Aflatoxin B(1) and Fumonisin B(1) Co-Contamination: Interactive Effects, Possible Mechanisms of Toxicity, and Decontamination Procedures." (1998)

Home , Fumonisin, Fumonisin B1, Fumonisin B2

Louisiana State University LSU Digital Commons

LSU Historical Dissertations and Theses Graduate School

1998 Aflatoxin B(1) and Fumonisin B(1) Co- Contamination: Interactive Effects, Possible Mechanisms of Toxicity, and Decontamination Procedures. Rebeca Lopez-garcia Louisiana State University and Agricultural & Mechanical College

Follow this and additional works at: https://digitalcommons.lsu.edu/gradschool_disstheses

Recommended Citation Lopez-garcia, Rebeca, "Aflatoxin B(1) and Fumonisin B(1) Co-Contamination: Interactive Effects, Possible Mechanisms of Toxicity, and Decontamination Procedures." (1998). LSU Historical Dissertations and Theses. 6746. https://digitalcommons.lsu.edu/gradschool_disstheses/6746

This Dissertation is brought to you for free and open access by the Graduate School at LSU Digital Commons. It has been accepted for inclusion in LSU Historical Dissertations and Theses by an authorized administrator of LSU Digital Commons. For more information, please contact [email protected]. INFORMATION TO USERS

This manuscript has been reproduced from the microfilm master. UMI films the text directly from the original or copy submitted. Thus, some thesis and dissertation copies are in typewriter face, while others may be from any type o f computer printer.

The quality of this reproduction is dependent upon the quality of the copy submitted. Broken or indistinct print, colored or poor quality illustrations and photographs, print bleedthrough, substandard margins, and improper alignment can adversely affect reproduction.

In the unlikely event that the author did not send UMI a complete manuscript and there are missing pages, these will be noted. Also, if unauthorized copyright material had to be removed, a note will indicate the deletion.

Oversize materials (e.g., maps, drawings, charts) are reproduced by sectioning the original, beginning at the upper left-hand comer and continuing from left to right in equal sections with small overlaps. Each original is also photographed in one exposure and is included in reduced form at the back o f the book.

Photographs included in the original manuscript have been reproduced xerographically in this copy. Higher quality 6” x 9” black and white photographic prints are available for any photographs or illustrations appearing in this copy for an additional charge. Contact UMI directly to order. UMI A Bell & Howell Information Company 300 North Zeeb Road, Arm Arbor MI 48106-1346 USA 313/761-4700 800/521-0600

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. AFLATOXIN B, AND FUMONISIN B, CO-CONTAMINATION: INTERACTIVE EFFECTS. POSSIBLE MECHANISMS OF TOXICITY. AND DECONTAMINATION PROCEDURES

A Dissertation

Submitted to the Graduate Faculty of the Louisiana State University and Agricultural and Mechanical College in partial fulfillment of the requirements for the degree of Doctor of Philosophy

The Department of Food Science

by Rebeca Lopez-Garcia B.S., Universidad La Salle, 1995 August, 1998

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. UMI Number: 9902648

This microform edition is protected against unauthorized copying under Title 17, United States Code.

UMI 300 North Zeeb Road Ann Arbor, MI 48103

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Dedication

To my father, Dr. Raul B. Lopez-Garcia “Toty”, who discussed science

with me at odd hours. His survival through struggle with honesty and courage has

taught me that the fight is worth the forging fire.

To my mother, Mrs. Rebeca Gomez-Deses de Lopez-Garcia ”E1 Lobito”,

whose unblemished soul has shined on me throughout life. Her unquestionable

integrity and uncompromised kindness are an example to live by.

To my niece. Marcela Lopez-Garcia Marcos, bom exactly when this project

was completed. Her light shines on the hope of new beginnings.

111

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Acknowledgements

I wish to express my deepest appreciation to all those people that

contributed to the completion of this project. Special gratitude is due to Dr.

Douglas L. Park, the director of this research, whose encouragement and constant

support were invaluable; and to Dr. Vince Wilson, my minor professor, for his

trust, strong support and priceless help. Special thanks are due to Dr. Wanda Lyon

whose fairness was a stronghold. This gratitude is extended to Dr. Leslie Plhak, Dr.

Barbara Shane and Dr. Don Labonte, who made me search, redefine and forge a

new person.

I would also like to thank Dr. Ralph Price (Department of Nutritional

Science, The University of Arizona), who believed in me, helped me through the

hardest moments and showed me the brightness of the future. Special thanks are

extended to Dr. William B. Richardson for his friendship and continuous support.

Gratitude is also due to Dr. Witoon Prinyawiwatkul who gave me the opportunity

to teach with him and supported me through my semester as a laboratory instructor.

The realization of this project would have been impossible without all the

people that contributed sharing their resources, their equipment and their

knowledge. Special acknowledgement is due to Dr. Bruce Ames and Ms. Carol

Wehr, University of California at Berkley, for providing the bacterial tester strains

and for their technical assistance; Dr. Mary Trucksess and Dr. Robert Eppley, Food

and Drug Administration for providing purified toxins; Mr. Herschel Morris and

Dr. Janet Simonson, LSU Agricultural Chemistry for their assistance in the use of

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. the HPLC; Dr. Ray Schneider, Dr. John Russin. Dr. Kenneth Damann and Ms.

Cheryl Giles, LSU Department of Plant Pathology, for their assistance in carrying

out the storage study; Dr. Leslie Plhak, LSU Food Science, and Dr. David York

and Ms. Gail Kilroy, LSU Pennington Biomedical Research Center for their

assistance in the cell-cell communication assay; Mr. Mike Hooks, LSU Campus

Safety Office for his invaluable assistance, patience and advice on the safe handling

of toxic reagents; Dr. Dennis Hsieh, Dr. Norman Kado and Mr. Paul Kuzmicky,

University of California at Davis, for receiving me in their laboratory for training in

the mutagenicity testing and for their continuous technical support; Dr. David

Kiang and Mr. Rahn Kollander, University of Minnesota Cancer Research Center,

for receiving me in their laboratory for training in the cell-cell communication

assay.

No words can express my gratitude to Mrs. Denise Craig, “The Fascist Lab.

Cop” and Mr. Brett Craig, “Big Man”, who were always there to share the good

and the bad times. In the same way, I wish to thank the mycotoxin team, Dr.

Armando Burgos-Hemandez and Mrs. Karla Zavalza de Burgos, Mr. Henry Njapau

and Mr. Syed S. Ali, for the late discussions, the long weekends in the lab., for

sharing their knowledge and experience, but most of all, for their friendship.

The financial support of El Consejo Nacional de Ciencia y Tecnologia

(CoNaCyT), The Institute of Food Technologists and the American Association of

Cereal Chemists is profoundly appreciated.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Finally, I wish to express my infinite gratitude to my family. I would no be

what I am without their love and guidance.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table of Contents

Dedication iii

Acknowledgements ...... iv

List of Tables xi

List of Figures xiii

Abstract xvii

I. Introduction 1

II. Literature review ...... 4 A. Historical background ...... 4 B. Aflatoxins 8 1. Mycology and occurrence ...... 8 2. Toxin production ...... 11 3. Chemical characteristics...... 12 4. Metabolism ...... 14 a. Absorption and distribution ...... 14 b. Phase I metabolism...... 16 c. Phase II metabolism...... 20 d. Competitive pathways: activation and detoxification ...... 21 5. Toxicity ...... 22 C. Fumonisins ...... 25 1. Mycology and occurrence ...... 25 2. Toxin production ...... 28 3. Chemical characteristics...... 32 4. Metabolism ...... 35 a. Absorption and distribution ...... 35 b. Mode of action ...... 36 c. Carcinogenicity ...... 38 5. Fumonisin Bi related toxicoses ...... 43 a. Hepatotoxicity ...... 43 b. Equine leukoencephalomalacia (ELEM) ...... 45 c. Porcine pulmonary edema (PPE) ...... 47 d. Human esophageal cancer ...... 49 e. Developmental toxicity ...... 51 f. Other diseases...... 55 D. Inhibition of gap-junction intercellular communication as a non-genotoxic mechanism in carcinogenesis ...... 56

vii

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. E. Effects of processing on mycotoxin-contaminated foods ...... 60 1. Physical methods of mycotoxin removal ...... 63 a. Cleaning and segregation ...... 63 b. Wet milling ...... 66 c. Dry milling ...... 68 2. Physical methods of detoxification ...... 70 a. Thermal inactivation ...... 70 b. Irradiation ...... 74 c. Adsorption from solution and covalent binding ...... 75 3. Biological decontamination ...... 76 4. Chemical inactivation ...... 78 5. Other industrial processes ...... 81

III. Preliminary evaluation of the production of aflatoxins and fiimonisins by Aspergillus flavus and Fusarium moniliforme during storage at optimal mold growth conditions ...... 85 A. Introduction ...... 85 B. Materials and methods ...... 87 1. Com, fungal and bacterial cultures ...... 87 2. Chemicals ...... 88 3. Com inoculation ...... 89 4. Safety precautions ...... 90 5. Toxin extraction ...... 91 6. Aflatoxin purification ...... 91 7. Quantification of aflatoxins by liquid chromatography (HPLC) 93 8. Fumonisin purification ...... 94 9. Quantification of fiimonisins by liquid chromatiography (HPLC) 94 10. Mutagenic potential of raw com extracts ...... 95 11. Toxicity to Salmonella typhimurium...... 97 12. Statistical analysis ...... 97 C. Results and discussion ...... 97 1. Aflatoxin production ...... 97 2. Fumonisin production ...... 106 3. Mutagenic potential ...... 108 4. Toxic potential ...... 120

IV. Interactive effects between aflatoxin Bi and fumonisin B i ...... 128 A. Introduction ...... 128 B. Materials and methods ...... 129 1. Bacterial cultures ...... 129 2. Chemicals ...... 130 3. Sample preparation ...... 130 4. Mutagenic potential of aflatoxin/fumonisin combinations ...... 131 5. Toxic potential of aflatoxin/fumonisin combinations ...... 133

viii

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. C. Results and discussion ...... 133 1. Aflatoxin Bi/fumonisin Bi mutagenicity ...... 133 a. Direct interaction hypothesis ...... 138 b. Mechanistic interaction hypothesis...... 139 2. Aflatoxin Bi/fumonisin B 2 mutagenicity ...... 146 3. Toxicity of AFB 1/FB 1 and AFB 1/FB2 ...... 147

V. Possible mechanisms of interaction between aflatoxin Bi and fumonisin Bi ...... 154 A. Introduction ...... 154 B. Materials and methods ...... 155 1. Bacterial cultures ...... 155 2. Chemicals ...... 157 3. Chemical interaction ...... 158 a. Effect of fumonisin Bi on aflatoxin Bi recovery after metabolic activation ...... 158 b. Thin layer chromatography (TLC) of recovered aflatoxin Bi and/or metabolites ...... 159 c. Effect of aflatoxin Bi on fumonisin Bi recovery after metabolic activation ...... 160 d. Statistical analysis ...... 162 4. Effect of fumonisin B 1 on 2-aminofluorene mutagenic potential 162 a. Mutagenic potential of 2-aminofluorene ...... 162 b. Effect of fumonisin B 1 on 2-aminofluorene mutagenicity 162 c. Safety precautions ...... 163 C. Results and discussion ...... 164 1. Effect of fumonisin B 1 on aflatoxin B 1 recovery after metabolic activation ...... 164 2. Effect of aflatoxin Bi on fumonisin Bi recovery after metabolic activation ...... 167 3. Effect of fumonisin Bi on 2-aminofluorene mutagenicity ...... 169

VI. Effect of fumonisin B| on gap-junction intercellular communication of clone-9 rat liver cells ...... 177 A. Introduction ...... 177 B. Materials and methods ...... 179 1. Chemicals ...... 179 2. Cell line ...... 180 3. Test solutions ...... 181 4. Growth curve ...... 181 5. Neutral red cytotoxicity assay ...... 182 6. Gap-junction intercellular communication ...... 183 7. Statistical analysis ...... 184

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. C. Results and discussion ...... 184 1. Growth curve ...... 184 2. Neutral red (NR) cytotoxicity assay ...... 184 3. Effect of fumonisin Bi on gap-junction intercellular communication ...... 189

VII. Effect of hydrogen peroxide and sodium bicarbonate on the toxin content and the toxic/mutagenic potential of aflatoxin- fumonisin- naturally co-contaminated com ...... 195 A. Introduction ...... 195 B. Materials and methods ...... 198 1. Com and bacterial cultures ...... 198 2. Chemicals ...... 198 3. Com treatments ...... 199 4. Aflatoxin analysis ...... 199 5. Fumonisin analysis ...... 200 6. Toxicity/mutagenicity testing ...... 200 7. Salmonella/microsomal mutagenicity assay ...... 200 8. Brine shrimp assay...... 201 9. Statistical analysis...... 201 C. Results and discussion ...... 201

VIII. Reduction of risks associated with mycotoxin contamination ...... 210 A. Introduction ...... 210 B. Integrated mycotoxin management programs ...... 210

IX. Summary and conclusions ...... 223

References 228

Appendix A: List of abbreviations ...... 267

Appendix B: Raw data for interaction study ...... 270

Appendix C: Letter of permission from Marcel Dekker, Inc ...... 276

Vita 277

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. List of Tables

Table 2.1 Summary of selected reports of potentially toxic molds from various food or agricultural commodities ...... 61

Table 3.1 Com inoculation design ...... 90

Table 3.2 Aflatoxin Bi (ng/g) production during storage ...... 99

Table 3.3 Aflatoxin B 2 (ng/g) production during storage ...... 100

Table 3.4 Aflatoxin Gi (ng/g) production during storage ...... 104

Table 3.5 Aflatoxin G 2 (ng/g) production during storage ...... 105

Table 3.6 Fumonisin Bi (pg/g) production during storage ...... 107

Table 3.7 Fumonisin B 2 (pg/g) production during storage ...... 108

Table 4.1 Combinations of AFBi and FBi tested for toxic/mutagenic potential ...... 131

Table 4.2 Combinations of AFB| and FB 2 tested for toxic/mutagenic potential ...... 132

Table 5.1 AFB1/FB 1 combinations tested for potential chemical interactions with metabolic activation ...... 158

Table 5.2 FB1/AFB1 combinations tested for potential chemical interactions with metabolic activation ...... 161

Table 5.3 2-Aminofluorene/fumonisin B| combinations used to evaluate the effect of FBi on 2-AF mutagenic potential ...... 163

Table 5.4 Rf values and description of TLC spots derivatized and underivatized samples ...... 166

Table 6.1 Fumonisin Bi and aflatoxin Bi samples tested for cytotoxicity and effect on gap-junction intercellular communication ...... 181

Table 6.2 Effect of fumonisin Bi on gap-junction intercellular Communication of Clone-9 rat liver cells ...... 189

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 8.1 Current aflatoxin action levels (total) established by the Food and Drug Administration ...... 220

Table 8.2 Medians and ranges in 1995 of maximum tolerated levels (ng/g) for aflatoxins and numbers of countries that have reglations for these ...... 220

Table 8.3 Aflatoxin residues in milk produced in Arizona ...... 222

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. List of Figures

Figure 2.1 Chemical structures of most common naturally-occurring aflatoxins ...... 13

Figure 2.2 Formation of the aflatoxin Bi-N7 guanidine adduct ...... 17

Figure 2.3 Biotransformation pathways for aflatoxin B | ...... 19

Figure 2.4 Structures of the fumonisin A and B analogues ...... 34

Figure 2.5 Structures of fumonisin Bi, sphinganine and sphingosine ...... 37

Figure 2.6 Sphingolipid biosynthesis and turnover ...... 39

Figure 3.1 Mutagenic potential of Aspergillus flavus inoculated com during storage. Salmonella tester strain TA-100 with metabolic activation ...... 109

Figure 3.2 Mutagenic potential of Aspergillus flavus inoculated com during storage. Salmonella tester strain TA-98 with metabolic activation ...... 110

Figure 3.3 Mutagenic potential of Fusarium moniliforme inoculated com during storage. Salmonella tester strain TA-100 with metabolic activation ...... 112

Figure 3.4 Mutagenic potential of Fusarium moniliforme inoculated com during storage. Salmonella tester strain TA-98 with metabolic activation ...... 113

Figure 3.5 Mutagenic potential of Aspergillus flavus/Fusarium moniliforme inoculated com during storage. Salmonella tester strain TA-100 with metabolic activation ...... 115

Figure 3.6 Mutagenic potential of Aspergillus flavus/Fusarium moniliforme inoculated com during storage. Salmonella tester strain TA-98 with metabolic activation ...... 116

Figure 3.7 Mutagenic potential of naturally contaminated com during storage. Salmonella tester strain TA-100 with metabolic activation ...... 118

xiii

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 3.8 Mutagenic potential of naturally contaminated com during storage. Salmonella tester strain TA-98 with metabolic activation ...... 119

Figure 3.9 Mutagenic potential of control com during storage. Salmonella tester strain TA-100 with metabolic activation ...... 121

Figure 3.10 Mutagenic potential of control com during storage. Salmonella tester strain TA-98 with metabolic activation ...... 122

Figure 3.11 Toxic potential of com samples during storage. Salmonella tester strain TA-100 with metabolic activation ...... 124

Figure 3.12 Toxic potential of com samples during storage. Salmonella tester strain TA-98 with metabolic activation ...... 126

Figure 4.1 Mutagenic potential of AFBi/FBj combinations. Salmonella tester strain TA-100 without metabolic activation ...... 134

Figure 4.2 Mutagenic potential of AFB i/FBi combinations. Salmonella tester strain TA-100 with metabolic activation ...... 136

Figure 4.3 Mutagenic potential of AFBj/FBi combinations. Salmonella tester strain TA-98 with metabolic activation ...... 137

Figure 4.4 Mutagenic potential of AFB 1/FB2 combinations. Salmonella tester strain TA-100 without metabolic activation ...... 148

Figure 4.5 Mutagenic potential of AFB 1/FB2 combinations. Salmonella tester strain TA-98 with metabolic activation ...... 149

Figure 4.6 Toxic potential of AFB 1/FB 1 combinations. Salmonella tester strain TA-100 with metabolic activation ...... 150

Figure 4.7 Toxic potential of AFB 1/FB 1 combinations. Salmonella tester strain TA-98 with metabolic activation ...... 151

Figure 4.8 Toxic potential of AFB 1/FB2 combinations. Salmonella tester strains TA-100 and TA-98 with metabolic activation 152

Figure 5.1 Metabolic transformation of 2-Aminofluorene ...... 156

xiv

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 5.2 Aflatoxin Bi recovery after reaction with fumonisin Bi with metabolic activation ...... 165

Figure 5.3 Fumonisin Bi recovery after reaction with aflatoxin Bi with metabolic activation ...... 168

Figure 5.4 Mutagenic potential of 2-aminofluorene. Salmonella tester strain TA-100 with and without metabolic activation 170

Figure 5.5 Mutagenic potential of 2-aminofluorene. Salmonella tester strain TA-98 with and without metabolic activation ...... 171

Figure 5.6 Mutagenic potential of 2-amino fluorene/fumonisin Bi combinations. Salmonella tester strain TA-100 with and without metabolic activation 173

Figure 5.7 Mutagenic potential of 2-aminofluorene/fumonisin Bi combinations. Salmonella tester strain TA-98 with and without metabolic activation ...... 174

Figure 6.1 Growth curves for Clone-9 rat liver cells ...... 185

Figure 6.2 Toxic potential of methanol to Clone-9 rat liver cells ...... 187

Figure 6.3 Toxic potential of fumonisin B[ and/or aflatoxin Bi to Clone-9 rat liver cells ...... 188

Figure 6.4 Effect of fumonisin Bi on gap-junction intercellular communication (GJIC) of Clone-9 rat liver cells ...... 190

Figure 6.5 Effect of aflatoxin Bi on the inhibition of gap-junction intercellular communication (GJIC) of Clone-9 rat liver cells caused by lxlO3 ng FBj/plate ...... 193

Figure 7.1 Chemical reduction of aflatoxin Bi and B 2 , and fumonisin Bi and B 2 naturally contaminated com exposed to H2 O2 and NaHCOs separate or in combination decontamination treatments ..... 202

Figure 7.2 Toxic potentials to brine shrimp of aflatoxin Bi and B 2 , And fumonisin Bi and B 2 naturally contaminated com exposed to H2 O2 and NaHCC >3 separate or in combination decontamination treatments ..... 204

Figure 7.3 Total toxin reduction (total aflatoxin^AFB 1 + AFB2 ; total fumonisin=FBi+FB 2 ) as determined by HPLC compared to the toxic potential of the raw com extracts as determined by brine shrimp after 48 hour exposure ...... 205

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 7.4 Mutagenic potential of com extracts from aflatoxin/fumonisin co-contaminated com exposed to decontamination treatments (H 2 O2 and NaHCC >3 separate or in combination) using tester strain TA-100 without metabolic activation ...... 207

Figure 7.5 Mutagenic potential of com extracts from aflatoxin/fumonisin co-contaminated com exposed to decontamination treatments (H 2 O2 and NaHCC >3 separate or in combination) using tester strain TA-100 with metabolic activation ...... 208

Figure 8.1 Risk assessment analysis and food safety program for naturally occurring toxicants ...... 212

Figure 8.2 Development of decontamination/detoxification procedures ...... 216

Figure 8.3 Integrated management for mycotoxin control ...... 218

xvi

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Abstract

The current study evaluated interactions between the mycotoxins aflatoxin

Bi (AFBi) and fumonisin Bi (FBi) in pure systems as well as in com. During

storage at 25°C with high moisture (ca. 50%), com contaminated with Aspergillus

flavus (LI3) and/or Fusarium moniliforme (A00102) presented different aflatoxin

and fumonisin levels. Toxin production/degradation varied over storage time (0-45

days). Evaluation of the mutagenicity of unifungal or bifungal com extracts with

theSalmonella/microsomal assay (tester strains TA-100 and TA-98) with metabolic

activation (S9) showed positive mutagenic responses.

Different ratios of pure AFB|/FBi produced divergent mutagenic responses

(tester strains TA-100 and TA-98) with S9. The mutagenicity of 103 ng AFBi/pIate

was reduced by addition of 1 to 102 ng FBi/plate. Higher FBi concentrations (103 to

103 ng/plate) caused an opposite effect. The mutagenic potential of 2-

aminofluorene (2-AF) was also affected by FB|. This suggested that FBi affects the

enzyme systems responsible for activating AFBi and 2-AF.

Analysis by liquid chromatography of AFBi/AFBi-metabolites after

reaction with FBi and S9 showed decreased recovery when 103 ng AFBi were

exposed to 1 to 102 ng FBi, but at higher FBi levels (103 to 103 ng), recovery was

increased. Thin layer chromatography analysis of these samples suggested that

unidentified AFBi-metabolites derivatized with trifluoroacetic acid were indirectly

detected as AFBi.

xvii

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. FBi (1 to 103 ng/plate) inhibited gap-junction intercellular communication

of Clone-9 rat liver cells. This could be a possible mechanism by which FBi acts as

a non-genotoxic carcinogen. Addition of 1 to 103 ng AFBi to 103 ng FBi did not

change the effect of FBi. An increase in the uptake of neutral red by Clone-9 rat

liver cells exposed to AFBj/FBi implied that peroxisome proliferation and

oxidative metabolism may be involved in AFBi/FBi interactions.

H2 O2 (3%) and NaHCC >3 (2%) were effective in decontaminating/

detoxifying aflatoxin-fumonisin naturally co-contaminated com without generating

mutagenic by-products.

This study showed preliminary evidence that simultaneous exposure to

AFB1/FB1 can give different toxicologicai responses than single exposure to either

of these compounds. More research is needed to better assess the toxicologicai

impact of AFB 1/FB1 co-contamination and to develop possible control strategies.

xviii

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. I. Introduction

Aflatoxins and fiimonisins are mold secondary metabolites produced by

Aspergillus flavus and Fusarium moniliforme. respectively. The potential risks

associated with these metabolites have long been recognized. However, recent

reports have shown that aflatoxins and fiimonisins can co-contaminate com. This

poses new research and regulatory questions because information about the

potential hazards associated with simultaneous exposure to both toxins is not

available.

This is not uncommon since availability data on exposure and toxicity of

most chemical mixtures is limited, and roughly 95% of the research resources in

toxicology is still devoted to the study of single chemicals (Krishnan and Brodeur,

1991; Yang, 1994). Concurrent or sequential exposure to a multitude of chemicals,

e., food additives, mycotoxins, pesticides, environmental and occupational

pollutants, etc., is a real issue and has pointed to the necessity of exposure

assessment, hazard identification, and risk assessment of chemical mixtures.

Furthermore, there has been increased public and regulatory concern regarding

exposure to complex systems that has triggered the interest of scientists and

toxicologists in this challenging vanguard of toxicology (Cassee, 1998).

The study of chemical mixtures is controversial and has exacerbated debates

on concepts such as similar joint action and independent joint action versus

interaction of chemicals, design of experimental studies and analysis of results.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. This project has focused on an initial in vitro approach in the study of a

mycotoxin mixture that frequently occurs in nature. This is a simplistic view

because these toxins co-exist in com, a more complex matrix that consists of tens,

hundreds or even thousands o f compounds. However, adequate testing of such

mixtures is virtually impossible; therefore, choosing a combination of two of the

compounds that have been recognized as toxic and that have been reported to exist

in the same commodity could give better information on the activity of the mixture.

Furthermore, if all possible combinations of all possible compounds present in the

mixture were to be evaluated, a complete experimental design would grow in

exponential proportions, making adequate testing virtually impossible.

One of the main interests in the field of combination toxicology is to find

out whether the toxicity of a mixture is different from the sum of responses to the

single compounds. The genotoxic nature of aflatoxin Bj (AFBi) coupled with the

non-genotoxic status of fumonisin Bi (FBi) and their ubiquitous, unpredictable and

unavoidable occurrence, triggered the interest in this area of research. The overall

objective of this project focused on studying the interactive effects of AFBi and

FBt in concentration ranges that can occur naturally as well as mechanisms of

toxicity and decontamination procedures.

The diverse components of this project concentrated on the study of

AFB i/FBi co-contamination from different perspectives. The first section focused

on the mycological aspects o f toxin production. The objective was to do a

preliminary evaluation of the production/degradation of both, aflatoxins and

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. fumonisins in the same environment under optimal mold growth conditions. The

second objective centered on the evaluation of potential toxin interaction by testing

whether FB| had an effect on the mutagenic potential of AFBi. From this study,

two approaches were taken and explored. If FB| interacted with AFBi, was it

directly by chemical reaction of the toxin and/or toxin by-products or were there

mechanistic implications?

Changes in gap-junction intercellular communication have been proposed

as a non-genotoxic carcinogenesis mechanism. Since the role of fumonisin in

carcinogenesis is still unclear, the third objective was to determine whether FBi

affected gap-junction intercellular communication, and the subsequent role of

aflatoxin interaction in this toxic response.

Finally, since information on the potential risks associated with AFBi and

FBi is still limited, decontamination treatments were explored as options in

mycotoxin risk management. Therefore, the fourth objective of this study was to

evaluate the effectiveness of H 2 O2 and/or NaHCC >3 in decontaminating/detoxifying

naturally contaminated com.

The toxicology of complex mixtures is still a new research area that poses

many new research and regulatory concerns. It is important to use an

interdisciplinary approach to better assess these issues and understand the risks

posed by simultaneous exposure to two or more mycotoxins.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. II. Literature review

A. Historical background

Mycotoxins have played an important role in the history of humankind.

