis methods, 3 11- 3 14 sulfated oligosaccharide chains, 'HNMR - spectral and mass-spectrometric analysis, 330-331 Glycosidation, benzylated 0-galactosyl trichloroacetimidates, 53 - 55 G1ycosides from acetylated glucosyl trichloroacetimidates, 41,43-44 fermentation, 10- 11 from mannosyl trichloroacetimidates, 58, 62-65 0-Glycosides, synthesis, 296 -297

382

SUBJECT INDEX. VOLUME 50

3-0-Glycosidic linkage formation, 287 -294 1-0-acetyl activation, 292 -293 electrophile-inducedlactonization of glycosy1 4-pentenoates, 293-294 glycosyl fluorides, 290 Koenigs-Knorr methods, 287-289 1-thioglycosides, 290-292 trichloroacetimidate method, 289 -290 GIycosphingolipids N-acetylgalactosamine-containing,structures, 84, 92 synthesis, lactosamine importance, 73 GIycosylation acetylated 0-galactosyl trichloroacetimidates, 53, 56-57 acetylated 0-mannopyranosyl trichloroacetimidates, 58,66-67 aldonolactones, 179 - 181 azidosphingosine derivatives, with trichloroacetimidates,4 1,45 -48 ceramides, by trichloroacetimidates,49 51 chitobiose derivatives, 8 1, 89 galactosaminetrichloroaetimidates, 98 102 0-galactosyl trichloroacetimidates,with sphingosine derivatives, 58 -59 0-glucosyl trichloroaetimidates, 39 -40 glycopeptide synthesis, 303-306 solvent effects, 35,39 steps, 26 trichloroacetimidatesof 2-azido-2deoxyD-mannose derivatives, 98, 103 Glycosyl carboxylates, base-catalyzed addition, I16 Glycosyl fluorides, 3-0-glycosidic linkage formation, 290 Glycosylic phosphate, lipid A backbone, 221 -222 Glycosyl imidates, base-catalyzed addition, 114, 116 Glycosyl oxides, base-catalyzed addition, 114 Glycosyl4-pentenoates, electrophile-induced lactonization, 293 - 294 Glycosyl phosphates, anomeric-oxygen activation, 116 Glycosyl region, lipid A backbone, 2 18 -22 1 Glycosyl sulfonates, base-catalyzed addition, 116

0-Glycosyl trichloroacetimidates, reactions with N-, S-,C-, and P-acceptors, 1 1 1, 114- 115 Glycoyl a-phosphate, lipid A substituents, 226-227 9-/3-~-Gulofuranosyladenine, 195

H Hanganutziu-Diecher antigen, 339,342 Hard-sphere exoanomeric algorithm, 333335 Hept-2-enono-1,4-1actone, 167 Heterocycles, lithiated, addition to aldonolactones, 138- 139 Hex-2-enono-I ,4-lactone 2-mesylates, 167 Hydrazine, lactone group reaction with, 155-157 Hydrogenation, catalytic, enonolactones, 170-173 Hydrogen bromide, reaction with aldonolactones, 134- 136 Hydrogen01ysis, bromodeoxyaldono-1,4-lactones, 173- 174 Hydrolysis enzymic, structure and configuration effects, 12 a-D-glucopyranoside by glucoamylase, 17- 18 a-Hydroxy acids, cyanohydrin synthesis, 6 8-(Hydroxyalkyl)adenines,153 1-Hydroxy-D-glycosamines,reactions with ammonium hydrogencarbonate, 282 283 Hydroxyl group, lipid A backbone position 4', 247-249 position 6', 249-252 I Induction, asymmetric, 2-7 discovery, 6 - 7 Invertase, 11 Invertin, 8, 12 1-Isocyanates,as precursors, 28 1-282 Isopropylidenation, aldonolactones, 127 2,3-~-Isopropylidene-~-ribonoI ,Cladone, 129

