JOURNAL OF CHROMATOGRAPHY LIBRARY- volume 54
chromatography of mycotoxins techniques and applications
This Page Intentionally Left Blank
JOURNAL OF CHROMATOGRAPHY LIBRARY- volume 54
chromatography of mycotoxins techniques and applications edited by
Vladimir Betina Department of Microbiology, Biochemistry and Biology, Slovak Technical University, Bratislava, Slovakia
ELSEVIER Amsterdam -London-New York -Tokyo
1993
ELSEVIER SCIENCE PUBLISHERS B.V. Sara Burgerhartstraat 25 P.O. Box 21 1,1000AE Amsterdam,The Netherlands
ISBN 0-444-81521-X
0 1993 Elsevier Science Publishers B.V. 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, photocopying, recording or otherwise, without the prior written permission of the publisher, Elsevier Science Publishers B.V., Copyright & Permissions Department, P.O. Box 521,1000AM Amsterdam,The Netherlands. Special regulations for readers in the U.S.A. - This publication has been registered with the Copyright Clearance Center Inc. (CCC), Salem, Massachusetts. Information can be obtained from the CCC about conditions under which photocopies of parts of this publication may be made in the U.S.A. All other copyright questions, including photocopying outside of the U.S.A., should be referred to the copyright owner, Elsevier Science Publishers B.V., unless otherwise specified. No responsibility is assumed by the publisher for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use or operation of any methods, products, instructions or ideas contained in the material herein. This book is printed on acid-free paper. Printed in the Netherlands
V
CONTENTS
..................................... ..................................................
List of contributors
xii
Preface
xiii
Part A . Techniques Chapter 1 . Sampling. sample preparation. extraction and clean-up V Betina 1.1 Introduction 1 . 2 Sampling and sample preparation 1 . 3 Sample extraction and clean-up 1 . 4 Illustrative example 1.5 Conclusions References
.
........................................ ............................................ ......................... .......................... .................................... ............................................. ..................................................
Chapter 2 . Techniques of thin layer chromatography R.D. Coker. A.E. John and J.A. Gibbs 2 . 1 Introduction 2.2 Clean-up methods 2 . 3 Normal phase TLC 2 . 3 . 1 Principles 2 . 3 . 2 Practical considerations
3
3 3
4
7 9 9
............. 1 2 ............................................ 12 ........................................ 12 ........................................ 16 ........................................ 16 .......................... 17 2 . 4 Reverse-phase TLC (RPTLC) ............................... 20 2 . 4 . 1 Principles ........................................ 20 20 2 . 4 . 2 Practical considerations .......................... 2 . 5 High performance thin layer chromatography (HPTLC) ...... 2 1 21 2 . 5 . 1 Principles ........................................ 22 2 . 5 . 2 Practical considerations .......................... 2 . 6 Preparative TLC ......................................... 23 2 . 6 . 1 Principles ........................................ 23 Practical considerations .......................... 25 2.6.2 27 2 . 7 Detection ............................................... 27 2 . 7 . 1 Fluorescence detection ............................ 21 2 . 7 . 2 Chemical derivatisation ........................... 2 . 7 . 3 Bioautographic methods ............................ 28 2 . 8 Quantitative and semi-quantitative evaluation ........... 2 8 2 . 9 Illustrative examples ................................... 30 2 . 1 0 Conclusions ............................................ 31 References .................................................. 32
vi
. Techniques of liquid column chromatography P. Kuronen ........................................ Introduction ............................................. Sample pretreatment ...................................... Column chromatography .................................... 3 . 3 . 1 Introduction .......................................
Chapter 3 3.1 3.2 3.3
..........................................
3.3.2
Procedure 3 . 4 Mini-column chromatography 3.4.1 Procedure 3.4.2 Illustrative example 3.5 High-performance liquid chromatography 3 . 5 . 1 Introduction Instrumentation and practice 3.5.2 3.6 Conclusion References
...............................
..........................................
36 36 37 40 40 40 44 45
............................... 46 ................... 46
....................................... ....................... ...............................................
...................................................
Chapter 4 . Techniques of gas chromatography R.W. Beaver 4 . 1 Introduction 4.2 Resolution in gas chromatography 4 . 2 . 1 Definition of resolution 4.2.2 Efficiency 4.2.3 Retention 4.2.4 Selectivity 4.3 Extracolumn resolution 4 . 3 . 1 Resolution through sample clean-up 4.3.2 Chemical derivatization 4.3.3 Resolution through detection 4.4 Conclusions References
....................................... ............................................. ......................... ........................... .........................................
46 48 71 72
78 78 79 79
79
.......................................... 83 ........................................ 84 ................................... 8 6 ................. 8 7 ............................
....................... .............................................. ...................................................
Chapter 5
. Emerging
91 91 96 96
techniques: immunoaffinity chromatography A . A . G . Candlish and W.H. Stimson 99 99 5 . 1 Introduction 101 5 . 2 Immunoaffinity chromatography theory 5 . 3 Practical aspects and instrumentation 103 111 5 . 4 Sample preparation 5.5 Illustrative examples 116 References 122
.................. ............................................. .................... ................... ...................................... ...................................
..................................................
vii
Chapter 6 . Emerging techniques: enzyme-linked immunosorbent assay (ELISA) as alternatives to chromatographic methods C.M. Ward. A.P. Wilkinson and M.R.A. Morgan ..... 6.1 Introduction 6.2 Principles of ELISA 6.2.1 The immune response and polyclonal antibodies 6.2.2 Monoclonal antibodies 6.2.3 Haptens 6.2.4 Specificity of anti-hapten antibodies 6.2.5 Principles of immunoassay 6.2.6 Enzyme immunoassays 6.2.7 Enzyme-linked immunosorbent assays (ELISAS)
........................................... ....................................
............................
....
.......................................... ............ ........................ .............................. ....... preparation .....................................
6.3 Sample 6.3.1 Extraction 6.4 Instrumentation and practice 6.4.1 Instrumentation 6.4.2 Practice 6.5 Illustrative examples 6.5.1 Aflatoxins 6.5.2 Other mycotoxins 6.6 Conclusions References
....................................... ........................... .................................. .........................................
.................................. ....................................... ................................. ............................................ .................................................
124 124 124 124 125 125 125 126 126 127 127 127 128 128 129 132 132 133 134 135
Part B . Applications Chapter 7 . Thin-layer chromatography of mycotoxins V Betina
.
7.1 7.2
7.3
....................................... Introduction ........................................... Aflatoxins ............................................. 7.2.1 Sampling and sample preparation .................. 7.2.2 Extraction and clean-up .......................... 7.2.3 Adsorbents and solvent systems ................... 7.2.4 Detection ........................................ 7.2.5 Selected applications ............................ Sterigmatocystin and related compounds ................. 7.3.1 Extraction and clean-up .......................... 7.3.2 Adsorbents and solvent systems ................... 7.3.3 Detection ........................................
............................ .........................................
1.3.4 Selected applications 7.4 Trichothecenes 7.4.1 Extraction and clean-up
..........................
141 141 143 143 144 149 152 153 162 162 164 166 166 168 169
viii
................... ........................................ ............................ ......................................... .......................................... .................................. ................................ ....................................... .................................... ................................... ...................................... .................................... ............................................ .......................... ................... ........................................ ............................ ............................................ .......................... ................... ........................................
7.4.2 Adsorbents and solvent systems 7.4.3 Detection 7.4.4 Selected applications 7.5 Small lactones 7.5.1 Patulin 7.5.2 Penicillic acid 7.5.3 Mycophenolic acid 7.5.4 Butenolide 7.5.5 Citreoviridin 7.6 Macrocyclic lactones 7.6.1 Zearalenone 7.6.2 Cytochalasans 7.7 Ochratoxins 7.7.1 Extraction and clean-up 7.7.2 Adsorbents and solvent systems 7.7.3 Detection 7.7.4 Selected applications 7.8 Rubratoxins 7.8.1 Extraction and clean-up 7.8.2 Adsorbents and solvent systems 7.8.3 Detection 7.8.4 Selected applications 7.9 Hydroxyanthraquinones .................................. 7.9.1 Extraction 7.9.2 Adsorbents and solvent systems 7.9.3 Detection 7.9.4 Selected applications 7.10 Epipolythiopiperazine-3. 6.diones 7.10.1 Extraction and clean-up 7.10.2 Adsorbents. solvent systems and detection 7.10.3 Selected applications 7.11 Tremorgenic mycotoxins 7.11.1 Adsorbents and solvent systems 7.11.2 Detection 7.11.3 Selected applications 7.12 Alternaria toxins ..................................... 7.12.1 Extraction and clean-up 7.12.2 Adsorbents and solvent systems 7.12.3 Detection 7.12.4 Selected applications
........................................ ............................ ...................... ......................... ....... ........................... ................................ .................. .......................................
169 170 173 178 178 181 183 184 184 186 186 191 196 196 196 197 198 199 199 199 199 200 200 201 201 201 201 203 203 204 204 205 205 205
......................... .................. ....................................... ...........................
207 209 209 209 210 210
............................
....................................... ...................
...........................
ix
............................................. ........................ ................. ...................................... .......................... ................................ ........................ ................. ...................................... ..........................
7.13
Citrinin 7 . 1 3 . 1 Extraction and clean-up 7 . 1 3 . 2 Adsorbents and solvent systems 7 . 1 3 . 3 Detection 7 . 1 3 . 4 Selected applications 7.14 a-Cyclopiazonic acid 7 . 1 4 . 1 Extraction and clean-up 7 . 1 4 . 2 Adsorbents and solvent systems 7 . 1 4 . 3 Detection 7 . 1 4 . 4 Selected applications 7 . 1 5 PR toxin and roquefortine 7 . 1 5 . 1 Extraction and clean-up 7 . 1 5 . 2 Adsorbents and solvent systems 7 . 1 5 . 3 Detection 7 . 1 6 Xanthomegnin. viomellein and vioxanthin 7 . 1 6 . 1 Extraction and clean-up 7 . 1 6 . 2 Adsorbents and solvent systems 7.16.3 Detection 7 . 1 6 . 4 Selected applications 7 . 1 7 Naphtho- y -pyrones 7 . 1 8 Secalonic acids 7 . 1 9 TLC of miscellaneous toxins 7 . 2 0 Multi-mycotoxin TLC 7 . 2 1 TLC in chemotaxonomic studies of toxigenic fungi 7 . 2 2 Conclusions References
............................
........................
210 211 212 212 213 214 214 215 215 216 217 217
................. 2 1 7 ...................................... 2 1 7 .............. 2 1 8 ........................ 218 ................. 2 1 8 ...................................... 218 .......................... 218 ...................................
219
...................................... 219 .......................... 220 .................................. 222 ..... 2 3 0 .......................................... 231 ................................................ 233
Chapter 8 . Liquid column chromatography of mycotoxins J.C. Frisvad and U Thrane 8 . 1 Introduction 8 . 2 Column chromatography 8.3 Mini-column chromatography 8 . 4 High performance liquid chromatography 8.4.1 8.4.2 8.4.3 8.4.4 8.4.5 8.4.6 8.4.7 8.4.8
. ..................... .......................................... ................................. ............................ ................ Aflatoxins ...................................... Sterigmatocystin and related compounds .......... Trichothecenes .................................. Small lactones .................................. Macrocyclic lactones ............................ Ochratoxins and related compounds ............... ..................................... .............
Rubratoxins Hydroxyanthraquinones and xanthones
253 253 287 289 290 290 295 296 299 302 303 306 307
X
................
8.4.9 Epipolythiopiperazine.3. 6.diones 8.4.10 Tremorgenic mycotoxins 8.4.11 Alternaria toxins 8.4.12 Toxic peptides 8.4.13 Fusarium toxins other than trichothecenes and zearalenones 8.4.14 Miscellaneous toxins 8.4.15 Multi-mycotoxin analyses by HPLC 8.5 Informative on-line detection methods 8.5.1 Applications of HPLC diode array detection 8.5.2 Applications of HPLC mass spectrometry 8.6 Conclusions References
......................... .............................. .................................
............................... ........................... ............... .................
Chapter 9
.
......
.......... ........................................... ................................................
Gas chromatography of mycotoxins P.M. Scott 9.1 Introduction 9.2 Trichothecenes 9.2.1 Introduction 9.2.2 Derivatization and detection procedures for trichothecenes 9.2.3 Methods for grains. grain foods and feeds 9.2.4 Methods for biological fluids 9.2.5 Methods for animal tissues 9.2.6 Methods for other foods 9.2.1 Additional applications 9.3 Zearalenone 9.3.1 Derivatization and detection procedures for
..................................... .......................................... ........................................ ....................................
308 309 312 314 315 311 319 321 321 354 355 356 313 313 373 313
.............................. 314 ....... 382 ................... 381 ...................... 389
......................... .........................
...........................................
............. ....... .............................. Moniliformin .......................................... Alternaria toxins ..................................... zearalenone and related metabolites 9.3.2 Methods for grains. grain foods and feeds 9.3.3 Other applications
9.4 9.5
9.5.1 Alternariol. alternariol monomethyl ether. altenuene and isoaltenuene 9.5.2 Tenuazonic acids 9.6 Slaframine and swainsonine 9.1 Patulin 9.7.1 Comparison of derivatization and detection procedures for patulin
......................
389 391 392 392 393 395 395 396
............................
...............................................
396 391 398 399
..........................
399
................................
xi
9.7.2 Methods for apple juice and other fruit products 9.7.3 Methods for other foodstuffs 9.8 Penicillic acid 9.8.1 Derivatization and detection procedures for penicillic acid 9.8.2 Methods for agricultural commodities 9.9 Sterigmatocystin 9.9.1 Comparison of detection procedures 9.9.2 Methods for grains 9.9.3 Dihydrosterigmatocystin 9.10 Aflatoxins 9.11 Ergot alkaloids 9.12 Miscellaneous mycotoxins 9.12.1 Sporidesmins 9.12.2 Butenolide 9.12.3 F-Nitropropionic acid 9.12.4 Fumonisins 9.12.5 Fusarin C 9.12.6 Griseofulvin and related compounds 9.12.7 Mycophenolic acid 9.12.8 Kojic acid, terreic acid and terrein 9.12.9 Oxalic acid 9.12.10 "Peptaibol" polypeptide antibiotics 9.12.11 Ochratoxin A 9.12.12 d-Cyclopiazonic acid 9.12.13 Loline alkaloids 9.12.14 Fusarochromanone 9.13 Conclusions References
........................................ .................... .......................................
............................. ............ ...................................... .............. .............................. ......................... ........................................... ...................................... ............................. ................................... ..................................... ......................... ..................................... ...................................... .............
.............................. ........... .................................... ........... .................................. .........................
Subject
.............................. .............................. .......................................... ................................................ index .............................................
400 401 401 401 403 404 404 404 405 405 406 408 408 409 409 409 410 411 411 411 412 412 412 412 413 413 414 414 427
xii
LIST OF CONTRIBUTORS R.W.
BEAVER
V. BETINA
A.A.G.
CANDLISH
R.D. COKER J.C. FRISVAD J.A. GIBBS A.E. JOHN P. KURONEN M.R.A. MORGAN P.M. SCOTT
W.H. STIMSON U. THRANE C.M. WARD A.P. WILKINSON
Coastal Plain Station, Department of Plant Pathology, College of Agriculture, The University of Georgia, Tifton, Georgia 31793, USA Department of Microbiology, Biochemistry and Biology, Faculty of Chemical Technology, Slovak Technical University, 812 37 Bratislava, Slovakia RhBne Poulenc Diagnostics Ltd., Montrose House, 187 George Street, Glasgow G1 lYT, Scotland Natural Resources Institute, Central Avenue, Chatham Maritime, Chatham, Kent ME4 4TB, United Kingdom Department of Biotechnology, The Technical University of Denmark, Bygning 221, DK-2800 Lyngby, Denmark Natural Resources Institute, Central Avenue, Chatham Maritime, Chatham, Kent ME4 4TB, United Kingdom Natural Resources Institute, Central Avenue, Chatham Maritime, Chatham, Kent ME4 4TB? United Kingdom Department of Chemistry, University of Helsinki, Vuorikatu 2 0 , SF-00100 Helsinki, Finland AFRC Institute of Food Research, Norwich Laboratory, Norwich Research Park, Colney, Norwich NR4 7UA, United Kingdom Sir F.G. Banting Research Centre, Health Protection Branch, Health and Welfare Canada, Tunney-s Pasture, Ottawa, Ontario K1A OL2 Canada Rhane Poulenc Diagnostics Ltd., Montrose House, 187 George Street, Glasgow G1 lYT, Scotland Department of Biotechnology, The Technical University of Denmark, Bygning 221, DK-2800 Lyngby, Denmark AFRC Institute of Food Research, Norwich Laboratory, Norwich Research Park, Colney, Norwich NR4 7UA, United Kingdom AFRC Institute of Food Research, Norwich Laboratory, Norwich Research Park, Colney, Norwich NR4 7UA, United Kingdom
...
Xlll
PREFACE Instruct a wise man and he will be wiser still. Proverbs, 9,9
The idea of this book has gradually developed during my of the Journal of membership of the editorial board Chromatography when I had to referee manuscripts dealing with chromatographic studies of mycotoxins. Another inspiration originated in reflections on a previous book on methods of production, isolation, separation and purification of mycotoxins which I edited in the early 1980s and which has been accepted very positively by many workers in the field. However, without a positive attitude of the Publishers and ethusiastic cooperation of the invited expert scientists the original idea would not have been transformed into reality. The book consists of two parts. In four chapters on Techniques, the most important principles of sample preparation, extraction, clean-up, and of established and prospective chromatographic techniques are discussed in relation to mycotoxins. Two shorter chapters deal with emerging techniques: immunoaffinity chromatography and enzyme-linked immunosorbent assay as alternative to chromatographic methods. In the Applications, the most important data, scattered in the literature, on thin-layer, liquid, and gas chromatography of mycotoxins have been brought together. Mycotoxins are mostly arranged according to families, such as aflatoxins, trichothecenes, lactones etc. Chromatography of individual important mycotoxins and multi-mycotoxin chromatographic analyses are also included. Applications are presented in three chapters devoted to thin-layer, liquid, and gas chromatography. Last but not least, I express my thanks to all the contributors for their excellent cooperation in preparing their manuscripts so that the book might become useful to the researchers who will use it. V. Betina
Editor
This Page Intentionally Left Blank
PART A
TECHNIQUES
This Page Intentionally Left Blank
3
Chapter 1 SAMPLING, SAMPLE PREPARATION, EXTRACTION AND CLEAN-UP
V. BETINA 1.1 INTRODUCTION The chromatography of mycotoxins is preceded by a sequence of operations which include sampling, sample preparation, extraction and clean-up. The results of the most sophisticated chromatographic procedure will be determined by the efficacy of these steps. Analytical methods must be extremely sensitive but each natural material which is expected to contain mycotoxins is composed of compounds which may interfere with the analysis, and thus specific methods must be used for a certain class of toxins present in a particular commodity. This chapter will focus upon a description of the sampling, sample preparation, extraction and clean-up procedures associated with chromatography of mycotoxins. 1.2 SAMPLING AND SAMPLE PREPARATION For the analysis of agricultural commodities good sampling techniques are of importance because the contamination of food products such as grain or nuts is most likely to occur in isolated "pockets" of mycotoxins (1). This may be due to mould proliferation and contamination of a few plants suffering from the stress of unfavourable conditions in a small portion of the field. Alternately, "pockets" of toxin may develop during storage of a larger quantity of a commodity because of localized conditions such as isolated areas of high moisture. Sampling, sample preparation, and sampling plans for foodstuffs for mycotoxin analysis have been published by Campbell et al. ( 2 ) , Dickens and Whitaker ( 3 ) and, more recently, by Park and Pohland ( 4 ) . These reviews provide lists of various types of equipment used for sample preparation and sources of supply. Sampling and subsampling procedures recommended for aflatoxins should be adequate for other mycotoxins (5). The problem of sampling is associated with the fact stressed above: there is not a normal distribution of
4
aflatoxins or other toxins within one batch. In general, more heterogeneous samples or food with larger particles require larger sample sizes. Thus, peanuts require a relatively large sample size, whereas progressively smaller sample sizes are needed for corn, wheat, rice, and millet products ( 6 ) . According to Davis et al. (l), a study investigating the presence of mycotoxins in crops in a field requires the geometric division of the field and the acquisition of representative samples from each sector. With a commodity such as corn, the sampling must be coordinated with harvesting so as to obtain kernels from a large number of ears. Sampling of stored crops with probes will only result in representative samples in the lot has been mixed by harvesting or some other mechanical operation. The increments taken from the lot should be mixed and the entire sample ground to reduce particle size and heterogeneity. Chapter 2 6 of the 14th edition of Official Methods of Analysis of the Association of Official Analytical Chemists (7) recommends grinding nuts in a large batch- type cutter to simultaneously mix and reduce particle size to a point at which the ground sample will pass through a No. 14 sieve. The subsample is obtained by systematically dividing the gross sample or utilizing a riffling device. The subsample is then more finely ground so that the particles will pass a No. 2 0 sieve. Analytical samples can then be withdrawn from this more representative subset by randomly dividing the subsample. In order to minimize the formation of mycotoxins during sampling, the analysis of the sample should be performed as rapidly as possible following collection. Warm, moist storage conditions should be avoided to prevent further mycotoxin production. Subsamples should be stored under refrigerated or dried conditions for future analysis. 1.3 SAMPLE EXTRACTION AND CLEAN-UP A variety of extraction and clean-up methods for mycotoxins have been employed. Since mycotoxins occur in a wide variety of commodities and products, the extraction from a sample depends on the physicochemical properties of the sample as well as those of the toxin. In general, the sample or ground sample is subjected to high-speed blending or mechanical shaking in the presence of the extraction solvent system. The slurry is then
5
filtered and is ready for subsequent purification procedures. Diatomaceous earth is sometimes included in the solvent system to speed the filtration step. The most efficient solvents for extracting mycotoxins are the relatively polar solvents such as methanol, acetone, acetonitrile, ethyl acetate, and chloroform. Modern techniques of mycotoxin extraction use water-organic solvent mixtures, e.g., chloroform- water (1O:l) (8). The water wets the substrate and increases penetration of the solvent mixture into the hydrophilic material. The aqueous phase can be an acid solution designed to break interactions between the toxins and sample constituents such as proteins. The small amount of the toxins taken up in the aqueous phase is immediately removed by the organic solvent, giving a rapid isolation procedure. Sodium chloride or other inorganic salts are often included in the aqueous phase to minimize the formation of emulsions during the extraction. The best known extraction and clean-up techniques, as published before 1985, were summarized by the present author (9). Examples of solvent systems utilized in the more recent literature for isolating a variety of mycotoxins are presented in Table 1.1. Extraction procedures employed for extraction of structurally- related families or individual mycotoxins are described in Chapter 7. As a large number of interfering compounds originally present in samples contaminate the primary sample extracts, these components must be removed as completely as possible. For this purpose, a variety of clean-up methods have been used. High levels of additional compounds ca be removed in several ways. For example, high levels of lipids present in certain commodities (cocoa beans, peanuts, peanut butter) would interfere with subsequent analytical procedures. For these foods, nonpolar solvents such as hexane can be included in the original solvent system (25), or they can be added after the homogenization and filtration steps to remove lipid constituents. Primary extracts in mixtures of acetone with water contain proteins that can be precipitated with lead acetate. Sometimes various pigments need to be removed from primary extracts. Scott ( 2 6 ) showed that theobromine could be removed from crude cocoa-bean extracts by treatment with silver nitrate
6
solution. It was also shown by the same author ( 2 7 ) that a coffee-bean extract could be purified by passage through a Florisil column and the unwanted contaminants eluted with tetrahydrofuran.
Aflatoxin B1 Aflatoxin M1 Aflatoxins
Deoxynivalenol Fusarochromanones Gliotoxin Nivalenol and deoxynivalenol Trichothecenes
Compounded feeds Cheese Peanuts, pistachio nuts, soya milk Feeds Maize Moist wheat or rice Agricultural commodities Barley, wheat, fusarium culture on rice Rice culture Cerea1s
Cereals and cereal products Culture filtrate Scirpentriol Sterigmatocystin Mouldy rice Culture filtrate Zearalenone Fermented corn Zearalenone and zearalenols
Chloroform, methanol10 water, acetonitrile-water Acetone-water (86:14) 11 12
Chloroform Acetone-water ( 8 0 : 2 0 ) Chloroform
13
Acetonitrile-water
16
14 15
(85:15)
Methanol
17
Chloroform Acetonitrile-water
18
19
(85:15)
Organic solvents
20
Ethyl acetate Ethyl acetate Chloroform Acetone
21 22 23
24
In addition to these preliminary clean-up procedures, other clean-up methods include column chromatography, liquid-liquid
7
extraction and commercially available solid-phase extraction (SPE) and chromatography cartridges. The sample extract is usually added to the cartridge in an appropriate solvent. The cartridge is then washed with one or more solvents in which the toxins are insoluble or less soluble than the impurities. The solvent composition is subsequently changed in such a way that the toxins are selectively eluted from the cartridge, and the eluate is collected. SPE techniques are increasingly utilized for the analysis of mycotoxins. Since these types of clean-up methods are sufficiently characterized in Chapter 2, they are not described here. Applications of clean-up procedures were reviewed recently (28) and are also included in Chapter 7 dealing with TLC of mycotoxins. Some more recent examples of clean-up techniques are presented in Table 1.2. The final step prior to analysis of the sample involves concentration of the cleaned-up extract. This is performed using a rotary evaporator operating under reduced pressure and slightly elevated temperature. After concentrating the extract to dryness a small volume of a solvent compatible with the subsequent chromatographic system is used to rinse out the flask and the final volume is adjusted with a gentle stream of nitrogen. The sample can also be placed in a steam bath under a stream of nitrogen for concentration of the sample. 1.4 ILLUSTRATIVE EXAMPLE
Extraction and clean-up of ochratoxin A can be described here as an example. Ochratoxin A present in acidified commomodities is readily soluble in many organic solvents. This charcteristic has been used in several methods. Egan et al., (37) extracted ochratoxin A from ground samples with chloroform, after acidification with aqueous phosphoric acid. Chloroform has also been used to extract the toxin from pig kidney (38), milk (39), and human plasma (40). When extracts of ochratoxin A are purified by immunoaffinity chromatography, methanol (41) or acetonitrile (42) are used. The usual next step is partial purification of the extract to remove lipids and other substances. This step can be sometimes ommited (43). In the method (37) ochratoxins are trapped in a laboratory prepared column containing diatomaceous
8
earth impregnated with sodium bicarbonate solution. Extraneous substances are washed off the column with hexane and chloroform, and the ochratoxins are eluted with benzene-acetic acid ( 9 8 : 2 ) . TABLE 1.2 Examples of clean-up techniques for mycotoxins
Aflatoxin B1
Phenyl non-polar bonded-phase 29 Reversed-phase disposable cartridges 10 Aflatoxins Silica gel column 12 Extract in aqueous methanol defatted 30 with hexane, toxins partitioned into chloroform, placed on silica gel column, eluted with chloroform-acetone Silica gel 60 column eluted with 31 chloroform-methanol ( 8 : 2 ) Sep-Pak Florisil and C18 cartridge 13 CB method 32 Alternaria toxins Liquid-liquid partition and 33 column chromatography Deoxynivalenol 16 Acetonitrile-water extract partially purified on a preparative minicolumn Fusarochromanones Thin-layer or column chromatography 17 18 Precipitation with petroleum ether and Gliotoxin gel permeation chromatography T-2 toxin C18 and silica gel column 34 Trichothecenes Precipitation of proteins with lead 20 acetate, purification of toxins with hexane and chloroform 35 Silica gel minicolumn Zearalenone Zearalenone and Silica gel column and elution with 24 hexane-ethyl acetate ( 8 : 2 ) zearalenols Multimycotoxin Gel permeation chromatography 36 (aflatoxin, ochratoxin and zearalenone)
* CB
=
Contamination Bureau.
9
In the procedure (38) for the determination of ochratoxin A in kidneys of swine, a liquid-liquid partitioning step is used instead. Most recently, first commercial prototypes of immunoaffinity cartridges for ochratoxin A have become available (44). These columns are composed of monoclonal antibodies specific for ochratoxin A and immobilized on Sepharose and packed into small plastic cartridges. The crude extract is forced through the column and ochratoxins are left bound to the immunoglobulin. Extraneous material is washed off the column with water or aqueous buffer, and the ochratoxins are finally eluted with acetunitrile. 1.5 CONCLUSIONS This short chapter was written with the aim to show the necessary operations which usually must precede analytical or preparative chromatography of mycotoxins: sampling, sample preparation, extraction, and clean-up procedures. Examples taken from recent literature concerning extraction were included to show the variety of materials in which the presence of mycotoxins has to be proved or disproved chromatographically. Some recent clean-up techniques were also described.
REFERENCES 1 N.D. Davis, J.W. Dickens, R.L. Freie, P.B. Hamilton, O.L. Shotwell, T.D. Wyllie and J.F. Fulkerson, J. Assoc. Off. Anal. Chem., 70 (1980) 95. 2 A.D. Campbell, T.B. Whitaker, A.E. Pohland, J.W. Dickens and D.L. Park, Pure Appl. Chem., 58 (1986) 305. 3 J.W. Dickens and T.B. Whitaker, in H. Egan, L. Stoloff, P. Scott, M. Castegnaro, I.K. O'Neil and H. Bartsch (Editors), Environmental Carcinogens - Selected Methods of Analysis. Vol. 5 : Some Mycotoxins. ARC, Lyon, 1982, p. 17. 4 D.L. Park and A.E. Pohland, J. Assoc. Off. Anal. Chem., 72 (1989) 399. 5 J.W. Dickens and T.B. Whitaker, in R.J. Cole (Editor), Modern
6 7 8 9 10
Methods in the Analysis and Structural Elucidation of Mycotoxins, Academic Press, New York, 1986, Ch. 2, p. 29. J.E. Smith and M.O. Moss, in Mycotoxins: Formation, Analysis and Significance, Wiley, New York, 1985, p. 104. Official Methods of Analysis of the Association of Official Analytical Chemists, AOAC, Arlington, VA, 14th. ed., 1984, Ch. 26. P.M. Scott, Adv. Thin Layer Chromatogr. (Proc. 2nd Bienn. Symp. 1980), 1982, p. 321. V. Betina, J. Chromatogr., 334 (1985) 211. H.P. van Egmond and P.J. Wagstaffe, Food Addit. Contamin., 7 (1990) 239.
LO 11 J.P. Bijl and C.H. van Peterghem, J. Assoc. Off. Anal. Chem., 70 (1987) 472. 12 R. Biffoli, F. Chiti and G. Modi, Riv. SOC. Ital. Sci. Aliment., 8 (1990) 19. 13 H.P. van Egmond, S.H. Heisterkamp, W.E. Paulsch and H.P. van Egmond, Food Addit. Contamin., 8 (1991) 17. 14 N. Bradburn, R.D. Coker, K. Jewers and K.I. Tomlins, Chromatographia, 29 (1990) 435. 15 M.A. Moreno, A. Olivares and G. Suarez, Mycotoxin Res., 5 (1989) 51. 16 W.C. Gordon and L.J. Gordon, J. Assoc. Off. Anal. Chem., 73 (1990) 266. 17 F.S. Chu, J. Assoc. Off. Anal. Chem., 74 (1991) 655. 18 J.L. Richard, R.L. Lyon, R.E. Fichtner and P.F. ROSS, Mycopathologia, 107 (1989) 145. 19 D.R. Lauren and R. Greenhalgh, J. Assoc. Off. Anal. Chem., 70 (1987) 479. 20 A.N. Kotik and V.A. Trufanova, Gig. Sanit., 1989, No. 9, 53. 21 K.E. Richardson, G.E. Toney, C.A. Haney and P.B. Hamilton, J. Food Protect., 52 (1989) 871. 22 Y. Horie, M. Miyaji, K. Nishimura, H. Toguchi, H. Yamaguchi and S. Udagawa, Proc. Jpn. Assoc. Mycotoxicol., 1989, No. 29, 21. 23 I.A. El-Kady, A.H. Moubasher and S.S.M. El-Maraghy, Egypt. J. Bot., 31 (1988) 99. 24 R. Vesonder and P. Golinski, Mycotoxin Res., 7A (Suppl.), Part I1 (1991) 175. 25 L. Stoloff, J. Assoc. Off. Anal. Chem., 66 (1983) 355. 26 P.M. Scott, J. Assoc. Off. Anal. Chem., 52 (1969) 72. 27 P.M. Scott, J. Assoc. Off. Anal. Chem., 51 (1968) 609. 28 v. Betina, J. Chromatogr., 477 (1989) 187. 29 N. Bradburn, R.D. Coker and R. Jewers, Chromatographia, 29 (1990) 177.. 30 J. Wu, J. Toxicol. Toxin Revs., 9 (1990) 120. 31 0. Sanchey, Food Lab. News, 1989, No. 17, 49. 32 R. W. Beaver, D.M. Wilson and M.W. Trucksess, J. Assoc. Off. Anal. Chem., 73 (1990) 579. 33 M. Kostecki, J. Grabarkiewicz-Szczesna and J. Chelkowski, Mycotoxin Res., 7 (1991) 3. 34 K.A. Koddington, S.P. Swanson, A.S. Hassan and W.B. Buck, Drug Metab. Disposit., 17 (1989) 600. 35 P. Lepom, Arch. Anim. Nutrit., 38 (1988) 799. 36 C. Dunne, M. Meaney and M. Smyth, J. Chromatogr., (In press). 37 H. Egan, L. Stoloff, M. Castegnaro, P. Scott, I.K. O'Neill
38
39 40 41
and €IBartsch . (Editors, Environmental Carcinogens: Selected Methods of Analysis, Vol. 5, Some Mycotoxins, IARC, Lyon, 1982, p. 255. W.E. Paulsch, H.P. van Egmond and P.L. Schuller, in Proceedings, V International IUPAC Symposium on Mycotoxins and Phycotoxins, September 1-3, 1982, Vienna, Austrian Chemical Society, Vienna, 1982, p. 40. M. Gareis, E. Martlbauer, J. Bauer and B. Gedek, Z . Lebens. Unters. Forsch., 186 (1988) 114. A. Breitholtz, M. Olsen, A. Dahlback and K. Hult, Food Addit. Contam., 8 (1991) 183. S.C. Lee and F.S. Chu, J. Assoc. Off. Anal. Chem., 6 7 (1984)
45. 42 N. Ramakrishna, J. Lacey, A.A. Candlish, J.E. Smith and I.A. Goddbrand, J. Assoc. Off. Anal. Chem., 73 (1990) 71. 43 K. Hult, R. Fuchs, M. Peraica, R. PleStina and S . Ceovic, J.
11
Appl: Toxicol., 4 (1984) 326. 44 J. Gilbert, in Proceedings of the International Conference on Fungi and Mycotoxins in Stored Products, Bangkok, Thailand, 23-26 April 1991 (in press).
12
Chapter 2 TECHNIQUES OF THIN LAYER CHROMATOGRAPHY R.D. COKER, A.E. JOHN and J.A. GIBBS
2.1 INTRODUCTION The use of horizontal thin layers as analytical tools was first described, in 1938, by the Russian workers Ismailov and The technique, known as "drop chromatography", Shraiber 1 1 1 . was largely ignored for the following 1 0 years until two American workers, Meinhard and Hall [2l, described the separation of metal ions in aqueous solution using microscope slides coated with an alumina-rich mixture. Of the several separations mentioned, the first and most simple was for aqueous solutions of Fe3+ and Zn2+ ions. The introduction of the thinlayer technique, as a routine analytical method, is generally attributed to the work of Kirchner and his associates in 1951 [3-51. Subsequently, thin layer chromatography (TLC) has been utilised for the separation and quantification of a wide range of compounds, including mycotoxins. The analysis of mycotoxins involves a sequence of discrete operations which includes sampling, sample preparation, extraction, clean-up, quantification and confirmation procedures [6,7]. Needless to say, the validity of the TLC quantification results will be determined by the efficacy of the sampling, sample preparation, extraction and clean-up steps [6-81. This chapter will focus upon a description of the clean-up and quantification procedures associated with thin layer chromatography. 2.2 CLEAN-UP METHODS Since mycotoxins occur in a wide variety of commodities and products, the analyst is faced with the problem of removing a large number of disparate, interfering compounds from the sample extracts. A variety of clean-up methods have been employed [6,81 including column chromatography [9-201, liquid-liquid extraction
and chemical adsorption [241 procedures (see also Chapter 1). Silica gel has been extensively used in column chromatography clean-up. Commodities (and mycotoxins) to which this method has been applied include cereals (aflatoxins) [251, oilseeds (aflatoxins) [ 1 2,15 I, vegetable oils (aflatoxins) [ 10 I, meats (aflatoxins) 1131, spices (aflatoxins) [16,171, dried fruits (aflatoxins) [11,26], wine (aflatoxins) [ l a ] , coffee (aflatoxins) [20], animal feeds (aflatoxins) [9,21 1, milk (aflatoxin M I ) [271, animal viscera (aflatoxins) [19,281, cereals (deoxynivalenol, zearalenone) [291 and porcine kidneys (ochratoxin A ) [301. Similarly, Florisil has been applied to the clean-up of cereals (trichothecenes, moniliformin, butenolide and zearalenone) [31]. Florisil, modified with oxalic acid, has also been used to clean up corn and groundnut meal (aflatoxins) [321 and cellulose columns have been used to clean up animal tissues (aflatoxins, including M i ) [281. Liquid/liquid extraction clean-up procedures, utilising acetonitrile/petroleum ether 1331 and chloroform/aqueous HC1 [34] have been used during the analysis of, for example, maize and barley (zearalenone) [331 and corn (citrinin) [341, whilst chemical adsorbents have been applied to blue cheese (roquefortine) 1351 and black olives (ochratoxin A ) [361. The clean-up methods described above are laborious, time[21-241
consuming and of limited efficiency. Because of these disadvantages, clean-up procedures using commercially available solid phase extraction (SPE) cartridges are increasingly utilised for the analysis of mycotoxins [61. S P E techniques involve the partitioning of analytes and interfering compounds between a mobile and stationary phase. The latter, contained within the cartridge, is composed of a solid adsorbent or an immobilised (bonded) liquid phase [61. Available bonded phases include ethyl (CZ), octyl (C8), octadecyl (C18), cyclohexyl ( C H ) , phenyl ( P H ) , cyanopropyl (CN), diol (20H), aminopropyl (NH2) and a selection of ion exchange phases. S P E clean-up has been applied to the analysis of aflatoxin in groundnuts [371, peanut butter [381, cottonseed [391, and corn [40,41] by utilising silica gel as the solid adsorbent.
14
Florisil has been used for the analysis of aflatoxins in sorghum [42] and green coffee. Bonded phases have been utilised for the analysis of aflatoxin Mi in milk [431, of aflatoxins B1 and M1 in animal tissues [44] and of aflatoxin Bl, ochratoxin A and citrinin in human urine [451. SPE clean-up, utilising the phenyl (PHI bonded-phase, is routinely used in the authors’ laboratory for the analysis of aflatoxins in a range of commodities including maize, cottonseed, peanut butter and palm kernel 146-49,921 (Figure 2.1). The estimation of the aflatoxin content of corn [461, for example, is initiated by extraction with aqueous acetone. A 5 ml aliquot of the filtered extract is diluted with aqueous methanol (6.7% v/v) and acetic acid (1% v/v) mixture (60 ml) and mixed in a reservoir attached to a solvated PH (phenyl) cartridge. Using a vacuum manifold, the sample mixture is drawn through the cartridge at a flow rate of approximately 7 ml rnin-’
Figure 2.1. The elution of solid phase extraction cartridges: aflatoxins are eluted from the phenyl cartridge, through a sodium sulphate drying column, with a volume of chloroform appropriate for the commodity.
15
thus retaining the aflatoxins in the stationary phase. The cartridge containing the aflatoxins is then dried, by pulling air through the cartridge, and, finally, the aflatoxins are eluted with chloroform ( 7 ml). The chloroform solution is drawn through a second cartridge, containing anhydrous sodium sulphate, before removing the solvent by evaporation under nitrogen. The resultant residue may then be stored, in the dark, at -2OOC before quantification. Commodities which produce excessive interfering compounds, such as cottonseed, require additional clean-up steps. Chemical adsorbents such as lead acetate solution may be used, [ 5 0 , 5 1 ] in addition to SPE, in such instances. After extracting with aqueous acetone, lead acetate solution (20% w/v; 2 ml) is added to the aqueous methanolfacetic acid mixture. Diatomaceous earth is also added to act as a filter aid. The SPE clean-up is then performed as previously described. The simplicity of the SPE clean-up makes it ideally suited to the analysis of large numbers of samples. In the authors' laboratory up to 60 samples per day are routinely prepared for quantification using this procedure. The introduction of commercially available liquid handling equipment has facilitated the automation of SPE clean-up procedures. One such application is the analysis of aflatoxin Mi in milk [521. Shepherd et al. [53], using HPLC quantification, compared six clean-up procedures for aflatoxin M1 in milk. Of these six methods, a procedure using a C 1 8 cartridge was the most efficient in terms of cost, analysis time and clean-up efficiency; 0.0005 pg kg-I of aflatoxin M I in whole milk was detected [ 5 4 1 . SPE clean-up procedures are not, of course, universally applicable. The technique, for example, failed to adequately clean-up extracts of sorghum, even when the lead acetate precipitation step was included. However, suitably clean extracts were successfully produced using a Florisil clean-up column [ 4 2 1 .
16 2 . 3 NORMAL PHASE TLC 2 . 3 . 1 Principles
A normal TLC plate consists of a thin, uniform layer of particulate adsorbent, the stationary phase, applied to a flat plate. Silica, alumina and cellulose are frequently used as stationary phases.
The chromatography
is performed
by
the
dropwise application (Figure 2 . 2 ) of microlitre quantities, of a solution of the cleaned-up analyte mixture, to one end of the TLC plate.
The mixture is drawn through the stationary phase,
by capillary action, within the developing solvent (the mobile phase), which is usually contained within a sealed glass tank (Figure
2.3).
During
this
chromatographic
process,
the
components of the analyte mixture are partitioned between the stationary
and
mobile
phases.
The
components
of
greater
polarity will have the greater affinity for the stationary phase and
will
travel
more
slowly
through
the
adsorbent,
thus
effecting the separation of the analyte mixture. In normal phase TLC the stationary phase is more polar than the mobile phase. Typically, the particle distribution will be
Figure 2.2, The application, by micro-pipette, of a sample extract to a two-dimensional TLC plate.
17
Figure 2 . 3 . The development of a two-dimensional TLC plate: the plate is developed in an appropriate solvent in the first direction, air dried and then rotated through 90' before development in the second direction with another solvent
5-80 pm with a mean particle size of approximately 20 pm. 2.3.2. Practical considerations
Normal TLC may take a variety of forms, the chosen method often depending upon the amount of additional sample clean-up required. 2.3.2.1 One and two dimensional TLC
When little additional clean-up is required, one dimensional TLC is often sufficient for the separation and quantification of mycotoxins. Using this method, multiple samples can be simultaneously developed and quantified. However, when the initial clean-up is inadequate, or when there are many sample components of interest, it may be necessary to extend the chromatographic separation into the second dimension. Here, the plate is dried after the first
18
development and is then rotated through 90' and developed in a different solvent, affording better resolution of the components and the removal of interfering compounds. However , one disadvantage of two dimensional (2D) TLC is that only one sample at a time can normally be evaluated (Figure 2 . 4 ) .
Figure 2 . 4 . Two dimensional thin layer chromatography: the plate shown is for a mixture of pure aflatoxins. It has been sprayed with 509 sulphuric acid and dried at 105'C to identify the aflatoxins which change colour from blue-green to light yellow. Both one dimensional and 2D TLC have been applied, with One dimensional great success to the analysis of mycotoxins. TLC, for example, has been applied [55,561 to the analysis of the aflatoxins, the ochratoxins, zearalenone, citrinin, patulin, the trichothecenes, cyclopiazonic acid, the rubratoxins, sterigmatocystin, penicillic acid, butenolide and citreoviridin. Two-dimensional TLC has been applied, for example, to the aflatoxins, ochratoxin A , cyclopiazonic acid and citrinin [ 5 6 1 . 2.3.2.2 Bi-directional development
to samples Bi-directional TLC is also applied [ 4 6 ] requiring substantial, additional clean-up. The first step, in this method involves the application of the samples approximately 3 cm from one long edge of a 1 0 x 20 cm aluminiumbacked plate. A preliminary development is performed to
19
transport interfering compounds into the area between the line of application and the edge of the plate.
This area of the
plate is then removed using a sharp knife.
The TLC plate is
then rotated through 1 8 0 ' phase.
and developed in a suitable mobile
Multiple developments may be employed if necessary.
Bi-directional TLC will be discussed further in Section 2.5. 2.3.2.3
Circular development
Two varieties of this technique are currently in use.
In
circular chromatography, the samples are applied at, or near, the centre of the plate and the development solvent is delivered centrally via a capillary or a wick. developed
in a so-called U
Typically, the plates are
chamber 1571.
separated into diffused arcs.
The analytes are
It is reported that separation
and resolution are better than that normally achieved by linear development 1581.
Figure 2 . 5 . Circular chromatography: the U-Chamber. (by permission of Camag, Switzerland)
In anti-circular chromatography, the samples are spotted around the circumference of the plate and the solvent moves from the edge to the centre of the plate. Theoretically, this technique should result in more compact spots since the area
20
into which the spots can move is restricted. In practice, resulting spots are elongated and some loss of resolution occur. Relatively expensive equipment is required for circular (Figure 2.5). The method has not been widely applied to
the can TLC the
analysis of mycotoxins. 2.3.2.4 Triangular TLC A further variation of TLC is performed using triangular plates as described by Issaq [591 for the separation of dye mixtures. The chromatography is performed on triangular plates, prepared from 5 X 20 cm or 5 X 10 cm plates; the samples are applied along the 5 cm base of the triangle. As in the case of anti-circular chromatography, the area into which the spots migrate is restricted thus reducing spot diffusion. Elongation of the spots occurs after the first development but a second development, in the same solvent system, transforms the spots to a circular shape. The technique has been assessed for the quantification of aflatoxins and cyclopiazonic acid [60]. These preliminary experiments suggested that the resolution was inferior to that obtained with HPTLC and that the geometry of the plates resulted in problems during automated densitometric quantification.
2.4 REVERSE-PHASE TLC (RPTLC) 2.4.1 Principles In RPTLC, the mobile phase is more polar than the stationary phase which is composed of silica to which non-polar groups (typically, C2, C8, C18 or phenyl) have been chemically bonded. Early attempts at RPTLC employed normal phase plates impregnated with paraffin [61I or silanized using a suitable alkyl chlorosilane. Acetylated cellulose plates were also employed [621. 2.4.2 Practical considerations Although RPTLC has not been widely applied to mycotoxins, recent reports include the analysis by RPTLC of ochratoxin A [ 63,641, of a range of toxins using C18 plates (aflatoxins B1 , B2, G I , G2, sterigmatocystin, ochratoxin A, citrinin, penicillic
21
acid, patulin, zearalenone, deacetoxyscirpenol, T-2 toxin, HT-2 toxin, nivalenol, neosolaniol, fusarenone-X, deoxynivalenol and 3-acetyl-deoxynivalenol) [651. High performance RPTLC C8 plates have been used to analyse aflatoxin M1 in milk [661 and, in the authors’ laboratory, C18 plates have been used for estimating zearalenone and alternariol monomethyl ether [671. 2.5 HIGH PERFORMANCE THIN LAYER CHROMATOGRAPHY (HPTLC) 2.5.1 Principles HPTLC has evolved with the development [681 of a) automated sample application and plate interpretation equipment and b) high quality TLC plates. Sample application equipment facilitates the accurate and precise application of nanolitre quantities of cleaned-up sample extracts using automated (or semi-automated) techniques (Figures 2.6a and 2.6b).
Figure 2.6a. The Nanomat 111: a semi-automatic spotting instrument which uses microcaps clamped in precise positions. (by permission of Camag, Switzerland)
High quality (performance) TLC plates are uniformly coated with adsorbents of small particle size, typically within the 2 1 0 pm range, and with a layer thickness of 0.1 - 0.3 mm. Both normal and reversed phase adsorbents are available.
22
Figure 2.6b. The Camag 27210 TLC autosampler: using a fine bore capillary, position on the plate, dosage and dosage rate can be specified for a number of samples and standards, with high precision.
2.5.2 Practical considerations Modern sample application equipment produces a spot of the order of 1 mm diameter which allows up to 30 samples to be
accommodated along the longer side of a 10 x 20 cm plate. The small particle sized adsorbents facilitate the rapid separation of the analyte components, sometimes after only 3 cm of travel. Although a selection of development chambers have been reported [68] for HPTLC, conventional 21 cm (h) x 28 cm (1) x 6 cm (w) glass tanks are successfully used in the authors' laboratory [691. One-dimensional, two-dimensional, circular, anti-circular and bi-directional development methods ( sections 2.3.2.(1-3)) have all been applied to HPTLC plates. In the authors' laboratory bi-directional normal HPTLC is routinely used for the
23
analysis of aflatoxin in a wide range of commodities including edible
nuts,
oilseeds,
[7,8,42,46-491
cereals,
root-crops
and
spices
(Figure 2.7).
Figure 2 . 7 . The analysis of aflatoxins by HPTLC: the plate shown is for samples of cottonseed. There are 25 sample spots and six mixed aflatoxin standards. Some of the contaminants have been removed by a preliminary development in the reverse direction in dry ether and then cutting away the band of impurities before two developments in the normal direction in 6 : 3 : 1 chloroform: xy1ene:acetone. A development tank, for the di-ethyl ether development stage of bi-directional HPTLC and a drying chamber, to remove residual solvent between developments, have been designed and fabricated
[701 in the author's
laboratory
(Figures 2.8a
and
2.8b). Attempts to improve the precision of the development step have
resulted
Chromatography
in
the
technique of Over
(OPTLC) [71,721.
Pressured
In OPTLC,
Thin-Layer
the TLC plate is
covered by a pressurised flexible membrane and the mobile phase is forced through the thin layer with the aid of a pump. A high separation efficiency is reported [71 I 2.9).
for this method
(Figure
TO date OPTLC has not been applied to the analysis of
mycotoxins. 2 . 6 PREPARATIVE TLC
2.6.1 Principles
The
increased
layer
thickness
(1-5
mm)
employed
in
preparative TLC (PTLC) facilitates the application and isolation
of milligram quantities of analyte.
24
Figure 2.8a. A vertical metal tank suitable for ether development. This has the advantages of much quicker development than in a glass tank, better retention of the very volatile solvent, and for aflatoxins, the exclusion of light.
Figure 2.8b.
A
fan-assisted plate dryer.
Figure 2.8. An HPTLC development tank and fan-assisted drying chamber.
25 6
\
1
3
2
10
4 1-
Figure 2 . 9 . The Chompres 1 0 OPLC development chamber. 1 . Bottom support block 7. Solvent inlet valve for 2. Polymethacrylate support plate controlled pumping of the 3 . External frame mobile phase 4 . Position for chromatographic plate 8. Solvent outlet 5. Clamp 9. Water outlet for release 6. Water inlet for supplying of cushion pressure 1 0 . Hydraulic system for pressure to upper side of plate through a plastic foil cushion operating blocks
(by permission of Dr Alfred Huethig Verlag, Heidelberg)
Typically, the sample extract is applied to the plate as a horizontal streak using a manual or automated applicator (Figure 2.10 ) . After development, the separated component "bands" may be located by their UV absorption or fluorescent properties. Alternatively, non-UV absorbing compounds may be visualised by carefully spraying the extremities of the plate with a suitable colour-producing reagent. The individual "bands" can then be carefully scraped from the plate and extracted from the adsorbent using a suitable solvent. 2.6.2 Practical considerations I n PTLC, the increased thickness of the adsorbent layer can lead to vertical band spreading. Tapered-bed plates have been
26
Figure 2.10. The application of sample to a PTLC plate: an even streak of the sample is delivered through a syringe, the plunger being depressed uniformly by the inclined sliding rail.
developed [73] to overcome this problem.
These plates feature a
pre-adsorbent layer of 0.7 mm thickness followed by a tapered separation area which increases in thickness from 0.3-1.7 mm. The tapered bed allows a more uniform mobile phase flow pattern which
reduces
the
vertical
band
spread
and
migration distance of the more polar components.
increases A
the
variety of
mycotoxins have been isolated using PTLC including cytochalasins H and J
[74], chaetoglobosins K and L
[751, proxiphomin and
protophomin [76], citreoviridin [771 and paspalitrem
A
[781, the
trans and cis isomers of zearalenone [791, isotopically labelled ochratoxin
A
zygosporins
[801, the rubratoxins [811, mycophenolic acid [821, [831,
sporidesmins A ,
deoxaphomin
[841,
janthitrems
[851,
C and G [86], PR toxin [87] and the methylated
derivative of territrem C
[88].
PTLC has been used
in the
authors’ laboratory for the partial purification of A l t e r n a r i a mycotoxins [891.
21
2.7 DETECTION After the development of the TLC plate, the separated components are located and quantified. If appropriate, the U V absorbance or fluorescent properties of the analytes may be used for this purpose. Alternatively, chromophores may be introduced into non-UV absorbing compounds by treating (by spraying or dipping) the plates with suitable reagents. Chemical derivatisation i n - s i t u may also be used to confirm the presence of a suspected mycotoxin. 2.7.1 Fluorescence detection The natural fluorescence of mycotoxins under ultra-violet light is widely used in their detection and quantification, allowing the detection, in some cases, of picogram quantities of these compounds. Naturally fluorescent mycotoxins include the aflatoxins (B1, B2, G1 and G2), aflatoxin Mi, zearalenone, ochratoxin A, sterigmatocystin, citrinin, patulin and penicillic acid [go]. The aflatoxins, for example, appear as characteristic blue (B1 and B2) and blue-green (GI and G2) fluorescent spots under long-wave ( 3 6 5 nm) UV light, zearalenone as a blue-green spot, citrinin as a yellow spot, sterigmatocystin as a dull brick-red spot and penicillic acid as a weak, light purple spot. 2.1.2 Chemical derivatisation The formation, in s i t u , of fluorescent derivatives can be used to a ) detect non-fluorescent mycotoxins, b) enhance the fluorescence of naturally fluorescing mycotoxins and c) confirm the presence of presumptive mycotoxins [ 5 6 , 9 0 1 . The non-fluorescent trichothecenes may be detected by chemical derivatisation. T - 2 toxin, for example, appears as a grey-blue fluorescent spot after spraying with 20 per cent concentrated sulphuric acid in methanol and heating at llO° C for 3 to 4 minutes. Alternatively, T-2 toxin will afford a bright blue fluorescence if treated with a mixture of aluminium chloride (in water:ethanol, 1:l) and chromotropic acid (in concentrated sulphuric acid:water, 5 : 3 ) followed by heating at 110OC.
The natural fluorescence of sterigmatocystin may be enhanced, to afford a bright yellow spot, by spraying with a 24 percent solution of aluminium chloride in 9 5 percent aqueous
28
ethanol and heating at 105'C for 10 minutes. The identity of sterigmatocystin may be confirmed by the formation of the acetate [911 or hemiacetal [921 derivatives. Similarly, the long-wave fluorescence of zearalenone and ochratoxin A may be enhanced if the plate is sprayed with aluminium chloride solution. The presence of ochratoxin A may be confirmed by the formation of the ethyl ester derivative. The natural, shortwave (254nm) fluorescence of patulin can be enhanced by treatment with 0.5 percent aqueous 3-methyl-2-benzothiazolinone hydrazone (MBTH) followed by heating at 130'C for 15 minutes. If penicillic acid is treated with MBTH, a visible pale yellow spot is produced. Citrinin tends to streak in many solvent systems and is probably best chromatographed on silica gel TLC plates impregnated with oxalic acid or ethylenediaminetetra-acetic acid (EDTA). The long-wave yellow fluorescence of citrinin may be converted to a green fluorescence by spraying the plate with 14 percent (w/w) boron trifluoride in ethanol. The presence of citrinin may be confirmed by the formation of the acetate derivative. 2.7.3 Bioautographic methods This procedure involves a combination of preparative TLC and biological detection. The trichothecenes T-2 and HT-2, for example, have been isolated by PTLC and detected by their toxicity towards the yeasts Kluyveromyces fragilis and Saccharomyces cerevisiae. The reported limit of detection was 0.2nM per spot [561. Bioautographic detection using Bacillus subtilis has been used in the PTLC of gliotoxin 1941. Similarly, bioautographic methods have also been applied to aflatoxin Bl, kojic acid and sterigmatocystin using Artemia (brine shrimp) larvae as the biological detection salina organism [951.
2.8 QUANTITATIVE AND SEMI-QUANTITATIVE EVALUATION The
quantitative
interpretation
of
developed
(HP))TLC
plates is performed by comparing the fluorescent intensities of standard mycotoxins extracts [90,911.
and
the
mycotoxin
components
of
sample
Using the automated and the semi-automated
29
Figure 2.11a. The Camag TLC I1 Scanner: the plate is secured on a graduated platen using magnetic spacers.
Figure 2.11b. Camag TLC I1 Scanner linked to a SP4270 Integrator and a PC. Figure 2.11. The densitometric quantification of a TLC plate.
30
equipment associated with HPTLC methodology (section 2 . 5 ) picogram quantities of aflatoxin, for example, can be precisely and accurately detected (Figure 2 . 1 1 ) . In the absence of densitometric equipment, the intensities of mycotoxin standard and sample spots may be compared visually. This semiquantitative interpretation is laborious and requires the services of a skilled technician. If mycotoxin standards are not available, the dilution to extinction method may be employed. Decreasing volumes of sample extract (typically 2 5 to 5 pl) are applied to the plate and the smallest volume (V) in which the mycotoxin is visible is then identified. If the minimum quantity of mycotoxin that can be visually detected is known, the concentration of mycotoxin in volume V and, consequently, in the original sample, can be calculated [ g o ] .
2.9
ILLUSTRATIVE EXAMPLES HPTLC methods have been developed in the authors' laboratory which facilitate the accurate and precise quantification of aflatoxin in edible nuts, oilseeds and their derivatives, cereals, root crops and spices [ 4 2 , 4 6 - 4 9 1 ; an HPTLC method has also been developed recently for cyclopiazonic acid in groundnut cake, Each newly developed TLC method must be validated by determining the accuracy, precision and limit of detection of that method for any given commodity (including different varieties). The following validation procedure is applied in the authors' laboratory. Mycotoxin-free extracts of the commodity are spiked to afford a range of mycotoxin concentrations, The choice of concentration range should depend upon the contamination levels of interest associated with that commodity. Table 2 . 1 shows suggested aflatoxin spiking levels for various commodity types. A second experiment should be performed using the same levels of artificial contamination, but using pure extraction solvent as opposed to the sample matrix. This will eliminate any possible bias in the methodology due to the components of the sample matrix. Six replicate analyses should be carried out Weighted regression at each of the contamination levels. analysis of the data will facilitate the calculation of the accuracy, precision and limit of detection for the method [ 9 2 1 .
31
Table 2.1 Suggested spiking levels for various commodity types. ~
Class of commodity
Typical permitted level (total) (pg/kg)
Suggested range for aflatoxin B1 (pg/kg)*
Foodstuffs
10
0, 2, 4, 8, 10 15, 30, 50, 100
Feedstuffs
50
0, 4, 8, 20, 40 50, 60, 100, 250
Raw materials for feedstuffs
0, 5, 10, 50, 100
200
150, 250, 500
*Concentrations of aflatoxins B1 and GI ; the concentrations of aflatoxins B2 and G 2 are approximately half those shown for B1 and G I .
The proposed method should then be further validated by comparison with an "official method" of analysis.
Naturally-
contaminated samples at two contamination levels are required for this comparison [921.
2.10 CONCLUSIONS
Modern HPTLC is a precise and accurate analytical tool with an efficiency which is comparable to that of high performance liquid assay phenyl
chromatography
(HPLC) and
(ELISA) methods. bonded-phase
enzyme
linked
immunosorbent
The application of a combination of
clean-up
and
HPTLC
to
the
analysis
of
aflatoxin in peanut butter has demonstrated [ 4 9 ] the efficiency of
HPTLC
quantification
as
compared
to
HPLC
and
ELISA
procedures. HPTLC is ideally suited to the analysis of large numbers of accumulated samples. Approximately thirty samples, for example, may be simultaneously chromatographed on a single 10 x 20 cm plate. The ability to perform TLC in a two-dimensional or bidirectional mode enables valuable, additional sample clean-up to be performed during the quantification step. the disposable nature of
TLC
plates
facilitates
Furthermore, the
use
of
32
extreme reaction conditions during the i n - s i t u chemical derivatisation of separated analytes. Undoubtedly, the combination of automated sample clean-up and modern HPTLC methodology can provide the mycotoxicologist with a very powerful analytical tool. Solid phase extraction clean-up combined with bidirectional HPTLC is the method of choice for the analysis of aflatoxins in the authors’ laboratory.
REFERENCES 1 N.A. Izmailov and M.S. Shraiber, Farmatsia, 3 (1938) 1 . 2 J.E. Meienhard and N.F. Hall, Anal. Chem., 21 (1949) 185. 3 J.G. Kirchner, J.M. Miller and K.J. Keller, Anal. Chem. , 23 (1951) 426. 4 J.M. Miller and J.G. Kirchner, Anal. Chem., 24 (1952) 1480. 5 J.M. Miller and J.G. Kirchner, Anal. Chem., 26 (1954) 2002. 6 R.D. Coker and B.D. Jones in R. Macrae (Editor), HPLC in Food Analysis, Academic Press Ltd., London, 1988, p. 335. 7 R.D. Coker in J. Gilbert (Editor), Analysis of Food Contaminants, Elsevier Applied Science Publications, New York, 1984, p. 207. 8 N. Bradburn, R.D. Coker, K. Jewers and K.I. Tomlins, Chromatographia, 29 (1990) 435. 9 G.N. Shannon, O.L. Shotwell and W.F. Kwolek, J. Assoc Off. Anal. Chem., 66(3) (1983) 582. 10 N. Miller, H.E. Pretorius and D.W. Trinder, J. Assoc Off. Anal. Chem., 68(1) (1985) 136. 1 1 D. Boyacioglu and M. Gonul, J. Assoc Off. Anal. Chem., 71(2) (1988) 280. 12 P.M. Scott, J. Assoc Off. Anal. Chem., 52(1) (1969) 72. 13 L.B. Bullerman, P.A. Hartman and J.C. Ayres, J. Assoc Off. Anal. Chem., 52(3) (1969) 638. 14 R.C. Shank, G.N. Wogan, J.B. Gibson and A . Nondastu, Food Cosmet. Toxicol., 10 (1972) 61. 15 J. Karnelic, M. Israel, S. Benado and C. Leon, J. Assoc Off. Anal. Chem., 56(1) (1973) 1452. 16 P.M. Scott and B.P.C. Kennedy, J. Assoc Off. Anal. Chem., 56(6) (1973) 1452. 17 P.M. Scott and B.P.C. Kennedy, Can J. Inst. Food Sci Technol. J., 8(2) (1975) 124. 18 D.M. Takahashi, J. ASSOC Off. Anal. Chem., 57(4) (1974) 875. 19 R.D. Stubblefield and O.L. Shotwell, J. ASSOC Off. Anal. Chem., 64(4) (1981) 964. 20 P.M. Scott, J. Assoc Off. Anal. Chem., 51(3) (1968) 609. 21 S-C. Chen and L. Friedman, J. ASSOC Off. Anal. Chem., 49(1) (1966) 28.
33 2 2 W.A. Pons and L.A. Goldblatt, J. Amer. Oil Chem. SOC., 4 2 ( 6 ) ( 1 9 6 5 ) 471. 2 3 W.A. Pons, A.F. Cucullu and A.O. Franz, J. Assoc Off. Anal. Chem., 5 5 ( 4 ) ( 1 9 7 2 ) 7 6 8 . 2 4 T.R. Romer, J. Assoc Off. Anal. Chem., 6 8 ( 1 9 7 5 ) 5 0 0 . 2 5 G.M. Shannon and O.L. Shotwell, J. Assoc Off. Anal. Chem., 5 8 ( 4 ) ( 1 9 7 5 ) 7 4 3 . 2 6 D. Boyacioglu and M. Gonul, Food Additives and Contaminants, 7 ( 2 ) ( 1 9 9 0 ) 2 3 5 . 2 7 R.D. Stubblefield, J. Amer. Oil Chem. SOC., 5 6 ( 9 ) ( 1 9 7 9 ) 8 0 0 . 2 8 N.L. Brown, S . Nesheim, M.E. Stack and G.M. Ware, J. Assoc Off. Anal. Chem., 5 6 ( 6 ) ( 1 9 7 3 ) 1 4 3 7 . 2 9 G.A. Bennet, S . E . Megalla and O.L. Shotwell, J. Amer. Oil Chem SOC., 6 1 ( 9 ) ( 1 9 8 4 ) 1 4 4 9 . 3 0 C. Wilkein, W. Battes, I. Mehlitz, R. Tiebach and R. Weber, 2 . Lebensm Unters Forsch, 1 8 0 ( 1 9 8 5 ) 4 6 9 . 31 H. Kamimura, M. Nishijima, K. Yasuda, K. Saito, A.
Ilbe, T. Nagayama, H. Ushiyama and Y. Naoi, J. Assoc Off. Anal. Chem., 6 4 ( 5 ) ( 1 9 8 1 ) 1 0 6 7 . 3 2 M.K.L. Bicking, R.N. Knisley and H.J. Svec, J. ASSOC Off. Anal. Chem., 6 6 ( 4 ) ( 1 9 8 3 ) 9 0 5 . 3 3 C.J. Mirocha, B. Schauerhamer and S.V. Pathre, J. Assoc Off. Anal. Chem., 5 7 ( 5 ) ( 1 9 7 4 ) 1 1 0 4 . 3 4 L.K. Jackson and A. Ciegler, Appl. Environ Microbiol.,
3 6 ( 3 ) ( 1 9 7 8 ) 408. 3 5 P.M. Scott and B.P.C. Kennedy, J. Agric Food Chem., 2 4 ( 4 ) (1976) 865. 3 6 B. le Tutour, A. Tantoui-Elaraki and A. Aboussalin, J. Assoc Off. Anal. Chem., 6 7 ( 3 ) ( 1 9 8 4 ) 6 1 1 . 3 1 W.J. Hurst, K.P. Snyder and R.A.Martin, Peanut Sci., 11 (1984) 21. 3 8 D. Tosch, A.E. Waltking and J.S. Schlesier, J. ASSOC Off. Anal. Chem., 6 7 ( 1 9 8 4 ) 3 3 7 . 3 9 J.D. McKinney, J. Amer. Oil Chem. SOC., 5 8 ( 1 9 8 1 ) 9 3 5 A . 4 0 J.E. Hutchins and W.M. Hagler, J. Assoc Off. Anal. Chem., 6 6 ( 1 9 8 3 ) 1 4 5 8 . 41 J.E. Thean, D.R. Lorenz, D.M. Wilson, K. Rodgers and R.C. Gueldner, J. Assoc Off. Anal. Chem., 6 3 ( 1 9 8 0 ) 631. 4 2 K. Jewers, A.E. John and G. Blunden, Chromatographia, 27(11/12) (1989) 617. 4 3 N. Takeda, J.Chromat., 2 8 8 ( 1 9 8 4 ) 4 8 4 . 4 4 G. Quan and G.C. Yang, J. Agric Food Chem., 3 2 ( 1 9 8 4 ) 1071. 4 5 D.L. Orti, R.H. Hill, J.A. Liddle and L.L. Neelham, J. Anal Toxicol., 1 0 ( 1 9 8 4 ) 9 7 3 . 4 6 K.I. Tomlins, K. Jewers and R.D. Coker, Chromatographia, 2 7 ( 1 9 8 9 ) 6 7 . 4 7 N. Bradburn, K. Jewers, B.D. Jones and K.I. Tomlins, Chromatographia, 2 8 ( 1 9 8 9 ) 5 4 1 . 4 8 N. Bradburn, R.D. Coker and K. Jewers, Chromatographia, 2 9 ( 1 9 9 0 ) 1 7 7 . 4 9 M.P.K. Dell, S.J. Haswell, O.G. Roch, R.D. Coker, V.F.P. Medlock and K.I. Tomlins, Analyst, 1 1 5 ( 1 9 9 0 ) 1435. 5 0 Official Methods of Analysis, 11th ed., Association of Official Analytical Chemists, Washington D.C., 1 9 7 0 , section 2 6 . 0 3 5 , p . 4 3 2 .
34 5 1 W.A. Pons and L.A. Goldblatt, J. Amer. Oil Chem. SOC., 4 2 ( 6 ) ( 1 9 6 5 ) 4 7 1 . 5 2 L.A. Gifford, C. Wright and J. Gilbert, Food Additives and Contaminants, 7 ( 6 ) ( 1 9 9 0 ) 8 2 9 . 5 3 M.J. Shepherd, M. Holmes and J. Gilbert, J. Chromat., 354 ( 1 9 8 6 ) 305. 5 4 N. Takeda, J. Chromatogr., 2 8 8 ( 1 9 8 4 ) 4 8 4 . 55 V. Betina, J. Chromatogr., 3 3 4 ( 1 9 8 5 ) 2 1 1 . 5 6 V. Betina, J. Chromatogr., 4 7 7 ( 1 9 8 9 ) 1 8 7 . 5 7 R.E. Kaiser, in A. Zlatis and R.E Kaiser (Editors),
HPTLC-High Performance Thin Layer Chromatography, Journal of Chromatography Library Vo1.9, Elsevier Scientific Publishing Company, Amsterdam, 1 9 7 7 , p. 7 3 . 5 8 J. Blome, in A. Zlatis and R.E Kaiser (Editors), HPTLC-High Performance Thin Layer Chromatography, Journal of Chromatography Library Vo1.9, Elsevier Scietific Publishing Company, Amsterdam, 1 9 7 7 , p. 3 9 and 51. 5 9 H.J. Issaq, J. Liq. Chromatogr., 3 ( 6 ) ( 1 9 8 0 ) 7 8 9 . 6 0 N. Bradburn, Natural Resources Institute, personal
communication.
6 1 K. Randerath, Chromatographie sur Couches Minces, Gauthier-Villars, Paris, 2nd ed. ( 1 9 7 1 ) 7 2 . 6 2 R.K. Gilpin and W.R. Sisco, J.Chromatogr., 1 2 4 ( 1 9 7 6 ) 257. 6 3 H.M. Stahr, M. Domoto, Bei Lei Zhu and R.Pfeiffer, Mycotoxin Research, 1 ( 1 9 8 5 ) 3 1 . 6 4 A.A. Frohlich, R.R. Marquardt and A. Bernatsky, J. Assoc Off. Anal. Chem., 7 1 ( 5 ) ( 1 9 8 8 ) 9 4 9 . 6 5 D. Abramson, T. Thorsteinson and D..Forest, Arch. Environ. Contam. Toxicol., 1 8 ( 1 9 8 9 ) 3 2 7 . 6 6 P.A. Biondi, L. Gavazzi, G. Ferrari, G . Maffeo and C. Secchi, J. of High Res. Chrom.& Chrom. Comms., 3 ( 1 9 8 0 ) 92. 6 7 V. Medlock, A.P. Dutta, Natural Resources Institute,
in
preparation. Hara, R.E. Kaiser and A. Zlatis, Instrumental HPTLC, Dr Alfred Huthig Verlag, Heidelberg, 1 9 8 0 . 6 9 R.D. Coker, K. Jewers, K.I. Tomlins and G. Blunden, Chromatographia, 2 5 ( 1 9 8 8 ) 8 7 5 . 7 0 S. Ganguli, Natural Resources Institute, unpublished material. 7 1 H.E. Hauck and W. Jost, J.Chromatogr., 2 6 2 ( 1 9 8 3 ) 1 1 3 . 7 2 S z . Nyiredi, C.A.J. Erdelmeier and 0. Sticher in R.E. Kaiser (Editor), Planar Chromatograhy, Vol. 1, Dr Alfred Huthig Verlag, Heidelberg, 1 9 8 6 , p. 1 1 9 . 7 3 Available from Analtech, Delaware, U S A. 7 4 S.A. Patwardhan, R.C. Sukh Dev Pandey and G.S. Pendse, Phytochemistry, 1 3 ( 1 9 7 4 ) 1 9 8 5 . 7 5 A. Probsl and Ch. Tamm, Helv. Chim. Acta., 6 5 ( 1 9 8 2 ) 6 8 W. Bertsch, S .
1543. 7 6 M. Binder and Ch. Tamm, Helv. Chim. Acta., 5 6 ( 1 9 7 3 ) 2387. 7 7 Y. Ueno in I.F.H. Purchase (Editor), Mycotoxins in Human Health, The Macmillan Press Ltd., London, 1 9 7 1 , p. 115.
35 7 8 R.J. Cole, J.W. Dorner, J.A. Lansden, R.H. Cox, C.
Pape, B. Cunfer, S.S. Nicholson and D.M. Bedel, J. Agric. Food Chem., 25 ( 1 9 7 7 ) 1 1 9 7 . 7 9 S.V. Pathre, C.J. Mirocha and S.W. Fenton, J. Assoc. Off. Anal. Chem., 62 ( 1 9 7 9 ) 1 2 6 8 . 8 0 A.E. de Jesus, P.S. Steyn, R. Vleggaar and P.L. Wessels, J. Chem. SOC., Perkin Trans., 2 ( 1 9 8 0 ) 5 2 . 8 1 C.O. Emeh and E.H. Marth, Arch. Microbiol., 1 1 5
(1977) 157. 8 2 D.F. Jones, R.H. Moore and G.C. Crawley, J. Chem. SOC. C., ( 1 9 7 0 ) 1 5 7 . 8 3 H. Minato, M. Matsumoto and T. Katayama, Annu. Rep. Shionogi. Res. Lab., 23 ( 1 9 7 3 ) 4. 84 M. Binder and Ch. Tamm, Helv. Chim. Acta., 5 6 ( 1 9 3 ) 966. 8 5 R.T. Gallagher, G.C.M. Latch and R.G. Keogh, Appl Environ. Microbiol., 39 ( 1 9 8 0 ) 2 7 2 . 8 6 P.J. Curtis, D. Greatbanks, B. Hesp, A.F. Cameron and A.A. Freen, J. Chem. SOC. Perkin. Trans., 1 ( 1 9 7 7 ) 1 8 0 . 8 7 P.M. Scott, B.P.C. Kennedy, J. Harwig and B.J. Blanchfield, Appl. Environ. Microbiol., 33 ( 1 9 7 7 ) 249. 8 8 K.H. Ling, H.H. Liou, C.M. Yang and C.K. Yang, Appl. Environ. Microbiol., 47 ( 1 9 8 4 ) 9 8 . 8 9 A.E. John, PhD Thesis 1990, CNAA, Portsmouth
Polytechnic.
9 0 R.D. Coker, B.D. Jones, M.J. Nagler, G.A. Gilman, A.J.
Wallbridge and S. Panigrahi, Natural Resources Institute, Mycotoxin Training Manual, 1 9 8 4 . 9 1 Official Methods of Analysis, 1 4 t h ed., Association of Official Analytical Chemists, Washington D.C., 1 9 8 4 , section 2 6 . 1 3 8 . 9 2 S. Nawaz, R.D. Coker and S.J. Haswell, Analyst, 1 1 7
(1992) 67. 9 3 N. Bradburn, Natural Resources Institute, in
preparation.
9 4 V. Betina and Z. Barath, J. Antibiot., 1 7 ( 1 9 6 4 ) 1 2 7 . 9 5 Z . Durackova, V. Bekina and P. Nemec, J. Chromatogr., 116 ( 1 9 7 6 ) 155.
36
Chapter 3 TECHNIQUES OF LIQUID COLUMN CHROMATOGRAPHY PIRJO KURONEN INTRODUCTION Mycotoxins are a heterogeneous class of toxic substances produced by various species of several genera of filamentous fungi. The main producers of mycotoxins belong to the Aspergillus, Fusarium, and Penicillium genera. They play an important role in foodborne diseases of humans and animals, with toxic effects as variable as their composition. There is an increasing concern over food safety. In 1981, Cole and Cox (1) compiled the properties of over 270 mycotoxins from filamentous fungi. It is worth noting that in 1983 altogether about 3000 structurally characterized fungal metabolites were known in the literature ( 2 ) . Watson (3) in 1981 gave a figure of 432 toxic fungal metabolites. Mycotoxins represent a wide range of compound types, various polarities, chemical structures, and acid-base properties. The aflatoxins have been and still are the most important group of mycotoxins because of their frequent contamination of foods and feeds and their extremely severe toxicological effects in animals and man. Therefore they occupy an important position in mycotoxin research, followed in importance by the trichothecenes which now number over sixty (4-6). The commodities with the most mycotoxin problems are maize (corn) and groundnuts (peanuts). This chapter will present the methodology of column liquid chromatography covering sample pretreatment, classical column chromatography, mini-column chromatography, and high-performance liquid chromatography (HPLC), with the main emphasis on HPLC. This is now the most widely used chromatographic method in the analytical laboratory. HPLC allows ultratrace analysis of a wide variety of compounds. Many analytes can be analyzed at a nanogram level, but detection limits of a picogram or even less have been demonstrated in certain cases using special techniques. The 3.1
31
applications of the column liquid chromatographic methods will be presented in Chapter 8. SAMPLE PRETREATMENT The aim of the sampling and sample preparation procedures is to produce as representative a laboratory sample as possible (7). This sample is then analyzed using an analytical procedure which comprises extraction, clean-up, qualitative, quantitative, and confirmatory steps (see Fig. 3.1). 3.2
I
SAMPLE PREPARATION
1
I
I
CHRCNATGRAPHIC
rnYS1S
Fig. 3.1. Flow diagram showing mycotoxin determination.
the
analytical
sequence
for
In the first step of chemical analysis sequence mycotoxins are removed from the food or feed sample by some means of extraction - either by blending or mechanical shaking with solvents such as methanol, acetonitrile, ethyl acetate, acetone, chloroform, dichloromethane, or water, used either singly or as mixtures. This step is illustrated more precisely in Chapter 1. The second step of chemical analysis includes the clean-up of the crude sample extract to remove lipids and other co-extractives which interfere with the detection of mycotoxins at trace-level concentrations and which may contaminate and damage the analytical chromatographic columns. Sample clean-up is normally by far the most time-consuming stage in the analysis sequence of mycotoxins.
38
The choice of sample clean-up method(s) depends critically on the type of mycotoxin(s) and matrix in question, the expected concentration of the mycotoxin(s), and the available final analytical method used for the detection and determination. Generally, analysis of agricultural commodities necessitates extensive clean-up before the final analysis is feasible. The diverse chemical compositions of agricultural commodities have hindered the development of any single method that can be uniformly applied to all products, and therefore a variety of methods is used for this purpose. The techniques used include liquid-liquid partitioning, chemical adsorption, dialysis, and various forms of chromatographic clean-up procedures. Column chromatographic methods are widely used as clean-up procedures using different column packing materials. These methods include classical open column chromatography (Section 3 . 3 ) , chromatography using small disposable pre-packed cartridges (e.g. Sep-Pak, Baker, Supelclean, and Bond Elut), and preparative HPLC (Section 3.5). Solid-phase extraction clean-up. Solid-phase extraction (7-9) has been one of the fastest growing sample (SPE) pretreatment methods. It is a more rapid, efficient, reproducible, and safer method than the traditional liquid-liquid extraction techniques and offers, in addition, a wider range of selectivity. A wide selection of different chromatographic sorbents makes it possible to utilize several extraction mechanisms, thus allowing mycotoxins to be extracted from complex sample matrices. The adsorbents in SPE cartridge columns are chemically similar to the column packing materials used in HPLC (Section 3 . 5 ) , but have a larger particle diameter (40 pm), which facilitates the sample handling process. Silica gel, silica-based non-polar and polar bonded phases , ion-exchange, and size-exclusion phases are available, packed into disposable polyethylene columns. Silica gel-filled cartridges and silica-based bonded-phase packings dominate the market, but polymeric-based cartridges ( 9 ) , which withstand a wider range of solvents and pH values, have become available, too. The standard cartridge sizes are 100 or 400 mg but, in addition, cartridges containing larger amounts of variety of stationary phases (up to 10 g ) are also available. A vacuum manifold is used to facilitate
39
rapid sample handling and solvent elution of the retained compounds. With the 400-mg column, sample sizes as large as 100 ml may be used, but most of the applications reported have used much smaller sample volumes. SPE contains four extraction steps which are illustrated in Fig. 3.2. First, the SPE column is prepared to receive a sample, using a proper solvent. Secondly, the sample extract is applied to the cartridge in a weak solvent, which results in a strong retention of the compounds of interest in the column. Thirdly, the column is washed with a solvent that elutes the less strongly retained sample components: the compounds of interest are finally eluted selectively in a small volume of stronger solvent (ideally about 0.6 ml per 100 mg adsorbent). Step 1
Step 2
Step 3
Step 4
Conditioning
Sample Application
Washing
Elution
Fig. 3.2. Solid-Phase Extraction steps. The SPE process can be used in three different modes: (1) sample clean-up, ( 2 ) sample concentration, and (3) matrix removal. In sample clean-up mode a SPE column retains the mycotoxin and allows impurities to pass through the column. In sample concentration mode large sample volumes are passed through the column and the retained mycotoxin is concentrated by eluting with a small volume of solvent. In the matrix removal mode the SPE column is used to retain interfering impurities and the mycotoxin is allowed to pass through the column. Clean-up and preconcentration procedures for mycotoxin-containing samples, including the use of silica and bonded-phase cartridges, have been reported in numerous references
40
(10-20). Hoke et al. (21) used SPE columns to concentrate mycotoxins (T-2, HT-2, DAS, and DON) so that the assay could respond to low aqueous mycotoxin concentrations after only 24 hours of exposure. Orti (16) has developed a multi-mycotoxin clean-up method which makes it possible to estimate simultaneously aflatoxin B1, ochratoxin A (OCH A), and citrinin (CIT) in human urine, using a sequence of different clean-up cartridges. The SPE technique can nowadays be automated with dedicated instruments available from a number of companies (9). Charcoal/Alumina Clean-up Column. The small charcoal/alumina clean-up columns (No. 213 and No. 215, available from Romer Labs) presented in 1981 by Romer et al. (22) have been effectively used to remove interfering materials from grain, feed, and food extracts prior to the final chromatographic determination of trichothecene mycotoxins (23-25). 3.3
COLUMN CHROMATOGRAPHY 3.3.1 Introduction Classical column chromatographic methods (5,26) have been used widely as clean-up procedures in trace analysis of mycotoxins, although today commercial pre-packed disposable cartridges (Section 3.2) are increasingly replacing these methods. Different modes of chromatography, such as adsorption, partition, ion-exchange, size-exclusion, and affinity Chromatography, have been used. The most widely used of these methods is, however, adsorption chromatography, on which in this section will be mainly focused . 3.3.2 Procedure The sample extract dissolved in a small volume of an appropriate solvent is added slowly and evenly to the top of the column after which the column is washed with one or more solvents in which the mycotoxins are insoluble or at least less soluble than the impurities. Thereafter the solvent composition is The changed so that the mycotoxins are eluted from the column. eluate is collected and concentrated, and the residue is redissolved in a small volume of solvent prior to the final chromatographic analysis or prior to further isolation by preparative HPLC for mass spectrometric confirmation of identity (27,28).
41
3.3.2.1 Stationary phase Silica gel is the most widely used adsorbent for classical column chromatographic clean-up of mycotoxins (29). Other adsorbents used include alumina, charcoal, cellulose powder, magnesium silicate (Florisil), diatomaceous earth (Celite), and macroreticular resin. Reference 29 includes silica gel column chromatographic clean-up methods for the aflatoxins in different commodities. Silica gel clean-up columns have also been used for patulin (PAT) in apple juice, zearalenone (ZEA) in maize, and sterigmatocystin (STE) in wheat and barley. Other column materials have been employed in some mycotoxin methods, but they have not found so widespread use as silica gel. The normal particle size range in traditional gravity column chromatography is 63-200 pm. In flash chromatography (30), where a slight gas (e.g. air, nitrogen) over-pressure is used to increase sample throughput, the optimum particle size is 40-63 pm. Smaller particles increase resistance to solvent flow, resulting in the need for pressurized systems. In addition to average particle size and particle size distribution, the other physical and chemical properties such as specific surface area, average diameter, pore diameter distribution, pore volume, packing density, pH, trace metal content, and activity, are important for adsorption chromatography. The specific surface area is particularly important because sample capacity is proportional to the total surface of the silica gel packed into the column. The activity of the adsorbent is also a very important consideration. It can be controlled by the deliberate addition of a known amount of water to the dried adsorbent, usually in the range of 2-10% water by weight. The bonded-phase silicas (31) are also available for flash chromatography, which has the advantages over traditional column chromatography including moderate resolution, rapid separation times, and reduced band broadening. 3.3.2.2 Solvent system A preliminary screening by TLC is used to establish the optimum mobile phase system for use in adsorption chromatography. A solvent composition is chosen that gives good separation and moves the toxins of interest to an RF value of approximately 0.3 Mixtures of methanol and chloroform (3:97) have to 0.4. frequently been used as mobile phase for aflatoxins and ZEA in
42
several matrices: 5-10% acetone in dichloromethane has also been a good solvent composition for aflatoxins and STE in many matrices. Good results have been obtained on a column for trichothecenes using a solvent mixture of about 5 % methanol in dichloromethane. For acidic mycotoxins such as OCH A, CIT, etc., an acidic mobile phase (e.g. acetic acid-benzene, 1:9) has been used. Mixtures of ethyl acetate or chloroform with hexane have often been used for column chromatographic separations of the less polar mycotoxins. 3 . 3 . 2 . 3 Column packing techniques Classical column chromatography requires only a glass column and a suitable packing material: in addition, flash chromatography requires a flow controller valve. Sometimes a solvent reservoir is added to the top of the column to contain a larger volume of elution solvent. Figure 3.3 presents typical column configurations for classical column chromatography, and flash chromatography. BleedPort
Needle valve
pressure
Flow controller
8
P Fig. 3.3. Typical all-glass equipment used for classical column chromatography and flash chromatography.
43
Depending on the amount of material to be purified, a column of appropriate diameter (usually 2 0 mm i.d.) is chosen. The adsorbent is supported by a glass frit or preferably a plug of glass wool placed at the bottom of the column (see Fig 3.3). Different methods can be used for column filling. The column can be filled to about 60-70% of its height with the solvent to be used in the separation. Thereafter adsorbent is added to the column in small increments through a filter funnel. The solvent is allowed to run out from the column at a rate not exceeding the addition rate of the adsorbent. This packing method is very time-consuming and may produce difficulties when large columns have to be packed. Alternatively, columns are dry-packed with gentle tapping of the side of the column, or slurry-packed by preparing a slurry of the adsorbent and the required mobile phase which is then poured into the column, and let to settle with the tap open. The column can be vibrated until the stationary phase has completely settled. In each case the column bed should be homogeneous and free of channels. If the column is not properly packed, the channels may result in irregular flow, leading to much band broadening, and the distorted bands are easily observed if coloured substances are chromatographed. In flash chromatography, the column is dry-packed with 15 cm of dry silica gel or bonded-phase silica. The column is then filled with solvent, and gas pressure ( - 2 0 psi) is used to push the air rapidly from the column (31). In many cases it is recommended that a 0.5-1.0 cm layer of anhydrous sodium sulphate is added to the top of the column to protect the adsorbent from water traces in the sample extract. 3.3.2.4 Fractionation and detection The solvent used to pack the column is drained off until it is just over the column bed. Next, the sample, dissolved preferably in a small volume of the mobile phase, is applied slowly and evenly with a pipette on the top of the column bed, and the column is refilled with the solvent. The sample band should be sharp at the top of the column. Some samples may not dissolve, however, in the mobile phase. In this case the sample is usually dissolved in a more polar solvent (solvent of greater elution strength in adsorption chromatography) that may affect column equilibration and may decrease resolution. In this case, the
44
sample dissolved in a more polar solvent is added to a small amount of column packing material and the solvent is removed. This is then packed on the top of the column bed. Alternatively, the sample, dissolved in a mobile phase, may be filtered through an adsorbent cake supported on a Buchner funnel, when the insoluble part of the sample remains on the adsorbent and the components of interest are eluted with the solvent. Excess of solvent is removed and the sample is now applied to the top of the column. If a less-polar sample solvent is used in adsorption chromatography, the sample is concentrated at the head of the column before elution begins. Suitable-sized fractions are collected either manually or with a fraction collector. Elution occurs in classical column chromatography at a flow rate of 0.5-5.0 ml/min depending on column dimensions. In flash chromatography, a single development with about 4 to 5 column volumes of the mobile phase is usually sufficient for simple separations. Elution occurs typically at a flow rate of 5 ml/min which is adjusted with a flow controller valve (see Fig. 3.3). Thus a typical flash chromatographic separation occurs very quickly, in 5-10 min. Elution of the mycotoxins can be monitored by a detuned UV or a refractive index detector. The components can also be determined at the end of the separation by spotting each collected fraction on a TLC plate using the techniques described in Chapter 2.
3.4
MINI-COLUMN CHROMATOGRAPHY One of the most widely used screening methods for certain mycotoxins (usually aflatoxins, OCH A, and ZEA) in contaminated samples before the examination by other analytical techniques has been mini-column chromatography. This technique, using the glass mini-column with an internal diameter of ca. 5 mm, is a special design of a classical column chromatography. Mini-column screening methods (32-43) include the following steps: extraction, purification of the extracts, concentration, and development on a mini-column for detection under UV light. These methods are rapid, simple, and require only little expertise and no sophisticated equipment. The mini-column methods are particularly useful for field analysis. The methods have limitations as well.
45
They are semi-quantitative, having a higher detection limit and less separation power, selectivity, and sensitivity than is obta ned by using HPLC. 3.4.1 Procedure The first mini-column method was introduced by Holaday (32) for detection of aflatoxins in peanuts. The column contained a 45-mm high silica gel layer between glass wool plugs. This column was dipped in the sample extract which drew up by capillary force. The sample components moved upwards in different extents with ascending solvent. After about 10-15 min the column was examined under UV light (365 nm). A blue fluorescent zone characteristic to aflatoxin was visible at the top of the column. Since then a number of refinements and improvements [e.g. by changing solvents (33,35,42) or the composition of the packing material] have been made in the Holaday method. In the method of Romer (36) descending chromatography with a mixture of chloroform and acetone is applied. In this method, packing of a mini-column contains successive zones of alumina, silica gel, and Florisil with calcium sulphate at both ends of the column and the packing materials are held in place by glass wool (Fig. 3.4). Calcium sulphate is a drier and the silica gel and neutral alumina perform clean-up functions.
-
Glass wool Calcium subhate
-
Calcium sulphate Glass wool
Fig. 3.4. Diagram of the Romer mini-column (36). The aflatoxins appear as a tight band at the top of
the
Florisil
46
layer, where they can be detected under UV light. By comparison of a sample column with a reference column containing a known amount of aflatoxin, it is possible to determine whether the sample column contains more or less aflatoxin than the standard column. of Romer was subjected to a The mini-column method collaborative study (36) and approved by the AOAC (44) as an official method for the detection of aflatoxins in the listed commodities (mixed feeds, corn, almonds, peanuts, peanut butter, pistachio nuts, cottonseed). The method does not distinguish between the different aflatoxins. Similar mini-column procedures to those for aflatoxins have been developed for some other fluorescing mycotoxins such as OCH A (37) and ZEA (39) in several commodities. The detection limits vary from 5-20 ng/g. 3.4.2 Illustrative example The mini-column screening method of Romer (36) for the detection of aflatoxins in a wide range of commodities contains the following steps. A 50-9 sample of a blended, ground commodity is extracted in a blender 3 min with acetone-water (85:15, v/v) and filtered. A 0.2 M sodium hydroxide and ferric chloride slurry is added to an aliquot of the filtrate and mixed. Thereafter basic cupric carbonate is added, mixed, and filtered. The acidified filtrate is extracted with chloroform. The chloroform phase is washed with 0.02 M sodium hydroxide. An aliquot of chloroform extract is transferred to a mini-column packed with calcium sulphate, Florisil, silica gel, neutral alumina, and calcium sulphate (see Fig. 3.4). The column is then developed with chloroform-acetone mixture (9:1, v/v) after which the column is viewed under a longwave UV light (365 nm). If aflatoxin is present, it can be seen as a blue or bluish-green fluorescent zone at the top of the Florisil layer. The mini-columns developed with an extract of aflatoxin-free commodity, and of aflatoxin-free commodity containing a known amount of aflatoxin are compared, and the aflatoxin level in the commodity can be estimated. 3.5
HIGH-PERFORMANCE LIQUID CHROMATOGRAPHY (HPLC) 3.5.1 Introduction HPLC became available for the analysis of foodstuffs nearly twenty years ago. The first published HPLC application f o r
41
mycotoxin research dates from 1973 (45). Since then, the trend has been towards increased use of HPLC for the ultimate separation, detection, and quantification of the mycotoxins in foods, after sufficient clean-up to remove interferences that could give rise to false positives. The important advantages of HPLC are its ability to handle thermally labile, poorly volatile, non-volatile, polar, and ionic compounds. The high resolving power between chemically similar compounds, the speed, increased sensitivity, accuracy and precision of the method, and the variety of detection systems now available make modern HPLC a more suitable technique than other liquid chromatographic techniques. In addition, HPLC is a quantitative technique and is suited for on-line clean-up of crude sample extracts, and finally, it can be automated quite easily. HPLC has limitations as well. The cost of modern HPLC equipment is quite high, and wide experience is necessary for obtaining the best possible benefit from an HPLC system. It must be pointed out that TLC, in its most sophisticated form, is as expensive as a simple HPLC system. There is still no sensitive universal or ideal detector for HPLC. In certain areas sensitivity in HPLC is not of course a problem, but in mycotoxin analysis sensitivity may be a serious limitation. For example, trichothecenes have been analyzed to a limited extent by conventional HPLC; sensitivity is limited, because most of them have weak or only end-absorption in the UV range. The development of efficient and reliable interfaces allowing HPLC to be coupled on-line with mass spectrometry (MS) will do much to overcome a sensitivity problem. The reviews devoted to the HPLC of mycotoxins have been written by Scott (46), Shepherd (47), and Coker and Jones (7). This section will present the instrumentation and practice of the HPLC method in mycotoxin analysis. The basic theory and practical considerations of HPLC will not be covered in detail in this context, because several good books have appeared, and the reader is directed to these. For example, the general principles (48-50), practical HPLC (51), recent progress (52-53), method development (54) and optimization ( 5 5 , 5 6 ) , and troubleshooting ( 5 7 ) in HPLC have been presented in several books mentioned.
48
3.5.2 Instrumentation and practice Requirements for the basic components of HPLC systems suitable for the analysis of mycotoxins are discussed. In addition, the areas where development has recently taken place are pointed out. There is a wide range of modes of chromatography in HPLC that can be employed, making possible the chromatography of many different compound classes. In most mycotoxin analyses, however, it will be profitable to use one of the three primary HPLC methods: reversed-phase (RP), ion-pair (IP), or normal-phase (NP) chromatography. RP-HPLC is nowadays the most commonly used method, and it is a potential technique for multimycotoxin
analysis under gradient elution conditions ( 5 8 - 6 4 ) . 3.5.2.1 Instrumentation Modern liquid chromatographs, which vary widely in sophistication, can be assembled from modular units designed to work independently of each other, or purchased as a single, integrated unit. Each system has its own advantages and drawbacks. Integrated HPLC systems have the advantage of being easier to automate but the drawback of being less flexible. Microcolumn HPLC (micro-HPLC) ( 6 5 , 6 6 ) , which employs columns with internal diameter < 1.0 mm and requires miniaturization of the whole system (pumping system, injector, gradient former, detectors etc.), is left outside this discussion. Micro-HPLC still faces many problems, but promises to be a significant technique in the future because of its low solvent consumption, simpler interfacing with mass spectrometers, lower detection limits when using concentration-dependent detectors (e.g. UV-vis and fluorescence detectors). Solvent reservoir. The solvent reservoir must hold a volume of solvent adequate for repeated analyses and it must be inert with respect to the solvents used, varying from aqueous buffers to hydrocarbons, depending on the mode of chromatography. Air bubbles are the major cause of problems with LC pumps. Problems with dissolved air are usually encountered with protic solvents such as water and alcohols. The best way to avoid bubbles is to thoroughly degas the solvents by heating, application of vacuum or ultrasound, or by sparging with helium. The latter is the most effective and convenient degassing method (65). After initial (a few minutes) vigorous bubbling of helium
49
through the solvents the helium flow is reduced to a trickle during the use of the LC system. Pumping systems. The pump is one of the most important components in HPLC equipment, because its performance directly affects the reproducibility of retention times, quantitative analysis, and detector sensitivity. Various pumps are available, but nowadays reciprocating piston pumps provide the basis f o r most pumping systems. Many pump problems (seal leakage, check valve failure, air bubbles etc.) can be avoided by using appropriate solvents, filtering the solvents and samples, degassing the solvents, and carefully flushing out the buffers. The pump must produce a wide range of flow rates for solvent delivery suitable for the various HPLC modes. F o r analytical and semi-preparative applications, where columns are 10-25 cm long with 1.0 to 10 mm internal diameter (i.d.) and packed with 3-10 pm particles, most modern pumps are able to produce flow rates from as low as 0.01-0.05 up to 5.0-10 ml/min. Although most HPLC pumps operate at pressures of 300-400 bar, the operating pressures should be less than 50% of the maximum capability of the pump (typically < 100-150 bar). Injection devices. Sample introduction is one of the critical steps in HPLC: even the best column will produce a poor separation if injection is not carried out carefully. The most widely used method uses an injection valve (e.g. Rheodyne Model 7010, 7125, or 7410), allowing reproducible volumes to be injected and good quantitative analyses. The six-port valve is the basis of all sample injection valves. The sample is injected with a syringe into a loop (either external or internal) and the solvent flow from the pump is then diverted via the loop to push the sample into the column. Loops can be filled in two ways: complete filling and partial filling. When the complete filling injection technique is used the loop is filled entirely with sample. When the loop is completely filled with the mobile phase, a sample amount equal to the loop volume cannot displace the existing solvent but tends to mix with it. Thus the sample volume must be about five times the volume of the loop (e.g. 100 pl of sample f o r a 2 0 7 1 loop). This is important for quantitative analysis using an external standard method. With the partial loop injection technique, the loop can be partly filled with the sample following
50
the manufacturer's guide. The sample size should not exceed about 50-60% of the loop volume. Although the partial filling method is less precise than the complete filling method, it is necessary in cases where no sample should be lost. Sample size can be adjusted by changing the sample loop volume. For analytical work, the typical sample loop volume is 5, 10, 20, or 50 p l . For very small amounts of sample the special valve (e.g. Rheodyne Model 7413) with an internal loop capacity between 0.5 and 5 p1 can be used. For semi-preparative work sample loops of 1000-2000 pl are appropriate. Automated injection system (autoinjector) is generally an automated version of the six-port injection valve used for manual injections. With the change of precision syringes, the autoinjector offers a wide range of fully programmable injection volumes (from 0.1 to 250 pl). The 25-pl syringe provides the highest accuracy and precision, but even a 1-pl injection is possible from as little as 5 pl of total sample. For large volumes a 250-pl syringe can easily be substituted. The autoinjector can be combined with the autosampler permitting unattended injection of samples into the liquid chromatograph. Column hardware. Columns are available in numerous different configurations and from about a hundred different suppliers. The columns vary in dimensions. They are typically 5-25 cm long when micro-particulate stationary phases of 3-10 pm are used. Longer columns are not worth considering. Columns of internal diameter (i.d.) 1-5 mm are used for analytical purposes in conventional HPLC. Columns of i.d. 1-2 mm (sometimes called microbore columns) increase detectability and reduce solvent consumption. However, while useful for trace analysis when the sample amount is very limited, or for interfacing with detectors such as a mass spectrometer, microbore columns do not give better resolution than regular analytical columns of 4-5 nun i.d. Micro-LC columns (i.d. < 0.5 mm) meet some interest, but most commercial LC instruments are not suitable for the optimum use of such columns. Wider columns of i.d. 10-20 mm are used for semi-preparative work with milligram quantities of sample. Large-scale purifications are performed using preparative liquid chromatography equipment (e.g. Waters Prep LC-500) removing the need for scraping the bands from the preparative plates.
51
Most HPLC column blanks are made of stainless-steel tubes and have compression fittings of various design (e.g. Swagelock, Valco) and steel frits at either end of the column to retain the column packing material. Stainless-steel is resistant to the pressures in HPLC and relatively inert to chemical corrosion. Inert and corrosion resistant glass tubes of various constructions have special uses as column material. Frits made of porous glass, alumina/ceramic, or titanium are also available (66). Recently, less-expensive cartridge columns of metal, glass, or plastic have become popular, since they can be easily replaced, and in many cases only the replacement of the small guard column at the head of the column is necessary. In addition, the cartridge columns can be easily connected into longer lengths for a particular separation. A variation of the cartridge column system is based on radial compression. The radial compression columns, which have been available for many years from Waters Chromatography Division, are loosely packed in a polyethylene sleeve, which in use is placed in a column holder that compresses the column packing material to form a stable bed. Column temperature control. The role of temperature is usually neglected in HPLC, and most HPLC separations are carried out at ambient temperature without the aid of a column oven. Temperature control is nevertheless an important part of separation efficiency. Reproducible retention times and retention indexes (63,67) are possible only if the column temperature is constant. Furthermore, at elevated temperature the viscosity of the mobile phase is lower, which is important with high viscosity mobile phases like the aqueous mixtures used in RP work. In addition, the effect of temperature can be exploited as a means to alter selectivity, since temperature change has a different effect on the retention of compounds (63,67). Further, the column temperature, like the type and composition of the mobile phase, affects the spatial arrangement of the bonded groups of the stationary phase (68,69). The analysis of cyclosporin A with RP-HPLC has required the use of column temperatures as high as 70-80 O C (70), but NP-HPLC using cyano column has allowed reproducible determinations at column temperature of 53 OC (71). The detector signals the presence of sample components and
52
measures their concentration in the mobile phase by producing electric signals. These signals are then conveyed to the recorder and/or display where they are shown as a deviation from the baseline, proportional to the compound concentration. Electronic inteqrators are needed to measure the detector signal. Sample pretreatment equipments. A wide variety of very complex sample clean-up or pretreatment procedures prior to the final HPLC analysis of mycotoxins (see Sections 3.2 and 3.3) are mostly necessary. And these procedures are often laborious, time-consuming, and subject to error and may be a major source of the imprecision of the total assay. Therefore, more and more laboratories use automated robotic arms (e.g. Zymark) or sample pretreatment equipments (e.g. Varian AASP System) to facilitate extraction, clean-up, enrichment, and application of the sample to the HPLC column and to provide for more reliable analysis results (9,71,72). The automated sample pretreatment can be performed also by using column switching technique in which two or more columns in series are connected by a switching valve so that the on-line selective transfer of a fraction or fractions from one column to one or more secondary columns for further separation is possible (72, 73-75). For example, the determination of STE in feed has been performed by HPLC with column switching (76). Smith and Robinson (77) have described a semi-automated HPLC method using column switching for assay of cyclosporin. Computer. Except for instrument control and data acquisition and analysis, the computers may be used for method development (56,78). 3.5.2.2 Normal-phase HPLC When the stationary phase is more polar than the mobile phase, the HPLC mode is called normal-phase (NP) chromatography or often adsorption chromatography. NP-HPLC is carried out with inorganic solids such as silica or alumina and various polar bonded phases (e.g. cyano, amino, diol, nitro) which have been reviewed in many sources (79,80). The polar bonded phases, which are formed from silica particles by binding covalently various polar silanes to the silica surface, are slowly replacing the traditional silica and alumina as packing materials for use in NP-HPLC, although silica and alumina still find widespread use. Silica, for example, has the unique ability to separate isomers,
53 and it is widely used in preparative chromatography. The most commonly used polar bonded phases are aminoalkyl, cyanoalkyl, and 1,2-dihydroxypropyl propyl ether (diol) phases (Fig. 3.5) of which cyanoand amino-derivatized silicas are the most popular. Amino-phases are also weak anion-exchangers, and therefore careful control of pH is important for ionic compounds. The polar bonded phases can be listed in order of increasing polarity: cyano < diol < nitro < amino. In NP-HPLC the compounds are generally eluted in the order of increasing polarity of the compounds. NP-HPLC uses mixtures of organic solvents as mobile phase. Hexane (or pentane, heptane) is generally preferred as the weak solvent and dichloromethane, methyl t-butyl ether, ethyl acetate, or acetonitrile are used as the stronger ones. The strength of the mobile phase in NP-HPLC is increased by raising the proportion of the more polar component in the mixture. Snyder et al. (81) have presented a useful table which shows solvent strength values for some useful organic solvent mixtures for NP-HPLC. The water content of the mobile phase must be carefully controlled to allow to be obtained reproducible results with silica ( 8 2 ) . In NP-HPLC sample-solvent interactions are relatively weak whereas sample-adsorbent or solvent-adsorbent interactions are strong, leading to a different classification of solvent selectivity as compared to RP-HPLC (83,84). The polar interaction between sample molecules and functional groups on the surface of NP packing plays a major role in determining separation selectivity. It must be pointed out that marked differences have been shown in the selectivity between different normal phases (84).
Earlier, NP-HPLC methods, mostly performed on silica columns, were used in mycotoxin analysis [e.g. aflatoxins (85-91), ZEA (92,93), PAT ( 9 4 ) ] , but nowadays RP-HPLC methods are favoured. NP-HPLC methods are desirable for samples dissolved in non-polar organic solvents (e.g. ether or chloroform extracts). If RP separation is used, however, less polar organic solvents should be evaporated to dryness, and the residue redissolved in an appropriate aqueous organic mixture. NP-HPLC (best with silica) is the first choice for preparative scale HPLC, too. 3.5.2.3 Reversed-phase HPLC In the technique of reversed-phase high-performance liquid
54
chromatography (RP-HPLC) the mobile phase is more polar than the stationary phase which is the opposite of NP-HPLC. The RP mode accounts for approximately 70-80% (possibly up to 9 0 % ) of all HPLC separations performed. This popularity depends on the simplicity, versatility, and broad applicability of the method. It has become the method of choice for the analysis of samples ranging from pharmaceutical and drug compounds to environmental pollutants and even large biological molecules. Detailed information on the RP-method may be found in several publications (48,54,55,57,95-97) . The RP techniques have found wide applications in the mycotoxin field, because mycotoxins are a highly miscellaneous set of compounds representing a wide range of polarities and chemical structures and different acid-base properties. RP methods have traditionally employed hydrocarbon-like stationary phases with polar aqueous organic mobile phases. The interaction between solute molecules and the stationary phase depends primarily on dispersion forces (non-specific hydrophobic interactions) and therefore compounds are separated according to their relative hydrophobicity, the most polar compounds being eluted first and the non-polar compounds being retained longer. Although great strides have been made in elucidating the retention mechanism in RP-HPLC, it is still only partially understood, because it is a complex function of the stationary phase, mobile phase, and solute. Stationary phase. The most commonly used stationary phases for RP separations have been and still are C-18 bonded phases, followed by C-8 and shorter n-alkyl, phenyl, or cyanopropyl bonded phases (Fig. 3 . 5 ) . Silica has been the most widely used base material for the aforementioned phases ( 9 8 - 1 0 0 ) . A variety of different procedures have been reported for the synthesis of chemically bonded silica-based packing materials during the last 20 years. Recently, several authors have exhaustively reviewed the preparation and characterization of bonded phases ( 1 0 1 - 1 0 3 ) . One great problem has been, and still is, the difficulty of ensuring reproducibility of the retention properties and selectivities from one commercial RP column to another, and even from one batch to another of the same product. These retention differences occur mainly for polar, particularly basic, compounds.
55
Methyl
Amlno
Hexyl
Q y 3 SI-O-SI-(CH2)nNH2 AH3
Octyl
Q y 3 SI-O-SI-(CH2)7CH3 AH3
+
NIt rl le
'
CH3
d-0-d I-(C
H2) ,C-N
AH3
Octadecyl
Fig. 3.5. The most common silica-bonded stationary phases. Cationic bases can interact quite strongly with silanols by hydrogen-bonding or with ionized silanols by ion-exchange (see The numerous variables involved in the preparation of Fig. 3.6). the RP, starting from the silica itself and ending with the column packing process, explain the great variations between commercial columns from different manufacturers. A standard silica as starting material, standard bonding reaction conditions, a standard procedure to characterize the phase, and standard column packing and testing procedures would be the prerequisites for a high batch-to-batch and column-to-column reproducibility. Another problem of commercially available RP stationary phases is their stability. With silica-based packing it is possible to use mobile phases with a pH between 2 and 8, because silica is soluble at high pH and the Si-C bond binding the hydrocarbon chain to the silica becomes labile at very low pH values. The more pH stable RP stationary phases (from pH 1 to 13) can be made from polymeric resin, but some of these phases may swell or shrink in contact with organic solvents. The pH stability of silica-based RP phase can be enhanced by substituting two bulky sterically protective groups (e.g. isopropyl or t-butyl) for the dimethyl groups on the silicon atom of silane (104).
56
Walters (105) has classified C-18 columns on the basis of two predominant RP retention mechanisms (hydrophobic and silanophilic interactions). This classification scheme will assist in selecting columns with similar performance from among the large number of C-18 brands on the market.
R
I+ I
n
R-N--R
I+
R-N;-R
s
'
,I
7
bH
Fig. 3.6. Interactions of cationic bases with alkyl bonded stationary phase: (a) hydrophobic interactions, (b) ion-exchange, (c) hydrogen bonding. Mobile phase. The properties of some LC solvents are listed in Table 3.1. It is generally accepted that retention in RP-HPLC is mainly controlled by the mobile phase, with the stationary phase playing the secondary role. Optimum selectivity is usually achieved by finding the right composition for the mobile phase. The preferred organic solvents for RP-HPLC are methanol (MeOH), acetonitrile (ACN), and tetrahydrofurane (THF), used in binary, ternary, o r sometimes in quaternary combinations with water. Organic solvents are strong and water is a weak solvent. Solvent strength (= chromatographic elution power) and selectivity are the properties of greatest chromatographic interest. In RP-HPLC solvent strength increases with the decrease in polarity. A change from methanol to acetonitrile or THF can result .in significant selectivity changes for various sample solutes. The Snyder triangle (48,106,107) is a widely accepted aid for characterizing solvent selectivity. Snyder has described a scheme
for classifying common LC solvents according to their polarity or chromatographic strength ( P I values) and according to their relative ability to engage in proton acceptor, proton donor and strong dipole interactions ( = selectivity). Thus solvents having similar functionalities tend to fall within the same selectivity group (see Table 3.1) and should have similar selectivity, while solvents from different groups should exhibit different selectivity for a given separation. However, several discrepancies in the triangle approach have been observed for the experimentally determined selectivities of some solvents (108-110). TABLE 3.1 Selected properties of some LC solvents (48,106,107)
SC
solvent
eoa
-n-Pentane -n-Hexane i-octane
0.00
pfb
0.01 0.01 i-propyl ether 0.28 Ethyl ether 0.38 CNorofonn 0.40 Dichloranethane 0.42 Tetrahydrofuran 0.45 Acetone 0.56 Dioxane 0.56 Mhyl acetate 0.58 Acetonitrile 0.65 n-Propanol 0.82 0.82 i-Prapanol
-
Ethanol Methanol Water
0.88 0.95
Very
large
( R P ~ )
0.0 0.1 0.1 2.4 2.8 4.1 3.1 4.0 5.1 4.8 4.4 5.8 4.0 3.9 4.3 5.1 10.2
-
-
4.4 3.4 3.5
3.1 -
4.2 3.6 2.6 0.0
Viscosity
Boiling W point Cutoff
(m~a,20 OC) (OC)
(mn)
0.23 0.33 0.50 0.37 0.24 0.57 0.44 0.46 0.32 1.54 0.45 0.37 2.30 2.30 1.20 0.60
195 190 200 220 205 245 230 220 330 220 260 190 210 210 210 205 490
1.00
36 66 99 68 34.5 61 40 66 56 101 77 82 97 82 78 65 100
Selectivity group
--
I I VIII V I11
VIa
VIa VIa
VIb
I1 I1 I1 I1 VIII
a Solvent strength parameter for liquid solid chrmtography (LSC) on & l d M (A1 0 )
solvent &?ty parameter calculated fran mhrscimeicierls ciata solvent strength weighting factor in W - H P X ;experimental value
A more precise solvent strength parameter ST has been defined for RP systems. ST for any solvent systems can be calculated from equation 1.
s T = I :i si@ i where
ST
is
the
total
solvent
strength
of
the
mixture, Si
58
(Table 3.1) is the solvent weighting factor, and #i is the volume fraction of solvent in the mixture. Approximately equal total solvent strengths will provide equal capacity factors (k') for different solvent mixtures in RP-HPLC. Other factors being equal, ACN has the following advantages over methanol: higher solvent strength, lower viscosity, and lower UV cut-off. Isocratic elution is useful only when toxins with similar retention behavior are to be studied, whereas gradient elution (111-113) is effective for the separation of samples containing compounds with a wide variety range of retention times. Clearly for screening or monitoring of several mycotoxins the only feasible approach is gradient elution (58-64), where great strides have been made in equipment, materials, and a better understanding of the technique. In addition, gradient elution is a valuable technique in concentrating the analyte into a narrow band for more sensitive detection, and gradient elution data can be applied for developing a final isocratic separation. Mobile phase additives. Non-ionic compounds can usually be chromatographed in RP-HPLC in the absence of mobile-phase additives (acids, buffers, ion-pairing reagents, or triethylamine). Ionic or ionizable compounds (e.g. moniliformin, ochratoxins, CIT) are chromatographed by RP-HPLC using one of the two techniques, ion-suppression and ion-pair chromatography ( I P C ) . In the former case the pH of the mobile phase is adjusted to suppress the ionization, which means about 1.0-2.0 pH units below and above the pKa value for an acid and a base, respectively, bearing in mind the pH stability of the stationary phase. It is worth noting that the degree of dissociation of acids and bases is highly solvent dependent. For example, the apparent pKa value of organic acids increases markedly with the organic solvent concentration of the mobile phase (114,115). The pH of the mobile phase also controls the ionization of acidic silanol groups in the RP packing. The pH adjustment usually is performed by using acetic acid (AcOH), phosphoric acid, trifluoroacetic acid (TFA), or different buffers (e.g. sodium or potassium phosphate, ammonium acetate) as mobile phase modifiers. Phosphoric acid is often preferred to acetic acid because of its non-aggressive behavior against the column and liquid chromatographic equipment (116) and a low UV cut-off value of 195 nm. TEA is sometimes
59
added as a silanol blocker to the mobile phase when basic compounds are to be separated. An acidic mobile phase is essential to ensure elutions of the acidic mycotoxins, e.g. OCH A and CIT (63). IPC is frequently, however, a more useful alternative for samples containing ionic or ionizable compounds, particularly if In this technique a the compounds are strong acids or bases. buffer and a so-called ion-pairing reagent is added to an aqueous organic mobile phase. A buffer controls the pH and ion-pairing reagent provides more retention and higher selectivity as compared to the chromatography without these additives. Negatively charged ion-pairing reagents [e.g. alkyl (usually C-5 to C-10) sulphonates] are used for the separation of protonated bases (cations), whereas cationic agents (e.g. tetrabutyl ammonium ion, TBA) are used for the ion-pair separation of carboxylate or other anions (117). For example, moniliformin (118), tenuazonic acid and 3-acetyl 5-substituted pyrrolidine-2,4-diones (119) have been analyzed using ion-pair chromatography. Reporting retention data. There is yet no standard method of reporting retention data in HPLC. The methods most in use today are retention times (t,) , retention volumes (V,) , and capacity factors (k') (Eqn 2), which are all strongly sensitive to variations in the chromatographic parameters. Relative retention expressions such as relative retention times (r) (Eqn 3) and relative capacity factors (r') (Eqn 4) have been used for some HPLC systems (120-122). Capacity factors (k') and relative capacity factors (r') suffer from the need for requiring measurement of the dead time (to), because there is no generally accepted method among numerous suggestions (123-130) for measuring this parameter. In addition, the relative methods involving comparison with an appropriate internal standard, depend on agreement among laboratories which standard to select. As a result the development of retention data libraries for comparison and identification purposes has not proceeded very far. k'
=
(tR
- to)/to
r = tR ( x ) / ~ R ( ~ ~ )
[21 [31
60
An alternative method of reporting retentions relies on the use of an appropriate series of homologous compounds that form a retention index scale. Retention indexes have been widely used in GC but infrequently in HPLC. Some efforts have, however, been made toward establishing retention index scales allowing better reproducibility and documentation of retention data. Baker and Ma (131) made the first proposal for a retention index series suitable for RP-HPLC, studying 2-alkanones as index standard compounds. However, 2-alkanones have only a weak chromophore and they have only limited use as index standard compounds for UV detection. Smith (132,133) and Kuronen (134) later introduced 1-phenyl-1-alkanones (Fig. 3.7a) as retention index standard compounds for RP-HPLC, Smith in an isocratic solvent system and Kuronen in gradient elution conditions. Gradient elution is, however, more applicable in allowing indexes to be determined for compounds with a wide range of polarities in a single chromatographic run (59,63,134-137). Further, a new series of (Fig. homologues 1-[4-(2,3-dihydroxypropoxy)phenyl)]-l-alkanones 3.7b) has been synthesized and evaluated as retention index calibrants in RP-HPLC under gradient elution conditions with UV and DAD (59,63,67,136). This series meets most of the essential requirements for a good reference series and it can serve as index standards for more polar solutes than the 1-phenyl-1-alkanone series. It is worth noting that the cubic spline interpolation (138) is a more precise method than the polygon method in calculating gradient-programmed retention indexes of the solutes because of the non-linearity of the calibration data (67). The gradient-programmed index is a complex function of the experimental conditions. Chromatographic parameters with greatest effect on the reliability of the gradient-programmed indexes are the source of the RP columns, column temperature, the organic modifier of the eluent, the pH of the eluent with ionizable compounds, and the exclusion of those members of the index series strongly determining the shape of the interpolation curve (67,136). The RI system can be used for tentative identifications under specified chromatographic conditions on an interlaboratory basis. The use of retention indexes in RP gradient elution HPLC
61
has been applied for mycotoxins (58,59,62-64)
n=l-ll
n = 1-11
Fig. 3.1. Structures of the homologous series of (a) l-phenyl-lalkanones, (b) l-[4-(2,3-dihydroxypropoxy)phenyl]-l-alkanones. 3.5.2.4 Detection Several sensitive and selective detectors capable of detecting only certain types of compounds have been developed for HPLC, whereas the lack of a sensitive universal detector has been to date one serious limitation of the method. The ideal HPLC detector possessing high sensitivity, low minimum detectability, wide linear dynamic range, good linearity, predictable and fast response, capability of being unaffected by changes in temperature, mobile phase composition and flow rate, capability of detecting all solutes or having predictable specificity, and providing qualitative information on the detected peak will Requirements for perhaps never be developed (95,139-142). detectors naturally vary with a particular separation problem. By far the most commonly used detectors in mycotoxin analysis have been conventional UV-vis and fluorescence detectors. Fortunately, very many mycotoxins (except many trichothecenes) display characteristic and strong UV absorptions at useful wavelengths (1,143,144), In addition, several toxins (e.g. aflatoxins, ZEA, OCH A, CIT) are naturally fluorescent: this property has offered a sensitive alternative to the UV detector. And the powerful combination of chromatography and spectroscopic techniques has become a reality also in HPLC analysis of mycotoxins with the development of the diode array detector (DAD) (59,60,62,63) and many interfacing techniques, especially thermospray (TSP) and dynamic fast atom bombardment (dynamic FAB), allowing HPLC to be coupled on-line with MS (60,64,145,146) .
62
3.5.2.4.1 "Classical" detections Refractive index detector. As a monitor of the refractive index of the eluate, the RI detector is a universal detector responding to all sample types. The pure mobile phase has a specific refractive index which changes when any compound elutes. The detector senses this difference and non-selectively records all peaks. To operate properly the RI detector requires excellent temperature, solvent composition and flow control. It is not amenable to gradient elution. Under favourable conditions the detection limit is about 0.5 pg, and the newer differential refractometers may allow quantitation of as little as 100 ng of most compounds. The RI detector is useful in preparative separation and routine quality control where ultratrace analysis or gradient elution is not required. Earlier, RI detection has been applied to the analysis of T-2, HT-2, and diacetoxyscirpenol (DAS) trichothecenes, with a detection limit of approximately 1 pg (147,148). Conventional UV-vis detector. The UV-vis detector is the most commonly used detector type in HPLC. This is the result of the vast number of UV-absorbing compounds and the great versatility and the excellent convenience and ruggedness of the detector. It can be highly sensitive, has a wide linear range, is unaffected by temperature fluctuation, and is very suitable for gradient elution. UV-vis detectors can be used for quantitation at the low-nanogram level. A primary requirement for successful UV detection is that the mobile phase system has been selected for optical transparency. All compounds absorbing UV or visible light are detected. Molecules absorb at a wavelength above 200 nm provided that they contain one of the following: an aromatic ring, a carbonyl group, a double bond adjacent to an atom with a lone pair of electrons, two conjugated double bonds, bromine, iodine, or sulphur. These groups of compounds do not absorb to the same extent or at the same wavelength. The absorption intensities, measured by molar absorptivity ( E ) , and wavelength maxima are also affected by neighboring groups in the molecule. Absorption increases with increasing conjugated unsaturation. Compounds with higher molar absorptivity produce larger peaks than those with a small molar absorptivity when identical amounts of compounds are injected. It
63
is useful to know the UV spectra of the various sample components (both analytes and interfering compounds) before the analysis, because it is then possible to choose the best detection wavelength. The UV-vis spectra with molar absorptivity values of most known mycotoxins are available from the literature (1). Many mycotoxins display characteristic and strong UV absorptions, allowing the detection of about 1 ppb of toxins with molar absorptivities 1-2.lo4 lmol-lcm-l. The Type A trichothecenes, T-2, HT-2, NEO, and DAS lack conjugated unsaturation and exhibit only end-absorption near 200 nm which means that they have low UV sensitivity at useful wavelength ranges, and can be detected and identified by UV detector only when present in relatively high concentrations. The preparation of p-nitrobenzoate derivatives of the Type A trichothecenes reportedly makes possible their ultratrace analysis in foods with W detection at 254 nm (149). The presence of a conjugated carbonyl in the Type B trichothecenes, DON, NIV, and FUS-X, generates a characteristic UV absorption near 220 nm. The minimum weight wm of a compound (in pg) giving a reasonable absorbance of peak maximum can be calculated from the following equation (150): wm
=
1000 MW(k‘+l) (S/N)(No)L0’5/~LcdcN 2 0.5
where MW is the molecular weight of the compound, k’ is the capacity factor, S/N is the required signal-to-noise ratio (usually > 2), No is the detector baseline noise in absorbance units, L is the column length (cm), E is the molar absorptivity (lmol-lcm-l), Lc is the length of the detector flow cell (cm), dc is the internal diameter of the column, and N is the column plate number. The simplest fixed wavelength UV detectors contain as source a low pressure mercury lamp which emits a sharp line spectrum with a strong line at 253.1 nm (254 nm) . Thus they are limited in their applications allowing only sample molecules absorbing near 254 nm to be detected. UV-vis detectors with a medium pressure mercury lamp as source can offer more fixed wavelengths, including 254, 280, 312, 365, 436, and 546 nm. Variable wavelength UV-vis detectors are the most common
64
absorbance HPLC detectors. Some of them are true recording spectrophotometers which allow a UV-vis spectrum to be generated from an eluting peak trapped in the flow cell (stop-flow of wavelength technique). Others require manual selection (usually 1 9 0 - 3 5 0 nm or 1 9 0 - 7 0 0 nm), allowing selection of a wavelength to maximize sensitivity or remove interfering peaks, thereby improving the accuracy of quantitative determination. Fluorescence detector. Compounds that naturally fluoresce or that can be made to fluoresce through chemical derivatization can be detected with high selectivity and sensitivity by this detector. The fluorescence detector is generally about 1000 times more sensitive than the UV detector. Laser-induced fluorescence (LIF) detector, which is one of the most promising applications of laser-based detection in HPLC, can further improve detectability even to femtogram level. High selectivity means that the compounds of interest can be readily distinguished from a complicated matrix of compounds that do not fluoresce. Both fixed wavelength and scanning fluorescence units are available. Fluorescence detectors can be used with gradient elution. The fluorescent compound being analyzed is excited by W radiation at UV maximum wavelength and the fluorescence energy emitted at a longer wavelength is detected. The intensity of the fluorescence and the position of the excitation/emission wavelength maxima depend on the mobile phase composition, pH, temperature, and dissolved gas content (particularly oxygen) of the mobile phase ( 9 7 ) . For example, halogenated solvents such as chloroform tend to quench or reduce dichloromethane or fluorescence. For example, the aflatoxins, STE, ZEA, ochratoxins, and CIT exhibit significant native fluorescence when subjected to UV irradiation. Under optimum conditions fluorescence detection is about 30-40 times more sensitive than UV detection for aflatoxins. The influence of chromatographic conditions on fluorescence intensity of mycotoxins have been discussed in the literature (7,45-46). Orti et al. (151) have given an excellent examination of chromatographic and spectroscopic properties of hemiacetals of aflatoxins and sterigmatocystins. Picogram quantities of the aflatoxins may be detected by fluorescence detection under appropriate conditions. The sensitivity of the fluorescence
65
detection of the aflatoxins can be further enhanced by several The aflatoxins can be converted to more techniques ( 7 , 4 5 , 4 6 ) . intensely fluorescent derivatives by using pre-column derivatization with TFA (7,45,46,152) or post-column derivatization with iodine or bromine ( 1 5 3 - 1 5 6 ) . A flow cell packed with silica may be used to intensify the fluorescence of the aflatoxins under normal-phase conditions ( 7 , 4 5 , 4 6 ) . In addition, the use of LIF detection enhances the sensitivity. Electrochemical detector. The electrochemical detector (ECD) provides a useful and highly selective and sensitive tool for the detection of readily oxidizing and reducing organic compounds. Examples of compounds that can be detected in oxidation mode are phenols, aromatic and aliphatic amines, thiols, thioketones, and thioethers. Aromatic nitro compounds, amides, oximes, alkyl and aryl halogen compounds, quinones, and amides are suitably detected by electrochemical reduction. Electrochemical reactions occur at the surface of a solid electrode which removes electrons for oxidation and supplies them for reduction. The operating potential of the ECD is set and the current due to oxidation or reduction is measured. The potential applied to the detection can be adjusted to allow discrimination between different electroactive compounds. The majority of applications are in the oxidation mode because dissolved oxygen in the mobile phase and the presence of heavy metals tend to cause problems in the reduction mode. ECD requires the use of a conducting mobile phase, containing aqueous organic (e.g. MeOH, ACN) mixtures and inorganic salts or acids (acetic acid, phosphoric acid), conditions which are compatible with RP-HPLC. These detectors are capable of femtomole sensitivity. ECD has been utilized in the analysis of roquefortine in blue cheese ( 1 5 7 ) and zearalenones in cereals (158) and edible animal tissue ( 1 5 9 ) . 3 . 5 . 2 . 4 . 2 Diode array detection. has The diode array detector (DAD) ( 1 4 0 - 1 4 2 , 1 6 0 - 1 6 4 ) established itself as a powerful LC detector during the last years. DAD uses a photodiode array (e.g. 2 5 6 elements) to detect many wavelengths simultaneously making it possible to provide both multlwavelength chromatographic and spectral information in a single chromatographic run, which makes DAD ideal for the screening and preliminary identification of mycotoxins
66
(59,60,62,63). The newest DAD instruments have competitive sensitivity with other UV detectors. Many graphical and numerical strategies have been developed for the presentation and analysis of the data (161,165,166). The most important capabilities of the DAD are the following: (1) on-the-fly UV-vis spectral scanning; (2) three-dimensional plots; (3) two-dimensional contour plots; (4) UV-spectral overlays; (5) absorbance ratioing (purity parameter) and absorbance ratio plots; (6) derivative spectra; and ( 7 ) recording of chromatograms simultaneously at several wavelengths. The most used function of the DAD is the generation of the on-the-fly UV-vis spectra of separated compounds. Comparison of peak spectra with reference spectra in the library can be used to confirm peak identity. UV-vis spectra can also aid in the identification of unknown peaks, or allow determination of at least the class of the compound. Figure 3.8 presents the on-line UV spectra of some mycotoxins produced with DAD under gradient elution RP-HPLC conditions (63). The spectra are practically identical with those produced off-line with UV-vis spectrophotometers and published in the literature (1,143,144). Insignificantly small shifts (-1-2 nm) are found in some cases. The UV-vis spectrum very often provides little structural information. Therefore qualitative features of the UV-vis spectra can be enhanced by generation of derivatives of the spectra. Verification of the peak homogeneity is provided by the coincidence of UV spectra taken at several points of the eluting peak (usually upslope, apex, and downslope). The three-dimensional presentation of wavelength ( h )-time (t)-absorbance (A) data ( = 3-D plot) is useful in selecting the optimal wavelength for detection sensitivity and selectivity. The two-dimensional contour plot of the absorbance contours in the wavelength-time plane gives symmetrical contours for pure peaks and skewed contours indicate co-elution. Absorbance ratios can also be used for peak homogeneity determinations requiring, however, carefully chosen wavelengths (167), which is easy only for known samples but very difficult if there are unknown impurities. Chromatograms can also be recorded simultaneously at several wavelengths, enabling resolution of co-eluted compounds and making it possible to screen the whole UV-vis region during an analysis, with no UV absorbing compound undetected.