Although the consumption of moldy food has not always been recognized as a

health hazard, mycotoxins have been intricately involved in the social, political,

and economical environment of cultures around the world. The earliest known

mold-associated disease is ergotism or “holy fire” that resulted from the

consumption of grains infected by the mold Claviceps purpurea. A gangrenous

form of ergotism was very common in Europe from the ninth to the fourteenth

centuries. The victims of this poisoning complained of numbness in the limbs,

intense sensations of heat and cold, followed by gangrene; and eventually the loss

of the extremity. At this point in history, such afflictions were associated with

religious events rather than dietary practices. Thus, this mycotoxicosis was known

as St. Anthony’s fire and was associated with sinful behavior (Smith and Moss,

1985). Later, between the sixteenth and the nineteenth centuries, a form of

convulsive ergotism was described. This type of intoxication was also noted in the

United States where it was implicated in the Salem witchcraft trials of 1692

(Matossian, 1982). During the seventeenth century, a form of cardiac beriberi

associated with yellow rice toxins was described in Japan (Ueno, 1983). Also, other

diseases such as “red mold” or “scabby disease” have also been associated with rice

and rice products (Beuchat, 1978).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Wars have shaped world history. As they bring about violent change, they

also make obvious alterations in the behavioral pattern of the population. Since

dietary practices are strongly related to this kind of event, some wars have shaped

our understanding of the impact of mycotoxins on human health. A particularly

severe mycotoxicosis, alimentary toxic aleukia (ATA), was first recognized in

Russia in 1913 where food shortages near Siberia forced the consumption of over

wintered grains (wheat, millet, and barely). The melting snow raised the moisture

content of the grains and favored mold growth and toxin production (Smith and

Moss, 1985). ATA has had a constant place in Russian history. In the twentieth

century, it reached epidemic proportions during and after World War II. In this

outbreak, mortality was as high as 60%. Highly toxic T-2 toxin and other

trichothecenes toxins were later confirmed to be the cause of the epidemic

(Bullerman, 1979; Ueno, 1987). Although mycotoxins have played an important

role in history, awareness of mold-related diseases was practically non-existent

until the 1960s. During this time, scientific attention focused in England where a

hepatotoxic disease in turkeys was reported. Thousands of poults died of an

unknown hepatic disease (Asplin and Camaghan, 1961; Blount, 1961; Lancaster et

al., 1961). This syndrome was dubbed as turkey "X” disease, and marked the

initiation of formal research in mycotoxicology. Shortly after the first report of the

“X” disease, there were other reports of poisoned chickens and ducklings. All of

these syndromes were characterized by hepatic necrosis, marked bile duct

hyperplasia, acute loss of appetite, wing weakness, and lethargy (Blount, 1961).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Later. Asplin and Camaghan (1961) identified the responsible toxins as metabolites

of Aspergillus flavus. These toxins were chemically characterized by Asao et al.

(1963) and designated with the term aflatoxins. After this incident, the field of

mycotoxicology gained the interest of the scientific community, and research on

this area proliferated. Thus, in the last three decades, many other mycotoxins have

been identified and related to human and animal health problems.

Fusarium moniliforme is one of the most prevalent fungi associated with

com ( Zea mays) throughout the world (Marasas et al., 1984). Although it has been

barely a decade since the fumonisins have been characterized as food-bome

mycotoxins, the toxicities associated with them have long been recognized. In

1881, Saccardo implicated Oospora verticilloides contaminated com with pellagra

in Italy. Years later, in 1902, Butler successfully reproduced equine Ieuko-

encephalomalacia (ELEM) in horses by feeding them with naturally contaminated

moldy com. Sheldon (1904) identified F. moniliforme as the causative agent of

com toxicoses in farm animals in the United States. Much later, in South Africa,

Kellerman et al. (1972) induced a hepatotoxic syndrome by feeding horses with F.

moniliforme culture material. In 1976, Marasas and collaborators reproduced

ELEM in horses by feeding pure F. moniliforme culture material. Marasas (1981)

also reported that F. moniliforme is significantly more prevalent in com from

geographical areas with a high incidence of human esophageal cancer than in those

with low incidence. Subsequently, Marasas et al. (1984) were able to induce liver

cancer in rats by feeding a pure culture of F. moniliforme from South African com.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Four years later, in 1988, Gelderblom demonstrated the cancer promotion activity

of F. moniliforme cultures in a rat short-term cancer initiation/promotion assay.

That same year, Gelderblom et al. isolated fumonisins Bi (FBi) and B 2 (FB2 ) from

cultures of F. moniliforme MRC 826. The chemical structures of these toxins were

elucidated by Benzuidenhout et al. (1988). In 1988, intravenous injection of FBi

resulted in a syndrome resembling ELEM (Marasas et al., 1988). Thus, there was

evidence that FBi was the etiological agent of ELEM. In 1989, confirmation was

provided when Laurent et al. isolated and characterized two new mycotoxins from

F. moniliforme: macrofusin and micromonilin. The structure of one these toxins,

macrofusin, proved to be identical to FBi. In 1989, com screenings caused

widespread ELEM and porcine pulmonary edema (PPE) outbreaks in the United

States (Harrison et al., 1990; Ross et al., 1990; Wilson et al., 1990). After these

outbreaks, the scientific community gained interest in the recently characterized

mycotoxins.

Molds and mycotoxins have a considerable impact worldwide in terms of

public health, agriculture and economics. Naturally occurring toxicants are

unavoidable, unpredictable and pose a unique challenge to food safety. The Food

and Agriculture Organization of the United Nations has reported that the

production of agricultural commodities is barely sustaining the world's increasing

population; and every year, at least 25% of the world’s food crops are contaminated

with mycotoxins. Mycotoxigenic fungi and their control pose a serious economic

impact around the world. They assert discernible costs to farmers, livestock and

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. poultry producers, grain handlers, and food and feed processors. However, it is

difficult to estimate the actual economical losses due to the large variability of

incidence of mycotoxins among commodities, years and seasons. Roughly, the

Council for Agricultural Sciences and Technology (CAST) has estimated that in the

United States alone, twenty million dollars is lost annually on peanuts

contaminated with aflatoxin (CAST, 1989). The ubiquity of toxigenic molds (i.e.,

A. flavus and F. moniliforme), the importance of com as human food and animal

feed, reports of possible co-contamination with aflatoxins and fumonisins and the

implication of these toxins with diverse toxicological syndromes, has provoked an

increased scientific and regulatory interest. To date, there are still many questions

without answers. Thus, research in this area is abundant.

B. Aflatoxins

1. Mycology and occurrence

The ubiquity of A. flavus, as well as, its ability to colonize multiple

commodities has prompted a vast amount of work on the causes, course and

prevention of contamination. Earlier studies assumed that invasion was a function

of inadequate storage conditions (i.e., drying, temperature, and moisture).

However, newer studies have shown that, invasion before harvest is equally or

more important (McDonald and Harkness, 1967; Petit et al., 1971; Cole et al.,

1982; Klich et al, 1984). In com, pre-harvest invasion is primarily dependent on

insect damage of the developing cobs. The fungus can also invade by growing

down the silks of the developing ears (Jones et al., 1980; Lillehoj et al., 1980). The

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ability to grow as a non-destructive pathogen in the tissues of a variety of plants

over the normal range of food storage temperatures as well as the capacity to grow

at low water activities give A. flavus the potential to grow in the vast majority of

commodities if pre-harvest, harvest and storage conditions are less than ideal.

Weather conditions also play an important role in aflatoxin formation. Since

A. flavus is considered an opportunistic pathogen, any factor that affects the normal

development of the plant, will favor fungal invasion; and subsequently, toxin

production. Payne et al. (1986) showed that irrigation helped reduce aflatoxin

formation; and reported that drought stress led to an increased number of infected

kernels in silk-inoculated ears. Other environmental factors that affect toxin

production and accumulation include lack or excess of nitrogen, excessive plant

populations, and poor irrigation practices (Payne et al., 1989; Fortnum and

Manwiller, 1985; Anderson et al., 1975).

The aflatoxigenic species of the Aspergillus group include A. flavus, A.

parasiticus, and A. nomius. A. flavus and A. parasiticus may colonize the food/feed

and produce mycotoxins. The role of A. nomius in aflatoxin contamination of crops

is still unknown. Specific strains of A. flavus and A. parasitucus are capable of

producing aflatoxins Bi, B 2 , G|, G2 , and Mi, and other related compounds (Wilson

and Payne, 1994). At the present time, the taxonomy of the A. flavus group is

uncertain. The differences between these fungi are often inconclusive, so it is

difficult to identify distinct species. According to Wicklow (1983) and Klich and

Pitt (1988), A. flavus and A. parasiticus are taxonomically distinct species.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. However, previous reports by Hesseltine et al. (1970) have described several

isolates with intermediate taxonomic characteristics between A. flavus and A.

parasiticus. In 1986 Kurtzman et al., complicated matters by proposing that both,

A. flavus and A. parasiticus were so closely related; so they should be considered

varieties of the A. flavus group. Therefore, they should be classified as A. flavus

var. flavus and A. flavus var. parasiticus. However, these classifications are only of

concern to the pure taxonomist. Most of the literature refers to the toxigenic species

as theA. flavus group without further specifications.

Aspergillus flavus and Aspergillus parasitucus grow in Czapek yeast extract

agar (CYA) showing colonies of 60-70 mm in diameter, plane, sparse to

moderately dense. The colonies are characteristically greyish green to olive yellow,

sometimes only yellow and become green with age. They are distinguished by their

rapid growth both at 25 and 37°C. The difference between these two species is that

A. flavus produces conidia which are variable in shape and size and have thin walls

that vary from smooth to moderately rough. In contrast, A. parasiticus conidia are

spherical with relatively thick, rough walls (Pitt and Hocking, 1997).

Growth temperatures for A. flavus vary from a minimum near 10-12°C, a

maximum near 43-48°C, and an optimum near 33°C (Domsch et al., 1980; ICMSF,

1996). The water activity needed for growth also varies in a range of 0.78 at 33°C

(Ayerst, 1969) to 0.84 at 30°C and 0.80 at 37°C (Pitt and Hocking, 1977). These

molds can grow in a very wide range of pH. It has been observed that growth can

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. occur from pH 2.1 to 11.2 at 25, 30 and 37°C. Optimal growth was recorded from

pH 3.4 to 10 and a peak near 7.5 (Wheeler et al., 1991; Olutiola, 1976).

A. flavus is adapted to use a broad assortment of organic resources as a

source of nutrients. It is considered a saprobe, as well as, an opportunistic pathogen

of plants, insects, domestic animals and humans. In the field, A. flavus populations

increase during hot and dry conditions. They grow especially well on crop debris,

on dormant tissues, and or damaged or weakened crops (Asworth et al., 1969;

Stephenson and Russel, 1974).

2. Toxin production

Aflatoxins B and G are produced by all toxigenic strains of A. parasiticus.

On the other hand, most, but not all A. flavus isolates produce only the B aflatoxins

(Pitt, 1989). In general, commodities contaminated with only the B aflatoxins may

be contaminated with A. flavus. However, there is no real evidence that suggests

thatA. parasiticus is the only source of the G aflatoxins (Wilson and Payne, 1994).

Some strains of A. flavus produce cyclopiazonic acid (CPA). This toxin has

been reported to be toxic in rats, dogs, pigs, and chickens (Domer et al., 1983;

Lomax et al., 1984; Nuehring et al., 1985; Purchase, 1971). CPA has metal

chelating properties, and it has been reported that it can bind physiologically

important cations such as calcium, magnesium and iron (Gallagher et al., 1978).

CPA occurs naturally in com, peanuts, cheese and millet and it can be found with

aflatoxins (Richard et al., 1989; Gallagher et al., 1978; Trucksess et al., 1987). Due

to this co-contamination, the toxicity of commodities contaminated with A. flavus

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. can be greater than expected from the levels of aflatoxins. The role of CPA and

aflatoxin co-contamination should be explored to further understand the

toxicological implications of consumption of foods/feeds contaminated with A.

flavus.

3. Chemical characteristics

Aflatoxins (Figure 2.1) are a group of bisfuran coumarin compounds

produced by mold secondary metabolism and have no obvious physiological role in

the primary growth and function of the mold (Bhatnagar, 1991). However,

byproducts of primary metabolism are generally precursors of secondary

metabolites. Thus, primary and secondary metabolisms are intricately related

(Drew and Demain, 1977). Aflatoxin biosynthesis is related to mold lipid

biosynthesis (Townsend et al., 1984). Its basic structure is derived from acetate

units from the polyketide pathway. Bhatnagar et al. (1991) have suggested that

AFB1 biosynthesis is based on a C 20 polyketide (decaketide) precursor (Bhatnagar,

1991). Polyketides are a large group of structurally diverse compounds that arise

from the head-tail condensation of acetate units (Applebaum, 1981). These

molecules then undergo several transformation reactions where the chain is folded,

condensed, oxidized, reduced, cleaved and rearranged to produce a wide variety of

molecules, including the aflatoxins (Bhatnagar, 1991). Hamid and Smith (1987)

have reported the degradation of aflatoxins in old or aging mycelium. There is

evidence that suggests the involvement of fungal cytochrome P-450

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. OCH3 0C H 3

OCH3 0C H 3

OH OH

OCH3 OCH3

Figure 2.1 Chemical structiares of most common naturally-occurring aflatoxins (Pavao et al., 1995)

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. monooxygenease enzyme systems in the degradation of aflatoxin Bi and Gi by

intact mycelium and cell-free extracts of A. flavus.

Each of the known aflatoxins has a different structure with variable activity

(Figure 2.1) (Pavao et al., 1995). The names of the most common aflatoxins, B and

G for blue and green, respectively, refer to their fluorescent colors under ultraviolet

light. In the case of aflatoxin M. an important aflatoxin metabolite, the letter refers

to its presence in milk (Bhatnagar et al., 1994).

Aflatoxin Bi is one of the most potent naturally occurring carcinogens.

Chemically, it has a carbonyl group in a five-membered ring and cross-conjugated

with an alpha, beta-unsaturated function. Its formula is C 17 H 12O6 with a molecular

weight of 312 (Pavao et al., 1995). Aflatoxins are soluble in a number of non-polar

solvents and insoluble in water.

4. Metabolism

a. Absorption and distribution

Human exposure to aflatoxins occurs mainly via ingestion of contaminated

foodstuffs. Thus, most of the pharmacokinetic considerations are associated with

the chronic administration of relatively low doses of AFBi through the oral route

(Hsieh and Wong, 1994). AFBi is a relatively low molecular weight (312 g/mole)

lipophilic compound. This suggests that its absorption after ingestion is efficient.

Wogan et al. (1967) reported that absorption after oral administration of 0.07

mg/kg [ 14 C] labeled AFBi in • male Fisher • rats was complete when compared to

intraperitoneal administration. They observed that up to 20 and 60% of the original

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. dose was eliminated via urinary and fecal routes, respectively. Slow passage

through the intestinal tract was noted since little radioactivity was excreted in the

first 8 hours. After 24 hours, the residual radioactivity in the carcass (7.6% of the

original dose) was found in the liver, indicating that it is the primary site of

accumulation of AFBi, its metabolites and/or bound material.

Absorption studies by Kumagai (1989) have shown that AFBi is rapidly

absorbed from the small intestines into the mesenteric venous blood. The

duodenum was considered the site of most efficient absorption. This study

suggested that species differences in AFBi uptake may be due to differences in the

composition of the intestinal epithelium. The author also reported that the degree of

aflatoxin uptake is proportional to AFBi concentration indicating that it is absorbed

by passive diffusion. AFGi, a less lipophilic analog, was absorbed at a lower rate,

indicating the importance of Iipophilicity in aflatoxin absorption.

Excretion of AFBi occurs primarily through the biliary pathway, followed

by the urinary pathway. From the intestine, AFBi apparently enters the liver

through the hepatic portal blood supply (Wilson et al., 1985). The liver readily

concentrates the toxin. Given the high efficiency of the liver to extract free AFBi

from the blood, the binding of AFBi to serum albumin at the site of intestinal

absorption has not been considered a major mechanism of detoxification. The

kidneys also concentrate aflatoxin, but to a lesser extent (Hsieh and Wong, 1994).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. b. Phase I metabolism

Cytochrome P-450 (P450) enzymes are involved in the oxidation of

numerous steroids, eicosanoids, alkaloids, and other endogenous substrates. These

enzymes also have major involvement in the oxidation of potential toxicants and

carcinogens such as those encountered among environmental pollutants, solvents,

pesticides and many natural products like aflatoxins (Guenguerich et al., 1996).

AFBi is considered the most potent of the aflatoxins and related

benzofurans in most genotoxicity assays; it is also the most hepatocarcinogenic

(Busby and Wogan, 1984). However, all evidence available to date shows that

AFBi by itself is not particularly genotoxic, and that the metabolically produced

8,9-epoxide is the only genotoxic product as evidenced by either base-pair

mutations, ffameshift mutations or induction of bacterial SOS response (Busby and

Wogan, 1984; Gamer et al., 1972; Baertschi et al., 1989). The stereochemistry of

AFBi epoxidation is of considerable significance in determining the course of its

toxicity. P450 enzymes produce two different stereomeric forms of the 8,9-

epoxide, the exo-epoxide in which the ring is positioned “below” the plane and the

endo- 8,9 epoxide in which the ring is positioned “above” the plane. Of these, the

exo isomer is at least 103 times more genotoxic than its endo counterpart

(Guenguerich et al., 1996). The bioactive form of aflatoxins is considered a strong

electrophile that can form covalent adducts with macromolecules such as proteins,

RNA and the N7 position of guanidine residues in DNA (Figure 2.2) (Foster et al.,

1983; Miller, 1991).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 0 C H 3

Aflatoxin B ,

O CH 3

Aflatoxin B ,-8,9-epoxide

H-.N O C H 3 N

Aflatoxin B ,-N7-Guanidine Adduct

Figure 2.2 Formation of the aflatoxin Bi- N7 guanidine adduct (Miller, 1991)

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Cytochrome P450 metabolic isozymes render a number of products that can

be considered activated metabolites (i.e., 8,9-epoxide) or detoxification products,

such as aflatoxins Mi, Pi and Qi (Figure 2.3) (Langouet et al., 1996). Although

peroxidases and lipoxygenases have been reported to oxidize AFBi (Battista and

Mamett, 1985), the P450s are more active in catalyzing the reaction. The relative

quantities of each metabolite depend largely on the relative activities of different

P450 isozymes. Among the P450s, CYP3A4, CYP1A2, CYPIA1 and CYP3A5

have been shown to transform AFBi. Cytochromes 1A2 and 3A4 have been

considered the most active. However, there is controversial information as to which

is the dominant enzyme. Ueng et al. (1995) have reported that CYP3A4 appears to

be the enzyme involved to the greatest extent in AFBi activation. Forrester et al.

(1990) and Aoyama et al. (1990) previously provided evidence that supports this

hypothesis. On the contrary, Gallagher et al. (1996) reported that CYP1A2

contributes to over 95% of AFBi activation in human liver microsomes at 0.1 juM

AFB i. The discrepancies between these studies may be a result of different sources

of P450’s used among the different experiments. The reconstitution system of Ueng

et al. (1995) was rather complex and cannot be considered similar to the in vitro

system employed by Gallagher et al. (1996). Also, the source of the human

CYP1A2 cDNAs used in the experiments of Ueng et al. (1995) contained a

modified N-terminus (Sandhu et al., 1994) and therefore may have exhibited

somewhat different catalytic activities from the CYP1A2 cDNA -

expressed microsomes used in the Gallagher et al. (1996) study. This study also

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 2.3 Biotransformation pathways for aflatoxin Bi (Eaton et al., 1994)

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. reported kinetic models that support CYPlA2’s dominant role in AFBi metabolism

in vivo.

c. Phase II metabolism

Phase II metabolism leads to the conjugation of phase I metabolites to make

them less toxic and more available for excretion. Species differences in the

susceptibility to the toxicity and carcinogenicity of AFBi are determined largely by

the organism’s ability to biotransform the toxin. The major phase II metabolites are

the glutathione, glucoronide and sulfate conjugates (Hsieh and Wong, 1994). The

glutathione (GSH) conjugate of the AFBi 8,9-epoxide has been identified as the

major phase II reaction. Thus, the sensitivity of a particular species has been related

to its ability to conjugate the epoxide with GSH. There is some evidence that

indicates that the various glutathione-S-transferases have differential activity

toward the endo and the exo forms of the 8,9-epoxide.

Glucoronide conjugation also plays a role in biotransformation and

excretion of AFBi and/or its metabolites. Loveland et al. (1984) reported that

glucoronides of aflatoxicol and aflatoxicol-Mi are the principal biliary metabolites

of AFBi in trout. Phase I metabolites such as aflatoxins Pi, Qi and Mi go through

glucoronide conjugation. The rate of conjugation of these three metabolites differs.

According to Metcalfe and Neal (1983), the phenolic hydroxyl group present in

AFPi is a much better site for glucoronide conjugation than the hydroxyl groups

present in AFMi and AFQi.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. d. Competitive pathways: activation and detoxification

The fate of AFBi in an organism depends on a delicate equilibrium between

several pathways. The quantity of aflatoxin available to exert toxic, mutagenic and

carcinogenic effects will depend on the amount converted to the various

metabolites and their relative biological activity. Thus, the hydroxylated

compounds (AFM|, AFPi and AFQO are considered detoxification products due to

their lower ability to react with DNA and proteins. The 8,9-epoxide is not always a

toxic product since detoxification of this reactive molecule may occur through

phase II conjugation with glutathione. Hydrolysis of the electrophilic epoxide to

form a dihydrodiol also represents a decrease in toxic potential. The dihydrodiol

can exist in a resonance form as a phenolate ion that is capable of forming SchifF

base adducts with protein amino groups, particularly lysine (Lin et al., 1978). Even

if the binding to protein can represent a toxicological risk, the relative toxic,

mutagenic and carcinogenic effects of protein conjugation is much lower than that

of the active epoxide.

It is difficult to discuss AFBi activation and detoxification pathways

without considering the kinetics involved in each reaction. Unfortunately, most in

vitro studies use AFBi concentrations that surpass the actual dietary human

exposure. It has been noted that humans and animals are chronically exposed to

low doses rather than to one acute single dose. Thus, it is difficult to extrapolate

from in vitro studies. Another factor to consider is that the activity of the different

metabolizing enzymes varies from species to species. Ramsdell and Eaton (1990)

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. demonstrated substantial variations in the patterns of liver microsomal AFBi

oxidation over a range of concentrations for four species: rat, mouse, monkey and

human. Affinity to detoxification reactions also plays an important role in AFBi’s

ultimate toxic effect. As with microsomal activation, the affinity of the enzyme

glutathione S-transferase to the AFBi-8,9-epoxide varies across species.

The information available to date suggests that the ultimate toxic response

to aflatoxin is not a univariate process. It is a multivariate equilibrium that depends

on external factors, such as, toxin concentration, presence of other xenobiotics,

environmental stress and nutritional status among others; and internal factors that

can include but are not limited to genetic predisposition, type and activity of

microsomal enzymes, and availability of efficient detoxification mechanisms.

5. Toxicity

The liver is considered the primary target organ for aflatoxin toxicity. Since

its characterization in the early 1960’s, acute structural and functional damage to

the liver has been reproduced in a wide variety of species. Acute aflatoxicosis has

been characterized by vomiting, abdominal pain, pulmonary edema, and fatty

infiltration and necrosis of the liver (Shank, 1977; 1981). Although there is ample

evidence for substantial human exposure in certain populations, information on

clinical aflatoxicosis in humans is still limited (Busby and Wogan, 1984). Review

of the effects of acute exposure to aflatoxin reveals that a wide variety of

vertebrates, invertebrates, plants, bacteria, and fungi are sensitive to these toxins,

but the range of sensitivity is wide. The basis for the species and strain variation in

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. the acute toxicity of aflatoxin is not fully understood. As discussed in the previous

section, two important factors are (1) the proportion of AFBi that is metabolized to

the 8,9-epoxide relative to other metabolites that are considerably less toxic and (2)

the relative activity of phase II metabolism, which forms non-toxic conjugates and

inhibits cytotoxicity (Cullen and Newbeme, 1994).

The aflatoxins have been studied extensively in vitro (IARC, 1987). It has

been reported that all the members of the family (AFBi, AFB2, AFGi and AFG 2)

produce genotoxic injury. AFBi is considered the most potent genotoxic agent in

vitro (as it is in vivo) and AFG 2 the least potent, but all naturally occurring

aflatoxins, as well as, the metabolite AFM|, have been reported to produce some

injury to DNA. AFBj produces chromosomal aberrations, micronuclei, sister

chromatid exchange, unscheduled DNA synthesis, and chromosomal strand breaks,

and forms adducts in rodent and human cells (IARC. 1987). AFGi produces

chromosomal aberrations in Chinese hamster bone marrow cells in vivo. as well as

chromosomal aberrations and sister chromatid exchange in Chinese hamster cells in

vitro. AFB2 binds covalently to DNA of rat hepatocytes, produces sister chromatid

exchange in Chinese hamster ovary cells, and stimulates unscheduled DNA

synthesis in rat hepatocytes but not in human fibroblasts. AFG 2 produces sister

chromatid exchange in Chinese hamster cells but does not induce unscheduled

DNA synthesis in human fibroblasts in vitro. Unscheduled DNA synthesis in rat

and in hamster hepatocytes results following exposure to AFG 2 and AFMi in vitro.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. The aflatoxins have been examined for potency in a range of primary

cultures, as well as, in established cell lines. Liver cell cultures from chick embryos

demonstrate cytotoxicity of AFBi to both mesenchymal and parenchymal cells

(Terao, 1967). Zuckerman et al. (1967) reported that 1 pg/ml exhibited cytotoxic

effects to 50% of human embryo liver cells following 24 hours exposure. Electron

microscopy revealed nucleolar capping of the chromatin, rounding of the cell, and

granulation of the endoplasmic reticulum (Zuckerman et al., 1967).

Experimental study of aflatoxin-induced carcinogenesis supports the

paradigm of cancer as a multistage process. Aflatoxin is considered a potent

tumorigenic agent, but it has also been shown to have promoting ability (Newbeme

and Butler, 1969; Busby and Wogan, 1984). Virtually all the toxic effects of AFBi

are now recognized to be attributable to the action of its electrophilic metabolites

on macromolecules (DNA and proteins). Aflatoxin carcinogenesis can be promoted

by a variety of factors, including an enhanced proliferative rate and alterations in

dietary intake (Rogers, 1994). Since Wogan (1966) reported that continuous

lifetime exposure to aflatoxin was unnecessary for the development of tumors,

shorter-term exposure to aflatoxin has been used to assess the risks associated with

this toxin. The early studies of aflatoxin-induced liver carcinogenesis in rats

indicated the presence and sequential development of precursor lesions (Lancaster

et al., 1961). Chronic and subchronic exposure to AFBi is an efficient method to

produce hepatic cancer. For example, 2 cycles each of 5 days duration with 25 mg

AFBi per day to young (approximately 100 g) Fischer 344 rats yielded putative

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. preneoplastic foci at 2-3 months postdosing (Appleton and Campbell. 1983;

Roebuck et al., 1991; Kensler et al., 1992). Hepatic cancers arise by approximately

1 year postdosing (Roebuck et al.. 1991).

C. Fumonisins

1. Mycology and occurrence

Fusarium is the main pathogenic fungal genus found in com. The most

commonly occurring species are F. graminearum, F. moniliforme and F.

subglutinans (Burgess et al., 1981; Marasas et al., 1984). Fusarium moniliforme

and its related species, F. proliferatum, are endemic in com with a worldwide

distribution occurring not only in the humid and sub-humid temperate zones, but

also extending into the subtropical and tropical areas of the world. It has been

isolated from Australia, Brazil, Canada. Central America, China, Egypt, Germany,

India, Indonesia, Israel, Italy, Jamaica, Japan, Hong Kong, Libya, Nepal, New

Zealand, Nigeria, Pern, Philippines, South Africa, Switzerland, Taiwan, Turkey, the

United States of America, Zambia and Zimbabwe (Cole et al., 1973; Marasas et al.,

1978; Marasas et al. 1979; Sydenham et al., 1991; Thielet al., 1991; Pittet et al.,

1992; Bacon and Nelson, 1994; Burgess et al., 1994; Visconti and Doko, 1994;

Yoshiazwa et al., 1996). It is found rarely in the cold temperate zones of the world,

but has been reported from Russia and Iceland (Bacon and Nelson, 1994).

F. moniliforme is an important pathogen of com and sorghum. In addition,

it is commonly isolated from several other grains including wheat, rice and oats.

Although there is no evidence that F. moniliforme is a pathogen of wheat, barley or

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. oats in nature (Cook, 1981), it can be found as an occasional contaminant of wheat

and barely stored seed (Bottalico et al., 1989). It also occurs, although infrequently,

on other food and non-food commodities, such as beans, cotton, green peppers,

flax, soybeans and sugar cane. F. moniliforme is also found on forage grasses such

as Panicum coloratum as well as alfalfa, red clover, and pearl millet (Bacon and

Nelson, 1994). This fungus can also decompose wool, cotton, filter paper and can

oxidize hydrocarbons and the herbicide, simanzine (Bacon and Nelson, 1994).