SUBJECT INDEX, VOLUME 50

K Ketones, aldonolactones, 127- 130 Kinetic anomeric effect, 29 Koenigs-Knom method, 21 3-0-glycosidic linkage formation, 287289 L Lactams, from aldonolactone intramolecular reaction, 154- 155 Lactase, 12 Lactone group, reaction with hydrazine and derivatives, 155- 157 Lactonization, electrophile-induced,glycosyl-4-pentenoates, 293 - 294 Lactosamine 2-azido-2-deoxytrichloroacetimidates,reaction with nucleophiles, 8 1, 85 as donors, 73,78-88 oligomers, synthesis, 8 1, 84 (+)-Lineatin, synthesis, 193 Lipid A, 2 1 1 - 265 amphiphilic nature, 2 13 amphoteric nature, 2 13 antibodies, 259-260 backbone, 2 16- 225,260 glycosyl region, 2 18-22 1 hydroxyl group at position 4’, 247 249,26 1 -262 hydroxyl group at position 6‘, 249252,262 phosphate groups, 221 -225 principles and structures, 216 structural elucidation, 2 17-2 18 conformation, 252-256 definition, 2 13 endotoxic activity, 263 - 264 fatty acids, see Fatty acids heptaacyl, 243 hexaacyl, 241 -242 as lipopolysaccharideendotoxic center, 256-258 molecular shape, 253-254 pentaacyl, 243-245 phase states, 258 phosphate substituents, 225 -227 properties, 2 13 - 2 14 serology, 258-260

383

structure, 260-26 1 chemical, 214-215 primary, 214-216 synthetic, 252 tetraacyl, 244 three-dimensional organization, 256-258 Lipopolysaccharides, 21 1-265, see also Lipid A core oligosaccharide, 2 12 lipid A as endotoxic center, 256-258 0-specific chain, 2 12 structural principles, 2 11 - 2 12 Lithium divinylcyanocuprate, 1,Cconjugate addition, 184- 185 Lithiumtrimethylsilylacetate, C- 1 elongation of aldonolactones, 144 Lock and key concept, 9- 13 9-a-~-Lyxofuranosyladenine, I 95 M 0-Mannopyranosyl trichloroacetimidates acetylated, glycosylation, 58,66 -67 as donors, 58,60-67 glycosides and saccharides from, 58, 62-65 reactions with, 58,6 1 Mannosamines, trichloroacetimidates, as glycosyl donors, 98, 103 Mass spectrometry mucin oligosaccharide chains, 322-328 oligosaccharide physicochemical analysis, 312-313 2’-C-Methyladenosine, 195 Methylation, oligosaccharide of glycoconjugates analysis, 3 1 1- 3 13 Methylenation, aldonolactones, 143- I46 Methyl D-gdactofuranosyl-(1 -+6)-/?-~galactofuranoside, 180 Methyl a-D-glucopyranoside, 16- 17 Methyl oxetane-2-carboxylate, 193 Mucins backbone and core sequences, 3 16- 322 oligosaccharide structure, 3 I6 - 322 IH-NMR- spectral and mass-spectrometric analysis, 322 - 328 Muramic acid, glycosides, synthesis, 8 1, 84, 90-91 Muramic acid, as donor, 8 1,84,90-9 1

384

SUBJECT INDEX, VOLUME 50

L-MYCXO~~, 158- 159 Mycoplasm pneumoniae, binding specificity, 339 N (+)-Negamycin, 201 Neplanocin A, 196 Nojirimycin, 154- 155 Nonglycosylicphosphate, lipid A backbone, 223- 224 j?-Noroxetan&n, 196 'H-Nuclear magnetic resonance, oligosaccharide physicochemical analysis, 3 12, 314 mucin oligosaccharidechains, 322-328 sulfated oligosaccharide chains, 330,333 0-Nucleophiles, alcohols and sugars as, 321 14; see also O-Glucosyl trichloroacetimidates, as donors 0-galactosyl trichloroacetimidates,49, 52-59 trichloroacetimidates of 6-deoxyhexoses, 98, 104- 110 of galactosaminederivatives, 84,92 102 of glucosamine derivatives, see Glucosamines, trichloroacetimidates of mannosamine derivatives, 98, 103 0-mannopyranosyl trichloroacetimidates, 58,60-67 Nucleophilicity, enhanced, /.?-oxides,29 Nucleosides, synthesis from aldonolactones, 195-201 0