67
Less chromatographic resolution is required with this multichannel detection than with the single-channel UV detector. Qualitative and quantitative analysis is possible f o r moderately overlapping spectra and chromatographic peaks with sophisticated data handling methods. A RP-HPLC gradient elution method has been applied as a multimycotoxin screening method where mycotoxins were characterized using retention indexes based on the 1-phenyl-1-alkanone ( 6 2 ) and l-[4-(2,3-dihydroxypropoxy)phenyl]-lalkanone (63) series and UV-vis spectral data produced with the DAD.
90%LAUl
0.
\
0-
- 10. 0.
-10.0
18
9 0 % - , 7
18
.o
289. 0
J
389.0
-lu.o-1.
184.0
. .
.
,
289.0
J
389.0
W o v e 1e n g t h
C n m l
W a v e l e n g t h
C,im3
W a v e l e n g t h
CnmJ
W a v e 1ength
Cnm3
68
90%-
,P\, P\
MU1
STE
-\‘I;
289.0
184.0
389.0
W a v e 1e n g t h
0-10.0184.0
289.0 W a v e 1 ensth
Cnml
389.0 Cnrnl
-10.0 184.0
289.0
389.0
W a v e 1e n g t h
W a v e 1e n g t h
Cnrnl
Cnrnl
90%-
9
:
:
0 -10.0-
184.0
;
-
y
1
289.0 W o v e 1e n g t h
.-.- .
189.0
289.0 W a v e 1 ength
hAU1
389.0 Cnml
0-10.0 .
ISB. 0
289.0
389.0
W a v e 1e n g t h
Cnrnl
389.0 C n r n l
Fig. 3.8. The UV spectra of some mycotoxins recorded from 190 to 400 nm with the diode array detector at 50 OC on LiChrosorb Hibar RP-18 column (5 pm, 2 5 0 ~ 4 . 0m m ) . Linear gradient from 20% ACN-HZO (pH=2.5) to 100% ACN in 40 min at the flow rate of 1 ml/min. 3.5.2.4.3 Mass spectrometry detection and identification
The combination of liquid chromatography and spectrometry (LC-MS) is an important technique that offers
mass high
69
sensitivity and selectivity in the analysis of a wide variety of compounds that is difficult or impossible with GC-MS - such as the analysis of many mycotoxins. The most important difference between MS and other LC detectors is its ability to provide structural and molecular weight information. Connecting MS to LC requires an interface device that will convert the liquid phase containing the analyte(s) to a gas phase in the presence of vacuum. The problems associated with interfacing the LC to the MS have been much greater than for the GC. Both quadrupole and magnetic sector MS instruments have been used for LC-MS. Recent developments in LC-MS interfaces have increased the reliability of the technique. Several interfacing techniques have been developed during the last few years including moving belt interface, direct liquid introduction (DLI) interface, thermospray (TSP) interface, particle beam interface, electrospray and atmospheric pressure type interfaces, monodisperse aerosol generation (MAGIC) interface, and dynamic fast atom bombardment (FAB) interfacing (168-171). All these techniques have their own strengths and weaknesses, depending on the LC-MS equipment and the results desired. The two interface techniques, TSP and FAB, offer most potential today and they are already routinely used in several laboratories. The popularity of the TSP interface is largely due to the fact that the total eluent (typically 1-2 ml/min) can be introduced into the ion source, the ionization of the sample may be obtained without the use of a filament, and its ability to operate under RP-HPLC conditions, working best with mobile phases containing a high proportion of water at flow rates between 0.5 and 2 . 0 ml/min (171), when many interfaces begin to fail. In TSP LC-MS, the mobile phase containing the separated analytes is introduced into the MS through a stainless steel capillary tube, which is directly heated by passing a current through it. The mobile phase i s converted by careful temperature regulation into a mist of droplets and carried into the ion source as a supersonic vapour jet. Excess of solvent vapour is removed by an extra vacuum pump, The detectable ions are usually produced by using filament-off ionization ( = buffer ionization). Volatile buffer, generally ammonium acetate, which is added to the mobile phase at a concentration of about 0.05 to 1.0 M, acts as the reagent gas
70
and produces CI-type MS spectra. These spectra usually give pseudomolecular ions [e.g. (M+H)+ and/or (M+NH4)+, in positive ion mode] together with a small degree of fragmentation limiting specificity and giving only little structural information. In the negative ion mode negative ions may be formed by proton abstraction or anion attachment. The amount of each ion species formed depends on the proton affinity of the gaseous analytes (gas-phase acidity). On the other hand little fragmentation may be an advantage when quantitating analytes in the selected ion mode, which produces greater sensitivity than the scanning mode. Furthermore, the pH can be adjusted using either ammonia or acetic acid, and volatile ion-pair reagents can also be used. A buffer ionization mode (pure TSP CI) cannot be used for low-polarity solvents of normal-phase LC-MS. In the cases where the analytes are not readily ionized, or NP-HPLC is used, some TSP devices include an electron filament and discharge ion source to assist in ionization or to make possible the use of normal-phase solvents. In addition, some TSP interfaces may also have adjustable fragmentor electrodes, which produce molecular fragments by increasing the rate of intermolecular collisions, therefore being useful for structure elucidation. Further, the combination of HPLC with MS-MS instrumentation, which is capable of fragmenting molecular ions into structurally significant daughter fragments under collision with an inert gas, can be used for structural studies. The response and sensitivity of TSP LC-MS using filament-off, filament-on, or discharge-on CI is very compound-dependent, and can be affected by several physical ionization factors. FAB has been an alternative ionization technique for several years. Nowadays dynamic FAB systems are available for use as interfaces in LC-MS. The flow rates in this system are in the order of 1-5 pl/min, and therefore use of microbore columns or post-column splitting before MS are necessary. Furthermore, the system requires the matrix (4-10% glycerol) in the mobile phase. The matrix can be added either to the mobile phase or using a post-column addition. The latter method has been found to have no significant effect on the retention time but may cause peak broadening (64). An argon or xenon molecular beam is used in the bombardment.
71
TSP LC-MS (60,64,145,146) have proven very useful for the
and also dynamic FAB LC-MS (64) analysis of a wide range of
mycotoxins; both being applicable as multimycotoxin methods. CONCLUSION HPLC plays an important role in the analysis of mycotoxins. It is a powerful analytical technique being able to separate a wide range of mycotoxins, being quantitatively precise, and in many cases a very sensitive technique. It demands, however, a good understanding of the problems involved in the application of this technique to the food analysis in order to produce reliable and accurate data. Mini-column chromatographic methods, especially developed for aflatoxins, are particularly useful for field analysis and as screening tests for agricultural commodities when rapid decisions have to be made for accepting or rejecting a lot. Efficient extraction and clean-up of the samples are very critical to successful HPLC. The purity of the residue obtained from the sample pretreatment will have a major influence on both detection sensitivity and degree of confidence in the result. Earlier, classical open column chromatographic methods have been widely used as a preliminary clean-up for trace analysis of mycotoxins by HPLC, but nowadays the replacement of classical laboratory-packed glass columns by commercially available cartridge clean-up columns has greatly simplified sample purification, making possible higher reproducibility between different laboratories. The dedicated TSP and FAB LC-MS instruments can now provide a powerful technique to the analysis of mycotoxins. In addition, the reliability of the detection can be greatly improved by the use of retention indexes, which offer an independent identification, additional to the data produced by DAD or the MS-data. In the near future the completely automated, unattended HPLC assay of mycotoxins, starting from the sample extraction and ending with the identification of the toxins and the calculation of the quantitative results, in one operation, will become a reality. 3.6
72
REFERENCES 1 R.J. Cole and R.H. Cox, Handbook of Toxic Fungal Metabolites, Academic Press, New York, 1981. W.B. Turner and D.C. Aldridge, Fungal Metabolites 11, 2 Academic Press, London, 1983. D.H. Watson, CRC Crit. Rev. Food Sci. Nutr., 22 (1985) 117. 3 C.W. Hesseltine, in: P.S. Steyn and R. Vleggaar (Eds.), My4 cotoxins and Phycotoxins, Elsevier, Amsterdam, 1986, p. 1. P.M. Scott, in: K. Robinson (Ed.), Developments in Food 5 Microbiology, Vol. 4, Elsevier, Amsterdam, 1988, p. 47. Y. Ueno, CRC Crit. Rev. Toxicol., 14 (1985) 99. 6 7 R.D. Coker and B.D. Jones, in: R. Macrae (Ed.), HPLC in Food Analysis, Academic Press, London, 1988, p. 335. 8 R.E. Majors, LC-GC, 4 (1986) 972. R.E. Majors, LC-GC Int., 2(2) (1989) 12. 9 J.E. Thean, D.R. Lorenz, D.M. Wilson, K. Rodgers, and 10 R.C. Gueldner, J. Assoc. Off. Anal. Chem., 63 (1980) 631. 11 J.D. McKinney, J. Am. Oil Chem. SOC., 58 (1981) 935A. 12 J.E. Hutchins and W.M. Hagler, J. Assoc. O f f . Anal. Chem., 66 (1983) 1458. W.J. Hurst, F.P. Snyder, and R.A. Martin, Peanut Sci., 11 13 (1984) 21. D. Tosch, A.E. Waltking, and J.F. Schlesier, J. Assoc. Off. 14 Anal. Chem., 67 (1984) 337. G.-S. Gian and G.C. Yang, J. Agric. Food Chem., 32 (1984) 15 1071. D.L. Orti, R.H. Hill, Jr., J.A. Liddle, L.L. Needham, and 16 L. Vickers, J. Anal. Toxicol., 10 (1986) 41. 17 N. Takeda, J. Chromatogr., 288 (1984) 484. H. Cohen and M. Lapointe, J. Assoc. Off. Anal. Chem., 69 18 (1986) 957. 19 H. Terada, H. Tsubouchi, K. Yamamoto, K. Hisada, and Y. Sakabe, J. Assoc. Off. Anal. Chem., 69 (1986) 960. K.C. Ehrlich and L.S. Lee, J. ASSOC. O f f . Anal. Chem., 67 20 (1984) 963. 21 S.H. Hoke, C.M. Carley, E.T. Johnson, and F.H. Broski, J. Assoc. Off. Anal. Chem., 70 (1987) 661. 22 T.R. Romer, D.E. Greaves, and G.E. Gibson, presented at 6th Annual Spring Workshop of the AOAC, Ottawa, Ontario, Canada, May 12-14, 1981. 23 R.E. Eppley, M.W. Trucksess, S. Nesheim, C.W. Thorpe, and A.E. Pohland, J. Assoc. Off. Anal. Chem., 67 (1984) 43. H.L. Chang, J.W. DeVries, P.A. Larson, and H.H. Patel, 24 J. ASSOC. Off. Anal. chem., 67 (1984) 52. T.R. Romer, J. ASSOC. Off. Anal. Chem., 69 (1986) 699. 25 C.P. Gorst-Allman and P.S. Steyn, in: V. Betina (Ed.), 26 Mycotoxins - Production, Isolation, Separation, and Purification. Developments in Food Science, Vol. 8, Elsevier, Amsterdam, 1984, p. 59. 27 W.C. Brumley, S. Nesheim, M.W. Trucksess, E.W. Trucksess, P.A. Dreifuss, J.A.G. Roach, D. Andrzejewski, R.M. Eppley, A.E. Pohland, C.W. Thorpe, and J.A. Sphon, Anal. Chern., 53 (1981) 2003. W.C. Brumley, M.W. Trucksess, S.H. Adler, C.K. Cohen, 28 K.D. White, and J.A. Sphon, J. Agric. Food Chem., 33 (1985) 326. L. Stoloff and P.M. Scott, in: S. Williams (Ed.), Official 29 Methods of Analysis, 14th ed., Association of Official Ana-
73
30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 51 50
59
lytical Chemists, Washington, D.C., 1984, p. 477. W.C. Still, M. Kahn, and A. Mitra, J. Org. Chem., 43 (1978) 2923. R.E. Majors and T. Enzweiler, LC-GC Int., 2(3) (1989) 10. C.E. Holaday, J. Am. Oil Chem. SOC., 45 (1968) 680. J. Velasco, J. Am. Oil Chem. SOC., 49 (1972) 141. P. Hald and P. Krogh, J. Assoc. Off. Anal. Chem., 58 (1975) 156. C.E. Holaday and J. Lansden, J. Agric. Food Chem., 23 (1975) 1134. T.R. Romer, J. Assoc. Off. Anal. Chem., 58 (1975) 500. T.R. Romer and A.D. Campbell, J. Assoc. Off. Anal. Chem., 59 (1976) 110. C.E. Holaday, J. Am. Oil Chem. SOC., 53 (1976) 603. T.R. Romer, N. Ghouri, and T.M. Boling, J. Am. Oil Chem. SOC., 56 (1979) 795. C.E. Holaday, J. Am. Oil Chem. SOC., 57 (1980) 491A. C.E. Holaday, J. Am. Oil Chem. SOC., 58 (1981) 931A. O.L. Shotwell and C.E. Holaday, J. Assoc. Off. Anal Chem., 64 (1981) 674. R.B. Sashidhar, R.B. Bhat, and S. Vasanthi, Curr. Sci., 58 (1989) 882. S . Williams (Ed.), Official Methods of Analysis, Association of Official Analytical Chemists, Washington, D.C., 1984. J.W. Seiber and D.P.H. Hsieh, J. ASSOC. Off. Anal. Chem., 56 (1973) 827. P.M. Scott, in: J.F. Lawrence (Ed.), Trace Analysis, Vol. 1, Academic Press, New York, 1981, p. 193. M.J. Shepherd, in: R.J. Cole (Ed.), Modern Methods in the Analysis and Structure Elucidation of Mycotoxins, Academic Press, New York, London, 1986, p. 293. L.R. Snyder and J.J. Kirkland, Introduction to Modern Liquid Chromatography, 2nd ed., John Wiley and Sons, Inc., New York, 1979. M.T. Gilbert, High Performance Liquid Chromatography, Wright, Bristol, 1987. P.R. Brown and R.M. Hartwick (Eds.), High Performance Liquid Chromatography, Wiley, New York, 1989. V.R. Meyer, Practical High Performance Liquid Chromatography, Wiley, Chichester, 1988. C. Horvath (Ed.), High Performance Liquid Chromatography, Advances and Perspectives, Vol. 5., Academic Press, San Diego, 1988. J.C. Giddings, E. Grushka, P. Brown (Eds.) , Advances in Chromatography, Vol. 28, Marcel Dekker, New York, 1989. L.R. Snyder, J.L. Glajch, J.J. Kirkland, Practical HPLC Method Development, Wiley, New York, 1988. S. Ahuja, Selectivity and Detectability Optimization in HPLC, Wiley, New York, 1989. J.C. Berridge, Techniques for the Automated Optimization of HPLC Separations, Wiley, New York, 1985. J.W. Dolan and L.R. Snyder, Troubleshooting LC Systems, Humana Press, Clifton, 1989. D.W. Hill, T.R. Kelley, K.J. Langner, and K.W. Miller, Anal. Chem., 56 (1984) 2576. P. Kuronen, in: M. Rautio (Ed.), Systematic Identification of Mycotoxins, B . 5 . Identification of Selected Trichothe-
14
60 61 62 63 64 65 66 67 68 69 70 71 72 13 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96
cenes, Aflatoxins, and Related Mycotoxins, The Ministry for Foreign Affairs of Finland, Helsinki, 1986, p. 21. E. Rajakyla, K. Laasasenaho, and J.D. Sakkers, J. Chromatogr., 384 (1987) 391. J.C. Frisvad, J. Chromatogr., 392 (1987) 333. J.C. Frisvad and U. Thrane, J. Chromatogr., 404 (1987) 195. P. Kuronen, Arch. Environ. Contam. Toxicol., 18 (1989) 336. R. Kostiainen and P. Kuronen, J. Chromatogr., 543 (1991) 39. J.W. Dolan, LC-GC Int., 2 (4) (1989) 22. R. Majors, LC-GC Int., 2(4) (1989) 14. P. Kuronen, Ann. Acad. Sci. Fenn., Ser. A2, 224, 1990. L. Trojer and L. Hansson, J. Chromatogr., 262 (1983) 183. M.L. Hunnicut and J.M. Harris, Anal. Chem., 58 (1986) 748. M. Furlanut, M. Plebani, and A. Burlina, J. Liq. Chromatogr., 12 (1989) 1759. D.R. Lachno, N. Patel, M.L. Rose, and M.H. Yacoub, J. Chromatogr., 525 (1990) 123. Ref. 57, p. 491. Ref. 48, p. 694. Ref. 54, p . 148. R.E. Majors, LC-GC Int., 3(9) (1990) 10. P. Lepom, J. Chromatogr., 354 (1986) 518. H.T. Smith, W.T. Robinson, J. Chromatogr., 305 (1984) 353. Ref. 54, p. 199. Ref. 55, p. 103. Ref. 48, p. 183. Ref. 54, p . 33. Ref. 48, p. 349. L.R. Snyder, J.L. Glajch, and J.J. Kirkland, J. Chromatogr., 218 (1981) 299. L.R. Snyder, in: C. HorvAth, (Ed.), High-Performance Liquid Chromatography, Advances and Perspectives, Vol. 3, Academic Press, New York 1983, p. 157. W.A. Pons and A.O. Franz, J. ASSOC. Off. Anal. Chem., 60 (1977) 89. T. Panalaks and P.M. Scott, J. ASSOC. Off. Anal. Chem., 60 (1977) 583. W.A. Pons, L.S. Lee, and L. Stoloff, J. Assoc. Off. Anal. Chem., 63 (1980) 899. M.J. Shepherd, T.D. Phillips, N.D. Heidelbauch, and A.W. Hayes, J. ASSOC. Off. Anal. Chem., 65 (1982) 665. A.D. Campbell, O.J. Francis, R.A. Beebe, and L. Stoloff, J. Assoc. Off. Anal. Chem., 67 (1984) 312. E.J. Tarter, J.-P. Itanchay, and P.M. Scott, J. Assoc. Off. Anal. Chem., 67 (1984) 597. J. Gilbert, M.J. Shepherd, Food Addit. Contam., 2 (1985) 171. M.E. Olsen, H.I. Petterson, K.A. Sandholm, and K.-H.C. Kiessling, J. Assoc. Off. Anal. Chem., 68 (1985) 632. N. Takeda, J. Chromatogr., 288 (1984) 484. H. Wisniewska and J. Piskorska-Pliszczynska, Bull. Vet. Inst. Pulawy, 25 (1982) 38. W.R. Melander, and C. Horvath, in: C. Horvath (Ed.), High Performance Liquid Chromatography, Advances and Perspectives, Vol. 2, Academic Press, New York, 1980. A.M. Krstulovic and P.R. Brown, Reversed-Phase High-Performance Liquid Chromatography. Theory, Practice, and Biomedical Applications, Wiley, New York, 1982.
75
97 98 99 100 101
102 103 104
105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129
C.F. Poole and S.A. Schuette, Contemporary Practice of Chromatography , El sevier , Amsterdam , 1984 . K.K. Unger, Porous Silica, Its Properties and Use as Support in Column Liquid Chromatography, J. Chromatogr. Library, Vol. 16, Elsevier, Amsterdam, 1979. H. Engelhardt and H. Elgass, in: C. Horvdth (Ed.), High-Performance Liquid Chromatography, Advances and Perspectives, Vol 2., Academic Press, New York 1980, p. 57. K.K. Unger and B. Anspach, Trends Anal. Chem., 6 (1987) 121. L.C. Sander and S.A. Wise, CRC Crit. Rev. Anal. Chem., 18 (1987) 299. P.J. Van den Driess, H.J. Ritchie, and S. Rose, LC-GC, 6 (1988) 124. J. Nawrocki and B. Buszewski, J. Chromatogr., 449 (1988) 1. J.J. Kirkland, J.L. Glajch, R.D. Farlee, Anal. Chem. , 61 (1989) 2. M.J. Walters, J. Assoc. Off. Anal. Chem., 70 (1987) 465. L.R. Snyder, J. Chromatogr., 92 (1974) 223. L.R. Snyder, J. Chromatogr. Sci., 16 (1978) 223. J.J. Lewis, L.B. Rogers, and R.E. Pauls, J. Chromatogr. , 264 (1983) 339. S.D. West, J. Chromatogr. Sci., 25 (1987) 122. S.D. West, J. Chromatogr. Sci., 27 (1989) 2. Ref.48, Chapter 16. P . Jandera and J. Churbcek, Gradient Elution in Column Liquid Chromatography, J. Chromatogr. Library, Vol. 31, Elsevier, Amsterdam, 1985. L.R. Snyder and M.A. Stadalius, in: C. Horvdth (Ed.), High Performance Liquid Chromatography, Advances and Perspectives, Vol. 4., Academic Press, New York, 1986, p. 195. G. Vigh, 2 . Varga-Puchony, A. Bartha, and S. Balogh, J. Chromatogr., 241 (1982) 169. D. Palalikit and J. H. Block, Anal. Chem., 52 (1980) 624. R. Schwarzenbach, J. Chromatogr. , 251 (1982) 339. A.P. Goldberg, E. Nowakowska, P.E. Antle, and L.R. Snyder, J. Chromatogr., 316 (1984) 241. M.J. Shepherd and J. Gilbert, J. Chromatogr., 358 (1986) 415. M.H. Lebrun, F. Gaudemer, M. Boutar, L. Nicolas, and A. Gaudemer, J. Chromatogr. , 464 (1989) 307. R. Gill, A.C. Moffat, R.M. Smith, and T.G Hurdley, J. Chromatogr. Sci., 24 (1986) 153. R. Gill, M.D. Osselton, R.M. Smith, and T.G. Hurdley, J. Chromatogr., 386 (1987) 65. R.M. Smith, T.G. Hurdley, R. Gill, and M.D. Osselton, J. Chromatogr., 398 (1987) 73. A.M. Krstulovic, H. Colin, and G. Guiochon, Anal. Chem., 54 (1982) 2438. H. Engelhardt, H. Muller, and B. Dreyer, Chromatographia, 19 (1984) 240. G.E. Berendsen, P.J. Schoenmakers, L. de Galan, E. Vigh, 2 . Varga-Puchony, and J. Inczedy, J. Liq. Chromatogr., 3 (1980) 1669. M.J.M. Wells, and C.R. Clark, Anal. Chem., 53 (1981) 1341. O.A.G.J. Van der Houwen, J.A.A. van der Linden, and A.W.M. Indemans, J. Liq. Chromatogr., 5 (1982) 232. K. Jinno, Chromatographia, 17 (1983) 367. R.M. McCormick and B.L. Karger, Anal. Chem., 52 (1980) 2249.
76
130 131 132 133 134
135
136
137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156
P.J.M. Van Tulder, J . P . Franke, and R.A. de Zeeuw, J. High Resolut. Chromatogr. Chromatogr. Comm., 10 (1987) 191. J.K. Baker and C.-Y. Ma, J. Chromatogr., 169 (1979) 107. R.M. Smith, J. Chromatogr., 236 (1982) 313. R.M. Smith, in: J.C. Giddings, E. Grushka, and P.R. Brown (Eds.), Advances in Chromatography, Vol. 26, Marcel Dekker, New York, 1987, p. 277. P. Kuronen, in: J. Enqvist and A. Manninen (Eds.), Systematic Identification of Chemical Warfare Agents, B.3. Identification of Non-PhosphOrus Warfare Agents, the Ministry for Foreign Affairs of Finland, Helsinki, 1982, p. 43. P. Kuronen, in: J. Enqvist and A. Manninen (Eds.), Systematic Identification of Chemical Warfare Agents, B.4. Identification of Precursors of Warfare Agents, Degradation Products of Non-Phosphorus Agents, and Some Potential Agents, the Ministry for Foreign Affairs of Finland, Helsinki, 1983, p. 51. P. Kuronen, in: M. Rautio (Ed.), Air Monitoring as a Means for Verification of Chemical Disarmamemnt, C.2. Development and Evaluation of Basic Techniques, Part I, the Ministry for Foreign Affairs of Finland, Helsinki, 1985, p. 162. P. Kuronen, Proc. 2nd Int. Symp. Protection Against Chemical Warfare Agents, National Defence Research Institute, NBS Research Dept., Umea, 1986, p. 261. W.A. Halang, R. Langlais, and E. Kugler, Anal. Chem., 50 (1978) 1829. Ref. 48, p . 125. R.P.W. Scott, Liquid Chromatography Detectors, J. Chromatogr. Library, Vol. 33, 2nd ed., Elsevier, Amsterdam, 1986. Ref. 55, p. 505. P.C. White, Analyst, 109 (1984) 667; 973. A.E. Pohland, P.L. Schuller, and P.S. Steyn, Pure Appl. Chem., 54 (1982) 2219. V. Betina (Ed.), Mycotoxins - Production, Isolation, Separation, and Purification. Developments in Food Science, Vol. 8, Elsevier, Amsterdam, 1984, pp. 87-485. R.D. Voyksner, W.M. Hagler, J r . , K. Tyczkowska, and C.A. Haney, J . High Resolut. Chromatogr. Chromatogr. Commun., 8 (1985) 119. R.D. Voyksner, W.M. Hagler, Jr., and S.W. Swanson, J. Chromatogr., 394 (1987) 183. R. Schmidt, E. Ziegenhagen, and K. Dose, J. Chromatogr., 212 (1981) 370. R. Schmidt and K. Dose, J. Anal. Toxicol., 8 (1984) 43. R. Maycock, and D. Utley, J. Chromatogr., 347 (1985) 429. Ref. 54, p. 87. D.L. Orti, J. Grainger, D.L. Ashley, and R.H. Hill, Jr., J . Chromatogr., 460 (1989) 269. D.L. Park, S. Nesheim, M.W. Trucksess, M.E. Stack, and R.F. Newell, J. Assoc. Off. Anal. Chem., 73 (1990) 260. H. Jansen, R. Jansen, U.A.T. Brinkman, and W. Fsei, Chromatographia, 24 (1987) 555. W.J. Hurst, F.P. Snyder, and R.A. Martin, J. Chromatogr., 409 (1987) 413. J.W. Dorner and R.J. Cole, J. Assoc. Off. Anal. Chem., I1 (1988) 43. W.T. Kok, T.C.H. van Neer, W.A. Traag, and
157 158
L.G.M.T. Tuinstra, J. Chromatogr., 3 6 3 ( 1 9 8 6 ) 2 3 1 . G.M. Ware, C.W. Thorpe, and A.E Pohland, J. Assoc. Off. Anal. Chem., 6 3 ( 1 9 8 0 ) 6 3 7 . M.R. Smyth and C.G.B. Frischkorn, Anal. Chim. Acta, 1 1 5 (1980) 293.
159 160 161 162
J.G. Roybal, R.K. Munns, W.J. Morris, J.A. Hurlbut, and W. Shimoda, J. Assoc. Off. Anal. Chem., 7 1 ( 1 9 8 8 ) 2 6 3 . D.G. Jones, Anal. Chem., 57 ( 1 9 8 5 ) 1 0 5 7 A . D.G. Jones, Anal. Chem., 57 ( 1 9 8 5 ) 1 2 0 7 A . T. Alfredson and T. Sheehan, J. Chromatogr. Sci., 2 4 ( 1 9 8 6 ) 473.
163 164 165 166 167 168 169 170 171
G.W. Schieffer, J. Chromatogr., 3 1 9 ( 1 9 8 5 ) 3 8 7 . Ref. 57, p . 3 2 1 . T. Alfredson, T. Sheehan, T. Lenert, S. Aamodt, and L. Correia, J. Chromatogr., 3 8 5 ( 1 9 8 7 ) 2 1 3 . S . Ebel and W. Mueck, Chromatographia, 2 5 ( 1 9 8 8 ) 1 0 3 9 . H. Cheng and R.R. Gadde, J. Chromatogr. Sci., 2 3 ( 1 9 8 5 ) 2 2 7 P.A. Ireland, in: R. Macrae (Ed.), HPLC in Food Analysis, Academic Press, London, 1 9 8 8 , p. 4 7 1 . P. Newton, LC-GC Int., 3 ( 9 ) ( 1 9 9 0 ) 2 8 . W.J.B. Lanchflower, Spectroscopy Int., 2 ( 4 ) ( 1 9 9 0 ) 3 7 . R.D. Voyksner and C.A. Haney, Anal. Chem., 5 7 ( 1 9 8 5 ) 9 9 1 .
78
Chapter 4 TECHNIQUES OF GAS CHROMATOGRAPHY R. W. BEAVER
4.1.INTRODUCTION The mycotoxins are of concern when they occur in animal or human feedstuffs. The extreme toxicity and/or carcinogenicity of many of the mycotoxins necessitates their detection and determination at very low levels. The combination of chemically complex matrices such as foodstuffs with the need to detect a single analyte at low levels presents a formidable challenge to the analytical chemist. Gas chromatography (GC) is an analytical technique which is, in many cases, capable of meeting this challenge. The utility of GC in the analysis of mycotoxins is dependent on the ability of the technique to resolve the mycotoxin of interest from other constituents in the matrix. GC, and capillary GC in particular, is often referred to as a high resolution chromatographic technique. While this is certainly true, contributions to resolution which occur prior to the GC column (such as clean-up and extraction schemes which serve to isolate the mycotoxins from interferences) and after the GC column (such as detectors which exhibit a high degree of specificity or selectivity for the mycotoxin of interest) are often overlooked in the search for high resolution. Since a broad view of resolution would refer to whatever means are employed to separate the analyte from interferences, both chromatographic resolution (which occurs on the GC column) and extracolumn resolution (such as through clean-up and detection) are important in analyses. In this chapter, the basic theory of GC, which leads to ways of controlling resolution on the GC column, will be discussed. Various techniques for achieving extracolumn resolution will also be examined. Where applicable, specific examples of mycotoxin determinations will be used to illustrate the discussion. The examples will be chosen for their pertinence to the discussion and are not intended to present a comprehensive review of the determination of mycotoxins by GC. The reader is referred to Chapter 9 in this book for such a review.
79
4.2.RESOLUTION IN GAS CHROMATOGRAPHY 4.2.1 Definition of Resolution In order to accurately identify and quantitate a chromatographic peak, the peak of interest must be adequately resolved, or separated, from all adjacent peaks. Resolution can be measured directly from the chromatogram (1) according to equation 111
where tR1and tR2are the retention times of the peaks of interest and W,, and W,, are the peak widths (same units as t,, and tR2) at the base line. Snyder and Kirkland (2) provided a set of fiqures which illustrate the effect that different R, values have on the appearance of chromatographic peaks. An R, value of 1.25 corresponds to essentially complete, or base line, resolution of peaks of approximately the same size. While equation [l] provides a means of measuring the resolution for two peaks in a chromatogram, it provides no insight into the physical and chemical parameters in the chromatographic process which affect resolution. An alternative expression for R, (3) is given in equation [2]
where N is the number of theoretical plates generated by the column (or, synonymously, the efficiency of the column), k is the capacity factor, and cx is the selectivity or separation factor. Equation [2] couches resolution in terms of the three fundamental factors over which the chromatographer has control: 1) efficiency (N); 2) retention (k); and 3) chemical interaction between analytes and the column stationary phase ( cx 1 . 4.2.2 Efficiency Efficiency in GC refers to the ability of the GC system to generate narrow peaks. Obviously, the narrower the peaks, the less likelihood that adjacent peaks will overlap. Efficiency is measured as a quantity known as a theoretical plate (N) and can be measured from the chromatogram as in equation [31 N
=
16
( 2)'
I31
80
where t, is the peak retention time and W, is the peak width at the base line (1). Column efficiencies are usually reported as plates per meter so that columns of different lengths can be compared. Alternatively, column efficiency can be expressed as height equivalent to a theoretical plate, or H: H = L/N
t41
where L is the column length. Thus, the more efficient the column the smaller the value of H. While equation [4] provides a method by which to measure H for a given column, it provides no insight into the physical parameters which affect H. The discussion which follows presents the factors which affect H in simple, intuitive terms which are easily visualized and which are sufficient for understanding GC efficiency in a broad sense. However, for the rigorous derivation of the van Deemter equation and for a theoretical discussion of the Golay equation the reader is referred to the text by Perry (4) and to the work of Ogan and Scott (5) and of Sandra ( 6 ) . The van Deemter equation can be written as follows:
where A , B, and C will be discussed below and 1 ~ .is the average linear velocity of the carrier gas (the terms carrier gas and mobile phase will be used interchangeably) through the column. For packed columns, A describes the contribution to peak broadening which results from the various paths which different analyte molecules take as they migrate through and around the particles comprising the column packing. It should be noted that the A term is independent of the carrier gas velocity. In packed columns, the A term of equation 151 is minimized by using small particles of the narrowest possible size distribution. In capillary columns the A term of equation [ 5 ] vanishes, and, with slight modifications, equation 1 5 1 becomes the Golay equation ( 5 , 6 ) . For both packed and capillary columns, the B and C terms of equation [5] describe the variation of plate height (H) with mobile phase linear velocity. Although, as previously noted, the true situation is complex, to a first approximation the B and C terms can be considered to be identical in both packed and capillary columns.
81
Examination of equation [ 5 1 shows that the B term contribution to peak height decreases with increasing mobile phase linear velocity. Thus, B contains the factors which relate to analyte longitudinal diffusivity in the mobile phase. A s a group of analyte molecules traverses the column, they will tend to diffuse longitudinally through the mobile phase. This results in an increase in H since the analyte molecules are dispersed over a greater length of the column. Thus, the faster the mobile phase flows through the column the less time analyte molecules have to diffuse. Since dispersion due to the B term is related to diffusivity in the mobile phase, peaks spread less due to the B term when nitrogen is used as a carrier gas and the most when hydrogen is used as the mobile phase (noting that nitrogen, helium, and hydrogen are the most commonly used GC carrier gases). In order for any separation to take place on a GC column, the analytes must be absorbed into the stationary phase have different affinities for the stationary phase so that different migration rates occur. The C term of equation 1 5 1 , which increases with increasing carrier gas velocity, accounts for non-equilibrium between the stationary and mobile phases. This term behaves conversely to the B term in that increased diffusivity of the analyte in the mobile phase results in a smaller value for C. Thus, hydrogen carrier gas minimizes C. The C term also contains contributions due to the thickness of the stationary phase (either the layer coated on the particles in packed columns or on the wall of capillary columns), contributions due to the diameter of capillary columns, and contributions due to analyte diffusivity in the stationary phase. Plots of H vs p (see for example reference 6) describe a flattened hyperbola. For a given column, i. e. fixed stationary phase film thickness and column diameter, minimum H values are obtained with hydrogen as the carrier gas. However, due to safety considerations, a small sacrifice in ultimate achievable plates is usually made and helium is most often the carrier gas of choice. For capillary columns, the choice of column diameter and stationary phase film thickness is a matter of compromise. Ettre and co-workers ( 7 , 8 ) and Leclercq &. (9) have investigated the effects of film thickness and column diameter on efficiency. In general, thin films (i.e. 0.1 pm) and small column diameters lead to higher efficiencies. However, columns of larger diameter and with
82
thicker stationary phase films (or packed columns) have higher sample capacities. It is often falsely assumed that capillary columns are inherently more efficient than packed columns. However, procedures for producingpacked columns with greater than 3000 platesjmeter have been published (10). Well made capillary columns also provide on the order of 3000 plates/meter. However, due to pressure constraints, the length of packed columns is limited to approximately 3 - 4 meters. Capillary columns of 5 0 or 100 meters are routinely used. Thus, capillary columns are capable, if required, of providing 20-30 times the theoretical plates of packed columns. It is well to examine, in practical terms, what effects all of the above discussed parameters have on actual GC separations. Despite the considerable theoretical interest in efficiency, most laboratories will perform GC analyses on commercially obtained columns, either packed or capillary, operated at close to the optimum carrier gas velocity and the carrier gas will be helium. Therefore, the only way to gain significant numbers of theoretical plates will be to increase the column length. With capillary columns especially, this is easily done. However, it must be remembered that doubling column length will double the analysis time. It is also important to remember that R, varies as the square root of N so that doubling column length (and analysis time) results in only about a 40% increase in R,. Despite the time penalty, difficult separations can often be achieved only by increasing N (other methods of controlling R, are discussed later) and the wide use and availability of capillary columns has made many previously difficult to achieve separations routine. Two reports by Bata and co-workers (11,121 provide excellent examples of the use of increased efficiency to enhance a mycotoxin determination. In the first report (ll), deoxynivalenol (DON) was extracted from wheat and, after clean-up and derivatization, the extract was separated on a 1 2 m x 0 . 2 5 nun i.d. capillary column. The method could reliably determine DON to approximately 100 ppb. Although no interferences were noted under the analysis conditions (i.e. the DON peak was sufficiently resolved from all other components), Bata and associates were able to reduce the quantifiable level of DON to 5 0 ppb by carrying out the analysis on a 0.13 mm i.d. column ( 12) . By reducing the band width of the DON peak ( increasing
83
efficiency) the height of the peak was enhanced and the mass of DON required to produce a recognizable signal was reduced. 4.2.3 Retention The second variable in equation 121 which affects R, is the capacity factor, symbolized by k and defined k
=-R+t
- t
where tR is the retention time of the peak of interest and to is the retention time of an unretained compound. (At reasonably high temperatures the retention time of methane gives a sufficiently accurate estimation of to). Upon examination of equation [2], it is apparent that only one value of k is used to calculate R,, while R, refers to the separation between two peaks. For closely spaced peaks (the value of R, becomes irrelevant for widely separated peaks), an average value of k is used in equation [21. The value of k has a profound effect on R,. Equation [ 2 ] shows that, for k = 0, R, = 0. In other words, without retention no resolution is possible. A l s o , for values of k < 10, small increases in retention have large effects on R,. If k is increased from 0.5 to 5, R, increases 250%. However, a further increase in k from 5 to 10 results in only an additional 9% increase in R,. Thus, the standard recommendation is that k be adjusted so that peaks elute with values of k ranging from G. 1 - 1 0 (2). Values of k greater than about 10 result in much longer analysis times with minimal improvements in R,. If k is expressed as in equation [71, ( 1 3 )
171 where C, is the concentration of analyte in the stationary phase, C, is the concentration of the analyte in the mobile phase, d, is the stationary phase film thickness, and r is the column radius, the factors which affect k are evident. As the temperature is increased, C, decreases and C, increases so that the ratio CJC, becomes smaller and k decreases. Thus, retention (k) is inversely proportional to temperature and k can be controlled by varying the separation temperature. Alternatively, equation [ 7 ] suggests that the ratio of stationary phase film thickness to column diameter affects k.
84
Since most mycotoxins are large molecules with numerous polar functional groups (14), the usual problem in GC separations of these compounds is one of too much retention ( k > 10) rather than too little. The first step in achieving a value of k<10 is to increase the separation temperature. However, most GC stationary phases have upper temperature limits of 250-350° C before they begin to decompose. If the mycotoxin of interest cannot be eluted with a reasonable value of k at the upper temperature limit of the stationary phase the ratio d,/r (equation 171) must be decreased. This can be accomplished by utilizing a column with a thinner stationary phase film or by increasing the column diameter. An alternative way to affect C,/C, (other than temperature) is to change the nature of the stationary phase so that the affinity of the analyte for the stationary phase is altered. This is discussed in more detail in the next section. 4.2.4 Selectivity The final factor of equation 121 by which the chromatographer can affect resolution is a , or the selectivity. Selectivity is defined as a =
LL kl
181
where k, and k, are the capacity factors for adjacent peaks and, by convention, k, is always greater than k, (i.e. a 2 1). Selectivity arises due to different chemical interactions between analytes and the GC stationary phase. Although these chemical interactions can be affected by temperature, the usual way of altering a is by changing the chemical nature of the stationary phase. In packed column GC, the stationary phase is deposited as a thin film over small particles of an inert material such as diatomaceous earth. In capillary GC, a thin film of stationary phase is deposited on the inner wall of the capillary column. Many phases used for capillary GC can be bonded, or cross-linked,by adding a small amount of organic free radical producers such as peroxides or azo-compounds (15-17) or by exposure to gamma radiation (18,191. Such bonded phases exhibit greater thermal stability, less bleed, and can even be rinsed with solvents in order to remove contaminants accumulated from many injections. Although there are literally hundreds of stationary phases available, only a few have found wide use in capillary GC. Leary and
85
co-workers (20) used statistical techniques to suggest a list of 12 phases which cover the entire range of selectivities available. Klee and co-workers (21) have published an excellent account of the factors from which selectivity arises. The most popular, and least selective, phases are probably the methyl silicones. These phases, consisting of high molecular weight silicone polymers substituted with methyl groups, are stable, easily cross-linked, and are easily coated to yield high efficiency columns. However, little chemical interaction occurs between analytes and the methyl silicone polymer so that separations are largely based solely on volatility. Molecules with very different functional groups but with similar boiling points are thus difficult to separate on the methyl silicones. Numerous phases are available where 5 - 7 5 % of the methyl groups on the silicone backbone are replaced with phenyl groups. The introduction of the phenyl groups with their polarizable electron clouds can enhance the selectivity obtainable for some separations. For example, the separation of a hydrocarbon from an aromatic molecule would be expected to be enhanced since the aromatic molecule would experience a specific interaction with the phenyl groups of the stationary phase. Polyglycol phases, with the capability of hydrogen bonding, yield still different selectivity. On these phases, analytes with hydroxyl, amine, or acid groups would be expected to be retained preferentially relative to non-polar molecules. The choice of stationary phase for a given separation is complex and is often based (successfully) on chemical intuition. However, in cases where a difficult separation cannot otherwise be achieved, several quantitative methods of stationary phase selection are available. McReynolds ( 2 2 ) has characterized phases based on a variety of chemical properties. Laub and Purnell ( 2 3 , 2 4 ) proposed a method for determining what mixture of phases would lead to optimum selectivity. Jones and Purnell ( 2 5 ) have demonstrated a method for predicting retention (and thus selectivity) in serially linked capillary columns containing different stationary phases. Even in cases where "good" selectivity is obtained and sufficient resolution is available on the first stationary phase used for a separation it is sometimes useful to use a second stationary phase with different selectivity in order to aid in confirmation of peak purity and identity. Davis and Giddings ( 2 6 ) have developed the
86
theory necessary to predict the peak capacity, or maximum number of peaks resolvable without coincident elution, for a given column. According to their calculations, the potential for co-elution of peaks is surprisingly high even for capillary columns generating high numbers of theoretical plates. Quantitation is obviously inaccurate if an unexpected peak coelutes with the peak of interest. Also, peak identification based on common retention time of a standard and a peak in the sample chromatogram, is, in the absence of a structure detector such as a mass spectrometer, not definitive. Additional evidence for the purity and identity of a given peak is obtained if the sample is run on two or more columns of widely different selectivity and both quantitation and standard retention times match on both columns. Several options exist for chromatography on different columns. Obviously, and this is the only practical option with packed columns, the sample can be chromatographed on one column. Then another column of different selectivity is installed and the sample rechromatographed. With flexible fused silica capillary columns, two columns can be inserted into the same injection port and routed to different detectors. This allows simultaneous chromatography (thus saving time) but requires two detectors. Gupta and Nikelly ( 27 ) have described a novel method for using a single detector and injector to perform simultaneous chromatography on two different columns. Numerous workers have exploited the availability of stationary phases of differing selectivities in the GC determination of mycotoxins. The work of Per0 and Harvan (28) provides an excellent example of the use of different stationary phase composition to alter selectivity. They examined the separation of Aspergillus sp., Alternaria sp. and Penicillium sp. fungal metabolites on four methyl silicone stationary phases substituted with 0%, 35%, 50%, and 75% phenyl groups. The 75% phenyl methyl silicone failed to resolve 5 of the 12 compounds studied. The 0% and 35% phenylmethyl silicones resolved all but two compounds (different pairs of compounds were unresolved on the different phases). Only the 50% phenyl methyl silicone was able to resolve all twelve compounds examined. 4.3.
EXTRACOLUMN RESOLUTION The preceding discussion has focused on the resolution of peaks as it occurs within (or on) the GC column. However, as was previously pointed out, there is a finite limit to the number of
peaks which a GC column is capable of resolving ( 2 6 ) . In consideration of the complex matrices in which mycotoxins occur, it is almost always advantageous t.o reduce the demand placed on the column for peak separation. We will refer to resolution which occurs due to factors other than efficiency, retention, or selectivity on the GC column as extracolumn resolution. Several alternatives exist for achieving extracolumn resolution. Perhaps the most obvious is to limit the number of compounds, and thus peaks, which the column will need to resolve. This objective can often be realized by various extraction or clean-up steps prior to GC analysis. Another means of obtaining extracolumn resolution is through the use of a detector which is specific for only the compound of interest (or as few compounds other than the one of interest as possible). In this way even if interferences are unresolved from the target analyte the detector generates no response due to the interference and the analyte can be accurately quantitated. 4.3.1 Resolution throuqh sample clean-up Generally, sample preparation for mycotoxin analysis includes at least two steps; extraction and clean-up. Ideally, the extraction solvent would remove only the mycotoxin of interest from the matrix. In the absence of such a completely specific extraction solvent, the solvent should be chosen so as to remove as much of the mycotoxin as possible while removing as little as possible of interfering compounds. In the case of the acidic mycotoxins, it is often possible to use acid/base extractions to effect considerable clean-up. Mirocha and co-workers (29) extracted zearalenone from grains using ethyl acetate. After evaporating the ethyl acetate, the residue was taken up in chloroform and partitioned with aqueous sodium hydroxide. The sodium salt of the deprotonated zearalenone was thus extracted into the aqueous phase. All non-acidic compounds which were extracted from the grab by the ethyl acetate were thus removed by this single step. Acidification of the zearalenone containing aqueous phase resulted in re-protonation of the zearalenone. The once again organic soluble mycotoxin was then extracted back into chloroform and subjected to subsequent clean-up and then to GC analysis. A similar acid/base extraction of blood serum allowed zearalenone in the serum to be determined by GC with no clean-up save for the acid/base partitioning (30).
88
Penicillic acid is another acidic mycotoxin for which numerous workers have employed acid/base partitioning for at least preliminary clean-up. For details and examples see (31). Following extraction fromthe matrix (and acid/base partitioning when applicable) most sample extracts require further clean-up prior to the GC analysis step. Column liquid chromatography (LC), either on laboratory prepared columns or on pre-packed commercially available columns, is often employed for this additional clean-up step. Especially when commercial columns are used this technique is alternatively referred to as solid-phase extraction (SPE). Clean-up of extracts on SPE columns can be approached from two perspectives. The extract can be applied to the column and a solvent of appropriate strength is used to elute the mycotoxin of interest while interferences remain on the column. Conversely, a solvent can be used to remove interferences after which a different solvent is then used to elute the mycotoxin. A combination of these techniques is also often used; i.e. an initial solvent removes some interferences and then an additional solvent is used to elute the mycotoxin while leaving other interfering substances on the column. SPE columns are available (or can be prepared in the laboratory) with a wide range of normal phase (polar) packings including silica gel, alumina, Florisil and bonded organic nitriles or amines. Nonpolar packings (often referred to as reversed-phase) consisting of octyl-, octadecyl-, or phenyl-silanes bonded to silica gel are also available. Various ion-exchange resins can also be used. The choice of the optimum SPE column packing and solvent or series of solvents is often a matter of trial and error. Obviously, the ultimate goal is to isolate the desired mycotoxin completely from all interferences. Selectivity in liquid chromatography, while arising from the same fundamental factors as selectivity in GC, can be manipulated to a much greater degree than in GC. The primary reason for the greater ability to alter selectivity in LC lies in the mobile phase. In GC, alterations in the mobile phase (carrier gas) result in essentially no change in selectivity. In LC, where the solvent as well as the stationary phase participate in the interactions which result in selectivity, a much larger range of selective interactions is possible. The classic text by Snyder and Kirkland (2) describes the ways in which LC selectivity arises and present general guidelines on the choice of solvents and stationary phases to optimize selectivity in LC.
89
Rood and co-workers have described GC methods for the determination of trichothecene toxins in feeds ( 3 2 ) and in urine and plasma (33) which used extensive LC clean-up. For the feeds ( 3 2 ) , extracts in acetonitrile-water were applied to a charcoal-aluminaCelite column and the toxins (including, among others diacetoxyscirpenol ( D A S ) , deoxynivalenol ( D O N ) and T-2 toxin) were eluted with acetonitrile-water. The eluate was evaporated, redissolved in ethyl acetate-methanol, applied to a Florisil column and the toxins were eluted with ethyl acetate-acetone. This eluate was evaporated, hydrolyzed, applied to a silica gel column, and then the toxins were eluted with methylene chloride-acetone. This extensive clean-up routine resulted in chromatograms almost free of interfering substances when soybeans, corn, and mixed feeds were analyzed by GC using a large diameter (0.53 m m ) , thick film ( 1 w ) capillary column. Analysis of blood plasma and urine ( 3 3 ) also used a complex clean-up scheme. T-2 toxin, DAS, DON, and nivalenol were extracted from the blood and urine by passing the sample through an octadecylsilane SPE column. The toxins were thus extracted into the non-polar stationary phase of the SPE column. Elution with methanol removed the mycotoxin fraction which was, after evaporation of the methanol and re-dissolution in ethyl acetate, applied to a Florisil column. Ethyl acetate was used to elute the toxins from the Florisil column and the evaporated eluate was hydrolyzed. The hydrolyzate was then further cleaned-up on a silica gel SPE column. GC analysis (after derivatization) on 6 ft. packed columns resulted in interference free chromatograms. An alternative to column chromatography for preliminary cleanup is thin-layer chromatography (TLC). In TLC, the sample is applied to a plate coated with a thin layer of sorbent and the plate is "developed" by submerging one edge in a pool of solvent so that capillary action causes the solvent to migrate up the plate. TLC plates are available with coatings covering nearly the entire range of types available for SPE columns (with the possible exception of ion-exchange materials), and solvents of many types are usable so that TLC offers a wide range of selectivity. In order to serve as a preliminary clean-up to GC, the area of sorbent containing the mycotoxin of interest must be scraped from the TLC plate. A strong solvent is then added to the scrapings and the resulting slurry is filtered. The filtrate can then be subjected to
90
derivatization or further clean-up prior to injection onto the GC column. TLC clean-up affords the opportunity to more positively identify, or confirm the identity of, mycotoxins. Coincidence of mobility of a standard and sample on a TLC plate followed by similar coincidence of retention times of standard and sample upon GC analysis serves to increase the confidence of peak identity. Also, if a non-destructive detection method (such as fluorescence) is available, the mycotoxin can be quantitated on the TLC plate prior to scraping for recovery and subsequent GC quantitation. Nesheim and Trucksess ( 3 4 ) have described the practical aspects of TLC of mycotoxins. Trucksess and co-workers (35) used sequential SPE and TLC prior to determining aflatoxins by GC/mass spectrometry (GC/MS). Corn or peanut butter was extracted with aqueous methanol and the aflatoxins were partitioned into chloroform. The chloroform was evaporated and the residue redissolved in methylene chloride. The methylene chloride was then added to a silica-gel SPE column which was eluted sequentially with hexane, diethyl ether, and methylene chloride. Aflatoxins were then eluted with chloroform-acetone. The eluate containing the aflatoxins was evaporated and the residue redissolved in chloroform. A portion of the chloroform was spotted on a silicagel TLC plate which was then developed with chloroform-acetone. The zone corresponding to af latoxin B, was removed and the af latoxins eluted from the silica with chloroform-acetone-isopropanol. After solvent evaporation, the residue in acetone was subjected to GCIMS with negative ion chemical ionization using a capillary GC column. No interferences were noted in the mass spectrum of aflatoxin B, which resulted. Patulin and penicillic acid havebeen determined by GC following TLC clean-up of sample extracts (36). Grains containing the mycotoxins were extracted with methanol-5% aqueous sodium chloride. The extracts were defatted with hexane and the methanol was removed by evaporation. The remaining saline solution was extracted with ethyl acetate to remove the toxins. The ethyl acetate extract was then evaporated and the residue dissolved in benzene-methanol. A portion of the benzene-methanol extract was applied to a silica-gel TLC plate which was then developed with benzene-methanol-acetic acid. The area of the plate containing the toxins was removed (patulin and penicillic acid migrated the same distance in the developing solvent
91
used) and the toxins were eluted with benzene-methanol. Following derivatization, the mycotoxins were chromatographed on packed GC columns with a variety of stationary phases. 4 . 3 . 2 Chemical derivatization Since most mycotoxins are large, polar molecules containing numerous polar functional groups, they are typically only very slightly volatile. Since a requirement of GC is that the analytes be vaporized at a temperature which does not result in decomposition of either the mycotoxin of interest or the GC stationary phase, the low volatility must be circumvented. Conversion of hydroxyl, amino, or acid groups to esters, amides, ethers, etc. will reduce hydrogen bonding and thus increase volatility and often increase thermal stability. Some of the many reagents and techniques available to effect the derivatization of various functional groups have been previously described (37). Of particular note to the present discussion is the use of halogenated reagents for derivatization of alcohols, amines, and acids. Alcohols and phenols can be converted to trichloroacetates or trifluoroacetates by r e a c t i o n w i t h t r i c h l o r o a c e t i c or trifluoroacetic anhydride. Perfluoroesters can be formed reaction with pentafluoropropionic or heptafluorobutyric anhydride. Similarly, amino groups can be converted to the halogenated amides by reaction with the halo-anhydrides. Reaction of carboxylic acids with pentafluorobenzyl bromide in basic medium results in the formation of pentafluorobenzyl esters. As will be discussed in the next section, the ability to form halogenated compounds from mycotoxins can tremendously increase the selectivity and sensitivity of the analytical method. 4 . 3 . 3 Resolution through detection The ideal detector for any GC determination would respond to the analyte of interest while producing no response due to other compounds present in the sample. The desired response would, in addition, result from the smallest possible amounts of the analyte. In other words, the detector would be both selective and sensitive. As the selectivity of a given detector increases the demand for resolution in the extraction and clean-up steps and in the chromatographic step decreases. In fact, with the ideal detector which responds only to analyte, no chromatographic step would be required. Thus, selectivity in the detector can be considered
92
equivalent to resolution on the GC column. Although the ideal detector obviously does not exist, efforts towards approaching the ideal are often rewarded in less stringent requirements for sample clean-up and chromatographic resolution. The most common GC detector (especially with capillary columns) is the flame ionization detector (FID) The FID works by burning the column eluate in a hydrogen/air flame and collecting the resulting ions on a charged electrode. The resulting current is amplified to yield the output signal. A well designed FID can detect as little as picograms/sec of some compounds (38). The FID is a rugged, sensitive, easy to operate detector. However, it is not a selective detector and produces a response for nearly all organic compounds (some small, highly halogenated or otherwise "nonflammable" compounds such as carbon disulfide or carbon tetrachloride produce negligible FID signals). Despite its lack of selectivity, the FID has seen extensive use in the determination of mycotoxins. An example is the rather unique determination of fumonisin B, in corn (39). Funonisin B1 contains two tricarballyic acid (TCA) groups joined by ester linkages to the backbone of the molecule. In a portion of corn extract the TCA was hydrolyzed from the backbone by 6 fl HC1 and esterified with isobutyl alcohol/HCl. Capillary chromatography with detection by FID resulted in detection of the isobutylester of tricarballyic acid but not of the parent backbone molecule. Although numerous other peaks were present in the chromatogram, the resolution of the column and the sensitivity of the detector was sufficient for quantitation to less than 500 ppb FBI in corn. The electron capture detector (ECD) is considerably more selective and can be more sensitive than the FID. The ECD measures the current produced by the ionization of carrier gas molecules by a B-radiation source. This continuous electrical signal is known as the standing current. When a compound with sufficient electron affinity passes through the detector, some of the electrons are "captured" and less carrier gas is ionized. A reduction in standing current results. Since the reduction in standing current is proportional to the amount of compound present in the detector, the reduction is measured and can be related to the amount of analyte. Compounds such as hydrocarbons which have very low electron affinities produce very little signal in an ECD. Conversely, the ECD is very sensitive to molecules with high electron affinities such as
.
93
halogenated compounds. The response of an ECD to a tri-brominated molecule can be as much as 1 million times as great as the response to a hydrocarbon (40). Ultimate sensitivity can range to as low as femtograms/sec for polyhalogen compounds (38). A recent report (33) found that, for some trichothecene toxins, forming the pentafluoropropionyl derivative resulted in approximately three times greater sensitivity than was obtained with trifluoroacetyl derivatives when using an ECD. Within the range of halocompounds to hydrocarbons, the ECD also responds to a greater or lesser degree to other types of compounds based on electronegativity, molecular size, conjugation, symmetry and bond strengths (4). For example the presence of nitrogen, oxygen, sulphur, or silicon can result in enhanced ECD response relative to hydrocarbons. In mycotoxin analysis, the ability to derivatize the toxins plays an especially important role when utilizing an ECD detector. If a derivatizing reagent can be found that both increases volatility (and thus suitability for GC analysis) and electron affinity, the potential for considerable enhancement of resolution exists. As was previously discussed, halogenated reagents capable of reacting with functional groups found in many mycotoxins are readily available. I n cases where the derivatizing reagent reacts relatively exclusively with the mycotoxin and not with other components of the extract matrix, the full advantages of the ECD are realized. A report by Kamimura and co-workers (41) on the determination of some Fusarium toxins in barley illustrated the additional resolution obtainable with an ECD relative to an F I D . After extracting the toxins and converting to the trimethysilyl derivatives, they chromatographed the resulting solution on packed columns using each detector. With the F I D , DON and nivalenol ( N I V ) (at 1.8 and 1.4 pg/g respectively) were present as small, barely distinguishable shoulders surrounded by other peaks on a much larger peak. With the ECD, both DON and N I V were completely resolved from each other and from all other peaks. The ECD peak heights were at least ten times the height of the F I D peaks. This analytical method illustrates how, despite the fact that the ECD was not operating at maximum sensitivity and selectivity since halogenated derivatives were not used, the choice of detectors can greatly influence the resolution which is attainable.
94
When a mass spectrometer (MS) is used as a GC detector, the combination of techniques is referred to as GC/MS. A mass spectrometer is close to being the ideal GC detector. It can provide sensitivity, selectivity, and, unlike either the FID or ECD, can provide structural information or confirmation. In mass spectrometry, a molecule is ionized with electrons or impact with charged reagent molecules. The ions are accelerated in an electric field and separated according to their mass to charge ratio by varying the magnetic field to which the moving ions are subjected. The mass and relative amounts of analyte ions formed from the ionization process is, in principle, unique to the structure of the analyte. In the case of analytes of unknown structure, the mass spectrum can be used to help deduce the molecular weight and structure of the molecule. McLafferty (42) has authored one of the classic texts on the use of mass spectra in structure determination. In the field of routine mycotoxin determinations, the MS is primarily used for peak identity confirmation. If a peak occurs at the proper retention time and the mass spectrum obtained for the peak matches the known spectrum for the compound the identity of the peak is then known to a high degree of confidence. Alternatively, a computerized library of known spectra can be assembled. Then, the spectra of unknown peaks eluting from the GC column can be compared to the library and the identity of the peak deduced. A technique known as select ion monitoring (SIM) can be used to increase the selectivity and sensitivity of the MS detector. When using SIM, the mass spectrometer monitors a single ion mass rather than scanning a range of masses. The mass which is monitored is usually one which is known to occur at high intensity in the spectrum of the analyte. By monitoring only a single mass for the entire time that a peak elutes from the GC column the sensitivity can be increased. Selectivity is improved in SIM by choosing a mass to monitor which is unique to the analyte. In this way interfering compounds which might co-elute with the analyte are completely "invisible" to the detector. A thorough discussion of GC/MS is far beyond the scope of this chapter. Aside from the interpretation of spectra and myriad techniques for obtaining spectra and quantitation, the interfacing of the MS to the GC instrument is a complex topic. The reader is
95
referred to the texts by Message ( 4 3 ) and McFadden ( 4 4 ) for practical, well-written overviews of the subject. GC/MS has been utilized extensively in the field of mycotoxin determinations. The Fusarium sp. toxins, especially the trichothecenes, have been widely determined by GC/MS. The work by Rood and associates (32,33), used mass spectrometry with negativeion chemical ionization (NICI) to confirm trichothecenes in feeds and animal body fluids. Detection levels of the trichothecenes investigated were as low as 10 pg in feeds (32) and 10 ng/mL in body fluids ( 3 3 ) by GC/MS-NICI. It was suggested the GC/MS-NICI always be used to confirm GC-ECD results for low levels of toxins. Krishnamurthy and co-workers ( 4 5 ) also determined some trichothecenes using GC/MS-NICI. They pointed out the need for derivatization to prevent degradation and to improve volatility and detected and quantitated as little as picogram amounts of the toxins after GC on a capillary column using MS-NICI with SIM. The three detectors discussed so far, the FID, ECD and MS, have been used for nearly all reported mycotoxin analytical methods. However, other types of detectors are available ( 4 0 ) and could potentially be useful in mycotoxin analytical schemes. The nitrogen-phosphorus detector (NPD) and flame photometric detector (FPD) are element specific detectors. The NPD operates similarly to an FID except that the flame conditions are such that the ionization of hydrocarbons is minimized. A bead coated with an alkali metal salt (Rb', CS', Na', K') is positioned in the flame and causes, depending on flame conditions and the salt, compounds containing N or P to selectively ionize. The NPD can detect less than 1 pg of some nitrogen or phosphorus containing compounds and exhibits selectivities as high as 100,OOO:l relative to hydrocarbons (40). The FPD burns the column eluate in a flame and measures the radiation emitted from sulphur or phosphorus containing compounds. FPD sensitivity can range to the low picogram range and selectivities relative to hydrocarbons are on the order of 10,OOO:l (40). Both the NPD and FPD are potentially useful mycotoxin analytical tools either for toxins that contain nitrogen, phosphorus or sulphur or for toxins that can be derivatized with reagents containing N, P or S. Infrared spectrophotometers have also been utilized as GC detectors. The infrared detector (IRD) yields specific structural information as does the MS. The IRD is probably at least an order of magnitude less sensitive than a MS and interpretation of IR
96
spectra is less apt to provide complete structural information than MS. However, IR detection is non-destructive so that an IRD can be coupled in-line prior to another detector, i.e. an MS. The information derived when I R D is serially coupled to an MS when combined with retention time information can provide nearly conclusive confirmation of analyte identity. CONCLUSIONS The ultimate goal of any analytical method is the precise and accurate detection, confirmation and quantitation of the target analyte. In order to achieve these goals the method must be capable of resolving the signal due to analyte from signals due to interferences. In GC, this resolution can arise from separation on the GC column, separation and clean-up prior to the gas chromatographic process, or through detection methods which respond selectively to the analyte. With analytes such as the mycotoxins which occur in complex matrices containing many potential interferences, some contribution to the ultimate resolution must be made by each of the processes. A report by Visconti and Palmisano (46) illustrates the difficulty often encountered in positively identifying a mycotoxin in a complex matrix. Samples of corn infected with Fusarium solani were extracted, cleaned-up on an SPE column and further cleaned-up by TLC. After this extensive clean-up, GC analysis of the silylated extract yielded an FID peak whose retention time matched that of standard DON. However, based on other techniques, g . solani was thought to be a non-producer of DON. Further investigation, including polarographic analysis and use of GC/MS, showed that the suspect peak was not due to DON. To quote the authors "This should focus attention on the complexity , and on the problems of erroneous identification which can arise with complex matrices such as fungal-contaminated foodstuffs. The use of different analytical techniques is always recommended for solving such particularly complex problems. I' Despite the complexity of his task, the mycotoxin analyst who perseveres and utilizes the powerful tools available will recognize the validity of the Latin phrase: "Magna est veritas & praevalebit." 4.4.
..
..
REFERENCES 1 P. Sandra, J. High Res. Chromatogr., 12 (1989) 82-86.
...
97
2 3 4 5 6 7 8 9 10
11
12 13 14 15 16 17
18 19 20 21
L. R. Snyder and T. J. Kirkland, Introduction to Modern Liquid Chromatography, Wiley, New York, 1 9 7 9 . B. L. Karger, L. R. Snyder and C. Horvath, An Introduction to Separation Science, Wiley, New York, 1 9 7 3 . J. A. Perry, Introduction to Analytical Gas Chromatography, Marcel Dekker, New York, 1 9 8 1 , pp. 8 2 - 1 1 3 . K. Ogan and R. P. W. Scott, J. High Res. Chromatogr. , 7 ( 1 9 8 4 ) 382-388.
P. Sandra, J. High Res. Chromatogr., 1 2 L. S. Ettre, Chromatographia, 1 7 ( 1 9 8 3 ) L. S. Ettre, G. L. McClure, and Chromatographia, 17 ( 1 9 8 3 ) 5 6 0 - 5 6 9 . P. A. Leclercq and C. A. Cramers, J. High 8 (1985) 764-771.
(1989) 273-277. 553-559.
J.
D.
Walters,
Res. Chromatogr. ,
R. J. Leibrand and L. L. Dunham, Research/Development (Hewlett-Packard) ( 1 9 7 3 ) 2 4 - 3 8 . A . Bata, A. Vanyi, and R. Lasztity, J. Assoc. Off. Anal. Chem., 6 6 ( 1 9 8 3 ) 5 7 7 - 5 8 1 . A. Bata, A. Vanyi, R. Lasztity, and J. Galacz, J. Chromatogr., 2 8 6 ( 1 9 8 4 ) 3 5 7 - 3 6 2 . W. Jennings, J. Chromatogr. Sci., 2 8 ( 1 9 9 0 ) 3 8 5 - 3 8 8 . R. J. Cole and R. H. Cox, Handbook of Toxic Fungal Metabolites, Academic Press, New York, 1 9 8 1 . V. M. Gandara, J. Sanz and I. Martinez-Castro, J. High Res. Chromatogr., 7 ( 1 9 8 4 ) 4 4 - 4 5 . S. R. Springston, K. Melda and M. V. Novotroy, J. Chromatogr., 2 6 7 ( 1 9 8 3 ) 3 9 5 - 3 9 8 . H. Liu, A. Zhang, Y. Jin and R. Fu, J. High Res. Chromatogr., 12 ( 1 9 8 9 ) 5 3 7 - 5 3 9 . M. A. Basso, M. J. T. F. dos Santos, K. E. Collins, and C. H. Collins, J. High Res. Chromatogr., 1 2 ( 1 9 8 9 ) 5 0 0 502.
J. A. Hubball, P. R. DiMauro, E. F. Barry, E. A. Lyons and W. A. George, J. Chromatogr. Sci., 2 2 ( 1 9 8 4 ) 1 8 5 - 1 9 1 . J. J. Leary, J. B. Justice, S. Tsuge, S. R. Lowry and T. L. Isehour, J. Chromatogr. Sci., 11 ( 1 9 7 3 ) 2 0 1 - 2 0 6 .
22 23
M. S. Klee, M. A. Kaiser and K. B. Laughlin, J. Chromatogr., 2 7 9 ( 1 9 8 3 ) 6 8 1 - 6 8 8 . W. 0. McReynolds, J. Chromatogr. Sci., 8 ( 1 9 7 0 ) 6 8 5 - 6 9 1 . R. J. Laub and J. H. Purnell, Anal. Chem., 4 8 ( 1 9 7 6 )
24
R. J. Laub and J. H. Purnell, Anal. Chem. 4 8 ( 1 9 7 6 )
25
J. R. Jones and J. H. Purnell, Anal. Chem., 6 2 ( 1 9 9 0 ) 2 3 0 0 -
26
J. M. Davis and J. C. Giddings, Anal. Chem., 5 5 ( 1 9 8 3 ) 4 1 8 -
27
P. K. Gupta and J. G. Nikelly, Abstracts of Papers, 200th American Chemical Society National Meeting, Abstract #ANYL
28
R. W. Per0 and D. Harvan, J. Chromatogr., 80 ( 1 9 7 3 ) 2 5 5 -
29
C. J. Mirocha, B. Schauerhamer and S. V. Pathre, J. Assoc. Off. Anal. Chem., 5 7 ( 1 9 7 4 ) 1 1 0 4 - 1 1 1 0 . H. L. Trenholm, R. Warner and E. R. Farnworth, J. Assoc. Off. Anal. Chem., 6 3 ( 1 9 8 0 ) 6 0 4 - 6 1 1 . C. W. Thorpe and R. L. Johnson, J. Assoc. Off. Anal. Chem. ,
30 31 32
799-803. 1720
-
1724.
2305.
424.
166.
258.
4 ( 1 9 7 4 ) 861-865.
H. D. Rood, Jr., W. B. Buck and S . P. Swanson, J. Assoc. Off. Anal. Chem., 7 1 ( 1 9 8 8 ) 4 9 3 - 4 9 8 .
98
33 34 35 36 37 38
39
40 41 42 43 44 45 46
H. D. Rood, Jr., W. B. Buck, and S. P. Swanson, J. Agric. Food Chem., 36 (1988) 74-79. S. Nesheim and M. W. Trucksess, in R. J. Cole (Editor), “Modern Methods in the Analysis and Structural Elucidation of Mycotoxins,” Academic Press, Orlando, 1986, p. 240-264. M. W. Trucksess, W. C. Brumley and S. Nesheim, J. Assoc. Off. Anal. Chem., 67 (1984) 973-975. T. Fujimoto, T. Suzuki and Y. Hoshino, J. Chromatogr., 105 (1975) 99-106. D. R. Knapp, Handbook of Analytical Derivatization Reactions, Wlley, New York, 1979. H. V. Drushel, J. Chromatogr. Sci. 21 (1983) 375-384. E. W. Sydenham, W. C. A. Gelderblom, P. G. Thiel, W. F. 0. Marasas, J. Agric. Food Chem., 38 (1990) 285-290. R. Buffington and M. K. Wilson, Detectors for Gas Chromatography - A Practical Primer, Hewlett-Packard Corporation, Avondale, PA, 1987. H. Kamimura, M. Nishijma, K. Yasuda, K. Saito, A. Ibe, T. Nagayama, H. Ushiyama and Y. Naoi, J. Assoc. Off. Anal. Chem., 64 (1981) 1067-1073. F. W. McLafferty, Interpretation of Mass Spectra, 2nd ed., W. A. Benjamin Inc., London, 1973. G. M. Message, Practical Aspects of Gas Chromatography/Mass Spectrometry, Wiley, New York, 1984. W. McFadden, Techniques of Combined Gas Chromatography/Mass Spectrometry, Wiley, New York, 1973. T. Krishnamurthy, E. W. Sarver, S. L. Greene and B. B. Jarvis, J. Assoc. Off. Anal. Chem., 70 (1987) 132-140. A. Visconti and F. Palmisano, J. Chromatogr., 252 (1982) 305-309.
99
CHAPTER 5
EMERGING TECHNIQUES: IMMUNOAFFINITY CHROMATOGRAPHY A. A. G. Candlish and W. H. Stimson
5.1.
INTRODUCTION
Several methods for the detection and quantification of mycotoxins have been developed since the discovery of these toxic fungal metabolites in the early sixties. In general , there are two approaches adapted for mycotoxin determination. These can be described as the following: 1.
Physicochemical methods, which include thin layer chromatography (TLC), high performance liquid chromatography (HPLC), gas-liquid chromatography (GLC) and mini-columns.
2.
Bioassay systems, which covers biological methods such as toxicity testing on mammalian cells and mutagenic effects on bacteria. However, since 1976 when the first immunoassays were described for aflatoxin B1 (AFBi), there has been rapid development of these assays for mycotoxin detection. The initial stages involve the production of antibody, but as mycotoxins are low weight molecules they are not immunogenic. However, following conjugation to a protein carrier the toxins may be used for immunisation to induce antibody product ion.
Using this approach a number of antibodies have been developed to mycotoxins such as the aflatoxins (AFs) ochratoxin A (OTA), zearalenone, T-2 toxin, and many more. With the availability of these antibodies, simple and rapid radioimmunoassays (RIA), enzyme-linked immunosorbent assays (ELISA), and immunoaffinity chromatography (IAC) columns have been developed for the determination of these mycotoxins in food samples, animal feed samples and also within biological fluids such as milk and urine. These techniques comply with the following criteria: 1.
Good sensitivity with levels of detection in the parts per billion (ppb) = pg of mycotoxin / Kg of sample and even in some cases as for AFMl in milk samples in the parts per trillion (ppt).
2.
Highly specific due to the nature of the interaction between antibody and antigen. The binding of antibody with mycotoxin is highly specific with high binding constants and can show little or no cross-reactivity with other metabolites or compounds that may be found in the sample extract.
100
3. Minimum sample preparation prior to the use of immunoassay systems. 4. Wide range of samples can be tested 5. Easy to perform by unskilled personnel.
6. Results can be obtained within a short period of time, usually less than one hour.
Once a useful antibody has been produced with the required characteristics then a choice must be made as to which of the three immunoassay types are to be developed. The RIA has the advantage of being very specific and is often the method of choice in clinical laboratories. However, the RIA has the disadvantages of requiring radioactive labelled reagents, expensive equipment for monitoring radioactivity and the disposal of radioactive waste. In general this technique has not been well accepted for mycotoxin detection although the first publication describing antibodies to AFBl in 1976 by Langone and van Vunakis (1) used a RIA. Conversely, the ELlSA has become widely accepted and used for the determination of mycotoxins. Many different laboratories have been actively researching this technique for mycotoxins; in addition, various commercial companies have developed products. The main advantages of this technique is its simplicity, ease of use and ability to be adapted into various formats such as 96 well microtitre plates, which allow large numbers of samples to be tested within a working day. The main disadvantages are that the method is generally only semi-quantitative and accurate quantitative results can only be obtained if complex standard dose-response curves are continually constructed. Perhaps one of the most applicable and adaptable procedures for mycotoxin detection is the IAC column. This method is simple, robust and can be used as a semi-quantitative method for the screening of mycotoxins or it can be coupled with physicochemical equipment such as HPLC for the complete and accurate estimation of mycotoxin quantities. Some of the advantages of this test are as follows: A test can be used at three levels depending on the requirement and facilities of the individual laboratories. A rapid screening assay can be developed using FlorisilB tips with the IAC columns requiring only an ultraviolet box to estimate the AFs in the tip. For more accurate results the AFs can be estimated using a fluorimeter which produces a digital reading of the level of toxin. Finally if complete accuracy and precision of toxin level and content is required then the IAC column may be linked to an HPLC.
101
5.2.
IMMUNOAFFINITY CHROMATOGRAPHY THEORY
IAC is a powerful and specific technique used for the separation and/or purification of biological compounds. The desired material binds specifically and reversibly to a ligand which has been fixed to an inert carrier (2). Ligands range from small molecules such as substrates for enzymes or mycotoxins to large molecules such as protein hormones or even whole bacterial cells. The interaction of a binding site with a ligand is determined by the overall size and shape of the ligand as well as the number and distribution of complementary surfaces. These complementary surfaces may involve a combination of charged and hydrophobic moieties and exhibit other short range molecular interactions such as hydrogen bonds and Vander Waals forces (3). The practical aspects of IAC development involves the following steps: (1) Choice of an appropriate ligand. (2) lmmobilisation of the ligand onto a support matrix.
(3) Contact of mixture for separation with the support matrix. (4) Removal of non-specific ally bound compounds. (5) Elution of the compound of interest in a purified form.
When selecting a particular ligand, there are five characteristics to consider. A. SDecificity: the ligand should recognise only the compounds to be purified. In general, group specific ligands allow purification of related compounds or compound families eg AFBl, B2, G I ,G2 and M I . 6.Reversibility: the ligand should form a reversible complex with the compound of interest in such a way that the complex does not dissociate under application of samples or washing buffers but easily dissociates without the need for denaturing the ligand. C. Stability: the ligand should be stable to the conditions used for immobilisation as well as the conditions of use. D. Size: the ligand should be large enough such that it contains several groups able to interact with the compound of interest resulting in sufficient stereoselectivity and affinity. E. Afflnitv: the energy with which the combining sites of the ligand binds to its ligate. Once the ligand has been chosen according to the characteristics needed from above, the next step is to covalently bind the ligand to a support matrix. First, the matrix should be selected according to the following properties(i) Chemically stable under conditions of activation, coupling and operation. (ii) Exhibit good physical resistance to withstand mechanical agitation. (iii) Perform under a wide range of pH and temperature conditions. (iv) Easily activated for coupling ligands at high concentrations. (v) Uniform in structure. There are several classes of matrices available for IAC (Table 5.1).
102
Table 5.1 Solid phase support matrices used for IAC Trade Name
Supplier
Support material
Ultrogel Sepharose Sephadex Sepharcryl Macrosorb Eupergit C Affi-gel
IBF, France Pharmacia LKB, Sweden Pharmacia LKB, Sweden Pharmacia LKB, Sweden Sterling Organic, UK Roehm Pharma, FRG Biorad. USA
Matrex Cellufine CPG Bakerbond Wide-Pore Prosep
Amicon, USA Pierce, USA J.T. Baker Inc, USA Bioprocessing, UK
Polyacrylamide/agarose Agarose Cross-linked dextran Polyacrylam ide/dextran Various organic powders Polyacrylamide Cross-linked polyacrylamide Cross-linked cellulose Controlled pore glass Polymer-Clad Silica Activated glass beads
Synthetic Trisacryl Dynospheres Separon H1000 TSK PW Fractogel TSWTovopearh
IBF, France Dyno Particles, Norway Tessek, Czechoslovakia Toson Inc, Japan E. Merk, FRG
Totally synthetic Synthetic polymer Synthetic polymer Polymer Vinyl polymer
Shandon, UK E. Merk, FRG Beckman, USA Waters, USA Pharmacia LKB, Sweden E.I. Dupont, USA
Silica Silica Silica Silica Cross-linked agarose Silica
Low to medium Dressure
Hiah pressure Hypersil WP300 Lichrospher Ultraspere Spheron Superose Zorbax
The most commonly used are the biopolymer agarose, cross-linked dextran, cellulose and starch together with the synthetic polymers polyacrylamide and hydroxyalkyl methacrylate. Among these the best known matrix in laboratories is agarose, commercially called Sepharose or Bio-gel. The selected matrix has to be activated to proceed to couple the ligand. Activation can be achieved in a number of ways. The most common method for agarose is activation with cyanogen bromide via the free OH groups of the polysaccharide (Fig. 5.1).
103
Immediately after the activation, the ligand is bound to the activated support by means of a primary aliphatic or aromatic amino group which should be present in on unprotonised form. This is done by mixing the ligand with the matrix and allowing the coupling to proceed for 2 hours at 25°C or 20 hours at 4°C. The unreacted active groups must be blocked to avoid non-specific binding of substances during chromatography. The conjugated matrix is usually packed in glass or plastic columns which may vary in length and diameter. Other activating procedures that can be used are indicated in Table 5.2. Conditions of adsorption where the sample mixture that contains the analyte comes into contact with the pre-packed column matrix should be previously chosen. The ionic strength and pH of the buffer used, flow rate through the column and operational temperature as well as washing conditions to remove non-specifically bound substances should also be determined. Finally, the molecule of interest that is bound to the antibody is eluted from the IAC column using one or more different conditions (4) such as change of ionic strength, change of pH, chaotropic agents, denaturants such as methanol, or polarity reducing agents. 5.3
PRACTICAL ASPECTS AND INSTRUMENTATION
Mycotoxins may be concentrated, purified and detected using IAC systems. In practice the first stage involves the development of specific antibody either of the polyclonal or the monoclonal formats (5). This aspect of mycotoxin research has been extensively described by other authors and will not be discussed within this review further, other than to direct the reader to references (6, 7, 8, 9, 10). Once a specific, high affinity MAB has been chosen it is possible to produce the MAB in large quantities using various techniques. Since IAC is extremely demanding on the quantity of antibody used per test ( = O.lmg) then it is advisable to use MAB wherever possible for this technique. The MABs may have to be produced in gram to kilogram quantities if IAC columns are to be manufactured on a large scale. Numerous companies are now actively involved in the contract production Gf MABs from hybridoma cell lines using both in vivo growth of hybridomas in the abdominal cavity of suitable animals ascites or in vitro cultivation of the hybridomas (see Table 5.3).
Once sufficient MAB has been produced it is then essential to purify and concentrate the antibody prior to coupling with the solid phase matrix. Three separate methods can be used for this process.
Activating Agent
Reagent toxicity
1.
Glutaraldehyde
Moderate
2.
CNBr
High
Activation tirne(h) 1-8 0.2-0.4
Ligand coupling time (h)
pH of coupling 6.5-8.5
2.4h at 25°C overnight at 4
8-10.0
"C Bisoxiranes
Moderate
5-18
4.
Periodate
Non-toxic
14-20
5.
Carbodirnide
Moderate
0.2-0.4
6.
Trichloro-Striazine
High
Overnight to 6 days 0.5-2.0 4-16
7.
Tresyl chloride
Moderate
0.5-0.8
8. 9.
Diazoniurn Hydrazine
Moderate High
1-3
0.5-1.0 3-16
2-4
Rapid
10 NHS
-
15-48 Overnight
Rapid
Stability of complex
$ 3 zg $.?
6-16
3.
Type of linkage
8.5-12.0 7.5-8.5
Michael's adduct, SchWs base lsourea irnido, carbarnate. Nsubstituted carbarnate Alkylarnide. ether, thio-ether Alkylarnine
8-9.5
N-substituted carbarnate 7.5-9.0 Triazunyl 7.5-10.5
Good Unstable at pH 4 or 210
Excellent
Good Unstable at pH >10 Good
Alkylarnine
Good
AZO
Moderate Excellent
6-8 7-9
Arnido
6.5-8.0
Arnido
Hydrolysis, pH sensitive
z
InN 0
CD
Q
5 $
c
rJY (D
Q
-.
3
52.
u"
s 0
3
? !0
5 U
5 Y
B
105
lmido carbonate t cyanate intermediate OH
+ INH2 I
O-C-N-R lsourea function
Figure 5.1. The cyanogen bromide activation of polysaccharides commonly used with agarose.
~
~~
Technical aspects
In vivo
In vitro
In vitro
In vitro
Mice
Labaraforyculture
Fermentor
Hollow fibre
intraperltoneal injection mice histocompatible. immunosupressedor nude
spenf/caf/ons inoculum (hybridorna cells) harvest
2 106 per mouse
static cultures in standard culture flasks batch culture
a
ascltic fluid up to 100mg MCA per mouse
0 5 106ml -1
supernatant 0 05mg MCA ml-1
20 - 30% purify (% MCA of total protein) Scaling up
Special points
0
1%
0
increase in number of llasks
increase In number of mice variable yjelds contamination: m u s e ascitic fluid components
a
sterile product contamination foetal calf serum
suspension (batch continuous. perfusion) encapsulation (batch
static culture in extracapillary space, medium circulated through fibres continuous culture
0 5 106ml - 1
I 106mi -1
supernatant 0 05mg MCA mi -1 perfusmn 0 5 m g MCA ml -1 encapsulation 0 3mg MCA mi-1 1-50%
extra-capillary supernatant 6mg MCA ml -1
increase in size or number
increase in size or number
sterile product contamination foetal calf serum
sterile product contamination no serum proteins
YO”.
I07
The first of these involves ammonium sulphate fractionation of ascites fluid. To an equal volume of ascitic fluid is added an equal volume of saturated ammonium sulphate solution (pH 6.5); this is left for 1h at 4°C to precipitate the MAB and immunoglobulins. The precipitated proteins are collected by centrifugation (l0,OOOg for 20 min) and the pellet washed twice with 50% saturated ammonium sulphate solution. The pellet is resuspended in a suitable buffer and its concentration adjusted to 10mg/ml. The ammonium sulphate fraction can be further purified by ion-exchange chromatography. The MAB solution is dialysed overnight against lOmM phosphate buffer, pH 7.5 and added to a "S" SepharoseO fast-flow cation exchange column (Pharmacia) to which the antibody binds. The column is washed with lOmM phosphate buffer, pH 7.5. The bound protein containing the specific MAB is eluted with lOmM phosphate buffer, pH 7.4 containing 50mM NaCI. Elution of protein is monitored by absorbance at 280nm and specific antibody activity estimated by immunological methods (1 1). The third method uses Protein A bound to Sepharose columns for the purification of specific MAB from large volumes of culture supernatant such as those produced with vitro fermentation procedures. This method is capable of purifying and concentrating of litres of antibody containing solution from pg/ml concentrations up to mg/ml levels. The steps involved in this procedure are as follows (12):
tens
-
1.
Swell 1.5g protein A Sepharose CL-4B (Pharmacia) in 0.05M Tris/HCI buffer pH 8.6 containing 0.15M NaCl and 0.1% sodium azide. The resin (5.6 ml) is then Dacked into a suitable column.
2.
The culture supernatant (1 litre) is applied to the Protein A column at a flow rate of 50ml per hour.
3.
The MAB is eluted from the Protein A column by stepwise pH elution with the following buffers at a flow rate of 50ml per hour for l h per buffer: 0.05M phosphate, pH 7.0 containing 0.15M NaCI. 0.05M citrate, pH 5.5 containing 0.15M NaCI. 0.05M acetate , pH 4.3 containing 0.15M NaCI. 0.05M glycine/HCI, pH 2.3 containing 0.15M NaCI.
4.
The MAB is monitored for activity in each of the different fractions by enzyme linked immunosorbent assay (ELISA).
5.
Fractions containing the antibody are pooled and dialysed against the appropriate buffer. For most MABs the fractions with pH values of 5.5 and 4.3 contain the appropriate activity.
6.
Columns can be regenerated by washing and equilibration with 0.05M Tris/HCI buffer, pH 8.6, containing 0.15M NaCl and 0.1% sodium azide.
108
Alternatively, Protein A may be attached onto controlled pore glass which is more rigid and durable than conventional polytheric matrices. The glass also has a very uniform interal pore size distribution which falls within a very narrow range, resulting in uniform rates of diffusion and hence rapid mass transfer between solution and solid phase. Protein A immobilised onto controlled pore glass is commercially available from Bioprocessing Ltd using a unique proprietary chemistry technique to satisfy four critical factors:
0 0 0
orientation of the ligand distribution of the ligand stability of the immobilised ligand elimination of non-specific surface interactions
The manner in which the Protein A molecules are orientated and immobilised on the surface is such that pairs of Protein A molecules can interact with the binding regions of each of the two Fc portions of the immunoglobulin at the same time. This results in very high binding capacities, especially for those immunoglobulins which typically do not bind well to conventionally immobilised Protein A and where the bivalent binding is an essential prerequisite, for example mouse IgGI. The increased capacity does not, however, result in a higher association constant (Ka) and therefore elution conditions remain unaltered. The combination of these factors means that Protein A attached to controlled pore glass is ideally suited for the efficient, cost effective and rapid purification of antibodies directly from bioreactor feedstock with minimal pre-treatment and particularly for large scale industrial use. Once sufficient specific antibody has been produced and purified it may then be immobilised onto a solid support. As indicated previously, the most commonly used solid support is agarose and the procedure used to immobilise MAB to AFs onto such is as follows (11): Freeze-dried CNBr-activated Sepharose 48 (log) is suspended in 300ml 1mM HCI over a scintered glass funnel. The gel is washed with 50ml 0.2M NaHC03 buffer, pH 8.7. containing 0.5M NaCI. To the washed gel is added purified MAB solution at 0.25mg ml in 70ml of the same buffer. The MAB solution and Sepharose are mixed for 2 hours at room temperature and then filtered through a sintered glass funnel. Any active groups remaining on the Sepharose are blocked by washing and mixing with 0.1M Tris/HCI buffer, pH 8.0, for 1-2 hours at room temperature. The mixture is filtered through a scintered glass funnel and washed with three to five cycles of two buffers of alternating pH. The first buffer is 0.1M Tris/HCI, pH 8.0, containing 0.5M NaCl and the second 0.1M sodium acetate/acetic acid, pH 4.0, containing 0.5M NaCI. The MAB bound Sepharose is washed and transferred to 70ml of 0.1M phosphate buffered saline, pH 7.4,
I09
containing 0.02% thimerasol, to give a total volume of 105ml. This can be stored at 4°C for up to 12 months without loss of activity. The affinity matrix can be used in an immunodiagnostic test for total AFs. Affinity columns are prepared by adding 0.5ml of MAB coupled SepharoseO to 55 x 5.5 mm columns. Test solution or sample is passed through the IAC column. As the solution passes through the column then AF molecules bind specifically to MABs attached to the solidphase Sepharose. Other components in the solution are unaffected by the antibody and inert solid-phase; these components therefore pass directly through the column. After washing with a buffer solution or distilled water at neutral pH to remove impurities, the AFs are desorbed by the use of an appropriate solution which causes antibody denaturation, eg methanol. The methanolic solution is then subjected to quantitative techniques for AFs such as fluorescence, using a florisilO tip, fluorimeter or HPLC. See Figure 5.2 for a diagramatic explanation of IAC:
4 4 A
ELUTION
B
r Figure 5.2 Schematic diagram for IAC for purification, concentration and detection of AFs. Samples containing AFs are first loaded onto the affinity gel column containing specific MABs against AFs (A). After washing to remove impurities (B), the AFs are eluted from the column with methanol (C) (adapted from ref 13).
110
Symbols:
0
solid-phase support (agarose) specific MAE to AFs attached to agarose
b 0
aflatoxins other substances
IAC columns are prepared by adding the Sepharose-linked MAB. Slurry (0.5ml) to small plastic columns (53 x 5.5mm; Jones Chromatography, Mid Glamorgan, UK). The slurry is held in place with porous frits (Porvair, Kings Lynn, UK) and top plus bottom stoppers are used to contain the slurry (Jones Chromatography, Mid Glamorgan, UK). Once the columns have been assembled it is possible to pass the sample extract through the column at a flow rate of 2-5mVmin using glass syringe barrels (loml) to load sample and fully assembled syringe (20mls) with rubber bung or air pump to pass the sample extract through the column. Detection of mycotoxins bound to the MAB columns involves elution in a small volume of alcoholic solution. The toxins present in this solution can be detected by various method for the particular mycotoxin eg thin layer chromatography, high performance liquid chromatography, gas-liquid chromatography or by aflatoxin natural fluorescence using a fluorimeter. Detection using the fluorescent properties of AFs can be achieved using three separate levels of instrumentation: 1.
The first of these involves a very simple and inexpensive technique. The eluting methanol solution from the affinity column is passed through a small Florisil tip (3 x 30mm) (200 mesh, Sigma) and the AFs are visualised by their natural fluorescent properties (blue or green fluorescence) under ultraviolet light in a dark room. This is a very robust but semi-quantitative method for the detection of the AFs. It can be used as a rapid screening method in a laboratory where only a very simple result is required and the budget is limited.
2
The second level involves the purchase of a fluorimeter that is capable of detecting AF fluorescence. Optical filters of 365nm for excitation and 400-420nm for emission of the AF molecules are required. The precision of these instruments is usually 0.5% with a claimed detection range of 0.1 to 1,000ppb for AFs. Such instruments are widely available from various instrumentation manufacturers but one ideal model is the Torbex FX-100 which can be specifically adapted for AFs. However, many difficulties may be encountered in making measurements under Sppb due to the non-specific fluorescence of contaminants.
3.
HPLC is the third and the most complicated of all the three methods. Consequently it is also by far the most expensive with an average system costing approximately f 15,000. However, the results obtained using HPLC are almost 100% confident and there is complete quantification of the individual mycotoxins
111
present. Figure 5.3 illustrates a typical HPLC system that can be used with IAC. The chromatograph was obtained using an IAC clean-up of ground peanut extract using the post-column system with the sample containing 53ppb total aflatoxins. The sample was extracted with lOOml of 80% (v/v) acetone dilute to 2% (vh) with water and then 40ml passed through an IAC column. The bound AFs were eluted with i m l methanol, diluted with l m l distilled water and 0.1ml injected into the HPLC. 5.4.
SAMPLE PREPARATION
After a sample has been received by the laboratory the first stage is to prepare the sample for analysis using IAC. This usually involves the exhaustive extraction of AFs from the food matrix with organic solvents such as methanol in a mixture with water. However, antibody activity and thus the performance of IAC columns are adversely affected with high concentrations of organic solvents. Thus, methods must be developed that allow for the adequate extraction of the mycotoxins from the sample combined with sample extracts that can be loaded onto IAC columns and still allow adequate antibody-antigen interaction. Using solid samples such as peanuts a number of extraction solvents have been developed and evaluated with IAC. There are four separate extractions that will be discussed: 1. 2. 3. 4.
80:20 (v/v) Acetone:water 60:40 (vh) Acetonitri1e:water 60:40 (vh) Methanokwater 100% chloroform
The first of these solvents, acetone, is used at a concentration of 80% in water to extract AFs from food at a samp1e:solvent ratio of 1.2. However, IAC columns have little or no activity with 80% acetone; thus, the sample extracts must be diluted to give a level of 2% acetone which will then allow antibody on the IAC columns to interact with the AFs. Furthermore, the volume of diluted acetone passed through the IAC column also alters the activity as can be seen from Table 5.4. The greater the volume of diluted acetone passed through the column then the lower is the recovery of AFs; this effect is most profoundly seen with the recovery of AFG2 to which the MAB has the lowest affinity.
112
Figure 5.3 Diagram of HPLC system with post-column derivatisation to determine aflatoxins. Chromatograph trace shows typical results obtained. NB: The analytical and guard columns are kept at a constant 40°C to produce constant and repeatable chromatographs.
113
Table 5.4 Assessment of the effect IAC column recovery with different volumes of 2% acetone
Volume (mi)
Percentage AFs Recovered G2
Total AFs
G1
10
76.0
96.5
88.0
105.0
91.4
20
72.5
115.5
82.0
975.0
91.9
40
68.5
116.5
101.0
110.0
99.0
50
56.0
101.5
101.5
95.0
88.5
80
48.0
113.0
92.0
109.0
90.5
100
30.5
98.5
84.5
88.5
75.5
200
14.0
77.5
61 .O
64.5
54.2
Methanokwater extraction procedures for AFs from peanut samples have been universally used in mycotoxicology laboratories for several years. In addition, 60:40(v/v) methanokwater using 250ml of solvent to 509 of sample, is the BF method as approved by the AOAC for extracting AFs from peanuts and peanut products. When this is applied to IAC the following procedure is used: Well ground sample (509) and 5g of NaCl in a solvent resistant blender jar are extracted with 250ml of methanol: water (60:40 (v/v) ) by blending for one minute at high speed. The extract is filtered through a Whatman No 4 filter paper and the filtrate is diluted with an equal volume of distilled water. The diluted extract is then passed through a column. Usually 10ml is passed through an individual column, this is equivalent to one gram of original sample. However, with IAC it is possible to increase the volume of sample extract passed through the column and thus increase the equivalent concentration of the sample. This will increase the detection limit of the method eg 50ml of diluted extract passed through a column is equivalent to five grams of sample and this is therefore five times more sensitive than passing lOml of sample extract through a column. The activity of IAC columns for total AFs in 30% methanol as determined by recovery is 90%. A column when challenged with 25ng of AFB1. 25ng of AFG1, 12.5ng of AFB2 and 6.25ng of AFG2 produced recoveries of 97%, loo%, 90% and 70%, of all four AFs respectively (11). When maize or peanut samples were artificially contaminated with the same concentrations of AFs the average recoveries were 94% and 83% respectively (Table 5.5).
I14
Table 5.5 Recovery of total AFs from artificially contaminated samples as determined using IAC
Sample Type
Percentage recovery Aflatoxin: Amount added (ng):
Maize Peanuts
B1
B2
G1
25.0
12.5
25.0
90.0 87.0
102.0 88.0
95.0 91.0
G2
Mean
6.25 89.0 68.0
94.0 83.0
Methanokwater (7030 (v/v) ) has also been used with IAC in an Association of Official Analytical Chemists (AOAC)/International Union of Pure and Applied Chemistry (IUPAC) collaborative study (14). This trial was conducted to evaluate the effectiveness of the IAC technique for the determination of AFs in maize, peanuts and peanut butter. The test sample (259) was transferred to a blender jar and 59 NaCl added. To this was added 125ml of methanokwater (7030) extraction solvent and blended for 2 minutes at high speed. The extract was filtered through a Whatman 2V filter paper and 15ml of filtrate was diluted with 30ml of water. The diluted filtrate was further filtered through a Whatman 934AH glass microfibre filter paper before IAC. Portions of 15ml second filtrate were then passed through IAC column. Chloroform can be used as an extraction solvent but the protocol with this solvent is longer and more complex, although it may provide the analyst with improved recoveries of AFs from certain types of samples. The method is performed by placing 50g ground sample and 25g celite or diatomaceous earth into an explosive proof container. Chloroform (250ml) and distilled water (25ml) is added and blended for one minute at high speed. The solvent extract is filtered through a Whatman No 4 filter paper and filtrate collected. The filtrate (14ml) is evaporated to dryness and the residue reconstituted in 12.7ml of methanokwater (60:40 (vh) ). The reconstituted sample (12ml) is then diluted with 60ml of distilled water and filtered through Whatman No 4 filter paper. The filtrate (7.5ml) is passed through an IAC column. The expected recoveries with this method are 30.4%, 92.8%, 80% and 89.6% for AFG2, GI, B2 and B1 when ground coffee is artificially contaminated with AFs. As can be seen from the above method, chloroform must be fully removed from the sample extract by evaporation. Chloroform is an organic solvent that is immicible with water and is also detrimental to antibody activity. The final sample extract to be passed through a column must contain a high percentage of water and for this reason chloroform can not be applied directly to the columns.
I15 A recent collaborative trial for IAC columns utilised 30ml acetonitri1e:water (60:40 (v/v)) to extract AFs from roasted peanut butter. The samples were extracted by shaking for 30 minutes at 10 oscillations/second at ambient temperature. Water (45ml) was added to the flask and shaken for a further 30 minutes. The sample extract was transferred to centrifuge bottles and centrifuged at 4,000rpm until sedimentation is complete. The supernatant is filtered and 15ml of filtered supernatant is diluted with 135ml PBS. Diluted filtrate (75ml) is the passed then passed through a column (15). With this extraction solvent the acetonitrile must be diluted to 2.5% such that it is compatible with the activity of the IAC columns. Milk and milk powder can contain a derivative of AFBl termed AFM1. This toxin is secreted in the milk of cows consuming feed contaminated with AFB1 and there have been several surveys in various countries that have detected AFMI in dairy products (16). However, the efficiency of conversion and secretion is not very high so that animals receiving a daily intake of about 20mg AFBl in their feed secrete milk containing about 10pg/l of AFM1. The maximum tolerated AFMl level in fluid milk within the Netherlands is 0.1ng/ml and, other countries have similar legal limits. Consequently, techniques for the detection of AFMl in milk and milk products must allow extremely low levels to be detected. The most ideal method is IAC coupled with HPLC. For the analysis of milk for AFM1, the milk is first warmed to 35-37°C and either filtered through a Whatman No 4 filter paper or centrifuged at lOOOg for 15 minutes. At least 50ml of the filtered milk is collected and passed through an IAC column at a slow, steady flow rate of 2-3ml per minute. If greater sensitivity is required then the volume of milk sample passed through the column can be increased up to 1,000ml (17). With milk powder or solid milk products then the sample (log) is added to 50ml of water warmed to 50°C mixed with a stirring rod, until a homogeneous mixture is obtained. The solution is allowed to cool to 20°C and then quantitatively transferred to a lOOml volumetric flask and the volume adjusted to 1OOmI. The reconstituted food product is filtered through a Whatman No 4 filter paper or centrifuged at 1,0009 for 15 min and 50ml of the solution is transferred into a syringe barrel and passed through an immunoaffinity column (17). Another method for the preparation of raw milk was described by Hansen, 1990 where 40ml portions of milk are mixed with l g NaCl and centrifuged at 2,000 x g for 5 min. The skim portion is filtered immediately before analysis and 25ml aliquots of prepared milk was passed through a column (18). IAC has also been applied to the in vitro isolation of AFBl from human urine, serum and milk samples (10, 19, 20). Freshly collected human urine (loml) was centrifuged or filtered through a 0.45pm filter. When the urine was applied directly to the immunoaffinity columns there was a loss of 40% activity of the columns. In this particular study human serum (loml) or human milk (loml) was applied directly to the antibody column without prior treatment. Quantification of the AFs was achieved using a competitive radioimmunoassay (RIA). Maize and wheat samples have also been analysed for their AF content using IAC combined with RIA (21). The samples were ground to a fine powder, defatted with 15ml hexane, extracted with 15ml of acetone/H20 (8515). After removal of the acetone by evaporation, the solution was extracted with 3ml of benzene 3 times. The benzene extracts were concentrated to
I I6
3ml of benzene 3 times. The benzene extracts were concentrated to dryness and redissolved in 0.2ml dimethylsulphoxide. Further dilutions were prepared in 0.1 M NaHC03. A clean-up method for ochratoxin A (OTA) using imrnunoaffinity columns has recently been developed (22) for the analysis of coffee products. Finely ground coffee beans or instant coffee powder (59) was extracted with 80ml of 1% (v/v) aqueous sodium bicarbonate. The suspension was sonicated for 15 min in an ultrasonic bath. The sonicated sample was adjusted to lOOml with 1% (v/v) NaHC03, filtered through a 6cm Whatman GF/B glass-fibre filter under reduced pressure, and followed by addition of equal volume of phosphate buffered saline. Canned coffee drink was filtered through a 6cm Whatman GF/B glass-fibre filter under reduced pressure. The sample solution (loml) of bean extract or 59 of the canned coffee drink filtrate was applied to the irnmunoaffinity columns. 5.5.
ILLUSTRATIVE EXAMPLES
An AOAC/IUPAC collaborative study has been conducted to evaluate the effectiveness of the IAC column method for the determination of AF (14). Samples of 11.4kg each of ground corn, raw peanuts, and peanut butter and a 4.5kg portion of ground corn, naturally contaminated with AFs were tested following the AOAC methods 26.026-26.031 and 26.058. The naturally contaminated corn sample was shown to contain 23ng/g, whereas, all other commodities were found to contain AFs at <2ng/g. Test portions were subsequently spiked with AFs B1, B2, G1 and G2 in a ratio of 7:1:3:1 at 30, 20 and 10ng/g. A total of 24 collaborating laboratories participated in the study from the United States, France, Canada and the Republic of South Africa. Each laboratory received artificially contaminated duplicates of corn at 30ng/g, peanuts at 20ng/g and peanut butter at 10ng/g, controls (duplicates at <2ng/g for peanuts and peanut butter); duplicates of naturally contaminated corn, and practice positive samples of all three commodities at 20ng/g and also practice negative samples of all 3 commodities containing <2ng/g. Twelve laboratories used solution fluorimetry with bromine (SFB) to detect total AFs, 9 used the liquid chromatography post-column derivatisation (PCD) with iodine to determine individual AFs, and 3 used both SFB and PCD methods. Reference standards, immunoaff inity columns, bromine developer solution, and fluorimeter calibration standards were provided. The principle of the method involved the extraction with rnethano1:water (7:3). The extract was filtered, diluted with water, and passed through an immunoaffinity column containing MAB specific for AFs B1, 62, G1 and G2. AFs were isolated, purified, and concentrated on the columns and then removed with a small volume of methanol. Total AFs were quantified by fluorescence measurement after reaction with bromine solution. Individual AFs were quantitated by liquid chromatography with fluorescence detection and post-column iodine derivatisation. The performance of IAC columns was assessed by estimating the recoveries for a standard aflatoxin solution of 15ml methanokwater (3:l) that contained 259 B1, 5ng B2,
1 I7
15ng G1 and 5ng G2. The expected recoveries were at least 90%, 8070, 90% and 60% respectively. The extraction of 259 test samples was achieved by adding 59 NaCl and 125ml of methanol: water 7:3). This was blended for 2 minutes at high speed and then filtered through a Whatman 2V filter paper. The filtrate (15ml) was diluted with water (30mI) and diluted the extract further filtered by passing through a glass microfibre filter paper prior IAC. If the filtrate is not clear it should be refiltered. IAC was performed by pipetting 15ml of the second filtrate (equivalent to l g of test sample) slowly through a column at a flow rate of 6ml/min. The column was then washed with 2 x lOml water and the water discarded. Finally the column was eluted with l m l of liquid chromatography grade methanol. Figure 5.4 illustrates the apparatus used for IAC. For quantitation by SFB the methanol eluate was collected in a fluorimeter cuvette and l m l of 0.002% (vh) bromine in water added. The tube was vortexed for 5 seconds and fluorescence measured with a fluorimeter at 450nm following excitation at 360nm (Sequioa Turner Model 450).
Pump Unit
H
- Glass syringe barrel
Figure 5.4 The IAC column system For quantitation by PCD, the methanol was collected in a 2ml volumetric flask and diluted to volume with liquid chromatography grade water and mixed. The AFs were quantified by liquid chromatography by injection of 50pl extract. Each AF peak was identified in the chromatogram from analysis of test sample by comparing retention times with those of corresponding reference standards. The quantity of each AF was
I I8
determined in the injected eluate from corresponding standard curves. The concentration of each AF in the test sample was calculated using the following formulas: W: 259 x (15m1/125ml) x (15mV45ml) =lg Aflatoxin, nglg = A x (TV/IV) x (1NV) =Ax40 Where W=weight of commodity represented by eluate, A=ng aflatoxin in eluate injected, TV=final eluate volume (2000pl), and IV=eluate injected (50~1). Concentration of all four AFs was added to obtain total AF concentration. Twenty of the 24 collaborators who received test samples completed the study (10 used SFB, 7 used PCD, and 3 used both SFB and PCD methods). When the "outliers" were excluded, the Horwitz ratio for the determination of added total AFs (10-30ng/g) by the SFB method ranged from 0.41 to 1.07 in corn, from 0.41 to 0.72 in peanut butter, and from 0.49 to 0.88 in peanuts. The RSDr (relative standard deviations for repeatability) of all spiked test samples ranged from 11.75 to 16.57%, and the RSDr (reproducibility) ranged from 10.97 to 33.09%. The Horwitz ratio for the determination of added individual AFs by the PCD method ranged from 0.13 to 1.85 in corn, 0.61 to 1.77 in peanut butter, and 0.3 to 3.63 in peanuts. The RSDr for all added individual AFs ranged from 5.45 to 79.060/0, and the RSDr ranged from 4.21 to 133.31%. The wide ranges of RSDr and RSDr values was thought to be due to the small amounts of 82 and G2 added to the test samples. Because a method with a Horwitz ratio <2.0is considered to have acceptable and typical precision (14), the immunoaffinity column coupled with the SFB or PCD method is effective for determining total AFs at levels 2 1Onglg. In general, recoveries of >loo% were achieved using SFB. The higher recoveries were thought to be caused by fluorescent interferences in the sample matrix with corn and peanut butter or due to trace contamination of AFs in the peanut butter. The collaborators commented that they had little difficulty with the method. The final recommendation of the AOAC/IUPAC was that IAC coupled with SBF for determination of total AFs and IAC coupled with liquid chromatography/postcolumn derivatisation with iodine for determination of individual AFs be adopted as Official First Action for the determination of AFs in corn, raw peanuts, and peanut butter at total AF concentration 2 10ng/g. This is now an additional method for AF determination. However, a second collaborative trial to evaluate an immunoaffinity column clean-up procedure with quantitation by fluorescence liquid chromatography (post-column derivatisation) for the determination of AFs in peanut butter failed to provide the results necessary to recommend the method for adoption as official. This method uses IAC columns from a difference source (Biocode) from the other trial. The reason for not
1 I9
receiving recommendation was due to low recovery levels for the two spiked samples of 51-67% (15). Other solvents have now been used for the extraction of AFs and the purification of the toxins with IAC. The environmentally more tolerable solvent of acetone has been assessed and can replace the more common extraction solvents such as chloroform and methylene chloride (23). Extraction of 259 sample with 125ml acetone:water (85:15) was achieved by shaking for 45 minutes or stirring with a magnetic stirrer. The liquid was filtered and if only AFBl and G1 are to be quantified 5ml filtrate is diluted with 45ml of distilled water (solution 1). If all 4 AFs are to be quantified, 3ml of the filtrate is diluted to lOOml (solution 2). A total volume of between 5-25ml of solution 1 or 10ml of solution 2 was passed through immunoaffinity columns. The columns were washed with water and bound AFs released by passing 3 x l m l methanol or acetonitrile. The solvent was evaporated to near dryness under nitrogen and resuspended in 0.5ml up to lOml with a solvent mixture, depending on the expected concentration or the sensitivity of the HPLC. Initial experiments with the acetone:water mixture produced recovery rates of AFB1 and AFG1 exceeding 90%, while the recovery rates of AFB2 and AFG2 were only 60% or 25%. Further, experiments on the water diluted extracts showed that the recovery rate of AF, especially AFB2 and AFG2, are influenced by the acetone content as well as by the quantity of the solution. It was discovered that lOml of an aqueous extract containing 8.5% acetone (solution 1) AFBl and G1 were almost quantitatively recovered against only 87% and 35% of B2 and G2 respectively. In order to achieve recoveries of 90% and more for all 4 major AFs, especially G2, a quantity of only lOml of an aqueous extract containing 2.6% acetone (solution 2) must be used. If 15ml or 20ml of these solutions were employed, AFB1, B2, and G1 could be recovered almost quantitatively the percentage of AFG2 was, however, only 85% or 70%. In order to check repeatability, 6 tests on 2 dairy cattle feeds and 1 supplementary feed for calves were performed. In one dairy cattle feed lOml of the extract was diluted to an acetone content of 8.5% (solution 1) and AFBl content determined. The average value was 2.71nglg f 3.1%. The second dairy cattle feed and the supplementary feed were tested with the same extract quantity, however, with an acetone content of 2.6% (solution 2). The following quantities of AFB1, B2. G1 and G2 were found 8.12nglg f 1.4%, 0.43nglg k 5.6%, 1.76nglg f 4.1% and
120
Table 5.6 Comparison of Total AF Recoveries Using IAC with Acetone or Methanol Extraction Solvents. Sample Type
Acetone'
Methanol"
AFs Recovered (ng/g) Ground Peanuts (A) Ground Peanuts (B) Ground Peanuts (C) Smooth Peanut Butter Peanut Butter (A) Peanut Butter (B) Coprah Paprika a
14.4 90.2 52.8 18.9 13.2 11.5 61.9 7.8
15.3 82.8 45.9 13.7 9.2 8.6 41.3 3.3
Acetone: water (80:20) diluted to 2% and 20ml, or 40ml passed through column. Methanol: water (60:40) diluted to 30% and lOml passed through column.
Compared to other methods, the procedure of the AF determination involving acetone extraction follow by IAC offers the following advantages: Extraction with an environmentally tolerable solvent mixture producing a good or very good recovery rate. 0
Low consumption of extractants (eg 25ml), with homogenised samples, as only a few ml of extract are required. Selective isolation of AFs is achieved to a large extent. Low threshold of determination and identification of AFs.
0
Less work and time is required compared with other methods which do not employ an immunochemical purification phase but comparable results are obtained.
A collaborative study has been completed to study a method based on IAC cleanup with HPLC determination. The aim of the study was to test a method applicable to the low AFMl concentrations found in milk (<50ng/kg milk) (17). The study involved 16 laboratories each receiving six randomly coded samples of milk powder with a low fat content (about 1'3'0) and six samples with a high fat content (about 28%). Furthermore, collaborators received a practice sample (209, sufficient for two analyses) with a known AFMl content, an unknown practice sample (log), fifteen IAC columns and standard AFMl solution in chloroform (lpglpl).
121
The method involved extraction of milk or milk powder samples for AFMl and then passing 50ml of extract through a column. The column was then washed with lOml water and the AFM1 eluted by the addition of 4ml acetonitrile. The volume was reduced to 300-5OOpl at 30°C using a stream of nitrogen and made to a final volume of 5ml with water. Finally the AFMl content was estimated by HPLC. Results, not corrected for recovery are illustrated in Table 5.7 for AFM1 content. Table 5.7: Reported AFMl Content in the Practice Samples (ng/kg milk powder) Laboratory Number
Known Sample
Unknown Sample ~
(580 f 170nSAFM I/kg) 1 2 3 4 5
6 7 8 9 10 11 12 13 14 15 16 17
1) 646 593 570 505 1) 645 686 530 716 588 586 608 500 485 2) 593
1) 184 200 190 315 1) 21 1 246 207 182 185 204 190 223 190 244
1) Results not included 2) Participation cancelled Sixteen participants from eleven countries participated in the collaborative trial for the determination of AFMl in milk powder. Five samples, containing AFM1 levels in the range of 80-600ng/kg milk powder, were analysed as blind duplicates. The overall RSD(R) was in the order of 20%, which is a very acceptable result. Hence, IAC columns should be considered as an appropriate tool for AFMl determination at low levels of contamination (17).
I22
References
1.
J.J. Langone, and H.Van Vunakis J. Natl. Cancer Inst., 50 (1979) 591
2.
W. Crueger, and A. Crueger (eds), Biotechnology: A Textbook of Industrial Microbiology, Science Technology Publishers, USA, 1990.
3. S. Angal, and P.D.G. Dean, In: Protein Purification Methods 245-259. IRL Press, Oxford, 1989
4.
E.L.V. Harris, and S. Angal (eds), Protein Purification Methods: A Practical Approach, IRL Press, Oxford, 1990
5.
0. Cattay (ed.), Antibodies, Volume 1: A Practical Approach, IRL Press, Oxford, 1988
6.
O.T. Schonherr, and E.H. Houwink, Ant. van Leeu., 50 (1984) 597
7.
A.A.G. Candlish. J.E. Smith, and W.H. Stimson, Biotech. Adv., 7 (1989) 401
8.
F.S. Chu, J. Food Prot., 47 (1984) 562
9.
A.A.G.Candlish, W.H. Stimson, and J.E. Smith, Lett. App;. Microbiol. 1 (1985) 57
10. J.D. Groopman, L.J. Trudel, P.R. Donahue, A. Marchak-Rothsteun, and G.N. Wogan, Proc. Natl. Acad. Sci. USA. 81 (1984) 7728 11. A.A.G.Candlish, C.A.Haynes, and W.H.Stimson, Int. J. Food and Tech. 23 (1988), 479 12. A.A.G. Candlish, W.H. Stimson, and J.E. Smith. In: J.Duncan, and L.Turrance (ods),Techniques for the Rapid Detection of Plant Pathogens. Blackwell Scientific Publications, Oxford, 1992.
13. J.M. Fremy, and F.S. Chu. 1n:H.P. Van Egmond (eds), Mycotoxins in Dairy Products. Elsevier Applied Science, Amsterdam 1989. 14. M.V.Trucksess, M.E. Stack, S. Nesheim, S.W. Page, R.H. Albert, T.J. Hansen, and K.F. Donahue, J.Assoc. Off. Anal. Chem. 74 (1) (1991) 81. 15. A.L. Patey, M. Sharman, and J. Gilbert. J.Assoc. Off. Anal. Chem. 74 (1) (1991) 76 16. H.P. van Egmond In: H.P. Van Egmond (eds),Mycotoxins in Dairy Products Elsevier Applied Science, Amsterdam 1989. 17. L.G.M.Th. Tuinstra, A.H. Roos, and van Trijp.. Report 92.14.(1992) Rikilt-dlo. 18. T.J. Hansen, (1990) J.Food Prot. 53 ( l ) ,75-77.
I23
19. F.D.Groopman, and K.F. Donahue. J.Assoc. Off. Anal. Chem. 71 (5) (1988) 861. 20. J.D. Groopman., W.F. Busby, P.R. Donahue, and G.N. Wogan, (1986) Syrn. Mol and Cell Biol. 32, 233-256. 21. P.S. Sun, and F.S. Chu, J. Food Safe. 1 (1982) , 67. 22. M. Nakajirna, H. Terada, K. Hisada, H. Tsubouchi, K.Yamamoto, T. Uda, Y. Itoh, 0. Kawamura, and Y. Ueno, Food and Agric lrnmunol2, (1990) 189 23. G. Werner, Agribiol. Res. 44 (4) (1991) 289 24. J.A. Egana, A.A.G. Candlish, In Prep. (1992)
124
Chapter 6 EMERGING TECHNIQUES: ENZYME-LINKED IMMUNOSORBENT ASSAY (ELISA) AS ALTERNATIVES TO CHROMATOGRAPHIC METHODS C.M. WARD, A.P. WILKINSON AND M.R. A. MORGAN
Food Molecular Biochemistry Department, AFRC Institute of Food Research, Norwich Research Park, Colney, Norwich, Norfolk NR4 7UA, United Kingdom 6.1 INTRODUCTION Mycotoxins, particularly aflatoxins, have biological potency in trace amounts. The foods, animal feedstuffs and biological tissues and fluids which may need to be analysed for mycotoxin contamination are chemically complex. To quantify mycotoxins using physicochemical techniques, they have to be extracted from the large amounts of protein, lipid and carbohydrate as well as structurally-related chemicals which will interfere with physicochemical analysis. Extraction is time-consuming and costly in terms of solvents and labour and severely limits the number of samples which may be analysed, an important consideration given that mycotoxin contamination is almost always non-uniform and that large sample numbers may have to be taken to obtain a reliable measure of the level of mycotoxin contamination in a sample lot. The ease of analysis is also of importance for commerce where rapid acceptancelrejection decisions have to be made. To overcome these problems, the ideal analytical technique should be highly sensitive so that trace levels of mycotoxin can be analysed, selective to avoid mis-classification and allow simple analysis procedures which in turn will facilitate a fast analysis rate with high sample through-put at low cost. Immunoassay is such an analytical technique and in the form of enzyme-linked immunosorbent assays (ELISAs) is being used routinely for mycotoxin analysis with test kits being commercially available from a large number of manufacturers in different countries. These kits are designed for simple yeslno, semi-quantitative or fully quantitative test procedures. Some of these procedures (including fully quantitative methods) have now been given official first action status by the Association of Analytical Chemists (AOAC).
6.2 PRINCIPLES OF ELISA 6.2.1 The immune response and polvclonal antibodies ELISAs, like all immunoassays, are analytical techniques relying on the specific recognition and binding of ligand or analyte by antibodies, which are animal-derived serum proteins (1-3). Antibodies are products of the immune system, an intricate cell-based system that higher vertebrates have evolved to protect themselves from disease causing microorganisms (4).
125
Antibody formation is induced in animals when organic material, of molecular weight greater than lo3 - lo‘ daltons, is recognised by antibody producing cells (B-lymphocytes)as being alien to normal body constituents. In the recognition process, or immune response, many small discrete areas of immunogen that are chemically diverse in structure and composition, are bound by membrane-bound antibodies present on the surface of 0-lymphocytes. This causes proliferation of the &lymphocytes and antibody secretion, with &ch lymphocyte producing only one particular antibody capable of binding to a discrete area of immunogen. Thus the result of the immune response is the stimulation of different B-lymphocyte cell-lines, or clones, and the secretion of many different antibodies, into the serum, each capable of binding to immunogen but with different specifications and affinities. Such anti-immunogen serum is termed polyclonal (5).
