摘要:

Subject Index

ISSN:0265-0568    

DOI:10.1039/NP99512000I1

出版商:RSC    

年代:1995

数据来源: RSC

摘要:

EditoriaI Natural Product Reports is now eleven years old and at regular intervals since its inception the Editorial Board has reviewed its role and areas of coverage. At the outset the journal took over the regular and comprehensive coverage of the area of natural product chemistry then covered by the Specialist Periodical Reports titles ‘Biosynthesis’ ‘The Alkaloids ’ ‘Ter-penoids and Steroids’ and ‘Aliphatic and Related Natural Product Chemistry’. In addition to the perceived advantages of providing a more regular and rapid publication of individual reviews we sought to provide articles of general interest across the whole range of natural products such as updates on developments in chromatographic and spectroscopic methods. At the same time we aimed to bring to the attention of our readership many of the new developments in enzymology genetics and the many other areas which have come collectively to be known as ‘bioorganic chemistry’.There is no questioning the burgeoning importance of bio- organic chemistry in the past ten years and we have had the increasing conflict between maintaining our original remit to provide a basic coverage of ‘natural products’ which we believe is essential for our long-standing and loyal subscribers and yet at the same time reflecting the increasing importance of ‘bioorganic’ disciplines to attract a wider readership. As all will be aware there are ever-present and increasing constraints on production costs of a specialist journal such as NPR and the Editorial Board with the help of the RSC Editorial Staff has constantly sought ways of squeezing a quart into a pint pot e.g.by reduction of structure sizes more efficient use of page lay-outs and by persuading our long- suffering authors on whom we all depend to say more in less space! The Board came to the conclusion that the only way in which we could include more of the essential coverage of developments in bioorganic chemistry was to seek an overall increase in the length of the journal. I am pleased to say that the RSC has agreed to this while keeping any consequent increase in subscription rates to a minimum. At the same time we have redesigned the cover for 1995 to improve the appearance of the journal and to provide more immediate access to information on the contents of each issue. We hope these changes will meet with the approval of our readers. The Editorial Board would be pleased to receive formal or informal comments on ways in which our ‘customers’ think the journal can continue to evolve and better serve the needs of the natural product and bioorganic community. T. J. Simpson

ISSN:0265-0568    

DOI:10.1039/NP995120X003

出版商:RSC    

年代:1995

数据来源: RSC

ISSN:0265-0568    

DOI:10.1039/NP99512BX004

出版商:RSC    

年代:1995

数据来源: RSC

摘要:

ISSN 0265-0568 NPRRDF 12 1-1-1-78 (1995) Natural Product Reports A journal of current developments in bioorganic chemistry Volume 12 Indexes CONTENTS ... 111 Preliminary pages for Volume 12 1-1 Subject Index 1-37 Index of Authors Cited ISSN 0265-0568 Coden NPRRDF Natural Product Reports A journal of current developments in bio organic chemistry Volume 12 1995 The Royal Society of Chemistry Cambridge Natural Product Reports (ISSN 0265-0568) @ The Royal Society of Chemistry 1996 All Rights Reserved No part of this publication may be reproduced stored in a retrieval system or transmitted in any form or by any means electronic mechanical photographic recording or otherwise without the prior permission of the publishers.Printed in Great Britain by the University Press Cambridge ISSN 0265-0568 NPRRDF 12 1-660; 1-1-1-78 (1995) Natural Product Reports A journal of current developments in bioorganic chemistry Volume 12 CONTENTS Editorial T. J. Simpson 1 The Biosynthesis of Fatty Acid and Polyketide Metabolites David O’Hagan Reviewing the literature published between mid-1992 and mid-1993 33 Modern Mass Spectrometry in Bioorganic Analysis M. A. Baldwin 45 The Chemistry of Heterocyclic o-Quinone Cofactors Shinobu Itoh and Yoshiki Ohshiro 55 The Biosynthesis of Plant Alkaloids and Nitrogenous Microbial Metabolites R. B. Herbert Reviewing the literature published in 1992 69 The Chemistry of Macrocyclic Bis(bibenzy1s) G. M. Keseru and M. Nogradi 77 Quinoline Quinazoline and Acridone Alkaloids J.P. Michael Reving the literature published between July 1992 and June 1993 91 Book Reviews Phytochemical Dictionary of the Leguminosae ed. F. A. Bisby J. Buckingham and J. B. Harborne (reviewed by Peter G. Waterman); Natural Products Their Chemistry and Biological SignlJicance by J. Mann R. S. Davidson J. B. Hobbs D. V. Banthrope and J. B. Harborne (reviewed by David O’Hagan) 93 Electron Transfer in Proteins Stephen K. Chapman and Andrew R. Mount 101 The Biosynthesis of Shikimate Metabolites P. M. Dewick Reviewing the literature published in 1993 135 Muscarine Oxazole Imidazole Thiazole and Peptide Alkaloids and Other Miscellaneous Alkaloids J. R. Lewis Reviewing the literature published between July 1992 and June 1993 165 Polyether Ionophores C.J. Dutton B. J. Banks and C. B. Cooper I83 Lignans Neolignans and Related Compounds R. S. Ward Reviewing the literature published in 1992 and 1993 207 Diterpenoids J. R. Hanson Reviewing the literature published in 1993 219 Book Reviews Biotransformations (Volume 6),ed. D. R. Hawkins (reviewed by D. H. G. Crout); Organic Synthesis Based on Name Reactions and Unnamed Reactions by A. Hassner and C. Stumer (reviewed by R. W. Alder) 22 1 Erratum 223 Marine Natural Products D. J. Faulkner Reviewing the literature published in 1993 271 Feverfew Chemistry and Biological Activity David W. Knight 277 Recent Progress in the Chemistry of Non-monoterpenoid Indole Alkaloids M. Ihara and K. Fukumoto Reviewing the literature published between July 1993 and June 1994 303 Natural Sesquiterpenoids B.M. Fraga Reviewing the literature published in 1993 321 Isoflavonoids and Neoflavonoids Naturally Occurring 0-Heterocycles D. M. X. Donnelly and G. M. Boland Reviewing the literature published between 1991 and 1993 339 Amaryllidaceae and Sceletium Alkaloids J. R. Lewis Reviewing the literature published in 1993 347 Book Reviews Enzymes in Synthetic Organic Chemistry by Chi-Huey Wong and George M. Whitesides (reviewed by C. L. Willis) ;Isopentenoids and Other Natural Products :Evolution and Function ed. W. David Nes (reviewed by J. MacMillan); Radical Chemistry by M. J. Perkins (reviewed by John A. Murphy) 349 The Phytochemistry of the Yew Tree Giovanni Appendino Reviewing the literature published between March 1992 and September 1994 361 Diterpenoid and Steroidal Alkaloids Atta-ur-Rahman and M.Iqbal Choudhary Reviewing the literature published between 1992 and 1994 381 The Viridin Family of Steroidal Antibiotics James R. Hanson 385 Recent Progress in the Chemistry of the Monoterpenoid Indole Alkaloids Reviewing the literature published between July 1993 and June 1994 413 Pyrrolizidine Alkaloids D. J. Robins Reviewing the literature published between July 1993 and June 1994 419 /3-Phenylethylamines and the Isoquinoline Alkaloids K. W. Bentley Reviewing the literature published between July 1993 and June 1994 J. E. Saxton 443 Book Reviews Modern Methods in Plant Analysis (Volume 15) Alkaloids ed.H. F. Linskens and J. F. Jackson (reviewed by Peter G. Waterman); The Practice of Peptide Synthesis (2nd Edition) by M. Bodanszky and A. Bodanszky (reviewed by Patrick Bailey) ;Principles of Bioinorganic Chemistry by Stephen J. Lippard and Jeremy M. Berg (reviewed by C. David Garner) 445 The Biosynthesis of Plant Alkaloids and Nitrogenous Microbial Metabolites Reviewing the literature published in 1993 465 Quinoline Quinazoline and Acridone Alkaloids J. P. Michael Reviewing the literature published between July 1993 and June 1994 477 Coumarins R. D. H. Murray Reviewing the literature published between 1988 and 1994 507 The Biosynthesis of C,-C, Terpenoid Compounds P. M. Dewick Reviewing the literature published between 1990 and 1992 535 Indolizidine and Quinolizidine Alkaloids J.P. Michael Reviewing the literature published between July 1993 and June 1994 R. B. Herbert 553 Book Reviews Organic Reactivity Physical and Biological Aspects ed. B. T. Golding R. J. Griffin and H. Maskill (reviewed by A. Williams); The Biochemistry of the Stilbenoids by John Gorman (reviewed by M. Nogradi) 555 The Biosynthesis of Pyridoxine Ian D. Spenser and Robert E. Hill 567 Steroids Reactions and Partial Synthesis James R. Hanson Reviewing the literature published in 1993 579 The Biosynthesis of Shikimate Metabolites P. M. Dewick Reviewing the literature published in 1994 609 Triterpenoids J. D. Connolly and R. A. Hill Reviewing the literature published between January 1992 and December 1993 639 Anthocyanins and Other Flavonoids J.B. Harborne and C. A. Williams Reviewing the literature published between January 1992 and December 1994 659 Book Reviews :Bioinorganic Chemistry :Inorganic Chemistry in the Chemistry of Life. An Introduction and Guide by W. Kaim and B. Schwederski (reviewed by F. A. Armstrong); Microbes Bugs and Wonder Drugs by Fran Balkwill and Mic Rolph (reviewed by John Mann); Alkaloids Chemical and Biological Perspectives (Volume 9) ed. S. W. Pelletier (reviewed by David J. Robins) Natural Product Reports Editorial Board Professor T. J. Simpson (Chairman) University of Bristol Dr J. R. Hanson University of Sussex Dr R. B. Herbert University of Leeds Professor J. Mann University of Reading Professor D.J. Robins University of Glasgow Dr C. J. Schofield University of Oxford Dr D. A. Whiting University of Notting ham Editorial Staff Editorial Office Dr Sheila R. Buxton The Royal Society of Chemistry Managing Editor Thomas Graham House Dr Roxane M. Owen Science Park Deputy Editor Milton Road Miss Nicoila P. Coward Cambridge Production Editor UK CB4 4WF Dr Anthony P. Breen Mr Michael J. Francis Telephone +44 (0) 1223 420066 Technical Editors Facsimile +44 (0) 1223 420247 Miss Daphne E. Houston E-mail rscl @rsc.org Miss Karen L. White RSC Server http://chemistry.rsc.org/rsc/ Editorial Secretaries Natural Product Reports is a bimonthly journal of critical reviews. It aims to foster progress in the study of bioorganic chemistry by providing regular and comprehensive reviews of the relevant literature published during well-defined periods.Topics include the isolation structure biosynthesis biological activity and chemistry of the major groups of natural products -alkaloids terpenoids and steroids aliphatic aromatic and 0-heterocyclic compounds. This is augmented by frequent reviews of the wider context of bioorganic chemistry including developments in enzymology nucleic acids genetics chemical ecology primary and secondary metabolism and isolation and analytical techniques which will be of general interest to all workers in the area. Articles in Natural Product Reports are commissioned by members of the Editorial Board or accepted by the Chairman for consideration at meetings of the Board.This journal includes reviews of books relating to natural products and bioorganic chemistry. Volumes for review should be sent to The Managing Editor Natural Product Reports The Royal Society of Chemistry Thomas Graham House Science Park Milton Road Cambridge UK CB4 4WF. Contributors to Volume 12 Appendino Giovanni 349 Atta-ur-Rahman 36 1 Baldwin M. A. 33 Faulkner D. J. 223 Fraga B. M. 303 Fukumoto K. 277 Lewis J. R. 135 339 Michael J. P. 77 465 535 Mount Andrew R. 93 Banks B. J. 165 Bentley K. W. 419 Boland G. M. 321 Chapman Stephen K. 93 Choudhary M. Iqbal 361 Connolly J. D. 609 Cooper C. B. 165 Dewick P. M. 101 507 579 Donnelly D. M. X. 321 Dutton C. J. 165 Hanson James R. 207 381 567 Harborne J. B. 639 Herbert R. B. 55 445 Hill R. A. 609 Hill Robert E. 555 Ihara M.277 Itoh Shinobu 45 Keseru G. M. 69 Knight David W. 271 Murray R. D. H. 477 Nbgradi M. 69 O’Hagan David 1 Ohshiro Yoshiki 45 Robins D. J. 413 Saxton J. E. 385 Spenser Ian D. 555 Ward R. S. 183 Williams C. A. 639 Nomenclature It is the policy of The Royal Society of Chemistry to en- courage the use of IUPAC* and IUBMB* Recommendations on nomenclature. Many of the appropriate nomenclature documents are included in the following compilations. Com pi lat ions 1 Nomenclature of Organic Chemistry Sections A B C D E F and H I979 edition a 550-page hardcover volume published in 1979 available from Pergamon Press Oxford. Section F of this volume covers general principles for the naming of natural products. 2 A guide to IUPAC Nomenclature of Organic Compounds (Recommendations Z993) a 182-page softcover volume published in 1993 available from Blackwell Scientific Publications Oxford to be used in conjunction with item 1.3 Biochemical Nomenclature and Related Documents A Compendium 2nd edition 1992 a 348-page softcover manual published in 1992 by Portland Press Ltd for IUBMB and available from the publisher (59 Portland Place London WIN 3AJ UK). 4 Compendium of Chemical Terminology IUPAC Recommendations a 456-page volume published in 1987 available in hardcover and softcover from Blackwell Scientific Publications Oxford. Specific Recommendations A selection of specific recent recommendations [many of which are included in Biochemical Nomenclature and Related Documents see above] that will be of particular interest to those who investigate bioorganic chemistry or the chemistry occurrence or biosynthesis of natural products includes Extension of Rules A- 1.1 and A-2.5 concerning numerical terms used in organic nomenclature (Recommendations 1986) Pure Appl.Chem. 1986 58 1693-1696. [The original versions of these Rules are in Nomenclature of Organic Chemistry 1979 edition see above] Nomenclature of steroids (Recommendations 1989) Eur. J. Biochem. 1989 186 429458. Nomenclature of tetrapyrroles (Recommendations 1986) Pure Appl. Chem. 1987 59 779-832. Nomenclature and symbols for folic acid and related compounds (Recommendations 1986) Pure Appl. Chem. 1987 59 833-836; Eur. J. Biochem. 1987 168 251-253.Nomenclature of prenols (Recommendations 1986) Pure Appl. Chem. 1987 59 683-689; Eur. J. Biochem. 1987 167 181-184. Nomenclature of retinoids (Recommendations 198 I) Pure Appl. Chem. 1983 55 721-726; Eur. J. Biochem. 1982 129 1-5. Nomenclature of vitamin D (Recommendations 198 l) Pure Appl. Chem. 1982 54 1511-1516; Eur. J. Biochem. 1982 124 223-227. Nomenclature of tocopherols and related compounds (Recommendations 1981) Pure Appl. Chem. 1982 54 1507-1510; Eur. J. Biochem. 1982 123 473475. Recommendations for the presentation of thermodynamic and related data in biology (1985) Eur. J. Biochem. 1985 153 429434. Enzyme Nomenclature (Recommendations 1984) Supplement 1 Corrections and additions Eur. J. Biochem. 1986 157 1-26. Enzyme Nomenclature 1992 (Recommendations of the Nomenclature Committee of the International Union of Biochemistry on the nomenclature and classification of enzyme-catalysed reactions) Academic Press Orlando Florida 1992.Nomenclature for multienzymes (Recommendations 1989) Eur. J. Biochem. 1989 185 485486. Symbolism and terminology in enzyme kinetics (Recommendations 1981) Eur. J. Biochem. 1982 128 281-291. Symbols for specifying the conformation of polysaccharide chains (Recommendations 1981) Pure Appl. Chem. 1983 55 1269-1272; Eur. J. Biochem. 1983 131 5-7. Polysaccharide nomenclature (Recommendations 1980) Pure Appl. Chem. 1982 54 1523-1526; Eur. J. Biochem. 1982 126 439441. Abbreviated terminology of oligosaccharide chains (Recommendations 1980) Pure Appl.Chem. 1982 54 15 17-1 522; Eur. J. Biochem. 1982 126 433-437. Nomenclature of glycoproteins glycopeptides and peptidoglycans (Recommendations 1985) Eur. J. Biochem. 1986 159 1-6. Nomenclature and symbolism for amino acids and peptides (Recommendations 1983) Pure Appl. Chem. 1984 56 595-624; Eur. J. Biochem. 1984 138 9-37 (see also Eur. J. Biochem. 1985 152 1 and the Newsletter 1985 of NC-IUB and JCBN ibid. 1985 146 pp. 238 and 239 and the Newsletter 1986 ibid. 1986 154 pp. 485 and 486). Nomenclature for incompletely specified bases in nucleic acid sequences (Recommendations 1984) Eur. J. Biochem. 1985 150 1-5 (see also Eur. J. Biochem. 1986 157 1). Abbreviations and symbols for the description of conformations of polynucleotide chains (Recommendations 1982) Pure Appl.Chem. 1983 55 1273-1280; Eur. J. Biochem. 1983 131 9-15 (see also the Newsletter 1984 of NC-IUB and JCBN Eur. J. Biochem. 1984 138 p. 7). A list of restriction endonucleases and their isoschizomers (updated annually) is given in R. J. Roberts and D. Macelis Nucleic Acids Res. 1991 19 2077-2109. Glossary for chemists of terms used in biotechnology Pure Appl. Chem. 1992 64,143. Selection of terms symbols and units related to microbial processes Pure Appl. Chem. 1992 64,1047. Organism Nomenclature Recent codes of nomenclature for organisms include International Code of Nomenclature of Bacteria and Statutes of the International Committee on Systematic Bacteriology (1989 Revision) ed.P. H. A. Sneath V. B. D. Skerman and V. McGowan American Society for Microbiology Washington DC USA 1976. [Appendix 2 of this publication (Approved Lists of Bacterial Names) appeared in Int. J. Syst. Bacteriol. 1989 39.1 International Code of Botanical Nomenclature (2988) ed. W. Greuter H. M. Burdett W. G. Chaloner V. Demoulin R. Grolle D. L. Hawksworth D. H. Nicholson P. C. Silva F. A. Stafleu E. G. Voss and J. McNeill Koeltz Scientific Books Konigstein Germany 1988. International Code of Zoological Nomenclature 3rd edn ed. W. D. L. Ride C. W. Sabrosky G. Bernardi R. V. Melville J. 0. Corliss J. Forest K. H. L. Key and C. W. Wright International Trust for Zoological Nomenclature in association with the British Museum (Natural History) London UK and the California Press Berkeley and Los Angeles USA 1985. * IUPAC International Union of Pure and Applied Chemistry. IUBMB International Union of Biochemistry and Molecular Biology.

ISSN:0265-0568    

DOI:10.1039/NP99512FP027

出版商:RSC    

年代:1995

数据来源: RSC

摘要:

Modern Mass Spectrometry in Bioorganic Analysis M.A. Baldwin Department of Neurology University of California at San Francisco CA 94743-0578,USA 1 Introduction and Historical Perspective 1.1 Electron Impact Ionization 1.2 The Analysis of Mixtures -Coupling with Separative Methods 1.3 Mass Analysis 2 Chemical Derivatization 3 Soft Ionization Methods 3.1 Chemical Ionization 3.2 Field Desorption 3.3 Fast Atom Bombardment 3.4 Ionization Methods for Intact Proteins 3.4.1 Matrix Assisted Laser Deso rption/I oniza tion 3.4.2 Electrospray Ionization 3.5 The Role of Soft Ionization in the Analysis of Mixtures 4 Mass Spectrometry Coupled with Separative Methods 4.1 Gas Chromatography-Mass Spectrometry 4.2 High Performance Liquid Chromatography-Mass Spectrometry 4.3 Other Separative Methods 5.Tandem Mass Spectrometry 6 Summary and Future Trends 6 Glossary of Terms 8 References. I Introduction and Historical Perspective This article is not intended to be a comprehensive review of the applications of mass spectrometry to bioorganic analysis. Rather it will summarize the currently available instrument- ation and methodology and hopefully it will act as a guide to what reasonably can and cannot be expected of the methods. It will explain the background to and development of the current state of the art so that potential users of mass spectrometry will understand the possible pitfalls as well as the advantages of various procedures.Mass spectrometry is already established as an extremely important technique in the context of natural product identification and characterization. Some selected examples of applications to natural product research will be mentioned. A glossary of relevant terminology appears at the end of this article. Mass spectrometry has played an important role as an analytical method ever since it was used by Aston to demonstrate the existence of isotopes and to measure their masses. The basic principles have changed little since Aston's experiments the sample must be volatilized and ionized not necessarily in that order; the ions are separated by magnetic or electrical fields in a vacuum chamber; and the ion current is detected and recorded as a function of mass.However the technology that allows each of these functions to be carried out has changed dramatically and is still changing at such a pace that we can carry out analyses today that would have been unthinkable only ten years ago. The unique character of many of the natural isotope patterns of common elements makes mass spectra of certain compound classes particularly characteristic. This is certainly true of organohalogen compounds containing chlorine or bromine which contrary to the belief of many environmentalists are diverse and widespread in nature even without the more recent contributions from the human race. These natural products include terpenes and nucleic acids from algae phenols from sponges and alkaloids and pyrroles from microbial sources.' The petroleum industry was the first community to recognize the value of mass spectrometry for the analysis of volatile organic compounds.It was largely to satisfy the needs of this industry that the first commercial mass spectrometers were produced. However as with so many other developments in scientific instrumentation the molecular nature of recent research in biological and medical sciences has acted as the hot- house that has forced innovative development. Instruments with the capability of measuring the masses of intact proteins with an accuracy of 0.01 % are now routinely available. 1.1 Electron Impact Ionization Electrical discharge provided an early method of dissociating and ionizing the sample and variants of this continue to be used today for inorganic and elemental analysis.Organic compounds required more gentle treatment and gas-phase electron impact (EI) became established as the standard method. Initially this relied upon the analyte being sufficiently volatile to give significant vapour pressure in an evacuated vessel from which it would slowly leak into the higher vacuum of the ionization chamber. Here the isolated molecules were bombarded with electrons emitted from a heated filament transferring more than enough energy for an electron to be ejected with the formation of a positive ion. Subsequent developments involved heating the sample expansion vessel for less volatile samples. Finally direct insertion systems allowed solid samples to be introduced directly into the ionization chamber on a heatable probe.Until the late 1960s virtually all organic mass spectrometers were completely reliant upon EI ionization. During this period the use of mass spectrometry became virtually routine for the characterization of natural products in large part due to the development of high resolution and accurate mass measure- menh2 Widely respected texts by Budzikiewicz Djerassi and Williams published in that period summarized the prevailing state of the art and made reference to the numerous classes of natural products that had been analysed successfully by mass spectrometry." These included separate volumes on steroids and alkaloids the latter covering more than 15 cla~ses.~ EI ionization was ideally suited to many of these compounds i.e.those for which the sample was sufficiently volatile and was not thermally labile. An EI ionization source is heated to about 200 "C by the filament and the hot metal surfaces readily catalyse thermal decomposition reactions such as the dehydration of alcohols. During the ionization step an electron is removed from the molecule forming a positive ion. Removing an electron from an organic molecule leaves an unpaired electron i.e. the newly formed species is a radical-cation and like all radicals is a highly reactive entity. The ionizing electron carries excess energy part of which is transferred to the newly formed ion as internal energy. This energy may be used for chemical reactions which for an isolated species in a vacuum will be unimolecular.Thus there are several reasons why the molecular ions formed in the ionization source rapidly undergo fragmentation reactions. This is both a strength and a weakness of EI ionization; the fragment ions observed in the mass spectrum can yield vital structural inf~rmation,~ but if the molecular ion is so unstable that it decomposes entirely the most vital piece of data is lost 33 i.e. the molecular mass The requirement for volatility and enhanced stability can often be satisfied by chemical derivatiz- ation of the polar functionalities (see Section 2). Ion/molecule reactions may occur if the pressure in the ion source becomes high enough for neutral molecules to collide with ions within the period (-1 ps) before they are accelerated into the flight tube.This phenomenon is the basis for chemical ionization the first ‘soft’ ionization method to become widely established. Several other soft ionization methods have been developed that do not rely on sample vapourization prior to ionization. 1.2 The Analysis of Mixtures -Coupling with Separative Methods The El mass spectrum of a pure compound generally displays a large number of peaks. The highest mass peak is usually attributable to the molecular ion and the remaining peaks correspond to fragment ions. The spectrum of a mixture is much harder to interpret as there is no obvious way to assign the various peaks to individual components.Consequently EI mass spectrometry is unsuitable for mixtures unless it is coupled with a separative technique. Gas chromatography (GC) and EI mass spectrometry were found to be ideally suited as they both rely on the volatility and thermal stability of the analyte. GC-MS became widely established allowing the analysis of extremely complex mixtures e.g. of essential oils from plant sources. With the development of high performance liquid chromatography (HPLC) in the 1960s and the dramatic increase in the range of compound types that could be separated by rapid and sensitive chromatography with high resolution the limitations of GC were soon apparent. Coupling the liquid flowing from the HPLC with the vacuum systems of the mass spectrometer was not trivial but 20 years of development have resulted in reliable sensitive and flexible instruments that allow mass spectrometric analysis of the full range of compounds that can be separated by HPLC.These issues will be discussed in more detail in Section 4.2. Another development that began in the 1960s that aided the growth of GC-MS was the replacement of stripchart recorders by computerized data acquisition systems. Initially these simply allowed the sequential recording of large numbers of spectra but subsequent hardware and software enhancements allowed the mass spectrometer to operate under computer control and the data can now be subjected to highly sophisticated real-time analysis and presentation routines. Computerized data ac-quisition also allows library searching an extremely effective means of identifying an analyte for which a spectrum already exists in the library or for identifying compound classes for novel unknowns.1.3 Mass Analysis There are several physical methods by which ions can be separated according to their mass-to-charge ratios (m/z)where z is the number of electronic charges e. Traditionally the application of a magnetic field (B) has been the most widely used technique. The equation for this is given by m/z = B2r2e/2V where r is the radius of the ion trajectory and V is the accelerating voltage. The mass spectrum which is a plot of ion current versus m/z,is usually obtained by scanning the current through the coils of an electromagnet. Mass spectrometers that combine both a magnetic and an electric field have ‘double- focusing’ properties and in favourable cases can separate species differing in mass by as little as 1 :100000.With such a high resolution instrument it is possible to measure the mass of an ion with an accuracy of about one ppm. The mass defects of the elemental isotopes that make up organic chemicals frequently allow molecular compositions to be determined from accurate mass measurements. On the atomic mass scale NATURAL PRODUCTS REPORTS 1995 lH = 1.0078 “C = 12.000 14N = 14.0031 and l60= 15.9959; thus the species CO N, CH,N and C,H have different precise molecular masses even though all have a nominal molecular mass of 28. They can be separated only if the mass spectrometer has a resolving power of 2700.The accurate mass measurement of a peak at m/z 28 will easily identify any of the above; this technique has become widely accepted to replace elemental analysis although it is effective only up to about 500 Da. The upper mass limit of mass spectrometers employing magnetic analysers usually ranges from about 1000 to 15000 Da. However the cost of buying and maintaining such a soph-isticated instrument is substantial and the majority of mass spectrometers now employ a much lower cost quadrupole mass analyser. Rather than magnetic fields the quadrupole uses a com- bination of radiofrequency and static electric fields to separate ions with lower resolution and with a maximum m/z range of 1000 to 3000. As will be described in Section 3 some ionization techniques result in highly charged ions (i.e.z 9 1) and this can increase the effective molecular mass range to > 100000.The quadrupole is a linear device in that ions travel through it from the ion source to the detector but it is also possible to trap ions by electrical fields within the aptly named quadrupole ion trap mass spectrometer. The ion cyclotron resonance (ICR) mass spectrometer also relies on ion trapping but the ions follow circular orbits in crossed magnetic and electric fields. It owes much to NMR technology as Fourier transformation of the output signal is required because the entire spectrum of ions is monitored simultaneously. Currently FT-ICR is the only method offering a realistic prospect of high resolution analysis of intact proteins.6 This instrument and the ion trap share a common advantage in that the selective ejection of ions allows the retention and study of ions of just one m/z value (see tandem mass spectrometry below).The simplest mass analyser is based on the principle of time of flight (TOF). Ions are formed virtually instantaneously by a very short burst of energy such as from a laser pulse and they are accelerated through a vacuum system by the high voltage V. All ions acquire equal kinetic energies (KE =zeV =mv2/2),but if they have different masses they travel at different velocities (v). The ions travel to a detector that is at a fixed distance from the ion source and their time of arrival is converted into mass.There is no upper mass limit for this analyser as higher mass ions simply take longer times. With the increasing availability of low-priced lasers for ionization and the recent optimization of sample handling procedures this technique has become established for the analysis of highly polar biomolecules including proteins and oligonucleotides. At this point brief reference will be made to tandem mass spectrometry or mass spectrometry-mass spectrometry (MS- MS) although this will be described in more detail in Section 5. This method uses combinations of mass analysers to isolate ions of one m/z value from the body of ions in a spectrum. The selected ions are subjected to reactions that generally result in fragmentation by collision induced dissociation (CID).2 Chemical Derivatization As has been referred to above samples of biological origin are frequently polar thermally unstable and difficult to volatilize. Compounds having multiple functional groups capable of hydrogen-bonding such as OH SH NH, and COOH are the most problematical. Sugars are excellent examples of such compounds as the multiplicity of hydroxyl groups and consequent high polarity results in intermolecular hydrogen bonding and renders even monosaccharides virtually involatile. In order to obtain sufficient vapour pressure for EI ionization it may be necessary to heat the sample to the point where it degrades usually by multiple dehydration reactions. This is most easily avoided by derivatization of the hydroxyls with one of several possible groups to reduce polarity.Examples of commonly used simple reactions for sugars include methylation NATURAL PRODUCT REPORTS 1995-M. A. BALDWIN acetylation trimethylsilylation and the formation of boronates. There are several well-known chemical reactions for each of these procedures; the choice of derivative and reaction is dependent on the specific analytical problem to be tackled.' There are several requirements for the ideal derivatization reaction. Most importantly it should be a simple reaction that gives a high yield of a single well-defined product with minimal side reactions if there are ten or more hydroxyl groups on a tri- or tetrasaccharide and each derivatization is only 90'/o complete the mass spectrum will be extremely complex and it will be very difficult to determine the original molecular mass.It may be undesirable to increase the molecular mass excessively as the mass range of the spectrometer may be limited. Methyl acetyl and trimethylsilyl groups each add 14 42 and 72 Da respectively ten methylations add only 140 Da whereas 10 trimethylsilylations will add 720 Da. On the other hand for a mass spectrometer with a high upper mass limit a large increase in molecular mass may be advantageous as background ions which are strongest at low mass may interfere with the ions of interest. Specialized derivatives may dramatically improve the properties for a particular type of analysis the sensitivity is substantially increased for negative chemical ionization by the introduction of a polyfluorinated substituent (see Section 4.1); several derivatizing agents add across double bonds and then undergo cleavage to define the chain position of the double bond.Isotopic labelling is often valuable for structural purposes and will distinguish between groups that were added by the derivatization and groups that were already present ;for example an oligosaccharide containing an amino sugar that may or may not be acetylated could be derivatized with trideuterioacetyl groups8 The use of a 1 1 H,:D acetylating mixture for peptides allows fragments correspond- ing to the N-terminus to be identified by the presence of doublets separated by 3 Da as distinct from C-terminal fragments that give single peaks.This technique was used in the sequencing of a C-terminally blocked nine-residue cardioactive peptide Met-Asn-Tyr-Leu-Ala-Phe-Pro-Arg-Met-NH,, iso-lated from 2500 Aplysia brasiiiana gastropod molluscs.s Derivatization of reducing sugars can give either a or p-anomeric that may separate chromatographically but reaction with the carbonyl of the open-chain form will prevent cyclization treatment with methoxylamine hydrochloride forms a methoxime at C-1 and the remaining hydroxyls can then be trimethylsilylated. The use of multiple derivatizations on a single molecule may reveal other structural information permethylation of an oligosaccharide followed by hydrolysis of the saccharide linkages to give hydroxyls borohydride reduction of the carbonyl groups of the monosaccharides and peracetylation of the resulting hydroxyls gives partially methylated alditol acetates.The mixture can then be separated and analysed by GC-MS.'O Identification of the particular positions in the sugar chain that have been derivatized with a methyl or an acetyl group reveals the original linkage positions between the individual monosaccharides. If the reduction is carried out with sodium borodeuteride rather than borohydride C-1 of a hexose becomes labelled with deuterium and is readily distinguished from C-6. This is illustrated for a 1,4-linked hexose in Scheme 1 the relative positions of the methyl and acetyl groups in the hexitol product being dependent on the original linkage positions.Terminal sugar residues will acquire an additional methyl group whereas chain branching will give a hexitol having one less methyl. EI analysis of the partially methylated alditol acetates does not give molecular ions and the interpretation is based entirely on GC retention times and the mass spectral fragmentation," whereas a soft ionization method such as chemical ionization gives molecular ions in addition to fragment ions.12 Reductive amination has been used to add various UV-absorbing groups to oligosaccharides offering several advantages including adding alkyl groups to facilitate reversed- phase HPLC separation allowing detection of the UV chromophore in HPLC and providing fragmentation-directing HO OH CH30 OCH3 1,blinkedhexose 1 ii FHO CH20CH3 7HOCH3 YHOCH3 HOGOH FHOH YHOH CH30 OCH3 CH20CH3 1 iii YHDOD YHDOAc 7HOCH3 YHOCH3 CHCXHa IV CHOCHa -1 ~HOH" YHOAc FHOH YHOAc CH20CH3 CH20CH3 1-deuterio-2,3,6-trimethylhexitol-l,4,5-triacetate Chemical derivatization of a 1,Clinked hexose from an oligosaccharide prior to determining the original linkage positions in the oligosaccharide by GC-MS.Reagents i CH,SO,CH; CHJ; ii AcOH 80 "C; iii NaBD, NaOH; iv Ac,O 100 "C Scheme 1 properties in mass spectra.13-15 Sugars may contain acid or amino groups and specific reactions may be employed to derivatize them permethylation of an amino sugar with methyl iodide in the presence of sodium hydroxide in dimethylsulfoxide adds three methyl groups to the amino function thus forming a quaternary ammonium cation.The introduction of the permanent positive charge greatly enhances the sensitivity for analysis by FAB and it also directs the fragmentation observed in tandem mass spectrometry.16 This is discussed in more detail in Section 5. 3 Soft Ionization Methods All of the new ionization methods in use in bioorganic mass spectrometry can be described as soft ionization methods; they impart relatively little internal energy to the analyte molecules and the ions observed in the spectra are predominantly species rather than fragment ions. 3.1 Chemical Ionization The term chemical ionization (CI) was coined to describe the interactions between the analyte and ions formed within an approximately lo4 molar excess of a reagent gas such as methane ionized by electron bombardment at a relatively high pressure of about 1 Torr (1 mmHg).At this pressure bimolecular ion/molecule processes predominate and the initial molecular ions of methane are rapidly converted to more thermodynamically stable ions such as protonated methane CHS. It is statistically improbable that many of the analyte molecules will be ionized by the initial electron impact process but multiple collisions will bring them into contact with the reagent gas ions. If the proton affinity (PA) of the analyte molecules (M) is greater than that of methane proton transfer will occur and the majority of analyte molecules may be come ionized. CH; +M +CH +MH' AH = PA(M)-PA(CH,) This technique which was pioneered by Munson and Field," is typical of the soft ionization methods in that the analyte ions formed by adduction of an ion such as a proton are inherently more stable even-electron species compared with the odd- electron radical-cations formed by EI.5 Furthermore because the amount of energy transferred on ionization (AH) is relatively small compared with bombardment by 70 eV electrons (1 eV -100 kJ mol-') most of the protonated mol- ecular ions do not fragment and they are the predominant ion species seen in the mass spectra.Thus CI is much more likely to provide molecular weight information than is EI. EI and CI are very compatible methods as they are both based on ionization in the gas phase by electrons emitted from a heated filament.It is common for ion sources to be capable of both modes of operation. Generally both methods are applied to volatile samples. Volatile organic samples can be sampled from aqueous solution through membranes that are permeable to organics but which protect the integrity of the vacuum a method first used in 1963. This simple and rapid sampling technique is ideal for process control and environ- mental monitoring but it has recently been used to screen volatile products from fermentation of the microorganisms Bacillus polymyoxa and Klebsia oxytoca to optimize production of butanedi01.l~~'~ The same method used to monitor fer- mentation of the fungus Bjerkandera adusta by EI CI and MS-MS showed that although benzaldehyde was the main product initially it dropped off after several days to be replaced by the previously unidentified chloromethoxybenzaldehyde.20 Because the ionizing species in CI is chemical in nature it is possible to vary the chemistry and thereby control the nature of the ionization process and the degree of any fragmentation.21 The use of isobutane as the reagent gas gives the t-butyl cation as the major reactive species.Isobutene the conjugate base has a higher PA than methane. Thus AH becomes smaller as less energy is transferred on proton transfer resulting in less fragmentation. Ammonia gives the ammonium ion which has an even higher PA; for some compound classes AH will be negative. Proton transfer is not favoured in such cases thus the ionization becomes compound selective.Another ionization mechanism often observed with ammonia is adduct formation i.e. addition of the intact ammonium ion to the analyte molecules. Electron bombardment of methane at lower pressures (0.1 Torr) gives rise to large numbers of free electrons having only thermal energies. These may be captured by analytes having high electron affinities forming negative ions. This is often referred to as negative CI although it is more correctly described as electron-capture ionization (ECI). This is par- ticularly favoured by certain compound classes such as amines for which it can be extremely sensitive and highly selective. The higher pressure conditions of CI compared with EI make the former highly compatible with GC (see Section 4.1).The analyte for CI is ionized in the gas phase and it was initially thought that the requirements for sample volatility would be similar to those for EI. However twenty years ago this author discovered that the reagent gas plasma had a scavenging action and would ionize peptides of four to six residues introduced directly into the ionization chamber on a glass rod.22 This became known as direct CI (DCI) and subsequently this technique achieved wider recognition when the samples were flash evaporated from an electrically heated wire very close to the ionization region. 3.2 Field Desorption A high voltage applied to a sharp tip or edge produces extremely high electrical field gradients strong enough to strip an electron from a gas-phase molecule that comes close to the tip.This is described as field ionization (FI). At the University of Bonn Beckey and Schulten grew carbon filaments on a very fine tungsten wire. They deposited involatile samples on the wire and applied a high voltage whilst simultaneously heating NATURAL PRODUCTS REPORTS 1995 OH yCH3 HO 0 I m/zllQ (1) Bruceantinol it. The action of the heat and the electric field released ions into the gas phase. This technique field desorption (FD) was used to study a wide range of compounds of low ~olatility.~~ In a study of bruceolides isolated from Brucea javanica FD was the only method by which the molecular mass of bruceantinol (1) could be assigned with confidence due to the facile loss of acetate from C-4' under all other ionizing condition^.^^ FI and FD are unusual amongst soft ionization methods in that they often give radical cations rather than protonated even-electron ions.FD gives almost no fragmentation and is useful for screening mixtures such as heterogeneous oligosaccharides released from proteins; it can be concluded that any peak observed corresponds to a component in the mixture. 3.3 Fast Atom Bombardment FD is a difficult technique to apply reproducibly and was largely superseded when Barber et al. demonstrated that involatile compounds could be analysed by fast atom bom- bardment (FAB).25 This was derived from secondary ion mass spectrometry (SIMS) in which a solid inorganic sample is bombarded with a stream of heavy ions.The primary ions ablate secondary analyte ions from the surface hence the name. Numerous workers had attempted to apply this technique to the analysis of organic compounds but the spectrum obtained was always transitory as the primary ions rapidly destroyed the analyte. Barber solved this problem by dissolving the sample in a viscous liquid of relatively low volatility such as glycerol and bombarding this with argon atoms in the vacuum system of the mass spectrometer. Diffusion of the sample within the liquid matrix constantly refreshed the surface and the matrix mediated the damage by the bombarding species. This allowed the analysis of much larger and completely involatile compounds ; for example insulin a small protein of mass 5800 having two amino acid chains linked by disulfide bonds became the standard by which the performance of FAB mass spectrometers was measured.New matrices heavier bombarding species including xenon atoms and caesium ions and new magnet technology were introduced to increase the upper mass limit to about 15000. The phrase 'liquid secondary ion mass spec-trometry ' (LSIMS) more correctly describes the technique when the bombarding species is ionic rather than atomic,26 but FAB and LSIMS are frequently used interchangeably. Most compounds yield molecular mass information from the (M+H)+ ions or perhaps (M+Na)+ or (M+K)+.Negative ions can also be studied as (M-H)- or sometimes as (M+ X)-where X is a group such as halide ion.The precise mechanism of FAB ionization is still open to debate some people believing that preformed ions are sputtered into the gas phase from solution whereas others maintain that neutrals are sputtered and ionized by CI-like processes in the high-pressure region immediately above the liquid surface.". 