Oligosaccharide chains, 0-linked, glycoproteins, 322, 324 Oligosaccharides conformations, 335 - 337 glucosyl trichloroacetimidates,reaction, 41 -42 with Le" determinants, synthesis, 8 1, 86 88 much structure, 3 16- 322 'H-NMR-spectral and mass-spectrometric analysis, 322 - 328 physiochemical analysis, 3 11 - 3 14 recognition specificity of proteins, 17

sequences related to Le" and Lebantigens, 337,340-341 sialylated, determinants, 337 - 342 Olivomycose, 158- 159 Organic chemistry, in 19th century, 5 Organolithium reagents, reaction with aldonolactones, 138- 143 Organomagnesium reagents, reaction with aldonolactones, I38 - 143 Orthoesters, derivatives, formation, aldonolactones reaction with alcohols, 149- 151 /.?-Oxides,enhanced nucleophilicity, 29

P Pasteur, Louis, tartaric acid preparation, 3 Penten-5-olides, 190 3-(~-gluco-Pentitol-1-yl)-Sphenyl-1,2,4-trazol[3,4-a]phthalazine, 156 Pentoses, trichloroacetimidates, 111 Peptide chain, elongation following selective deprotection, 294-298 N-glycopeptides, 283-287 Phosphate groups, lipid A backbone, 22 1 -225 substituents, lipid A, 225-227 4'-Phosphate, non-glycosylic,lipid A substituents, 226 2-N-Phthaloyl-2deoxytrichloroacetimidate, reaction with nucleophiles, 8 1 - 83 2-N-Phthaloyl trichloroacetimidate, reaction with nucleophiles, 68,72-76 2-Polyhydroxyalkylimidazoles,152 Proteins glycopeptide binding, 298-299 oligosaccharide recognition, specificity, 17 Pseudomonas aeruginosa, lipid A, 244- 245 cdyrones, /.?-elimination,165- 166

Q 0-Quinovopyranosyl trichloroacetimidates, 105, 110 Quinovosides, 105

R Ranunculin, synthesis, 183 Reformatsky-type reactions, aldonolactone carbonyl group, 136- 138

SUBJECT INDEX, VOLUME 50 0-Rhamnopyranosyl trichloroacetimidates, 105, 108-109 Rhamnosides, 105 Rhodinose, 158 Rhodobacter capsulatus, lipid A, 232-234 Rhodobacter sphaeroides, lipid A, 232 - 234 L-Ribofuranose, derivatives, synthesis, 132 Ribonolactone, p-elimination, I89 D-Ribonolactone 0-benzylidene derivatives, physical constants for reassigned structures, 126127 syntheses from, 192- 195 S Saccharides, synthesis, 21 -22 from acetylated glucosyl trichloroacetimidates, 4 1,43- 44 from mannosyl trichloroacetimidates, 58, 62-65 Salmonella typhimurium, lipid A, 217 Serology, lipid A, 258-260 Sialic acid, substitution patterns, 'H-NMR analysis, 331 -332 Sodium amalgam, aldonolactone reduction, I57 Somatostatin, glycosylated,299 Sphingosines, derivatives, trichloroacetimidate glycosylation with, 58-59

385

1-Thioglycosides, 3-0-glycosidic linkage for-

mation, 290-292 2,5,6-Tri-0-acetyl-3-deoxy-~-rib~hexono1,4-lactone, 176 Tri-0-benzylfucosyl donor, 98, 104 Trichloroacetimidate, 0-glycosyl, formation, 25 - 30 PTrichloroacetirnidate, a-selective glycosidation, 13, 77 Trichloroacetimidate method, 25 - 115 alcohols and sugars as 0-nucleophiles, see 0-Nucleophiles kinetic anomeric effect, 29 3-0-glycosidic linkage formation, 289 290 0-glycosyl trichloroacetimidate formation, 25 - 30 reaction with Brernsted acids, 30-32 2,2,2-Trichloroethoxycarbonylgroup, 286287

U Umbelactone, 188- 189

V van't Hoff-Le Be1 theory, asymmetric carbon atom, 3 -4

swFs

derivatives, synthesis, aldonolactones acetals use, 130-132 fermentation, 8 - 10 labeled at anomeric center by aldonolactones, 161- 162 as 0-nucleophiles, see 0-Nucleophiles relative configurations,proof by Fisher, 2

T Thioglycosides,as glycosyl donors, 22

X Xylosides, methyl, lack of effect by emulsin, 13

Y Yeast enzyme presence, 11 fermentations, 7 - 9

CUMULATIVE AUTHOR INDEX FOR VOLS. 46-50 Ions with Carbohydrates: Use of Relaxation Probes, 47, 125- 166