F. moniliforme and F. proliferatum in the section Liseola are characterized

by conidiophores bearing abundant straight, fusiform to obovoid microconidia in

long chains. The conidia are produced from branched and non-branched

polyphialides and monophialides in F. moniliforme and in F. proliferatum. These

fungi also produce falcate microconidia, but do not produce chlamydospores. There

are four taxa within this section of which F. moniliforme is the most common.

Another cosmopolitan species of this section is F. proliferatum. Both fungi are

similar morphologically, but may be readily distinguished by the presence of

polyphialides in F. proliferatum and their absence in F. moniliforme (Bacon and

Nelson, 1994).

F. moniliforme grows at a maximum temperature of 32-37°C and a

minimum temperature of 2.5-5°C with an optimum of 22.5-27.5°C (Armolik and

Dickson, 1956). The minimum water activity for F. moniliforme growth is 0.87;

and Marin et al. (1995) have reported that there is production of fumonisins at 0.92

aw. Moisture and aeration are important factors for fumonisin production. It has

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. been observed that fiimonisin production is minimal at low oxygen tension. Thus,

storage of grains under modified atmosphere conditions, and low kernel moisture

<22% should reduce or prevent toxin formation (Pitt and Hocking, 1997).

Although F. moniliforme is one of the earliest known examples of a

mycotoxic and phytotoxic species, its relationship to com is not fully understood

(Bacon et al., 1992). F. moniliforme is seed and soil bome (Riley, et al., 1993); and

the association of this fungus in the com plant is considered endophytic, while on

com kernels it is both external and systemic. The systemic kernel infection does not

affect the seed germination, but rather seedling vigor and growth can be reduced.

Within the com kernel, the fungus is associated with the tip cap end. the point of

attachment to the cob. The fungus is located within this area just below the vascular

tissue of each kernel. Only small amounts of hyphae are found in sound kernels.

However, from this location and under improper storage conditions, the small

amount of inoculum can become highly invasive to the kernel (Bacon and Nelson,

1994). In addition to kernels, the fungus is also found in com plant residues,

especially stems and cobs where it can survive in the soil at least a year or more

(Liddell and Burgess, 1985; Nyvall and Kommedahl, 1970). The systemic location

of this fungus within the seed provides favorable conditions to maintain virulence

and viability for years. Bacon and Nelson (1994) reported that in laboratory studies,

com seeds can still harbor viable F. moniliforme after 8 years. The visual results of

ear infection vary, some kernels are symptomless while others are obviously

damaged and infected. F. moniliforme is also believed to infect com by growing

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. along the silks, through tip ends of ears, over the kernel surface and into the

pedicels, into the vascular cylinder of the cob, and finally into the kernels

(Koehler, 1942; Salama and Mishricky, 1973). In the case of ear infection via an

external route, insect damage may play an important role by causing wounds on the

kernel pericarp thus allowing fungal penetration (Koehler, 1942). In kernels

associated with animal toxicity, the fungus is usually found as an extensive mass of

sporulating hyphae which colonizes most of the internal section of the kernel,

including the embryo (Bacon and Nelson,1994).

Both F. moniliforme and F. proliferatum produce two kinds of spores, and

both of these spores function in the survival and infection aspect of their life cycle.

These fungi do not produce chlamydospores as survival structures, but they can

survive in the soil as thickened hyphae within a soil moisture range of 5 to 35% at 5

to 10°C for 12 months (Bacon and Nelson, 1994). Macroconidia can survive the

longest. Thus, they might function to carry these fungi over the seasons while

microconidia, which are windblown, function as secondary infective propagules,

infecting plants during the growing season. Dormant microconidia can survive for

900 days although survivability is inversely related to both high temperatures and

relative humidity (Liddell and Burgess, 1985).

2. Toxin production

F. moniliforme consists of a complex of biological species or mating

populations. Currently, it is defined as having six different mating populations

among which exist bisexual, self-sterile groups (Riley et al., 1993). Only mating

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. populations "A" and "F" of this fungus have been found to be toxigenic, producing

the mycotoxins fumonisins, fusaric acid and fusarin C. These mating populations

are distributed worldwide and are found primarily on cereal grasses in the warmer

regions of the world. Although members of all six mating populations can be

recovered from both com and sorghum, the distribution appears to be host species

specific. It has been observed that fungi of mating population "A" , "D" and “E”

are found in com and of mating population “F” are primarily found on sorghum

(Jardine and Leslie, 1992; Leslie et al., 1992a; Leslie et al., 1992b). Although it has

not been established whether these two species vary in their pathogenicity or

specificity for each host, it has been found that the "A" mating population is a

better producer of fumonisin Bi than the "F" population (Riley et al., 1993). Also,

the mating population "D" of Giberella fujikori has been reported as being a good

producer of fumonisin Bi (FBi) (Riley et al., 1993). Recently, it has been reported

that Alternaria alternata produces FBi (Chen et al.. 1992); and this suggests that F.

moniliforme might produce the AAL toxins as well. In addition to F. moniliforme

and F. proliferatum, several other newly described species, not yet placed in any

Fusarium section, also produce these toxins and include: Fusarium nygamai,

Fusarium napiforme, Fusarium anthophilum and Fusarium dlamini (Nelson et

al., 1992). However, since these species are not usually associated with food and

feed products, they are not considered important as far as the natural production of

fumonisins on agricultural commodities.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Little is known about the biology of F. moniliforme and F. proliferatum in

terms of food contaminants and physiology of toxin production. These fungi

produce several mycotoxins including: fusarins, fusaric acid and fumonisins

(Chamberlain et al., 1993). Nelson and collaborators (1983) have discounted

reports of production of zearalenone and deoxynivalenol by F. moniliforme. More

recent reports require confirmation Fusarium proliferatum has been found to

invade com and to produce fumonisins and moniliformin (Bullerman and

Draughon, 1994). However, due to the increasing concern about fumonisin in foods

and feeds, more research is being conducted to understand these factors. Since both

fungi are plant pathogens, they also produce phytotoxins and several of the

mycotoxins are also phytotoxins. Thus, these fungi produce a series of substances

toxic to plants and animals. Some of the important mycotoxins/phytotoxins

produced by F. moniliforme and F. proliferatum include: fusaric acid: A

phytotoxic, mycotoxic, hypertensive agent as well as a synergist for other

mycotoxins; fusarin C: A mutagen and macrophage inhibitor; moniliformin: A

phytotoxic and mycotoxic agent; and fumonisins: A class of compounds that have

shown to be hepatotoxic, nephrotoxic and hepatocarcinogenic in rats, neurotoxic in

equines and pulmonary edema inducers in porcine species, and suspected

carcinogens in man. The fumonisins are also herbicidal to several plant species and

can induce adventitious root on non-sensitive detached tomato shoots (Bacon and

Nelson, 1994).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. The fumonisins have been detected in com cobs, bee wings (glumes of old

com florets) and com screenings. The highest concentrations have been found in

the bee wings, although com screenings have higher concentrations than whole

com and have produced the most toxicological concern, particularly to swine and

horses (Bacon and Nelson, 1994; Ross, 1994). Desjardins et al. (1997), have shown

that Giberella fiijikori isolated from rice from various geographic areas is capable

of producing significant levels of FBi (>5 pg/g) and moniliformin (>100 pg/g)

when grown in autoclaved maize substrate. To date, the occurrence of these

mycotoxins in rice and rice-based foods has received little attention (Pestka, 1994).

Other reports indicate that the highest fumonisin concentrations are found in com

screenings (also called fines) which are generated as com moves through various

distribution channels (Ross, 1994). Once a plant is infected, the precise timing and

accumulation pattern of the fumonisins in planta are unknown. Bacon and Nelson

(1994) have indicated that although a range of 9 to 91% infection is usually

reported for some seed lots of com, in some years com kernels may be 1 0 0 %

infected by F. moniliforme indicating that abiotic factors, primarily rainfall might

be responsible. Biotic factors, such as com cultivar and fungal isolate, might also

be responsible.

Because Fusaria are generally considered to develop as a result of weather

and crop stress, local weather conditions were studied by several laboratories. The

researchers concluded that no single set of weather circumstances were responsible,

but general drought stress, followed by wet and warm weather was the pattern most

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. likely to be involved in fumonisin production (Ross, 1994). There are some

important factors that have to be considered for post-harvest fumonisin control such

as:

1. Aeration: There is low production of fumonisins under a low oxygen tension

and a modified nitrogen, carbon dioxide atmosphere.

2. Moisture: A kernel moisture content of less than 22% will prevent toxin

production (Riley et al., 1993)

3. Temperature: Le Bars et al. (1994) reported that FBi production rate was

maximal at 20°C and decreased sharply according to temperature following the

order: 25, 15, 30 and 10°C

Control of soil-borne and kernel infection may be achieved with the use of

microorganisms (soil and endophytic bacteria). Control of F. moniliforme on

foodstuff should focus on its complex infection cycle and/or potential mechanisms

for developing com resistance (Riley et al., 1993). The successful utilization of

biocontrol measures must be based on detailed knowledge of the basic biology of

F. moniliforme and F. proliferatum and their associations with com and other

agricultural commodities, information, which is not yet available.

3. Chemical characteristics

Chemically, the fumonisins consist of several structurally related

homologues of which fumonisin Bi (FBi), 2-amino-12,16-dimethy 1-3,5,10-

trihydroxy- 14,15-propane-1,2,3-tricarboxy eicosane, is of primary concern since it

is considered the most active cancer promoting component within the fumonisins

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. (Figure 2.4) (Bacon and Nelson, 1994). The other fumonisins are FB 2, FB3. FB4 ,

FCi, FAi and FA 2 and each differ by lacking one of the hydroxyl groups at either

the C-5 or C-10 position of the 10 carbon aminopentol backbone. The latter three

fumonisins may be artifacts of isolation. The A series differs from B series by

possession of a N terminal acetyl group (Tanaka et al., 1993). Hydrolytic removal

of one of the propanetricarboxylic acid moeities from fumonisins Bi and B 2 yields

partially hydrolyzed fumonisins (PHj and PH 2, respectively). The removal of both

moeities yields the aminopentol fraction (AP[ and AP 2, respectively) (Sydenham,

1994).

Gelderblom et al. (1993) studied the relationship between molecular

structure and biological activity. Their results suggest that the fumonisin molecule

must be both, intact and have a free amino group to exert its effects. Independent

studies with synthesized analogues of FBi also indicate that these same structural

properties are necessary for biological activity of the toxins (Kraus et al., 1992).

Fumonisins Bi, B 2 and B 3 exist as colorless compounds that are soluble in

polar solvents such as water, methanol, or acetonitrile and insoluble in non-polar

solvents (i.e., hexane). FBi has a molecular formula of C34H59NO14 and a molecular

mass of 721 g/mole, while both, FB 2 and FB 3 contain one oxygen atom less, having

similar molecular formulae of C34H59NO14 and molecular mass of 705 g/mole

(Sydenham, 1994). None of the fumonisin mycotoxins possess chromophores, and

therefore they do not absorb either ultra-violet (UV) or visible light, nor do they

fluoresce (Sydenham, 1994).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. COOH

COOH

CH3

COOH

Ri r 2 R3 FB! OH OH H f b 2 H OH H f b 3 OH H H f b 4 H H H FA! OH OH CH3OH f a 2 H OH CH3OH

Figure 2.4 Structures of the fumonisin A and B analogues (Sydenham, 1994)

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 4. Metabolism

a. Absorption and distribution

To date, there is limited information on FBi absorption, distribution and

excretion. However, all studies available (Shephard et al., 1992; Prelusky et al.,

1994; Vudathala et al., 1994; Norred et al., 1996) have indicated that it is poorly

absorbed and rapidly cleared from the blood. It undergoes both, biliary and urinary

excretion and it tends to persist in relatively small quantities in the liver and

kidneys for prolonged periods (>96 hours). It is still unclear if these low persisting

quantities are responsible for the cascade of biochemical events prompted by FB|,

or if the short residence time of the higher doses is responsible for eliciting the

toxic responses. Furthermore, if humans are exposed to FBi through daily

consumption of contaminated com, there would always be a persisting low dose

with longer hepatic and renal residence times. This idea is supported by an in vivo

study in male Sprague-Dawley rats where it was shown that only prolonged

exposure to an hepatotoxic dietary level (250 mg FBi/kg) that induced

hepatotoxicity over a period of 14 to 21 days resulted in cancer initiation

(Gelderblom etal., 1992).

Studies regarding the metabolism of fumonisins by various enzyme

preparations have shown that they are not substrates for the microsomal

cytochrome P450, estereases, or hepatic triglyceride lipase systems (Cawood et al.,

1994). Fractionation, purification and analysis of the hepatocyte incubation

medium treated with fumonisins failed to indicate the presence of any metabolites.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. This information shows that unlike aflatoxin. the intact fumonisin molecule is

needed for its biological activity (Gelderblom et al., 1993).

b. Mode of action

Chemically, fumonisins have been found to be similar to sphinganine. an

intermediate in the biosynthesis of complex sphingolipids (Figure 2.5) (Sweeley.

1991; Bell et al., 1993). Sphingolipids have been reported to play an important role

in cellular membrane function, regulation of cell growth, differentiation, cellular

signaling and apoptosis (Mato, 1990; Merrill et al., 1997). This similarity has led to

the hypothesis that fumonisins interact with sphingolipid metabolism enzymes

(Merrill et al., 1996). To date, numerous studies have reported that fumonisins are.

indeed, potent inhibitors of the biosynthesis and turnover of complex sphingolipids

(Norred et al., 1992; Wang et al., 1992; Merrill and Shroeder, 1993; Merrill et al.,

1993; Abbas et al., 1994; Merrill et al., 1995a; Merrill et al., 1995b; Schroeder et

al., 1994; Schwartzet al., 1995; Smith and Merrill, 1995; Wu et al., 1995; Yoo et

al., 1996; Merrill et al., 1997). It has been reported that FBi interferes with

sphingolipid metabolism by competitively inhibiting the enzyme sphinganine

(sphingosine) N-acyltransferase (ceramide synthase) with respect to both of its

substrates, sphinganine and fatty acyl-CoA (Merrill et al. 1995b). In vitro studies

with diverse cells such as hepatocytes (Wang et al., 1991), neurons (Merrill et al.,

1993a), kidney cells (Riley et al., 1994) and plant cells (Abbas et al., 1994) have

shown that this enzyme plays a key role not only in sphingolipid biosynthesis, but

also in their turnover. The inhibition of the turnover pathway results in sphinganine

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. COOH

COOH

CH3

COOH

Fumonisin Bi

Sphinganine

Sphingosine

Figure 2.5 Structures of fumonisin Bi, sphinganine and sphingosine (Merrill et al.. 1998)

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. accumulation. In the presence of FBi, a portion of the accumulated sphinganine is

phosphorylated to sphinganine-1-phosphate and then degraded to hexadecanal and

ethanolamine phosphate, which is incorporated into phosphatidylethanolamine

(Merrill et al., 1993b) (Figure 2.6). This raises the possibility that fumonisin also

affects the metabolism of phosphoglycerolipids via this indirect route.

Some of the accumulated cellular sphinganine is released from cells and

appears in the bloodstream and urine. Therefore, it has been hypothesized that the

increase of sphinganine in blood and/or urine could be used as a biomarker for

exposure to fumonisin. The accumulated bioactive compounds, as well as, the

depletion of complex sphingolipids, may account for the toxicity, and perhaps the

carcinogenicity, of fumonisins (Merrill et al., 1996). The determination of the

change in the sphinganine/sphingosine ratio may indicate the amount of liver

damage and regeneration caused by FBi. The inhibition of sphingolipid

biosynthesis and turnover has also been proposed as an important factor in FBi

tumorigenic potential,

c. Carcinogenicity

Gelderblom et al. (1989) observed that fumonisin B( (FBi) is a cancer

promoter in a short-term cancer initiation-promotion assay using

diethylnitrosamine (DEN)-initiated rats. FBi is also a cancer initiator as evidenced

by the induction of resistant hepatocytes in rat liver (Gelderblom et al., 1992; 1994).

Although FBi has been shown to be a complete carcinogen in rat liver, it is

considered a poor cancer initiator. In a study reported by Gelderblom et al. (1994),

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1996) et et al., CoA) CoA) to yield 3-ketosphinganine (Ksa), hihydroceramides by which ceramide synthase using is various long-chain reduced fatty acyl to Co-A’s. Glycosidic sphinganine groups and (Sa). the Sphinganine is acylated to plasma membranes. Sphingolipid turnover is thought to involve the internalization of the sphingolipids, followed by their hydrolysis in acidic sphingosine compartments (So). (lysosomes Sphingosine and endosomes) is to cither ceramides reacylated (N-Acyl-So), or then phosphorylated to In to the presence of sphingosine-phosphate fumonisin B|, (FBi) (So-P) sphinganine accumulates and and phosphate there (Sa-P). is also an increase in sphinganine 1- Figure 2.6 Sphingolipid biosynthesis and turnover (Merrill The first steps occur in the endoplasmic reticulum (ER), where serine is condensed with palmitoyl-CoA (Pal- 4,5-trans-double bond (of sphingosine - So) are subsequently added to the 1-hydroxyl group in the Golgi and cleaved to fatty aldehyde and ethanolamine phosphate, which is incorporated into phosphtidylethanolamine (PE).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. prolonged exposure (21 days) to a relatively high dietary level (250 mg/kg) resulted

in tumorigenesis. Thus, it was concluded that a time and dose dependent threshold

level exists for cancer initiation by FBi. At a dietary level of 1000 mg/kg for 21

days, only the B fumonisins (FBi, FB?, and FB3) initiated cancer in rat liver,

whereas, the N-acetylated analogues, FAj and FA 2 , and the hydrolysis products

(aminopentol), AP| and AP 2 , and tricarballilic acid (TCA) did not (Gelderblom et

al., 1993). Thus, it has been considered that the free amino group and the intact

molecule are required for cancer initiation. However, the mechanism by which

fumonisin acts as a tumorigenic agent is still unknown.

The role of fumonisin in carcinogenesis is still unclear. Although FBi has

been reported as a complete carcinogen in rats (Gelderblom et al.. 1988), it is not

positive results in any of the traditional mutagenicity assays. Several authors

(Gelderblom et al., 1991, 1992; Park et al., 1992), have reported negative results in

the Salmonella/microsomal mutagenicity assay with and without metabolic

activation. FBi does not induce unscheduled DNA synthesis in isolated rat

hepatocytes (Norred et al., 1991a), and is not genotoxic in the in vivo and in vitro

DNA repair assays in primary rat hepatocytes (Gelderblom et al., 1992a; Norred et

al., 1992a,b). The fumonisins also failed to induce any genotoxic effects in the

DNA-repair assay using Escherichia coli (strains 3431753. uvrB/vecA and

3431765.uvr+/vec+) and the umu-microtest (with Salmonella TA 1535/pKS 1002)

with and without rat liver S9 mix (Gelderblom et al., 1996). No micronuclei

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. induction by FB| has been observed, and its mitogenic activity is still under debate

(Gelderblom et al., 1996).

Although the fumonisins were negative in various in vitro genotoxicity and

mutagenicity assays, short term in vivo studies have shown that FB| mimics

genotoxic carcinogens with respect to the induction of resistant hepatocytes in rat

liver (Gelderblom et al., 1992; 1994). Furthermore, it has been observed that FBi

induces two important enzymes, gamma glutamyltranspeptidase (GGT) and the

placental form of glutathione S-transferase (GST), which are accepted histological

markers for preneoplastic lesions initiated by genotoxic carcinogens. Feeding

experiments with fumonisins in rats indicated that an increase in cell proliferation

is also likely to play a critical role in the induction of resistant hepatocytes, and the

resultant regenerative cell proliferation is a prerequisite for initiation (Gelderblom

et al., 1994). In this regard, regenerative cell proliferation is also a critical event

during FBi carcinogenesis (Columbano et al, 1991). However, single and/or

multiple dosages of fumonisins in the presence of a stimulus for regenerative cell

proliferation fail to cause initiation (Gelderblom et al., 1992).

It has been widely accepted that carcinogenesis is a multiphase process

where many factors are involved. It would be unreasonable to consider a single

event on the ultimate carcinogenic outcome. In a real human diet, exposure to FBi

would not be considered the only event leading to cancer of the gastrointestinal

tract. An example is the epidemiological and molecular evidence from studies on

cancer patients, where a close association between aflatoxin B\ (AFBi) and

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. hepatitis B virus (HBV) infection with hepatocellular carcinoma was shown. It has

been reported that a mutation at codon 249 in the p53 gene is commonly detected in

this type of cancer (Bressac, 1991). FBj has been shown to co-contaminate grains

with aflatoxin Bi under natural conditions (Chamberlain et al., 1993; Chu and Li,

1994), so potential interaction and/or synergistic activities could also be related.

Therefore, the aflatoxin/hepatitis environment could be further complicated by the

presence of other toxins, such as, the fumonisins. Potential interactive effects need

to be determined. Studies currently in progress, are investigating the role of p53

associated mutations in early hepatocyte lesions and the ultimate cancer induced by

FBi in rats chronically fed 50 mg FBj/kg for two years. DNA from rat liver

nodules was amplified by polymerase chain reaction (PCR) with gene specific

primers, corresponding to codon 249 of the human gene. Results from this study

show that the positive control, AFBi, induced mutations at a very low frequency at

rat codon 243, the equivalent to codon 249 in humans. No mutations were detected

after cancer initiation with FBi. Only one rat fed 50 mg/kg for 12 months, showed

a mutation at codon 243 while none of the rats with hepatic carcinomas after 18 to

26 months showed any mutations in p53 (Sheffield et al., 1993). More information

is needed on the mechanism by which fumonisin acts as a tumorigenic agent; the

factors that affect this process and its relationship with other hepatic carcinogens.

Furthermore, it has been reported that many of the natural composition diets fed to

laboratory animals potentially contain fumonisins (Merrill et al., 1997). Merrill et

al. (1997) performed a pilot feeding study in which female Sprague-Dawley adult

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. rats were fed with several standard laboratory diets for 2-3 weeks. Urine was

collected and analyzed for sphinganine and sphingosine as biomarkers for exposure

to fumonisin. This study revealed a statistically significant correlation (p=0.042)

between sphinganinersphingosine ratio and the fumonisin content of the diets; even

when the fumonisin levels in the diets were fairly low (<4 pg/g). These results

suggested that the presence of fumonisin in natural composition diets should be

considered when feeding studies that might be affected by the presence of this

mycotoxin are designed and interpreted. Furthermore. Martinova and Merrill

(1995) have reported that fumonisins alter immune function. Thus, the content of

fumonisin in laboratory diets might be of particular concern in studies that use

transgenic and/or immunocompromised animals (Merrill et al, 1997).

Sphingolipids are not considered “essential” nutrients and provide few calories.

However, recent findings have shed new light on the nutritional significance of

these highly bioactive components of food (Merrill et al., 1997). As new

information is revealed, it is important to consider that dietary agents that alter

sphingolipid metabolism, such as the fumonisins, could be implicated in diseases

other than the epidemiologically related esophageal cancer.

5. Fumonisin Bi related toxicoses

a. Hepatotoxicity

Hepatic injury is a common manifestation of fumonisin toxicoses in many

species, and seems compatible with the interference of normal biochemical

processes such as sphingolipid metabolism (Ledoux et al., 1992). This condition

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. has been primarily reported in animals with the experimentally produced disease

(Ledoux et al., 1992). The fumonisin induced liver lesion differs from that of most

hepatotoxicants including aflatoxins (Ledoux et al., 1992). In this case,

fumonisins are diffusely distributed throughout the liver. A study conducted in rats

by Norred and Plattner (1993) found that radioactively tagged fumonisin was

accumulated in the liver and kidneys. The concentration of radioactivity was low

but remained constant for at least 96 hours. The sub-acute pathological changes in

the livers of rats caused by FBi were also similar to those caused by culture

material of F. moniliforme MRC 826 (Voss, 1994). In a 90-day feeding study

involving the purified toxin, mouse liver lesions were observed (Gelderblom et al.,

1991). Increased incidence (10 of 15 rats) of hepatocellular carcinoma was shown

on rats fed FB1 (>90% pure) at 50 pg/g in a semi-purified diet for 18-26 months

(Jaskiewicz et al., 1987). Hepatotoxicity was evident after 6 months and

progressed in severity with time. Rats fed com either contaminated with F.

moniliforme or F. moniliforme culture materials developed liver tumors (Hascheck

and Haliburton, 1984). The sub-acute pathological changes in the livers of rats

caused by FBi were also similar to those caused by culture material of F.

moniliforme (Voss, 1994) In equines, hepatotoxicity is manifested by icterus,

hemorrhages and edema (Harrison et al., 1990). Hepatic damage has not been

reported in field cases of fumonisin toxicoses in swine. This may be due to the

rather subtle nature of the liver lesions compared to the grossly evident pulmonary

edema (Harrison et al., 1990). Liver lesions were seen only in the swine

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. experimentally fed com screenings (Colvin and Harrison, 1992). In a feeding study

reported by Colvin et al. (1992), liver changes were observed in all pigs. The mild

lesions were characterized by centrolobular and random hepatocellular cytoplasmic

vacuolar change, hepatocellular cytomegalia, disorganized hepatic cords and early

perilobular fibrosis. At lower doses of fumonisin in swine and equines, the most

prominent feature is slowly progressive hepatic disease. Additional research is

needed before FBi can be considered a hepatotoxin for swine (Colvin and Harrison,

1992). Liver is a target organ in most of the affected animals and is involved in all

fumonisin associated diseases, so analysis of liver enzymes may serve as a

nonspecific indicator of fumonisin toxicoses,

b. Equine leukoencephalomalacia (ELEM)

ELEM is a syndrome clinically characterized by an acute neurological

disorder preceded by lethargy and inappetence and presence of Iiquefactive necrotic

lesions in the white matter of the cerebrum. Necrosis of the gray matter may also

be involved (Harrison et al., 1990; Shieret al., 1991). It appears that this disease

occurs only in equines.

ELEM has been recognized since the nineteenth century as a sporadic,

seasonal, "epidemic-like" condition (Ross et al., 1992). Since its first report in the

United States in 1891, ELEM has been recognized in South America, China,

Greece, Egypt, South Africa and Germany (Ross et al., 1992).

Historically, the diagnosis of ELEM has been based on tissue lesions in

conjunction with clinical signs and the isolation of F. moniliforme from suspected

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. feeds (Marasas et al., 1988). Since the recognition of F. moniliforme as the

causative fungus, ELEM has been reproduced several times under diverse

experimental conditions. Marasas et al. (1988) produced ELEM in a horse by

intravenous administration trying to avoid hepatotoxicity as much as possible.

Research shows that ELEM has been produced in horses given pure FBi by

stomach tube and by feeding naturally contaminated screenings (monitoring for

liver toxicity) (Shier et al., 1991). Because of the uncertain reports, it is not

possible to speculate on what levels of fumonisin contamination may be a problem.

When FBi levels of feed associated with ELEM were compared to those that were

not related, a pattern emerged suggesting that concentrations above 10 pg/g in

horse feed could be involved with ELEM (Marasas et al., 1992). This pattern has

remained constant in cases where the feed is representative. Marasas et al. (1988)

succeeded in reproducing ELEM in a horse by intravenous injection of 7 daily

doses of 0.125 mg FBj/g live mass spread over 10 days. Kellerman et al. (1990)

induced ELEM in two horses with oral daily dose of 1.25-4 mg FBi/g body weight;

both animals developed typical symptoms of ELEM in 25 days. Wilson et al.