28 Limited fragmen- tation may be observed sometimes sufficiently for structural elucidation such as the sequencing of pep tide^,^^ or oligo-saccharides."O The development of FAB opened a new era in mass spectrometry. It transformed a method ideally suited to the analysis of volatile organic compounds to one equally ap- plicable to polar involatile biological samples. Most import- antly it was easy to use and anyone could do it. During the NATURAL PRODUCT REPORTS.1995-M. A. BALDWIN (2) (M+H)+ dl449 m/2 431 -CH&OOH -CH3CSH -CH&OOH 4 m/z 27 1 m/z 371 Fragmentation of the protonated molecular ion of 4a-phorbol ester diacetate (2) monitored by MS-MS Scheme 2 1980s the analysis of underivatized sugars oligonucleotides salts peptides and antibiotics became almost routine. Soft ionization methods such as FAB often reveal stereochemical differences that would be masked by high-energy ionization. The stereochemistry of the 4a-~/~-cis-and 4P-~/~-trans-phorbol esters members of the tigliane class of diesters can be differentiated by FABMS. The presence of m/z 27 1 comprising rings B and c with a lactone bridging carbons 3 and 20 (the hydroxymethyl group) is a powerful diagnostic test for the 401 configuration.:” The formation of this ion which is illustrated in Scheme 2 for the diacetate (2) demonstrates some general points concerning fragmentation after soft ionization the protonated molecular ions are relatively stable even-electron species compared with the radical-cations formed by EI.These ions largely undergo the elimination of small neutral molecules such as water or acetic acid thereby preserving the even-electron character; and the energy imparted by CID in MS-MS is sufficient to bring about substantial skeletal rearrangements. The electron movements are illustrated by full arrows rather than ‘fishhooks ’ indicating movement of electron pairs rather than radical reactions. The authors were able to compare the spectra from a series of related compounds and the reaction mechanisms were supported by a series of tandem experiments carried out in a double-focusing mass spectrometer using specialized scan modes.Different scan modes were used to show (i) the fragmentation pathways from the protonated molecular ion leading to m/z 271 ; (ii) the precursor ions that dissociated to form m/z 271 ;and (iii) the fragment ions formed by the further dissociation of m/z 27 1. The addition of metal cations such as two Li’ ions to the deprotonated anions of long-chain fatty acids encourages charge-remote fragmentations in CID-MS-MS that allow the identification of positions of substituents along the chain an example being the locating of the double bonds in homo- conjugated octadecadienoic acids.32 Organometallics proved to be highly amenable to FAB analysis including porphyrins that mimic photosynthetic reaction centre^.^" An example of peptide sequencing was provided by analysis of a novel phytotoxin pentadepsipeptide named pholamide isolated from the ‘black-leg’ fungus that attacks canola.The cyclic molecule was ring opened by methanolysis at room temperature prior to MS-MS sequencing whereas the stereochemistry of the original cyclic molecule was determined by NMR.34 CH~O‘ (3) 3,5-dimethoxy-4-hydroxycinnamicacid (sinapinic acid) (4) a-cyano-4hydroxycinnamicacid (5) 2-amino-5-nitropyridine Intact proteins are not included in the above list of compounds analysed by FABMS because with a few exception^,^^ the majority of them never really succumbed to FAB ionization despite the best efforts of a large number of mass spectroscopists.However enzymic and chemical degra- dation with peptide mapping by FAB has become established as a vital component in the elucidation of the primary structure and posttranslational modifications of FAB also became the basis for a microbore HPLC-MS interfacing method in which the LC eluent flows through a capillary or a frit onto the FAB ta~get.~’,~~ An application of this is described in Section 4.2. In the development of any technology certain techniques are employed simply to fill a gap until something better comes along whereas others become established as permanent additions to a growing array of methods.EI falls into the group of permanent methods and undoubtedly it has been joined by FAB despite its limitations and the fact that at least two other extremely successful ionization methods were subsequently introduced that in some respects are superior. These will be described below. The modern arsenal of ionization techniques is becoming even more diverse. 3.4 Ionization Methods for Intact Proteins Brief reference should be made to plasma desorption mass spectrometry (PDMS). This technique developed by MacFarlane is based on the bombardment of the sample on a thin foil by a nuclear fragment from the fission of 2s52Cf thus sputtering protonated ions into the gas phase. Simultaneously another fragment from the same fission process is detected which starts a clock and the ions are mass analysed by TOF.39 Experimentally this technique is one of the easiest to carry out by a non-specialist and it proved to be more successful than FAB for the analysis of proteins.4o It can also be used to monitor smaller species than intact proteins; in a comparison between PDMS and radiolabelling of high-mannose glyco- peptides from the variable surface glycoprotein of trypanosomes PDMS was judged to offer significant advantages .4 3.4.1 Matrix Assisted Laser Desorption/Ionization Employing a laser for ionizing samples for mass spectrometric analysis is not a new idea.The novel aspect of matrix assisted laser desorption/ionization (MALDI) introduced by Hillenkamp and Karas that resulted in an extremely sensitive and widely applicable technique for protein analysis was the dispersal of the sample in a large excess of a UV-chromophoric matrix compound.42-34 The precise ionization mechanism is not known but intuitively this is analogous to the use of the reagent gas in CI or the liquid matrix in FAB.The first successful matrix was nicotinic acid which absorbs the laser energy and transfers it to the analyte giving protonated molecular ions that are sputtered into the gas phase. The maximum resolution of the TOF analyser of -400 (defined as MIAM,where AM is the width at half-height of a peak of mass M) is poor in relation to the maximum mass range (m/z > lOOOOO) but most proteins give only a protonated molecular ion and perhaps doubly and triply protonated ions that appear at m/z = M/2 and M/3 respectively.Thus there is no problem in separating these species even though the peaks may be several hundred mass units wide. Unfortunately some matrix compounds tend to form photochemical adducts with the analyte and further broaden the peaks 3,5-dimethoxy-4-hydroxycinnamic acid (sinapinic acid) (3) forms adducts of mass M+ 206 Da. This characteristic is less prevalent with some more recent matrices such as a-cyano-4-hydroxycinnamic acid (4).45 Basic compounds such as 2-amino-5-nitropyridine (5) are also effective as matrices particularly for the analysis of compounds such as oligonucleotides that are more stable as anions than as cations.46 The accuracy of mass measurement depends upon being able to measure the arrival time of the ions at the detector.To calibrate the mass scale it is essential to introduce another protein as an internal standard together with the analyte. With careful measurements the mass of a homogeneous protein can be specified to within 0.1 YO,i.e. 1 Da in 10000or 10 in 100000. A molecular mass measurement can frequently identify a posttranslational modification such as the presence of a phosphate group adding 80 Da to the molecular mass. MALDI has been used to study a wide variety of proteins much of the pioneering work being carried out by Beavis and Chait at Rockefeller University New York with home-made instr~ments.~’ Now numerous instrument companies offer 48 commercial versions.The sensitivity is extremely high no more than one pmol being required in most instances which for a protein of mass 25000 is only 25 ng. This is dissolved in 1 pl of a solution containing 10 pg of the matrix compound dried on a target inserted into the vacuum and irradiated by pulses of laser radiation usually from a nitrogen laser at 337 nm or from a neodymium-YAG laser at 266 or 355 nm. A complete mass spectrum is recorded for each laser pulse although the total time required for the experiment may be longer as several scans may be acquired and averaged to optimize the data. The time- to-mass conversion is rapidly carried out by a personal computer. Posttranslational modifications can result in substantial microheterogeneity in proteins particularly when the modification involves sugar residues or lipids.Sugars may be N-linked to asparagine or 0-linked usually to serine or threonine and there are glycosylphosphatidylinositol anchors that contain both sugars and lipids (see Section 5). There are many diverse structures due to processing by different cellular enzymes e.g. asparagine-linked oligosaccharides may range from high mannose species which vary only in the number of mannose residues to hybrid and complex structures that differ widely in structure and the nature of the individual sugar residues,49 the degree of processing being partly dependent For upon the protein ~tructure.~~ the intact protein this diversity of structures gives rise to an array of molecular ions that may not be resolved.The separation may be improved by reducing the mass of the analyte which may be achieved by digesting the protein and isolating a smaller peptide containing the substituted amino acid. Enzymes are available commercially (endo H and peptide N-glycosidase F) that will remove the oligosaccharides and allow the analysis of the peptide alone51 or chemical methods may be employed such as hydrazinolysis of N-linked structures or base hydrolysis of 0-linked forms. The oligosaccharides can be isolated by high pH anion exchange chromatography for separate analysis.52 The extremely high sensitivity of MALDI and consequent small sample size allows numerous enzyme digests to be carried out on small aliquots of a glycoprotein or glycopeptide.Thus the entire array of oligosaccharides can be analysed by judicious use of various glycosidases with different specificities separately analysing the products from each reaction.53 Peptides oligosaccharides glycolipids and oligonucleotides have all been analysed by MALDI. NATURAL PRODUCTS REPORTS 1995 Figure 1 A schematic representation of an ESI source The resolution of the time-of-flight analyser can be improved by the use of an electrostatic refocusing device called a reflectron. Ironically this is of little utility for the largest ions as these tend to be ‘metastable’ and to dissociate in the flight path. Metastable dissociation is not a problem with the linear TOF analyser as the ions do not change their velocities on fragmentation; ions of any given mass and ionic or neutral fragments formed by the in-flight dissociation of ions of the same mass will all arrive at the detector simultaneously.The reflectron however slows deflects and reaccelerates the ions causing a separation of the original ions and its fragments. A TOF mass spectrometer with a reflectron may have a resolving power as high as 10000.54For the analysis of lower mass species such as peptides and oligosaccharides the enhanced resolving power provided by the reflectron can be a substantial advantage. 3.4.2 Electrospray Ionization (ESI) An alternative ionization method for large involatile analytes generally described as electrospray or sometimes as ionspray was originally introduced in 1970 but has been extensively refined in recent year^.^^.^^ The ESI source is illustrated in Figure 1.The sample is injected into a flowing liquid stream frequently 1 1 methanol-water or acetonitrile-water acidified with 1 YOacetic acid which protonates basic sites on the analyte molecules. This is sprayed from a fine nozzle into a chamber at atmospheric pressure forming an aerosol of minute droplets. The nozzle is held at a high potential causing the droplets to carry an electric charge. A stream of nitrogen gas encourages the evaporation of the solvent from the droplets until the coulombic repulsion of the charges on the ions causes them to fly apart leaving isolated ions in the gas phase. Ions are drawn through differentially pumped apertures into the high vacuum system and focused by lenses into the mass spectrometer analyser which is usually a quadrupole analyser but may be magnetic.The ESI mechanism has been reviewed ~ecently.~’ The upper m/z limit of a quadrupole is usually 2000-4000 and yet because each ion can carry multiple charges this method can give the molecular mass of a protein of 100000 Da or more. If every basic site on such a protein becomes protonated it is likely to carry 100 charges or more. As the mass spectrometer separates ions according to m/z rather than mass such an ion will appear at m/z 1000. In practice there is usually a distribution of charge states giving a range of mass spectral peaks of m/z = MH,/n MH,+,/n+ 1 etc. From the spacing of these peaks a simple algorithm allows the molecular mass to be calculated by the data system which can also deconvolute or transform the array of peaks corresponding to the various multiply charged ions into a single molecular peak.The statistics for mass measurement are improved by the NATURAL PRODUCT REPORTS 1995-M. A. BALDWIN presence of multiple peaks allowing molecular masses to be determined to better than 0.01 %. CID-MS-MS of multiply charged ions can be complex as there may be several permutations for sharing the charge between the fragments.58 It is frequently extremely difficult to isolate and purify samples of biological origin such as proteins therefore great emphasis is placed on the sensitivity of detection. ESIMS is highly sensitive although perhaps not as sensitive as MALDI.A single scan may be recorded while less than a picomole of material is sprayed from the nozzle but in general the total amount of sample required for a particular assay may be 10 times greater. In terms of the accuracy of mass measurement and the ability to resolve closely spaced ions ESIMS is generally superior to MALDI. ESI is in more common use than MALDI because it can be applied with a 'conventional' mass spectrometer such as a quadrupole or magnetic sector instrument with an upgrade to the original equipment whereas MALDI generally requires a dedicated TOF mass spectrometer. A further advantage of ESI comes from the fact that the analyte is delivered to the mass spectrometer in a flowing liquid stream and thus it is readily compatible with HPLC as will be discussed below in the next section.Both techniques have found rapid acceptance in the biotechnology industry. They are ideally suited to address such questions as to whether recombinant proteins from various expression systems carry the normal glycosylation and other posttranslational modifications that may be vital for biological action. It is also becoming increasingly apparent that ESI and to a lesser extent MALDI are extremely valuable for smaller analytes such as pep tide^.^^ They may offer advantages over longer established methods such as FAB in terms of speed sensitivity flexibility of mass range and sample type and ease of operation. Both methods can operate effectively in negative- ion mode ; the negative-ion analysis of several compound classes including antibiotics drugs nucleotides and peptides is excellent under solution conditions that promote anion formation.6o Both positive and negative-ion modes were employed in an ESI study of anticancer drugs including tamoxifen steroids.hexaglutamates and organometallics related to cisplatin. Negative-ion analysis proved most suc-cessful for steroid sulfates such as 5-androstene-3,17-diol- 17- sulfate." Negative ion ESI is ideally suited to oligonucleotide analysis ; extensive calculations have been made to correlate molecular weight information with oligonucleotide compo- sition for both DNA and RNA.'j2 When faced with the requirement for a mass spectrometer and ionization system to achieve the widest possible range of analyses ESI offers many very compelling arguments.3.5 The Role of Soft Ionization in the Analysis of Mixtures It is common for an EI mass spectrum to exhibit numerous strong ions arising from fragmentations of a single molecular species whereas soft ionization mass spectra that show multiple strong ions are most likely due to mixtures. Glycopeptides consisting of a common peptide core carrying heterogeneous arrays of oligosaccharides give complex spectra but the mass differences between the various molecular ions can often be interpreted in terms of the oligosaccharide compositions. Derivatization of hydroxyl groups will change the observed masses and aid the spectral interpretati~n.~~ Mixtures may be created deliberately to aid the sequencing of peptides.A number of methods have been used to generate peptide 'ragged ends ' to produce families differing by successive amino acids. Most recently a technique called ladder sequencing has employed a stepwise chemical digestion with the Edman reagent phenylisothiocyanate (PITC) containing a small proportion of a terminating agent such as phenylisocyanate (PIC). At each cycle the majority of the peptide molecules are derivatized by the PITC and then cyclized and cleaved by trifluoroacetic acid. Thus the PITC causes successive amino acid cleavages from the N-terminus and would eventually cleave the entire peptide except that at each cycle a small fraction of the peptides react with a PIC molecule which forms a stable phenylcarbamyl peptide that is resistant to further reaction.Thus after many cycles a mixture of concatenated peptides is obtained and analysed by a soft ionization method such as MALDI. The spectra show families of peaks differing by single amino acids allowing the peptide sequence to be read directly. The sequencing of peptides of molecular mass greater than 7 kDa generated in the synthesis of HIV-1 protease has been described.'j4 4 Mass Spectrometry Coupled with Separative Methods 4.1 Gas Chromatography-Mass Spectrometry When gas chromatographs were first interfaced with mass spectrometers packed columns were the standard with gas flows of 30 ml min-' which was excessive for the vacuum system of the mass spectrometer ;therefore various separating systems were devised.These included porous frits that selectively passed the carrier gas and membranes that selectively passed the organic analyte. The most successful separator developed by Ryhage employed two opposing jets in an evacuated envelope. The momentum of the relatively large analyte molecules carried them across the gap between the two jets in a molecular beam and into the mass spectrometer whereas the lighter carrier gas atoms (helium) diffused out into the envelope and were pumped away.65 Using this approach GC-MS became widely established and was used successfully for a very diverse range of analyses. However separators were never 100% efficient and as they had to be heated to prevent condensation they were potential sites for sample decompo- sition especially in the early models that were made of metal before the widespread adoption of glass.The introduction of high resolution capillary columns operating at flow rates of 1-5 ml min-' resolved this difficulty and it became standard practice to introduce the entire gas flow directly into the ionization chamber. This was also compatible with the higher ion source pressures used for CI. GC-MS is in very wide use as a routine method for trace analysis of mixtures. The US Environmental Protection Agency introduced protocols for the analysis of environmental pollutants in air soil and water and thereby established a major market for bench-top GC-MS instruments. There have been numerous examples of natural products being separated and analysed by GC-MS; in fact it is impossible to do justice to the scale of the applications of GC-MS.In reviewing the GC-MS literature published between July 1974 and June 1976 Brooks and Middleditch acknowledged that they had deliberately excluded many references even though they cited 647.'j6 A decade later in a comparable two-year review Evershed cited 472 papers.6' Head-space analysis of volatiles from decomposing oranges infected with Peniciffium revealed 23 compounds of potential interest. When ranked by retention times these ranged from acetaldehyde to geranio1.68 Cucurbitine which has anthelmintic activity was assayed in extracts from the seeds of cucurbita ssp. A chiral stationary phase in the GC column was used to determine the correct stereochemistry and the samples were analysed as methyl esters and di(trifluoroacety1)amides.The fragmentation patterns observed from EIMS allowed structural analysis.69 Several derivatization procedures have been adopted for long-chain alcohols and acids that otherwise suffer from dehydration and rarely give molecular ions. Analogous long chain aldehydes encountered as insect pheromones were converted to Schiff bases with amino pyridines or amino pyrimidines. These were readily separated by GC and gave both molecular and fragment peaks by EI ; radical-induced cleavages revealed chain structure and double-bond positions.io GC-MS analysis of the fatty acids and aldehydes obtained from the methanolic hydrolysis of phospholipid fractions from the NATURAL PRODUCTS REPORTS 1995 n I The pyrrolidine derivatives of (5E,92)-6-bromoheptacosa- octacosa- and nonacosa-5,9-dienoic acids (6a) and the products from reduction with deuterium and Wilkinson's catalyst [RhCl(PPh,)] in methanol (6b) Scheme 3 sponges Petrosiaficiformis and P.hebes revealed 54 fatty acids including novel brominated species. The interpretation of the mass spectra of the bromo compounds was aided by the presence (or absence) of the characteristic 1 1 isotope pattern.71 Cleavage of the C-7-C-8 bond in (6a) is favoured energetically being allylic to both double bonds (Scheme 3). This gave two peaks at mlz2.58 and 260 showing that the ion retained the bromine.The same bond cleavage occurring in concert with the loss of bromine and a hydrogen transfer gave a single peak at m/z 180.By reducing with deuterium to give (6b) and observing the mass differences for the various methylene groups i.e. 14 for CH, 15 for CHD and 16 for CD, the position of the bromine atom was found to be C-6 rather than C-5. A less prosaic application of GC-MS was the identification of C, to C, alkanes in the analysis of resins used in the preservation of an Egyptian mummy thereby implicating the addition of bitumen to the wrappings. Acidic components not amenable to GC-MS required the use of soft ionization methods. Low resolution FAB revealed a multitude of molecular species and MS-MS was used as the separative method (see below).This confirmed that many of the components were oxidation products of abietic acid indicating the employment of a conifer resin as the embalming fluid." GC-MS is readily combined with CI as GC is more compatible with the higher ion-source pressures. Selective CI of nitrogen and sulfur heterocycles in petroleum fractions was achieved using ammonia as a reagent gas in an ion trap mass spectrometer. Ion-molecule reactions at -1 mmHg gave a number of product ions including the ammonium ion. An advantage of the ion trap mass spectrometer was that unwanted species from competing charge-exchange reactions were ejected giving rise to much cleaner spectra than could be obtained by conventional GC-MS.'3 The selectivity of electron-capture ionization for organochlorine compounds was used to identify many of these in the blubber of the Beluga whale (Delphinapterus leucas) from Hudson Bay.After solvent extraction and liquid chromatography GC coupled with EIMS and negative-ion ECIMS were employed. Octa- and non-achlorobornanes were identified from the strong (M-Cl)- ions under ECI conditions and the complementary (M-CH,CI)+ ions from EI; these compounds are amongst the more than 670 previously found in the insecticide Toxaphene which was banned in North America a decade ago but which continues to be used in many parts of the globe.74 ECI is particularly effective for polyfluorinated compounds and appropriate Ion repeller electrode Liquid flow ......-I...... -vacuum pump -Hot vapour -1 mi min-' (HPLC erc.) -vacuumpump To mas Figure 2 A schematic representation of a TSP source derivatization procedures have been developed.Pentafluoro- benzoylation was employed for GC-MS analysis of 1-0-alkyl-2-acetyl glycerols derived from platelet activating factor by treatment with phospholipase C giving compound (7) which was detectable at the level of 100 fg and was assayed by comparison with a trideuteriated internal 4.2 High Performance Liquid Chromatography-Mass Spectrometry Compared with GC-MS considerably greater problems were encountered with the first attempts to couple HPLC with mass spectrometry. The typical liquid flow of 1 ml min-' from an analytical HPLC column is equivalent to -1000 ml min-' of gas at normal temperature and pressure.Removing the liquid by external evaporation and transferring the analyte into the mass spectrometer on a moving belt was one approach. It required the sample to be somewhat involatile so that it would remain on the belt as it passed through the vacuum system to the ion source where it then had to be sufficiently volatile and thermally stable to evaporate under the influence of heat and be ionized. Over the years this introduction method was refined to the stage that less volatile samples could be ionized directly from the belt by FAB76 or laser de~orption.'~ Another approach was to split the liquid flow and introduce a small fraction (< 1 %) directly into the vacuum system but with a consequent reduction in the sensitivity of detection.Normal phase HPLC with an organic solvent such as hexane allowed the residual solvent to act as a CI reagent gas." This strategy enjoyed some success before the later development of thermospray (TSP) ion sources that could cope with the entire 1 ml min-' of an aqueous liquid from an analytical HPLC column. For TSP the column eluent contains a buffer such as ammonium acetate and is weakly acidic protonating basic molecules. It is sprayed into a heated tube that is evacuated with a high capacity rotary pump and vapour phase ions are sampled through an aperture into a mass analyser usually a quadrupole. This is illustrated in Figure 2. This method is particularly suited to polar compounds such as drug metabolites thus it was the first mass spectrometric method that was truly compatible with the types of compounds that were best separated by HPLC rather than GC." LC-TSPMS has been used to assay natural products such as quinine and related Cirzchona alkaloids from cell cultures,'' and flavones from fermentation broths.s1 In the latter study daidzein NATURAL PRODUCT REPORTS 1995-M.A. BALDWIN m/z 137*$i-=H) 153 (R = OH) (8)Daidzen; R = H (9)Genistein; R=OH (8) and genistein (9) were monitored and identified by their (M+H)+ ions at m/z 225 and 271 and by the fragmentation across the pyran ring induced by CID which gave m/z 1 19 and the complementary ions m/z 137 or 151. It was demonstrated that (8) and (9) could be identified from the MS-MS spectra obtained by DCI after the probe had been dipped directly into the crude extracts.In a study of flavonoids from Camellia sinensis LC-TSPMS-MS was identified as the method of choice.82 LC-MS is particularly valuable for compounds lacking chromophores such as saponins. Several of these including glucuronides were identified in extracts from the fruits of Tetrapleura tetraptera giving (M+H)+ ions as well as am- monium and acetonitrile adducts. CID induced cleavages along the sugar chain down to the aglyc~ne.