A J., Complexes of Metal ANGYAL, STEPHEN Cations with Carbohydrates in Solution, 47, 1-43; The Composition of Reducing Sugars in Solution: Current Aspects, 49, 19 - 35 AuGB, CLAUDINE. See David, Serge

B BIERMANN, CHRISTOPHER J., Hydrolysis and Other Cleavages of Glycosidic Linkages in Polysaccharides, 46,25 1 27 1

BLEHA,TOMAS.See TvaroSka, Igor C

CARTER, DOUGLAS R. See Dill, Kilian CLARKE, RONALD J., COATES, JOHNH., F., Inclusion and LINCOLN, STEPHEN Complexes of the Cyclomalto-oligosaccharides (Cyclodextrins), 46,205 - 249; Addendum, 46, 333-335 COATES, JOHNH. See Clarke, Ronald J. CSUK,RENB,and GLANZER, BRIGITTE I., N.M.R. Spectroscopy of Fluorinated Monosaccharides, 46,73 - 177; Addendum, 46,331 -332

D

F FERRIER, ROBERT J. See Somdk, Liszl6

G GARG,HARIG., and LYON,NANCYB., Structure of Collagen Fibril-Associated, Small Proteoglycans of Mammalian

Origin,49,239-261 GARG,HARIG., VON DEM BRUCH,KARSTEN,and KUNZ,HORST,Developments in the Synthesis of Glycopetides Containing Glycosyl L-Asparagine, LSerine, and L-Threonine, 50,271 - 3 10 GAUTHERON, CHRISTINE. See David, Serge GLANZER, BRIGITTE I. See Csuk, Rent G. See de Lederkremer, GROS,EDUARDO Rosa M. H

HARDING, STEPHEN E., The Macrostructure of Mucus Glycoproteins in Solution, 47,345-381

HEMMER, REINHARD. See Stoss, Peter HICKS,KEVINB., High-Performance Liquid Chromatography of Carbohydrates, 46, 17 - 72; Addendum, 46,327 - 329 HORTON,DEREK.,[Obituary of] R. Stuart Tipson, 50, xiii-xxi; see also Tsuchiya, Tsutomu HOUNSELL, ELIZABETH F., Physicochemical Analyses of Oligosaccharide Determinants of Glycoproteins, 50,3 1 1 - 350

DAVID,SERGE,AuGB, CLAUDINE, and GAUTHERON, CHRISTINE, Enzymic Methods in Preparative Carbohydrate Chemistry, 49, 175-231 DE LEDERKREMER, ROSAM., and GROS, EDUARDO G., [Obituary of] Venancio Deulofeu, 46, 1 1 - 15 K DE LEDERKREMER, ROSAM., and VARELA, OSCAR,Synthetic Reactions of AldonoKASHIMURA, NAOKI.See Komano, Tohru KINZY,WILLY.See Schmidt, Richard R. lactones, 50, 125 -209 DILL,KILIAN,and CARTER,R. DOUGLAS, KNIREL,YURIYA,, VINOGRADOV, EVGENY ')C-Nuclear Magnetic Resonance-SpecV., and MORT,ANDREWJ., Applicatral Studies of the Interactions of Metal tion of Anhydrous Hydrogen Fluoride

* Startingwith Volume 30, a Cumulative Author Index covering the previous 5 volumes has been published in every 5th volume. That listing the authors of chapters in Volumes 1 - 29 may be found in Volume 29. 386

CUMULATIVE AUTHOR INDEX FOR VOLS. 46-50 in the Structural Analysis of Polysaccharides, 47, 161-202 KOMANO, TOHRU,and KASHIMURA, NAOKI,[Obituary ofj Konoshin Onodera, 46, 1 -9 KUNZ,HORST.See Garg, Hari G.

L LEGLER, GUNTER,Glycoside Hydrolases: Mechanistic Information from Studies with Reversible and Irreversible Inhibitors, 48, 3 19- 384 LEMIEUX, R. U.,and SPOHR,U., How Emil Fischer Was Led to the Lock and Key Concept for Enzyme Specificity,50, 1 20 LINCOLN, STEPHEN F. See Clarke, Ronald J. LINDBERG, BENGT,Components of Bacterial Polysaccharides,48,219- 3 18 LINDNER, BUKO.See Ziihringer, Ulrich LIPTAK,ANDRAS,Ninfisi, Pa, and Sztaricskai, Ferenc, [Obituary of] Rezsd Bognk,49,3-9 LYON,NANCYB. See Garg, Hari G.