(1992) reported that for levels above the 10-20 pg/g range, there should be concern

for toxicity in horses. However, ponies fed 8 pg/g (Osweiler et al., 1992) did show

minor non-specific lesions in liver, kidney and brain stem. Current evidence

suggests that horses that consume feed containing levels as low as 8 pg/g of FB1

may be at risk for developing ELEM (Wilson et al., 1991), but further evaluation of

diets at 8 pg/g FBi is needed. Up to this point, researchers have found that in

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. equines, high dosage levels cause fatal hepatotoxicity and mild brain lesions while

low dosage levels produce mild hepatotoxicity and severe brain lesions (Shier et

al., 1991).

c. Porcine pulmonary edema (PPE)

PPE is a syndrome in which swine experience acute, interstitial pulmonary

edema, hydrothorax and death. Thoracic cavities are filled with a golden-yellow

liquid that clots when exposed to atmosphere; trachea and bronchi contain a clear

foamy liquid (Haliburton and Buck, 1986). Microscopical examination of sections

of lung tissue reveal edema so severe that individual lobules and pleura are

separated from the parenchyma. The alveolar septa are congested but without

hyperplasia or fibroplasia (Haliburton and Buck, 1986). When the dead swine were

analyzed, the absence of epithelial hyperplasia or fibroplasia suggested that these

conditions were perhaps due to an unusual toxin (Colvin and Harrison, 1992). The

lesions induced by FBi are remarkably distinctive and should not generally be

confused with other conditions that induce pulmonary and/or thoracic effusion.

A possible mechanism of the development of PPE is that altered

sphingolipid metabolism causes hepatocellular damage resulting in release of

membranous material into circulation. This material is phagocytosed by the

pulmonary intravascular macrophages (PIM's) thus, triggering the release of

mediators which ultimately results in pulmonary edema. This is emphasized by the

fact that studies in swine indicate that at lower doses of FBi, the most prominent

feature is slowly progressive hepatic disease, while at higher doses, acute

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. pulmonary edema is superimposed on hepatic injury and may cause death (Ledoux

et al., 1992).

The pancreas appears to be similar to the lung in its sensitivity to FBi.

Pancreatic lesions were present in three cases of experimentally induced acutely

fatal FBi toxicoses (Colvin and Harrison, 1992). In cases of suspected FBi

toxicoses in swine, examination of the pancreas should provide valuable diagnostic

information. In 1989-1990 outbreaks of pulmonary porcine edema (PPE) were

reported in different parts of USA (Shier et al., 1991; Bane et al., 1992; Rheeder

et al., 1992). Two workers from Georgia (Marasas et al., 1988) published the

initial report on PPE and its association with FBI. In these two field cases, pigs

consumed com screenings containing 162 and 99 pg/g FBi. Com from the 1989

crop was associated with deaths of 34 mature swine in Southwest Georgia. To

confirm that FBi was the causative agent, the syndrome was reproduced using the

screenings and then intravenous injections of pure FB|. The results of these studies

showed a strong association between F. moniliforme and PPE (Rheeder et al.,

1992). Chemical and mycological investigations have revealed the presence of FBi

at concentrations of 20 to 360 pg/g in suspect swine feeds (Marasas et al., 1988).

Animals fed with com screenings naturally contaminated with 175 pg/g

fumonisins developed PPE, but presented hepatotoxicity when fed doses higher

than 23 pg/g for 14 days (Ledoux et al., 1992; Colvin and Harrison, 1992; Ross et

al., 1991). Purified FBj administered intravenously has been shown to produce

PPE (Ledoux et al., 1992; Colvin and Harrison, 1992; Ross et al., 1991).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Harrison et al. (1990) induced pulmonary edema in a pig by injection of 0.4 mg

FBi of body weight daily for 4 days. As with ELEM. a strong correlation exists

between the fumonisin content of feed obtained from different farms and outbreaks

of PPE. More studies are required in order to obtain the necessary information to

treat and to prevent the disease (Haschek et al., 1992).

d. Human esophageal cancer

Fusarium moniliforme infection of com has been correlated with human

esophageal cancer (EC) risk in Transkei, South Africa where com is a dietary staple

and in China (Haschek et al., 1992). Esophageal cancer also occurs in higher than

normal concentrations in portions of Iran (Kmet and Mahboubi, 1972) and the

Charleston, South Carolina area of the United States (Fraumeni and Blot, 1977).

Transkei has been the most studied region due to the fact that its rate of EC is 50-

100 per 100,000 population while normal occurrence for esophageal cancer

elsewhere is less than 5 cases per 100,000 population (Shier et al., 1992). In this

region, the com is grown locally, harvested and stored in open cribs; and

frequently, it is visibly moldy. A retrospective study of com samples collected over

the period 1976-1989 from Transkenian areas of high esophageal cancer showed

higher levels of fumonisins in com than samples from areas with low incidence

(Rheeder et al., 1992). Contamination of com with Fusarium moniliforme and

fumonisins in the areas of high EC is statistically higher than that found in areas of

the Transkei with lower cancer incidence (Hascheck et a l, 1992; Sydenham et al.,

1990; Bane et a l, 1992). Thiel et al. (1992) reported that based on a daily

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. consumption of 460 g com/day by a 70 kg person, the estimated intake of FBi was

0.44 mg/kg/day. This daily dose is alarmingly high when compared to daily

intakes of 0.6-4.0 mg/kg/day that give rise to ELEM in horses or 3.75 mg/kg/day

that produce hepatocarcinogenesis in rats (Hascheck, et al., 1992). It is important

to point out that com samples from high and low cancer rate areas in the Transkei

are not commercial com but rather home-grown, produced, harvested, stored and

consumed by the people living in rural areas. Laboratory studies have shown a

high correlation between the cancer promoting activity and toxicity of different

strains of F. moniliforme. There is evidence that fumonisins are neither genotoxic

nor mutagenic and do not appear to be metabolized to a significant extent in vivo,

so the mechanism of carcinogenicity is difficult to ascertain (Shier et al., 1992).

Nonetheless, the carcinogenic potential has been shown in laboratory rats (Voss,

1994; Jaskiewicz et al., 1987). Gelderblom et al. (1992) demonstrated

hepatocarcinogenicity in rats by incorporating 50,000 ng/g to the diet in a 26 month

period. In other cancer experiments in rats, approximately 66% of the animals

developed hepatocellular carcinoma while a cancer rate as low as 0.02% can be

considered a significant risk to a human population (Haschek et al., 1992). Results

of experiments with rats imply that FBi is a complete carcinogen and, together with

other related compounds, might be responsible for the hepatocarcinogenicity of F.

moniliforme (Voss, 1994). FBi and FB 2 pose a potential risk to humans as they

occur in food and have been shown to be statistically correlated with the prevalence

of human esophageal cancer in Transkei. However, it is important to establish that

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. cancer manifestation may take a lifetime and the presence of fumonisin has only

been analyzed recently. It is necessary to follow the EC rate and fumonisin

contamination in time to assess if the population develops this condition. On the

other hand, there are social economic, cultural, genetic and a diverse scope of

factors that can contribute to a high cancer rate. Whether the high rate of EC in

regions of the Transkei is caused by fumonisins cannot be determined conclusively,

but the available evidence suggests that their presence may be at least partly

responsible.

e. Developmental toxicity

The developmental toxicology of FBi has been evaluated in different

species. In vivo studies with culture material extracts have reported embryotoxic

effects in hamsters (Floss et al., 1994a,b), mice (Gross et al., 1994), and rats

(Lebepe-Mazur et al., 1995. Studies using purified FBi rather than culture material

reported embryotoxicity to hamsters in vivo (Floss et al., 1994b) and to chick

(Zacharias et al., 1996) and rat embryos in vitro (Johnson et al., 1993; Flynn et al.,

1996). However, FBi embryotoxicity has been questioned by other studies that

have shown negative results. LaBorde et al. (1997) reported lack of embryotoxicity

in New Zealand white (NZW) rabbits at 0.5-1.0 mg/kg/day doses. However, they

reported that FBi was quite toxic to NZW rabbits at doses as low as 0.5 mg/kg/day.

In this study, adverse developmental effects (decreased fetal weight) were observed

only at doses producing maternal toxicity. Another study by Ferguson et al. (1997)

reported minimal maternal or offspring toxicity in rats. This study suggested that

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. doses <9.6 mg purified FB|/kg body weight and/or <1.6 mg FBi/kg body weight

obtained from culture material caused minimal maternal toxicity and produce few

developmental functional alterations. In addition, potential FBi related functional

alterations were prevalent only in males suggesting that there is a mild sex-specific

effect of fumonisin. In developing rats, Kwon et al. (1997) observed that FBi alters

brain sphinganine levels and myelination. These diverse reports emphasize the need

to do more research to understand FB| mechanism of action and its potential risk to

human fetuses.

In addition to high rates of esophageal cancer, it has been reported that the

incidence of neural tube defects (NTDs) in the Transkei is significantly higher than

in other areas of South Africa (Ncyiyana, 1986; Viljoen et al., 1995). The incidence

of NTDs has also been reported to be higher in the Hebei province of the People’s

Republic of China than the average for the rest of the country. This province

contains the Cixian county that is also one of the areas of China having a high

incidence of esophageal cancer (Chinese Birth Defects Monitoring Program, 1990).

In the United States, high rates of NTDs have been observed in Cameron County,

Texas from 1990 to 1991 (Texas Department of Health, 1992) and in Harris

County, TX from 1989 to 1991 (Canfield et al., 1996). The prevalence of these

defects was high among Hispanics (Canfield et al., 1996) for whom com and com

products represent a dietary staple. In 1989-1990 there were reports of clusters of

horse fatalities due to ELEM in this same area in Texas. This event established that

the com crop in that area was contaminated with fumonisins during this period.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Moreover, recent findings have reported that fumonisin Bi induces depletion of

cellular sphingolipids and interferes with folate uptake (Stevens and Tang, 1997).

Folate and its derivatives are vitamins that function as coenzymes in several

metabolic reactions. They are considered essential for the biosynthesis of molecules

such as thymidine and purines, as well as, the conversion of homocysteine to

methionine (Wolf, 1998). It has been reported that dietary folate is taken up by

cells by two mechanisms. The first one uses a high capacity, low affinity protein

transmembrane transporter known as the “reduced folate carrier”. The second one

is a membrane-anchored receptor protein with high affinity for both, folic acid and

reduced folates denominated as “membrane folate receptor” (MFR). The MFR is

important in tissues, such as, the kidneys, the placenta and breast where high folate

levels are required. This receptor is attached to the plasma membrane through a

covalent link to a glycosylphosphatidylinositol anchor (GPI) in a membrane

domain that is rich in cholesterol and sphingolipids. Folate uptake by MFRs

appears to take place by the classical endocytosis process. The MFR on the plasma

membrane is bound and the folate is internalized by invagination that forms

vesicles that are pinched off and enter the cell. After the folate has been unloaded,

the MFR is recycled back to the membrane. It has been suggested that

sphingolipids, particularly glycosphingolipids interact through hydrogen bonding

with the GPI anchor of the membrane protein to localize it in a particular

membrane domain (VanMeer, 1992).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. The involvement of sphingolipids in the folate uptake mechanism led to the

hypothesis that factors affecting sphingolipid metabolism could be involved in

defective cellular folate uptake. Stevens and Tang (1996) reported that intestinal

cells (Caco-2 cells) are depleted of sphingosine after treatment with FBi. Uptake

into the cells of the folate derivative 5-methyltetrahydrofolate was virtually

eliminated by FBi. When ceramide was added to these cells, the folate uptake was

renewed. In this study, the authors concluded that the inhibition of the folate uptake

was caused by the loss of the receptor rather than by its redistribution within the

membrane. The authors showed that the MFR is critically dependent on the

sphingosine and cholesterol levels on the membrane. There is evidence that the GPI

anchor requires a sphingolipid/cholesterol-enriched domain, possibly to target the

receptor to a particular site on the membrane. These results are important since the

risk of fetal NTDs is known to be associated with dietary folate deficiency (Creizel

and Dudas. 1992). Stevens and Tang (1996) considered that inhibition of folate

uptake through dietary exposure to FBi could be related to the high incidence of

NTDs among Texas’ Hispanic population. Furthermore, D’Angelo and Selhub

(1997) have reported a correlation between increased risk of heart disease and

elevated blood levels of homocysteine associated with folate depletion. These

findings have focused attention on fumonisins’ potential role in other human health

problems.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. f. Other diseases

There are several non-specific syndromes that affect other species exposed

to fumonisin. Many reports implicate F. moniliforme with poultry diseases and wild

bird mortality (Voss et al., 1989; Shier et al.. 1992). A 21-day feeding study in

broiler chicks reported that dietary fumonisin resulted in depressed weight gains,

increased organ weights, diarrhea, thymic cortical atrophy, multifocal hepatic

necrosis, biliary hyperplasia, and rickets (Voss et al., 1989). Other researchers

report that com cultures fed to chicken and birds produce immunosuppression.

According to the actual evidence, the level necessary to promote toxicosis in

poultry is below 100 mg/kg (Voss et al., 1989). During 1988 and 1989, a distinct

swine disease syndrome (Mystery Swine Disease - MSD), characterized by

epidemics of prenatal and neonatal mortality and respiratory diseases and immune

system dysfunction in older swine was reported in North America (Gelderblom et

al., 1992). The etiology of this disease is unknown at this time, but Bane et al.

(1991) reported that the distribution of MSD in Illinois was the same as for PPE.

The results of one study (Gelderblom et al., 1992) do not indicate a direct cause

and effect relationship, but show a strong relationship between MSD and feeding

com contaminated with fumonisin at concentrations of 20 pg/g or greater.

Fumonisin may be the initiating "trigger" factor leading either directly or indirectly

to MSD.

Studies in rats show that the kidney is also a sensitive target organ for FB|

fed either as contaminated com or pure FBi (Fincham et al., 1992; Shier, et al.,

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1991). Other studies suggest that the kidney is not an important target organ for

this toxin (Jaskiewicz et al., 1987). The reason for this difference has not yet been

explained. Recently, it has been reported that intravenous injection of FBi at doses

of 1 mg/kg for five days can cause severe and acute kidney damage in rabbits

(Gumprecht et al., 1994). Vervet monkeys, that were fed low levels (0.25-1%) of

F. moniliforme culture material in a diet based on com, low in fat and high in

carbohydrates for up to two years, showed atherosclerosis (Riley et al., 1993).

Research showed that atherogenic potential is secondary to chronic hepatotoxicity

(liver fibrosis) (Riley et al., 1993).

Fumonisins seem to affect an important metabolic pathway in several

animals. This is evidenced by the generalized presence of several degrees of

hepatotoxicity in all the fumonisin-associated diseases that have been studied.

Each species will react in different ways to these alterations, so most of the

toxicoses are species-specific and present a wide variety of syndromes in different

target organs.

D. Inhibition of gap-junction intercellular communication as a non- genotoxic mechanism in carcinogenesis

The general concept of traditional chemical carcinogenesis mechanisms has

relied on the genotoxicity of xenobiotics. The idea that non-genotoxic pathways

can induce cancers has not even been considered until recently. Recently, a

considerable amount of data has been presented on the carcinogenic potential of

compounds with no apparent genotoxic potential. Thus, the scientific community

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. has been faced with the challenge to understand the process of carcinogenesis from

a non-genotoxic point of view, since these compounds pose significant regulatory

dilemmas relating to the risk assessment of human exposure (Ricci and Melton,

1985).

In a simplistic approach, cancer can be regarded as a rebellion in an orderly

cellular society. Cancer cells disregard their neighbors and grow autonomously

over surrounding normal cells. Intercellular communication is considered to play an

important role in maintaining an orderly society through control of cell growth, cell

differentiation and homeostasis (Yamasaki, 1987). Thus, it has been proposed that

disruption of cell-cell communication could be involved in carcinogenesis

(Loewenstein, 1979; Murray and Fitzgerald, 1979; Yotti et al., 1979).

Most cells have two different ways to communicate with other cells. One is

communication mediated by growth factors or hormones where direct cell contact

is not needed. Since this type of communication depends on extracellular factors

that are easily identified, the involvement of these factors in carcinogenesis has

been recognized. There have been several reports showing that many oncogenes

encode growth factors or growth factor receptors (Bradshaw and Prentis, 1987).

The second type of communication is cell contact-mediated intercellular

communication or gap junction intercellular communication (GJIC). During this

type of contact, cells exchange signals directly from the inside of one cell to the

inside of a second cell (Yamasaki, 1996, 1990; Bennett et al., 1991).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. In tissues, cells communicate directly by exchanging low molecular weight

molecules (<1000 daltons) through gap junctions (Simpson et al., 1977). Some of

these exchanged molecules may include sugars, nucleotides, amino-acids and ions

that are important in normal cell function. Second messengers, i.e. calcium, cAMP

and inositol triphosphate can also be exchanged through gap junctions and

participate in tissue control (Saez et al., 1989).

GJIC is mediated by gap junction channels that are composed of protein

channels comprised of connexin molecules. Six connexin molecules form a

connexon in a cell. A gap junction channel is formed when a connexon from a

neighboring cell is aligned and linked (Yamasaki, 1996). These six monomers form

a channel that closes by twisting of the six units (Zampighi and Unwin, 1979). Gap

junctions in different tissues are not always identical. So far, 12 cDNAs coding for

different connexin molecules have been cloned (Beyer, 1993). Connexins are

composed of four transmembrane domains with their C- and N- terminals in the

cytoplasm. The amino acid sequences on the terminal cytoplasm portions vary

significantly between different connexins while the transmembrane and

extracellular portions are more conserved (Zhang and Nicholson, 1989;

Goodenough et al., 1988; Milks et al., 1988). Protein kinase regions are present at

the C-terminal end of some of the connexins, and phosphorylation has been

considered to play an important role in regulating gap junction function (Moreno et

al., 1992). Since connexins are proteins, they are regulated by the usual regulatory

mechanisms that apply for most proteins. These include transcription, mRNA

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. stabilization and translational control, and post-translational phosphorylation. Other

control levels that affect connexin function include, change in cell adhesion

molecules, change in membrane structure and conformation, and growth factors

(Musil and Goodenough, 1995; Meyer et al., 1992; Jongen et al., 1991; Spray et

al., 1987; Zampighi and Unwin, 1979).

The discovery that many tumor promoting agents can block GJIC is the

most supportive evidence of its involvement in carcinogenesis (Trosko and Chang,

1983; Enomoto et al, 1981; Bradshaw and Prentis, Murray and Fitzgerald, 1979;

Yottie et al., 1979). In addition, it has also been shown that inhibition of GJIC by

phorbol esters can be antagonized by various mouse skin anti-promoting agents

such as retinoic acid, glucocorticoids and cAMP (Yamasaki and Enomoto, 1985).

To further support this concept, there are also other lines of evidence that tumor

promoter mediated inhibition of GJIC is associated with enhancement of cell

transformation (Yamasaki and Katoh, 1988). Although there have been studies that

contradict the involvement of GJIC in carcinogenesis, it can be considered as a

plausible mechanism for certain tumor promoters. It is conceivable that multiple

mechanisms are involved and that different chemicals will affect cells through

diverse pathways that will ultimately result in cancer.

While the relevant mechanism that causes the opening and closing of gap

junction channels remains obscure, the involvement of protein kinase C action in the

regulation of intercellular communication has been implicated. It has been

hypothesized that the intercellular protein kinase C is related to the inhibition of gap

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. junction function. Moreover, complex sphingolipids have many important functions in

cell membranes. The regulated breakdown and turnover of complex sphingolipids,

such as, sphingomyelin, are known to result in the formation of lipid "second

messengers". These second messengers act as intracellular signals for turning on and

off cellular processes including the expression of genes and activation or inactivation

of specific proteins, such as, protein kinase C (Riley et al., 1994). The fumonisin-

related inhibition of de novo sphingolipid biosynthesis and transformation may play

an important role in disrupting the production of protein kinase C. Which in turn,

results in disrupted gap-junction cell communication and, ultimately, in cancer

promotion. The simultaneous exposure to a genotoxic carcinogen (aflatoxin) and a

non-genotoxic carcinogen (fumonisin) may have a synergistic effect on the complex

carcinogenesis process.

E. Effects of processing on mycotoxin-contaminated foods*

Mycotoxins comprise a structurally diverse and chemically complex group

that can contaminate a wide variety of food products. Table 2.1 presents a summary

of some of the most commonly affected commodities, the potential toxic mold

genera/species, and the mycotoxins that contaminate them.

Most of the research on the effects of processing has been related to the

aflatoxins, but some processing and stability experiments have been done on other

* This section has been reprinted from Ref. (98-160), p. (407) by courtesy of Marcel Dekker, Inc. (See Appendix C)

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 2.1 Summary of selected reports of potentially toxic molds from various food or agricultural commodities (CAST, 1989)

Commodity Potentially toxic genera/species P o ten tia l fou n d m y co to x in s Wheat, flour, bread, commeal, Aspergillus Penicillium aflatoxins popcorn, tortillas, chips Flaws citrinum ochratoxin A Ochraceus citreo-viride sterigmatocystin Versicolor cyclopium patuiin martensii penicillic acid Fusarium patulum deoxynivalenol Moniliforme pubertum zearalenone fum onisin Peanut, in-shell pecans, nuts Aspergillus Penicillium aflatoxins Flavus cyclopium ochratoxin A Ochraceus expansum patuiin Versicolor citrinum sterigmatocystin trichothecenes Fusarium cytochalasins sp. oosporein Meat pies, cooked meats, cocoa Aspergillus Penicillium aflatoxins powder, hops. Cheese Flaws viridicatum ochratoxin A roqueforti patuiin patulum penicillic acid commune Aged meat products, cured ham Aspergillus Penicillium aflatoxins moldy meats, cheese Flaws viridicatum ochratoxin A Ochraceus cyclopium patuiin Versicolor penicillic acid sterigmatocystin penitrem

Black and red pepper, pasta Aspergillus Penicillium A flatoxins noodles Flavus sp. ochratoxin A Ochraceus (continued)

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Commodity Potentially toxic genera/species P o ten tia l fou n d m ycotoxin s Dry beans, soybeans, com, Aspergillus Penicillium afatoxins sorghum, barley Flavus cyclopium ochratoxin A Ochraceus viridicatum sterigmatocystin Versicolor citrinum penicillic acid Alternaria sp. expansum patuiin Fusarium islandicum citrmtn moniliforme urticae griseofulvin altem ariol altenuene Refrigerated and frozen pastries Aspergillus Penicillium aflatoxins flavus cyclopium sterigmatocystin versicolor citrinum ochratoxin A martensii patuiin citreo-viride penicillic acid palitans citrinin puberulum penitrem roqueforti urticae viridicatum Moldy supermarket foods Penicillium Aspergillus penicillic acid cyclopium sp. trichothecenes Fusarium aflatoxins oxysporum possibly other solani Aspergillus and Penicillium toxins Foods stored in homes, both Penicillium Aspergillus aflatoxins refrigerated and nonrefrigerated sp. sp. kojic acid ochratoxin A penitrem patuiin penicillic acid Apples and apple products Penicillium patuiin expansum

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Aspergillus and Penicillium metabolites, such as, ochratoxin A, citrinin, penicillic

acid and patuiin. More recently, research has been conducted on the effect of

processing on the Fusarium toxins, i.e.. trichothecenes, fumonisins, deoxynivalenol

(DON), T-2 toxin and zearalenone (Scott, 1984). Research in this area is extensive

and beyond the scope of this publication. This section will only discuss a few

examples pertinent to the effectiveness of normal processing operations in reducing

the risks associated with mycotoxin contamination and the industrial applicability

of selected decontamination procedures.

1. Physical methods of mycotoxin removal

a. Cleaning and segregation

Once a contaminated product has reached a processing facility, cleaning and

separating are the first alternatives of control. These procedures are usually non-

invasive; and, except for milling, will not significantly alter the product. In some

cases, these are the best methods to reduce mycotoxin presence in the finished

products, i.e., a significant amount of aflatoxins in shelled peanuts can be removed

by electronic sorting and hand-picking (Dickens and Whitaker, 1975). However,

complete removal of all contaminated particles or aflatoxin cannot be expected

with physical methods of separation. Since the toxin can diffuse into the interior of

the kernel, residual contamination may be present at very low levels in the final

product. If there is a high level of residual contamination, other procedures must be

used to manage the contamination in the final product.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Floatation and density segregation of aflatoxin-contaminated com and

peanuts have been reported to significantly reduce aflatoxin concentrations (Cole,

1989). Kirskey et al. (1989) reported that 95% of the aflatoxin in 21 of 29 samples

of peanuts was contained in kernels that floated in tap water. Furthermore, Phillips

et al. (1994) reported that the mean aflatoxin level was decreased from 301 ng/g to

20 ng/g using floatation as a separation mechanism. Although there is still an

aflatoxin residue, separation has been shown to significantly reduce aflatoxin

contamination and can be considered as an effective first line of defense for certain

products.

Separation of fumonisin-contaminated com, particularly screenings, based

on size has been reported as a possible candidate for decontamination. Com

screenings are broken com kernels that usually contain about 10 times the

fumonisin content of intact com (Murphy et al., 1993). However, screenings

represent a significant proportion of the com used for feed, so it is not possible to

segregate fumonisin contents by screening size without compromising the feed

supply. Further decontamination procedures need to be used to treat com

screenings used for feed.

Although significant quantities (ca. 25%) of deoxynivalenol (DON) can be

removed by cleaning and polishing, the toxin remains in wheat flour at levels

ranging from 60-80% of original toxin levels from the starting wheat (Seitz et al.,

1986; Trenholm et al., 1991). Another study on wheat samples from several states

showed a reduction in deoxynivalenol content in flour from spring wheat that had

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. been cleaned by aspiration prior to milling into flour (Bennett et al., 1992).

However, the toxin residues found in the flour indicate that physical separation

should only be considered a step in a more elaborate risk management plan.

If apples rotted with Penicillium expansum are used for commercial juice

production, the final product may contain patuiin. It has been reported that

trimming of apples to remove the rot reduces the patuiin content by 93 - 99%

(Lovett et al., 1975). In general, physical separation or trimming of apples before

processing is the best method to reduce patuiin contamination. In this case, physical

separation should be considered the focal point of the risk management plan. The

correct use of trimming and separation will render safe high quality commercial

juice.

Cleaning and segregation represent a primary source of the removal of

hazard in mycotoxin contaminated food. In some cases, these procedures will

eliminate the problem. However, in most cases, further processing is needed to

reduce the risks associated with mycotoxin contamination. Although some

contamination may persist, physical separation is a good alternative for industry.

These processes require an initial investment to purchase adequate equipment, but

their maintenance is rather inexpensive. Furthermore, physical removal will not

affect the product's characteristics; so after the initial separation stage, there is no

other significant alteration of the product. With well-calibrated separating

equipment and appropriate planning, this method is a good choice for products such

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. as peanuts, and other tree nuts (aflatoxin), com (aflatoxin and fumonisin), and

apples (patuiin).

b. Wet milling

Wet milling is a process widely used for com. Thus, there have been several

studies that evaluate the effects of this process on mycotoxins associated with this

grain. In this case, an established processing method can be used to manage

contaminated grains. It is important to identify the fraction(s) that remain toxic.

These fractions may then be diverted to less risk uses or subjected to

decontamination procedures. Laboratory studies have reported that during wet

milling of inoculated com, aflatoxin Bi was distributed in the milling fractions. It

went primarily into the steep water (39 to 42%) and fiber (30 to 38%), with the

remainder found in gluten (13-17%), germ (6 - 10%) and starch (only 1%) (Bennet

et al., 1978; Wood et al., 1982). The distribution of mycotoxins in the different

fractions poses new concerns. In a good risk management plan, an individual

contaminated fraction needs to be considered as a new product because it will have

its own unique characteristics. Some of the factors to be considered for a particular

milling product include: the end use, the chemical properties, and the level and type

of contamination. Each product or by-product should be approached differently,

and risk management decisions should be made on an individual basis.

Other mycotoxins are also distributed during wet milling of com.

Zearalenone concentrates in the gluten (49 to 56%) and milling solubles (17 to

26%), but is absent in the starch (Bennett et al., 1978; Bennett and Anderson,

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1978). Recent wet milling laboratory-scale studies on naturally-contaminated com

have shown that zearalenone remains mostly in the gluten and the germ remains

relatively free of the toxin. The starch fractions, which account for 65-71% of

milled products, were free of detectable zearalenone. Milling solubles contained

one to four times the levels of zearalenone in the starting com and gluten fractions.

The most contaminated fraction contained two to seven times the levels in starting

com. Although these fractions account for only 14-19% of the com, they account

for 72-75% of the zearalenone. Fiber and germ fractions contained up to three times

the levels of zearalenone in the starting com and these fractions account for 15-

16% of the milled product. These results show that if a mycotoxin-contaminated lot

of com is processed by wet milling, toxin free starch is produced; however, other

products generally used in animal feed have much higher levels of toxin than the

starting com (Bennett and Anderson, 1989). The concentration of toxin in a

particular fraction simplifies the process of risk management. The other milling

fractions can be considered safe, and decisions need to be taken on a single product.