~~ In the negative-ion mode LC-TSPMS was employed to study the products of enzymic digestion of plant cell-wall polysaccharides. The predominant adduct ion species included OAc- HSOj NaSO; SO:- and combinations thereof.84 In recent years microbore and capillary HPLC techniques have become widely available with flow rates dcwn to a few pl min-l requiring only very small samples.Cytokinins are N6-substituted derivatives of adenine which occur as free bases ribosides ribotides or glucosides. These species which induce cell division in plants have previously been analysed by GCMS after formation of trimethylsilyl or trifluoroacetyl derivatives. However these derivatives are unstable in aqueous media whereas cytokinins were quantified in the needles of Picea abies by capillary LC-FABMS which proved to be 10-100 times more sensitive than static FAB due to constant refreshment of the sample and reduced ion suppression.85 Concurrent with the introduction of micro LC techniques new versatile ionization methods such as ESI have been developed for the direct introduction of samples dissolved in liquid eluants thus optimizing MS compatibility with HPLC.Most importantly with these new techniques there is no requirement whatsoever for the sample to be volatile and thus HPLC-MS is now suited to the many compound classes that are separated most effectively by HPLC. A negative ion LC-ESI study of phosphopeptides has demonstrated that fragmentation induced by collisions between the ESI source and a simple quadrupole analyser gives ions of m/z 63 and 79 corresponding to PO; and PO; respectively. By stepping the collisional excitation voltage between scans so as to yield (M-H)- ions and fragment ions in successive scans it is possible to identify and characterize phosphopeptides in mixtures of peptides generated by enzymic digestion of proteins.86 4.3 Other Separative Methods Supercritical fluid chromatography (SFC) is often useful when analytes are too polar and involatile for GC separation.It has been coupled with MS and there are many examples of its application but it has yet to find wide acceptance. It can be interfaced directly with EI;8' with CI using gases such as methane isobutane or ammonia;8s or with ESI.89A separative technique familiar to organic chemists and widely used in natural product chemistry is thin layer chromatography (TLC). Collision Ion source c-I Gas Figure 3 Schematic representation of an MS-MS experiment. Ions of one m/z value are selected in MSl and fragmented by collisions with an inert gas.The mass spectrum of the fragment ions is obtained by scanning MS2 There have been examples of the application of TLC-MS; TLC spots have even been removed from the plate and attached to a FAB target with double-sided adhesive tape to analyse antimicrobial components in the bleomycin complex,go but the two methods are not readily compatible. Capillary electro- phoresis (CE) has been combined with mass spectrometry to give very high resolution separation and high sensitivity analysis. The ionization method is either flow-FAB or ESI both of which are best suited to flow rates of -5 ,u1 min-' which is as much as 100 times greater than CE flow rates so the column eluent is supplemented with a higher flow of a liquid of a composition that is optimized for the ionization method.The sample sizes are typically in the fmol range so the mass spectrometer must be highly sensitive for successful CE-MS. The separation of macrolide antibiotics and mass spectrometric analysis by ESI was compared for nano-scale LC-MS and CE- MS. Both methods worked well but the higher sample loading of LC gave improved mass spectrometric signals and limits of dete~tion.~~ Sodium dodecylsulfate-polyacrylamide gel electro- phoresis (SDS-PAGE) is widely used for the separation of biochemical samples particularly peptides and proteins. In 2D SDS-PAGE complex mixtures are separated by molecular weight in one dimension and by isoelectric point in a second dimension.The analytes can be electroblotted onto a membrane made of an inert material such as polyvinyldifluoride (PVDF) coated with a UV absorbing matrix and rastered with a laser beam for MALDI analysis. It is also possible to carry out chemical or enzymic reactions on the PVDF membrane and to monitor the products without separation or purifi~ation.~'! 5 Tandem Mass Spectrometry It was stated in an earlier section that EI mass spectrometry is not ideally suited to the analysis of mixtures unless it is coupled with a separative method. On the other hand the ion fragmentation observed in EI mass spectra is very important for structural characterization. Soft ionization methods that give almost no fragmentation are able to display the various molecular species in a mixture but they reveal little structural information.Mass spectrometry is itself a separative method and by connecting two mass spectrometers in series it is possible to select a single component of a mixture in MS1 dissociate the selected ions by collision and analyse the fragments in MS2. This procedure described as tandem mass spectrometry is often abbreviated to MS-MS by analogy with other 'hyphenated techniques' that couple a separative method with a mass spectr~meter.~" It is illustrated schematically in Figure 3. The instrumentation for MS-MS can be a relatively modest triple quadrupole a highly sophisticated and extremely ex- pensive four-sector high resolution magnetic instrument or a hybrid of the two. In the triple quadrupole (Q) the initial ion separation is effected by Q1.The selected ions experience collisions with an inert gas in 42 and a fraction of their kinetic energy is converted to internal energy causing dissociation and other reactions. Q2 operates in a 'radio frequency only' mode i.e. without the DC voltage. This transmits all ions irrespective of mass without discriminating against ions that change their mass by fragmentation. The newly formed ions are then analysed by scanning Q3. The kinetic energy of the ions in a quadrupole mass spectrometer is low (10-100 eV) but the energy transferred on collision is sufficient to cause extensive fragmentation. The ion currents emerging from 43 are weak compared with a simple mass spectrum obtained from a single quadrupole.However for the analysis of a minor component in a complex mixture MS-MS often gives a higher sensitivity of detection because the ratio of signal to noise is improved by filtering out all ions other than those of interest. The principle of the four-sector tandem mass spectrometer is basically the same although both the precursor and product ions can be selected with high mass resolution which further improves the signal-to-noise ratio. The collision process occurs at much higher energies usually 2-10 keV which increases the range of possible reaction types. The most advanced examples of these instruments are equipped with array detectors that simultaneously detect all ions within a certain mass window which can be up to 50% of the mass range.Such instruments enjoy the Fellgett advantage which is familiar from IR and NMR spectroscopy comparing scanning instruments with more modern interferometers and Fourier transformation. Array detection is much more efficient than scanning across the selected mass range and thereby provides a further enhancement to the sensitivity of detection. The major disadvantage of such an instrument is the cost in the region of $1000000 which effectively restricts the availability of these instruments to major multinational chemical and pharmaceutical companies or to centralized national facilities for non-commercial users. By contrast triple quadrupoles cost about one-third of this price and are in widespread use. Other alternatives for tandem mass spectrometry include the use of a two-sector high resolution double focusing instrument in various specialized scanning modes or to carry out collisions in an ion trap or an ion cyclotron resonance instrument and to select only the fragment ions for further analysis.The latter method can be used to study sequential reactions sometimes NATURAL PRODUCTS REPORTS 1995 f f HOI?CH&H2NH2 Protein-Ser-NHCH2CH20POH I ? ? [Man]-Man-Man-Man-GIcN-lnositol-OPOH [&alNAc-Gal-Sia] &H2$H$H Figure 4 The prion protein GPI anchor. (Gal = galactose GalNac = N-acetylgalactosamine GlcN = glucosamine Man = mannose Ser = Serine Sia = sialic acid.) selecting the molecular ion for each monosaccharide in turn in MS1 fragmenting the ions by collisions and scanning the product ion spectra in MS2.A comparison of the CID spectra showed common features that revealed the branching patterns. As anticipated there was a linear sequence of at least three hexoses (sometimes four) attached to hexosamine and inositol. The hexose closest to the reducing terminus carried a second chain of up to three sugar residues N-acetylhexosamine- hexose-sialic acid and in a separate experiment using ESIMS of the intact GPI (not hydrolysed with HF and not per-methylated) it was shown that a further phosphoethanolamine was also attached to this same hexose. As yet mass spectrometry is not capable of distinguishing between isomeric sugar residues such as mannose glucose and galactose in oligosaccharides but the sensitivity of the oligosaccharide to specific glycosidases such as a-mannosidase and neuraminidase together with chromatographic analysis of monosaccharides produced by acid hydrolysis allowed the individual residues to be identified.98 The structure of the GPI anchor linked to serine-231 at the C-terminus of the prion protein is illustrated in Figure 4.There is considerable interest in determining linkage positions in oligosaccharides by CID-MS-MS information that is readily available from NMR studies only if sufficient sample is available. Many authors have demonstrated the ability of FABMS-MS and CID to distinguish between some linkage referred to as MS" where n = the number of reaction ~teps.'~.~~ positions especially where the bonds are hindered.This may be The following example of MS-MS analysis demonstrates the ability of this technique to identify all the components in a mixture with picomolar sensitivity and to measure the molecular mass and carry out the structural analysis of each component. Some proteins have an oligosaccharide-containing phospho-lipid group termed a glycosylphosphatidylinositol (GPI) linked to the C-terminus through pho~phoethanolamine.'~. GPI-9i anchored proteins are relatively rare in mammals but occur frequently in proteins of bacteria yeast and other lower organisms. The function of the GPI is to anchor the protein to cellular membranes. All GPI anchors studied to date contain a linear chain of three mannose residues linked to the protein C- terminus through phosphoethanolamine and linked at the reducing terminus through glucosamine to phosphatidyl-inositol.Other sugar residues may be present in GPIs causing significant heterogeneity . The complex oligosaccharides forming the core of the GPI anchor of the scrapie prion protein were released by hydrolysis of the phosphodiester bonds with aqueous HF and then were permethylated. Ionization by LSIMS and mass separation in MS1 of a high-performance four-sector tandem mass spec- trometer revealed species having five separate molecular masses. From amino acid analysis of the protein it was known that the major components amounted to only 30 pmol of material. The molecular masses allowed a prediction of the compositions in terms of hexoses hexosamines N-acetyl hexosamines etc.but the scrapie prion protein GPI proved to be unique in that the two largest species carried sialic acid. The characterization of an oligosaccharide requires the determination of the sequence and branching of the sugar residues. This was achieved by enhanced by derivatization and/or production of adduct ions such as alkali-metal cationized species. It was demonstrated in a recent study that collisions with a more passive gas such as argon rather than the more commonly used helium were more productive in terms of cleavages across saccharide rings. A single CID spectrum from a four-sector tandem mass spec- trometer for a permethylated bianntenary heptasaccharide revealed extensive diagnostic fragment ions.99 One of the most highly developed applications of tandem mass spectrometry is the sequencing of peptides.Most of the amino-acid residues (amino acid minus water) have charac- teristic masses although there are some ambiguities e.g. lysine and glutamine at m/z 128. Cleavage of a peptide chain at a particular point gives an N-terminal and a C-terminal fragment either of which may carry the charge thus providing comp- lementary information. Furthermore each amino acid can cleave at up to three positions; the C-1-N amide bond the N-C-2 bond and the C-2-C-1 bond.29 Thus there can be up to six different signals for the cleavage at any one amino acid and it is also possible to generate further ions via side-chain reactions particularly in the high-energy regime characteristic of sector tandem mass spectrometers.100 Also there are characteristic low mass ions due to single amino acids usually observed as immonium ions and other small fragments such as dipeptides.l*' Thus the spectra are quite complex and require experience for successful interpretation although several computer programs are now available for automated analy- sis.lo2.103 Advantages of peptide sequencing by tandem mass spectrometry include the ability to analyse peptide mixtures the provision of C-terminal data as well as N- NATURAL PRODUCT REPORTS 1995-M.A. BALDWIN terminal the ability to deal with N-terminally blocked peptides 7 Glossary of Terms and the speed of analysis; with computerized interpretation the CE sequence may be available within minutes rather than the CI several hours required for Edman N-terminal analysis.CID Weaknesses of the method relate to an upper size limit of DCI around 20 residues for routine sequencing and the potential ECI ambiguities resulting from certain mass combinations. EI An example of peptide sequencing that has caused con-ESI siderable excitement in the field of immunology demonstrates FAB the power of MS-MS to isolate and provide structural data on FD trace quantities of individual components in extremely complex F1 mixtures. Peptides derived from exogenous proteins that enter FT the cell are presented to the immune system by class I1 major GC histocompatability complex (MHC) molecules. MHC I-Ad GPI molecules were purified from mouse lymphoma cells.The HPLC peptides were released by acid filtered fractionated by HPLC ICR and then further fractionated by HPLC-ESIMS-MS. At least LSIMS 664 peptides were detected and as many as 2000 more were mlz present at or near the detection limit of 30 femtomole; the most MALDI abundant did not exceed 10 picomole. Sequencing revealed MS conserved common motifs which allowed binding studies with MS-MS synthetic peptides. The crucial binding region was determined PA to have the sequence Gln-Met-Val-Arg-Thr-Ala-Ala-Glu-Val-PD Capillary electrophoresis Chemical ionization Collision induced dissociation Direct chemical ionization Electron capture ionization Electron impact Electrospray ionization Fast atom bombardment Field desorption Field ionization Fourier transform Gas chromatography Glycosylphosphatidylinositol High performance liquid chromatography Ion cyclotron resonance Liquid secondary ion mass spectrometry Mass-to-charge ratio Matrix assisted laser desorption/ionization Mass spectrometry Tandem mass spectrometry Proton affinity Plasma desorption Sodium dodecylsulfate polyacrylamide gel electrophoresis Supercritical fluid chromatography Secondary ion mass spectrometry Time of flight Thermospray Ala.lo4 This study was carried out using a low resolution triple quadrupole MS-MS instrument.6 Summary and Future Trends If the past is an accurate guide to predicting the future mass spectrometry will come to play an even greater role in bioorganic analysis.It is common for analytical methods to reach a peak of activity and then to wane as they are superseded by better alternatives. That this fate has not befallen mass spectrometry is entirely due to the unprecedented and dramatic improvements in the technology both for producing ions and for analysing them. Thirty years ago the author of this article was involved in the mass spectrometric analysis of alkaloids steroids triterpenes and other natural products. In 1963 we were delivering these to the ion source by heating milligram quantities in an evacuated reservoir at about 250 “C. In the majority of cases our largest peak was at m/z44,i.e. CO, but by 1964 we had our first direct insertion probe.High resolution instruments were in their infancy EI was the only ionization method and the first tentative steps were being taken to interface GC and MS. Now we have techniques sensitive to fmol quantities of impure involatile ionic and thermally unstable analytes. There is almost no upper limit to the size of compounds that we can analyse. The information available is far more sophisticated and numerous experimental parameters can be optimized for the particular assay required. Current trends can be seen to be moving in two different directions. In the 1960s virtually every British university acquired a high-resolution magnetic mass spectrometer that represented the state of the art at that time. This could not happen today; in fact the company that made these instruments has recently ceased production of magnetic mass spectrometers.Some of the instrumentation is becoming ever more complex and expensive to the point that it will be routinely available only through a few national centres. On the other hand most of the instrument companies are already supplying simpler yet extremely powerful mass spectrometers for operation by non- specialists. These machines will sit in chemistry and bio- chemistry labs and they will be treated like fluorimeters spectrophotometers or any other item of routine equipment. The chemist who has synthesized a peptide or the biochemist who has isolated a new protein can now obtain his own ESI or MALDI spectrum and determine the molecular weight within a few minutes.The technology for this already exists and the wider application is limited only by cost; when such instruments are mass-produced these costs will decrease. SDS-PAGE SFC SIMS TOF TSP 8 References 1 J. W. Gribble J. Nat. Prod. 1992 55 1353. 2 J. H. Beynon ‘Mass Spectrometry and its Application to Organic Chemistry’ Elsevier Amsterdam 1960. 3 H. Budzikiewicz C. Djerassi and D. H. Williams ‘Mass Spectrometry of Organic Compounds’ Holden-Day San Francisco 1967. 4 H. Budzikiewicz C. Djerassi and D. H. Williams ‘Structural Elucidation of Natural Products by Mass Spectrometry. Vol. 1 Alkaloids’ Holden-Day San Francisco 1964. 5 F. W. McLafferty and F. Turecek ‘Interpretation of Mass Spectra’ University Science Books Mill Valley California 4th ed.1993. 6 K. D. Henry J. P. Quinn and F. W. McLafferty J. Am. Chem. Soc. 1991 113 5447. 7 D. R. Knapp ‘Handbook of Analytical Derivatization Reactions’ Wiley New York 1979. 8 A. Dell and P. R. Tiller Biochem. Biophys. Res. Commun. 1986 135 1126. 9 H. R. Morris M. Panico A. Karplus P. E. Lloyd and B. Riniker Nature 1982 300 643. 10 B. Lindberg and J. Lonngren Meth. Enzymol. 1978 50 3. 11 T. J. Waeghe A. G. Darvill M. McNeil and P. Albersheim Carbohydr. Res. 1983 123 28 1. 12 R. A. Laine Anal. Biochem. 1981 116 383. 13 C. C. Sweeley and H. A. Nunez Annu. Rev. Biochem. 1985 54 765. 14 J. W. Webb K. Jiang B. L. Gillece-Castro A. L. Tarentino T. H. Plummer J. C. Byrd S. J. Fisher and A. L. Burlingame Anal.Biochem. 1988 169 337. 15 C. M. John and B. W. Gibson Anal. Biochem. 1990 187 281. 16 M. A. Baldwin N. Stahl L. G. Reinders B. W. Gibson S. B. Prusiner and A. L. Burlingame Anal. Biochem. 1990 191 174. 17 B. Munson and F. H. Field J. Am. Chem. Soc. 1966 88 2621. 18 M. J. Hayward T. Kotiaho and A. K. Lister Anal. Chem. 1991 62 1798. 19 F. R. Lauritsen T. Katahio T. K. Choudhury and R. G. Cooks Anal. Chem. 1992 64 1205. 20 F. R. Lauritsen T. Kotiaho and D. Lloyd Bid. Mass Spectronz. 1993 22 585. 21 A. G. Harrison ‘Chemical Ionization Mass Spectrometry’ CRC Press Boca Raton Florida 2nd ed. 1992. 22 M. A. Baldwin and F. W. McLafferty Org. Mass Spectrom. 1973 7 1353. 23 H. D. Beckey ‘Principles of Field Ionization and Field Desorption Mass Spectrometry’ Pergammon New York 1977.24 M. A. Baldwin D. M. Carter F. A. Darwish and J. D. Phillipson Biomed. Mass Spectrom. 1981 8 362. 25 M. Barber R. S. Bordoli R. D. Sedgewick and A. N. J. Tyler J. Chem. Soc. Chem. Commun. 1981 325. 26 W. Aberth K. Straub and A. L. Burlingame Anal. Chem. 1982 54 2029. 27 C. Fenselau and R. J. Cotter Chem. Rev. 1987 87 501. 28 J. A. Sunner R. Kulatunga and P. Kebarle 1986 Anal. Chem. 58 1312. 29 P. Roepstorf and J. Fohlman Biomed. Mass Spectrom. 1984 11 601. 30 B. Domon and C.E. Costello Glycoconjugate J. 1988 5 397. 31 N. Takeda A. Tatematsu M. Suzuki H. Ohigashi and K. Koshimuzu Biol. Mass Spectrom. 1993 22 251. 32 J. S. Crockett M. L. Gross W. W. Christie and R. T. Holman J.Am. Soc. Mass Spectron?. 1990 1 183. 33 S. Naylor J. A. Cowan J. H. Lamb C. A. Hunter and J. K. M. Sanders J. Chem. Soc. Perkin Trans. 2 1992 41 I. 34 M. Soledade C. Pedras J. L. Taylor and T. T. Nakashima J. Org. Chew?. 1993 58 4778. 35 M. Barber and B. N. Green Rapid Commun. Mass Spectrom. 1987 1 80. 36 ‘Methods in Enzymology’ ed. J. McCloskey Academic San Diego 1990 Vol. 193. 37 Y.Ito. D. Takeuchi and M. J. Goto J. Chromatogr. 1985 346 161. 38 R. M. Caprioli T. Fan and J. S. Cottrell Anal. Chem. 1986 58 2949. 39 R. D. MacFarlane and D. F. Torgerson Science 1976 191 920. 40 R. D. MacFarlane J. C. Hill D. L. Jacobs and R. G. Phelps in ‘Mass spectrometry in the analysis of large molecules’ ed. C. J. McNeal Wiley Chichester 1986 p.1. 41 M. F. Bean J. D. Bangs T. L. Doering P. T. Englund G. W. Hart C. Fenselau and R. J. Cotter. Anal. Chem. 1989 61 2686. 42 M. Karas and F. Hillenkamp Anal. Chem. 1988 60 2299. 43 M. Karas U. Bahr and F. Hillenkamp Int. J. Mass Spectrom. lon Processes 1989 92 23 1. 44 M. Karas U. Bahr A. Ingedoh and F. Hillenkamp Angeiz). Chein. Intl. Ed. Engl. 1989 28 760. 45 R. C. Beavis T. Chaudhary and B. T. Chait Org. Mass Spectrotn. 1992 27 156. 46 M. C. Fitzgerald G. R. Parr and L. M. Smith Anal. Chem. 1993 65 3204. 47 R. C. Beavis and B. T. Chait Anul. Chern. 1990 62 1836. 48 B. T. Chait and S. B. Kent Science 1992 257 1885. 49 T. W. Rademacher R. B. Parekh and R. A. Dwek Ann. Rev. Biochem. 1988 57 785. 50 S. C. Hubbard J. Biol.Chew. 1988 263 19303. 51 A. L. Tarentino C. M. Gomez and T. H. Plummer Jr. Bio-chemistrj>,1985 24 4665. 52 R. R. Townsend and M. R. Hardy Glycobiologj. 1991 1 139. 53 K. K. Mock M. Davey and J. S. Cottrell Biochen?. Biophjis. Res. Commun. 1991 177 644. 54 K. Walter U. Boesl and E. W. Schlagg Int. J. Mass Spectrom. Ion Processes 1986 71 309. 55 S. F. Wong C. K. Meng and J. B. Fenn 1988 J. Phys. Chem. 92 546. 56 J. B. Fenn M. Mann C. K. Meng and S. K. Wong Mass Spectrotn. Rev. 1990 9 37. 57 P. Kebarle and L. Tang Anal. Chem. 1993 65 972A. 58 X.-J. Tang P. Thibault and R. K. Boyd Anal. Chem. 1993 65 2824. 59 A. L. Burlingame T. A. Baillie and D. H. Russell Anal. Chem. 1992 64 467R. 60 R. F. Straub and R. D. Voyksner J. Am. Soc.Mass Spectrom. 1993 4 578. 61 G. K. Poon G. M. F. Bisset and P. Mistry J. Am. Soc. Mass Spectrotn.. 1993 4 588. 62 S. C. Pomerantz J. A. Kowalak and J. A. McCloskey J. Am. Soc. Mass Spectrom. 1993 4 204. 63 A. Dell N. H. Carman P. R. Tiller and J. Thomas-Oates Bionied. Envirorr. Mass Spectrom. 1988 16 19. 64 B. T. Chait R. Wang R. C. Beavis and S. B. H. Kent 1993 Science 262 89. 65 R. Ryhage Anul. Chem. 1964 36 759. 66 C. J. W. Brooks and B. S. Middleditch in ‘Mass Spectrometry’ NATURAL PRODUCTS REPORTS 1995 Specialist Periodical Reports ed. R. A. W. Johnstone Royal Society of Chemistry London 1977 Vol. 4 p. 146. 67 R. P. Evershed in ‘Mass Spectrometry’ Specialist Periodical Reports ed. M. E. Rose Royal Society of Chemistry London Vol.9 1987 p. 196. 68 D. V. McCalley and J. F. Torres-Grifol Analyst 1992 117 121. 69 E. Schenkel P. Duez and M. Hanocq J. Chromatogr. 1992,625 289. 70 D. J. Harvey J. Am. Soe. Mass Spectrom. 1991 2 245. 71 N. Carballeira and F. Shalabi J. Nut. Prod. 1993 56 739. 72 M. L. Proeflce and K. L. Rhinehart J. Am. SOC.Mass Spectrom. 1992 3 582. 73 C. S. Creaser F. Krokos K. E. O’Neill M. J. C. Smith and P. G. McDowell J. Am. Soc. Mass Spectrom 1993 4 322. 74 G. A. Stern D. C. G. Muir 3. B. Westmore and W. D. Buchanon Bid. Muss Spectrom. 1993 22 19. 75 C. S. Ramesha and W. C. Pickett Biomed. Environ. Mass Spectrotn. 1986 13 107. 76 J. G. Stroh J. C. Cook R. M. Milberg L. Brayton T. Kihara Z. Huang K. L. Rhinehart Jr. and I. A. S. Lewis Anal.Chem. 1985 57 985. 77 T. P. Fan E. D. Hardin and M. L. Vestal Anal. Chem. 1984,56 1870. 78 P. Arpino M. A. Baldwin and F. W. McLafferty Biomed. Mass Spectrom. 1974 1 80. 79 C. R. Blackley and M. L. Vestal Anal. Chem. 1983 55 750. 80 S. Abdulrahman M. E. Harrison K. J. Welham J. D. Phillipson M. F. Roberts and M. A. Baldwin J. Chromatogr. 1991 562 713. 81 M. S. Lee. D. J. Hook E. H. Kerns K. J. Volk and I. E. Rosenberg Biol. Mass Spectrom. 1993 22 84. 82 Y. Y. Lin K. J. Ng and S. J. Yang J. Chromatogr. 1993 629 389. 83 M. P. Maillard and K. Hostettmann J. Chromatogr. 1993 647 137. 84 W. M. A. Niessen R. A. M. van der Hoeven and J. van der Greef J. Chromutogr. 1993 647 319. 85 N. Imbault T. Moritz 0.Nilsson H.-J. Chen M.Bollmark and G. Sandberg Bid. Mass Spectrom. 1993 22 201. 86 M. J. Huddleston R. S. Annan M. F. Bean and S. A. Carr J. Am. Soc. Mass Spectrom. 1993 4 710. 87 R. D. Smith and H. R. Udseth Anal. Chern. 1983 55 2266. 88 R. D. Smith H. T. Kalinoski J. R. Udseth and B. W. Wright Anal. Chem. 1984 56 2476. 89 E. C. Huang J. J. Henion and T. R. Covey J. Chrornatogr. 1990 511 257. 90 S. F. Grappel A. J. Giovenella D. J. Newman and L. J. J. Nisbet J. Antihiot. 1984 37 1098. 91 C. E. Parker J. R. Perkins K. B. Tomer Y. Shida K. O’Hara and M. Kono J. Am. Soc. Mass Spectrom. 1992 3 563. 92 M. M. Vestling and C. Fenselau in ‘Techniques in Protein Chemistry’ ed. J. Crabb Academic Press New York Vol. V 1994 p. 59. 93 F. W. McLafferty ‘Tandem Mass Spectrometry’ Wiley New York 1983.94 J. N. Louris R. G. Cooks J. E. P. Syka P. E. Kelley G. C. Stafford Jr. and J. F. J. Todd Anal. Chem. 1987 59 1677. 95 A. G. Marshall and P. B. Grosshans Anal. Chem. 1991 63 215A. 96 M. A. J. Ferguson and A. F. Williams Annu. Rev. Biochem. 1988 57 265. 97 M. G. Low and A. Saltiel Science 1988 239 268. 98 N. Stahl M. A. Baldwin R. Hecker K.-M. Pan A. L. Burlingame and S. B. Prusiner Biochemistry 1992 31 5043. 99 J. Lemoine B. Fournet D. Despeyroux K. R. Jennings R. Rosenberg and E. De Hoffman J. Am. Soc. Mass Spectrom 1993 4 197. 100 K. Biemann and S. A. Martin Mass Spectrom. Rev. 1987 6 I. 101 A. M. Falick W. M. Hines K. F. Medzihradszky M. A. Baldwin and B. W. Gibson J. Am. Soc. Mass Spectrom. 1993 4 882. 102 R. S. Johnson and K. Biemann Biomed. Environ. Mass Spectrorn. 1989 18 945. 103 W. M. Hines A. M. Falick A. L. Burlingame and B. W. Gibson J. Am. Soc. Muss Spectrotn. 1993 3 326. 104 D. F. Hunt H. Michel T. A. Dickenson J. Shabanowitz A. L. Cox K. Sakaguchi E. Appella H. M. Grey and A. Settle 1992 Science 256 18 17.