6.2.2 Monoclonal antibodies Monoclonal antibodies (6) are the product of a single antibody producing cell-line or clone. Such antibodies can be produced in virro by fusing antibody-secreting spleen cells that do not normally survive in virro with tumor cells. The resultant hybridoma is immortal and will increase in number whilst still producing antibody. Monoclonal antibodies are identical molecules able only to react with a specific target site and they can be produced in large quantities ensuring their widespread dissemination (2). 6.2.3 Haptens Compounds of low molecular weight, less than l@- lo‘ daltons, such as mycotoxins, cannot by themselves act as immunogens. Such low molecular weight compounds are termed haptens. However, anti-hapten antibodies can be formed providing the hapten is chemically linked to an immunogenic macromolecule such as a protein. The chemistries used for conjugation of mycotoxins to proteins have recently been reviewed (7). 6.2.4 SDecificitv of anti-hapten antibodies In an immune response to a hapten-protein conjugate such as aflatoxin B,carboxymethyloxime-bovine serum albumin (8) various antibodies will be produced, some recognising a) hapten alone, b) hapten-linkage chemistry for instance aflatoxin B,carboxymethyloxime or c) hapten-linkage-carrier (9,lO). Antibodies binding the carrier protein represent the largest group of antibodies. Those that can bind non-conjugated or ‘free’ hapten will be a small fraction of the total antibodies produced and in some cases they are rare. Furthermore, antibodies binding hapten alone have been shown by Landsteiner (1 1) to recognise those entities of the hapten molecule that are distal to the point of hapten-protein linkage. This property can be exploited to generate immunoassay systems that are highly specific, recognising one compound amongst others of similar structure. For example, antibodies can be generated that distinguish aflatoxin B, from aflatoxin b,two compounds which only differ in the degree of saturation of the C 8 - C9 bond of the bifuran moiety (8,12). Anti-hapten antibodies, as well as being highly specific for target compounds could, with judicious design of hapten-linker group
126
protein camer, be used to generate antisera of broad specificity able to recognise groups or classes of molecule. However, for mycotoxins, such antisera seem difficult to produce (8,13,14). Princides of immunoassay The general principle of an immunoassay is to quantify high affinity antibody-analyte interactions, attaining an assay that can detect very low concentrations of analyte when highsensitivity detection systems are utilised. Yalow and Berson (15) were the first to describe such a system by making use of radioactively labelled compounds. Such assays involve an unknown quantity of analyte competing with a fixed amount of labelled analyte for a limiting number of analyte binding sites provided by the antibody. At equilibrium, antibody can exist in two phases, one bound to analyte and the other analyte-free according to the equation:
6.2.5
+
+
An An' 2Ab (free phase)
* AbAn
+ AbAn'
(bound phase)
where An' = labelled analyte, Ab = antibody, An = analyte (unknown). If the amount of unknown analyte is relatively low, more label is able to bind to the antibody. If there is a lot of unknown analyte, little label is bound to antibody. Provided free and bound phases can be separated and label can be quantified in either of the phases, then the amount of unknown analyte can be determined by reference to the behaviour of known standards (13,16,17). Radioimmunoassays (RIAs) are very sensitive analytical procedures but have severe disadvantages such as the need for radiological protection procedures, generation and safe disposal of radioactive waste, short shelf-life of labelled reagents and the expensive equipment needed for radioactivity measurements. Such problems have led to investigations into the use of alternative systems for measuring the primary reaction between.antibody and analyte of which the use of enzyme labels either conjugated to analyte or antibody has proved very successful (1,17,18). 6.2.6 Enzvme immunoassavs Enzyme immunoassays can be categorized as either "heterogeneous" or "homogeneous". Heterogeneous enzyme immunoassays are of the same principle as RIA, requiring the separation of free and bound enzyme-labelled molecules (analyte or antibody). In homogeneous assays the antibody-analyte interaction influences the activity of an enzyme such that no separation step is required but this type of assay is limited by the availability of suitable enzyme conjugates which are very difficult to synthesise and the need to have samples free of substances such as enzyme inhibitors, activators and substrates, that will influence the activity of the indicator enzyme. As a result, homogeneous enzyme immunoassays have not been used for routine mycotoxin analysis (1,181.
127
6.2.7 Enzvme-linked immunosorbent asSavs (ELISAS) ELISAs (heterogeneous enzyme immunoassays) have been used extensively for mycotoxin analysis because of the advantage resulting from the use of a separation step which, depending on assay format, either eliminates completely or diminishes the possibility of modification of enzyme activity by compounds in the sample being analysed. The assay formats are often termed direct or indirect (see section 6.4.2.2). In both formats the enzymes employed are available in high purity, have high turnover number and act upon a substrate giving a readily measurable reaction product such as a fluorophore or most commonly a chromophore (1,18). Phase separation is effected by immobilizing one of the assay components on to a solid phase which in the case of ELISAs is most commonly a 96-well polystyrene microtitration plates. In the direct assay format enzymes can be labelled to either mycotoxin or anti-mycotoxin antibody using chemistries similar to those employed for immunogen synthesis (7). If enzyme is labelled with mycotoxin, then the anti-mycotoxin antibody is the component immobilized to the solid phase. If the anti-mycotoxin antibody is labelled, then the solid phase needs to be a mycotoxin-protein conjugate using a different protein to the one used for immunogen synthesis so that only antimycotoxin antibodies participate in the assay. The disadvantage of the direct assay format is that sample components can still come into contact with the detection enzyme and interfere in the assay. The indirect assay format (see section 6.4.2.2.2) is a simple way to overcome possible enzyme interference because the primary antibody-analyte interaction is detected using a second, species-specific enzyme-conjugated antibody. Thus to detect anti-mycotoxin sera raised in rabbits, the second enzyme conjugated antibody must recognise rabbit antibodies. Such antibodies are produced in another animal species, for example goat. In this way there will be no direct contact between detection enzyme and sample. 6.3 SAMPLE PREPARATION The localised Occurrence of mycotoxins in high, biologically active concentrations is of great concern to the analyst. To derive a representative sample that will take into account the scattered ‘hot spots’ in a bulk of uncontaminated commodity requires complex statistically designed sampling plans (19). These are described in a number of reviews (20-22). To meet the statutory and voluntary guidelines for permissible levels of mycotoxins laid down in most countries (23) a large number of samples are generated. Immuno-analysis is highly suited to batch-wise processing of large numbers of samples. Sample preparation is usually very simple often requiring little more than the solubilization of the mycotoxin. Chromatographic techniques may require extensive, time-consuming clean up procedures using specialised and expensive technology and reagents. This is a severe limitation on the number of samples which can be routinely analysed. 6.3.1 Extraction Should the sample already be in a liquid form then direct analysis may be possible. For example, milk has been analysed for aflatoxin M, (24) and by prior passage through a Sep-Pak C,, cartridge the sensitivity of detection has been improved (25). With undiluted human serum
128
interference was overcome by precipitation and centrifugationwith an equal volume of methanol for the analysis of aflatoxin (26). Urine is also a difficult matrix but a direct RIA for aflatoxin B, in rat urine has been described (27). However, use of an affinity column to clean up human urine may be required (28), a ClScartridge for aflatoxin (29) or XAD-2 for T-2 toxin (30). With a solid sample, both immunologist and chromatographer face the same problem - that of reproducible solubilization and extraction of the maximum amount of the target analyte. This is greatly assisted if the sample is first finely ground. Organic solvents, usually in the presence of water are used and solvents which are not water miscible tend to be avoided because they involve the extra step of evaporating to dryness before dilution in assay buffer although sensitivity may be improved by concentration of extract. Methanol is a suitable solvent for some toxins and was used to extract the trichothecene diacetoxyscirpenol (DAS) from wheat (31). Other trichothecenes, 3-acetyldeoxynivalenol (ADON) and deoxynivalenol (DON) from rice and wheat respectively (32,33), were extracted after the addition of water to methanol (40 : 60, v : v). The efficiency of extraction of aflatoxin was also improved by the addition of water although the ratio of water to solvent and the proportion of extractant to sample needs to be optimized (34,35). Nevertheless it has been shown that chloroform gave consistently higher recoveries of aflatoxins than methanol from cereal products. Extraction of aflatoxins from peanut butter (12) was found to be most efficient using acetonitrile : water (1 : 1, v : v). Acidified chloroform was used for the extraction of barley for sterigmatocystin and ochratoxin A (36,37). Thus the choice of solvent depends on the nature of the sample matrix and the relative solubilities of the mycotoxin. Also the susceptibility of each antibody to interference by the solvent and coextracted components of the matrix needs to be taken into account. It is clear that the appropriate solubilization procedures for mycotoxins prior to immunological analysis can combine simplicity with speed and result in reduced variability. 6.4 INSTRUMENTATION AND PRACTICE At its simplest ELISA requires little apparatus. For semi-quantitative measurement sample and reagent may be dispensed from a plastic dropper. Subsequently distilled water from a squeeze bottle (or in the case of dipstick assays, tap water) can be used for washing. The end result can then be determined by eye comparing the colour intensity of the sample in relation to a control. 6.4.1 Instrumentation In quantitative analysis a range of both positive and air displacement automatic pipettes are essential, and the use of a multi-channel pipette will speed up the application of reagents to microtitration plates. Dilution and pipetting are the most crucial steps in a quantitative ELISA, determining precision and reproducibility. Accordingly pipettes should be well maintained and the operator experienced in their correct operation, Reaction times may be reduced if plates are incubated at 37°C or if they are gently rotated on a shaker and equipment is available which can combine both functions.
129
The simpler plate washers are manually operated, filling under gravity and emptying via a water vacuum pump, one row at a time. Electrically operated machines may also wash row by row or deal with all 96 wells simultaneously. These machines can be programmed with the number of wash cycles, volumes and dwell times. The most sophisticated models also have a stacking and feeding device so that a batch of plates can be processed without supervision. To determine the optical density of the end product of enzyme activity there are hand-held, battery operated readers which can be used outside the laboratory. These are able to measure colour intensity one column at a time in a plate divided into strips. Readers of increasing sophistication are marketed which read complete plates at visible or fluorescent wavelengths. The ultimate version is a computer operated model which can read a plate in seconds at wavelengths throughout the visible spectrum. For laboratories where large numbers of samples are routinely analysed then fully automated, computer controlled robotic systems are available capable of carrying out the complete sequence of analysis during a 24 hour operation. Whereas acceptable results can be calculated from standard curves fitted on log graph paper with a flexible ruler, data processing using a software package is more versatile. The computer will operate the plate reader and receive data directly from it. The most appropriate curve is usually a 4-parameter log plot which can be fitted to the data and confidence limits calculated. Sample values can be read off the graph and further processed with the determination of standard errors and coefficients of variation. The greater number of replicates possible using ELISA permits more thorough statistical analysis. Statistical design of ELISA protocols (38) can eliminate systematic errors. Calculation of a precision profile (39) will show the useful working range of the assay. 6.4.2 Practice It is the antibody which confers the extreme specificity and sensitivity to the ELISA technique. 6.4.2.1 Solid Phase Free and bound phases are usually separated by prior immobilisation of one of the reagents onto a solid support. Many ingenious separation systems have been designed particularly for commercial kits. Some use dipsticks or probes of different types which can be transferred from reagent to reagent between washing steps. One test uses a card in which antibody is immobilised on to a spot with an absorbent layer beneath. Sample, enzyme-labelled analyte and substrate are successively added. The colour intensity by reference to a control spot indicates the presence of analyte above a predetermined concentration. Some assays have been developed using beads of Sepharose, cellulose or polystyrene. A sensitive assay for aflatoxin M, in milk (40,41) used polystyrol beads but this requires an additional centrifugation step. However reagents linked to magnetic particles can be temporarily immobilised onto a magnetic surface. Most assays use the polystyrene surface of a tube or plate onto which the analyte-protein conjugate or specific antibody is bound. More recently the 96-well microtitration plate has
I30
predominated as the system of choice. This has many advantages of compactness, ease of handling and is suitable for automatic handling. Indeed plates are now available in which protein and even haptens can be covalently linked directly to spacer groups on the plate surface without the need to synthesise a conjugate. 6.4.2.2 Formats Several alternative methods are suitable for mycotoxin analysis. 6.4.2.2.1 Direct ELISA The wells of the microtitration plate are coated with the anti-mycotoxin antibody. Then sample or standard analyte are added at the same time as an amount of the enzyme conjugated analyte. The latter is partitioned between free and bound antibody during incubation and the unbound phase is washed away. Upon addition of, for instance, a chromogenic substrate for the enzyme the optical density of the developed colour can be measured spectrophotometrically. An alternative format is to immobilise an analyte-protein conjugate onto the plate, adding enzyme labelled antibody and sample/standard before developing with substrate. 6.4.2.2.2 Indirect ELISA A further variant is to introduce an additional step by using a labelled anti-species antibody rather than enzyme labelling the specific antibody. This has the advantage that there is no direct contact between the enzyme and the sample, which can cause interference with the previous methods. Whereas the analyte-enzyme conjugate is relatively unstable and would need to be resynthesised at intervals, the labelled anti-species antibodies are available commercially. Any of these formats can be used for semi-quantitativeanalysis and are especially useful when large numbers of samples need to be screened for the presence of mycotoxin contamination. However assays based on microtitration plates are eminently suited to fully quantitative analysis. 6.4.2.3 Antiserum titre It is a general principle that the more dilute the antisera, the more sensitive the assay. The limiting factors are the affinity of the antibody and the limit of detection of the signal is generated by the enzyme. To determine the optimal dilution of antiserum the bound and free reagents are titrated against each other at a range of concentrations in a two-dimensional "chequerboard". The target optical density is about 1.5 and several permutations of concentrations may achieve this; that using the least antibody is preferable. 6.4.2.4 Enzyme Many different enzymes such as fl-galactosidase and urease have been used but two others predominate. Alkaline phosphatase has been widely adopted with p-nitrophenol phosphate as substrate. The uptake of horse radish peroxidase (HRP)was held back due to lack of suitable substrates which were initially light-sensitive or putative carcinogens. However excellent substrates such as 2,2' azinobis (3-ethylbenzthiazoline) sulphuric acid (ABTS) and 3,3',5,5'
131
tetramethyl benzidine (TMB) are now available. As HRP has a high turnover rate, assays are more sensitive and it can be conveniently conjugated to haptens and proteins and this enzyme is now widely used. 6.4.2.5 Enhanced Systerns Assay sensitivity can be improved still further by using substrates which can be detected at lower concentrations. Thus fluorescent substrates are sometimes used or those based on chemiluminescence such as luminol or acridinium ester and appropriate plate readers are now becoming available. Radioisotopic labels such as ''I are being replaced due to the inherent hazards in handling isotopes and the short shelf life of the labelled compounds. Improved enzyme assays can now match and exceed the sensitivity of radioimmunoassays. An alternative strategy is to amplify the signal generated by the colorimetric immunoassay. It is possible to utilise the very high affinity of biotin for streptavidin which has four binding sites, with appropriate labelling to substantially increase the signal generated. Another approach is to enhance the endpoint detection with a second enzyme system. With NADP as substrate and alkaline phosphatase label the NAD generated is converted to NADH by a dehydrogenase. This is oxidised back to NAD in a colorimetric reaction. As each NAD molecule is recycled many times the signal is greatly magnified. 6.4.2.6 Validation Before an assay can be used quantitatively a series of rigorous tests must be conducted to validate and optimise the assay system. An ideal immunoassay for mycotoxins should have the following characteristics: Polyclonal antisera at < 1 : 10,OOO or Monoclonal antibody < 1 : 100 Limit of detection < 10 pg Specificity for desired mycotoxin(s) No interference from sample matrix or extraction solvent Quantitative recovery Reproducible in different laboratories
Correlate with alternative methods 6.4.2.6.1 Sensitivity A basically sigmoidal calibration curve is drawn plotting optical density against log concentration of a mycotoxin standard. Many mycotoxins can now be purchased commercially pure enough to serve as a standard or it may be possible to derive the material from a related toxin or to purify from fungal culture. In these cases purity can be checked by TLC or HPLC and concentration confirmed by U.V. absorption. The limit of detection of an assay usually refers to the lowest concentration of analyte which can be reliably differentiated from zero and is the point on the standard curve which lies at two or three times the standard deviation of the zero value. It should be noted however that this
132
value may not coincide with the working range of the assay as determined by the precision profile. The assay sensitivity usually refers to the minimum amount of analyte in the sample that can be determined. For an immunological assay, sensitivities to 1 ppb and even below can be routinely achieved. 6.4.2.6.2 Specificity The range of compounds detected in the assay is determined by the specificity of the antibody either for a single compound or a broader group of related mycotoxins, for instance the aflatoxins. To determine the cross-reactivity of an antibody the compound is substituted in the assay and the percent cross reaction calculated as the ratio of the amount of toxin required to give 50% inhibition of maximum binding to the amount of reference compound giving the same inhibition. 6.4.2.6.3 Ouantitative Recovery A number of factors can influence the apparent recovery of an assay, for instance interference from the food matrix or extract. This can be checked by testing for the superimposibility of standard curves in assay buffer and in blank matrix extract. Should there be a divergence then running the standards in a blank sample extract or in the extraction solvent can often solve the problem. Linearity can be tested for by assaying different volumes of the sample extract and plotting against mycotoxin recovery. The result should be a straight line of 45" slope. Recovery is best determined by spiking a blank sample preparation at several concentrations and is usually close to 100%without the need for correction factors. Repeating this analysis on several plates on different days gives a measure of the reproducibility of the assay. Less than 10%coefficient of variation is achievable in most assays. Finally, immunoassay results should correlate with other established techniques usually HPLC or TLC. 6.5 ILLUSTRATIVE EXAMPLES Over the last decade enzyme immunoassays have been developed for most of the major mycotoxins and increasingly for the minor toxins and putative carcinogens as they are identified. These have been extensively reviewed (2,7,42,43). 6.5.1 Aflatoxins To be of value an assay must have been fully validated for the analysis of a food product and found to correlate closely with at least one of the established techniques. Thus ELISAs for aflatoxin were compared with TLC and HPLC for cotton seed, corn and milk (44), TLC and LC for corn and mixed feed (45); TLC for figs (46) and HPLC and HPTLC for peanut butter (47). In each case immunoassays compared favourably with chromatographic measurements for quantitative analysis. In a review of the approaches to the rapid analysis of aflatoxins in food (48) immunoassays were compared and contrasted to TLC and HPLC not only on performance
I33 but also on cost in time and materials. In terms of cost effectiveness ELISA predominated due to its speed and the large numbers of sample throughput.
Immunoassays are usually developed in specialid rather than general analytical laboratories where the technique is required for practical application. Therefore the uptake of ELISA for routine analysis may have been held back due to the lack of availability of the antibody. However a growing number of self-contained kits are now commercially available for both semiqualitative and quantitative analysis. Literature is now appearing in which these kits have been critically evaluated and compared between brands and with established methods. There has been a particular interest in the use of qualitative kits for large scale screening of nuts and cereals especially for corn (maize) in the United States of America. Therefore two kits (49,50), three (51,52) and four (53) were compared. These rapid tests, some taking only a few minutes, indicate whether the test sample contain above or below a predetermined level, often 10 ppb. The focus of interest is on the level of false negatives or positives around the cut-off point. Generally the accuracy is better than 95%. These screening methods are often integrated with conventional chromatographic procedures (5435) to confirm the level of contamination reducing considerably the number of samples for full analysis. Only those few tests which fail to identify levels slightly above the cut-off are of concern. However by setting the screening assay to a lower level, say 5 ppb, even this small number of erroneous results is prevented but at the expense of additional samples for full analysis. Due to the promise shown by ELSA screening methods collaborative trials have been conducted each at a number of laboratories. A trial of compound feed reference materials compared two immunoassay methods with TLC and HPLC (56). Another compared five immunoassay kits with methods previously adopted (57). As a result of collaborative trials ELISA kits have been adopted official first action screening methods by AOAC for cotton seed products and mixed feed at 2 15 pg g-l and corn and roasted peanuts at 20 pg g-l (58,59)and cotton seed and peanut butter at 2 30 pg g-l (60). Although ELISAs are highly suited as a screening method, this use greatly undervalues their role in quantitative analysis. A quantitative commercial ELISA kit gave reproducible results correlating with HPLC for aflatoxin in nuts (61) and peanut butter (62). In collaborative trials ELISA kits gave quantitative results for aflatoxins in peanut butter (63-65). Following an international collaborative trial (66) the first ELISA kit was adopted first action by AOAC for the quantitative analysis of aflatoxin in peanut butter at concentrations of 9 to 90 pg kg-I. It seems therefore inevitable that ELISAs will be used in the near future as widely for quantitative analysis as they are at present for screening purposes.
6.5.2 Other mycotoxins Although kits are produced for several other mycotoxins (ochratoxin A and trichothecenes) they have not yet been so widely investigated as those for aflatoxins. However they will certainly be examined as ELISAs become accepted as a quantitative method. The trichothecenes are a group of mycotoxins which has proved difficult to analyse by
134
conventional chemical methods although specific immunoassays have been described for many of the trichothecenes. For instance, a sensitive monoclonal antibody has been raised to T-2 toxin (63). An assay for deoxynivalenol (DON) has been applied to wheat (33) and to the acetylated derivative 3-acetoxynivalenol (ADON) to rice (32). Validated in wheat, an assay for diacetoxyscirpenol (DAS) has also been described (31). Attempts have been made to raise a generic antibody to a ‘type’ molecule, with broad specificity recognition to trichothecenes ( 13,68). A monoclonal antibody specific for zearalenone has been used in an ELISA (69) with negligible cross reactions to zearalenol and zearalanols or the trichothecenes. An assay for malenone and related metabolites has been used to monitor for these mycotoxins in porcine urine (70). An ELISA for ochratoxin A has been developed for analysis of barley (37) and porcine kidney (36). The survey of cereals, feedstuffs and porcine kidney using a radioimmunoassay (71) has shown widespread levels of contamination. Immunoassays to a number of other mycotoxins have been developed; sterigmatocystin (72); roridin A (73); cyclopiazonic acid (74); kojic acid (75,76);rubratoxin (77) and to the recently identified fumonisins (78). New approaches to the application of immunotechniques are constantly being devised. Currently efforts have been directed towards broad spectrum analysis either by the use of broad specificity antibodies or multi analyte systems. Using antibodies immobilized onto a nylon membrane test strip, a method has been developed for the simultaneous detection of aflatoxin, trichothecenes, ochratoxin and zearalenone (79). Another system envisaged by Ekins (80) uses tiny discs onto which many antibodies are placed as dots. After exposure to the test sample the disc is developed using analytes labelled with fluorescent markers each emitting at a different wavelength. The ELISAGRAM (81) is a technique which combines the ability of HPTLC to separate closely related mycotoxins with the sensitivity of ELISA by the process of immunoblotting. After development, the HPTLC plate is overlain with a sheet of nitrocellulose coated with mycotoxin specific antibody. These sites can then be quantified by densitometry after reaction with substrate. The difference to the control is a measure of the mycotoxin present. 6.6 CONCLUSIONS Chromatographicand immunoanalytical techniques may therefore be complementary. Certainly, ELISAs, as they become available, are an alternative to the more time consuming chromatographic process for routine analysis. Particularly for large scale screening, antibody techniques will replace chromatography. Frequently however the techniques will be supplementary; for instance immunoaffinity columns are used as a clean up step before chromatography or as the specificity of immunoassays is confirmed by chromatography (see also Chapter 5).
135
REFERENCES 1 M. Ollerich, in H.U. Bergemeyer (Editor in Chief), Methods of enzymatic analysis, Vol. I Fundamentals, Verlag Chemie GmbH, Weinheim, (1983) p 233. 2 M.R.A. Morgan, Tetrahedron, 45 (1989) 2237. 3 J.H. Rittenburg, in J.H. Rittenburg (ed), Development and application of immunoassay for food analysis, Elsevier, London, (1990) p 29. 4 N.A. Staines, J . Brostoff and K. James, Introducing immunology, Gower Medical Publishing, London, (1985). 5 W.E. Paul (ed), Fundamental immunology, Raven, New York, (1984). 6 G. Kholer and C. Milstein, Nature, 256 (1975) 495. 7 A.P. Wilkinson, C.M. Ward and M.R.A. Morgan, in H.F. Linskens and J.F. Jackson (eds), Plant toxin analysis, Modern methods of plant analysis, New series Vol 13, Springer-Verlag, Heidelberg, (1992) 185. 8 C.M. Ward, A.P. Wilkinson, S. Bramham, H.A. Lee, H.W-S. Chan, G.W. Butcher, A. Hutchings and M.R.A. Morgan, Mycotox. Res., 6 (1990) 73. 9 J.J. Pestka, J. Assoc. Off. Anal. Chem., 71 (1988) 1075. 10 L.H. Stanker, B.E. Watkins and M. Vanderlaan, in H. Frehse (ed), Pesticide chemistry. Advances in international research, development and legislation. Proceedings of the Seventh International Congress of Pesticide Chemistry (IUPAC), Hamburg 1990, VCH, Weinheim (1992) p 397. 11 K. Landsteiner, The specificity of serological reactions, Harvard Univ. Press, Boston (1945). 12 M.R.A. Morgan, A.S. Kang and H.W-S. Chan, J . Sci. Food Agric., 37 (1986) 908. 13 R.D. Wei and F.S. Chu, Anal. Biochem., 160 (1987) 399. 14 S.L. Hefle and F.S. Chu, Food Agric. Immunol., 2 (1990) 181. 15 R.S. Yalow and S.A. Berson, Nature, 184 (1959) 1648. 16 J. Daussant and D. Bureau, in R.D. King (ed), Development in food analysis techniques Vol 3, Elsevier Applied Science, London (1984) p 175. 17 C.J. Smith, in J.H. Rittenburg (ed), Development and application of immunoassay for food analysis, Elsevier, London (1990) p 3. 18 J.P. Gosling, Clin. Chem., 36 (1990) 1408. 19 K. Jewers, R.D. Coker, G. Blunden, M.J. Jazwinsky and A.J. Sharky, in: B. Flanigan (ed), Spoilage and mycotoxins of cereals and other stored products, Int. Bidet. Supp., 22 (1986) 83. 20 P.L. Schuller, W. Horwitz and L. Stollof, J. Assoc. Off. Anal. Chem., 59 (1976) 1315. 21 D.L. Park and A.E. Pohland, J . Assoc. Off. Anal. Chem., 72 (1989) 399. 22 F.M. Garfield, J. Assoc. Off. Anal. Chem., 72 (1989) 405. 23 H.P. Van Egmond, Food Addit. Contam., 6 (1989) 139. 24 J.J. Pestka, Y. Li, W.O. Harder and F.S. Chu, J. Assoc. Off. Anal. Chem., 64 (1981) 294.
I36
25 J.M. Fr6my and F.S. Chu, J. Assoc. Off. Anal. Chem., 67 (1984) 1098. 26 A.P. Wilkinson, D.W. Denning and M.R.A. Morgan, Food Addit. Contam., 5 (1988)
609. 27 P. Sizaret, C. Malaveille, R.L. Monsanto and C. Frayssinet, J. Natl. Cancer Inst., 69 (1982) 1375. 28 C.P. Wild, D. Umbenhauer, B. Chapot and R. Montesano, J. Cell Biochem., 30 (1986) 171. 29 R.D. Stubblefield, J.I. Greer, O.C. Shotwell and A.M. Aikens, J. Assoc. Off. Anal. Chem., 74 (1991) 530. 30 R.C. Lee, R.D. Wei and F.S. Chu, J. Assoc. Off. Chem., 72 (1989) 345. 31 E.N.C. Mills, J.M. Johnston, H.A. Kemp and M.R.A. Morgan, J. Sci. Food Agric., 42 (1988) 225. 32 H.A. Kemp, E.N.C. Mills and M.R.A. Morgan, J. Sci. Food Agric., 37 (1986) 888. 33 E.N.C. Mills, S.M. Alcock, H.A. Lee and M.R.A. Morgan, Food Agric. Immunol., 2 (1990) 109. 34 T.B. Whitaker and J.W. Dickens, J. Assoc. Off. Anal. Chem., 69 (1986) 508. 35 M.D. Friesen, J. Toxicol. Toxin Rev., 8 (1989) 363. 36 M.R.A. Morgan, A.S. Kang and H.W-S. Chan, J. Sci. Food Agric., 37 (1986) 873. 37 M.R.A. Morgan, R. McNerney and H.W-S. Chan, J. Assoc. Off. Anal. Chem., 66 (1983) 1481. 38 D.S. Bunch, D.M. Rocke and R.O. Harrison, J. Immunol. Meth., 132 (1990) 247. 39 R.P. Ekins, in W.M. Hunter and J.E.T. C o m e (eds), Immunoassays for clinical chemistry, Churchill-Livingstone, 76 (1983). 40 J. von Steiner, G. Hahn and W. Heeschen, Milchwissenshaft, 43 (1988) 772. 41 F.F.J. Nieuwenhof, J.D. Hoolwerp and J.W. van den Bedem, Milchwissenshaft, 45 (1990) 584. 42 M.R.A. Morgan and H.A. Lee, in J.H. Rittenburg (ed), Development and application of
immunoassays for food analysis, Elsevier, New York, (1990) 143. 43 F.S. Chu, in M. Vanderlaan, L.H. Stanker, B.E. Watkins and D.W. Roberts (eds), Immunoassays for trace chemical analysis, Amer. Chem. Soc. Symp., 45 1, Washington (1990) 140. 44 D.E. Dixon-Holland, J.J. Pestka, B.A. Bidigare, W.L. Casale, R.L. Warner, B.P. Ram and L.P. Hart, J. Food Protect., 51 (1988) 201. 45 K-I. Hongyo, Y. Itoh, E. Hifumi, A. Takeyasu and T. Uda, J. Assoc. Off. Anal. Chem., 75 (1992) 307. 46 N. Reichert, S. Steinmeyer and R. Weber, Z. Lebenzm. unters. Forsch., 186 (1988) 505. 47 M.P.K. Dell, S.J. Haswell, O.G. Roch, R.D. Coker, V.F.P. Medlock and K. Tomkins, Analyst, 115 (1990) 1435. 48 M.J. Shepherd, D.N. Mortimer and J. Gilbert, J. Assoc. Publ. Anal., 25 (1987) 129. 49 M.W. Trucksess, K. Young, K.F. Donahue, D.K. Moms and E. Lewis, J. Assoc. Off. Anal. Chem., 73 (1990) 425.
137
50 W.F. Wilke, M.F. Ehrich and K.W. KO,App. Agric. Res., 5 (1990) 32. 51 J.W. Dorner and R.J. Cole, J. Assoc. Off. Anal. Chem., 72 (1989) 962. 52 A.L. Patey, M. Sharman, R. Wood and J. Gilbert, J. Assoc. Off. Anal. Chem., 72 (1989) 965. 53 W.P. Cochrane, in M. Vanderlaan, L.H. Stanker, B.E. Watkins and D.W. Roberts (eds), Immunoassay for trace chemical analysis, Amer. Chem. SOC.,Washington, (1990) 40. 54 L.S. Lee, J.H. Wall, P.J. Cotty and P. Bayman, J. Assoc. Off. Anal. Chem., 73 (1990) 581. 55 R.W. Beaver, M.A. James and T.Y. Liu, J. Assoc. Off. Anal. Chem., 74 (1991) 827. 56 H.P. van Egmond and P.J. Wagstaff, Food Addit. Contam., 7 (1990) 239. 57 D.E. Koeltzow and S.N. Tanner, J. Assoc. Off. Anal. Chem., 73 (1990) 584. 58 D.L. Park, B.M. Miller, L.P. Hart, G. Yang, J. McVey, S.W. Page, J.J. Pestka and L.H. Brown, J. Assoc. Off. Anal. Chem., 72 (1989) 326. 59 D.L. Park, B.M. Miller, S. Nesheim, M.W. Trucksess, A. Vekich, B. Bidigare, J.L. McVey and L.H. Brown, J. Assoc. Off. Anal. Chem., 72 (1989) 638. 60 M.W. Trucksess, M.E. Stack, S. Nesheim, D.L. Park and A.E. Pohland, J. Assoc. Off. Anal. Chem., 72 (1989) 957. 61 M. Azer and C. Cooper, J. Food Protect., 54 (1991) 291. 62 C.M. Ward and M.R.A. Morgan, Food Addit. Contam., 8 (1991) 9. 63 D.N. Mortimer, M.J. Shepherd, J. Gilbert and C. Clark, Food Addit. Contam., 5 (1988) 601. 64 A.L. Patey, M. Sharman, R. Wood and J. Gilbert, J. Assoc. Off. Anal. Chem., 72 (1989) 965. 65 A.L. Patey, M. Sharman, and J. Gilbert, J. Mycotox. Res., 6 (1990) 2. 66 A.L. Patey, M. Sharman, and J. Gilbert, J. Assoc. Off. Anal. Chem. Int., 75 (1992) 693. 67 R. Hack, E. Martlbauer and G. Terplan, Lett. Appl. Microbiol., 9 (1989) 133. 68 R.C. Lee, D.L. Bunner, R.W. Wannermacher and F.S. Chu, J. Agric. Food. Chem., 38 (1990) 444. 69 R. Teshima, M. Kawase, T. Tanaka, K. Hirai, M. Sato, J. Sawada, H. Ikebuchi, M. Ichinoe and T. Terao, J. Agric. Food Chem., 38 (1990) 1618. 70 O.A. MacDougal, A.J. Thulin and J.J. Petska, J. Assoc. Off. Anal. Chem., 73 (1990) 65. 71 L. Fukal, Food Addit. Contam., 7 (1990) 253. 72 M.R.A. Morgan, R. McNerney and H.W-S. Chan, J. Sci. Food Agric., 37 (1986) 475. 73 R. Hack, E. Martlbauer and G. Terplan, Appl. Environ. Microbiol., 54 (1988) 2328. 74 S. Hahnau and E.W. Weiler, J . Agric. Food Chem., 39 (1991) 1887. 75 D.W. Lawellin, D.W. Grant and B.K. Joyce, Bacteriological Proceedings, 75 (1975) 209. 76 A.E. Abdalla and D.W. Grant, Sabouraudia, 18 (1980) 191. 77 R.M. Davis and S.S. Stone, Mycopathologia, 67 (1979) 29. 78 J.I. Azcona-Olivera, M.M. Abouzied, R.D. Plattner and J.J. Pestka, J. Agnc. Food Chem., 40 (1992) 531.
138
79 E. Schneider, R. Dietrich, E. Malbauer, E. Usleber and G . Terplan, Food Agric. Immunol., 3 (1991) 185. 80 R.P. Ekins, F . W . Chu and E.M. Biggart, Clin. Immunoassay, 13 (1990) 169. 81 J.J. Pestka, J. Immunol. Meth., 136 (1991) 177.
PART B
APPLICATIONS
This Page Intentionally Left Blank
141
Chapter 7 THIN-LAYER CHROMATOGRAPHY OF MYCOTOXINS
V. BETINA 7.1 INTRODUCTION
Of the chromatographic techniques applied to mycotoxins, thin-layer chromatography (TLC) has been and remains by far the most widely used in the detection, analysis and characterization of fungal toxins. However, in the early stages of mycotoxicology it was paper chromatography (PC) that preceded its application. PC was used by Sargeant et al. ( 1 ) in the first separation of aflatoxins from a crude extract of groundnuts. The toxic component produced a single spot that fluoresced under UV light. Several other applications of PC in mycotoxicology may be found in ref. 2. In studies of the metabolites of the toxigenic fungus Penicillium islandicum, PC was adopted for the identification of its anthraquinones (islandicin, iridoskyrin, catenarin, chrysophanol, erythroskyrin, luteoskyrin, flavoskyrin, rubroskyrin and oxyskyrin) and applied to the detection of anthraquinone-producing strains (3). Another mycotoxin, cyclochlorotine or chlorine-containing peptide, isolated from P. islandicum has been characterized by PC in several solvent systems ( 4 ) . The historical precedence of PC in the field of mycotoxins has been discussed in a recent review ( 5 ) . A classification of antibiotics by means of “salting-out paper chromatograms” ( 6 ! 7 ) was extended to fungal metabolites, including the following compounds currently classified as mycotoxins: alternariol, aspergillic acid, citrinin, cyanein (brefeldin A), gliotoxin, griseofulvin, kojic acid, mycophenolic acid, neohydroxyaspergillic acid, patulin, penicillic acid, rugulosin, terreic acid and trichothecin ( 8 ) . Another principle of classification by means of the so-called ntpH-chromatogramsll has been introduced (9, 10). Using this PC technique, the ionic character of unknown antibiotics and also the general possibilities of their isolation could be determined when only their crude concentrates from Petri dish cultures were available. The principles of “pH-chromatographyIl have been reviewed elsewhere (11, 12). Several fungal metabolites belonging now to mycotoxins, such as citrinin, mycophenolic acid, gliotoxin, rugulosin, neohydroxyaspergillic acid and trichothecin, have been studied by llpH-chromatographyft (13, 14).
I42
PC studies of compounds in several solvent systems for the purpose of their classification and identification have proved useful in systematic analysis. A combination of the sequential and simultaneous analysis of antibiotics was elaborated by the present author. In addition to a series of antibiotics, the following mycotoxins were characterized by the method: kojic acid, penicillic acid, cyanein (brefeldin A ) , alternariol, rugulosin, citrinin, trichothecin, patulin, gliotoxin, mycophenolic acid, aspergillic acid and griseofulvin (15). In our laboratory, the three above-mentioned PC techniques were found to be helpful in the isolation and identification of citrinin produced by a penicillium strain (16). Several strains of a collection of fungi have been found to produce substances that inhibited mycobacteria. When the active substances, present in crude extracts from agar cultures of four aspergilli, were analysed by systematic PC analysis (15), all of them belonged to subclass IIa and their Ifsummarized chromatogramsll were similar to that of kojic acid of the same subclass. The identity to kojic acid was confirmed after the isolation and purification of the active products (17). However, TLC superseded PC after Stahl standardized the procedure at the end of the 1950s and showed its wide applicability. With advances in techniques, TLC has become the method of choice for some mycotoxins. It was shown as early as in 1962 by Nesbitt et a l . (18) and, in 1963, by Hartley et al. ( 19) , that the single blue-f luorescing spot of aflatoxin-containing extract, observed by PC, could be split into four main components when the extracts were chromatographed on silica gel TLC plates developed in chloroform-methanol. Two of these components, fluorescing blue under UV light, were designated aflatoxin B1 and B and the other two components, fluorescing turquoise under UV fight, were designated aflatoxins G1 and G2. Since then various combinations of silica gel and solvent systems have been proposed for separating aflatoxins by TLC Pure aflatoxins for structural determination were obtained by preparative TLC (PLC), and quantitative methods have been developed. In some instances the aflatoxins served as the model compounds for the development of reliable methods to measure trace levels in foods and feeds. The TLC of aflatoxins has received the most attention over the years (20): consequently, it is the most refined and generally serves as a model for other mycotoxins. In a chapter on TLC and high-performance TLC (HPTLC) of mycotoxins, the chief advantages of TLC have been characterized by Nesheim and Trucksess ( 2 1 ) as follows: "It is simple and economical. Generally, it is nondestructive, so that compounds may be recovered for further analysis. The adsorbent can be impregnated with a variety of chemicals so as to achieve specific separations. Quantitation
.
143
can be carried out either visually or instrumentally, and to achieve a number of reagents are available visualization of specific compounds. The capacity of one-dimensional TLC can be as many as 25 discrete sample extract spots, and for two-dimensional TLC it can be 100 to 4 0 0 spots. Until recently the precision of measurement of the spots, expressed as coefficients of variation (CVs) . , was in the range of 5 to 35%. Instrumentation that greatly improves this precision has become available. The absorbance of aflatoxin spots can now be determined automatically with CVs less than
. .
.
.
.
1%.
The advantages and disadvantages of the use of TLC in mycotoxicology have been discussed by Shepherd (22) and Rogers (23). In the 1980s, numerous reviews and book chapters on the chromatography of mycotoxins in general, and on TLC in particular, have been published (2, 24-38). Techniques of TLC in mycotoxicology are described in Chapter 2 of the present book. Their applications are presented in this chapter. Here, mycotoxins are mostly classified according to their structures in ltfamiliesll. When possible, data are arranged in the following order: (i) sampling and sample preparation, (ii) extraction and clean-up, (iii) adsorbents and solvent systems, (iv) detection, (v) applications of one-dimensional, bi-directional and two-dimensional TLC, (vi) HPTLC, (vii) quantitation, and (viii) preparative TLC (PLC). 7.2 AFLATOXINS
The four naturally-occurring aflatoxins, B1, B2, G1 and G , acutely toxic and carcinogenic metabolites produced 2y Aspergillus flavus and the closely related species A. parasiticus. Other members of the group are derived from these four toxins as metabolic products of microbial and animal systems (such as M1, M2, P1, Q and aflatoxicol) or are produced spontaneously in response to the chemical environment (such as Baa, G2 and D1). The aflatoxins are highly fluorescent, highly oxygenated , heterocyclic compounds characterized by dihydrodifurano or tetrahydrofurano moieties fused to a substituted coumarin moiety. Aflatoxin B is the most prevalent naturally-occurring member of the group that has had the most profound impact on the development of mycotoxicology. A survey of applications of TLC is given below on sampling and sample preparation, extraction and clean-up techniques, adsorbents and solvent systems, detection, qualitative and quantitative analyses, and preparative TLC (PLC) of aflatoxins. 7.2.1 Sampling and sample preparation According to Dickens and Whitaker (39), sampling and subsampling for aflatoxin analyses is more difficult than for other known mycotoxins in agricultural products, and sampling and subsampling procedures recommended for aflatoxins should be are
144
adequate for other mycotoxins. The problem of sampling is associated with the fact that there is not a normal distribution of aflatoxins within one batch. Thorough reviews for sampling and subsampling for mycotoxin analyses have been published by Dickens and Whitaker (40) and, more recently, by Park and Pohland (41). Both reviews provide lists of various types of equipment used for sample preparation and sources of supply. Some data may also be found in Chapter 1 of the present book. 7.2.2 Extraction and clean-up Extraction and clean-up procedures for aflatoxins have been described in refs. 5, 42-52. As many additional compounds present in samples contaminate primary aflatoxin extracts, these components must be removed as completely as possible in order to prevent interference with the final purification and assay. In addition to lipids, some commodities contain other components that interfere with subsequent analyses by TLC and should be removed from the primary extracts. Primary extracts in mixtures of acetone with water contain proteins that can be precipitated with lead acetate. Extracts from cotton-seed contain interfering gossypol pigments: extracts from cereals contain fluorescing substances with TLC properties similar to those of aflatoxins. These pigments need to be removed by column chromatography or by other means. It was noticed by Heathcote and Hibbert (53) that one of the difficulties encountered in the analysis of extracts from mycelia of Aspergillus flavus was the occurrence of a dark pigment, which remained immobile on the TLC plate and interfered seriously with the chromatography. This pigment was formed during air-drying of the mycelium. BY using freeze-drying and Soxhlet extraction at a temperature below 35OC, a considerably cleaner aflatoxin extract was obtained with a consequent improvement in the quality of the chromatogram. In coffee-beans it is caffeine and other compounds whereas in cocoa-beans it is chiefly theobromine that must be removed. Scott (54) showed that a coffee-bean extract could be purified by passage through a Florisil column and the unwanted contaminants eluted with tetrahydrofuran. It was also shown by the same author (55) that theobromine could be removed from crude cocoa-bean extracts by treatment with silver nitrate solution. A variety of extraction and clean-up procedures for aflatoxins have been published. Schuller et al. (56) reviewed Sampling procedures and collaboratively studied methods for the analysis of aflatoxins. A n exhaustive review of older extraction and clean-up procedures may be found in the classic book by Heathcote and Hibbert (46). They include extractions from chemically defined media, mycelia, as well as from natural commodities and derived products (peanuts, cotton-seed, cereals, milk, meats, coffee, cocoa-beans, coconut, copra, copra meal and
145
animal feedstuffs). These procedures were given in a more condensed form in a book devoted to production, isolation, separation and purification techniques for mycotoxins (47). The determination of aflatoxins in vegetable oils is usually based on partition between two immiscible solvents. Aflatoxins are extracted from the oil into a polar solvent and subsequently partitioned into chloroform. Athough the recoveries are acceptable, these methods are time consuming, require large volumes of solvents, and frequently involve troublesome emulsions. Miller et a1.(57) proposed a simple method for the determination of aflatoxins which was successfully applied to both crude and degummed oils. The oil sample, dissolved in hexane, was applied to a silica gel column and washed with diethyl ether, toluene and chloroform. Aflatoxins were eluted from the column with chloroform-methanol (97:3). Quantitation was then performed by TLC or HPLC. Lovelace et al. (58) published a screening method for the detection of aflatoxins and metabolites in human urine. In the clean-up, Celite 545 filter aid was added to the samples, followed by acetone and filtration. To the stirred filtrate was added water followed by 20% lead acetate solution. After coagulation, a saturated solution of sodium chloride was added with stirring, followed by Celite. The mixture was filtered, the filtrate stirred, and freshly prepared iron(II1) hydroxide slurry added, followed by Celite 545. After filtration, 0.1% sulphuric acid was added and the filtrate was extracted twice with chloroform. The combined extracts were washed with 5% NaCl solution and evaporated to dryness under nitrogen. The residue was dissolved in chloroform and again evaporated. The final residue was re-dissolved in chloroform and analysed by TLC. Chapter 26 of the 14th edition of the Official Methods of Analysis (26) contains AOAC approved analytical methods for the determination of aflatoxins involving sample preparation, TLC quantitation, and employ column chromatography, liquid-liquid extraction or chemical adsorption methods for removing interfering compounds. Further methods can be found in Chapter 49 of the 15th edition of the Official Methods of Analysis (44). Modifications and improvements of extraction and clean-up techniques have been published (59-64). Modern methods of aflatoxin extraction use water-organic solvent mixtures, e . g . , chloroform-water (1O:l) (26). The water wets the substrate and the small amount taken up in the aqueous phase is immediately removed by the solvent, giving a rapid isolation procedure. Velasco (65) proposed the replacement of benzene as a solvent for aflatoxin standards. The best known extraction and clean-up techniques, as published before 1985 were summarized by the present author (30). More recent techniques are described below.
I46
Reichert et al. (38) isolated aflatOXin B1 from dried figs as follows. The figs were cut into smaller pieces, and immersed in 50 mL of methanol for 16 h. From the extract 2 5 mL were taken, 20 mL of water were added and the mixture was extracted twice with 50 mL of dichloromethane. The extracts were combined, dried over sodium sulphate, the solvent was removed under reduced pressure and the residue was dissolved in 1 mL of dichloromethane. The figs with the rest of the methanol were homogenized for 5 min in a Waring Blendor with a gradual addition of 75 mL of methanol and 4 0 mL of water. After filtration, 70 mL of the filtrate were removed, 2 0 mL of water were added and the mixture was extracted twice with 50 mL of dichloromethane. Both dichloromethane phases were filtered through sodium sulphate, combined and evaporated under reduced pressure. To remove residues of water, 10 mL of dichloromethane were added, shaken and filtered through sodium sulphate. The solvent was evaporated and the residue was dissolved in 1 mL of dichloromethane. The ability of different concentrations of aqueous acetone, aqueous methanol and aqueous acetone-methanol (1:l) extract aflatoxin from naturally contaminated maize was assessed by Bradburn et al. ( 4 5 ) . With each system, the amount of aflatoxin extracted increased as the ratio of organic so1vent:water progressed from 5 0 : 5 0 to 8 0 : 2 0 and then decreased or remained constant when the composition of the extraction solvent was 80% Aqueous acetone was found to extract 27% altered to 9O:lO. more aflatoxin than the corresponding aqueous methanol with the mixed solvent system extracting an intermediate amount of toxin. Clean-up of 8 0 % aqueous acetone extracts of maize by elution through phenol (PH) bonded-phase cartridges resulted in poor aflatoxin retention when the proportion of acetone in 5 mL aliquots of extract was greater than 70%. This could be increased to 100% retention by the addition of an equal amount of methanol prior to dilution with 1% aqueous acetic acid and elution through the cartridge. The recent development of bonded phase clean-up methods (66) using disposable cartridges offers the possibility of rapid and automated clean-up for use in aflatoxin assays. Polar and non-polar bonded phases can be used. When polar bonded phases (67, 68) in aflatoxin analysis, (Si) have been employed liquid-liquid extraction, followed by solvent evaporation and reconstitution with a non-polar solvent was necessary before the extract could be added to the Si cartridge. Some authors have included further clean-up steps, such as defatting using n-hexane (67) or a preliminary treatment with ammonium sulphate solution (68). Non-polar bonded phases can considerably simplify the clean-up sequences. For example, in the assay of aflatoxin MI in milk (69), the diluted milk is applied directly to an octadecyl (C18) cartridge, interfering compounds are removed by washing
147
with water, basic and then acidic acetonitrile. Aflatoxin is eluted with further additions of acidic acetonitrile. Other assays utilizing non-polar bonded phase clean-up of sample extracts containing the aflatoxins include either liquid-liquid extraction ( 7 0 ) or a sequence of extraction cartridges, for example, either two C18 cartridges (71), or Si, octyl (C8) and cyanopropyl (CN) cartridges (72). Tomlins et al. (73) tested the suitability of a variety of commercially available non-polar bonded phase cartridges octadecyl (C18), octyl (C8), ethyl (C 2): cyclohexyl (CH) and phenyl (PH) for the rapid clean-up of maize extracts prior to the determination of aflatoxin levels by HPTLC. Their procedures were as follows. Extraction. Either aflatoxin-free or contaminated ground maize was blended with water (1:1.25 w/v) in a blender at low speed for 2 min. Methanol (200 mL) was added to a portion (100 g) of the resultant slurry and the mixture was blended at high speed in a blender for 3 min. The product was filtered through a Whatman No 1 filter paper and about 50 mL of the crude filtrate collected. Clean-up. A bonded-phase cartridge (500 mg) was solvated eluted with water (10 mL). Maize with methanol (10 mL) and extract (10 mL) diluted to a solution containing 11.1% methanol by 1% aqueous acetic acid (v/v; 60 mL) was passed through a solvated cartridge at a flow rate of 10 mL/min. The cartridge was washed with distilled water ( 1 0 mL) and dried by passing air through it for 5 min. The cartridge was connected to the top of a 4 mL reservoir fitted with a 20 mm frit containing granular anhydrous sodium sulphate ( 3 9). Elution of the cartridge with chloroform (7 mL; 1 mL/min) produced a solution of aflatoxins in the solvent, which was evaporated to dryness at 5OoC under a gentle stream of nitrogen in a sample concentrator. Of the non-polar bonded phases examined, the PH phase has been found to afford the best recovery of aflatoxin (97-102%) when aqueous solutions (1% v/v) of acetic acid containing 5, 10, 20 and 3 0 % of methanol were used as the mobile phase and chloroform ( 7 mL) was employed to elute aflatoxins. Bradburn et a l . ( 7 4 ) compared a modified version of the method of Tomlins et a l . (73) with the first action AOAC and Romer methods for determining aflatoxin levels in maize by bi-directional HPTLC. The modified phenyl non-polar bonded phase clean-up procedure was found to recover significantly more aflatoxin from a sample of naturally contaminated maize, and to have accuracy better than, and precision equivalent to the other methods. In addition, it was shown to be more rapid and cost effective than the AOAC methods. Another modification of the PH non-polar bonded phase clean-up procedure ( 7 3 ) , in the clean-up of cottonseed extracts,
I48
involves an addition of lead acetate solution to the extract prior to its passage through the bonded phase cartridge (75). The lead acetate solution was prepared as described in the AOAC method for cottonseed (26). Comparison of the former with the latter method showed that PH cartridge method recovered 80% more aflatoxin B1 and 5 2 % more B2 from a sample naturally contaminated with aflatoxin. It was also more precise for aflatoxin B1 as well as being more rapid and cost effective. Kamimura et a l . (76) used Florisil columns for clean-up procedures. The procedure was tested by Jewers et al. (77) for its suitability for determining aflatoxin levels in sorghum grains by bi-directional HPTLC and was compared with the CB (Contamination Bureau) method. Both the accuracy and precision of Kamimura's method were found to be comparable to, or better than, the AOAC procedure. Most recently, an intercomparison of methods for aflatoxin B1 in compound-feed reference materials was summarized by van Egmond and Wagstaffe (78). The study involved 24 European laboratories which analysed two pairs of compounded feedstuffs. The participants used a variety of methods of analysis, including chloroform, methanol-water, and acetonitrile-water for extraction, TLC, reversed phase disposable cartridges, and immunoaffinity columns for extract clean-up, and TLC, HPLC and ELISA in the determinative step. Extraction and clean-up procedures used prior to a TLC determinative step are given in Table 7.1. Extraction and clean-up procedures for aflatoxins have been reported by many workers (79-114). Culture filtrates have been extracted with chloroform (83, 84) and the extracts washed with light petroleum or hexane (84). Chloroform has also been used to extract mycelium (53), agar medium (85) and mycelium plus culture filtrate (86). Extraction and clean-up procedures for aflatoxins in groundnuts and products thereof have been described (87-99). For extractions, mixtures of solvents included hexane-methanol-water, hexane-water-acetone, aqueous acetone, chloroform-methanol-hexane, etc. Most clean-up procedures have been based on precipitations with lead acetate or iron(II1) hydroxide and column chromatography. Similar techniques have been applied for analysis of cotton seeds (100-103), cereals and flours (97, 104-108), coffee-beans and cocoa-beans (99), and other other plant products (109). Aflatoxins were extracted from meats (110) with chloroform, purified by CC on silica gel and eluted with chloroform-methanol (97:3). Mixed feeds were extracted with 20% citric acid solution-methylene chloride (1:lO) followed by clean-up using CC on silica gel, washed with glacial acetic acid-toluene (1:9), tetrahydrofuran-hexane (1:3), or acetonitrile-diethyl ether-hexane (1:3:6) and eluted with acetone-methylene chloride (1:4) (111). Powdered milk was extracted with acetone-chloroform-water, impurities were
149
precipitated with lead acetate, the filtrate defatted with hexane and aflatoxin extracted with chloroform (112, 113). Liquid milk was mixed with acetone, dialysed against aqueous acetone, extracted with chloroform and dried (114). Lovelace et al. (58) extracted urine with chloroform after precipitation with lead acetate solution and iron(II1) hydroxide. The extract was washed with NaCl solution and evaporated. TABLE 7.1 Extraction and clean-up methods for aflatoxin B1 animal feedstuffs used prior to TLC determination Selected from ref. 78.
in compound
Extractiona
Clean-upa 1’
20 g sample, 200 mL CHC13, 20 mL H20, 20 g Celite
sio, column. Final solvent: CHC13
50 g sample, 250 mL CHC13, Solvent partitioning. Final 50 mL saturated NaCl solution solvent CH2C12 50 g sample, 250 mL CHC13,
sio,
25 g sample, 125 mL CHCl 12.5 mL H20, 12.5 g Celii!;!
Florisil Sep-Pak + C18 Sep-Pak. H20-Me2C0 (85:15). Extracted with CHC13
25 mL H20, 25 g Celite
co umn. Final solvent: CHC13
a Abbreviations: CHCl , chloroform: Si02, silica gel: H 0, water: NaCl , sodium ch?oride; CH2C12, methylene chloriAe; Me C acetone. sequence of additions.
grin
7.2.3 Adsorbents and solvent systems is the mostly used adsorbent in TLC of Silica gel aflatoxins. The first successful1 separation of aflatoxins into the main components, B l r B2, G1 and G2, was carried out on silica gel plates using chloroform-methanol (98:2) by de Iongh et al. (79). TLC on plates of rice starch and CaS04 were used to separate aflatoxins and other mycotoxins using three solvent systems (80). Numerous combinations of silica gel and solvent systems have been proposed in efforts to improve the separation and to obtain more reproducible RF values. Most of the solvent systems were based on chloroform plus 2-7% of methanol. Subsequently, methanol was replaced with 10-15% of acetone. Improvements proposed by Kozloski (81) and by Issaaq and Cutchin (82) are described below. The use of non-chloroform based solvent systems
150
has been suggested Table 7.2.
by several
workers. Examples
are given in
TABLE 7.2 Selected solvent systems for TLC of aflatoxins SystemX CHC13-MeOH (98:2) CHC13-MeOH (94:6 or 49:l) CgHg-EtOH-H20 (46:35:19)
NotesX
Ref.
Separation of Blr B2, G1 and G2 47 Aflatoxins plus sterigmatocystin 53 and versicolorin group on SilicAR Tol-EtOAc (8:l) TLC-7G CHC13-Me2CO-Hex (85:15:20) On SilicAR AGF 115 CHC13-Me2CO-iPrOH (825:150:25) 91 CHC13-Me CO-(85:15) PLC of B1 116 117 Et O-MeOi-H20 (93:3 :1) Csfi6-EtOH-H 0 (46:35:19, Sensitive to changes in 118 upper layer? humidity B2, G1 and G2 117 H 0-MeOH-Et20 (1:3:96) Separation of B i8rOH-Me CO-H20 (5:10:85) M1 in milk and iAiry products 119 CHCl3-Me8H (9:l) B1 in extracts from A. flavus 86 cultures CH2Cl -MeOH (95:5) D1 in ammoniated corn extracts 120 Quantitation of Blr B2, and G2 109 Et 0 8ollowed by CH$13-Me CO-H20 (88:12 :1.5 ) in groundnuts Et20 folfowed by CHCl As above 121 Me CO-CgHg (9:l:l) and by CHZ1 -Me2CO-Hex (71:?2.5: 16.5) CHC13-EtOAc-THF (8:12:0.6) Over-pressured liquid chroma108 tography on HPTLC Kieselgel 60 CHC13-Me2CO-NH3 (90:10:0.25) Separation of B1, B2' G1 and G2 82 on six commercial silica gel plates CHC13-Me CO-Hex (85:15:20) As above 82 CHC13-Xyf-Me2C0 ( 6 :3 :1 ) HPTLC on silica gel glass122 backed plates and aluminiumbacked sheets Two-dimensional 1st: CH3CN-Me2CO-C6H6 On Silufol sheets 123 ( 9 : 1:1); 2nd: Et20-MeOH-H20 (96:3 : 1) 1st: Tol-EtOAc-90% FA B1, B2, G1 and G2 in figs on 124 (5:4:1); activated silica gel 60 2nd: CHC13-Me CO (9:l) 1st: CHC13-Me CO ?88:12) Aflatoxins in corn, 125 2nd: 95% jenaturated on SILG-HR-25 plates EtOH 1st: CHCl -Me CO-iPrOH M1 on silica gel 60 126 (80:?5:15); 2nd: Tol-EtOAc-90% FA
151
TABLE 7.2 Continued SystemX
NotesX
Ref.
1st: EtOAc (H20 sat.); 2nd: CHC13-TCE-AmOH-FA (80:15:4:1); 3rd (same dimension as 2nd): C6H7Me2CO-To1 (60:20:10:10 1st: CHC13-Me2C0 (9:l): 2nd: Tol-EtOAc-FA (5:4:1) 1st: (H20 sat.) CHC13-Me CO (88:12): 2nd: Tol-E$OAc-FA (48:40:12) 1st: CHC13-Me2C0 (88:12); 2nd: Tol-EtOAc-FA (48:40:12) 1st: Et20-MeOH-H 0 (94:4.5: 1.5f: 2nd: CHC13-Me2C0
B1 in compound animal feedstuffs
78
As above (TLC; HPTLC)
78
As above
78
As above
78
As above
78
Abbreviations: FA, formic acid; NH3, ammonia solution; iPrOH, 2-propanol; Toll toluene; EtOAc, ethyl acetate: TCE, trichloroethylene; AmOH, amyl alcohol: Xyl, xylene; others as in Table 7.1. In an attempt to overcome the lack of reproducibility of TLC resolution, a number of silica gel preparations and solvent systems were investigated by Heathcote and Hibbert (53). They found that the neutral SilicAR TLC-7G (Mallinckrodt) gave an excellent resolution of aflatoxins in the solvent systems chloroform-methanol (97:3), toluene-ethyl acetate (8:l) and benzene-ethanol-water (46:35:19). Problems concerning the solvent systems, adsorbents and environmental effects (especially relative humidity) were discussed by Heathcote and Hibbert (46) and summarized more recently by Heathcote (47). Kozloski (81) described procedures for improving aflatoxin spot size and fluorescence intensity. By using strong eluting solvents, diffuse spots could be reduced in size and poorly resolved chromatograms returned to their original state for re-development. The separation of four aflatoxins on six commercial silica gel plates in twelve solvent systems for aflatoxins frequently mentioned in the literature were compared (82). Two of the solvent systems resolved the four aflatoxins on all the tested plates (see Table 7.3). The results showed that the use of ammonia solution as a solvent modifier at a concentration of 0.5% played an important role in achieving good resolution of the four aflatoxins on silica gel TLC plates, that HPTLC plates
152
gave more compact spots than separation can be achieved when selected. TABLE 7.3 Comparative separation of plates Data from ref. 82
the other plates and that the optimum solvent system is
aflatoxins on
commercial silica gel
RF x 100
Plates
Solvent F~
Silica gel 60 K5F HPTLC Sil G 25 HR Adsorbosil-1 Silica gel IBF
33
87 48 87 44 63
29
84 44 82 37 57
solvent G~
23
20
76 37 72
71 33xxxx 65
31 52
46
25
48 59
48
54 80 68
43
36
30
54 41 46
46
43xxx
36
71 61
63 52
30 36 54
41
45
CHC13-Me2CO-ammonia solution (90:10:0.25).
xx CHClj-Me2CO-Hex (85:15:20). xxx Diffuse spots. xxxx Compact spots.
Solvents used for application of samples to adsorbent layers are also of importance. Coker et al. (122) compared benzene-acetonitrile (98:2), chloroform and dichloromethane as solvents for applying the aflatoxins to HPTLC layers at flow rates between 20 and 500 nL/sec. Benzene-acetonitrile (98:2) and dichloromethane gave the smallest spot diameters. The larger spot diameters obtained using commercial chloroform were attributed to the presence of 1% ethanol in the solvent. Benzene-acetonitrile (98:2) was selected because it produced compact spots whose diameters were the least affected by variations in the solvent delivery rate: it readily dissoves aflatoxin, either as a dry film or as crystals. In addition, collaborative studies have confirmed its suitability for both the preparation and storage of standard solutions of aflatoxin (127). 7 . 2 . 4 Detection The detection of aflatoxins is mostly based on their fluorescence under W light. Aflatoxins Blr B2, M and MZa give blue fluorescence; fluorescing turquoise are G1, G , GZal GM and GMZa (47) Techniques of quantitation are descrded in Chapter 2. When aflatoxins were quantified by UV fluorescence in the reflectance mode either a filter densitometer (360 nm) or a monochromatic densitometer (366 nm) were compared and the
.
153
latter was shown to be more suitable (122). With the 365 nn excitation-emission fluorescence, excitation at and emission at 430 nm were used by Lee et al. (128). Ripphahn and Halpaar (129) quantitated aflatoxins usingexcitation at 366 nm andemission at 460 nm with satisfactory results. In addition to detection of fluorescence under U V light, buraekova et al. (130) used p-anisaldehyde as a spray reagent. In the visible range, aflatoxins B1 and B2 gave green and blue colour, respectively; but at a wavelenght of 366 nm the colours of the spots were as follows: orange for Bl, green for B2, pale pink to orange for G1, green-blue for G2, and pink to orange for M . According to Jesenska et a l . (131) the identification of aBlatoxin B1 on chromatograms (Silufol sheets developed with CHC13-Me2C0 /9:1/) was possible under UV light at 366 nm by comparing with standards (a violet fluorescence and, after spraying the spots with 20% solution of H2S0 , a yellow fluorescence). Aflatoxins B and G1 can also be confirmed by the formation of their water adAucts B2a and Gja, respectively, with trifluoroacetic acid (132). This reagent was also used by van Egmon et al. (133) in a confirmation test for aflatoxin MI (see Section 2.5.2). buraekova et al. (134) described a bioautographic method using larvae of Artemia salina (brine shrimp) for testing aflatoxin B1 and other mycotoxins after TLC on Silufol sheets. 7.2.5 Selected applications 7.2.5.1 Biosynthetic studies. Biosynthetically, the aflatoxins are acetate-derived decaketides that are formed via polyhydroxyanthraquinone intermediates. The proposed aflatoxin biosynthetic pathway now consists of the following steps: acetate --> decaketide --> norsolorinic acid --> averantin --> averufanin --> averufin --> versiconal hemiacetal acetate --> A --> sterigmatocystin --> versicolorin 0-methylsterigmatocystin -- aflatoxin B (135). TLC has been extensively applied in studies on biosyn6hesis of af latoxins by Aspergillus flavus and A. parasiticus. Earlier applications may be found in the book by Heathcote and Hibbert (46). Some most important further applications are presented here. Aflatoxin-producing strains of A. flavus were detected by Hara et al. (85) by a fluorescence method utilizing the ultraviolet induced fluorescence of aflatoxins in agar medium. The presence of aflatoxins was confirmed by TLC of chloroform extracts of the fluorescing agar. The developing solvent was chloroform-acetone-water (88:12:1.5) and the toxins were determined quantitatively with a densitometer. Singh and Hsieh (136) studied aflatoxin bios nthesis by A. parasiticus using Y4C-labeled blocked mutants of intermediates. Cells and spent media from incorporation studies were extracted with acetone and chloroform, respectively. The extracts were pooled, evaporated to dryness under vacuum at room temperature and the residues were taken up in 0.2 mL of chloroform or acetone. The products in crude extracts were
154 separated from residual precursors by two-dimensional TLC on precoated silica gel 60 plates developed with hexane-acetonediethyl ether (70:30:20) and chloroform-acetone-2-propanol (85:15:2.5), respectively. For the determination of the specific activity the metabolites were repeatedly purified using Adsorbosil-1 plates developed with three systems: hexane-acetone-diethyl ether, chloroform-acetone-2-propanol and ethyl acetate-2-propanol-water (10:2:1). The RF values of aflatoxins B1 and B2 and of intermediates are presented in Table 7.4. TABLE 7.4 RF values of aflatoxins and their intermediates Data from ref. 136. Compounds
R~ values in systems'
HAE
CAI
EIW
0.12
0.66 0.59 0.76 0.32
0.86 0.82
_ _ _ _ _ _ ~
Aflatoxin B1 Aflatoxin B2 Averuf in Versiconal acetate Versicolorin A Sterigmatocystin
0.08 0.63
0.22 0.54
0.43
0.74 0.93
0.97 0.93 0.95 0.98
Abbreviations: HAE , hexane-acetone-diethyl ether (70:30 :20 ) ; CAI, chloroform-acetone-2-propanol (85:15:2.5); EIW, ethyl acetate-2-propanol-water. In a study on kinetics of fungal growth and aflatoxin production, Yanagita et al. (137) detected and quantitated aflatoxins by means of TLC. The chloroform extracts were evaporated and the residues were dissolved in 0.5 mL of benzene-acetonitrile (98:2). Mallinckrodt Silic AR TLC-7G plates were used and were developed with chloroform-acetone (9:l). Aflatoxins in samples were identified by the agreement of RF values of spots with those of parallel spots of reference aflatoxins on the same plate. When necessary, suspected aflatoxin spots were rechromatographed with four solvent systems, i.e., hexane-ethyl acetate (1:3), chloroform-ethyl acetate (3:1), benzene-acetone (3:1), and benzene-ethyl acetate-ethanol (30:19:1). Bennett (138) had prepared spore colour and auxotrophic mutants from an aflatoxinogenic strain of Aspergillus parasiticus. These mutants and heterozygous diploids formed by pairwise combinations of auxotrophs were assayed for aflatoxin and its anthraquinone precursors production. Samples of extracts were subjected to TLC on Adsorbosil-1 silica gel plates in
155
chloroform-acetone (95:5) and aflatoxins were quantified densitometrically. Aflatoxin formation in strains of A. flavus isolated from formic acid-treated hay was investigated by Clevstrom et al. (139). Acidified samples from liquid cultures were extracted with chloroform and were concentrated prior to TLC. Separation was performed on Kieselgel plates in chloroform-acetone (9:1), and quantifications were made by TLC fluorimetry (excitation at 360 nm and emission at 455 nm). For confirmation, the TLC plates were sprayed with concentrated H2S04 and derivatization was performed. Wicklow and Shotwell (140) found relatively high aflatoxin levels in sclerotia and conidia of a variety of strains of A. flavus and A. parasiticus. Substantial aflatoxin levels in conidia could place at risk those agricultural workers exposed to dust containing large numbers of conidia, e.g., the airborne dust around combine harvesters of cereals. Chloroform extracts from spores or conidia were cleaned-up on a silica gel minicolumn and aflatoxins were measured in solutions of residues by TLC using a densitometer. Biosynthesis of aflatoxin was found to be suppressed by nitrate (141). Aflatoxin was extracted from the whole broth by chloroform and assayed by TLC on silica gel plates developed with chloroform-acetone (9:l). Aflatoxin spots were visualized under UV light, scraped off the plate, eluted from the silica gel with 5 mL of methanol, and measured quantitatively on a spectrophotometer at 363 nm. In contrast to the previous observation, production of aflatoxin by A. flavus can be stimulated by sorbic acid, which is employed as an effective fungal inhibitor to protect food. Gareis et al. (86) extracted mycelium and broth with chloroform, the combined extracts were evaporated to dryness, the residues were dissolved in benzene-acetonitrile (98:2) and separated by TLC. Plates were developed twice with chloroform-acetone (9:l) and aflatoxin B was detected by comparison with standard toxin. The identity o& aflatoxin B1 was confirmed by derivatization to aflatoxin with trifluoroacetic acid. Af latoxin B1 was quantitated'i? fluoridensitometry. In a study of aflatoxin biosynthesis, Sharma et al. (142) used chloroform extracts of Aspergillus flavus and A. parasiticus for quantitation of individual aflatoxins as follows. A portion of the concentrated extract was developed on precoated silica gel plates (TLP-102, Anasil; 250 mm; 5 x 10 cm; Analabs). The plates were given a clean-up run in distilled peroxide-free diethyl ether. The RF of aflatoxin in diethyl ether was zero, whereas most of the interfering pigments moved with the solvent. After the clean-up run, the plates were removed, dried, and developed in chloroform-acetone (9:l). Aflatoxins B1, B2, G1, and G2 were identified with reference standards of authentic aflatoxins on viewing under UV light. The
I56
developed plates were scanned in a dual-wavelength TLC scanner equipped with a fluorescence detector. TLC has also been used by Valcarcel et al. (143) in studies on effects of selected inhibitors (benzoic acid and some of its derivatives, chlorox, cinnamon, dimethyl sulfoxide and sodium acetate) on growth, pigmentation and aflatoxin production by A. parasiticus. A TLC method for aflatoxin analysis has been applied in investigations of stimulatory effects of rubratoxin B (144) and sodium biselenite (145) on aflatoxin production. Buchanan et al. (146) investigated the effect of miconazole, an imidazole derivative effective against a variety of yeasts and A. parasiticus. At fungi, on aflatoxin biosynthesis by sub-inhibitory concentrations, miconazole stimulated aflatoxin biosynthesis. Chloroform extracts from media and mycelia were used for TLC analysis. Separation of extracts from peptone-based media was modified such that the TLC plates were predeveloped with anhydrous diethyl ether in order to eliminate interfering fluorescent pyrazines. The aflatoxins (Blr B , G1, G 2 ) were then separated using chloroform-acetone-water ( 9 5 :7 :3 ) and quantitated with a fluorodensitometer (excitation at 365 nm and emission at 436 nm). Similar procedures were used in a study on the role of cellular energy status as a regulator of the induction of aflatoxin production (147). The concentrated combined chloroform extract of A. flavus mycelium and culture medium was spotted and separated on silica gel 7G (Baker) with 90% toluene-EtOAc-formic acid (6:3:1) simultaneously with known quantities of aflatoxins B1 and G1. The developed plates were scanned fluorometrically for quantitation of aflatoxins (148). Quantitative TLC of aflatoxins has been used in screening aflatoxin production under various conditions (149, 150). Leitao et al. (151) used TLC for the identification of aflatoxins in extracts from cultures of Aspergillus strains isolated from foodstuffs. The aflatoxins were then quantitated by HPLC. TLC has been applied in a series of studies on biosynthetic routes leading to aflatoxins. Such studies have been undertaken by Dutton et al. (152), Cleveland et al. (153), Bhatnagar et al. (154), Townsend et al. (155), Henderberg et al. (156) and Bhatnagar et al. (157). The most important data concerning aflatoxin and its anthraquinone intermediates are presented in Section 7.3. 7.2.5.2 TLC of aflatoxins in foods and feeds. Only data published after 1970 are included in this section. For older data, the book by Heathcote and Hibbert (46) may be consulted. Chang et al. (158) compared two methods for extraction of aflatoxin from peanut meal and peanut butter. Both TLC and reversed-phase HPLC were used to quantitate the extracted aflatoxins. The HPLC and TLC methods correlated well, particularly for aflatoxins B1 and B2. Quantitative determination of aflatoxins in peanut products using sequential TLC was reported by Klemm (121) The method involves double
.
157
development with diethyl ether followed by chloroform-acetone-water (88:12:1.5) and triple development with diethyl ether followed by chloroform-acetone-benzene (9O:lO:lO) and chloroform-acetone-n-hexane (71:12.5:16.5). The aflatoxins could be detected spectrometrically (325 nm) at levels equal or > 0.05 ng per spot. Adoption as official first action of the solvent-efficient TLC method with densitometric determination (159) was recommended for determination of aflatoxins B1, B2, G I I and G in corn, peanut butter and raw peanuts at concentrations equaf and > 26 ng total aflatoxins/g for peanut butter and equal or > 13 ng total aflatoxins/g for corn and raw peanuts; however, variation for aflatoxin G1 was generally greater compared to aflatoxin B1. With visual comparison, concentrations should be equal or > 26 ng total aflatoxins/g for raw peanuts and > 26 ng/g for corn and peanut butter. Most recently, an inexpensive and rapid screening method for aflatoxins in peanuts and peanut products was published (160). The AOAC Romer method has been used and foung highly reliable. However, the clean-up step utilized anhydrous FeC13 and basic CuC03. The extraction (with a mixture of 270 mL of methanol plus 30 mL of 40 g/L aqueous KC1) and clean-up (150 mL 100 g/L aqueous CuS04) steps were combined with the AOAC minicolumn to provide a screening technique. The two methods gave results in complete agreement, also when the extracts were submitted to quantitative TLC. Hsieh et al. (161) employed a sequence of solvent systems for the TLC of aflatoxin Bland its metabolites. The silica gel plate was first developed in diethyl ether, thereby mobilizing aflatoxicol and completely separating it from other metabolites. After quantitation for aflatoxicol, the same TLC plate was developed in chloroform-acetone-2-propanol (85:15:15). Aflatoxin Q1 and aflatoxicol H1 were completely separated. Final separation of aflatoxins M1 and B2a was effected by a third development in benzene-ethanol (40:4) or chloroform-methanol (9:1). Kostyukovskii and Melamed (162) determined aflatoxins B1, B2! G1 and G2 on Silufol sheets and silica gel L 5/40 plates using benzene-diethyl ether-hexane (1:1:1)1 chloroform-benzeneacetone (9:l:l) or chloroform-benzene-ethyl acetate-acetone (10:4:4:3) as the solvent systems and detecting the separated spots under UV light at 365 nm. The detection limits were 0.2-0.5 ng per spot. Maize in flood-affected areas in the Bhagalpur district of India in 1985 demonstrated heavy infestation of Aspergillus flavus and aflatoxins (163). Out of 60 samples positive in the bright greenish yellow fluorescence test,only 42 were found to be contaminated with deleterious levels of aflatoxins. Qualitative analysis of aflatoxins was done on TLC plates by using toluene-isoamyl alcohol-methanol (90:32:2). The presence of aflatoxins in the contaminated samples was chemically confirmed by using trifluoroacetic acid or by spraying with 25% H2S04.
158
Greater than 100% recoveries using instrumental HPTLC were observed by Zemnie (164) for aflatoxin analyses in spiked corn samples. Spots overlying aflatoxins B1 and B2 were identified by GLC as C16-C18 free fatty acids. These fatty acids enhanced the fluorescence of aflatoxin B from 13.7 to 35.7% greater than controls, resulting in > 100% recoveries. The inclusion of acetic acid in the mobile phase resulted in an increased motility of the free fatty acids, which eliminated the positive interference on aflatoxin fluorescence. Miller et al. (57) described a simple method for the TLC of aflatoxins in vegetable oils. Their extraction and clean-up procedures were already described in Section 7.2.2. Quantitation was performed by HPLC and TLC. In TLC, two-dimensional development was used with chloroform-acetone (9:l) and diethyl ether-methanol-water (96:3:1). Amounts of aflatoxin B1 were determined by viewing under 360 nm UV light and comparing with standard aflatoxin spots. Two-dimensional TLC on Silufol sheets with acetonitrileacetone-benzene (9:l:l) and diethyl ether-methanol-water (96:3:1) as the solvent systems, visualization with nitric acid-water (1:2) and quantitation by fluorescence under long-wave UV light was reported by Eller et al. (123). Two-dimensional TLC of four aflatoxins in feed extracts was conducted by Jain and Hatch (165) on pre-coated silica gel plates. Excellent separation of aflatoxins from impurities was achieved and all four aflatoxins were well resolved using chloroform-acetone-water for the first development and toluene-ethyl acetate-formic acid (30:15:5, or 24:20:6 for samples containing citrus pulp) in the second direction and UV detection. Van Egmond et al. (166) compared six methods of analysis for the determination of aflatoxin B in feeding stuffs containing citrus pulp. These methods inclde the official 1976 European Community (EC) procedure, 4 procedures proposed in the EC to replace this method, and a new procedure. In all procedures, chloroform is used for initial extraction. Various clean-up systems are then applied. The ultimate separation and detection is by HPLC in 3 procedures and two-dimensional TLC in 2 procedures. One method allows either HPLC or TLC. All experiments were carried out with samples artificially contaminated with aflatoxins B1, B2, G1 and G at approximate levels of 13, 5, 10 and 4 pg/kg, respectivefy. Three methods were suitable: a procedure in which gel permeation clean-up and two-dimensional TLC are used: a procedure in which TLC clean-up and reversed-phase HPLC with post-column derivatization are used: a procedure in which cartridge clean-up and either HPLC or TLC are used. The latter method is preferred because its efficient clean-up yields a very clean extract, allowing the application of various systems of HPLC or TLC. A method was developed by Shannon et al. ( 105) for the determination of aflatoxin B1 in commercially prepared mixed feeds. Eluates from CC were evaporated, dissolved in
I59
benzene-acetonitrile (98:2) and applied onto Sil G-25 HR plates. Two solvent systems were used for TLC. The routine solvent system was chloroform-acetone-water (90:10:1.5) and in the second system methylene chloride wasused instead of chloroform, which resulted in a less polar system that left interferences in lower RF, range than aflatoxin B1. Methylene chloride could not be used if aflatoxins G1 and G2 were present because this system Visual changed the order of resolution to B1, G1, B2, then G and densitometric quantitation was used. In the same la2oratory, another method was developed (111) which incorporates methylene chloride and citric acid solution extraction, clean-up on a small silica gel column and TLC for quantitation. The routine solvent system is chloroform-acetone-water (90:10:1.5). Another system merely substitutes dichloromethane for chloroform, which results in a less polar system that, consequently, leaves interferences at a lower RF range than aflatoxin Dichloromethane substitution should not be used if aflatoxinBk' and G are present because this system changes the order 0% resolufion to B1, G1, B2, then G2. The lower RF toxins may be masked by interferences. In addition to TLC, HPLC methods for screening aflatoxins in feeds have been published by Kozloski (167) and by Simonella et al. (168). Aflatoxins have been detected by TLC in corn by Chakrabarti (169), in black olives by Le Tutour et al. (170), in sliced bread by Reiss (171), in soy sauce by Li (172), and in eggs by Trucksess et al. (173). TLC was one of the steps leading to the detection of aflatoxin D1 in ammoniated corn (120). Corn cultured with Aspergillus f l a w s to produce a high level of aflatoxin was ammoniated to reduce the high level. An extract of the ammoniated corn was separated by TLC on silica gel 60 plates in methylene chloride-methanol (95:5) and detected under short-wave W light (RF = 0 . 4 ) , followed by reversed-phase HPLC. Examination of the fractions by tandem mass spectrometry led to the detection of aflatoxin D1 as a product of the ammoniation process. Bicking et al. (174) determined aflatoxins in air samples of refuse-derived fuel by TLC with laser-induced fluorescence spectrometric detection. In an collaborative study, Stubblefield et al. (175) reported the determination and TLC confirmation of aflatoxins B1 and MI in artificially contaminated beef livers. 7.2.5.3 Aflatoxins in human and animal samples. Lovelace et al. (58) published a screening method for the detection of aflatoxins and metabolites in human urine. The extraction and clean-up procedures are described in Section 7.2. The final residue was dissolved in chloroform and analysed by TLC on activated silica gel G-HR plates using chloroform-acetone-2-propanol (85:10:5) with equilibration. The recoveries and R values for aflatoxins B l r Gl, M1, BZa, G aflatoxicols and Eetrahydrodeoxyaflatoxin B1 are given in Tag?;
.
7.5.
I60
TABLE 7.5 Recovery of aflatoxin metabolites from human urine and RF values Adapted from ref. 58. Metabolites Aflatoxin B1 Aflatoxin G1 Aflatoxin M1 Aflatoxin BZa Aflatoxin G2 Af latoxicol f (natural isomer) Aflatoxicol I1 (unnatural isomer) Tetrahydrodeoxyaflatoxin B,
Recovery
(%)
RW
70 76 75
0.73
16
0.39 0.29 0.69 0.62 0.79
55 35
48
60
0.59 0.44
In contrast to most other mycotoxins, aflatoxin B1 requires biotransformation before exerting its action. This bioactivation is performed by liver microsomes and the resulting derivative, aflatoxin B -2,3-epoxidet reacts with the nucleophilic sites of macromolecules. It may also be rapidly hydrolysed , enzymically or nonenzymically, to form 2,3-dihydro-2,3-dihydroxyaflatoxin Bl. The formation of the latter metabolite in vitro by rat liver microsomes was confirmed by Neal and Colley (176) using HPLC and TLC. In TLC, silica gel TLC plates were used without activation and developed in either chloroform-acetone (9:l) or chloroform-methanol (9:l). Spots were visualized by their fluorescence under long wavelenght W light. 7.2.5.4 Aflatoxin M1. Several analytical papers concerning aflatoxins M have been published in the 1980s. Increasing interest in ahlatoxin M1 has been caused by its heat stability in milk, presence in consumer milk and cheeses, and possible liver cancer risk ( 3 4 ) . TLC, HPLC, enzyme-linked immunosorbent assays (ELISA) and immunoaffinity chromatography are presently used for determination of aflatoxin M1. Several TLC methods are described here. Aflatoxin M1 has been determined in milk and various dairy products by means of TLC (177-192). Serralheiro and Quinta (178) extracted the toxin from fluid or powdered milk with chloroform, evaporated the extract, and the residue was partitioned between carbon tetrachloride and an aqueous saline-methanol solution. The toxin was once again extracted from the methanol solution with chloroform and analyzed by TLC on Sil G-25HR plates. Three solvent systems were compared: diethyl ether-methanol-water (95:4:1), toluene-ethyl acetate-formic acid (50:45:5), and toluene-ethyl acetate-chloroform-formic acid (40:50:10:5). The chromatograms were sprayed with H2S04;water (1:3). This spray converts the blue fluorescent aflatoxin M1 to a compound with a yellowish fluorescence.
161
Bijl et al. (180) described a two-dimensional TLC for cheese with a detection limit in the range of 10 ng/kg. An improved, rapid method for the routine determination of aflatoxin M1 in milk, based on TLC followed by fluorimetric analysis, was published by Kubicek et al. (185). Extraction and clean-up basically consisted of two successive sorbent chromatography steps on Bond-Elut C18 and Bond-Elut SI columns. TLC analysis was Carrie out on Kieselgel 60 aluminium foils, from which a margin of 2 cm, parallel to the baseline, had been cut off, thus yielding a 20 x 18 cm size. The whole extract (60 pL, corresponding to 20 g milk), as well as three different aflatoxin M1 standards of 0.08, 0.20 and 0.60 ng per spot (twice each) were applied pointwise 10 cm above the baseline of the plate by means of an applicator. The plate was developed with diethyl ether twice, and subsequently with toluene-formic (6:1:3). After drying the plate, the acid-ethyl acetate positions of the standards were observed under W light, marked on the vertical margin of the plate with a pencil and the plate then cut parallel to the base line 4-6 cm above the markings. The resulting smaller plate was now turned by 180' and developed chloroform-acetone-2-propanol (87:10:3). After first with development, standards were again observed under W light, and the plate cut off 5 mm below them parallel to the baseline. This procedure of development and cutting was repeated further twice. Thereafter the plate was dried and aflatoxin M1 quantified densitometrically by measuring its fluorescence in a TLC scanner (excitation at 365 nm, emission at 420 nm). The concentration of aflatoxin M1 was calculated from a calibration curve which was prepared by applying different amounts of aflatoxin M1 standard (0.08, 0.2, and 0.6 ng per spot: each twice). The method has a limit of determination of 5 ng/kg with a recovery of 90%. TLC data for aflatoxin M2 have been reported (186). Koch and Kroes (187) described the quantitative determination of harmful aflatoxins in selected cheese samples of food for the military. A newly characterized hydroxy derivative of aflatoxin B , designated aflatoxin M , has been detected by Lafont et ai. (188) in commercial mifk samples in France using a TLC method. Reversed-phase HPTLC with fluorimetric detection was used by Blanc et al. (189) in their studies of binding aflatoxin M1 to milk proteins. A rapid and reproducible method for the extraction and determination of aflatoxin M in milk and dairy products was published by Cirilli (190)- A h e r extraction and clean-up, the aflatoxin was detected by TLC or HPLC. In TLC on silica gel, hexane-acetone (9:l) was used for the first development followed chloroform-acetone-2-propanol (85:10:5) in the same by direction. Fluorescence at 365 nm was observed after spraying the plates with nitric acid and fluorodensitometry was carried out at 440 nm.
I62
Bauer et al. (191) described a standardizable test method for the determination of aflatoxin M in milk and milk products ( e . g . , reconstituted milk) by exhaction with chloroform, solvent evaporation, dissolving in benzene-acetonitrile (9:l) and TLC on Silufol sheets. Three successive solvent systems were used: diethyl ether, chloroform-acetone-hexane (35:50:15), and diethyl ether-hexane-methanol-water (80:10:8:2) with detection at 360 nm at each stage. A standard curve and formula for calculation of aflatoxin amounts were discussed. Van Egmond et al. (133) reported on a collaborative study of four methods of analysis of aflatoxin M1 in milk powders. Two TLC and two HPLC methods were tested. In TLC, two-dimensional development was used and aflatoxin M1 was quantified either visually or fluorodensitometrically. Specific confirmation procedures were applied using trifluoroacetic acid. Stubblefield (192) published a paper on the optimum conditions for formation of the aflatoxin M1-trifluoroacetic acid derivative. According to Scott (34), the AOAC procedure for preparation of aflatoxin M standards for TLC (26.008-26.011) and the TLC methods for de$ermination of aflatoxin M1 in dairy products (26.090-26.094) and for determination and confirmation of aflatoxins B1 and M1 in liver (26.101-26.109) be made official final action. 7.2.5.5 Preparative TLC. PLC has been used as a final step in purification of aflatoxins (193) because of the poor separation of aflatoxins B1 (1941, B2 (195), G1 (196) and G ( 197) by conventional column chromatography. However , the use 03 preparative TLC plates becomes cumbersome with larger amounts of material. Recently, the isolation and purification of gram quantities of the four aflatoxins was published by De Jesus et al. (193). The method involves final purification on a Waters Prep LC-500 instrument, loaded with silica cartridges, and elution with chloroform. During the purification, fractions were monitored by TLC according to Gorst-Allman and Steyn (197). 7.3 STERIGMATOCYSTIN AND RELATED COMPOUNDS Earlier TLC data for sterigmatocystin and the polyhydroxyanthraquinone intermediates of aflatoxin biosynthesis may be found in the book by Cole and Cox (186). Other anthraquinone derivatives are included in Section 7.9. 7.3.1 Extraction and clean-up A procedure for the isolation of pure sterigmatocystin was described by Steyn and Rabie (198). It involved cultivation of appropriate strains of Aspergillus versicolor, A. nidulans or A . bipolaris on sterile maize for about 21 days. The contents of the flasks were dried at 4OoC and milled. This material was extracted with an azeotropic mixture of chloroform-methanol for 16 h, the extract filtered and the solvent removed (rotary evaporator). The residue was dissolved in chloroform and shaken
I63
while heating in a boiling water bath. After cooling, proteins were precipitated after addition of filter aid. The procedure was repeated and the precipitate was washed with chloroform. The filtrates were combined and evaporated to dryness. The residue was dissolved in chloroform and used for column chromatography or TLC. Vesonder and Horn (199) extracted fermented cracked corn or feed or liquid culture mycelia in a Waring Blendor for 3 min (9:l). Partial with methanol-4% aqueous KC1 solution purification was achieved on a Florisil column. The column was eluted with hexane followed by acetone-methylene chloride (5:95). The eluates were examined by TLC. Abramson and Thorsteinson (200) isolated sterigmatocystin from barley as follows. Samples were extracted with acetonitrile-water, the extract was then washed with hexane, transferred to chloroform, and eluted from a silica gel column with cyclohexane-ethyl acetate. When HPLC had to be used, the sterigmatocystin in the eluate was acetylated in order to eliminate the background interference in fluorescence measurement. After a large-scale fermentation, sterigmatocystin and 5-methoxysterigmatocystin were extracted from the mycelium of Aspergillus versicolor with acetone. After filtration, the extract was concentrated to an essentially aqueous residue that contained some solid precipitated material. The entire mixture was extracted with methylene chloride at pH 10 to 10.5. The extract was concentrated, dried over MgS04, and boiled in methanol until crystallization began. After cooling and standing overnight at -lO°C, the products were obtained, either sterigmatocystin as silky off-white needles or 5-methoxysterigmatocystin as yellow needles, depending on the original culture (201). In a study of biotransformation of sterigmatocystin by blocked mutants of A. parasiticus, Floyd et a l . (202) extracted the filtrate from a submerged culture with chloroform. Mycelial pellets were soaked in acetone and the acetone extract was decanted, adjusted with water to approximately 30% and extracted three times with chloroform. The chloroform extracts from the culture filtrate and the mycelial pellets were combined, evaporated to near dryness, resuspended in methylene chloride, diluted and used for TLC analysis. When biosynthetic relationships between sterigmatocystin and the four main aflatoxins were investigated by Dutton et a l . (152) two procedures were used. Mycelia of A . f l a w s from cultures in a growth medium were washed with acetone until the washings became colourless. A 0.5 volume of water was added to the extract, and the aqueous acetone was extracted successively with two 0.5 volumes of hexane and then chloroform until the aqueous portion became colourless. The pooled chloroform extracts were dried over anhydrous sodium sulphate and reduced
164
to about 25 mL. The culture filtrate was extracted with 2 equal volumes of methylene chloride-acetone (9:1), and then the pooled extracts were dried over anhydrous sodium sulphate and reduced to about 5 mL. Mycelia from a replacement medium were washed with acetone, chloroform, and then acetone again. All washings were combined with the culture filtrate and then transferred to a separatory funnel. The organic layer was passed through a bed of anhydrous sodium sulphate. The remaining aqueous layer was further extracted with two portions of chloroform and these were run through the sodium sulphate bed. The total extract was evaporated, and the residue was transferred in a small amount of acetone to a vial. The acetone was then evaporated under a stream of nitrogen, and the final residue was dissolved in a 1:l chloroform-acetone mixture. Bennett e t a l . (203) soaked wet mycelial pellets in acetone for 2 h. The acetone extract was filtered, diluted to 70% with deionized water, extracted twice with chloroform and evaporated to dryness. The resultant dried mycelial extract was resuspended in 1 mL of chloroform or acetone prior to chromatographic analysis. For norsolorinic acid isolation, the mycelium was separated from the culture filtrate, soaked overnight in acetone and then the acetone/mycelium mixture was macerated in a Waring Blender for 1 min. The resultant slurry was filtered and maceration was repeated with additional acetone until the mycelium was colourless. An equal volume of water was added to the final acetone solution and this mixture was extracted twice with chloroform (138). In biotransformation experiments, Henderberg e t a l . (156) extracted entire cultures of A. f l a w s or A. p a r a s i t i c u s (mycelium plus culture filtrate from submerged cultivation) with aqueous acetone (70%). The mycelial pellets were then removed by filtration and the culture filtrate was extracted twice with chloroform. The pooled, non-aqueous fractions from entire cultures were filtered fthrough sodium sulphate, then concentrated or air dried. Dried samples were resuspended in chloroform for subsequent TLC. For PLC the resuspended samples were concentrated under nitrogen. Bhatnagar e t a l . (157) purified extracts from A. p a r a s i t i c u s cultures primarily by preparative TLC with diethyl ether-methanol-water (96:3:1). Further purification of sterigmatocystin, 0-methylsterigmatocystin and aflatoxins was carried out by Sephadex LH-20 column chromatography (column size, 3.5 by 15 cm) with methanol or acetone as the developing solvents. 7.3.2 Adsorbents and solvent systems In TLC of sterigmatocystin and other intermediates, a variety of solvent systems has been used (see Table 7.6).
I65 TABLE 7.6. TLC of sterigmatocystins Adsorbent
Silica gel G
Silica gel Silufol
Silica gel
Silica gel Adsorbosil-1 silica gel Silica gel G Silica gel G Silica gel 60
Solvent system
C6H6-Me CO (10:0.2) C6HsmMe8H (10S0.2) CC1 -Me2C0 (10:0.2) CHCf3-Me2C0 (10:0.5) CC1 -MeOH (10:2) CH2812-MeOH ( 10 :0.5 ) CHC13-MeOH (10:0.5) Tol-MeOH (99:l) C6H6-MeOH-HOAc (24:2:1) Tol-EtOAc-FA (6:3:1) C6H -EtOH (95:5) CHCf3-MeOH ( 4 :1) CHC13-MIBK((4:1) CHC13-Me CO (9:l) CHC13-HOic-Et 0 (17:1:3) nBuOH-HOAc-H26 ( 4 :1 :4 , upper layer) CHC13-MeOH (97:3) CHC13-Me2CO-Hex (7:2:1) CHC13-Me CO (9:l) C6H6-CHCf3-Me2C0 ( 9 :8 :3 ) C6H6-EtOH-HOAC (90:5:5) Tol-MeOH (99:l) Et20-MeOH-H20 (96:3:1) Et 0-MeOH-H20 (96:3:1) To?-EtOAc-HOAc ( 50 :30 :4 ) Et20-MeOH-H20 (96:3:1) Et 0-MeOH-H20 (96:3:1) Tof-EtOAc-HOAc ( 50 :30 :4 ) Tol-EtOAc-Me CO (60:25:15) CHC13-Me CO ?10:0.5) CHCl3-Me6H ( 10 :0.5 ) CClq-MeOH (10:2)
R,
X 100
Ref.
Sx
MSX
OMSX
26 48 60 58 71 85 92 43 51 49 46 81 80 56 73 83
18 12 17 24 44 52 65 34
01 25 32 23 38 19 23
201 130
197
67 35 27 24 43
204
199 201 202
34
97
44 27
97 75 86 74 93 81
44 43 29 24 69 42
153 156 154
Abbreviations: S, sterigmatocystin; MS, 5-methoxysterigmatocystin; OMS, 0-methylsterigmatocystin; FA, 90% formic acid; EtOH, ethanol; MIBK, methyl isobutyl ketone; nBuOH, n-butanol; CC14, carbon tetrachloride; others as in Table 7.4.
166
7.3.3 Detection Sterigmatocystin exhibits brick-red fluorescence under long wave W light (153), this changes to yellow when sprayed with aluminium chloride solution and heated for a few minutes in an oven at 12OoC (205). Other detection methods include panisaldehyde or iron(II1) chloride spray (130) and cerium(1V) sulphate (197). 0-Methylsterigmatocystin demonstrates a characteristic pale blue fluorescence under W illumination. Fluorescence of this compound intensifies and turns yellow-green when sprayed with aluminium chloride (153). In bioautography, buraekova et al. (134) detected sterigmatocystin with Artemia salina larvaeafter TLC on Silufol sheets. Most of the versicolorin metabolites can be detected under visible light as yellow-orange (versicolorin A), yellow and C), red (averufin), and orange-red (versicolorins B (norsolorinic acid and versiconal hemiacetal acetate). 7.3.4 Selected applications Sterigmatocystin and its derivatives have often been included in multimycotoxin TLC analyses (see Section 7.15). Hu et a l . (206) used a two-dimensional TLC determination of sterigmatocystin in cereal grains. TLC determinations of the toxin in cheese have been reported by van Egmond et al. (205) and Francis et a l . (207). Selected TLC data for the versicolorin metabolites are given in Table 7.7. TABLE 7.7 TLC of versicolorins Data from ref. 11. Metabolite
Adsorbent
Solvent systemX
RF -
Versicolorin A Versicolorin B Versicolorin C Averufin
Adsorbosil-1 Silica gel Silica gel Silica gel
0.32 0.23 0.23
Norsolorinic acid
Adsorbosil-1
C6H6-HOAC (95:s) C6H6-HOAC (95:s) C6H -HOAC (95:5) CHCf -Me2CO-HOAc (97: 1) CHCl -Me CO-Hex (85: 203 Tol-EtOAc (27:12) CHC13-Me2C0 (85:15)
Versiconal hemiacetal SilicAR TLC-7G acetate
2: 3:
0.50 0.69
0.32 0.33
Abbreviations as in Table 7.6. TLC methods have been helpful in studies of biosynthetic routes leading to aflatoxins. Dutton et al. (152) separated aflatoxins, sterigmatocystins and anthraquinones by two-dimensional TLC on aluminium-backed silica gel 60 plates. The solvent system used for the first dimension was chloroform-acetone (85:15). For separation of aflatoxins and
167
sterigmatocystins, the second system was diethyl ether-methanol-water (96:3:1). For the separation of anthraquinone pigments, the second solvent system was toluene-ethyl acetate-acetone-acetic acid (60:25:15:2). After development, the plates were inspected under long-wavelength W light, and the various fluorescent spots were marked. Sterigmatocystin and its derivatives were detected by spraying the plates with aluminium chloride reagent. After TLC in the appropriate solvent system, a visual estimate of the amount of metabolites was made, and this was used to estimate the appropriate amount of extract to spot for one-dimensional TLC on prescored silica gel G plates for aflatoxins and densitometry readings. For estimation of 0-methylsterigmatocystin plates were developed in the diethyl ether-methanol-water system: for estimation of sterigmatocystin, they were developed in ethanol-carbon tetrachloride (2:98) and then sprayed with aluminium chloride. Plates were scanned for fluorescent materials by using a recording densitometer and 360 nm. The quantities of aflatoxins, excitation at sterigmatocystin, and 0-methylsterigmatocystin were calculated based on comparisons of areas of standards run on the same plate. Similar chromatographic procedures were described by Bennett et al. (203). One-dimensional TLC in diethyl ether-methanol-water (96:3:1) was used by Bhatnagar et a l . (154) for detection and quantitation of sterigmatocystin, 0-methylsterigmatocystin and aflatoxins in their work in which 0-methylsterigmatocystin was identified as a precursor of aflatoxins Bland G1 (see Table 7.8). TABLE 7.8 TLC separation of sterigmatocystin (ST), O-methylsterigmatocystin (OMST), aflatoxins B1 and G1 and the metabolite CP461 Modified from ref. 154. Solvent systema
Et 0-MeOH-H 0 (96:3:1) To?-E~OAC-H~AC (50:30:4) Tol-EtOAc-Me CO (60:25:15) CHC13-Me CO ?10:0.5) CHC13-Me6H ( 10:0.5 ) CC14-MeOH (10:2)
RF x 100 ST
OMST
B1
B2
CP461b
97 75
44 43
37 35 41 22
28 24
44 43 29 24 69 41
86
74 93 81
29
24 69 42
63 39
30
11
55
Time minC 45 45 45 40 40 45
a Abbreviations as in Table 7.6. Identical with ST. The TLC plates were spotted with approximately 50 ng of various compounds and developed for a distance of nearly 14 cm
.
168
The first two solvent systems in Table 7.8 were also used in an enzymological study of the conversion of sterigmatocystin to aflatoxin B1 (153). In a similar study (157), TLC with diethyl ether-methanol-water (96:3:1) as the solvent system separated sterigmatocystin, 0-methylsterigmatocystin and aflatoxin B from chloroform extracts of enzyme assays conducted with celi-free extracts prepared from A s p e r g i l l u s p a r a s i t i c u s . TLC was also succesfully applied by Floyd e t a l . (202) to detect sterigmatocystin and aflatoxins in their study with blocked mutants in attempts to elucidate the later steps of aflatoxin biosynthesis. In a paper by Henderberg e t a l . (156) analytical TLC and PLC were described as applied in a study of the biosynthetic origin of aflatoxins G1 and Gz. Sterigmatocystin was confirmed to be a precursor of aflatoxins Bl, G1 and G but no evidence for the conversion of aflatoxin B1 to G1 was gound. Townsend e t a l . (155) synthesized labelled averufin, which was subsequently incorporated by mycelial suspensions of A s p e r g i l l u s p a r a s i t i c u s . The intact incorporation of averufin into versiconal acetate, versicolorin A and aflatoxin B1 was demonstrated. Analytical and preparative TLC proved to be extremely useful methods in both the synthesis and incorporation of labelled averufin. When combined medial and mycelial extracts were subjected to PLC on silica gel 60 plates, labelled averufin and aflatoxins were separated using hexane-acetone-diethyl ether (7:6:4). The aflatoxin mixture was further separated on a column (97:3). In TLC, of silica gel with chloroform-methanol chloroform-acetone (9:l) provide much cleaner separation of aflatoxin B from the other aflatoxins, with the following RF values: wilh hexane-acetone-diethyl ether (7:3 :1 ) , averufin 0.40, all aflatoxins 0.18; with chloroform-methanol (97:3), averufin 0.55, aflatoxins B1 and B2 0.52, aflatoxins G and G2 0.44 ; and with chloroform-acetone ( 9:1), aflatoxin 0.54, aflatoxin B2 and averufin 0.50, aflatoxin G1 0.40, aflatoxin G 2
A,
0.35. 7.4 TRICHOTHECENES
The trichothecenes represent a family of about 130 structurally related natural substances produced mostly by fungi, but in some instances ( e . g . , baccharinoids) isolated also from higher plants. The trichothecenes can be divided into two groups, one consisting of the alcoholic derivatives of the trichothecene nucleus and their simple esters and the other of the more complex macrocyclic di- and triesters. According to differences in the trichothecene nucleus, the trichothecenes are further divided into four types. The first type (type A, characterized by a hydrogen atom or a hydroxyl at the 8-position) is represented by T-2 toxin and diacetoxyscirpenol. Type B has a ketone group at the 8-position and is represented by nivalenol and deoxynivalenol. Type C is characterized by the
169
second epoxy function at C-7,8 or C-9,lO. Type D includes macrocyclic trichothecenes. Another classification is as follows: (i) trichothecens without an oxygen function at the C-8 position ( e . g . , scirpentriol and its derivatives) , (ii) trichothecenes with an oxygen function other than ketone at the C-8 position ( e . g . , T-2 toxin and its derivatives) , (iii) trichothecenes with a ketone at the C-8 position ( e . q . , nivalenol and its derivatives), and (iv) macrocyclic trichothecenes ( e . g . , verrucarins, roridins, baccharinoids). TLC together with GC have been widely applied in the qualitative and quantitative analysis of trichothecenes and have been described in several reviews and books ( e . q . , refs. 5, 30, 208-217). 7.4.1 Extraction and clean-up The general procedures were summarized by Takitani and Asabe (208) as follows. The trichothecenes are extracted with methanol or acetonitrile (or their mixtures with water) from food or feed samples. The lipids are removed from the extracts with n-hexane or isooctane. In many instances, the mycotoxins are re-extracted from the original extracts with chloroform. After washing with water, the solutions are applied to a column of silica gel or Florisil and the eluates containing the toxins are submitted to TLC. Romer (218) described the use of small charcoal/alumina clean-up columns in determination of trichothecenes in foods and feeds. Harrach et al. (219) extracted satratoxins from a sample of straw with methanol, the methanol was evaporated and the residue was partitioned between light petroleum and water. The aqueous layer was extracted with methylene chloride and the residue from the organic layer was placed on a silica gel column. Elution with ethyl acetate gave a residue that was subjected to PLC. According to another procedure (220) satratoxins can be extracted from straw with methylene chloride and, after concentration, are directly applied to TLC plates. 7.4.2 Adsorbents and solvent systems Most TLC studies of trichothecenes have used silica gel as the adsorbent. In a recent review (30), the following solvent systems were compiled from the literature dealing with TLC of A, chloroform-methanol (95:5); B, trichothecenes: chloroform-methanol (7:l): C, benzene-acetone (1:l); D, chloroform-acetone (3:2); E, benzene-tetrahydrofuran (85:15); F , ethyl acetate-toluene (3:l): G, ethyl acetate-n-hexane (3:l): H, chloroform-2-propanol-ethyl acetate (95:5:5): 1, chloroform-ethanol-ethyl acetate (90:5:5); J, n-butanol-acetic acid-water (4:1:4, upper layer): K, chloroform-methanol (98:2): L, benzene-methanol-acetic acid (24:2:1); M, toluene-ethyl acetate-90% formic acid (6:3:1): N, benzene-ethanol (95:5): 0, chloroform-methanol (4:l): P, chloroform-methyl isobutyl ketone Q, chloroform-acetone (9:l): R, chloroform-acetic (4:l):
I70
acid-diethyl ether (17:1:3); S , acetone-n-hexane (1:l); T I chloroform-methanol (9O:lO); U, toluene-ethyl acetate-90% formic acid (5:4:1); V, ethanol-ethyl acetate-acetone (1:4:4): W, ethanol-benzene-acetone (1:3:3); X I ethanol-chloroform-acetone (1:4:4); Y, ethyl acetate-ethanol (6:l): and 2 , ethyl acetate. Thus, the whole alphabet was needed to cover this purely empirical approach to solvent systems. 7.4.3 Detection Except for type D (group iv) macrocyclic trichothecenes, trichothecene mycotoxins show almost no absorption bands under ultraviolet or visible light. Standard and less frequent detection methods for the TLC of trichothecenes have been described (27, 221-223). Detection requires the TLC plates to be developed with suitable solvents so that the spots can be detected subsequently by colour or fluorescence (223). Different reagents work best with the different types of trichothecenes. Aluminium chloride is relatively specific for type B trichothecenes whereas type A can be detected with sulphuric acid (21) or chromotropic acid (221). Both of these reagents have a poor structural affinity for the 12,13-epoxy group in the trichothecene nucleus (224). 4-(pnitrobenzy1)pyridine is reported to interact with the trichothecene nucleus and has been used for the detection of types A , B and D. These reagents react with a wide range of extraneous compounds. Unless the samples are put through several clean-up steps, these reactions can obscure the toxins (224-226). Some macrocyclic trichothecenes fluoresce naturally, but others are detected as fluorescent spots with sulphuric acid and heating. Other detection reagents, such as nicotinamide-2-acetylpyridine, have been reported (21). A new, sensitive TLC-HPLC method for detection of et a l . (227). trichothecenes was published by Yagen Diphenylindenone sulphonyl (Dis) esters of trichothecenes, when sprayed with sodium methoxide, showed fluorescent spots on silica gel TLC under longwave W light. The detection limit for trichothecene esters in TLC was 20-25 ng per spot for T-2 toxin, HT-2 toxin, diacetoxyscirpenol, T-2 triol, T-2 tetraol and iso-HT-2 toxin. A quantitative HPLC analysis of Dis trichothecene esters using W detection at 278 nm was also developed. Other current procedures employ reagents such as panisaldehyde (130, 228, 229) and cerium(1V) sulphate (197). Ehrlich and Lillehoj (230) monitored triacetyldeoxynivalenol in CC eluates by TLC. As the latter compound is not stained with an aluminium chloride spray (2311, portions of the fractions from CC were treated with 0.5 M NaOH in 90% aqueous ethanol to regenerate deoxynivalenol. The samples were then submitted to TLC and deoxynivalenol was detected by spraying with 20% aqueous aluminium chloride and heating to produce a characteristic blue fluorescence.
171
Baxter et a l . (221) developed a procedure employing chromotropic acid (disodium 4,5-dihydroxynaphthalene-2,7-disulphonate dihydrate) as a sensitive and specific spray reagent to detect trichothecenes on TLC plates. They found that aluminium chloride was relatively specific for type B trichothecenes such as vomitoxin (deoxynivalenol) and that the type A trichothecenes, which do not react with aluminium chloride, can be rendered visible with chromotropic acid. In their procedure, following TLC development, the dried plate was sprayed with aluminium chloride reagent, heated and then viewed at 365 nm to determine the presence of vomitoxin (bright blue colour). The same plate was then sprayed with chromotropic acid reagent (1 part of a 10% aqueous solution of chromotropic acid mixed with 5 parts of concentrated sulphuric acid-water /5:3/) and heated at llO° C for 5-15 min until all the reference standards appeared as dark spots against a light mauve background. After cooling, the plate was re-examined at 365 nm. The above-mentioned and less frequently used detection reagents are summarized in Table 7.9. Some macrocyclic trichothecenes can be detected by absorption or fluorescence under UV light (80). Standard procedures using sulphuric acid, panisaldehyde, aluminium chloride, 4(p-nitrobenzyl)pyridine and nicotinamide-2-acetylpyridine have been described in Ueno's book (27) together with less frequent methods. Kroll et a l . (228) described a screening method for the detection of type A and type B trichothecenes in extracts from cereals and cereal products. Trichothecenes in the cleaned-up extracts are first transformed to their respective alcohols and then submitted to TLC separation and detection as follows. Fifty mL of a chloroform extract are evaporated to dryness. To the residue 5 mL of a mixture of methanol-25% ammonia solution (4:l) are added, the flasks are covered and incubated at 6OoC for 24 h. After evaporation of the solvent mixture in Vacuo, the residue is dissolved in 1 mL of a chloroform-methanol (2:l) mixture and volumes of 10 pL are spotted on a Kieselgel G TLC plate. The plate with samples and standards is developed twice with benzene-acetone (1:l) as the solvent system to a height of about 15 cm. After spraying the plate with 20% methanolic 1 5 min, the sulphuric acid and heating at llO°C for trichothecene alcohols of the type A are detected at 360 nm. The trichothecene alcohols of both types A and B are detected at 360 nm after spraying the plates with the nicotinamide-2-acetylpyridine reagent. The detection limit was 20 and 6 0 ng/spot for the standards and the samples, respectively. The trichothecene alcohols are visible as blue green (with sulphuric acid as the detection reagent) or pale blue spots (after the nicotinamide-2-acetypyridine spray) against a dark background. Othe mycotoxins do not disturb the detection.
172
TABLE 7.9 Detection of trichothecenes on TLC plates Adapted from ref. 11. ReagentX Trichothecene'
Visible light Colour
SA
AN
AC NBP NAP CA
cs
Type Type Type Type
A B A B
DAS Nivalenol T-2 toxin Trichothecin DON Fusarenon-X Type B DON A1 1 types Types A and B T-2 toxin DAS HT-2 toxin DON T-2 trio1 T-2 tetraol T-2 toxin Roridin A
Limit Colour (nglXX
250 Greyish black 250 Brown Pinkish violet 250 Yellowish brown-greenish Violet Pale grey Violet Beige Yellow
Blue-violet Purple Brown Purple Grey Purple Purple Grey-black Grey-black
Long-wave W light
20-200
100
200
100 100 100 100 100
Limit (nglXX
Blue
50
Blue Yellow-orange Grey Yellow Blue Blue
50
Bright blue
50
Blue 20-50 Bright blue 50 Blue-white 100 Bright blue 50 Intense black 100 Bright blue 50 Bright blue 50
Abbreviations: SA, sulphuric acid: AN, panisaldehyde; AC, aluminium chloride: NBP, 4-(pnitrobenzyl)pyridine: NAP, nicotinamide-2-acetylpyridine; CS, cerium (IV) sulphate. xx Detection limit (ng per spot). Unlike chemical detection, bioautography is based on biological effects of the substances to be detected. As TLC procedures would not alter the chemical structure and hence the biological activity of trichothecenes and other bioactive compounds, bioautography overcomes the visualization problems found when using chemical detection methods and is a suitable alternative. Bioautography with bacteria and fungi in paper chromatography and TLC was reviewed by Betina (232).
I73
Bioautography using the yeast Candida pseudotropicalis to detect trichothecin in TLC of extracts from a culture of Trichothecium roseum was published (233). The study of the toxicity of trichothecenes has shown that several genera of yeasts are sensitive (234). Koshinsky et al. (235) described a simple method for the detection and quantitative determination of T-2 toxin and its separation from HT-2 toxin on silica gel layers based on growth inhibition of Kluyveromyces fragilis and Saccharomyces cerevisiae. The detection limit for T-2 toxin was 0.2 nM per spot. The area of growth inhibition corresponded logarithmically to the toxin concentration. T-2 toxin could be quantitatively detected from 0 . 2 to 160 nM per spot. These tests were performed as follows. After development of 20 cm glass or plastic-backed pre-coated silica gel 60 TLC plates (Merck) in chloroform-methanol (9:1), the plates were removed and dried overnight at 6OoC. Before the bioautography, both sides of TLC plates were sterilized by a 60-s exposure to W light (General Electric, ge5micidal 30-W bulb, G30T8) from 60 cm with a flux of 24 ergs/s/cm , The plastic-backed plates were secured to pieces of glass with double sided tape. To contain the agar a 3-mm ridge was formed with TygonR tubing running around the plate. Both glass and plastic-backed plates were placed on a horizontal bench and evenly layered with growth media, either sterile molten (80' C) YPG (1.75% agar) or YPG agar (1% yeast extract, 1% peptone, 4% glycerol, and 1.75% agar). After solidification, the surface of the agar growth medium was swabbed with an inoculum made of a standardized culture of K. fragilis or S. cerevisiae. The plates were placed in a sterile, tightly sealed plastic container and incubated at 35OC and 24 h for K. fragilis and 3OoC and 48 h for S. cerevisiae. After incubation, the swabbed TLC plates were layered with a 10% (w/v) triphenyltetrazolium chloride agar overlay (236). The presence of the trichothecenes could be observed as areas where there was no growth. The use of tetrazolium with K. fragilis aids the visualization of the zones of growth inhibition as it stains the colonies red. Zones of growth inhibition were measured to the nearest 0.2 mm using calipers. Using bioautography the detection limit for T-2 toxin was lowered when compared with some chemical detections. (From the experience of the present author, it could be noted that two simplifications in the above procedures would be possible. First, the yeast inoculum could be added to the agar medium at about 45OC immediately before layering. Second , triphenyltetrazolium chloride can be added directly to the agar medium at much lower concentration before its inoculation, or the plates after incubation can be carefully sprayed with a solution of this indicator.) 7.4.4 Selected applications analyses of Many references to TLC (and also GC) trichothecenes can be found in Ueno's monograph (27) in the sections devoted to natural occurrence of these toxins in European and other countries.
174
Baldwin et al. (237) demonstrated the presence of seven trichothecene mycotoxins. Ma] or metabolites were 7a,8a-dihydroxycalonectrin, with 3-acetyldeoxynivalenol and 3,15-diacetyldeoxynivalenol, deoxynivalenol, calonectrin, 12,13-epoxytrichothec-9-ene as minor isotrichodermin and products. The crude ethyl acetate extract from the culture filtrates was fractionated on a silica gel column by eluting with diethyl ether-acetone (9:l). Fractions were analysed by TLC and combined as appropriate. Merck silica gel F plates were developed with diethyl ether-acetone (9:l) and &?e spots were revealed using 20% sulphuric acid or 4-(pnitrobenzyl)pyridine spray reagents. Unfortunately, RF values of the seven trichothecenes were not in the paper. TLC data on trichothecenes have been reported in studies on mycotoxins in natural products (238-244). Analytical and preparative TLC have been used in studies of the bioconversion of T-2 toxin into 3'-hydroxy-T-2 toxin and 3'-hydroxy-HT-2 toxin (245). A rapid method for the determination of trichothecenes was developed by Bata et al. (246). The trichothecenes occurring in purified extracts of food and feed samples were converted into the corresponding free alcohols by transesterification and then analysed by HPTLC or GC. As already mentioned in Section 7.4.3, Yagen et al. (227) prepared diphenylindenone sulphonyl (Dis) esters of trichothecenes which were then analysed by TLC or HPLC. The following procedures were used. Stock standard solutions of trichothecenes in acetone were prepared at concentrations of 100 and 1 pg/mL. Diphenylindenone sulphonyl chloride (Dis-C1) was synthesized and purified and an acetone stock solution at a concentration of 1 mg/mL was prepared. 4-Dimethylaminopyridine was dissolved in acetone at a concentration of 1 mg/mL. All stock solutions were sealed and stored at ~ O C . Appropriate volumes of trichothecene solutions and 200 pL of Dis-C1 solution were placed into centrifuge tubes, and the solvent was evaporated under a gentle stream of nitrogen. The residue was dissolved in 100 pL of the solution of 4-dimethylaminopyridine in acetone and the reaction was carried out in sealed tubes in an oven at 6OoC for 30 min. The tubes were allowed to cool and the solvent was evaporated. A 1-mL volume of ethyl acetate was added and the mixture was extracted with 2 mL of brine solution. Aliquots ( 5 pL) of the ethyl acetate layer were applied directly to the silica gel plate. Alternatively, the ethyl acetate layer was separated, concentrated, redissolved in 10 pL ethyl acetate and then placed on the TLC plate. The plate was developed with toluene-ethyl acetate-acetone (7:2:1). The developed plate was dried in a stream of air and sprayed with sodium methoxide or butoxide (8 g sodium in 100 mL methanol or butanol). The fluorescent
175
spots were examined under long-wavelength W light while the plate was still wet. The chromatographic data are summarized in Table 7.10. TABLE 7.10 Chromatographic data for Dis-trichothecenes using toluene-ethyl acetate-acetone (7:2:1) as solvent system Adapted from ref. 227. Compound
RF
Dis-T-2 toxin Dis-HT-2 toxin (upper spot) (lower spot) Dis- DASX Dis-Iso-HT-2 toxinXX Dis-T-2 triolXX Dis-T-2 tetraolXX
0.64 0.48
0.39 0.57
0.37 0.38
0.085
Detection limit (ng per spot) 20 25 30
20
25
25 25
DAS, diacetoxyscirpenol.
xx RF of the prominent spot was taken. TLC plates coated with silica gel GH-R and developed with either chloroform-acetone (97:3) or toluene-ethyl acetate-formic acid (5:4:1) were used to characterize neosolaniol monoacetate (247) which was detected by spraying with 50% ethanolic sulphuric acid and heating for 10 min at 12OoC. Lepom and Baath (248) described an efficient method for producing and purifying gramme quantities of T-2 toxin. The purity of the crystallized product was verified by TLC on Silufol sheets developed with chloroform-methanol (97:3) and using various spray reagents for detection. The bioconversion of T-2 toxin by a Curtobacterium sp. was investigated by Ueno et al. (249). In TLC analysis, three solvent systems were used: A, chloroform-methanol (95:5), B, chloroform-ethyl acetate-methanol (140:5:5), and C, ethyl acetate-hexane (3:2). The extract from the culture medium gave two metabolites. One of them had RF values of 0.18, 0.15, and 0.09 in solvent systems A , B, and C, respectively. These values coincided with those of HT-2 toxin. The second metabolite had lower RF values: O,ll, 0.08, and 0.05 in solvent systems A, B, and C, respectively. It was identified as T-2 triol. TLC has also been used to characterize T-2 toxin metabolites in excreta of rats (250) and chicken (251). Gordon and Gordon (252) described a rapid screening method for deoxynivalenol in agricultural commodities based on its selective adsorption on a novel detector minicolumn giving
I76
a blue fluorescence when heated 5 min at 100°C. Selective adsorption assays agreed with TLC results. Separation and purification of products of biotransformation of 3-acetyldeoxynivalenol by a Fusarium strain was achieved by means of TLC and four compounds were identified: deoxynivalenol, 15-acetyldeoxynivalenol, 3,15-diacetyldeoxynivalenol and fusarenon-X (237). TLC has also been helpful in the identification of nivalenol and its putative precursor, ( 253) and of synthetic 7-deoxynivalenol 8-oxo-12,13-epoxytrichothecenes related to 4-deoxynivalenol (254). Analytical and preparative TLC has been applied in the field of macrocyclic trichothecenes (219, 220). Bata et a l . (255) described an improved three-step (TLC, GC and HPLC) procedure for the determination of the macrocyclic satratoxins G and H and verrucarin J in cereals. RF values of macrocyclic verrucarins and roridins are given in Table 7.11. TABLE 7.11 RF x 100 values for verrucarins and roridins Data from refs. 256 and 257. Trichothecene
Verrucarin A 2'-Dehydro A B C
D E
F G H
J
Roridin A D E H
Adsorbent Alumina
Silica gel
AX
AX
BX
CX
70
28 58 47
59
68 69
47
82 83
0
28 28 0
74 70
18
54 49 59 59
70 35
40 59
52 55 9
DX
63
37
72 64
51 42 14 18 24 24
20 29 35
72
Solvent systems: A, chloroform-methanol (98:2); B, chloroform-methanol (97:3); C, benzene-tetrahydrofuran (85:15); D, diethyl ether (2x).
177
Harrach et a l . (219) subjected cleaned-up concentrates of satratoxins G and H to PLC on silica gel. The band with RF values identical with those of standards was collected, extracted with acetone and used for comparison with satratoxin standards by HPLC. A method for small routine laboratories for the detection of satratoxins in straw samples was also published by Harrach (220). TLC and other chromatographic techniques have been used in recent isolations and characterizations of macrocyclic trichothecenes, baccharinoids, from the Brasilian shrub, Baccharis megapotamica. Most probably, these trichothecenes are originally produced by fungi in the rhizosphere of the plant which transforms them into their final structures. TLC data for 13 baccharinoids are given in Table 7.12. In this work, a model 7942 Chromatotron has been used for PLC with plates prepared as circular glass disks. TABLE 7.12 RF x 100 values for baccharinoids on silica gel TLC Data from refs. 258 and 259. Baccharinoid
B1 B2 B3 B7 B9 B10 B12 B13 814 B16 B17 B20 B21 B23 B24 B25 B27
RF x 100 in solvent systems
0.50 0.48 0.52 0.54 0.40 0.40 0.54 0.19 0.19 0.10 0.66 0.43 0.50 0.17 0.16 0.31 0.44
0.18 0.19 0.30 0.29 0.43 0.36 0.76 0.28 0.21 0.17 0.74 0.38 0.57 0.20 0.18 0.64 0.44
0.51 0.47 0.58 0.55
0.48
0.43 0.65 0.43 0.42 0.29 0.54 0.46 0.48 0.31 0.28 0.58 0.51
Solvent systems: A, dichloromethane-methanol (95:5 for B1 to B7, 96:4 for B9 to B27); B, ethyl acetate; C, hexane-2-propanol (60:40 for B1 to B7, 70:30 for B9 to B27).
I78
7.5
SMALL LACTONES This group includes mycotoxins having five- or six-membered cyclic lactone rings in their structures. Patulin, penicillic acid, ascladiol, mycophenolic acid and butenolide contain a five-membered lactone; citreoviridin has a six-membered cyclic lactone in its molecule. TLC and other chromatographic techniques for patulin and other lactones have been reviewed by Scott (260), Engel and Teuber (261) and Betina (5, 30). 7.5.1 Patulin 7.5.1.1 Extraction and clean-up. This mycotoxin is a natural contaminant of apple juice. According to Scott and Kennedy (262), the samples are extracted with ethyl acetate and the extract is concentrated, diluted with a four-fold volume of benzene and transferred on to a silica gel column. After elution with benzene, patulin is eluted from the column with ethyl acetate-benzene (25:75). The eluate is evaporated to dryness under nitrogen. For TLC evaluation, the residue is dissolved in ethyl acetate. Siriwardana and Lafont (263) extracted patulin from apple juice with isopropanol-ethyl acetate, cleaned-up the extract on a silica gel column, eluted with ethyl acetate-benzene, evaporated the eluate to dryness and dissolved the residue in chloroform. Leuenberg et a l . (264) described an apparatus with which apple juice is directly applied on to a column of diatomaceous earth, eluted with toluene-ethyl acetate (3:1), the eluate is transferred on to a Kieselgel 60 F column and patulin is eluted with the same solvent system and evaporated under nitrogen. After adjusting the pH of the culture filtrates to 2.0 with diluted HC1, patulin and intermediates of its biosynthesis were extracted twice with equal volumes of ethyl acetate. The combined extracts were dried over anhydrous sodium sulphate, filtered, and the solvent removed by rotary evaporation (265). Miguel and de Andres (266) extracted patulin and penicillic acid from cereals, legumes, and sunflower seeds with acetonitrile-water (4:l). Lead acetate was added to remove pigments and lipids from the extract prior to liquid-liquid partition in chloroform. 7.5.1.2 Adsorbents and solvent systems. Silica gel has been used in most instances of patulin analysis. Selected data on adsorbents, solvent systems and R values are given in Table 7.13.In a screening for the agility of 850 strains of Micromycetes to produce patulin, Steiman et a l . (270) carried out TLC on 60 F 254 silica gel plates (Merck) using chloroform-ethyl acetate-90% formic acid (60:30:10) as development solvent. Two other sets of solvent systems were used for two-dimensional TLC: 1) benzene-ethyl acetate (9O:lO) and dichloromethane-methanol (9O:lO); 2) benzene-dioxan-ethyl acetate (95:24:4) and dichloromethane-methanol (9O:lO). RF values were not given in this paper.