The management of only one toxic fraction can be simpler than the handling of

several contaminated milling products.

An advantage of wet milling is that additional chemicals can be added to the

steeping solution. So, a modified process could be used to ensure the safety of the

final products. Bennet and Richard (1996) have conducted traditional wet milling

(using lactic acid-sulfurous acid steeping) of two samples of fumonisin B1

naturally-contaminated com. In com containing 13.9 pg/g of fumonisin B|, a

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. portion (22%) of recoverable toxin was found in the steep and process waters.

Other fractions contained fumonisins with gluten having the highest concentration

followed by fiber and finally germ with the lowest concentration. The gluten and

fiber from com contaminated at 13.9 gg/g could pose a risk because they contain

toxins at levels considered to be harmful to certain animals (Ross et al., 1991).

In wet milling of deoxynivalenol-contaminated com, much of the DON

went into the steep liquor, although measurable amounts remained in the starch

(Scott, 1984). Experimental wet-milling of com contaminated with T-2 toxin

showed that 67% of the toxin was removed by the steep and process water (T-2

toxin is more water soluble than aflatoxin or zearalenone). Wet-milling of com

containing ochratoxins A and B showed that 43% of ochratoxin A remaining in the

steeped com went into the process water and solubles on subsequent processing,

with 4% remaining in the germ and 51% in the grits (Wood, 1982).

Mycotoxins have diverse chemical characteristics, so the distribution during

wet milling will vary according to the commodity and the type and level of

contamination. Each case should be considered separately when developing a risk

management plan for wet milling operations,

c. Dry milling

Dry milling will also fractionate the toxin(s). Thus, the identification of the

toxic fractions is needed to adequately use milling as a decontamination procedure.

In dry milling of naturally contaminated com, aflatoxin B| concentrated in the

germ and hull fractions; but distribution varied with the contamination levels. The

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. grits, low-fat meal and low-fat flour (the prime products) contained only 6 to 10%

of the aflatoxin Bi. In artificially contaminated rice aflatoxin was greatly reduced

by milling, with more than 95% in the bran and polish fraction (Achroder, 1968). In

durum wheat, aflatoxin Bj was determined to be in peripheral parts of the kernel.

Upon milling, the aflatoxin concentration in the flour varied according to the

quality of the final product and was maximal in the bran. Prior conditioning

reduced the aflatoxin content by 47% (Scott, 1984).

For zearalenone-contaminated com, all dried milled com fractions

contained zearalenone and only 3 to 10% was removed (Scott, 1984). Dry cleaning

(screening) of the zearalenone-naturally contaminated com does not significantly

reduce toxin levels in com lots. Bennet et al. (1976) reported that all mill fractions

from laboratory scale and commercial processes were contaminated with

zearalenone. The highest levels of toxin were found in the hull and high fat

fractions. Prime product mix (grits, low-fat meal, and flour, which account for 57-

63% of product yield) contained about 20% of the zearalenone from the starting

com samples (Bennett et al. 1976).

No specific dry milling studies have been conducted on fumonisin-

contaminated grains. However, assays on com based foods have shown that the

toxin is present in human foodstuffs, i.e., com grits and com meal (Bennett and

Richard, 1996; Sydenham et al., 1991; Stack and Eppley, 1992). On the industrial

perspective, milling can be used as an effective method of separation. If the

distribution of the toxin is determined, and alternate management of the toxic

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. fraction is developed, milling can be considered a good cost/benefit method. A

combination of cleaning, sorting and milling can significantly reduce the risks

associated with mycotoxins.

2. Physical methods of detoxification

a. Thermal inactivation

Thermal inactivation is a good alternative for products that are usually heat

processed. However, some of the mycotoxins are chemically stable at processing

temperatures. Thermal inactivation should be evaluated for the conditions of a

particular process. Additional processing steps, i.e., use of compounds that would

promote the chemical degradation of a toxin, could be employed to ensure the

safety of the procedure. In some cases, a change in conditions such as pH or

moisture content will contribute to the chemical modification of a particular toxin.

Aflatoxins are resistant to thermal inactivation, and are not destroyed

completely by boiling water, autoclaving or a variety of food and feed processing

procedures (Christensen et al., 1977). Aflatoxins may be destroyed partially by oil-

and dry roasting of peanuts. However, Stoloff (1982) reported that aflatoxins are

generally stable in peanut materials at room temperature. Baur (1975) found no

significant changes in levels of aflatoxins Bi, B 2, G| and G 2 in peanut meals, and

raw and roasted peanut butter stored at 23° C for two years. Lutter et al. (1982)

reported that alternative peanut roasting methods, i.e., microwave roasting, destroy

aflatoxins almost completely. Unfortunately, this process would involve an

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. increase in processing cost, making it commercially impractical. Thus, microwave

roasting is not a good industrial solution, even if it yields a safe product.

Roasting has been a good method for certain commodities, i.e., peanuts. As

mentioned before, if an operation that is regularly used in processing is an

effective decontamination procedure, then that operation would be the first choice

to manage that particular product. This would provide a safe, inexpensive risk

management method. An example of roasting as an effective means of control is its

use in coffee processing. It has been reported that roasting green coffee beans

containing added aflatoxin Bi at 200° C produced a 79% toxin loss after 12 min.

and more than 94% loss after 15 min. (Levi, 1980). Furthermore, roasting of green

coffee beans at 200° C for 20 min. destroyed 6 8 % of added sterigmatocystin under

laboratory conditions (Levi et al., 1975). It is important to mention that these

studies were performed with spiked coffee beans. The use of spiked material for

experimental processes gives useful information. However, the effects of roasting

should be confirmed with naturally-contaminated green coffee beans because

naturally occurring toxins may have different interactions with the food matrix. It is

important to confirm the effectiveness of a processing operation under industrial

conditions before using it as part of a risk management plan.

Several studies have reported that aflatoxin Bi is moderately stable in

heated peanut oil, com oil and coconut oil. Therefore, flying in unrefined oils

could add aflatoxin during processing (Dawarkanath et al., 1969; Peers and Linsell,

1975). Fortunately, oils are not of concern because only a small percentage of

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. aflatoxin Bi present in oilseeds passes into extracted or pressed oil; and refining

and bleaching operations essentially eliminate it (Parker and Melnick, 1966).

Gracian (1980) reported that the average loss of total aflatoxin concentration in

pressed olive oil was 76% .

Conway et al. (1978) reported that roasting com (145-165° C) yielded a

reduction of 40-80% of the original aflatoxin Bj. Studies on cooking of aflatoxin-

contaminated rice found that normal cooking of rice destroyed about 49% of

aflatoxin B§; pressure cooking and cooking with excess water destroyed more of

the aflatoxin Bi in the rice (73 and 82%, respectively). These results provide

evidence of the influence of water on stability of aflatoxin during heat processing.

Furthermore, when pasta products contaminated with aflatoxins were cooked for 10

min. in boiling water, 29% of the aflatoxin was found in the drained water and 6 6 %

in the cooked pasta (Stoloff et al., 1978).

Surveys have shown that thermally processed com products, i.e., canned

com, tortillas, and ready to eat cereals generally have lower concentrations of FB|

than do unprocessed products, i.e., com meal and grits (Parker and Melnic, 1966;

Pittet et al., 1992; Doko and Visconti, 1994). Since most corn-based foods receive

some sort of thermal processing, it has been suggested that this kind of procedure

could have an effect on the fumonisin content in food. However, results reported

by Jackson et al. (1995), Alberts et al. (1990), Dupuy et al. (1993), and Scott and

Lawrence (1994) indicate that the fumonisins are fairly heat stable compounds. In

general, it has been reported that the loss of FBi and FB 2 was more rapid and

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. extensive in alkaline or acidic environments than at neutral pH. Thermal processing

operations, such as, boiling or retorting, which occur at temperatures <125°C, have

little effect on the fumonisin content of food; however, foods that reach

temperatures >150°C during processing may have significant losses of fiimonisins.

Although some of these studies show decomposition of fiimonisins, the toxic

potential is not necessarily eliminated (R. Lopez-Garcia and D. L. Park,

unpublished data); therefore, detoxification of these compounds cannot be

assumed. Further work is needed to determine the toxicity of thermally processed

fumonisin-contaminated foods (Jackson et al., 1996). In this case, thermal

processing is not a good choice for industry. Alternative methods for fumonisin

decontamination need to be applied.

Studies on the effect of heat processing of ochratoxin A-contaminated flour

have reported that considerable reduction of ochratoxin A (76%) occurred in

samples of white flour heated to 250°C for 40 minutes. Ochratoxin A appears to

be more readily destroyed in dry cereal than in the presence of water (unlike

aflatoxin Bi and patuiin) (Scott, 1984). This was also evident in a study reported by

Osbome where ochratoxin A was not degraded during breadmaking, but 62% was

lost after baking of biscuits, which have a lower water content (Osbome, 1979).

Processing of other commodities naturally contaminated with ochratoxin A

have shown that after blanching, salting and heat processing beans, about 53% of

ochratoxin A remained (Harwig, 1974). In this case, it is important to determine if

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. the heat processing will yield products with unchanged technological

characteristics.

Several authors have studied the effect of making bread on aflatoxin present

in w'heat flour or in maize flour (El-Banna et al., 1983; Reiss, 1978). Variable

destruction levels have been observed, but there are still conflicting results. Bennett

and Richard (1996) reported that baking does not destroy or significantly reduce

levels of deoxynivalenol. DON has proven to be more heat stable during food

processing than any other mycotoxin tested. No DON was destroyed during the

baking of different ethnical products such as Egyptian bread. Western style bread

or cookies baked from hard wheat flour (Scott, 1984).

b. Irradiation

The effects of irradiation on most mycotoxins is relatively unknown.

However, with the state of today's processing facilities, irradiation does not seem to

be a good commercial method. The cost involved in setting up an irradiation

facility and the reluctance of the consumer to accept irradiated products makes this

method inaccessible to most industries.

Feuell (1966) reported that gamma irradiation (2.5 Mrad) did not degrade

aflatoxin in contaminated peanut meal, and UV light produced no observable

change in fluorescence or toxicity of the treated sample. However, exposure to

contaminated peanut oil to short-wave and long-wave UV light resulted in a

reduction of aflatoxin levels (Feuell, 1966; Shantha and Sreenivasamurthy, 1977).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. c. Adsorption from solution and covalent binding

Adsorption from solution or covalent binding of aflatoxins to non-biological

materials are good methods for aflatoxin decontamination. In industries such as the

oil refineries, the use of adsorbent materials is readily available and is part of

normal processing operations. A variety of adsorbent materials; i.e., activated

carbon and clays, has been shown to bind aflatoxins in aqueous solutions, and

certain aluminosilicates have been reported to bind aflatoxins in peanut oil and

animal feeds. The long-term effects and safety of the aluminosilicates has not been

determined (Mansimanco et al., 1973; Decker, 1980; Machen et al., 1988).

However, a phyllosilicate clay (HSCAS or NovaSil™) that is currently used as an

anti-caking agent in the feed industry has been studied for its ability to bind

aflatoxin and reduce the toxicity/mutagenicity of the final product. Phillips et al.

(1994) have reported that this clay [1] tightly binds aflatoxins in aqueous

suspensions; [ 2 ] markedly diminishes aflatoxin uptake by the blood and

distribution to target organs; [3] prevents acute aflatoxicosis in farm animals; and

[4] decreases the levels of aflatoxin Mi residues in milk. In this case, phyllosilicate

clay is an effective decontamination agent that is readily used in the feed industry.

However, its only approved use is as an anti-caking agent and its use for removal of

aflatoxin from feed is not permitted in certain countries including the United States

of America. Thus, its beneficial effects are limited to a particular industrial process.

The bentonite clays used in oil purification (to remove pigments) seem to adsorb

aflatoxins present in the unrefined oil. Activated charcoal has also proven to be

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. effective in reducing the patuiin in naturally contaminated fruit juices (Sands et al.,

1976; Chu et al., 1975). More research is needed for the application of this method

with other mycotoxins and commodities.

3. Biological decontamination

Biological methods are a good option for the fermenting industry, i.e.,

breweries. In this case, however, it is important to determine the yeast's viability in

the presence of mycotoxins. It is also important to determine the effect of

mycotoxins on the yield and development of secondary characteristics associated

with the final product. It is relevant to note that biological methods demonstrating

effective decontaminating properties are usually the result of specific compounds

produced by selected microorganisms. When this occurs, it is usually more efficient

and economical to add the active agent directly.

Chu et al. (1975) reported that aflatoxin Bj is not completely removed

during the beer brewing process, i.e., 18 and 27% remained from two

contamination levels of starting material, 1 and 10 ng/g, respectively. It was

relatively stable in the cooker mash treatment, but significantly lost in the wort

boiling and final fermentation steps. Dam et al. (1977) found that aflatoxin B[ is

reduced by 47% after cooking and fermentation of com, wheat or com:milo.

Following distillation, a concentration effect was observed more aflatoxin B| was

present in the wet solids (dry weight basis) than in the starting com (Lindroth,

1980). Aflatoxin from contaminated com does not go into the alcoholic distillate,

but accumulation of aflatoxin in spent grains is a potential problem when using this

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. material as animal feed. Further decontamination procedures must be used if these

by-products are to be used as animal feed (Dam et al.,\911\ Lindroth, 1980).

Bennet and Richard (1996) reported that although pure fumonisins exhibit

considerable water solubility, residues of toxin remain in the distillers’ dried grains.

This fermentation product accounts for 31-35% of the total fumonisin B|, in the

starting com. Facilitated by solubility in water, 51-54 % of fumonisin Bi, in

starting com was extracted into whole stillage. No toxin could be detected in

distilled ethanol. The fermentation process did not destroy fumonisin and 85% of

the toxin could be recovered in the products. Products from ethanol fermentation of

fumonisin-contaminated com generally used as animal feeds could be detrimental if

consumed by pigs or horses, animals sensitive to relatively low levels of this toxin.

The use of zearalenone contaminated com for ethanol production was

investigated by Bennet et al. (1981). The presence of toxin had no apparent effect

on ethanol yields and there was no carry-over from zearalenone to distilled ethanol.

However, the fermentation process did not destroy the toxin and recovered solids

contained 2 to 2.5 times the level of zearalenone in the starting com. Traditional

fermentation for brewing of com beer has shown 51% carry-over of zearalenone

into the finished product (Okoye, 1987). Fermentation by Saccharomyces

cerevisiae of wort containing zearalenone resulted in conversion of 69% of the

toxin to beta-zearalenone, a metabolite with less activity than the parent compound

(Scott et al., 1992). Chu et al. (1975) reported that 91 to 96% of ochratoxin A

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. remained after the cooker mash treatment of the beer brewing process and none

was destroyed during pasteurization and boiling of beer itself (Chu et al., 1975).

Almost complete destruction of patuiin occurs during alcoholic

fermentation of apple juice (Burroughs, 1977). Fermentation of juice containing

patuiin at a level of 40 pg/mL using S. cerevisae effectively eliminated patuiin in 2

weeks (Harwig et al., 1973). Winemaking has also been reported to destroy patuiin

(Scott, 1977).

In a survey performed by Scott et al. (1993), deoxynivalenol was found in

29 of 50 samples of Canadian and imported beers. Nine samples contained greater

than 5 ng/mL deoxynivalenol. There are conflicting results on DON stability during

the beer making process. Data is dependent upon the extraction and the analytical

method.

4. Chemical inactivation

Ammoniation of com, peanuts, cottonseed, and meals to alter the toxic and

carcinogenic effects of aflatoxin contamination has been subject to intense

research efforts. Results of these studies have been summarized by Park et al.

(1988). Reports on the extensive evaluation of this procedure demonstrate support

for the efficacy and safety of ammoniation as a practical solution to aflatoxin

decontamination in animal feeds. This process has been successfully used for many

years in the U.S., France, Senegal, Sudan, Brazil, Mexico and South Africa. The

practical applications along with research results strongly support the use of the

ammonia treatment to reduce risks posed by aflatoxin contamination. Particularly

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. in the high temperature/high pressure where process parameters are clearly defined

and followed. With the ambient temperature/atmospheric pressure method, caution

must be exercised to assure even distribution of ammonia gas/water or aqueous

ammonia solution in the product. This process is limited to whole kernel seed/nut

products and has not been successful for ground or meal products due to the

inability of the ammonia to penetrate the commodity. Norred et al. (1991) found

that atmospheric pressure/ambient temperature ammoniation reduced fumonisin

content of F. moniliforme culture material but did not reduce the toxicity of

material when fed to rats. Park et al. (1992) reported 79% reduction of fumonisin

levels in com after high pressure/ambient temperature ammoniation followed by a

low-pressure/high temperature treatment.

Nixtamalization, the traditional alkaline heat treatment of com used in the

manufacture of tortillas, reduces significantly the levels of aflatoxin (Ulloa and

Herrera, 1970; Ulloa Sosa and Schroeder, 1969). However, subsequent studies have

shown that much of the original aflatoxin is reformed on acidification of the

products (Price and Jorgensen, 1985). This treatment has also been proposed as a

method to detoxify fumonisin-contaminated com (Sydenham et al., 1991). A

feeding study reported by Hendrich et al. (1993) showed that nixtamalization

greatly reduces FB i content. However, when nixtamalized com was fed to weaning

rats, the toxicity of the inoculated com was not reduced. The major toxic product of

fumonisin hydrolysis during nixtamalization appears to be hydrolyzed fumonisin

Bi (HFBO, but it is possible that other breakdown products, such as calcium-FBi

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. complexes and other products may also play a role in the toxicity of nixtamalized

com. The hydrolysis of FB| by treatment with Ca(OH )2 and heat does not render a

detoxified product. It produces a more toxic final product. Feeding weaning rats

with culture material containing 8-11 pg/g of HFBi produced some toxicity

symptoms as severe as did 45-48 pg/g FBi. These data indicate that the

production of hydrolyzed fumonisins in the processing of goods for human

consumption may be of significant concern. Park et al. (1996) have reported a

"modified nixtamalization" procedure that involves the addition of hydrogen

peroxide and sodium bicarbonate to the traditional process. Results from this study

show 40% reduction of brine shrimp mortality when comparing the products of

traditional nixtamalization and modified nixtamalization. Traditional

nixtamalization is probably not a useful strategy for fumonisin detoxification

(Murphy et al., 1995). However, the modified procedures where other agents, such

as, hydrogen peroxide and/or sodium bicarbonate are used, pose a good alternative

for Latin American countries where nixtamalized com is a major component of

the diet.

It has been shown that sodium chloride has marked influence on loss of

aflatoxins in artificially contaminated unshelled peanuts. Cooking in water in a

pressure cooker for 0.5 h caused a loss in concentration of 80-100% with sodium

chloride, but only 35% without the salt (Scott, 1984). Sodium bisulfite has been

shown to react with aflatoxins (Bi, Gi, Mi, and aflatoxicol) under various

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. conditions of temperature, concentration, and time to form water-soluble products

(Hagler et al., 1982).

Non-enzymatic browning has been proposed as a decontamination

procedure for fumonisin Bi. When FBi was incubated with glucose or fructose to

mimic non-enzymatic browning conditions, a first order rate reaction was observed.

This showed that the amine group of FBi was derivatized by the sugar. Preliminary

cell tissue culture tests, with these adducts suggest they are less toxic than FBi

(Murphy et al., 1995).

Both, patulin and penicillic acid are moderately stable in apple juice, with a

half-life of several weeks (Pohland and Allen, 1970). The rate of disappearance of

these toxins increased markedly with the addition of ascorbate or a mixture of

ascorbate and ascorbic acid to the juice (Brackett and Marth, 1979).

Processed foods are very complex systems. Processing not only alters the

food but also adds new ingredients and conditions. These new factors change the

environment, so many new interactions can occur. Exploring the application of

known food additives to mycotoxins as an integral part of the process can open a

lot of opportunities for risk management.

5. Other industrial processes

Since molds and their secondary metabolic products are ubiquitous in foods

and feeds, there have been several reports on the effects of processing and storage

on mycotoxins present in commodities other than grains and oilseeds, i.e., cured

meats, dairy products, and fruit products. It has been reported that recovery of

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. aflatoxins added to meats is only 16% from raw ham, and 19% from salami after

storage for 6 to 8 weeks at 5 to 37° C (Stott and Bullerman, 1976). Storage at 5° C

for shorter periods of time (up to 1 week) did not destroy aflatoxins Bi and Gi in

bologna or cheese (Lieu and Bullerman, 1977), nor did curing and cooking bacons

and hams greatly reduce aflatoxin levels (Furtado et al., 1981). Van Egmond

(1983) reported that sterigmatocystin was stable in cheese for at least 3 months at

temperatures of -18° to 16°C. Aflatoxins B| and Gi were completely unstable in

berry jams stored for 6 months at 22°C, although 26 to 38% of aflatoxins B 2 and G 2

were recovered (Pensala et al., 19?).

Aflatoxin Mi is stable during pasteurization (Wiseman et al., 1983).

Aflatoxin Mi has been reported to be more stable in naturally contaminated milk

than in freeze-dried spiked milk. Evidence has been obtained that aflatoxin Mj

binds to milk protein casein (Scott, 1984)

In most cases, there is no single method that will insure the complete

removal of the contaminant. Integrated management with a combination of

different methods can significantly reduce the risks associated with mycotoxin

contamination and yield products that are safe and acceptable to the consumer.

The ubiquity of Aspergillus, Fusarium and Penicillium in food and feed

commodities, as well as, their ability to produce a wide variety of toxic compounds,

makes mycotoxicology a complex field. There is great need for more research in

areas, such as, toxin production, identification, toxicological properties, potential

interactions and control alternatives.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. More information is needed on the ecological relationships between

different mold species and their effect on mycotoxin production. This information

could lead to a better understanding of the role of mycotoxins in fungal physiology

as well as in plant pathology. Since there have been reports on the co

contamination of different mold species in a wide range of commodities, their

interactions need to be understood to more clearly define the factors that affect

toxin production in the field, during harvest and in storage. This information would

be useful in the development of preventive measures and possible biological

control mechanisms. Furthermore, the understanding of fungal interactions could

also lead to a better assessment on types and quantities of toxins produced under

certain conditions. Therefore, facilitating the evaluation of potential human and

animal exposure to different toxic agents.

Toxicological perception of the effect of exposure to a single compound

gives valuable information on its mode of action and potential effects. However, in

reality, humans and animals get exposed to a wide variety of xenobiotics that can

have potential interactions. These effects can have positive or negative influences

on the ultimate toxic outcome. Although the study of the toxicological effect of

complex chemical mixtures is a difficult area, research on the potential interactions

between agents that are known to co-exist in nature (i.e., aflatoxins and fumonisins)

could give valuable information on the toxicological significance of more complex

mixtures.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. It would be unreasonable to believe that all toxic outcomes are the result of

single events rather than a combination of factors. If a "healthy'’ status is

considered to be a physiological equilibrium or homeostasis, we need a better

understanding of the factors that upset this state. Intrinsic individual factors, such

as, genetic predisposition, general health status, age and gender, as well as extrinsic

factors such as diet and environment among others play an important role on the

equilibrium. All of these are very complex areas that interrelate on several levels.

From the food toxicology point of view, research in mycotoxicology could lead to a

better understanding of the role of diet in health and disease.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. III. Preliminary evaluation of the production of aflatoxins and fumonisins by Aspergillus flavus and Fusarium moniliforme during storage at optimal mold growth conditions

A. Introduction

Although it is generally assumed that cereal grains are pristine in nature and

that agricultural commodities are a source of good nutrition and health, the

contamination of food and feeds with naturally-occurring toxins, i.e., mycotoxins,

has presented new concerns. These contaminants are persistent, unavoidable and

unpredictable. Thus, every year, large amounts of food and feed are contaminated

with a diverse range of mycotoxins. These toxins differ in chemical nature and

biological activity; and in some cases, {i.e., aflatoxins) can present a serious health

risk to humans and animals. Therefore, the understanding of the conditions that

affect their production and stability is an essential tool for the design of strategies

that could prevent their formation or control their proliferation.

Aspergillus flavus and Fusarium moniliforme are widely distributed

throughout the world. Although each mold has its unique mode of action, their co

existence in plants, such as, com indicates that there could be an interactive

relationship between both species. In the past it was widely believed that one mold

species persisted over the other; however, more recent reports have suggested that

they can co-exist in the same environment (Chamberlain et al., 1993). This concept

leads to new research questions. If these species co-exist, how does this affect the

biosynthesis and degradation of secondary metabolites like mycotoxins? If there is

an interaction, are mycotoxins produced as a defense mechanism? Does the

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. presence of other mold species trigger the production of certain biologically active

compounds?

Chamberlain et al. (1993) reported that A. flavus and F. moniliforme infect

com by different routes. While A. flavus is considered a non-pathogenic fungus that

colonizes com through kernel damage into the outer seed covering; F. moniliforme

is recognized as a pathogen that enters the kernel through the pedicle and occupies

the internal space (Zummo and Scott, 1990). Thus, since in theory, these molds

occupy different microhabitats within the com plant, it could be assumed that they

do not interact. However, Zummo and Scott (1992) reported that inoculation of

kernels with both fungi resulted in inhibition of A. flavus and decrease in aflatoxin

concentration. Conversely, Chamberlain et al. (1993) observed that in com

contaminated with both, aflatoxigenic and fumonisin-producing fungi, the

concentrations of aflatoxin and fiimonisin were independent. These reports

suggested that when conditions within the kernels are favorable, both, aflatoxins

and fumonisins can be produced.

After harvesting, and during storage the plant structure is altered allowing

more opportunity for mold interaction. It has been observed (Lopez-Garcia and

Park, unpublished data; Price, personal communication) that during storage of

ground aflatoxin-fumonisin- co-contaminated com, changes in toxin concentrations

occur. During storage under refrigerated conditions (ca. 4°C), aflatoxin

concentrations decreased while fumonisin concentrations increased. This suggested

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. a dynamic system where fungi were interacting. Thus, new research concerns were

presented.

The objective of this study was to provide preliminary information on the

relationships between toxigenic strains of A. flavus and F. moniliforme in cracked

com during a short storage period under moisture and temperature conditions that

favor mold growth. Also, the toxin-producing potential of untreated com under the

same moisture and temperature conditions was determined. Since mold growth is a

complex process, one can expect the production of innumerable compounds other

than aflatoxins and fumonisins. Thus, the toxic/mutagenic potential of raw com

extracts from different sampling periods was also determined. This study will only

provide initial data on the toxin production by the A. flavus and F. moniliforme.

This information can be used to develop other more detailed studies.

B. Materials and methods

1. Corn and fungal and bacterial cultures

Non-contaminated whole kernel com samples were obtained from Cargill

(Port Allen, LA). Fusarium moniliforme (A00102, Leslie et al., 1992) was kindly

provided by Dr. Raymond Schneider (Louisiana State University, Department of

Plant Pathology, Baton Rouge, LA) and Aspergillus flavus (13L, Cotty, 1989; 1990)

isolated from Arizona agricultural soil was provided by Dr. Robert Brown (United

States Department of Agriculture, Southern Regional Research Center, New

Orleans, LA). Salmonella typhimurium tester strains TA-100 and TA-98 were

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. kindly provided by Dr. Bruce N. Ames (University of California Berkley, Berkley,

CA).

2. Chemicals

Acetic acid, ammonium sulfate, ampicillin, D-biotin, disodium

hydrogen phosphate, glucose- 6 -phosphate, L-histidine, magnesium sulfate, 2-

mercaptoethanol, ortho-phtaldehide, sodium ammonium sulfate, disodium

tetraborate, sodium chloride, sodium dihydrogen phosphate, 6 -nicotinamide

adenine-dinucleotide phosphate (NADP) and trifluoroacetic acid were all purchased

from Sigma Chemical Co. (St. Louis, MO). Chloroform (HPLC grade), citric acid

monohydrate, dimethyl sulfoxide, ethyl ether anhydrous, hexane (HPLC grade),

methanol (HPLC grade), potassium chloride, potassium phosphate dibasic

anhydrous and water (HPLC grade) were acquired from Mallinckrodt Baker, Inc.