ISSN:0265-0568    

DOI:10.1039/NP9951200033

出版商:RSC    

年代:1995

数据来源: RSC

摘要:

Natural Product Reports Editorial Board Professor T. J. Simpson (Chairman) Dr J. R. Hanson Dr R. B. Herbert Professor J. Mann Professor D. J. Robins Dr C. J. Schofield Dr D. A. Whiting Editorial Staff Dr Sheila R. Buxton Managing Editor Dr Roxane M. Owen Deputy Editor Miss Nicola P. Coward Production Editor Dr Anthony P. Breen Mr Michael J. Francis Technical Editors Miss Daphne E. Houston Miss Karen L. White Editorial Secretaries University of Bristol U n iversity of Sussex University of Leeds University of Reading University of Glasgow University of Oxford U n iversi ty of N ott ing ha m Editorial Office The Royal Society of Chemistry Thomas Graham House Science Park Milton Road Cambridge UK CB4 4WF Telephone +44 (0)1223 420066 Facsimile +44 (0)1223 420247 E-mail rscl@rsc.org RSCServer http://c hem istry.rsc.org/rsc/ Natural Product Reports is a bimonthly journal of critical reviews. It aims to foster progress in the study of bioorganic chemistry by providing regular and comprehensive reviews of the relevant literature published during well-defined periods. Topics include the isolation structure biosynthesis biological activity and chemistry of the major groups of natural products -alkaloids terpenoids and steroids aliphatic aromatic and 0-heterocyclic compounds. This is augmented by frequent reviews of the wider context of bioorganic chemistry including developments in enzymology nucleic acids genetics chemical ecology primary and secondary metabolism and isolation and analytical techniques which will be of general interest to all workers in the area.Articles in Natural Product Reports are commissioned by members of the Editorial Board or accepted by the Chairman for consideration at meetings of the Board. Natural Product Reports (ISSN 0265-0568) is published bimonthly by The Royal Society of Chemistry Thomas Graham House Science Park Milton Road Cambridge UK CB4 4WF. 1996 Annual Subscription Price EEA f325.00 USA $615.00 Rest of World f333.00. Change of address and orders with payment in advance to The Royal Society of Chemistry The Distribution Centre Blackhorse Road Letchworth Herts. UK SG6 IHN. Air Freight and mailing in the USA by Publications Expediting Service Inc.200 Meacham Avenue Elmont NY 11003. US Postmaster send address changes to Natural Product Reports Publications Expediting Service Inc. 200 Meacham Avenue Elmont NY 1 1003. Second-Class postage paid at Jamaica NY 11431-9998. All other despatches outside the UK are by Bulk Airmail within Europe and Accelerated Surface Post outside firope. Printed in the UK. Members of the Royal Society of Chemistry should order the journal from The Membership Manager The Royal Society of Chemistry Thomas Graham House Science Park Milton Road Cambridge UK CB4 4WF. 0 The Royal Society of Chemistry 1996 All Rights Reserved No part of this publication may be reproduced stored in a retrieval system or transmitted in any form or by any means electronic mechanical photographic recording or otherwise without the prior permission of the publishers. Printed in Great Britain by the University Press Cambridge Subscription rates for 1996 EEA f 325.00 USA $615.00 Rest of World f333.00

ISSN:0265-0568    

DOI:10.1039/NP99512FX034

出版商:RSC    

年代:1995

数据来源: RSC

ISSN:0265-0568    

DOI:10.1039/NP99512BP035

出版商:RSC    

年代:1995

数据来源: RSC

ISSN:0265-0568    

DOI:10.1039/NP99512BX036

出版商:RSC    

年代:1995

数据来源: RSC

摘要:

The Chemistry of Heterocyclic o-Quinone Cofactors Shinobu ltoh and Yoshiki Ohshiro Department of Applied Chemistry Faculty of Engineering Osaka University Yamada- oka 2-I Suita Osaka 565 Japan I Introduction 2 Syntheses of PQQ and Related Quinones 3 Adduct Formation Reactions and Redox Reactions 4 Function as Efficient Redox Catalysts 5 Synthesis and Characterization of a TTQ Model Compound 6 Conclusion 7 References I Introduction In the early 1980s it was established that in addition to the well-known NAD(P)(H)-and flavin-dependent dehydro-genases there is a third class of dehydrogenases that contain PQQ (pyrroloquinolinequinone) as the redox cofactor.' Although the existence of this novel cofactor in certain enzymes had been suggested in the early 1960s the exact structure was only elucidated by two independent laboratories in 1979.2.3 The finding of this novel cofactor led to the discovery by many researchers in related areas of several kinds of quinone- containing enzymes (referred to as quinoproteins) from both eukaryotic and prokaryotic organisms.4 During the 1980s PQQ or a closely related compound had been believed to be the cofactor of all kinds of quin~proteins.~.~ However recent advances in analytical methods in biology have demonstrated that PQQ is not the only one; amino acid derived cofactors such as trihydroxyphenylalanine (TOPA),' tryptophan trypto- phylquinone (TTQ),' and a covalently modified tyrosine cross- linked to a cysteinyl sulfur (Tyr-Cys)8 are involved in bovine COOH HOOC HN HOOC 0 PQQ COOH HOOC HN do 0 serum amine oxidase methylamine dehydrogenase and galactose oxidase respectively.In addition to its enzymological importance PQQ itself has been demonstrated to be a growth-stimulating substance for microorganismss and a nutritionally important compound for mammals.'* Furthermore the biomedical significance of PQQ has been discussed in relation to its facial electron-transfer ability," and several types of pharmacological activities of PQQ and its derivatives have been demonstrated.12 Quino- proteins have also been applied as biocatalystsl" or biosens~rs.'~ In this article we survey the chemistry of the newly found heterocyclic o-quinone cofactors PQQ15 and TTQ.To date several kinds of biologically important heterocyclic quinones are known,16 but PQQ and TTQ are the only two that act as redox catalysts in (several) biological systems. 2 Syntheses of PQQ and Related Quinones To date PQQ has been synthesized in six different way~.l'-~~ Among them Corey's method1' is worth noting because his synthetic strategy has been widely applied to the syntheses of several PQQ model compounds as mentioned below. PQQ can be prepared on a kg scale by modified Corey procedure^.^^ Buchi's method2' is also interesting because it is accomplished by using tyrosine and glutamic acid derivatives the starting materials in PQQ biosynthesis. In order to study the substituent effects several PQQ derivatives have been prepared.Decarboxylated PQQs [(1)-(3)] the so-called decarboxymethoxatins were synthesized by Bruice and coworkers with strategies basically similar to the Corey peptide !J& 'peptide H 0 TTQ ,COOH ,COOH 0 NATURAL PRODUCT REPORTS 1995 COOMe COOH R' HM HM fJ$ '0 '0PiOOC N HOOC '0N R2 OMe OMe OMe (8)80% (conversion yield) (5)99% (5) R' (6)R' = R2 = R3 = COOH = COOMe; R2 = R3 = COOH (7) R' = R3 = COOMe; R2 = COOH COOMe (8) R' = COOMe; R2 = COOP$; R3 = COOH i ii R7 R7 OMe OMe (4) (6)97% vii OMe 0 COOMe (15) R7 = R8 = Rg= CONMe,; 94% (19) 65% (16) R7 = COOMe; R8 = Rg = CONMe2; 68% (20) 68% (17) R7= R9 = COOMe; R8 = CONMe,; 55% (21) 58% Mw HOOC OMe (7) Reagents i TFA-H,O (2 l) 60 "C 1 h; ii (COCl), benzene rt 24 h; iii Pr'OH DMAP CH,Cl, rt 10 min; iv 0.1 M K,CO,-CH,CN (1 l) rt 2 h; v 0.1 M K2C03-CH,CN (2 l) 60 "C 9 h; vi 0.1 M K,CO,-CH,CN (2:l) rt 2 h; vii TFA-H,O (2:l) 60 "C 1 h Scheme 1 R1 OMe (6)R' = COOMe; R2 = R3 = COOH (7) R' = R3 = COOMe; R2 = CWH (8) R' = COOMe; R2 = COOPr'; R3= COOH li R4 R4 ___c ' R5 N 0 OMe 0 (9) R4=COOMe; R5=R6=H; 94% (12) 71% (10) R4 = R6 = COOMe; R5 = H; 64% (13) 56% (11) R4=COOMe; R5=COOPri; R6=H; 73% (14) 39% Reagents i PhNO, 200°C 1-2 h; ii CAN CH,CN-H,O OOC 10 min Scheme 2 method although several steps were required in each syn- thesi~.~~,~~ The decarboxylated models can also be synthesized very easily using (4) a synthetic intermediate of a total synthesis of PQQ," as a common starting *' Regioselective hydrolysis of the three methyl ester groups of (4)under varying conditions produces the tri- di- or monocarboxylic acid (18) R7 = COOMe; R8 = COOPr'; R9 = CONMe2; 97% (22) 47% Reagents i (COCI), benzene; ii Me,NH CH,Cl,; iii CAN MeCN-H,O Scheme 3 R" 0 (23) R'O = R'' = R12 = CONHCH(Me)COOMe (24) RIO = R" = R12 = CONHCH(COOMe)CH&HMe2 (25) R'O = R" = Ri2 = CONHC12H25 (26) R'O = COOMe; R" = Ri2 = CONHC12H25 (27) R'O = R12 = COOMe; R" = CONHC12H25 (28) Ri0 = COOMe; R" = COOPS; R12 = CONHCI2Hs derivatives (5) (6) or (7) and (8) respectively (Scheme 1).Thermal decarboxylation of (6) (7) and (8) and subsequent oxidation by CAN affords the ester derivatives of (l) (2) and (3) i.e.(12) (13) and (14) respectively (Scheme 2). Regio- selective functionalization through an amide linkage has also been accomplished starting from the carboxylic acid derivatives (5)-(8) as shown in Scheme 3. This methodology was applied for the preparation of PQQ model compounds having amino acid residues such as -NH-Ala-OMe (23) and -NH-Leu-OMe (24) or a long chain alkyl group [(25)-(28)]. Mono- and dimethyl ester derivatives of PQQ i.e. (30) (31) and (32) have also been prepared directly from the trimethyl ester of the quinone (29) (PQQTME) by applying a similar strategy to that mentioned above regioselective hydrolysis and regioselective esterification (Scheme 4).28These mono- and dimethyl ester derivatives may be useful compounds for studying the metal coordination chemistry of PQQ.l-Methylated PQQ (33) has been prepared by applying the Corey method during which the indole intermediate was N-methylated by the phase-transfer system of iodomethane-18-crown-6-postassium t-butoxide.26 Examination of the electronic effects of the pyrrole and pyridine nuclei of PQQ is another interesting subject. In order to explore this issue benzoquinolinequinone derivative (35) NATURAL PRODUCT REPORTS 1995-S. ITOH AND Y. OHSHIRO 47 COOMe COOMe COMe i’ii * 0 (29)1ii (31) 67% COOMe COOMe - \ MeOOC N 0 0 0 (30) 64% (32) 41% Reagents i TFA-H,O (1 :l) 60 “C 12 h; ii 0.1 M K,CO,-MeCN (1 I) rt 4 h; iii MeOH H,SO,(cat.) 60 “C 4 h N Scheme 4 COOH MeOOC 0 (33) H2N 0 MeOOC II MeOOC*COOMe OH (34) 28% I ii MeOOC 0 (35) 67% Reagents 1 PTSA (cat.) CH,CI, reflux 17 h; ii Fremy’s salt KH,PO, CH,CN rt 8 h Scheme 5 and 6-deaza derivative (41) have been synthesized.The benzoquinolinequinone (35) was synthesized by a Doebner-von Miller-type annulation between 3-amino-2-naphthol and dimethyl 2-oxoglutaconate and a subsequent oxidation with Fremy’s salt [(KSO,),NO] (Scheme 5).29The synthesis of (41) was achieved by constructing 2-carbomethoxybenz[indole (38) from 1-aminonaphthalene by a JappKlingemann reaction with methyl 2-methylacetoacetate and a subsequent Fischer indolization reaction.The quinone function was introduced by a Fremy’s salt oxidation of 5-aminoindole (40) which was prepared from (38) by regioselective nitration and catalytic (36) 93%(crude) iii J COOMe P & d=ccmMe ~ N 0 \0 (38) 93% (37) 54% 1. COOMe COOMe __c i “ \0 d \/ NO2 NH2 (39)82% (40)47% Jvii qkJ-0 (41) 44% Reagents i NaNO, HCl(aq.) 0 OC 10 min; ii methyl 2-methyl- acetoacetate KOH MeOH-H,O (1 l) 0 “C 5 h; iii H,PO, MeOH reflux 15 min; iv anhydrous HCl MeOH reflux 30 min; v HNO, AcOH rt 8.5 h; vi H,-PtO, MeOH rt 38 h; vii Fremy’s salt KH,PO,(aq.) acetone rt. 7 h Scheme 6 COOR ROOC (42)R=H (43)R=Me hydrogenation (Scheme 6).,O The physicochemical properties and the reactivities of these deaza derivatives have been examined in detail to explore the electronic effect of the heterocyclic skeletons and the substituents.”* 30 3 Adduct Formation Reactions and Redox Reactions One of the most intriguing aspects of the chemical properties of PQQ is its high reactivity toward nu~leophiles.~~ Acetone adds easily to the quinone carbonyl carbon of PQQ and its trimethyl ester derivative under weakly alkaline conditions to form respectively the aldol-type adducts (42) and (43).,., In fact PQQ was first isolated and crystallized from the methanol dehydrogenase of methylotrophic bacteria as the acetone adduct.2 The addition position of C-5in both (42) and (43)was COOMe COOMe Me@o -H+ ~ MeOOC HO OMe H20+OMe (44) COOMe MeH 40 -OH KHlN$' MeOOC OMe 0 0 (29) (45) 11..H IN*;;. 0 'H (454 (46) J COOMe 0 (47) Scheme 7 determined by X-ray crystallographic analysis. Acetone adduct formation of the other model compounds was also stud- ied.24.29.30 Addition of water and alcohols to the quinone function of PQQ has also been studied spectrophototometrically in Covalent addition of hexanol to PQQ was applied to the development of the detection method for PQQ (the so-called hexanol extraction pr~cedure).~~ The addition position of water and alcohols was always considered to be C-5 but no direct evidence had been reported. Recently the authors isolated the C-5 hemiacetal (44) in the reaction of (29) with methanol under neutral conditions and determined its crystal structure for the first time (Scheme 7).On the other hand treatment of (29) in methanol under acidic conditions gave dimethyl acetal (47) as the major product for which the addition position of methanol was determined to be C-4 by X-ray crystallographic analysis. Studies of the adduct formation reactions with methanol using a series of PQQ model compounds (12) (33) (35) and (41) and molecular orbital calculations provided a clear-cut explanation for the difference in position for hemiacetal and acetal formation which is as fOlI0 w s.32 The calculated values of the heat of formation clearly indicate that the C-5 hemiacetal is more stable than the C-4 adduct (by 1-3 kcal mol-l depending on the method used).Because hemiacetal (44) is readily converted into the original quinone in solution the hemiacetal formation step is considered to be completely reversible. Under such circumstances the reaction can be regarded as being thermodynamically con-trolled. Thus it is reasonable that hemiacetal (44) is formed as the only isolable product under neutral conditions. In the presence of acid elimination of water from the protonated NATURAL PRODUCT REPORTS. 1995 COOMe ,COOMe OMe 0 Me0 OMe COOMe Me0 OMe (50) 6"" No (51) COOMe COOR (54) R = Me R (55) R = H (52) R=NH2 (53) R=Me intermediate (45a) proceeds much faster than from protonated hemiacetal (44a) because of the favoured concomitant release of the pyrrole proton (H-1) to give conjugated (46) (Scheme 7).Attack by a second molecule of the solvent gives C-4 acetal (47). There is no conjugative effect upon loss of water from C- 5 hemiacetal (44) making loss of the pyrrole proton less favoured. Once C-4 acetal(47) is formed it cannot revert to the C-4 hemiacetal(45) in the presence of excess methanol ([MeOH $-[H,O]). Therefore C-4 acetal (47) is gradually accumulated under the acidic conditions. This mechanism is supported by the results of the acetal formation reactions of other quinones. The C-4 acetal(48) was obtained from (41) which has the same indole-4,5-quinone skeleton as (29) but the acetal formation occurred at the opposite site of the quinone i.e. C-5 and C-10 for (33) and (35) to give acetals (49) and (50) respectively.As the acetals (49) and (50) were derived from the corresponding C-5 and C-I0 hemiacetals these must be more stable than the corresponding C-9 and C-4 adducts as discussed above. These results might give us important information about the action of quinoprotein methanol dehydrogenases. Hydrazines are often used as carbonyl reagents in order to convert the carbonyl cofactors into the more stable hydrazone derivative^.^^ The trimethyl ester of PQQ is easily converted into the C-5 hydrazone derivatives (51) (52) and (53) by treatment with hydrazino compounds such as 2,4-dinitro-phenylhydrazine semicarbazide and acetohydrazide respec- ti~ely.~~.~~ The addition position of C-5 was confirmed by X-ray crystallographic analysis of the ethyl ester derivative of (5l).38 When aminoguanidine was used intramolecular cyclization and then aromatization of the corresponding hydrazone took place to give the triazine derivative (54).37 It is interesting to note that the redox reaction (reduction of NATURAL PRODUCT REPORTS 1995-S.ITOH AND Y.OHSHIRO 49 COOR RR'CHNH2 2RR'C=O + NH3 R=Ph; R'=H:5450% (based on PQQ) Reagents i PQQ ( 1 mol "/a) cetyltrimethylammonium bromide pH 8.1-30 "C 24 h aerobic conditions 0 Scheme 9 quinone IRNHNH. OH quinol RR'C=NH 9 :$H quinol COOH -IN@: RR'CHNH2-HOOC 0 HO NH PQQ R+H R' cahinolamine Scheme 8 NH2 the quinone and oxidation of the hydrazine) occurs depending + upon the acidity of the reaction mixture and the natures of the RR'C=O substrates or quinones.For example aminoguanidine gave aminophenol R iminoquinone triazine (55)in the reaction with PQQ under acidic conditions although it produced PQQH (quinol) when treated with the Scheme 10 same substrate in an alkaline solution. The electronic nature of the substituent attached to the hydrazino group also altered the COOMe COOMe reaction course (redox reaction vs.adduct formation) the more elec tron-wi thdrawing the subs ti tuen t was the more favoura ble MeM Me@ was adduct formation. On the other hand the electron-withdrawing nature of the pyridine nucleus conjugated to the u-\ quinone ring enhanced the redox reaction. The trimethyl ester MeOOC N 0 MeOOC ''OH N of PQQ (29) was mainly converted into the quinol derivative (29)H in the reaction with phenylhydrazine hydrochloride in NR methanol although benzindolequinone (41) and phenanthrene- (57) R = H (58)R=Bd quinone gave the corresponding hydrazone adduct under the (59) R = cyclopropyl same conditions.All these results could be interpreted by an ionic mechanism through the C-5 carbinolamine-type adduct COOMe intermediate (56) from which both the redox reaction (path a) and the adduct formation reactions (path b) proceed depending on the factors mentioned above (Scheme 8). Because PQQ or a structurally related compound was MeOOC MwoH formerly believed to be the organic cofactor of copper-containing amine oxidases bacterial methylamine dehydrogen- ases and methylamine oxidase many investigations have been MeOOC carried out on the reaction of PQQ with amines during the 1980s.Although the idea of PQQ being the cofactor of the amine oxidizing enzymes has now turned out to be incorrect the reaction of PQQ with amines is still noteworthy because PQQ has been reported to play a fundamental role in the shown in Scheme Also in this case several products are crosslinking of collagen and elastin during connective tissue produced from PQQ depending on the substrates and on the biogenesis and in the regulation of intracellular spermine and reaction conditions. The iminoquinone derivatives (57) (58), spermidine levels.39 In fact PQQ has been demonstrated to be and (59) are easily isolated from the reaction of (29) with an excellent turnover catalyst for the oxidative deamination of ammonia t-butylamine and cyclopropylamine respectively.amines even in non-enzymatic systems; this was the first The corresponding aminophenol (57)H can be generated by example of quinone-catalysed amine oxidation (Scheme 9).40 the reduction of (57) with methylhydrazine. In the reaction of Kinetic studies and product analyses indicated that the (29) with benzylamine in CH,CN under anaerobic conditions amine oxidation reaction proceeds viaan ionic (transamination) (57)H is directly formed together with quinol (29)H,. More mechanism through the C-5 amine adduct (carbinolamine) as interestingly alkylaminophenol (60) is produced together with NATURAL PRODUCT REPORTS 1995 COOR COOH HOOC R (61) (62) R = H CH3 C6H5 CH&OOH CH ,C H2COOH 5-imidazolymethyI CHflH etc CONMe2 MeHo Me2NOC Me (63) (29)H and (57)H2 in the reaction with n-propylamine under the same conditions where the product ratio of these products is quite different depending upon the amine c~ncentration.~~ In the case of N-methylpropylamine only (29)H is isolated but its reactivity is relatively low as expected and no redox reaction occurs in the case of tertiary amines such as triethylamine.On the other hand the pyrazine derivative (61) is obtained in the reaction of PQQ with eth~lenediamine."~~~ All these results strongly support the proposed ionic mechanism46 and are consistent with the mechanism for the reaction with hydrazines mentioned above.As an extension of amine substrates the reaction of coenzyme PQQ with several amino acids has also been investigated in ~itro.~~ PQQ catalyses the oxidative decarboxylation of a-amino acids to afford the corresponding aldehydes under aerobic conditions (Scheme 11). During the catalytic cycles PQQ is i RCHO + Cop + NH3 R = Ph; 3400% (based on PQQ) Reagents i PQQ (1 mol "/o) cetyltrimethylammonium bromide pH 7.0 30 OC 5 h aerobic conditions Scheme I1 gradually deactivated by conversion into the oxazolopyrrolo- quinoline (OPQ) derivative (62). Product analyses indicate that the reaction proceeds via an ionic mechanism that involves a carbinolamine-type intermediate as in the case of hydrazines and amines. From this intermediate direct decarboxylation (to give the quinol) dehydration followed by decarboxylation and hydrolysis (to give the aminophenol) and intramolecular cyclization (to give the OPQ derivative) competitively occur ;the first of these is the major pathway the last two being minor.In the reactions with the P-hydroxy amino acids oxidative dealdolation ((2,-C fission) proceeds effectively (Scheme 12). The C,-C fission was also observed for tyrosine and COOH -RCHO + HCOC02H + NH3 R R = Ph; 678% (based on PQQ) Reagents i PQQ (3.3 mol YO),cetyltrimethylammonium bromide pH 10.7 30 "C 5 h aerobic conditions Scheme 12 Scheme 13 7- PhCH=NH HO HN A H OH HXPh quinol Scheme 14 A trypt~phan.~~similar ionic mechanism that involves a carbinolamine-type intermediate is suggested by the product analysis under both aerobic and anaerobic conditions.Nitroalkanes react with o-quinones to form I ,3-dioxolane derivatives. For example the 1,3-dioxolane derivative (63) is easily obtained in the reaction of (19) with nitroethane under weakly alkaline condition^.^^ One plausible mechanism is that electron transfer which results in formation of a radical pair between the semiquinone and the substrate is followed by radical coupling in the solvent cage. Subsequent intramolecular cyclization affords the 1,3-dioxolane derivatives as shown in Scheme 13. This novel reaction is interesting in relation to the action of nitroalkane oxidase because this enzyme also contains a quinonoid cofactor in addition to fla~in.~O Recently Martin et al.developed several kinds of chemical derivatization of PQQ to apply as a sensitive detection proced~re.~~,~~ 4 Function as Efficient Redox Catalysts As already mentioned above PQQ and the closely related o-quinone compounds act as efficient turnover catalysts in the aerobic oxidation of amines and amino acids (Schemes 9 11 and 12). This finding has been applied to the development of a detection method for PQQ based on glycine-dependent redox cycling.52 The method is sufficiently sensitive for PQQ because of its effectiveness in catalytic amine oxidation. Studies of structure-reactivity relationships using several model compounds indicated that such efficient catalysis in amine oxidation is derived from the pyrroloquinolinequinone skeleton of pQQ.29.30.42 The pyridine nucleus having an electron-withdrawing nature facilitates the nucleophilic addition of amines and stabilizes the carbinolamine intermediate thus formed (see Scheme 10).We propose that the conjugated system formed by the C-4 carbonyl and the pyrrole ring which has a dissociable proton at the 1-position is essential for the operation of intramolecular general-base catalysis whereby an a-proton is abstracted from the amine substrate (Scheme 14). 51 NATURAL PRODUCT REPORTS 19954. ITOH AND Y. OHSHIRO n Catalase Diaphorase HLADH Scheme 15 PQQ displays catalytic efficiency in other systems. For example PQQ and related heterocyclic o-quinones have been used as efficient catalysts for the regeneration of NAD' from NADH by molecular oxygen which can be adapted to the synthetically useful alcohol-dehydrogenase-catalysed oxidation reaction (Scheme 15).53-55 It is interesting to note that this system works very well even under heterogeneous conditions in non-polar organic solvents with a tiny amount of water.56 The heterocvclic o-quinones also function as electron transfer ~atalys<s~~ for the electrochemical regeneration of NAD'.PQQ has also been found to be a redox cofactor for glucose dehydrogenases from various microorganisms. Even in the absence of the enzyme the redox reaction between PQQ and glucose proceeds in an alkaline aqueous solution (Scheme 16). CH20H CH20H HO OH LQWH Q HO OH OH Reagents PQQ OH- H,O Scheme 16 Kinetic studies of the redox reaction using glucose and related substrates indicated that the active species is the 1,2-ene-diolate formation of which is the rate-determining step (Scheme 17).The mechanism of enediolate-oxidation by PQQ has not yet been clarified but an electron-transfer mechanism seems to be plausible; PQQ acts as an efficient turnover catalyst when the reaction is carried out under aerobic condition^.^^ Oxidation of thiols by PQQ to the corresponding disulfides also proceeds very efficiently. Kinetic studies of the reaction under anaerobic conditions provide a bell-shaped pH-rate profile having a maximum rate at around the pK of the thiol. Interestingly the rate-determining step is different on each side of the profile indicating the existence of at least one intermediate in the course of the reaction.A covalent adduct is also proposed to be the key intermediate for the thiol oxidation (Scheme PQQ acts as an efficient turnover catalyst in the autoxidation of thiols where asymmetric disulfide is obtained predominantly in the reaction of thiophenol and t-butylthiol (Scheme 19).'j0 Interactions between PQQ and metal ions are thought to be involved in certain enzymatic systems. It appears from recent research that the presence of Ca2+ is a common feature of PQQ- containing alcohol and glucose dehydrogenases.61 The authors have been trying to make Ca2+ complexes of (31) and (32) to obtain information about the enzyme active sites.62 Na+ was shown to coordinate to PQQ at N-6 and the carboxylate group at C-2 in a crystal made in a phosphate buffer solution.63 Coordination of PQQ with transition metal ions has also been st~died~~-~~ and applied to facilitate some transition-metal catalysed reaction^.^^ %O H+OH Rate limiting H+OH CH20H CHflH 1,P-Enediolate Scheme 17 H"\ ""1 PQQ -kl *lNQO -0 SPh HO SPh OH + PtSSPh Scheme 18 + 0S-SBu' 66% + BU'S -SBU' 0% Reagents i PQQ (1 mol YO), cetyltrimethylammonium bromide pH 7 30 OC 1 h aerobic conditions Scheme 19 5 Synthesis and Characterization of a TTQ Model Compound Tryptophan tryptophylquinone (TTQ) is a novel amino-acid- derived cofactor that was found in bacterial methylamine dehydrogenases (MADH; EC 1 .4.99.3) in 1991.i.68 It also has a unique heterocyclic o-quinone structure i.e.6,7-indole-quinone with a 2-indolyl group at the 4-position. It is however very difficult to isolate the cofactor intact and study its chemistry because TTQ is tightly associated in the enzyme matrix through a peptide linkage. In order to obtain in- formation on the structure and reactivity of the active-site NATURAL PRODUCTS REPORTS 1995 6 Conclusion In this article the chemistry of the newly found heterocyclic o-quinone cofactors PQQ and TTQ and their related compounds Me bR2 H have been reviewed and several interesting aspects of their OMe &€I (64) (65) R2 = C02Me (66) R2=C02H (67) R2=H Me (68)R3=Me (69) R3= H Table 1 Comparison of the physical properties of model compound (70) and native TTQ (70) TTQ Dihedral angle of the two indole rings 46.9” 42b Resonance Raman/v,, (cm-’)d E,, vs SCE (mV) at pH 7.3 1628 -188 162Y -148c UV/VIS/~,, (nm) 407 429 Molecular orbital calculations were performed with the MOPAC program (version 6.1 AM1 method) using a CAChe Worksystem (SONY Tektronix); M.J. S. Dewar E. G.Zoebisch E. F. Healy and J. J. P. Stewart J. Am. Chem. SOC.,1985 107 3902. *Ref. 68. ‘M. Husain and V. L. Davidson Biochemistry 1987,26,4139. KBr disk. ‘G. Backes V. L. Davidson F. Huitema J. A. Duine and J. Sanders-Loehr Biochemistry 1991 30 9201. ’W. C. Kenny and W. McIntire. Biochemistrv 1983 22 3858. cofactor model compound (70) was synthesized by the Friedel-Crafts acylation with propionyl chloride on indole derivative (64) by the standard method gave the 4-acylated compound (65) (93 YO),which was converted into (67) by ester hydrolysis (79 YO)followed by thermal decarboxylation using CuCrO in quinoline (59%).The second indole ring was constructed in 71 O/O yield by Fischer indolization with phenylhydrazine hydrochloride on (67). Deprotection of the methoxy group of (68) by trimethylsilyl iodide gave the 7-hydroxy derivative (69) (93 YO),which was finally converted into the expected quinone (70) in 57 O/O yield by oxidation with Fremy’s salt. Table 1 shows representative data of the physical properties of the model compound for comparison to those of native TTQ.Theoretical calculations using the AM1 method for compound (70) indicate that its molecular geometry (the dihedral angle of the two indole rings) is very close to that of TTQ in the active site of the enzymes. This conclusion is supported by the ‘H NMR analysis (NOE) of compound (70). The redox potential and spectral characteristics such as UV/VIS and resonance Raman of (70) are also very similar to those of the native enzymes. Furthermore model compound (70) acts as an efficient turnover catalyst in the aerobic oxidation of benzylamine in methanol indicating that it possesses the same chemical function of methylamine dehydrogenases as well as the same physicochemical properties. chemical functions have been introduced. Although many issues remain to be resolved interesting characteristics of the cofactors have been identified.Further studies of the reaction mechanisms structure-reactivity relationships and bio-synthesis will provide much more information not only for quinoprotein chemistry but also for general organic and synthetic organic chemistry. Acknowledgements. The authors acknowledge with gratitude the contributions of Dr Mitsuo Komatsu Dr Toshikazu Hirao Dr Minae Mure and the graduate students listed in our reference papers. The authors also thank Prof. Shunichi Fakuzumi for his valuable discussions. The work of the authors and their coworkers carried out in this field was partially supported by grants from the Ministry of Education Science and Culture of Japan. 7 References I J.A. Duine and J. Frank Jzn Trends Biochem. Sci. 1981,6 278. 2 S. A. Salisbury H. S. Forrest W. B. T. Cruse and 0. Kennard Nature 1979 280 843. 3 J. Westerling J. Frank and J. A. Duine Biochem. Biophys. Res. Commun. 1979. 87 719. 4 J. A. Duine and J. A. Jongejan Annu. Rev. Biochem. 1989 58 403. 5 C. Hartmann and J. P. Klinman BioFuctors 1988 I 41. 6 S. M. Janes D. Mu D. Wemmer A. J. Smith S. Kaur D. Maltby A. L. Burlingame and J. P. Klinman Science 1990 248 981. 7 W. S. McIntire D. E. Wemmer A. Chistoserdov and M. E. Lidstrom Science 1991 252 817. 8 N. Ito S. E. V. Phillips C. Stevens Z. B. Ogel M. J. McPherson J. N. Keen K. D. S. Yadav and P. F. Knowles Narure 1991 350 87. 9 K. Matsushita and 0.Adachi in ‘Principles and Applications of Quinoproteins’ ed.V. L. Davidson Marcel Dekker New York 1993 p. 355 and references cited therein. 10 F. M. Steinberg C. Smidt J. Kilgore N. Romero-Chapman D. Tran D. Bui and R. B. Rucker in ‘Principles and Applications of Quinoproteins’ ed. V. L. Davidson Marcel Dekker New York 1993 p. 367 and references cited therein. 11 M. A. Paz R. Fliickiger and P. M. Gallop in ‘Principles and Applications of Quinoproteins ’ ed. V. L. Davidson Marcel Dekker New York 1993 p. 381 and references cited therein. 12 A. Watanabe T. Tsuchida H. Nishigori T. Urakami and N. Hobara in ‘Principles and Applications of Quinoproteins’ ed. V. L. Davidson Marcel Dekker New York 1993 p. 395 and references cited therein. 13 A. Netrusov in ‘ Principles and Applications of Quinoproteins’ ed.V. L. Davidson Marcel Dekker New York 1993 p. 409 and references cited therein. 14 I. Karube K. Yokoyama and Y. Kitagawa in ‘Principles and Applications of Quinoproteins’ ed. V. L. Davidson Marcel Dekker New York. 1993 p. 429 and references cited therein. 15 Y. Ohshiro and S. Itoh Bioorg. Chem. 1991 19 169; Y. Ohshiro and S. Itoh in ‘Principles and Applications of Quinoproteins’ ed. V. L. Davidson Marcel Dekker New York 1993 p. 309. 16 M. Tisler Adv. Heterocycl. Chem. 1989 45 37. 17 E. J. Corey and A. Tramontano J. Am. Chem. Soc. 1981 103 5599. 18 J. A. Gainor and S. M. Weinreb J. Org. Chem. 1981 46,4319; J. A. Gainor and S. M. Weinreb J. Org. Chem. 1982 47 2833. 19 J. B. Hendrickson and J. G. deVries J.Org. Chem. 1982 47 1148; J. B. Hendrickson and J. G. deVries J. Org. Chem. 1985 50 1688. 20 A. R. MacKenzie C. J. Moody and C. W. Rees J. Chem. Soc. Chem. Commun. 1983,1372; A. R. MacKenzie C. J. Moody and C. W. Rees Tetrahedron 1986 42 3259. 21 G. Buchi. J. H. Botkin G. C. M. Lee and K. Yakushijin J. Am. Chem. Soc. 1985 107 5555. 22 P. Martin Helv. Chim. Acta 1993 76 988. NATURAL PRODUCT REPORTS 1995-S. ITOH AND Y. OHSHIRO 23 P. Martin E. Steiner K. Auer and T. Winkler Helv. Chitn. Acta 1993 76. 1667. 24 P. R. Sleath. J. B. Noar G. A. Eberlein and T. C. Bruice J. Am. Chem. So(... 1985 107 3328. 25 J. B. Noar and T. C. Bruice J. Org. Chem. 1987 52 1942. 26 S. Itoh J. Kato. T. Inoue Y. Kitamura M. Komatsu and Y. Ohshiro Sjsnthesis 1987 1067.27 S. Itoh T. Inoue Y. Fukui X. Huang M. Komatsu and Y. Ohshiro Chem. Lett. 1990 1675. 28 S. Itoh. X. Haung M. Komatsu and Y. Ohshiro. unpublished results. 29 S. Itoh Y. Fukui S. Haranou M. Ogino M. Komatsu and Y. Ohshiro J. Ot-g. Chem. 1992 57 4452. 30 S. Itoh Y. Fukui. M. Ogino S. Haranou M. Komatsu and Y. Ohshiro. J. Org. Chern. 1992 57 2788. 31 J. A. Duine. J. Frank and J. A. Jongejan Adv. Enzymol. Relar. Subj. Biochem.. 1987 59 169. 32 S. Itoh M. Ogino Y. Fukui. H. Murao M. Komatsu Y. Ohshiro T. Inoue Y. Kai and N. Kasai. J. Am. Chem. Soc. 1993 115 9960. 33 R. H. Dekker. J. A. Duine J. Frank P. E. J. Verwiel and J. Westerling Eut-. J. Biochem. 1982 125 69. 34 R. A. van der Meer A. C. Mulder J. A. Jongejan and J.A. Duine FEBS Lett. 1989 254 99. 35 Duine and his coworkers developed the ‘so-called hydrazine method’ as a detection method for PQQ. Their procedure led to incorrect conclusions in the case of covalently bound cofactors but their strategy has been widely applied to several enzymes to discover interesting new cofactors; see V. L. Davidson in ‘Principles and Applications of Quinoproteins’ ed. V. L. Davidson Marcel Dekker New York 1993 p. 3 and references cited therein. 36 M. Mure K. Nii T. Inoue S. Itoh and Y. Ohshiro J. Chem. Soc. Perkin Trnn.~.I 1990 315. 37 M. Mure K. Nii S. Itoh and Y. Ohshiro Bull. Chem. Soc. Jpn. 1990 63 41 7. 38 H. van Koningsveld J. C. Jansen J. A. Jongejan J. Frank and J. A. Duine Actci Crj-stallogr.Section C Cr).stal Struct. Commun. 1985 41 89. 39 J. Killgore. C. Smidt. L. Duich N. Romero-Chapman D. Tinker K. Reiser. M. Melko D. Hyde. and R. B. Rucker. Science 1989 245 850. 40 Y. Ohshiro. S. Itoh K. Kurokawa J. Kato T. Hirao and T. Agawa. Tc2truhedron Lett. 1983 24 3465. 41 Y. Ohshiro and S. Itoh Bioorg. Chem. 1991 19 169 and references cited therein. 42 S. Itoh M. Mure M. Ogino. and Y. Ohshiro J. Org. Chem. 1991 56 6857. 43 M. Mure S. Itoh and Y. Ohshiro Chem. Lett. 1989 1491. 44 S. N. Gacheru P. C. Trackman S. D. Calaman F. T. Greenaway. and H. M. Kagan J. Biol. Chem. 1989 264 12963. 45 Derivatization to the pyrazine-type chromophores was developed as a detection method for PQQ; P. M. Gallop E. Henson M. A. Paz S. L. Greenspan and R.Fliickiger Biochem. Biophys. Res. Comtnun. 1989 163 755; P. Martin and T. Winkler Helv. Chim. Actci 1993 76. 1678. 46 Mechanistic details are also discussed in the literature; E. J. Rodriguez and T. C. Bruice J. Am. Chem. Soc. 1989 11 1 7947. 47 S. Itoh N. Kato Y. Ohshiro and T. Agawa Terrahedron Lett. 1984 25 4753; M. Mure A. Suzuki S. Itoh and Y. Ohshiro J. Chem. SOC.,Chem. Common. 1990. 1608; S. Itoh M. Mure A. Suzuki H. Murao and Y. Ohshiro J. Chem. Soc. Perkin Trans. 2 1992 1235. 48 M. A. G. Van Kleef J. A. Jongejan and J. A. Duine Eur. J. Biochem. 1989 183 41. 49 S. Itoh K. Nii M. Mure and Y. Ohshiro Tetrahedron Left. 1987 28 3975. 50 T. Kido and K. Soda Seikagaku 1985 57 1065. 51 P. Martin and T. Winkler. Helv.Chim. Actu 1993 76. 1674. 52 R. Fliickiger M. A. Paz E. Henson and P. M. Gallop in ‘Principles and Applications of Quinoproteins’. ed. V. L. Davidson Marcel Dekker New York 1993. p. 331. 53 S. Itoh M. Kinugawa N. Mita and Y. Ohshiro J. Chem. Soc. Chem. Commun. 1989 694. 54 S. Itoh N. Mita and Y. Ohshiro Chem. Lett. 1990. 1949. 55 K. Ichinose F. Leeper J. Finian and A. R. Battersby J. Chem. Soc. Perkin Trans. 1 1993 1213. 56 S. Itoh T. Terasaka M. Matsumiya M. Komatsu and Y. Ohshiro J. Chem. Soc. Perkin Trans. 1 1992 3251. 57 S. Itoh H. Fukushima M. Komatsu and Y. Ohshiro Chem. Lett. 1992 1583. 58 S. Itoh M. Mure and Y. Ohshiro J. Chem. Soc. Chem. Commun. 1987 1580. 59 S. Itoh N. Kato M. Mure and Y. Ohshiro Bull. Chem. Soc. Jpn.1987 60 420. 60 S. Itoh N. Kato Y. Ohshiro and N. Kato Chem. Lett. 1985 135. 61 S. White G. Boyd F. S. Mathews Z. Xia W. Dai Y. Zhang and V. L. Davidson Biochemistry 1993 32 12955 and references cited therein. 62 S. Itoh X. Huang M. Komatsu and Y. Ohshiro unpublished results. 63 T. Ishida M. Doi K. Tomita H. Hayashi M. Inoue and T. Urakami J. Am. Chern. Sue. 1989 111 6822. 64 J. B. Noar E. J. Rodriguez and T. C. Bruice J. Am. Chem. SOL‘. 1985 107 7198. 65 S. Suzuki T. Sakurai S. Itoh and Y. Ohshiro Inorg. Chem. 1988 27 591; S. Suzuki T. Sakurai S. Itoh and Y. Ohshiro Chem. Lett. 1988 777. 66 B. Schwederski V. Kasack W. Kaim E. Roth and J. Jordanov Angew. Chem. Int. Ed. Engl. 1990 29 78. 67 T. Hirao T. Murakami T. Ohno and Y.Ohshiro Chem. Lett. 1989,785;T. Hirao M. Ohno and Y. Ohshiro Tetrahedron Lett.. 1990 31 6039; T. Hirao T. Murakami M. Ohno and Y. Ohshiro Chem. Lett. 1991 299. 68 L. Chen F. S. Mathews V. L. Davidson E. G. Huizinga F. M. D. Vellieux J. A. Duine. and W. G. J. Hol FEBS Lett. 1991 287 163; L. Chen F. S. Mathews V. L. Davidson E. G. Huizinga F. M. D. Vellieux and W. G. J. Hol Proteins 1992 14 288. 69 S. Itoh M. Ogino M. Komatsu and Y. Ohshiro J. Am. Chem. Soc. 1992 114 7294.