179
TABLE 7.13 TLC data f o r patulin Adapted from ref. 30. Adsorbent
Solvent systemX
Silica gel G-HR Silica gel
Tol-EtOAc-FA(5:4:1) Tol-EtOAc-90% FA(6:3:1) C6H6-MeOH-HOAC (24:2:1) C6H6-MeOH-HOAC (24:2:1) Tol-EtOAc-90% FA(6:3:1) C6H -EtOH (95:5) CHCt3-MeOH (4:1 ) CHC13-MIBK (4:l) CHC13-Me CO (9:l) CHC13-HOic-Et20 (17:1:3) nBuOH-HOAc-H20 (4:1:4, upper layer) CHC13-MeOH (97:3) CHC13-Me2CO-Hex (7:2:1) CHC13-Me2C0 (9:l) EtOAc-Hex (1:l) CHC13-Me CO-PrOH (85:15:20) C H6-CHCf -Me CO (45:40:15) E ~ O H - Ho ?4:if Tol-Et&ic-90% FA (6:3:1) C6H6-MeOH-HOAc (24:2:1) C6H -HOPr-H20 (2:2:1) CHCP CHC13-MeOH (1:l) MeOH CH2C12-EtOAc (95:45) iPr20-Pen-EtOH-Pyr (84:12:4:0.8) Tol-HOAc-90% FA (50:40:10) CHC13-Me CO (9O:lO) C H C ~ ~ - M( 95 ~ ~ :5H1 Pen-EtOAc (96:4) Tol-EtOAc-85% FA (50:40:10) Tol-EtOAC-95% FA (5:4:1)
Silufol
Silica gel F254
Silica gel
Kieselgel 60 F Silica gel 60
Kieselgel 60 G Silica gel K5 ~~
RF x 100
Ref.
58 41 21 24 27 14 61 19 17 24 70
260 229
22 27 16
197
56 20 71 37 13 64 4 71 66 15 32
130
18 267
264 26 3
39 42 35 0
39 60
268 269
~
Abbreviations: FA, formic acid; EtOH, ethanol; MIBK, methyl isobutyl ketone: nBuOH, n-butanol; PrOH, propanol; HOPr, propionic acid; iPr20, diisopropyl ether; Pen, pentane; others as in previous tables.
I80
7.5.1.3 Detection. Scott and Sommers (271) detected patulin using phenylhydrazinium chloride as a spray reagent (detection limit 100-300 kg/L of juice). Later, the sensitivity was improved using 0.5% 3-methyl-2-benzothiazolinone hydrazone (MBTH)-hydrochloric acid solution, permitting the detection of 20-25 kg/mL or 10 ng of patulin per spot (262, 272). This method of detection has also been used by other workers (128, 263, 264, 270). After spraying with MESTH, patulin appears as a yellow fluorescent spot and can be detected by fluorodensitometry. Roland and Beuchat (269) sprayed the air-dried plates with 4% phenylhydrazine hydrochloride and heated at llO°C for 2 to 3 min; patulin appeared as yellow spot under visible light. 7.5.1.4 Selected applications. Leuenberg et a l . (264) reported RF values of 0.15 for patulin and 0.75 for its acetylated product on Kieselgel 60 F with methylene chloride-ethyl acetate (95:5) as the solvent system. Meyer (268) published a TLC method for the quantitation of patulin in fruit and vegetable products. After extraction and clean-up using CC, patulin was chromatographed using toluene-ethyl acetate-85% formic acid (50:40:10) and detected with a fresh 4% solution of o-dianisidine in 85% formic acid. Quantitation was based on the yellow fluorescence under long-wave W light (limit 10 ng spot). Meyer also identified patulin after acetylati:;: On Kieselgel 60 G plates and using toluene-ethyl acetate-65% formic acid (50:40:10), the RF values of patulin and of the acetylated product were 0.39 and 0.54, respectively. TLC of patulin and intermediates in its biosynthesis by Penicillium urticae has been successfully applied by Bu’Lock et al. (273) and Gaucher and co-workers (265, 274-279). In a study of the conversion of 6-methylsalicylic acid into patulin, Forrester and Gaucher (278) used preparative TLC on silica gel to isolate several metabolites of P. urticae. The various metabolites were identified by comparison of their physical properties including RF data from both paper and thin-layer chromatography. The TLC data are given Table 7.14. Analytical and preparative TLC has been used to characterize the intermediates of the patulin pathway: phylostine, isopatulin and isoepoxydon (275, 277, 278). In the last paper, TLC radiochromatography was perfo ed with samples of culture filtrates after an addition of “C-isoepoxydon. In a work dealing with the effect of patulin on aminoacyl-tRNA synthetases, the purity of patulin samples was checked by chromatography on Kieselgel sheets with toluene-ethyl acetate-formic acid (6:3:1) and visualization by its absorption of W light. When patulin was preincubated at alkaline or neutral pH, in the presence of reducing agents before the chromatography, its absorption in W light disappeared linearly with the concentration of the reducing agent, whereas the preincubation at acidic pH did not affect
181
this absorption property whatever agents (280).
the concentration of reducing
TABLE 7.14 TLC data for P. urticae metabolites Adapted from ref. 278. Metabolite
6-Methylsalicylic acid m-Creso1 Toluquinol m-Hydroxybenzyl alcohol ni-Hydroxybenzaldehyde m-Hydroxybenzoic acid Gentisyl alcohol Gentisaldehyde Gentisic acid Patulin
R~~ in systems A
B
0.75 0.71 0.29 0.17 0.64 0.43 0.04 0.52
0.71 0.29 0.25 0.46 0.21 0.08 0.45
0.20 0.32
0.63
0.32
0.20
A 10-mL sample of filtered medium was acidified to pH 2.0 and extracted with two 10-mL portions of ethyl acetate. The volume of the combined extract was reduced to 1 mL on a steam bath and a 5-kL aliquot was spotted onto a 20-cm 250-km thick, neutral silica gel plate which was then developed in solvents:(A), chloroform-acetic acid (9:l) or (B) ethyl acetate-petroleum ether (bp 60-70° C)-acetic acid (60:90:2). After development the chromatogram was sprayed with either 10% potassium permanganate in distilled water (white spots on pink) or diazotized pnitroaniline (brown to orange spots on white). 7.5.2 Penicillic acid 7.5.2.1 Extraction and clean-up. Extraction and clean-up and chromatographic methods for penicillic acid were reviewed by Engel and Teuber (281). Extraction from culture filtrates of Penicillium cyclopium and clean-up techniques prior to TLC have been described by Axberg and Gatenbeck (282) and Reimerdes et a1.(283). The former workers extracted penicillic acid from acidified cell-free preparations with ethyl acetate and subsequently extracted with aqueous NaHC03. After acidification and re-extraction with ethyl acetate the extract was used for TLC A variety of procedures for extraction from foods and feedstuffs have been published. Penicillic acid was extracted with chloroform-methanol (9:l) from corn (2841, with methylene chloride-methanol (1:l) from peas, rice, oats and crushed
.
I82
coconut (285) and with acetonitrile-4% KC1 (9:l) from cheese (286) and from raw sausages (287). Thorpe and Johnson (288) used extraction with ethyl acetate from corn, dried beans and apple juice. Penicillic acid was then extracted with 3% aqueous NaHC03 and, after acidification to pH 3 with HC1, re-extracted with ethyl acetate, dried and evaporated under nitrogen. Further purification was achieved on a silica gel column using hexane-ethyl acetate-formic acid (750:250:1) as the eluent. The eluate was concentrated and evaporated to dryness under nitrogen. For chromatography, the residue was dissolved in chloroform. 7.5.2.2 Adsorbents and solvent systems. Adsorbents and solvent systems listed in Table 7.15 have been used for the TLC detection and assay of penicillic acid. TABLE 7.15 TLC data for penicillic acid Adapted from ref. 30. Adsorbent
Solvent systemX
Silica gel
Silica gel Silufol
Silica gel Silica gel
+
oxalic acid
RF x 100
iPr20-Pen-EtOH-Pyr 23 (84:12:4:0.8) Tol-EtOAc-90% FA 41 (50:40:10) CHC13-Me CO (9O:lO) 40 CHCl -Me8H (95:5) 35 47 Tol-&OAc-90% FA (6:3 :1) 22 C6H6-MeOH-HOAC (24:2:1) 27 C6H6-MeOH-HOAC (24:2:1) 31 Tol-EtOAc-90% FA (6:3:1) C6H -EtOH (95:s) 14 66 CHCf3-MeOH (4:l) 17 CHC13-MIBK (4:l) 15 CHC13-Me CO (9:l) 26 CHC13-HOic-Et 0 ( 17 :1 :3 ) nBuOH-HOAc-H,8 ( 4 :1 :4, 76 upper layer) CHC13-MeOH (98:2) 16 CHC13-Me CO (9:l) 20 CHC13-Me8H-H 0-FA (250:24:25:1? CHC13-EtOAc-90% FA (60:40:1)
Ref. 263
229 130
197 282 283
Abbreviations: FA, formic acid; EtOH, ethanol; MIBK, methyl isobutyl ketone; nBuOH, n-butanol; PrOH, propanol; HOPr, propionic acid; iPr20, diisopropyl ether; Pen, pentane; others as in Table 7.14.
183
7.5.2.3 Detection. Penicillic acid can be rendered visible by several methods: (i) at 254 nm as a light purple spot (197); (ii) after spraying with p-anisaldehyde it develops a green colour under visible light and a blue fluorescence under long-wave W light (229); (iii) spraying with diphenylboric acid-2-ethanolamine (285) yields a blue fluorescence with excitation at 365-370 nm and emission at 440 nm, with a detection limit of 5 ng; (iv) application of ammonia fumes (283, 289) induces a blue fluorescence with an excitation at 350 nm and emission at at 440 nm; (v) with cerium(IV)sulphate, penicillic acid gives a light orange spot (197); (vi) it may be detected by W densitometry at 234 nm (283); (vii) spraying with 3-methyl-2-benzthiazolinone hydrazone hydrochloride solution and heating produces a yellow fluorescence under W light (290). 7.5.3 Mycophenolic acid 7.5.3.1 Extraction and clean-up. Mycophenolic acid can be extracted from acidified culture filtrates with chloroform, the extract being dried and evaporated to dryness. The residue is dissolved in hot acetone and filtered. Crystallization is achieved on addition of cold n-hexane (281). Mycophenolic acid has been reported in blue cheese and starter cultures of Penicillium roqueforti (291). Jones et al. (292) investigated microbial modification of mycophenolic acid which was added to various microbial cultures. The biotransformed products and the original compound were extracted from culture filtrates and mycelia with ethyl acetate at pH 2. The organic extracts were separated into acidic, phenolic, and neutral fractions by extractions with aqueous sodium hydrogen carbonate and aqueous sodium hydroxide. The alkaline extracts were acidified with aqueous hydrochloric acid and extracted with ethyl acetate.The organic extracts were washed with water, dried (MgS04), and evaporated, and the residues were processed by CC and/or PLC. For CC the fermentation products were adsorbed on silica gel and placed on a column of silica gel in light petroleum. The column was eluted with increasing proportions of chloroform in light petroleum and then with increasing proportions of ethyl acetate in chloroform. PLC was carried out on layers 1 mm thick, except where the chromatographic separations were small and thin-layers were used. The chromatograms were developed in either benzene-ethyl acetate-formic acid (66:33:1) or benzene-acetic acid (9:l). From cheese samples, mycophenolic acid is extracted with methanol-acetone at pH 6. After filtration and precipitation of casein, the supernatant is concentrated, defatted with hexane and extracted with chloroform, chloroform-ethyl acetate (1:l) and ethyl acetate. The combined extracts are dried and evaporated to dryness and the residue is dissolved in chloroform and used for TLC (263, 293). 7.5.3.2 Adsorbents and solvent systems. The best known adsorbents and solvent systems for TLC of mycophenolic acid are given in Table 7.16.
I84
TABLE 7.16 TLC data for mycophenolic acid Adapted from ref. 30. ~
~~
Adsorbent
Solvent systemX
Silica gel
AmOAc-PrOH-HOAc-H20 (4:3:2:1) C6H6-EtOAc-FA (66:33:1) C6H6-HOAc (9:l) C6H6-MeOH-HOAC (24:2:1) Tol-EtOAc-90% FA (6:3:1) C6H -EtOH (95:s) CHCP3-MeOH (4:l) CHC13-MIBK (4:l) CHC13-Me CO (9:l) CHCl3-HOic-Et 0 ( 17 :1 :3 ) nBuOH-HOAc-H26 ( 4 :1 :4 , upper layer) Et20-Hex-90% FA (60:20:0.4) iPr 0-Pen-EtOH-Pyr (42:6:2:0.4) CHCf3-Me2CO-H20 (93:7:1)
Silufol
Silica gel
R,
x 100
Ref.
65
294 292
57 67
130
30
82 90
43 72 87 39
15 22
263
Abbreviations: AmOAc, amyl acetate; PrOH, n-propanol; EtOH, ethanol; Hex, n-hexane; iPr20, diisopropyl ether; others as in Table 7.15. 7.5.3.3 Detection. Mycophenolic acid can be detected with p-anisaldehyde (130, 229), giving a grey spot under visible light and pale blue fluorescence at 366 nm, with ethanolic iron(II1) chloride, giving a grey-brown spot under visible light (130), with fumes of ammonia or diethylamine, giving immediately an unstable bright sky-blue spot under long-wave W light (263, 295, 296), or by fluorescence quenching of a fluorescence indicator (261) incorporated into the TLC plates (excitation at 254 nm). 7.5.4 Butenolide Agar cultures of Fusarium nivale can be extracted with diethyl ether, ethanol-water (80:20) or methylene chloride. From liquid media or mouldy grain, butenolide is extracted with ethyl acetate (297). TLC data of the toxin are listed in Table 7.17. The toxin is detected by spraying with p-anisaldehyde, showing a grey reaction product under visible light (130). Spraying with 2 ,4-dinitrophenyl hydrazine and heating to 100°C produces yellow spots (297). 7.5.5 Citreoviridin According to Engel and Teuber (281), mouldy rice is extracted with ethanol, the extract is evaporated to dryness and
185
the residue is dissolved in benzene and precipitated with n-hexane. The precipitate is applied to a silica gel column, which is eluted with n-hexane-acetone (2:l). The citreoviridin-containing fraction is evaporated and the toxin is crystallized from methanol. TABLE 7.17 TLC data for butenolide Adapted from ref. 30. RF x 100
Adsorbent
Solvent systemX
Silica gel
Tol-EtOAc-FA (6:3:1) CHC13-iPrOH-EtOAc (40:5:5) CHCl -MeOH (93:7) Tol-atOAc-FA (6:3 :1) CHC13-MeOH (4:l) nBuOH-HOAc-H20 (4:1:4, upper layer)
Silufol
Ref.
10
--
229 298
10 41 43
130
Abbreviations: FA, 90% formic acid, iPrOH, 2-propanol; others as in Tables 7.15 and 7.16. The separation of citreoviridin from natural extracts has been achieved only by means of TLC. Chromatographic data are given in Table 7.18. TABLE 7.18 TLC data for citreoviridin Modified from ref. 30. ~~
~~
Adsorbent
Solvent systemX
Silica gel
Me CO-Hex (1:l) Et8Ac-To1 ( 1:1 ) CHC13-MeOH (9:l) CHC13-MeOH-Me2C0 (45:3:2) C6H -MeOIi-HOAc ( 24 :2 :1 ) CHCt3-MeOH (4:1 ) CHC13-HOAc-Et 0 (17:1:3) nBuOH-HOAc-H26 ( 4 :1:4 I upper layer) EtOAc-To1 (3:l)
Silufol
Kieselgel G 1500 ~~
Rv x 100
Ref.
45 50
299
74
300
85 23 60
130
10 79
-
301
~
Abbreviations: Me2COl acetone: Hex, n-hexane: EtOAc, ethyl acetate; Tol, toluene: CHC13, chloroform: MeOH, methanol; HOAC, acetic acid; Et2O, diethyl ether: nBuOH, n-butanol.
I86
Citreoviridin appears as a yellow spot under visible light and shows yellow fluorescence under long-wave W light. W densitometric and fluorodensitometric evaluations on TLC were described (301). On TLC plates, the W maximum and the emission maximum were at 360 and 525 nm, respectively. The former evaluation was found to be the more reliable. Citreovirin was characterized by Cole et al. (302) by TLC on silica gel 60 F plates with toluene-ethyl acetate-formic acid ( 5 : 4 : 1) as $32 solvent system. The toxin was isolated from cultures of Penicillium charlesii and mouldy pecan fragments. 7.6 MACROCYCLIC LACTONES In this section, the most important zearalenone and cytochalasans are included. Zearalenone and related mycotoxins are produced by fusaria whereas cytochalasans are metabolies of a variety of fungal genera and species. Several cytochalasans have been first characterized as cytotoxic antibiotics. Another macrocyclic antibiotic, brefeldin A (known also as cyanein, decumbin or ascotoxin), has become of interest as an antifungal and cytotoxic antibiotic. It also inhibits intracellular transport of proteins. Hence some its chromatographic data are included here. Preparative TLC of the antibiotic (as cyanein) combined with bioautographic detection has been used in our laboratory (303). Analytical and preparative TLC of radioactive brefeldin A and its degradation products was carried out on Kieselgel G and Kieselgel PF254, respectively and the spots were detetcted with iodine vapour ( 3 0 4 ) . In small-scale incubations of Penicillium brefeldianum (305), chloroform extracts of the filtrate were first purified by PLC on silica gel using chloroform-methanol (9:l) as the solvent system. Brefeldin A (RF = 0.25) gave a violet coloration on spraying the plate with 5% sulphuric acid in methanol and heating at ca. 12OoC. The band of brefeldin A was eluted from the adsorbent with chloroform-methanol ( 8 : 2 ) , the solvent removed in vacuo and the solid residue recrystallized from methanol-ethyl acetate or methanol-water. In this series of experiments, derivatives of brefeldin A were characterized or isolated by PLC and TLC. Other chromatographic data on brefeldin may be found elsewhere (306). 7.6.1 Zearalenone Zearalenone often occurs with trichothecene toxins produced by fusaria colonizing maize, oats, barley, wheat and sorghum. Analytical, quantitative and preparative TLC of zearalenone have been employed by many workers. An excellent review of assay procedures was given by Shotwell et al. (307). Methods for the production, isolation, separation and purification of zearalenone, including chromatographic methods, have been reviewed more recently (308). TLC methods for zearalenone have
I87
also been reviewed (5, 30). Extraction, clean-up, detection and TLC techniques for zearalenone are summarized here. 7.6.1.1 Extraction and clean-up. A versatile method for the isolation, detection and quantitation of zearalenone in maize and barley was developed by Mirocha et al. (309). The method employs either TLC, GC or GC/MS or their combinations. Two extractions and two clean-up techniques were used. Either extraction was carried out in a Soxhlet apparatus or batch extraction was used, in both instances with ethyl acetate as the extraction solvent. The extracts were concentrated nearly to dryness and re-dissolved in chloroform. In the first clean-up, zearalenone was extracted from chloroform with 1 M NaOH and, after adjusting the pH of the aqueous phase to 9.5 with phosphoric acid, the toxin was re-extracted with chloroform. The extract was dried with sodium sulphate and concentrated nearly to dryness. The residue was dissolved in acetone and used for TLC or GC analysis. Several workers have used Eppley's extraction procedure (61) with chloroform-water (1O:l) as the extraction mixture. The clean-up procedure consists in chromatography on a sodium sulphate-silica gel-sodium sulphate column with sequential elution with n-hexane followed by benzene, both washes being discarded. Zearalenone is eluted with benzene-acetone (95:5), aflatoxins with chloroform-ethanol (97:3) and ochratoxins with benzene-glacial acetic acid (9:l). A modification of the procedure was published by Ishii et a l . (310). Gimeno (311) proposed another extraction and clean-up procedure. Ground samples are extracted with acetonitrile-4% KC1 (9:l) in 0.1 M HC1 and the extract is defatted with isooctane. The acetonitrile layer is filtered through anhydrous sodium sulphate and the sodium sulphate is washed repeatedly with chloroform, which is added to the filtrate already collected. After evaporation under vacuum, the residue is dissolved in chloroform and used for TLC analysis. et al. (312) developed a method for the Swanson determination of zearalenone and zearalenol in grains and animal feeds. The method involved extraction with 75% methanol, precipitation of pigments with lead acetate and defatting with light petroleum. The toxins were subsequently partitioned into toluene-ethyl acetate and chromatographed on HPTLC plates. Liu et al. (313) initially extracted whole corn, wheat kernels and pig feed (containing ground shell corn and soybean meal) by blending samples (50 g) for 5 min in 200 mL of methanol-water ( 6 0 : 4 0 ) . The mixture was filtered through filter paper, methanol was evaporated, and the sample was then extracted three times with equal volumes of ethyl acetate. Ethyl acetate fractions were dried over sodium sulphate and evaporated to dryness. The residue was re-dissolved in 100 pL of ethyl acetate, and 1 to 5 pL was spotted onto silica gel plates.
I88
7.6.1.2 Adsorbents and solvent systems. Silica gel is mostly used as the adsorbent. A selection of solvent systems is presented in Table 7.19. It was found by Gimeno (315) that solvent systems containing formic acid were not satisfactory when Fast Violet B salt spray detection was used. TABLE 7.19 TLC data for zearalenone Adsorbent Silica gel
Silica gel G
Silica gel Silica gel Silica gel
Solvent systemX CHC13-MeOH (97:3) EtOAc-Hex (1:l) CHC13-Me CO-iPrOH (85:15:20) C6H6-CHcf3-Me2CO ( 45 :45 :15 ) C6H6-MeOH-HOAC (24:2:1) Tol-EtOAc-90% FA (6:3:1) C H -EtOH (95:5) C&!!?3-MeOH ( 4 :1 ) CHC13-MIK (4:l) CHC13-Me CO (9:l) CHC13-HO$c-Et20 (17:1:3) BuOH-HOAC-H 0 (4:1:4, upper layer) Tol-EtOAc-CHC1 (2:l:l) Tol-EtOAc-90% $A( 6 :3 :1 ) Et 0-CHX (3:l) Tof-CHC13-Me2C0 (3:15:2) Tol-EtOAc (1:3)
RF x 100
Ref.
40 41 71 44 57 58 40 88
197
130
61
61 64 84 64 66 52
52
314 315 313 ~
Abbreviations: CHC13 , chloroform; MeOH, methanol , Me2C0, acetone; Hex, n-hexane; EtOAc, ethyl acetate; iPrOH, 2-propanol; C6H , benzene; HOAc, acetic acid; Toll toluene; FA, formic acid; EeOH, ethanol; MIK, methyl isobutyl ketone; Et2O, diethyl ether; BuOH, n-butanol; CHX, cyclohexane. 7.6.1.3 Detection. Zearalenone appears as a greenish blue fluorescent spot under short-wave (254 nm) W light, but the fluorescence is less intense under long-wave (61). Two other detection methods were used by Mirocha et a l . (309): (a) The plate is sprayed with fresh 50% sulphuric acid in methanol and then heated for 10-20 min at 12OoC. Zearalenone turns yellow and then brown. (b) The freshly developed and dried plate is sprayed with a freshly prepared solution of 1% aqueous K3Fe(CN)6-2% aqueous iron(II1) chloride (l:l), followed by 2 M HC1. Zearalenone appears as an intense blue spot. Pathre et a l . (316) sprayed TLC plates with concentrated sulphuric acid and heated for 10 min at llO° C, giving charred spots. Swanson et a l . (312) sprayed HPTLC plates with Fast
I89
Violet B salt solution. The sensitivity of the method was > 8 0 ng/g for zearalenone and 200 ng/g for zearalenol. Scott et a l . (317) sprayed plates with 0.7% aqueous Fast Violet B salt solution followed by pH 9.0 buffer solution (a mixture of 50 mL of 0.025 M sodium borate and 4.6 mL of 0.1 M HC1) until the silica gel layer appeared wet. After drying in an air current, zearalenone gave pink spots under visible light. The plates were then sprayed with 50% sulphuric acid and heated for 5 min at 120OC; zearalenone gave mauve spots under visible light. With this spray reagent, 5 ng of zearalenone on a TLC plate could be detected, compared with 2 ng of the toxin claimed et al. (318). The latter workers used by Malayandi a bis-diazotized benzidine spray reagent, which forms a brick-red derivative with zearalenone. Eppley et a l . (319) and later Martin and Keen (320) used an aluminium chloride spray reagent to enhance the fluorescence of zearalenone. Gimeno (311) sprayed plates with a 20% ethanolic solution of aluminium chloride and observed them under 366 and 254 nm W light; zearalenone showed a bright blue spot fluorescence, especially under 366 nm W light. The plate was then heated for 10 min at 105°C, cooled and sprayed again with aluminium chloride solution (not heated) and observed under 366 and 254 nm W light; zearalenone showed a bright blue-violet fluorescence spot with improved contrast with respect to the background after this second spraying. Gorst-Allman and Steyn (197) examined the developed plates under W light at wavelengths of 254 and 366 nm and the following spray reagents gave characteristic colours with (a) cerium(1V) zearalenone after heating for 10 min at llO°C: sulphate (1% solution in 3 M sulphuric acid); (b) 2,4-dinitrophenylhydrazine (1 g), concentrated sulphuric acid (7.5 mL), ethanol (75 m L ) and water (170 mL); and (c) iron(II1) chloride (3% solution in ethanol). The characteristic colours of zearalenone were purple at 254 nm, white at 366 nm, yellow-brown with reagent ( a ) , dark orange with reagent (b) and light purple with reagent (c). The most sensitive detection was at 254 nm and the detection limit was 1 kg. 7.6.1.4 Selected applications. Ishii et a l . (310) detected zearalenone in fractions from CC using Kieselgel G TLC plates and the solvent systems benzene-acetic acid (9:1), benzene-acetone (9:l) and chloroform-ethanol (95:5) with detection at 254 nm. A two-dimensional TLC method for the detection of zearalenone in animal feeds was developed by Jemmali (321). Two-dimensional TLC with benzene-acetone (60:35) and toluene-ethyl acetate-formic acid (60:30:10) proved to be insufficient for resolving a- and p-zearalenol, which appeared as a single spot (322), but the diastereomeric mixture was resolved into two components by HPLC and C. In a study of the biosynthesis of "C-zearalenone from the the radiochromatographic homogeneity of the 14C-acetate,
190
isolated 14C-zearalenone was determined by TLC on silica gel G plates using chloroform-methanol (97:3) as the solvent system. Among fluorescent bands, only that corresponding to zearalenone was radioactive (323). I studies of the bioconversion of a- and p-14C-zearalenol into P4C-zearalenone by Richardson et al. (324), the recovery from the silica gel columns was ascertained by TLC of acetone and methanol eluates. Later on, reversed-phase TLC of extracts from zearalenone-producing Fusarium isolates revealed several more polar compounds that were similar to zearalenone. GC-MS analysis of these extracts revealed 6 estrogenic compounds not previously reported as natural products, namely cis-zearalenone, cis-a-zearalenol, cis- and trans-6-zearalenol, and a- and p-zearalanol (325). Kamimura (326) used TLC and PLC in studies of the microbial conversion of zearalenone. A strain of Rhizopus sp. produced zearalenone 4-p-D-glucopyranoside in addition to aand p-zearalenol. A micro-method was used to extract 14C-zearalenone and to separate it by means of TLC on silica gel using chloroform-ethanol (97:3). After detection under W light, the zearalenone-containing area was scraped from the plate and used to count the radioactivity directly or after purification by GC (327). A rapid TLC quantitation of zearalenone in corn, sorghum and wheat was described (311). After extraction and clean-up, zearalenone was separated by TLC and its identity was confirmed with nine solvent systems and two spray reagents. The toxin was then quantitated by the limit of detection method. The minimal detectable concentration was 140-160 pg/kg with aluminium chloride solution as the spray reagent and 85-110 pg/kg with Fast Violet B salt as the spray reagent. Pathre et al. (316) used PLC with light petroleum-diethyl ether-glacial acetic acid (70:30:2) as the solvent system to provide trans-zearalenone free from cis-zearalenone and other detectable impurities. To determine the chromatographic purity, zearalenone was dissolved in toluene to give a ca. 4 mg/mL were spotted and developed in solution of which 5 pL chloroform-methanol (97:3). PLC was also used by Thouvenot and Morfin (328) to obtain zearalenone and zearalanone (internal standard) for GLC. Tanaka et a l . (329) used PLC to confirm the presence of zearalenone in zearalenone-positive samples. The sample solution was linearly spotted on silica gel 60 TLC plates, developed with chloroform-acetone (9:l) until the front reached a height of 10 cm. Zearalenone gave a blue fluorescence band at RF 0.63 at 254 nm. The zone corresponding to zearalenone was scraped, eluted with chloroform, filtered, evaporated nearly to dryness for mass spectral analysis. In most instances, TLC of zearalenone has been included in multi-mycotoxin screening methods. Examples are given in Section 7.15.
191
7.6.2 Cytochalasans The cytochalasans are secondary metabolites that have peculiar effects on mammalian cells. Their trivial names are (e.g. , derived either from their biological effects cytochalasins) or from their producing fungi ( e . g ., zygosporins). Their systematic nomenclature as compounds with a common skeleton (cytochalasans) was proposed by Binder et a l . (330) and its rules were summarized by T a m (331). For practical reasons, the cytochalasans discussed here are grouped according to their trivial names. Procedures used for the production, isolation, separation and purification of 37 known cytochalasans were reviewed elsewhere (332). 7.6.2.1 Cytochalasins. Analytical and preparative TLC of cytochalasins has been carried out on silica gel by most workers. Examples of solvent systems are given below. Detection can be effected with iodine wapour (3331, ethanolic sulphuric acid, vanillin-phosphoric acid and Dragendorff reagent (334), or panisaldehyde followed by examination under long-wave W light (130). Padhye et al. (334) achieved good TLC separations of 7 cytochalasins on silica gel G plates in solvent systems A , C and E (Table 7.20). Cytochalasins C and D with almost identical RF values were well distinguished from each other by spraying with ethanolic sulphuric acid and observing their fluorescence under w light, C giving dull orange and D weak yellow spots. TABLE 7.20 R x 100 values of cytochalasins on silica gel plates dapted from ref. 334. CytoSolvent systemX chalasin A B C D A B C D
H
63 58 46 45 63
E
79
J
44
53 38 34 32
55
38 57
62 53 44
42
64 56 63
39 36
E
36 30 26 23 1gXx 2OXx 46 50 35 34 40 48
F
G
H
I
0 0 0 0
93 92
67 63
79 78 78 76 80 72 82
o
0 0
90
89 a7 75 90
59
58 50
55 62
J
K
95 95 92 93 92 90 96
56 47 30 34 60
46 7lXx
Solvent systems: A , chloroform-methanol (95:5); B, chloroform -methanol-formic acid (95:5:5); C, chloroform-diethylamine (9O:lO): D, diisopropyl ether-ethyl acetate (9O:lO): E, F, cyclohexane-ethyl acetate-diethylamine (60:30:10); benzene-chloroform (50:50); G I n-butanol-formic acid-water (8O:lO:lO); H, benzene-methanol (70:30); I , 2-propanol; J, acetone: K , benzene-acetone (70:30) xx Tailing.
I92
The
mobilities of
cytochalasins A,
B and
C on silica gel
G plates and pre-coated Silufol sheets in 8 solvent systems were
compared by buraekova et a l . (130). In most systems, the mobilities of the cytochalasins decreased in the order A > B > C (see Table 7.21). TABLE 7.21 RF x 100 values of cytochalasins on Silufol sheets Adapted from ref. 130. cytochalasin
SorbentX RF x looxx
A
Sil G Fo1 Sil G Fo1 Sil G Fol
B C
silica gel
G plates and
DetectionXXX
A
B
C
D
E
F
G
H
1
41 34
58
25
85
17
33
23
83
20 08
31 16
99 95 98 94
Pale Blue Beige Violet Blue
0
08
04
09
46 44 24 26 14 20
28
34 19 32
42 46 34 24 27
13 13 10 10
78 74 75 74
07
14
90
85
2
Beige
Orange
Sorbents: Sil G, silica gel G; Fol, Silufol sheets. xx Solvent systems: A, benzene-methanol-acetic acid (24:2:1); B, toluene-ethyl acetate-90% formic acid (6:3:1); C, chloroform-methanol (4:l): El benzene-ethanol (95:5); D, chloroform-methyl isobutyl ketone (4:l); F, chloroform-acetone (9:l): G, chloroform-acetic acid-diethyl ether (17:1:3); H I n-butanol-acetic acid-water (4:1:4, upper layer). xxx Detection: 1, panisaldehyde; 2, W at 366 nm after panisaldehyde. TLC on silica gel G plates of several derivatives of cytochalasin B was carried out with chloroform containing 0.5-20% as the solvent system (333). Lees and Lin (335) used TLC to purify 7,20-diacetylcytochalasin B on silica gel GF plates. In the system chloroform-ethyl acetate (l:l), the RF values of cytochalasin B, dihydrocytochalasin B and diacetylcytochalasin B were 0.4, 0.5 and 0.6, respectively. Aldridge and Turner (336) separated cytochalasins C and D on silica gel G plates using chloroform-methanol-formic acid (90:5:5) as the solvent system. The plates were then s rayed with 5% ethanolic sulphuric acid and heated at ca. 110 C for a few minutes. Cytochalasin C gave an orange and cytochalasin D a yellow fluorescence under W light. Hayakawa et a l . (337) carried out TLC of cytochalasin D on silica gel GF with (or benzene-ethyl acetate (7:3) or chloroform-methanol both) and the spots were rendered visible by spraying with concentrated sulphuric acid followed by heating at ca. 180 C.
(854)
193
A variety of solvent systems have been used by Chappuis and T a m (338) in the analytical and preparative TLC of derivatives and degradation products of cytochalasin D. In PLC, Kieselgel 60 PF25 was the sorbent, and analytical TLC was carried out on Fertigpfatten 60 F2 4. In the course isolation of cytochalasins H and J, TLC and PLC were carried out on silica gel layers containing 15% of gypsum (339). Chromatography of crude diethyl ether extracts from Phomopsis p a s p a l i with chloroform-methanol (9:l) as the solvent system and spraying with concentrated sulphuric acid revealed four spots with RF values 0.56 (yellow, minor), 0.49 (red, major), 0.36 (yellow, minor) and 0.32 (red, major). The red spots corresponded to cytochalasin H (kodo-cytochalasin-1) (kodo-cytochalasin-2). In a typical and cytochalasin J experiment, 337 mg of the ether extract on PLC gave 132 mg of cytochalasin H and 28 mg of cytochalasin J. A TLC method for the determination of cytochalasin H production was reported by Mujumdar e t a l . (340). Capasso e t a l . (341) showed that the phytopathogenic fungus, Ascochyta heteromorpha, produces cytochalasins A and B both in v i t r o and in v i v o . Elegant HPLC, TLC and HPTLC techniques were used in their work. In TLC and HPTLC, 7 cytochalasins were compared. Data for HPTLC are shown in Table 7.22.
02
TABLE 7.22 HPTLC data for cytochalasins Adapted from ref. 341. cytochalasin
RF x 100 in systemX
A
B
C
0.70 0.55 0.59 0.54
0.60 0.48 0.40
0.73 0.55 0.58
0.62 0.60
0.37
0.56
0.23
0.55
0.23 0.45
0.59
0.42
0.23
Detection limit (ng) 320
160 40 40 80 120 230
Solvent systems on Kieselgel 60 F254: A, chloroform-methanol ethyl acetate-n-hexane (70:30); c , D, chloroform-2propanol (90:lO). (92:8); B,
From
Phomopsis
sp.
six
new
cytochalasans,
named
194
cytochalasins N, 0, PI Q, R and S were isolated, together with the four known compounds, epoxycytochalasins H and J and cytochalasins H and J (342). The dichloromethane extract of the culture on wheat was separated by silica gel CC and HPLC. The fractions containing cytochalasins were detected as fluorescent spots under an W light on TLC plates after spraying with 50% sulphuric acid and heating. Kieselgel 60 F 54 precoated plates were used for TLC. The identities 0 % the four known cytochalasins were confirmed by the direct TLC and IR comparisons. 7.6.2.2 Zygosporins. PLC was used in the isolation and purification of zygosporins from a culture of Zygosporium masonii (343). The culture filtrate was extracted with ethyl acetate, the washed and dried extract was evaporated to about one-third of its volume and the separated product was filtered off. The filtrate was evaporated in vacuo to give crude cytochalasin D (zygosporin A ) and a paste ( A ) . Recrystallization of the crude cytochalasin D from acetone gave the pure compound and a residue (B). From the combined residues A and B zygosporins were isolated using CC and PLC. Fraction 2 , eluted from the silica gel column with chloroform, was crystallized from ethyl acetate to give a crystalline product (C) and a paste (D). The latter was chromatographed on alumina to give an oil and a paste (E), eluted with chloroform-methanol. Fraction 3, eluted from the silica gel column with chloroform-methanol (9:1), was dissolved in light petroleum and the precipitate (F) was collected. The crystalline product C was separated into cytochalasin D (R 0.40) and zygosporin E (RF 0.48) by PLC with ethyl acetate as tie solvent. The paste E was re-chromatographed on silica gel to give an amorphous powder, which was separated into zygosporin G (R 0.35) and zygosporin F (RF 0.28) by PLC using tofuene-methanol (10:1) as the solvent system. The precipitate F was separated into cytochalasin D (RF 0.50) and zygosporin D (RF 0.40) by PLC using chloroform-methanol (1O:l) as the solvent system. In addition to the use of PLC in the isolation of zygosporins, TLC has been used to characterize degradation products and derivatives of the four zygosporins (343, 344). RF values of zygosporins on silica gel plates developed with chloroform-methanol (9:l) were reported (345): 0.50 for zygosporin A (cytochalasin D), 0.40 for zygosporin D, 0.55 for zygosporin E and 0.57 for zygosporin F and G. 7.6.2.3 Aspochalasins. Keller-Schierlein and Kupfer (346) isolated aspochalasins A, B, C and D from Aspergillus parasiticus. TLC of the aspochalasins was performed on Kieselgel 60-Fertigplatten F254 and the spots were rendered visible by spraying with 50% sulphuric acid and heating at 200° C, with iodine vapour or fluorescence under W light. RF values of 0.35, 0.27 and 0.54 were obtained for aspochalasins C, D and B,
195
respectively (in ethyl acetate, blue fluorescence). In chloroform-methyl acetate (4:1), aspochalasins A and B had 0.26, respectively. TLC data for RF . values of 0.53 and derivatives and degradation products of aspochalasins were also given by these workers. 7.6.2.4 Deoxaphomin, proxiphomin and protophomin. Binder and Tamm (347) isolated deoxaphomin by PLC from mother liquors after crystallization of phomin (cytochalasin B ) as follows. The mother liquors were combined and deoxaphomin was separated on preparative plates with chloroform-acetone (3:l). The crude product was further purified using four preparative separations (chloroform-acetone, 3:l; twice with chloroform-acetone-formic acid, 90:5:5; chloroform-acetone, 3:l). The compound was extracted with chloroform-acetone (1:l) and the extracts were checked for their purity by means of TLC (chloroform-acetone, 3:l; chloroform-acetone-formic acid, 90:5:5). Proxiphomin and protophomin were isolated by Binder and Tamm as follows (3481.The residue after isolation of phomin and deoxaphomin was chromatographed on a Kieselgel column. The fractions, eluted with methylene chloride-methanol (9:l) and containing several non-polar components, were combined and chromatographed again on Kieselgel. From the eluate in methylene chloride crude proxiphomin was obtained and the methylene chloride-methanol (98:2) fractions contained protophomin. The crude preparation of proxiphomin was purified using PLC in methylene chloride-methanol. Extraction of the main zone with chloroform-acetone (4:l) resulted in 55 mg of chromatographically pure proxiphomin. TLC was carried out with methylene chloride-methanol (98:2) and methylene chloride-ethyl acetate (9:l). The protophomin-containing fractions were chromatographed on PLC layers, yielding crude protophomin, which was submitted to further purification by PLC : twice with chloroform-acetone-formic acid (96:2:2) and once with methylene chloride-methanol (98:2). Extraction of the zones with chloroform-acetone (3:l) yielded almost pure protophomin. 7.6.2.5 Chaetoglobosins. With extracts from Diplodia macrospora cultures, Probst and Tamm (349) used TLC to check the presence of compounds with a positive reaction to phenols and indoles - spraying with 5% solutions of ammonium cerium(1V) nitrate in acetone and of hydroxylammonium chloride in 80% aqueous acetone. The extracts with positive reactions were cleaned up on a silica gel column. The fractions containing chaetoglobosins K and L were purified by PLC on silica gel plates using toluene-ethyl acetate-formic acid (5:4:1) as the solvent system. Chaetoglobosins A , B , C, D and E were analysed on silica gel using benzene-ethyl acetate (1:1) and F254 benzene-chloroform-methanol (10:10:3) as the solvent systems.
196
Metabolites were detected by W irradiation at 254 and 365 nm and by spraying with Ehrlich's reagent and coloration after heating (350). The RF value of chaetoglobosin K was 0.53-0.56 on silica gel 6 0 plates developed with toluene-ethyl acetate-formic acid (5:4:1) and it was observed by Cutler e t a l . (351) as a dark spot under short-wave W light. Probst and Tamm (352) reported decreasing RF values of five chaetoglobosins on the same sorbent and with methylene chloride-methanol (95:5), showing increasing polarity from left to right, as follows: 19-0-acetylchaetoglobosin A > chaetoglobosin C > 19-0-acetylchaetoglobosin B > 19-0-acetyl chaetoglobosin D > chaetoglobosin A. Sekita e t a l . (353) used TLC in their work on chaetoglobosins A-J. TLC was also used by Cole e t a l . (354) in the isolation and identification of two new cytochalasans from Phomopsis s o j a e . 7.7 OCHRATOXINS The ochratoxin group consists of ochratoxin A and its methyl and ethyl (ochratoxin C) esters, ochratoxin B, its methyl and ethyl esters, and 4-hydroxyochratoxin A. Ochratoxin A and its esters are the toxic members of the group. 7.7.1 Extraction and clean-up Extraction and clean-up procedures for ochratoxins were reviewed by Steyn (355). Mouldy material can be extracted with various solvents and their combinations, such as methanol-water, acetonitrile-aqueous KC1, chloroform-methanol, or mixtures of organic solvents with diluted phosphoric acid. Ochratoxin A has to be determined in various materials. According to a Steyn's review, the problem of ochratoxin A contamination has been brought closer to home by reports of its occurence in barley, corn, swine tissue, pig serum, pig kidneys, sausages, commercial roast coffee, and human serum, kidneys or milk. A significantly higher incidence of ochratoxin A has been found in blood serum of patients with urinary systemic tumours and/or endemic nephropathy living in an endemic area of Bulgaria than in people from a non-endemic area. Clean-up procedures for ochratoxins include CC, gel filtration chromatography, solvent partition or dialysis. One of the recent methods, published by Cohen and Lapointe (356), employs a new extraction solvent (ethanol-chloroform-5% aqueous acetic acid) and clean-up using a Sep-Pak silica cartridge followed by a cyan0 cartridge. Another method (357) includes the use of a C Sep-Pak cartridge. 7.7.2 hisorbents and solvent systems. Silica gel, oxalic acid-treated silica gel, and rice starch have been reported as adsorbents for TLC of ochratoxins. Solvent systems and other TLC data are given in Table 7.23.
197
TABLE 7.23 TLC data for ochratoxins Adsorbent
Silica gel Silica gel
Rice starch Silica gel G
Oxalic acidtreated silica gel Silica gel
RF X 100
Solvent systemX
AX
BX
C6H6-HOAC (3:l) Tol-EtOAc-HOAc (5:4:1) C6Hs-HOAC ( 4 ~ 1 ) Tol-TCE--OH-HOAc (80:15:4:1) Tol-HOAc (20:0.15) C Hs-MeOH-HOAC (24:2:l) Tol-EtOAc-FA (6:3:1) C6H -EtOH (95:5) CHCP3-MIBK (4:l) CHC13-Me CO (9:l) CHC 1 -HOlc-E t 0 nBuOH-HOAc-H26 (4:1:4, upper layer) CHC13-MeOH (98:2) CHC13-Me2C0 (9:l)
50 70 40 60
35
43 52
30 41
59 34xx llXXX 23xxx 56 95
46 72 12xx 75 0 53xx 02 73 33 86 79 87
CsH6-HOAC (3:l)
50
CHC13-HOAc (4:l) C6H6-HOAC (25:l)
Cx
Ref. HAx 358 359
80 130
80
32 34
197
35
25 55
358 , 360 358, 361 362
Abbreviations: A, ochratoxin A: B, ochratoxin B: C, ochratoxin C; HA, 4-hydroxy-ochratoxin A: C6H6, benzene: HOAc, acetic acid: Toll toluene: EtOAc, ethyl acetate; TCE, trichloroethylene: AmOH, amyl alcohol: MeOH, methanol: FA, 90% formic acid; EtOH, ethanol: CHCl chloroform; MIBK, methyl isobutyl ketone: Me2C0, acetone: EZiO, diethyl ether: nBuOH, n-butanol. xx Tailing. xxx Elongated spot. 7.7.3 Detection A generally used technique is to view the plate under long-wave (366 nm) W light; ochratoxin A appears as a green fluorescent spot (blu-green on acidic plates) and ochratoxin B has blue fluorescence. The fluorescence of the ochratoxins changes to purple blue on exposure to ammonia fumes or Spraying with aqueous sodium hydrogen carbonate or sodium hydroxide (358, 359). The presence of ochratoxin A on chromatograms can also be confirmed by boron trifluoride derivatization (360).
I98
7.7.4 Selected applications In a report on TLC systematic analysis of 37 fungal metabolites in eight solvent systems, data for ochratoxins A, B and C were included (130). A very efficient separation of ochratoxins A and B was achieved by impregnation of the silica gel with oxalic acid (197). The TLC plates were then developed with the neutral solvent systems: chloroform-methyl isobutyl ketone (4:1), chloroform-methanol (98:2) or chloroform-acetone (9:l). Semi-quantitative and quantitative methods for the determination of low levels of ochratoxin A have been developed ( e . g . , refs. 363-365) and have been reviewed (355, 366-374). Patterson and Roberts (375) applied two-dimensional TLC to the analysis of feedstuffs. The chromatograms were developed with toluene-ethyl acetate-90% formic acid (6:3:1) (first direction) and chloroform-acetone (9:l) (second direction) and then examined at 366 nm. Quantitation of ochratoxin A was described by Johann and Dose (376). In a study on postharvest production of ochratoxin A inbarley, Haggblom and Ghosh (377) used DC Alufolien Kieselgel 60 with benzene-acetic acid (9:l) as the solvent system. Quantitation at 365 nm was carried out by fluorodensitometry. TLC has been applied in the quantitative determination of ochratoxin in vegetable foods by Asensio e t a l . (378). TLC remains one of the chief methods for the detection, identification and quantitation of ochratoxin A. Stahr e t a l . (379) included TLC among methods of chemical analysis for ochratoxin poisoning. Problems of streaking of ochratoxin A and B spots in neutral solvent systems accompanied by increasing RF values with increasing amount applied and the effects of acidic modifiers on these values have been discussed by Nesheim and Trucksess (21). Tsubouchi e t a l . (380) tested the heat stability of ochratoxin A in contaminated coffeee beans. The method developed by Nesheim e t a l . (371) for the determination of ochratoxins A and B in barley is very sensitive and specific for ochratoxin A. The method was adopted by the Association of Official Analytical Chemists as an official, first action method (369). It was also used by PleStina e t a l . (370) in the analysis of food samples from areas in Yugoslavia where Balkan endemic nephropathy is a major problem. The fluorescence intensity can change when ochratoxin A is exposed to ammonia-methanol vapour and the magnitude of the change is influenced by the residual mobile phase. This observation was exploied by Nesheim e t a l . (371). Samples are spotted on TLC plates in benzene-acetic acid (9:l) and benzene-acetic acid-methanol (90:5:5) is used as the mobile phase. The developed plate is exposed to ammonia-methanol vapour and then was covered with another glass plate to prevent evaporation of the ammonia-methanol. If the ammonia-methanol
199
does escape and the fluorescence intensity drops, it can be restored by re-exposure to fresh ammonia-methanol. The fluorescent spots under these conditions are stable for several days, whereas they occasionally fade in a few minutes on acidic plates. The method is recommended for most commonly contaminated commodities such as corn, barley and pig tissue. The method includes a confirmatory step. Methyl esters are prepared with boron trifluoride as a catalyst. The esters are identified by comparing the RF values of standard and analyte derivatives. PLC with benzene-acetic acid (4:l) as the solvent system was used for the purification of isotopically labelled ochratoxin A (372). Conversion of ochratoxin C into ochratoxin A in rats was studied by Fuchs et a l . (373) and ochratoxin A-containing fractions from a silica gel column were purified by PLC in toluene-dioxane-acetic acid (95:35:4). Ochratoxin A has been included in multi-mycotoxin analytical methodology (374). Other multi-mycotoxin analyses, in which ochratoxins have been included, are described in Section 7.15. RUBRATOXINS Rubratoxins A and B are structurally related toxins. Their production, physical, chemical and biological properties were summarized by Davis and Richard (381). 7.8.1 Extraction and clean-up The more toxic rubratoxin B can be extracted after concentrating the culture filtrate and mycelial washing, the concentrate being acidified with HC1 and extracted with diethyl ether. The ether extract is evaporated and the residue is dissolved in acetone and analysed by TLC (382). For corn, extraction with ethanol, acetone and ethyl acetate yields the maximum amount of rubratoxin A, whereas refluxing with diethyl ether yields the maximum amount of rubratoxin B. For rice, extraction with ethyl acetate in benzene yields the maximum amount of rubratoxin A, whereas extraction with ethyl acetate-benzene and diethyl ether yields the maximum amount of rubratoxin B (381). Hayes and McCain (383) reported that acetonitrile was satisfactory for extracting rubratoxin B from corn. 7.8.2 Adsorbents and solvent systems TLC of rubratoxin can be accomplished according to Cottral ( 3 8 4 ) as follows. Spotting of the silica gel plates should be carried out under nitrogen to prevent oxidation and internal and external standards should be included on the plates. The solvent system is chloroform-methanol-glacial acetic acid-water
7.8
(80:20:1:1). 7.8.3 Detection
Rubratoxin adopts a greenish fluorescence after heating the
200
plate at 2OO0C for 10 min. The intensity of the fluorescence can be increased by subsequently spraying the plate with 21,71-dichlorofluorescein; however, the background will also have a yellow-green fluorescence (383). Whidden et al. (64) quantitated rubratoxin B according to Hayes and McCain (383) and described the following confirmatory tests. The fluorescent derivatives, which were formed from rubratoxin B on a TLC plate after heating at 2OO0C for 10 min, were exposed to ammonia vapour for 10 min. Examination under long-wave W light revealed a change in the intensity and colour of the fluorescence. Rubratoxin was then more easily observed as a light blue spot, although the detection limit remained the same. Further, the fluorescence intensity of fluorescent greatly reduced, which compounds near rubratoxin B was considerably improved the contrast and thereby the ease of detecting rubratoxin. Also, after prolonged heating of the TLC plates at 100°C for 2-10 h with ammonium hydrogen carbonate, rubratoxin became visible under W light. The reactions of ammonia and ammonium hydrogen carbonate with rubratoxin B both produced very similar fluorescent derivatives on the TLC plates. The ammonium ion apparently combined with the anhydride derivative of rubratoxin to produce an amide or imide, which reacted with chlorine fumes and a spray reagent to produce a colour reaction. The spray reagent was prepared by mixing equal volumes of a 0.2 M pyridine solution of l-phenyl-3-methyl-2-pyrazolin-5-one and 1 M aqueous potassium cyanide. Subsequently, rubratoxin first turned pink under visible light, then quickly changed to blue and subsequently brown. The detection limit was 10 pg. 7.8.4 Selected applications TLC data for rubratoxins reported by Hayes and Wilson (385) were as follows: on silica gel HF254 plates with chloroform-methanol-glacial acetic acid (80:20:2) the R values for rubratoxin A and B were 70 and 56, respectively. With six of the eight solvent systems used by buraekova et a l . (130) no migration of rubratoxin B was observed on Silufol plates. With chloroform-methanol (4:l) and n-butanol-acetic acid-water (4:1:5, upper layer) its RF values were 0.28 and 0.88, respectively. Emeh and Marth (382) used PLC on freshly activated plates prepared with silica gel HF 4+ and developed the plates with 137. ethyl acetate-acetic acid
(82:
7.9 HYDROXYANTHRAQUINONES
The most important hydroxyanthraquinone mycotoxins are emodin, luteoskyrin and rugulosin. TLC of these and related mycotoxins has been reviewed (5, 30, 386, 387).
20 I
7.9.1 Extraction Anke et al. (388) extracted the mycelia of aspergilli with acetone (50 mL/g mycelium) and the culture broth with ethyl acetate (1:l). The extracts were concentrated to 5% of their volumes and aliquots were used directly for TLC. Ethyl acetate was also used to extract culture filtrates of Trichoderma viride (389). After drying with anhydrous sodium sulphate, the solvent was evaporated under reduced pressure and the residue was dissolved in acetone prior to TLC. 7.9.2 Adsorbents and solvent systems Silica gel is usually used as the adsorbent, sometimes impregnated with oxalic acid. For PLC, 58 g of silica gel PF 45 (Machery, Nagel and Co.) were mixed with 120 mL 0.2 M oxaiic acid and poured on glass plates (20 x 40 cm); after drying the plates were activated for 3 h at 13OoC (388). The following solvent systems were used by Anke et al. (388): (a) chloroform-methanol (97:3); (b) carbon (90:10) : (C) benzene-ethyl tetrachloride-chloroform acetate-acetic acid (45:55:1); ( d ) light petroleum (b.p. 40-60° C)-ethyl formate-formic acid (90:4:1). 7.9.3 Detection The hydroxyanthraquinones give yellow, orange or red spots on TLC plates. They are also detected by spraying the plates with a saturated solution of magnesium acetate in methanol or 5% potassium hydroxide in methanol (386). Varna et al. (387) compared detection with methanolic solutions of magnesium acetate and copper acetate. The colour obtained with 0.2% copper acetate was more stable than that with magnesium acetate. The colour obtained with copper acetate increased for 2 h and then remained stable for 24 h. buraekova et al. (130) detected luteoskyrin and rugulosin with panisaldehyde reagent. Spots of two hydroxyanthraquinones from Trichoderma viride on Silufol plates became intensely orange and violet, respectively, when the plate was exposed to ammonia fumes (389). 7.9.4 Selected applications Analytical TLC was used to characterize emodin on silica-7GF plates developed with (a) toluene-ethyl acetate-formic acid (5:4:1) and (b) chloroform-acetone (83:7). Orange-red spots in visible light had R values of 0.80 in the former system and 0.45 in the latter (380). On silica G plates impregnated with 0.5 M oxalic acid and developed with benzene-hexane (l:l), rugulosin gave an RF value of 0.25 (391). An RF value of 0.40 was reported (392) for luteoskyrin chromatographed on silica gel G plates impregnated with 0.5 M oxalic acid using acetone-n-hexane-water (6:3:1.5) as the solvent system. for hydroxyanthraquinones from Penicillium TLC data islandicum are given in Table 7.24. The separation of skyrin, (rugulin), rugulosin and 2,2-dimethoxy-4a,4a-dehydrorugulosin
202
a minor metabolite from Penicillium rugulosum, obtained by CC was monitored by TLC on Silufol plates developed with chloroform-ethyl acetate (2:l). Detection was carried out at 366 nm and by bioautography using Bacillus subtilis (393). Two main anthraquinones from a colour mutant of Trichoderma viride, 1,3,6,8-tetrahydroxyanthraquinone and l-acetyl-2,4,5,7-tetrahydroxy-9,10-anthracenedioneI were purified by PLC on Silufol plates using benzene-acetone (75:25) for repeated development (394). Quantitation of emodin and its major hepatic metabolite, w-hydroxyemodin, was performed by TLC as described by Murakami et a l . (395). TABLE 7.24 TLC data for hydroxyanthraquinones from Penicillium islandicum Adapted from ref. 386. Compound
RF X 100
A Islandicin Chrysophanol Iridoskyrin Roseoskyrin Dianhydrorugulosin Catenarin Punicoskyrin Rhodoislandin A Rhodoislandin B Auroskyrin Emodin Skyrin Aurantioskyrin Dicatenarin Luteoskyrin Deoxyluteoskyrin 4a-Oxyluteoskyrin Rubroskyrin Deoxyrubroskyrin
82
65(Y)"
B
C
D
85
95
84
90
70
Solvent systems: A, benzene-hexane (1:1); B, benzene-acetone (20:l); C , benzene-acetone (4:l); D, acetone-n-hexane-water (5:5:3.5, upper layer). xx Y, yellow on spraying with magnesium acetate reagent. The remaining pigments red or purple using the same detection.
203
7.10 EPIPOLYTHIOPIPERAZINE-3,6-DIONES This class of fungal secondary metabolites includes compounds with various biological activities such as mycotoxins ( e . g . , sporidesmins) or antibiotics ( e . g . , gliotoxin). The isolation, separation, purification and chemical and biological properties have been summarized by Nagarajan (396). TLC of these metabolites has been reviewed by the present author (30). 7.10.1 Extraction and clean-up Most data on extraction and clean-up procedures included in this section are taken from the above review (396) where references to the original literature may be found. Hyalodendrins are extracted from the filtered culture broth with chloroform and the extract is evaporated to dryness. The residue is chromatographed on silica gel. Benzene-chloroform (65:35) eluates afford hyalodendrin. Later fractions give bisdethiodi(methy1thio)hyalodendrin. Gliotoxins can be extracted from filtrates with chloroform or benzene. CC on silica gel or crystallization without chromatography have been described. Most recently, screening techniques of A s p e r g i l l u s f u m i g a t u s isolates for gliotoxin production were described by Richard e t a l . (397) as follows. The liquid cultures were harvested by filtering and the filtrate was extracted three times with chloroform. The extracts were combined and placed at 4OC overnight. The chloroform layer was absorbed on a hydrophilic matrix column (Chem Tube CT-2050) and gliotoxin was eluted with chloroform. The eluate was evaporated to dryness and redissolved in 2 m L of methylene chloride. The 2 mL of extract from each isolate was placed on a silica gel Sep-Pak primed with 5 mL of methylene chloride. Each Sep-Pak was eluted with 2 mL of each of the following solvents and saved separately: hexane, ether, ethyl acetate, chloroform, and methanol. Each eluate was evaporated to dryness and redissolved in 100 pL methylene chloride for use in TLC analysis. The latter workers described the extraction and clean-up procedure for gliotoxin from rice culture as follows: (1) Extract rice in flask with 250 mL chloroform (overnight, static). ( 2 ) Filter extract into evaporating flask. (3) Extract rice a second time with 200 mL chloroform (8-12 h, static). (4) Filter and combine with extract above. (5) Evaporate combined extracts to dryness and redissolve in 10 mL chloroform. (6) Add 250 mL light petroleum (b.p. 3O-6O0C) and place at 4OC for 12-24 h. (7) Filter solvent, save and evaporate to dryness. Discard precipitate. (8) Redissolve residue in 5 mL of methylene chloride-cyclohexane ( 5 0 : 5 0 ) and inject onto gel permeation column. (9) Elute with methylene chloride-cyclohexane ( 5 0 : 5 0 ) at 5 mL/min discarding first 100 mL fraction and collecting four 20-mL fractions. (10) Combine fractions 2-4, evaporate to dryness. Redissolve for TLC or HPLC analysis.
204
Aranotin and related compounds are extracted from filtered broth with ethyl acetate and the extract dried over anhydrous sodium sulphate. Evaporation of the solvent under reduced pressure affords the crude antibiotic complex. The individual metabolites are separated by CC and the separation is monitored by TLC. Sporidesmins which are responsible for facial eczema in grazing animals in New Zealand were isolated from cultures of the fungus P i t h o m y c e s c h a r t a r u m . A mixture of the culture and water-methanol (2:3) was stirred for 24 h and filtered. The residue was extracted again with water-methanol. The aqueous methanol extracts were combined and concentrated. The concentrate was diluted with water, extracted with diisopropyl ether and the extract evaporated to dryness. The residue was washed with light petroleum. The lipid-free residue was dissolved in benzene and separated on a silica gel column using a benzene-ethyl acetate gradient. Sporidesmin B eluted first, followed by sporidesmin and sporidesmin E. The next fractions contained sporidesmin G and D. Sporidesmins H and J were isolated by PLC from the next eluates. 7.10.2 Adsorbents, solvent systems and detection Silica gel is usually used as the adsorbent. Some solvent e t a l . (398) detected systems are given below. Hodges sporidesmins by spraying with 5% aqueous silver nitrate or viewing under reflected short-wave W light. In TLC of melinacidin, bioautography with B a c i l l u s s u b t i l i s was employed (399). Sirodesmins were detected by spraying with chromic acid and heating ( 4 0 0 ) . Gliotoxin was visualized with a spray reagent of 5% silver nitrate in 90% ethanol. Other detections are given below. 7.10.3 Selected applications PLC has been used in the preparation of sporidesmin H and J (396). Hodges e t a l . (398) characterized sporidesmin A on silica gel F plates with benzene-ethyl acetate (4:l) and chloroform-me%nol ( 19: 1 ) as the solvent system, resulting in RF values of 0.38 and 0.57. respectively. The melinacidin factors were differentiated from each other (399) on silica gel G plates using the solvent systems toluene-ethyl acetate (1:l or 3:2) and methylene chloride-ethyl acetate (7:3). Gliotoxin was analysed by Richard e t a l . ( 3 9 7 ) on silica gel 60 plates including internal and external standards of gliotoxin (at least 500 ng of gliotoxin per spot). The plates were developed 10 cm in an unlined tank with methylene chloride-methanol (97:3). Repeated PLC of fractions from a silica gel column afforded sirodesmins A, C and G (400). Analytical TLC was performed on (1:2), silica gel GF with toluene-ethyl acetate chloroform-methanof5formic acid (95: 4 :1 ) and chloroform-methanol
205
(95:5) as the solvent systems. Elution of hyalodendrin from a silica gel column was monitored by TLC on silica gel GF254 and detection under W light, giving an RF value of 0.60. Hyalodendron tetrasulphide was obtained from an enriched CC fraction by PLC (401). The latter compound also gave an RF value of 0.5 on Kieselgel plates developed in benzene-acetone (9:l) (402). TLC followed by bioautography has been used for the antibiotic A30641 (403), antibiotics of the A26771 series (404), aranotin and its derivatives (405). Epicorazines A and B were purified by means of PLC (406). 7.11 TREMORGENIC MYCOTOXINS Except for the territrems, the known tremorgenic mycotoxins have in common an indole moiety and can be placed into the following groups : the paspalitrem group, the fumitremorgin-verruculogen group, the penitrem group, the janthitrems and the tryptoquivaline group. TLC has been used in monitoring the CC separation and purification of most of the tremorgens, and also in preparative and qualitative separations. 7.11.1 Adsorbents and solvent systems Silica gel has been used in most TLC studies of the tremorgens. Solvent systems are mentioned in the applications. Penitrems A-F are unstable in chloroform when exposed directly to light, presumably as a result of acid formation in the solvent. Hence, its use must be avoided in work with these toxins (407). 7.11.2 Detection Detection methods used in TLC of indole-derived tremorgens include short- and long-wave W light and the following spray reagents: 50% sulphuric acid in ethanol without and with heating, cerium(1V) sulphate in sulphuric or phosphoric acid, phosphomolybdic acid, iron(II1) chloride, aluminium chloride, m-dinitrobenzene, 2,4-dinitrophenylhydrazine and Van Urk reagent. The following results have been obtained with these detections. 7.11.2.1 Paspalitrem group. Aflatrem appeared as a dark spot under long-wave W light; spraying with m-dinitrobenzene caused the spots to turn a non-specific brown colour, but spraying with phosphomolybdic acid with applied heat turned the spots an orchid to violet colour (408, 409). Paspaline and paspalicine were detected as pale green spots with Van Urk reagent (410). Paspaline and paspalitrem A were revealed as grey-blue spots in visible light after spraying with 50% ethanolic sulphuric acid and heating for 5 min at 150 C and were fluorescent under long- and short-wave W light. Under the same conditions, paspalitrem B was visible as a green spot immediately after spraying (411).
206
Paxilline was detected after spraying TLC plates with 50% ethanolic sulphuric acid or 3 % phosphomolybdic acid and heating for 5 min at 100°C. With the latter treatment paxilline gave a dark blue spot and with the former a greenish grey spot. It was also revealed under long-wave W light as a blue-grey fluorescent spot after the former but not latter treatment (412). Cockrum et a l . (413) detected paxilline as spots showing a characteristic colour (purple-blue fading through yellow with a blue border to salmon pink) when sprayed with a 10% solution of cerium(1V) sulphate in concentrated phosphoric acid, diluted immediately before use with acetone (1:4). 7.11.2.2 Fumitremorgin-verruculogen group. Fumitremorgin A develops a slate grey-blue spot under visible light or immediately after a mustard-coloured spot under W light spraying with 50% ethanolic sulphuric acid (414). Fumitremorgin C develops a bright orange spot immediately after spraying with the same reagent and minimal heating (415). Fumitremorgin B was detected under W light and with the following spray reagents: (a) cerium(1V) sulphate (1% solution in 3 M sulphuric acid): (b) 2,4-dinitrophenylhydrazine (1 g), concentrated sulphuric acid (7.5 mL), ethanol (75 mL) and water (170 mL); (c) iron(II1) chloride ( 3 % solution in ethanol). Characteristic colours of fumitremorgin B were light purple at 254 nm, yellow-brown with rea ent (a) immediately and also after heating for 10 min at 1108C, light orange with reagent (b) after heating and orange with reagent (c) after heating. The most sensitive detection was at 254 nm with reagent (a). The lowest detectable amount of fumitremorgin B was 1 pg (197). Verruculogen (416) and 15-acetoxyverruculogen (414) become visible immediately after spraying with 5 0 % ethanolic sulphuric acid as slate-grey spots under visible light. When sprayed with a 10% solution of cerium(1V) sulphate in concentrated phosphoric acid, diluted immediately before use with acetone (1:4), verruculogen produced pinkish blue spots, fading to yellow-green (413). Mycotoxin TR-2 produced a light-brown fluorescent spot after spraying with 50% ethanolic sulphuric acid and heating for 5 min at 100°C (414). 7.11.2.3 Penitrem group. Penitrem A was revealed as a blue spot after spraying with 50% ethanolic sulphuric acid and heating (417). Penitrems A and B produce stable green spots after spraying with 1-2% iron(II1) chloride in butanol and gentle heating (418). Penitrems A-F give blue spots immediately after spraying with cerium(1V) sulphate, which become stable dark purple after heating (407). 7.11.2.4 Janthitrems. Unlike all previously discovered Penicillium tremorgens, the janthitrems are highly fluorescent under long-wave W light. The intense blue fluorescence is reminiscent of that of the aflatoxins. They can be also detected by spraying the TLC plates with Ehrlich reagent and exposure to HC1 vapour for 5-10 min, resulting in grey-green spots (419).
201
7.11.3 Selected applications Aflatrem on silica gel G plates developed in chloroform-methanol ( 9 5 : 5 ) was characterized by an RF value of about 0.8 (408). TLC was applied in monitoring the CC purification of paspalicine (410) and paxilline (412), TLC of paspaline and paspalicine carried out on Kieselgel HF plates using chloroform as the solvent gave RF values of 0.35 and 0.7, respectively. PLC was used to isolate and to purify paspalinine, paspalitrem A and paspalitrem B. The three tremorgens appeared on silica gel GH-R plates, developed in chloroform-acetone (93:7), at RF 0.60 (paspalitrem A), 0.52 (paspalinine) and 0.20 (paspalitrem B) (411). The RF values of paxilline on silica gel GH-R and on silica gel 60 F254 were 0.75 and 0.52, respectively (412, 413), when developed in toluene-ethyl acetate-formic acid (5:4:1). TLC was used to check paxilline in fractions from CC during purification of the toxin. Spraying with Ehrlich's reagent followed by heating revealed paxilline by its colour, yellow becoming green. Complementary detection involved spraying with 50% ethanolic sulphuric acid and heating at 100°C for 5 min (420 ) . Analytical or preparative TLC has been applied in studies on the role of paxilline in the biosynthesis of lolitrem B (421) and penitrems A and E (422, 423). The RF values of fumitremorgin A on silica gel GH-R plates in chloroform-acetone (97:3) and toluene-ethyl acetate-formic acid (5:4:1) were 0.30 and 0.65, respectively (414). In the latter system, the RF value of fumitremorgin C was 0.55 (415). Using the same adsorbent, fumitremorgin B had an RF value of 0.67 in diethyl ether and 0.38 in acetone-methylene chloride (5:95) as the solvent system (424). Mean RF values of fumitremorgin B on Merck pre-coated silica gel F 54 plates in six solvent systems were reported (197) as folfows: 0.51 in chloroform-methanol (97:3); 0.36 in chloroform-acetone-n-hexane (7:2:1); 0.28 in chloroform-acetone (9:l); 0.14 in ethyl acetate-n-hexane (1:l); 0.71 in chloroform-acetone-2-propanol (85:15:20); and 0.30 in benzene-chloroform-acetone (45:40:15). For verruculogen chromatographed on either MN-Kieselgel GH-R (416) and silica gel (413) plates developed in toluene-ethyl acetate-formic :zidF2?g:4: 1) , RF values of 0.65 and 0.48, respectively, have been reported. Recently, effects of temperature , light , and water activity on growth of a heat-resistant mould, Neosartorya f i s c h e r i , and production of fumitremorgins A and C and verruculogen were investigated by Nielsen and co-workers (425). Mycotoxins were analyzed by TLC on silica gel plates developed in chloroform-acetone (93:7). Fractions having the same secondary metabolite profile on thin-layer chromatograms were combined, concentrated under vacuum, and analyzed by HPLC. PLC has been used as a purification step for penitrem A (417). TLC data for the toxin have been reported by Gorst-Allman and Steyn (197), Ciegler (417) and Wilson et al.
208
procedure for the quantitative detection of penitrems (then called tremortins) in agricultural products involved extraction with chloroform-methanol (2:l) followed by TLC and Richard and Arp (4281, using colorimetric assay (427). extraction and TLC analysis, reported on the occurrence of penitrem A in mouldy cream cheese. Simple HPLC and TLC systems for the separation, identification and quantitation of the various penitrems in culture extracts were devised by Maes et a l . (407). As the penitrems are unstable in chloroform when exposed directly to light, all contact of the penitrems with this solvent was avoided. The most efficient solvent systems for the TLC separation of the penitrems were found to be (a) n-hexane-ethyl acetate (70:30), (b) dichloromethane-acetone (85:15) and (c) benzene-acetone (85:15). In solvent system (a) penitrems B and F and penitrems C and D still overlapped, whereas penitrems C and E overlapped in system (b). The only system that gave a complete separation of all the penitrems was (c) when the chromatogram was developed twice. The order of decreasing RF values for the penitrems was F, B, A, E l C and D (see Table (426). A
7.25).
TABLE 7.25 TLC of penitrems Data from refs. 407 and 429 Penitrem
A B C D E F
R, x loox A
B
C
D
16 18 9 9 13 18
49 53 39 37 46 55
37 39 28 26 33 42
32 36 22 22 28 36
E 46 32 29 50
Solvent systems: A, n-hexane-ethyl acetate (70:30); 8, dichloromethane-acetone (85:15); C, benzene-acetone (85:15); D, n-hexane-ethyl acetate (6:4); E, methylene chloride-ethyl acetate (9:l). PLC has been used in the purification of the janthitrems but CC on Mallincrodt Silica ARCC-7 silica gel was more succesful (419 ) . The three major janthitrems have the following RF values on silica gel 60 F254 pre-coated plates, developed in toluene-ethyl acetate-acetone (3:2:1): janthitrem A 0.61, janthitrem B 0.54 and janthitrem C 0.74.
209
Territrems are metabolites of Aspergillus terreus (430, 431). Territrems A and B were separated by TLC in the following solvent systems (431): (a) benzene-ethyl acetate (1:l); (b) toluene-ethyl acetate-658 formic acid (5:4:1); and (c) benzene-ethyl acetate-acetic acid (55:40:5). Detection is based on blue fluorescence of the territrems (430). Territrem C exhibited light-blue fluorescence on silica gel 60 F254 pre-coated plates at RF values of 0.25 in system (a), 0.43 in system (b) and 0.42 in system (c). The intensity of fluorescence was quenched when the concentration was higher than 20 pg per spot. The fluorescence intensity also gradually faded after development in system (a), but was enhanced and turned greenish in acidic solvent systems. PLC was used to isolate the methylation product of territrem C and its identity with territrem B was proved (431). More recently, Peng et al. (432) succeded in isolating another related metabolite. As its RF values in TLC were between those of territrems B and C , the compound was designated territrem B'. 7.12 ALTERVARIA TOXINS Alternaria mycotoxins and phycotoxins have received much interest in recent years. Production, isolation, clean-up procedures and chromatographic techniques (TLC, GC and HPLC) for the determination of alternariols, altenuene and tenuazonic acid were reviewed (433). TLC is the most widely used technique for the detection of these mycotoxins (for a review, see ref. 434). 7.12.1 Extraction and clean-up Palmisano et al. (435) extracted undried cultures of rice, maize and tomatoes or naturally contaminated samples (50 g ) in a blender with 75 mL of methanol and filtered. A 40-mL portion of the filtrate was clarified by addition of 80 mL of 5% aqueous ammonium sulphate and filtered. A 90-mL volume of the filtrate (corresponding to 20 g of the original substrate) was extracted twice with 5 mL of methylene chloride. For oleaginous samples (sunflower seeds or corn kernels) a defatting step, with 30 mL of phexane, preceded the methylene chloride extraction. The combined extracts containing the dibenzo-a-pyrone and perylene derivatives were evaporated to dryness and reconstituted with 1 mL of methanol. According to another technique (436), ground samples or kernels or chaffs (4-10 g) were extracted with 50 mL of methanol, filtered, evaporated to small volume, and, if needed, purified on Celite 545 column. 7.12.2 Adsorbents and solvent systems With Alufolien Kieselgel 60 F254 (Merck), the solvent systems used (436) were: (a) toluene-ethyl acetate-formic acid (6:3:1) and/or (b) chloroform-ethanol-ethyl acetate (90:5:5). The obtained results are given in Table 7.26.
210
TABLE 7.26 TLC data for Alternaria toxins Adapted from ref. 436. Toxin
Alternariol Alternariol methyl ether A1tenuene Altertoxin Tenuazonic acid
R,
x 100x
A
B
44
32 78
59 20 34 29
15 28 10
Solvent systems: A, toluene-ethyl acetate-formic B, chloroform-ethanol-ethyl acetate (90:5:5).
acid
(6:3:1):
7.12.3 Detection Alternaria toxins can be detected by quenching of fluorescence under UV light at 254 nm (tenuazonic acid) or by their fluorescence at 365 nm after spraying with a 20% aluminium chloride in ethanol. Yellow-orange fluorescence is characteristic for altertoxin and violet-blue for alternariol, alternariol methyl ether, and altenuene (436). 7.12.4 Selected applications TLC data for alternariol, alternariol monomethyl ether, ltertoxin I and I1 and tenuazonic acid were published (437). %-labelled alternariol and alternariol monomethyl ether were isolated from ethyl acetate extracts of conidia of A. alternata by PLC. Two solvent systems were used (438): (a) toluene-dioxane-acetic acid (95:25:4) and (b) methanol-2 M HC1 (5:l). Visconti et al. (439) used TLC to determine alternariol, alternariol methyl ether, altenuene, and tenuazonic acid in olives. Altenuene, alternariol, alternariol methyl ether (dibenzo-a-pyrone derivatives), altertoxin-I and altertoxin-11 (perylene derivatives) were found in extracts of artificially infected maize, rice and tomato samples and naturally contaminated sunflower sedds (435). Natural occurrence of Alternaria toxins (alternariol and alternariol methyl ether) in the grain and chaff of cereals was detected (436). 7.13 CITRININ TLC has been used by many workers to characterize, identify and quantitate citrinin in various commodities and also in preparative work. Chromatographic methods, including TLC, were reviewed (11, 30, 440).
21 1
7.13.1 Extraction and clean-up Extraction solvents and clean-up techniques for citrinin are given in Table 7.27. Chloroform, ethyl or butyl acetate and methanol are the most commonly used solvents for extraction. Originally, precipitation from culture filtrates with concentrated hydrochloric acid was applied (441). In clean-up procedures, silica gel CC, Extrelut columns or partition at different pH values between aqueous and organic phases have been used. TABLE 7.27 Extraction and clean-up of citrinin Materia1
Extraction solvent( s
Clean-upx
Culture filtrate
Precipitation with conc. HC1 (12.5 mL/L)
Culture filtrate
EtOAc at pH 2.5
Culture filtrate
CHC13 followed by EtOAc from conc. filtrate
Culture filtrate
BuOAc at pH 2.5
Corn
CHC13
Culture filtrate Culture filtrate
CHCl3
Crude CIT dissolved in 441 CHC13 crystallization from h O H Partition into buffer pH 442 8.5, re-extraction with CHC13 at pH 2.5, evaporation, partition between CC14 and (CH20H)2, CC14 phase evaporated, crystallization from Me CO cc silica gel, efution 443 with CHC13, partition into 0.2 M NaHCO , acidification , crysta? 1ization of precipitate from EtOH Evaporation, dissolution 444 in C6H6, partition into sat. aq. KHC03, re-extraction with c H at pH 3.8, evaporation ,6dfssolution in EtOH Extract rinsed with dil. 445 HC1 and H20, partition into 0.1 M NaHC03, reextraction with CHCl at pH 2.5 and concentragion, partition into 0.1 M NaHC03, precipitation (pH 2.5) Concentration and TLC 446
Static culture Culture filtrate and mycelia
EtOAc
CHC13 at pH 1.5
EtOAc (filtrate) Hot EtOAc (mycelium)
Ref.
Evaporation, dissolution 447 in CHC13 or 0.1 M buffer (PH 10) Concentration 443 Extract passed through Na SO4 and concentrated u d e r N2
448
212
TABLE 7.27 Conth u e d ~
~~
~
Ref.
Materia1
Extraction solvent( s
Clean-upx
Tomatoes
MeOH and Hex
Maize
MeOH-CHC13 (1:l)
Centrifugation, 5 M 448 H2S04 added, partition into CHC13, evaporation, dissolution in CHCl Filtration , evaporazion , 197 partition Hex-90% MeOH (l:l), MeOH layer evaporated, partition CHC13H20 (l:l), CHC13 layer extracted with sat. NaHC03, re-extraction with CHC13 at pH 2, concentration CHC13 layer passed 449 through Extrelut column
Cereal grains CHC13-0.1 M H3P04 (15:2)
Abbreviations: CIT, citrinin: EtOAc, ethyl acetate: CHC13, chloroform: CC14, carbon tetrachloride: (CH20H)2, ethylene glycol: Me2C0, acetone: BuOAc, butyl acetate: C6H6, benzene: EtOH, ethanol: MeOH, methanol: Hex, n-hexane. 7.13.2 Adsorbents and solvent systems Silica gel is the most often used adsorbent in the TLC of citrinin. Better results were obtained on oxalic acid pre-treated plates. Silufol plates were impregnated with 0.25 M oxalic acid in methanol by developing the plates in the solution. The plates were then dried in air and spotted (450). Marti et a l . (451) dipped inactivated silica gel 60 in 10% oxalic acid. Gorst-Allman and Steyn (197) immersed the plates in a 10% solution of oxalic acid in methanol for 2 min. After heating at llO°C for 2 min and cooling, the plates were immediately spotted and developed. Gimeno (452) found glycolic acid to be better then oxalic acid because of reduced diffusion of the citrinin spots and hence enhanced detectability. A variety of solvent systems have been used by various workers: some of them are mentioned in 7.13.4. 7.13.3 Detection Citrinin can be observed on chromatograms under UV light owing to its yellow fluorescence. In addition, several spray reagents have been employed. Curtis et a l . (453) used a freshly prepared solution of a stabilized diazonium salt of o-dianisidine (0.05 g in 4 0 mL of methanol-water, l:l), followed by methanol-aqueous ammonia (1:l) to promote the coupling reaction. Citrinin produced a pale pink colour. Improved colour
213
resolution was obtained if the TLC plates were allowed to dry overnight before spraying. After spraying with a 3% solution of iron(II1) chloride in methanol, citrinin is detected as a brown spot (450). Citrinin was also detected with p-anisaldehyde spray reagent (130). Gorst-Allman and Steyn (197) detected citrinin and other acidic mycotoxins under short- and long-wave UV light or by spraying with cerium(1V) sulphate, 2,4-dinitrophenylhydrazine and iron(II1) chloride reagents. Marti et a l . (451) obtained a detection limit of 20 ng per spot of citrinin by measuring the yellow-green fluorescence under UV light. 7.13.4 Selected applications Curtis et a l . (453) examined phenolic metabolites including citrinin using Kieselgel G plates and benzene-methanol-acetic acid (10:2:1) as the solvent system. Betina and Binovska (444) monitored the production of citrinin in the course of a submerged fermentation. The cleaned-up samples (see Table 7.27, ref. 444) were spotted on oxalic acid-impregnated Silufol plates. The most suitable solvent systems were benzene-methanol-acetic acid (10:2:1) and benzene-methanol (95:5). In producing 14C-citrinin by P. citrinum, Phillips et a l . (454) isolated and purified the compound by the method of Davis et a l . (441). The identity and purity of citrinin were established by TLC using diethyl ether-hexane-formic acid (75:25:1) and ethyl acetate-acetone-0.1 M (40:40:20) as the solvent systems. A single peak of radioactivity appeared, which co-chromatographed with authentic, chemically pure citrinin. The production of citrinin in corn was monitored by TLC on silica gel F2 using the solvent system chloroform-methanol (75:25) and dezection under 366 nm UV light (445). Harwig et al. (448) detected citrinin in extracts from Penicillium spp. cultures isolated from decaying tomato fruit, and also in tomato extracts, using silica gel plates and the solvent systems toluene-ethyl acetate-formic acid (5:4:1) and ethyl acetate-acetone-water (5:5:2). TLC analysis and chemical confirmation of citrinin in barley were reported by Hald and Krogh (455). TLC quantitations of citrinin have been published. Wu et a l . (446) separated citrinin-containing extracts by TLC on Adsorbosil-1 using toluene-ethyl acetate-formic acid (6:3:1) as the developing solvent and fluorodensitometry. Damodaran et a l . (447) reported a procedure for the isolation and quantitation of citrinin in culture filtrates. Cleaned-up samples were spotted on to silica gel plates and developed in toluene-ethyl acetate-formic acid (5:4:1). The fluorescent portions were citrinin was extracted with scraped off , carbonate-hydrogencarbonate buffer pH 10 and the determinations were carried out using Folin's reagent. Further quantitations have been reported by Ciegler et a l . (443) and Damoglou et a l .
2 14
(456). The latter procedures were shown to be of importance in the separation and identification of dihydr itrinone and ochratoxin A as products of conversion of “C-citrinin by Penicillium viridicatum (457). The presence of radiolabelled products on TLC plates formed by the breakdown of citrinin was assessed by autoradiography. 7.14 a-CYCLOPIAZONIC ACID Of the known tetramic acids, a-cyclopiazonic acid is the most studied. Data concerning the production, isolation, separation and purification of this and related toxins were reviewed by Cole (458). 7.14.1 Extraction and clean-up Gorst-Allman and Steyn (197) carried out extraction of Penicillium cyclopiumcontaminated maize with methanolchloroform (1:1), the mixture was filtered and the filtrate evaporated to dryness. The residue was partitioned between n-hexane and 90% methanol (1:l) and the methanol layer was evaporated to dryness. The solid was partitoned between chloroform and water (1:l) and the chloroform layer was extracted with saturated sodium hydrogencarbonate solution. The aqueous layer was acidified to pH 2 and extracted with chloroform. The chloroform extract was concentrated and contained a-cyclopiazonic acid. LeBars (459) extracted cheese samples with azeotropic chloroform-methanol. The filtered and evaporated extract was dissolved in acetone-water-lead acetate solution. A saturated solution of sodium sulphate and Celite were added and the suspension was filtered. The filtrate was defatted by partition against hexane, acidified to pH 3 and extracted with chloroform. The centrifuged and dried extract was evaporated to dryness and dissolved in the minimum volume of chloroform for TLC. Benkhemmar et al. (460 ) extracted cyclopiazonic acid from culture filtrates a modified Le Bars technique. A 40-mL portion of culture FXltrate, adjusted to pH 3 with aqueous HC1 ( 5 0 : 5 0 ) , was extracted with four volumes of methanol-chloroform (1:4). The methanol-chloroform layer was decanted and retained, and then it was dried with Na2S04, filtered, and vacuum concentrated to dryness. The crude extract was taken up in chloroform for characterization by TLC to discrimiate cyclopiazonic acid-producing from non-producing strains of Aspergillus oryzae. According to Lansden (461), samples of peanuts or corn were extracted with methanol-chloroform (20:80); the extract was stripped of most interferences by partitioning with aqueous sodium hydrogencarbonate followed by acidification and repartitioning with chloroform. Rao and Husain (462) extracted the toxin from culture filtrates and solid substrates as follows. Twenty five mL of the
215
filtrate was extracted twice with equal amounts of chloroform. The pooled extract was dried over anhydrous Na2S04 and evaporated to dryness. The residue was dissolved in one mL chloroform and used for TLC. The solid substrates were first defatted with petroleum ether by thorough extraction using a mechanical shaker. The defatted substrate was extracted with 200 mL of chloroform twice for 24 h on a mechanical shaker. The extracts were pooled, filtered, dried over anhydrous Na2S04 and evaporated to dryness. The residue was dissolved in one mL of chloroform and used for TLC. Extraction procedures of Hermansen e t al. (463 ) started by homogenization of culture broths. The homogenate was acidified with 2 M HC1 to pH 2 and extracted with chloroform-methanol (4:l) by shaking for 14 h. After filtration the phases were separated, and the organic phase was dried (Na2S04) 2a;; evaporated to dryness. The residue was dissolved in chloroform. Isolates of Aspergillus and Penicillium species from dried beans, corn meal, macaroni and pecans were examined for their ability to produce cyclopiazonic acid. From static fungal cultures in 100 mL volumes of a culture medium, the culture broth and the mycelial mat were extracted by adding 100 mL chloroformin the flask and soaking with occasionally shaking for 24 h (464). The mixture was heated in a steam bath until it boiled. After cooling, 10 mL of the chloroform layer was withdrawn and filtered through 5 g anhydrous Na2S04. The filtrate was collected in a vial. The sodium sulphate was washed with 2 mL chloroform and was collected in the same vial. The extract was evaporated to dryness on a steam bath under a stream of nitrogen. 7.14.2 Adsorbents and solvent systems Silica gel TLC plates have been impregnated with oxalic or tartaric acid (197, 459). A variety of solvent systems have been used, e . g . , (a) chloroform-methyl isobutyl ketone (4:1), (b) chloroformmethanol (98:2), (c) chloroform-acetone (9:1), (d) ethyl acetate-2-propanol-ammonia solution (20:15:10), (el chloroform-acetone (95:5), or (f) toluene-ethyl acetate-formic acid (5:4:1). Systems (b) and (c) are recommended for acidic mycotoxins. Systems (a), (d) and (f) were used by Le Bars (459) for the quantitation of cyclopiazonic acid from commercial cheese samples. 7.14.3 Detection Cyclopiazonic acid can be detected by derivatization with Ehrlich's reagent. Lansden (461) recommended the following preparation of this spray reagent: 1 g 4-dimethylaminobenzaldehyde is dissolved in 75 mL ethanol and 25 mL concentrated HC1 are added. The dried plates are sprayed with the reagent until first appearance of blue spot among cyclopiazonic acid standards. Colour is developed within 10 min, without heating.
2 16
In addition to Ehrlich's reagent, cyclopiazonic acid can be detected with either iron(II1) chloride or concentrated sulphuric acid and heating (465). Other detection methods were reported by Gorst-Allman and Steyn (197). 7.14.4 Selected applications A semi-quantitative TLC in the presence of appropriate internal and external standards was published by Hermansen et al. (463). The analyses were performed on silica gel 60 precoated on glass. Before use the TLC glates were dipped in 0.3 M aqueous oxalic acid and dried at 110 C for 2 h. Standards and samples (2 pL) were applied on the plate and developed in toluene-ethyl acetate-formic acid (5:4:1) followed by drying at room temperature. Cyclopiazonic acid was detected with Ehrlich's reagent and showed as a spot with RF 0.70. Using Lansden's procedure (461) and detection with Ehrlich's reagent, the toxin was quantitated by reflection densitometry at 540 nm. The detection limit was 25 ng per spot. A simple determination of cyclopiazonic acid in contaminated food and feeds was described by Rathinevalu et al. (466). Semi-quantitative TLC of the toxin in extracts from culture media has been reported by Trucksess et al. (464) and Hermansen et al. (463). al. (460) applied TLC to discriminate Benkhemmar et cyclopiazonic acid-producing (CPA ) from non-producing (CPA-) strains of Aspergillus oryzae. TLC was performed on oxalic acid-impregnated silica gel plates and chloroform-methyl isobutyl ketone (4:l) as the solvent system. After detection with Ehrlich's reagent, the toxin from CPA+ strains revealed a blue-violet spot at R 0.75. TLC has been appfied in studies on the production of cyclopiazonic acid by Penicillium verrucosum var. cyclopium (467). TLC was performed on silica gel G-1500 LS 254 with ethyl acetate-2-propanol-25% ammonia solution (20:15:10). The toxin was measured quantitatively with a spectrodensitometer with a digital counter and integrator at 282 nm. It was detected as a violet spot under visible light after spraying with Ehrlich's reagent. In screeing the toxin in agricultural commodities, Rao and Husain (462) applied PLC to chloroform extracts from culture filtrates. The standard was spotted at one end of the plate. After development (the same system as in the latter paper), the standard was detected with Ehrlich's reagent (the remainder of the plate being covered with a glass plate). When the standard was detected, the covering plate was removed and the TLC plate exposed to iodine vapour. The area with an R value corresponding to the standard spot and coloured witK iodine vapour was scraped off, eluted with methanol, and used for colorimetric determination of the toxin using a modification of Ehrlich's reagent. In addition to TLC methods, cyclopiazonic acid in agricultural products and foods can be successfully determined by HPLC (see Chapter 8).
217
7.15 PR TOXIN AND ROQUEFORTINE PR toxin and roquefortine are secondary metabolites of strains of Penicillium rogueforti and have been isolated from fungal isolates from blue cheese and other sources. The production, isolation and chromatographic techniques were reviewed (30, 468). 7.15.1 Extraction and clean-up Still (469) extracted PR toxin from culture filtrates with chloroform and Scott et al. (470) used ethyl acetate. Two basic procedures for extraction and clean-up from blue cheese were published by Scott and Kanhere (471). In the first procedure, the sample was extracted with a mixture of methanol-water and hexane and centrifuged. After filtration, the methanol-water layer was extracted with chloroform, the extract was evaporated, the residue was dissolved in chloroform and immediately analysed by TLC for PR toxin and/or PR imine. In the second procedure, cheese was blended with ethyl acetate and centrifuged. The extract was evaporated and partitioned between hexane and acetonitrile. The acetonitrile layer was evaporated and the residue was dissolved in chloroform for immediate TLC analysis. Roquefortine is present mainly in the mycelium of P. roqueforti. Extraction and clean-up procedures were summarized by Scott (472). CC procedures for the separation of roquefortine from other alkaloids isolated from P. rogueforti or other penicillia have also been described. It has been found that roquefortine could be eluted with chloroform-methanol-25% ammonia solution (70:10:0.5) from silica gels (473) and with chloroform-ethanol (95:5) from basic alumina (474). Fractions from CC columns were monitored by TLC. 7.15.2 Adsorbents and solvent systems Solvent systems for the TLC of PR toxin on silica gel include chloroform-methanol (96:4), chloroform-2-propanol (1O:l or 4:1), toluene-ethyl acetate-formic acid (5:4:1 or 6:3:1) and toluene-ethyl acetate (30:70) saturated with water (471,475, 476). Solvent systems for roquefortine that have been used with silica gel TLC plates (477, 478) include chloroform-methanol-28% ammonia solution (90:10:1), chloroform-methanol (9:1), chloroform-re-distilled diethylamine (8:2), chloroform-ethanol (10:1.5), acetone-chloroform (3:2) and benzene-methanol (93:7). 7.15.3 Detection PR toxin can be detected by its green fluorescence under long-wave UV light following exposure of the chromatograms to short-wave UV light for about 0.5 min (471, 475). After spraying the chromatograms with 50% sulphuric acid, the toxin appears as a yellow spot (475). The toxin was quantitated in situ by
218
fluorodensitometry
after
spraying the plates with 1% in concentrated HC1-acetone (1:lO) or in ethanol with subsequent exposure to HC1 fumes for 10 mini the latter is the preferred method (476). Roquefortine on TLC plates can be detected as a blue-grey spot after spraying with 5 0 % sulphuric acid and heating at llO° C for 10 min (479). Other spray reagents are Pauli reagent (480), Van Urk reagent (473) and Ehrlich's reagent (474, 478). pdimethylaminobenzaldehyde
7.16 XANTHOMEGNIN, VIOMELLEIN AND VIOXANTHIN
Xanthomegnin, viomellein and vioxanthin are toxic metabolites of a number of fungi including Aspergillus and Penicillium species: these micromycetes are of particular interest because they are routinely implicated in toxin contamination of foods and feeds. 7.16.1 Extraction and clean-up Wall and Lillehoj (481) used the following extraction and clean-up procedure. A strain of A. ochraceus was cultivated on rice for 10 days. The mouldy rice was extracted by suspension in methylene chloride and grinding. The extract was filtered and the solvent removed by vacuum evaporation. The crude oil was subsequently extracted three times with acetonitrile and the acetonitrile solutions were used for chromatography. 7.16.2 Adsorbents and solvent systems TLC methods for the detection of xanthomegnin and viomellein utilize silica gel plates and benzene-methanol-acetic acid (18:l:l) or toluene-ethyl acetate-formic acid (6:3:1) as the solvent systems (482). 7.16.3 Detection After standing for 6 h, the spots of xanthomegnin turn from yellow to orange and those of viomellein turn from yellowish green to yellowish brown. Exposure to ammonia fumes turns the compounds from yellow to purple (482). The detection limits were 0.1 pg for xanthomegnin and 0.3 kg for viomellein. 7.16.4 Selected applications Standards of xanthomegnin and viomellein were prepared by Wall and Lillehoj (481) by PLC on silica gel plates that were developed in benzene-methanol-acetic acid (18:l:l). Appropriate bands were scraped off the plates and the compounds were eluted with methylene chloride. The solvent was removed under a stream of nitrogen and standards were stored as dry films in a freezer. Purity was determined by TLC and HPLC comparisons with reference compounds. In a screening for toxigenic isolates of Aspergillus ochraceus from green coffee beans, Stack et al. (483) applied TLC in detecting xanthomegnin, viomellein and vioxanthin in addition to ochratoxins.
219 7.17
NAPHTHO-r-PYRONES
Monomeric and dimeric naphtho-r-pyrones have been isolated from the mycelium of Aspergillus niger by several groups of workers. Ehrlich et al. (484) subcultured an A. niger isolate on rice, corn, cottonseed and two liquid media. After incubation, the culture (in the case of culture on liquid media, the mycelial mat) was extracted with methylene chloride. The solvent was evaporated and the residual red paste was treated with 9 volumes of cold hexane and kept at 5OC overnight. The red precipitate was collected, dissolved in methylene chloride and filtered. Samples were examined by HPTLC. HPTLC was carried out on LHP-KF plates (Whatman) and developed with benzene-ethyl acetate-formic acid (10:4:1). Components were identified by their colour, fluorescence under long-wave UV light and colour after spraying with Gibbs reagent. HPTLC showed that the mixture contained more than 18 components, contained but only the material migrating at RF 0.5-0.8 naphtho-r-pyrones. The results are given in Table 7.28. TABLE 7.28 HPTLC data for naphtho-r-pyrones Adapted from ref. 484. Compound Flavasperone Fonsecin monomethyl ether Rubrofusarin Aurasperone A Isoaurasperone A Aurasperone B Aurasperone D Aurasperone C
RF x loox 81 76 72 67 61 56 53 49
Gibbs test
Fluorescence
Blue Brown Blue-green Violet Red-violet Brown Violet Brown
Violet Violet Orange Yellow Yellow Yellow Yellow Yellow
With benzene-ethyl acetate-formic acid (100:40:10) on Whatman LHP-KF 7.18 SECALONIC ACIDS
The secalonic acids are xanthone dimers with identical molecular masses and molecular formulae, differing in their stereochemistry. Secalonic acid D is the most studied member of this group (485). Methods used for the production, isolation, separation, purification and detection of secalonic acid D have been summarized (486). TLC and HPLC techniques were also included. On TLC plates, secalonic acids can be detected by quenching fluorescence (487) or by spraying with cerium(1V) sulphate
220
reagent (488), iron(II1) chloride, or panisaldehyde reagent (130). RF values of secalonic acids in a variety of solvent systems are given in Table 7.29. Ciegler et al. (487) quantitated secalonic acid D on pre-coated silica gel F254 plates using benzene-ethyl acetate-formic acid (100:40:10) as the solvent system. TABLE 7.29 RF x 100 values for secalonic acids SorbentX
OA-treated silica gel TA-treated silica gel Silufol
Solvent systemX
CHC13-MP (9:l)
Secalonic acid A
B
D
23
46
23
CHC13-Pen
17
C6H6-MeOH-HOAC (24:2:1) Tol-EtOAc-FA (6:3:1) CHC1,-MeOH (4:l)
28 32 68
Ref. F 488, 489 29
490
130
Abbreviations: OA, oxalic acid; TA, tartaric acid; CHC13, chloroform; MP, 4-methyl-2-pentanone; CgHg, benzene; MeOH, methanol; HOAc, acetic acid; Toll toluene; EtOAc, ethyl acetate; FA, 90% formic acid; Pen, 2-pentanone. 7.19 TLC OF MISCELLANEOUS TOXINS In this section, TLC data for the following compounds are included: moniliformin, wortmanin, echinulin, fusaric acid analogues, fusarin C, viridin and toxic peptides. Jansen and Dose (491) described a quantitative TLC determination of moniliformin in vegetable foods and feeds. Crude acetonitrile extracts of Fusarium moniliforme cultures were checked for moniliformin (492, 493) by spotting, together with a standard, on pre-coated thin layers of silica gel 60 and developing in chloroform-methanol-formic acid (70:30:0.16). The toxin was detected by spraying and heating with 0.5% aqueous 3-methyl-2-benzothiazolinone hydrazone hydrochloride. The limit of detection was approximately 8 wg/g in corn culture. Most recently, Chelkowski et al. (494) published a simple TLC method for moniliformin detection. Rice cultures were dried and powdered. Moniliformin was extracted from 3 g of powdered culture with 6 mL of water to form a slurry, diluted after 15 min with 34 mL of ethanol. This suspension was kept overnight in a refrigerator and filterd the next day. The filtrate (1-20 pL) was spotted on to Merck 5553 TLC plates. A moniliformin standard in methanol (100 pg/mL) in amounts of 0.2, 0.5, and 1 pg in
22 1
a spot was placed on the same plate. Plates were developed in chloroform-methanol (6:4) as the solvent system and moniliformin 0.5% water solution of MBTH Aldrich visualized with (3-methyl-2-benzothiazoline-hydrazone hydrochloride, freshly prepared) after heating for 10 min at 14OoC. The spots which appeared were red-violet, with detection limit 0.1 pg in each spot. In acidic atmosphere the colour of spots turns into brown-grey and even green, so it is necessary to avoid contact of chromatograms with vapours of volatile acids (HC1 and others). Ammonia vapours intensify formation of the carmine-red colour. The authors recommended placing developed plates for 5 min into a tank with ammonia vapours before they are sprayed with MBTH. Shepherd and Gilbert (495) developed an effective HPLC method for moniliformin. PLC on silica gel plates developed with chloroform-methanol (97:3) was used to purify a haemorrhagic factor from Fusarium oxysporum identical with the antibiotic wortmannin (496). Echinulin was isolated by means of PLC from acetone extracts of feed refused by swine. The solvent system was ethyl acetate-hexane (8:2) and the toxin turned blue in the presence of panisaldehyde reagent at llO°C. The anisaldehyde-reactive material from the PLC was identified with echinulin by its UV and IR spectra (497). Viridin, a steroid-like antibiotic, is converted by viridin-producing fungi intc its dihydro derivative, viridiol, which is ineffective as an antibiotic but is a potent phytotoxin. Both metabolites were isolated from culture extracts by means of PLC (498). TLC was used to characterize two new fusaric acid analogues from Fusarium moniliforme (499). Fusarin C is a mutagenic mycotoxin produced by Fusarium moniliforme. Its natural occurrence in corn was reported by Gelderblom et al. (500). Corn samples were extracted with water and methylene chloride-2-propanol (1:l). After filtering, drying the extract and evaporating to dr ness, the residue was extracted with petroleum ether (60-808C) and chloroform. The petroleum ether was re-extracted with acetonitrile and the residues from the chloroform and acetonitrile extracts were chromatographed on a column of silica gel with methylene chloride-methanol (19:l) as the eluent. Scott et a l . (493) used acetonitrile to extract ground corn, corn meal, or wheat flour. After filtration and evaporation of the solvent, clean-up was carried out on small disposable amino bonded phase or silica gel columns with methylene chloride-methanol (9:l) as eluting solvent. TLC on silica gel has been used by Farber and Sanders (501) or by Wiebe and Bjeldanes ( 5 0 2 ) using chloroform-methanol (9:l) or chloroform-2-propanol (9:l) as the solvent
222
systems.Standards and positive samples were identified by the presence of bright yellow spots under visible light ( R F in the former system, 0.32 to 0.35). Jackson e t al. (503) assessed fusarin C standard purity by TLC, nuclear magnetic resonance, and mass spectral analysis. TLC has also been used to characterize cyclosporin A extracted from rice (504). Silica gel plates were developed in 3 solvent systems: (1) n-butanol-acetic acid-water (4:1:1), (2) chloroform-acetic acid-methanol (85:10:5), and (3) ethyl acetate-hexane-acetone (2:l:l). TLC plates were dried with a hot-air blower gun placed in an iodine chamber for 15 min to detect iodine-reactive substances. Iodine was sublimed from the plates by placing them in an oven at llO°C for 15 min. The plates were sprayed with 6 M HC1 and oven-dried at llO°C for 30 min. Dried plates were sprayed with 0.1% ninhydrin solution in n-butanol. Orange to brown spots had RF values: 0.83 in system 1: 0.92 in system 2: 0.81 in system 3. 7.20 MULTI-MYCOTOXIN TLC Various multi-mycotoxin methods have been published for the simultaneous detection of a number of mycotoxins, which differ in the extraction solvents, clean-up procedure and final detection TLC procedure. In clean-up techniques, mini-column chromatography has been used by several workers ( e . g . , refs. 60, 61, 505-507). Patterson et al. (508) used a dialysis clean-up procedure. A final TLC analysis has been adopted in the following selected instances. Originally, Eppley (61) described a screening method for zearalenone, aflatoxin and ochratoxin. His techniques were subsequently used or adapted by various workers. Steyn (465) reported a TLC system for the simultaneous separation and detection of eleven mycotoxins, in which extensive purification of acidic mycotoxins was achieved by removal of the neutral material. The procedure used silica gel G TLC plates impregnated with oxalic acid, with development in chloroform-methyl isobutyl ketone (4:l). The mobility of the neutral mycotoxins was essentially unaffected when oxalic acid was omitted, whereas the acidic mycotoxins, e . g . , cyclopiazonic acid and secalonic acid, and also ochratoxins remained at the origin. The mycotoxins were detected by examination of TLC plates under long-wave UV light and spraying with 1% cerium(1V) sulphate in concentrated sulphuric acid or 1% ethanolic iron(II1) chloride. Later, Gorst-Allman and Steyn (197) used the following spray reagents: (a) 2,4-dinitrophenylhydrazine (1 9)-concentrated sulphuric acid (7.5 mL)-ethanol (75 mL)-water (170 mL); (b) hydrazono-2,3-methylbenzothiazole hydrochloride (0.5% aqueous solution); (c) iron(II1) chloride ( 3 % solution in ethanol): (d) aluminium chloride (1% solution in chloroform): (e) Ehrlich reagent: (f) cerium(1V) sulphate (1% solution in 3 M sulphuric acid): (9) vanillin (1% in 50%
223
phosphoric acid). The plates were sprayed, the immediate effects noted, and they were then heated at llO°C for 10 min. Iodine and ammonia fumes were also used for some plates. Characteristic colours were reported. Whidden et al. (507) developed a method for simultaneous extraction, separation and qualitative analysis of rubratoxin B, aflatoxin B1! diacetoxyscirpenol, ochratoxin A, patulin, penicillic acid, sterigmatocystin and zearalenone in corn. Mycotoxins were extracted with acetonitrile, sequentially eluted from a silica gel mini-column and rendered visible by TLC. A flow chart for the extraction and separation of the eight mycotoxins is presented in Fig. 7.1. Fractions 11-IV were analysed on the same TLC plate using external and internal standards and the solvent system toluene- ethyl acetate-formic acid (6:3:1). Fraction V (containing rubratoxin B) was applied to a separate TLC plate together with external standards (five concentrations of the toxin) and developed in acetonitrile-acetic acid (100:2).
Ground sample Acetonitrile Residue
Filtrate Wash with
II
isooctane
I
I
Isooctane
Acetonitrile
(g5:r
acetone
ether
methanol
Fig. 7.1. Flow chart for the extraction andseparation of mycotoxins Adapted from Whidden et al. (ref. 507).
224
A multi-mycotoxin method involving a membrane clean-up step two-dimensional TLC was published by Patterson et a l . (508). Fishbein and Falk (509) developed TLC procedures for five types of mycotoxins (aflatoxins, ochratoxins, aspertoxin, 0-methylsterigmatocystin and sterigmatocystin) and some other fungal metabolites. Stoloff et a l . (510) described a multi-mycotoxin TLC method for aflatoxins, ochratoxins, zearalenone, sterigmatocystin and patulin in a number of agricultural products. They used silica fluorophores and 9e1 plates with internal benzene-methanol-acetic acid (18:l:l) or hexane-acetone-acetic acid (18:2:1) as the solvent system. The developed plates were viewed under both short- and long-wave UV light. The limits of detection ranged from 20 (aflatoxin) to 450 wg/kg (patulin). Joseffson and Moller (505) reported detection limits of aflatoxin 5, ochratoxin 10, patulin 50, sterigmatocystin 10 and zearalenone 35 pg/kg by using gel filtration on Sephadex LH-20 as a clean-up procedure prior to TLC. Wilson et a l . (290) published a method for the detection of aflatoxins, ochratoxins, zearalenone, citrinin and penicillic acid. Mycotoxins in chloroform extracts were isolated by CC and then separated by TLC on Adsorbosil-1 pre-coated plates. Moubasher et al. (511) evaluated the toxin-producing potential of fungi isolated from blue-veined cheese. The toxins tested for were aflatoxins, patulin, versicolorin, sterigmatocystin, ochratoxin A, kojic acid and penicillic acid. Coman et a l . (512) reported a TLC analysis of feed samples in which four aflatoxins, ochratoxin A, zearalenone, sterigmatocystin and T-2 toxin were detected. Zearalenone, T-2 toxin, neosolaniol and HT-2 toxin were detected in grains of barley, wheat and oats by Ilus et a l . (513) as follows. Toxins were extracted with ethyl acetate, purified on a Kieselgel TLC plate and analysed by TLC using acetone-hexane as the solvent with detection at 3 6 6 nm or with panisaldehyde reagent. Nowotny et a l . (514) detected citrinin, ochratoxin A and sterigmatocystin in samples of commercial cheese using TLC and HPLC. Gimeno and Martins (515) described a rapid TLC determination of mycotoxins which can often be found in fruits and fruit products. The method was tested for patulin, citrinin and aflatoxin in apples and pears and their juices and jams. The mycotoxins were extracted with a mixture of acetonitrile and 4% aqueous KC1 (9:l). The extract was cleaned up with water and then acidified, and the toxins were recovered with chloroform and separated by TLC. Toxin identity was confirmed with various developing solvents, spray reagents and chemical reactions and then quantitated by the limit of detection method. The minimal detectable concentrations were: patulin 120-130, citrinin 30-40, aflatoxin B1 and G1 2-2.8 and aflatoxin B2 and G2 2 wg/kg * and
22 5
A method for the routine examination of mouldy rice, wheat bread and other vegetable foodstuffs was published by Johann and Dose (376). The mycotoxins are first extracted with acetonitrile-4% KC1 and cyclohexane and then transferred from acetonitrile into a methylene chloride phase and separated by two-dimensional TLC. Aflatoxins are determined fluorimetrically after development in chloroform-acetone (9:l) and methylene chloride-acetonitrile (8:2). Other mycotoxins (ochratoxin A, patulin, penicillic acid, and sterigmatocystin) are analysed on separate plates with toluene-ethyl acetate-acetic acid (6:3:1) and benzene-acetic acid (8:2). Citrinin is chromatographed on a plate pre-treated with oxalic acid. Citrinin and ochratoxin A, like the aflatoxins, can be immediately determined by fluorimetry, whereas the other toxins have to be converted into fluorescent derivatives using spray reagents (penicillic acid using diphenylboric acid-2-ethanolamineI patulin using N-methylbenzthiazolone-2-hydrazone and sterigmatocystin using aluminium chloride) for quantitative determination. Gorst-Allman and Steyn (197) separated 13 mycotoxins as neutral (aflatoxin B1, sterigmatocystin, zearalenone, patulin, T-2 toxin, roquefortine, penitrem A , fumitremorgin B and roridin A) and acidic (citrinin, ochratoxin A , a-cyclopiazonic acid and penicillic acid) metabolites. Mean values of the neutral mycotoxins are presented in Table '7.30 and those of acidic mycotoxins in Table 7.31. The acidic mycotoxins were well separated on silica gel plates pre-treated with oxalic acid.
TABLE 7.30 Mean R x 100 values of neutral mycotoxins Adaptes from ref. 197. Mycotoxin
Aflatoxin B1 Sterigmatocystin Zearalenone Patulin T-2 toxin Roquefortine Penitrem A Fumitremorgin B Roridin A
Solvent system'
A
B
C
D
E
44
35 53
27 55
03
65 74 71 56 68 13 76 71 61
67
40 22 45
51 27 36
16 22
03
01
02
40 51 31
51 36 22
38
34 28 13
41 41 18 13 01 49 14 09
F
24 56
44 20
22 02
45 30
14
Solvent systems: A, chloroform-methanol (97:3); B, chloroform-acetone-nhexane (7:2:1): C, chloroform-acetone (9:l): D, ethyl acetate-nhexane (1:l): E l chloroform-acetone2-propanol (85:15:20); F, benzene-chloroform-acetone (45:40:15).
226
TABLE 7. 31 Mean RF x 100 values of acidic mycotoxins using pre-treated with oxalic acid Adapted from ref. 197. Mycotoxin
Citrinin Ochratoxin A a-Cyclopiazonic acid Penicillic acid
TLC plates
R F X 100
Chloroform-methanol
Chloroform-acetone
(98:2)
(9:1)
52 32 52 16
51 34 44 20
buraekova e t al. (130) presented a TLC systematic analysis for 37 mycotoxins and 6 other fungal metabolites in which Inchromatographic spectra" were generated for each toxin from their R values in eight solvent systems. The advantage of this system Ties in the comparisons of relative rather than absolute RF values, as the latter show greater variations than the former with changes in the conditions of the environment. This method was developed for the identification of known mycotoxins. The chromatographic spectrum of an unknown substance provides a preliminary identification by comparison with known chromatographic spectra or eliminates the known metabolites from the unknown. The method was extended to the detection of unknown mycotoxins by combining it with a bioassay to yield a bioautographic detection method (134). Lee e t al. (128) described a method for the simultaneous determination of thirteen mycotoxins by HPTLC. With seven continuous multiple developments with two solvent systems of different polarity, a baseline separation of sterigmatocystin, zearalenone, citrinin, ochratoxin A, patulin, penicillic acid, luteoskyrin and aflatoxins B1, B , G1, G2, M1 and M 2 was obtained. About 1 h was require3 for the separation and quantitation of all 13 mycotoxins from one spot. By using in s i t u scanning of the HPTLC plate, detection limits in the low nanogram range were obtained by UV-visible absorption and in the low picogram range by fluorescence, with a relative standard deviation of 0.7-2.2% in the nanogram range. Chromatography was performed on 10 x 10 cm HPTLC plates coated with silica gel 60 and impregnated with EDTA. The development stage and spectroscopic properties used for quantitative determination of the individual mycotoxins are given in Table 7.32. The mobile phase migration distance was 4 cm and was fixed by arranging for a portion of the plate to protrude through the top of the
221
saturated development chamber, at which point the solvent could freely evaporate. For the very complex sample of 13 mycotoxins, the use of continuous multiple development offerred certain advantages, such as the possibility of quantifying the components as they were separated, the use of more than one solvent system, and natural refocussing of the sample spot, which occurred when the plate was dried between developments. The resolution of sterigmatocystin, zearalenone and citrinin was obtained in the first continuous development. The plate was removed from the chamber and air-dried prior to making the quantitative measurement of the three separated toxins. The other toxins remained close to the origin. After a second and third development, ochratoxin A was separated sufficiently to be determined. A fourth development enabled penicillic acid, patulin and luteoskyrin to be determined. For the separation of aflatoxins, still remaining close to the origin, a second, more polar, solvent system was used. After three continuous developments with this new solvent system, the six aflatoxins were completely separated. TABLE 7.32 Development stages and spectroscopic detection of mycotoxins by HPTLC Adapted from ref. 128. Development stageX
Time Mycotoxin (min) separated
methods
for
the
Spectral characteristic used for detectionX
Tol-EtOAc-FA (30:6:0.5) 5.0 Sterigmatocystin 1st development Zearalenone Citrinin 2nd development 5 . 0 No measurement 3rd development 6.0 Ochratoxin A 4th development
used
6.0 Penicillic acid
Patulin Luteoskyrin
Tol-EtOAC-FA (30:14:4.5) 5th development 8.0 No measurement 6th development 8.0 No measurement 8.0 Aflatoxins B1, B2, 7th development G1, G2, MI and M2
Ref. Flu.,, Flu.,,
at 324 nm at 313 nm at 460 nm
Flu.,, Flu.,, Ref. Ref. Ref.
at at at at at
Flu.,, Flu.,,
at 365 nm at 430 nm
313 460 240 280 440
nm nm nm nm nm
Abbreviations: Tol, toluene: EtOAc, ethyl acetate: FA, formic acid; Ref., reflectance; Flu., fluorescence: em, emission: ex, excitation.
228
At each scanning stage, the migration distance of the spot to be measured was maintained between 1 and 3 cm. Only patulin and luteoskyrin slightly overlapped each other, but as patulin does not show any absorption at the absorption maximum for luteoskyrin (440 nm), this was no problem. Hence the method described is capable of providing good resolution of complex mycotoxin mixtures. However, the authors used standard mycotoxin solutions and did not show whether comparable results could be obtained with samples extracted from natural commodities. HPTLC and reversed-phase TLC of 10 mycotoxins (ochratoxin A, aflatoxins Bl, B , G1 and G2, zearalenone, sterigmatocystin, T-2 toxin, diacegoxyscirpenol and vomitoxin) with the use of various normal- and reversed-phase solvents and UV detection were reported by Stahr and Domoto (516). Golinski and Grabarkiewicz-Szczesna (517) published chemical confirmatory tests for ochratoxin A, citrinin, penicillic acid, sterigmatocystin and zearalenone that are performed directly on TLC plates. Later Grabarkiewicz-Szczesna et a l . (518) reported a multi-detection procedure for the determination of 11 mycotoxins in cereals. A simultaneous TLC detection of aflatoxin B1 and zearalenone in mixed feed for pigs was described by Fulgeira and de Bracelenti (519). A quantitative TLC method for the analysis of aflatoxins, ochratoxin A, zearalenone, T-2 toxin and sterigmatocystin in foodstuffs was published by Tapia (520). Detections of Fusarium moniliforme toxins (493) and toxigenic Fusarium isolates (521) have been reported. A simple screening method for moulds producing the intracellular mycotoxins brevianamide A, citreoviridin, cyclopiazonic acid, luteoskyrin, penitrem A, roquefortine c, sterigmatocystin, verruculogen, viomellein and xanthomegnin was developed by Filtenborg et a l . (522). After removing an agar plug from the mould culture, the mycelium on the plug is wetted with a drop of methanol-chloroform (1:2). By this treatment the intracellular mycotoxins are extracted within a few seconds and transferred directly to a TLC plate by immediately placing the plug on the plate while the mycelium is still wet. After removal of the plug, known TLC procedures are carried out. The same procedure was applied to detect aflatoxins, ochratoxin A, citrinin, patulin and penicillic acid in solid substrates. In most screening procedures, extraction and clean-up techniques are applied prior to the TLC analysis. Krivobok et a l . (523) described rapid and sensitive methods for detecting toxigenic fungi producing aflatoxins, ochratoxin A, sterigmatocystin, patulin, citrinin, penicillic acid and zearalenone. The toxin-producing moulds tested produced detectable amounts of their respective mycotoxins within 2-4 days of incubation in a liquid medium. Sterigmatocystin had to be extracted from the mycelium and the rapid roduction of zearalenone needed to be temperature programmed (24I3C for growth
229
and 10°C for toxin production). Detection of the toxins by means of TLC was possible without extraction of the medium or after extraction without purification. The sensitivity of TLC detection and the recovery after extraction were good. An extraction, purification and separation diagram for mycotoxins from contaminated food was proposed by Hadidane et al. (524). Non-oleaginous and oleaginous solid samples and also oils were analysed. Four solvent systems and four detection techniques were used. Data for four aflatoxins, citrinin, and four Fusarium toxins were reported. A modification of these techniques and its use in monitoring and identification of fungal toxins in food products, animal fed and cereals in Tunisia was published more recently by Bacha et al. (525). The Fusariummycotoxins most frequently encountered in corn and often implicated in the natural causes of mycotoxicoses include: zearalenone, zearalenols, some trichothecenes and moniliformin. A multi-mycotoxin method for Fusarium isolates from corn kernels or tissues was published by Bottalico et al. (526). The isolates were grown on autoclaved corn kernels at 27OC for 4 weeks. Then the cultures were dried at 6OoC and finely ground. Samples (50 9 ) of dried corn (kernels or vegetative parts) or dried Fusarium cultures (20 9) were extracted with methanol-aqueous NaC1, defatted with hexane, and partitioned with dichloromethane. After the evaporation of the solvent, the residue was brought up to 2 mL with methanol-water (40:60), passed through a Sep-Pak C-18 cartridge, and eluted with a new portion (2 mL) of the methanol-water mixture representing the first pure fraction (fraction A ) . Further elution with methanol (2 x 2 mL) yielded fraction B. The two fractions were separately evaporated to near dryness and reconstituted with methanol (0.5 mL). Fraction A was examined for nivalenol, fusarenone, deoxynivalenol, 15-acetyldeoxynivalenol; and 3-acetyldeoxynivalenol, and fraction B was examined for diacetoxyscirpenol, T-2 toxin, zearalenone, and zearalenols (a and p ) . Analyses of zearalenone and trichothecenes were performed by TLC and GLC. Trichothecenes eluted in fraction A, as well as zearalenols, were confirmed and quantitated by HPLC. The separation of 3-acetyldeoxynivalenol and 15-acetyldeoxynivalenol was only possible by TLC or capillary GLC, and not by HPLC. Due to the low recovery of the extraction procedure for polar trichothecenes, particularlynivalenol, a HPLC method for nivalenol was used in few cases. Analysis of moniliformin was carried out in accordance with the method previously employed (527). Chakrabarti and Ghosal (528) used PLC in a study of the occurrence of free and conjugated 12,13-epoxytrichothecenes and zearalenone in banana fruits infected with Fusarium moniliforme. TLC was used by Mirocha et al. (529) to detect mycotoxin production by Fusarium oxysporum and F. sporotrichoides isolated
230
from Baccharis spp. from Brazil (known to produce macrocyclic trichothecenes). Physical and chemical properties that may be used to determine the purity of several Fusarium mycotoxins have been investigated by Bennett and Shotwell (530). A combination of analytical procedures, which include HPTLC, liquid chromatography, gas chromatography, gas chromatography/mass spectrometry, ultraviolet spectrometry, and nuclear magnetic resonance spectrometry have been used to examine mycotoxin standards obtained from commercial sources and from laboratory fermentations. Data for the following mycotoxins were reported by these workers: a-zearalenol, f3-zearalenol, deoxynivalenol, T-2 toxin, HT-2 toxin, diacetoxyscirpenol, neosolaniol, nivalenol and fusarenone X. 7.21 TLC IN CHEMOTAXONOMIC STUDIES OF TOXIGENIC FUNGI Multi-mycotoxin TLC has been introduced in chemotaxonomic studies of penicillia and other micromycetes. In addition to mycotoxins, other secondary metabolites have also been detected in such studies. The development of the methodology is given in this section. In addition to several physiological criteria, the pattern of extracellular metabolites (mostly mycotoxins) after TLC was used by Frisvad (531) as a chemotaxonomic criterion in identification of common asymmetric penicillia. Mycotoxin analyses were performed using an "agar plug methodmm.Agar plugs were cut out at the border of a colony grown on an agar medium with a flamed stainless-steel tube (inner diameter 0.4 cm). The plugs were placed directly on the pre-coated TLC plates and were removed after 10 s , and the application spots were allowed to dry. After spotting toxin standards (patulin, ochratoxin A, citrinin, penicillic acid, griseofulvin and penitrem A), the plates were developed in four solvent systems and each toxin was visualized in at least two ways. Later, Frisvad and Filtenborg (532) reported a classification of terverticillate penicillia based on profiles of mycotoxins and other secondary metabolites. In this study, extracellular and intracellular mycotoxins and other metabolites were analysed by means of TLC. Griseofulvin was used as an external standard in all analyses. Four solvent systems were used. The TLC plates were examined before and after chemical treatment under visible light and UV light at 366 nm. Frisvad (533) included TLC analyses of extracellular (ochratoxin A and citrinin) and intracellular mycotoxins (xanthomegnin and viomellein) in a screening of groups of toxigenic Penicillium viridicatum. Profiles of primary and/or secondary metabolites have been used by the same author in classification of various Penicillium and Emericella species (534-537). More recently, Frisvad and Thrane (538) published a general
23 1
standardized method for the analysis of 182 mycotoxins and other fungal metabolites, based on HPLC and combined with TLC in two different solvent systems using R values relative to griseofulvin. Data for the 182 metabolifes may be found in their paper. These metabolites include the best known mycotoxins, penicillin G , many alkaloids, polyketides and terpenes. A similar approach has been undertaken by Paterson (539). He presented standardized TLC data in two solvent systems for secondary metabolites of Penicillium and other fungi to assist in the identification of products of Penicillium species. Of 107 metabolites detected with TLC system /toluene-ethyl were named and 27 acetate-90% formic acid (5:4:1)/ 8 0 unidentified compounds were allotted reference numbers; in the case of the metabolites detected by system 2/chloroform-acetone2-propanol (85:15:20)/ the equivalent figures were 79 and 18, respectively. A chemotaxonomic study, confirming the production of a range of important mycotoxins by certain species of Penicillium, was reported by El-Banna et al. (540). One thousand four hundred Penicillium isolates were identified according to Pitt's classification. To confirm which species produce which mycotoxins, representative isolates were investigated for the synthesis of 18 mycotoxins. Isolates were grown on malt extract agar incubated for one to three weeks at 25OC. Thereafter the medium was extracted with chloroform, and the filtered, concentrated extracts used for mycotoxin analysis by TLC. The production of any particular mycotoxin was confirmed by using external standards in optimal developing systems with toxins visualized by the best visualizing methods. The following 18 mycotoxins investigated were produced by one or more Penicillium species: brevianamid A , citreoviridin, citrinin, cyclopiazonic acid, fumitremorgin B, griseofulvin, luteoskyrin, ochratoxin A , patulin, penicillic acid, penitrem A , PR toxin, roquefortine, rugulosin, verrucosidin, verruculogen, viridicatumtoxin and xanthomegnin. Procedures for the detection of the mycotoxins used in this work were described and included adsorbents, solvent systems, detection methods and colours of the mycotoxins after treatment. Mycotoxin production by the various species of Penicillium is quite distinctive and may be used as a valuable aid in their identification. 7.22 CONCLUSIONS
This chapter was written with the aim of demonstrating the scope of applications of TLC in the still developing field of mycotoxins. The "mycotoxin era" had its origins in the early sixties when the gradual decline of applications of paper chromatography was due to the rapid development of TLC. Hence, applications of PC in mycotoxicology are now interesting mostly
232
from a historical point of view and only some typical examples were mentioned in the Introduction. TLC is by far the most widely used chromatographic technique applied to mycotoxins owing to its relatively simple, fast and inexpensive character. As in most instances the mycotoxins to be analysed or purified by means of TLC are present in contaminated samples, they must be extracted and cleaned up prior to TLC if reliable results are to be obtained. Extraction procedures, reviewed in this chapter, include extractions of mycotoxins from feeds and foodstuffs, cultivation media and/or mycelia of toxigenic fungi. Extraction solvents include chloroform, methylene chloride, ethyl acetate, acetone, acetonitrile, methanol and their combinations. Clean-up procedures include CC (mostly using silica gel columns), gel-permeation chromatography, liquid-liquid partition and precipitation techniques. In these procedures, contaminating lipids, fatty acids, proteins and various pigments are mostly removed from the mycotoxin samples. Silica gel is the most commonly used adsorbent in the TLC of mycotoxins. With acidic toxins, better results are obtained when the silica gel plates are pre-treated with oxalic acid, tartaric acid or EDTA. Chemically bonded reversed-phase layers can be used in special applications. The variety of solvent systems is enormous. The most often used solvents combined in various ratios include benzene, chloroform, toluene, ethyl acetate, methylene chloride, acetone, methanol, formic acid and acetic acid. The detection techniques vary with the mycotoxins to be detected. Coloured toxins are examined under visible light, and fluorescent ones are revealed under short- and/or long-wave UV light. Colourless and non-fluorescent compounds can be detected by means of appropriate spray reagents producing colours or fluorescence. Bioautographic detections have also been described, using mostly Artemia salina larvae or microbial cultures. In addition to the classical one-dimensional TLC, two-dimensional TLC and HPTLC have been used by many researchers. With HPLC and in quantitations, TLC becomes more expensive owing to the need of densitometers and spectrophotometers. PLC has been used in the initial preparation of several mycot.oxins belonging to the aflatoxins, cytochalasans, hydroxyanthraquinones, indole-derived tremorgens, zearalenone and its derivatives, etc. The reviewed applications of the TLC of aflatoxins, sterigmatocystins and other aflatoxin intermediates, ochratoxins, rubratoxins, small lactones, trichothecenes, cytochalasans, tremorgenic mycotoxins, hydroxyanthraquinones, epipolythiopiperazine-3,6-diones, zearalenones, citrinin, a-cyclopiazonic cid, secalonic acids, PR toxin, roquefortine, xanthomegnin, viomellein, naphthopyrones and some peptidic mycotoxins emphasize the great importance of thin-layer
233
chromatography in the relatively young field of mycotoxicology. However, other chromatographic techniques may be useful in such instances where TLC provides insufficient results. Examples may be found in chapters 8 and 9 of the present book. REFERENCES 1 K. Sargeant, A. Sheridan, J. OmKelly and R.B.A. Carnaghan, Nature (London), 192 (1961) 1095. 2 V. Betina (Editor), Mycotoxins - Production, Isolation, Separation and Purification, Elsevier, Amsterdam, 1984. 3 Y. Ueno, in V. Betina (Editor), Mycotoxins - Production, Isolation, Separation and Purification, Elsevier, Amsterdam, 1984, Ch. 15, p. 329. 4 Y. Ueno, in V. Betina (Editor), Mycotoxins - Production, Isolation, Separation and Purification, Elsevier, 1984, Ch. 24, p. 475. 5 V. Betina, J. Chromatogr., 477 (1989) 187. 6 J. Miyazaki, K. Omachi and T. Kamata, J. Antibiot., 6 (1953) 6. 7 J. Uri, Nature (London), 183 (1959) 1188. 8 P. Nemec, V. Betina and L. Kovaeieova, Folia Microbiol., 6 (1961) 277. 9 V. Betina, Nature (London), 182 (1958) 796. 10 V. Betina and P. Nemec, Nature (London), 187 (1960) 1111. 11 V. Betina, Chromatogr. Rev., 7 (1965) 119. 12 V. Betina, Methods Enzymol., 43 (1975) 100. 13 V. Betina, Chem. Zvesti (Bratislava), 15 (1961) 750. 14 V. Betina, Chem. Zvesti (Bratislava), 15 (1961) 859. 15 V. Betina, J. Chromatogr., 15 (1964) 379. 16 V. Betina, P. Nemec, M. Kutkova, J. Balan and 5 . Kovae, Chem. Zvesti (Bratislava), (1964) 128. 17 V. Betina, Antimicrobial Agents and Chemotherapy 1966, American Society for Microbiology, Washington, 1967, p. 637. 18 B.F. Nesbitt, J. O'Kelly, K. Sargeant and A. Sheridan, Nature (London), 195 (1962) 1062. 19 R.D. Hartley, B.F. Nesbitt and J. OIKelly, Nature (London), 198 (1963) 1056. 20 R.J. Cole (Editor), Modern Methods in the Analysis and
-
21 22 23 24 25
Structural Elucidation of Mycotoxins, Academic Press, New York, 1986. S. Nesheim and M.W. Trucksess, in R.J. Cole (Editor), Modern Methods in the Analysis and Structural Elucidation of Mycotoxins, Academic Press, New York, 1986, p. 239. M.J. Shepherd, in R.J. Cole (Editor), Modern Methods in the Analysis and Structural Elucidation of Mycotoxins, Academic Press, New York, 1986, p. 293. D. Rogers, Int. Lab., June (1985) 12. C.P. Gorst-Allman and P.S. Steyn, in V. Betina (Editor), - Production, Isolation, Separation and Mycotoxins Purification, Elsevier, Amsterdam, 1984, Ch. 9, p. 59. L.S. Lee and D.B. Skau, J. Liq. Chromatogr., 4 (Suppl. 1) (1981) 43.