(Paris, KT). Bacto agar, potato dextrose agar and crystal violet were ordered from

Difco Laboratories (Detroit, MI); glucose was from Fisher Scientific (Fair Lawn,

NJ). Acetonitrile (HPLC grade) was from Baxter Co. (Muskegon, MI). Ethanol was

obtained from Aaper Alcohol and Chemical Co. (Shelbyville, KT). Oxoid nutrient

broth No. 2 was purchased from Unipath LTD (Basingstoke, Hampshire, England).

Post-mitochondrial supernatant (S9 mix) was purchased from Molecular

Toxicology, Inc. (Boone, NC).

Pure fumonisins Bi and B 2 were kindly provided by Dr. Robert M. Eppley

(Center for Food Safety and Applied Nutrition - CFSAN, U. S. Food and Drug

Administration-FDA, Washington, D.C.). Pure aflatoxin Bj was generously

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. provided by Dr. Mary Trucksess (CFSAN-FDA, Washington, D. C.). Aflatoxin Bi,

Bi, Gi, and G 2 HPLC standard 50 pg/g mixture was purchased from Romer

Laboratories, Inc. (Union, MO).

3. Corn inoculation

Inoculation and com culture was performed according to the method

reported by Leslie et al. (1992). Briefly, yellow aflatoxin/fumonisin free whole

com kernels were coarsely cracked in a Willey mill, then homogenized in a tumble

blender. Samples (300 g.) were placed in sterile Fembach flasks, and 33 ml of

sterile distilled water were added to each flask. The flasks were covered with a

cheesecloth plug and aluminum foil; then autoclaved at 121°C for 30 minutes.

After autoclaving, another 33 ml of sterile distilled water were aseptically added to

each flask.

Fungi {A. flavus and F. moniliforme) were cultured on potato dextrose agar.

After colonies had covered the entire petri dish, culture material was aseptically

scraped off and rinsed with sterile distilled water (ca. 10 ml/plate). The mold/water

suspensions from each plate were collected to make a stock inoculating suspension.

In the case of the combined sample, equal amounts of Aspergillus and Fusarium

suspensions were mixed thoroughly. This procedure ensured that com samples

would have equal mold inoculi. Seventy five ml of the conidia suspension were

added to their respective flasks (ca. 10 7 conidia/flask) (Table 3.1).

Flasks were incubated in the dark for 45 days at 25°C. The cultures were

thoroughly mixed with a sterile metal spatula every other day to evenly distribute

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. the inoculum. Different spatulas were used for each sample and the hood was

thoroughly disinfected with 70% ethanol between samples to avoid cross

contamination. For the naturally contaminated samples, the same amount o f total

sterile distilled water was added (141 ml), but the samples were not autoclaved.

Therefore, the naturally occurring fungi and other micro flora, i.e., yeast and

bacteria were allowed to grow. The control samples had the same amount of water

added, but they were not inoculated. Samples (approximately 100 g.) were removed

from of each flask at 15, 30 and 45 days of incubation. The samples were placed in

sealed plastic bags and stored at -20°C until analysis.

Table 3.1 Com inoculation design

Sample N Heat treatment Water added Inoculum A 3 121°C; 30 min 141 ml A. flavus F -» 121°C; 30 min 141 ml F. moniliforme A+F 3 121°C; 30 min 141 ml A. flavus and F. moniliforme

NC J No treatment 141 ml No inoculum Control 3 121°C; 30 min 141 ml No inoculum All samples were analyzed in triplicate

4. Safety precautions

Since highly toxigenic strains of Aspergillus and Fusarium were used in this

study, several safety measures were taken to avoid exposure to toxic material and

mold spores. A half-face negative pressure respirator equipped with HEPA filters

was used every time the com samples were handled. Also, protective eye wear and

nitrile gloves were worn at all times. All samples were handled in a biological

laminar flow hood. During the first portion of the storage spore high production

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. was high; so special care was taken to prevent airborne spore contamination by

thoroughly cleaning the hood after each flask was handled.

5. Toxin extraction

Com samples (100 g.) were placed in a blender with 300 ml MeOH:HzO

(77:23) and blended at high speed for 5 minutes. The extracted slurry was filtered

through Whatman # 1 filter paper. The filtrate was considered the crude com

extract.

6. Aflatoxin purification

For the 15 day samples, aflatoxin clean-up and purification was performed

according to the method reported by Thean et al. (1980). Seventy five ml of the

crude com filtrate were placed in a 250 ml beaker with 75 ml of 0.61 M

(NH4 )aS0 4 solution and 5 g of celite 545 analytical filter aid. The solution was

mixed for 2 minutes and then filtered through Whatman #2 fluted filter paper.

Then, 100 ml of this filtrate were placed in a 250 ml separatory funnel and 5 ml

CHCI3 were added. The mixture was shaken vigorously for 2 min. The phases were

allowed to separate and the CHCI 3 fraction was drained into a 20 ml scintillation

vial. This process was repeated with a second 5 ml CHCI 3 . Both CHCI3 fractions

were collected in the same vial. The CHCI 3 extract was evaporated to complete

dryness under a N 2 stream.

To purify the aflatoxin from the CHCI 3 extract, commercially available C-

18 solid phase extraction columns (Bond Elut, Varian, Harbor City, CA) were used.

The columns were pre-conditioned with 3 ml hexane. The dry extract was

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. dissolved in 0.5 ml CHCI 3 using a Vortex mixer. Care was taken to wash all the

vial walls. The extract was then diluted with 0.5 ml hexane, and applied to the pre

conditioned C-18 solid phase extraction column. The vial was rinsed with 1 ml

CHCL3 :C6H6 (1:1) and the rinse solution was also applied to the column. The

column was then washed with 5 ml hexane and 5 ml ethyl ether anhydrous. Finally,

the aflatoxins were eluted with 3 ml chloroformrethanol (95:5) into clean

scintillation vials. The eluted extracts were evaporated to dryness in a warm water

bath (ca. 40°C) under N 2 .

The 30 and 45 day samples had copious mold growth. When this extraction

procedure was used, the presence of interfering compounds was observed. Mold

apparently produced a multitude of unidentified compounds, some of these seem to

have emulsifying properties that prevented the separation of the organic phase

(CHCI3). Several attempts were made to break the emulsion (/. e., addition of NaCl,

increasing the volume of CHCI 3, adding octanol, freezing to -20°C and

centrifuging). However, even when part of the CHCI 3 phase was separated,

recovery from extracts spiked with 100 ng pure AFBiwas below detectable levels.

Therefore, it was decided to use a purification method that did not involve

liquid: liquid extraction. These samples had to be extracted and purified following

the method reported by Wilson and Romer (1991) with slight modifications. The

recovery of spiked solutions using this method was 90-95%. Briefly, 0.5 ml of the

crude methanokwater extract were mixed with 4.5 ml acetonitrile:water (9:1) in a

capped culture tube and vortexed for 1 min. Multifunctional cleanup (MFC)

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. columns (Mycosep #224-Romer Labs., Inc.. Washington. MO) were used to purify

this extract. The Mycosep column was fitted over the culture tube and the flanged-

end was pushed into the solution, letting the crude extract pass through the column.

Approximately 1 ml of the purified extract was collected and stored (4°C) until

HPLC analysis.

7. Quantification of aflatoxins by liquid chromatography (HPLC)

The samples were analyzed following the method reported by Wilson and

Romer (1991). The 15 day dry extracts were reconstituted in 1 ml acetonitrilerwater

(9:1). Thirty and 45 day samples were already in an acetonitrilerwater (9:1)

solution. To derivatize the aflatoxins, a 200 pi aliquot of the purified extracts was

transferred into an HPLC auto-injector vial (Waters, Milford, MA), and 700 pi of

trifluoroacetic acid derivatizing reagent (distilled water:trifluoroacetic acid:acetic

acid - 7:2:1) were added. The samples were then placed in a 65°C water bath for

8.5 min. The vials were then cooled in an ice-water bath.

The samples were placed in an HPLC autosampler (Waters 717 plus

autosampler - Waters, Milford, MA). The analysis was carried out using a Waters

modular system with a Waters 470 scanning fluorescence detector (360 nm

excitation and 440 nm emission) and a Novapak C |8 reverse phase column. Fifty

pL of sample were injected and analyzed using a watenacetonitrile (8:2) mobile

phase with a 2.0 ml/min. flow rate. Approximate retention times for pure aflatoxins

Gi (G2a), Bi (B2a), G2 and B 2 were 2.6, 3.5, 7 and 9.8 min., respectively under these

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. particular experimental conditions. Aflatoxin concentrations were calculated using

the Millenium Chromatography Manager Software (Waters, Milford, MA).

8. Fumonisin purification

Fumonisins were purified following the AOAC official method 995.15

(1996). Solid-phase extraction (SPE) was performed using 500 mg silica-based

strong anion exchange (SAX) columns (Varian BondElut - Varian, Harbor City,

CA). The SPE cartridges were fitted to a SPE manifold and conditioned by washing

with 5 ml methanol followed by 5 ml methanol .-water (3:1). A 10 ml aliquot of the

crude com extract (pH adjusted ton 5.8-6.5 if necessary) was applied to the

cartridge while maintaining a flow rate < 2.0 ml/min. The cartridge was then

washed with 5 ml methanol:water (3:1) followed by 3 ml methanol. Special care

was taken to avoid cartridge drying at this stage. Fumonisins were eluted with 10

ml methanol:acetic acid (99:1) at a flow rate <1.0 ml/min. The eluate was collected

in a 20 ml scintillation vial. The solvent was evaporated to dryness under N 2 at ca.

60°C. The vial was then washed twice with 1 ml methanol and the resulting

solution was transferred to a 4 ml amber vial. The methanol was evaporated to

dryness. Dried sample residues were stored at -20°C until liquid chromatography

(LC) analysis.

9. Quantification of fumonisins by liquid chromatography (HPLC)

Fumonisins were quantified using the same HPLC system described for

aflatoxin analysis. The fluorescence detector was set at 335 nm excitation and 440

nm emission wavelengths. The analysis was carried out using a Novapak Cis

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. reverse-phase (3.9x150 mm) column with a reverse-phase guard column. The flow

rate was set at 1 ml/min. using a acetonitrile:water:acetic acid (50:50:1) mobile

phase.

The dried purified residues were redissolved in 200 pi methanol. A 50 pi

aliquot was transferred into an autoinjector vial and 450 pi of orto-pthaldehide

(OPA) derivatizing reagent (40 mg ortho-pthalehide + 1 ml methanol + 5 ml 0.1 M

disodium tetraborate solution + 50 pi mercaptoethanol). The derivatized solution

was mixed and injected (10 pi) within 1 min. of adding the OPA solution. The

approximate retention times for fumonisin Bi an B 2 were 4.1 and 12.8 min.,

respectively.

10. Mutagenic potential of raw corn extracts

Composite samples were prepared by mixing 1 ml of each of the three raw

extracts from samples described in Table 3.1. The mixture was then filter-sterilized

using a 0.22 p-Millipore syringe filter (Millipore, New Bedford, MA) and

sequentially diluted 3 times using dimethyl sulfoxide (DMSO) to obtain a total of four

solutions test solutions ( 1 0 °, 1 0 ' 1, 1 0 '2and 1 0 '3).

The samples were tested using Salmonella typhimurium tester strains TA-100

and TA-98 according to the standard plate incorporation method described by Maron

and Ames (1983). Frozen cultures were plated onto ampicillin master plates and

incubated for 48 h at 37°C. A single colony was then inoculated into 50 ml sterile

Oxoid Broth No. 2; and incubated in the dark at 37°C for 10-14 h using a gyratory

(200-250 rpm) shaker water bath. After incubation, growth was confirmed by

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. checking the turbidity using a spectrophotometer (Spectronic 20D, Spectronic

Instruments, Rochester, NY) at 650 nm using Oxoid Broth No. 2 as a blank.

Absorbance readings ca. 0.8 indicated an optimal cell density of 1-2 * 10 9 cells/ml.

The tester strains were tested to confirm the presence of the genetic markers using the

following assays:

Histidine requirement: A positive result of this test (growth in

histidine/biotin plates; no growth in non-histidine/biotin plates) showed that

bacteria had not been altered as a result of handling or storage conditions and that

the mutant form was present.

Crystal violet: The results of this test showed a zone of inhibition around a

disk impregnated with crystal violet. The positive test results indicated that the

permeability of the membrane was present.

Ampicillin resistance (R-factor): Tester strains that grew on ampicillin

plates indicating that the genetic markers were intact.

After the bacteria tested positive for all the genetic markers, the assay was

performed. Two ml top agar were placed in sterile culture tubes and kept at

approximately 40° C in a water bath. One hundred pi of sample were placed in

each tube, 100 pi of bacteria (tester strain TA-100 or TA-98) and 500 pi S-9 mix

were added. The tubes were mixed and poured into minimal glucose agar plates.

All samples were tested in triplicate. DMSO was used as a negative solvent control

and pure AFBi (1, 10, 100 and 1000 ng/plate) was used as a positive control. The

plates were left at room temperature until the top agar solidified; then were inverted

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. and incubated at 37° C for 48 hours. After incubation, revertant colonies were

counted. Samples that presented over double the number of natural revertant

colonies were considered mutagenic.

11. Toxicity toSalmonella typhimurium

The most concentrated samples (10°) described in the previous section

were also tested to determine if the extracts were toxic to the S. typhimurium tester

strains. The assay was performed following a modified mutagenicity assay

protocol. After the 10-14 h incubation, the 10 9 cells/ml bacterial suspension was

serially diluted with sterile nutrient broth to obtain a 1 0 2 cells/ml suspension..

Samples were tested using nutrient agar plates instead of minimal glucose agar

plates.

12. Statistical analysis

Statistical analysis was performed using the SAS software (SAS, 1988).

One-way analysis of variance (ANOVA) was used to determine differences in toxin

production and toxicity to S. typhimurium. Scheffe’s test with an alpha value of

0.05 was used to compare differences among the different treatments.

C. Results and discussion

1. Aflatoxin production

Within the first days abundant mold growth was evident in all flasks except

for the control. A. flavus produced a very high amount of aflatoxin Bi in the first

15 days (Table 3.2). When A. flavus and F. moniliforme were together, the amount

of AFBj produced during the first days was significantly (p<0.05) lower than when

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. A. flavus grew by itself. It is interesting to observe that the naturally contaminated

sample had an apparently higher toxin production than the combined A. flavus!F.

moniliforme sample. However, these values were not significantly different

(p>0.05). There are several explanations why there was higher toxin production in

naturally contaminated com. One explanation could be that the naturally

contaminated com may have contained a higher initial amount of toxigenic A.

flavus and/or A. parasiticus spores than the inoculated sample. On the other hand,

the presence of other mold species may have had an influence on the production of

aflatoxin in the naturally contaminated com. However, since the difference was not

significant, it could be assumed that AFBi production was independent of the mode

of infection {i.e., naturally occurring or inoculated).

It is interesting to observe that the concentration of AFBi decreased over

time. When the results were analyzed, a pattern was observed. Every 15 days, the

toxin concentrations were reduced by approximately 90-95% in all samples. This

was more evident in the A. flavus samples were AFBi was detected in all three time

periods.

The constant decrease in AFBi concentration over time suggests that the A.

flavus by itself degraded AFBi. The same pattern was observed in the

production/degradation of AFB 2 (Table 3.3). The significant difference (p<0.05)

between the initial aflatoxin production by A. flavus and the other samples excludes

the possibility that the aflatoxins (Bi and B2) reached a toxic level beyond which,

they were degraded. Therefore, the degradation pattern may be attributed to the

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. mold metabolizing the toxin when it reached a certain growth stage or to the

presence of another metabolite interfering with AFBi and promoting its

degradation.

Table 3.2 Aflatoxin B| (ng/g) production during storage

Sample 15 days 30 days 45 days A. flavus 10.5xl04 ±0.6xl04 4.8x1 03±0.4x103 2 .2 x 1 0 2 a b b F. moniliforme ND ND 1.9±1.9 b b b A. flavus+F. 5.2xl0'±0.4xl0‘ ND ND moniliforme b b b

Naturally 5.6xl0z±0.6xl02 3.8x10^1.4x10' ND contaminated b b b Control ND ND ND b b b Values are mean±standard error of three replications ND: Non-detectable (Detection limit=lng/g) Values with the same letter are not significantly different (p<0.05)

These results are consistent with earlier observations reported by Doyle and

Marth (1978a; 1978b; 1978c; 1978d). The reports by these authors showed that the

ability of Aspergilli to degrade aflatoxin depended on several factors. They reported

that 8-10 day old mycelia were the most effective in degrading aflatoxin. This is

consistent with the fact that the aflatoxin concentration decreased after 15 days

incubation. Huyn and Lloyd (1984) also reported that Aspergillus parasiticus

synthesized large amounts of aflatoxin when young mycelia (4 days old) were

present. This synthesis peaked and then started to decline when mycelia aged (14

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. days old). It would be interesting to quantify aflatoxin production during the first

15 days under the same conditions, using the same A. flavus strain.

Table 3.3 Aflatoxin B 2 (ng/g) production during storage

Sample 15 days 30 days 45 days A. flavus l.lx l 0 4±0 .6 x l 0 J 4.4x10"±2xl0z 4.1xl0'±3.8xl0‘ a a,b b F. moniliforme l.SxlO'iO.OxlO1 2.2xl0'±0.5xl0‘ 0 .2 x l 0 '± 0 .2 x l 0 * b b b A. flavus+F. 5.8xl0'±lxl0‘ ND ND moniliforme a,b B b Naturally 1.3xl0z±0.14xl02 0.5xl0'±0.3xl0' ND contaminated a,b b b Control ND ND ND b B b Values are mean±standard error of three replications. ND: Non-detectable (Detection limit=lng/g) Values with the same letter are not significantly different (p<0.05)

Doyle and Marth (1978) and Hyun and Lloyd (1984) also reported that

disrupting the fungi, blending or fragmenting the mycelia, actively increased the

degradation of aflatoxin, while leaving the mycelia intact did not. In the current

study, the samples were mixed thoroughly with a metallic spatula every other day,

this might have damaged the mycelia. Therefore, aflatoxin degradation could have

been promoted in this experiment.

It has also been reported that substrates that support substantial mold

growth yield mycelia with the greatest ability to degrade aflatoxin. The current

study used com as the substrate. Since com is frequently colonized by A. flavus, it

would be considered ideal to promote copious mold growth. In this study,

100

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. substantial mold growth was evident within the first 48 hours of incubation;

especially in the A. flavus samples.

From a metabolic point of view, there are literature reports that state that

fungi have developed survival mechanisms that allow them to rapidly adapt their

metabolism to varying environmental conditions. This metabolic flexibility is

achieved through the synthesis of a large set of intra-and extracellular enzymes that

are able to degrade complex biopolymers. Filamentous fungi are also capable of

secreting a diverse range of primary and secondary metabolites and performing

several complex transformations. There is biochemical evidence that some of these

transformations are performed by fungal cytochrome P450 enzymes (Van der Brink

et al., 1998). There is limited information on the specific activities of fungal P450s

because purification of these enzymes is hampered by protein instability and low

expression levels (Van der Brink et al., 1998). However, it has been determined

that some A. flavus and A. parasiticus cytochromes (cyp64 and cyp60Al,

respectively) are actively involved in aflatoxin biosynthesis and degradation

pathways. Hamid and Smith (1987) reported that an increase in aflatoxin

production was paralleled by an increase in NADPH-cytochrome P450 reductase

and cytochrome P450 levels in fungal microsomes. The use of cytochrome P450

inhibitors, such as, SKF 525-A, metyrapone and cyanide inhibited aflatoxin

biosynthesis. These results suggested that aflatoxin biosynthesis by A. flavus is

mediated by cytochrome P450 monooxygenase enzymes. These authors also

studied the ability of mycelial homogenates to degrade aflatoxin. In this case the

101

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. addition of P450 inhibitors, cycloheximide, SKF 525-A and metyrapone to 6 - and

8 -day old cultures inhibited aflatoxin degradation. Thus, it was suggested that the

enzymes involved in the degradation of aflatoxin belong to the cytochrome P450

enzymes.

The significant difference (p<0.05) between aflatoxin production by A.

flavus as compared to the combined samples and the naturally contaminated

samples, suggested that the presence of F. moniliforme or other fungal species

affected the aflatoxin synthesis or its degradation. In this study, a significant lower

amount of AFBi and AFBi was noted in the combined samples in the first 15 days.

However, there is not enough information to determine what happened between

days 1 and 15. Therefore, the lower amount could be due to decreased aflatoxin

biosynthesis or enhanced degradation. It is important to mention that in a

mammalian system (male Wistar rats), intraperitoneal exposure to fumonisin B|

induced cytochromes P4501 A, P4503A and P4504A subfamilies with 1 Al and 4A1

predominating (Martinez-Larranaga et al. , 1996). Although fungal and mammalian

metabolism differ considerably, this evidence suggests that it would be interesting

to investigate the role of fumonisin in the ability of Aspergillus to synthesize and

degrade aflatoxins. This information brings out the importance of understanding the

role of mycotoxins in fungal physiology and plant pathogenesis.

Furthermore, Doyle and Marth (1978d) reported that (a) steaming A.

parasiticus mycelium for 6 minutes resulted in loss of ability to degrade aflatoxin

(b) broth mold cultures were not able to degrade aflatoxin; and (c) the supernatant

102

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. fluid from mycelial homogenates degraded aflatoxin. These results suggested that a

heat-labile, intracellular factor was responsible for toxin degradation. Doyle and

Marth (1978d) reported that although conditions affecting degradation of aflatoxins

by mold mycelia implied that an enzyme might be involved. However, the rates at

which aflatoxins were degraded did not support the idea that the toxin was being

degraded directly by enzymatic activity. Therefore, it was hypothesized that one or

more enzymes produced by-products that would react with aflatoxin. Doyle (1979)

speculated that peroxidase may be such an enzyme since it catalyzes decomposition

of hydroperoxides resulting in the generation of free radicals. Hence, Doyle and

Marth (1982) designed a study to determine peroxidase activity in mycelia of A.

parasiticus that degraded aflatoxin. Results from this study showed that substantial

amounts of fungal peroxidase were present in cultures where aflatoxin was

degraded. These results were supported by Applebaum et al. (1980) and Doyle and

Marth (1978) who showed that lactoperoxidase is involved in aflatoxin

degradation. Doyle and Marth (1978h) observed that at a given concentration of

this enzyme, aflatoxin degradation was independent of initial aflatoxin

concentration. The authors reported the formation of aflatoxin derivatives that co

chromatographed with aflatoxin B 2a, and other water soluble by-products.

This information also relates to the findings of fumonisin Bi toxicity in

vivo. Martinez-Larraiiaga (1996) reported that 0.25 and 2.5 mg/kg body weight FBi

administered intraperitoneally to male Wistar rats, produced an increase in the

peroxisomal P-oxidation of fatty acids and induction of the cytochrome P4504A

103

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. subfamily. Although, these studies show metabolic alteration in a mammalian

system, the involvement of peroxidase in aflatoxin degradation by Aspergillus

suggests the possibility of similar effects in a fungal system.

Tables 3.4 and 3.5 show the production of aflatoxins Gi and G 2 ,

respectively. In this case, no significant toxin production was observed. This

indicates that the A. flavus strain used in this study does not produce G aflatoxins.

This is consistent with the fact that onlyA. parasiticus produces G aflatoxins. The

low levels of G aflatoxins detected in several of the samples, including the F.

moniliforme samples, could be due to original com contamination. This emphasizes

the fact that aflatoxin contamination is heterogenous and that “hot spots” may be

present. Even when the cracked com was thoroughly homogenized before

sampling, some contaminated kernels could have gone to different flasks. This

could explain why the Fusarium samples had low levels of aflatoxins Bi and G 2 .

Table 3.4 Aflatoxin Gi (ng/g) production during storage

Sample 15 days 30 days 45 days

A. flavus ND ND 6 ± 6 F. moniliforme ND ND ND A. flavus+F. moniliforme ND ND ND

Naturally contaminated 28±14 ND 2 ± 2 Control ND ND ND Values are mean±standard error of three replications ND: Non-detectable (Detection limit=lng/g) Non of the samples are significantly different (p<0.05)

104

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 3.5 Aflatoxin G 2 (ng/g) production during storage

Sample 15 days 30 days 45 days A. flavus NDNDND

F. moniliforme 4±1 2 ± 1 ND A. Jlavus+F. moniliforme 12±5 ND ND

Naturally contaminated 1 0 + 1 ND ND Control NDND ND Values are mean±standard error of three replications. ND: Non-detectable (Detection limit=lng/g) Non of the samples are significantly different (p<0.05)

It is also important to consider the com substrate. It is possible that there is

more than a bilateral interaction between molds. Com is a complex chemical

matrix, therefore, multilateral chemical interactions might have occurred during

this storage period. It could be possible that fungal metabolites interacted with com

compounds and that the new by-product(s) inhibited or degraded the aflatoxin.

Also, there could have been bacterial activity that affected toxin production or

degradation, or bacterial metabolism could have also influenced the toxin

degradation patterns.

It was also observed that the 30 and 45 days samples had copious mold

growth that in some cases interfered with the aflatoxin extraction and purification

procedures. This led to the use of a different purification method (described in the

materials and methods section). This may have contributed to incomplete toxin

extraction, and may also be a reason for the lower quantities detected in these

samples. The difficulty in extraction resided in the fact that there was some

105

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. interference from compound(s) that acted as surfactants; therefore, when organic

extraction (CHCI3) was attempted, a very stable emulsion was formed. It would be

interesting to characterize these compounds and determine if they play a role in

aflatoxin degradation.

It is important to point out that even if the combined and the naturally

contaminated samples had much lower toxin production, they still had high levels

of aflatoxins within the first 15 days. This shows that under favorable conditions,

aflatoxins can be produced regardless of the presence of other molds, i.e., F.

moniliforme. This emphasizes the fact that co-contamination can occur and that

simultaneous exposure to several toxins is possible.

2. Fumonisin production

FBi production followed a different pattern than aflatoxin. Fumonisins are

produced in higher quantities usually 1 0 3 times more than aflatoxins, so they are

usually quantified in pg/g instead of ng/g. Table 3.6 shows FBi production over

time. Unlike aflatoxin, these samples do not show a defined pattern. In the F.

moniliforme sample, a significant increase (p<0.05) of FB[ was observed over time.

The combined samples and the naturally contaminated com did not produce

significantly different amounts (P<0.05) of FB| during the first 15 days. These

results contrast with those obtained from aflatoxin. FBj production was not initially

affected by the presence of Aspergillus or other mold species.

106

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 3.6 Fumonisin Bi (jj.g/g) production during storage

Sample 15 days 30 days 45 days A. flavus 15±12 8±4 ND b b B F. moniliforme 17±3 254±90 361±131 b a a A. flavus+F. moniliforme 8±3 8 ± 2 7±1 b b b Naturally contaminated 18±6 38±4 6±3 b b b

Control ND 1 ± 1 ND b b b Values are meantstandard error of three replications ND: Non-detectable (Detection limit=Tpg/g) Values with the same letter are not significantly different (p<0.05)

It was also interesting to observe that there were different toxin production

patterns. In the F. moniliforme samples, a constant increase of FB i was observed.

However, in the combined samples (A. flavus+F. moniliforme) FBi concentration

was not significantly (p<0.05) altered. In the naturally contaminated com, an

apparent increase followed by a decrease were observed; however this was not

significant (p>0.05).

The FBi concentrations detected in the Aspergillus samples and in the

control, may have been from original contamination. This emphasizes again the

heterogeneity of mycotoxin contamination, and the need to find control

mechanisms that will prevent exposure to these toxins. The consumption of one

contaminated portion ( > 8 pg/g) is sufficient to elicit ELEM in a horse. The fact that

these samples originated from a commercial source stresses the need of a better

understanding of factors that affect toxin production and degradation. This

107

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. knowledge would help in designing better control programs that would minimize

animal and human exposure.