ISSN:0265-0568    

DOI:10.1039/NP9951200045

出版商:RSC    

年代:1995

数据来源: RSC

摘要:

The Biosynthesis of Plant Alkaloids and Nitrogenous Microbial Metabolites R. B. Herbert School of Chemistry University of Leeds Leeds LS2 9JT neviewing the literature published between January and December 1992 (Continuing coverage of literature in Natural Product Reports Vol. 10 p. 575) 1 Pyrrolidine and Piperidine Alkaloids 1.1 Pyrrolidine Alkaloids 1.2 Deoxynojirimycin and Related Metabolites 1.3 Pyrrolizidine Alkaloids 2 Isoquinoline Alkaloids 3 Indole Alkaloids 3.1 Terpenoid Indole Alkaloids 3.2 Indigo and lndoxyl 3.3 Ergot Alkaloids 3.4 Sparsomycin 3.5 Chaetoglobosins 4 Other Metabolites Derived from the Shikimate Pathway 4.1 Ansatrienins and Rifamycins 4.2 Erbstatin 4.3 Cyanogenic Glycosides Betalains and Taxanes 4.4 Pyrroloquinoline Quinone 4.5 GIyantrypine 4.6 Obafluorin 4.7 Pyoverdins Desferriferribactins Lincomycins and Sirodesmins 5 p-Lactams 5.1 Clavulanic Acid 5.2 Penicillins and Cephalosporins 6 Miscellaneous Metabolites 6.1 Isocyanides Cyanides and Cyanogenic Glycosides 6.2 Anatoxin a(s) Capreomycin Acivicin Bellenamine and 3-Nitropropanoate 6.3 Valanimycin 6.4 Dynemicin A and Domoic Acid 6.5 Antibiotics A10255 6.6 Nikkomycins and Griseolic Acids 6.7 Aminoglycoside-aminocyclitol Antibiotics 7 References Following past practice appropriate reference is made in the following to earlier reports and reviews which provide useful background information.'-8 In part access to the literature was obtained through the IS1 Data Service at Bath.1 Pyrrolidine and Piperidine Alkaloids 1.1 Pyrrolidine Alkaloids [13C2]Acetatehas been found in root cultures of Hysoscyamus albus to be incorporated into hyoscyamine (8) and 6p-hydroxyhyoscyamine (9) in an unexpected manner i.e. labelling equally split between (1 1) and (lq9 (ref. 4 p. 508). This has been confirmed for 6p-hydroxytropine (10) in Datura stra-moniurn plants.lo Further it was found that [13C4]acetoacetate failed to act as an intact precursor the same result being observed as for [13C2]acetate.This result was surprising as acetoacetate (2) has hitherto been held8 to be a key precursor which condenses with (1) as shown to give hygrine (3) (Scheme I path a).It was also foundlo that [2-*H]-N-methy1-A1-pyrrolinium chloride [as (l)] yielded (10) labelled equally on 55 both bridgehead positions. This and the acetate result lead to the conclusion that there is a racemic intermediate which may be (3) or (6) in tropane alkaloid biosynthesis (see ref. 9 for a similar conclusion). The acetoacetate result indicates a pathway for tropane alkaloids which is similar to that of cocaine (16) (ref. 5 p. 576) and which involves two sequential additions of acetate with (5) as an intermediate. Further results are awaited with keen interest. Resolution at the enzymic level would be most welcome and probably definitive. Two tropinone reductases have been identified invivo (ref.5 p. 575). One transforms tropinone (4) into tropine (7) and the other transforms it into Y-tropine [C-3 epimer of (7)]. Further work with cultured roots of Hyoscyamus niger has been published." This formation of Y-tropine is essentially irreversible whilst that of (7) is reversible. The two enzymes have been purified to near homogeneity. Both are Class B oxidoreductases that stereospecifically transfer the pro-S hydrogen of NADPH to their substrates. Possible roles of these reductases in alkaloid biosynthesis have been discussed." Nortropane sulfur analogues (ref. 5 p. 575) have been used to study the effect of the reductases on the spectrum of tropane esters produced in root cultures of D. stramonium.12 The characteristics of growth and tropane alkaloid pro-duction in hairy-root cultures of H.albus which had been transformed with Agrobacterium rhizogenes have been stud-ied.13The conversion of hyoscyamine (8) into scopolamine and 6-hydroxyhyoscyamine (9) has been noted in regenerating shoot cultures of a Duboisia hybrid.14 H-B ;c= 0 IF CHQCSCOA 0 0 (3) (4) COAS Me OR2 OH HOHS H (7) (8) R' = H; R2 = P,+ n 0 HOH& H CoAS (9) R' = OH; R2 = Ph* 0 00 (10) R' =OH; R2= H Scheme 1 NATURAL PRODUCT REPORTS 1995 +t clio CH20H l=o *H+OH *H+OH HO* HO Ho HA OH HO -HO& HA OH (20) (19) Dehydration and reduction I HO HO%HA HO*HO HA HoH OH ', Scheme 2 Putrescine is an important early intermediate in nicotine biosynthesis.8 Diamine oxidase is involved in its oxidation and ornithine decarboxylase and arginine decarboxylase in its formation.The levels of these enzymes have been studied using transformed tobacco ~1ants.l~ N-Methylputrescine oxidase has been purified from Nicotiana rustica.16 The effects have been studied of nicotinic acid N-methylpyrrolinium salt (l) and A'-piperideine on alkaloid production in root cultures of Nicotiana ala ta. The benzoyl moiety in cocaine (16) originates in phenyl- alanine (13)? and probably involves the CoA ester of benzoic acid (from results with the N-acetylcysteamine thioester) (ref. 2 p. 105). It has now been shown that the N-acetylcysteamine thioester of [3-13C,14C]-trans-cinnamic acid [as (14)] is a specific and efficient precursor for (1 6) in Erq'tho.~ylumcoca.'8 The next precedented steplg is hydration to give 3-hydroxy-3-phenyl- propanoic acid [as (15)] or its CoA ester the (R)-(+)-acid (15) gave a satisfactory incorporation and was found to be a much better cocaine precursor than the (S)-isomer." It is to be noted that the configuration of (15) is the same as that which CO2H H Q&R1 R2 (23) (24) R' =OH; Rz = Me (25) R'=Me; R2=OH "'0 MmQ HO I Meon HO Meon HO Meon HO Me0 0 .OH results from double-bond hydration in the course of P-oxidation of fatty acids.1.2 Deoxynojirimycin and Related Metabolites Cultures of Streptomyces subrutilis produce nojirimycin (20) deoxynojirimycin (22) mannonojirimycin (19) and deoxy- mannonojirimycin (2 1),2" which are representatives of a group of alkaloids in which there is currently great biological interest.The biosynthesis of these compounds has been the subject of a preliminary publication2' (ref. 5 p. 576) and the economical and elegant results that have been obtained are now the subject of a full paper.'" A set of specifically deuteriated samples of glucose (17) have been used as precursors and the labelling patterns that result are summarized in Scheme 2. The shift of deuterium from C-2 of (17) to appear at C-6 in (21) and (22) implicates fructose ( 18) as an intermediate the isomerization of (17) to (18) involving a proton shift from C-2 to C-1.[6,6-'H,]Glucose [as (17) A = 2H]was incorporated into (21) and (22) with retention of a single deuterium atom. This indicates NATURAL PRODUCT REPORTS 1995-R. B. HERBERT that oxidation of C-6 of (1 7)/( 18) occurs (to the aldehyde level) and that (19) and (20) are intermediates. Deuteriated nojirimycin (20) labelled both (21) and (22) which supports this deduction and also indicates that epimerization at C-2 occurs at the (19)/(20) stage as shown [a small amount of epimerization was observed with (2 I)]. Epimerization results in loss of a label originally present at C-5 in (17) (21) but not (22) was labelled. This places (19) as the first alkaloid formed from (17)/(18). 1.3 Pyrrolizidine Alkaloids The stereochemistry of enzymic processes in the biosynthesis of pyrrolizidine alkaloids has been authoritatively reviewed.22 The cooccurrence of 2-pyrrolidineacetic acid (23) with the pyrrolizidine alkaloids tussilaginic acid (24) and isotussilaginic acid (25) (also C-1 epimers) in several Arnica species suggests a possible biogenesis for these offbeat pyrrolidine types with (23) as an intermediate.2s It is to be noted that the N-methyl CoA derivative of (23) is implicated in the biosynthesis of cocaine (1 6) and possibly also of tropane alkaloids (Section 1.1 Scheme 1) (ref.5 p. 576). 2 lsoquinoline Alkaloids The biosynthesis of benzylisoquinoline alkaloids,R e.g. reticuline is universally from (S)-norcoclaurine (26) (ref. 5 p. 579 and earlier reports"). Thus biosynthesis naturally affords (S)-reticuline (27).However it is (R)-reticuline (29) which is used for the biosynthesis of morphinan alkaloids e.g. thebaine (32). The necessary inversion of configuration occurs via 1,2-dehydroreticuIine (28).' The enzyme responsible for the reduction 1.2-dehydroreticuline reductase has been isolated from seedlings of the opium poppy (Papaver sornniferurn) purified to apparent homogeneity and characterized.24 The enzyme mediates the transfer of the pro-S hydride of NADPH to C-1 of (28). It shows high substrate specificity (neither 1,2- dehydronorreticuline nor 1,2-dehydrococlaurine are substrates) and is present only in plants containing morphinan alkaloids. Phenol oxidative coupling in vivo on (29) affords salutaridine (30) (ref.4 p. 51 1 ; ref. 5 p. 579) chemical reduction of which affords two epimeric alcohols [as (31)]. One of these alcohols and only one is then an in vivo precursor for morphinan alkaloids e.g. (32).6,8 An oxidoreductase has been isolated from P. somnferurn purified and characterized; it shows high substrate specificity and stereospecificity."j It catalyses in the presence of NAPDH the reduction of salutaridine to the alcohol (3 I) which is a very efficient precursor for morphinan alkaloids in vivo (its epimer is not incorporated). This alcohol was quite unequivocally established to have the structure and absolute configuration shown in (3 1). This is important because the biologically active alcohol had originally been assigned the opposite stereo-chemistry.'6 Further the alcohol (3 l) now called salutari- dinol,25 has the 'proper -stereochemistry for the S,2' reaction leading to (32).These alcohols were previously and now confusingly called salutaridinol I and 11. 3 lndole Alkaloids 3.1 Terpenoid Indole Alkaloids The complete mRNA sequence for the enzyme strictosidine synthase from Cutharanthus roseus has been determined ; isozymes appear to occur post-translationally.2i The synthase gene from Rau~tdfiaspecies has been subject to detailed genetic analysis.2H The gene encoding strictosidine synthase (from C. roseus) and that encoding tryptophan decarboxylase are co- ordinately regulated and a first important regulating event in the biosynthesis of terpenoid indole alkaloids is transcription of these biosynthetic genes.27 29 Analysis of the gene for strictosidine synthase in ten Rauwolfia species reveals high (34) R = H (35)R =Ac (37) HO (39)R' = R2 = H (40)R' = Me; R2= H (41) R' =Me; R2= OH II OH CHpOH 'N H (42) conservation of sequence.:3o This indicates that unless these species have only diverged recently there is a stringent selection pressure for strictosidine synthase.Cell suspension cultures of plants from three different families which produce terpenoid indole alkaloids have been examined for alkaloid production and activity of chorismate mutase isochorismate synthase anthranilate synthase tryptophan decarboxylase and strictosidine synthase.31 Further work on the production of alkaloids in C.roseus cultures has been reported.s2 Tabersonine (33) is an intermediate in the biosynthesis of vindoline (35).H The last step in the biosynthesis of (35) is an acetylation which is catalysed by an 0-acetyltransferase which converts (34) into (35) (ref. 4 p. 514). Following earlier findings that the appearance of the enzyme is light induced it has been found that phytochrome is involved in this induction in C. roseus seedling^.:'^ In the quest for intermediates in the biosynthesis of vindoline the metabolism of tabersonine (33) in cell cultures of C. roseus has been studied.:34 It was found that lochnericine (36) and lochnerinine (37) were metabolites which were presumably sequentially formed. Metabolism of ajmaline (38) by cell cultures of R.serpentina has been found to yield raumacline (39) and its derivatives (40) and (41) which are new alkaloids.35 Illustration is hereby provided of the ability of cell cultures to make new compounds which may be of biological interest and application. 3.2 Indigo and Indoxyl Indigo in plants is formed by air oxidation of indoxyl which is itself the product of the action of P-glucosidases on indican (42) or isatan B (43). In rigorous experiments with four plant species from different families it has been showns6 using [3-'"C]indole and ~-[ring-3-TC]tryptophan that (42) and (43) derive from indole (from a very efficient and specific incorporation) but not tryptophan. This is consistent with the reported absence of NATURAL PRODUCT REPORTS 1995 XOPP y.H Scheme 3 tryptophanase in plants which was confirmed in these experiments. Previous evidence that tryptophan is a precursor is shown to be wrong (ref. 4 p. 519). 3.3 Ergot Alkaloids The first pathway-specific step in the biosynthesis of ergot alkaloids e.g. elymoclavine (47) is the alkylation of tryptophan (45) with dimethylallyl pyrophosphate (DMAPP) (44) which yields dimethylallyltryptophan (46). This reaction is catalysed by dimethylallyltryptophan synthase. The enzyme has been isolated from Claviceps purpurea purified and characteri~ed.~’ The mechanism by which the enzyme works has been probed with a set of analogues of DMAPP and L-tryptophan.3* It was concluded from the results obtained that the reaction catalysed by the synthase is an electrophilic aromatic substitution (Scheme 3) mechanistically similar to the electrophilic alkylation catalysed by farnesyl diphosphate synthase the prenyl-transferase for which detailed mechanistic results are available.Unnatural amino acids have been incorporated into ergopeptine alkaloids. New unnatural alkaloids namely egorine (48) (an analogue of ergovaline) ergonirine (49) (an analogue of ergocornine) and ergonorinine (50) have been isolated following the administration of L-norvaline to cultures of a strain of C. p~rpurea.~’ 3.4 Sparsomycin The biosynthesis of the unusual antibiotic sparsomycin (51) in Streptomyces sparsogenes has been the subject of a preliminary report4(’ (ref. 2 p. 117). Further clarification is now available in a full paper.41 From a set of experiments with l3C-labe1led precursors which gave sparsomycin with good enrichment it is now clear that the dithioacetal moiety (55) originates from serine sequentially through L-cysteine L/D-S-methylcysteine (52) (shown to be a natural constituent of the culture by isotope dilution) L/D-S-(methylthiomethy1)cysteine (53) and L/D-S-(methylthiomethyl) cysteinol(54).(It was shown by 2H-labelling that incorporation was not by oxidation back to the acid but incorporation was lower than for the corresponding acid.) The origin of the uracil moiety of sparsomycin (51) is quite unusual in being from trypt~phan.~~.~~ A normal metabolism of tryptophan (45) involves cleavage of ring B with kynurenine and N-formylanthranilic acid as the resulting metabolites.Failure of the latter compound to act as an intact precursor for sparsomycin and negative results with corresponding hydroxy derivatives indicate that this is not the route from tryptophan to sparsomycin.41 It was suggested therefore that ring A is cleaved before ring B but further attempts to define the actual route with indole tryptophan and their hydroxy derivatives (48) R’=Me; R2=Pf (49) R‘ = PS; R2= PP (50) R’ = R2 = PP c- c- fi OH c- D ’0 (57) failed. The specific incorporation of labelled (56) shows that the last biosynthetic step in the formation of the uracil moiety of (51) involves oxidation at C-8.41 3.5 Chaetoglobosins Chaetoglobosin A (64) is biosynthesized in Chaetomium globosum from tryptophan nine acetate/malonate units and three C units derived from methionine (ref.7 p. 214). The biosynthesis of (64) has been examined further using cultures of C. s~bafine.~~.~~ In the first of these papers the authors have exploited their previous ingenious and notable use of P-450 inhibitors to accumulate new intermediates. Use of the inhibitor metapyrone led to the accumulation of the (less oxidized) prochaetoglobosins [(59)-(62)] plus two others. Not only can the steps of oxidation be sequenced in this way but it is also possible to use the inhibited cultures to make labelled precursors for feeding experiments.42 The origins of the oxygen atoms in the chaetoglobosin A (64) have been mapped using lSO2 and [l-13C,1*02]acetateas The results are summarized in (57) which also includes the results obtained for deuterium incorporation with [l-13CC,2H3]acetate.Deuterium retention could only be seen for the first four units of the nonaketide framework and the origin of the oxygen atom at C-19 could not be decided due to NMR line broadening. A Diels-Alder reaction on (58) is proposed for the biosynthesis of the chaetoglobosins. Evidence was obtained that thermolysis of (59) leads to a retro Diels-Alder process NATURAL PRODUCT REPORTS 1995-R. B. HERBERT L-Tryptophan + 9 x CH3COzH -+ 3xc R R 0 0 OH R = mcH2-H v (61) Scheme 4 CO2H COSCoA POTC02H + OH :I rH0 OH -7- NADPH CH2W (67) (65) n 1 "+NHR COSCoA COSCoA Glutamate Scheme 5 H rTco2t (68) -H2N$H H CH20P OH which provides some support for the forward reaction in viv~.~~ (69) The combined eviden~e~~p~~ leads to the biosynthetic grid to chaetoglobosin A (64) shown in Scheme 4.43It should be noted that in a feeding experiment labelled (59) was not incorporated into (64).4 Other Metabolites Derived from the Shikimate Pathway 4.1 Ansatrienins and Rifamycins The ansatrienins contain a cyclohexanecarboxylic acid residue (66) which has been shown to arise from shikimic acid (65) (ref. 5 p. 584; ref. 2 p. 120). Careful experiments with a cell-free preparation of Streptomyces collinus have given results which define the pathway to ansatrienin A as that shown in Scheme 5.44Two reductases are involved which both require NADPH as coenzyme and reduction occurs on carboxylic acids activated as coenzyme A thioester~.~~ Contained within the framework of a number of antibiotics e.g.ansatrienins and rifamycins is a C,N unit which derives by way of 3-amino-5-hydroxybenzoicacid (72)7,s,45,4s (ref. 5 J OH (71) Scheme 6 p. 584). The way in which this unusual aromatic amino acid is biosynthesized has received welcome new Although (72) is formed viathe shikimate pathway the evidence indicates that branching into the biosynthesis of (72) occurs at a stage prior to shikimic acid (65). A reasonable hypothesis has been proposed in which amination occurs early that is to say on erythrose-4-phosphate (67) yielding amino DAHP (69) via (68) (Scheme 6).45 Key support for this hypothesis has been NATURAL PRODUCT REPORTS 1995 HO I OH (73) (74) NHCHO (75) (76) HO C02H (77) R = H 0 Med obtained using cell-free extracts of Nocardia mediterranei amino DAHP (70) was efficiently converted (45%) into 3- amino-5-hydroxybenzoic acid (72) (significantly added DAHP with and without added glutamine was not converted) as was aminodehydroshikimic acid (7 1) (95 "/o) (aminoshikimic acid showed negligible con~ersion).~~ The incorporation of racemic [1,2-13C,]glycerol into (73) in cultures of a N.mediterranei mutant has been e~amined.~' The results [thickened bonds in the C,N unit in (73)] show that it is C-1 of erythrose-4-phosphate (67) [cf.(68)] which is aminated [equivalent to C-3' in (73)]. Quinic acid was tested as a precursor and was found to be a poor one. These results are consistent with the pathway to (72) illustrated in Scheme 6. 4.2 Erbstatin Results of experiments with [G-14C]shikimic acid [U-14C]phenylalanine and [U-14C]tyrosine show that tyrosine is by far the best precursor for erbstatin (74) in cultures of Streptomyces sp. MH435-hF3.49 Biosynthesis would seem to be similar to that of tuberin (75),50 but in the case of (74) further reaction occurs in which the chain undergoes a 1,2-shift concomitant with the introduction of the second hydroxy group. 4.3 Cyanogenic Glycosides Betalains and Taxanes Evidence has been obtained that the meta-hydroxylated cyanogenic glycoside zierin (76) derives from L-phenylalanine and not tyrosine or meta-tyrosine in Xeranthernum ~ylindraceum~' (for further results on the biosynthesis of cyanogenic glycosides see Section 6.1).A 4,5-dioxygenase has been isolated which catalyses the ring scission of DOPA; this is a key step in betalain biosynthesis (ref-5 P. 581). An acyltransferase has been isolated from ChenoPodium rubrum which catalyses the transfer of feruloyl or (79) (83) coumaroyl residues from acylated glucose to amaranthin (77) to give e.g. celosianin I1 (78).52 Satisfactory incorporations into taxol (79) of the expected precursors phenylalanine (cf. ref. 54) mevalonate and acetate have been reported using preparations of Taxus bac~ata.~~ 4.4 Pyrroloquinoline Quinone Pyrroloquinoline quinone (PQQ) is biosynthesized from glutamate and tyrosine (ref.5 p. 582). At least four genes are required for biosynthesis in Actinobacter calcoaceticus. The DNA region where one of these genes was mapped codes for a polypeptide of only twenty-four amino acids. This polypeptide is essential for PQQ synthesis and site-directed mutagenesis shows that at least one glutamate and one tyrosine residue are essential for its function. One or both of these amino acid residues may it is hypothesized be used as actual precursors for PQQ.55 4.5 Glyantrypine The incorporation of radioactive precursors into glyantrypine (80) in cultures of Aspergillus clavatus is consistent with the expected biosynthetic origins in anthranilic acid tryptophan and gly~ine.~~ 4.6 Obafluorin Results of a study into the biosynthesis of the curious p-lactone obafluorin (83) in Pseudomonasfluorescens previously published in preliminary form (ref.4 p. 520; ref. 3 p. 194) are now reported in full.57358 The amino acid moiety (84) of obafluorin (83) arises from L-p-aminophenylalanine (81) and glyoxylate (82)/glycine. A pathway consistent with the results obtained has been proposed (ref. 4 Scheme 16) ;simple adaptation leads to other antibiotics such as bestatin amistatin and the amiclenomycins. In essence new results58 concern (85) and (86) which are analogues of (81). The former gave traces of a new metabolite in cultures of P.fluorescens and in resting cells one of two products could be identified as (87) i.e.formed by condensation of (85) with 2,3- dihydroxybenzoic acid before or after oxidation of the aromatic amino group. The ring-opened form (88) of obafluorin (83) appears after and at the expense of obafluorin in the cultures so the sequence of late biosynthesis is one of hydrolysis not lactonization. The failure Of (85) to label (88) is consistent with this and also indicates that hydroxylation is not a late NATURAL PRODUCT REPORTS 1995-R. B. HERBERT (100)Proclaviminic Acid J 0flyNH2 -0 G-?HfiNH2 C02H cod (87) R=H (88) R =OH (102) Claviminic Acid (101) 0gyoH HO-NH Pep tide% k02H (103) ClavulanicAcid NH HO Scheme 7 01" 0i" NH2 NH2 (89) The pyrroline (91) has been isolated from a strain of Streptomyces lincolnensis which does not produce lincomycins.60 Addition of (91) to cultures of another non- producing strain of S.lincolnensis results in lincomycin production. Thus (91) is deduced to be an intermediate in the biosynthesis of the lincomycin unit (92). Ultimately this unit originates in the aromatic ring scission of dopa.7 0 Tyrosine serine and mevalonate are precursors for the sirodesmin skeleton and diketopiperazine (93) is a biosynthetic inter~nediate.~ into The incorporation of [3,5-3H,U-14C]tyrosine sirodesmins A-C [(96)-(98)] in Sirodesmium diversum confirms that all nine of the tyrosine carbons are retained in the 0 (94) biosynthesis of the sirodesmins." The pattern of acetate (93) I incorporation into the dimethylallyl-derived moiety is con- sistent with a stereospecific Claisen-type rearrangement of (93) leading to (94) which is followed by ring closure to (95).5 P-Lactams 5.1 Clavulanic Acid The biosynthesis of clavulanic acid (103) proceeds as shown in (96)n=2 (97)n=3 Scheme 7 (ref. 5 p. 584); ornithine (99) is a principal precursor (98)n=4 with C-5 of the amino acid becoming C-9 of (103) (ref. 3 p. 200). A number of other clavams of which (104) and (105) are representative were hypothesized62 to arise from ornithine in the opposite regiochemical sense i.e. C-5 of (99) became C- biosynthetic step. ~~-(4-aminophenyl)glycine (86) failed to give 3 of (104) and (105). However in cultures of Streptomyces any new detectable metabolites in normal cultures or resting clavuligerus which produced both (I 03) and (104) it was found that cells.with (2S,4S)-[4-2H,5-13C]ornithine no incorporation of either label occurred into (104); normal 13C incorporation into C-9 of (103) was observed. On the other hand [3-13C]ornithine 4.7 Pyoverdins Desferriferribactins Lincomycins and gave (103) and (104) which were both labelled on C-2. Thus Sirodesmins ornithine incorporation into both types of clavam occurs in the When grown in an iron-deficient medium Pseudomonas aptata same regiochemical sense. Moreover ~~-[2,3-~~CJproclavirninic produces both a desferriferribactin i.e. (89) and a pyoverdin acid [as (loo)] served as an intact and equally efficient precursor i.e.(90).59 The identical sequence of the peptide chain in each for both (103) and (104). It is suggested62 that the aldehyde supports the hypothesis that desferriferribactins are the (106) which has been proposed as an intermediate for the biosynthetic precursors for pyoverdins. (For earlier work see extraordinary 'enantiomerization ' of (102) which gives (103) ref. 4,p. 525). The ultimate origins for these metabolites in may also serve as an electron sink for decarboxylation leading tyrosine is apparent. to (104). Substrate tolerance by the synthase responsible for the conversion of (100) into claviminic acid (102) has been examined.":' Incubation of Ihreo-( 107)with claviminate synthase afforded the y-lactam analogue (109) of claviminic acid and (1 10)[cfi (IOl)].The erythro-compound (108) failed to give any bicyclic y-lactams on incubation with the synthase. On the basis of the structure of proclaviminic acid (100) the reverse order of reactivity might have been expected. Neither (109) nor (1 10) showed significant activity as antibacterial agents or p-lactamase inhibitors. 