234 26 P.M. Scott, Adv. Thin Layer Chromatogr. (Proc. 2nd Bienn. Symp. 1980), 1982, p. 321. 27 Y. Ueno (Editor), Trichothecenes Chemical, Biological and Toxicological Aspects, Elsevier, Amsterdam, 1983. 28 Official Methods of Analysis of the Association of Official Analytical Chemists, AOAC, Arlington, VA, 14th ed., 1984, Ch. 26. 178 29 P. Majerus and R. Wollen, Z. Lebensm.-Unters.-Forsch., (1984) 79. 30 V. Betina, J. Chromatogr., 334 (1985) 211. 31 D.S.P. Patterson, in H.J.M. Bowen (Editor), Specialist Periodical Reports, Environmental Chemistry, Vol. 2, Mycotoxins, Royal Society of Chemistry, London, 1982, p. 183. 32 K.H. Nahm and K.S. Nahm, Han'guk Ch'uksan Hakhoe Chi, 28 (1986) 426; C.A., 106 (1987) 1 7 4 7 1 9 ~ . 33 P.M. Scott, J. Assoc. Off. Anal. Chem., 69 (1986) 240. 34 P.M. Scott, J. Assoc. Off. Anal. Chem., 71 (1988) 70. 35 N.V. Howell and P.W. Taylor, J. Assoc. Off. Anal. Chem., 64 (1981) 1356. 36 S.R. Tonsager, D.A. Maltby, R.J. Schock and W.E. Braselton, Adv. Thin Layer Chromatogr. (Proc. Bienn. Symp.), 2nd, 1980, (1982) 389. 37 P.M. Scott, J. Assoc. Off. Anal. Chem., 72 (1989) 75. 38 N. Reichert, S. Steinmeyer and R. Weber, Z . Lebens. -Unters. -Forsch., 186 (1988) 505. 39 J.W. Dickens and T.B. Whitaker, in R.J. Cole (Editor), Modern
-
Methods in the Analysis and Structural Elucidation of Mycotoxins, Academic Press, New York, 1986, Ch. 2, p. 29. 4 0 J.W. Dickens and T.B. Whitaker, in H. Egan, L. Stoloff, P. Scott, M. Castegnaro, I.K. O'Neil and H. Bartsch (Editors), Environmental Carcinogens - Selected Methods of Analysis. vol. 5:Some Mycotoxins, ARC, Lyon, 1982, p. 17. 41 D.L. Park and A.E. Pohland, J. Assoc. Off. Anal. Chem., 72 (1989) 399. 4 2 K. Saito, M. Nishijima, K.
Shokuhin Eiseigaku
Yasuda, H. Kamimura and A. Ibe Zasshi, 25 (1984) 112; C.A., 101 (1984
149940t. 43 V.K. Mehan, T.H. Bhavanishankar and J.S. Bedi, J. Food Sci Technol., 22 (1985) 123. 44 Official Methods of Analysis, 15th ed., AOAC, Arlington, VA 1990, Ch. 49. 45 N. Bradburn, R.D. Coker, K. Jewers and K.I. Tomlins, Chromatographia, 29 (1990) 435. 46 J.G. Heathcote and J.R. Hibbert, Aflatoxins: Chemical and Biological Aspects, Elsevier, Amsterdam, 1978, Ch. 4, p. 54. 47 J.G. Heathcote, in V. Betina (Editor), Mycotoxins
-
Production, Isolation, Separation and Purification, Elsevier, Amsterdam, 1984, Ch. 7, p. 89. Touchstone and D. Rogers (Editors), Thin Layer 4 8 J.C. Chromatography: Quantitative, Environmental and Clinical Applications (Proceedings of 1979 Symp. ) , Wiley, New York, 1980. 49 H.S. Stahr, in J.C.
Touchstone and D. Rogers (Editors), Thin Layer Chromatography: Quantitative, Environmental and Clinical Applications (Proceedings of 1979 Symp.), Wiley, New York, 1980, p. 177.
235 50 S. Nesheim, in J.C.
Touchstone and D. Rogers (Editors), Thin Layer Chromatography: Quantitative, Environmental and Clinical Applications (Proceedings of 1979 Symp.), Wiley, New York, 1980, p. 194. 51 J.A. Miguel, V. de Andres and M. Carballo, Invest. Agric. Prod. Prot. Veget., 2 (1987) 305. 52 K. Jewers, R.D. Coker, B.D. Jones, J. Cornelius, M.J. Nagler, N. Bradburn, K. Tomlins, V. Medlock, P. Dell, G. Blunden, O.G. Roch and J. Sharkey, J. Appl. Bacteriol. Symp. Suppl., 1989, 105s. 53 J.G. Heathcote and J.R. Hibbert, J. Chromatogr., 108 (1975) 131. 54 P.M. Scott, J. Assoc. Off. Anal. Chem., 51 (1968) 609. 55 P.M. Scott, J. Assoc. Off. Anal. Chem., 52 (1969) 72. 56 P.J. Schuller, W. Horwitz and L. Stoloff, J. Assoc. Off. Anal. Chem., 59 (1976) 1315. 57 N. Miller, H.E. Pretorius and D.W. Trinder, J. Assoc. Off. Anal. Chem., 68 (1985) 136. 58 C.E.A. Lovelace, H. Njapau, L.F. Salter and A.C. Bayley, J. Chromatogr., 227 (1982) 256. 59 S.V. Pathre, C.J. Mirocha and S.W. Fenton, J. Assoc. Off. Anal. Chem., 62 (1979) 1268. 60 H.P. van Egmond, W.E. Paulsch, E.A. Sizzo and P.L. Schuller, Report No. 152/79 LCO, Rijksinstituet voor Volksgesondheid, 1979. 6 1 R.M. Eppley, J. Assoc. Off. Anal. Chem., 51 (1968) 74. 6 2 B.G.E. Josefsson and F.E. M ller, J. Assoc. Off. Anal. Chem., 60 (1977) 1369. 63 Y. Takeda, E. Isohata, R. Amano and M. Uchiyama, J. Assoc. Off. Anal. Chem., 6 2 (1979) 573. 64 M.P. Whidden, N.N. Davis and B.A. Roberts, J. Agric. Food Chem., 28 (1980) 784. 6 5 J. Velasco, J. Amer. Oil Chem. SOC., 58 (1981) 938A. 66 Anonymous, in K.C. Vanhorne (Editor), Sorbent Extraction 67 68 69 70 71 72 73
Technology, Analytichem. International Institute, Harbor City, USA, 1985. M.W. Trucksess, W.C. Brumley and N. Stanley, J. Assoc. Off. Anal. Chem., 67 (1984) 973. J.E. Thean, E.R. Lorenz, D.M. Wilson, K. Rodgers and R.C. Gueldner, J. Assoc. Off. Anal. Chem., 63 (1980) 631. N. Takeda, J. Chromatogr., 288 (1984) 484. G.E. Rottinghaus, B. Olesen and G.E. Osweiler, American Association of Veterinary Laboratory Diagnosticians, 25th Annual Proceedings 1982, p. 485. G. Quan and G.C. Yang, J. Agric. Food Chem., 32 (1984) 1071. D.L. Orti, R.H. Hill, J.A. Liddle and L.L. Neelham, J. Anal. Toxicol., 10 (1986) 41. K.I. Tomlins, K. Jewers and R.D. Coker, Chromatographia, 27
(1989) 67. 74 N. Bradburn, R.D. Coker and K. Jewers, Chromatographia, 29 (1990) 177. 75 N. Bradburn, K. Jewers, B.D. Jones and 1.1. Tomlins, Chromatographia, 28 (1989) 541. 76 H. Kamimura, M. Nishijima, K. Yasuda, H. Ushiyama, S . Tabata, S . Matsumoto and T. Nishima, J. Assoc. Off. Anal. Chem., 68 (1985) 458.
236 77 K. Jewers, A.E. John and G. Blunden, Chromatographia, 27 (1989) 617. 78 H.P. van Egmond and P.J. Wagstaffe, Food Addit. Contamin., 7 (1990) 239. 79 H. De Iongh, R.H. Beurthuis, R.O. Vles, C.B. Barrett and W.D. Ort, Biochim. Biophys. Acta, 65 (1962) 548. 80 D. Miskovic, Chromatographia, 13 (1980) 342. 81 R.P. Kozloski, J. Assoc. Off. Anal. Chem., 64 (1981) 1263. 82 H.J. Issaaq and W. Cutchin, J. Liq. Chromatogr., 4 (1981) 1097. 83 N.F. Dutton and J.G. Heathcote, J. S. Afr. Chem. Inst., 22 (1969) S107. 84 H. De Ionah, J.C. van Pelt, W.O. Ord and C.B. Barrett, Vet. Rec. , 76 71964) 901. 85 S. Hara, D.I. Fennel1 and C.W. Hesseltine, Appl. Microbiol., 27 (1974) 1118. 86 M. Gareis, J. Bauer, A von Montgelas and B. Gedek, Appl. Environ. Microbiol., 47 (1984) 416. 87 S. Nesheim, J. Assoc. Off. Anal. Chem., 47 (1964) 586. 88 J.A. Robertson, Jr., L.S. Lee, A.F. Cucullu and L.A. Goldblatt, J. Amer. Oil. Chem. SOC., 42 (1965) 467. 89 M.R. Heusinkveld, C.C. Shera and F.J. Baur, J. Assoc. Off. Anal. Chem., 48 (1965) 448. 90 W.A. Pons and L.A. Goldblatt, J. Amer. Oil Chem. SOC., 42 (1965) 471. 91 W.A. Pons, Jr., A.A. Cucullu, L.S. Lee, R.Y. Mayne and L.A. Goldblatt, J. Assoc. Off. Anal. Chem., 49 (1966) 554. 92 A.F. Cucullu, L.S. Lee, R.Y. Mayne and L.A. Goldblatt, J. Am. Oil. Chem. SOC., 43 (1966) 89. 93 R.M. Eppley, J. Assoc. Off. Anal. Chem., 49 (1966) 1218. 94 A.E. Waltking, J. Assoc. Off. Anal. Chem., 53 (1970) 104. Shotwell, G.M. Shannon, D.W. 95 R.D. Stubblefield, O.L. Weisleder and W.K. Rohwedder, J. Agric. Food Chem., 18 (1970) 391. 96 J. Velasco, J. Assoc. Off. Anal. Chem. 58 (1975) 757. 97 L.J. Vorster, Analyst (London), 94 (1969) 136. 98 J. Velasco and S.L. Morris, J. Agric. Food Chem., 24 (1976 86. 99 Official Methods of Analysis, 12th ed., AOAC, Arlington, VA 1976, Ch. 26. 100 W.A. Pons, Jr., A.F. Cucullu, A.O. Franz and L.A. Goldblatt J. Amer. Oil Chem. SOC., 45 (1968) 694. 101 L. Stoloff, A. Graff and H. Rich, J. ~ s s o c .off. Anal. Chem., 49 (1966) 740. 102 W.A. Pons, Jr., J. Assoc. Off. Anal. Chem., 52 (1969) 61. 103 W.A. Pons, Jr., J. Assoc. Off. Anal. Chem., 58 (1975) 746. 104 J. Velasco, J. Amer. oil Chem. soc., 49 (1972) 141. 105 G.M. Shannon, R.D. Stubblefield and O.L. Shotwell, J. Assoc. Off. Anal. Chem., 56 (1973) 1024. 106 M.T. CutUli de Simon and F.G. Suarez, Ann. Bromatol., 35 (1983, Publ. 1984) 93; C.A., 101 (1984) 7 1 1 5 7 ~ . 107 T.A. Gbodi, N. Nwude, Y.O. Aliu and C.O. Ikediobi, Food Chem. Toxicol., 24 (1986) 339. 108 H. Gulyas, J. Chromatogr., 319 (1985) 105. 109 Ya.L. Kostyukovskii, D.B. Melamed and M.F. Nesterin, Prikl. Biokhim. Mikrobiol., 17 (1981) 759.
237 110 L.B. Bullerman, P.A. Hartman and J.C. Ayres, J. Assoc. Off. Anal. Chem., 52 (1969) 638. 111 G.M. Shannon, O.L. Shotwell and W.F. Kwolek, J. Assoc. Off. Anal. Chem., 66 (1983) 582. 112 L.F.H. Purchase and M. Steyn, J. Assoc. Off. Anal. Chem., 50 (1967) 363. 113 M.S. Masri, J.R. Page and V.C. Garcia, J. Assoc. Off. Anal. Chem., 51 (1968) 594. 114 C.E.A. Patterson and B.A. Roberts, Food Cosmet. Toxicol., 13 (1975) 541. 115 J.I. Teng and P.C. Hanzas, J. Assoc. Off. Anal. Chem., 52 (1969) 83. 116 K.L. Hanna and T.C. Campbell, J. Assoc. Off. Anal. Chem., 51 (1968) 1197. 117 J. Velasco, J. Amer. Oil Chem. SOC., 46 (1969) 105. 118 J.H. Broadbent, J.A. Cornelius and G. Shone, Analyst (London), 88 (1963) 214. 119 R.D. Stubblefield, G.M. Shannon and O.L. Shotwell, J. Assoc. Off. Anal. Chem., 52 (1969) 669. 120 M.D. Grove, R.D. Plattner and R.E. Peterson, Appl. Environ. Microbiol., 48 (1984) 887. 121 U. Klemm, GIT Fachz. Lab. (Suppl. Chromatogr.), 9 (1982) 12; C.A., 97 (1982) 1802904. 122 R.D. Coker, K. Jewers, K.I. Tomlins and G . Blunden, Chromatographia, 25 (1988) 875. 123 K.I. Eller, L.V. Maksimenko and V.A. Tutelyan, Vopr. Pitan., No. 6 (1982) 62. 124 L. Allen, J. Assoc Off. Anal. Chem., 57 (1974) 1398. 125 R.J. Alexander and M.C. Baur, Cereal Chem., 54 (1977) 699. 126 D.S.P. Paterson, E M. Glancy and B.A. Roberts, Food Cosmet. Toxicol., 16 (1978 49. 127 J.W. Rodricks and L. Stoloff, J. Assoc. Off. Anal. Chem., 53 (1970) 92. 128 K.Y. Lee, C.F. Po0 e and A. Zlatkis, Anal. Chem., 52 (1980) 837. 129 J. Ripphahn and Halpaar, J. Chromatogr., 112 (1975) 81. 130 Z. buraekova, V. Betina and P. Nemec, J. Chromatogr., 116 (1976) 141. 131 2 . Jesenska, M. Polster, J. Matyakova and 0. Polakova, Zbl. Bakt. Hyg., I. Abt. Orig. B, 171 (1980) 408. 132 Official Methods of Analysis, 13th ed., AOAC, Washington, 1980, Sect. 26.051. 133 H.P. van Egmond, A.B. Leussik and W.E. Paulsch, Int. Dairy Fed. Bull., No. 207 (1986) 150. 134 Z. JhraCkova, V. Betina and P. Nemec, J. Chromatogr., 116 (1976) 155. 135 V. Betina, Mycotoxins Chemical, Biological and Environmental Aspects, Elsevier, Amsterdam, 1989, p. 120. 136 R . Singh and D.P.H. Hsieh, Arch. Biochem. Biophys., 178 (1977) 285. 137 T. Yanagita, T. Sakai, K. Ageishi, H. Uesora and S . Moriya, J. Gen. Appl. Microbiol., 23 (1977) 261. 138 J.W. Bennett, J. Gen. Microbiol., 113 (1979) 127. 139 G. Clevstrom, B. Goransson, B. Hlodversson and H. Pettersson, J. Stored Prod. Res., 17 (1981) 151.
-
238 140 D.T. Wicklow and O.L. Shotwell, Can. J . Microbiol., 29 (1983) 1. 141 T. Kachholz and A.L. Demain, J. Nat. Prod., 46 (1983) 499. 142 A. Sharma, S.R. Padwal-Desal and G.B. Nadkarni, Appl. Environ. Microbiol., 49 (1985) 79. 143 R. Valcarcel, J.W. Bennett and J. Vitanza, Mycopathologia, 94 (1986) 7. Moss and F. Badii, Appl. Environ. Microbiol., 43 144 M.O. (1982) 895. 145 F. Badii, M.O. Moss and K. Wilson, Lett. Appl. Microbiol., 2 (1986) 61. 146 R.L. Buchanan, S.B. Jones and H.G. Stahl, Mycopathologia, 100 (1987) 135. 147 R.L. Buchanan, S . B . Jones, W.V. Gerasimowicz, L.L. Zaika, H.G. Stahl and L.A. Ocker, Appl. Environ. Microbiol., 53 (1987) 1224. 148 J. Coallier-Ascah and E.S. Idziak, Appl. Environ. Microbiol., 49 (1985) 163. 149 D.T. Wicklow, B.W. Horn and O.L. Shotwell, Mycologia, 79 (1987) 679. Davis, S.K. Iyer and U.L. Diener, Appl. Environ. 150 N.D. Microbiol., 53 (1987) 1393. 151 J. Leitao, G. de Saint Blanquart, J.R. Reilly and Ch. Paillas, J. Chromatogr., 435 (1988) 229. 152 M.F. Dutton, K. Ehrlich and J.W. Bennett, Appl. Environ. Microbiol., 49 (1985) 1392. 153 T.E. Cleveland, A.R. Lax, L.S. Lee and D. Bhatnagar, Appl. Environ. Microbiol., 53 (1987) 1711. 154 D. Bhatnagar, S.P. McCormick, L.S. Lee and R.A. Hill, Appl. Environ. Microbiol. 53 (19879 1028. 155 C.A. Townsend, S.B. Christensen and S.G. Davis, J . Chem. SOC., Perkin Trans., I, (1988) 839. 156 A. Henderberg, J.W. Bennett and L.S. Lee, J . Gen. Microbiol., 134 (1988) 661. 157 D. Bhatnagar, T.E. Cleveland and E.B. Lillehoj, Mycopathologia, 107 (1989) 75. 158 H.H.L. Chang, J.W. de VrieS and W.E. Hobbs, J. Assoc. Off. Anal. Chem., 62 (1979) 1281. 159 D.L. Park, M.W. Trucksess, S. Nesheim and M. Stack, Abstracts lOlst AOAC Annual International Meeting, September 14-17, 1987, San Francisco, CAI p. 66. 160 C.M. Sylos and D.B. Rodriguez-Amaya, J . Sci. Food Agric., 49 (1989) 167. 161 D.P. Hsieh, M.A. Fukayama, D.W. Rice and J . J . Wong, in ref. 48, p. 241. 162 Ya.L. Kostyukovskii and D.B. Melamed, Zh. Anal. Khim., 39 (1984) 2229. 163 K.K. Sinha, Appl. Environ. Microbiol., 53 (1987) 1391. 164 T.M. Zemnie, J. Liq. Chromatogr., 7 (1984) 1383. 165 A.V. Jain and R.C. Hatch, in J.C. Touchstone (Editor),
Advances in Thin Layer Chromatography (Proceedings of 2nd Bienn. 1980 Symp.), Wiley, New York, 1982, p. 363. 166 H.P. van Egmond, W.E. Paulsch and E.A. Sizoo, Food Addit. Contam., 5 (1988) 321. 167 R.P. Kozloski, Bull. Environ. Contam. Toxicol., 36 (1986)
815. 168 A. Simonella, L. Toretti, C. Fillipponi and L. Ambrosii, J . High. Resolut. Chromatogr., Chromatogr. Commun., 10 (1987) 626.
239 169 A.G. Chakrabarti, Bull. Environ. Contam. Toxicol., 33 (1984 515. 170 B. Le Tutour, A. Tantauci-Elaraki and A. Aboussalim, J. Assoc Off. Anal. Chem., 67 (1984) 611. 171 J. Re s s , Food Cosmet. Toxicol., 13 (1975) 325. 172 X. Li, Gongye Weishengwu, 18 (1988); C.A., 110 (1989) 133789a. 173 M.W. Trucksess, A.F. Cucullu, L.S. Lee and A.O. Franz, J. Assoc. Off. Anal. Chem., 60 (1977) 795. 174 M.K.L. Bicking, R.N. Kniseley and H.J. Svec, Anal. Chem., 55 (1983) 200. 175 R.D. Stubblefield, W.F. Kwolek and L. Stoloff, J. Assoc. O f f . Anal. Chem., 65 (1982) 1435. 176 G.E. Neal and P.J. Colley, FEBS Lett., 101 (1979) 382. 177 D. Tikhova and V. Peneva, Khranit. Promst., 33 (1984) 20. 178 M. Serralheiro and M.L. Quinta, J. Assoc. Off. Anal. Chem., 69 (1986) 886. 179 0. Destro and L. Gelosa, Ind. Aliment. (Pinerolo), 25 (1986) 868; C.A., 106 (1987) 100922~. 180 J.P. Bijl, C.-H. Peteghem and D.A. Dekeyser, J. Assoc. Off. Anal. Chem., 70 (1987) 472. 181 L. Dominiquez, L.J. Blanco, E. Gomez-Lucia, E.F. Rodriguez and G. Suarez, J. Assoc. Off. Anal. Chem., 70 (1987) 470. 182 S . Quintavalla and A. Casolari, Ind. Conserve, 60 (1985) 85. 183 V.I. Mochalov, L.N. Semenova and L.A. Eremina, Moloch. Promst., 11 (1981); C.A., 96 (1982) 50808~. 184 V.A. Tutel'yan, K.I. Eller, N.V. Rubakova, J.C.V. Correa and V.G. Stoyanova, Zh. Anal. Khim., 42 (1987) 2062. 185 E.M. Kubicek, F.K. Vojir and H.G. Hulzer, Ernahrung/Nutrition, 12 (1988) 302. 186 R.J. Cole and R.H. Cox, Handbook of Toxic Fungal Metabolites, Academic Press, New York, 1983. 187 W. Koch and W. Kroes, Wehrmed. Monatschr., 30 (1986) 205; C.A., 105 (19861 7767n. 188 P. Lafont, 'M.G..Siriwardana, J. Sarfati, J.P. Debeau uis and C.A., J. Lafont, Microbiol. Aliment. Nutr., 4 (1986) 141 105 (1986) 170705n. 189 B. Blanc, E. Lauber and R. Sieber, Microbiol. A iment Nutr., 1 (1983) 163. 190 G . Cirilli, Microbiol. Aliment. Nutr., 1 (1983) 199. 191 S . Bauer, D. Schindler and J. Kroll, Nahrung, 33 (1989) 217; C.A., 110 (1989) 230266k. 192 R.D. Stubblefield, J. Assoc. Off. Anal. Chem., 70 (1987) 1047. 193 A. de Jesus, C.P. Gorst-Allman, R.M. Horak and R. Vleggaar, J. Chromatogr., 450 (1988) 101. 194 G.E. Neal, Chem. Ind. (London), (1984) 542. Nesheim and W.C. Bramley, J. Amer. Oil Chem. SOC., 58 195 S. (1981) 945A. 196 H.A. Koch, J. Weisser and E. Skuras, Pat. Appl.; C.A., 104 (1986) 67499d. 197 C.P. Gorst-Allman and P.S. Steyn, J. Chromatogr., 175 (1979) 325. 198 M. Steyn and C.J. Rabie, J. Assoc. Off. Anal. Chem., 58 (1975) 622. 199 R . F . Vesonder and B.W. Horn, Appl. Environ. Microbiol., 49 (1985) 234. 200 D. Abramson and T. Thorsteinson, J. Assoc. Off. Anal. Chem., 72 (1989) 342.
.
240
R.W. My lymaki, D.E. Netleton, Jr. and F.A. O'Heron, Antimicrob. Ag. Chemother., 8 (1975) 159. 202 J.C. Floyd, J.C. Mills I11 and J W. Bennett, Exp. Mycology,
201 W.T. Bradner, J.A. Bush,
11 (1987) 109. 203 J.W. Bennett, S. Kofsky, A. Bulb n and M. Dutton, Dev. Ind. Microbiol.. 26 (1985) 479. 204 G. Sullivan, D:D. Maness, G.J. Yakatan and J. Scholler, J . Chromatogr., 116 (1976) 490. 205 H.P. van Egmond, W.E. Paulsch, E. Deijl and P.L. Schuller, J. Assoc. Off. Anal. Chem., 63 (1980) 110. 206 W. Hu, C. Thian, X. Luo and Y. Wang, Weishengwuxue Tongbao, 10 (1983) 265; C.A., 100 (1984) 1 7 3 2 3 9 ~ . 207 O.J. Francis, G.M. Ware, A.S. Carman and S . S . Kuan, J. Assoc. Off. Anal. Chem., 68 (1985) 643. 208 S. Takitani and Y. Asabe, in Y. Ueno (Editor), 209 210 211 212 213 214
Trichothecenes - chemical, Biological and Toxicologica Aspects, Elsevier, Amsterdam, 1983, p. 113. Ch. Tamm and M. Tori, in V. Betina (Editor), Mycotoxins - Production, Isolation, Separation and Purification Elsevier, Amsterdam, 1984, Ch. 8, p. 131. J. Chelkowski (Editor), Fusarium Mycotoxins, Taxonomy and Pathogenicity, Elsevier, Amsterdam 1989. J.R. Bamburg, Clin. Toxicol., 5 (1972) 495. R.M. Eppley, J. Assoc. Off. Anal. Chem., 58 (1975) 906. R.M. Eppley, J. Amer. Oil Chem. SOC., 56 (1979) 824. S. Takitani, Maikotokishin (Tokyo), 13 (1981) 24; C.A., 97
(1982) 1603p. 215 J.R. Bamburg and F.
Strong, in S . Kadis, A. Ciegler and S.J. Ajl (Editors), Microbial Toxins, Vol. 7, Academic Press, New York, 1971, p. 272. 216 E.B. Smaley and F.M. Strong, in I.F.H. Purchase (Editor), Mycotoxins, Elsevier, Amsterdam, 1974, p. 217. 217 S.V. Pathre and C.J. Mirocha, in J.V. Rodricks, C.W. Hesseltibne and M . A . Mehlman (Editors), Mycotoxins in Human and Animal Health, Pathotox, Park Forest South, IL, 1977, p. 238. 218 T.R. Romer, J. Assoc. Off. Anal. Chem, 69 (1986) 699. 219 B. Harrach, A. Bata, E. Bajmocy and M. Benko, Appl. Environ. Microbiol., 45 (1983) 1419. 220 B. Harrach, Mycotoxin Res., 4 (1988) 20. 221 J.A. Baxter, S.J. Terhune and S.A. Qureshi, J . Chromatogr., 261 (1983) 130. 222 Y. Ramakrishna and R.V. Bhat, Curr. Sci., 56 (1987) 524. 223 G.V. Rao, P.S. Rao, S . Girisham and S.M. Reddy, Curr. Sci., 54 (1985) 507. 224 Y. Ueno (Editor), Trichothecenes - Chemical, Biological and Toxicological Aspects, Elsevier, Amsterdam, 1983, Ch. IV, p. 113. 225 A. Sano, Y. Asabe, S . Takitani and Y. Ueno, J. Chromatogr., 235 (1982) 257. 226 B. Harrach, C.J. Mirocha, S.V. Pathre and M. Palyusik, Appl. Environ. Microbiol., 41 (1981) 1428. 227 B. Yagen, A. Sintov and M. Bialer, J. Chromatogr., 356 (1986) 195. 228 J. Kroll, Ch. Giersch and S. Guth, Nahrung, 32 (1988) 75. 229 P.M. Scott, J.W. Lawrence and W. van Walbeck, Appl. Microbiol., 20 (1970) 839.
24 1
230 K.C. Ehrlich and E.B. Lillehoj, Appl. Environ. Microbiol., 48 (1983) 130. 231 H. Kamimura, M. Nishijima, K. Yasuda, K. Saito, A. Ibe, T. Nagayama, H. Ushiyama and Y. Naoi, J. Assoc. Off. Anal. Chem., 64 (1981) 1067. 232 V. Betina, J. Chromatogr., 78 (1973) 41. 233 V. Betina and M. Vankova, Biologia (Bratislava), 32 (1977) 943. 234 S . Sukroongreung, K.T. Schappert and G.G. Kchachatourians, Appl. Environ. Microbiol., 48 (1984) 416. 235 H. Koshinsky, S. Honour and G. Kchachatourians, J. Chromatogr., 463 (1989) 457. 236 N. Ogur, R. St. John and S Nagai, Science (Washington), 125 (1957) 928. 237 N.C.P. Baldwin, B.W. Bycroft, P.M. Dewick, D.C. Marsh and J. Gilbert, 2. Naturforsch., Tei C, 42 (1987) 1043. 238 R.M. Eppley, M.W. Trucksess, S. Nesheim, C.W. Thorpe, G.E. Wood and A.E. Pohland, J. Assoc. Off. Anal. Chem., 67 (1984) 43. 239 J. Schultz, R. Motz, I. Kliche and K. Lehmann, Monatsh. Veterinermed., 38 (1983) 777. 240 M.W. Trucksess, S. Nesheim and R.M. Eppley, J. Assoc. Off. Anal. Chem., 67 (1984) 40. 241 M.W. Trucksess, M.T. Foold, H.M. Mossoba and S.W. Page, J. Agric. Food. Chem., 35 (1987) 445. 242 K.H. Richardson, W.M. Hagler and P.B. Hamilton, Appl. Environ. Microbiol., 47 (1984) 643. Ghosal, D.K. Chakrabarti, A.K. Srivastava and R.S. 243 S. Srivastava, J. Agric. Food Chem., 30 (1982) 106. 244 W.M. Hagler, K. Tyczkowska and P.B. Hamilton, Appl. Environ. Microbiol., 47 (1984) 151. 245 T. Yoshizawa, T. Sakamoto and K. Okamoto, Appl. Environ. Microbiol., 47 (1984) 130. 246 A. Bata, A. Vanyi and R. Lasztity, Acta Vet. Hung., 32 (1984) 51. 247 J.A. Lansden, R.J. Cole, J.W. Dorner, R.H. Cox, H.G. Cutler and J.D. Clark, J. Agric. Food Chem., 26 (1978) 246. 248 P. Lepom and H. Baath, J. Basic. Microbiol., 29 (1989) 215. 249 Y. Ueno, K. Nakayama, K. Ishii, F. Tashiro, Y. Minoda, T. Omori and K. Komegata, Appl. Environ. Microbiol., 46 (1983) 120. 250 T. Yoshizawa, T. Sakamoto and K. Kuwamura, Appl. Environ. Microbiol., 50 (1985) 676. 251 A. Visconti and C.J. Mirocha, Appl. Environ. Microbiol., 49 (1985) 1246. 252 W.C. Gordon and L.J. Gordon, J. Assoc. Off. Anal. Chem., 73 (1990) 266. 253 K.C. Ehrlich, Mycopathologia, 107 (1989) 111. 254 K.C. Ehrlich and K.W. Daigle, Biochim. Biophys. Acta, 923 (1987) 206. 255 A. Bata, J. Fekete and B. Harrach, Symp. Biol. Hung., 31 (1986) 325. 256 E. Harri, W. Loeffler, H.P. Sigg, H. Stahelin, Ch. Stoll, Ch. Tamm and D. Weisinher, Helv. Chim. Acta, 45 (1962) 840. 257 B. Bohner, E. Fetz, E. Harri, H.P. Sigg, Ch. Stoll and Ch. Tamm, Helv. Chim. Acta, 48 (1965) 1079.
242 258 B.B. Jarvis, S.N. Comezoglu,
H.L. Ammon, C.K. Breedlove, R.F. Bryan, R.W. Miller, M.K. Woode, D.R. Strelman, A.T. Sneden, R.G. Dailey and S.M. Kupchan, J. Nat. Prod., 50
(1987) 815. 259 B.B. Jarvis, S.N.
Comezoglu, M.M. Rao, N.B. Pena, F . E . Boettner, T.M. Williams, G. Forsythe and B. Epling, J. Org. Chem., 52 (1987) 45. 260 P.M. Scoot, in J.F.H. Purchase (Editor), Mycotoxins, Elsevier, Amsterdam, 1974, p. 383. 261 G. Engel and M. Teuber, in V. Betina (Editor), Mycotoxins - Production, Isolation, Separation and Purification, Elsevier, Amsterdam, 1984, Ch. 13, p. 291. 262 P.M. Scott and B.P. Kenneedy, J. Assoc. Off. Anal. Chem. 56 (1973) 813. 263 M.G. Siriwardana and P. Lafont, J. Chromatogr., 173 (1979 425. 264 U. Leuenberger, R. Gauch and E. Baumgartner, J. Chromatogr. 161 (1979) 303. 265 J. Sekiguchi and G.M. Gaucher, Biochem., 182 (1979) 445. 266 J.A. Miguel and V. de Andres, Invest. Agrar.: Prod. Prot Veget., 2 (1987) 225; C.A., 111 (1989) 6010k. 267 J. Reiss, Chromatographia, 4 (1971) 576. 268 R.A. Meyer, Nahrung, 26 (1982) 337. 269 J.O. Roland and L.R. Beuchat, Appl. Environ. Microbiol., 4 (1984) 205. 270 R. Steiman, F. Seigle-Murandi, L. Sage and S. Krivobok, Mycopathologia, 105 (1989) 129. 271 P.M. Scott and E. Sommers, J. Agric. Food Chem., 16 (1968 483. 272 P.M. Scott, J. Assoc. Off. Anal. Chem., 57 (1974) 621. 273 J.D.Bu'Lock, D. Shephard and D.J. Winstanley, Can. J. Microbiol., 15 (1969) 279. 274 J. Sekiguchi and G.M. Gaucher, Appl. Environ. Microbiol., 33 (1977) 147. 275 J. Sekiguchi and G.M. Gaucher, Can. J. Microbiol.. 25 (1979) 881. 276 J.W.D. Grootwassink and G.M. Gaucher, J. Bacteriol., 14 (1980) 443. 277 J. Sekiguchi, G.M. Gaucher and Y. Yamada, Tetrahedron Lett 1 (1979) 41. 278 P.I. Forrester and G.M. Gaucher, Biochemistry, 11 (1972 1102. 279 J. Sekiguchi and G.M. Gaucher, Biochem. J., 182 (1979) 445 280 W. Arafat, D. Kern and G. Dirheimer, Chem.-Biol. Interact I 56 I19851 333. 281 G . 'Engel and M. Teuber, in V. Betina (Editor), Mycotoxins
- Production, Isolation, Separation and Purification, Elsevier, Amsterdam, 1984, Ch. 13, p. 297. 282 K. Axberg and S . Gatenbeck, FEBS Lett., 54 (1975). 283 E.H. Reimerdes, G. Engel and J. Behnert, J. Chromatogr.,
110 (1975) 361. 284 C.P. Kurtzman and A. Ciegler, Appl. Microbiol., 20 (1970) 204. 285 M. Ehnert, A. Popken and K. Dose, 2. Lebensm. -Unters. -Forsch., 172 (1981) 110. 286 G. Engel, Milchwissenschaft, 3 3 (1978) 201. 287 A. Ciegler, H.-J. Mintzlaff, W. Machnik and L. Leistner, Fleischwirtschaft, 52 (1972) 1311 and 1317. 288 C.W. Thorpe and R.L. Johnson, J. Assoc. Off. Anal. Chem., 57 (1974) 861.
243 289 A. Ciegler and C.P. Kurtzman, J. Chromatogr., 51 (1970) 511. 290 D.M. Wilson, W.H. Tabor and M.W. Trucksess, J. Assoc. Off. Anal. Chem., 59 (1976) 125. 291 P. Lafont, M.G. Siriwardana, I. Combemale and J. Lafont, Food Cosmet. Toxicol., 17 (1979) 147. 292 D.F. Jones, R.H. Moore and G.C. Crawley, J. Chem. SOC., C, (1970) 1725. 293 G. Engel, K.E. Milczewski, D. Prokopek and M. Reuber, Appl. Environ. Microbiol., 43 (1982) 1031. 294 R.H. Williams, L.D. Boeck, J.C. Cline, D.C. De-Long, K.
Herzon, R.S. Gordee, M. Gorman, R.E. Holmes, S.H. Larsen, D.H. Lively, T.R. Matthews, J.D. Nelson, G.A. Poore, W.M. Stark and M.J. Sweeney, Antimicrob. Ag. Chemother.-1968
(1969) 229. 295 P. Lafont, J.-B. Debeaupuis, M. Gaillardin and J. Payen, Appl. Environ. Microbiol., 37 (1979) 365. 296 G. Engel, J. Chromatogr., 207 (1981) 430. 297 S.G. Yates, H.L. Tookey, J.L. Ellis and H.J. Burkhardt, Tetrahedron Lett., 24 (1967) 346. 298 Y. Ueno, N. Sato, K. Ishii, K. Sakai and M. Enomoto, Jpn. J. Exp. Med., 42 (1972) 461. 299 Y. Ueno and I. Ueno, Jpn. J. Exp. Med., 42 1972) 91. 300 Y. Ueno, in I.F.H. Purchase (Editor), Myco oxins, Elsevier, Amsterdam, 1974, p. 283. 301 G. Engel, J. Chromatogr., 130 (1977) 293. 302 R.J. Cole, J.W. Dorner, R.H. Cox, R.A. Hill H.G. Cutler and J.M. Wells, Appl. Environ. Microbiol., 42 (1981) 677. 303 V. Betina and Z. Barath, J. Antibiot., 17 (1964) 127. 304 U. Handschin, H.P. Sigg and Ch. Tamm, Helv. Ch m. Acta, 51 (1968) 1943. 305 C.T. Mabuni, L. Garlaschelli, R.A. Ellison and C.R. Hutchinson, J. Amer. Chem. SOC., 101 (1979) 707 Betina, in V. Betina (Editor), Mycotoxins Production, 306 V.
-
307 308 309 310 311 312 313 314 315 316
Isolation, Separation and Purification, Elsevier, Amsterdam, 1984, Ch. 11, p. 251. O.L. Shotwell, in J.V. Rodricks, C.W. Hesseltine and M.A. Mehlman (Editors), Mycotoxins in Human and Animal Health, Pathotox, Park Forest South, IL, 1977, p. 403. V. Betina, in V. Betina (Editor), Mycotoxins - Production, Isolation, Separation and Purification, Elsevier, Amsterdam, 1984, Ch. 11, p. 237. C.J. Mirocha, B. Schauerhamer and S.V. Pathre, J. Assoc. Off. Anal. Chem., 57 (1974) 1104. K. Ishii, M. Sawano, Y. Ueno and H. Tsunode, Appl. Microbiol., 27 (1974) 625. A. Gimeno, J. Assoc. Off. Anal. Chem., 66, 1983, 565. S.P. Swanson, H.A. Corley, D . G . White and W.B. Buck, J. Assoc. Off. Anal. Chem., 67 (1984) 580. M.-T. Liu, B.P. Ram, L.P. Hart and J.J. Pestka, Appl. Environ. Microbiol., 50 (1985) 332. B.A. Roberts and D . S . P . Patterson, J. Assoc. Off. Anal. Chem., 58 (1975) 1178. A. Gimeno, J. Assoc. Off. Anal. Chem., 6 2 (1979) 579. S.V. Pathre, C.J. Mirocha and S.W. Fenton, J. Assoc. Off. Anal. Chem., 6 2 (1979) 1268.
244 317 P.M. Scott, T. Panalaks, S. Kanhere and W.F. Miles, J. ~ssoc.Off. Anal. Chem., 61 (1978) 593. 318 M. Malayandi, J.P. Barrette and P.L. Wawrock, J. Assoc. Off. Anal. Chem., 59 (1976) 959. 319 R.M. Eppley, L. Stoloff, M.W. Trucksess and C.W. Chung., J. Assoc. Off. Anal. Chen., 57 (1974) 632. 320 P.M.D. Martin and P. Keen, Sabouraudia, 16 (1978) 15. 321 M. Jemmali, Arch. Inst. Pasteur Tunis, 3-4 (1977) 249. 322 A. Bottalico, A. Visconti, A. Logrieco, M. Solfrizzo and C.J. Mirocha, Appl. Environ. Microbiol., 49 (1985) 547. 323 W.M. Hagler and C.J. Mirocha, Appl. Environ. Microbiol., 39 (1980) 668. 324 K.E. Richardson, W.M. Hagler, Jr., P.B. Hamilton, Appl. Environ. Microbiol., 47 (1984) 1206. 325 K.E. Richardson, W.M. Hagler, Jr. and C.J. Mirocha, J. Agric. Food Chem., 33 (1985) 862. 326 H. Kamimura, Appl. Environ. Microbiol., 5 2 (1986) 515. 327 J.C. Wolf and C.J. Mirocha, Appl. Environ. Microbiol., 33 f19771 546. 328 D.R. 'Thouvenot and R.F. Morfin, J. Chrornatogr., 170 (1979 165. 329 T. Tanaka, A. Hasegawa, Y. Matsuki, U.-S. Lee and Y. Ueno J. Chromatogr., 328 (1985) 271. 330 M. Binder, Ch. Tamm, W.B. Turner and H. Minato, J. Chem S O C . , Perkin Trans. I, (1973) 1146. . . Tanenbaum (Editor), Cytochalasins, 331 Ch. -Tamm, in S.
Biochemical and Cell Biological Aspects, Nort-Holland, Amsterdam, New York, Oxford, 1978, Ch. 2, p. 15. 332 V. Betina, in V. Betina (Editor), Mycotoxins - Production, Isolation, Separation and Purification, Elsevier, Amsterdam, 1984, Ch. 12, p. 259. 333 W. Rothweiler and Ch. Tamm, Helv. Chim. Acta, 53 (1970) 696. 334 S.B. Padhye, C.R. Joshi and G.S. Pendse, J. Chromatogr., 108 (1975) 405. 335 A. Lees and S. Lin, J. Supramol. Struct., 1 2 (1979) 185. 336 D.C. Aldridge and W.B. Turner, J. Chem. SOC., C, (1969) 923. 337 S. Hayakawa, T. Matsushima and T. Kimura, Annu. Rep. Shionogi Res. Lab., No. 23 (1973) 2. 338 G. Chappuis and Ch. Tamm, Helv. Chim. Acta, 6 5 (1982) 521. 339 S.A. Patwarhan, R.C. Sukh Dev. Pandey and G.S. Pendse, Phytochemistry, 13 (1974) 1985. 340 A.M. Mujumdar, A.GH. Kapadi, G.R. Desphande and G.S. Pendse, Indian J. Pharm. Sci., 49 (1987) 107; C.A., 108 (1988) 11296a. 341 R. Capasso, A. Evidente and G. Randazzo, J. Chromatogr., (in
.
press ) Isawa, T. Hirose, T. Shimizu (nee' Tonioka), K. Koyaha and S. Natori, Tetrahedron, 45 (1989) 2323. 343 H. Minato, M. Matsumoto and TI. Katayama, Annu. Rep. Res. Lab., 23 (1973) 4. 344 H. Minato and T. Katayama, J. Chen. SOC., C, (1970) 45. 345 H. Minato, M. Matsumoto and T. Katayama, Paper presented at 167th Meeting of the Japan Antibiotics Research Association, May 28th, 1969. 342 Y.
245 346 W. Keller-Schierlein and E. Kupfer, Helv. Chim. Acta, 6 2 (1979) 1501. 347 M. Binder and Ch. Tamm, Helv. Chim. Acta, 56 (1973) 966. 348 M. Binder and Ch. Tamm, Helv. Chim. Acta, 56 (1973) 2387. 349 A. Probst and Ch. Tamm, Helv. Chim. Acta, 65 (1982) 1543. 350 S. Udaqawa, T. Muroi, H. Kurata, S . Sekita, K. Yoshihira, S . Natori and M. Umeda, Can J. Microbiol., 25 (1979) 170. 351 H.G. Cutler, F.G. Grumley, R.H. Cox, R.J. Cole, J.W. Dorner,
J.P. Springer, F.M. Latterell, J.M.E. Thean and A.E. Rossi, J. Aqric. Food Chem., 28 (1980) 139. 352 A. Probst and Ch. Tamm, Helv. Chim. Acta, 64 (1981) 2056. 353 S. Sekita, K. Yoshihira, S . Natori, S.-I. Udaqawa, F. Sakabe, H. Kurata and M. Umeda, Chem. Pharm. Bull., 30
(1980) 1609. 354 R.J. Cole, D.M. Wilson, J.L. Harper, R.H. Cox, T.W. Cochran, H.G. Cutler and D. K. Bell, J. Aqric. Food Chem., 30 (1982) 301. 355 P.S. Steyn, in V. Betina (Editor), Mycotoxins Production,
-
Isolation, Separation and Purification, Elsevier, Amsterdam, 1984, Ch. 9 , p. 183. 356 H. Cohen and M. Lapointe, J. Assoc. Off. Anal. Chem., 69 (1986) 957. 357 H. Terada, H. Tsubouchi,
K. Yamamoto, K. Hisada and Y. Sakabe, J. Assoc. Off. Anal. Chem., 69 (1986) 960. 358 P.S. Steyn and K.J. van der Merwe, Nature (London), 211
(1966) 418. 359 J. Piskorska-Pliszcynska and T. Juskiewicz, Arch. Inst. Pasteur Tunis, 3-4 (1977) 279 360 Official Methods of Analysis, 13th ed., AOAC, Washington, 1980, p. 426. 361 R.D. Hutchison, P.S. Steyn and D.L. Thompson, Tetrahedron Lett., (1971) 4033. 362 P.S. Steyn and C.W. Holzapfel, J. S . Afr. Chem. Inst., 20 (1967) 186. Wylie and L.G. Morehouse (Editors), 363 P.M. Scott, in T.D.
Mycotoxic Fungi. Mycotoxins and Mycotoxicoses. Vol. I, Marcel Dekker, New York, 1977, Ch. 2, p. 283. 364 W.W. Carlton and P. Kroqh, Food Drug Admin. Bur. Vet. Med., (1979) 165. 365 S . Nesheim, in J.V. 366 367 368 369 370 371 372 373 374
Rodricks (Editor), Mycotoxins and Other Fungal Related Food Problems, Advan. Chem. Ser., Vol. 149, 1976, p. 165. L. Fishbein and M.L. Falk, Chromatoqr. Rev., 12 (1970) 42. H.-M. M ller, Landwirtsch. Forsch., 33 (1976) 282. S.F. Nesheim, N.F. Hardin, O.J. Francis and W.S. Lanqham, J. Assoc. Off. Anal. Chem., 56 (1973) 817. S.F. Nesheim, J. Assoc. Off. Anal. Chem., 56 (1973) 822. R. Plestina, S. Ceovic, S . Gatenbeck, V. Habazin-Novak, K. Hult, E. Hokby, P. Kroqh and B. Radic, Proc. 4th Int. IUPAC Symp. Mycotoxins and Phycotoxins, Lausanne, 1979. S. Nesheim, M.W. Trucksess and C.W. Thorpe, AOAC 98th Annu. Meet., Washington, DC, 1984, Abstr. No. 126. A.E. de Jesus, P.S. Steyn, R. Vleqqar and P.L. Wessels, J. Chem. SOC., Perkin Trans. 2, (1980) 52. R. Fuchs, K. Hult, M. Peraica, B. RadiC and R. Plestina, Appl. Environ. Microbiol., 48 (1984) 41. P.S. Steyn, Pure Appl. Chem., 53 (1981) 891.
246 375 D.S.P. Patterson and B.A. Roberts, J. Assoc. Off. Anal. Chem., 6 2 (1979) 1265. 376 H. Johann and K. Dose, Z. Anal. Chem., 314 (1983) 139. 377 P.E. Haggblom and J. Ghosh, Appl. Environ. Microbiol., 49 (1985) 787. 378 E. Asensio, I. Sarmiento and K. Dose, Fresenius' Z. Anal. Chem., 311 (1982) 511. 379 H.M. Stahr, M. Domoto, B.L. Zhu and R. Pfeiffer, Mycotoxin Res., 1 (1985) 31. 380 H. Tsubouchi, K. Yamamoto, K. Hisada, Y. Sakabe and S. Udagawa, Mycopathologia, 97 (1987) 111. 381 R.M. Davis and J.L. Richard, in V. Betina (Editor),
- Production, Isolation, Separation and Mycotoxins Purification, Elsevier, Amsterdam, 1984, Ch. 14, p. 315. 382 C . O . Emeh and E.H. Marth, Arch. Microbiol., 115 (1977) 157. Hayes and H.W. McCain, Food Cosmet. Toxicol., 13 383 A.W. (1975) 221. 384 G.E. Cottral, Manual
of Standardized Methods for Veterinary Microbiology, Cornell University Press, Ithaca, N.Y., 1978, p. 659. 385 A.W. Hayes and B.J. Wilson, Appl. Microbiol., 16 (1968)
1163. 386 Y. Ueno,
in V. Betina (Editor), Mycotoxins - Production, Isolation, Separation and Purification, Elsevier, Amsterdam, 1984, Ch. 15, p * 239. 387 P.N. Varna, D.R. Lohar and A.K. Satsangi, Pharm. Weekbl., 118 (1983) 432. 388 H. Anke, I. Kolthoum, H. Zahner and H. Laatsch, Arch. Microbiol., 126 (1980) 223. 389 V. Betina, M. Liskova, S. Betina and V. Frank, Folia Microbiol. 34 11989) 252. 390 J.M. Wells, R:J. Cole and J.W. Kirksey, Appl. Microbiol., 30 (1975) 26. 391 N. Takeda, S. Seo, Y. Ogihara, U. Sankawa, I. Iitaka, I. Kitagawa and S . Shibata, Tetrahedron, 29 (1973) 3703. 392 Y. Ueno and I. Ishikawa, Appl. Microbiol., 18 (1969) 406. 393 P. Sedmera, M. Podojil, J. Vokoun, V. Betina and P. Nemec, Folia Microbiol., 23 (1978) 64. 394 V. Betina, P. Sedmera, J. Vokoun and M. Podojil, Experientia, 42 (1986) 196. 395 H. Murakami, J. Kobayashi, T. Masuda, N. Morooka and Y. Ueno, Mutation Res., 180 (1987) 147. Nagarajan, in V. Betina (Editor), Mycotoxins 396 R.
-
397 398 399 400 401 402
Production, Isolation, Separation and Purification, Elsevier, Amsterdam, 1984, Ch. 16, p. 351. J.L. Richard, R.L. Lyon, R.E. Fichtner and P.F. Ross, Mycopathologia, 107 (1989) 145. R. Hodges, J.M. Ronaldson, J.S. Shannon, A. Taylor and E.P. White, J. Chem. SOC., (1974) 26. A.D. Argoudelis and F. Reusser, J. Antibiot., 24 (1971) 383. P.J. Curtis, D. Greatbanks, B. Hesp, A.F. Cameron and A.A. Freer, J. Chem. SOC., Perkin Trans. I, (1977) 180. M.A. Stillwell, L.P. Magasi and G.M. Strunz, Can. J. Microbiol., 20 (1974) 759. G.M. Strunz, M. Kakushima and M.A. Stillwell, Can. J. Chem.,
53 (1975) 295. 403 D.H. Berg, R.P. Massing, M.M. Hoehn, L.D. Boeck and R.L. Hamill, J. Antibiot., 29 (1976) 394. 404 F. Dorn and D. Arigoni, Experientia, 30 (1974) 134.
241
405 D.C. DeLong, D.H. Lively and N. Neuss, U.S. Pat., 3 745 158, 1973. 406 M.A. Baute, G. Deflieux, R. Baute and A. Neveu, J. Antiblot., 31 (1978) 1099. 407 C.M. Maes, P.S. Steyn and F.R. van Heerden, J. Chromatogr., 234 (1982) 489. 408 B.J. Wilson and C.H. Wilson, Science (Washington), 144 (1964) 177. 409 B.J. Wilson, in A. Ciegler, S . Kadis and S.J. Ajl (Editors), Microbial Toxins, Vol. VI, Fungal Toxins, Academic Press, New York, 1971, p. 207. 410 Th. Fehr and W. Acklin, Helv. Chim. Acta, 49 (1966) 1907. 411 R.J. Cole, J.W. Dorner, J.A. Landsen, R.H. Cox, C. Pape, B. Cunfer, S.S. Nicholson and D.M. Bedell, J. Agric. Food Chem., 25 (1977) 1197. 412 R.J. Cole, J.W. Kirksey and J.M. Wells, Can. J. Microbiol., 20 (1974) 1159. 413 P.A. Cockrum, C.C.J. Culvener, J.A. Edgar and A.L. Payne, J. Nat. Prod., 42 (1979) 534. 414 R.J. Cole and R.H. Cox, Handbook of Toxic Fungal Metabolites, Academic Press, New York, 1981. 415 R.J. Cole, J.W. Kirksey, J.W. Dorner, D.M. Wilson, J. Johnson, Jr., N. Johnson, D.M. Bedell, J.P. Springer, K.K. Chexal, J. Clardy and R.H. Cox, J. Agric. Food Chem., 25 (1977) 826. 416 R.J. Cole, J.W. Kirksey, J.H. Moore, B.R. Blankenship, U.L. Diener and N.B. Davis, Appl. Microbiol., 24 (1972) 248. 417 A. Ciegler, Appl. Microbiol., 18 (1969) 128. 418 C.T. Hou, A. Ciegler and C.W. Hesseltine, Can. J. Microbiol., 17 (1971) 599. 419 R.T. Gallagher, G.C.M. Latch and R.G. Keogh, Appl. Environ. Microbiol., 39 (1980) 272. 420 M. Ibba, S.J.C. Taylor, Ch.M. Weedon and P.G. Mantle, J. Gen. Microbiol., 133 (1987) 3109. 421 Ch.M. Weedon and P.G. Mantle, Phytochemistry, 26 (1987) 969. 422 P.G. Mantle and J. Penn, J. Chem. SOC., Perkin Trans. I, (1989) 1539. 423 I. Laws and P.G. Mantle, J. Gen. Microbiol., 135 (1989) 1679. 424 H.W. Schroeder, R.J. Cole, J. Hein, Jr. and J.W. Kirksey, Appl. Microbiol., 29 (1975) 857. 425 P.V. Nielsen, L.R. Beuchat and J. Frisvad, Appl. Environ. Microbiol., 54 (1988) 1504. 426 B.J. Wilson, C.H. Wilson and A.W. Hayes, Nature (London), 220 (1968) 77. 427 C.T. HOU, A. Ciegler and C.W. Hesseltine, J. Assoc. Off. Anal. Chem., 54 (1971) 1037. 428 J.L. Richard and L.H. Arp, Mycopathologia, 67 (1979) 107. 429 A.E. de Jesus, P.S. Steyn, F.R. van Heerden, R. Vleggaar, P.L. Wessels and W.E. Hull, J. Chem. SOC., Chem. Commun., (1981) 289. 430 K.H. Ling, C.-K. Yang and F.-T. Peng Appl. Environ. Microbiol., 37 (1979) 355. 431 K.H. Ling, C.-M. Yang, C.-A. Kuo and M.-D. KUO, Appl. Environ. Microbiol., 44 (1982) 860. 432 F.-C. Peng, K.H. Ling, Y. Wang and G.-H. Lee, Appl. Environ. Microbiol., 49 (1985) 721. 433 L.M. Seitz, in V. Betina (Editor), Mycotoxins - Production, Isolation, Separation and Purification, Elsevier, Amsterdam, 1984, Ch. 20, p. 443.
248 434 D.E. Schade and A.D. King, J. Food Protect., 47 (1984) 978. 435 F. Palmisano, P.G. Zambonin, A. Visconti and A. Bottalico, J. Chromatogr., 465 (1989) 305. 436 J. Grabarkiewicz-Szczesna, J. Chelkowski and P. Zajkowski, Mycotoxin Res., 5, (1989) 77. 437 M.E. Stack, M.S. Thesis, American University, Washington, DC, 1984. 438 P. H ggblom, J. Gen. Microbiol., 133 (1987) 3527. 439 A. Visconti, A. Logrieco and A. Bottalico, Food Addit. Contam., 3 (1986) 323. 440 V. Betina, in V. Betina (Editor), Mycotoxins - Production,
Isolation, Separation and Purification, Elsevier, Amsterdam,
1984, Ch. 10, p. 217.
441 N.D. Davis, D.K. Dalby, U.L. Diener and G.A. Sansing, Appl Microbiol., 29 (1975) 118. 442 V. Betina, P. Nemec, M. Kutkova, J. Balan and $. KovaC, Chem. Zvesti (Bratislava), 18 (1964) 128. 443 A. Ciegler, R.F. Vesonder and L.K. Jackson, Appl. Environ Microbiol., 33 (1977) 1004. 444 V. Betina and 2. Binovska, Biologia (Bratislava), 34 (1979 461. 445 L.K. Jackson and A. Ciegler, Appl. Environ. Microbiol., 36 (1978) 408. 446 M.T. Wu, J.C. Ayres and P.E. Koehler, Appl. Microbiol., 27 (1974) 427. 447 C. Damodaran, C.S Ramadoss and E.R.B. Shanmugasundaram, Anal. Biochem., 52 (1973) 482. 448 J. Harwig, P.M. Scott, D.R. Stolz and B.J. Blanchfield, Appl. Environ. Microbiol., 38 (1979) 267. 449 R. Dick, U. Baumann and B. Zimmerli, Mitt. Geb. Lebensm.-Unters. Hyg., 79 (1988) 159. 450 V. Betina, 5 . Balint, V. Hajnicka and A. Nadova, Folia Microbiol. 18 (1973 40, 451 L.R. Marti, D.M. Wilson and B.D. Ewans, J. Assoc. Off. Anal. Chem., 6 1 (1978) 1353. 452 A. Gimeno, J. Assoc. Off. Anal. Chem., 6 7 (1984) 194. 453 R.F. Curtis, P.C. Harries, C.H. Hassall and J.D. Levi, Biochem. J., 90 (1964) 43. 454 R.D. Phillips, W.O. Berndt and A.W. Hayes, Toxicology, 12 (1979) 285. 455 B. Hald and P. Krogh, J. Assoc. Off. Anal. Chem., 56 (1973) 1440. 456 A.P. Damoglou, G.A. Downey and W. Shannon, J. Agric. Food Sci., 35 (1984) 395. and A.P. Damoglou, Appl. Microbiol. 457 M .A. Patterson Biotechnol., 27 (1987) 574. 458 R.J. Cole, in V. Betina (Editor), Mycotoxins Production,
-
459 460 461 462 463 464
Isolation, Separation and Purification, Elsevier, Amsterdam, 1984, Ch. 18, p. 405. J. Le Bars, Appl. Environ. Microbiol., 38 (1979) 1052. 0. Benkhemmar, F. Gaudemer and I. Bouvier-Fourcade, Appl. Environ. Microbiol., 50 (1985) 1087. J.A. Lansden, Assoc. Off. Anal. Chem., 69 (1986) 964. B.L. Rao and A. Husain, Mycopathologia, 97 (1987) 89. K. Hermansen, J.C. Frisvad, C. Emborg and J. Hansen, FEMS Microbiol. Lett., 21 (1984) 253. M.W. Trucksess, P.B. Mislivec, K. Young, V.R. Bruce and S.W. Page, J. Assoc. Off. Anal. Chem., 70 (1987) 123.
249 465 P. S. Steyn, J. Chromatogr., 45 (1969) 473. 466 A. Rathinevalu, E. Rajabhavani and B. Shanmugasundaram, J. Assoc. Off. Anal. Chem., 67 (1984) 38. 467 R.K. Malik, G. Engel and M. Teuber, Appl. Microbiol. Biotechnol., 24 (1986) 71. 468 P.M. Scott, in V. Betina (Editor), Mycotoxins Production,
-
Isolation, Separation and Purification, Elsevier, Amsterdam, 1984, Ch. 23, p. 469. 469 P.E. Still, Thesis, University Microfilms, Ann. Arbor, MI,
1973. 470 P.M. Scott, B.P.C. Kennedy, J. Harwig and B.J. Blanchfield, Appl. Environ. Microbiol., 3 3 (1977) 249. 471 P.M. Scott and S.R. Kanhere, J. Assoc. Off. Anal. Chem., 62 (1979) 141. 472 P.M. Scott, in V. Betina (Editor), Mycotoxins - Production,
Isolation, Separation and Purification, Elsevier, Amsterdam,
1984, Ch. 22, p. 463.
473 A.G.
Kozlovskii, T.A. Reshetilova, T.N. Medvedeva, M.U. Arinbasarov, V.G. Sakharovskii and V.M. Adanin, Biokhimiya,
44 (1979) 1335. 474 S. Ohmomo, T. Sato, T. Utagawa and M. Abe, J. Agric. Chem. SOC. Jap., 49 (1975) 615. 475 R.-D. Wei, P.E. Still, E.B. Smalley, H.K. Schnoes and F.M. Strong, Appl. Microbiol., 25 (1973) 111. 476 P. Lafont and J.P. Debeaupuis, J. Chromatogr., 198 (1980) 481. 477 P.M. Scott, M.A. Merrien and J. Polonsky, Experiential 32 (1976) 140. 478 R.E. Wagener, N.D. Davis and U.L. Diener, Appl. Environ. Microbiol., 39 (1980) 882 479 P.M. Scott and B.P. Kennedy, J. Assoc. Off. Anal. Chem., 56 (1973) 813. 480 S. Ohmomo, T. Atagawa and M. Abe, Agric. Biol. Chem., 41 (1977) 2097. 481 J.H. Wall and E.B. Lillehoj, J. Chromatogr., 268 (1983) 461. 482 J.E. Robberts, S. Hong, J. Tuite and W.W. Carlton, Appl. Environ. Microbiol., 36 (1978) 819. 483 M.E. Stack, P.B. Mislivec, T. Denizel, R. Gibson and A.E. Pohland, J. Food Prot., 46 (1983) 965. 484 K.C. Ehrlich, A.J. DeLucca and A. Ciegler, Appl. Environ. Microbiol., 48 (1984) 1. 485 R.J. Cole and R.H. Cox, Handbook of Toxic Fungal Metabolites, Academic Press, New Yor, 1983, p. 646. 486 V. Betina, in V. Betina (Editor), Mycotoxins - Production, 487 488 489 490 491
Isolation, Separation and Purification, Elsevier, Amsterdam, 1984, Ch. 25, p. 481. A. Ciegler, A.W. Hayes and R.F. Vesonder, Appl. Environ. Microbiol., 39 (1980) 285. P.S. Steyn, Tetrahedron, 26 (1970) 51. P.S. Steyn, J. S. Afr. Chem. Inst., 22 (1969) 520. R. Andersen, G. B chi, B. Kobbe and A.L. Demain, J. Org. Chem., 42 (1977) 352. Dose, Fresenius' 2. Anal. Chem., 319 Ch. Jansen and K.
(1984) 60. 492 J.M. Farber,
G W. Sanders, G.W. Lawrence and P.M. Scott, Mycopathologia, 101 (1988) 187. 493 P.M. Scott, G.A Lawrence and T.I. Matula, in P.S. Steyn and Editors), Bioactive Molecules, Vol. 1, R. Vleggaar Mycotoxins and Phycotoxins, Elsevier, Amsterdam, 1986, p. 305.
250 494 J. Chelkowski, M. Zawadski, P. Zajkowski, A. Logrieco and A. Bottalico, Mycotoxin Res., 6 (1990) 41. 495 M.J. Shepherd and J. Gilbert, J. Chromatogr., 358 (1986) 415. 496 H.K. Abbas and C.J. Mirocha, Appl. Environ. Microbiol., 54 (1988) 1268. 497 R.F. Vesonder, R. Lambert, D.T. Wicklow and M.L. Biehl, Appl. Environ. Microbiol., 56 (1988) 830. 498 R.W. Jones and J.G. Hancock, Can. J. Microbiol., 33 (1987) 963. 499 H.R. Burmeister, M.D. Grove, R.E. Peterson, D. Weisleder and P.D. Plattner, Appl. Environ. Microbiol., 50 (1985) 311. Gelderblom, P.G. Thiel, W.F.O. Marasas and K.J. van 500 W.C.A. der Merwe, J. Agric. Food Chem., 32 (1984) 1064. 501 J.M. Farber and G.W. Sanders, Appl. Environ. Microbiol., 51 (1986) 381. 502 L.A. Wiebe and L.F. Bjeldanes, J. Food Sci., 46 (1981) 1424. 503 M.A. Jackson, P.J. Slininger and R.J. Bothast, Appl. Environ. Microbiol., 55 (1989) 649. 504 J.V. Edwards and E.B. Lillehoj, J. Assoc. Off. Anal. Chem., 70 (1987) 126. M ller, J. Assoc. Off. Anal. 505 B.G.E. Josefsson and T.E. Chem., 6 0 (1977) 1369. 506 Y. Takeda, E. Isohata, R. Amano and M. Uchiyama, J. Assoc. Off. Anal. Chem., 6 2 (1979) 573. 507 M.P. Whidden, N.N. Davis and B.A. Roberts, J. Agric. Food Chem., 28 (1980) 784. 508 D.S.P. Patterson, S.P. Derycks and U.L. Diener, J. Agric. Food Chem., 28 (190) 1265. 509 L. Fishbein and M.L. Falk, Chromatogr. Rev., 12 (1970) 42. 510 L. Stoloff, L. Nesheim, L. Yin, J.V. Rodricks, M. Stack and A.D. Campbell, J. Assoc. Off. Anal. Chem., 54 (1971) 91. Abdel-Kader and I.A. El-Kady, 511 A.H. Moubasher, M.I.A. Mycopathologia, 66 (1978) 187. 512 I. Coman, 0. Popescu, P. Halga, N. Grunberg and E. Ciudin, Rev. Cresterea Anim., 28 (1978) 58; C.A., 93 (1980) 68758d. 513 T. Ilus, M.L. Niku-Paavola and T.M. Enari, Eur. J. Appl. Microbiol., Biotechnol., 11 (1981) 244. 514 P. Nowotny, W. Baltes, W. Kroenert and R. Weber, Lebensmittelchem. Gerichtl. Chem., 37 (1983) 71. 515 A. Gimeno and M.L. Martins, J. Assoc. Off. Anal. Chem., 66 (1983) 85. Stahr and M. Domoto, Advances in Thin Layer 516 H.M. Chromatography (Proceedings of 2nd Bienn. Symp. 1980), 1982, p. 405. 517 P. Golinski and J. Grabarkiewicz-Szczesna, J. Assoc. Off. Anal. Chem., 6 7 (1984) 1106. 518 J. Grabarkiewicz-Szczesna, P. Golinski, J. Chelkowski and K. Szebiotko, Nahrung, 29 (1985) 229. Fulgeira and B.J. de Bracelenti, Rev. Latinam. 519 C. Microbiol., 29 (1987) 216. 520 M.O. Tapia, Rev. Argent. Microbiol., 17 (1986) 183. 521 U. Thrane, Lett. Appl. Microbiol., 3, No. 5 (1986) 93.
25 I 522 0. Filtenborg, J.C.
Frisvad and J.A. Svendsen, Appl. Environ. Microbiol., 45 (1983) 581. 523 S . Krivobok, F. Seigle-Murandi, R. Steiman and D. Marzin, J. Microbiol. Methods, 7 (1987) 29. 524 R. Hadidane, C. Roger-Regnault. H. Bouattour, F. Ellouze, H. Bacha, E.E. Creppy and G. Dirheimer, Human. Toxicol., 4 (1985) 491. 525 H. Bacha, R. Hadidane,
E.E. Creppy, C . Regnault, F. Ellouze and G. Dirheimer, J. Stored Prod. Res., 24 (1988) 199. 526 A. Bottalico, A. Logrieco and A . Visconti, Mycopathologia, 107 (1989) 85. 527 A. Bottalico, A. Visconti and M. Solfrizzo, Phytopath. Medit., 21 (1982) 105. 528 D.A. Chakrabarti and S . Ghosal, Appl. Environ. Microbiol., 51 (1986) 217. 529 C.J. Mirocha, H.K. Abbas, T. Kommendahl and B.B. Jarvis, Appl. Environ. Microbiol., 55 (1989) 254. 530 G.A. Bennett and O.L. Shotwell, J. Assoc. Off. Anal. Chem., 73 (1990) 270. 531 J.C. Frisvad, Appl. Environ. Microbiol., 41 (1981) 568. 532 J.C. Frisvad and 0. Filtenborg, Appl. Environ. Microbiol., 46 (1983) 1301. 533 J.C. Frisvad, J. Appl. Bacteriol., 54 (1983) 409. 534 R.A. Samson and J.I. Pitt (Editors), Advances in Penicillium
and Aspergillus Systematics, Plenum Press, Mew York, London,
1985. 535 J.C. Frisvad, 536 537 538 539 540
in R.A. Samson and J.I. Pitt (Editors), Advances in Penicillium and Aspergillus Systematics, Plenum Press, New York, London, 1985, p. 311. J.C. Frisvad, in R.A. Samson and J.I. Pitt (Editors), Advances in Penicillium and Aspergillus Systematics, Plenum Press, New York, London, 1985, p . 327. J.C. Frisvad, in R.A. Samson and J.I. Pitt (Editors), Advances in Penicillium and Aspergillus Systematics, Plenum Press, New York, London, 1985, p . 437. J.C. Frisvad and U. Thrane, J. Chromatogr., 404 (1987) 195. R.R.M. Paterson, J. Chromatogr., 368 (1986) 249. A.A. El-Banna, J.I. Pitt and L. Leistner, System. Appl. Microbiol., 10 (1987) 42.
This Page Intentionally Left Blank
253
Chapter 8 LIQUID COLUMN CHROMATOGRAPHY OF MYCOTOXINS J.C. FRISVAD and U. THRANE 8.1. INTRODUCTION
Liquid column chromatographic methods are by far the most important in preparative and analytical chemistry of non-volatile and non-antigenic natural products (1-10), even though developments in thin-layer techniques have been remarcable in recent years (1 1-13). Both types of methods are now used extensively in natural products chemistry and the combination of them is often rewarding. Being natural products from filamentous fungi that evoke a toxic response in vertebrates when introduced in low concentration by a natural route (14), mycotoxins are chemically very diverse. They may be polar (e.g. patulin), non-polar (e.g. aflatrem), acidic (e.g. citrinin), basic (e.9. roquefortine C), and they may contain chlorine (e.g. penitrem A), a nitro group (e.9. P-nitropropionic acid), a characteristic chromophore (e.g. luteoskyrin), be strongly fluorescing (e.g. territrems) etc. which are all attributes that are very important in the selection of separation and detection methods. Extraction and clean-up from a complex matrix (especially foods and feedstuffs) also depend heavily on the chemical nature of the mycotoxins. Thus because most of the known mycotoxins are present in the fungal membrane in conidia, ascomata, sclerotia, mycelium (15,16) they contain a non-polar moiety in the molecule and are often soluble in organic solvents, while other important very polar mycotoxins may have been partially neclected, because they were retained in the water or waterhnethanol fraction. Good examples of very polar compounds are the important carcinogens fumonisins (17) and islanditoxin and cyclochlorotine (18). However a substancial part of known mycotoxins have polar and non-polar portions in the molecule and will be
present both in the fungal thallus and excreted into the growth medium (i.e. food, feedstuff or fermentation broth). Most of those secondary metabolites are soluble in
254
chloroform or ethyl acetate. In the case of chloroform extractable mycotoxins fungal and food lipids will be a major analytical problem, while amino acids, carbohydrates, organic acids, nucleic acids etc. cause problems in water/methanol or acetonitrile extractions. Thus the structural diversity of the mycotoxins render the design of good general multi-mycotoxin methods difficult. In recent years a clear connection between fungal species and profiles of mycotoxins and other secondary metabolites have been apparent (16, 19-28). Furthermoreeach food commodity has its own associated spoilage mycoflora (29), i.e. fungi that actually grow on the product under natural conditions. This means that only few mycotoxins will be probable contaminants in different foods under specified environmental conditions. Such knowledge should be used more extensively in new multi-mycotoxin methods, but should of course be used with caution in compound feeds and blended foods. An updated list of the producers of important mycotoxins and other secondary metabolites is given in Table 8.1. A large number of producers of fungal metabolites have been misidentified (19, 26, 30) and the metabolites named after fungi that did not produce them. Together with data on the associated mycoflora of different types of foods and feedstuffs (29, 31), valuable information on the possible occurrence of mycotoxins may be drawn and this could help in simplifying clean-up procedures and HPLC methods for mycotoxins. A major part of known fungal secondary metabolites (32-34) are not classified as mycotoxins. They may have toxic effects on insects (insecticides), plants (herbicides) or microorganisms (antibiotics) or they may have pharmacological effects on vertebrates or act synergistically with known mycotoxins on vertebrates. Furthermore they may be good indicators of fungal contamination of foods and feeds or of mycotoxins produced in smaller amounts. Thus some of these fungal secondary metabolites will also be treated in this review. Several excellent reviews have been written on liquid chromatography of mycotoxins (3, 35-40), especially the aflatoxins, so recent advances in applications of liquid chromatography of mycotoxins will be emphasized in this chapter, especially analytical separation and detection methods.
255
TABLE 8.1. An updated list of mycotoxins and other secondary metabolites and their producers' Fungal metabolite
Known producers
4-acetamido-4-hydroxy-2-butenoic acid y-lactone (butenolide)
Fusarium acuminatum Fusarium cerealis Fusarium culmorum Fusarium graminearum Fusarium poae Fusarium sambucinum var. coeruleum Fusarium sporotrichioides Fusarium tricinctum
1'-acetoxypaxilline
Emericella striata
8a-acetoxyverrucarinJ
Myrothecium verrucaria
19-acetylchaetoglobosinA, B, D
Chaetomium globosum
3-acetyldeoxynivalenol
Fusarium cerealis (=F. crookwellense) Fusarium culmorum Fusarium graminearum Alternaria citri Fusarium culmorum Penicillium chrysogenum var. chrysogenum
Aculeasin A y Aflatoxiwl
B
Aspergillus aculeatus 2
Aflatoxin B,, B,
Aspergillus flaws Aspergillus nomius Aspergillus parasiticus
Aflatoxin G,, G2
Aspergillus nomius Aspergillus parasiticus
Aflatoxin
Aspergillus nomius Aspergillus parasiticus
G,.
Aflatoxin M,, M2
Aspergillus nomius Aspergillus parasiticus
Aflatrem
Aspergillus flaws
a-acetyl-y-methyl tetronic acid Altenuene
Alternaria alternata
Altenuisol
Alternaria alternata
Altenusin
Alternaria alternata
Alternariol
Alternaria alternata Alternaria brassicicola Alternaria cheiranti Alternaria citri Alternaria cucumerina
256
TABLE 8.1. (continued) Fungal metabolite
Known producers
Alternariol (continued)
Alternaria dauci Alternaria kikuchiana Alternaria longipes Alternaria porri Alternaria raphani Alternaria tenuissima Botrytis aclada (= 6. allii) Corynespora smithii Penicillium diversum Pleospora scrophulariae Talaromyces flavus
Alternariol-monomethylether
see alternariol
Altertoxin I, II
Alternaria alternata Alternaria cassiae Alternaria mali Alternaria tenuissima
Anhydrofusarubin
Fusarium solani Fusarium verticillioides
Antibiotic Y
Fusarium acuminatum Fusarium avenaceum Fusarium chlamydosporum Fusarium tricinctum
Aranotins
Amauroascus aureus Aspergillus terreus
Ascladiol
Aspergillus clavatus
Ascochalasin
Ascochyta heteromorpha
Ascochitine
Ascochyta fabae Ascochyta pisi
Asperfuran
Aspergillus oryzae Penicillium clavigerum Penicillium glandicola Neosartorya fennelliae
Asperlicin
Petromyces alliacea
Aspergillic acid
Aspergillus flavus Aspergillus nomius Aspergillus parasiticus Aspergillus sojae
Asperthecin
Emericella nidulans Emericella quadrilineata Emericella rugulosa
Aspochalasin A-D
Aspergillus microcysticus
257
TABLE 8.1. (continued) Fungal metabolite
Known producers
Asteltoxin
Emericella variecolor
Asterric acid
Aspergillus terreus Penicillium glabrum Penicillium vulpinum
Auranthine
Penicillium aurantiogriseum var. aurantiogriseum chernotype I
Aurantiarnine
Penicilliurnaurantiogriseurn var. aurantiogriseum chernotype I var. aurantiogriseum chernotype I I var . neoechinulaturn
Aurovertin B
Calcariosporium arbuscula
Austalides
Aspergillus ustus
Austamide
Aspergillus ustus
Austin
Aspergillus ustus Emericella variecolor Penicillium diversum
Austdiol
Aspergillus ustus
Austocystins
Aspergillus puniceus Aspergillus ustus
A verufin
see sterigmatocystin Cercospora arachidicola Cercospora smilacis
Barnol
Eupenicillium baarnense
Benzoic acid
Aspergillus raperi Fusarium oxysporum Rhizoctonia leguminiwla Rhizoctonia solani
Bostrycin
Alternaria eichhorniae Bostriconema alpestre
Bostrycoidin
Fusarium oxysporum Fusarium solani
Botryodiploidin
Apiosordaria sp. Botryodiplodia theobromae Lacunospora tetraspora Penicillium brevicompactum Penicillium roqueforti var. carneum Talaromyces stipitatus Triangularia bambusae Zopfiella maisushimae
258
TABLE 8.1. (continued) Fungal metabolite
Known producers
Brefeldin A
Alternaria carihami Curvularia lunata var. lunata Cylindrocarpondestructansvar .destructans Eupenicillium brefeldianum Eupenicillium ehrlichii Eupenicillium ludwigii Penicillium cremeogriseum Penicillium onobense Penicillium piscarium Phoma medicaginis var. medicaginis
Brevianarnide A, B
Penicillium aurantiogriseum var. viridicatum chemotype I Penicillium brevicompactum
Byssochlamic acid
Byssochlamys fulva Byssochlamys nivea Paecilomyces variotii
Byssotoxin
Byssochlarnys fulva
Canadensolide
Aspergillus tamarii Neosariorya stramenia Penicillium arenicola
Calonectrin
Monographella nivalis
Canescin
Aspergillus fumigatus Penicillium canescens Penicillium smithii
Carlosic acid
Penicillium charlesii
Carolic acid
Penicillium charlesii
Catenarin
Drechslera catenaria Drechslera grarninea Drechslera tritici-repentis Eurotium acutum Eurotium amstelodami Eurotium chevalieri Eurotium cristatum Eurotium echinulatum Eurotium glabrum Eurotium leucocarpum Eurotium niveoglaucum Eurotium repens Eurotium rubrum Helminthosporium velutinum Penicillium islandicum Pyrenophora graminea Pyrenophora tritici-repentis Talaromyces stipitatus
259
TABLE 8.1. (continued) Fungal metabolite
Known producers
Chaetochromin
Chaetomium caprinum Chaetomium gracile Chaetomium tetraspermum Chaetomium thielavioideum
Chaetocin
Chaetomium minutum Chaetomium tenuissimum Chaetomium thielavioideum Farrowia sp.
Chaetoglobosin A-E
Chaetomium cochliodes Chaetomium globosum var. globosum Chaetomium globosum var. rectum Chaetomium mollipilium Chaetomium subaffine Penicillium echinulatum var. discolor Pencillium expansum
Chaetoglobosin F, G, J
Chaetomium globosum var. globosum
Chaetoglobosin K, L, M
Diplodia macrospora
Chetomin
Chaetomium cochliodes Chaetomium funicola Chaetomium globosum var. globosum Chaetomium subglobosum Chaetomium tenuissimum Chaetomium umbonatum
6a-chlamydosporol
Fusarium acuminatum Fusarium avenaceum Fusarium chlamydosporum
6p-chlamydosporol
see 6a-chlamydosporol
Chromanol 1, 2, 3
Aspergillus duricaulis
Chrysarin Chrysogine
Alternaria citri Aspergillus parasiticus Penicillium chrysogenum var. chrysogenum
Chrysophanol
Ascochyta pisi Chaetomium elatum Drechslera catenaria Hypocrea austro-grandis Penicillium islandicum Phoma foveata Pseudospiropes simplex Sepedonium ampullosporum Trichoderma hamatum
260
TABLE 8.1 . (continued) Fungal metabolite
Known producers
Citreomontanin
Penicillium manginii
Citreoviridin A
Eupenicillium ochrosalmoneum Penicillium citreonigrum Penicillium rnanginii Penicillium miczynskii Penciillium smithii
Citreoviridin
see citreoviridin A
Citrinin
Aspergillus carneus Aspergillus terreus Ceuthospora sp. Clavariopsis aquatica Penicillium citrinum Penicillium expansum Penicillium hirsutum var. albocoremium Penicillium lividum Penicillium verrucosum chemotype II Penicillium westlingii Pythium ultimum
Citromycetin
lllosporium olivatrum Penicilliurn glabrum Penicillium roseopurpureum Penicillium steckii
Clad0fulvin
Fulvia fulva
Cladosporin
Aspergillus proliferans Cladosporium cladosporioides Eurotium glabrum Eurotium pseudoglaucum Eurotium repens Penicillium daleae Penicillium selandiae
Clerocidin
Oidiodendron truncatum
Cochliodinol
Chaetomium cochliodes Chaetomium elatum Chaetomium globosum
Compactin
Aspergillus terreus Penicillium solitum
Crotocin
Acremonium crotocinigenum Trichothecium roseum
Culmorin
Fusarium cerealis Fusarium culmorum Fusarium graminearum
26 1
TABLE 8.1. (continued) ~~
Fungal metabolite
Known producers
Curvularin
Alternaria cinerariae Alternaria zinniae Bipolaris nodulosa Bipolaris papendorfii Bipolaris spicifera Drechslera australiensis Eupenicillium senticosum Penicillium roseopurpureum Penicillium steckii Pseudodiplodia obiones
Curvulinic acid
Bipolaris ellisii Bipolaris papendorfii Microdiplodia microsporella Penicilliumjanczewskii Penicillium novae-zeelandiae
Cyclochlorotine
Penicillium islandicum
Cyclopaldic acid
Aspergillus duricaulis Aspergillus puniceus Penicillium commune chernotype I & II Penicillium mononematosum Pestalotia palmarum Neosadorya quadricincta
Cyclopenin
Penicillium aurantiogriseum var. aurantiogriseum chernotype II var. neoechinulatum var. polonicum var. viridicatum chernotype II Penicillium commune chernotype II Penicillium crustosum Penicillium echinulatum var. echinulatum var. discolor Penicillium hirsutum var. albocoremium var. allii var. venetum Penicillium solitum Penicillium vulpinum
Cyclopenol
see cyclopenin
Cyclopeptin
see cyclopenin
Cyclopiamide
Penicillium griseofulvum var. griseofulvum
Cyclopiamine
Aspergillus caespitosus Penicillium griseofulvum var. griseofulvum
262
TABLE 8.1. (continued) Fungal metabolite
Known producers
Cyclopiazonic acid
Aspergillus flavus Aspergillus otyzae Aspergillus tamarii Penicillium camemberti Penicillium commune chemotype I 8 II Penicillium griseofulvum var. griseofulvum
Cyclosporin A
Beauveria nivea Cylindrocarpon lucidum Tolypocladium inflatum
Cynodontin
Bipolaris cynodontis Sipolaris euchlaenae Bipolaris oryzae Bipolaris sorokiniana Bipolaris speciiera Bipolaris victoriae Curvularia lunata var. lunata Curvularia pallescens Cycloconium olieagineum Drechslera avenae Exserohilum rostratum Mycocentrospora acerina Phoma terrestris
Cytochalasin A
Ascochyta heteromorpha Curvularia lunata var. lunata Drechslera biseptata Drechslera dematioidea Gnomonia erythrostoma Hypomyces odoratus Phoma exigua var. exigua
Cytochalasin B
Ascochyta heteromorpha Curvularia lunata var. lunata Drechslera dematioidea Hormiscium sp. Phorna exigua var. exigua
Cytochalasin C
Metarrhizium anisopliae Hypoxylon terricola
Cytochalasin D
Coriolus vernicipes Engleromyces goetzii Metarrhizium anisopliae Zygosporium masonii Hypoxylon terricola Microporus afinis
263
TABLE 8.1. (continued) Fungal metabolite
Known producers
Cytochalasin E
Aspergillus clavatus Aspergillus terreus Drechslera dernatioidia Rosellinia necatrix
Cytochalasin F
Drechslera dernatioidia
Cytochalasin G
Nigrosabulum sp.
Cytochalasin H
Phomopsis paspali
Cytochalasin K, L, M
Chalara microspora
Cytochalasin K
Aspergillus clavatus
Cytochalasin N, 0, P, Q, R, S
Phomopsis sp.
Cytochalasin N’, 0:P’, Q’, R’
Hypoxylon terricola
bis-dechlorogeodin
Penicillium glabrum
Dechlorogriseofulvin
see griseofulvin
Dechloronidulin
Emericella unguis
Dehydrocarolic acid
Penicillium adametzii Penicillium charlesii
Dehydrocurvularin
Alternaria cinerariae Alternaria citri Alternaria cucumerina Alternaria dauchi Alternaria macrospora Alternaria scirpimla Alternaria tomato Alternaria zinniae Aspergillus aureofulgens Drechslera australiensis Penicillium restricturn Penicillium steckii Pseudodiplodia obiones
Dehydrocyclopeptin
see cyclopenin Emericella striafa
Dehydropaxilline Dehydroustic acid
Aspergillus puniceus Aspergillus ustus
Z’-dehydroverrucarin A
Myrothecium verrucaria
Demethoxyviridiol
Nodulisporium hinnuleum Trichoderma viride
Deoxybostrycin
Alternaria eichhorniae
Deoxaphomin
Phoma exigua var. exigua Ascochyta heteromorpha
264
TABLE 8.1. (continued) Fungal metabolite
Known producers
Deoxynivalenol
Fusarium cerealis Fusarium culmorum Fusariurn graminearum
1-deoxypebrolide
Penicillium brevicompactum
12,13-deoxyverrucarinA
Myrothecium verrucaria
Dermoglaucin
Cortinarius sanguineus
Desacetylpebrolide
Penicillium brevicompactum
Desertorin A, B, C
Emericella desertorurn
Destrutoxins
Metarrhizium anisopliae
Dethiosecoemestrin
Emericella striata
Diacetoxyscirpenol
Fusarium acuminatum Fusarium equiseti Fusarium sambucinum var. sambucinum Fusarium sporotrichioides
2',3', ip,8p-diepoxyroridin H
Cylindrocarpon sp.
Diethylphthalate
Penicillium funiculosum
Dihydrocytochalasin B
Drechslera dematioidea
22,23-dihydr0-24,25-dehydr02 1-oxo-aflavinine
Aspergillus niger
2,3-dihydro-3,6-dihydroxy2-methyl-4-pyrone
Penicilliurn restricturn
Dihydroergotamin
Claviceps paspali Claviceps purpurea
cis-dihydrofusarubin
Fusarium solani Fusarium verticillioides
trans-dihydrofusarubin
see cis-dihydrofusarubin
5,6-dihydro-4-methoxy-2Hpyran-2-0118
Penicillium italicum
2 ',3'-dihydrosorbicillin
Verticillium intertexturn
Dihydrosterigmatocystin
Aspergillus versicolor
Dihydroxyaflavinine
Aspergillus flavus
2,4-dihydroxy-6-(1,2-dioxopropyl) benzoic acid
Penicillium brevicompactum
2,4-dihydroxy-6-(1-hydroxy2-oxopropyl) benzoic acid
Penicillium brevicompactum
2,4-dihydroxy-6-(2-oxopropyl) benzoic acid
Penicillium brevicompactum
265
TABLE 8.1. (continued) Fungal metabolite
Known producers
2,7-dimethoxy-6-( 1-acetoxyethyl)-juglone 2,7-dimethy1-6-ethyljuglone
Nattrassia mangiferae
3,5-dimethyl-6-hydroxyphthalic acid
Penicillium gladioli
Dimethylphthalate
Penicillium funiculosum
Dipicolinic acid
Fusarium reticulaturn Beauveria bassiana Paecilomyces furnosoroseus Penicillium citreonigrum Verticilium lecanii
Diploidiatoxin
Diplodia maydis
Dithiosilvatin
Aspergillus silvaticus
Dothistromin
Cercospora arachidicola Cercospora microsora Cercospora rosicola Cercospora smilacis Cercosporidium personaturn Mycovellosiella ferruginea Scirrhia pini Sirosporium di ffusum
Duclauxin
Penicillium duclauxii Penicillium herquei Talaromyces macrosporus Talaromyces stipitatus
Echinulin
Eurotium amstelodami Eurotium chevalieri Eurotium echinulatum Eurotium heterocaryoticum Eurotium repens Eurotium rubrum
Emestrin
Emericella acristata Emericella foveolata Emericella parvathecia Emericella quadrilienata Emericella striata
Emestrin B
Emericella quadrilineata Emericella striata
Emindol DA
Emericella desertorurn Emericella quadrilineata
Emindol SA
Emericella striata
Nattrassia mangiferae
266
TABLE 8.1. (continued) Fungal metabolite
Known producers
Emodin
Aspergillus aculeatus Aspergillus ochraceus (= A. alutaceus) Aspergillus wentii Acroschyphus sphaerophoroides Caloplaca sp. Cetraria cullulata Cortinarius sanguineus Drechslera catenaria Eurotium chevalieri Eurotium cristatum Eurotium echinulatum Fulvia fulva Hamigera avellanea Nephroma laevigata Peniciliopsis clavariaeformis Penicillium brunneum Penicillium islandicum Penicillium tardum Hypocrea austro-grandis Phoma foveata Pyrenochaeta terrestris Talaromyces stipitatus Valsonia rubricosa Xanthoria fallax
Engleromycin
Engleromyces goetzii
Enniatins
Fusarium acuminatum Fusarium avenaceum Fusarium oxysporum Fusarium sambucinum var. sarnbucinum
€pi- & fagi-cladosporic acid
Cladosporium herbarum
Epicorazine A, 6
Epicoccum nigrum
1Cepi- 14-hydroxy-10,23-dihydro24,25-dehydro-aflavinine
Aspergillus f l a w Aspergillus niger Aspergillus parasiticus
€pi- 10-verruculotoxin
Penicillim brasilianum
Epoxycytochalasin H, J
Phomposis sojae
$,@-epoxyisororidin E
Cylindrocarpon sp.
Tp,Bp-epoxyroridinH
Cylindrocarpon sp.
Equisetin
Fusarium equiseti Fusarium pallidoroseurn
Eremofortins
Penicillium roqueforti
261
TABLE 8.1. (continued) Fungal metabolite
Known producers
Ergocristine
Claviceps paspali Claviceps purpurea
Ergocryptin
see ergocristine
Ergometrin
see ergocristine
Ergotamin
see ergocristine
Ergosterol
nearly all fungi
Erythroglaucin
Alternaria porri Dermocybe cinnabarina Drechslera catenaria Eurotium acutum Eurotium appendiculatum Eurotium chevalieri Eurotium cristatum Eurotium echinulatum Eurotium glabrum Eurotium herbariorum Eurotium intermedium Eurotium leucocarpum Eurotium niveoglaucum Eurotium pseudoglaucum Eurotium repens Eurotium rubrum Eurotium spiculosum Talaromyces stipitatus Xanthoria fallax Xanthoria mandschurica
Eryirhoskyrine
Penicillium islandicum
Ethisolide
Micropera caespitosa Penicillium decumbens
Expansolide
Penicillium expansum
Ferulic acid
Rhizoctonia leguminicola Rhizoctonia solani
Flavipin
Acrospheira sp. Aspergillus flavipes Epicoccum nigrum
Flavoglaucin
Eurotium amstelodami Eurotium chevalieri Eurotium echinulatum Eurotium herbariorum Eurotium heterocaryoticum Eurotium niveoglaucum
268
TABLE 8.1. (continued) Fungal metabolite
Known producers
Flavoglaucin (continued)
Eurotium pseudoglaucum Eurotium repens Eurotium rubrum
Frequentin
Penicillium commune chemotype I & II Penicillium ierlikowskii
Fructigenine A
Penicillium vulpinum
Fulvic acid
Eupenicillium brefeldianum Eupenicillium ehrlichii Penicillium cremeogriseum Penicillium glabrum Penicillium griseofulvum var. griseofulvum Penicillium hirsutum var. allii Penicillium piscarium
Fumagillin
Aspergillus fumigatus Penicillium scabrosum
Fumigaclavine A, B, C
Aspergillus fumigatus
Fumigatin
Aspergillus fumigatus
Fumitremorgin A, B, C
Aspergillus caespitosus Aspergillus fumigatus Neosartorya fischeri var. fischeri Penicillium brasilianum Penicillium graminicola Penicillium mononematosum
Fumonisin B,, B,
Fusarium proliferatum Fusarium verticillioides
Fusarenone X
s0e 3-acetyldeoxynivalenol
Fusaric acid
Fusarium lateritium Fusarium oxysporum Fusarium solani Fusarium verticillioides Peziza atrovinosa
Fusarin C
Fusarium avenaceum Fusarium cerealis Fusarium culmorum Fusarium graminearum Fusarium oxysporum Fusarium poae Fusarium sambucinum var. sambucinum Fusarium sporotrichioides Fusarium tricinctum Fusarium verticillioides
269
TABLE 8.1. (continued) Fungal metabolite
Known producers
Fusarochromanone
Fusarium equiseti
Fusarubin
Fusarium solani Fusarium verticillioides
Fusidic acid
Acremonium fusioides Acremonium strictum Calcarisporium arbuscula Gabarnaudia tholispora lsaria kogane Mortierella ramanniana Verticillium lamellicola
Gallic acid
Phycomyces blakesleanus
Gentisylalcohol
see patulin
Gibberellic acid
Gibberella fujikuroi
Gladiolic acid
Penicillium gladioli
Glauconic acid
Penicillium purpurogenum Talaromyces assiutensis Talaromyces ohiensis Talaromyces panasenkoi Talaromyces trachyspermus
Gliotoxin
Aspergillus fumigatus Aspergillus terreus Aspergillus ustus Eurotium chevalieri Eurotium rubrum Gliocladium virens Penicillium adametzii Penicillium turbatum Rosellinia necatrix Trichoderma lignorum Trichoderma hamatum
Gregatins
Aspergillus panamensis Phialophora gregata
Griseofulvin
Khuskia oryzae Khuskia sacchari Penicillium aethiopicum Penicillium canescens Penicillium coprophilum Penicillium griseofulvum var. griseofulvum var . dipodomyimla Penicillium janczewskii Penicillium jensenii Penicillium lanosum
270
TABLE 8.1. (continued) Fungal metabolite
Known producers
Griseofulvin (continued)
Penicillium nodusitatum Penicillium raistrickii Penicillium sclerotigenum
Griseophenone C
see griseofulvin
Hadacidin
Byssochlamys nivea P enicillium camemberti Penicillium crustosum Penicillium glabrum Penicillium hispanicum Penicillium lividum P enicillium purpurescens Penicillium simplicissirnum Penicillium spinulosum Penicillium turbatum
Helminthosporin
Bipolaris cynodontis Drechslera catenaria Drechslera graminea Drechslera tritici-repentis
Helvolic acid
Aspergillus fumigatus Emericellopsis pusilla Emericellopsis terricola Gliocladium sp. Mammaria echinobotryoides Metarrhizium anisopliae Neosartorya aurata Sarocladium oryzae Stilbella eryihrocephala Verticillium epiphytum Verticillium lecanii
Hyalodendrin
Hyalodendron sp.
5'-hydroxyasperentin
see cladosporin
para-hydroxybenzoic acid
Eurotium echinulatum Lambertella corni-maris Penicillium griseofulvum var. griseofulvurn Polyporus tumulosus Rhizoctonia leguminicola Rhizoctonia solani
Hydroxyisocanadensicacid
see canadensolide
5-hydroxymaltol
Penicillium sp.
4-hydroxymellein
Apiospora camptosporas Aspergillus melleus Cercospora taiwanensis Lasiodiplodia theobromae
27 I
TABLE 8.1. (continued) Fungal metabolite
Known producers
o-hydroxVpachybasin
see pachybasin
Bp-hydroxyroridin E
Myrothecium roridum
8a-hydroxyverrucarinJ
Myrothecium verrucaria
2’-hydroxy-2’-(E)-verrucarinJ
Myrothecium roridum
lndoleacetic acid
Aureobasidium pullulans Cladosporium herbarum Epicoccum nigrum Fusarium spp.
lpomeamarones
Ceratocystis fimbriata Fusarium oxysporum (both on sweet poatatoes)
lslandicin
Penicillium islandicum
Isochromantoxin
Penicillium mononematosum Penicillium steckii
Isocochliodinol
Chaetomium murorum
lsoemodin lsomarticin
Neocosmospora sp.
Isororidin E
Cylindrocarpon sp. Myrothecium roridum Myrothecium verrucaria
lsosatratoxin H
Stachybotfys atra
ltalicic acid
Penicillium italicum
ltalinic acid
Penicillium italicum
ltalinic acid methylester
Penicillium italicum
Janthitrem B
Eupenicillium zonatum Penicillium piscarium
Javanicin
Fusarium solani Fusarium verticillioides
Kojic acid
Aspergillus flavus Aspergillus oryzae Aspergillus nomius Aspergillus parasiticus Aspergillus sojae Penicillium lanosum
Kotanin
Aspergillus clavatus Eurotium sp.
Lambertellin
Lambertella corni-maris Lambertella hicoricae Pseudospiropes simplex
212
TABLE 8.1. (continued) fungal metabolite
Known producers
Lapidosin
Eupenicillium lapidosum
Leucinostatins
Paecilomyces silacinus
Lichexanthone
Penicillium griseofulvum var. griseofulvum
Luteoskyrin
Penicillium islandicum
Macrosporin
Alternaria porri Alternaria solani
Malformin C
Byssochlarnys nivea Aspergillus niger Thielava sepedonium
Maltoryzin
Aspergillus flavus
Marticin
see javanicin
Meleagrin
Penicillium chrysogenum Penicillium conferturn Penicillium coprophilum Penicillium glandicola var. glandicola var. glaucovenetum Penicillium hirsuturn var. albocoremiurn
Melinacidins
Chaetomium retardaturn Verticillium cinnabarinum Verticillium tenerum
6-methoxymellein
Aspergillus caespitosus Penicillium thomii Sporormia affinis Sporormia bipartis
Methoxysterigmatocystin
see sterigmatocystin see cyclopenin
3-methoxyviridica tin
Methylhydroquinone
Nectria erubescens
6-methylsalicylic acid
see patulin
Mevinolin
Aspergillus terreus Monascus purpureus Monascus ruber
Mitorubrin
Hypoxylon fragiforme Penicillium crateriforme Talaromyces flavus Talaromyces rnacrosporus Talaromyces rnimosinus Talaromyces udagawae Talaromyces wortmannii
Mitorubrinic acid
s0e mitorubrin
273
TABLE 8.1. (continued) Fungal metabolite
Known producers
Mitorubrinol
see mitorubrin
Mitorubrinol acetate
see mitorubrin
Mollicellins
Chaetomium amygdalisporum Chaetomium mollicellum
Mollisin
Mollisia caesia Mollisia gallens
Moniliformin
Fusarium anthophilum Fusarium avenaceum Fusarium chlamydosporum Fusarium oxysporum Fusarium proliferaturn Fusarium sacchari Fusarium verticillioides
Mono-methoxycurvulinicacid
see curvulinic acid
Monorden
Cylindrocarpon destructans Monocillium nordinii Penicillium resedanum Verticillium chlamydosporum
Mycelianamide
Penicillium griseofulvum var. griseofulvum
Mycochromenic acid
Penicillium brevicompacturn
Mycophenolic acid
Leptographium abientinum Penciillium brevicompactum Penciilliurn raciborskii Penicillium roqueforti var. roqueforti var. carneum Phaerosphaeria nodorum
Myrotoxin A, B, C, D
Myrothecium roridum
Myioxin A, B, C
Myrothecium roridum
Naphthalic anhydride
Aspergillus silvaficus Godronia cassandrae Penicillium herquei Roesleria pallida
"Naphthoy-quinones, toxic"
Aspergillus carbonarius Aspergillus niger
Nalgiolaxin
Penicillium nalgiovense
Nalgiovensin
Penicillium nalgiovense
Nectriafurone
Fusarium solani
Neocochliodinol
Chaetomium amydalisporum
214
TABLE 8.1. (continued) Fungal metabolite
Known producers
Neoxaline
Aspergillus aculeatus
Neosolaniol
Fusarium acuminatum Fusarium sporotrichioides
Nidulin
Emericella unguis
Nidulotoxin
Aspergillus sydowii Aspergillus versicolor Emericella nidulans
Nigragillin
Aspergillus niger
P-nitropropionic acid
Arthrinium phaeospermum Arthrinium sacchari Arthrinium saccharicola Aspergillus avenaceus Aspergillus flavus Aspergillus oryzae Aspergillus wentii Penicillium atrovenetum
Nivalenol
see 3-acetyldeoxynivalenol
Nominine
Aspergillus nomius
Norjavanicin
see javanicin
Norlichexanthone
see griseofulvin
Norsolorinic acid
see sterigmatocystin
Nortryptoquivaline
Aspergillus clavatus Aspergillus furnigatus
Ochratoxin A,
B, C
Aspergillus ochraceus Aspergillus melleus Aspergillus petrakii Aspergillus sclerotiorum Aspergillus fresenii Penicillium verrucosum chemotype I & II Petromyces alliacea
Oosporein
Acremonium sp. Beauveria bassiana Chaetomium aureum Chaetomium trilaterale Phlebia albida Phlebia mellea Verticillium psalliotae
Orsellinic acid
widespread precursor
275
TABLE 8.1. (continued) Fungal metabolite
Known producers
Oxalic acid
Aspergillus niger Penicillium oxalicum Penicillium verrucosum Whetzelinia sclerotiorum and many other fungi
Oxaline
Penicillium atramentosum Penicillium aurantiogriseum var. melanoconidium Penicillium coprophilum Penicillium glandicola var. glandicola Penicillium oxalicum Penicillium vulpinum
Pachybasic acid
see pachybasin Aspergillus crystallinus Trichoderma viride
Pachybasin Palitantin
Eupenicillium brefeldianum Eupenicillium ehrlichii Penicillium commune Penicillium echinulatum var. echinulatum var. discolor Penicillium solitum
Paraherquamide
Penicillium brasilianum Penicillium charlesii
Parasitic01
Aspergillus nomius Aspergillus parasiticus
Paspaline
Aspergillus clavatoflavus Aspergillus leporis Claviceps paspali
Paspalinine
Aspergillus flavus Claviceps paspali
Paspalitrems
Claviceps paspali
Patulin
Aspergillus clavatus Aspergillus giganteus Aspergillus terreus Byssochlamys fulva Byssochlamys nivea Paecilomyces variotii Penicillium clavigerum Penicillium coprobium Penicillium expansum
216
TABLE 8.1. (continued) Fungal metabolite
Known producers
Patulin (continued)
Penicillium glandicola var. glandicola var. glaucovenetum Penicillium griseofulvum var. griseofulvum var. dipodomyicola Penicillium melinii Penicillium novae-zeelandiae Penicillium selandiae Penicillium vulpinum
Paxillin
Acremonium loliae Aspergillus clavatoflavus Emericella desertorum Emericella striafa Eupenicillium tularense Penicilliurn paxilli
PD 113,325
Myrothecium roridum
Pebrolide
Penicillium brevicompactum
Penicillic acid
Aspergillus ochraceus Aspergillus auricomus Aspergillus fresenii Aspergillus melleus Aspergillus ostianus Aspergillus sderotiorum Eupenicillium baarnense Eupenicillium ehrlichii Paecilomyces lilacinus Penicilliurn aurantiogriseurn var. aurantiogriseurn var. melanoconidium var. neoechinulatum var. polonicum var. viridicatum Penicillium brasilianum Penicillium fennelliae Penicillium hirsuturn var. albocoremium Penicillium janczewskii Penicillium matriti Penicillium megasporurn Penicillium pulvillorum Penicillium raistrickii Penicillium rolfsii Penicillium roqueforti var. carneum Petromyces alliacea
277
TABLE 8.1. (continued) Fungal metabolite
Known producers
Penicillin G
Acremonium chrysogenum Aspergillus caespitosus Emericella nidulans Penicillium chrysogenum var. chrysogenum var. dipodomyis Penicillium matriti Penicillium turbatum
Penitrem A, 8, C, 0,E, F
Penicillium aurantiogriseum var. melanoconidium Penicillium clavigerum Penicillium crustosum Penicillium glandicola var. glandicola var. glaucovenetum Penicillium hirsutum var. albocoremium Penicilliumjanczewskii
Phoenicin
Eupenicillium cinnamopurpureum Penicillium chermesinum Penicillium crateriforme Penicillium atrosanguineum
Phomarin
Phoma foveata
Phomopsins
Phomopsis lepstromiformis
Physcion
Achaetomium cristalliferum Alternaria porri Aspergillus wentii Caloplaca murorum Cetraria cullulata Dermocybe cinnabarine Eurotium acutum Eurotium amstelodami Eurotium appendiculatum Eurotium carnoyi Eurotium chavalieri Eurotium cristatum Eurotium echinulatum Eurotium glabrum Eurotium herbariorum Eurotium intermedium Eurotium leucocarpum Eurotium niveoglaucum Eurotium pseudoglaucum Eurotium repens Eurotium rubrum Eurotium spiculosum
278
TABLE 8.1. (continued) Fungal metabolite
Known producers
Physcion (continued)
Eurotium tonophilum Penicillium herquei Physcia sp. Xanthoria fallax Xanthoria mandschurica
PI-3
Penicillium italicum
PR-toxin
Penicillium roqueforti var. roqueforti
PR- 1636
Aspergillus candidus Aspergillus ustus
Preechinulin
Eurotium amstelodami Eurotium chevalieri Eurotium repens
Protophomin
Phoma exigua var. exigua
Proxiphomin
Phoma exigua var. exigua
Puberulonic acid
Penicillium aurantiogriseum var. aurantiogriseum
Purpurogenone
Penicillium purpurogenum
Pyrichalasin H
Pyricularia grisea
Pyrogallol
Penicillium griseofulvum
2-pyrovo ylaminobenzamide
Alternaria citri Fusarium culmorum Neosartorya fischeri var. spinosa Penicillium chrysogenum
Questin
Aspergillus terreus Chrysosporiurn merdarium Dermocybe cinnarnomeolutea Eurotium cristatum Eurotium glabrum Eurotium repens Eurotium rubrum Monascus ruber Penicillium glabrum
Questinol
see questin
Ravenelin
Bipolaris ravenelii
Regulin
Aspergillus restrictus
Restrictocin
Aspergillus restrictus
Roquefortine A, B
Penicillium roqueforti
219
TABLE 8.1. (continued) Fungal metabolite
Known producers
Roquefortine C
Penicillium atramentosum (trace? Penicillium aurantiogriseum var. melanoconidium (trace) Penicillium chrysogenum var. chrysogenum Penicillium confertum (trace) Penicillium coprobium (trace) Penicillium coprophilum Penicillium crustosum Penicillium expansum Penicillium glandicola var . glandicola var glaucovenetum Penicillium griseofulvum var. griseo fulvum Penicillium hirsutum var. hirsutum var. albocoremium var. allii var. hordei var. venetum Penicillium oxalicum (trace) Penicillium roqueforti var. roquefodi var. carneum Penicillium sclerotigenum Penicillium vulpinum
.
Roquefortine D Roridin A
Roridin D
see roquefortine C Cryptomela acutispora Cylindrocarpon sp. Dendrodochium toxicum Myrothecium roridum Myrothecium verrucaria Phomposis lepstromiformis Cryptomela acutispora Cylindrocarpon sp. Myrothecium roridum Myrothecium verrucaria
Roridin E
Myrothecium roridum Myrothecium verrucaria Stachybotrys atra Stachybotrys karnpalensis
Roridin t i
Cylindrocarpon sp. Myrothecium roridum Myrothecium verrucaria
Roridin J
Myrothecium verrucaria
280
TABLE 8.1 . (continued) Fungal metabolite
Known producers
Roridin K acetat
Myrothecium verrucaria
Roritoxin A, B, C, D
Myrothecium roridum
Roseopurpurin
Penicillium roseopurpureum
Roseotoxin B
Trichotheciurnroseum
Rubratoxin A, B
Penicillium crateriforme
Rugulosin
Acroschyphus sphaerosporoides Cryphonectria parasitica Endothia coccolobii Endothia fluens Endothia gyrosa Endothia japonica Endothia macrospora Endothia viridistroma Penicillium brunneum Penicillium concavorugulosum Penicillium islandicum Penicillium piceum Penicillium rugulosum Penicillium tardum Penicillium variabile Sepedonium ampullosporum Talaromyces rotundus Talaromyces wortmannii
Rugulovasine A
Gloeophyllum trabeum Pellicularia filamentosa Penicillium atramentosum Penicillium corylophiloides Penicillium commune Penicillium concavorugulosum Penicillium crateriforme Pulcherricium caeruleum
Satratoxin F, G, H
Stachybotrys albipes Stachybotrys atra Stachybofrys kampalensis Stachybotrys microspora
Scytalidine
Scytalidium album
Scytalone
Penicillium aurantiogriseum Phialophora lagerbergii Scytalidium album Thielaviopsis basicola Verticillium dahlia8
28 I
TABLE 8.1. (continued) Fungal metabolite
Known producers
Secalonic acid A
Aspergillus ochraceus Claviceps purpurea Parmelia entotheiochroa Phoma terrestris
Secalonic acid B
Aspergilus aculeatus Claviceps purpurea
Secalonic acid D
Aspergillus aculeatus Claviceps purpurea Penicillium isariiforme Penicillium oxalicum
Secalonic acid F
Aspergillus aculeatus Claviceps purpurea
Shikimic acid
widespread precursor
Simatoxin
Penicillium islandicum
Sirodesmin
Sirodesmium diversum
Skyrin
Acroscyphus sphaerophoroides Cryphonectria parasitica Endothia fluens Endothia gyrosa Endothia havananensis Endothia japonica Endothia longirostris Endothia rnacrospora Endothia radicalis Endothia singularis Endothia tropicalis Hypomyces lactifluorurn Hypomyces trichothecioides Penicilliopsis clavariaeforrnis Penicillium brunneum Penicillium concavorugulosum Penicillium islandicum Penicillium piceum Penciillium rugulosum Penicillium variabile Physcia obscura var. endococcina Pyxine endochrysina Sepedonium ampullosporum Trypetheliopsis boninensis
Slaframin
Rhizoctonia leguminicola
Solaniol
see fusarubin
Soranjidiol
282
TABLE 8.1. (continued) Fungal metabolite
Known producers
Sorbicillin
Penicillium chrysogenum Verticillium intertexturn
Spiculisporic acid
Penicillium crateriforme Penicillium minioluteum Talaromyces panasenkoi Talaromyces trachyspermus
Spinulosin
Aspergillus fumigatus Penicillium spinulosum
Sporidesmin
Leptosphaerulina chartarum Pithomyces chartarum
Steckiin
Penicillium steckii
Sterigmatocystin
Aspergillus flavus Aspergillus multicolor Aspergillus nomius Aspergillus parasiticus Aspergillus versicolor Bipolaris nodulosa Chaetomium thielavioideum Chaetomium udagawae Emericella acristata Emericella aurantiobrunnea Emericella bicolor Emericella cleistominuta Emericella corrugata Emericella dentata Emericella echinulata Emericella falconensis Emericella foveolata Emericella heterothallica Emericella lata Emericella navahoensis Emericella nidulans Emericella parvathecia Emericella purpurea Emericella quadrilineata Emericella rugulosa Emericella spectabilis Emericella striata Emericella unguis Emericella variecolor Farrowia malaysiensis Monocillium nordinii
Stipitatic acid
Talaromyces stipitatus
283
TABLE 8.1. (continued) Fungal metabolite
Known producers
Sulochrin
Aspergillus fumigatus Aspergillus terreus Aspergillus wentii Oospora sulphurea-ochracea Penicillium glabrum
Sydowic acid
Aspergillus sydowii
T-2 toxin
Fusarium acuminatum Fusarium poae Fusarium sporotrichioides
Talaromycins
Talaromyces stipitatus
Tenuazonic acid
Alternaria alternata Alternaria brassicae Alternaria brassicicola Alternaria cheiranti Alternaria citri Alternaria japonica Alternaria kikuchiana Alternaria longipes Alternaria mali Alternaria oryzae Alternaria porri Alternaria raphani Alternaria solani Alternaria tenuissima Aspergillus nomius Phoma sorghina Pyricularia oryzae
Terphenyllin
Aspergillus candidus
Terrecyclic acid
Aspergillus terreus
Terreic acid
Aspergillus terreus
Terrein
Aspergillus terreus Neosartorya fischeri var. spinosa Penicillium soppii
Terrestric acid
Penicillium aurantiogriseum var. aurantiogriseum chernotype I Penicillium crustosum Penicillium hirsutum var. hirsutum var. albocoremium var. hordei var. venetum Pyricularia oryzae
Terretonin
Aspergillus terreus
284
TABLE 8.1. (continued) Fungal metabolite
Known producers
Territrerns
Aspergillus terreus Penicilliurn echinulaturn var. echinulaturn
para- toluquinone
Nectria erubescens
Torreyol
Clitocybe illudens
TR-2
see verrucologen
Trichoderrnin
Trichoderma viride
TrichorzianinesA
Trichoderma haRianUm
Trichorzianines B
Trichoderma harzianum
Trichothecolone
Trichotheciurnroseum
3,4,5-trihydroxy-7-rnethoxy 2-methylanthraquinone
Alternaria porri
Trypacicin
Aspergillus furnigatus Aspergillus ochraceus Neosartorya fenneliae
Tryptoquivalines
Aspergillus furnigatus Neosartorya aureola Neosartorya fischeri var. fischeri var. glabra chemotype 111 var. spinosa Penicillium aethiopicurn Penicilliurn digitaturn
Tryptoquivalones
see tryptiquivalins
Tubingensin A, 6
Aspergillus niger
Ustic acid
Aspergillus puniceus Aspergillus ustus
Verrnicellin
Penicillium aculeaturn Penicillium panamense Talarornyces flaws
Verrniculin
Penicilliurn crateriforrne Penicilliurn pinophilurn Talarornyces flaws Talarornyces ohiensis
Verrucarin A
Dendrodochiurn toxicurn Myrotheciurn leucotrichurn Myrotheciurn roridum Myrotheciurn verrucaria
Verrucarin 6
Myrotheciurn roridurn Myrotheciurn verrucaria Stachybotrys atra
285
TABLE 8.1. (continued) Fungal metabolite
Known producers
Verrucarin J
Myrothecium roridum Myrothecium verrucaria Stachybotrys albipes Stachybotrys atra Stachybotrys kampalensis Stachybotrys microspora
Verrucofortine
Penicillium aurantiogriseum var. aurantiogriseum var. polonicum var. viridicatum
Verrucolon
Penicillium verrucosum
Verrucosidin
Penicillium aurantiogriseum var. aurantiogriseum var. melanoconidium var. polonicum
Verrucologen
Aspergillus caespitosus Aspergillus fumigatus Eupenicillium crustaceum Penicillium brasilianum Penicillium graminicola Penicillium mononematosum
Verruculotoxin
Penicillium brasilianum
Versicolorin A
Aspergillus flavus Aspergillus multicolor Aspergillus parasiticus Aspergillus puniceus Aspergillus ustus Aspergillus versicolor Drechslera sorokiniana Emericella nidulans
Verticillin A
Verticillium sp.
Vertinolide
Verticillium intefiextum
Vertisporin A
Verticimonosporium diffractum
Violaceic acid
Emericella striata
Viomellein
Aspergillus alutaceus Aspergillus auricomus Aspergillus melleus Aspergillus ostianus Aspergillus sulphureus Eupenicillium javanicum
286
TABLE 8.1. (continued) Fungal metabolite
Known producers
Viomellein (continued)
Penicillium aurantiogriseum var. aurantiogriseum chernotype II var. melanoconidium var. viridicatum Penicillium clavigerum Penicillium mariaecrucis Penicillium simplicissimum Nannizzia cajetanii Trichophyion megninii Trichophyton rubrum Trichophyton violaceurn
Viridamine
Penicillium aurantiogriseum var. viridicatum
Viridic acid
Penicillium aurantiogriseum var. viridicatum
Viridicatic acid
see terrestric acid
Viridicatin
see cyclopenin
Viridicatol
see cyclopenin
Viridicatumtoxin
Penicillium aethiopicurn Penicillium brasilianum Neosartorya fennelliae
Viriditoxin
Aspergillus viridi-nutans Neosartorya aureola Paecilomyces variotii Penicillium mononematosum
Wentilacton
Aspergillus wentii
Wortmannin
Aspergillus janus Fusarium sarnbucinum var. coeruleum Myrothecium roridum Penicillium proteolyticum Talaromyces flavus
Xanthoascin
Aspergillus candidus
Xanthocillin X
Eupenicillium egyptiacum Eurotium chevalieri Neosartorya spathulata Penicillium chrysogenum Penicillium italicurn
Xanthomegnin
see viomellein
Zearalenol
see zearalenone
287
TABLE 8.1. (continued) Fungal metabolite
Known producers
Zearalenone
Fusarium cerealis Fusarium culmorum Fusarium equiseti Fusarium graminearum Fusarium pallidoroseum
Zygosporin 0,E, F, G
Zygosporium masonii
' List of references on data presented in this tabel can be obtained from the authors upon request
* Presumably a fungal metabolite and/or unknown producer Production of roquefortine C in trace amounts
8.2. COLUMN CHROMATOGRAPHY In most cases column chromatography has been applied for large scale separation of standards of mycotoxins, for analytical standards, or for toxicological testing (41-44). The principles of low, medium or high pressure liquid chromatography are quite analogeous and one of the most important principles in these techniques for producing pure analytical standards is the combination of different types of columns. Of the four most important principles in column chromatography, adsorption, partition, ion-exchange and gel filtration, the first is most widely used. lon-exchange chromatography is suited for acidic mycotoxins like P-nitropropionicacid, ochratoxin A, citrinin, penicillic acid, terrestric acid, secalonic acid, cyclopiazonic acid, fumonisins, mycophenolic acid, viridicatumtoxin and tenuazonic acid, but may also be used for ionic mycotoxins like moniliformin or basic mycotoxins like roquefortine C and ergot alkaloids. These ionic molecules have been quite difficult to analyze using common TLC or HPLC systems, e.g. some acids (secalonic acid, viridicatumtoxin) or bases (roquefortine C. meleagrin, oxaline) will not elute on silica gel using the common TLC eluent toluene/ethyl acetate/formic acid (45) or they will show tailing spots (cyclopiazonic acid, citrinin, rugulosin, luteoskyrin). Other acidic (terrestric acid) or basic (roquefortine C, meleagrin, oxaline) fungal secondary metabolites will give broad
288
peaks, even in acidic HPLC eluents (42-48). As many filamentous fungi are able to produce either acidic or basic metabolites or both (16, 19-27, 31, 49) this fact should be considered carefully when extracting and purifying secondary metabolites from fungal cultures or foods. It is preferable to use two different types of columns (especially normal phase
followed by reversed phase column material) instead of two consequetive elutions on the same column material with different eluents. Column material of many different kinds are now available among them silica, silicic acid, alumina, magnesium oxide, magnesium silicate, calcium hydrogenphosphate, calcium sulphate, charcoal, diatomaceous earth, cellulose, and silica bonded with non-polar phases e.g. phenyl, cyclohexyl, octyl (CJ, octadecyl (C,J, ethyl (C,); polar phases, i.e. diol, cyanopropyl (CN), aminopropyl, N-propylethylenediamine or
ion-exchange phases, e.g.
benzenesulphenylpropyl, trimethylaminopropyl and carboxymethyl. Multi-mycotoxin methods may require a combination of two extraction methods, one for polar mycotoxins, usually using methanol/water or ethanol/water and one for non-polar mycotoxins usually chloroform, dichloromethaneor ethylacetate. The polar fraction will also contain carbohydrates, amino acids, organic acids, purines and pyrimidines etc., while the non-polar fraction will contain lipids such as mono-, di- and tri-glycerides, sterols, waxes and phospholipids, carotenes etc. The polar compounds may be fractionated by different ion-exchangers(50) and many lipids in non-polar solvents may be removed for example by partition between methanol (after evaporation of first extraction solvent and redissolving in the alcohol) and hexane. Most known fungal secondary metabolites can be extracted by chloroform or ethyl acetate in an acidified system, but important compounds such as the fumonisins and P-nitropropionic acid will remain in the water/alcohol phase (17). Much more specific extraction and purification procedures may be selected for a single rnycotoxin or a group of chemically related mycotoxins. Flash column chromatography has been used for the purification and separation of trichothecene mycotoxins (51, 52) Dry column chromatography has also been used instead of preparative TLC. Much less organic solvent is used and the individual toxins can be cut as slices of the columns. In a separation of aflatoxin B,,B, G, and G, Megalla (53) used a layer of neutral alumina (5 cm) and silica gel (25 cm) in a cellophane bag. A similar method was used by McKinney (54) for the purification of aflatoxins. The use of polythene bags
289
also allows a chemical confirmation by dipping the columns in reagent solutions such as mineral acids (55). 8.3. MINI-COLUMN CHROMATOGRAPHY Mini-columns have been used for clean-up of many mycotoxins and they are part of some of the analytical methods published by the AOAC (Association of Official Analytical Chemists)(56). The Romer and Holaday-Velasco mini-columns are, for example, packed with calcium sulphate, florisil, silica gel, neutral alumina and calcium sulphate (5657). Often these procedures are necessary because of the high content of lipid in the foods and feeds they are used for, e.g. groundnuts, corn and milk, or because of coloured interfering compounds. The many types of pre-packed cartridges with the different types of adsorbents mentioned above have raised the efficiency, repeatability and quality of clean-up procedures and the final analytical result. A combination of these mini-columns may be used for a very efficient clean-up of different types of mycotoxins, especially from complex matrices such as foods, feedstuffs, blood and urine. Mini-columns are used in many methods for aflatoxins including aflatoxin M,. Both silica gel (58-73) and reversed phase (66,68,74-80), florisil (70,81) and gel-permeation (82) mini-columns have been used. Often silica gel rinse up involve application of a chloroform or dichloromethane fraction to a column (often hexane-solvated), washing with hexane or diethyl ether and eluting with strong eluents such as chloroform or dichloromethane
- ethanol or acetone mixtures. Differenttypes of (mini)-columnshave also been used for other mycotoxins such as sterigmatocystin (cupric carbonat
- diatomaceous earth, florisil and polyamide,
69,83), trichothecenes (silica gel (84), reversed phase (85-89), cyano (71,go), charcoal (91), florisil (92) or other phases (93)), zearalenone (amino (94) or florisil (95-96)), ochratoxin A (cupric carbonate - diatonaceous earth (70), cyano (81), reversed phase (97) or XAD-2 (98)) and moniliformin (amberlite IRC50 (99)). Combinations or sequences of charcoal, alumina, florisil, CN and C, or C,, mini-columns have been used for an efficient clean-up of primarily aflatoxins and trichothecenes from biological material (40, 100-103).