M

MAEDA,KENJI.See Tsuchiya, Tsutomu MORT,ANDREWJ. See Knirel, Yuriy A.

N NANAsI,PAL. See Liptik, Andris NELSON,DAVIDA. See Theander, Olof 0

OGAWA, SEIICHIRO. See Suami, Tetsuo

P PALASINSKI, MIECZYS~AW. See Tomasik, Piotr PERCHERON, FRANCOIS, [Obituary of] Jean Emile Courtois, 49, 1 1 - 18

387

R RIETSCHEL, ERNSTTH. See Ziihringer, Ulrich

S SCHMIDT, RICHARD R., and KINZY, WILLY,Anomeric-Oxygen Activation for Glycoside Synthesis: The Trichloroacetimidate Method, 50,21 - 123 SOMSAK,LAsz~6,and FERRIER, ROBERTJ., Radical-Mediated Brominations at Ring Positions of Carbohyrdrates, 49, 37-92 SPOHR,U. See Lemieux, R. U. STOSS,PETER,and HEMMER, REINHARD, 1,4:3,6-Dianhydrohexitols, 49,93- 173 SUAMI,TETSUO,and OGAWA,SEIICHIRO, Chemistry of Carba-Sugars (PseudoSugars) and Their Derivatives, 48,21 90 SZTARICSKAI, FERENC. See Liptik, Andris.

T THEANDER, OLOF,and NELSON,DAVIDA. Aqueous, High-Temperature Transformation of Carbohydrates Relative to Utilization of Biomass, 46,273- 326 TOMASIK, PIOTR,PALASINSKI, MIECZYSLAW,and WIEJAK,STANISLAW, The Thermal Decomposition of Carbohydrates. Part I. The Decomposition of Mono-, Di-,and Oligosaccharides,47, 203-278 TOMASIK, PIOTR,WIEJAK,STANISLAW, and PALASINSKI, MIECZYS~AW, The Thermal Decomposition of Carbohydrates. Part 11. The Decomposition of Starch, 47,219-343 TSUCHIYA, TSUTOMU, Chemistry and Developments of Fluorinated Carbohydrates, 48, 91 -271. TSUCHIYA, TSUTOMU, MAEDA,KENJI,and HORTON,DEREK,[Obituary of] Hamao Umezawa, 48, 1-20 TVAROSKA, IGOR,and BLEHA,TOMAS,Anomeric and Exo-anomeric Effects in Carbohydrate Chemistry, 47,45- 123

388

CUMULATIVE AUTHOR INDEX FOR VOLS. 46-50 V

VARELA, OSCAR.See de Lederkremer, Rosa M. VINOGRADOV, EVGENY V. See Knirel, Yuriy A. VON DEM BRI-H, KARSTEN.See Garg, Hari G. W

WEIJAK,STANISLAW. See Tomasik, Piotr

Z

Z~~HRINGER, ULRICH,LINDNER, BUKO,and RIETSCHEL, ERNSTTH., Molecular Structure of Lipid A, the Endotoxic Center of Bacterial Lipopolysaccharides, 50,2 1 1 - 276 ZEHAVI,URI, Applications of Photosensitive Protecting Groups in Carbohydrate Chemistry, 46, I79 -204

CUMULATIVE SUBJECT INDEX FOR VOLS. 46-50 A

Collagen fibril-associatedsmall proteoglycans of mammalian origin, structure of, 49, 239-261 Complexes of metal cations with carbohydrates in solution, 47, 1-43 Courtois, Jean Emile, obituary of, 49, 1 1- 18 Cyclomalto-oligosaccharides(cyclodextrins), inclusion complexes of, 46,205-249 addendum to, 46,333-33s

Aldonolactones, synthetic reactions of, 50, 125-209 Anhydrous hydrogen fluoride, application of for the structural analysis of polysaccharides, 47, 167-202 Anomeric and exo-anomeric effects in carbohydrate chemistry, 47,45- 123 Anomeric-oxygen activation for glycoside synthesis, trichloroacetimidate method, 50,21-123