5.2 Penicillins and Cephalosporins Isopenicillin N (1 12) is biosynthesized from s-aminoadipate cysteine and valine by way of the LLD-ACV tripeptide (1 11). The conversion of (1 11) into (1 12) is catalysed by isopenicillin N synthase (IPNS). Information on the incorporation of oxygen from valine into (1 11) using lnO-labelled amino acid which had previously been published in preliminary form (ref. 4 p.524; ref. 5 p. 584),is now available in a full paper.64 A full paper65 has also appeared on the conversion catalysed by IPNS of the fluorinated analogue (1 13) of ACV (1 11) into the novel product (1 14) (preliminary publication ref. 5 p. 586). A wide range of ACV analogues have been tested as substrates for IPNS. The outcome of these experiments judiciously honed in further experiments with more subtly modified analogues has often provided valuable mechanistic insights into the way in which IPNS works (ref. 5 p. 586 and earlier reports1-'). In a workman-like study a wide range of ACV analogues have been further examined as substrates for highly purified IPNS from Penicillium chrysogenum and for enzyme also highly purified which was expressed in Escherichia coli viu a cloned gene derived from CephaIosporiuiiz u~rernonium.~~ Changes in the a-aminoadipic acid and valine moieties gave substrates which were less efficiently processed than ACV itself.Modifications to the cysteine moiety gave NATURAL PRODUCT REPORTS. 1995 peptides which were unable to serve as substrates. Conversion of the valine residue into an aromatic amino acid or one bearing a highly electronegative residue r.g. trifluorovaline. resulted in loss of substrate activity and enzyme inhibition. Work on the genes of p-lactam biosynthesis has been reviewed.6i The pchC gene which encodes for IPNS. has been subcloned from Streptomyces clavuligerus into E. coli that had been engineered to allow high-level expression of IPNS.68The enzyme was obtained in inactive form but could be renatured.IPNS genes from Flavohacteriirnl sp. and Strc~i~foiii!'i'c's junionjinensis have been expressed in E. coli.69 The procedure employed is suitable for large-scale production of IPNS. Oxygen utilization by IPNS (from P. C'~YI'SO~~~'~ZN~II). when catalysing the conversion of ACV (1 11) into isopenicillin N (1 12) has been measured; it is first order in respect of oxygen concentration." The glucose-regulated expression of genes responsible for isopenicillin N (1 12) biosynthesis in .4spc~rgillus nidulans has been studied in detail." The effect of L-lysine has been similarly examined.'2 The regulation by lysine of x-aminoadipate reductase from P. i~hrj~sogenum has been studied in relation to the flux from x-aminoadipate into penicillin biosynthesis.'3 X-Ray studies (EXAFS)'* have provided good evidence that in the absence of substrate (ACV) and oxygen thc iron at the active site of IPNS is coordinated by (N.0)-containing ligands in an approximately octahedral arrangement.Two or three of the Fe-(N.0) interactions are likely to be due to histidyl imidazole ligation both in the presence and absence of ACV. Upon anaerobic binding of ACV the iron coordinating environment changes considerably and Fe-S coordination occurs. Further spectroscopic studies,'j which included site- specific Cys -,Ser mutated enzymes gave results which show that Fe-S coordination is to the thiol group in ACV (1 11). This provides support for the proposed mechanisms of IPN (1 12) formation65 (ref.5 p. 586 Scheme 10; also earlier reports' '1). The biosynthesis of cephalosporins e.g. cephalosporin C NATURAL PRODUCT REPORTS 1995-R. B. HERBERT 63 H T f (122) + 1-Fe-Enz + Enz-Fe'"=O* CH3 COJi r I (115) DAOC I Scheme 8 C02H (115) DAOC; R&HS (116) R =CDs (117) DAC; R =CH20H (1 18) Cephalosponn C; R = CH20Ac C02H (121) R =H (122) R =D (1 18) begins with the isomerization of isopenicillin N (1 12) to penicillin N. Ring expansion affords deacetoxycephalosporin C (DAOC) (1 15) and hydroxylation then yields deacetyl- cephalosporin C (DAC) (1 17).8 Both of these latter steps are catalysed by a single enzyme in Cephalosporium acremonium. H RNn&!:-0 I Enz CO2H J (117) DAC Scheme 9 The mechanism of hydroxylation has been probed by incubating an equal mixture of DAOC (1 15) and its trideuterio derivative (1 16) with partially purified DAOC/DAC syntha~e.~~ A competitive kinetic isotope effect was observed in the conversion of DAOC into DAC which is consistent with a mechanism in which the first irreversible step takes place at the C-3'-methyl group (R) of (115) as might be expected.A mechanism involving direct insertion of a ferry1 species into a C-H bond at C-3' is proposed (Scheme 8). Incubation of DAC (1 17) with the C. acremonium DAOC/ DAC synthase results remarkably in a third step of oxidation to yield the aldehyde (119) which undergoes hydrolysis to ( The affinity of DAC for the acetyltransferase is higher than for the DAOC/DAC synthase however so under normal conditions the end product of cephalosporin biosynthesis is (1 18) not (1 19).The unnatural cephalosporin (121) is enzymically convertible into DAC (1 17). Incubation of the deuteriated compound (1 22) with DAOC/DAC synthase gave in addition to (1 17) some of the spiro-epoxide (125) which was not seen with (l2l).'* On further incubation the aldehyde (120) was obtained as product. An equal mixture of (121) and (122) gave DAC without a kinetic isotope effect indicating that there must be an intermediate which is formed prior to loss of the hydrogen atom at C-4. The intermediate may be (123)/(124) which in the absence of deuterium at C-4 goes by proton loss along a single path to DAC (1 17); when deuterium is present it is lost less readily so the epoxide (125) can accumulate as an alternative to formation of DAC (Scheme 9).This epoxide it is suggested reacts further to give the aldehyde (120). 6 Miscellaneous Metabolites 6.1 Isocyanides Cyanides and Cyanogenic Glycosides Following on from an earlier review of naturally occurring iso~yanides,~~ natural products containing isocyano and cyano groups have been the subject of a review that includes discussion of biosynthesis.80 Cyanogenic glycosides are biosynthesized from the cor-responding a-amino acid and the metabolic steps for each are similar (e.g. ref. 5 p. 581 ; this report Section 4.3). Thus the C02H I '02'ANH2 HNkk2 +NH2 HN H2N.NH2 (135) R=H (136) R=OH 0 H (133) R=H (137) R = H (134) R=OH (138) R = OH cyclopentenoid deidaclin (126) is derived from the unusual amino acid (128) in Turnera ulmifolia.E1 This has been confirmed for (126) in Turnera angustifoliaE2 It has been shown also that linamarin (1 27) is specifically derived in Passijlora morifolia from L-valine.82 Evidence was obtained that the corresponding nitrile in each case is involved in cyanohydrin biosynthesis and T. angustifolia was able to transform 2-methylpropionitrile into linamarin (127) even though the plant does not make (127) naturally. L-Valine was not converted into (127) in T. angustifolia nor was (128) converted into (126) in P. morifoliae2 On the other hand (128) is metabolized in cassava which naturally produces linama~in.~~ 6.2 Anatoxin a(s) Capreomycin Acivicin Bellenamine and 3-Nitropropanoate Anatoxin a(s) (130) which is one of the most potent anticholinesterases known is one of the neurotoxic alkaloids produced by Anabaena Jos-aquae.The 0-and N-methyl groups originate in the methyl group of methionine (labelled glycine and serine were also incorporated appropriate to metabolism via tetrahydr~folate).~~ Following satisfactory preliminary results with arginine and ornithine bearing radioactive labels ~-[U-'~C]arginine was examined it provided C-2-C-6 as an intact unit. ~-Erythro-4-hydroxyarginine (1 29) was found to be a minor constituent of A. Jos-aquae cultures; it may well be an intermediate in anatoxin a(s) biosynthesis.[I3C],Acetate was not a specific precur~or.~~ Because of the possible relevance of the moiety (131) in capreomycin to the biosynthesis of streptothricin F from arginine (ref. 5 p. 590) the way in which capreomycin IA (133) and IB (134) originate from this a-amino acid has been examined in Streptomyces capre01us.~~ Initial results with "C- NATURAL PRODUCT REP0 labelled material confirmed the origin of (13 1) as arginine as for streptothricin F. The pentadeul derivative (132) gave (133) and (134) with retentic the H-3 and both of the H-5 labels. This is cor biosynthesis via an a$-dehydroarginyl intermedia. a peptide). Notably the results for streptothricin I exclude such an intermediate (ref. 5 p.590). It i (1 33) is a precursor for (1 34). Results on the biosynthesis of acivicin (137 published in preliminary forme6 (ref. 1 p. 53 available in a full paper.87 In summary the skelet in ornithine via N-hydroxyornithine (1 36) (the o> from molecular oxygen). Ring closure involves dis the 3S-hydrogen in ornithine; the C-2 proton is lo during biosynthesis. The erythro-and thre ornithines did not serve as precursors. Acivicin (1 hydroxyacivicin (1 38). Bellenamine (141) is formed from L-lysine (139' P-lysine (1 40) in Streptomyces nashvillensis.sE reaction i.e. (139) to (140) has a stereocherr opposite to that which affords L-P-lysine in str biosynthesi~.~~ Interestingly both C-1' and the a. function in (14 1) derive from glycine ;the amide n origin in general nitrogen metabolism.88 The biosynthesis of 3-nitropropanoic acid in be,..atrovenetum which has been the subject of preliminary accounts (ref. 3 p. 206 and earlier reports) is now the subject of a full paper.go Mature cells of P. atrovenetum produce a dehydro- genase which catalyses the interconversion of 3-nitro-propanoate and 3-nitroacrylate an apparently futile cycle. 6.3 Valanimycin Following on from research into the biosynthesis of elaiomycin (142)' the in vivo formation of valanimycin (143) in Streptomyces viridifaciens has been explored.g1 Previous results indicated that alanine and valine could act as precursors for (143) (ref. 1 p. 536). The results with alanine have been confirmed but serine has now been shown to be a much more immediate precur~or.~~ Serine is also a specific precursor for elaiomycin (142) and a further parallel with elaiomycin biosynthesis where C-5-C- 12 plus the P-nitrogen derive from n-octylamine was found in the intact and efficient incorporation of [1-13C,15N]iso- NATURAL PRODUCT REPORTS 1995-R.B. HERBERT / O ooo m 4 OH 0 OH EnzS-<\ EnzS-<\ 0 0 Scheme 10 Scheme 11 butylamine into valinimycin. Even more efficient incorporation of [ 1-13C,15N]isobutylhydroxylamine locates it as a probable intermediate necessary for N-N bond formation. 6.4 Dynemicin A and Domoic Acid Dynemicin A (144) is a potent antibacterial and antitumor antibiotic isolated from Micromonospora chersina. It combines within its structure characteristics of the enediyne and anthracycline classes of antibiotics.Its biosynthetic origins have been probed and the results show that the bicyclic enediyne and anthraquinone moieties are separately formed each from a heptaketide (Scheme The deduced pathway could be extended to a hypothesis for the biosynthesis of the esperamicin/calicheamicin class of antibiotics. Domoic acid (145) is a neuroexcitatory amino acid produced by the marine diatom Nitzschia pungens forma multiseries. Its biosynthesis has been neatly and thoughtfully probed with I3C- labelled acetate.93 The results reveal that (145) is formed from two separate units one of which is an isoprenoid the other an activated glutamate derivative (Scheme 11). This is a notable and novel biosynthesis.rn = ~-[l-'~C]Threonine + = ~~-p-%]~ysteine A = ~~-[~-'~~]~erine = [2-13CJGlycine A = ~~-[3-'~~]~1anine Figure 1 6.5 Antibiotics A10255 The A 10255 antibiotic complex is produced by Streptomyces gardneri. The biosynthesis of the antibiotics has been investigated. Fifteen of the seventeen amino acid residues present in factor G (146) (Figure 1) were labelled by [l- 13C]serine (the same residues were labelled by [2-13C]glycine presumably by conversion first into serine; there is one masked glycine residue though which is also labelled by ~erine).~~ Cysteine appears to be the immediate precursor for the cysteine residues. Threonine is the source of two residues and alanine one. Complementary results were obtained for factors B and E with the inclusion now of methyl groups from methionine.Very similar results were obtained for the biosynthesis of the structurally related peptide antibiotics thiostrepton (ref. 4 p. 518) and nosiheptide (ref. 3 p. 204). There is much fascinating mechanism to be unravelled for the biosynthesis of these antibiotics. 6.6 Nikkomycins and Griseolic Acids The nikkomycins which are metabolites of Streptomyces tendae are potent inhibitors of chitin synthases from fungi and insects. Nikkomycin X (147) contains the curious heterocyclic residue 4-formylimidazolin-2-one which appears to originate in histidine (148).95 It is not easy to formulate a mechanism based on histidine metabolism which allows for the conversion of the amino acid into this heterocyclic appendage in (147) but evidence has been obtained using a mutant of S.tendae that histidine aminofransferase activity may be associated with nikkomycin X production i.e. the keto-acid (149) may be a biosynthetic intermediate.96 The biosynthesis of polyoxins e.g. polyoxin C [(150) R = CH20H] is clearly similar to that of nikkomycins e.g. (147) and for the polyoxins the amino- hexuronic acid part originates in ribose and PEP (Scheme 12; * label from [l-13C]glucose) with (151) being a shunt from this path~ay.~' Four new metabolites nikkomycins S (1 52) S (1 53) So (1 54) and So (159,which are similar to (1 5 l) have been isolated from S. tendae cultures.98 Their formation is dependent on iron concentration and the amount of S plus S formed bears an inverse relationship to the amount of X plus Z produced.These are apparently metabolites at a shunt from the biosynthesis of nikkomycins X (147) and Z and it is most evident that the aminohexuronic acid part of nikkomycins is biosynthesized in a similar way to that of the polyoxins. OHC OH NH2 HO OH (147) -NATURAL PRODUCT REPORTS 1995 The griseolic acids e.g. griseolic acids A (158) and B (159) which are produced by Streptomyces griseoaurantiacus are cyclic-nucleotide-phosphodiesterase inhibitors. The bio-synthesis of griseolic acid A (1 58) has been studied with a range of 13C-labelled compound^.^^ The pattern of incorporation of [1 -13C,15N]glycine [1 -13C]glucose [6-13C]glucose and [2-13C]ribose indicates that adenosine (1 57) is a part biosynthetic source for (158); the ability of adenosine but not adenine to support recovery of griseolic acid production in a mutant indicates that adenosine is an intact precursor.The pattern of incorporation of [2-13C]acetate [13C,]acetate [2-13C]pyruvate and [1,4-13C2]succinate indicates that the remainder of the molecule originates in the tricarboxylic acid cycle possibly by way of oxaloacetate (156). 6.7 Aminoglycoside-Aminocyclitol Antibiotics The aminocyclitol moieties (1 60) and (1 6 l) present in respectively streptomycin and e.g. the neomycins both derive from D-glucose but by different pathways. The accumulated evidence for the biosynthesis of 2-deoxystreptamine (1 6 1) (ref.3 p. 202; ref. 1 p. 538) has now been capped with further evidence on the biosynthesis of neomycins in Streptomyces fradiae.loO(Some of the work reported has been the subject of a preliminary publication,lo1 see ref. 1 p. 538). Both protons at C-6 of D-glucose (164) are retained in the formation of (161) (and with retention of stereochemical integrity ref. 3 p. 202) as is the C-3 proton but those at C-5 and C-4 are lost. In feeding experiments with labelled samples 6-deoxy-5-keto- glucose (1 63) and 6-deoxyglucose (1 62) both gave labelled (161). The results indicate that (162) is incorporated via (163) which is used for cyclization via the enol (165); it certainly appears that removal of the C-6 substituent of glucose precedes cyclization.The results taken together with electronic con- siderations (cyclization would be aided by a hydroxy group at C-4 rather than a carbonyl) lead to the pathway illustrated in Scheme 13 with the immediate product of cyclization being 2- deoxy-scyllo-inosose (1 66). The enzyme responsible is now called 2-deoxyinosose synthase ; the proposed mechanism (Scheme 13) is reminiscent of that for dehydroquinate synthase (see refs. cited in ref. 102). The deoxyinosose synthase has been identified in S. fradiae cultures that produce neomycins and has been partially purified.lo2 The substrate for formation of (166) is ~-glucose-6- 0 .R C02H *OdH2 PEP 7 OH 6H Uridine -'@ H OH OH -(150) Scheme 12 NATURAL PRODUCT REPORTS 1995-R.B. HERBERT 0 oHc,$ Y (152) R=H (154) R= H (153) R=OH (155) R=OH OH OH (157) J y42 R ,I (158) R =OH (159) R=H (160) X= OH (161) X= H Oxidation os HHO O SOH-HO OH OH OH (164) f3-Gl~o~; R = H -ROH 1R-Pi Reduction HO OH OH 0 Scheme 13 Q H2N OXNH2 (167) R = H (168) R = CH=NH phosphate (not glucose) and apparently the preferred co-enzyme is NAD' rather than NADP'. The results of a study of the streptomycin biosynthetic genes in Streptomyces griseus and S. glaucescens have revealed new insights into their gene orders structures evolution and functions.103 The enzyme which converts fortimicin A (161) into its N-formidoyl derivative (168) has been isolated from Micro-monospora olivasterospora purified and characterized.lo4 The carbon atom of the formimido group was shown to derive from C-2 of glycine.8 References 1 R. B. Herbert Nat. Prod. Rep. 1988 5 523. 2 R. B. Herbert Nat. Prod. Rep. 1990 7 105. 3 R. B. Herbert Nat. Prod. Rep. 1991 8 185. 4 R. B. Herbert Nat. Prod. Rep. 1992 9 507. 5 R. B. Herbert Nat. Prod. Rep. 1993 10 575. 6 R. B. Herbert In 'Rodd's Chemistry of Carbon Compounds' ed. S. Coffey Elsevier Amsterdam 2nd edn. 1980 vol. IV part L p. 291. 7 R. B. Herbert in 'Rodd's Chemistry of Carbon Compounds' ed. M. F. Ansell Elsevier Amsterdam 2nd edn. 1988 vol. IV part L Supplement p. 155. 8 R. B. Herbert 'The Biosynthesis of Secondary Metabolites' Chapman and Hall London 2nd edn. 1989.9 U. Sankawa H. Noguchi H. Hashimoto and Y. Yamada Chem. Pharm. Bull. 1990 38 2066. 10 T. Hemscheidt and I. D. Spenser J. Am. Chem. SOC.,1992 114 5472. 11 T. Hashimoto K. Nakajima G. Ongena and Y. Yamada Plant Physiol. 1992 100 836. 12 B. Drager A. Porsteffen A. Schaal P. H. McCabe A. C. J. Peerless and R. J. Robins Planta 1992 188 581. 13 P. Christen T. Aoki and K. Shimomura Plant Cell Rep. 1992 11 597. 14 W. Gritsanapam and W. J. Griffin Phytochemistry 1992 31 3069; see also G. D. Lin and W. J. Griffin ibid. 41 51. 15 M. Mengoli D. Chriqui and N. Bagni J. Plant Physiol. 1992 140 153. 16 S. C. Haslam and T. W. Young Phytochemistry 1992 31 4075. 17 J. B. Friesen P. C. Burkhouse D. D. Biesboer and E. Leete Phytochemistry 1992 31 3059.18 J. A. Bjorklund and E. Leete Phytochemistry 1992 31 3883. 19 M. H. Zenk Rec. Adv. Phytochem. 1979 12 139. 20 D. J. Hardick D. W. Hutchinson S. J. Trew and E. M. H. Wellington Tetrahedron 1992 48 6285. 21 D. J. Hardick D. W. Hutchinson S. J. Trew and E. M. H. Wellington J. Chem. SOC. Chem. Commun. 1991 729. 22 D. J. Robins Experientia 1991 47 1118. 23 C. M. Passreiter Phytochemistry 1992 31 4135. 24 W. De-Eknamkul and M. H. Zenk Phytochemistry 1992,31,813. 25 H. Lotter J. Gollwitzer and M. H. Zenk Tetrahedron Lett. 1992 33 2443. 26 D. H. R. Barton D. S. Bhakuni R. James and G. W. Kirby J. Chem. SOC. C 1967 128. 27 G. Pasquali 0.J. M. Goddijn A. deWaal R. Verpoorte R. A. Schilperoort J. H. C. Hoge and J.Memelink Plant Molecular Biology 1992 18 1121. 28 D. Bracher and T. M. Kutchan Arch. Biochem. Biophys 1992 294 717. 29 I. A. Roewer N. Cloutier C. L. Nessler and V. De Luca Plant Cell Rep. 1992 11 86. 30 D. Bracher and T. M. Kutchan Plant Cell Rep. 1992 11 179. 31 C. Poulsen and R. Verpoorte Plant Physiol. Biochem. 1992 30 105. 32 L. Toivonen S. Laakso H. Rosenqvist Plant Cell Rep. 1992 11 390; 395; A. Decendit D. Liu L. Ouelhazi P. Doireau J. M. Merillon and M. Rideau ibid. 400; J. M. Merillon D. Liu Y. Laurent M. Rideau and C. Viel Phytochemistry 1992,31 1609. 33 R. J. Aerts and V. DeLuca Plant Physiol. 1992 100 1029. 34 T. Furuya K. Sakamoto K. Iida Y. Asada T. Yoshikawa S. Sakai and N. Aimi Phytochemistry 1992 31 3065 35 H.Takayama M. Kitajima S. Suda N. Aimi S. Sakai S. Endress and J. Stockigt Tetrahedron 1992 48 2627; L. Polz J. Stockigt H. Takayama N. Uchida N. Aimi and S. Sakai Tetrahedron Lett. 1990 31 6693. 36 Z.-Q. Xia and M. H. Zenk Tetrahedron 1992 31 2695. 37 J. C. Gebler and C. D. Poulter Arch. Biochem. Biophys. 1992 296 308. 38 J. C. Gebler A. B. Woodside and C. D. Poulter J. Am. Chem. SOC.,1992 114 7354. 39 N. C. Perellino J. Malyszko M. Ballabio B. Gioia and A. Minghetti J. Nut. Prod. 1992 55 424. 40 R. J. Parry and E. E. Eudy J. Am. Chem. SOC. 1988 110 2316. 41 P. J. Parry Y. Li and E. Eudy Gomez J. Am. Chem. SOC. 1992 114 5946. 42 H. Oikawa Y. Marakami and A. Ichihara J. Chem. SOC. Perkin Trans. I 1992 2949. 43 H. Oikawa Y.Murakami and A. Ichihara J. Chem. SOC. Perkin Trans. I 1992 2955. 44 K. A. Reynolds P. Wang K. M. Fox and H. G. Floss J. Antibiot. 1992 45 41 1. 45 H. G. Floss and J. M. Beale Angew. Chem. Int. Ed. Engl. 1989 28 146. 46 R. W. Rickards Stud. Nut. Prod. Chem. 1991 9 431. 47 R.-M. Meier and Ch. Tamm J. Antibiot. 1992 45 400. 48 C.-G. Kim A. Kirshning P. Bergon Y. Ahn J. J. Wang M. Shibuya and H. G. Floss J. Am. Chem. SOC. 1992 114 4941. 49 Y. Tabata M. Imoto and K. Umezawa J. Antibiot. 1992 45 1382. 50 K. M. Cable R. B. Herbert and J. Mann J. Chem. Soc. Perkin Trans. I 1987 1593. 51 A. Nahrstedt and P. Schwind Phytochemistry 1992 31 1997. 52 M. Bokern S. Heuer and D. Strack Botanica Acta 1992 105 146. 53 L. 0.Zamir M. E.Nedea and F. X. Garneau Tetrahedron Lett. 1992 33 5235. 54 E. Leete and G. B. Bodem Tetrahedron Lett. 1966 7 3925. 55 N. Goosen R. G. M. Huinen and P. van de Putte J. Bacteriol. 1992 174 1426. 56 J. Penn M. Purcell and P. G. Mantle FEMS Microbiol. Lett. 1992 92 229. 57 R. B. Herbert and A. R. Knaggs J. Chem. SOC.,Perkin Trans. I 1992 103. 58 R. B. Herbert and A. R. Knaggs J. Chem. SOC. Perkin Trans. I 1992 109. 59 H. Budzikiewicz H. Schroder and K. Taraz Z. Naturforsch. C Biosci. 1992 47 26. 60 M. S. Kuo D. A. Yurek J. H. Coats S. T. Chung and G. P. Li J. Antibiot. 1992 45 1773. 61 J. D. Bu’Lock and L. E. Clough Aust. J. Chem. 1992 45 39. 62 D. Iwata-Reuyl and C. A. Townsend J. Am. Chem. Soc. 1992 114 2762. 63 J. E.Baldwin R. M. Adlington J. S. Bryans M. D. Lloyd T. J. Sewell C. J. Schofield K. H. Baggaley and R. Cassels J. Chem. SOC. Chem. Commun. 1992 877. 64 J. E. Baldwin R. M. Adlington J. W. Bird R. A. Field N. M. O’Callaghan and C. J. Schofield Tetrahedron 1992 48 1099. 65 J. E. Baldwin G. P. Lynch and C. J. Schofield Tetrahedron 1992 48 9085. 66 G. W. Huffman P. D. Gesellchen J. R. Turner R. B. NATURAL PRODUCTS REPORTS 1995 Rothenberger H. E. Osborne F. D. Miller J. L. Chapman and S. W. Queener J. Med. Chem. 1992 35 1897. 67 G. Turner in ‘Secondary Metabolites their Function and Evolution,’ ed. D. J. Chadwick and J. Whelan Ciba Foundation Symposium No. 171 J. Wiley and Sons Chichester 1992 p. 113. 68 M. Durairaj J. L. Doran and S. E. Jenson Appl.Environ. Microbiol. 1992 58 4038. 69 0. Landman D. Shiffman Y. Av-Gay Y. Aharonowitz and G. Cohen FEMS Microbiol. Lett. 1991 84 239. 70 Z. A. Bainbridge R. I. Scott and D. Perry J. Chem. Tech. Biotechnol. 1992 55 233. 71 A. A. Brakhage P. Browne and G. Turner J. Bacteriol. 1992 174 3789. 72 A. A. Brakhage and G. Turner FEMS Microbiol. Lett. 1992,98 123. 73 Y. Lu R. L. Mach K. Affenzeller and C. P. Kubicek Can. J. Microbiol. 1992 38 758. 74 R. A. Scott S. Wang M. L. Eidsness A. Kriauciunas C. A. Frolik and V. J. Chen Biochemistry 1992 31 4596. 75 A. M. Orville V. J. Chen A. Kriauciunas M. R. Harpel B. G. Fox E. Munck and J. D. Lipscomb Biochemistry 1992,31,4602. 76 J. E. Baldwin R. M. Adlington R. T. Aplin N. P. Crouch and R.Wilkinson Tetrahedron 1992 48 6853. 77 J. E. Baldwin K.-C. Goh and C. J. Schofield J. Antibiot. 1992 45 1378. 78 J. E. Baldwin R. M. Adlington N. P. Crouch and I. A. C. Pereira J. Chem. SOC. Chem. Commun. 1992 1448. 79 M. S. Edenborough and R. B. Herbert Nut. Prod. Rep. 1988 5 229. 80 P. J. Scheuer Acc. Chem. Res. 1992 25 433. 81 I. Tober and E. E. Conn Phytochemistry 1985 24 1215. 82 E. S. Olafsdottir L. B. Jsrgensen and J. W. Jaroszewski Phytochemistry 1992 31 4129. 83 B. Koch V. S. Nielsen B. A. Halkier C. E. Olsen and B. L. Msller Arch. Biochem. Biophys. 1992 292 141. 84 B. S. Moore I. Ohtani C. B. de Koning R. E. Moore and W. W. Carmichael Tetrahedron Lett. 1992 33 6595. 85 S. J. Gould and D. A. Minott J. Org. Chem. 1992 57 5214.86 S. Ju V. A. Palaniswamy P. Yorgey and S. J. Gould J. Am. Chem. SOC. 1986,108,6429; S. J. Gould and S. Ju J. Am. Chem. SOC.,1989 111,2329. 87 S. J. Gould and S. Ju J. Am. Chem. SOC. 1992 114 10166. 88 Y. Ikeda H. Naganawa S. Kondo and T. Takeuchi J. Antibiot. 1992 45 1919. 89 T. K. Thiruvengadam S. J. Gould D. J. Aberhart and H.-J Lin J. Am. Chem. SOC. 1983 105 5470. 90 R. L. Baxter A. B. Hanley H. W.3. Chan S. L. Greenwood E. M. Abbot I. J. McFarlane andK. Milne J. Chem. SOC. Perkin Trans. I 1992 2495. 91 R. J. Parry Y. Li and F.-L. Lii J. Am. Chem. SOC. 1992 114 10062. 92 Y. Tokiwa M. Miyoshi-Saitoh H. Kobayashi R. Sunaga M. Konishi T. Oki and S. Iwasaki J. Am. Chem. SOC. 1992 114 4107. 93 D. J. Douglas U. P.Ramsey J. A. Walter and J. L. C. Wright J. Chem. SOC. Chem. Commun. 1992 714. 94 M. E. Favret J. W. Paschal T. K. Elzey and L. D. Boeck J. Antibiot. 1992 45 1499. 95 R.-M. Schmidt H. Pape and M. Junack Z. Naturforsch. C Biosci. 1986 41 135. 96 U. Roos S. Mattern H. Schrempf and C. Bormann FEMS Microbiol. Lett. 1992 97 185. 97 K. Isono T. Sato K. Hirasawa S. Funayama and S. Suzuki J. Am. Chem. SOC. 1978 100 3937. 98 T. C. Schiiz H.-P. Fiedler H. Zahner M. Rieck and W. A. Konig J. Antibiot. 1992 45 199. 99 S. Miyakoshi H. Huruyama T. Shioiri S. Takahashi A. Torikata and M. Yamazaki J. Antibiot. 1992 45 394. 100 S. K. Goda and M. Akhtar J. Antibiot. 1992 45 984. 101 S. K. Goda and M. Akhtar J. Chem. SOC. Chem. Commun. 1987 12. 102 N. Yamauchi and K. Kakinuma J Antibiot. 1992 45 774. 103 J. Distler K. Mansouri G. Mayer M. Stockmann and W. Piepersberg Gene 1992 115 105. 104 T. Dairi K. Yamaguchi and M. Hasegawa Mol. Gen. Genet. 1992 236 49.

ISSN:0265-0568    

DOI:10.1039/NP9951200055

出版商:RSC    

年代:1995

数据来源: RSC