Table 3.7 Fumonisin B 2 (pg/g) production during storage

Sample 15 days 30 days 45 days

A. flavus 1 ± 1 14±6 4±2 F. moniliforme 10±7 109±37 104±41

A. flavus+F. moniliforme 8 ± 1 8 ± 6 4±1 Naturally contaminated 9±3 13±3 3±1

Control 1 ± 1 9±9 ND Values are mean±standard error of three replications ND: Non-detectable (Detection limit=lpg/g) Non of the samples are significantly different (p<0.05)

3. Mutagenic potential

Figures 3.1 and 3.2 show the mutagenic potential of the raw com extracts

from the A. flavus inoculated com. A dose-response relationship was observed. As

the samples were diluted, the mutagenicity decreased. Hence, it can be considered

that the compound(s) responsible for the mutagenic response are either not present

in large quantities or their mutagenic potential is relatively low. Also, it can be

observed that mutagenicity decreased over time. This observation is consistent with

the fact that AFBi decreased over time. So, the decline in mutagenicity could be

due to the decline in aflatoxin concentration. However, the 30 and 45 day samples

(initial AFBi=484 and 218 ng/plate) showed much higher mutagenicity than the

positive control, pure AFBi. It was observed that 100 ng/plate AFBi resulted in

108

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. and and 3 45 45 days tester strain ,4.8xl0 4 —♦ _ days 15 _ « - days 30 Salmonella 1.00E+00 inoculated corn during storage. Dilution Aspergillusflavus .00E-02 .00E-02 1.00E-01 0 1.00E-03 500 766 766 and 0 ng/plate for 15, 30 and 45 days, respectively. Natural 34±9.Values revertants=l are mean±SEM of Figure 3.1 Mutagenic potential of three three replicates. TA-100 with metabolic activation. Approximate undiluted (1.00E+00) amounts ofAFB|=10.5xl0 218 ng/plate for 15, 30 and 45 days, respectively. Approximate undiluted (1.00E+00) amounts of FBi=l500, 1000 2000 2500 ra ra 1500 vO o

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. and 3 45 45 days tester strain _ « _ days 30 — days 15 Salmonella 1.00E+001.00E-03 1.00E-01 inoculated corn during storage. Dilution Aspergillus flavus 1.00E-02

. . Figure Figure 3.2 Mutagenic potential of 218 218 ng/plate for 30 45 and 15, days, respectively. Approximate FB|=1500, 766 and 0 ng/plate for 30 and 45 15, days, respectively. Natural revertants=29±2. Values are thre mean±SEM ofreplicates. 1000 TA-98 with metabolic activation. Approximate undiluted (1.00E+00) amounts ofAFBi=10.5xl04, 4.8xt0 2000 2500 §? 3 1500 t*

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 173±2 revertants per plate; therefore, the results shown in Figure 3.1 could be the

result of AFBi mutagenic potential combined with the response elicited by other

compound(s). This effect was not as evident in TA-98 (Figure 3.2) where 100

ng/plate pure AFBi produced 532±91 revertants/plate. These differences could be

explained by the different sensitivities of each organism.

In the case of the F. moniliforme com extracts (Figures 3.3 and 3.4), an

interesting fact was observed. These samples did not contain any detectable AFBi;

however, a mutagenic response was observed for all three time periods. Even if

there were high concentrations of FBi in the extracts, this toxin is not considered to

be mutagenic in this assay (Gelderblom et al„ 1991, 1992; Park et al., 1992).

Furthermore, a FBi control was used in this experiment. No mutagenic response

was observed in a concentration ranging from 1 to 100,000 ng FB i/plate.

The mutagenic response of F. moniliforme raw com extract was more

evident on the 15 day samples indicating that the unknown factors that elicited this

response were degraded over time. A dose response was also observed. Dilution of

the samples decreased their mutagenic potential. These results show that a diverse

variety of chemically active compounds can be produced at the same time. F.

moniliforme is known to produce mutagenic toxins such as fiisarin C and

moniliformin (Bacon and Nelson, 1994). These toxins may have been responsible

for this response. However, it is possible that other unidentified mutagenic

compounds may have been produced. This emphasizes the fact that when a raw

extract is studied, multiple interactions may take place. The ultimate response may

111

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. tester _ days 30 .. 45 days - a -#— 15 days -#— 15 Salmonella 1.00E+00 inoculated corn during storage. 1.00E-01 Dilution Fusarium moniliforme 1.00E-02 0 1.00E-03 500 Figure Figure 3.3 Mutagenic potential of three replicates. strain strain TA-100 with metabolic activation. No AFBi was detected on any of the samples. Approximate 254 and 361 pg/plate FB|=17, for 45 30 and days, respectively. 15, Natural revertants=139±9. Values are mean±SEM of 1000 2500 2000 5 1500 to

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. tester Salmonella — — 30 days — .. ..45 days — days #— 15 1.00E+00 inoculated corn during storage. Dilution Fusarium moniliforme 1.00E-02 1.00E-01 1.00E-03 100 200 250 Figure Figure 3.4 Mutagenic potential of strain TA-98 with metabolic activation. No AFBi was detected on any 254 and of361 ng/plate the for samples. 30 and 15, 45 Approximate days, respectively. FBi=17, Natural revertants=29±2. Values are mean±SEM of three replicates. « 150 U>

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. depend on an equilibrium of the effect of all compounds present in the system.

Some of these compounds may have a direct effect while others may interact and

either inhibit each other or have a synergistic response.

Figures 3.5 and 3.6 show the mutagenic response to raw A. flavus/F.

moniliforme com extracts. These samples also showed a mutagenic response to

both, TA-100 and TA-98. A dose-response type of relationship was also observed,

and the mutagenic potential tended to decrease over time.

These results also show several interesting facts. AFBi concentrations were

relatively low at 15 days, and there was no detectable aflatoxin in the 30 and 45

days samples. So, these mutagenic responses bring out the same issues as the F.

moniliforme samples. Are other mutagens being produced, or is fumonisin

interacting with other compound(s) to produce a mutagenic by-product? Another

important factor to point out is that all of these assays were carried out with

metabolic activation (S9). Therefore, these responses reflect the effect of

compound(s) that may have been metabolized. Some of these activated compounds

may have caused this response, or they may have reacted with other compounds

present in the system to render a mutagenic by-product.

The raw com extract is considered a complex mixture, so a multitude of

reactions may have taken place during this assay. These results indicate that

Aspergillus and Fusarium may produce more than one toxic compound when they

grow together. Understanding the relationship of these molds in the com plant and

114

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 8 , , 8 .. 45 days * • - _ -« __ -« days 30 —«— days 15 1.00E+00 inoculated corn during storage. 1.00E-01 Dilution Aspergillus jlavus/Fusarium moniliforme tester strain TA-100 with metabolic activation. Approximate undiluted (1.00E+00) amounts of ; . 1.00E-03 1.00E-02 AFBj=5.2, 0 and 0 ng/plate for 30 15, and 45 days respectively. Approximate undiluted amounts of FB|= Figure 3.5 Mutagenic potential of replicates. and 7 pg/plate for 30 45 and days, 15, respectively. Natural revertants=139±9. Values are mean±SEM ofthree Salmonella 200 8 0 0 1000 1200 03 t 600 . * 400 . Ul

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 8 , , 8 . . .45 days - a . - . - — •— days 15 _ « _ days 30 inoculated corn during storage. 1.00E-01 Dilution Aspergillus flavus/Fusarium moniliforme 1.00E-02 tester strain TA-98 with metabolic activation. Approximate undiluted (1.00E+00) amounts of o o : 1.00E-03 50 replicates. AFB|= 5.2, 0 and 0 ng/plate for 30 and 45 15, days respectively. Approximate undiluted amounts ofFB| = and 7 jig/plate for 30 and 45 15, days, respectively. Natural revertants = 29±2. Values are mean±SEM of three Figure Figure 3.6 Mutagenic potential of 150 100 Salmonella 250 200

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. during storage, may help in elucidating some of the potential toxic responses that

would result from dietary exposure to com contaminated with these fungi.

Figures 3.7 and 3.8 show the mutagenic potential of naturally contaminated

com extracts. These results are similar to those discussed for Figures 3.5 and 3.6. In

this case, there is a higher initial AFBi concentration that could be responsible for

the mutagenicity. However, the AFBi content decreased over time. Another

possible explanation for this mutagenic response may be that some of the AFBi

fungal degradation by-products retain some mutagenic potential. It would be

interesting to follow AFBi degradation over shorter periods of time (<15 days) and

isolate and identify some of the by-products, subsequently testing their mutagenic

potential.

These results show the need for more research in this area. The possible

effect of FBi on AFBi degradation was discussed earlier. If fumonisin is

responsible for enhancing aflatoxin degradation, does it affect the amount and type

of residual by-products? Is it possible that the presence of fiimonisins shift

aflatoxin degradation pathways to yield different by-product(s) than those produced

by Aspergillus itself? Is Fusarium metabolizing some of the aflatoxins? The answer

to some of these questions may help shape our understanding of the relationship

and interactive potential between these com contaminants, and perhaps help define

better mechanisms of biological control to prevent toxin formation.

Since com is a complex matrix and is considered a living system,

incubating it in the presence of moisture may have also contributed to the responses

117

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 45 45 days pg/plate 6 —«— days 15 _ * _ days 30 tester strain TA-100 M 1.00E+00 / / Salmonella / Dilution .00E-02 .00E-02 1.00E-01 0 1.00E-03 50 50 , 15, 30 15, and 45 days respectively. Approximate undiluted (1.00E+00) amounts of FB| = 18, 38 and for 15, for 30 45 and days, respectively. 15, Natural revertants = are Values mean±SEM ofthree 139±9. replicates. Figure 3.7 Mutagenic potential of naturally-contaminated corn during storage. with with metabolic activation. Approximate undiluted (1.00E+00) amounts of AFBi = 563, 38 and 0 ng/plate for 100 150 350 500 200 450 400 oo

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 30 days — ... 45 days 4 - 9 — days «— 15 — .. tester strain TA-98 / 1.00E+00 Salmonella pg/plate for 15, 30 and 45 days, 6 1.00E-01 Dilution 1.00E-02 .

0 80 1.00E-03 20 60 40 Figure Figure 3.8 Mutagenic potential of naturally-contaminated corn during storage. with metabolic activation. Approximate undiluted (1.00E+00) AFBi = 563, 38 and 0 ng/plate for 30 15, and 45 respectively. respectively. Natural revertants = 30±2. Values are mean±SEM three ofreplicates. 140 120 100 days respectively. Approximate undiluted amounts of FBi = 18, 38 and P. u >

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. shown previously. Therefore, it was important to determine if the control com

extracts elicited a response on this assay.

Figures 3.9 and 3.10 show that the control com extracts did not present

mutagenic potential. This suggests that the responses observed in the other samples

were due to fungal activity in the com. It is important to note that some

compound(s) in the com may have been altered by the mold growth. Therefore,

consideration of factors present in the com is also important when considering the

risk associated with fungal growth in com even if the com by itself did not show a

response.

4. Toxic potential

The raw com extracts were complex mixtures. Therefore, the evaluation of

their mutagenic potential may have been hampered by other responses. If the

samples were toxic to the bacteria, then some of the results could have been false

negatives. A reduction in response could have been due to a toxic response rather

than to a lack of mutagenicity. Furthermore, toxic responses could give false

evidence of interaction between compounds in the extract. Thus, it was important

to determine if these samples exerted toxic effects on either of the Salmonella tester

strains. Since most of the mutagenic differences were observed in the most

concentrated samples (10°), only this dilution was tested for toxic potential. Instead

of challenging the bacteria to mutate by using minimal glucose agar plates, the

assay was carried out using nutrient broth.

120

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. • • 45 45 days • • - a .. — days *— 15 - « _ days 30 tester strain TA-100 with metabolic Salmonella Dilution 1.00E-02 1.00E-01 1.00E+00 .

0 1.00E-03 50 Figure Figure 3.9 Mutagenic potential of control corn during storage. activation. activation. No AFBi was detected ng/plate in the for samples. 15, 30 and Approximate 45 undiluted days, replicates. amounts respectively. of FB| Natural revertants = = 0, 134±9. Values 73 are and mean±SEM 0 of three 150 100 200 300 250 ed 3 04 'H, to

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. - • 45 45 days • - - a — — 30 days — . - — days «— 15 1.00E+00 tester strain TA-98 with metabolic Salmonella Dilution 1.00E-02 1.00E-02 1.00E-01 ; 0 1.00E-03 Figure Figure 3.10 Mutagenic potential of control corn during storage. activation. No AFBi was detected in the samples. Approximate undiluted (1.00E+00) amounts of FB| = 0, 73 replicates. and and 0 ng/plate for 30 and 45 15, days, respectively. Natural revertants = 30+2. Values are mean±SEM ofthree 100 150 300 200 250 ) (U > 1

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 3.11 shows Salmonella TA-100 toxic response to the different com

extracts. In the Fusarium and the combined samples {Aspergillus+Fusarium), the

toxic potential seemed to increase with storage time. These results could explain

part of the reduction in mutagenicity observed previously. However, none of these

samples were significantly different (p<0.05) form the control. Conversely, the

toxicity of the Aspergillus samples seemed to decrease over time. In this case, the

15 day storage extract was significantly different from the control. This implies that

under these experimental conditions, this sample was toxic to Salmonella TA-100.

Although the toxicity seemed to decrease in the 30 and 45 day samples, these

results were not significantly (p<0.05) different from the 15 day extract. It can be

noted that these same samples had an apparent reduction in mutagenicity over

storage time (Figure 3.1). Even if the decrease in toxicity is not significant, the

results suggest that detoxifying reactions may have taken place as the Aspergillus

culture grew older. The decrease in AFBi content discussed earlier, coupled with

the decline in mutagenic potential and the apparent decline in toxic potential

suggest that A. flavus shifted to detoxifying pathways that may have included toxic

compounds other than aflatoxin.

An interesting fact is that the control com extracts (15 and 30 days)

exhibited significant toxicity when compared to the control (p<0.05). These

samples did not have any fungal activity. These results imply that com by itself is a

dynamic system and that different reactions may have taken place in the com.

Some of the by-products may have been toxic to the Salmonella TA-100. This

123

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

=Fusarium, | Control H i days 5 □ days 30 □45 days F fi I corn Control Aspergillus, NC tester strain TA-100 with metabolic A+F Salmonella Sample NC=Naturally contaminated. Values mean±SEM are ofthree replicates. Fusarium, + Negative Positive spergillus 1.00E+08 7.00E+08 3.00E+08 6.00E+08 5.00E+08 2.00E+08 0.00E+00 Figure Figure 3.11 Toxic potential of corn samples during storage. A+F=/l activation. activation. Negative control=DMSO, Positive 10,000 control= ng AFB,, A= g> 4.00E+08 o 2 is, U to

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. toxicity decreased significantly (p<0.05) over time. The toxic potential shown by

the control com extracts brings attention to the fact that reactions not related to

fungal growth, i.e. lipid oxidation, could have also occurred in the com. These

reactions could have generated by-products that were toxic, but not mutagenic. The

involvement of other microorganisms such as bacteria is also possible. Bacteria

could have produced bacteriocins that inhibited Salmonella growth.

Salmonella TA-98 was less sensitive to the com extracts. Thus, it showed

higher colony counts than TA-100 (Figure 3.12). In this case, the 15 day Fusarium

extract was significantly toxic. This sample had a high FBi concentration (ca.

1.7xl0 3 pg/plate). However, even if the FBi concentration increased over storage

time (ca. 2.5x104 and 3.x 10 4 pg/plate for 30 and 45 days, respectively), these

samples were not significantly different from the control. Therefore, the toxicity

cannot be attributed to FBi but to other compounds present in the extract. Another

hypothesis is that maybe the higher concentrations of fumonisin trigger

detoxification reactions in older cultures. More research is needed to understand the

role of fumonisin from a fungal physiology and plant point of view.

Another significantly toxic sample was the naturally contaminated extract

(45 days). In this case, aflatoxin and fumonisin concentrations decreased over

time. So, this extract had the least amount of AFBi and FBi when compared to the

15 and 30 days samples. Again, other compounds may be responsible for the

toxicity. Moreover, aflatoxin and fumonisin may have an interactive relationship

125

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

=Fusarium, , , F A=Aspergillus □ 45 days g Control U days 15 □ days 30 I tester strain TA-98 with metabolic corn NC NC Control Salmonella F F A+F Sample NC=Naturally contaminated. Values are mean±SEM ofthree replicates. e Negativ Positive A i i =Aspergillus+Fusarium, 1.00E+08 1.00E+08 i 6.00E+08 3.00E+08 2.00E+08 5.00E+08 4.00E+08 0.00E+00 Figure Figure 3.12 Toxic potential of corn samples during storage. A+F A+F activation. activation. Negative control=DMSO, Positive 10,000 control= ng AFBi,

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. that changes depending on the concentration ratio or at different concentration

levels.

No definite pattern to relate toxin production, mutagenic potential and toxic

potential could be established. However, there is some evidence that indicates that

many factors can be simultaneously involved in these processes. This experiment

did show that A. flavus and F. moniliforme can co-exist in the same environment

and produce aflatoxins and fumonisins, respectively. Thus, they present a risk to

consumers and new research concerns. If these toxins co-contaminate com and

there is evidence of fungal interaction that shifts toxin concentrations over storage

time, what are the potential toxic effects of simultaneous exposure at different

concentration levels. Do they interact; if so, what relationship do they have? Is

there potential for inhibitory or synergistic reactions? These questions prompted

further research in this area. Therefore, the next sections will explore some of these

possible interactions.

127

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. IV. Interactive effects between aflatoxin Bi and fumonisin Bi

A. Introduction

Results from the previous section suggested that AFBi and FBi can be

produced simultaneously by A. flavus and F. moniliforme. Furthermore, the data

obtained suggested that mold growth and toxin production are dynamic processes

where multiple interactions are possible. These interactions yielded different levels

of toxins at different storage times. This implies that when com is consumed,

humans and animals could be exposed to a wide range of toxic compounds that

could be present in different combinations or ratios.

The previous experiment concentrated on studying toxin production under

optimal mold growth conditions. The high moisture levels yielded copious mold

growth, and may have promoted some of the interactive reactions. However, these

reactions could take place over longer periods of time with com held at storage

moisture and temperature conditions. This is evidenced by several reports of

AFB1/FB 1 co-contamination of commercial com that did not have apparent mold

growth (Park et al., 1996; Chamberlainet al., 1993; Zummo and Scott, 1992).

Aflatoxins and fumonisins are produced in nature in a different range of

concentrations. Aflatoxins usually occur in the ng/g range (Schroeder and Botler,

1973) while fumonisins generally occur at higher levels, in the |ig/g range. Thus,

simultaneous exposure to a combination of these toxins at these levels should be

explored. The assessment of the potential effects of complex mixtures is a difficult

task. However, the potential of joint action of chemicals has long been recognized

128

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. by physicians, public health specialists, toxicologists, health risk assessors and

statisticians (Mumtaz, 1995). There are several methods to assess the effects of

diverse mixtures encountered in the environment. One approach often used to

evaluate chemical mixtures is to consider the effect of known xenobiotics present

in the mixture. In this case, although all the information may not be available for

the mixture (i.e.. raw com extract) toxicological properties of some of the

compounds (i.e.. aflatoxins) have been investigated. In the current study, both,

aflatoxins and fumonisins are known to be present in the com. and the relative

concentrations in which they occur has also been reported. Therefore, the use of a

combination of these two compounds may yield more useful information on the

toxic potential of the mixture.

The objective of this phase of the study was to determine if different levels

of pure AFBi (0-1000 ng/plate) combined with varied levels of FBi (0-100,000

ng/plate) show different effects in the Salmonella/microsomal mutagenicity assay

using tester strains TA-98 and TA-100. Since fumonisin does not yield positive

results in this assay, this study mainly evaluated the effect of fumonisin on

mutagenic potential of aflatoxin.

B. Materials and methods

1. Bacterial cultures

Salmonella typhimurium tester strains TA-100 and TA-98 were kindly

provided by Dr. Bruce N. Ames (University of California Berkley, Berkley, CA).

129

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 2. Chemicals

Ampicillin, D-biotin, disodium hydrogen phosphate, glucose-6-phosphate,

L-histidine, magnesium sulfate, sodium ammonium sulfate, sodium chloride,

sodium dihydrogen phosphate, and B-nicotinamide adenine-dinucleotide phosphate

(NADP) were purchased from Sigma Chemical Co. (St. Louis, MO). Citric acid

monohydrate, dimethyl sulfoxide, potassium chloride, potassium phosphate dibasic

anhydrous and water (HPLC grade) were acquired from Mallinckrodt Baker, Inc.

(Paris, KT). Bacto agar and crystal violet were ordered from Difco Laboratories

(Detroit, MI); glucose from Fisher Scientific (Fair Lawn, NJ). Ethanol was

obtained from Aaper Alcohol and Chemical Co. (Shelbyville, KT). Oxoid nutrient

broth No. 2 was from Unipath LTD (Basingstoke, Hampshire, England). Post-

mitochondrial supernatant (S9 mix) was purchased from Molecular Toxicology,

Inc. (Boone, NC).

Pure fumonisins Bi and B 2 were kindly provided by Dr. Robert M. Eppley

(Center for Food Safety and Applied Nutrition - CFSAN, U. S. Food and Drug

Administration-FDA, Washington, D.C.). Pure aflatoxin Bj was generously

provided by Dr. Mary Trucksess (CFSAN-FDA, Washington, D. C.).

3. Sample preparation

Toxin dilutions were always prepared fresh immediately before running an

assay. Stock solutions of AFBi (lpg/pl) and FBi (10(ig/pl) were prepared by

diluting dry pure toxins in DMSO. These solutions were then combined and diluted

130

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. to obtain solutions that would follow the experimental design presented in Table

4.1.

Table 4.1 Combinations of AFBi and FB| tested for toxic/mutagenic potential

Aflatoxin E ti (ng/plate) Fumonisin B| (1) (2) (3) (4) (ng/plate) 10° 101 102 103 (A) (1A) (2A) (3A) (4A) 10° (B) (IB) (2B) (3B) (4B) 10‘ (C) (1C) (2C) (3C) (4C) 102 (D) (ID) (2D) (3D) (4D) 103 (E) (IE) (2E) (3E) (4E) 104 (F) (IF) (2F) (3F) (4F) 105 Values in parenthesis indicate the labels used to identify each sample. Combinations are expressed as ng AFBj/plate / ng FBj/plate. Samples were prepared and analyzed in triplicate

When FB2 was also evaluated and the concentration of AFBi was fixed at

1000 ng/plate and FB 2 was varied from 0 to 100, 000 ng/plate. Samples were also

prepared fresh immediately before the assay. AFB 1/FB2 combinations were

evaluated using the experimental design presented in Table 4.2.

4. Mutagenic potential of aflatoxin/fumonisin combinations

The samples were tested using Salmonella typhimurium tester strains TA-

100 and TA-98 according to the standard plate incorporation method described by

131

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Maron and Ames (1983). A detailed description of this method was presented in

Chapter III. An S9 cocktail without the postmitochondrial supernatant was used for

the samples without metabolic activation. Although the S-9 cocktail did not contain

the rat liver metabolizing enzymes, it included all the other co-factors including

NADP and glucose-6-P. These compounds could have been incorporated to

bacterial metabolism and affected the response. This control only tests whether the

metabolic activation by mammalian liver enzymes is required to generate a

mutagenic metabolite. For the purpose of this experiment, all other potential

interactions with the components of the S-9 cocktail were disregarded.

Table 4.2 Combinations of AFBi and FB 2 tested for toxic/mutagenic potential

Aflatoxin Bi (ng/plate) Fumonisin B 2 (4) (ng/plate) 103 (G) (4G) 10° (H) (4H) 101 (I) (41) 102 (J) (4J) 103 (K) (4K) 104 (L) (4L) 105 Values in parenthesis indicate the labels used to identify each sample. Combinations are expressed as ng AFBi/plate / ng FB 2/plate. All samples were prepared and anlalyzed in triplicate.

132

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 5. Toxic potential of the aflatoxin/fumonisin combinations

Salmonella TA-98 and TA-lOO’s toxic response to the aflatoxin/fumonisin

combinations was evaluated using the same modified procedure described in

Chapter III. Statistical analysis of toxicity results was carried out using SAS

software (SAS, 1988). Scheffe’s test was used to determine differences between

controls and samples.

C. Results and Discussion

1. Aflatoxin Bt/fumonisin B( mutagenicity

The representative results of a series of experiments are presented in this

section. Raw data and results from repetitions of this experiment are presented in

Appendix A. Figure 4.1 shows the mutagenic potential of AFBi/FBi combinations

using Salmonella tester strain TA-100 without metabolic activation. These samples

were not tested with Salmonella tester strain TA-98. The results indicated that none

of the samples were mutagenic. This was expected since AFBi needs metabolic

activation to become a mutagen, and FBt is not considered a mutagen in this assay

with or without metabolic activation. However, this results show that when both

toxins are in the same system without metabolic activation, they do not interact to

produce mutagenic by-products at any of the combinations tested. This indicates

that metabolic activation is essential for the potential of aflatoxin to react with other

biological molecules. Therefore, the potential to interact with fumonisin will

depend on activation of AFBi to more reactive compounds. Since these results

were negative, all further testing was performed with metabolic activation.

133

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. IngAFBl IngAFBl AFBI ng 100 lOngAFBl lOngAFBl -OngAFBl •••*•• 1000 ng AFBI ng 1000 •••*•• tester strain TA-100 without metabolic Salmonella V. FBI (ng'plate) FBI o...... o. Figure 4.1 Figure 4.1 Mutagenic potential ofAFBi/FBj combinations. activation. Natural revertants=154±5. are Values mean±SEM ofthree replicates. 50 O.OQBOO O.OQBOO l.OQBtOO 1.0QBO1 1.00BQ2 1.0QEKB 1.0QBW 1.00BO5 150 300 300 , 250 200 „ „ -▻> u>

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. When the samples were tested with metabolic activation, a pattern emerged

(Figures 4.2 and 4.3). No mutagenic response was observed at AFBi levels (10° to

101 ng/plate) alone or in combination with the whole range of FBi concentrations.

In general such low amounts of AFBi do not show a response in this assay.

Therefore, FBi did not show potential for increasing the mutagenicity of low levels

of AFBi. However, at doses where mutagenicity is usually observed (102 and 103

ng/plate), an interesting pattern was noted. In both tester strains, it was observed

that the mutagenic potential varied as fumonisin concentrations increased. These

results suggested the possibility of an interaction between both toxins. However,

the pattern implied that the potential interaction is complex in nature. It was

observed that the mutagenic potential decreased in a dose-dependent fashion when

10°-102 ng FBi were present. However in the 102-105 ng FBj/plate range, the

mutagenicity increased back to its original level; and sometimes the increase

surpassed the original level (data not shown). This behavior was consistent in both

tester strains.

These results suggested that fumonisin may affect one or more of the steps

involved in the mutagenicity process. This assay does not give information on the

mechanism involved except for the fact that metabolic activation is needed to

observe an effect. This leads to several hypothesis that can be divided in two

general groups. The first one is that one or more reactions directly involving the

toxins are taking place; and the second one is that FBi is affecting AFBi’s

metabolic transformation.

135

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. IngAFBl lOOngAFBl —*— 1000ngAFBI • • * • • * • • KhgAFBl • • -*-• — OngAFBl l.OOEKB s s tester strain TA-100 with metabolic 1.00EKW Salmonella / FBl(ngftlate) 1.00B01 1.00EKE l.OOEKB l.OOEKX) ■ ■ — — - i — ■ ------...... A 0 Figure 4.2 Mutagenic potential of AFB|/FB| combinations. activation. activation. Natural revertants=153±l are Values replicates. mean±SEM ofthree QOQEtOO 5 0 0 * 1000 1500 2000 3000 2500

„ „ ts £ Os

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. - - lOng AFBI lOng - - + —•— 0 ng AFBI « ng AFBI 1 — — • - —*— ng AFBI 1000 - * - • ng AFBI 100 tester strain TA-98 with metabolic i -f i Salmonella FBI (ng/plate) ------I-.. 1.00E+00 1.00E+00 1.00E+01 1.00E+02 1.00E+03 1.00E+04 1.00E+05 0 0.00E+00 400 200 Figure Figure 4.3 Mutagenic potential of AFBi/FB| combinations. activation. activation. Natural revertants=32+3 are Values ofthree mean+SEM replicates. 1400 1200 1000 u

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. a. Direct interaction hypothesis

AFBi biotransformation (Figure 2.3) yields several by-products. Some of

them are considered detoxification compounds since their mutagenic potential is

lower than that of the parent compound. However, virtually all the toxic effects of

AFBi are now recognized to be attributable to the action of activated AFBi

metabolites (i.e., AFB1-8 ,9-epoxide) that are capable of reacting with several

macromolecules. The cytochrome P450 epoxidation of AFBi's terminal fiiran ring

double bond yields a very potent electrophile (Essigmann et al., 1977; Swenson et

al., 1977; Baertschi et al., 1989). This compound is not stable and is rapidly bound

to a nucleophilic species or transformed to the AFB[-8.9- dihydrodiol. The diol can

exist in a resonance form as a phenolate ion that is capable of forming Schiff base

adducts with amino groups (i.e., lysine in protein). On the other hand, FBi contains

one such primary amine group (Figure 2.4). Dantzer et al. (1997) and Murphy et

al. (1996) reported that FB| can undergo Maillard type reactions to form Schiff

base adducts, and these adducts can undergo an irreversible dehydration reaction .