D Decomposition of mono-, di-, and oligosaccharides, 47,203-278 Decomposition of starch, 47,279-343 Deulofeu, Venancio, obituary of, 46, 1 1 - 15 1,4:3,6-Dianhydrohexitols, 49, 93- 173 Disaccharides, decomposition of, 47,203278

B Bacterial polysaccharides, components of, 48,279-318 Bognir, RezsB, obituary of, 49, 3 - 9

C Carba-sugars (pseudo-sugars) and their derivatives, chemistry of, 48,21-90 Carbohydrate chemistry anomeric and exo-anomeric effects in, 47, 45- 123 applications of photosensitive protecting groups in, 46, 179-204 Carbohydrates aqueous, high-temperature transformation of relative to utilization of biomass, 46,273- 326 with complexes of metal cations in solution, 47, 1-43 "C-nuclear magnetic resonance-spectral studies of the interactions of metal ions with, 47, 125-166 high-performance liquid chromatography of, 46, 17-72 addendum to 46,327- 329 radical-mediated brominations at ring positions of, 49, 37-92 thermal decomposition of, 47,203-278 '3C-nuclear magnetic resonance-spectral studies of the interactions of metal ions with carbohydrates, 47, 125- 166

E Enzymic methods in preparative carbohydrate chemistry, 49, 175-237 Enzyme specificity,how Emil Fischer was led to the lock and key concept for, 50, 1-20 F Fluorinated carbohydrates, chemistry and developments of, 48,91-277 Fluorinated monosaccharides, N.M.R. spectroscopy of, 46,73- 177 addendum to, 46,33 1-332 G Glycopeptides containing glycosyl L-asparagine, L-serine, and L-threonine, developments in, 50,277- 3 10 Glycoside hydrolases, mechanistic information from studies with reversible and irreversible inhibitors, 48,319- 384 Glycoside synthesis, anomeric-oxygen activation for (trichloroacetimidate method), 50,21 - 123

* Startingwith Volume 30,a Cumulative Subject Index covering the previous 5 volumes has been published in every 5th volume. That listing the chapters in Volumes 1 -29 maybe found in Volume 29. 389

390

CUMULATIVE SUBJECT INDEX FOR VOLS. 46-50

H High-performanceliquid chromatography of carbohydrates, 46, 17- 72 addendum to, 46,327 - 329

Inclusion complexes of cyclomaltc-oligosaccharides (cyclodextrins), 46,205 -249 addendum to, 333-335

L Lipid A, molecular structure of, 50,2 11 -276 Liquid chromatography, high-performance, of carbohydrates, 46,17-72 addendum to, 46,327- 329 Lock and key concept for enzyme specificity, how Emil Fischer was led to, 50, 120

Oligosaccharides,decomposition of, 47, 203-278 Onodera, Konoshin, obituary of, 46, 1- 9 P Photosensitiveprotecting groups in carbohydrate chemistry, applications of, 46, 179-204 Physicochemical analyses of oligosaccharide determinants of glycoproteins, 50,3 11350 Polysaccharides application of anhydrous hydrogen fluoride for the structural analysis of, 47, 167-202 hydrolysis and other cleavages of glycosidic linkages in, 46,25 l -27 l Preparative carbohydrate chemistry, enzymic methods in, 49, 175- 237

R M Macrostructure of mucus glycoproteins in solution, 47,345 - 38 1 Metal cations, complexes of, with carbohydrates in solution, 47, 1-43 Molecular structure of lipid A, 50,211-276 Monosaccharides, decomposition of, 47, 203-278 Mucus glycoproteins, macrostructure of, in solution, 47,345 - 38 1

N N.M.R. spectroscopy of fluorinated monosaccharides, 46,73 - 177 addendum to, 46,33 1 - 332 0

Oligosaccharide determinants of glycoproteins, physicochemical analyses of, 50, 311-350

Radical-mediated brominations, at ring positions of carbohydrates, 49,37-92 Reducing sugars in solution, composition of, 49, 19-35 S

Starch, decomposition of, 47,279-343 Synthetic reactions of aldonolactones, 50, 125-209 T Thermal decomposition of carbohydrates, 47,203-278 Tipson, R. Stuart, obituary of, 50,xiii-xxi Trichloroacetimidate method, anomeric-oxygen activation for glycoside synthesis, 50.2 1- 123

U Umezawa, Hamao, obituary of, 48, 1-20

This Page Intentionally Left Blank

I S B N 0-12-007250-5 90051