290
A very simple and efficient method for the determination of aflatoxin M, in milk was based on immunoaffinity column clean-up (104) and this kind of method is especially suited for a single mycotoxin often giving very "clean chromatograms" (105-106). lmmunoaffinitycleanup has also been applied to aflatoxin B, (107-108) and ochratoxin A (109) analysis. There is little doubt that a large number of future mycotoxin analyses in foods and feeds will involve clean-up with disposable mini-columns. It is not clear yet which clean-up methods are most efficient in multi-mycotoxin methods. The sample may be subdivided into for example acidic, neutral/polar, neutraVapolar and basic fractions at all stages and extracted, cleaned-up, separated and detected accordingly. This approach may be an advantage for detectors such as diode array detectors (DAD) or mass selective detectors (MSD) (see later). However in many cases only a single analytical procedure is feasible and an analytical compromise between chemically quite different mycotoxins is necessary. More research is needed to evaluate which types of minicolumns and eluents are best in separating co-occurring mycotoxins from the background matrix in one general procedure. 8.4. HIGH PERFORMANCE LIQUID CHROMATOGRAPHY 8.4.1. Aflatoxins Determination of aflatoxins by HPLC has been reviewed extensively by Scott (110) Shepherd (37), Betina (39), Coker and Jones (40),Beaver (111) and Shepherd
ef a/. (1 12) and determination of aflatoxin M, by Scott (1 13), so this topic will not be reviewed extensively here. However some important applications of HPLC in aflatoxin analysis will be summarized below. 8.4.1 . I . HPLC solvents and sample introduction. It is generally recognized that the injection solvent should be close or equal to the eluent. In the case of aflatoxin M, (convertedto the more fluorescing aflatoxin M2,), Beaver
(1 14)
has
shown
that
in
reversed-phase
(RP)-HPLC
using
water/isopropanoI/acetonitrile as the eluent, the aflatoxin M, peak exhibited 25.000
theoretical plates when injected dissolved in 30% aqueous acetonitrile and 10.000 plates when injected dissolved in water alone, compared to for example only 2000
29 1
plates when injected into a water/ acetonitrile/methanoIeluent in a 30% acetonitrile in water. solution. This clearly shows that careful choice of sample solvent can improve the quality of an analysis considerably. In broader analysis, involving several chemically different mycotoxins the choice of solvent are more restricted because of poor solubility in for example water. Also degradation of aflatoxins depend heavily on the solvent used (1 15). Exposure to light in solvents without acetic acid resulted in significant aflatoxin degradation and even at -1 8 "C, aflatoxin degraded when acetic acid was absent (1 15). An optimally stable solvent was acetonitrile/water/aceticacid
(50:50:0.5),but aflatoxins were also stable in crude peanut extracts (1 15). Our experience with crude fungal extracts dissolved in methanol and stored at -1 8 "C has been very good, probably because of "co-stabilization" caused by the many different secondary metabolites in one extract (Frisvad and Thrane, unpublished). A new method for aflatoxin M, in milk, which can be automated, using on-line dialysis and trace enrichment on a RP column, and back-flush to the analytical column, has been developed by Tuinstra et a/. (1 16-1 17). The recovery was over 50% at the 50 ng/kg level in automated analysis and is proposed for automated screening of large
numbers of milk samples at 100 ppt level or higher. The combination of flow injection analytical (FIA) methods and HPLC in the analysis of mycotoxins may be the basis of several future methods. The FIA technique may be used for detection of the total of several related mycotoxins (e.g. total aflatoxins) and as a post-column reactor/detector (1 18-119). 8.4.1.2. Analytical separation of aflatoxins.
A large number of column types and materials have been used for an efficient separation of the aflatoxins. In most applications normal phase (NP) or reversed phase (RP) silica gel based materials in 10-30 cm long columns have been used, but rapid methods using radial compression columns have also been applied with success for separation of the aflatoxins (1 20-1 21). Effective separation of the aflatoxin have been achieved both by using NP- and RP-HPLC. A major problem of the latter methods are that eluents always contain water, which will quench the fluorescence of aflatoxin B, and G,. Therefore pre- or post-column reactions or silica gel-filled detector cells are often a part of such systems (see below). NP-HPLC systems have been used in several methods (e.g. 59, 122-142). One advantage of the NP systems is that
292
transformation to water adducts (from B,, G, and M, to B,
G, and M ,, respectively)
is avoided, but the chloroform/cyclohexane/acetonitrile/isopropanol(73:22:3:2.5) eluent most often used (134) may cause separation problems because of chloroform humidity fluctuations and isopropanol content (142). Tutelyan et a/. (141-142) suggested a low viscosity eluent, ether/methanol/water(95:4:1) causing better separations and shorter retention times than the chloroform containing eluent mentioned above. The method of Tutelyan eta/. (141-142) allows determination of both B,, B, G,, G, and the more polar M,, M ,,
and B,= Shorter retention times of the latter polar aflatoxins could be
obtained with a more polar eluent ether/methanol/water (90:8:2). The stability of dynamically modified silica gel (much less dependent on brand
of stationary phase) can be used to reduce inter-laboratory performance differences (143). Unfortunately the equilibration time can be 12 to 18 hours, but after further
method development these methods may be of great value in the development of highly reproducible mycotoxin analysis. The number of applications of RP-HPLC methods in aflatoxin analysis is now much greater than the number of NP applications (121, 123-124, 126, 129, 132, 138, 144-191). These methods often involve use of acetonitrile and/or methanol and water,
in isocratic or gradient elutions. Because of water quenching of aflatoxin B, and G, fluorescence these are often treated with acid to give the B, and G,
derivatives.
However, B, and G, derivatives are unstable in methanol (191), which is a very much used as part of many eluents. Among eluents used, the most common are
water/acetonitrile/methanoImixtures (75, 126, 138, 144, 146-148, 151-152, 156, 162163, 167-168, 170, 172, 175, 180, 182, 186-187, 190), water/acetonitrile (121, 146, 148, 155, 159-160, 165, 169, 171, 174) and water/methanol (148, 153-154, 157-158, 173, 179, 185, 188). Some of these eluents are added acetic acid (121, 171, 186),
phosphate buffers (182, 188) or sodium chloride (152, 167). Few of these analysis are gradient elutions (4, 47, 156-157, 165). The gradient elutions are required in multimycotoxin analysis, but excellent separation of the aflatoxins have been achieved with the isocratic eluents refered to above. Injection with a water/acetonitrile soluent and elution with either water/acetonitrile or water/acetonitrile/methanol seem to be the best choices for general aflatoxin analysis. The reversed phase column brand seems to be less important, however a new column should always be tested with a mixture of the important aflatoxins and separation optimized by minor adjustments in the eluent
293
composition. Flow rate is mostly dictated by optimization due to a common wish for short analysis time (high flow rate), use of as little eluent as possible (e.g. microbore columns) and possible interface to a mass selective detector, which usually requires low flow rates. In the applications above flow rates from 0.5 to 3 ml/min have been used. 8.4.1.3. Detection of aflatoxins. Most detectors for aflatoxins take advantage of the strong fluorescence, but if aflatoxins are present in more than trace levels UV detection is an alternative or both methods can be used for confirmation of identity. Diode array detection will reveal the very characteristic UV spectra of the aflatoxins (4,47), but at a cost of sensitivity. Fluorescence detection of aflatoxins in NP-HPLC applications was intially used by e.g. Hsieh el a/. (193) and Johnson eta/. (194). Manabe eta/. (195) examined the influence of different eluents on fluorescence quenching and recommended an eluent containing toluene, ethyl acetate, formic acid and methanol. Panalaks and Scott (124) and Zimmerli (149-150) developed silica gel-packed flow cells for the sensitive detection of aflatoxins for NP-HPLC applications, because aflatoxin B, and B, fluoresce poorly compared to GIand G,in NP systems (196-197). The methods of Panalaks and Scott (124) has since been used by Pons (130) and Thean etal. (59) for determination of aflatoxins in corn and Francis etal. (134) for determination of aflatoxins in peanut butter. The major disadvantage of the silica gel-packed flow cells is the instability of the packed cell with time (191). The fluorescence of aflatoxin B, and G, is poor in aqueous systems (used in RP-HPLC systems), so treatment with hydrochloric acid (167) or trifluoroacetic acid (TFA) to convert aflatoxin B, and G, to their B, recommended. Aflatoxin B, and
and G ,,
derivatives has been
G, are left unaltered. Diebold and co-workers (144-
145) proposed hydrochloricacid as the derivatisation reagent, because TFA according to them caused additional unidentified peaks and they used a very sensitive laser
fluorometric detection method for the aflatoxins. However in most cases TFA has been used for the formation of highly fluorescent hemiacetal derivatives (62, 75, 121, 127, 138, 147-148, 151, 156, 159-160, 162, 164, 166, 168, 171, 178, 186, 187, 198-199).
Davis and Diener proposed that iodine could be used for post-column derivatisation of aflatoxins for enhanced flourescence (158). This principle is now used
294
in several methods for aflatoxin using either iodine (174, 176, 180, 182, 165,200-204), bromine (1 19, 181) or chloramine (1 19) as the oxidizing agent. The methods based on TFA or iodine require an extra chemical conversion step and an extra pump respectively, but at this stage these two methods appear to be the best available RPHPLC methods concerning accurate and sensitive detection of aflatoxins (189). Cyclodextrin has also been used to enhance the fluorescence of the aflatoxins (205) and synchronous fluorescence spectrophometry has been introduced (206). Another method such as electrochemical detection (207) may also be basis for new methods, even though the latter method is not particularily sentitive. HPLC methods have been compared to ELSA (enzyme linked immuno sorbent assay) techniques and for screening purposes (89, 189, 203, 208-210). The ELISA techniques appear to be very good and simple, but they compared most favorably to HPLC methods with lipid containing products (corn, nuts, peanuts) and less favorably for cereals and grain samples (210). Immunological methods seem to be of use both in the clean-up phase (211) and in the final confirmation phase, but use of ELISA cards is also as very simple screening technique, which can be used in "the field". 8.4.1.4. Aflatoxin determination in different products
Most methods for aflatoxins have been developed for their determination in lipid containing products such as corn (59, 121, 130, 136, 159, 175, 182, 185, 187, 192, 204,212), cottonseeds (131,186,214-215) and peanuts, peanut butterand nuts (121, 125, 134, 151, 152, 154, 175, 180, 182, 187, 203-204, 209-210, 212, 215). However
methods have also been proposed for
COCO
beans (167), feedstuffs (126, 128, 159,
174, 189, 210, 217), eggs (216), wine (147-148),soy products (132) and spices (131).
Finally methods have been proposed for human and animal tissue both for metabolic and medical studies, but also for the analysis of meat (76, 135, 157, 164, 166, 218), serum (138, 168, 173, 188) and urin (137, 165, 168, 190). Studies on milk and milk products have been reviewed extensively by Scott (89) and include ref. 75, 139-140, 155-156, 160, 163, 169-172 and 219. Of particular interest is the new hydroxy derivative of aflatoxin M,, aflatoxin M, (220-221). This metabolic product is more carcinogenic in rainbow trout than aflatoxin B, or M, and more emphasis should be given to analytical methods forthis important new derivative.
295
8.4.2. Sterigmatocystin and related compounds Sterigmatocystin is a carcinogenic mycotoxin produced by a series of both related and quite unrelated fungi (Table 8.1). Fortunatelyonly three species (Aspergilus flavus, A. nomius and A. parasiticus) have the enzymes needed for the production of the next biosynthetic steps towards aflatoxin G, (via methoxy-sterigmatocystin and aflatoxin B,).There is an interest in sterigmatocystin both because it is a precursor of aflatoxin nut also because it is produced by common fungi in foods, notably Aspergillus versicolorand Emericella nidulans. For the first objective analysis of other biosynthetic intermediates are also interesting and the HPLC methods should be developed accordingly. For the second objective HPLC methods may be more directed towards very sensitive and specific analysis, even though sterigmatocystin is among the mycotoxins which quite often have been included in multi-mycotoxin HPLC methods. Few NP- and several RP-HPLC applications have been developed for the analysis of sterigmatocystinin foods and feedstuffs and fungal cultures (205,222-235). Furthermore several multi-mycotoxin methods including sterigmatocystin have been proposed, either comprising both related and unrelated mycotoxins (4, 47, 236-238) or secondary metabolites biosyntheticallyrelated to sterigmatocystin (239-243). Some of the early methods employed NP separations using eluents containg two or more of the following eluents hexane, chloroform, dichloromethane, ethyl acetate, n-propanol combined with acetic acid (e.g. 239-240) after silica gel (240) rinse up. The RP methods employ simple silica gel (232) or gel permeation rinse up (224) and the same kind of eluents as those used for aflatoxin analysis, i.e. methanol and or acetonitrile combined with water, the latter often acidified with acetic acid (234,237,242-243) or buffered with phosphate or phosphoric acid (224,230). Other methods employ
methanoI/tetrahydrofuran/aceticacid (242). Hurst et a/. (238) obtained good results for several mycotoxins using a cyano column and hexaneh-propanoVglacial acetic acid. Water/acetonitrile gradients have been shown to be of general use (4,47,226) for a large number of mycotoxins, especially in acidic gradients (4,47). Most detection methods for sterigmatocystin have been based on the UV maximum for sterigmatocystin at 325 nm (e.g. 230), but interfering compounds from foods or feeds reduce the reproducibility of the methods. Two derivatisation methods clearly improve the specific detection of sterigmatocystin:Abramson and Thorsteinson (232) acetylated sterigmatocystinin pyridin and acetic anhydride for 3 hours at 100 "C
296
and they achieved to diminish the observed background interference from barley considerably. Neely and Emerson (235) considered the "relatively long reaction time, the sensitivity of the reation to water and the gradual decomposition of the acetyl derivative" a problem and suggested the use of an aluminium chloride post column derivatisation of sterigmatocystin and fluorescence detection (exitation 254 nm and emission 455 nm). The latter method was developed for fermentation broth analysis, but it may be combined with parts of the method of Abramson and Thorsteinson (232) to a good general method for the detection of sterigmatocystin in cereals.
Two papers on the NP-HPLC separation of secondary metabolites related to sterigmatocystin were published in 1976 (239-240), but interestingly only the most predominant metabolites (sterigmatocystin, demethylsterigmatocystin
and
5-
methoxysterigmatocystin) were analysed by both groups of researchers, while sterigmatin, 6-deoxyversicolorinA, 6,8-O-dimethylaverufin,6,8-O-dimethylversicolonn A and aversin were analysed by Ito et a/. (240), the metabolites also found in Aspergillus parasiticus, versicolorin A & C, averufin and avermutin were analysed by Kingston et a/. (239). 8.4.3. Trichothecenes The trichothecenes are among the most important mycotoxins, but the poor UV absorption of tnchothecenes without an enone chromophore (type A trichothecenes, e.g. T-2 toxin, HT-2 toxin diacetoxyscirpenol (DAS)) makes HPLC analysis a less applicable methodthan gas chromatography or mass spectrometry (39-40,89,244-248).
8.4.3.1. Non-macrocyclictrichothecenes At least 80 different non-macrocyclic trichothecenes have been structureelucidated (249). It has been possible to analyze underivatized T-2 toxin and othertype A trichothecenes by HPLC (4,47), but such methods are only applicable to extracts containing very high concentrations of type A trichothecenes, and are thus only of practical use in few cases. Sensitive HPLC methods for type A trichothecenes require effective clean-up and derivatisation. Type B trichothecenes (e.g. nivalenol (NIV), deoxynivalenol (DON) = vomitoxin, fusarenone X (FUS-X), 3-acetyl deoxynivalenol(3AC DON)) have an absorption maximum at 219-221 nm (34,47) and may be more
297
easily detected by HPLC-UV. Lanin et a/. (250-252) have examined the influence of different eluents on the separation of five type B trichothecenes (NIV, DON, 3-AC DON, 15-AC DON and 7desoxy DON) on a RP-column and found that the best separation under isochratic conditions was achieved with waterltetrahydrofurane (76:24). Eluents with acetonitrile gave better separations than eluents with ethanol, but both gave insufficient separations of DON and 7-desoxy DON and 3-AC DON and 15-AC DON. Electrochemical detection has been used for the detection of DON (253-255). The electrochemical detection method used by Sylvia et a/. (254) improved sensitivity 12 times compared to UV detection. A C,, reversed phase system was used with
methanoVwater (35:65) as the eluent. The method could not be used for T-2 toxin. The method of Childress eta/. (255) employed photolysis before electrochemical detection. Because of the poor UV absorption of type A trichothecenes and the difference between type A and B trichothecenes, several chemical post-column derivatization methods have been developed. The post-column derivatization method developed by Sano et a/. (256) can only be used for type B trichothecenes however, because it is based on production of formaldehyde from the ketogroup after warm alkali treatment, followed by reactions with methyl acetoacetate and ammonium acetate. Other postcolumn derivatization methods have been based on p-nitrobenzylchloride (257), diphenylindenonesulphonyl esters (258)or anthracene-9-carbonylchloride (259). The method of Yagan et a/. (258) is very sensitive and is applicable to type A trichothecenes. The method of Bayliss eta/. (259) is also sensitive, but anthracene-9carbonyl chloride also react with the hydroxy groups in for example (fungal) sterols and may be a problem in extracts which will usually contain several such compounds. An elegant method based on ELlSA as an post column monitoring system for type A trichothecenes has been developed by Chu and Lee (260). Coupled with non-aqueous size-exclusion chromatography rinse up systems (261) or other column clean-up systems (262), immunological detection methods may result in very sensitive and specific analyses. Thus post-column derivatization or sensitive immunological monitoring has meant a significant improvement of analysis of type A trichothecenes compared to earlier methods (263). It is typical for most applications of HPLC in trichothecene analysis that reversed phase systems (&,
or occasionally C),
and
gradients of methanoVwater or acetonitrile /water are used, even though
298
tetrahydrofuran may have the best separation ability at isocratic conditions (252). The most common type B trichothecenes were baseline separated using acetonitrile /water gradients (4,47). Cereals are often invaded by Fusarium species both before and after harvest. The risk of trichothecene contamination of cereals is therefore of great concern. The methods available are often based on column clean-up (93,100,253,256,264-275) and have been developed both for carbohydrate rich (wheat, rice) and fat rich cereals, especially corn (271-275). Lauren and Agnew (275) suggested to hydrolyze the major trichothecenes to four basic families of trichothecens with the basic trichothecene skeletons NIV, DON, scirpentriol and T-2 tetraol respectively and they improved and extended the methods developed by Rood et a/. (276-277) and Kroll (278) and included moniliformin and zearalenone in their analysis. Methods have been developed for the analysis of trichothecenes in animal tissues (279-281) and plasma and urine (258,276). Like the methods for cereal products the trichothecene analysis are based on column chromatography clean-up and chemical derivatizations. 8.4.3.2. Macrocyclic trichothecenes A large number of the macrocyclic trichothecenes have absorption maxima in the range of 217-259 nm (34) and can thus be detected by UV detectors. Roridin A and verrucarin A had retention indices of 1013 and 1022, respectively, in the acidic water/acetonitrilegradient used by Frisvad and Thrane (47) and could be determined at their absorption maxima of 245 and 259 nm. HPLC and flash chromatography have been used extensively in the purification and monitoring of the synthesis of baccharins and related toxic secondary metabolites from Brazilian plants (52, 282-283). Normal phase systems using ethyl acetate and hexane were used for the analysis of verrucarins and roridins (284-285), but the same toxins are also easily separated by
RP systems using water/acetonitrile gradients (47). Satratoxin G and H and trichoverrols were determined in conidia and cereals grains respectively by Sorenson et a/. (286) and Stack and Eppley (287) respectively. Being very important in the
airspora of houses and factories Stachybotrys atra conidia may be of more concern as airborne contaminants. The same fungus has only been sporadically reported from cereals. It is expected that more multi-toxin methods will be developed for the
299
macrocyclic trichothecenes, probably based on water/acetonitrile or water/methanol gradients and diode array detection. 8.4.4. Small lactones A quite large number of filamentous fungi produce small lactones or related
compounds (Table 8.1) which are either generally toxic, e.g. patulin and penicillic acid (34,42-43), neurotoxins, such as citreoviridin and verrucosidin (34), good chelators of metal ions, such as the Raistrick phenols (288), mycophenolic acid, terrein, terrestric acid, kojic acid, cyclopaldic acid etc. (48), or they have other occasionally unknown biological effects. Most HPLC methods have been developed for patulin and penicillic acid, but all the fungal secondary metabolites listed above can be analyzed by general methods such as that of Frisvad and Thrane (47). 8.4.4.1. Patulin Patulin is produced by Penicillium expansum in fruits and fruit products, but may also be produced by P. griseofulvum in cereals and P. glandicola, P. roqueforti var.
carneum,Paecilomyces variotii, Byssochlamys nivea, B. fulva,Aspergillus clavatus and
A. ferreus in silage, malt or airtight storage (31). Several TLC and GC methods are available for patulin analysis, but most methods are now based on HPLC, occasionally with TLC or GC-MS confirmation. In general the strong UV absorbtion at 275-276 nm is used for detection. Patulin and griseofulvin produced in culture by Penicillium griseofulvum (= P.
urticae = P. pafulum) can be analyzed directly from the fermentation broth after SepPak NP clean-up (289-290), after chloroform/ethyl acetate and/or ethyl acetate extraction (47) or diethylether extraction (290). The method of Priest and Light (291) includes an effective separation of a series of the biosynthetic intermediates in the patulin biosynthesis by using RP-HPLC and a gradient of buffered methanoVwater. Several of the same intermediatesare also separated in the acidified acetonitrilelwater gradient used by Frisvad and Thrane (47). Most HPLC methods for patulin have been developed for its determination in apple juice and other fruit products. The first methods were based on NP columns (292-305), but since 1980 RP columns have been used almost exclusively (305-319). The separation from 5-hydroxymethylfurfuralis important in these analyses (308,318-
300
321) and Sep-Pak clean-up is very often employed after extraction with ethyl acetate. In the reversed phase applications,water (306), water/acetonitrile (4,47,307,313,316317), water/tetrahydrofuran (312,315) or watedmethanol (314) are all used with success. Patulin is unstable in cheese, other milk products, and meat (322), but has been included in several multi-mycotoxinscreening methods in fungal cultures or foods such as cocoa beans (4,47, 236-238,323-324). 8.4.4.2. Penicillic acid The most important penicillic acid producers in food are Penicillium
aurantiogriseum (var. aurantiogriseum,var. polonicum, var. melanoconidiumand var. viridicatum) and members of Aspergillus subgenus Circumdati section Circumdati (formerly the Aspergillus ochraceus group). Other reported producers are quite uncommon and apparently only as superficial contaminants in foods and feedstuffs (Frisvad and Filtenborg, unpublished data). Penicillic acid has been isolated from corn (325-326), poultry feed (327), dried beans (328) and tobacco products (329), but interestingly not from cereals with a low lipid content such as wheat and barley. The poor stability of penicillic acid in the presence of -SH groups may explain the absense from both wheat, meat and cheese (330-333). Penicillic acid is best determined at its absorption maximum at 226 nm (34,47), but in several applications its UV absorbance is determined at 254 nm (323-324,334336) or at 245 nm (238). The reason for this may be that a large number of compounds have a maximum absorbance at 225 nm, while fewer compounds have a maximum at 254 nm. It has been analyzed together with patulin, zearalenone, and sterigmatocystinon a cyano column with hexane/l -propanol/acetic acid as the eluent (238) or on RP columns using acidified acetonitrile/water as eluent (47,237, 334) or neutral acetonitrile/water(238,335-336).Extraction methods and purificationof penicillic acid from biological tissues were developed by Chan et a/. (334) and Hanna et al. (336), but were as simple as those used for fungal cultures (47). However, for cocoa beans a silica Bond-elut columns clean-up step was included (238). As Penicillium
aurantiogriseum varieties (Table 8.1) are extremely common in cereals (31), more cereal samples should be screened for the presence of penicillic acid.
30 1
8.4.4.3. Mycophenolic acid
Mycophenolic acid is produced by three species of Penicillium and Lepfographium abienfinum (Table 8.1). It is however only P. brevicompacfum and P. roqueforti var. roqueforti and var. carneum that are relevant producers of this
apparently only weakly toxic secondary metabolite concerning foods and feedstuffs. It elutes as a sharp peak in the general HPLC screening system of Frisvad and Thrane (47, illustrated in ref. 45) and it is easily separated from the many other secondary
metabolites produced by Penicillium species. Neely and Parks (336a) developed a simple HPLC method for analysis mycophenolic acid in fermentation broth. 8.4.4.4. Butenolide
Butenolide is a short and too general abbreviation for 4-acetamido-4-hydroxy-2butenoic acid y lactone. This mycotoxin is also included in the general HPLC screening system of Frisvad and Thrane (47), but poor absorption at 225 nm and high polarity require a better HPLC method for an optimal detection, especially in foods and feedstuffs. 8.4.4.5. Verrucosidin
Verrucosidin, a neurotoxin, is produced by Penicillium auranfiogriseum var. auranfiogriseum chemotype I, var. polonicum and var. melanoconidium. It was first
described as a tremorgen from a strain of P. verrucosum var. cyclopium (337-338). This strain was later examined taxonomically and found to be P. auranfiogriseumvar. polonicum (16). Verrucosidin is also included among the mycotoxins in the multi-
mycotoxin method of Frisvad and Thrane (47), but again it seems likely that other more specific analytical LC methods can be developed for this mycotoxin. It should be among the mycotoxins screened for in cereals, as the three varieties of P. aurantiogriseum,very common in cereals, are consistent producers of this toxin (16). 8.4.4.6. Citreoviridin
Citreoviridin is produced by Penicillium citreonigrum, P. manginii, P. miczynskii, P. smifhii and Eupenicillium ochrosalmoneum (Table 8.1). The first and the latter
species and their neurotoxins may be of importance as they occur frequenly in rice in Taiwan and Japan (43) and pecans in U.S.A. (P. citreonigrum) (15,339) and corn in
302
U.S.A (Eupenicillium ochrosalmoneum)(l5,19,340,341). Cole et a/. (339) used a reversed phase column and methanol/water (65:35)at 1.5 ml/min as eluent, but Stubblefield et a/. (342)developed a normal phase method for determination of citreoviridin in corn and rice. This method was based on dichloromethane extraction, silica and amino-column clean-up and ethyl acetatdn-hexane (7:3)elution at 1.5 mllmin. The neurotoxin was detected by UV absorbtion at 388 nm. The chemically related secondary metabolites citreoviridin, citreomontanin, asteltoxin and aurovertin B were also included in the multi-mycotoxinHPLC method of Frisvad and Thrane (47).
The very characteristic UV spectra of these metabolites make their analysis specific.
8.4.5.Macrocyclic lactones The most important mycotoxin in this group of secondary metabolites are zearalenone and zearalenol, but several other compounds have been described in this The producers of these estrogenic compounds are biosynthetic family (43,p. 275,47). listed in Table 1 , and they are all quite frequently occuring and actively growing in cereals. Other compounds in this group include monorden, brefeldin A and curvularin. These compounds are produced by certain Penicillium species that are quite uncommon in foods and feedstuffs, but other species, often "field fungi" may produce them (see Table 8.1).It is not clear whether any of these compounds are mycotoxins sensu stricto, but they can all be separated by HPLC in an acidic water/acetonitrile gradient (47). Several methods have been developed for the detection of zearalenone in different foods, feeds and animal tissues, blood and urine. In most cases fluorescence detection has been used (exitation wavelenth 236 or 280 nm, emmission 41 8 or 470 but electrochemicaldetection (343),voltametric detection (344)and nm)(356,372,363), UV detection have also been used (4,47,237,275). Most methods have been developed for corn and other cereals (266,275,323,345-363). Trenholm et a/. (354) used zearalenone oxime as an internal standard for their analysis of zearalenone in wheat. However because of its hormone like activity and carry-over of the toxin, analytical methods have also been developed for the detection of zearalenone in milk (368),blood (369-372)and urine (372-374). animal tissue (364-367),
303
Rinsing up of zearalenone has included addition of diatomaceous earth, extraction with chloroform or dichloromethane, occasionally silica cartridge rinse up (372), extraction into base, acidification of the water phase and reextraction with chloroform or dichloromethane (356,363,372). NP applications have included eluents such as water-saturated dichloromethane containing 2 Yo 1-propano1 (372) and RP applications have included eluents such as acetonitrile/water (94:6) (363), acetonitrile/water/acetic acid (55:45:2) (237), methanol/acetonitrile/water(5:8:10) (356), methanol/ 1 Yo acetic acid (62:38) (275) and methanol/water (7:3) or acetonitrile /water/acid gradients (4,47,236). An amino column was used by Rannft et a/. (363).
8.4.6. Ochratoxins and related compounds 8.4.6.1. Ochratoxin A Being both a nephrotoxin and a carcinogen, ochratoxin A is considered one of the most important mycotoxins (42-43). Ochratoxin A has been found naturally occuring in barley, wheat, rye, oats, corn, sorghum, peanuts, coffee beans in Denmark, Canada, U.S.A., France, Sweden, Poland, Yugoslavia, Great Britain and India (43). In addition to Aspergillus species (Table 8.1) several species of Penicillium have been reported to be producers of ochratoxin A, but only one species have been found to produce this nephrotoxin: Penicillium verrucosum (16). The culture ex type of this species and cultures of its synonyms, such as P. caseiand P. nordicum are very good producers of this toxin. P. verfucosum has been isolated from all barley samples of 77 tested leading to porcine nephropathy in Denmark (Frisvad, unpublished results) and no other species were able to produce ochratoxin A. Other Penicillium species reported to produce ochratoxin A were misidentified (16,19,375-377) or the metabolite detected was another bluish green flourescing secondary metabolite (19). It should be further investigatedwhether simultaneous productionof ochratoxin A and citrinin by Penicillium
verrucosum chemotype I in cereals or ochratoxin A production by P. vefrucosum chemotype I in meat products such as salami are bofh causes of ochratoxin A contamination and human and animal health problems. Like in the case of zearalenone, rubratoxin and citrinin, the analysis of ochratoxin A in biological matrices is greatly improved by using acids in the eluent.
304
Severe peak broadening and/or binding to the column, often dependent on the batch
of reversed phase column (236) havelhas been observed in neutral eluents because of the carboxylic and/or phenolic groups in these mycotoxins. Therefore most multimycotoxin methods have included acids like phosphoric acid, acetic acid, formic acid or triflouracetic acid (4,47). The acidic groups in these molecules also suggest a partition into bicarbonate solution after organic solvent extraction, and a re-extraction into organic solvent after acidification of the water phase as a very efficient clean-up method. Several general multi-mycotoxin methods that include ochratoxin A have been developed (e.g. 4,47,236-237). These methods often depend on a general detection method, i.8. UV detection. For most dedicated applications of ochratoxin A analysis, however, fluorescence detection is much more sensitive and has been applied almost exclusively. Applications have been developed for cereals and feedsstuffs (90,347, 362,378-388), coffee and cocoa beans (389-392), foods and feedsstuffs (393-399), cheese (400), eggs (401), tissues, liver and kidneys (402-410), milk (411-412), serum and blood (413-415), urine (416), and rumen fluid (417). Reversed phase columns (C,, C, (362,413) or C, (403)) have also been used nearly universally for ochratoxin A (all above except 395) with an eluent of either acidified methanowwater (350,362,392, 394,396), buffered methanol/water (415), acidified acetonitrilelwater (394,387388,400,403-405,413,416-417, isopropanol added also in ref. 387-388 and 400),
bufferedacetonitrile/water (391,397-398,402) or acidified acetone/water(310). Gradient elution has been used when several mycotoxins have been analyzed together with ochratoxin A (4,47,386). Chamkasem e l al. (386) used phosphate buffer and methanoVacetonitrile in their gradient elution method for aflatoxins, ochratoxin A and zearalenone in grains, oilseeds and animal feeds. 8.4.6.2. Citrinin Citrinin is also a nephrotoxic mycotoxin and it is produced simultaneously in several cases with ochratoxin A by P. verrucosum. However several other Penicilium species have been shown to be producers of citrinin. Of these (see table 1) only Penicillium verrucosum, P. expansum, P. hirsutum var. albocoremium, P. citrinum and Aspergillus terreus are known to be active colonizers of foods and feeds (16,19,29,49).
305
Citrinin has been found as a natural contaminant of barley, wheat, rye, oats in Canada (418), barley and oats in Denmark (419) and rice in Japan (420).
Only quite few analytical HPLC methods exist for citrinin compared to the large number of HPLC methods for ochratoxin A. Citrinin is a stonger acid than ochratoxin A and is more difficult to analyze without ion-pairing agents, buffers or acids. In most
cases citrinin is extracted by organic solvents and rinsed up by partition into bicarbonate. Most application have been based on reversed phase columns but Dick et a/. (421) developed a sensitive NP-HPLC method for citrinin in cereals using
hexane/chloroform (6:4). RP-HPLC applications include eluents such as 0.25 N phosphoric acid and methanol or acetonitrile (422), 0.25N phosphoric acid/acetonitrile/ isopropanol (387,423), water/acetic acid/acetonitrile (40:59:1) containing 0.025 M tetrabutylammonium phosphate (424) or ion-pair partition chromatography (425) or other acidic eluents (4,47,426). However Zimmerli et a/. (427) claimed that these methods worked poorly for them, except the ion-pairing method (425). The latter method gave problems with lost fluorescence, which could be overcome by post column addition of acid (427). Zimmerli etal. (427) thus developed a sensitive method based on an acid-buffered silica gel column (428) using the same eluent as in their earlier method (421). UV detection at 340 nm has been used in some cases (423), but fluorescense detection (387,421-422,427) (exitation 340-360 nm, emission 500 nm) is much more sensitive (424). It seems that either a acid-buffered NP column (427) or a tetramethylammonium phosphate buffered RP system (424) are the only analytical systems giving consistently sharp peaks of citrinin independent of the brand of column. HPLC methods for citrinin has been developed for cereals (387,421 -422,425427), fermentation broth (424) or biological fluids (423). For broth and fluids, rinsing
up may not be necessary at all. Improved methods for citrinin and other acidic mycotoxins such as ochratoxin A, terrestric acid, penicillic acid, rubratoxin B and zearalenone in cereals may be based on the method developed by Vail and Homann (424) or the NP method of Zimmerli etal. (427). 8.4.6.3. Xanthomegnin, viomellein and related compounds.
In foods the most important producers of xanthomegnin and viomellein and the related viriditoxin are several varieties of Penicillium aurantiogriseum, Aspergillus ochraceus and Paecilomyces variotii (Table 8.1). Xanthomegnin and viomellein has
306
been found to occur naturally in barley in Denmark (429) and wheat in Great Britain
(430). Xanthomegnin and viomellein have been considereddifficult to analyze because of binding to HPLC columns (431). Earlier methods were based on normal phase separations. Because of the acidity of the phenolic groups in xanthomegnin and viomellein acid should be added to the eluent (432-435).The first methods developed used NP columns and either chloroform/methanol/ acetic acid (98:l:l)(432-434) or toluene with 1% acetic acid/ methanol (493:7)(435).Reversed phase applications were developed by Carmen el a/. (436-438)using acidified water/acetonitrile as eluent and modified by Wall and Lillehoj (431) by adding sodium dodecyl sulphate to avoid irreversible binding to the RP column. Preparative HPLC methods for ochratoxin A, viomellein and xanthomegnin have also been developed (439-442). Xanthomegnin and viomellein have been detected by UV absorption at 405 nm giving a detection limit of 12 ng xanthomegnin (431-436), electrochemical detection, with 0.5 ng xanthomegnin as the detection limit (438) or fluorometric detection after reaction with ammonia and hydrogen peroxide (exitation 340 nm and emission 445
nm)(437). Applying the last method as little as 0.1 ng of xanthomegnin could be detected (437). It is known that Penicillium verrucosum producing ochratoxin A and citrinin and varieties of P. auranfiogriseum producing penicillic acid, xanthomegnin and viomellein are co-occurring in cereals (1 6,49),but until now no multi-mycotoxinmethod has been developed for all these five nephro- and hepato-toxins in cereals. All these toxins were detected by Frisvad (48) in cultures of Aspergillus ochraceus by an acidic gradient elution method and diode-array detection, but this method has not been further developed for cereals.
8.4.7. Rubratoxins The only confirmed producer of rubratoxin is Penicillium crateriforme (formerly called P. rubrum)(49).This species also produce another acidic mycotoxin spiculisporic acid (49).The closely related species P. purpurogenum produce other chemical related acids (glauconic and glaucanic acid) but not rubratoxins (1 9,20). P. crateriforme has been found in corn and is probably able to produce rubratoxin B on that substrate
(444-445). Rubratoxin elutes as a sharp peak in RP systems with acidic acetonitrile/water
307
gradients (47,236-237) using UV detection at 254 nm or diode-array detection. Unger and Hayes (446) developed a RP-HPLC method for rubratoxin B in plasma and urine using an eluent of water/acetonitrile/ethyl acetate (9.9:11:3). Engstrom and Richard (447) developed a NP-HPLC method for rubratoxin B in mixed feed based on acidic
ethyl acetate extraction, cool and dark handling and storage using ethyl acetatekh loroform/acetic acid (80:20:1) as the eluent.
8.4.8. Hydroxyanthraquinonesand xanthones
Several monomeric anthraquinones have been characterized from filamentous fungi (Table 8.1), but only emodin (448) and physcion (449-451) have been suggested as mycotoxins. Penicillium islandicum and several other penicillia produce bianthraquinone mycotoxins such as rugulosin and luteoskyrin ("Yellow rice toxins"), while other species produce a family of bixanthones including the mycotoxin secalonic acid D. Other related anthraquinonesand xanthones are treated under Alfemaria toxins (see below). 8.4.8.1. Emodin and physcion
These anthraquinonesare producedby several species of filamentous fungi, but also by lichens and plants (Table 1). Most HPLC applications have been developed for anthraquinones as extracted from plants such as rhubarb (452-457), sometimes as glycosides (458). However they are also easily determined by the general HPLC method of Frisvad and Thrane (47), even though the anthraquinones elute quite late in that system. Matthees (459) developed a NP and RP analytical HPLC system for emodin in feeds based on extraction into aqueous acetonitrile, partitioning into chloroform and NP-HPLC using isooctan/isopropanol/acetic acid (95:5:1) or water/methanol/acetic acid (20:80:1) for RP-HPLC and UV detection at 280 nm. 8.4.8.2. Rugulosin and luteoskyrin
Rugulosin is produced by food and feed-borne fungal species such as Penicillium islandicum, P. rugulosum and P. piceum (19) and luteoskyrin is produced
by P. islandicum common species in rice (29). Rugulosin and luteoskyrin eluted quite
308
late in the HPLC method of Frisvad and Thrane (47), and dedicated methods based on HPLC should be developed for these important carcinogenic mycotoxins. 8.4.8.3. Secalonic acids Secalonic acid D is produced in corn by Penici//iumoxalicum, but it may also be produced naturally by other species, e.g. by Claviceps purpufea in rye. A HPLC method for this mycotoxin was developed by Reddy et a/. (460) and secalonic acid D was detected for the first time a natural contaminant of corn dust in 1982 by Ehrlich
eta/. (461). Two eluents were used in a RP-HPLC system using UV detection at 340 nm: water/acetonitrile/ acetic acidnetrahydrofuran (6:lO:l:l) or (6:8:1:1) and applied on biological fluids (460). Secalonic acid D could also be determined in fungal cultures by the HPLC method of Frisvad and Thrane (47), but like other quinones and xanthones it also eluted quite late. 8.4.9. Epipolythiopiperazine-3,6-diones. The most important toxins in this class are gliotoxin, sporidesmin and emestrin. The most important producers of these toxins are Aspergillus fumigatus (gliotoxin),
Pithomyces chartarum(sporidesmin)and Emericella striata (emestrin). HPLC methods have been developed for gliotoxin and sporidesmin. 8.4.9.1. Gliotoxin Gliotoxin extracted from fungal cultures eluted as a sharp peak (47), but a dedicated method has also been developed for gliotoxin in rice (462). After chloroform extraction and partial clean-up by petroleum benzine precipitation and gel permeation chromatography, gliotoxin was analyzed by RP-HPLC using water/methanol(57:43) as eluent and UV detection at 254 nm. 8.4.9.2. Sporidesmin Sporidesmin has been detected by UV at 254 nm by a cyano column and hexane/isopropanolor hexane/chloroform (for preparative purposes) (463) or on RP columns (464-466)using water/methanolas eluent for analytical separations. Different very complex clean-up procedures have been suggested to avoid sporidesrnolides, polyphenoles and other interfering substances (463-466), including ethyl acetate
309
extraction and clean-up by chloroform/hexaneelution on a Lipidex 5000 column (463), acetonitrile/benzene extraction, evaporation to
dryness and dissolving in
methanoVwater, removing lipids with hexane and reextraction of the water phase with benzene (464), extraction with diethylether, partitioning into hydrochloric acid, addition of water and back-extraction into diethylether, followed by partitioning into sodium bicarbonate/hydroxide solution, neutralized, evaporated to dryness and dissolved in chloroform followed by preparative HPLC with chloroform with 0.8% ethanol as eluent (465). Separation from the interfering substances may also be obtained by gradient
elution and less complex rinsing up. 8.4.10. Tremorgenic mycotoxins
A large number of tremorgenicfungalsecondary metabolites have been isolates, including those with a tryptophan nucleus, penitrems, janthitrems, lolitrems, aflatrems, paxilline, paspaline, paspalicine, paspalitrems, verrucologens, and the tryptoquivalins, but also the territrems and verrucosidin, lacking any nitrogen in the molecule. 8.4.10.1. Penitrems
The most important producer of penitrem A in foods and feeds is Penicillium crustosum, but other producers such as f .glandicola (formerly f .granulatum) and f . aurantiogfiseumvar. melanoconidiummay also play a role (16). Penitrem A has been
found naturally occuring in refrigerated cream cheese, where it caused intoxication of two dogs (467) or mouldy walnuts, where it caused toxicosis in a dog (468) and it may also have been present in a sample of beer which caused tremors in a man (469). Even though the toxin was not found originally in the isolate of f . cfustosum, we have later examined the strain and found that it produced large amounts of penitrem A. Even though the penitrems were nor actually found in the samples all evidence indicates that these tremorgens were involved in a toxic syndrome of sheep and horses (470) and corn infected with f . cfustosum caused a natural intoxication of cattle (471). Maes et a/. (472) developed a RP method for penitrem A to F using water/methanol (22:78) at a flow rate of 1.5 ml/min (and a column temperature of 40
"C).As an internal standard they prepared penitrem A monoacetate. Even though UV absorbance is higher at 233 nm than at 296 nm, the latter wavelength was selected
310
for monitoring because of the greater selectivity. The method of Maes eta/. (472) was also used by Dorner et a/. (471) and in a modified form by Mantle et a/. (473). The latter authors used water/methanol (1:5) as an eluent at a flow rate of 2.5 ml/min and UV detection at 335 nm. di Menna et a/.(474) used a combination of a RP system (C,)
with a gradient from water/methanol (28:72) to (8:92) at a flow rate of 1.2 ml/min and UV detection at 230 nm with a NP system monitored at 290 nm and an eluent
consisting of dichloromethane/acetonitrile (92.5:7.5). The latter NP system was used mainly for confirmation of identity of the penitrems. The penitrems appear to be sensitive to light and acids (472). 8.4.1 0.2. Janthitrems The janthitrems are produced by Eupenicillium zonatum and Penicillium janfhinellum, which are not particularily common in feeds or foods. They were
considered to be involved in ryegrass staggers, but this neurological disease in cattle and sheep is now believed to be caused by endophytes producing lolitrems. One method has been developed for the HPLC determination of the janthitrems (473). RPHPLC (C,) (preferred for a NP and a CN column) (473) was used to separate the janthitrems produced in laboratory media using an eluent of water/methanol (20:80) and UV detection at 265 nm. In this system janthitrem A, B, C, verrucologen and fumitremorgen
D, penitrem A,
B could be separated. For fungal extracts
water/methanol (36:64) for 10 min followed by a linear gradient over 5 min to water/ methanol (20:80) was used with detection at 330 nm. Fluorescencedetection was used to confirm identity of the janthitrems and increased (50 fold) sensitivity (exitation 254
nm and cutoff emission at 370 nm). The fluoresecence was only high in the RP system, and poor in the NP system (eluent hexane/ethylacetate/methanol,85:14.7:0.3) or the CN system (eluent hexane/isopropanol,9:l).The janthitrems appear quite late in the HPLC system of Frisvad and Thrane (47). 8.4.1 0.3. Lolitrems and paxilline The lolitrems are important tremorgens involved in ryegrass staggers (474) and paxilline, produced by several filamentous fungi (Table 8.1), appears to be a precursor of lolitrems.Weedon and Mantle (475) used the HPLC system of Gallagher eta/. (476) to quantify lolitrem 8, i.e. a NP silica column and a mobile phase of
31 1
dichloromethane/acetonitrile (1 5 : l ) at 2 ml/min and fluorescence detection (exitation
268 nm, emission 450 nm). Paxilline was anlyzed on a NH, column using dichloromethane/isopropanol (10O:l)at a flow rate of 4 ml/min and UV detection at 281 nm. Paxilline and l-acetoxy paxilline had retention indices of 1291 and 1386 in an acidified water/ acetonitrile gradient HPLC system (47).
8.4.10.4. Aflatrem, paspaline, paspalicine and paspalitrem A series of indol and carbazole alkaloids have been isolated from Aspergillus
flaws (477-481), A. nomius (482), A. leporis (483) and A. tubingensis (484-486). All these metabolites have been separated using HPLC. Cole et a/. (477) separated aflatrem, paspalinine and dihydroxyaflavinine by RP-HPLC gradient elution (water/acetonitrile 80:20 to 20:80). Gloer and TePaske and co-workers (478,481-486) used RP-HPLC to separate several aflavinines, nominine, leporin, aflavazol and tubingensins. The conditions were alike: Watedmethanol (30:70 or 10:90) was used as eluent at 2 ml/min and the metaboliteswere monitored at 215 nm, occasionally also employing diode array detection to find new metabolites with similar chromophores. Nozawa and coworkes used either NP separation (using hexane/ ethylacetate, 4:l)
(479) of the aflatrem, paspalinine and aflavinines, but later employed RP- HPLC for separation (480). Paspalinin, paspalin and aflatrem had retention indices of 1332,161 7 and 1514, respectively in a acidic water/acetonitrile gradient (47).
8.4.10.5. Fumitremorgins and verrucologen Verrucologen and other fumitremorgins are produced by Aspergillus fumigatus,
Neosartorya fischeri var. fischeri, Penicillium brasilianum,P. graminicolaand other less common species (Table 8.1). These tremorgens and fumigaclavine A, B, and C may have been implicated in mouldy corn silage intoxication of cattle, in which A. fumigatus is particularily common (487). Di Menna et a/. (474) used HPLC to separate verrucologen and fumitremorgen B. The fungal cultures were analyzed by a RP HPLC column using water/methanol (28:72) as eluent and monitored at 230 nm. The results were
validated
by
employing
NP-HPLC
of
the
same
extracts
in
dichloromethane/acetonitrilecontaining 0.5 % acetic acid and UV detection at 230 nm
and by comparison to standards. Nielsen eta/. (488-490) used RP-HPLC to separate verrucologen and fumitremorgen A, B and C. Their method was modified from the
312
method described by Frisvad and Thrane (47) by avoiding trifluoroacetic acid, which is not necessary for good separation and peak shape of these tremorgens. 8.4.10.6. Tryptoquivalins Tryptoquivaline and tryptoquivalone may be implicated in Aspergillus clavatus malt intoxication of different animals. HPLC has been used in the separation of these tremorgens (47,491-492), but no dedicated method has been developed for them. They are separated easily by using gradient elution using water/acetonitrile (47). 8.4.10.7. Territrems The territrems are some of the few known fungal tremorgens without nitrogen in the molecule. They were originally isolated from a strain of Aspergillus ferreus (493496), and apparently only produced by the original isolate (497), but it appears to be consistently produced by Penicillium echinulatum var. echinulatum (498), a species common on lipid-containing foods. The territrem are strongly fluorescing like the aflatoxins and therefore Ling et al (499) used both TLC and HPLC to differentiate between these toxins. Using a NP column and water saturated chloroform/cyclohexane/acetonitrile(25:7.5:1) with 0.25% ethanol at a flow rate of 2 ml/min, aflatoxin B, and B, could not be fully separated from territrem A and B, and the authors adviced to verify the presence of the aflatoxins by measuring the ratio of peak heights at 365 nm compared to 335 nm (UV detection). Later Ling (personal communication) used RP-HPLC using 60% acetonitrile in water with 0.1 N acetic acid at a flow rate of 1 ml/min to separate territrem A,
B and C
(retention time 9.36, 8.37, and 5.3, respectively). 8.4.1 1. Alternaria toxins Species of Alternaria produce a series of chemically different secondary metabolites (Table 8.1) and several of these are considered as mycotoxins (500-501).
Alternaria alternata, a very common fungus in plants, is often capable of producing large amounts of tenuazonic acid, an important mycotoxin (500).
A series of host-selectivesecondary metabolites have been analyzed by HPLC, e.g. macrosporin, altersolanol A and alterporriols (502-504) but several eluents were used to separate these metabolites on RP columns (502): 0.05 M ammonium
313
dihydrogen phosphate and phosphoric acid (pH 2.5) in water/acetonitrile (7:3, 4:l or
1:l) at a flow rate of 1 ml/min. Maleyl amide derivatives of some host selective phytotoxins from Altefnaria alternata fsp. lycopefsici were prepared and analyzed at
250 nrn in a RP gradient system (503). However most interest have been invested in the mycotoxic secondary metabolites of Alfernaria.
8.4.11 . 1 Tenuazonic acid Tenuazonic acid, cyclopiazonic acid and terrestric acid will give quite broad peaks in most chromatographicsystems (47,505)and may cause trouble because they are strong acids and extremely efficient metal chelators (505). Several methods for tenuazonic acid have been suggested, often in analytical procedures involving other
Ahernaria toxins such as alternariol (AOH), alternariol monornethylether (AME), altenuene (ALT) and altertoxin I and II (ATX-I & ATX-11). Scott and Kanhere (505) tested several HPLC systems for the analysis of tenuazonic acid. A RP system using methanoVwater with 0.1 % phosphoric acid gave very broad peaks. Two other RP systems were also tested but interfered with constituents from tomato paste. The HPLC system advocated was based on RP column coated with C12 dien (4-
dodecyldiethylene-triamin) and a eluent consisting of methanoVwater with 0.001 M ZnSO,. However Heisler and co-workers reported on good results for tenuazonic acid using a RP separation based on a water/ methanol (1 :9) at a flow rate of 2 ml/rnin in fruit and vegetable products (506-508).Later Stack eta/. (509) developed a RP HPLC method for tenuazonic acid (and alternaflol) using rnethanoVwater (85:15) containing
300 rng ZnSO,/I as eluent. Tenuazonic acid eluted as a broad but symmetric peak in the system of Frisvad and Thrane (47, see ref. 510, fig. 3 for illustration). Some of the most efficient analytical RP-HPLC systems for tenuazonic acid were developed by Lebrun eta/. (51 1-512). Ion-pair (5 mM cetrimid in water/methanol (45:55))and ligandexchange chromatography (5 mM C, dien and 5 mM ZnSO, added to waterimethanol
(25:75)bufferedwith 30 mM ammonium acetate, pH 6) could be applied for an efficient quantification of tenuazonic acid in Pyricularia oryzae infected leaves and were preferrred for anion-exchange chromatography. However the latter methods may not be suitable for LC-MS because several constituents are not volatile.
3 14
8.4.1 1.2. Alternariols, altenuenes and altertoxins.
Alternariols and related mycotoxins are more easily analyzed by HPLC than tenuazonic acid (513). Chu and Bennett (514) developed a method for preparing large amounts of alternariol by preparative NP-HPLC using different ratios of hexane and ethyl acetate as eluents. However for analytical HPLC methods RP columns have been used in most cases. However,Ozcelik eta/. (515) preferred an NP-HPLC system after comparing with several RP techniques for tenuazonic acid and several alternariols. This system involved chloroform/methanol(95:5)at a flow rate of 0.7 ml/min and UV detection at 280 nm as in many other applications. However RP systems has been advocated by
other
authors
(506-507, 509316-520).
MethanoVwater or
acetonitrilelwater eluents have often been used in RP applications of analysis of AMernaria toxins (47,506-507,509,516-518). Frisvad and Thrane (47) used a water/acetonitrile gradient with trifluoroacetic acid, while Palmisano el a/. (518) used a water/methanol gradient with phosphoric acid for diode array applications. Both systems are generally applicable for all kinds of mycotoxins, but the former has the advantage of low UV absorbtion and volatility and low corrosiveness of trifluoroacetic acid in contrast to phosphoric acid. A HPLC system of high selectivity was developed by Palmisano and Visconti and co-workers (519-520) using electrochemical detection after post-column addition of bromine (519) for altenuene and isoaltenuene. The methods involved addition of either sodium nitrate, sodium bromine and nitric acid (519) (altenuenes) or just nitrate and nitric acid (520) (altertoxins) to an eluent of water and methanol. 8.4.12. Toxic peptides
Only few analytical methods have been proposed for other peptide mycotoxins than cyclosporin and toxins from fleshy fungi. The latter toxins were reviewed in detail by Betina (39). The method of Edwards and Lillehoj (521) for cyclosporin in rice was based on gel permeation chromatography, followed by RP HPLC using water/acetonitrile (1:l) at a flow rate of 1 ml/min, and monitored at 212 nm. Cyclosporin was also analyzed by TLC and the identity was further confirmed by infrared spectroscopy. Samuels et a/. (522) used a related method for cyclodepsipeptides from Metarhizium anisopliae but they used cation exchange chromatography followed by a similar RP HPLC analysis and confirmed their results
315
by TLC and fast-atom bombardment mass spectrometry. A method developed for phomopsinA in lupin stubble also employedcation exchange chromatography followed by RP HPLC after methanoVwater extraction and purification by partitioning between n-butanol and water (523). These three methods have a lot in common and could be used for important toxins such as cyclochlorotine, for which only TLC methods exist
(524) to the authors knowledge. The nephrotoxic glycopeptides from Penicillium auranfiogriseum var. auranfiogriseum possibly involved in Balkan endemic nephropathy were purified by a
procedure used for proteins, i.e. employing water extraction, cation exchange, anion exchange, size exclusion chromatography, reverse phase Sep-Pak mini column chromatography, followed by RP gradient HPLC and finally isocratic RP HPLC (525). For RP-HPLC water/acetonitrile with trifluoroacetic acid were used as eluents and UV monitoring was at 226 or 210 nm. Analysis for penicillin often follow the same kind of analytical RP-HPLC procedures as those outlined above, often using eluents containing water with phosphate buffer and acetonitrile (526-527) ocasionally using post-column reactions
(527-528). 8.4.13. Fusarium toxins other than trichothecenes and zearalenones A series of toxins have been proposed to be implicated in equine leukoencephalomalaciaand other diseases (moniliformin,fusarins and fumonisins) and several HPLC methods have been developed for these mycotoxins. It is now believed that the cancerogenic fumonisins are the principal causes of several diseases and the fumonisins have been found to occur naturally (529). However other mycotoxins have also been analyzed by HPLC such as the fusarochromanones and fusaric acid.
8.4.13.1. Fumonisins The fumonisins may be purified by ion exchange followed by preparative RP HPLC using watedmethanol containing trifluoroacetic acid and/or acetic acid (530) and confirmed analytically by a series of chromatographic and spectroscopic techniques
(17,531),such as TLC, GC-MS and HPLC. Shepherd eta/.(532) developed a HPLC method for the fumonisins based on methanoVwater (3:l) extraction, ion-exchange mini-column chromatography and pre-column derivatizationwith ortho-phthaldialdehyde
316
followed by separation by RP-HPLC using water with 0.1 M sodium dihydrogen phosphate adjusted to pH 3.3 with oftho-phosphoric acid/methanol (20:80)at a flow rate of 1 ml/min. The derivatives of the fumonisins were detected by fluorescence (exitation 335
nm
and
emission
440 nm).
This
method
and
other
chromatographidspectroscopicmethods were used by Gelderblom and co-workersand Plattner and co-workers to analyze for fumonisins in foods and feeds (17,533-536).
8.4.13.2.Fusarochromanone Fusarochromanone has been found naturally occuring in cereal feed associated with tibia1 dyschondroplasia(537)and it was produced by isolates of fusafium equiseti
(26,538).The fusarochromanones could be separated in the RP-HPLC system of Frisvad and Thrane (47),and similar methods were used by Wu eta/. (539)and Yu and Chu (540).Yu and Chu (540)used water with trifluoroacetic acid/acetonitrile (4:6) at 1 ml/rnin and immunodetection for determination of fusarochromanone in cereals
and Wu el a/. (539)used water/methanol/acetic acid (20:120:1)at 1 ml/min and UV detection at 254 nm.
8.4.13.3. Fusarins The fusarins, especially fusarin C, are mutagenic mycotoxins once believed to be involved in leukoencephalomalaciain horses, esophageal cancer in humans and hepatocarcinomasin ducks and mice, but these clinical effects are now believed to be caused by the fumonisins. The fusarins have been found to be naturally occuring in corn (541)and have been analyzed by TLC and HPLC (541-549). Gelderblom e l a/.
(548)and Jackson et a/. (547)used a NP-HPLC method both for preparative and analytical HPLC. They employed chloroform/methanol (19:l)as eluent at 1.5 ml/min. This eluent or methylenechloride/methanoI(l9:1) have been used for the determination of fusarin C in cereals. Fusarin C was detected at 365 nm or 350 nm. However fusarin C is quite unstable at some conditions and should be analyzed accordingly (545,550). In the system of Frisvad and Thrane (47)the UV break-down products reported by Scott eta/. (545)were not observed (546).The break-down products have another UV spectrum which could be detected easily by diode array detection, if they were present. The break-down products, never observed by us, may have been caused by chloroform.
317
8.4.1 3.4. Moniliformin Moniliformin has a characteristic UV spectrum (maxima at 227 and 261 nm) and can be detected by RP-HPLC-diode array detection using water/acetonitrile with trifluroacetic acid (47). However in that HPLC system the retention is weak and more dedicated methods have been developed for this mycotoxin, especially based on ionpairing extraction and chromatography (275,549,551-552). Eluents based on water/methanol or water/acetonitrile have been used together with different ion-pairing reagents (tetra-n-butyl-ammoniumhydroxide), and phosphate buffers (551-552). 8.4.13.5. Fusaric acid Like moniliformin, cyclopiazonic acid, terrestric acid, tenuazonic acid and dipicolinic acid, fusaric acid has also been analysed by ion-pairing HPLC (553-554). It is not known if fusaric acid is an important in any mycotoxicosis. 8.4.13.6. Gibberellins The gibberellins are phytotoxic secondary metabolites from Gibberella and
Fusarium species, but it is not known, whether they have any role in mycotoxicosis. HPLC methods for gibberellins have been summarized by Lin eta/. (555), which used a gradient RP-HPLC method from 35
methanol in water, containing 0.05 % acetic
acid to 100 Yo methanol at a flow rate of 1 ml/min. They included 66 different gibberellins in their assay. 8.4.14. Miscellaneous toxins 8.4.14.1. Fumagillin Fumagillin, a secondary metabolite of strains of Aspergi//usfumigaius, has been analyzed by RP HPLC using water/acetonitrile/ acetic acid (500:500:1.5), and UV detection at 351 nm (556-557). The HPLC method of Assil and Sporns (557) also included an ELSA screening technique for this antiprotozoan metabolite in honey. 8.4.1 4.2. P-nitropropionic acid This mycotoxin has recently been reported to be produced by Arthrinium
sacchari, A. saccharicola and A. phaerospermum in sugarcane causing severe
318
poisoning in humans (558), and it may also be produced by an artificially inoculated strain of Aspergillus oryzaeon cooked sweet potato, white potato, banana and cheddar cheese (559). RP-HPLC was employed for the analysis of P-nitropropionic acid in plasma after perchloric acid treatment, using an isocratic 0.15% phosphoric acid eluent at 0.9 ml/min followed by rinsing of the column by a methanol (0-35 %) gradient and detected at 210 nm (560-561), but the method was later modified to use a rinsing gradient of acetonitrile (040%) (562). 8.4.14.3. Cyclopiazonic acid The principal producers of cyclopiazonic acid in foods and feedstuffs are Penicillium commune, P. griseofulvum, Aspergillus flavusand A. tamarii.These species
are very common and cyclopiazonic acid has been found as a natural contaminant of corn, cheese, peanuts and millet (563). Peterson et al. (563) developed a HPLC method for purification of cyclopiazonic acid and they used NP-HPLC for preparative purification based on an eluent of chloroform/methanol(99:1or 99.5:0.5) on a silica gel column preparated with oxalic acid. For analytical HPLC they used an amino column and an eluent consisting of 25 mM potassium dihydrogen phosphate/methanol (223). UV detection at 282 nm was necessary using the chloroform containing eluents,
whereas the stronger absorption at 225 nm could be used with the buffer-methanol eluent. Lansden (564) developed a RP HPLC method for cyclopiazonic acid, based on the method for tenuazonic acid (505) and this method was later modified by Norred et al. (565). However the peak of cyclopiazonic acid is still quite broad in the system of Lasnsden and co-workers (564-565) and Frisvad and Thrane (47). Goto et al. (566) developed a dedicated sensitive NP- HPLC method for cyclopiazonic acid, using an eluent more like those used in NP-TLC: ethyl acetate /isopropanol/25% aqueous ammonia . This is one of the few systems employing bases in the eluent, but it may be efficient for alkaloids. However for LC-diode array detection or LC-MS RP-HPLC systems with volatile main eluents and acids, bases or buffers are preferred. 8.4.14.4. Roquefortine C Roquefortine C is produced by a large number of Penicillium species, some of which occur very regularily in foods and feedstuffs (Table 8.1). It has been found in feed grain causing mycotoxicosis (567). Roquefortine C and the related secondary
319
metabolites meleagrin and oxaline are difficult to analyze by TLC and HPLC (47). For TLC the eluents chloroform/acetone/isopropanol or chloroform/ammonia/methanolare very efficient for separation of roquefortine C,meleagrin and oxaline (45) but for diode array detection RP system are preferred. Ware et a/. (568) developed a RP HPLC system for roquefortine C in blue cheese using ethyl acetate extraction of melted cheese added diatomaceous earth and partition into a hydrochloric acid solution. After neutralization roquefortine C was extracted back with ethyl acetate and the toxin was analyzed by HPLC in an eluent of a 0.05 M monobasic ammonium acetate in water/methanol (1 :I). Danieli eta/. (569) used RP-HPLC to analyze for roquefortine C in cheese employing a gradient of buffered water (pH 4)/acetonitrile. Experience in our laboratory has shown that trifluoroacetic acid in both the water and acetonitrile part of the gradient will give a good peak shapes of roquefortine C, meleagrin and oxaline. 8.4.14.5. PR-toxin PR-toxin has only been found in Penicillium roqueforti var. roqueforti (16). Moreau el a/. (570) developed a NP-HPLC method for PR-toxin and eremofortins A,
B and C using chloroform as the eluent and thsi method was also used later (571-572). Danieli el a/. (573) later developed a RP-HPLC method for PR-toxin using water/acetonitrile (65:45) as eluent and UV detection at 250 nm. Frisvad and Thrane (47) (see illustration in ref. 45) used a water/acetonitrile gradient to separate several toxins from P. roquefortivar. roqueforti. 8.4.15. Multi-mycotoxin analyses by HPLC
A series of papers have been published on HPLC multi-mycotoxin analysis and several
of
those
have
been
mentioned
above
(4,46-47,161,236-238,
386,392,396,545,574-577). They often cover chemically related mycotoxins or mycotoxins that are present in the same commodities. The associated mycoflora (29) of different foods and feeds may help in determining the mycotoxins that should be included in a particular HPLC multi-mycotoxinmethod (31). It is characteristic for most of these methods that they employ acidic extraction with either acetonitrile, chloroform or ethyl acetate, followed by partition with petroleum benzine or hexane to remove lipids and often mini-column clean-up steps. Most applications use RP determinations with either methanol or acetonitrile with acidified water (4,46,47,161,236,237). The
320
method of Hurst et a/. (238,392)for patulin, penicillic acid, zearalenone, sterigmatocystin and ochratoxin A in cocoa beans employs a cyano column and an eluent consisting of hexaneln-propanollaceticacid with UV detection at 245 and 280 nm. Most multi-toxin methods should cover several chemically different mycotoxins and therefore a diode-array detector or a mass selective detector may be particularily well suited for these analysis. This require, however, volatile buffers or acids and eluents with low absorbtion in the spectral range from 200-600 nm, hence in general one should chose among the following eluents and buffers: methanol, ethanol, tetrahydrofuran, acetonitrile, water, trifluoroacetic acid, acetic acid, triethylamin and ammonium acetate (see below). The HPLC method of Frisvad and Thrane (47)now include approximately400 standards of mycotoxinsand secondary metabolites (Table
8.2)and is thus of very general applicability. Retention times of the different mycotoxins will vary significantly between different batches and brands of columns (46).Hill et a/. (236)therefore suggested to use an alkylphenone retention index system for mycotoxins. This was taken up by Frisvad and Thrane (47)and later Paterson and Kemmelmeier (578)and Kuronen (4). Kuronen (4)pointed out that alkylphenone retention indices may be less accurate with compounds outside the range of the index compounds. The compounds he suggested, 1-[4-(2,3-dihydroxy-propoxy)phenyl]-I-alkanones, may be slightly better for the
purpose, but are not commercially available. Experience has shown that the alkylphenones work excellently, but a little less precise for very fast and slow eluting compounds. However such compounds are always those for which better analytical methods could be developed anyway. In practice the compounds with less precise retention indices may then be recognized by their UV-VIS spectra (diode array detection). The method of Frisvad and Thrane (47)also included a confirmation of identity by using NP-TLC in two different eluents. Even though retention indices may be more stable than retentiontimes, an extra correction may be necessary (579).Sole reliance on retention indices may cause misidentification of unknown secondary metabolites and mycotoxins. For example Paterson and Kemmelmeier (578)reported, based on retention index data, that the unrelated Penicillium brevicompactum and P. citrinum produced mycotoxins such as ochratoxin A, viridicatum-toxin, griseofulvin,
xanthomegnin, viomellein, and several other metabolites which are not produced by these species (16,19,21). This emphasizes the importance of using standards and
32 1
confirmatory tests. Inter-laboratorystudies may be necessary to standardize retention index data to a level where preliminary identification can be suggested. However confirmations of identity are still of major importance. 8.5. INFORMATIVE ON-LINE DETECTION METHODS 8.5.1. Applications of HPLC diode array detection Diode array detection (DAD) giving full UV-VIS spectra in the range of (190-) 200-600 nm, have put a new dimension into HPLC analysis. A large number of mycotoxins and other fungal secondary metabolites, but also food constituents, have very characteristic spectra (47,538) which can be used for confirmation of identity, peak purity determination, peak "unmixing" and optimal selection of detection wavelength. Fortunatelymany of the mycotoxins with weak uncharacteristic UV spectra can often be analyzed more efficiently by e.g. gas chromatography (e.g. the type A trichothecenes). HPLC-DAD has been used to analyze very complex mixtures of secondary metabolites from fungi (1 6,20-27,47-46,510,518,546) but can also be used for foods and feedstuffs (518,577, Frisvad, unpublished). Frisvad and Thrane (47) reported on TLC data, HPLC retention indices and UV maxima for 182 mycotoxins and fungal secondary metabolite data. This data base has now been expanded considerably,including more precise UV-VISdata. These data are presented in Table 8.2, which includes retention indices and UV spectral data reported as all maxima, minima and shoulders and their relative absorptions related to the largest absorbtion (100%). Many of the spectra have been plotted in ref. 538. The original data-base (47) was quite meagre concerning UV data as only maxima and an indication of the largest and next largest absorption were given. It is well known that UV spectra, in contrast to IR and MS spectra, are often dependent of pH and type of eluent (580-582). This make comparisons between literature data taken in methanol, ethanol and data recorded by the diode array detector difficult in several cases. Again it is ernphazised that standards should always be used to verify identity of a mycotoxin and not just by comparison to literature data on UV spectra. The gradient used by Frisvad and Thrane (47) will start with a pH value of ca. 3.5 and end with a pH of ca. 2.7, but this little change does not affect the chromophore absorbtions for compounds eluting early and late in the 50 min analytical run-time. However a change from acetonitrile to methanol or especially neutral or
322
TABLE 8.2. Retention indices (RI) and UV data of fungal metabolites as measured on-line with a diode array detector in the HPLC system of Frisvad and Thrane (47). Absorbing wavelengths are given in nm. Wavelengths marked are maxima, those marked 's' are shoulders and the other figures are minima. After each wavelength (i.e. after ' I ') the relative absorption in % of the maximal absorption is given. In case the only absorption maximum is below 200 nm, UV data is presented as 'end'. I*'
Fungal metabolite 4-ace famido-4-hydroxyFbutenoic acid y-lactone (butenolide)
7 '-acetoxypaxilline
RI
673
7386
UV Data
........
........
230'1 700
2651 77 280'1 27
201') 700
2061 45
15-acetoxyscirpenol
783
end
3-acetyldeoxynivalenol
747
279'1 700
33-acetyldiacetoxyscipenol
986
end
a-acetyl-y-methyl-tetronic acid
669
2071 74
237'195
244155 263'1 700
756
207'1 700
2394 35
2671 79 306'1 29
Acetyl T-2 toxin
7 757
end
Aculeasin A y
7216
220sI 28
250)3
274') 4
AflatoxicolB
929
207'1 700
2391 75
257~127 261 '1 26 2751 4 333'1 37
Aflatoxin 8,
895
2721 57
225'1 68
257 I26
267'1 44
2881 6 362'1 74
Aflatoxin 8,
867
2121 59
278'1 67
2334 54 253 I 28 267'1 44 2881 6 364'1 82
323
TABLE 8.2. (continued) Fungal metabolite
RI
UV Data
........
....,...
G,
865
201'1 100 2101 82
218'1 87 242s I 36 251 I28 265'1 37 2831 3 368'1 64
Aflatoxin G2
834
2081 85
216'1 93
242'1 44 2531 34 265'1 40 2841 3 368'1 79
Aflatoxin GZa
769
2081 83
216'1 88
242'1 41 2551 32 265'1 34 2841 3 366'1 71
Aflatoxin M,
820
2161 55
229'1 71
251 I28 265' I 42 2851 7 358*161
Aflatoxin Mz
785
216'1 53
2481 21
251 *I 23 2531 21 263'1 29 2851 4 356'1 53
1514
2141 65
231'1 100
2651 27 283'1 30
Agroclavine
754
2081 79
223'1 100
2461 6 275~121 281 '1 22 290s I 18
Altenuene
839
214 I 24
240'1 100
261 I20 281 *I 34 3041 16 319'1 18
Aflatoxin
Aflatrem
324
TABLE 8.2. (continued) Fungal metabolite Aiternariol
RI
........
........
276~149 237132
255'1 700 2751 77 288'1 27 2941 79 298'1 27 3081 70 337'1 23
7074
2331 22
257'1 55
2791 72 287'1 74 2941 72 298'1 73 3701 7 339'1 74
902
203165
273'187
237 I 30 259'1 700 2791 45 285'1 48 3731 6 355'1 76
Anhydrofusarubin
7766
278155
237'169
2551 44 290'1 68 245s I 7 7 3991 0 540'1 33
para-anisaide hyde
798
208137
223'157
257 I 2 288'1 79
Antibiotic Y
957
276.~173 227167
243'1 83 2631 67 273s I 67 287'1 77 306s I 43 3791 38 347'1 57 3541 48 364' I 53
Ascochitine
7080
2031 76
2331 59 263 *I 92 279'1 700 341'1 75 3561 70 476'1 39
A Iternariol-monomethylether
Alterfoxin I
935
UV Data
275'190
325
TABLE 8.2. (continued) ~
~~~
RI
UV Data
........
........
1881
2161 51
221 *( 51
2701 6 316'1 16
Asperthecin
885
2181 49
237'1 88
2461 68 263'1 100 2791 52 287'1 56 3021 31 317'1 35 351 16 482'1 55 509s 1 46
As teltoxin
983
272~131 2351 18
Fungal metabolite Aspergillic acid
273 I 93
2961 15
366'1 700
Asterric acid
969
212'1 100
239 I 23
251'1 27 2791 4 315'1 15
Aurantiamine
868
2121 42
231 *I 54
269 1 24
Aurantioclavine
758
2121 75
225'1 100
251 15 285'1 24 292s I 23
1129
272s 1 33
2351 20
273'1 84 2941 7 364'1 100
Austamide
907
2031 64
214sl 72
233'1 100 2551 47 265'1 48 285s I 35 341 14 395'1 11
Austdiol
700
203'1 34
225)8
257'1 51 3101 3 381'1 100
Aurovertin B
321'1 100
326
TABLE 8.2. (continued) RI
UV Data
........
Averufin
7 342
2091 60
224'1 700 247 I36 257s I 48 269'1 55 2731 54 293' I 98 3751 26 322'1 26 3531 70 453'1 37
Barnol
837
207'1 700 225~127 267 I 2 275'1 2
Benzoic acid
747
2701 32
237'1 97
2591 6 275'1 8
Bostrycin
750
208'1 60
2721 59
227'1 700 2651 74 302'1 30 257 12 480 I 24 503 I 26 540'1 76
Fungal metabolite
........
Bostrycoidin
1046
207'1 700 2251 37
257'1 95 2921 72 323 I 22 3871 5 488'1 28 522.~178
Brassicasterol
2060
2721 65
2221 65 227'1 66
Brefeldin A
97 7
214sl 65
Brevianamide A
869
2071 66
233'1 700 265s I 22 2881 7 407'1 73
7097
205s I 95
233s I 4 7
897
207'1 700
Byssochlamic acid Canadensolide
276'1 66
253s I 28
327
TABLE 8.2. (continued) RI
UV Data
........
........
Canescin
912
2071 15
244'1 100
2771 74 277'1 15 287s I 11 2981 4 337'1 13
Carlosic acid
690
2031 72
233'187
246154 265'1 100
Carolic acid
677
2001 15
231'1 90
2441 51 265'1 700
Catenarin
7 798
2081 39
231'1 100
2441 41 257*150 2631 46 277'1 54 2921 31 304'1 33 3431 4 477s I 36 488' I 42 520s I 27
Chaetoglobosin C
7176
2081 79
221'1 700
240126 251 7 27 279~117 288~114
706
208 I 78
223'1 100
244 I 6 273~127 281 *I 23 290s I 20
7 137
201'1 100
223s 64
2631 72
6a-chlamydosporol
729
205'1 100
233s 8
24214 285* I 30
6Q-chlamydosporol
720
205'1 100
233s 8
2421 4 285'1 30
Chromanol 7
917
227'1 100
2441 16
267'145
Chromano12
909
223'1 100
2441 76
269'1 45
Chromanol3
835
227'1 100
2441 16
269'147
Fungal metabolite
Chanoclavine
Chetomin
287'1 13
328
TABLE 8.2. (continued) RI
UVData
,
.......
........
1124
2101 58
223' I 79
235 I 26 255'1 44 283'1 21 3081 3 428'1 19
719
201 *I 83
2121 52
229'1 100 251 I 19 265' 1 25 273s 1 23 2871 10 304'1 16 313sl 13
Chrysophanol
1244
2081 51
225') 100
237133 259'1 65 2751 30 279'1 31 2831 30 287'1 31 3081 2 428'1 30
Citreomontanin
1679
229s 1 28
2481 18
267'121 2851 15 317*122 3411 18 413'1 84
Citreoviridin
1051
205'1 58
2231 21
239'1 27 2591 21 288s I 89 294'1 97 321 I24 388'1 100 403s I 94
Citreoviridin A
1070
201'1 53
221 I 16
237'1 21 261 I 12 285s I 48 294 I 53 3191 13 366.~173 387'1 100
Fungal metabolite
Chrysazin
Chrysogine
329
TABLE 8.2. (continued) RI
UV Data
........
........
7074
2081 22
276'125
2351 77 273'1 87 3001 13 368'1 700
Citrinin
867
2031 78
214'1 100
2784 97 246s I 56 2791 9 327'1 44
Citromycetin
697
214'1 700 2421 52
251 *I 53 283)27 302'1 37 323 I 26 358'1 48
2791 53
2471 47 265sl 72 269'1 75 294s I 38 3331 5 445. I 33
Fungal metabolite Citreoviridin X
Clad0fulvin
7260
Cladosporin
986
274'1 700 2401 72
Clerocidin
960
2031 76
235'1 700
Compactin
7208
203 1 24
237'1 92
23-3197 237') 700 244s I 68
Curvularin
976
2731 62
227'1 65
233~157 257 I20 277'1 34 2881 25 300'1 26
Cyclochlorotine
828
207'1 100
Cyclopaldic acid
833
2151 9
245'1 100
273.~129 323s'l 7
233'1 74
267'1 63 288 1 24 298 I 26
330
TABLE 8.2. (continued) RI
UV Data
........
......
Cyclopenin
863
203) 93
210'1 100
231~159 251s 1 25 2751 5 288'1 7
Cyclopenol
771
201'1 100
214.~167
233~135 2691 5 285'1 8
Fungal metabolite
Cyclopiazonic acid
1169
2081 65
225'1 100
246123 281 '1 51
Cynodonfin
1369
2231 22
239'1 35
2514 16 2691 5 290'1 6 3681 1 516'1 12 5281 10 542'1 1 1 5481 1 1 550'1 72
Cytochalasin A
1129
216s I 48
229s I 32
Cytochalasin B
1015
212.~166 231~124
Cytochalasin C
1074
end
Cytochalasin D
1004
end
Cytochalasin E
1058
end
Cytochalasin H
1004
end
Cytochalasin J
900
end
bis-dechlorogeodin
919
207'1 100
216~182 2511 15 285'1 61 339s I 13
1076
205'1 100
255~130
201 122
248'1 100
Dechloronidulin Dehydrocarolic acid
68 1
271 I34 296 I 49
33 I
TABLE 8.2. (continued) Fungal metabolite
RI
Dehydrocurvularin
854
UV Data
........
........
216~163 233~146 2531 21 2831 54 304'1 70
Dehydropaxilline
1398
2081 49
232'1 100
248~148 279s I 23
Demethoxyviridiol
809
2191 21
253'
100
2831 15 321'1 49
Deoxybostrycin
84 1
2121 51
227'
100
2671 12 304'1 29 3521 2 471~123 501'1 27 536~117
Deoxynivalenol
685
218'1 100
Dermoglaucin
1078
2031 85
211'1 100
233~148 2461 38 265s I 56 283' I 92 3371 5 430'1 37
Desacetylpebrolide
9 6
2101 5
232'148
26313 273'1 4
Desertorin A
958
210*) 100
237~142 2651 12 294s I 36 308'1 48 319~144
Desertorin 6
1044
210'1 100
237s 41
2621 13 298s I 39 308') 48 319~144
Desertorin C
1111
210'1 100
223s 78
237~147 2651 16 296~141 308'1 50 319~143
Desmosterol
1920
end
332
TABLE 8.2. (continued) RI
UV Data
........
........
Dethiosecoemestrin
1568
2121 57
225'1 78
2481 5 283'1 13 290.~112
Diacetoxyscirpenol
866
end
4,15-diacetylverrucarol
95 1
end
Diethylphthalate
996
205s I 94
Fungal metabolite
Dihydrocytochalasin 6
1127
208'
I 100
2161 83 221 *I 87 223 I 86 231 * ( 93 2631 12 277'1 17
end
Dihydroergotamin
96 1
2141 72
219'1 74
2441 5 279'1 14
cis-dihydrofusarubin
803
208s I 76
227 I 44
244'1 86 261 I25 277'1 34 302'1 24 3331 9 391'1 41
trans-dihydrofusarubin
846
see above
2',3'-dihydrosorbicillin
1194
2031 74
216'1 100
23 1s55 2481 7 285'1 76 3171 28 329 I 29
Dihydroxyaflavinine
1056
21 1 I 71
225'1 100
2571 10 277.~116 283'1 18 288s I 76
2,4-dihydroxy-6-(1,Pdioxopropyl) benzoic acid
680
214'1 100 2391 18
261 *I 46 2791 15 296'1 23
333
TABLE 8.2. (continued) RI
UV Data
........
........
2,4-dihydroxy-6-(1,2-dioxopropyl) benzoic acid, lactol
855
221 150
239'1 100
253~167 2751 18 298') 25 3231 18 347'1 22
2,4-dihydroxy-6-(1-hydroxy-2. oxopropyl)benzoic acid
717
212'1 100
240122
261 *I 33 2871 18 298' I 20
2,4-dihydroxy-6-(1-hydroxy-2oxopropyl) benzoic acid, lactol
698
212'1 100
227~157 2421 15 269'1 48 2881 24 302'1 28
2,4-dihydroxy-6-(2-oxopropyl) benzoic acid
719
214'1 100
2391 18
2,4-dihydroxy-6-(2-oxopropyl) benzoic acid, lactol
807
214'1 100
228.~163 2401 13 271 *I 63 2921 26 300'1 28
Dihydroxysterigmatocystin
1069
207.~176
221 I59
233' I 77 2351 77 249'1 100 2791 11 327' I 50
2,7-dimethoxy-6-(1-acetoxyethyl)-juglone
1015
218'1 100
239126
263'1 51 2791 9 308'1 33 3431 4 427') 13
1150
218'1 100
237125
263 I 59 2791 9 310'1 33 3451 3 426'1 15
Fungal metabolite
902
205~194 209'1 100
263'1 37 2871 18 298'1 21
229 1 23 241 *I 33 261 I 1 298'1 14
334
TABLE 8.2. (continued) Fungal metabolite
RI
UV Data
........
........
Dimethylphthalate
851
2031 93
208'1 100
2161 85 223s 1 88 231 *I 95 2631 12 277'1 18
Dipicolinic acid
675
223s I 26
2461 6
267.~112 272'1 14 281sJ10
Dithiosilvatin
1152
221 I33
227'1 36
233s I 3 1 251 1 10 273'1 21
Dothistromin
1061
2161 41
223'1 42
2421 16 257sl 18 267s I 20 292') 33 317sl 10 3491 5 455'1 12
Duclauxin
1137
203' I 76
2111 68
229'1 100 265s I 33 3061 11 321'1 13 345s I 11
Echinulin
1370
2141 80
231'1 95
2571 1 1 283 I 22 294s) 18
693
2071 86
223'1 100
2461 8 275s 1 24 281'1 25 288s I 2 1
Emestrin
1036
227s I 51
259s I 27
267s I 25 283~116
Emestrin B
1050
225s)53
248)28
265') 34 285s I 25
Emindole DA
1560
211162
225'1 95
2481 5 279~116 283') 16 288s)15
Elymoclavine
335
TABLE 8.2. (continued) RI
UV Data
........
........
Emindole DB
1557
201'1 85
2081 69
223'1 100 2481 6 275s I 20 283'1 22 290s I 20
Emodin
1132
2071 58
223'1 100
2371 37 255s I 51 267'1 56 275)52 288'1 60 331 14 441 *I 33
Epicorazine A
847
2081 94
218'1 99
271~15
Epoxyagroclavine
704
2071 55
223'1 100
2461 12 275s I 22 281'1 23 288s I 20
Epoxyagroclavine-AgroclavineN,N-dimer
850
221sl 78
2481 15
265s I 23 273'1 26 283s I 24 292s I 21
Epoxyagroclavine-N,N-dimer
806
216~170
2481 12
265~119 273'1 21 283s I 19 292s I 16
Epoxysuccinic acid
677
end
Fungal metabolite
1381
221 I 28
235'1 37
261 I25 294 1 59
Ergocristine
991
239s I 4 1
271 I 4
319'1 17
a-ergokryptin
970
221.~163 239s I 53
271 15 319'1 22
Ergometrin
715
2141 88
227'1 96
237s I 88 2691 8 313'1 36
Equisetin
336
TABLE 8.2. (continued) Fungal metabolite
RI
Ergosterol
1585
Ergotamin
947
Eryf hroglaucin
7428
UV Data
........
.......
2461 9
265sl 7 7
273'1 f 4 2771 13 283'1 14 294~19
237~147
271 I 4
319'1 77
2741 44
237'1 68
2441 28 257'1 34 265)30 275'1 34 2921 79 303 I 20 347 13 488' I 28
2271 60 237'1 66 263 I 24 296s I 87 323'1 100
Ethisolide
73 1
203'1 700
Expansolide A
978
end
Expansolide 6
1008
end
I
Ferulic acid
749
2071 57
278'1 71
Festuclavine
762
2081 60
223'1 100 2441 8 273s I 20 279'1 21 288~178
Flavoglaucin
1557
2271 27
239'1 37
9 76
207 I 34
237'1 700
Frequentin Fructigenine A
1728
205~199 2271 72
2561 14 277'1 30 3081 7 389'1 74
246'125 279s 1 4 285s) 4
337
TABLE 8.2. (continued) RI
UV Data
........
Fulvic acid
939
2031 96
208'1 700 2271 62 233 1 64 271s I 22 288)15 337'1 42 3451 41 389'1 78
Fumigaclavine A
725
2071 53
223'1 100
Fumigaclavine B
690
2071 52
223'1 700 2421 6 279'1 79 290sl 15
Fumigaclavine C
881
2701 55
229'1 100 24818 283'1 25
Fumitremorgen A
7387
2191 68
225'169
Fungal metabolite
Furnitremorgen B
7 797
2751 77
Fumitremorgen C
956
278'1 73
Fusarenone X
706
227'1 100
Fusaric acid
77 7
Fusarin C Fusarochromanone
229'1 86
........
2421 5 273~179 279'1 20 288~117
2551 273'1 2831 292'1
70 14 72
2571 279'1 2851 296'1
77 78
72
17 20
269'1 77
290'1 75
274 I 28
227'1 34
2461 70 273'1 46
7 060
2571 73
368'1 700
727
272'1 82
2331 37
257'1 700 3691 35 281'1 42 370sJ 75 3331 5 387'1 55
338
TABLE 8.2. (continued) Fungal metabolite Fusarubin
RI
UV Data
........
........
90 1
212157
227'1 100
2671 13
302') 32 3521 2 469s I 23 496'1 26 530s I 16
1313
218.~165
Gallic acid
68 1
214'1 100
2401 10
271'141
Genfisylalcohol
669
221sI 26
2531 2
292'1 17
Gibberellic acid
736
203'1 100
Gladiolic acid
771
2051 32
231 *I 100
2571 31 267'1 35 302.~113
Glauconic acid
93 1
212'1 100
Gliotoxin
833
2441 27
269'1 34
Griseofulvin
999
212'1 98
2271 81
Griseophenone C
936
205'1 100
227~144 2481 5 297'1 50 337sl 15
HT-2 toxin
928
end
Hadacidin, Na'
669
end
Fusidic acid
Helminthosporin
1317
211145
231'1 100
Helvolic acid
1260
214 I 68
233'1 81
237'1 88 251~169 271 I47 292'1 100 325s I 23
242134 255'1 49 281 I20 288 * I 23 3191 3 484'1 30 518sl 18
339
TABLE 8.2.(continued) RI
UV Data
........
........
5 '-hydroxyasperentin
825
274'1 700
2401 72
267'1 63 288 1 24 300'1 26
para-hydroxybenzoic acid
676
208s I 89
2271 78
255'1 97
Hydroxyisocanadensicacid
728
225'1 700
5-hydroxymalfol
692
276s I 59
2351 72
263~142 287'1 66
4-hydroxymellein
755
208'1 700
2291 72
246'1 22 2751 2 373'1 75
w -hydroxypachybasin
978
274 I 58
227'1 60
237 I48 257s I 88 259'1 700 279s I 42 3041 6 335s I 7 0 403'1 20
lndolacetic acid
766
2041 67
278'1 700
2451 8 280'1 27 289~177
7349
2741 48
237'1 83
2421 46 253'1 56 2751 75 290'1 20 3251 3 465s I 24 488 I 29 524s I 78
Isochromantoxin
882
227s I 20
2531 7
283'16
lsoemodin
985
2081 46
225'1 700
2371 34 257'1 65 279s 1 29 287s I 29 2081 2 429 I 30
Fungal metabolite
lslandicin
340
TABLE 8.2. (continued)
.,......
RI
UV Data
........
lsomarticin
972
210158
227'1 100 2671 74 304'1 29 357 I2 480s I 24 498'1 25 534~175
ltalicic acid
907
270120
237~142 240'150 2571 42 269'1 45 292 I 29 335'1 700 352sl 65
ltalicic acid-methylester
7052
272123
239'150
Janfhifrem 6
7172
237~133 240126
Fungal metabolite
257139 271 *I 47 285s I 38 294 I 29 335' I 700 352s I 68 263'157 287 I 9 331 *I 34
Ja vanicin
968
212157
227'1 100 267)73 305'1 37 3561 2 475s 1 23 502'1 27 536s I 7 7
Kojic acid
673
203165
276'1 700 235140 244s145 269'1 73
1097
2031 92
208'1 700 2374 35 2591 75 296s)36 306'1 41 377~135
Kotanin
341
TABLE 8.2. (continued) Fungal metabolite Larnbertellin
RI
UV Data
........
........
961
2051 92
211'195
233~171 261 I48 283~167 290 I 69 298s I 27 3351 4 432'1 27
Lanosterol
1962
Lapidosin
898
216'1 100
231 I 78
239'1 81 2651 46 273'1 48 321 I 13 335'1 14
Lichexanthone
1377
207sl 83
221 151
242'1 100 267~133 281 I22 309 I 63 339s I 24
Luieoskyrin
1269
2331 12
253.91 17
261'120 2671 18 273'1 19 2851 14 296*1 17 3251 9 445'1 100
Macrosporin
1133
2181 53
225'1 56
Malformin A
1031
end
243 I 27 285'1 100 306s I 38 341 I 14 381'1 21
Malforrnin B
1047
end
Malformin C
1036
end 227'1 100
2671 13 304'1 31 351 12 4 75s I 24 498'1 26 532s I 16
Marticin
942
end
2121 58
342
TABLE 8.2.(continued) Fungal metabolite Meleagrin
Methoxysterigmatocystin
RI
UV Data
........
........
849
2181 60
229'1 67
259 I 20 283s I 25 329'1 66
1072
207'1 82
2781 50
239'1 700 2751 70 315'1 42
3-methoxyviridica tin
995
203'1 100 2121 88
221 '1 98 2631 16 281 *I 21 3001 14 315.~179 323 I 23 335s I 17
6-methylsalicylic acid
760
207'1 100
238~177
2651 2 300'1 9
Mevinolin
7274
2031 27
233'1 89
2351 88 239'1 700 246s I 67
Mitorubrin
7 098
2031 74
273'1 85
2351 42 267'1 100 292s I 63 3701 47 351'1 88
Mitorubrinic acid
946
212'1 78
239) 47
273'1 700 300s I 59 3171 49 349'1 66 366s I 52 395s I 27 422s I 27 455sl9
Mitorubrinol
936
2031 74
213'1 85
2351 45 265'1 100 292s 1 64 3701 46 351 *I 86 364~179
343
TABLE 8.2. (continued) RI
UV Data
........
........
Mitorubrinol acetate
7 059
2041 72
274'1 83
2371 52 265'1 700 292s)62 3061 46 349'1 84
Mollisin
7 756
207' 700 233130
Fungal metabolite
259'1 67 277s I 42 3061 3 478'1 72
Moniliformin
670
2031 75
Monorden
923
205'1 100 274sl 77
2421 27 275'1 52
2294 31
2651 75
379'1 55
274'1 700 2351 74
249'1 22 2731 2 303'1 7 7
7442
2701 72
274'1 75
2371 37 255'1 65 288.~176 3731 6 356' I 34 374s I 27
Nectriafurone
949
2781 57
240'1 74
257 I 77 259'1 73 2871 72 323' 1 23 3771 7 1 443'1 49
Neosolaniol
723
end
205'1 700 2764 89
2571 78 267'1 79
Mycelianamide Mycophenolic acid
Naphthalic anhydride
Nidulin
7202 977
7487
Nigragillin
773
269'1 700
P-nitropropionicacid
678
208'1 700
Nivalenol
676
227'1 700
227'1 700 2451 26 267'1 30
344
TABLE 8.2. (continued) Fungal metabolite Norninine
Norjavanicin
RI
UV Data
........
........
7 620
2721 57
227'1 82
2481 4 277sl 73 284'1 74 290s 1 13
89 1
2741 57
223'1 62
2671 72 298'1 27 34.31 3 497'1 77
Norlichexanthone
7 000
205.~168 2781 47
Norsolorinic acid
7524
2731 69
Nortryptoquivaline
1763
208'1 700 2271 75
229'1 87 253s 1 38 267s I 24 275s 1 20 2981 7 304'1 8 377sl 7
Ochratoxin A
7091
2051 94
275'1 98
248s I 2 7 2831 2 332'1 77
Oosporein
667
205'1 99
237 I37
257~139 289'1 700
Orsellinic acid
777
272'1 700 2351 79
Oxalic acid
676
203'1 99
247'1 700 267s I 26 2871 79 373'1 60 346s I 20
235'1 700 2551 56 275~179 304'1 87 352sl 78 366sl 77 3971 74 463'1 36
246sl 7
253'1 33 2771 9 292'1 74
345 TABLE 8.2. (continued) RI
UV Data
........
........
883
2181 59
229'1 65
2571 21 283s I 26 327'1 66 339s I 57
Pachybasic acid
1004
205'1 87
221 I69
225'1 69 2371 56 259'1 100 279s I 36 3001 6 337'1 10 3491 10 403'1 21
Pachybasin
1232
2141 56
223'1 60
231 I52 248~184 259 I 98 277s I 47 3021 6 333'1 10 351 19 405'1 20
Fungal metabolite Oxaline
Palitantin
886
231'1 100
Parasiiicol
880
205'1 100 216~157 2421 15 255s I 21 263'1 23 2751 4 329'1 27
Paspaline
1617
2081 53
231'1 100 251 18 281 *I 24
Paspalinin
1332
2101 59
231'1 100 273~126
680
2051 10
230~124 277'1 100
1291
2081 49
231'1 100 2691 20 281 *I 22
Penicillic acid
715
2051 45
229'1 100
Penicillin G
669
208'1 100 2441 4
259'1 4
Penitrem A
1342
225 I 46
259 I 7 296'1 17
Patulin Paxilline
235'1 50
346
TABLE 8.2. (continued) Fungal metabolite
RI
UV Data
........
L,L-phenylalanine anhydride
867
2441 1
259'1 2
Phoenicin
721
2051 74
214'1 79
........
2351 43 267'1 100
3391 1
488'1 7
Phomarin
1097
2051 59
218'1 94
239 I 30 269'1 100 294s I 48 3251 7 413'1 23
Physcion
1340
2101 63
223'1 82
2391 37 253s I 40 267'1 43 2751 40 287'1 43 300s 1 29 3331 5 439'1 27
218'1 100
257~130 2671 18 292s I 30 323'1 40 354s I 7 1
PI-3
PR-1635
802
PR-toxin
86 1
2781 23
249'1 100
Preechinulin
890
2121 70
225'1 85
2531 10 283'1 20 288~118
1319
212.5 73
2231 53
239'1 73 2481 66 267'1 77 3131 11 351'1 17 395)4 501'1 17 536sl 11
Purpurugenone
2831 79 363'1 100
347
TABLE 8.2. [continued) Fungal metabolite
RI
UV Data
........
........
Pyrogallol
679
203~197 207'1 100
2-pyruvoylaminobenzamide
679
218'1 100
244.~128 27314 296'1 8
1027
2081 67
223'1 100
239136 251s I 39 285. I 62 3231 6 436'1 26
Questinol
867
2071 67
223'1 700
239141 246'1 44 2551 42 269s I 52 285') 63 3231 6 434'1 28
Ravenelin
1089
221 1 45
233'1 62
2371 61 261'1 700 2881 10 339'1 37 3771 70 397'1 11
Riboflavin
786
201 I41
218'1 100
231 I45 248s I 84 261'1 100 2851 2 352'1 30 387s I 22
Roquefortine A
743
2071 55
223'1 100
2441 5 281 * ) 22
Roquefortine B
686
2071 55
223'1 100
2421 6 281 *I 21 288.~118
Roquefortine C
922
205~192 2231 33
Quesfin
225.~154 2481 4 269'1 6
233') 34 2631 19 304'1 54
348
TABLE 8.2. (continued) Fungal metabolite Roquefortine D
Roridin A Roseopurpurin
........
RI
UV Data
........
686
2031 67
218'1 100 251 ( 2 1
1013 866
283s143 288'1 45 294 I 44 302'1 45
245'1 100 2071 71
221'1 100
2371 47 249'1 56 2591 53 269s I 56 285'1 65 3251 4 434 I 29
Rubratoxin B
1076
205~195 2351 23
Rugulosin
1132
2291 49
251'1 83
273~164 3151 38 391 *I 81
Rugulosuvine
859
208184
216'190
24817 273.~114 281'1 14 288~112
Rugulovasine A
711
216' 100
24616
285'1 17
Scytalidine
1301
Scytalone
711
Secalonic acid D
Shikimic acid
1190
678
249'1 29
end 218' 93
231~166 24718 283'1 100 317.~139
216~155 233~140 2551 30 263'1 31 2871 19 337'1 71 377s I 17 210'1 700
349
TABLE 8.2. (continued) RI
UV Data
........
........
1349
2161 39
223'1 40
2421 23 255'1 27 2791 15 298'1 17 331 13 457'1 12
958
2121 50
229'1 100
2691 11 308'1 29 3561 2 480s I 22 505'1 27 5301 17 540'1 18
Soranjidiol
1119
2051 59
219'1 90
2391 31 269'1 100 292s I 50 3191 7 413'1 25
Sorbicillin
1172
203'1 100
233s I 30
2551 13 325*1 76
Spinulosin
700
208s I 80
2531 11
296'1 100
Steckiin
792
2051 97
21091 100
2251 56 233 I 64 2501 10 283'1 76 304s I 47
Sferigmafocystin
1104
205s I 80
2181 54
233s I 84 248'1 roo 2791 10 327'1 46
Stigmasterol
1916
end
Stipitatic acid
686
2081 19
259'1 100
2941 10 327'1 13 3391 12 355'1 14
Sulochrin
919
207'1 100
2431 16
283'1 39 319sl 17
Fungal metabolite Skyrin
Solaniol
350 TABLE 8.2.(continued) Fungal metabolite T-2 toxin
RI 1025
UV Data
........
........
2331 44 248.~150 279'1 100
end
Tenuazonic acid
810
2081 41
223'1 47
Terrein
682
216) 9
281.1 100
Terrestric acid
711
212s I 35
231'1 64
Terretonin
1043
2331 19
277'1 100
Territrern A
1135
2071 85
216'1 96
2751 7 339'1 39
Territrern B
1114
212.~161
2751 5
331'1 16
Territrern C
1033
210.~165
2831 4
341'1 14
2081 43
221.1 51
242s I 18 2651 8 288 I 22
Toluhydroquinone
Torreyol Trichoderrnin
689
1299
end
993
end
2441 39 269'1 100
Trichorzianines A'
1379
2161 43
221'144
24813 281'1 7 290sl6
Trichorzianines B Ila
1377
221.5 41
24813
271.~15 279'1 6 290.~15
Trichorzianines B lllc
1393
2161 42
221 '1 42
251 I 3 281 *I 6 290s I 5
Trichorzianines B IVb
1375
223~135 25113
Trichorzianines B Vb
1383
end
Trichorzianines B Vla
1468
251 13
283'1 3
271~15 281 '1 5 290~14
288.~12
35 I TABLE 8.2. (continued) Fungal metabolite
RI
........
........
2141 60
229'1 80
2481 48 261~152 281'1 100 313~136 3451 8 430'1 41 287'1 57 321sl 18
UV Data
Trichonianines B Vlb
1415
end
Trichonianines 6 Vll
1446
end
Trichothecin
1004
214'1 100
730
231'1 100
Trichothecolone 3,4,5-trihydroxy-7-methoxy2-methyl-anthraquinone
1155
Trypacidin
988
207'1 100
2491 11
Vermiculin
839
2051 86
221'1 100
1032
221 I25
261 *I 100
Verrucarin A Verrucarol
715
end
1072
205s I 91
2271 14
68 1
2031 13
237'1 100
Verrucosidin
1214
2251 53
239'1 61
269 I 25 294 * I 39
Verruculogen
1137
2161 75
225'1 76
2551 11 275'1 15 2871 13 294'1 13
Verruculotoxin
766
205s I 86
2401 1
257'1 1
epi- 10-verruculotoxin
715
end
Vertinolide
85 1
2071 20
235'1 50
2461 45 279'1 100
Violaceic acid
838
2181 72
229'1 79
2451 52 263') 66 283s I 52
Verrucofortine Verrucolon
246'1 26 275~16 283s)5
352
TABLE 8.2. (continued) RI
UV Data
........
........
Viomellein
1235
2031 42
223'1 60
2371 52 265'1 100 294s I 21 3231 8 371 *I 22 409s I 10
Vioxanthin
1369
2121 31
221 '1 32
233 I 27 269'1 100 306sl 13 3331 8 374'1 24
Viridamine
897
216)48
225'1 49
2631 23 373'1 100
Viridicatic acid
687
2051 17
233'1 87
244 I 53 265'1 100
Viridicatin
988
205'1 83
2101 81
223'1 100 239s 1 55 2651 14 288'1 21 2971 20 308s I 24 318'1 29 329s I 22
Viridicatumtoxin
1206
221 I47
239'1 75
251 I62 267sl 76 285'1 94 331sl 1 I 3581 5 434'1 28
Viriditoxin
1286
2161 51
221 *I 51
2371 36 263'1 91 3001 12 377'1 26
Woltmannin
946
2351 31
259 * 55
283 I 29 295'1 35
Xanthocillin X
1110
221 I 18
239' 21
2651 5 294s I 12 362'1 100
Fungal metabolite
353
TABLE 8.2. (continued) Fungal metabolite Xanthomegnin
Zearalenol
Zearalenone
RI
UV Data
........
..,.....
1110
203152
237'1 100
281129 292'1 31 337 16 403'1 19
973
1075
208~183 221 I77
237'1 94 2591 44 267'1 45 308.~115
2071 30
255129 273'1 44 2991 17 315'1 19
237'1 100
' Major of several peaks
slightly basic conditions may change some UV spectra drastically. These problems were misinterpreted by Paterson and Kemmelmeier (583), when they claimed that the small change in pH during the elution should affect the UV spectra of similar chromophores. Their comparisons of spectra taken in neutral or basic solvents taken by a stand-alone spectrophometer to spectra taken on-line by the diode array detector in acidic water/acetonitrile mixtures as occuring in gradients were simply not relevant as such spectra are known to be different in many cases (580-582). Furthermore neutral solvents or even worse very basic eluents could not be used generally because of problems with peak broadening of basic or acidic mycotoxins and corrosiveness of sodium hydroxide (pH approx. 12.4). We have made a library of the many compounds (authentic standards) listed in Table 2 and this work excellently for peak identification in the system we use. A large change system would mean that the standards should be run again and new UV spectra taken in that system. The differences between spectrataken in acidic methanoVwatergradients and acidic acetonitrile/ water gradients are very small however (Frisvad and Thrane, unpublished observations). Diode array
354
detection, especially in connection with retention index data and confirmed by TLC data, using authentic standards is a very efficient method for reliable mycotoxin analysis. After screening using such a system, more dedicated sensitive methods may developed for those mycotoxins that are considered a problem.
8.5.2.Applications of HPLC mass spectrometry HPLC mass spectrometry (HPLC-MS) have a great potential for very specific analysis but may be more difficult to apply to more broad screening-like analysis (584-
586). Five interfaces between the liquid chromatograph and the mass selective detector are of interest for mycotoxins at present: Direct inlet, thermospray/ plasmaspray, fast atom bombardment, particle beam and electrospray. Particle beam are giving the most informative spectra but are best for quite apolar molecules whereas electrospray is most suited for very polar molecules (584-586). Thermospray is in an intermediate position between those extremes, but can be used for "normal" flow rates at 1 mWmin. However the spectra from thermospray applications only contain little structural information and they are dependent of the eluent (586a). Tiebach et a/. (587-588)used a direct inlet technique to analyze aflatoxin, nivalenol and deoxynivalenol by micro HPLC-MS, using acetonitrile/water (1 :I ), flow rate 5 pl/min, as eluent and using chemical ionisation and both positive and negative ion detection. The separation of compounds was poor but the mass spectra quite informative and the authors claimed that the method could be used for foods. Thermospray have been the most widely used interface between the liquid chromatograph and the mass selective detector (e.g. 577,589-593).Usually acidified (phosphoric acid) water/ acetonitrile (577)or water/methanol isocratic runs (589), added ammoniumacetate to generate ions, on RP columns have been used. Carlson
et a/. (592)used a plasmaspray interface and a isocratic waterlacetonitrile (30:70)at 1 ml/min for prehelminthosporol. RajakylP et a/. (577) used an acidified water
acetonitrile gradient and thermospray and combinedthe HPLC-MS analysis with HPLCDAD.
One of the most interesting interfaces for LC-MS is fast atom bombardment which have been used directly on a mixture of fungal metabolites (594)or after HPLC separation. Kostiainen et a/. (595)used a water/methanol gradient and post-column addition of glycerol and obtained very informative mass spectra with glycerol adducts
355
for trichothecenes. A large number of new developments will probably be seen in the years to come in the LC-MS area of mycotoxin analysis, especially the electrospray interface (596). 8.6. CONCLUSIONS
HPLC is probably the most valuable method for mycotoxin analysis, however both selectivily, sensitivity and confirmation of identity should be considered. Based on the many applications listed above it may be concluded that the most general applications are those that involve gradient elutions using acidified water/acetonitrile or acidified waterlmethanol, especially if diode array detection and/or mass spectrometric detectors are available. In the latter case acids like acetic acid, trifluroacetic acid or other volatile acids should be used. It is also recommended to used a retention index series and to use authentic standards, and a great number of standards are now commercially available. Identifications should be confirmed by normal phase TLC if RP-HPLC is used or vice versa, rather than using a series of eluents. It is more difficult to propose a general method for individual mycotoxins in differentkinds of foods. Here one should consider all the available chemical information and design an optimally sensitive method accordingly. Knowledge of the associated mycoflora of the foods or feedstuffs may help in deciding which method should be used. For these more dedicated HPLC analysis both reversed phase, normal phase, cyano, amino etc. columns could be considered and several ion-pairing reagents, eluents etc. Also the actual extraction method and final detection method may be based on chemical and biological knowledge of the fungi, their toxins and the commodity they grow in.
356
REFERENCES 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27
C.K. Lim, HPLC of small molecules, a practical approach, IRL Press, Oxford, 1986. H.F. Linskens and J.F. Jackson (Eds.), High performance liquid chromatography in plant sciences, Modern methods of plant analysis, New series volume 5, SpringerVerlag, Berlin, 1987. D.B. Drucker, Microbiologicalapplications of high-performance liquid Chromatography, Cambridge University Press, Cambridge, 1987. I.N. Papadoyannis, HPLC in clinical chemistry, Marcel Dekker, New York, 1990. W.S. Hancock and J.T. Sparrow, HPLC analysis of biological compounds, A laboratory guide, Marcel Dekker, New York, 1984. P.R. Brown and R.A. Hartwick (Eds.), High performance liquid chromatography, Chemical analysis vol. 98, John Wiley and Sons, New York, 1989. A.M. Krstulovic and P.R. Brown, Reversed-phase high performance liquid chromatography. Theory, practice and biomedical applications, Wiley-lnterscience, New York, 1982 S . Ahuja, Selectivity and detectability optimizations in HPLC, Chemical analysis vol. 104, John Wiley and Sons, New York, 1989. R. Macrae (Ed.), HPLC in food analysis, Academic Press, London, 1988. A. Fox, S.L. Morgan, L. Larsson and G. Odham (Eds.), Analytical microbiology methods. Chromatography and mass spectrometry, Plenum Press, New York, 1990 H. Jork, W. Funk, W. Fischer and H. Wirnmer, Thin-layer chromatography. Reagents and detection methods, vol. 1-3, VCH, Weinheim, 1990-1991. N. Grinberg (Ed.), Modern thin-layer Chromatography. Chromatographic Science Series vol. 52., Marcel Dekker, New York, 1990. J. Sherma and B. Fried (Eds.), Handbook of thin-layer Chromatography. Chromatographic Science Series vol. 55, Marcel Dekker, New York, 1991. J.W. Bennett, Mycopathologia, 100 (1987) 3. D.T. Wicklow, in: K.A. Pirozynski and D.L. Hawksworth (Eds.): Coevolution of fungi with plants and animals, Academic Press, New York, London, 1988, pp. 173. J.C. Frisvad and 0. Filtenborg, Mycologia 81 (1989) 837. W.C.A. Gelderblom, K. Jaskiewicz, W.F.O. Marasas, P.G. Thiel, R.M. Horak, R. Vleggaar and N.P.J.Kriek, Appl. Environ. Microbiol., 54 (1988) 1806. M. Enomoto and I. Ueno, in: I.F.H. Purchase (Ed.), Mycotoxins, Elsevier, Amsterdam, 1974, pp. 303. J.C. Frisvad, Arch. Environ. Contam. Toxicol. 18 (1989) 452. J.C. Frisvad, 0. Filtenborg, R.A. Samson, A.C. Stolk, Antonie van Leeuwenhoek 57 (1990) 179. J.C. Frisvad and 0. Filtenborg, in: R.A. Samson and J.I. Pitt (Eds.), Modern concepts in Penicillium and Aspergillus classification, Plenum Press, New York, 1990, pp. 159. J.C. Frisvad and R.A. Samson, in: R.A. Samson and J.I. Pitt (Eds.), Modern concepts in Penicillium and Aspergillus classification, Plenum Press, New York, 1990, pp. 201. J.C. Frisvad, R.A. Samson and A.C. Stolk, in: R.A. Samson and J.I. Pitt (Eds.), Modern concepts in Penici//ium and Aspergillus classification, Plenum Press, New York, 1990, pp. 445. R.A. Samson, P.V. Nielsen and J.C. Frisvad, in: R.A. Samson and J.I. Pitt (Eds.), Modem concepts in Penicillium and Aspergillus systematics, Plenum Press, New York, 1990, pp. 455. R.A. Samson, A.C Stolk, and J.C. Frisvad, Stud. Mycol. (Baarn) 31 (1989) 133. U. Thrane, in: J. Chelkowski (Ed.), Fusarium Mycotoxins, taxonomy and pathogenicity, Elsevier, Amsterdam, 1989, pp. 199. K. Liljegren, A. Svendsen and J.C. Frisvad, Proc. Jpn. Assoc. Mycotoxicol. Suppl. 1 (1988) 35.
351
28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50
51 52 53 54 55 56 57 58 59 60 61 62
J.W. Dorner, R.J. Cole and U.L. Diener, Mycopathologia 87 (1984) 13. J.C. Frisvad and 0. Filtenborg, Proc. Jpn. Assoc. Mycotoxicol. Suppl. 1 (1988) 163. W.F.O. Marasas, P.E. Nelson and T.A. Toussoun, Toxigenic Fusarium species. Identity and mycotoxicology, Pennsylvania State University Press, University Park, 1984. J.C. Frisvad and R.A. Samson, In: D.K. Arora, K.G. Mukerji and E.H. Marth (Eds.), Handbook of Applied Mycology, Vol. 3: Foods and feeds. Marcel Dekker, New York, 1991, pp. 31. W.B. Turner, Fungal metabolites, Academic Press, London, 1971. W.B. Turner and D.C. Aldridge, Fungal metabolites II. Academic Press, London, 1983. R.J. Cole and R.H. Cox, Handbook of toxic fungal metabolites, Academic Press, New York, 1981. P.M. Scott,in: J.F. Lawrence (Ed.), Trace analysis, Academic Press, New York, 1981, pp. 193. R.D. Coker, in: J.Gilbert (Ed.), Analysis of food contaminants, Applied Science Publ., London, 1984, pp. 207. M.J. Shepherd, in: R.J. Cole, Modern methods in the analysis and structural elucidation of mycotoxins, Academic Press, New York, 1986, pp. 293. D.C. Hunt, in: R. Macrae (Ed.), HPLC in food analysis, Academic Press, London, 1982, pp. 271. V. Betina, J. Chromatogr. 477 (1989) 187. R.D. Coker and B.D. Jones, in: R. Macrae (Ed.), HPLC in food analysis, Academic Press, London, 1988, pp. 335. P.J. Martin, H.M. Stahr, W. Hyde and M. Domoto, J. Liq. Chromatogr. 9 (1986) 1591. V. Betina (Ed.), Mycotoxins. Production, isolation, separation and purification, Elsevier, Amsterdam, 1984. V. Betina, Mycotoxins, Elsevier, Amsterdam, 1989. M. RarnachandraPai, N. Jayanthi Bai and T.A. Venkitasubramaniana, Anal. Biochem. 63 (1975) 274. J.C. Frisvad, 0. Filtenborg and U. Thrane, Arch. Environ. Contam. Toxicol. 18 (1989) 331. J.C. Frisvad, J. Chromatogr. 392 (1987) 333. J.C. Frisvad and U. Thrane, J. Chromatogr. 404 (1987) 195. J.C. Frisvad, Bot. J. Linn. SOC.99 (1989) 81. J.C. Frisvad and R.A. Samson, in: J. Chelkowski (Ed.), Cereal grain. Mycotoxins, fungi and quality in drying and storage, Elsevier, Amsterdam, 1991, pp. 441. B. Bjerg, 0. Olsen, K.W. Rasmussen and H. Serensen, J. Lq. Chromatogr. 7 (1984) 691. H. Cohen, G.A. Neish, and H.L. Trenholm, Microbiol. Aliment. Nutr. 3 (1985) 133. G.G. Habermehl, L. Busam, P. Heydel, D. Mebs, C.H. Tokarnia, J. Dobereiner and M. Spraul, Toxicon 23 (1985) 731. S.E. Megalla, Mycopathologia 84 (1983) 45. J.D. McKinney, J. Assoc. Oil Chem. SOC. (1981) 935A. M.S. Madhyastha and R.V. Bhat, Curr. Sci. 52 (1983) 222. S.Williams (Ed.), Official methods of analysis, 14th edition, Association of Official Analytical Chemists, Washington, D.C. G.L. Stanley, V.P. DiProssimo and A.C. Koontz, J. Assoc. Off.Anal. Chern. 62 (1979) 136. R. Knutti, K.Sutter and C. Schlatter, Swiss Food 1 (1979) 17. J.E. Thean, D.R. Lorenz, D.M. Wilson, K. Rodgers and R.C. Gueldner, J. Assoc. Off. Anal. Chem. 63 (1980) 631. J.-M. Fremy and B. Boursier, J. Chromatogr. 219 (1981) 156. J.D. McKinney, J. Amer. Oil Chem. SOC.58 (1981) 935A. J.E. Hutchins and W.M. Hagler, J. Assoc. Off. Anal. Chem. 66 (1981) 1458.
358 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 74 75 76 77 78 79
80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95
A. Riberzani, S.Castelli, A. del Vo and C. Pedretti, Ind. Aliment. 22 (1983) 342. W.J. Hurst, K.P. Snyder and R.A. Martin, Peanut Sci. 11 (1984) 21. D. Tosch, A.E. Waltking, and J.F. Schlesier, J. Assoc. Off. Anal. Chem. 67 (1984) 337. J.P. Bijl and C. van Peteghem, Anal. Chim. Acta 170 (1985) 149. A.E. Yousef and E.H. Marth. J. Assoc. Off. Anal. Chem. 68 (1985) 462. J.P. Bijl, C.H. van Peteghem.and D.A. Dekeyser, J. Asoc. Off: Anal: Chem. 70 (1987) 472. O.J. Francis, G.E. Ware, A S . Carman and S.S. Kuan, J. Assoc. Off. Anal. Chem. 68 (1985) 643. O.L. Shotwell, in: R.J. Cole (Ed.), Modern methods in the analysis and structural elucidation of mycotoxins, Academic Press, New York, 1986, pp. 51. H. Cohen and M. Lapointe, J.Assoc. Off.Anal. Chem. 67 (1984) 1105. M.W. Trucksess, W.C. Brumley and S. Nesheim, J. Assoc. Off. Anal. Chern. 67 (1984) 973. R.D. Stubblefield and H.P. van Egmond, in: H.P. van Egmund (Ed.), Mycotoxins in dairy products, Elsevier, London, 1989, pp. 57. W. Steiner and R. Battaglia, Mitt. Geb. Lebensm. Hyg. 74 (1983) 140. N. Takeda, J. Chromatogr. 288 (1984) 484. G.-S. Qian and G.C. Yang, J. Agric. Food Chem. 32 (1984) 1071. H. Hisada, H. Terada, K. Yamamoto, H. Tsubouchi and Y. Sakabe, J. Assoc. Off. Anal. Chem. 67 (1984) 601. J. Ferguson-Foss and J.D. Warren, J. Assoc. Off. Anal. Chem. 67 (1984) 1111. J.-M. Fremy and F.S. Chu, J. Assoc. Off. Anal. Chem. 67 (1984) 1098. E. MBrtlbauer and G. Terplan, Arch. Lebensmittelhyg. 36 (1985) 53. A. Carisano and G. della Torre, J. Chromatogr. 355 (1986) 340. Anonymous, J. Assoc. Off. Anal. Chem. 69 (1986) 361. D.L. Orti, R.H. Hill, J.A. Liddle, L.L. Needham and L. Vickers, J. Anal. Toxicol. 10 (1986) 41. C.P. Wild, F.A. Pionneau, R. Montesano, C.F. Mutiro and C.J. Chetsanga, Intl. J. Cancer 40 (1987) 328. H. Karnimura, M. Nishijima, K. Yasuda, H. Ushiyama, H. Tabata, S. Matsumoto and T. Nishima, J. Assoc. off. Anal. Chem. 68 (1985) 458. M.T. Hetmanski and K.A. Scudamore, Food Add. Contam. 6 (1988) 35. H.P. van Egmond, W.E. Paulsch, E. Deijill and P.L. Schuller, J. Assoc. Off. Anal. Chem. 63 (1980) 110. J.D. Rosen, R.T. Rosen and T.G. Hartman, J. Chromatogr. 335 (1986) 241. S.P. Swanson, A.M. Dahlem, H.D. Rood, Jr., L. Cote, W.B. Buck and T. Yoshizawa, J. Assoc. Off. Anal. Chem. 69 (1986) 41. G.M. Shannon, R.E. Peterson and O.L. Shotwell, J. Assoc. Off. Anal. Chem. 68 (1985) 1126. S.C. Lee and F.S. Chu, J. Assoc. Off. Anal. Chem. 64 (1981) 156. Y.C. Xu, G.S. Chang and F.S. Chu, J. Assoc. Off. Anal. Chem. 69 (1986) 967. C.J. Mirocha, R.J. Pawlosky and H.K. Abbas, Arch. Environ. Contam. Toxicol. 18 (1989) 349. H. Cohen and M. LaPointe, J. Assoc. Off. Anal. Chem. 69 (1986) 957. R.M. Eppley, M.W. Trucksess, S. Nesheim, C.W. Thorpe and A.E. Pohland, J. Asooc. Off. Anal. Chem. 69 (1986) 37. T. Tanaka, A. Hadegawa, Y. Matsuki, K. lshii and Y. Ueno, Food Add. Contam. 2 (1985) 125. Y. Ramakrishna, R.B. Sashidhar and R.V. Bhat, Bull. Environ. Contam. Toxicol. 42 (1989) 167. P.M. Scott and G.A. Lawrence, J. Assoc. Off. Anal. Chem. 71 (1988) 1176. T. Tanaka, A. Hadegawa, Y. Matsuki, US. Lee and Y. Ueno, J. Chromatogr. 328
359
96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138
(1985) 271. S.P. Swanson, R.A. Corley, D.G. White and W.B. Buck, J. Assoc. Off. Anal. Chern. 67 (1984) 580. H. Terada, H. Tsubouchi, K. Yarnarnoto, K. Hisada and Y. Sakabe, J. Assoc. Off. Anal. Chem. 69 (1986) 960. R.C. Lee and F.S. Chu, Food Agric. Imrnunol. 1 (1989) 127. M.J. Shepherd and J. Gilbert, J. Chromatogr. 358 (1986) 415. T.R. Rorner, J. Assoc. Off. Anal. Chern. 69 (1986) 699. H.L. Chang, J.W. de Vries, P.A. Larson and H.H. Patel, J. Assoc. Off. Anal. Chern. 67 (1984) 52. A. Sano, S.Matsutani, M. Suzuki and S.Takitani, J. Chrornatogr. 410 (1987) 427. L. Soltes, Chern. Listy 84 (1990) 1215. D.N. Mortirner, J. Gilbert and M.J. Shepherd, J. Chrornatogr. 407 (1987) 393. M.J. Shepherd, M. Holrnes and J. Gilbert, J. Chrornatogr. 354 (1986) 305. J.M. Frerny and F.S. Chu, in: H.P. van Egrnond (Ed.), Mycotoxins in dairy products, Elsevier, London, pp. 97 J.D. Groopman and T.W. Kernsler, Pharrnacol. Ther. 34 (1987) 321. J.D. Groopman and K.F. Donahue, J. Assoc. Off. Anal. Chern. 71 (1988) 861. A.A.G. Candlish, W.H. Stimson and J.E. Smith, J. Assoc. Off. Anal. Chem. 71 (1988) 961. P.M. Scott, J. Assoc. Off. Anal. Chern. 68 (1985) 242. R.W. Beaver, Arch. Environ. Contarn. Toxicol. 18 (1989) 315. M.J. Shepherd, D.N. Mortirner and J. Gilbert, J. Assoc. Publ. Analysts 25 (1987) 129. P.M. Scott, Food Addit. Contarn. 6 (1989) 283. R.W. Beaver, J. Assoc. Off. Anal. Chern. 73 (1990) 69. R.W. Beaver, J. High Res. Chrornatogr. 13 (1990) 833. L.G.M.T. Tuinstra, P.G.M. Kienhuis, W.A. Traag, M.M.L. Aerts and W.M.J. Beek, J. High Res. Chrornatogr. 12 (1989) 709. L.G.M.T. Tuinstra, P.G.M. Kienhuis and P. Dols, J. Assoc. Off. Anal. Chern. 73 (1990) 969. M. Valcarcel and M.D. Luque de Castro, J. Chrornatogr. 65 (1982) 3. F. LAzaro, M.D. Luque de Castro and M. ValcArcel, J. Chrornatogr. 448 (1988) 173. E.C. Shepherd, T.D. Phillips, N.D. Heidelbaugh, J. Assoc. Off. Anal. Chem. 65 (1982) 665. J.W. DeVries and H.L. Chang, J. Assoc. Offic Anal. Chern. 65 (1 982) 206. R.C. Garner, J. Chrornatogr. 103 (1975) 186. L.M. Seitz, J. Chrornatogr. 104 (1975) 81. T. Panalaks and P.M. Scott, J. Assoc. Off. Anal. Chern. 60 (1977) 583. W.A. Pons and A.O. Franz, J. Assoc. Off. Anal. Chern. 61 (1978) 793. H. Schweighardt, J. Bohm and J. Leibetseder, Ernahrung 2 (1978) 3. M. Blanc, Ann. Fals. Exp. Chirn. 72 (1979) 427. T. Goto, M. Manabe and S. Matsuura, Agric. Biol. Chern. 43 (1979) 2591. P.J. Colley and G.E. Neal, Anal. Biochern. 93 (1979) 409. W.A. Pons, J. Assoc. Off. Anal. Chern. 62 (1979) 586. W.A. Pons, L.S. Lee and L. Stoloff, J. Assoc. Off. Anal. Chem. 63 (1980) 899. R.-D. Wei, S.-C. Chang and S.-S. Lee, J. Assoc. Off. Anal. Chern. 63 (1980) 1269. M.J. Awe and J.L. Schranz, J. Assoc. Off. Anal. Chem. 64 (1981) 1377. O.J. Francis, L.J. Lipinski, J.A. Gaul and A.D. Campbell, J. Assoc. Off. Anal. Chern. 65 (1982) 672. D.M. Miller, D.M. Wilson, R.D. Wyatt, J.R. McKinney, W.A. Crowell and B.P. Stuart, J. Assoc. Off. Anal. Chem. 65 (1982) 1. J. Bohrn, C. Noonpugdee, J. Leibetseder and M. Schuh, Ernahrung 8 (1984) 675. T. Anukarahanonta and C. Chudhabuddhi, J. Chromatogr. 275 (1983 387. S.Tsuboi, T. Nakagawa, M. Tornita, T. Seo, H. Ono, K. Kawarnura and N. Iwamura,
360
139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181
Cancer Res. 44 (1984) 1231. G.-S. Qian, P. Yasei and G.C. Yang, Anal. Chern. 56 (1984) 2079. T. Goto, M. Manabe and S.Matsuura, Agric. Biol. Chern. 46 (1982) 801. V.A. Tutelyan, K.I. Eller and V.S. Sobolev, Food Addit Contarn. 6 (1989) 459. V.A. Tutelyan, V.S. Sobolev, N.V. Rybakovaand K.I. Eller, J. Toxicol. - Toxin Reviews 8 (1989) 375. S.H. Hansen, P. Helboe and M. Thornsen, J. Pharrn. Biorned. Anal. 2 (1984) 165. G.J. Diebold and R.N. Zare, Nature 196 (1977) 1439. G.J. Diebold, N. Kamy, R.N. Zare and L.M. Seitz, J. Assoc. Off. Anal. Chem. 62 (1979) 564. R.D. Stubblefield and O.L. Shotwell, J. Assoc. Off. Anal. Chern. 60 (1977) 784. D.M. Takahashi, J. Assoc. Off. Anal. Chern. 60 (1977) 799. D.M. Takahashi, J. Chrornatogr. 131 (1977) 147. 8. Zirnrnerli, J. Chrornatogr. 131 (1977) 458. B. Zirnrnerli, Mitt. Gebiete Lebensrn. Hyg. 68 (1977) 36. R.M. Beebe, J. Assoc. Off. Anal. Chern. 61 (1978) 1347. W.J. Hurst and P.B. Toorney, J. Chrornatogr. Sci. 16 (1978) 372. E. Koch, A. Nothhelfer and H. Treiber, Lebensrnitttelchern. Gerichtl. Chern. 32 (1978) 78. R. Knutti, C. Balsiger and K. Sutter, Chrornatographia12 (1979) 349. W. Winterlin, G. Hall and D.P.H. Hsieh, Anal. Chern. 51 (1979) 1873. R.M. Beebe and D.M. Takahashi, J. Agric. Food Chern. 28 (1980) 481. P.J. Henog, J.R.L. Smith and R.C. Garner, Carcinogenesis 1 (1980) 787. N.D. Davis and U.L. Diener, J. Assoc. Off. Anal. Chern. 63 (1980) 107. H. Cohen and M. Lapointe, J. Assoc. Off. Anal. Chern. 64 (1981) 1372. J.-M. Frerny and B. Boursier, J. Chromatogr. 219 (1981) 156. M.V. Howell and P.W. Taylor, J. Assoc. Off. Anal. Chern. 64 (1981) 1356. 8. Haghighi, C. Thorpe, A.E. Pohland and R. Barnett, J. Chrornatogr. 206 (1981) 101. R.L. Price, K.V. Jorgensen and M. Billotte, J. Assoc. Off. Anal. Chern. 64 (1981) 1383. J.F. Gregory 111 and D. Manley, J. Assoc. Off. Anal. Chern. 64 (1981) 144. P.R. Donahue, J.M. Essigrnann and G.N. Wogan, Banbury Report 13 (1982) 221. J.F. Gregory 111 and D.B. Manley, J. Assoc. Off. Anal. Chern. 65 (1982) 869. W.J. Hurst, L.M. Lenowich and R.A. Martin, Jr., J. assoc. Off. Anal. Chern. 65 (1982) 888. T. Seo, H. Ono, E. Cho, T. Nakagawa, M. Tornita, S. Tsuboi, N. Iwamura, M. Rornero and P. Abadi, ICMR Annals 2 (1982) 121. L.G.M.T. Tuinstra and W. Hassnot, Fresenius Z. Anal. Chern. 312 (1982) 622. P. Charnbon, S.D. Dano, R. Charnbon and A. Geahchan, J. Chrornatogr.259 (1983) 372. H.L. Chang and J.W. DeVries, J. Assoc. Off. Anal. Chern. 66 (1983) 913. G. Cirilli, Microbiol. Alirn. Nutr. 1 (1983) 199. S.M. Larnplugh, J. Chrornatogr. 273 (1983) 442. L.G.M.T. Tuinstra and W. Haasnoot, J. Chrornatogr. 282 (1983) 457. K. Kawarnura, S.Tsuboi, N. Iwarnura, Y. Irnanaka, T. Wada, N. Kohno, M.L. Cruz, ICMR Annals 4 (1984) 153. M.J. Shepherd and J. Gilbert, Food Addit. Contam. 1 (1984) 325. J.E. Hutchins, Y.J. Lee, K. Tyczkowska and W.M. Hagler, Jr., Arch. Environ. Contarn. Toxicol. 18 (1989) 319. E.J. Tarter, J.P. Hanchay and P.M. Scott,J. Assoc. Off. Anal. Chern. 67 (1984) 597. X.B. Xu and Z.L. Jin, J. Chrornatogr. 317 (1984) 545. J. Gilbert and M.J. Shepherd, Food Addit. Contarn. 2 (1985) 171. W.T. Kok, T.C.H. Van Neer, W.A. Traag and L.G.M.T. Tuinstra J. Chrornatogr. 367 (1986) 231.
36 1 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 21 1 212 213 214 215 216 217 218
P.G. Thiel, S. Stockenstrom and P.S. Gathercole, J. Liq. Chromatogr. 9 (1986) 103. W.E. Paulsch, H.P. Van Egmund and E.A. Sizoo, J. Assoc. Off. Anal. Chem. 71 (1988) 957. H.P. Van Egmund and P.J. Wagstaffe, Food Addit. Contam. 6 (1989) 397. R.W. Beaver, D.M. Wilson and M.W. Trucksess, J. Assoc. Off. Anal. Chem. 73 (1990) 579. L.S. Lee, J.H. Wall, P.J. Cotty and P. Bayman, J. Assoc. Off. Anal. Chem. 73 (1990) 581. D.L. Park, S. Nesheim, M.W. Trucksess, M.E. Stack and R.F. Newell, J. Assoc. Off. Anal. Chem. 73 (1990) 260. G. Sabbioni, S. Ambs, G.N. Wogan and J.D. Groopman, Carcinogenesis 11 (1990) 2063. H.P. Van Egmund and P.J. Wagstaffe, Food Addit. Contam. 7 (1990) 239. T.A. Bean, D.M. Yourtee, B. Akande and J. Ogunlewe, J. Toxicol. - Toxin Reviews 8 (1989) 43. D.M. Wilson, Arch. Contam. Environ. Toxicol. 18 (1989) 308. G.D. Wachob, LC Mag. (LC-GC) 4 (1989) 42. D.P.H. Hsieh, D.L Fiizell, J.L. Miller and J.N. Seiber, J. Chromatogr. 117 (1975) 474. E. Johnson, A. Abu-Shumays and S.R. Abbott, J. Chromatogr. 134 (1977) 107. M. Manaba, T. Goto and S.Matsuura, Agric.Biol Chem. 42 (1978) 2003. J.A. Robertson, W.A. Pons and L.A. Goldbtatt, J. Agric. Food Chem. 15 (1967) 798. J. Chelkowski, Photochem. Photobiol. 20 (1974) 279. G. Buchi, D.M. Foulkes, M. Kurono and G.F. Mitchell, J. Am. Chem. SOC.88 (1966) 4534. K.K. Maggon, S. Gopal, L. Viswanathan, T.A. Venkitasubramanian and S. Rathi, Ind. J. Biochem. Biophys. 9 (1972) 195. H. Jansen, R. Jansen, U.A.T. Brinkman and R.W. Frei, Chromatographia 24 (1987) 555. J. W. Dorner and R.J. Cole, J. Assoc. Off. Anal. Chem. 71 (1988) 43. C.W. Thorpe, G.M. Ware and A.E. Pohland, in: W. Pfannhauser and P.B. CzedicEysenberg (Eds.), Proceedings of the Vth international IUPAC Symposium on Mycotoxins and Phycotoxins, Technical University, Vienna, 1982, pp. 52. J.W. Dorner and R.J. Cole, J. Assoc. Off. Anal. Chem. 72 (1989) 962. M.W. Trucksess, M.E. Stack, S. Nesheim, S.W. Page, R.H. Albert, T.J. Hansen and K.F. Donahue, J. Assoc. Off. Anal. Chem. 74 (1991) 81. O.J. Francis, G.P. Kirschenheuter,G.M. Ware, A.S. Carmen and S.S. Kuan, J. Assoc. Off. Anal. Chem. 71 (1988) 725. C.C. Harris, G. La Veck, J. Groopman, V.L. Wilson and D. Mann, Cancer Res. 46 (1986) 3249. B.T. Duhart, S. Shaw, M. Wooley, T. Allen and C. Grimes, Anal. Chirn. Acta 208 (1988) 343. R.J. Cole, J.W. Dorner, J.W. Kirksey and F.E. Dowell, Peanut Sci. 15 (1988) 61. M.P.K. Dell, S.J. Haswell, O.G. Roch, R.D. Cooker, V.F.P. Medlock and K. Tornlins, Analyst 115 (1990) 1435. M. Azer and C. Cooper, J. Food Prot. 54 (1991) 291. A.L. Patey, M. Shaman and J. Gilbert, J. Assoc. Off. Anal. Chem. 74 (1991) 76. J.A. Lansden, J. Agric Food Chem. 25 (1977) 969. W.A. Pons and A.O. Franz, Jr., J. Assoc. Off. Anal. Chern. 60 (1977) 89. L.S. Lee, L.V. Lee, Jr. and T.E. Russell, J. Am. Oil Chem. SOC.63 (1986) 530. V.K. Mehan, T.N. Bhavanishankarand J.S. Bedi, J. Food Sci. Technol. 22 (1985) 123. R. Onori, Food Addit. Contam. 4 (1987) 407. W.A. Traag, J.M.P. van Trijp, L.G.M.T. Tuinstra and W.T. Kok, J. Chromatogr. 396 (1987) 530. M.Y. Siraj, A.W. Hayes, P.D. Unger, G.R. Hogan, N.J. Ryan and B.B. Wray, Toxicol
362
21 9 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257
Appl. Pharmacol. 58 (1981) 422. J.M. Fremy and T. Cariou, Analusis 12 (1984) 103. P. Lafont and M.G. Siriwardana, J. Chromatogr. 219 (1981) 162. P. Lafont, M.G. Siriwardana, J. Sarfati, J.P. Debeaupuis and J. Lafont, Microbiol. Nutr. Alim. 4 (1986) 141. E. Isohata, Y. Takeda and M. Ushiyama, Shokuhin Eiseigaku Zasshi 17 (1976) 392.(In Japanese) H.S. Schroggins, EDRO SARAP Res. Tech. Rep. 1 (1976) 1. (CA 088(05)035956). M.E. Stack, S. Nesheim, N.L. Brown and A.E. Pohland, J. Assoc. Off. Anal. Chem. 59 (1976) 966. R. Schmidt, J. Mondani, E. Zeigenhagen and K. Dose, J. Chromatogr. 207 (1981) 435. O.J. Francis, L.L. Lipinski, J.A. Gaul and A.D. Campbell, J. Assoc. Off. Anal. Chem. 65 (1982) 672. P. Nowotny, W. Baltes, W. Kroenert and R. Weber, Lebensm. chem. Gerichtl. Chem. 37 (1983) 71. D. Abramson, R.N. Sinha and J.T. Mills, Cereal Chem. 60 (1983) 350. J.T. Mills and D. Abramson, Can. J. Plant Pathol. 8 (1986) 151. P. Lepom, J. Chromatogr. 354 (1986) 518. A. Hasegawa, T. Tanaka, S. Yamamoto, N. Toyazaki, Y. Matsuda and S.Udagawa, Proc. Jpn. Assoc. Mycotoxicol. 26 (1987) 37. D. Abramson and T. Thorsteinson, J. Assoc. Off.Anal. Chem. 72 (1989) 342. D.L. Orti, J. Grainger, D.L. Ashley and R.H. Hill, J. Chromatogr. 462 (1989) 269. J. Leitao, G. de Saint Blanquat and J.R. Bailly, J. Liq. Chromatogr. 11 (1988) 2285. F.L. Neely and C.S. Emerson, J. Chromatogr. 523 (1990) 305. D.W. Hill, T.R. Kelley, K.J. Langner and K.W. Miller, Anal. Chem. 56 (1984) 2576. G.W. Engstrom, J.L. Richard and S.J. Cysewski, J. Agric. Food Chem. 25 (1977) 833. W.J. Hurst, K.P. Snyder and R.A. Martin, J. Chromatogr. 392 (1987) 389. D.G.I. Kingston, P.N. Chen and J.R. Vercellotti, J. Chromatogr. 118 (1976) 414. K. Ito, A. Yamane, T. Hamasaki and Y. Hatsuda, Agr. Biol. Chem. 40 (1976) 2099. R.K. Berry, M.F. Dutton and M.S. Jeenah, J. Chromatogr. 283 (1984) 421. S.P. McCormick, E. Bowers and D. Bhatnagar, J. Chromatogr. 441 (1988) 400. J.A. Billington and D.P.H. Hsieh, J. Agric. Food Chem. 37 (1989) 676. R. Kostiainen, A. Rizzo and A. Hesso, Arch. Environ. Contam. Toxicol. 18 (1989) 356. R.D. Planner, In: R.J. Cole (Ed.), Modern methods for the analysis and structural elucidation of mycotoxins, Academic Press, New York, 1986, 240. P.M. Scott, J. Assoc. Off. Anal. Chem. 65 (1982) 876. Y.Ueno, In: Y. Ueno (Ed.), Trichothecenes. Chemical, biological and toxicological aspects, Elsevier, Amsterdam, 1983, pp. 113. J. Gilbert, In: M.O. Moss and J.E. Smith (Eds.), The applied mycology of Fusafium, Cambridge University Press, Cambridge, 1984, pp. 175. J.F. Grove, Nat. Prod. Rep. 5 (1988) 187. S.N. Lanin, V.V. Petrenko, A.N. Leonov, G.P. Kononenko and N.A. Soboleva, Khim. Prir. Soedin. 1989 (1989), 861. (CA 112(19)173676) S.N. Lanin, Y.S. Nikitin and V.V. Petrenko, Zh. Anal. Khim. 44 (1989) 2235. (CA 112(13)113675) S.N. Lanin and Y.S. Nikitin, J. Chromatogr. 558 (1991) 81. C. Noonpugdee, J. Boehm, J. Leibetseder and M. Schuh, Ernaehrung (Vienna)9 (1985) 622. V.L. Sylvia, T.D. Phillips, D.Timothy, B.A. Clement, J.L. Green, L.F. Kubenaand N.D. Heidelbaugh, J. Chromatogr. 362 (1986) 79. W.L. Childress, I S . Krull and C.M. Selavka, J. Chromtogr. Sci. 28 (1990) 76. A. Sano, S. Matsutani, M. Suzuki and S. Takitani, J. Chromatogr. 410 (1987) 427. R. Maycock and D. Utley, J. Chromatogr. 347 (1985) 429.
363 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299
B. Yagen, A. Sintov and M. Bialer, J. Chrornatogr. 356 (1986) 195. M.A.J. Bayliss, R.B. Homer and M.J. Shepherd, J. Chrornatogr. 445 (1988) 393. F.S. Chu and R.C. Lee, Food Agric. Irnrnunol. 1 (1989) 127. R.A. Williams, R. Macrae and M.J. Shepherd, J. Chromatogr. 477 (1989) 315. Q. Ma and G. Gu, Zhongcaoyao 20 (1989) 349. (CA 111(23)210163). R. Schmidt, E. Ziegenhagen and K. Dose, J. Chrornatogr. 212 (1981) 370. H.L. Chang, J.W. DeVries, P.A. Larson and H.H. Patel, J. Assoc. Off.Anal. Chern. 67 (1984) 52. R. Schmidt, Mycotox. Res. 2 (1986) 39. H. Schweighardt, J. Boehrn, A.M. Abdelharnid, J. Leibetseder, M. Schuh and E. Glawishnig, Chrornatographia 13 (1980) 447. D.R. Lauren and R. Greenhalgh, J. Assoc. Off. Anal. Chern. 70 (1987) 479. H. Cohen and M. Lapointe, Microbiol. Alirn. Nutr. 2 (1984) 325. A. Visconti and A. Bottalico, Chrornatographia 17 (1983) 97. R. Schmidt and K. Dose, J. Anal. Toxicol. 8 (1984) 43. R. Schmidt, K. Lenz, P. Flesch and K. Dose, Fres. Z. Anal. Chern. 317 (1984) 665. C. Noonpugdee, Chromatographia 23 (1987) 47. G.A. Bennett, R.E. Peterson, R.D. Plattner and O.L. Shotwell, J. Am. Oil Chern. SOC. 58 (1981) 1002A. K.C. Ehrlich, L.S. Lee and A. Ciegler, J. Liq. Chromatogr. 6 (1983) 833. D.R. Lauren and M.P. Agnew, J. Agric. Food Chern. 39 (1991) 502. H.D. Rood, W.B. Buck and S.P. Swanson, J. Agric. Food Chern. 36 (1988) 74-79. H.D. Rood, W.B. Buck and S.P. Swanson, J. Assoc. Off. Anal. Chern. 71 (1988) 493498. K. Kroll, Nahrung 32 (1988) 75. J. Boehrn, C. Noonpugdee, J. Leibetseder and M. Schuh, Ernaerung (Vienna) 8 (1984) 601. L.-M. Cote, J. Nicoletti, S.P. Swanson and W.B. Buck, J. Agric. Food Chern. 34 (1986) 458. R.A. Corley, S.P. Swanson, G.J. Gullo, L. Johnson, V.R. Beasley and W.B. Buck, J. Agric. Food Chern. 34 (1986) 868. B.B. Jarvis, S.N. Gornezoglu, M.N. Rao, and N.B. pens, J. Org. Chem. 52 (1987) 45. B.B. Jarvis, in: R.P. Sharma and D.K. Salunkhe, Mycotoxins and phytoalexins, CRC Press, Boca Raton, 1991, p. 361. A. Bata, J. Fekete and B. Harrach, Syrnp. Biol. Hung. 31 (1986) 325. B.B. Jarvis, J.O. Midiwo, D. Tuthill and G.A. Bean, Science 214 (1981) 460. W.G. Sorenson, D.G. Frazer, B.B. Jarvis, J. Simpson and V.A. Robinson, Appl. Environ. Microbiol. 53 (1987) 1370. M.E. Stack and R.M. Eppley, J. Assoc. Off. Anal. Chern. 63 (1980) 1278. B. Andersen, Antonie van Leeuwenhoek 60 (1991) 115. M. Torres, V. Sanchis, M. Riba and R. Canela, Appl. Environ. Microbiol. 51 (1986) 209. M. Jirnenez, V. Sanchis, R. Meteo and E. Hernandez, Mycotoxin Res. 2 (1988) 59. J.W. Priest and R.J. tight, J. Chrornatogr. 513 (1990) 237. G.M. Ware, C.W. Thorpe and A.E. Pohland, J. Assoc. Off. Anal. Chern. 57 (1974) 1111. G.M. Ware, J. Assoc. Off. Anal. Chern. 58 (1975) 754. H. Tanner and C. Zanier, Schw. Z. Obst- Weinbau 112 (1976) 656. H. Woidich, W. Pfannhauser and G. Blaicher, Lebenrnittelchern. Gerichtl. Chern. 32 (1978) 61. U. Leuenberger, R. Gauch and E. Baumgartner, J. Chrornatogr. 161 (1978) 303. H. Stray, J. Assoc. Off. Anal. Chern. 61 (1978) 1359. S. Kubacki and H. Goszcz, Przern. Spozyw. 33 (1979) 427. (CA 092(19)162261) R. Adam and W.-D. Koller, Deutch. Lebensrn. Rundsch. 75 (1979) 254.
364
300 30 1 302
336a 337 338
G. Balicher, H. Woidich and W. Pfannhauser, Ernaerung (Vienna) 4 (1980) 201. A. Anderson and E. Josefsson, Var Fiida 31 (1979) 365. U. Pechanek, G. Blaicher, W. Pfannhauser and H. Woidich, Chromatographia 13 (1980) 421. R.E. Brackett and E.H. Marth, 2.Lebensm. Unters. Forsch. 169 (1979) 92. R.E. Brackett and E.H. Marth, J. Food Prot. 42 (1979) 862. R.E. Brackett and E.H. Marth, In: G. Charalambous (Ed.), Liquid chromatographic analysis of food and beverages, Vol. 2, Academic Press, New York, pp. 499. T.D. Moller and E. Josefsson, J. Assoc. Off. Anal. Chem. 63 (1980) 1055. M. Geipel, W. Baltes, W. Kronert and R. Weber, Chem. Mikrobiol. Technol. Lebensm. 7 (1981) 93. G. Ruggieri and P. Ruggieri, Rass. Chem. 34 (1982) 121. B. Albneyer, K.W. Eichom and R. Plapp, Z. Lebensm. Unters. Forsch. 175 (1982) 172. H.P. Neukom, A. Romann, D. Froehlich, 2.Lebensm. Unters. Forsch. 175 (1982) 342. A. Cavallaro and D. Carreri, Boll. Chim. Unione Ital. Lab. Prov., Parte Sci. 33 (1983) 527. M. Valletrisco, G. Ruggieri, 1. Niola and P. Ruggieri, Ind. Aliment. 22 (1983) 636. R. Battaglia, U.P. Buxtorf, A. Etournaud, H. Guggisberg, U. Leuenberger, J. Luthy, W. Steiner, W. Stutz, A. Cominoli and C. Wyss, Milt. Gebiete Lebensm. Hyg. 75 (1984) 506. R.D. Wilson, Food Technol. New Zealand 16 (1981) 27. P.R. Forbito and N.E. Babsky, J. Assoc. Off. Anal. Chem. 68 (1985) 315. D.N. Mortimer, 1. Parker, M.J. Shepherd and J. Gilbert, Food Addit. Contam. 2 (1985) 165. D. Ehlers, Lebensmittelchem. Gerichtl. Chem. 40 (1986) 2. S.J. Kubacki and H. Gostcz, Pure Appl. Chem. 60 (1988) 871. L. Adensam, H. Lebedova, J. Pavlovsek and B.Turek, Prum. Potravin 40 (1989) 127. F. Vilbois, Bios 12 (1981) 26. R. Woller and P. Majerus, Wein-Wissenschaft 41 (1986) 205. W.T. Stott and L.B. Bullerman, J. Food Sci. 41 (1976) 201. D.C. Hunt, A.T. Bourdon and N.T. Crosby, J. Sci. Food Agric. 29 (1978) 239. E. Isohata, Y. Takeda and M. Uchiyama, Shokuhin Eiseigaku Zasshi 17 /I 976) 308. (CA 086(01)003640) C.P. Kurtzman and A. Ciegler, Appl. Microbiol. 20 (1970) 204. C.W. Bacon, J.G. Sweeney, J.D. Robbins and C. Burdick, Appl. Microbiol. 26 (1973) 155. R.W. Pero, D. harvan, R.G. Owens and J.P. Snow, J. Chromatogr. 65 (1972) 501. C.W. Thorpe and R.L. Johnson, J. Assoc. Off. Anal. Chem. 57 (1974) 861. J.P. Snow, G.B. Lucas, R.W. Harven and R.G. Owens, Appl. Microbiol. 24 (1972) 34. F.Y. Lieu and L.B. Bullerman, J. Food Sci. 42 (1977) 1222. G. Engel, Milschwissenschaft 33 (1978) 201. A. Ciegler, H.J. Mintzlaff, D. Weisleder and L. Leistner, Appl. Microbiol. 24 (1972) 114. L. Leistner and C. Eckardt, Fleischwirtschaft 59 (1979) 1892. P.K. Chan, M.Y. Siraj and A.W. Hayes, J. Chromatogr. 194 (1980) 387. T.D. Phillips, G.W. Ivie, N.D. Heidelbaugh, L.F. Kubena, S.J. Cysewski, A.W. Hayes and D.A. Witzel, J. Assoc. Off. Anal. Chem. 64 (1981) 162. G.D. Hanna, T.D. Phillips, L.F. Kubena, S.J. Cysewski, G.W. Ivie, N.D. Heidelbaugh, D.A. Whihel and A.W. Hayes, Poultry Sci. 60 (1981) 2246. F.L. Neely and R.J. Parks, J. Chromatogr. 540 (1991) 383. B.J. Wilson, C.S. Byerly and L.T. Burka, J. Amer. Vet. Med. Assoc. 179 (1981) 480. L.T. Burka, M. Ganguli and B.J. Wilson, J. Chem. SOC.Chem. Commun. 1983 (1983)
339
R.J. Cole, J.W. Dorner, R.H. Cox, R.A. Hill, H.G. Cutler and J.M. Wells, Awl. Environ.
303 304 305 306 307 308 309 310 31 1 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336
544
365
340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377
Microbiol. 42 (1981) 677. D.T. Wicklow, B.W. Horn and R.J. Cole. Can. J. Bot. 60 (1982) 1050. D.T. Wicklow, R.D. Stubblefield, B.W. Horn and O.L. Shotwell, Appl. Environ. Microbiol. 54 (1988) 1096. R.D. Stubblefield, J.L. Greer and O.L. Shotwell, J. Assoc. Off. Anal. Chem. 71 (1988) 721. G.M. Ware, O.J. Francis, S.S.Kuan and S.A. Carmen, Anal. Lett. 22 (1989) 2335. M.R. Smyth and C.G.B. Frischkorn, Anal. Chim. Acta. 115 (1980) 293. F. Kovacs, C. Szathmary and M. Palyusik, Acta. Vet. Acad. Sci. Hung. 25 (1975) 223. P.M. Scott, T. Panalaks, S.Kanhere and W.F. Miles, J. Assoc. Off. Anal. Chem. 61 (1978) 593. G.M. Ware and C.W. Thorpe, J. Assoc. Off. Anal. Chern. 61 (1978) 1058. M. Malaiyandi and J.P. Barrette, J. Environ. Sci. Health, Part B 13 (1978) 383. T.E. Moiler and E. Josephsson, J. Assoc. Off. Anal. Chem. 61 (1978) 789. E. Josephson and T.E. Moiler, J. Assoc. Off. Anal. Chem. 62 (1979) 1165. R. Fankel and I. Blutsch, Landwirtsch. Forsch. 32 (1979) 250. H. Cohen and M.R. Lapointe, J. Assoc. Off. Anal. Chem. 63 (1980) 642. L. Helan, V. Kahle and K. Tesarik, Vet. Med. (Prague) 28 (1983) 111. (CA 098(21)177588). H.L. Trenholm, R.M. Warner and D.W. Fitzpatrick, J. Assoc. Off. Anal. Chem. 67 (1984) 968. H.L. Chang and J.W. DeVries, J. Assoc. Off. Anal. Chem. 67 (1984) 741. G.A. Bennett, O.L. Shotwell and W.F. Kwolek, J. Assoc. Off.Anal. Chem. 68 (1985) 958. R.W. Bangeris, J.A. Gaul and G.M. Ware, J. Assoc. Off. Anal. Chem. 69 (1986) 894. T. Tanaka, A. Hasegawa, Y. Matsuki, U. Lee and Y. Ueno, J. Chrornatogr. 328 (1985) 271. Y.K. Kim and J.K. Roh, Han'guk Ch'uksan Hakhoechi 28 (1986) 99. (CA 105(19)170721) P. Csokan and F. Kovacs, Magy. Allatorv. Lepja 42 (1987) 205. (CA 107(13)114352). P. Lepom, Arch. Anim. Nutr. 38 (1988) 799. W. Langseth, Y. Ellingsen, U. Nymoen and E.M. IZlkland, J. Chromatogr. 478 (1989) 269. K. Ranfft, R. Gerstl and G. Mayer, 2. Lebensm. Unters. Forsch. 191 (1990) 449. C.L. Holder, C.R. Nony and N.C. Bowman, J. Assoc. Off.Anal. Chern. 60 (1977) 272. J.E. Roybal, R.K. Munns, W.J. Morris, J.A. Hurlbut and W. Shimoda, J. Assoc. Off. Anal. Chem. 71 (1988) 263. M.B. Medina and J.T. Sherman, Food Adit. Contam. 3 (1986) 263. G.V. Turner, T.D. Phillips, N.D. Heidelbaugh and L.H. Russell, J. Assoc. Off. Anal. Chem. 66 (1983) 102. P.M. Scott, and G.A. Lawrence, J. Assoc. Off. Anal. Chem. 71 (1988) 1176. H.L. Trenholm, R.M. Warner and E.R. Farnworth, J. Assoc. Off. Anal. Chem. 64 (1981) 302. D.B. Prelusky, R.M. Warner and H.L. Trenholm, J. Chromatogr. 494 (1989) 267. G.V. Turner, T.D. Phillips, N.D. Heidelbaugh and L.H. Russell, J. Environ. Sci. Health, Part B17 (1982) 297. M.E. Olsen, H.I. Pettersson, K.A. Sandholm and K.-H. C. Kiessling, J. Assoc. Off. Anal. Chem. 68 (1985) 632. L.J. James, L.G. McGirr and T.K. Smith, J. Assoc. Off. Anal. Chem. 65 (1982) 8. N. Verbiese-Genard and M. Hanocq, Anal. Lett. 19 (1986) 1229. J.C. Frisvad and 0. Filtenborg, Appl. Environ. Microbiol. 46 (1983) 1301. J.C. Frisvad, in: R.A. Samson and J.I. Pitt (Eds.), Advances in Penici//ium and Aspergillus systematics, Plenum Press, New York, pp. 327. J.C. Frisvad, in: R.J. Cole (Ed.), Modern methods in the analysis and structure
366
378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 3 96 397 398 399 400 401 402 403 404 405 406 407 408 409 410 41 1 412
elucidation of mycotoxins, Academic Press, New York, pp. 415. B. Hald and P. Krogh, J. Assoc. Off. Anal. Chem. 58 (1975) 156. R. Tresel, E. Hommel and B. Helak, Monatsschr. Brauwiss. 42 (1989) 331. K. Rannft, R. Gerst and G. Mayer, Agribiol. Res. 43 (1990) 44. M. Shimomura and E. Ishikuro, Shiryo Kenkyo Hokoku 14 (1989) 1. (CA 111(11)095674) B.G. Osborne, J. Sci. Food Agric. 30 (1979) 1065. A. Ibe, M. Nishishima, K. Yasuda, K. Saito, H. Kamimura, T. Nagayarna, H. Ushiyama, Y. Naoi and T. Nishima, Shokuhin Eiseigaku Zasshi 25 (1984) 334. (CA 102(01)004533). H. Schweighardt, M. Schuh, A.M. Abdelhamid, J. Boehm and J. Leibetseder, Z. Lebensm. -Unters. Forsch. 170 (1980) 355. A.A. El-Banna and P.M. Scott, J. Food Prot. 47 (1984) 189. N. Chamkasem, W.Y. Cobb, G.W. Latimer, C. Salinas and B.A. Clement, J. Assoc. Off. Anal. Chem. 72 (1989) 336. P. Lepom, J. Chromatogr. 355 (1986) 335. P. Nowotny, W. Baltes, W. Kronert and R. Weber, Chem. Mikrobiol. Technol. Lebendsm. 8 (1983) 29. A. Cantafora, M. Grossi, M. Miraglia and L. Benelli, Riv. SOC.Ital. Sci. Aliment. 12 (1983) 103. C. Micco, M. Grossi, M. Miraglia and C. Brera, Food Addit. Contam. 6 (1989) 333. H. Terada, H. Tsubouchi, K. Yamamoto, K. Hisada and Y. Sakabe, J. Assoc. Off. Anal. Chem. 69 (1986) 960. W.J. Hurst and R.A. Martin, J. Chromatogr. 265 (1983) 353. A.M. Abdel-Hamid; H. Schweighardt, J. Boehm, J. Leibetseder and M. Schuh, Agric. Res. Rev. 61 (1988) 91. H. Schweighardt, M. Schuh, A.M. Abdel-Hamid, J. Bohm and J. Leibetseder, Z. Lebensm. -Unters. Forsch. 170 (1980) 355. D.C. Hunt, A.T. Bourdon, P.J. Wild and N.T. Crosby, J. Sci. Food Agric. 29 (1978) 234. H. Schweighardt and J. Leibetseder, Wien. Tierartzl. Mschr. 68 (1981) 302. M. Nakajima, H. Terada, K. Hisada, H. Tsubouchi, K. Yamamoto, Y. Itoh, T. Uda, 0. Kawamura and Y. Ueno, Proc. Jpn. Assoc. Mycotoxicol. 32 (1990) 50. U. Baumann and 8. Zimmerli, Mitt. Geb. Lebendsm. Hyg. 79 (1988) 151. L. Adensam, M. LebedovA and B. Turek, Ces. Hyg. 34 (1989) 39. P.Nowotny,W. Baltes, W. Kronert and R. Weber, Lebensmittelchem. Gerichtl. Chem. 37 (1983) 70. J. Piskorska-Pliszczynska, Bromatol. Chem. Toxicol. 17 (1984) 327. T.D. Phillips, A.F. Stein, G.W. Ivie, L.F. Kubena, A.W. Hayes and N.D. Heidelbaugh, J. Assoc. Off. Anal. Chem. 66 (1983) 570. D.C. Hunt, L.A. Philip and N.T. Crosby, Analyst 104 (1979) 1171. D.C. Hunt, B.R. McConnie and N.T. Crosby, Analyst 105 (1980) 89. J. Bauer, M. Gareis and B. Gedek, Bed. Munch. Tierartzl. Wochenschr. 97 (1984) 279. W. Unglaub and F. Holl, Fleischwirtschaft 70 (1990) 406. H. Huang, Z. Zhao, H. Sun and Z. Niu, Sepu 8 (1990) 189. (CA 113(11)096233) H.M. Stahr, M. Domoto, B.L. Zhu and R. Pfeiffer, Mycotoxin Res. 1 (1985) 31. C. Wilken, W. Baltes, I. Mehlitz, R. Tiebach and R. Weber, 2. Lebensm. Unters. Forsch. 180 (1985) 496. E. Josefsson, Vhr Fijda 31 (1979) 415. M. Gareis, E. Martlbauer, J. Bauer and B. Gedek, Proc. Jpn. Assoc. Mycotoxicol. Suppl. 1 (1988) 61. M. Gareis, E. Martlbauer, J. Bauer and B. Gedek, 2. Lebensm. Unters. Forsch. 186 (1988) 114.
361 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 43 1 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454
E. Martlbauer and G. Terplan, Arch. Lebensrnittelhyg. 39 (1988) 133. K. Hult, R. Fuchs, M. Peraica, R. Plestina and S. Ceovic, J. Appl. Toxicol. 4 (1984) 326. A. Breitholh, M. Olsen, A. Dahlback and K. Hult, Food Addit. Contarn. 8 (1991) 183. M. Castagnero, V. Maru, G. Maru and M.-D. Ruiz-Lopez, Analyst 115 (1990) 129. C.M. Lerch and H.-M. Muller, Chrornatographia 30 (1990) 424. P.M. Scott, W. van Walbeek, 8. Kennedy and D. Anyeti, J. Agric. Food Chern. 20 (1972) 1103. P. Krogh, B. Hald and E.J. Pederson, Acta Path. Microbiol. Scand. 818 (1973) 689. M. Saito, M. Enornoto, and T. Tatsuno, in: A. Ciegler, S. Kadis and S.J. Ail (Eds.), Microbial toxins Vol. VI, Academic Press, New York, 1971, p. 357. R. Dick, U. Baurnann and 6. Zirnrnerli, Mitt. Geb. Lebensrn. Hyg. 79 (1988) 159. L.R. Marti, D.M. Wilson and B.D. Evans, J. Assoc. Off.Anal. Chern. 61 (1978) 1353. R.D. Phillips, A.W. Hayes and W.O. Berndt, J. Chrornatogr. 190 (1980) 419. R.B. Vail and M.J. Hornann, J. Chrornatogr. 535 (1990) 317. T. Nakagawa, T. Kawarnura, Y. Fujirnoto and T. Tatsuno, Shokuhin Eiseigaku Zasshi 23 (1982) 297. D.L. Orti, R.H. Hill, J.A. Liddle, L.L. Needharn and L. Vickers, J. Anal. Toxicol. 10 (1986) 41. B. Zirnmerli, R. Dick and U. Baurnan, J. Chrornatogr. 462 (1989) 406. R. Schwarzenbach, J. Chrornatogr. 334 (1985) 35. 8. Hald, D.H. Christensen and P. Krogh, Appl. Environ. Microbiol. 46 (1983) 1311. K.A. Scudarnore, P.M. Atkin and A.E. Buckle, J. Stored Prod. Res. 22 (1986) 81. J.H. Wall and E.B. Lillehoj, J. Chrornatogr. 268 (1983) 461. M.E. Stack, N.L. Brown and R.M. Eppley, J. Assoc. Of.Anal. Chern. 61 (1978) 590. M.E. Stack, P.B. Mislivec, T. Denizel, R. Gibson and A.E. Pohland, J. Food Prot. 46 (1983) 965. M.E. Stack and P.B. Mislivec, Appl. Environ. Microbiol. 36 (1978) 552. A. Ciegler, L.S. Lee and J.J. Dunn, Appl. Environ. Microbiol. 42 (1981) 446. A S . Carrnan, S.S.Kuan, O.J. Francis, G.M. Ware, J.A. Gaul and C.W. Thorpe, J. Assoc. Off. Anal. Chern. 66 (1983) 587. S.S. Kuan, AS. Carrnan, O.J. Francis and G.M. Ware, Anal. Lett. 18 (1985) 837. AS. Carrnan, S.S.Kuan, O.J. Francis, G.M. Ware and A.L. Luedtke, J. Assoc. Off. Anal. Chern. 67 (1984) 1095. R.E. Peterson and A. Ciegler, Appl. Environ. Microbiol. 36 (1978) 613. R.E. Peterson and M.D. Grove, Appl. Environ. Microbiol. 45 (1983) 1937. I. Delgadillo, Mycotoxin Res. 2 (1986) 9. C.M. Jansen and K. Dose, Mycotoxin Res. 1 (1985) 11. J.E. Burnside, W.L. Sippel, J. Forgacs, W.T. Carll, M.B. Atwood and E.R. Doll, Arner. J. Vet. Res. 18 (1957) 817. B.J. Wilson and C.H. Wilson, J. Bacteriol. 84 (1962) 283. B.J. Wilson and C.H. Wilson, J. Bacteriol. 83 (1962) 693. P.D. Unger and A.W. Hayes, J. Chrornatogr. 153 (1978) 115. G.W. Engstrorn and J.L. Richard, J. Agric. Food Chern. 29 (1981) 1164. J.M. Wells, R.J. Cole and J.W. Kirksey, Appl. Microbiol. 30 (1975) 26. H. Anke, 1. Kolthourn, H. Zahner and H. Laatsch, Arch. Microbiol. 126 (1985) 223. M. Bachrnann, J. Luthy and C. Schlatter, J. Agric. Food Chern. 27 (1979) 1342. P. Blaser, H. Ramstein, W. Schmidt-Lorenz and C. Schlatter, Lebensrn. Wiss. Technol. 14 (1980) 66. C.-H. Chen, P.-F. Cheng and L.-H. Wang, Yao Hsueh T’ung Pao 16 (1981) 64. (CA 095(16)138706) L. Zhao, S. Lin, Y. Wang and S. Dong, Zhongguo Yaoke Daxue Xuebao 19 (1988) 266. (CA 110(16)141633) M. Kurnano, S. Handa and F. Hirayarna, Nagasaki-ken Eisei Kogai Kenkyushoho 30
368
455 456 457 458 459 460 46 1 462 463 464 465 466 467 468 469 470 47 1 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 49 1
(1988) 129. (CA 111(11)095688). Y. Sun and 0. Chen, Yaowu Fenxi Zazhi 5 (1985) 294. (CA 104(06)039830). P.P. Rai, T.D. Turner and S.A. Matlin, J. Chrornatogr. 110 (1975) 401. A.J.J. van den Berg and R.P. Labadie, J. Chromatogr. 329 (1985) 311. V. Quercia, Pharmacology 20 (1980) 76. D. Matthees, J. Agric. Food Chem. 31 (1983) 453. C.S. Reddy, R.V. Reddy and A.W. Hayes, J. Chrornatogr. 208 (1981) 17. K.C. Ehrlich, L.S. Lee, A. Ciegler and M.S. Palrngren, Awl. Environ. Microbiol. 44 (1982) 1007. J.L. Richard, R.L. Lyon, R.E. Fichtnerand P.F. Ross, Mycopathologia 107 (1989) 145. D.R. Lauren and R.J. Fairclough, J. Chrornatogr. 200 (1980) 288. C.A. Halder, R.A. Taber and B.J. Camp, J. Chrornatogr. 175 (1979) 356. J.W. Ronaldson, J. Chern. Tech. Biotechnol. 32 (1982) 556. M. Bonnefoi, P. Sauvagnac, F. Massat and J. Le Bars, Rev. Med. Vet. (Toulouse) 138 (1987) 991. J.L. Richard and L.H. Arp, Mycopathologia 67 (1979) 107. J.L. Richard, P. Bacchetti and L.H. Arp, Mycopathologia 76 (1981) 55. R.J. Cole, J.W. Dorner, R.H. Cox and L.W. Raymond, J. Agric. Food Chern. 31 (1983) 657. B.J. Wilson, C.H. Wilson and A.W. Hayes, Nature (London) 220 (1968) 77. J.W. Dorner, R.J. Cole and R.A. Hill, J. Agric. Food Chem. 32 (1984) 411. C.M. Maes, P.S. Steyn and F.R. van Heerden, J. Chrornatogr. 234 (1982) 489. P.G. Mantle, K.P.W.C. Perera, N.J. Maishrnan and G.R. Mundy, Appl. Environ. Microbiol. 45 (1983) 1486. M.E. di Menna, D.R. Lauren and P.A. Wyatt, Appl. Environ. Microbiol. 51 (1986) 821. C.M. Weedon and P.G. Mantle, Phytochernistry 26 (1987) 969. D.R. Lauren and R.T. Gallagher, J. Chrornatogr. 248 (1982) 150. R.J. Cole, J.W. Dorner, J.P. Springer and R.H. Cox, J. Agric. Food Chern. 29 (1981) 293. J.B. Gloer, M.R. TePaske, J.S. Sirna, D.T. Wicklow and P.F. Dowd, J. Org. Chem. 53 (1988) 5457. K. Nozawa, S.Sekita, M. Harada, S.Udagawa and K. Kawai, Chem. Pharrn. Bull. 37 (1989) 626. T. Tanaka, A. Hasegawa, N. Aoki, S.Yarnarnoto, S.Udagawa, S.Sekita, M. Harada, K. Nozawa and K. Kawai, Proc. Jpn. Assoc. Mycotoxicol. 30 (1989) 19. M.R. TePaske, J.B. Gloer, D.T. Wicklow and P.F. Dowd, J. Org. Chern. 55 (1990) 5299. J.B. Gloer, B.L. Rinderknecht, D.T. Wicklow and P.F. Dowd, J. Org. Chern. 54 (1989) 2530. M.R. TePaske, J.B. Gloer, D.T. Wicklow and P.F. Dowd, Tetrahedron Lett. 32 (1991) 5687. M.R. TePaske, J.B. Gloer, D.T. Wicklow and P.F. Dowd, J. Org. Chern. 54 (1989) 4743. M.R. TePaske, J.B. Gloer, D.T. Wicklow and P.F. Dowd, Tetrahedron Lett. 30 (1989) 5965. M.R. TePaske, J.B. Gloer, D.T. Wicklow and P.F. Dowd, Tetrahedron 45 (1991) 4961. R.J. Cole, J.W. Kirksey, J.W. Dorner, D.M. Wilson, J.C. Johnson, A.N. Johnson, D.M. Bedell, J.P. Springer, K.K. Chexal, J.C. Clardy and R.H. Cox, J. Agric. Food Chern. 25 (1977) 826. P.V. Nielsen, L.R. Beuchat and J.C. Frisvad, Appl. Environ. Microbiol. 54 (1988) 1504. P.V. Nielsen, L.R. Beuchat and J.C. Frisvad, J. Appl. Bact. 66 (1989) 197. P.V. Nielsen, L.R. Beuchat and J.C. Frisvad, J. Food Sci. 54 (1989) 679. A.L. Dernain, N.A. Hunt, V. Malik, B. Kobbe, H. Hewkins, K. Matsuo, and G.N. Wogan, Appl. Environ. Microbiol. 31 (1976) 138.
369
492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525
P.S. Steyn and R. Vleggaar, in: W. Hen, H. Grisebach, G.W. Kirby and C. Tamm (Eds.), Progress in chemistry of organic natural products. Vol. 111, Springer Verlag, Wien, 1985, p. 1. K.H. Ling, C.-K. Yang and F.-T. Peng, Appl. Environ. Microbiol. 37 (1979) 355. K.H. Ling, H.-H. Liou, C.-M. Yang and C.-K. Yang, Appl. Environ. Microbiol. 47 (1984) 98. F.4. Peng, K.H. Ling, Y. Wang and G.-H. Lee, Appl. Environ. Microbiol. 49 (1985) 721. K.H. Ling, in: G.R. Waller (Ed.), Phytochemicalecology: allelochemicals, mycotoxins and insect pheromones and allomones. Academia Sinica Monograph Series No. 9. Institute of Botany, Taipei, 1989, p. 355. K.H. Ling, E.4. Wang, M.-S. Chiang and L.-J. Chen, J. Biomed. Lab. Sci. 2 (1989) 143. J.C. Frisvad, Chemometrics Intell. Lab. Syst. 12 (1992) In press K.H. Ling, C.-K. Yang and H.-C. Huang, Awl. Environ. Microbiol. 37 (1979) 358. D.E. Schade and A.D. King, J. Food Prot. 47 (1984) 266. D.J. Harvan and R.W. Pero, in: J.V. Rodricks (Ed.): Mycotoxins and other fungal related problems. Adv. Chem. Ser. Vol. 149 (12976) 344. R. Suemitsu, K. Horiuchi, K. Ohnishi and S. Yanagawase, J. Chromatogr. 454 (1988) 406. R. Suemitsu, K. Horiuchi, K. Ohnishi and M. Kubota, J. Chromatogr. 503 (1990) 282. D.J. Siler and D.C. Gilchrist, J. Chromatogr. 238 (1982) 167. P.M. Scott and S.R. Kanhere, J. Assoc. Off. Anal. Chern. 63 (1980) 612. E.G. Heisler, J. Siciliano, E.E. Stinson, S.F. O m a n and D.D. Bills, J. Chromatogr. 194 (1980) 89, E.E. Stinson, D.D. Bills, S.F. Osman, J. Siciliano, M.J. Ceponis and E.G. Heisler, J. Agric. Food Chem. 28 (1980) 960. E.E. Stinson, S.F. Osman, E.G. Heisler, J. Siciliano and D.D. Bills, J. Agric. Food Chem. 29 (1981) 790. M.E. Stack, P.B. Mislivec,J.A.G. Roach andA.E. Pholand, J. Assoc. Off. Anal. Chem. 68 (1985) 640. R.A. Samson and J.C. Frisvad, Proc. Jpn. Assoc. Mycotoxicol. 32 (1990) 3. M.H. Lebrun, F. Gaudemer, M. Boutar, L. Nicolas and A. Gaudemer, J. Chromatogr. 464 (1989) 307. M.H. Lebrun, F. Dutfoy, F. Gaudemer, G. Kunesh and A. Gaudemer, Phytochemistry 29 (1990) 3777. L.M. Seitz, in: V. Betina (Ed.): Mycotoxins - Production, isolation, separation and purification., Elsevier, Amsterdam, 1984, p. 443. F.S. Chu and S.C. Bennett, J. Assoc. Off.Anal. Chem. 64 (1981) 950. S.Ozcelik, N. Ozcelik and L.R. Beuchat, Intl. J. Food Microbiol. 11 (1990) 187. M. Wittkowski, W. Baltes, W. Kronert and R. Weber, Z. Lebensm. Unters. Forsch. 177 (1983) 447. A. Visconti, A. Logrieco and A. Bottalico, Food Addit. Contam. 3 (1986) 323. F. Palmisano, P.G. Zambonin, A. Visconti and A. Bottalico, J. Chromatogr.465 (1989) 305. F. Palmisano, A. Sibilia and A. Visconti, Chromatographia29 (1990) 333. A. Visconti, A. Sibilia and F. Palmisano, J. Chromatogr. 540 (1991) 376. J.V. Edwards and E.B. Lillehej, J. Assoc. Off. Anal. Chem. 70 (1987) 126. R.J. Samuels, A.K. Charnley and S.E. Reynolds, J. Chromatogr. Sci. 26 (1988) 15. G.R. Hancock, P. Vogel and D.S. Petterson, Aust. J. Exp. Agric. 27 (1987) 73. Y. Ueno, in: V. Betina (Ed.): Mycotoxins - Production, isolation, separation and purification, Elsevier, Amsterdam, 1984, p. 475. S.E. Yeulet, P.G. Mantle, M.S. Rudge and J.B. Greig, Mycopathologia 102 (1988) 102.
370 526 527 528 529 530 531 532 533 534 535 536 537 538
539 540 541 542 543 544 545 546 547 548 549
550 551 552 553 554
W.A. Moats, J. Chromatogr. 507 (1990) 177. J.O. Boison, C.D.C. Salisbury, W. Chan and J.D. MacNeil, J. Assoc. Off.Anal. Chem. 74 (1991) 497. R. Berger and M. Petz, Deutche Lebensm. Rundschau 87 (1991) 137. J.I. Azcona-Olivera, M.M. Abouzied, R.D. Plattner and J.J. Pestka, J. Agric. Food Chem. 40 (1992) 531. R. Vesonder, R. Peterson, R. Plattner and 0 .Weisleder, Mycotoxin Res. 6 (1990) 85. P.F. Ross, P.E. Nelson, J.L. Richard, G.D. Osweiler, L.G. Rice, R.D. Plattner and T.M. Wilson, Appl. Environ. Microbiol. 56 (1990) 3225. G.S. Shephard, E.W. Sydenham, P.G. Thiel and W.C.A. Gelderblom, J. Liq. Chromatogr. 13 (1990) 2077. J.F. Alberts, W.C.A. Gelderblom, P.G. Thiel, W.F.O. Marasas, D.J. van Schalwyk and Y. Berend, Appl. Environ. Microbiol. 56 (1990) 1729. E.W.Sydenham, P.G.Thiel, W.F.O. Marasas,G.S. Shephard, D.J.vanSchalwykand K.R. Koch, J. Agric. Food Chern. 38 (1990) 1900. P.G. Thiel, G.S. Shephard, E.W. Sydenham, W.F.O. Marasas, P.E. Nelson and T.M. Wilson, J. Agric. Food Chem. 39 (1990) 109. T.M. Wilson, P.F. Ross, L.G. Rice, G.D. Osweiler, P.A. Nelson, D.L. Owens, R.D. Plattner, C. Reggiardo, T.H. Noom, and J.W. Pickrell, J. Vet. Diagn. Invest 2 (1990) 213. P. Krogh, D.H. Christensen, 8 . Hald, 8. Harlou, C. Larsen, E.J. Pedersen and U. Thrane, Appl Environ. Microbiol. 55 (1989) 3184. K. Singh, J.C. Frisvad, U. Thrane and S.B. Mathur, An illustrated manual on identification of some seed-borne Aspergilli, Fusaria, Penicillia and their mycotoxins. Danish Government Institute of Seed Pathology for Developing Contries,- Hellerup, 1991. W. Wu, P.E. Nelson, M.E. Cook and E.B. Smalley, Appl. Environ. Microbiol. 56 (1990) 2989. J. Yu and F.S. Chu, J. Assoc. Off. Anal. Chem. 74 (1991) 855. W.C.A. Gelderblom, P.G. Thiel, W.F.O. Marasas and K.J. van der Merwe, J. Agric. Food Chem. 32 (1984) 1064. W.C.A. Gelderblom, W.F.O. Marasas, P.S. Steyn, P.G. Thiel, K.J. van der Merwe, P.H. van Rooyen, R. Vleggaar and P.L. Wessels, J. Chem. SOC.Chem. Commun. (1984) 122. L.A. Wiebe and L.F. Bjeldanes, J Food Sci. 46 (1981) 1424. J.M. Farber and G.W. Sanders, Appl. Environ. Microbiol. 51 (1986) 381. P.M. Scott, G.A. Lawrence and T.I. Matula, in: P.S. Steyn and R. Vleggaar (Eds.), Mycotoxins and phycotoxins '86, Elsevier, Amsterdam, 1986, p. 305. U. Thrane, Mycotoxin Res. 4 (1988) 2. M.A. Jackson, J.N. Stewart, R.E. Peterson and P.J. Slininger, J. Agric. Food Chem. 38 (1990) 1511. W.C.A. Gelderblom, P.G. Thiel, K.J. van der Merwe, W.F.O. Marasas and H.S.C. Spies, Toxicon 21 (1983) 467. P.S. Thiel, W.C.A. Gelderblom, W.F.O. Marasas, P.E. Nelson and T.M. Wilson, J. Agric. Food Chem. 34 (1986) 773. J.M. Farber and P.M. Scott, in: J. Chelkowski (Ed.), Fusarium- Mycotoxins, taxonomy and pathogenicity. Elsevier, Amsterdam, 1989, p. 41. M.J. Shepherd and J. Gilbert, J. Chromatogr. 358 (1986) 415. P.M. Scott and G.A. Lawrence, J. Assoc. Off. Anal. Chem. 70 (1987) 850. H.R. Buneister, M.D. Grove, R.E. Peterson, D. Weisleder and R.D. Plattner, Appl. Environ. Microbiol. 50 (1985) 31 1. C. Dore, M. Jullien, D. Boussotrot and R. Cassini, in: E.C. Lougheed and H. Tiessen (Eds.), Proceedings of the sixth International Asparagus Symposium, Guelph, Canada, 1985, University of Guelph Press, Guelph, 1985, p. 246.
371 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 586a 587 588 589 590
J.-T. Lin, A.E. Stafford, G.L. Steffens and N. Murofushi, J. Chromatogr. 543 (1991) 471. J.M. Brackett, M.D. Arguello and J.C. Schaar, J. Agric. Food Chem. 36 (1988) 762. H.I. Assil and P. Sporns, J. Agric. Food Chem. 39 (1991) 2206. X.J. Liu, X.Y. Luo and W.J. Hu, in: S. Natori, K. hashimoto and Y. Ueno (Eds.), Mycotoxins and phycotoxins '88. Elsevier, Amsterdam, 1989, p. 109. A.J. Penel and F.V. Kosikowski, J. Food Prot. 53 (1990) 321. A.D. Muir and W. Majak, Toxicol. Lett. 20 (1984) 133. A.D. Muir, W. Majak, M.A. Pass and G.S. Yost, Toxicol. Lett. 20 (1984) 137. W. Majak and R.E. McDiarrnid, Toxicol. Lett. 50 (1990) 213. R.E. Peterson, G.M. Shannon and O.L. Shotwell, J. Assoc. Off. Anal. Chem. 72 (1989) 332. J.A. Lansden, J. Assoc. Off. Anal. Chem. 67 (1984) 728. W.P. Norred, R.J. Cole, J.W. Dorner and J.A. Lansden, J. Assoc. Off.Anal. Chem. 70 (1987) 121. T. Goto, E. Shinshi, K. Tanaka and M. Manabe, Agric. Biol. Chem. 51 (1987) 2581. P. Haggblom, Appl. Environ. Microbiol. 56 (1990) 2924. G.M. Ware, C.W. Thorpe and A.E. Pohland, J. Assoc. Off. Anal. Chem. 63 (1980) 637. B. Danieli, M. Magri and D. Lavezzari, in: XXI International Dairy Congress. Vol. 1. Book 2. Moscow, USSR, 1982, p. 170. S. Moreau, A. Masset and J. Biguet, Appl. Environ. Microbiol. 37 (1979) 1059. S. Moreau, A. Combier-Lablanche and J. Biguet, Appl. Environ. Micobiol. 39 (1980) 770. S.-C. Chang, Y.-H. Wei, D.-L. Wei, Y.-Y. Chen and S.-C. Jong, Appl. Environ. Microbiol. 57 (1991) 2581. B. Danieli, B. Bianchi-Salvadori and A.V. Zambrini, Milchwissenschaft 35 (1980) 423. M. Li and F. Jia, Shipin Kexue 23 (1981) 33. G.F. Griffin, S.C. Bennett and F.S. Chu, J. Chromatogr. 280 (1983) 363 F.G. Thiel, C.J. Meyer and W.F.O. Marasas, J. Agric. Food Chem. 30 (1982) 308. E. Rajakyla, K. Leasasenako and P.J.D. Sakkers, J. Chromatogr. 384 (1987) 391. R.R.M. Paterson and C. Kemmelmeier, J. Chromatogr. 483 (1989) 153. M. Bogusz, J. Chromatogr. 387 (1987) 401. C. Reichardt, Solvent and solvent effects in organic Chemistry, VCH, Weinheim, 1988. A.I. Scott, Interpretation of the ultraviolet spectra of natural products, Pergamon Press, Oxford, 1964. M. Jaquet and P. Laszlo, in: A. Weissberger (Ed.), Techniques of chemistry, Vol. VIII, Solutions and solubilities, John Wiley and Sons, New York, 1975. R.R.M. Paterson and C. Kemmelmeier, J. Chromatogr. 511 (1990) 195. A.L. Yergey, C.G. Edmonds, I.A.S. Lewis and M.L. Vestal, Liquid chromatographylmass spectrometry - Techniques and applications, Plenum Press, New York, 1990. W.M.A. Niessen and J. van der Greef, Liquid chromatography -mass spectrometry Principles and applications, Marcel Dekker, New York, 1990. M.A. Brown (Ed.), Liquid chromatographylmass spectrometry - Applications in agricultural, pharmaceutical and environmental chemistry, ACS Symposium Series 420, American Chemical Society, Washington D.C., 1990. E.R.J. Wils and A.G. Hulst, Fres. 2. Anal. Chem. 342 (1992) 749. R. Tiebach, W. Blaas and M. Kellert, J. Chromatogr. 323 (1985) 121. R. Tiebach, W. Blaas, M. Kellert, S. Steinmeyer and R. Weber, J. Chromatogr. 318 (1985) 103. T. Krishnamurthy, D.J. Beck and R.K. Isensee, Biomed. Environ. Mass Spetrom. 18 (1989) 287. R.D. Voyksner, W.M. Hagler and S.P. Swanson, J. Chromatogr. 394 (1987) 183.
372 591 592 593 594 595 596
R.D. Voyksner, W.M. Hagler, K. Tyczkowska and C.A. Haney, J. High Resolution Chromatogr. Chromatogr. Commun. 8 (1985) 119. H. Carlson, P. Nillson, H.-B. Jansson and G. Odham, J. Microbiol. Methods 13 (1991) 259. T. Krishnamurthy, D.J. Beck, R.K. lsensee and B.B. Jarvis, J. Chromatogr. 469 (1989) 209. J.R.J. Pare, R. Greenhalgh, P. Lafontaine and J.W. Apsimon, Anal Chem. 57 (1985) 1472. R. Kostiainen, K. Matsuura and K. Nojima, J. Chromatogr. 538 (1991) 323. W.P. Korfmacher, M.P. Chiarelli, J.O. Lay, J. Bloom and M. Holcomb, Rapid Comm. Mass Spetrometry 5 (1991) 463.
313
Chapter 9 GAS CHROMATOGRAPHY OF MYCOTOXINS PETER M. SCOTT 9.1. INTRODUCTION
Gas chromatography (GC) may be used as an analytical technique for mycotoxins that can be volatilized in a heated GC column or possess at least one functional group allowing conversion of the mycotoxin into a volatile derivative. In practice, the reactive group is in almost all cases a hydroxyl group and the derivatives formed are usually trimethylsilyl (TMS) ethers or heptafluorobutyryl (HFB), pentafluoropropionyl (PFP) or trifluoroacetyl (TFA) esters. The first mycotoxin (apart from oxalic acid) to be analysed by GC was zearalenone (F-2) as reported in 1967 by Mirocha et al. (1), followed by patulin, mycophenolic acid, griseofulvin, koj ic acid, terreic acid and terrein in 1970 (2-5) and in 1971 by certain ergoline alkaloids, sterigmatocystin, alternariol and related Alternaria toxins, penicillic acid and various trichothecenes (610). Since then the application of GC to mycotoxin analysis has grown tremendously, mainly because of the interest in trichothecenes, butthere appears to be only one review specifically devoted to this topic (11). Reviews on chromatographic analysis of mycotoxins in general have, of course, included GC (12-15). GC has a major advantage over other forms of chromatography, liquid chromatography (LC)-mass spectrometry (MS) notwithstanding, in that it can be readily coupled to a mass spectrometer to enable more specific detection and determination of mycotoxins, as well as their identification. Vesonder and Rohwedder (16) have reviewed this specialized technique of GC-MS and its application to mycotoxin analysis. Flame ionization detection (FID) and electron capture detection (ECD) are the main other techniques used for GC of mycotoxins. The trichothecenes are the only mycotoxins for which GC is widely used. There is one Official Method of the Association of Official Analytical Chemists (AOAC) that employs GC for mycotoxin determination, viz. for deoxynivalenol (DON) in wheat. 9.2. TRICHOTHECENES 9.2.1 Introduction
The total number of trichothecenes isolated from natural sources (mainly fungal) was 148 at last count (17), comprising 83 non-macrocyclic and 65 macrocyclic compounds. Only a few of these, in particular DON, nivalenol (NIV), T-2 toxin (T-2), HT-2 toxin (HT2) and diacetoxyscirpenol (DAS) (Fig. 9.1) have been detected so far as naturally occurring contaminants in foodstuffs (18).
374
CH20H
I1
I
Fig. 9.1. Examples of trichothecenes. Type A: diacetoxyscirpenol ( I , R'=H; R2=OCOCH,) , T-2 toxin (I,R'= (CH,) ,CHCH,COO : R2=OCOCH,) , HT-2 toxin (I,R1=(CHS)2CHCH2COO;R2=OH). Type B: nivalenol (11, R=OH) , deoxynivalenol (11, R=H)
.
GC is the most commonly used means of separating and identifying trichothecenes, not only in extracts of foodstuffs but also in biological fluids and tissues. These applications will be discussed later (see Sections 9.2.3-9.2.6). Several types of internal standards (e.9. deuterated DAS and HT-2, methoxychlor, 7hydroxy DAS, 16-hydroxyverrucarol, isoT-2) have been used in GC determinations of trichothecenes (19-25, inter alia) ; they have been added to the sample before extraction, before derivatization or after derivatization. Derivatization and detection Drocedures for trichothecenes 9.2.2.1 No derivatization. It is normal to derivatize the hydroxyl group(s) of trichothecenes for GC in order to attain the volatility and sensitivity needed for trace analysis. However, several workers have omitted the derivatization step. Stahr et al. (26,27) demonstrated that GC of underivatized T-2, DAS and other trichothecenes was possible, with a sensitivity of 10-100 ng on a packed column using FID. Bijl et al. (28) also detected ng amounts of T-2 and DAS, as well as trichothecin (which has no free hydroxyl group), by capillary GC with FID. D'Agostino et al. (29) performed capillary GC on underivatized verrucarol, DON, DAS, T-2, HT-2 and T-2 triol using both FID and ammonia chemical ionization (CI) MS detection. Co-injection of standards with an acetone plug improved peak shape. T-2, HT-2 and T-2 triol did not separate on DB-1 but there were slight differences in retention time on a DB-5 column. MS detection, particularly negative ion (NI) CI (30-33), has in fact been the preferred means of detection for GC of underivatized trichothecenes. Detection limits for 9 trichothecenes detected by oxygen NICI capillary GC-MS ranged from 50 pg for DON, monitored at masses of 284 and 295, to 9.2.2
3 75
375 pg for neosolaniol (NS) and T-2 trio1 at a signal-to-noise ratio of 1O:l (32). Lau et al. (34) determined eight underivatized trichothecenes by capillary column GC-CI tandem mass spectrometry (MS/MS) with detection limits in the range 10-67 pg except for HT-2 (305 pg): HT-2 and T-2 did not separate on the 15 m DB-5 capillary column used. GC of underivatized trichothecenes has been particularly useful for their characterization in extracts of fungal cultures. Capillary columns have been employed, with detection by FID or MS (operated in the electron impact (EI) mode) (35-38). The GC-MS technique was most useful for identification of new trichothecenes. However, Plattner et al. (39) noted losses and reproducibility problems with underivatized T-2 and NS, but not with DAS, when these trichothecenes were introduced into the mass spectrometer by GC for MS/MS analysis. On-column injection of underivatized DON caused up to 8% degradation, principally to isoDON, when assay was made by capillary GC-MS (35). As previously mentioned, not all trichothecenes possess derivatizable hydroxyl groups. Thus when GC of acetyl T-2 in the presence of trichothecenes that had been trimethylsilylated (40,41), heptafluorobutyrylated (41) or trifluoroacetylated (42) was reported, it was of course not being chromatographed as a derivative. Similary, triacetoxyscirpenol, trichodermin, crotocin and trichothecin were necessarily chromatographed underivatized, with FID detection, in the presence of TMS derivatives of trichothecenes possessing free hydroxyl groups (10,43). Trichothecin, together with its de-esterified analog trichothecolone, has been chromatographed in the absence of derivatizing agent and detected by FID (28,44,45). Ishii et al. (46) measured trichothecin in wheat by GC-MS (EI mode) using ions at m/z 246, 203 and 175. 9.2.2.2 Trimethylsilylation. The very first report on GC of trichothecenes was by Ikediobi et al. (10) who formed TMS derivatives of a number of trichothecenes with derivatizable hydroxyl groups, plus four that would not have derivatized (see section 9.2.2.1) and were not affected by the silylation reagents used. These reagents were (i) hexamethyldisilazane (HMDS) trimethylchlorosilane(TMCS)-pyrid~ne(2+1+7) and (ii) N,Obis(trimethylsily1)acetamide (BSA) - pyridine (4+1), used at room or refrigerator temperature. The second reagent was preferred as the derivatives were stable for at least two weeks at room temperature and much longer at -2OOC. Low pg quantities of trichothecenes were detected by FID. Subsequently, TMS derivatives have been the ones most frequently used for GC of trichothecenes (particularly for type B trichothecenes possessing 7-hydroxyl and conjugated 8-carboxyl groups) (Fig. 9.1). Various reagent mixtures have been employed. Type A trichothecenes such as DAS and T-2 are readily derivatized, even with BSA alone (47.48). However, Tanaka et al. (49) showed
376
that HMDS-TMCS-pyridine gave two peaks with diacetyl NIV and fusarenone-X (FX), both type B trichothecenes, after 0.25 or 6 hours at room temperature. Nakahara and Tatsuno (50) trimethylsilylated NIV, another type B trichothecene, with a BSA-TMCS (1+1) reagent and also obtained two peaks due to incomplete derivatization after 20 minutes at 75'C. None of these type B trichothecenes had been tested by Ikediobi et al. (10). Bis(trimethylsily1)trifluoroacetamide (BSTFA), a reagent used by some researchers (51,52), did not give any fully derivatized DON, nor did a mixture with TMCS (3+2), even after heating at 100°C for 30 minutes (53). Variable results with BSA-TMCS (5+1) (Tri-Silo BT) and BSTFA-TMCS (5+1) for DON and NIV were also reported by Kientz and Verweij (54). Gilbert et al. (53) theorized that in type B trichothecenes it is the 7-hydroxyl group that is difficult to derivatize because of hydrogen bonding to the adjacent 8-carbonyl group. In support of this, tris-TMS NIV was shown to contain an unreacted 7-hydroxyl group based on nuclear magnetic spectroscopic evidence (50). However, reaction of DON with BSTFA and BSTFA-TMCS (4+1) formed two bis-TMS derivatives, identified by GC-MS, at least one of which must logically have had a derivatized 7-hydroxyl group (54). Tanaka et al. (49) were the first to show that trimethylsilylimidazole (TMSI) was a necessary ingredient for complete trimethylsilylation of type B trichothecenes; they chose TMSI-TMCS-pyridine (5+1+45) as reagent. A mixture of TMSI, TMCS and a suitable solvent has been the preferred reagent of a number of workers subsequently (55-65). Gilbert et al. (53) studied the optimum conditions for forming TMS derivatives of trichothecenes. Whereas TMSI alone brought about complete derivatization of DON, it was not a convenient reagent to use, causing damage to the capillary column unless removed by washing the reaction mixture with water (56). Regisil 323 (BSTFA-TMCS-TMSI, 3+2+3) gave 100% tris-TMS DON at room temperature (53). Tri-Sil. TBT is another commercial formulation, consisting of TMSI-BSA-TMCS (3+3+2), that readily derivatizes type B and other trichothecenes (66-75): Ohta et al. (76) used a reagent ratio of (5+5+1). Again it is preferable to wash the reaction mixture with an aqueous solution before GC (72). Trichothecolone, possessing only a 4-hydroxyl group, and T-2 have been trimethylsilylated with N-methyl-N-(trimethylsilyl) trifluoroacetamide (MSTFA) (77,78). TMS derivatization of macrocyclic trichothecenes - verrucarin A, roridins A and E, satratoxins G and H and baccharin B5 - was accomplished with BSTFA (at 90°C) by Rosen et al. (79). Except for roridin E and satratoxin G, they could be detected in the 1-10 ng range by GC-MS with selected ion monitoring (SIM) using a short (1 m) capillary column. A more general approach to detection of macrocyclic trichothecenes is alkaline hydrolysis to the parent alcohol verrucarol which is treated with BSTFA or Tri-Silo BT to form the TMS derivative and determined by GC-FID or GC-MS (80-82) (see also Section 9.2.2.3).
377
TMS ethers of trichothecenes are generally stable once formed (10,58,60). They can even be purified by thin layer chromatography (TLC) (83). TMS ethers of trichothecenes are usually detected by FID, ECD or MS (SIM) Kuroda et al. (60) were the first to show that ECD is more sensitive than FID for determining TMS ethers of both type A and B trichothecenes, but particularly so for the latter because of their conjugated 8-carbonyl group. A s little as 2-4 pg of DON, NIV or FX separated on a packed column could be detected by ECD, compared to 5-10 ng with FID: corresponding detectable quantities for the type A trichothecenes DAS, NS and T-2 were 200-400 pg by ECD and 30-60 ng by FID. In other laboratories, the superiority of ECD over FID for detecting trichothecene TMS derivatives (59) and the increased ECD sensitivity of TMS derivatives of type B compared to type A trichothecenes have been confirmed (59,84,85). Detection and determination of TMS derivatives of trichothecenes by MS, particularly after capillary GC, is the favoured technique if the instrumentation is available (86). The specificity of MS gives reliable identifications in grains, biological fluids and other matrices and it has been often used to confirm results obtained with other detection techniques. The EI mode is commonly employed with single or multiple ion monitoring (19,53,55,57,75,78,84,87-97). Examples of ions that may be monitored are m/z 512, 497, and 422 for DON TMS ether: 600, 510, 482, and 379 for NIV TMS ether: 480 for FX TMS ether: 436 for T-2 TMS ether: and 496, 481 and 406 for the TMS ether of the DON metabolite DOM-1. GC-MS of macrocyclic trichothecene TMS ethers has been previously mentioned (79). Full EI mass spectra of trichothecene TMS ethers have been published in several of these papers. The positive or negative ion CI MS modes have also been used for GC analysis of trichothecene TMS ethers (69,87,98-101). Characteristic ions are 513, 497 and 423 for DON TMS ether and 601, 585, 511 and 289 for NIV TMS ether in the positive ion (PI) CI mode: and 512, 305 and 297 for DON TMS ether and 600, 303, 298 and 297 for NIV TMS ether in the NICI mode (87,98,99,101). GC-tandem MS of T-2 TMS ether was carried out by Desjardins et al. (102). TMS derivatives of DON and NIV have been also identified by matrix isolation/Fourier transform infrared spectrometric analysis following capillary GC (103). Separation of TMS derivatives of several trichothecenes of both A and B types is readily achieved on a packed or capillary GC column (preferablytemperatureprogrammed) (19,49,51,56,59,60,84,85). Some variation of elution order for five type A and five type B trichothecene TMS derivatives with column polarity was noted by Scott and Kanhere (85) in a study using six different fused silica capillary columns. Generally, complete resolution was obtained, with a few exceptions: e.g. on DB-1701, 3- and 15-acetyl DON were not separated and FX, NIV and 15-monoacetoxyscirpenol (MAS) formed an almost unresolved peak. Separation on SE-30 is shown in Fig. 9.2.
.
378
SE- 30 TMS 3-ADON
0'
I
I
10
I
I
20
I
RETENTION TIME(MIN)
I
I
30
Fig. 9.2. Ca illary GC-ECD of trichothecene TMS ethers on SE-30; ca. 30 p g eaci of DON, 3-acetyl DON (3-ADON), 15-acetyl DON (15ADON) X and NIV and 1000 pg each of 15-MASI DAS, NS, T-2 and HT-2 injecked (85). 9.2.2.3 Hevtafluorobutvrvlation. In order to utilize the sensitivity of ECD, heptafluorobutyrate (HFB) derivatives are commonly used for determination of trichothecenes. The first application was with T-2 and DAS, which react very readily at room temperature with heptafluorobutyrylimidazole (HFBI) (21,104). Later, DON HFB was formed with this reagent by heating at 60'C for one hour (105) Other trichothecenes that have been derivatized with HFBI include HT-2, verrucarol, 4- and 15-acetoxyscirpendiol (monoacetoxyscirpenol, MAS), NS, NIV and FX (22,23,72,105-111). Reaction with type A trichothecenes proceeds at room temperature as indicated above for T-2 and DAS, while temperatures used for DON and other type B trichothecenes have ranged from 45' (112) to 110' (23). Luo et al. (108) found that l0O'C caused loss of NIV HFB. Reaction mixtures are washed with aqueous sodium bicarbonate (which may be followed by a water wash) , water (113) or phosphate buffer (pH 7.0)
.
379 (72) in order to remove excess reagent; Muszkat et al. (114) noted less interference after two washes with sodium bicarbonate solution. Heptafluorobutyrylation of trichothecenes is also carried out using heptafluorobutyric anhydride (HFBA) with 4-dimethylaminopyridine (4-DMAP) or trimethylamine as catalysts dissolved in an organic solvent (24,25,115). Again the reaction mixture is usually washed with aqueous sodium bicarbonate solution. Faster derivatization of DON at 6OoC is achieved with HFBA/4-DMAP than with HFBI (115). HT-2 and a demethylated analogue have been derivatized with HFBA at 6OoC without a catalyst (116). No catalyst was used by Muiioz et al. (117) to derivatize DON either although the extent of heptafluorobutyrylation was not indicated. Partial derivatization of DON and NIV could be observed in a recent report on the use of a polymer-bound 4-(N-benzyl-N-methylamino)pyridine solid catalyst with HFBA: DON bis-HFB and NIV tris-HFB were identified by GC-MS (118). The unreacted hydroxyl group was presumed to be the 7-hydroxyl group by analogy with acetylation studies on DON and 3-acetyl DON (119,120) and the similar slow heptafluorobutyrylation of 3,15-diacetyl DON (118). HFB derivatives of trichothecenes are determined by ECD or MS (SIM) Low picogram quantities can be detected by both techniques, even on a packed column. Using ECD, sensitivities are generally worse for derivatized DAS and T-2, which are later eluting and only contain one HFB grouping per molecule, than for derivatized DON and NIV, which elute early in the chromatograms and possess three and four HFB groups, respectively (85). Heptafluorobutyrylation is also advantageous for MS detection of trichothecenes because the high molecular weight of the HFB derivative offers greater specificity for GC-MS(S1M) than the TMS ether. For example, DON tris-HFB has a molecular ion at m/z 884 in the EI mass spectrum and the limit of detection is 1-3 pg on a packed column (105). Other packed column detection limits for GC-MS have been reported as 13-80 pg for DAS and HT-2 but about 1 ng for T-2 (monitored at m/z 602) (105). EI mass spectra of HFB derivatives of 12 trichothecenes have been published by Krishnamurthy et al. (23) , in addition to the PI- and NICI mass spectra. The latter mode of ionization gives very high sensitivity with the electronegative HFB groups and is about 5000 times more sensitive than PIC1 for DON HFB derivative (0.1 pg was measurable) (121). Minimum amounts of HFB derivatives detectable by capillary GC-NICI MS as reported by Krishnamurthy et al. (23) ranged from 0.1 pg (DON) to 2.0 pg (T-2) and confirmable limits were 1-5 pg using five or six ions. Macrocyclic trichothecenes (see also Section 9.2.2.2) were analyzed, after alkaline hydrolysis, by capillary GCNICI MS of the resulting verrucarols as their HFB derivatives (201000 pg) (107). Detailed studies on capillary GC-NICI tandem MS of HFB derivatives of 7 trichothecenes were reported by Kostiainen et al. (122) and Kostiainen and Rizzo (24); high selectivity and
.
380
sensitivity down to 0.1-2 pg were achieved by this MS detection technique also. HFB derivatives of trichothecenes are generally stable for several days, with the exception of FX (23), NS (123) and NIV (110); deterioration of HT-2 HFB was mentioned in an earlier publication (105) but was not a problem according to others (23, 110). Increased stability of DON HFB derivative was observed if silylated glassware was used (124). Double peaks with NIV and DON HFB have been encountered under certain conditions (22,72,110,111) as well as a shoulder on the DON HFB peak (110). The formation of two isomeric tetrakis-HFB derivatives of NIV has been shovn by GC-MS (72,125). Separation of all of several trichothecene HFB derivatives may not be complete on a packed column (126) but generally is on a capillary column (85,108,127), although 3- and 15-acetyl DON derivatives do not resolve on some phases (85). Elution order can vary according to column polarity; separation on DB-1701 is shown in Fig. 9.3 (85).
25 DON
DB-1701 HFB
MAS 15-ADON 3.ADON
HT.2 I
20
cl
x
z
15
u)
g 2a a
0 UJ w
10
'I
0
I
I
10
I
I
20
TIME (MIN)
I
I
30
1
Fig 9.3. Capillar GC-ECD of trichothecene HFB derivatives on DB1701; 20 p of eacg injected, except for DAS and T-2 (40 pg) (85). Peak markel MAS is for 15-MAS.
38 1
9.2.2.4 PentafluoroDroDionvl derivatives. Analogous to HFB derivatives, pentafluoropropionyl (PFP) derivatives of trichothecenes are formed with pentafluoropropionic anhydride and triethylamine or 4-DMAP (123,128) or with pentafluoropropionylimidazole (129). They have the advantage over HFB derivatives for MS detection using instruments with an upper mass limit of about m/z 1000 that the molecular ions are lower (
382
capillary GC was found to be exceptional (142). The analyst also needs to be aware of the possibility of chemical reaction in addition to trifluoroacetylation: two metabolic derivatives of T-2, designated TC-1 and TC-3, underwent dehydration in the hydroxylated 8-isovaleryl group with trifluoroacetic anhydride and each formed two isomers separable by capillary GC (132). This was one of the reasons why the TMS derivatives of 3I-hydroxyl metabolites of T-2 were preferred over the TFA derivatives in metabolism studies (143). An oxidation product of DON proposed for confirmation purposes was quantitated by GC-ECD as its TFA derivative (131). 9.2.2.6 Otherde Simple riv acetylation ativof es. trichothecenes for GC with acetic anhydride in the presence of pyridine has been reported (49,144,145). Detection was by FID, MS, or a photoionization detector. One disadvantage of acetylation is that in a series such as NIV, FX and diacetyl NIV, they all form the same compound, namely tetra-acetyl NIV in this case. Perfluorobenzoyl and pentafluorodimethylsilyl derivatives were briefly investigated by Begley et al. (129) but volatility was too low for the limits of column temperature. 9.2.3
Methods for arains. arain foods and feeds m n . An assessment of quantitative methods for determination of trichothecenes in grains and grain products published in 1982 (86) concluded that GC, with ECD or MS detection, was the best technique for quantitating trichothecenes. Eight methods, of which 6 used packed column GC and one was a capillary GC method, met criteria of 270% recovery (with 130% coefficient of variation, if given) and reported detection limits below 100 ng/g for at least one of the trichothecenes studied. Since 1982, a large number of GC methods for trichothecenes has been published and the proportion of packed column to capillary column methods has markedly decreased. Nevertheless, packed columns have proved especially useful for determination of DON in grains. The AOAC Official Method for GC determination of DON in wheat uses a packed column (3% OV-101) (146) The method involves extraction with chloroformethanol (8+2), chromatographic cleanup on a column of small particle size silica gel, and determination of the HFB derivative of DON by GC-ECD. Within the authors’ laboratory, recoveries of DON added to wheat at levels of 118-1184 ng/g averaged 88% with a coefficient of variation of 8.6% (115). The method is the only GC method for trichothecenes that has been collaboratively studied in several laboratories: the average recovery was 92%, and the mean repeatability (variation within laboratories) and reproducibility (variation among laboratories) were respectively 32 and 41% for spiked samples and 31 and 48% for naturally contaminated samples. The method was found to be applicable at 2350 ng/g (146). Several of the more useful principles for extraction of samples and their cleanup were established using packed column GC for the determinative step. A study of the extraction efficiency for DON, 9.2.3.1
.
383
NIV, FX and 3-acetyl DON in artificially moldy rice using acetonitrile, methanol, and aqueous mixtures of these solvents concluded that acetonitrile-water (3+1) was the most efficient (91,147). Methanol-water (3+1) recovered only 75-88% of the trichothecenes by comparison with this extraction solvent. Use of water to extract DON from wheat, corn and feed gave recoveries of >85% for spiked samples in the range 50-1000 ng/g and compared well with the method of Scott et al. (105) when tested in three laboratories on a naturally contaminated wheat sample (148). The method of Scott et al. (105) served well to analyse wheat for DON when this trichothecene was first found in Canadian wheat and has been applied to various grain foods also (126). It employed GC-ECD on a 3% OV-3 column with GC-MS (SIM) for confirmation of DON HFB derivative; recoveries from wheat averaged 72 and 80% in two different laboratories with coefficients of variation of 10%. Good recoveries of NIV from cereal products were also achieved when partition from aqueous methanol into ethyl acetate was carried out on a hydrophilic solid matrix rather than in a separatory funnel, a procedure that shortened the method (72). An important advance in cleanup of grain and feed extracts was the introduction of a small charcoal/alumina column and its application to packed column GC analysis of type B trichothecenes as their HFB derivatives (149). Originally, normal silica gel column chromatography was one of the most commonly used cleanup procedures for packed column GC of trichothecenes (21,62,67,72,83,105,150-152). This became replaced by the silica Sep-Pake cartridge (78,91,112,148,153). Florisile (55,59,60,62,65,83,91,95,147), Amberlitee XAD-4 or XAD-2 (59,60,65) and C,, (73,75) are other cleanup column materials that have been employed. The combination of a Florisil. column and a silica Sep-Pak. cartridge has been successfully used for cleanup of grain extracts for packed column GC by Tanaka et al. (91) in surveys for DON and NIV in samples from 19 countries (94). Detection limits of this method were 2 ng/g for DON and NIV with recoveries averaging 87 and 86%, respectively, from spiked wheat, corn and polished rice (91). Recently, the silica cartridge was omitted in analyses of Dutch cereals (95). In all these surveys, DON and NIV were determined as their TMS ethers by ECD with confirmation by GC-MS (SIM), using a 1.5% OV-17 column phase. The studies of Tanaka et al. (94,95) represent the most sustained application of packed column GC to analysis of grains for trichothecenes. Other use of packed column GC for survey work has been referred to in reviews (18,154) and additional papers (155157). It has also found considerable application in studies on the effects of milling and food processing on trichothecenes and on detoxification of trichothecenes in grains (158). Interferences can give rise to problems of misidentification of trichothecenes by packed column GC. Visconti and Palmisano (159) noted a peak that interfered with DON TMS ether on a GC column of 3%
384
OV-1 when cultures of Fusarium solani on corn were analysed by FID. Also Mirocha et al. (160) reported two substances in feed extracts, identified as 1-glyceryl monooleate and linoleate, whose TMS ethers (detected by FID) were not resolved from T-2 TMS ether on 3% OV-1; TFA and PFP derivatives of T-2 and the monoglycerides were well resolved. Scott et al. (161) observed that HFB derivatization of extracts of pre-harvest wheat gave 30% higher DON levels than TMS derivatization but 36% lower levels for wheat at harvest. These examples show the need for GC-MS confirmation, particularly for packed column GC. 9.2.3.2 CaDillarv column. For multiple trichothecene determination, capillary GC is essential, particularly with ECD. Many laboratories worldwide have used capillary GC, with ECD or M S detection, for determination of three or more trichothecenes directly in grains, grain foods, and feeds (22,24,28,32,42,51,69, 84,88,106,109,110,122,127,162-166). Capillary GC has also been used extensively for determination of just one or two trichothecenes (usually DON and NIV) in these commodities (30,57,61,68,87,106,111, 113,117,124,167-171). These are direct determinations of individual trichothecenes, usually as their HFB esters or TMS ethers with ECD or MS detection. An alternative method is to carry out alkaline hydrolysis of purified extracts and standards to parent trichothecene alcohols (T-2 tetraol, scirpentriol, DON and NIV), which can then be determined by capillary GC-ECD of their PFP or TFA derivatives (128,172) or capillary GC-FID of their TMS ethers (173) For macrocyclic trichothecenes the parent alcohol is usually verrucarol (80). This approach offers a greater analytical sensitivity, as all trichothecenes within a given group are converted to one compound, the parent alcohol, and the PFP derivatives of the parent alcohols have greater response factors by ECD and NICI MS than the original esterified trichothecenes (128); it is, of course, only a screening method for groups of trichothecenes. The preferred GC method for trichothecene determination in grains and related foodstuffs is capillary GC-MS(S1M) and for unequivocal identification a complete mass spectrum is essential (98,99). Unfortunately, MS instrumentation may not always be available for dedicated trichothecene analysis and is often reserved for confirmation purposes only, although one solution may be the mass selective detector or I1benchtop1@ mass spectrometer (96,127,162,165). Thus an important goal of trichothecene GC methodology for grains is to determine trichothecenes by ECD with a minimum incidence of false positives due to interferences, using GC-MS for confirmation of identity. The efficacy of the method can only be assessed after field use, as for example was reported by Scott et al. (110) concerning a capillary GC-ECD method for DON, NIV, T-2, HT-2 and DAS in wheat. The keys to attaining low levels of interferences are good cleanup and choice of derivative. These will be discussed below.
.
385
Small cleanup columns continue to be extensively used in capillary GC methods for grains and feeds. Charcoal is either mixed with alumina only (84,109,110,174) as in the original procedure of Romer (149), with alumina and Celite. (57,69,103,128), with alumina and capped with cation exchange resin (175) or used alone (163). The eluting solvent is acetonitrile-water (generally 84+16) and no recovery problems of trichothecenes, even the more polar NIV, have been reported in the above-mentioned papers. Scott et al. (110) noted that an alumina-carbon cleanup column gave cleaner capillary chromatograms than silica gel for trichothecenes in wheat determined as their HFB derivatives, although an interference for NIV HFB was still present. A s mentioned under Section 9.2.3.1, the silica Sep-Pak. or similar cartridge is a useful cleanup column and it has also been incorporated into capillary GC methods (106,128,167,170). Florisil. cartridges gave improved cleanup over silica cartridges for capillary GC-ECD (22). However, a high polarity solvent such as chloroform-methanol (7+3) or ethyl acetate-acetone (4+1) was necessary to give good recoveries of the more polar trichothecenes NIV and T-2 tetraol (22,128). Variable recovery of trichothecenes from different lots of Florisil. used in a mini-column cleanup procedure was noted by Rood et al. (128) who did not recommend activated Florisil.. A C,, Sep-Pak. or similar cartridge (22,69,84 ,103) and a cyano cartridge (106,167) have been employed in conjunction with other cleanup procedures. Conventional Florisil. (24,84,87,111,122)and silica gel (32,51,110,124)columns containing 2-25 g adsorbent remain useful cleanup procedures in capillary GC methods, even for sophisticated GC-MS methods. Other cleanup procedures that have been used include preparative TLC (30,164), preparative LC (113,176) and gel permeation chromatography (165,167). Which derivative to use for analysis of grains for trichothecenes usually depends on the presence of any interferences and whether just type B or both type A and type B trichothecenes are to be determined. The first factor is influenced by the combination of cleanup procedures used and the polarity of the capillary column (separation of standard trichothecenes themselves is not a problem in most cases). For determination of DON or DON and NIV only, the TMS derivatives are often preferred as there may be fewer interferences and GC-ECD sensitivity is good, e.g. limits of determination of 12 ng/g for DON and NIV in wheat and barley after two column cleanup steps (57) and 1 ng/g for DON and NIV in feeds after three column cleanup steps (84) have been reported. Fused silica capillary columns used in these two studies were 5% phenylmethylsilicone and OV-1, respectively. Detection limits for DON and NIV TMS ethers in grain foods have been found to depend on the mode of injection (168). A s expected, higher detection limits are seen with type A trichothecenes determined as TMS ethers by ECD (84). Therefore, fluorinated derivatives such as HFB's have been explored for multitrichothecene GC-ECD determination (types A and B)
386 (108-110,163). Still, detection limits for type A trichothecenes are of the order of 2 0 0 ng/g using only one chromatographic cleanup column step ( 1 1 0 , 1 6 3 , 1 7 0 ) and somewhat lower ( 5 0 - 1 0 0 ng/g) for a method employing two chromatographic cleanup procedures ( 1 0 6 ) . Surprisingly, capillary GC-FID can offer a similar detection limit for T-2 and other type A trichothecenes a6 shown by Eller and Sobolev ( 4 2 ) , who determined them as TFA derivatives on an OV-101 column. Capillary GC-MS (SIM) offers at least equivalent detection limits for grain and feed analysis. DON and NIV could be detected in concentrations of 1 0 ng/g in cereals by GC-NICI MS of their TMS ethers using a DB-5 fused silica capillary column ( 8 7 ) . Hussein et al. ( 8 8 ) reported limits of detection for DON, DAS and T-2 in the range 5-10 ng/g in a survey of New Zealand corn using GC-MS (SIM) (EI mode) of the TMS ethers separated on a capillary column of BP1 (cross linked methyl silicone). For HFB derivatives: capillary GC-NICI MS could determine scirpentriol, 15-MAS and NIV naturally occurring in sorghum in the range 3-70 ng/g, with monitoring of 3-6 ions for confirmation ( 2 2 ) : DON was confirmed in barley down to 1 3 ng/g by GC-EI MS on a CP-Sil 5 (methly silicone) glass capillary column ( 1 2 4 ) : and T-2, DAS, 15-MAS and DON could be detected down to 1 ng/g in spiked wheat samples by GC-NICI tandem MS ( 1 2 2 ) . One disadvantage of using HFB derivatives is poor sensitivity for NS, which appears to be unstable in the GC-MS system ( 2 2 ) . Kostiainen and Nokelainen (137) described the use of nalkylbis(trifluoromethy1)phosphine sulfides as retention index standards in the identification of several trichothecenes in porridge flakes by capillary GC-EI MS of their TFA derivatives. Detection limits for the trichothecenes were 5-50 ng/g. Recoveries of trichothecenes added to grains, grain foods, and feeds and determined by capillary GC have generally been reported as good. Detailed results are too numerous to list here. In some cases, recoveries below 7 0 % have been observed: e.g. 6 5 and 5 7 % for DON in corn and mixed feeds, respectively, and 5 6 % for T-2 in barley ( 1 0 6 ) : and 41-55% for NIV in sorghum ( 2 2 ) . Analysts should be aware that trichothecenes are unstable in cereals and feeds over a several months' storage period ( 8 4 ) . Application of capillary GC methods to worldwide surveys of grains, grain foods and feeds for trichothecenes has been significant (18,68,69,88,110,117,154,162,169~175,177-179). As use of capillary GC expands so will our confidence in the levels and incidences of trichothecenes reported to occur naturally in these commodities, particularly if GC-MS is used for determination and confirmation. Capillary GC has also been applied in a few studies on the effects of food processing on trichothecenes as in tortilla making ( 1 8 0 ) , bread baking ( 1 8 1 ) and processing and cooking of spaghetti and noodles ( 1 8 2 ) : to date nearly all studies in this area ( 1 5 8 ) have made use of other analytical procedures, particularly packed column GC.
387
Methods for biolosical fluids GC has been extensively applied to analysis of trichothecenes in biological fluids in the following main areas: metabolism studies in animals, particularly farm animals (183); diagnostic screening in livestock to determine exposure to trichothecenes from feed (135); carryover of trichothecenes and their metabolites into milk (47,90,97 ,184-187) : and assessment of human exposure to trichothecenes suspected of use as chemical warfare agents ("yellow rain") (31). The latter provided considerable stimulus to the development of methods for trichothecenes in biological fluids such as urine and blood, and highlighted necessary aspects of quality assurance in these analyses such as frequent use of blank samples, spiked controls, possible toxin degradation during sample handling and confirmation of results by other laboratories (31,188). GC analysis of biological fluids uses the same techniques as outlined previously (Section 9.2.2): GC-MS is preferred for these determinations. The analyses are complicated by the increased number of compounds that may be present due to metabolism, including deepoxy derivatives (183,189). Thus the principle of hydrolysing purified extracts of plasma and urine to parent alcohols is a valuable way of indicating exposure of animals to trichothecenes. Rood et al. (135) quantitated T-2 tetraol, scirpentriol, DON and NIV, together with the deepoxy derivatives of the first three parent alcohols, as their TFA derivatives by GC-ECD with GC-NICI MS confirmation; the detection limit was below 25 ng/ml. Another difference between analysis of grains and animal fluids such as urine, bile and milk is that the trichothecenes may be largely present as glucuronide conjugates, which are not extracted unless the fluid is first incubated with p-glucuronidase (135,185,186,190-193). Neither this enzyme pretreatment nor alkaline hydrolysis to parent alcohols has been applied to analysis of human urine samples for trichothecenes (125,194). The analytical scheme for analysis of biological fluids may be simplified compared to grain analysis in that solid phase extraction on a C,, cartridge or mini-column (20,47,135,186,195,196), absorption onto a hydrophilic matrix such as a Clin E l u P (Chem Elutn) tube (31,125,134,138,193,197-200) or adsorption onto a resin such as Amberlitem XAD-2 (136,194) can be used to replace direct extraction with an organic solvent, although ethyl acetate extraction has been used by some researchers (25,140,184,188). Others have added a water-miscible solvent such as methanol, acetonitrile or acetone (for blood or plasma) before proceeding with cleanup step(s) 9.2.4
(20,29,89,122,129,186,187,194,197,198).
Types of cleanup for blood, serum, urine, bile and milk are similar to those used for grains and include common use of a C,, SepPak. cartridge or mini-column (20,25,29,89,125,129,135,136,183, A silica cartridge (138,199), Florisil. mini-column 194,200). (135,195,201) or charcoal-alumina column (31,186,202) are also among the main cleanup procedures that have been used.
388
The most sensitive method for determination of trichothecenes in blood was developed in response to the Ityellow rain" situation. An outline of this GC-MS (SIM) method, described as "very arduoustt, was given by Reutter et al. (188) who quoted detection limits (signal-to-noise ratio 25:l) for human blood of 0.01 ng/ml (T-2 tetraol) to 0.5 ng/ml (T-2, HT-2) and verification limits of 0.1 to 4 ng/ml. A well-planned quality assurance program was necessary and proved that the method was accurate and reliable. Several of the other published methods for blood or plasma have reported detection limits of the order of 1 ng/ml or lower. Begley et al. (129) used capillary GC of PFP derivatives and NICI MS detection to achieve detection limits for human blood (signal-to-noise ratio 5:l) of <0.1 ng/ml (DON) and 0.1 to 2 ng/ml for other trichothecenes except T-2 (10 ng/ml), but recoveries were only 20-50%. Using underivatized T-2 and HT-2, sensitivities of 1 ng/ml plasma were obtained by Greenhalgh and Ashley (31) by capillary GC-NICI MS, but again recoveries were low (25 and 50%) at 10 ng/ml spiking levels. Detection limits of 0.7 ng/g and 3.6 ng/g in human blood were estimated for underivatized T-2 and DAS, respectively, by a capillary GC-ammonia CI MS method (29); recoveries continued to be low - 50 and 30%. Application of capillary GC-tandem MS (EI or NICI modes) to analysis of human plasma and blood allowed detection limits of about 1 ng/g for T-2, HT-2, DAS, 15-MAS and DON as TFA or HFB derivatives (20,122); recoveries of 70-80% for T-2 and 90-140% for HT-2 were reported (20). Prelusky et al. (186) also reported good recovery (98%) of DON from (bovine) serum with a detection limit of 1 ng/ml by a GC-MS method. Instability of HT-2 and T-2 in blood and plasma has been observed by Begley et al. (129) and Greenhalgh and Ashley (31), which may account for recovery problems experienced by some workers. The most sensitive method for determination of trichothecenes in urine (125) employed capillary GC-MS (EI and NICI) and had limits of detection for eight trichothecenes (as HFB derivatives) varying from 1 ng/ml for T-2 tetraol to 2-5 ng/ml for T-2, based on signalto-noise ratios of s3:l for the most intense ions (EI). Recoveries ranged from 36% for T-2 tetraol to 84% for 15-MAS. There has been some interest in whether trichothecenes could carry over from feed to milk of cows and sheep (97,184,186,187,192, 193,200). A method for analysing T-2 in milk by GC-MS (EI or CI) had a detection limit of 3-6 ng/g after TLC cleanup (184). DON and its deepoxy metabolite DOM-1 have been determined in milk as TMS derivatives by GC-ECD with a detection limit of 1 ng/ml and recoveries of 82 and 85%, respectively (200). HFB derivatives of milk extracts were more complex than TMS derivatives of the same extracts, and with frozen milk samples interferences at the retention time of DON HFB were occasionally observed. This problem was overcome by using GC-MS (SIM) with multiple ions; a detection limit of 1 ng/ml (signal-to-noise ratio 3:l) was found for DON in cow's milk (186). Only trace levels of DON (<4 ng/ml) were
389
transmitted to cow's milk from an oral dose of 920 mg DON (186). Transmission of DON to milk of pigs (gilts) has also been studied by GC (203). Although not strictly speaking a biological fluid, feces has also been analysed by GC for trichothecenes and their metabolites in various metabolism and transmission studies with chickens, cows, rabbits and rats (89,198,199,204-209). 9.2.5 Methods for animal tissues Food safety authorities are concerned over transmission of trichothecenes from feeds to food-producing animals and GC methods have been able to measure the extent of possible occurrence of residues. Methods must be able to detect low ng/g levels. El-Banna et al. (210) found that acetonitrile extraction of chicken eggs or meat, cleanup on an alumina-charcoal column and packed column GC-ECD after HFB derivatization allowed detection of as little as 10 ng DON/g while for GC-MS(S1M) the detection limit was 5 ng/g: recoveries were 78-106%. DON residues in chickens were not detected in this or other transmission studies using the same method (204,211), although at a very high level of DON in feed (83 pg/g) about 20 ng/g of DON was detected in the gizzard (204). Coppock et silica and C,, cartridges al. (174) added additional cleanup steps - in a packed column GC-ECD method for DON in swine tissue that had a detection limit of 20 ng/g and recovery of 90%. Another method for determination of DON in animal tissues (liver and kidney) at low ng/g levels used C,8 Sep-Pak and LC cleanup steps with capillary GC of the HFB derivative (212). Metabolites of T-2 found in chicken liver 18 hours after intraperitoneal injection were 3,-hydroxy-HT-2, HT-2, T-2 trio1 and four others, which were determined and characterized by GC-MS of their TMS and TFA derivatives (206). Traces of T-2 (in the heart) and HT-2 (in the lung and kidney) were detected as TFA derivatives by capillary GC-PIC1 MS in tissues of a cow orally dosed with T-2: cleanup of an acetonitrile extract was on an XAD-2 column (197). Beasley et al. (195) described a method for T-2 in swine tissues with average recoveries of 105% and a limit of reliable quantitation of 40 ng/g. Cleanup included a lead acetate treatment of the acetone extract and a Florisil. column: GC of T-2 HFB was on a packed column with ECD. DAS was determined in tissues of pigs and calves by Coppock et al. (213). The method of analysis included cleanup on magnesium silicate and Biobeads. SX-3 columns, with capillary GC of DAS HFB. Recoveries averaged 83% and the detection limit was 10 ng/g. Surprisingly high levels of DAS and HT-2 (2.5 pg/g and up to 4.0 pg/g, respectively) have been found in body tissues of humans who were victims of chemical attack in Southeast Asia (194); analyses of TFA derivatives were performed by GC-PIC1 MS. 9.2.6 Methods for other foods In general, methods for trichothecenes in other agricultural products have not been developed specifically for that commodity.
-
390
Thus packed column GC was applied to extracts of soybean, red bean and buckwheat flour obtained by the same extraction and cleanup procedures used for grains in order to determine FX, DAS and T-2 as their TMS ethers ( 8 3 ) ; recoveries ( 6 6 - 8 6 % ) and detection limits ( 1 0 7 0 ng/g) were similar to those for grains. Furthermore, T-2 and DAS have been detected using packed column GC as natural contaminants in peanuts by the same method used for corn and sorghum ( 1 5 5 ) ; natural occurrence of NIV and DON in the Japanese health food Job's-tears has been shown by a packed column GC-ECD method, with GC-MS confirmation, which was the same as that used for analysis of various grain foods ( 9 2 ) ; and 68-70% recovery of NIV and DON from soybeans was reported by Isohata et al. ( 1 5 6 ) by a packed column GCECD method for grains, which was also applied to various baby foods and fruit juices. Soybeans and soybean meal have been analysed by known GC methods for DON and T-2 with confirmation of DON by capillary GC-MS ( 2 1 4 ) . Soybean seeds were also analysed by Clear et al. ( 2 1 5 ) for DON, DAS, T-2 and HT-2 following the extraction procedure described (146) with an additional gel permeation by Ware et al. chromatographic cleanup step and determination of the HFB derivatives by capillary GC-MS(S1M). Neither of the foregoing reports on soybeans included recovery studies. A capillary GC-ECD method for 6 trichothecenes in both wheat and soy flour ( 1 6 3 ) had average recoveries of 8 3 - 1 0 5 % for both commodities but interferences were noted for NIV and DAS in soy flour: detection limits ranged from 2 0 ng/g (for DON) to 700 ng/g for DAS. A capillary GC-ECD screening procedure that included soybeans as a substrate involved alkaline hydrolysis of trichothecenes in extracts cleaned up on charcoal-alumina-Celite. and Florisil. mini-columns to the parent alcohols, scirpentriol, T-2 tetraol, DON and NIV, which were determined as their PFP derivatives; recoveries of DAS, T-2 and DON from soybeans averaged 8 4 , 8 8 and 6 5 % respectively, with a detection limit of 2 0 ng/g ( 1 2 8 ) . Potato is one foodstuff that gives rise to interferences when trichothecene HFB derivatives are determined by capillary GC-ECD (216). In our laboratory we have found that aqueous acetonitrile extraction gives rise to less interference. Ethyl acetate was the extraction solvent used by Desjardins and Plattner ( 2 1 7 ) , with cleanup on a charcoal column followed by GC-MS of the TMS derivatives; 1 0 0 % recovery of DAS from potato tuber was achieved by this method. GC-MS(S1M) of trichothecene HFB derivatives and GC-MS/MS of underivatized trichothecenes were compared for 4 8 inoculated potato tubers by El-Banna et al. ( 2 1 6 ) with generally good agreement between the two techniques, although three false positives for DON were observed with the MS/MS system; overall method recovery (determined by GC-ECD) was in the range of 6 9 - 1 0 9 % for DAS, T-2, HT-2, DON, acetyl DON and FX but only 2 0 % for NIV. GC-FID of underivatized DON, DAS and T-2 has been applied to detect these trichothecenes in beer at levels of < 1 0 to 4 2 ng/ml
39 1 (218). Cleanup of a chloroform extract of beer plus added methanol was on a column of basic alumina. Capillary GC-FID methods have been developed for determination of underivatized trichothecin and trichothecolone in grape juice and wine with a detection limit of 50 ng/ml (44,45). Using a C18solid phase extraction tube, mean recoveries of trichothecin were 77 and 93% from grape juice and wine, respectively, but a separatory funnel extraction procedure was necessary to achieve reasonable recoveries of trichothecolone (65 and 7 5 % ) (45). Black et al. (22) analysed pollen and honey for various trichothecenes after alkaline hydrolysis to T-2 tetraol, scirpentriol, NIV, DON and verrucarol, which were detected by capillary GC-ECD and GC-MS(S1M) of their HFB derivatives. 9.2.7 Add itional a m 1ications GC has been applied to in vitro metabolism studies involving microorganisms (including rumenmicroorganisms) (90,100,219,220)and enzyme preparations and tissue fractions (76,143,191,221,222). Only papers giving information on GC separations or mass spectral information on derivatized metabolites are referenced here as examples. A further important application of GC is the identification and determination of trichothecenes in fungal cultures both solid and liquid. These papers are too numerous to fully review here but they often contain useful information on GC of trichothecenes. A s recent examples of such applications: retention times for the TMS ethers of the Fusarium sambucinum metabolites scirpentriol and all seven acetylated derivatives on packed and capillary columns were given by Richardson et al. (43); a variety of trichothecenes, some of which were new compounds, were characterized either without derivatization or as TFA derivatives by capillary GC-MS in cultures of E. sDorotrichioides (36,133); 17 trichothecenes were identified in liquid cultures of E. tricinctum by GC-MS of their TFA derivatives (223); isoverrucarol was characterized in E. OxvsDorum cultures by GC-MS of its TMS derivative (224); and distribution of DON and 15-acetyl DON in corn ears inoculated with E. sraminearum and grown in a controlled-environment facility was determined by packed column GC-ECD (225). Desjardins et al. (102) demonstrated the use of GC-tandem MS to determine ”0 labeling in T-2 during biosynthetic studies with Fusarium sDorotrichioides grown in the presence of HZ1’O and “Oz. An interesting application of GC to test for macrocyclic trichothecenes was the analysis of a sample of ceiling fiber board contaminated with Stachvbotrvs atra spores from a house where the inhabitants had chronic health problems (226). A 50% methanol-water extract was hydrolyzed, the hydrolysate cleaned up on a silica gel column and verrucarol identified by GC-MS.
-
392 9 . 3 . ZEARALENONE 9.3.1 Derivatization
and detection Drocedures for zearalenone and related metabolites Although LC is considered the usual choice for determination of zearalenone ( 2 2 7 ) , applications of GC to its determination and confirmation, particularly in studies of its natural occurrence, mycology and metabolism. are numerous. Most GC methods for zearalenone (ZEN) and =- and p-zearalenol (ZEL) (Fig. 9 . 4 ) make use of trimethylsilylation. Kientz and Verweij ( 5 4 ) studied the activity of several silylating agents on ZEN. BSTFA and BSTFA-TMCS ( 4 + 1 ) derivatized ZEN at room temperature within 3 0 minutes. This ready reaction with BSTFA, either at room temperature or after 1 0 - 1 5 minutes at 6OoC, had previously been observed by others (19,51,173,228-232). Suzuki et al. ( 2 3 0 ) studied the reaction in detail and proved by GC-MS that complete derivatization of ZEN
HO ‘-OH
I
II
Fig. 9 . 4 . Structures of zearalenone (I) and =-zearalenol (11): Pzearalenol has the opposite configuration for the secondary hydroxyl group. occurred, as did Ueno and Tashiro ( 2 3 1 ) for =- and p-ZEL. Thouvenot and Morfin ( 2 3 3 ) used BSTFA-TMCS ( 9 9 + 1 ) . BSA-TMCS ( 5 + 1 ) (Tri-Silo BT) also gave satisfactory derivatization of ZEN ( 5 4 , 1 7 3 , 2 3 4 - 2 3 9 ) . However, if TMSI were included in the reagent mixture, as in TriSil. TBT, loss of response was observed according to Kientz and Verweij ( 5 4 ) . This discouraging finding does not support satisfactory experience of others with the Tri-Silo TBT reagent for derivatization of ZEN, --ZEL and p-ZEL ( 7 0 , 1 6 6 , 2 4 0 , 2 4 1 ) . Other reagents that have been used for trimethylsilylation of ZEN and = and p-ZEL are BSA ( 1 6 4 , 2 4 2 , 2 4 3 ) , MSTFA ( 2 3 6 , 2 3 9 , 2 4 4 ) , TMSI-TMCSethyl acetate ( 5 + 1 + 4 5 ) ( 5 9 ) and HMDS-TMCS ( 9 + 1 in chloroform or 2 + 1 in pyridine) ( 1 , 2 3 9 ) . As indicated above by Suzuki et al. ( 2 3 0 ) and others ( 2 3 6 , 2 3 8 , 2 4 3 , 2 4 5 ) , TMS derivatization of ZEN goes to completion with formation of bis-TMS ZEN. Zearalenone TMS derivative is usually detected by FID, with a sensitivity limit of below 0.5 ng injected onto a packed column ( 2 3 9 ) ; Thouvenot and Morfin ( 2 3 3 ) reported a detection limit of 1
393
ng, using a capillary column. By GC-MS (SIM), 1 0 - 2 0 pg amounts of ZEN TMS ether were detectable monitoring the ions m/z 4 6 2 and 4 2 9 ( 1 9 ) . ECD is not useful for detection of ZEN TMS ether ( 5 9 ) . Since ZEN is often analysed at the same time as trichothecenes, it is worth noting that its TMS ether is retained more than any of the commonly looked-for trichothecenes (DON, NIV, DAS, NS, HT-2 and T-2) on packed columns of 2 % OV-17 ( 5 9 ) , and mixed 4 % OV-1 + 3% OV-17 + 4 % OV-101 ( 2 2 9 ) , as well as on capillary columns of BP1 ( 8 8 ) , CP-Sil 8 CB (cross linked phenyl methyl silicone) ( 5 4 ) , SE-52 ( 5 1 , 5 2 , 1 7 3 , 2 3 2 ) , OV-17 ( 8 2 , 1 6 6 ) , OV-1 ( 1 9 ) and OV-101 ( 1 6 6 ) . The elution order ZEN, =-ZEL and p-ZEL was observed for their TMS derivatives on capillary columns of DB-5 ( 2 3 8 , 2 4 0 ) and DB-1 ( 2 4 3 ) while the order =-ZEL, p-ZEL and ZEN was recorded for a packed 3 % QF-1 column ( 2 3 1 , 2 3 5 , 2 3 8 ) ; =- and p-ZEL were not separated on 3% OV1 , 3 % OV-17 or 3% OV-101 packed columns ( 2 3 5 , 2 4 6 ) . Cis- and transZEN have been separated as TMS ethers by GC, as have cis- and trans isomers of =-ZEL and P-ZEL ( 2 3 7 , 2 3 8 ) . The derivatives dimethoxyzearalenone, formed by a methylation with methanol-trimethylanilinium hydroxide in the injector ( 2 3 6 ) , and methyloxime-bis-TMS ZEN, formed with methoxyamine hydrochloride then trimethylsilylation ( 2 3 3 , 2 3 6 ) , have been prepared for confirmation of identity of ZEN and related compounds ( 2 4 7 ) . Two peaks, stated to correspond to syn and anti epimers, were observed with the latter derivative by capillary GC ( 2 3 3 ) . The TFA derivative of ZEN was formed with trifluoroacetic anhydride-sodium bicarbonate but two peaks resulted which were identified by GC-MS as cis and trans forms of bis-TFA ZEN and were prone to decomposition ( 5 4 ) ; unsuitability of the TFA derivative for quantitation of ZEN because of instability was also noted by Wreford and Shaw ( 1 3 9 ) . The PFP derivative of ZEN, formed by reaction of ZEN with pentafluoropropionic anhydride and trimethylamine, was determined by GC-ECD with good sensitivity ( 5 0 0 pg ZEN gave a moderately large Surprisingly peak after separation on a packed column) ( 2 4 8 ) . little use has been made of the HFB derivative, which has been prepared by using HFBI or HFBA ( 1 0 8 , 2 4 9 ) , and of which as little as 5 0 pg could be detected by ECD ( 1 0 8 ) . In our laboratory, heptafluorobutyrylation of ZEN with HFBI was shown by GC-MS to result in both mono- and bis-HFB derivatives. Underivatized ZEN has been detected by GC-MS ( 2 5 0 ) and GC-MS/MS (34) * 9.3.2
Methods for mains. m a i n foods and feeds Several GC methods have been published for determination of ZEN in grains and feeds. Although ZEN (then referred to as F-2) was identified in corn and hay by GC in the late 1 9 6 0 ’ s ( 1 , 2 4 2 ) , the first detailed GC method was that of Mirocha et al. ( 2 3 6 ) , who employed alkaline or acetonitrile-petroleum ether cleanup procedures and formation of the TMS ether for determination. The limit of detection was < 5 0 ng/g by FID and < 1 0 ng/g by MS(S1M) with
394
recoveries averaging 84% from corn by the base cleanup procedure. Dimethoxy and methyloxime-bis-TMS derivatives were used for GC confirmation. This method was collaboratively studied by Thaler (251) for corn and mixed feed with good results. Scott et al. (244) incorporated a silica gel minicolumn cleanup in an LC-TLC-GC method for corn-based foods that had a detection limit of 0.1 ng/g using GC-MS(S1M) of ZEN TMS ether. Comparison of chromatograms of ZEN methyloxime and ZEN TMS derivatives confirmed identity of ZEN in a capillary GC-FID method for corn that had a detection limit of 100 ng/g (233). Other GC-FID methods (both using packed columns) for determination of ZEN in grains as its TMS ether were reported by Jemmali et al. (252), Sugimoto et al. (253) and Suzuki et al. (230). The second method claimed a detection limit of 50 ng/g in corn or milo and 90-97% recoveries of ZEN while the latter method recovered 72-88% from cereals. Holder et al. (248) found that GC-ECD determination of ZEN in animal chow extracts as its PFP derivative was not satisfactory below 1 pg/g because of interferences. However, Luo et al. (249) determined ZEN in grains as its HFB derivative by GC-ECD with a detection limit of <20 ng/g and recoveries above 80%. In TLC or LC methods for determination of ZEN in grains and in studies on its natural occurrence in grains by these other procedures, GC-MS has been used for confirmation of identity of ZEN (244,254,255). GC-FID has also been employed for confirmation purposes (256,257). Determination of ZEN in grains is often made in conjunction with analysis for trichothecenes. In some cases the fusariotoxins have been simultaneously determined by GC in one injection. This approach necessitates a common derivative, the TMS ether, except where the HFB (108) or no derivative (34) has been used, and also a common means of detection, FID (51,59,166,232,258), GC-MS (70,95,259), GC-ECD (108) or GC-MS/MS (34). Uniform cleanup procedures for both trichothecenes and ZEN were used by Hungarian workers (51,166,232). Overall average recoveries from cereals of 78-90% for ZEN have been reported (51,59,232) and limits of detection for ZEN in corn, barley and wheat were 100 ng/g by the multi-fusariotoxin method of Kamimura et al. (59). In other multifusariotoxin methods, where ZEN has been determined by a chromatographic technique other than GC, usually LC, GC-MS of the TMS ether is the preferred choice for confirmation by a complete mass spectrum, total ion chromatography, multiple ion monitoring or single ion monitoring at m/z 462 (88,93-95,180,260-264). It is quite possible to carry out GC-MS confirmation of both ZEN and trichothecenes in one injection of the TMS ethers (88). Natural occurrences of --ZEL in animal feed and =- and $-ZEL in Fusariuminfected corn stalks have also been confirmed by GC-MS of their TMS ethers (246,259).
395
Other amlications Two other plant-derived foods have been analysed for ZEN by GC methods: red pepper (228,265) and beer (266). The detection limit was 10 ng/g and recoveries were 78 and 70%, respectively. GC-FID or GC-MS confirmation has been applied to ZEN in walnuts (267), in the Japanese health food Job’s tears (92) and in New Zealand pastures 9.3.3
(268).
A sensitive packed column GC-FID method for determination of ZEN as its TMS ether in blood serum was described by Trenholm et al. (239). Extraction took place on a hydrophilic matrix with elution by dichloromethane-2-propanol (9 + 1) and base cleanup procedure ‘was used to give a detection limit of 100 ng/ml with 68-74% recoveries. Confirmation was by GC-MS, which has also been used to confirm ZEN and =- and/or p-ZEL (as their TMS derivatives) in edible animal tissues (243), milk (269,270), chicken liver (271) and turkey blood plasma and tissues (272) as part of transmission studies, and in other metabolism studies (231,273). Determination of ZEN and a-ZEL in the urine of ruminants by GC-MS/MS of their TMS ethers has recently been reported (274); the detection limit for both was 1 ng/ml The effects of food processing and food additives on ZEN were determined using a GC method by Matsuura et al. (275,276) and GC-MS was again used as a confirmation procedure in studies on the distribution of ZEN during wet-milling and fermentation of contaminated corn (277,278). Mycological applications of GC include use of GC-MS to characterize (as TMS ethers) metabolites related to ZEN in Fusarium cultures (235,238,279-281) ; to confirm production of ZEN by Fusarium strains isolated from plant materials such as grains, soybeans, clover, alfalfa and feed (245,282-284) ; and in studies on substrates and culture conditions supporting laboratory production of ZEN by Fusarium isolates (241). GC-FID has also been used for quantitation and confirmation purposes in these last two areas of research (285-
.
287). 9.4. MONILIFORMIN
Derivatization is necessary f o r GC of moniliformin (Fig. 9.51. According to Gilbert et al. (288), trimethylsilylation, methylation
I
II
Fig. 9.5. Structures of moniliformin (I) and the derivatized free acid (11).
396
and trifluoroacetylation were not suitable for a GC-MS procedure. However, by reacting the free acid with N-methyl-N-(tertbutyldimethylsily1)trifluoroacetamide (MTBSTFA) containing 1% tertbutyldimethylchlorosilane (TBDMCS), they quantitatively formed a derivative with a molecular weight of 453 and characteristic EI and CI mass spectra. The proposed structure (Fig. 5 ) was elucidated by 'H and 13C nmr and mass spectroscopy, although the substitution pattern was not unequivocally established. As little as 0 . 0 0 5 ng moniliformin could be determined by GC-MS with SIM at m/z 396 and calibration was linear at least up to 0 . 5 ng. Although no cereal analysis was actually carried out, Gilbert et al. (288) projected that the procedure would allow sub-ng/g sensitivity. 9.5. ALTERNARIA TOXINS 9.5.1 Alternariol. alternariol monomethyl ether. altenuene and
isoaltenuene Alternariol, alternariol monomethyl ether, altenuene and isoaltenuene are phenolic compounds (Fig. 9.6) which readily form TMS derivatives with BSA-TMCS-pyridine (8,289), BSA-TMCS-acetonitrile (290), BSA-TMCS-tetrahydrofuran (291) or BSA-TMCS-TMSI (TriSilo TBT) (292). Confirmation of the structures of these derivatives for the first three of these metabolites, with
& / \
HO
/ \
H
-
OR
CH3
/ \
Ho&H+
MIRROR IMAGE
\
\
OH
OCH3
I &H
HO OH
'OCH~
Structures of alternariol (I, R=H alternariol Fig. 9.6. monomethyl ether (I, R=CH,) , altenuene (11) and isoalkenuene (111).
391
introduction of respectively three, two and three TMS groups, as would be expected, was achieved by GC-MS (both EI and CI) (290). The derivatives were stable in a refrigerator for several weeks when precautions were taken to exclude moisture ( 8 ) . Minimum detectable quantities of these toxins by FID were about 100 ng on a packed column of 3% OV-17, which was considered the most useful stationary phase by Per0 et al. (8). There was no problem with separation of these three toxins on a number of phases, except for 3% Dexsil-300 (8,289). Altenuene and isoaltenuene TMS ethers were separated on an HP-1 capillary column (292). There are no data on detection limits for GC-MS of TMS derivatives of alternariol, alternariol monomethyl ether and altenuene. However, one may conclude that GC-MS is much more sensitive than GC-FID: a 30 ng altenuene peak in the reconstructed ion current chromatogram ( 6 ions) on a DB-1 fused silica capillary column was shown as prominent and well separated from other peaks in an extract of moldy tomato (290). This amount corresponded to 300 ng altenuene/g tomato. The general limit of detection for alternariol, alternariol methyl ether and altenuene in juices of stone fruits, berries, citrus fruits and tomatoes, as well as in the moldy fruits, was 20 ng/g by this GC-MS method, which involved extraction with methylene dichloride and column chromatographic cleanup on silica gel. Alternariol and alternariol methyl ether have been determined in tobacco by GC of their TMS derivatives with FID (291). Samples were extracted with boiling acetone and the extracts purified by column chromatography on silica gel. Neither of the toxins were detected in forty commercial tobacco samples analysed (detection limit 250-625 ng/g). However, shredded tobacco inoculated with Alternaria alternata supported production of about 2 pg/g OP both alternariol and alternariol methyl ether. Simultaneous GC-FID determination of alternariol, alternariol methyl ether and altenuene in cultures of Alternaria tenuis ( 8 . alternata) grown in a rice medium was readily possible on a 3% OV-17 column (8). 9.5.2 Tenuazonic acids Tenuazonic acid (Fig. 9.7) has been detected without derivatization by capillary GC on a methyl silicone solumn with EI MS(S1M) at m/z 141 [M-561 (293) as well as PI- and NICI MS (294). It has been determined by GC of its TMS derivative. With the reagent mixture BSA-TMCS-pyridine (6+2+9), warmed at 5OoC for 10 minutes, the sodium and N,N,-dibenzylethylenediamine salts could be derivatized as well as the free acid (295). The limit of detection by FID was 100 ng tenuazonic acid on a 3% OV-17 packed column (295). Scott and Kanhere (296) showed by GC-MS that trimethylsilylation under the conditions of Harvan and Per0 (295) yielded a tris-TMS derivative with a molecular ion at m/z 413. The two isomers in socalled “isotenuazonic acid”, one of which is the original L-tenuazonic acid and the other is D-allo tenuazonic formed by epimerization at C-5 on long standing, were just separable by GC-MS
398
Fig. 9.7. Structure of L-tenuazonic acid. (SIM): both isomers were detected in samples of tomato paste at levels of 10-100 ng/g (296). Grain sorghum is at present the only other commodity to which GC has been applied for determination of tenuazonic acid. Again the TMS derivative was used (297). FID allowed detection of tenuazonic acid to 1 pg/g but none was found in 12 samples of weathered grain sorghum extracted with 0.01 N hydrochloric acid and cleaned up by partition from ethyl acetate into sodium bicarbonate solution. A GC column phase of 3 % OV-101 minimized interferences in analysis of sorghum compared to 3 % OV-17. GC has also been used to study transportation of tenuazonic acid into cigarette smoke (295) and to analyze fungal cultures for tenuazonic acid (295). SLAFRAMINE AND SWAINSONINE The first identification of slaframine (Fig. 9.8) in toxic red clover hay infested with Rhizoctonia leauminicola that caused salivary syndrome in horses was made by GC-MS, following preparative TLC cleanup of a 95% ethanol extract. The hay contained 50-100 pg slaframine/g based on a guinea pig bioassay. The N-TMS derivative of slaframine, formed with Tri-Silo BT, was selected for this analysis and a full EI mass spectrum was obtained (298). The following additional derivatives have been made and detected by GCFID on 3% OV-17 under the same conditions: TMS 0-deacetyl-Nacetylslaframine, N-acetylslaframine, N-acetyl-0-deacetylslaframine, TMS 0-deacetylslaframine and slaframine HFB (298). With the exception of the latter derivative, all these derivatives as well as underivatized slaframine and 0-deacetylslaframine were used to give presumptive GC identification of slaframine in the toxic hay. In addition to slaframine, the red clover hay referred to above also contained a structurally related metabolite of Rhizoctonia . 1.co1a, swainsonine (Fig. 9 . 8 ) (299). An ethanol extract Of 1euumin the hay was cleaned up by base/solvent partition and cation exchange chromatography. After conversion to swainsonine triacetate, GC was carried out on OV-1 or SP-2100 packed columns with identification by EI and CI mass spectra. Stahr et al. (300) detected acetylslaframine and acetylswainsonine chromatographed on a 3 % OV-17 column by using a nitrogen selective alkali halide detector, with 0.1 pg sensitivity. 9.6.
399
I Fig. 9 . 8 .
I1
Structures of slaframine (I) and swainsonine (11).
9 . 7 . PATULIN 9 . 7 . 1 ComDarison o f
derivatization and detection Drocedures for patulin When patulin was found for the first time as a natural contaminant of apple juice it was determined by GC without any derivatization ( 3 0 1 ) . Suzuki et al. ( 3 0 2 ) remarked on the poor sensitivity (minimum detectable amount 0.1 pg) for free patulin compared to the TMS derivative while Ralls and Lane ( 3 0 3 ) reported that underivatized patulin was unstable during GC-MS. Otherwise, GC of patulin has only been carried out after derivatization of the secondary alcohol group (Fig. 9 . 9 ) . The first GC method described for quantitative analysis of patulin in apple juice explored the use
Fig. 9 . 9 . Structure of patulin. of the TMS ether, acetate and chloroacetate ( 2 ) . The most commonly used derivative for GC of patulin is the TMS ether. Silylating reagents that have been employed are BSA ( 2 , 3 0 4 - 3 1 1 ) , BSA-TMCSpyridine ( 2 8 9 , 3 1 2 ) , BSA-TMCS ( 3 0 6 , 3 1 1 ) , BSTFA ( 3 1 1 , 3 1 3 - 3 1 7 ) , MSTFA ( 3 1 8 ) , TMSI ( 3 1 1 ) , HMDS-TMCS ( 3 1 1 , 3 1 9 ) and HMDS-TCMS-pyridine (302,320-322). Heating was used in some cases, particularly for BSTFA and MSTFA, and organic solvents such as benzene were often included in the reaction mixture. Suzuki et al. ( 3 1 1 ) observed slower reaction of patulin at room temperature with HMDS-TMCS and TMSI than with BSTFA, BSA or BSA-TMCS. The TMS derivative of patulin is generally regarded as stable ( 3 1 1 , 3 1 2 , 3 2 1 ) and can be stored for at least one month at 5OC or 2OoC ( 3 1 1 ) . However, Pohland
400
et al. (2) and Suzuki et al. (302) stated (without quantitative supporting data) that it was unstable. The reason for these different observations is not apparent. Patulin TMS ether may be detected by FID (2,289,302,312315,319,320,322). Detection limits ranged from 10-100 ng (2,302,312,322), all determined by packed column GC. Detection by ECD (304,306,307,310,311) is more sensitive and offers sub-nanogram detection for patulin as would be expected from the conjugated carbonyl grouping: the lowest minimum determinable amount was 0.1 ng (306,310), again using a packed column. MS techniques have also been used for determination of patulin as its TMS derivative. GC-EI MS with single or multiple ion monitoring of one or more of the ions m/z 226, 211 and 183 has usually been employed (305,307The ratio of the m/z 226 and 227 ions was used 309,316,318,323). for confirmatory purposes by Rosen and Pareles (309). Use of a perdeuterated analog of trimethylsilylated patulin as an internal standard was not satisfactory due to exchange of deuterium and hydrogen from excess BSA reagent (305,323). Sensitivity by GC-MS was generally good and as little as 0.2 ng patulin, depending on the GC column phase, could be detected by high resolution SIM (323). Chemical ionization increased the sensitivity of the GC-MS technique for patulin TMS ether (317,323); using a fused silica capillary column (CP Sil 5CB), Mortimer et al. (317) obtained a large peak for only 0.4 ng patulin, monitored at m/z 227. The acetate has been the preferred derivative for some analysts and may be prepared by reaction of patulin with acetic anhydride and pyridine at room temperature (2,302,303,324) or with acetic anhydride alone at high temperature (303). GC analysis must be performed immediately after derivative formation (2). With FID the detection limit was given as 6 0 ng (2) or 20 ng (302). GC-MS with selected ion monitoring at m/z 154, 137 and 136 (325) or with total ion current detection (303) has been described; a detection limit of 50 ng for patulin as its acetate was reported by Ware et al. (324) (ions monitored were not specified). GC of patulin chloroacetate, prepared with chloroacetic anhydride and pyridine, allowed detection of 40 ng patulin using FID and 12 ng by EC detection (2). This derivative was more stable than the acetate. The HFB derivative of patulin, prepared using HFBI as reagent, offered very good sensitivity for patulin analysis by capillary GC using ECD and <0.05 ng could be detected (326). 9.7.2 Methods for avvle juice and other fruit vroducts GC has been employed for determination of patulin in apple juice and other fruit products by a number of laboratories using the techniques outlined in Section 9.7.1. Extraction with ethyl acetate and cleanup on a silica gel column was used in virtually all methods. Where no cleanup was done, the detection limit for apple juice was only 700 ng/ml for the acetate with FID (2) but 1 ng/ml by MS(S1M) of the TMS derivative (309). Otherwise, reported detection limits for apple juice were in the range of 1-10 ng/ml by ECD
401
methods
(304,310,326) and 0.2-10 ng/ml by MS methods (307,316,317,323,325). The most sensitive method employed high resolution MS (SIM) of the TMS derivative and reliably measured 0.2 ng patulin /ml apple juice on a 3% SP 2250 packed column (323). The foregoing methods used both packed and capillary columns. With a capillary column, a detection limit of 10 ng/ml could be attained even using FID (314).
Overall method recoveries for patulin added to apple juice and other fruit juices are generally of the order of 80%. However, patulin was unstable in cider vinegar and very low recoveries were determined by GC-MS after one or more days storage (303). No collaborative studies have been carried out on GC methods for patulin. In addition to the original demonstration of natural occurence of patulin in apple juice (301), GC has been used to detect and determine patulin in surveys of apple juice or apple juice concentrate (304,307,314,317,325), other fruit juices (325,327) and various fruits (including apples) and berries (304,315,327). Ogawa et al. (319) did not detect patulin in a survey of apple, orange, lemon, grapefruit, pineapple and grape juices (detection limit 200 ng/ml) 9.7.3 Methods for other foodstuffs Although natural occurrence of patulin is not known for commodities other than fruit products, with the exception of corn and sorghum silages (328), GC methods for grains, grain foods, soybeans and cocoa beans have been developed. All make use of the patulin TMS derivative on packed columns with FID (302,312,322), ECD (306) or MS detection (318). The method of Fujimoto et al. (306) for rough rice, flour and soybean involved extraction with methanol5% aqueous sodium chloride (1+1) or, preferably, ethyl acetate, cleanup by solvent partition and preparative TLC; recoveries of patulin ranged from 8 5 to 92% using ethyl acetate and the detection limit was 20 ng/g. For analysis of unroasted cocoa beans by GC-MS, the same limit of detection (i.e. 20 ng/g) was achieved following ethyl acetate extraction and minimal cleanup (hexane-water partition); however, overall recovery of patulin was only 50% (318). Application of GC to determination of patulin in culture filtrates of Pencillium urticae (E. patulum) has also been reported
-
(320,321). 9.8. PENICILLIC ACID 9.8.1 Derivatization and detection procedures for Denicillic acid
Analysis for penicillic acid has in several cases been described concurrently with patulin, although the two mycotoxins are known to occur naturally in quite different types of agricultural commodities. Underivatized penicillic acid was detectable in amounts of >300 ng by GC on packed columns of 5% Apiezon L and other phases with FID (9,312). However, as with patulin, penicillic acid has usually been
402
determined as its TMS derivative. It was not until 1985 that capillary column GC on SE 3 0 showed that this derivative consisted of 2 epimeric forms with retention times of 11.6 and 11.8 minutes and virtually identical mass spectra (329). Penicillic acid is optically inactive but the carbon bearing the derivatizable hydroxyl group is chiral (Fig. 9.10). All other work on GC of penicillic acid TMS derivative has been carried out on packed columns, resulting in a single peak which elutes before the corresponding patulin derivative (9,306,311,312, 318,319,322). It is of interest to note that the TFA derivative of penicillic acid gave two well separated peaks (which were not characterized by MS) on an 8% QF-1 stationary phase according to Suzuki et al. ( 9 ) , although Thorpe and Johnson (330) reported a single peak for this derivative using a 3% OV-1 stationary phase.
Fig. 9.10.
Structure of penicillic acid.
The following silylating reagents have been used to make penicillic acid TMS ether: BSA (306,311), BSA-TMCS (306,311), BSATMCS-pyridine (289,312), BSTFA (311), BSTFA-TMCS (329), MSTFA (318), TMSI (311), HMDS-TMCS (311,319) and HMDS-TMCS-pyridine (322) virtually the same as for patulin (Section 9.7.1). As with patulin, reaction of penicillic acid with HMDS-TMCS or TMSI was slower than with BSTFA, BSA or BSA-TMCS at room temperature (311). However, unlike patulin, the TMS derivative of penicillic acid was not stable at room temperature, although it was stable for 15 days (but not for 30) at 5Oc (311). The usual means of detection of penicillic acid TMS ether has been FID (9,289,312,319,322), which allows detection of 10-30 ng penicillic acid. As would be expected, ECD (306,311) offers better sensitivity: a minimum of 0.05 ng penicillic acid was determinable. GC-MS with monitoring of the ions m/z 227 and 214 and their ratios gave quantitation at the 10 ng level (318). As mentioned above, the TFA derivative of penicillic acid has been observed as either one or two peaks. Thorpe and Johnson (330) noticed that losses of the TFA derivative occurred on evaporation of excess trifluoroacetic anhydride reagent. They used ECD but did not report detection limits. For confirmation, 8 ions of penicillic acid TFA were monitored by GC-MS; 100 ng could be confirmed by this procedure. The two penicillic acid TFA peaks measured with FID by Suzuki et al. (9) each corresponded to a minimum determinable amount
403
of 100 ng penicillic acid. The acetate derivative, which formed one peak, had a similar detection limit, with only one third the response of the TMS derivative (9). Phillips et al. (331) found that penicillic acid rapidly reacted with diazomethane to give first a methyl ester then a pyrazoline derivative, which could be detected down to 10 ng by FID; GS-MS confirmed this reaction sequence. Formation of penicillic acid PFP to obtain greater sensitivity than the TFA derivative by ECD was mentioned by Thorpe and Johnson (330). However, there has been no further method development in this direction. 9.8.2 Methods for aaricultural commodities In an analytical method for penicillic acid in foodstuffs selected for inclusion in a book of methods published by the International Agency for Research on Cancer, GC was one of three determinative procedures (332). Penicillic acid was extracted from ground corn, oats, barley, dried beans and Swiss cheese with ethyl acetate, purified by partition into 3% sodium bicarbonate solution followed by silica gel chromatography and determined as the TFA derivative by ECD. The original method was applicable to corn, beans and apple juice (330). The limit of detection was 4 ng/g and recoveries averaged 76% (332) or, in the earlier study, 81-97% at spiking levels above 10 ng/g but only 56-66% at lower levels (330). The method was used to detect and estimate penicillic acid naturally occurring in corn and dried beans (330). Preparative TLC was used to clean up chloroform-methanol (97+3) extracts of moldy corn and rice in a GC method described by Per0 et al. (312) that formed the TMS derivative of penicillic acid for FID. This GC procedure was applied to chloroform-methanol (1+1) extracts of moldy feed cleaned up by sod.ium bicarbonate solution (333), chloroform extracts of moldy tobacco and acetone extracts of tobacco smoke condensate (334). Natural occurrence of penicillic acid in moldy tobacco at 110 and 230 ng/g was demonstrated. Other GC methods have been developed for determination of penicillic acid in rice, flour and soybeans (9,306,311,322), unroasted cocoa beans (318) and fruit juices (319). Extraction solvents for grains and soybeans were methanol-5% sodium chloride solution (1+1) or ethyl acetate-water (7+1) and for cocoa beans and fruit juices ethyl acetate was used. The lowest detection limit for rice was 20 ng/g by the ECD procedure of Fujimoto et al. (306) , but this included a preparative TLC cleanup step in the method; overall recoveries averaged 89%. The detection limit for penicillic acid in cocoa beans using GC-MS of the TMS ether was 10 ng/g (recoveries 7576%) and for fruit juices was 100 ng/ml (also by FID of the TMS ether) with 64-90% recoveries. Penicillic acid has never been found as a natural contaminant in any of these commodities. The most sensitive method for determination of penicillic acid in foodstuffs remains that of Thorpe (332) referred to earlier.
404 9.9. STERIGMATOCYSTIN 9.9.1 CornDarison of detection
Drocedures Although sterigmatocystin is generally determined by TLC or LC methods (335), GC methods do exist. Most of the development work for GC of sterigmatocystin has originated in Japan. Sterigmatocystin (Fig. 9.11) can be determined directly without Using FID, the minimum detectable derivatization (7,336-338) quantity was 20 ng on a 0.53% SE-30 column (7) or on various other silylated and conditioned columns (338). Variation of FID responses on different column phases was demonstrated by Manabe et al. (7) and detection limits on 1.5% OV-1, 1.5% OV-17 and 1.5% or 5% SE-30 were relatively high (100 ng) (7,339). Better sensitivity, down to 4 ng,
.
Fig. 9.11.
Structure of sterigmatocystin.
was achieved by GC-MS with monitoring of the ions at m/z 324, 306 and 295 but quantitation was not possible with this low an amount (338). Saito (337) reported a detection limit of 10 ng for GC-MS. Nanogram amounts of free sterigmatocystin were very labile under certain chromatographic conditions exposure to metal surfaces, some solid supports, and higher temperatures (>250"C): glass capillary columns and quartz columns with quartz solid supports were not satisfactory (338). Short glass columns (25 cm x 2 mm i.d.) were best (338). Silylation of the injector, column, GC-MS interface and stationary phase supports was obviously necessary (7,338); maximum sensitivity was obtained only after repetitive injection of sterigmatocystin. Determination of sterigmatocystin as its TMS ether, formed at room temperature with BSTFA, was much more convenient and sensitive (340). Using ECD, as little as 0.5 ng sterigmatocystin could be detected on a 2% OV-17 packed column. To date there has been no extension of this procedure to capillary column GC. 9.9.2 Method s for arains The above GC detection procedures have been applied to determination of sterigmatocystin in various grains. The two most sensitive methods employed GC-MS of underivatized sterigmatocystin
-
405
(338) and GC-EC of the TMS ether (340). The former could detect, but not quantify, 1 ng/g and quantitate at 5 ng/g for wheat, rice, barley and corn. Gel permeation chromatography and solvent partition were the cleanup procedures used for acetonitrile-4% potassium chloride (9+1) or methanol-chloroform (3+17) extracts and recoveries determined by using "C-labelled sterigmatocystin were excellent (86-99%). This method suffered from the problems associated with GC of underivatized sterigmatocystin (Section 9.9.1.). The second method (340) had a limit of detection of 10 ng/g with 82-92% average recoveries of sterigmatocystin added to brown rice, barley and wheat. However, cleanup of methanol-1% sodium chloride (55+45) extracts was extensive and included column chromatography on silica gel and Florisil. as well as gel permeation chromatography. One other method, which again included a gel permeation chromatography cleanup step, had a detection limit of 50 ng/g rice by FID of underivatized sterigmatocystin (336). This method was used to detect sterigmatocystin in stored brown rice (341). Saito (337) tested three solvent mixtures for extraction and GC-MS determination of sterigmatocystin (and dihydrosterigmatocystin) from rice and obtained the best recovery (98%) with methanol-1% aqueous sodium chloride (55+45). 9.9.3. pihvdrosteriamatocvstin Dihydrosterigmatocystin is related to sterigmatocystin in that the furan double bond (Fig. 11) is reduced. It is also a mycotoxin produced by ASDerUillUS VersicoloX and was included in the GC-MS method of Saito (337). Separation of underivatized dihydrosterigmatocystin and sterigmatocystin in rice extracts was achieved on a column of 2% OV-17 and the detection limit for both was about 10 ng. Ions monitored for dihydrosterigmatocystin were m/z 326 and 297 (cf. 324 and 295 for sterigmatocystin). 9.10.
AFLATOXINS Analytical chromatography of aflatoxins in foodstuffs always meant TLC or LC until Friedli (342) reported that underivatized aflatoxin B, could be analysed by fused silica capillary GC-MS with on-column injection and direct introduction from the column into the A quantity of 10 ng was ion source of the mass spectrometer. The injected and monitored at m/z 312 (the molecular ion). technique was subsequently applied for confirmation of aflatoxin B, separated from extracts of corn and peanut butter by preparative TLC (343) and of aflatoxins B, and B, in peanut extracts cleaned up on a silica cartridge (344). NICI MS with a full spectral scan confirmed aflatoxin B, in the first case; single ion monitoring (EI) of the molecular ions at a resolution of 3000 was used for detection of aflatoxins B, and B, in the second case. The limits of detection for aflatoxins B, and Bz in peanuts were 0.1 ng/g at a signal-tonoise ratio of 2.5 (344) but aflatoxins G, and G, did not come through the DB-5 fused silica capillary column used in this work.
406 Goto et al. (345,346) succeeded in determining the four major aflatoxins (B,, B,, G, and G,) (Fig. 9.12) and separated them on a 10 meter DB-5 fused silica capillary column. This time FID was used but sensitivity was lower for aflatoxins G, and G, than for aflatoxins B, and B,. The limit of quantification was 1 ng for aflatoxins B, and B, and 2 ng for aflatoxins G, and G2 (346). The minimum detectable amounts by GC-MS were 0.3 ng for aflatoxins B, and B, and 1 ng for aflatoxins G, and G, (346). Aflatoxins B1, B, and G, were determined by GC-MS in fungal culture extracts (aflatoxin G, was not present) (345). It is possible that the lower ion source temperature (250° C) in these studies compared to 310'C in that of Rosen et al. (344) accounted for the different results for aflatoxins G. Also in the latter study a much higher initial GC column temperature was used (200'C compared to 50° C). The Japanese
0
0
H
I
II
Fig. 9.12. Chemical structures of aflatoxin B, (I) and aflatoxin G, aflatoxins B, and G, have a reduced double bond in the terminal ring.
Kk
method for determination of aflatoxin in food by capillary GC-MS was recently patented (347). The chemical structures of aflatoxins B,, B,, G, and G, (Fig. 9.12) unfortunately do not permit direct derivatization for GC. No work has been done on derivatization of the hemiacetals of aflatoxins B, and G1 (aflatoxins BZa and G,.) formed by acid-catalyzed water addition to the terminal double bond. 9.11 ERGOT ALKALOIDS In underivatized ergot alkaloids of the peptide type, e.g. ergotamine (Fig. 9.13), the lysergic acid and peptide parts of the molecule are cleaved at the amide nitrogen in a hot GC injector (225'C-3OO0C). Peptide fragments are separated by capillary or packed column GC and can be identified by MS. Characteristic ions for SIM under NICI conditions were 280/281 for ergosine, 314/315 for ergotamine, 294/295 for ergocornine, 308/309 for --and pergokryptine and 342/343 for ergocristine; a DB-1 fused silica
407
Fig.
9.13.
Structures of ergotamine (I) and ergometrine (11).
capillary column enabled detection of picogram quantities of alkaloids ( 3 4 8 , 3 4 9 ) . Ergometrine (ergonovine) (Fig. 9.13) did not decompose under these conditions (injector 25OoC) and could be monitored at m/z 323/324 (molecular weight 3 2 5 ) . The only other natural lysergic acid-derived ergot alkaloid to survive GC without thermal decomposition is lysergic acid amide ( 6 ) . Van Mansvelt et al. ( 3 5 0 ) found that six naturally occurring ergot alkaloids ergotamine, ergosine, ergostine, ergocristine, ergokryptine and ergocornine - all gave two or three decomposition peaks on GC. These consisted of one or two cyclic lactams containing two amino acids and a precursor of these lactams which also contained a deaminated third hydroxyamino acid. Detection of the peaks was by GC-FID on a 3% SE-30 column and identifications were made by GC-MS. Reproducible and instantaneous degradation was achieved in the injection port at an optimum temperature of 30OoC. The same principle has been applied to GC determination of synthetic dihydro ergot alkaloids by FID, with identification of peaks by GC-MS; only one peptide decomposition product was observed in these studies (351,352). One to 1 0 ng of alkaloid could be detected by FID and <0.5 ng by GC-MS ( 3 5 1 ) . More indirectly, and to a more limited extent, the thermal decomposition products of ergotamine have been obtained by acid hydrolysis, which was followed by GC-MS ( 3 5 3 ) . Derivatization of ergonovine to its TMS ether has also been employed for direct GC of this ergot alkaloid on a 1% OV-1 column with detection by FID ( 3 5 4 ) . N-trimethylsilyldiethylamine and TMSI in pyridine was the reagent mixture used and it was tentatively assumed that three TMS groups were introduced into the ergonovine molecule.
-
408
Clearly, GC is not very useful for determination of ergot alkaloids in foodstuffs. The thermal decomposition technique only identifies the peptide part of the molecule, so there is no differentiation between peptide alkaloid stereoisomers with differences in the lysergic acid moiety, e.g. ergotamine and ergotaminhe (350). Ergot alkaloids of the clavine type do not have amide linkages to peptide moieties. Examples of clavine alkaloids are agroclavine, elymoclavine, chanoclavine-I, festuclavine and pyroclavine (all ClaviceDs alkaloids) and fumigaclavines A and B (Penicillium and Asveraillus alkaloids). They can be chromatographed on stationary phases of 3% JXR, 5% SE-30 and 5% XE-60 (6). Application of GC to ergot alkaloid analysis has been in the pharmaceutical and forensic areas (353,354) , and for confirmation of results obtained by LC analysis of cereal products (348,349). 9.12. MISCELLANEOUS MYCOTOXINS 9.12.1 Svoridesmins Sporidesmin (Fig. 9.14), one of a group of related metabolites
produced by Pithomvces chartarum, causes facial eczema disease in sheep and cattle. Sporidesmin itself could not be gas chromatographed because of thermal decomposition; trimethylsilylation did not solve this problem (355). However, sporidesmin reacted with trifluoroacetic anhydride to give anhydrodethiosporidesmin (Fig. 9.14), which could be determined by GC on a 2% OV-1 Column, presumably by FID judging from use of a C32 hydrocarbon internal standard, although the type of detector was not specifically mentioned (355). The identity of the GC peak was
OH
CH2 Fig. 9.14. sporidesmin
I Structures (11).
II of sporidesmin
(I) and anhydrodethio-
confirmed by MS. Sensitivity was 10 p g by FID and the GC procedure was applied to determine total sporidesmins (A, El D and F, but
409
probably not B) in spores of p- chartarum and in rat liver microsomal suspensions. 9.12.2 Butenolide The Fusarium toxin commonly known as butenolide has the structure 4-acetamido-4-hydroxy-2-butenoic acid y-lactone (Fig. 9.15). Without derivatization it showed a moderately strong ECD response and 2 ng could be detected (356). The GC peak was identified by GC-MS.
Fi
,
9.15.
Structure of butenolide.
The GC procedure was the determinative step in a method for rice, wheat and barley (356). The sample was extracted with acetonitrile-5% lead acetate solution (3 + 1); water was added, the aqueous acetonitrile was defatted with n-hexane and the toxin was extracted into chloroform. Cleanup consisted of column chromatography on Wakogel S-1 and silica gel. The GC column stationary phase was 2% DEGS + 0.5% H,PO,. Recoveries of butenolide by this method were 71-87% and the detection limit was about 10 n9/9. 9.12.3 4 - J Gilbert et al. (357) described a sensitive method for determination of p-nitropropionic acid as its pentafluorobenzyl derivative using ECD, with GC-MS confirmation. The derivatizing reagent was pentafluorobenzyl bromide at 65OC with triethylamine as catalyst. The procedure was applied to analysis of cheese and fungal culture filtrates, for which limits of detection were 3 and 1 rg/g, respectively. Average recovery of p-nitropropionic acid from cheese extracted with acetone-50% sulfuric acid (97 + 3) was 86%; from culture filtrate of AsDeraillus Orvzae acidified and extracted with ether-ethyl acetate, the recovery averaged 96%. 9.12.4 Fumonisins Fumonisin B, (Fig. 9.16) is the major cancer-promoting metabolite of Fusarium moniliforme. Its natural occurrence in moldy corn has been demonstrated by LC procedures, with indirect confirmation by capillary GC of tricarballylic acid present in esterified hydrolysates of the fumonisins (358). The tricarballylic acid was esterified with isobutyl alcohol and 3M hydrochloric acid then determined by capillary GC-FID (25-150 ng gave a linear
410
COOH I
OH
0'
OH
H O O C Y COOH Fig. 9.16.
Structures of fumonisin B, (R=OH) and fumonisin B, (R=H).
calibration curve) and capillary GC-MS, with measurement of the total ion current and the mass spectrum. It should be noted that related compounds containing the tricarballylic acid grouping, such as fumonisin B, (Fig. 9.16), would be included in the total determination by this procedure. Jackson and Bennett (359) have determined fumonisins B, and B, in liquid culture of moniliforme by GC-FID and GC-MS of the TMS derivatives of the aminopolyol moieties obtained after hydrolysis. In this case peaks for both fumonisins B, and Bz were observed, well separated on a DB-5 capillary column. Plattner et al. (360) used TFA as well as TMS derivatives of the aminopolyols for GC-MS determination of fumonisins in corn, feeds and fungal cultures. 9.12.5 Fusarin C Analysis of fusarin C (Fig. 9.17) is usually carried out by Lc.
COOCHS I
\OH Fig. 9.17.
Structure of fusarin C.
41 1
Recently a GC method was published by Tseng et al. (361) for its determination in corn cultures following trimethylsilylation with Tri-Sila TBT and FID detection. A single peak was observed on an SE-30 column and the limit of detection was about 0.01 pg/g. 9.12.6 Griseofulvin and related comDounds The antifungal antibiotic griseofulvin (Fig. 9.18) has been assayed in pharmaceutical preparations and culture extracts of Penicillium urticae by GC-FID on 1% OV-17, 1% QF-1 or 1-2% SE-30 columns (4,362) and by GC-MS on 3% OV-101 or 1% OV-17 (362). Griseofulvin was well separated from dechlorogriseofulvin, dehydrogriseofulvin and isogriseofulvin.
CI I Fig. 9.18.
CH3
Structure of griseofulvin.
9.12.7 Mvcovhenolic acid The TMS derivatives of mycophenolic acid (Fig. 9.19) and its methyl and ethyl esters have been prepared with BSTFA at 80°C in a GC-FID determination procedure (3). Radio GC and radio GC-MS of fully methylated mycophenolic acid were used in biosynthetic studies on Penicillium brevicomDactum (363).
co OH
HOOC
0
CH3O
Fig. 9.19. Structure of mycophenolic acid. 9.12.8
Kojic acid (5-hydroxy-2-hydroxyethylpyran-4-one), terreic acid (5,6-epoxy-3-hydroxy-~-toluquinone) and terrein (2,3-dihydroxy-4propenyl-4-cyclopenten-1-one) are three Aweraillus mycotoxins that have been separated as their TMS derivatives on 3% OV-101 and other
412
phases (5). The lower practical limit for determination of kojic acid by FID was 100 ng and its presence in an A. flavus extract was confirmed by this procedure. With the derivatizing reagent employed (HMDS and TMCS in pyridine), a second peak for terreic acid slowly formed as the first peak disappeared. In what has been one of the few attempts to detect different kinds of mycotoxins by a GC procedure, Per0 and Harvan (289) used TMS derivatives and FID for kojic acid and terrein, in addition to other fungal metabolites. Several stationary phases were used, of which 3% OV-17, OV-101 and OV-11 gave the best results. Recently, Goto et al. (346) detected as little as 1 ng of underivatized kojic acid by capillary GC-FID; the minimum detectable amount by GC-MS monitoring the molecular ion at m/z 142 was 0.5 ng. This is another example of multimycotoxin GC as the kojic acid was co-chromatographed and separated from cyclopiazonic acid and aflatoxins B,, B,, G, and G,. 9.12.9 Oxalic acid Although commonly regarded as a plant toxin, oxalic acid is also, strictly speaking, a mycotoxin. A number of GC procedures for the dimethyl ester have been reported (364,365, and references cited therein). Diazomethane is the most efficient and reliable method for forming this derivative and the' limit of determination was 500 pg by capillary GC on an OV-1 fused silica column using FID (364). 9.12.10 llPevtaibolll volvvevtide antibiotics Polypeptides containing unusual a,a-dialkyl =-amino acids, in particular a-aminoisobutyric acid (2-methylalanineI Aib), are of widespread occurrence within different genera of fungi, including Trichoderma, Gliocladium, and Penicillium (366,367). Those peptides that are N-terminally acetylated and contain a high proportion of Aib and a C-terminally bonded amino alcohol are termed llpeptaibolsll. The broad range of biological activity of these metabolites includes toxicity to experimental animals. An example is paracelsin A (368). For detection of peptaibol mycotoxins or antibiotics in culture broths, they are hydrolysed with 6N hydrochloric acid at llO° C and the amino acids are analysed by LC, ion exchange chromatography or capillary GC after derivatization (with pentafluoropropionic anhydride) of the n-propyl esters (366,369). Use of the chiral phase Chirasil-L-Val for capillary GC of the amino acids (and amino alcohols) in pentaibols has the advantage that Aib elutes first and is not masked by other compounds (367). FID and MS are used for GC detection of the derivatized amino acids. 9.12.11 Ochratoxin A Broce (370) found that GC of ochratoxin A was not feasible for quantitative determination. Rather surprisingly, there have been no published reports on GC of this important mycotoxin. 9.12.12 --CvcloDiazonic acid According to Stahr et al. (300), cyclopiazonic acid (Fig. 9.20) eluted from a column of 3% OV-17 at 190° C and <1 p g could be
413
Fig. 9.20.
Structure of a-cyclopiazonic acid.
detected with a nitrogen selective alkali halide detector. GC of underivatized cyclopiazonic acid has now been confirmed by Goto et al. (346): the minimum detectable amounts by capillary GC-FID and GC-MS were 0.5 ng and 0.1 ng respectively and separation from aflatoxins was achieved on a DB-5 fused silica column. 9.12.13 Loline alkaloids Tall fescue infected with the endophytic fungus Acremonium coenoDhialum may contain ergot-type alkaloids such as ergovaline and ergonovine as well as saturated pyrrolizidine alkaloids of the loline type (Fig. 9.21). Packed column GC of N-formyl- and N-acetyllolline with a nitrogen-phosphorus detector was described by Belesky et al. (371). Separation by capillary GC of these two alkaloids and several related loline alkaloids was achieved by Yates et al. (372) using FID with MS confirmation. The lower limit of detection was 10 ng of alkaloid by FID. A method for infected tall fescue seed and forage had mean recoveries of 82-111% of
Fig. 9.21. Structures of loline (R = H) , N-formylloline (R = CHO) , and N-acetylloline (R = COCH,). N-acetylloline and N-formylloline; an ion-exchange solid phase extraction tube was used to clean up forage extracts. It was advised that GC columns should be conditioned with standard loline alkaloids and dedicated solely to analysis of these alkaloids. 9.12.14 Fusarochromanone Fusarochromanone (Fig. 9.22), a mycotoxin produced by Fusarium eauiseti, has been detected by TLC in Danish feed samples associated
414
.
with tibia1 dyschondroplasia in broiler chickens (373) Confirmation of identity was by capillary GC-MS (EI mode). GC-MS has also been carried out on the TMS derivatives of monoacetylfusarochromanone (374) and a related compound, TDP-6 (375) (Fig. 9.22). As these compounds are fluorescent it is unlikely that they will be routinely determined by GC.
Fig. 9.22. Structures of fusarochromanone (R1=H,R'=NH,) , acetylfusarochromanone (R1=H, R2=NHCOCH3) and TDP-6 (R'=CH,, R2=OH)
.
CONCLUSION As capillary GC columns, with their higher resolving power, gradually replace packed columns, and other advances are made in column technology and injection, as well as in detection and data processing systems, wider applications of GC to mycotoxin analysis can be expected. This particularly applies to the trichothecenes, many of which can not be conveniently determined at trace levels by TLC or HPLC. It has been indicated in this review that some underivatized mycotoxins can pass through a fused silica capillary column unchanged following on-column injection. Although sensitivity is lost by dispensing with an electron-capturing derivative , this is balanced by the convenience of not carrying out a derivatization step and the wider range of compounds, not necessarily derivatizable at all, that may potentially be analysed. Thus multimycotoxin GC analysis, which up till now has been explored very little except for the trichothecenes, could be a viable proposition. Promising applications include the analysis of fungal cultures and veterinary diagnostic work, particularly if MS detection is used. Nevertheless, for trace analysis of individual mycotoxins or groups of mycotoxins derivatization will continue to be used in order to achieve low limits of detection and quantitation. 9.13.
REFERENCES 1
2 3 4
.
C.J. Mirocha, C.M. Christensen and G.H. Nelson, Appl. Microbiol , 15 (1967 A.E. Pohland, K. San ers 497-503. and C.W. Thorpe, J. Assoc. Off. Anal. Chem., 53 (1970) 692-695. F.E. Gainer and H.J. Wesselman, J. Pharm. Sci., 59 (1970) 11571159. R.J. Cole, J.W. Kirksey and C.E. Holaday, Appl. Microbiol. 19 (1970) 106-108.
d
415
5
6
7
8 9
10
11 12 13
14 15 16 17 18 19 20
21 22 23
24 25 26
27
28
R.G. Owens, R.E. Welty and G.B. Lucas, Anal. Biochem., 35 (1970) 249-258. S. Agurell and A. Ohlsson, J. Chromatogr., 61 (1971) 339-342. M. Manabe, M. Minamisawa and S. Matsuura, J. Agric. Chem. SOC. Jpn., 45 (1971) 565-570. R.W. Pero, R.G. Owens and D. Harvan, Anal. Biochem., 43 (1971) 80-88. T. Suzuki, M. Takeda and H. Tanabe, J. Food Hyg. SOC. Jpn., 12 (1971) 495-500. C.O. Ikediobi, I.C. HSU, J.R. Bamburg and F.M. Strong, Anal. Biochem., 43 (1971) 327-340. R.W. Beaver, in: R.J. Cole Ed.), Modern Methods in the Analysis and Structural Elucidation of Mycotoxins, Academic Press, Orlando, Florida, 1986, pp. 265-292. V. Betina, J. Chromatogr., 477 (1989) ,187-233. C.P. Gorst-Allman and P . S . Steyn in: V. Betinp (Eda;$ Mycotoxins - Production, Isoiation, Separation Purification, Elsevier, Amsterdam, 1984, pp. 59-85. T. Romer, in: J.F. Lawrence (Ed.), Food Constituents and Food Residues. Their Chromatographic Determination, Marcel Dekker, New York, 1984, pp,. 355-393. H.P. van Egmond, in: R.D. King (Ed.), Developments in Food Analysis Techniques -3, Elsevier Applied Science, Barking, United Kin dom, 1984, pf. 99-144. R.F. Vesonler and W.K ohwedder, in: R.J. Cole (Ed.), Modern Methods in the Analysis and Structural Elucidation of Mycotoxins, Academic Press, Orlando, Florida, 1986, pp. 335351. J.F. Grove, Nat, Prod. Rep., 5 (1988) 187-209. P.M. Scott, in: V.R. Beasley (Ed. Trichothecene Mycotoxicosis: Pathophysiologic Effects, Volhme I, CRC Press, Boca Raton, Florida, 1989, pp. 1-26. J. Gilbert, in: M.O. Moss and J.E. Smith (Eds.), The Applied Mycology of Fusarium, Cambridge University Press, Cambridge, 1984, pp. 175-193. R.J. Pawlosk C.J. Mirocha, Y. Wen and H.K. Abbas, J. Assoc. Off. Anal. Ckm., 72 (1989) 807-812. T.R. Romer, T.M. Boling and J.L. MacDonald, J. Assoc. Off. Anal. Chem., 61 (1978) 801-808. R.M. Black, R.J. Clarke and R.W. Read, J. Chromatogr., 388 (1987) 365-378. T. Krishnamurthy, M.B. Wasserman and E.W. Sarver, Biomed. Mass Spectrom., 13 (1986) 503-518. R. Kostiainen and A. Rizzo, Anal. Chim. Acta, 204 (1988) 233246.
B. Yagen, M. Bialer and A. Sintov, J. Chromatogr., Biomed. Appl., 343 (1985) 67-75. H.M. Stahr, P.F. ROSS, W. Hyde and W. Obioha, Appl. Spectrosc., 32 (1978) 167-174. H.M. Stahr, A.A. Kraft, and M. Schuh, Appl. Spectrosc., 33 1'1979) 294-297. \--.-- - -- - -
31
J. Bi'l, M. .Van Roelenbosch, P. Sandra, M. Schelfaut and C. Van Petegiem, in: Proceedin s of the V. International IUPAC Sym osium on Mycotoxins an3 Phycotoxins, Vienna, Austria, 1-3 September 1982, Austrian Chemical Society, Vienna, 1982, pp. 12-15. P.A. D'Agostino, L.R. Provost and D.R. Drover, J. Chromatogr., 367 (1986) 77-86. W.C. Brumley, M.W. Trucksess, S.H. Adler, C.K. Cohen, K.D. White and J.A. Sphon, J. Agric. Food Chem., 33 (1985) 326-330. R. Greenhalgh and A. Ashley, Comments Toxicology, 2 (1988) 17-
32
W.F. Miles and N.P.
33
A. Pohland, C. Thorpe and J. Sphon, in: H. Kurata and Y. Ueno (Eds.), Toxigenic Fungi - Their Toxins and Health Hazard, Kodansha, Tokyo and Elsevier, Amsterdam, 1984, 217-230. B. P.-Y. Lau, P.M. Scott and T. Sakuma, Abstractf:'32nd Annual Conference on Mass Spectrometry and Allied Topics, San Antonio, Texas, 27 May - 1 June 1984, no. 789. R. Greenhalgh, D. Levandier, W. Adams, J.D. Miller, B.A. Blackwell, A.J. McAlees and A. Taylor, J. Agric. Food Chem. , 34
29 30
34 35
36.
(1985) 652-658.
(1986) 98-102.
Gurprasad, Biomed. Mass Spectrom., 12
416
36 37 38 39 40 41 42 43 44 45 46 47
R. Greenhalgh, B.A. Blackwell, M. Savard, J.D. Miller and A. Taylor, J. Agric. Food Chem., 36 (1988) 216-219. D.R. Lauren, A. Ashley, B.A. Blackwell, R. Greenhalgh, J.D. Miller and G.A. Neish, J. Agric. Food Chem. , 35 (1987) 884-889. S.P. McCormick, S.L. Ta lor, R.D. Plattner and M.N. Beremand, Appl. Environ. Microbiof. , 55 (1989) 2195-2199. R.D. Plattner. M.N. Beremand and R.G. Powell, Tetrahedron, 45 (1989) 2251-2262. E.-L. Hintikka, B. Westerling, L. Saari, S. Berg and A. Rizzo, Microbiol. Alim. Nutr., 6 (1988) 259-261. E. Lindfors, S. Berg and A. Rizzo, in: K. Aibara, S.Kuma ail IUPAC '88 and ICPP ?88 K. Ohtsubo and T. Yoshizawa (Eds. Mycotoxins and Phycotoxins, JApanese Association of Mycotoxicology, Tokyo, 1988, pp. 57-58. K.I. Eller and V.S. Sobolev, J. Anal. Chem. USSR, 38 (1983) 690-694; Zh. Anal. Khim., 38 (1983 903-907. K.E. Richardson, G.E. Toney, C.A. f!l aney and P.B. Hamilton, J. Food Prot., 52 (1989) 871-876. S. Schwenk and B. Altmayer, Z. Lebensm. Unters. Forsch. , 181 (1985) 304-307. S. Schwenk, B. Altmayer and K.W. Eichhorn, J. Chromatogr., 448 (1988) 424-427.
K. Ishii J. Kobayashi, Y. Ueno and M. Ichinoe, Appl. Environ. Microbioi. , 52 (1986) 331-333. J. Gore, A. Rougereau and 0. Person, J. Chromatogr., 291 (1984) 404-408.
50
R. Siegfried, Landwirtsch. Forsch., 31 (1978) 141-144. K. Tanaka, R. Amano, K. Kawada and H. Tanabe, J. Food Hyg. SOC. Jpn., 15 (1974 195-200. Y. Nakahara and T. Tatsuno, Chem. Pharm. Bull., 21 (1973) 1267-
51
A.
52
A. Vanyi
53
J. Gilbert, J.R. Startin and C. Crews, J. Chromatogr., 319
54 55
C.E. Kientz and A. Verweij, J. Chromatogr. , 355 (1986) 229-240. Y. Dohi, F. Watanugi, H. Kitai, K. Kosaka, M. Ichinoe and K.
48 49
56 57 58 59 60 61 62 63 64 65 66 67 68 69
1271.
Bata, A. Vanyi and R . Lasztity, J. Assoc. Off. Anal. Chem.,
66 (1983) 577-581. A.
Bata and R. Lasztity, Acta Vet. Acad. Sci. Hung.,
30 (19825 65-69. (1985) 376-381.
.
Ohba, J . Food Hy Soc. Jpn., 25 (1984) 1-9. A.F. Rizzo, L. 8aar1, and E. Lindfors, J. Chromatogr., 368 (1986) 381-386.
T. Sugimoto, K. Yamamoto and M. Minamisawa, J. Food Hyg. S O C . Jpn., 28 (1987) 169-174. H. Kamimura, M. Nishijima, K. Saito, S. Takahashi, A. Ibe, S . Ochiai and Y. Naoi J.. ,Food Hyg. SOC. 19 (1978) 443-448. H. Kamimura, M. ,Nishi]ima, K. Yasud
H. Kuroda, T. Mori, C. Nishioka, H. Okasaki and M. Takagi, J. Food Hyg. SOC. Jpn., 20 (1979) 137-1.42. Y. Onji, Y. Aoki, Y. Yamazoe, Y. Dohi and T. Moriyama, J. Liq. Chromatogr., 11 (1988) 2537-2546. T. Suzuki, M. Kurisu, Y. Hoshino, M. Ichinoe, N. Nose, Y. Tokumaru and A. Watanabe, J. Food Hyg. SOC. Jpn., 21 (1980) 434_ 9_ H.L. Trenholm, W.P. Cochrane, H. Cohen, J.I. Elliot, E.R. Farnworth, D.W. Friend, R.M.G. Hamilton, G.A. Neish and J.F. Standish, J. Amer. Oil Chem. SOC., 58 (1981) 992A-994A. H.L. Trenholm, W.P. Cochrane, H. Cohen, J.I. Elliot, E.R. Farnworth, D.W. Friend, R.M.G. Hamilton, J.F. Standish and B.K. Thompson J. Assoc. Off. Anal. Chem., 66 (1983) 92-97. H. Yazaki and H. Takahashi, Chiba-ken Eisei Kenkyusho Kenkyu Hokoku, (1981) 11-15. G.A. Bennett, R.E. Peterson, R.D. Plattner and O.L. Shotwell, J. Amer. Oil Chem. SOC., 58 (1981) 1002A-1005A. G.A. Bennett, S.E. Megalla and O.L. Shotwell, J. Amer. Oil Chem. SOC., 61 (1984) 1449-1451. J. Gilbert, M.J. Shepherd and J.R. Startin, J. Sci. Food Agric., 34 (1983) 86-92. A.I. Leonov, G.P. Kononenko, V.K. Shevtsov and N.A. Soboleva, Khim. Prir. Soed., (1988) 313-314.
417
75
C.J. Mirocha, S . V . Pathre, B. Schauerhamer and C.M. Christensen, A pl. Environ. Microbiol., 3 2 ( 1 9 7 6 ) 553-556. H. Schweighardt, J. Bohm, A.M. Abdelham+d, J. Leibetseder, M. Schuh and E. Glawischnig, Chromatographia, 13 ( 1 9 8 0 ) 447-450. P.M. Scott, S.R. Kanhere and E.J. Tarter, J. Assoc. Off. Anal. Chem., 6 9 ( 1 9 8 6 ) 889-893. T.-C. Tseng, G.-F. Yuan, J. Tseng, I-W. Hsiao and C.J. Mirocha, Bot. Bull Acad. Sinica, 26 ( 1 9 8 5 ) 83-95. R.F. Vesonder, A. Ciegler, W.K. Rohwedder and R. Eppley, Toxicon, 17 ( 1 9 7 9 ) 658-660. T. Yoshizawa and H. Hosokawa, J. Food Hyg. SOC.Jpn., 24 ( 1 9 8 3 )
76
M. Ohta, K. Ishii and Y. Ueno, J. Biochem. (Tokyo),
77
P. Flesch, G. Stockinger and I. Voigt-Scheuermann, Wein-Wiss.,
78
J.P. Chaytor and M.J. Saxby, J. Chromatogr.,
79
J. D. Rosen, R.T. Rosen and T.G. Hartman, J. Chromatogr.,
80
A. Bata, B. Harrach, K. Ujszaszi, A. Kis-Tamas and R. Lasztity, Appl. Environ. Microbiol., 49 ( 1 9 8 5 ) 678-681. B. Harrach, C.J. Mirocha, S.V. Pathre and M. Palyusik, Appl. Environ. Microbiol., 4 1 ( 1 9 8 1 ) 1428-1432. C.I. Szathmary, C.J. Mirocha, M. Palyusik and S . V . Pathre. Appl. Environ. Microbiol., 32 ( 1 9 7 6 ) 579-584. K. Yamamoto, J. Nara Med. ASSOC., 26 ( 1 9 7 5 ) 245-255. E. Karppanen, A. Rizzo, S. Berg, E. Lindfors and R. Aho, J. Sci. Agric. SOC. Finland, 57 ( 1 9 8 5 ) 195-206. P.M. Scott and S.R. Kanhere, J. Chromatogr., 368 ( 1 9 8 6 ) 374-
70 71 72 73 74
81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98
99 100 101 102 103 104 105 106
413-415.
1591-1598.
4 1 ( 1 9 8 6 ) 182-189. 113.
( 1 9 8 6 ) 241-251.
380.
237
82 (1977)
(1982)
107355
P.M. Scott, J. Assoc. Off. Anal. Chem., 65 ( 1 9 8 2 ) 876-883. W. Blaas, M. Kellert, S. Steinmeyer, R. Tiebach and R. Weber, 2 . Lebensm. Unters. Forsch., 1 7 9 ( 1 9 8 4 ) 104-108. H.M. Hussein, R.A. Franich, M. Baxter and I.G. Andrew, Food Addit. Contam., 6 ( 1 9 8 9 ) 49-58. T. Sakamoto, S.P. Swanson, T. Yoshizawa and W.B. Buck, J. Agric. Food Chem., 34 ( 1 9 8 6 ) 698-701. S.P. Swanson, J. Nicoletti, H.D. Rood, Jr., W.B. Buck and L.M. C6te, J. Chromatogr., Biomed. Appl., 414 ( 1 9 8 7 ) 335-342. T. Tanaka, A. Hasegawa, Y. Matsuki, K. Ishii and Y. Ueno, Food Addit. Contam., 2 ( 1 9 8 5 ) 125-137. T. Tanaka, A. Hasegawa, Y. Matsuki and Y. Ueno, Food Addit. Contam., 2 ( 1 9 8 5 ) 259-265. T. Tanaka, A. Hasegawa, Y. Matsuki, U.-S. Lee and Y. Ueno, Food Addit. Contam., 3 ( 1 9 8 6 ) 247-252. T. Tanaka, A. Hasegawa, S. Yamamoto, U.-S. Lee, Y. Sugiura and Y. Ueno, J. Agric. Food Chem., 3 6 ( 1 9 8 8 ) 979-983. T. Tanaka, S . Yamamoto, A . Hasegawa, N. Aoki, J.R. Besling, Y. Sugiura and Y. Ueno Mycopathologia, 1 1 0 ( 1 9 9 0 ) 19-22. R. Tiebach and W. Biaqs,,J. Chromatogr., 454 ( 1 9 8 8 ) 372-381. T. Yoshizawa, L.-M. Cote, S.P. Swanson and W.B. Buck, Agric. Biol. Chem. , 5 0 ( 1 9 8 6 ) 227-229. C.J. Mirocha, S.V. Pathre, R.J. Pawlosky and D:W. Hewetson, in: R.J. Cole (Ed.), Modern Methods in the Analysis and Structural Elucidation of M cotoxins, Academic Press, Orlando, Florida, 1 9 8 6 , pp. 353-39Y. C.J. Mirocha, R.J. Pawlosky and D.W. Hewetson, in: J.L. Richard and J.R. Thurston (Eds.) Diagnosis of Mycotoxicoses, Martinus Nijhoff, Dordrecht, 1 9 8 6 , pl. 305-322. S.P. Swanson, H.D. Rood, Jr., J.C. ehrens and P.E. Sanders, Appl, Environ. Microbiol., 53 ( 1 9 8 7 ) 2821-2826. R. Tiebach, W. Blaas, M. Kellert, S. Steinmeyer and R. Weber, J. Chromato r., 318 ( 1 9 8 5 ) 103-111. A.E. Desjar%ins, R.D. Plattner and F. Vanmiddlesworth, Appl. Environ. Microbiol., 5 1 ( 1 9 8 6 ) 493-497. M.W. Trucksess, M.T. Flood, M.M. Mossoba and S.W. Page, J. Agric. Food Chem., 3 5 . ( 1 9 8 7 445-448. N. Milama and H. Lelievre, nalusis, 7 ( 1 9 7 9 ) 232-235. P.M. Scott, P.-Y. Lau and S.R. Kanhere, J. Assoc. Off. Anal. Chem. , 64 ( 1 9 8 1 ) 1364-1371. H. Cohen and M. Lapointe, J. Assoc. Off. Anal. Chem. , 67 ( 1 9 8 4 )
a
i i n 6 - i i n7
418
107 108 109 110
T. Krishnamurth E.W. Sarver, S.L. Greene and B.B. Jarvis, J. Assoc. Off. Anal: Chem., 70 (1987) 132-140. Y. Luo, J. Zheng and J. Yang, Environ. Chem., 8 1989 27-31. Y. Luo, J. Zheng and J. Yang, Environ. Chem., 9 11990l 43-50. P.M. Scott, G.A. Lombaert, P. Pellaers, S . Bacler, S.R. Kanhere, W.F. Sun, P.-Y. Lau and D. Weber, Food Addit. Contam. , 6 (1989) 489-500.
112
Steinme er R. Tiebach and R. Weber, Z. Lebensm. Unters. Forsch. , 1ll (1985) 198-199. B.G. Osborne and K. H. Willis, J. Sci. Food Agric., 35 (1984)
113
C. Noon ugdee, J. Bohm, J. Leibetseder and M. Schuh, Ernahrung,
114
L. Muszkat, N. Paster and R. Barkai-Golan, Phytoparasitica, 16
111
S.
579-583.
9 (19857 622-625. (1988) 275-280.
G.M. Ware, A. Carman, 0. Francis and S . Kuan, J. Assoc. Off.Ana1. Chem. , 67 (1984) 731-734. 116 A. Visconti, C.J. Mirocha and R.J. Pawlosky, J. Assoc. O f f . Anal. Chem., 70 (1987) 193-196. 117 L. Mudoz, M. Cardelle, M.Pereiro and R. Riguera, J. Agric. Food Chem. , 38 (1990) 1004-1006. 118 S.R. Kanhere and P.M. Scott, J. Chromatogr., 511 (1990) 384115
119
389.
J.F. Grove, A.J. McAlees and A. Taylor, J. Org. Chem., 53 (1988) 3860-3862.
T. Yoshizawa, T. Shirota and N. Morooka, J. Food Hyg. SOC. Jpn., 19 (1978) 178-184. J.L. MacDonald and J.C. Swims, in: J.W. Finley 121 J.M. Rothber and D.E. Sctwass (Eds.) , Xenobiotics in Foods and Feeds, American Chemical Society, Washington, D.C., 1983,.pp. 271-281. 122 R. Kostiainen, A. Rizzo and A. Hesso, Arch. Environ. Contam. Toxicol., 18 (1989) 356-364. 123 T. Krishnamurthy and E.W. Sarver, J. Chromatogr., 355 (1986) 120
253-264.
E.J. Mulders and H.A.M. van Impelen-Peek, Z. Lebensm. Unters Forshch., 183 (1986 406-409. 125 R.M. Black, R.J. Clarke and R.W. Read, J. Chromatogr., 367
124
126 127 128 129 130 131 132 133 134 135 136 137
(1986) 103-115.
P.M. Scott, S.R. Kanhere and P.-Y. Lau, in: Proceedings of the V. International IUPAC S posium on M cotoxins and Ph cotoxins, Vienna, Austria, 1-3 Sepgmber 1982 , lustrian Chemical Society, Vienna, ;982,, pp. 44-47. T.W. Nowicki, Proceedings of the 21st Annual Western Trace Organic Analysis Workshop, Vancouver, British Columbia, 23-25 April 1985, pp. 27-32. H.D. Rood, Jr. , W.B. Buck and S . P . Swanson, J. Assoc. Off. Anal. Chem., 71 (1988) 493-498. P. Begley, B.E. Foulger, P.D. Jeffery, R.M. Black and R.W. Read, J. Chromatogr., 367 (1986) 87-101. D.W. Hewetson and C.J. Mirocha, J. Assoc. Off. Anal. Chem., 70 (1987) 647-653.
R.R. King, R. Greenhalgh and B.A. Blackwell, J. Agric. Food Chem. , 32 (1984) 72-75. R.J. Pawlosky, C.J. Mirocha and T. Yoshizawa, J. Agric. Food Chem. , 32 (1984) 1423-1425. A. Visconti, C.J. Mirocha, A. Bottalico and J. Chelkowski, Mycotoxin Res., 1 (1985) 3-10. A. Visconti, L.M. Treeful and C.J. Mirocha, Biomed. Mass Spectrom., 12 (1985) 689-694. H.D. Rood, Jr., W.B. Buck and S . P . Swanson, J. Agric. Food Chem., 36 (1988) 74-79. T. Yoshizawa, T. Sakamoto and K. Kuwamura, Appl. Environ. Microbiol., 50 (1985) 676-679. 5; Kostiainen and S. Nokelainen, J. Chromatogr. , 513 (1990) 31J I .
H:D. Rood, Jr., S.P. Swanson and W.B. Buck, J. Chromatogr., Biomed. Appl. , 378 (1986) 375-383. 139 B.J. Wreford, and K.J. Shaw, Food Addit. Contam., 5 (1988) 141-
138
- _ .. 147.
J. Bauer, W. Bollwahn M. Gareis, B. Gedek and K. Heinritzi, Appl. Environ. Microbial. , 49 (1985) 842-845. 141 S.V: Pathre, C.J. Mirocha, C.M. Christensen and J. Behrens, J. Agric. Food Chem. , 24 (1976) 97-103. 140
419
142 14 3
C.E. Kientz and A. Verweij, J. Chromatogr., 407 (1987) 340-342. C.A. Knu p, S.P. Swanson and W.B. Buck, J. Agric. Food Chem.,
14 4
H.M. Stahr, D. Lerdal, W. Hyde and R. Pfeiffer, Appl. Spectrosc., 37 (1983) 396-400. H.M. Stahr, M. Domoto, W. Hyde and P.J. Martin, Microchem. J.,
14 5 14 6 147 148 149 150 15 1
34 (1986y 865-868. 32 (1985) 266-271.
G.M. Ware, O.J. Francis, A.S. Carman and S . S . Kuan, J. Assoc. Off. Anal. Chem., 69 (1986) 899-901. T. Tanaka, A. Hasegawa, Y. Matsuki, K. Ishii and Y. Ueno, Proc. Jpn. Assoc. Mycotoxicol., (1984) 34-37. S.J. Terhune, N.V. Nguyen, J.A. Baxter, D.H. Pryde and S.A. Qureshi, J. Assoc. Off. Anal. Chem., 67 (1984) 1102-1104. T.R. Romer, J. Assoc. Off. Anal. Chem., 69 (1986) 699-703. G.A. Bennett, R . D . Stubblefield, G.M. Shannon and O.L. Shotwell, J. Assoc. Off. Anal. Chem., 66 (1983) 1478-1480. R.F. Vesonder, A. Ciegler, R. F. Rogers, K.A. Burbridge, R.J. Bothast and A.H. Jensen, Appl. Environ. Microbiol., 36 (1978) 885-888.
H. Tanaka and K. Ishizaki, Shiryo Kenkyu Hokoku (Tokyo Hishiro Kensasho) , 13 (1988) 1-9. 153 A. Cisti, J. Faure, J.F. Narbonne and J.G. Faugere, Ann. Fals. Exp. Chim., 76 (1983) 107-110. 154 M. Gareis, J. Bauer, C. Enders and B. Gedek, in: J. Chelkowski (Ed.), Fusarium. Mycotoxins, Taxonomy and Pathogenicity, Elsevier, Amsterdam, 1989, pp. 441-472. 155 T.N. Bhavanishankar and T. Shantha, J. Sci. Food Agric., 40 152
156
(1987) 327-332.
E. Isohata, M. Toyoda and Y. Saito, Eisei Shikenjo Hokoku, 104 (1986) 113-118.
.
Y. Luo, T. Yoshizawa and T. Katayama, Appl. Environ. Microbiol , 56 (1990) 3723-,3726. 158 A.L. Patey and J. Gilbert, in: J. Chelkowski (Ed.), Fusarium. Mycotoxins, Taxonomy and Pathogenicity, Elsevier, Amsterdam, pp. 399-420. 159 A. Visconti and F. Palmisano, J. Chromatogr., 252 (1982) 305157
160 161 162
309.
C.J. Mirocha, S.V. Pathre and J. Behrens, J. Assoc. Off. Anal. Chem. , 59 (1976) 221-223. P.M. Scott, K. Nelson, S.R. Kanhere, K.F. Kar inski, S . Hayward, G.A. Neish and A.H. Teich, Appl. Environ. Ificrobiol., 48 (1984) 884-886. D. Abramson, R.M. Clear and T.W. Nowicki, Can. J. Plant Sci., 67 (1987) 611-619.
Health Protection Branch, Laboratory Procedure LPFC-155, Ottawa , 1989. 164 T. Ilus, M.-L. Niku-Paavola and T.-M. Enari, Eur. J. Appl. Microbiol. Biotechnol. , 11 (1981) 244-247. 165 T.W. Nowicki and D. Gaba, personal communication, 1990. 166 C. Szathmary, J. Galbcz, L. Vida and G. Alexander, J. Chromatogr. 191 (1980) 327-331. 167 H. Cohen and M. Lapointe, J. Assoc. Off. Anal. Chem. , 65 (1982)
163
168 169 170 171 172 17 3 174 17 5
1429-1434.
Y. On]i, M. Uno, H. Nagami, Y. Dohi and T. Moriyama, J. Food Hyg. SOC. Jpn., 28 (1987) 50-54. H. Pettersson, H.-K. Kiessling and K. Sandholm, Swedish J. Agric. Res., 16 (1986) 179-182. E.W. Sydenham and P.G. Thiel, Food Addit. Contam., 4 (1987) 277-284. A. Visconti, J. Chelkowski and A. Bottalico, Mycotoxin Res., 2 (1986) 59-64. K.I. Eller and V.A. Tutelyan, in: K. Aibara, S. Kumagai, K. Ohtsubo and T. Yoshizawa {Eds.), IUPAC '88 and ICPP '88. Mycotoxins and Phycotoxins, Japanese Association of Mycotoxicology, Tokyo, 1988, pp. 59-60. A. Bata, A. Vanyi, R. Lasztity and J. Galacz, J. Chromatogr., 286 (1984) 357-362. R.W. Coppock, S.P. Swanson, H.B. Gelberg, G.D. Koritz, W.E. Hoffman, W. B. Buck and R. F. Vesonder, Amer. J. Vet. Res. , 46 (1985) 169-174. C.C. Hsia, J.L. Wu, X.Q. Lu and Y.S. Li, Cancer Det. Prev., 13 (1988) 79-86.
420
176 177 178
H. Cohen and M. Lapointe, in: Proceedings of the. V. International IUPAC Sym osium on Mycotoxins and Ph cotoxins, Vienna, Austria. 1-3 SepEember 1982, Austrian Chemica]l Society, Vienna’ 1982 , p. 20-21. H.K. &bas, C.3: Mirocha, R.A. Meronuck, J.D. Pokorny, S.L. Gould and T. Kommedahl, Appl. Environ. Microbiol, 54 (1988) __ 1930-1933.
179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 2 02 203 204 205 206 2 07
14 16-1420. T.W. Nowicki, D.G. Gaba, J.E. Dexter, R.R. Matsuo and R.M. Clear, J. Cereal Sci., 8 (1988) 189-202. J. Bauer, M. Gareis and B. Gedek, in: J. Chelkowski (Ed.), Fusarium. M cotoxins, Taxonomy and Pathogenicity, Elsevier, Ams er am f989, pp. 139-165. G.Jf Cdollins and J.D. Rosen, J. Assoc. Off. Anal. Chem., 62 (1979) 1274-1280. L.M. CBte, A.M. Dahlem, T. Yoshizawa, S.P. Swanson and W.B. Buck, J. Dairy Sci., 69 (1986) 2416-2423. D.B. Prelusky, H.L. Trenholm, G.A. Lawrence and P.M. Scott, J. Environ. Sci. Health, B19 (1984) 593-609. T.S. Robison, C. J. Mirocha, H. J. Kurtz, J: C. Behrens, M. S. Chi, G.A. Weaver and S.D. Nystrom, J. Dairy S c i . , 62 (1979) 637-641. D.J. Reutter, S.F. Hallowell and E.W. Sarver, U.S. A r m Chemical Research, Development and Engineering Center Reporf No. CRDEC-TR-87088, Aberdeen Proving Ground, Maryland, 1987. S.P. Swanson and R.P. Corley, in: V.R. Beasley (Ed.), Trichothecene Mycotoxicosis: Pathophysiologic Effects, Volume I, CRC Press, Boca Raton, Florida 1989, pp. 37-61. R.A. Corley, S . P . Swanson, G.J. Guilo, L. Johnson, V.R. Beasley and W.B. Buck, J. Agric. Food Chem., 34 (1986) 868-875. M. Gareis, A. Hashem, J. Bauer and B. Gedek, Toxicol. Appl. Pharmacol., 84 (1986) $68-172. D.B. Prelusky, D.M. Veira and H.L. Trenholm, J. Environ. Sci. Health, B20 (1985) 603-624. D.B. Prelusky, D.M. Veira, H.L. Trenholm and B.C. Foster, J. Environ. Sci. Health, B22 (1987) 125-148. C.J. Mirocha, R.A. Pawlosky, K. Chatterjee, S. Watson and W. Hayes, J. Assoc. Off. Anal. Chem., 66 (1983) 1485-1499. V.R. Beasley, S.P. Swanson, R.A. Corley W.B. Buck, G. D. Koritz and H.R. Burmeister, Toxicon, 24 (1986j 13-23. S.P. Swanson, L. Terwel, R.A. Corley and W.B. Buck, J. Chromatogr., 248 (1982) 456-460. K. Chatterjee, R.J. Pawlosky, L. Treeful and C.J. Mirocha, J. Food Safety, 8 (1986) 25-34. K. Chatter’ee, A. Visconti and C.J. Mirocha, J. Agric. Food Chem. , 34 71986) 695-697. A.M. Dahlem, S.P. Swanson, L.-M. CBte, T. Yoshizawa and W.B. Buck, J. Chromatogr., Biomed. Appl., 378 (1986) 226-231. S.P. Swanson, A.M. Dahlem, H.D. Rood, Jr. , L.-M. C B t e , W.B. Buck and T. Yoshizawa, J. Assoc. O f f . Anal. Chem., 69 (1986) 41-43. S. P. Swanson, V. Ramaswamy, V. R. Beasley, W. B. Buck and H. R. Burmeister, J. Assoc. Off. Anal. Chem., 66 (1983) 909-912. A.K. Lun, E.T. Moran, Jr., L.G. Young and E.G. McMillan, Bull. Environ. Contam. Toxicol., 40 (1988) 317-324. D.W. Friend, B.K. Thompson, H.L. Trenholm, K.E. Hartin and D.B. Prelusky, Can. J. Anim. Sci., 66 (1986) 229-236. A.K. Lun, L.G. Young, E.T. Moran, Jr. , D.B. Hunter and J.P. Rodriquez, Poultry Sci., 65 (1986) 1095-1099. Y. On]i, Y. Dohi, Y. Aoki, T. Moriyama, H. Nagami, M. Uno, T. Tanaka and Y. Yamazoe, J. Agric. Food Chem., 37 (1989) 478-481. A. Visconti and C.J. Mirocha, Appl. Environ. Microbiol., 49 (1985) 1246-1250. R.D. Voyksner, W.M. Hagler, Jr. and S.P. Swanson, J. Chromatogr., 394 (1987) 183-199.
42 I
T. Yoshizawa, S.P. Swanson and C.J. Mirocha, Appl. Environ. Microbiol., 39 (1980) 1172-1177. 209 T. Yoshizawa, H. Takeda and T. Ohi, Agric. Biol. Chem., 47 (1983) 2133-2135. 210 A.A. El-Banna, R.M.G. Hamilton, P.M. Scott and H.L. Trenholm, J. Agric. Food Chem., 31 (1983) 1381-1384. O.J. Fletcher, L.D. 211 L.F. Kubena, S.P.. Swanson, R.B. Harve Rowe and T.D. Phillips, Poultry Sci., (198511 1649-1655. 212 J. Bohm, C. Noonpugdee, J. Leibetseder and M. Sc uh, Ernahrung, 208
&
8 (1984) 601-605.
R.W. Coppock, S.P. Swanson, H.B. Gelberg and W.B. Buck, Amer. J. Vet. Res., 49 (1988) 1997-1999. 214 W.M. Ha ler, Jr., F.T. Jones and D.T. Bowman, Biodeter. Res.,
213
2 (19893 337-349.
215
R.M. Clear, T.W. Nowicki and J.K. Daun, Can. J. Plant Pathol.,
216
A.A. El-Banna, P.M. Scott, P.-Y. Lau, T. Sakuma, H.W. Platt and V. Campbell, Appl. Environ. Microbiol., 47 J1984) 1169-1171. A.E. Desjardins and R.D. Plattner, J. Agric. Food Chem., 37
217 218 219 220 221 222 223 224 225 226
11 (1989
308-312.
388-392. 5 1989) .
Payen, T. Girard, M. Gaillardin and P. Lafont, Microbiol. Alim. Nutr., 1 (1983) 143-146. L.-M. Gate, J. Nicoletti, S.P. Swanson and W.B. Buck, J. Agric. Food Chem., 34 (1986) 458-460. S.P. Swanson, C. Helaszek, W.B. Buck, H.D. Rood, Jr. and W.M. Haschek, Food Chem. Toxicol., 26 (1988) 823-829. J. Kobayashi, T. Horikoshi, J.-C. Ryu, F. Tashiro, K. Ishii and Y. Ueno, Food Chem. Toxicol., 25 (1987) 539-544. T. Yoshizawa, S.P. Swanson and C.J. Mirocha, Appl. Environ. Microbiol., 40 (1980) 901-906. W.W.A. Bergers, J.G.M.M. van der Stap and C.E. Kientz, Appl. Environ. Microbiol., 50 (1985) 656-662. K. -H. Kim, Y. -W. Lee, C.J. Mirocha and R. J. Pawlosky, Appl. Environ. Microbiol., 56 (1990) 260-263. G.A. Bennett, D.T. Wicklow, R.W. Caldwell and E.B. Smalley, J. Agric. Food Chem., 36, (1988) 639-642. W.A. Croft, B.B. Jarvis and C.S. Yatawara, Atmos. Environ., 20 (1986) 549-552.
228
P.M. Scott, in: J.F. Lawrence (Ed.), Trace Analysis, volume 1, Academic Press, New York, 1981, 193-266. 2. Ormai-Cserhalmi, A. Bata angpf. Sarudi, Acta Aliment. 13
229
R. Siegfried and H.K. Frank, Z. Lebensm. Unters. Forsch., 166
230
T. Suzuki, Y. Hoshino, M. Kurisu, N. Nose and A. Watanabe, J.
227
(1984) 183-188.
(1978) 363-367.
233
Food Hyg. SOC. Jpn., 19 (1978) 201-207. Y. Ueno and F. Tashiro, J. Biochem. (Tokyo), 89 (1981) 563-571. A. Van i, A. Bata and R. Lasztity, in: K. Naguib .M.M. Naguib, D.L. $ark and A.E. Pohland (Eds.) , Proceedings of the International Symposium on Mycotoxins, Cairo, 6-8 September, 1981, National Research Centre, Cairo, Etygf:11983, p. 357-361. E.R. Thouvenot and R.F. Morfin, J. Chromatogr. , 170 (1979) 165-
234
A. Bata, R. Ldsztity and J. Balla, Acta Aliment., 9 (1980) 99-
231 232
173.
106.
W.M. Hagler, C.J. Mirocha, S.V. Pathre and J.C. Behrens, Appl. Environ. Microbiol., 37 (1979) 849-853. 236 C.J. Mirocha, B. Schauerhamer and S.V. Pathre, J. Assoc. Off. Anal. Chem., 57 (1974) 1104-1110. 237 S.V. Pathre, C.J. Mirocha and S.W. Fenton, J. Assoc. Off. Anal. Chem. , 62 (1979) 1268-1273. 238 K.E. Richardson, W.M. Hagler, Jr. and C.J. Mirocha, J. Agric. Food Chem., 3 3 (1985) 862-866. 239 H.L. Trenholm, R.W. Warner and E.R. Farnworth, J. Assoc. Off. Anal. Chem., 63 (1980) 604-611. 240 G.A. Bennett and O.L. Shotwell, J. Assoc. Off. Anal. Chem., 73 235
241 242
(1990) 270-275.
K.E. Richardson, Hamilton, J. Food C.J. Mirocha, J. Appl. Microbiol.,
W.M. Hagler, Jr., C.A. Haney and P.B. Prot., 48 (1985) 240-243. Harrison, A.A. Nichols and M. McClintock, 16 (1968) 797-798.
422
256
J.E. Roybal, R.K. Munns, W.J. Morris, J.E. Hurlbut and W. Shimoda, J. Assoc. Off. Anal. Chem., 71 (1988) 263-271. P.M. Scott, T. Panalaks, S. Kanhere and W.F. Miles, J. Assoc. Off. Anal. Chem. , 61 (1978) 593-600. K.E. Richardson, W.M. Hagler, Jr., C.L. Campbell and P.B. Hamilton, Mycopathologia, 90 (1985) 155-160. A. Bottalico A. V+conti, A. Lo rieco, M. Solfrizzo and C.J. Mirocha, A pi. Environ. Microbio?. , 49 (1985) 547-551. W.J.A. Vansenheuvel, Sep. Sci. , 3 (1968) 151-163. C.L. Holder, C.R. Nony and M.C. Bowman, J. Assoc. Off. Anal. Chem., 60 (1977) 272-278. Y. Luo, J. Zheng and J. Yang, personal communication, 1989. P.F. ROSS, in: J.L. Richard and J.R. Thurston (Eds.), Diagnosis of Mycotoxicoses, Martinus Nijhoff, Dordrecht, 1986, pp. 323-329. M. Thaler Landwirtsch. Forschung, 37 (1980) 677-684. M. Jemmali, Y. Ueno, K. Ishii, C. Frayssinet and M. Etienne, Experiential 34 (1978) 1333-1334. T. Sugimoto, M. Minamizawa, K. Takano and Y. Furukawa, J. Food Hyg. SOC. Jpn., 17 (1975) 12-18. N. Kazansas, R.W. Ely, M.L. Fields and J.W. Erdman, Jr., Appl. Environ. Microbiol., 47 (1984) 1118-1125. M.S. Palmgren, L.S. Lee, A.J. DeLucca, I1 and A. Ciegler, Amer. Ind. Hyg. Assoc. J., 44 (1983) 485-488. M. Jemmali, Ann. Microbiol. (Inst. Pasteur) , 124B (1973)
257 258
R. Lasztity and L. Woller, Nahrung, 19 (1975) 537-546. A. Bottalico, P. Lerario and A. Visconti, Phytopathol. Med. , 20
2 59
C.J. Mirocha, B. Schauerhamer, C.M. Christensen, M.L. NikuPaavola and M. Nummi, Appl. Environ. Microbiol., 38 (1979)
260
H.K. Abbas, C.J. Mirocha and J. Tuite, Appl. Environ. Microbiol., 51 (1986) 841-843. W.M. Hagler, Jr., K T czkowska and P.B. Hamilton, Appl. Environ. Microbiol. , b7 T1984) 151-154. W.M. Hagler, Jr., D.T. Bowman, M. Babadoost, C.A. Haney and S.P. Swanson, Crop Sci., 27 (1987) 1273-1278. U.-S. Lee, H.-S. Jang, T. Tanaka, A. Hasegawa, Y.-J. Oh and Y. Ueno, Food Addit. Contam., 2 1985) 185-192. U.-S. Lee, H.-S. Jang, T. Tana a, A. Hasegawa, Y.-J. Oh, C.-M. Cho, Y. Sugiura and Y. Ueno, Food Addit. Contam., 3 (1986) 253-
243 244 245 246 247 248 249 250 251 252 2 53 254 2 55
261 262 263 264
109-114.
(1981) 1-6.
749-750.
Ic
261.
266
Bata, R. Lasztit I. Sarudi and Z.O. Cserhalmi, Z. Lebensm. Unters. Forsch. , 1% (1983) 203-204. A. Bata, I. Sarudi and 2 . Ormai-Cserhalrni, Acta Aliment., 10
267
M. Jemmali and C. Mazerand, Ann. Microbiol. (Inst. Pasteur) ,
268
M.E. di Menna, D.R. Lauren, P.R. Poole, P.H. Mortimer, R.A. Hill and M.P. Agnew, New Zealand J. Agric. Res., 30 (1987) 499-
269
C.J. Mirocha, S.V. Pathre and T.S. Robison, Food Cosmet. Toxicol., 19 (1981) 25-30. M. Palyusik, B. Harrach, C.J. Mirocha and S.V. Pathre, Acta Vet. Acad. Sci. Hung., 28 (1980) 217-222. C.J. Mirocha, T.S. Robison, R.J. Pawlosky and N.K. Allen, Toxicol. Appl. Pharmacol., 66 1982) 77-87. M. Olsen, C.J. Mirocha, H.K. &bas and B. Johansson, Poultry Sci. 65 (1986) 1905-1910. M. Oisen, H. Pettersson, K. Sandholm, A. Visconti and K.-H. Kiessling, Food Chem. Toxicol., 25 (1987) 681-683.
265
270 271 272 273
A.
(1981) i62-163.
131B (1980) 319-321. 5nA -_ _.
R.J. Plasencia, C.J. Mirocha, R.J. Pawlosky and J.F. Smith, J. Assoc. Off. Anal. Chem., 73 (1990) 973-980. 275 Y. Matsuura, T. Yoshizawa and N. Morooka, J. Food Hyg. SOC. Jpn., 20 (1979) 385-390. 276 Y. Matsuura, T. Yoshizawa and N. Morooka, J. Food Hyg. SOC. Jpn., 22 (1981) 293-298. 277 G.A. Bennett, E.E. Vandegraft, O.L. Shotwell, S.A. Watson and B.J. Bocan, Cereal Chem., 55 (1978) 455-461. 278 G.A. Bennett, A.A. Lagoda, O.L. Shotwell and C.W. Hesseltine, J. Amer. Oil. Chem. SOC., 58 (1981) 974-976. 274
42 3
279 280 281 282 283 284 285 286 287
R.A. Jackson, S.W. Fenton, C.J. Mirocha and G. Davis, J. Agric. Food Chem., 22 (1974) 1015-1019. S.V. Pathre, S.W. Fenton and C.J. Mirocha, J. Agric. Food Chem., 28 (1980) 421-424. J.A. Steele, C.J. Mirocha and S.V. Pathre, J. Agric. Food Chem., 24 (1976) 89-97. D.R. Lauren, M.E. di Menna, R. Greenhalgh, J.D. Miller, G.A. Neish and L.W. Burgess, New Zealand J. Agric. Res., 31 (1988) 219-225.
U.-S. Lee, H.-S. Jan T. Tanaka, N. Toyasaki, Y. Sugiura, Y.J. Oh, C.-M. Cho ani'Y. Ueno, Appl. Environ. Microbiol., 52
1 1258-1260.
(1986
C.S. rempus, W.M. Hagler, Jr. , F.T. Jones and P.E. Nelson, Biodeter. Res., 2 (1989 351-362. D. Cullen, R.W. Caldwel! and E.B. Smalley, Phytopathology, 72 1982) 1415-1418.
k.E. Richardson W.M. Hagler, Jr. and P.B. Hamilton, Appl. Environ. Microbiol. , 47 (1984) 643-646. T. Yoshizawa, Y. Matsuura, Y. Tsuchiya, N. Morooka, K. Kitani, M. Ichinoe and H. Kurata, J. Food Hyg. SOC. Jpn., 20 (1979) 2126.
J. Gilbert, J.R. Startin, I. Parker, M.J. Shepherd, J.C. Mitchell and M.J. Perkins, J. Chromatogr., 369 (1986) 408-414. 289 R.W. Per0 and D. Harvan, J. Chromatogr. 80 (1973,) 255-258. Fresenius Z. Anal. 290 M. Kellert, W. Blaas and M. Wittkowsk:, Chem., 318 (1984) 419-424. 291 G.B. Lucas, R.W. Pero, J.P. Snow and D. Harvan, J. Agric. Food Chem. , 19 (1971) 1274-1275. 292 A. Visconti, A. Bottalico, M. Solfrizzo and F. Palmisano, Mycotoxin Res., 5 (1989) 69-76. 293 M.H. Lebrun, F. Dutfoy, F. Gaudemer, G. Kunesch and A. Gaudemer, Ph tochemistry, 29 (1990) 3777-3783. 294 M.E. Stack, 5.B. Mislivec, J.A.G. Roach and A.E. Pohland, J. Assoc. Off. Anal. Chem., 68 (1985) 640-642. 295 D.J. HaNan and R.W. Pero, J. Chromatogr., 101 (1974) 222-224. 296 P.M. Scott and S.R. Kanhere, J. Assoc. Off. Anal. Chem. , 63 288
(1980) 612-621.
297
D. B. Sauer, L.M. Seitz, R. Burroughs, H. E. Mohr , J. L. West , R.J. Milleret and H.D. Anthony, J. Agric. Food Chem. , 26 (1978)
298
W.M. Hagler and R.F. Behlow, Appl. Environ. Microbiol., 42
1380-1383.
(1981) 1067-1073.
H.P. Broquist, P.S. Mason, W.M. Hagler and T.M. Harris, Appl. Environ. Microbiol.,, 48 (1984) 386-388. 300 H.M. Stahr, P. Martin, W. Hyde and M. Domoto, in: J.L. Richard and J.R. Thurston (Eds.), Diagnosis of Mycotoxicoses, Martinus Nijhoff, Dordrecht, The Netherlands, 1986 p. 283-297. 301 P.M. Scott, W.F. Miles, P. Toft and J.G. buge, J. Agric. Food Chem., 20 (1972) 450-451 302 T. Suzuki, M. Takeda and H. Tanabe, J. Food Hyg. SOC. Jpn., 12 299
303 304 305
306
(1971) 489-494.
J.W. Ralls and R.M. Lane, J. Food Sci., 42 (1977) 1117-1119. A. Andersson and E. Josefsson, Vbr Foda, 31 (1979) 365-374. D.T. Coxon and K.R. Price, in: G.A. Pe in, D.S.P. Patterson and B.J. Shreeve (Eds.), Proceedings oP a Third Meeting on Mycotoxins in Animal Disease, Weybrid e United Kingdom, Ministry of Agriculture, Fisheries and Zobd, Pinner , United Kingdom, 1979, pp. 23-24. Y. Fujimoto, T. Suzuki and Y. Hoshino, J. Chromatogr., 105
3 07 3 08
99-106. ?E.A1975) . Josefsson and A. Andersson, Vbr Fdda, 28 (1976) 189-196. Josefsson and A. Andersson, Arch. Inst. Pasteur Tunis,
309
J.D. Rosen and S.R. Pareles, J. Agric. Food Chem., 22 (1974)
3 10
D. Skierska and W. Martinek, Rocz. Panstw. Zakl. Hig., 35
(1977) 261-268.
1024-1026.
(1984) 69-75.
T. Suzuki, Y. Fu'imoto Y. Hoshino and A. Tanaka, J. Chromatogr., 105 (1475) 96-98. 312 R.W. Pero, D. HaNan, R.G. Owens and J. P. Snow, J. Chromatogr. ,
311 313
65 (1972) 501-506.
A . Bata and R. Lasztity, Elelm. Kozl., 27 (1981) 45-54.
424
Bpta, R. Lasztity and G.S. Sandor, in K. Nacpib, M.M. Na uib, D.L. Park and.A.E. Pohland (Eds.), Proceedings of the Inzernational Symposium on Mycotoxins , Cairo , 6-8 September, 1981, National Research Centre, Cairo, Et;g!sf1983 , pp. 567-572. 3 15 I. Freisz-Gyalmos, G. Babay, K. Abrai and A. Bata, Egeszs>udomdny, 29 (1985 3lZ36. 316 M. Kellert, W. Baltes, W. Biaas and M. Wittkowski, Fresenius Z. Anal. 315 (19831 245-246. . . ..- - . Chem.. -. .- ... . 317 D.N. Mortimeri-I. Parker, M.J. Shepherd and J. Gilbert, Food Addit. Contam., 2 (1985) 165-170. 3 18 J.P. Chaytor and M.J. Saxby, J. Chromatogr., 214 (1981) 135-
3 14
A.
. - - - - I
~~~
~~~~
139.
328 329
H. Ogawa, J. Suzuki, S. Ochiai, M. Nishi'ima, Y. Naoi and T. Nishima, Ann. Rep. Tokyo Metr. Res. Lab. Publ. Health, 35 1984) 224-230. Ehman and G.M. Gaucher, J. Chromato r., 132 (1977) 17-26. S.Z. Lezi, P. Thonart, J.-P. Wathele?, P. Blanpain and J. Bechet, Rev. Fern. Ind. Aliment. , Bruxelles, 36 (1981) 122-129. T. Suzuki, Y. Fujimoto, Y. Hoshino and A. Tanaka, Agric. Biol. Chem., 38 (1974) 1259-1260. K.R. Price, Biomed. Mass Spectrom., 6 1979) 573-574. G.M. Ware, C.W. Thorpe and A.E. Pohlan6, J. Assoc. Off. Anal. Chem. , 57 (1974) 1111-1113. B. Bergner-Lang, M. Kachele and E. Stengel, Dtsche. Lebensm. Rundsch., 79 (1983) 400-404. E.J. Tarter and P.M. Scott, J. Chromatogr., 538 (1991) 441-446. A, Andersson, E. Josefsson, G. Nilsson and K. Akerstrand, VAr Foda, 29 (1977) 292-298. L. Escoula, Ann. Rech. Vbt., 5 (1974) 423-432. J. Keromnes and D. Thouvenot, Appl. Environ. Microbiol., 49
330
C.W. Thorpe and R.L. Johnson, J. Assoc. Off. Anal. Chem., 57
3 19
320 321 322 323 324 325 326 327
5.
(1985) 660-663. (1974) 861-865.
T.D. Phillips, G.W. Ivie, N.D. Heidelbaugh, L.F. Kubena, S.J. C sewski, A.W. Hayes and D.A. Witzel, J. Assoc. Off. Anal. Cgem., 64 (1981) 162-165. 332 C.W. Thorpe, in: H. Egan, L. Stoloff, M. Castegnaro, P. Scott, I.K. O'Neill, H. Bartsch and W. Davis (Eds.) , Environmental Carcinogens. Selected Methods of Analysis. Volume 5 - Some Mycotoxins, International Agency for Research on Cancer, Lyon, France, 1982, pp. 333-348. 3 3 3 C.W. Bacon, J.G. Sweeney, J.D. Robbins and D. Burdick, Appl. Microbiol., 26 (1973) 155-160. 334 J.P. Snow, G,.B. Lucas, D. Harvan, R.W. Per0 and R.G. Owens, Appl. Microbiol., 24 (1972) 34-36. 335 D. Abramson and T. Thorsteinson, J. Assoc. Off. Anal. Chem., 72 331
(1989) 342-344.
341
M. Manabe, M. Minamisawa and S. Matsuura, J. Agric. Chem. SOC. Jpn., 47 (1973) 209-215. Y. Saito, Proc. Jpn. Assoc. Mycotoxicol., (1980) 14-17. A.S. Salhab, G.F. Russell, J.R. Cou hlin and D.P.H. Hsieh, J. Assoc. Off. Anal. Chem., 59 (1976) 7037-1044. Y. Takeda, E. Isohata, R. Amano, H. Tanabe and H. Kurata, J. Food Hyg. SOC. Jpn,., 15 (1974) 363-367. T. Suzuki, Y. Hoshino, Y. Kikuchi, N. Nose and A. Watanabe, J. Food Hy SOC. Jpn., 17 (1976) 253-257. M. Manaze and 0. Tsuruta, Trans. Mycol. SOC. Jpn., 16 (1975)
342
F. Fried1i, J
336 337 338 339 340
.
399-405.
.
(1981) 495-499.
High Res
. Chromatogr., Chromatogr . Commun . , 4
M.W. Trucksess, W.C. Brumley and S. Nesheim, J. Assoc. Off. Anal. Chem., 67 (1984) 973-975. R.T. Rosen J.D. Rosen and V.P. DiProssimo, J. Agric. Food Chem., 32 (1984) 276-278. 345 T. Goto, M. Matsui and T. Kitsuwa, J. Chromatogr., 447 (1988) 343
344 346 347
410-414.
T. Goto, M. Matsui and T. Kitsuwa, Proc. Jpn. Assoc. Mycotoxicol. (1990) 43-47. M. Matsui and 0. Kitsuwa, Jpn. Kokai T o k k p Koho JP 01,219,666 [89,219,666] 1 September 1989, 4 pp.; Chem. Abstr., 112 (1990) 137845e.
425
348
C. Klug, W. Baltes, W. Kronert and R. Weber, in: Proceedings Euro Food Tox 11, 15-18 October 1986, Zurich, Switzerland, pp.
349
C. Klug, W. Baltes, W. Kronert and R. Weber, Z. Lebensm. Unters. Forsch. , 186 (1988) 108-113. F.J.W. van Mansvelt, J.E. Greving and R.A. de Zeeuw, J. Chromatogr., 151 (1978) 113-120. T.A. Plomp, J.G. Leferink and R.A.A. Maes, J. Chromatogr., 151
350 351 352 353 354 355 356 357 358 359 360 361
306-311.
(1978) 121-132.
G. Sze esi and M. Gazdag, J. Chromatogr., 122 (1976) 4.79-485. R.A. Be Zeeuw, F.J.W. van Mansvelt and J.E. Greving, J. Forensic Sci., 22 (1977) 550-557. D.L. Sondack, J. Pharm. Sci., 6 3 1974) 584-587. C.A. Pyle, S.D. Aust, J.W. Ronaidson, P.T. Holland and E. Payne, New Zealand J. Sci., 21 (1978) 379-382. T. Suzuki, M. Kurisu, N. Nose and A. Watanabe, J. Food Hyg. SOC. J n., 22 (1981) 197-202. M. Gidert A. Penel, F.V. Kozikowski, J.D. Henion, G.A. Maylin and D.J. Lisk, J. Food Sci. , 42 (1977) 1650-1653. E.W. Sydenham, W.C.A. Gelderblom, P.G. Thiel and W.F.O. Marasas, J. Agric. Food Chem., 38 (1990) 285-290. M.A. Jackson and G.A. Bennett, Appl. Environ. Microbiol., 56 (1990) 2296-2298.
R.D. Plattner, D.C. Shackelford, L.G. Rice and P.F. Ross 104th Annual International Meeting of the Association of Official Analytical Chemists, 10-13 September 1990, New Orleans, LA, Abstract no. 209. T.-C. Tseng, C.-S. Chung and I. Li, Bot. Bull. Acad. Sinica, 31 (1990) 169-174.
364
M. Margosis, J. Chromatogr., 70 (1972) 73-80. D.L. Doerfler, C.P. Nulton, C.D. Bartman, F.J. Gottlieb and I.M. Cam bell, Can. J. Microbiol., 24 (1978) 1490-1501. H.D. Got?stein, M.N. Zook and J.A. Kuc, J. Chromatogr., 481
365 366
J.M.L. Mee and R.W. Stanley, J. Chromatogr. , 76 (1973) 242-243. H. Bruckner and M. Przybylski, Chromatographia, 19 (1984) 188-
367
H. Bruckner and C. Reinecke, J. High Res. Chromatogr., 12 ~.
362 363
(1989) 55-61.
199.
(1989) 113-116.
H. Bruckner, B. Gehrlein and M. Kiess, in: H. Chmiel, W.P. Hammes and J.E. Bailey (Eds.), Biochemical Engineering, Gustav Fischer Verlag, Stuttgart, 1987, pp. 450-453. 369 H. Bruckner and C. Reinecke, J. High Res. Chromatogr. Chromatogr. Commun. 11 (1988) 735-738. 370 D. Broce, Diss. Abstr. Int. B., 30 (1970) 3059-3060. 371 D.P. Belesky, J.D. Robbins, J.A. Stuedemann, S.R. Wilkinson and O.J. Devine, Agron. J., 79 (1987) 217-220. 372 S.G. Yates, R.J. Petroski and R.G. Powell, J. Agric. Food Chem. , 38 (1990) 182-185. 373 P. Krogh, D.H. Christensen, B. Hald, B.Harlou, C. Larsen, E.J. Pedersen and U. Thrane, Appl. Environ. Microbiol. , 55 (1989) 368
3184-3188.
W. Xie, C.J. Mirocha, R.J. Pawlosky, Y. Wen and X. Xu, Appl. Environ. Microbiol., 55 (1989) 794-797. 375 W. Xie C.J. Mirocha, Y. Wen and R.J. Pawlosky, Appl. Environ. Microbiol. , 56 (1990) 2946-2948. 374
This Page Intentionally Left Blank
427 SUBJECT INDEX 3-AC DON (see 3-acetyldeoxynivalenol) 15-acetoxyverruculogen - TLC 206 3-acetyldeoxynivalenol - HPLC 296, 297 - TLC 174 15-acetyldeoxynivalenol - TLC 176 aflatoxicol - TLC 157, 159, 160 aflatoxin - biosynthesis - - TLC 153-156 - solid phase extraction 14-15 aflatoxin B1 - HPLC 292, 293 - HPTLC 30-31 - TLC 156, 157, 158,159, 160, 167, 168, 223 - - preparative 162 aflatoxin B2 - HPLC 291, 292, 293 - TLC 156, 157, 158,159, 167 aflatoxin BZa - HPLC 292, 293 - TLC 157, 159, 160 aflatoxin D1 - TLC 159 aflatoxin G1 - HPLC 291, 292, 293 - TLC 157, 158, 159, 160 aflatoxin G2 - HPLC 292, 293 - TLC 157, 158, 159 aflatoxin G2a - HPLC 292, 293 - TLC 159, 160 aflatoxin M1 - clean-up 144-146 - HPLC 291, 292 - immunoaffinity chromatography (see also IAC) 120 - TLC 157, 160, 161, 162 aflatoxin M2 - TLC 161 aflatoxin Ma, - HPLC 290, 292 aflatoxin M4 - HPLC 294 aflatoxin Q1 - TLC 157 aflatoxins - clean-up 90, 147-149 - - by IAC 110, 113-116
-
ELISA 132, 294 extraction 144-147 GC 405 - HPLC 290-294, 304 - - detection 293 - IAC 116-121 - TLC 143-162, 224, 225, 226, 227, 228 - - adsorbents 149 - - detection 152 - - in foods and feeds 156 - - solvent systems 150 aflatrem - HPLC 311 - TLC 205, 207 aflavazol - HPLC 311 aflavinines - HPLC 311 a1tenuene - GC 396 - TLC 210 altenuenes - HPLC 313, 314 Alternaria toxins - extraction and clean-up 209 - GC 396 - HPLC 312-314 - TLC 209 alternariol - GC 396 - TLC 210 alternariol methyl ether - GC 396 - TLC 210 - GC 396 alternariols - HPLC 313, 314 altertoxin - HPLC 313, 314 - TLC 210 antibodies - monoclonal 125 - polyclonal 124 aranotin - extraction and clean-up 204 - TLC 205 aspertoxin - TLC 224 aspochalasins - TLC 194 aurantioskyrin - TLC 202 aurasperones - TLC 219
428
auroskyrin - T L C 202 averantin - T L C 153 avermutin - HPLC 296 aversin - HPLC 296 averufanin - T L C 153 averufin - HPLC 296 - T L C 153, 166, 168 baccharin B5 - GC 376 baccharinoids - T L C 177 baccharins - HPLC 298 brefeldin A - HPLC 302 - T L C 186 brevianamide A - T L C 228 butenolide - GC 409 - HPLC 301 - T L C 184 calonectrin - T L C 174 catenarin - T L C 202 chaetoglobosins - T L C 195 chromatography - column 40-44, 287 - flash 43 - gas (see also G C ) 78-96 - high performance liquid (see also H P L C ) 46-71 - immunoaffinity ( I A C ) 99-116 - - sample preparation 112 - liquid column 40-44 - mini-column 44-46 - paper 141 chrysophanol - T L C 202 citreoviridin - H P L C 301 - T L C 184, 228 citrinin - extraction and clean-up 211 - H P L C 304, 305, 306 - T L C 210, 212, 213, 224, 225, 226, 227, 228, 229 - - detection 212 clean-up 4-7, 12-16, 37-40, 87-91 - aflatoxins 147-149 - mycotoxins 12-16, 37-40,
87-91 patulin 90 penicillic acid 90 trichothecenes 169 curvularin - H P L C 302 cyanein (see brefeldin A ) cyclochlorotine - T L C 315 a-cyclopiazonic acid - E L I S A 134 - extraction and clean-up 214 - GC 412 - HPLC 318 - T L C 214, 216, 222, 225, 226 , 228 - - detection 215 cyclosporin A - H P L C 314, 315 - T L C 222 cytochalasans - T L C 191 cytochalasins - H P T L C 193 - T L C 191 DAS (see also diacetoxyscirpenol) - GC 375, 378, 381, 384, 388, 389, 390, 393 deoxaphomin - T L C 195 deoxyluteoskyrin - T L C 202 deoxynivalenol (see also DON and vomitoxin) - H P L C 296 - T L C 174, 175 deoxyrubroskyrin - T L C 202 6-deoxyversicolorin A
-
-
H P L C 296
derivatization - in G C 91 detection - bioautography 172 - in GC 91-96 - in H P L C 61-71 - in T L C 27-28 - trichothecenes in T L C 170-173 diacetoxyscirpenol (see also D A S ) - GC 374, 378 - T L C 223 3,15-diacetyldeoxynivalenol
-
T L C 176 dianhydrorugulosin - T L C 202 dicatenarin - T L C 202
429
dihydrosterigmatocystin - GC 405 dihydroxyaflavinine - HPLC 311 Dis-trichothecenes - TLC 175 DON (see also deoxynivalenol and vomitoxin) - GC 374, 376, 377, 378, 379,
-
380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 393 HPLC 296, 297, 298
echinulin - TLC 221 ELISA 124-134 - aflatoxins 132, 294 - direct 130 - fumonisins 134 - indirect 130 - instrumentation 128 - mycotoxins 132 - ochratoxin A 134 - practice 129 - rubratoxin 134 - sample preparation 127 - trichothecenes 133 emodin - HPLC 307 - TLC 201, 202 enzyme immunoassays 126 epicorazines - TLC 205 epipolythiopiperazine3,6-diones - HPLC 308 - TLC 203
12,13-epoxytrichothec-9-ene
- TLC
174
ergot alkaloids - GC 406 extraction 87 - aflatoxins 144-147 - liquid/liquid 13 - solid phase 13-14, 38-40, 88
flash chromatography 43 flavasperone - TLC 219 fonsecin monomethyl ether - TLC 219 fumagillin - HPLC 317 fumitremorgen A - HPLC 311 fumitremorgen B - HPLC 310, 311 fumitremorgen C - HPLC 311 fumitremorgins
- TLC 206, 207, fumonisins - ELISA 134 - GC 409 - HPLC 315
225
fusarenone X (see also FUS-X and FX) - GC 376, 377 - HPLC 296 - TLC 176 fusaric acid - HPLC 317 - TLC 221 fusarin C - GC 410 - TLC 221 fusarins - HPLC 315, 316 fusarochromanone - GC 413 - HPLC 316 FX (see also fusarenone-X) - GC 378, 380, 381, 382, 383, 390
gas chromatography (see also GC) 78-96 - mycotoxins 373-414 GC - aflatoxins 405 - altenuene 396 - Alternaria toxins 396 - alternariol 396 - alternariol monomethyl ether 396 - baccharin B5 376 - butenolide 409 - a-cyclopiazonic acid 412 - DAS 375, 378, 381, 384, 388, 389,390, 393
- derivatization 9 1 - - trichothecenes 374-382 - detection 91-96
- -
by ECD 9 2
- - by FID 9 2 - diacetoxyscirpenol -
-
-
(see also DAS) 374, 378 dihydrosterigmatocystin 405 DON 374, 376, 377, 378,
379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 393 ergot alkaloids 406 fumonisins 409 fusarenone-X 376, 377, 378, 380, 381, 382, 383, 390 fusarin C 410 fusarochromanone 413 griseofulvin 411 HT-2 toxin 374, 375, 378, 379, 388, 381, 384, 388,
430
389, 390, 393
- isoaltenuene
-
-
-
-
-
396
kojic acid 411 loline alkaloids 413 moniliformin 395 15-monoacetoxyscirpenol 377, 378, 380 mycophenolic acid 411 p-nitropropionic acid 409 ochratoxin A 412 oxalic acid 412 patulin 399 penicillic acid 401 rrpeptaibol"polypeptides 412 roridins 376 satratoxins 376 scirpentriol 381, 384, 387, 388, 391
- select ion monitoring (SIM)
-
-
94 slaframine 398 sporidesmins 408 sterigmatocystin 404 swainsonine 398 T-2 tetraol 381, 384, 387,
388, 391 T-2 toxin 374, 375, 376, 378, 379, 381, 382, 384, 386, 389, 390, 391, 393
T-2 trio1 374
tenuazonic acid 397 terreic acid 411 terrein 411 triacetoxyscirpenol 381 - trichothecenes 373-391 - - derivatization 374-382 - trichothecin 374, 375 - trichothecolone 376 - verrucarin A 376 - verrucarol 376, 378, 384, 391
- zearalenols
392, 393, 394 zearalenone 392, 393, 394, 395 GC-MS 94
-
- trichothecenes 374 - trichothecin 375 gibberellins - HPLC 317 gliotoxin - HPLC 308 - TLC 204
gliotoxins - extraction and clean-up 203 glycopeptides - HPLC 315 griseofulvin - GC 411 - HPLC 299
haptens 125 high performance - thin-layer chromatography (see HPTLC) - liquid chromatography (see HPLC) HPLC 46-71 - aflatoxin B1 292, 293 - aflatoxin B2 291, 292, 293 - aflatoxin B2a 292, 293 aflatoxin G1 291, 292, 293 - aflatoxin G2 292, 293 aflatoxin G2a 292, 293 - aflatoxin M1 291, 292 aflatoxin M2a 290, 292 aflatoxins 290-294, 304 aflatrem 311 aflavazol 311 aflavinines 311 - altenuenes 313, 314 Alternaria toxins 312-314 - alternariols 313, 314 - altertoxins 313, 314 avermutin 296 - averufin 296 brefeldin A 302 - butenolide 301 - citreoviridin 302 - citrinin 304, 305, 306 - curvularin 302 - a-cyclopiazonic acid 318 - cyclosporin A 314, 315 - deoxynivalenol (see also DON and vomitoxin) 296 6-deoxyversicolorin A 296 - detection 61-71 - dihydroxyaflavinine 311 - diode array detection 65 - DON (see also deoxynivalenol and vomitoxin) 296, 297, 298 emodin 307 epipolythiopiperazine3,6-diones 308 - fluorescence detection 64 - fumagillin 317 - fumitremorgen A 311 - fumitremorgen B 310, 311 - fumitremorgen C 311 - fumitremorgins 311 - fumonisins 315 - fusaric acid 317 - fusarins 315, 316 - fusarochromanones 316 - gibberellins 317 - gliotoxin 308 - glycopeptides 315 - griseofulvin 299 - hydroxyanthraquinones 307,
-
-
-
308
43 1
- instrumentation 48-52 - janthitrems 310 - leporin 311
- verrucosidin 301
- luteoskyrin 307 - macrocyclic lactones 302 - mass spectrometry detectioi1
-
- lolitrems 310
68 moniliformin 315, 317 monorden 302 multi-mycotoxin analyses 319-320 - mycophenolic acid 301 - p-nitropropionic acid 317 - nivalenol (NIV) 296, 297, 298 - nominine 311 - normal-phase (NP-HPLC) 52-53 - 1-0-acetyl paxilline 311 - ochratoxins 303 - paspaline 311 - paspalinine 311 - patulin 299, 320 - paxilline 311 - penicillic acid 3 0 0 , 305, 306 , 320 - penitrems 309, 310 - phomopsin A 315 - physcion 307 - PR-toxin 319 - refractive index detection 62 - retention indices - - mycotoxins 322-353 - - secondary metabolites 322-353 - reversed-phase (RP-HPLC) 53, 60 - roquefortine C 318 - roridins 298 - rubratoxin B 305, 307 - rubratoxins 306 - rugulosin 307 - satratoxins 298 - secalonic acid D 308 - small lactones 299-302 - sterigmatin 296 - sterigmatocystins 295 - T-2 toxin 296, 297, 298 - tenuazonic acid 313, 314, - terrestric acid 305 - territrems 312 - tremorgenes 309-312 - trichothecenes 296-299 - trichoverrols 298 - tryptoquivalins 312 - tubingensins 311 - UV-vis detection 62 - verrucarins 298
-
-
verruculogen 310, 311
- versicolorin A 296 - versicolorin c 296
viomellein 305, 306 xanthomegnin 305, 306 zearalenone 302, 304, 305, 306, 320 HPLC-MS - mycotoxins 354 HPTLC - 21-23, 31 - aflatoxin B1 30-31 HT-2 toxin - GC 374, 375, 378, 379, 380, 381, 384, 388, 389, 390, 393 - TLC 175, 224 hyalodendrin - TLC 205 hyalodendrins - extraction and clean-up 203 4-hydroxy-ochratoxin A - TLC 197 hydroxyanthraquinones - HPLC 307, 308 - TLC 200 immunoaffinity chromatography (IAC) 99-116 - practice 105 - sample preparation 112 - theory 101 immunoassays 126 iridoskyrin - TLC 202 islandicin - TLC 202 isoaltenuene - GC 396 isotrichodermin - TLC 174 janthitrems - HPLC 310 - TLC 206, 208 kojic acid - ELISA 134 - GC 411 - TLC 224 lactones - macrocyclic - - TLC 186-196 - small - - TLC 178-186 LC-MS 68 leporin - HPLC 311 liquid column chromatography (see also LC) 36-77 loline alkaloids - GC 413
432
lolitrem - TLC 207 lo1itrems HPLC 310 luteoskyrin HPLC 307 TLC 202 TLC 226, melinacidins TLC 204
5-methoxysterigmatocystin - TLC 163, 165 0-methylsterigmatocystin - TLC 224
mini-column chromatography 44-46 - principles 289 mnorden - HPLC 302 moniliformin GC 395 HPLC 315, 317 TLC 220, 229, 15-monoacetoxyscirpenol (see also 15-MAS) - GC 377, 378, 380, 386, 388 monoclonal antibodies 125 multi-mycotoxin analyses HPLC 319-320 TLC 222 mycophenolic acid clean-up 183 extraction 183 - GC 411 HPLC 301 TLC 183 mycotoxins clean-up 12-16, 37-40, 87-91 gas chromatography (see also GC) 373-414 HPLC retention indices 322-353 HPLC-mass spectrometry (HPLC-MS) 354 liquid column chromatography 253-355 - list of 255-287 paper chromatography 141 UV spectral data 322-353 naphtopyrones TLC 219 neosolaniol - TLC 224 p-nitropropionic acid - GC 409 - HPLC 317 N I V (see also nivalenol) - GC 376, 377, 378, 379, 380, 381, 382, 383, 384, 385,
-
-
-
-
386, 387 390 , 391, 393 HPLC 296 297 , 298 niva lenol see a so N I V ) - HPLC 296 297, 298 nominine - HPLC 311 norsolorinic acid - TLC 153, 164, 166 1-0-acetyl paxilline HPLC 311
-
~
- TLC 153, 164, ochratoxin A - clean-up 7-9
0-methylsterigmatocystin
-
165, 167
extraction 7 GC 412 HPLC 303, 304, 305, 306, 3 20 TLC 197, 223, 224, 225, 226, 227, 228 ochratoxin B TLC 197 ochratoxin C TLC 197 ochratoxins HPLC 303 TLC 196, 222, 224 - - clean-up 196 detection 197 - - extraction 195 quantitation 198 oxalic acid GC 412 4a-oxyluteoskyrin TLC 202 paper chromatography mycotoxins 141 paspalicine TLC 205, 207 paspal ine HPLC 311 TLC 205 paspalinine HPLC 311 TLC 207 paspalitrems TLC 205, 207 patulin clean-up 90, 178 - extraction 178 - GC 399 - HPLC 299, 320 - TLC 178, 223, 224, 225, 226, 227 paxi11ine - HPLC 311 - TLC 206, 207 penicillic acid - clean-up 90, 181 - extraction 88, 181
-
---
433
- GC
-
rugulosin
401
- HPLC
HPLC 3 0 0 , 3 0 5 , 3 0 6 , 3 2 0 TLC 1 8 1 , 2 2 3 , 2 2 4 , 2 2 6 , 227,
-
228
penitrems - HPLC 3 0 9 , 3 1 0 - TLC 2 0 6 . 2 0 7 . 2 0 8 , 2 2 5 , pepta ibol polypeptides . - GC 4 1 2 phomopsin A - HPLC 3 1 5 physcion - HPLC 3 0 7 polyclonal antibodies 1 2 4 PR toxin - HPLC 3 1 9 - TLC 2 1 7 protophomin - TLC 1 9 5 proxiphomin - TLC 1 9 5 PTLC 2 3 - 2 7 punicoskyrin - TLC 2 0 2 quantitation - in TLC 2 8 - 3 0 rhodoislandin A - TLC 2 0 2 rhodoislandin B - TLC 2 0 2 roquefortine C - HPLC 3 1 8 roquefortine - TLC 2 1 7 , 2 2 5 roquefortins - TLC 2 2 8 roridin A - ELISA 1 3 4 roridins - GC 3 7 6 - HPLC 2 9 8 - TLC 1 7 6 , 2 2 5 roseoskyrin - TLC 2 0 2 RPTLC 2 0 - 2 1 rubratoxin - ELISA 1 3 4 rubratoxin B - HPLC 3 0 5 , 3 0 7 - TLC 2 2 3 rubratoxins - HPLC 3 0 6 - TLC 1 9 9 rubrofusarin - TLC 2 1 9 rubroskyrin - TLC 2 0 2 rugulin - TLC 2 0 2
228
307
TLC 2 0 1 sample - extraction 4 - 7 preparation 3 - 4 , 3 7 sampling 3-4 satratoxins - GC 3 7 6 - HPLC 2 9 8 - TLC 1 7 7 scirpentriol - GC 3 8 1 , 3 8 4 , 3 8 7 , 3 8 8 , 3 9 1 secalonic acid - HPLC 3 0 8 - TLC 2 2 2 secalonic acids - TLC 2 1 9 secondary metabolites - HPLC retention indices 322-353
- UV spectral data
322-353
sirodesmins - TLC 2 0 4 skyrin - TLC 2 0 2 slaframine - GC 3 9 8 small lactones - HPLC 2 9 9 - 3 0 2 - TLC 1 7 8 - 1 8 6 sporidesmin - HPLC 3 0 8 sporidesmins - extraction and clean-up - GC 4 0 8 - TLC 2 0 4 sterigmatin - HPLC 2 9 6 sterigmatocystin - clean-up 1 6 2 - ELISA 1 3 4 - extraction 1 6 2 - GC 4 0 4 - HPLC 3 2 0 - TLC 1 5 3 , 1 6 5 , 1 6 7 , 1 6 8 , 224,
225,
204
227, 228 166
- detection
sterigmatocystins - HPLC 2 9 5 - TLC 1 6 2 - 1 6 8 swainsonine - GC 3 9 8 T-2 tetraol - GC 3 8 1 , 3 8 4 , 3 8 7 , T-2 toxin - GC 3 7 4 , 3 7 5 , 3 7 6 , 381, 390,
- HPLC
382, 384, 391,393 296, 297,
388,
391
378, 379, 386, 389, 298
434
-
TLC 174, 175, 224, 225, 228 T-2 trio1 - GC 374 TLC 175 tenuazonic acid GC 397 - HPLC 313 314 - TLC 210 terreic ac d GC 411 terrein GC 411 terrestric acid HPLC 305 territrems HPLC 312 territrems TLC 209. tetrahydrodeoxyaf latoxin B1 TLC 160 thin-layer chromatography (see also TLC) techniques 16-31 TLC 3-acetoxydeoxynivalenol 174 15-acetoxyverruculogen 206 15-acetyldeoxynivalenol 176 aflatoxicol 159, 160 aflatoxicol H 157 aflatoxin B1 $56, 157, 158, 159, 162 160, 167, 168, 223 aflatoxin B2 156, 157, 158, 159, 167 aflatoxin B2a 157, 159, 160 - aflatoxin D1 159 aflatoxin G1 157, 158, 159, 160 aflatoxin G2 157, 158, 159 - aflatoxin M1 157, 160, 161, 162 aflatoxin M2 161 - aflatoxin Q1 157 aflatoxins 143-162, 224, 225, 226, 227, 228 aflatoxins - clean-up 147-149 - extraction 144-147 - aflatrem 205, 207 altenuene 210 - Alternaria toxins 209-210 - alternariol 210 alternariol methyl ether 210 - altertoxin 210 - anti-circular 19-20 - aranotin 205 aspertoxin 224 - aspochalasins 194 - aurantioskyrin 202 - aurasperones 219
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
auroskyrin 202 averantin 153 averufanin 153 averufin 153, 166, 168 baccharinoids 177 bi-directional 18-19 bioautographic detection 28 brevianamide A 228 butenolide 184 calonectrin 174 catenarin 202 chaetoglobosins 195 chemotaxonomy 230 chrysophanol 202 circular 19-20 citreoviridin 184, 228 citrinin 210-213, 224, 225, 226, 227, 228, 229 cyclochlorotine 315 a-cyclopiazonic acid 214-216, 225, 226, 228 cyclosporin A 222 cytochalasans 191 deoxaphomin 195 deoxynivalenol 174, 175 deoxyrubroskyrin 202 detection 27-28 3,15-diacetyldeoxynivalenol 176 dianhydrorugulosin 202 dicatenarin 202 dicatenarin 202 Dis-trichothecenes 175 echinulin 221 emodin 201, 202 epicorazines 205 epipolythiopiperazine3,6-diones 203-204 12,13-epoxytrichothec-9-ene 174 flavasperone 219 fonsecin monomethyl ether 219 fumitremorgins 206, 207 fumitremorgins 225 fusarenone-X 176 fusaric acid 221 fusarin C 221 gliotoxin 204 high performance (see also HPTLC) 21-23 HT-2 toxin 175, 224 hyalodendrins 205 4-hydroxy-ochratoxin 197 hydroxyanthraquinones 200-202 iridoskyrin 202 islandicin 202 isotrichodermin 174 janthitrems 206, 208
435
- kojic acid
-
224
lolitrem 207 luteoskyrin 202, 226 macrocyclic lactones 186-196
- macrocyclic trichothecenes 176 - melinacidins 204
-
5-methoxysterigmatocystin
-
0-methylsterigmatocystin
163, 165 153, 164, 165, 167, 224 220, 229
- moniliformin
-
multimycotoxin methods 222-230
mycophenolic acid 183 naphtopyrones 219 neosolaniol 224 normal phase 16-20 ochratoxin A 197 ochratoxin A 225, 227, 228 - ochratoxin B 197 - ochratoxin C 197 - ochratoxins 196-198, 224 - one dimensional 17 , - 4a-oxyluteoskyrin 202 - paspalicine 205, 207 - paspaline 205 - paspalinine 207 - paspalitrems 205, 207 - patulin 178, 223, 224, 225,
-
226, 227
paxilline 206, 207
- penicillic acid
-
-
-
-
-
-
-
181, 223, 224, 226, 227, 228 penitrems 206, 207, 208, 225, 228 PR toxin 217 preparative (see also PTLC) 23-27 protophomin 195 proxiphomin 195 punicoskyrin 202 quantitative 28-30
reverse-phase (see also
RPTLC) 20-21
rhodoislandin A 202 rhodoislandin B 202 roquefortine 217, 225, 217, 228
roridins 176, 225 roseoskyrin 202 rubratoxin B 223 rubratoxins 199-200 rubrofusarin 219 rubroskyrin 202 rugulin 202 rugulosin 201 satratoxins 177 secalonic acid 219, 222
- semiquantitative - sirodesmins 204 - skyrin 202
-
28-30
small lactones 178-186
- sporidesmins 204 - sterigmatocystin -
-
-
-
153, 165, 167, 168, 224, 225, 227, 228 sterigmatocystins 162-167 T-2 toxin 174, 175, 224, 225, 228 T-2 trio1 175 tenuazonic acid 210 territrems 209 tetrahydrodeoxyaflatoxin B1 168 TR-2 206 tremorgenic toxins 205-208 triangular 20 trichothecenes 168-177, 229, 230 two-dimensional 17-19 verrucarins 176 verruculogen 206, 207 versicolorin A 153, 166 versicolorin B 166 versicolorin C 166
versiconal hemiacetal acetate 153, 166 viomellein 218 viomellein 218 vioxanthin 218 viridin 221 wortmannin 221 xanthomegnin 218 xanthomegnin 218 zearalenols 229 zearalenone 186, 188, 190,
223, 224, 225, 227, 228, 229 zygosporins 194 TR-2 - TLC 206
-
tremorgenes
-
HPLC 309-312
-- TLC 205 - detection
205
triacetoxyscirpenol
-
GC 381
trichothecenes - bioautography in TLC 172 - chemical classification 168 - clean-up 89, 169 - Dis esters
- TLC 174, - ELISA 133 - extraction - GC 93, 9 5
-
175 169
derivatization 374-382
436
- detection without
-
derivatization 374 heptafluorobutyrylation 378
pentafluoropropionyl derivatives 381 - trifluoroacetates 381 - trimethylsilylation 375 - mass spectrometry (see also GC-MS) 374 - macrocyclic
-
HPLC 298 TLC 176
- non-macrocyclic - HPLC 296-298 - TLC 168-177, 229, 230 - detection 170-173 - type A - ELISA after HPLC 297 trichothecin - GC 374, 375 - GC-MS 375 trichothecolone - GC 376 trichoverrols
- HPLC
298
tryptoquivalins
- HPLC 312 tubingensins - HPLC 311
UV spectral data
- mycotoxins 322-353 - secondary metabolites 322-353
verrucarin A
-
GC 376
verrucarins
- HPLC 298 - TLC 176
verrucarol
-
GC 376, 378, 384, 391
verrucosidin HPLC 301
verruculogen
- HPLC 310, 311 - TLC 206, 207 versicolorin A - HPLC 296
-
TLC 153, 166
-
TLC 166
-
HPLC 296 TLC 166
versicolorin B
versicolorin C
-
versiconal hemiacetal acetate
-
TLC 153, 166
viomellein
- HPLC 305 - TLC 218 vioxanthin - TLC 218 viridin - TLC 221
306
vomitoxin see also deoxynivalenol and DON)
-
HPLC 296, 297
wortmannin
- TLC
221
xanthomegnin
-
HPLC 305, 306
- TLC
218
zearalenols
- GC
392, 393, 394
- TLC 229 zearalenone - clean-up 187 - extraction 87, 187 - GC 392, 393, 394, 395 - HPLC 302, 304, 305, 306, 320 - TLC 186, 223, 224, 225, 227, 228, 229 - detection 188 - quantitation 190 zygosporins - TLC 194
437
JOURNAL OF CHROMATOGRAPHY LIBRARY A Series of Books Devoted to Chromatographic and Electrophoretic Techniques and their Applications Although complementary to the Journal OfChromatography,each volume in the Library Series is an important and independent contribution in the field of chromatography and electrophoresis. The Library contains no material reprinted from t h e journal itself.
Other volumes in this series Volume 1
Chromatography of Antibiotics (seealso Volume 26) by G.H. Wagman and M.J. Weinstein
Volume 2
Extraction Chromatography edited by T. Braun and G. Ghersini
Volume 3
Liquid Column Chromatography.A Survey OfModernTechniques and Applications edited by Z. Deyl, K. Macek and J . Janak
Volume 4
Detectors in Gas Chromatography by J . SevEik
Volume 5
Instrumental Liquid Chromatography.A Practical Manual on HighPerformance Liquid Chromatographic Methods (see alsoVolume 27) by N.A. Parris
Volume 6
Isotachophoresis. Theory, Instrumentation and Applications by F.M. Everaerts, J.L. Beckers andTh.P.E.M.Verheggen
Volume 7
Chemical Derivatization in Liquid Chromatography by J.F. Lawrence and R.W. Frei
Volume 8
Chromatographyof Steroids by E. Heftmann
Volume 9
HPTLC -High Performance Thin-LayerChromatography edited by A. Zlatkis and R.E. Kaiser
Volume 10
Gas Chromatography of Polymers byV.G. Berezkin,V.R. Alishoyev and I.B. Nemirovskaya
Volume 11
Liquid Chromatography Detectors (see also Volume 33) by R.P.W. Scott
Volume 12
Affinity Chromatography by J.Turkova
Volume 13
Instrumentation for High-Performance Liquid Chromatography edited by J.F.K. Huber
Volume 14
Radiochromatography.The Chromatography and Electrophoresis of Radiolabelled Compounds byT.R. Roberts
Volume 15
Antibiotics. Isolation, Separation and Purification edited by M.J. Weinstein and G.H. Wagman
438 Volume 16
Porous Silica. Its Properties and Use as Support in Column Liquid Chromatography by K.K. Unger
Volume 17
75Years of Chromatography -A Historical Dialogue edited by L.S. Ettre and A. Zlatkis
Volume 18A
Electrophoresis. A Survey of Techniques and Applications. Part A. Techniques edited by Z. Deyl
Volume 18B
Electrophoresis. A Survey of Techniques and Applications. Part B: Applications edited by Z. Deyl
Volume 19
Chemical Derivatization in Gas Chromatography by J. Drozd
Volume 20
Electron Capture.Theory and Practice in Chromatography edited by A. Zlatkis and C.F. Poole
Volume 21
Environmental Problem Solving using Gas and Liquid Chromatography by R.L. Grob and M.A. Kaiser
Volume 22A
Chromatography. Fundamentals and Applications of Chromatographic and Electrophoretic Methods. Part A Fundamentals andTechniques edited by E. Heftmann
Volume 22B
Chromatography. Fundamentals and Applications of Chromatographic and Electrophoretic Methods. Part B: Applications edited by E. Heftmann
Volume 23A
Chromatography of Alkaloids. Part A Thin-Layer Chromatography by A. Baerheim Svendsen and R. Verpoorte
Volume 23B
Chromatography of Alkaloids. Part B: Gas-Liquid Chromatography and High-Performance Liquid Chromatography by R. Verpoorte and A. Baerheim Svendsen
Volume 24
Chemical Methods in Gas Chromatography by V.G. Berezkin
Volume 25
Modern Liquid Chromatography of Macromolecules by B.G. Belenkii and L.Z. Vilenchik
Volume 26
Chromatography of Antibiotics. Second, Completely Revised Edition by G.H. Wagman and M.J. Weinstein
Volume 27
Instrumental Liquid Chromatography. A Practical Manual on HighPerformance Liquid Chromatographic Methods. Second, Completely Revised Edition by N.A. Parris
Volume 28
Microcolumn High-Performance Liquid Chromatography by P. Kucera
Volume 29
Quantitative Column Liquid Chromatography. A Survey of Chemometric Methods by S.T. Balke
Volume 30
Microcolumn Separations. Columns, Instrumentation and Ancillary Techniques edited by M.V. Novotny and D. Ishii
Volume 31
Gradient Elution in Column Liquid Chromatography. Theory and Practice by P. Jandera and J. ChuraEek
Volume 32
The Science of Chromatography. Lectures Presented at the A.J.P. Martin Honorary Symposium, Urbino, May 27-31,1985 edited by F. Bruner
Volume 33
Liquid Chromatography Detectors. Second, Completely Revised Edition by R.P.W. Scott
Volume 34
Polymer Characterization by Liquid Chromatography by G. Glockner
Volume 35
Optimization of Chromatographic Selectivity. A Guide to Method Development by P. J . Schoenmakers
Volume 36
Selective Gas Chromatographic Detectors by M. Dressler
Volume 37
Chromatography of Lipids in Biomedical Research and Clinical Diagnosis edited by A. Kuksis
Volume 38
Preparative Liquid Chromatography edited by B.A. Bidlingmeyer
Volume 39A
Selective Sample Handling and Detection in High-Performance Liquid Chromatography. Part A edited by R.W. Frei and K. Zech
Volume 39B
Selective Sample Handling and Detection in High-Performance Liquid Chromatography. Part B edited by K . Zech and R.W. Frei
Volume 40
Aqueous Size-Exclusion Chromatography edited by P.L. Dubin
Volume41A
High-Performance Liquid Chromatography of Biopolymers and Bio-oligomers. Part A: Principles, Materials and Techniques by 0.MikeS
Volume 41B
High-Performance Liquid Chromatography of Biopolymers and Bio-oligomers. Part B: Separation of Individual Compound Classes by 0.MikeS
Volume 42
Quantitative Gas Chromatography for Laboratory Analyses and On-Line Process Control by G. Guiochon and C.L. Guillemin
Volume 43
Natural Products Isolation. Separation Methods for Antimicrobials, Antivirals and Enzyme Inhibitors edited by G.H. Wagman and R. Cooper
Volume 44
Analytical Artifacts. GC, MS, HPLC,TLC and PC by B.S. Middleditch
440
Volume 45A
Chromatography and Modificationof Nucleosides. Part A: Analytical Methods for Major and Modified Nucleosides- HPLC, GC, MS, NMR, UV and FT-IR edited by C.W. Gehrke and K.C.T. Kuo
Volume 45B
Chromatography and Modificationof Nucleosides. Part B: BiologicalRoles and Function of Modification edited by C.W. Gehrke and K.C.T. Kuo
Volume 45C
Chromatography and Modification of Nucleosides. Part C: Modified Nucleosides in Cancer and Normal Metabolism- Methods and Applications edited by C.W. Gehrke and K.C.T. Kuo
Volume 46
Ion Chromatography. Principles and Applications by P.R. Haddad and P.E.Jackson
Volume 47
Trace Metal Analysis and Speciation edited by I S . Krull
Volume 48
Stationary Phases in Gas Chromatography by H. Rotzsche
Volume 49
Gas Chromatography in Air Pollution Analysis byV.G. BerezkinandYu. S. Drugov
Volume 50
Liquid Chromatography in Biomedical Analysis edited byT. Hanai
Volume 51A
Chromatography, 5th edition. Fundamentals and Applications of Chromatography and Related Differential Migration Methods. Part A: Fundamentals andTechniques edited by E. Heftmann
Volume 51B
Chromatography, 5th edition. Fundamentals and Applications of Chromatography and Related Differential Migration Methods. Part B: Applications edited by E. Heftmann
Volume 52
Capillary Electrophoresis. Principles, Practice and Applications by S.F.Y. Li
Volume 53
HyphenatedTechniques in Supercritical Fluid Chromatography and Extraction edited by K. Jinno
Volume 54
Chromatography of Myocotoxins.Techniques and Applications. edited by V. Betina