These authors report that formation of Maillard-type compounds by this reaction

blocks the toxic potential of fumonisin. This implies that, theoretically, if the

AFBi-8,9-dihydrodiol and FBi are present in the same system they could undergo a

Maillard-type reaction. This reaction would block the functional groups considered

most responsible for the toxic potential of these xenobiotics.

In the current study, both toxins were present at the same time in the same

solvent system. So, in this case, this reaction could be possible. However, in an in

138

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. vivo system pharmacokinetic factors such as absorption and residence time should

be considered for both toxins. The possibility that they can have direct chemical

reaction in vivo is less than that of thein vitro system. Direct chemical interaction

will be explored further in the next chapter,

b. Mechanistic interaction hypothesis

The results shown in Figures 4.2 and 4.3 suggest that there is more than one

mechanism responsible for the shift in the mutagenic potential of aflatoxin Bt.

Therefore, it is possible that fumonisin interacted with the activation/detoxification

pathways rather than directly with AFBi. There are several possible mechanisms

that could be either inhibited or induced by fumonisin B|.

Cytochrome P450 enzymatic activity is considered the main pathway

responsible for aflatoxin biotransformation. Several of the P450 enzymes have been

reported to be involved in AFBi metabolism. The relative quantities of different

metabolites produced depend largely on the relative activities of each isozyme. The

major AFBi detoxification metabolites include aflatoxins Mi, Pi and Qi as

detoxification by-products; and AFBi-8,9-epoxide as an activation product.

Furthermore, the 8,9-epoxide has been shown to exist in two stearically different

forms: the 8,9 exo-epoxide and the 8,9-endo epoxide. These molecules have been

shown to have different affinity for DNA and glutathione (Gallagher et al., 1996).

Guenguerich et al. (1996) reported that the exo isomer is 10 3 times more mutagenic

than the endo-epoxide. Furthermore, Gallagher et al. (1992) reported that the 1A2

and 3A4 isozymes appear to be the predominant catalysts of AFBi epoxidation in

139

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. human liver in vitro. Relative to 3A4, 1A2 produces a higher ratio of activation to

detoxification products. The differences in potential activation and detoxification

metabolites suggest that there may be some agents that could shift the equilibrium

to either a detoxification or an activation pathway.

In the case of FBi, its role in cytochrome P450 activity is not clear. It has

been reported that it is not a substrate for the microsomal P450’s, estearases or

hepatic lipase systems. Therefore, direct competition with AFBi should not be an

issue. However, Martinez-Larranaga et al. (1996) reported a marked induction of

hepatic and renal P450 1A, 3A and 4A families in rats fed with FBi. However,

although in Western blot analyses of both 1A isozymes (1A1 and 1A2) showed

marked increase in P450 enzymes, only the activity of the 1A1 isozyme was

increased. The amount of 1A2 increased, but its activity was not different from the

control. This effect on 1A2 was could not be explained by the authors. Relating

these results to the present study, it cannot be assumed that induction was a

mechanism in an in vitro system. However, the possibility that FBi affects

cytochrome P450 1A2 activity could be relevant to the changes in mutagenicity

observed here.

Studies on primary hepatocyte cultures have shown that FBi alters the

incorporation of 14C palmitic acid into cellular lipids indicating, that added to its

effect on sphingolipid metabolism, it also affects synthesis of cellular lipids

(Gelderblom et al., 1996). In addition to fatty acid changes, membrane associated

cholesterol was markedly decreased. This study suggested that FBi has important

140

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. effects on the structure of the major membrane components. Changes in the

membrane structure and function could lead to several toxic effects, and could

finally lead to the disintegration of membrane continuity. From the xenobiotic

metabolism point of view. Kim (1987) and Meydahl et al. (1985) reported that

changes in fatty acid content of hepatic endoplasmic membranes have been

accomplished by altering dietary lipids. It has also been reported that phospholipids

may enhance the binding of substrates to cytochrome P450’s (Tsai and Pence,

1992). Therefore, changes in membrane lipid composition may alter substrate

specificities and participation in various pathways. Hence, the activation and

detoxification of numerous xenobiotics in liver microsomes may be altered.

Whether FBi can exert a similar effect in an in vitro system, such as. the

Salmonella/microsomai mutagenicity assay is yet to be determined.

Although P450 activity is considered the major enzyme system involved in

the metabolic activation of AFBi, other enzyme families may also be involved.

There is an increasing number of reports that have documented that, in general,

polyunsaturated fatty acids cause a marked reduction in the rate of the NADPH-

dependent cytochrome P450 oxidation of aflatoxin (Burgos-Hemandez, 1998;

Nemoto and Takayama, 1983; Stohs and Wu, 1981). Specifically, linoleic and

linolenic acids were reported to strongly inhibit AFBi activation by rat liver

microsomes (Burgos-Hemandez, 1998; Firozi et al., 1986). Also, an inhibition of

AFBi activation mediated by hepatic microsomes in the presence of arachidonic

acid has been reported (Ho et al., 1992). These data have led to question the role of

141

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. P450 as the sole catalyst of AFBi activation. Battista and Mamett (1985) reported

that the enzyme prostaglandin synthetase is capable of mediating the epoxidation of

AFBi through a co-oxidation mechanism. Thus, the relative contribution of other

pathways to AFBi's ultimate mutagenic outcome has gained interest from the

scientific community.

Enzymes, such as, prostaglandin synthetases (PHS), lipoxygenases (LO)

(cyclooxygenases) and peroxidases can have dual functions, they can act as

dioxygenases and as co-oxidases. Furthermore, studies have shown that soybean

LO can co-oxidize a wide range of endogenous chemicals and xenobiotics in the

presence of suitable polyunsaturated fatty acids (Kulkami, 1996). Regarding animal

systems, several studies have reported that LO are present in the liver of several

mammalian species (Roy and Kulkami, 1997; Rosello-Catafau, et al., 1994; Roy

and Kulkami, 1994; Shiratori et al., 1992; Hughes et al., 1992; Liu and Massey,

1992; Pereira and Das, 1991; Kawada et al.. 1990; Maruyama et al., 1989; Tiegs

and Wendel, 1988; Hagmann et al., 1987; Macias et al., 1987; Roll et al., 1986;

Perez et al., 1984; Nemoto and Takayama, 1984). Liu and Massey (1992)

presented evidence that guinea-pig liver cytosolic LO can bioactivate AFBi, and

when compared to the P450-catalyzed reaction, a relatively low Km for the DNA

binding of AFBi was observed. These results implied the importance of the LO

pathway in AFBi bioactivation.

On the other hand, there have been reports that FBi can induce peroxisomal

and microsomal fatty acid-oxidizing enzyme activities. Peroxisomes (or

142

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. "microbodies”) are single membrane-limited cytoplasmic organelles, which are

characterized by their content of catalase and a number of hydrogen peroxide-

generating enzymes. Peroxisomes contain the same fatty acid beta-oxidation cycle

enzymes as mitochondria; therefore, the are metabolically competent to oxidize this

type of compounds. While enzymes of the P-oxidation cycle (normally assessed as

cyanide insensitive palmitoyl-CoA oxidation) are markedly induced, only small

changes are observed in other peroxisomal activities such as catalase and D-amino

acid oxidase. Apart from stimulating peroxisomal fatty acid metabolism,

peroxisome proliferators also increase microsomal fatty acid (co-1)- and particularly

co-hydoxylase activities. This is due to induction of cytochrome p-450 isoenzymes

in the CYP4A subfamily and is normally measured as lauric acid 12-hydrolase

(Lake, 1995). Martinez-Larranaga et al. (1996) reported that treatments of rats with

FBi gave rise to a concentration-dependent increase in hepatic microsomal lauric

acid hydroxylase activity, which is closely associated with the CYP4A subfamily

of hemoproteins. Furthermore, the role of FBi in fatty acid metabolism has already

been discussed. Short-term in vivo and in vitro studies have shown that exposure to

FBi caused major changes in the phospholipid fraction of cells. There was a

significant increase of hepatic C l8:2 (linoleic acid) and arachidonic acid (AA)

(Gelderblom et al., 1995). In vitro studies in different cell lines have shown that

FBi interferes with arachidonic acid conversion to other metabolites, such as

prostaglandins. In primary hepatocyte cultures, exposure to FBi caused an

accumulation of AA suggesting that FBi inhibits the release and/or the

143

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. transformation of AA in primary hepatocytes. an effect that is greatly enhanced at

high (cytotoxic) doses of FBi (Gelderblom et al.. 1996). This fact could be related

to the relative increase in mutagenicity observed at the higher FB i doses.

The studies described above provide evidence that primary hepatocytes

exposed to FBi show an accumulation of polyunsaturated fatty acids (PUFA)

(Gelderblom et al., 1996). The cytotoxic effects of PUFA have been studied

extensively (Gavino et al., 1981; Begin et al., 1988). The data provided by these

studies suggest that the cytotoxicity of PUFA is associated with an increase in the

extent of lipid peroxidation. In conjunction, these observations imply that FBi

treated cells could be more susceptible to lipid peroxidation than normal cells

(Gelderblom et al., 1996).

The role of peroxidases in aflatoxin degradation was discussed in Chapter

III. Prostaglandin H synthetase (PHS) is the most extensively studied peroxidase. It

has been reported that it possesses two distinct catalytic activities. A

cyclooxygenase that converts AA to a cyclic endoperoxide-hydroperoxide (PGG 2 ),

and a peroxidase that converts the hydroperoxide to the corresponding alcohol

(PGH2 ) (Parkinson, 1996).

In summary, cytochrome P450 and peroxyl-radicals formed by PHS,

lipoxygenase, and/or lipid peroxidation may all play a role in activating aflatoxin

Bj. Fumonisin has been shown to be involved in several of these oxidation

mechanisms. More information on the mechanistic implications of FBi’s role in the

144

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. multiple pathways that can oxidize AFBi could help understand the changes in

mutagenicity observed in this study.

Another possibility is that FBi affected the activity of phase II

conjugation/detoxification pathways rather than the P450 enzymes. If FBi affects

enzymes, such as, glutathione-S-tranferases, then the ultimate mutagenic outcome

could be affected. At some point (approximately 100 ng FBi/plate), the system

could have become saturated; therefore, the mutagenicity could return to its original

level. In vivo studies have suggested that FBi changes the level and activities of

glutathione-S-transferases (Lebepe-Mazur et a l 1995). These results give

supportive evidence for this hypothesis. However, since the metabolic activation in

this study was provided by the S9 mix, the main role of fumonisin in this study

needs to be related to enzyme activity rather than to enzyme induction.

In addition, glutathione plays a key role in detoxification by reacting with

hydrogen peroxide and organic peroxides. This reaction is catalyzed by glutathione

peroxidase. This enzyme needs two glutathione molecules to complete its cycle. Its

active site contains a covalently attached Selenium ion (Se) which forms a

selenium analog of cysteine (the sulphur has been replaced by Se). The reduced

form of this enzyme (E-Se') reduces the peroxide substrate to an alcohol and in turn

is oxidized to selenic acid (E-SeOH). Glutathione is then conjugated to form a

selenosulfide adduct (E-Se-S-G). Subsequently, a second glutathione molecule

regenerates the active form of the enzyme by attacking the selenosulfide and

forming oxidized glutathione (Stryer, 1988). This information can also be related to

145

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. the results observed in this study. Since glutathione plays an important role in

detoxification of both, electrophiles and free radicals, the generation of free radicals

caused by higher levels of FBi, may have depleted glutathione. Therefore, free

radicals may have interfered with the major AFBi detoxification pathway. This is a

plausible account for the increase in mutagenicity observed at higher FB| levels.

Initially, the lower FBi levels may have generated some free radicals that were

able to react with a multitude of compounds, including the AFB 1-8 ,9-epoxide. This

could explain the decrease in mutagenicity. Then, at higher concentrations, the

generation of higher quantities of free radicals, coupled with enhancement of co

oxidation mechanisms and glutathione depletion may have promoted the return to

increased mutagenicity.

The involvement of FBi in the oxidation pathways should be explored to

determine its ability to inhibit and/or enhance toxic responses. This information

would give better tools to estimate the risk of dietary exposure to fumonisin and

evaluate it from the perspective of food safety.

2. Aflatoxin Bi/fumonisin B2 mutagenicity

Since in Chapter III a high production of FB 2 by Fusarium moniliforme was

reported, and chemical differences between FBi and FB 2 are minimal (Figure 2.4).

The ability of FB2 to affect AFBi mutagenicity was also tested. Due to the dose-

dependent fashion of the shift in mutagenicity, it was considered that if FB 2 exerted

a similar effect, an additive response could occur during dietary exposure to co

contaminated com.

146

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figures 4.4 and 4.5 show the effect of FB 2 on AFBi mutagenicity. It can be

observed that the effect was less marked than that of FBi. Salmonella tester strain

showed a tendency to exhibit the same behavior, but a definite effect could not be

established. However, there is some evidence that FB 2 could elicit the same

response as FBi even if the response is weaker. Potential synergistic or additive

effects should be explored.

3. Toxicity of AFB1/FB1 and AFB1/FB2

The toxicity of all of the samples tested for mutagenicity was evaluated to

determine if the decrease in mutagenicity was indeed due to an interaction that

favored detoxification, or was it due to toxicity to the tester strains. Results from

the toxicity assay are presented in Figures 4.6-4.8. Although some variation was

observed, none of the values were significantly different when compared to the

control. This assay had high variability that could be attributed to the fact that at

some concentrations, the bacterial suspension was highly diluted. Each plate

presented between 1 and 10 colonies. When the dilution factor was considered,

difference in one colony/plate in reality meant 1x10s colonies. Although the results

from this assay may be considered inconclusive due to high variability, the lack of

significant difference may imply that the effect observed in the mutagenicity assays

was due more to an interactive behavior that enhanced detoxification rather than to

an acute toxicity response. Other acute toxicity assays should be employed to

determine the toxic potential of these samples.

147

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. —■— ng AFBI 1000 - 0 ng AFBI tester strain TA-100 with metabolic Salmonella combinations. 2 /FB 1 FBI (ng/plate) 0.00E+00 1.00E+00 1.00E+01 1.00E+02 1.00E+03 1.00E+04 1.00E+05 500 Figure Figure 4.4 Mutagenic potential of AFB activation. activation. Natural revertants=153±l are Values three replicates. mean±SEM of 3500 3000 2500

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1000ng AFBI O ngA FBl tester strain TA-98 with metabolic Salmonella combinations. 2 /FB 1 1.00B 02 l.OQEKB 1.00E+O4 1.00E+O5 FBI (n^plate) Figure 4.5 Mutagenic potential of AFB activation. activation. Natural revertants=32±3. Values are mean±SEM three ofreplicates. O.OOBOO O.OOBOO 1.00BO0 l.OQBOl 200 800 400 1000 1400 1200 600 • a > ID cd (U $ •£» o

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ■One AFBI H I ng AFBI □ ng AFBI 10 ■ ng AFBI 100 □ ng AFBI 1000 tester strain TA-100 with metabolic Salmonella FBI (ng/plate) . 0.00E+00 1.00E+00 1.00E+01 1.00E+02 1.00E+03 1.00E+04 1.00E+05 1.00E+08 1.00E+08 9.00E+08 8.00E+08 Figure Figure 4.6 Toxic potential of AFB|/FB| combinations. 7.00E+08 6.00E+08 5.00E+08 3.00E+08 activation. Negative control, DMSO=6xl08colony forming ±2x108 units/plate. Values are mean±SEM ofthree replicates. 0.00E+00 2.00E+08 4.00E+08

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ] 10 ng AFBI AFBI ng 10 ] 2 gO AFBI ng □ 1 □AFBI ng 1 [ g AFBI ng 100 AFBI ng 1000 gg tester strain TA-98 with metabolic activation. Salmonella FBI FBI (ng/plate) O.OOE+OO l.OOE+OO 1.00E+01 l.OOE+OO 1.00E+02 O.OOE+OO 1.00E+03 1.00E+04 1.00E+05 1.60E+09 Figure Figure 4.7 Toxic potential ofAFB|/FB| combinations. 1.40E+09 Negative control, colony units/plate. Values forming DMSO=9xl08are mean±SEM of ±3x108 three replicates. 2.00E+08 2.00E+08 : , O.OOE+OO o, £ 6.00E+08 o o « 1.20E+09 S 1.00E+09 00 2 <2 u 4.00E+08 I 8.00E+08 L fl

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. H TA-98 TA-98 H without AFBI □TA-100 with ng AFBI 1000 ■TA-98 with ng AFBI 1000 ■TA-100 without AFBI tester strains TA-100 and TA-98 with 1.00E+05 Salmonella combinations. 2 /FB FB2 (ng/p!ate) FB2 1 1.00E+01 1.00E+01 1.00E+03 O.OOE+OO Figure 4.8 Toxic potential of AFB metabolic activation. Negative control, DMSO=6xl08±2xl08 and colony 9xl08 forming±3x108 units/plate for TA-100 TA-98, and respectively. are Values ofthree replicates. mean+SEM 1.40E+09 1.20E+09 O.OOE+OO 8.00E+08 2.00E+08 00 o 4.00E+08

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. This experiment showed that simultaneous exposure to AFBi and FB| may

elicit different responses than exposure to the individual toxins. This effect could

be due to a combination of a multitude of factors that may include direct chemical

interaction or enhancement/inhibition of different metabolic pathways. The overall

response may be due to an equilibrium of all these reactions. The possible chemical

interaction as well as the involvement in mechanistic pathways will be further

explored in subsequent sections.

153

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. V. Possible mechanisms of interaction between aflatoxin Bi and fumonisin Bi

A. Introduction

In the previous chapter, it was observed that FBi changed the mutagenicity

elicited by AFBi in a dose-dependent fashion. In an attempt to understand these

results, several hypothesis were presented. Therefore, further experiments were

performed to get more information on the possible interactions that might have

taken place. The first approach was to determine if the interaction was due to a

direct chemical reaction between both toxins.

The objective of this study was to determine the recovery of a fixed amount

of AFBi (1 x 103 ng=3 nmol) after reaction with a range of different amounts of FBi

(1 to IxlO5 ng= 1.4x1 O'3 to 1.4xl02 nmol) with metabolic activation. In addition, the

recovery of a fixed amount of FBi (lxlO' ng=l.4xl02 nmol) after exposure to a

range of AFBi (3x1 O'3 to 3 nmol) was also determined. The null hypothesis on this

experiment was that the recovery of both AFBi and FBi decreases as the

concentration of the other toxin increases. These results would show that there is a

direct chemical reaction either between the two toxins or with other toxin

metabolite(s).

A second approach was taken to determine if the changes were due to a

direct effect on the metabolic pathways rather than to direct chemical reactions.

The objective of this portion of the study was to determine the effect of FB i on the

mutagenic potential of a different mutagen, 2-aminofluorene (2-AF). This

154

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. compound is considered a positive control for the Salmonella!microsomal

mutagenicity assay with metabolic activation (Maron and Ames. 1983).

2-AF was considered the ideal candidate since its mutagenicity requires

metabolic activation and involves a different mechanism than that of AFBi- 2-AF

does not form DNA adducts through the formation of an epoxide (Figure 5.1). This

fact was important since that would rule out the potential Maillard type reaction

with fumonisin, and therefore, give a better understanding on the effect of FBi

directly on activation/detoxification pathways, without direct chemical reaction

interference.

It has been reported that in mammalian systems, the in vitro N-oxidation of

2-AF is mediated through pathways involving cytochrome P450, specifically P450

1A2 (Wolfel et al., 1991), flavin containing monooxygenase (FMO) (Hammons et

al., 1985), and co-oxidation via the peroxidase, prostaglandin synthetase (Wolfel et

al., 1991). These mechanisms are similar to some of the mechanisms shown to

bioactivate AFBi. Therefore, it was considered that the evaluation of FBj's role in

2-AF mutagenicity would be a useful tool to elucidate potential mechanisms

involved in the change of AFBi mutagenicity observed in Chapter IV.

B. Materials and methods

1. Bacterial cultures

Salmonella typhimurium tester strains TA-100 and TA-98 were kindly

provided by Dr. Bruce N. Ames (University of California Berkley, Berkley, CA).

155

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 2-Nttronuorene

2-Arnrcfluorcne

NHAc

N-Hydroxy-2-Amir)fijorene N . , 2-Acetybrrnnonuorene OH

N- Hydro^,-2- Arrmonuorene

N-(Deox>guarr)sirh8-y[)-2-amaiofluorene

N -( Deoxygjanosin- &-y[>- 2- acetytaminofLiorene

Figure 5.1 Metabolic transformation of 2-Aminofluorene. R=Fluorene group (Kriek, 1992) 156

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 2. Chemicals

Acetic acid. 2-aminofluorene, ammonium sulfate, ampicillin, D-biotin,

disodium hydrogen phosphate, gIucose-6-phosphate, L-histidine, magnesium

sulfate. 2-mercaptoethanol, ortho-phtaldehide, sodium ammonium sulfate,

disodium tetraborate, sodium chloride, sodium dihydrogen phosphate, B-

nicotinamide adenine-dinucleotide phosphate (NADP) and trifluoroacetic acid were

purchased from Sigma Chemical Co. (St. Louis, MO). Chloroform (HPLC grade),

citric acid monohydrate, dimethyl sulfoxide (DMSO), ethyl ether anhydrous,

hexane (HPLC grade), methanol (HPLC grade), potassium chloride, potassium

phosphate dibasic anhydrous and water (HPLC grade) were acquired from

Mallinckrodt Baker, Inc. (Paris, KT). Bacto agar and crystal violet were ordered

from Difco Laboratories (Detroit, MI); glucose from Fisher Scientific (Fair Lawn,

NJ). Acetonitrile (HPLC grade) was from Baxter Co. (Muskegon, MI). Ethanol was

obtained from Aaper Alcohol and Chemical Co. (Shelbyville, KT). Oxoid nutrient

broth No. 2 was from Unipath LTD (Basingstoke, Hampshire, England). Post-

mitochondrial supernatant (S9 mix) was purchased from Molecular Toxicology,

Inc. (Boone, NC).

Pure fumonisin Bi was kindly provided by Dr. Robert M. Eppley (Center

for Food Safety and Applied Nutrition-CFSAN-, U. S. Food and Drug

Administration-FDA-, Washington, D.C.). Pure aflatoxin B| was generously

provided by Dr. Mary Trucksess (CFSAN-FDA. Washington, D. C.). Aflatoxin Bi,

157

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. B2 , Gi, and Gt HPLC standard 50 pg/g mixture was purchased from Romer

Laboratories, Inc. (Union, MO).

3. Chemical interaction

a. Effect of fumonisin B| on aflatoxin B| recovery after metabolic activation

Pure AFBi and FBi stock solutions (lpg/pl and lOpg/pI, respectively) were

prepared by solubilizing pre-weighed purified toxins in a methanol:water (30:70)

solution. The samples were prepared and placed in amber vials immediately before

the assay. Stock solutions were combined and diluted to obtain concentrations that

would yield the final test amounts shown on Table 5.1.

Table 5.1 AFBi/FBi combinations tested for potential chemical interactions with metabolic activation Aflatoxin Bi (nmol) Fumonisin Bi (nmol) 0 3 0 0/0 3/0 1.4x1 O'3 0/1.4x10° 3/1.4x10° 1.4x1 O'2 0/1.4x1 O'2 3/1.4x10° 1.4x10'' 0/1.4x10 '1 3/1.4x10'* 1.4 0/1.4 3/1.4 1.4x10' 0/1.4x10* 3/1.4x10* 1.4xl02 0/1.4xl02 3/1.4x10" Combinations are expressed as nmol AFBi/nmol FBi. Samples were prepared in triplicate.

100 pi of test solutions were placed in sterile disposable capped culture

tubes and 500 pi S9 cocktail prepared according to Maron and Ames (1983) were

added unless otherwise indicated (negative control). Pure AFBi (3 nmol) without

metabolic activation (500 pi S9 cocktail without postmitochondrial supernatant

fraction) was used as a recovery control. Sample 0/0 only contained 500 pi S-9 to 158

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. determine if the S-9 mix contained compounds that would interfere with the liquid

chromatographic analysis.

The tubes were incubated in a water bath at 37°C with shaking (180-200

rpm) for 90 minutes. After incubation, the samples were placed in ice to stop the

reactions. Aflatoxin and/or metabolites were extracted by adding 1 ml CHCU to the

culture tubes and vortexing. The aqueous and organic phases were allowed to

separate and the CHCI3 phase was separated and collected in an amber vial. This

procedure was repeated three times to obtain ca. 3 ml CHCI 3 extract. The samples

were evaporated to dryness under a N 2 stream. AFBi was then purified according to

the method reported by Thean et al. (1980). The purified extracts were analyzed by

liquid chromatography following the method reported by Wilson and Romer

(1991). The purification and quantification methods were described in detail in

Chapter III.

b. Thin layer chromatography (TLC) of recovered aflatoxin Bi and/or metabolites

Since the HPLC analytical method used in the previous section requires

AFBi derivatization with trifluoroacetic acid (TFA) to yield the more fluorescent

AFB2a, thin layer chromatography was used to qualitatively determine the presence

of metabolites that would be converted to AFB2a when exposed to TFA.

Pre-coated aluminum silica gel plates (0.25 mm) were activated by placing

them in a dry incubator at 60°C for at least 3 hours. A 1 pg AFBi/pi

benzene:acetonitrile (98:2) solution was used as a control. 5 pi of pure AFBi and 5

pi of the same samples that were injected into the HPLC (described in the previous

159

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. section) were spotted at 1.0 cm distance and at 1.5 cm. from the lower edge of the

TLC plate. The spots were allowed to dry. Then, 5 pi of TFA derivatizing agent (1

ml trifluoroacetic acid+4 ml hexane) were applied directly over the samples

previously spotted. After the spots dried, the plate was placed in an incubator at

65°C for 10 minutes to allow for derivatization reaction. The plates were allowed to

cool down, and 5 pi of underivatized pure AFBi and underivatized samples were

spotted at 1.0 cm from each other and from the derivatized spots. The plate was

developed in a TLC chamber using chloroform.acetic acid:water (98:1:1) as a

mobile phase. The solvent front was marked and the plate was removed from the

chamber and allowed to dry under the hood. The dried plate was observed under a

long wavelength UV light. Fluorescent spots were marked to calculate the Rf

values.

c. Effect of aflatoxin Bi on fumonisin Bi recovery after metabolic activation

Pure AFBi and FBi stock solutions (lpg/pl and lOpg/pl, respectively) were

prepared by solubilizing pre-weighed purified toxins in a methanol:water (3:1)

solution. The samples were prepared in amber vials immediately before the assay.

Stock solutions were combined and diluted to obtain concentrations that would

yield the final test amounts shown in Table 5.2.

1 0 0 pi of test solutions were placed in sterile disposable capped culture

tubes and 500 pi S9 cocktail prepared according to Maron and Ames (1983) were

added unless otherwise indicated (negative control). Pure FBj (1.4xl0 2 nmol)

without metabolic activation (500 pi S9 cocktail without postmitochondrial

160

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. supernatant fraction) was used as a recovery control. Sample 0/0 only contained

500 pi S-9 to determine if the S-9 mix contained compounds that would interfere

with the liquid chromatographic analysis.

Table 5.2 FBi/AFBi combinations tested for potential chemical interactions with metabolic activation Fumonisin Bi (nmol)

Aflatoxin B[ (nmol) 0 140 0 0 / 0 3/0 3x1 0° 0/3x10° 140/3x10° 3x1 O’2 0/3x1 O' 2 140/3xl0‘2 3x10'' 0/3x10' 1 140/3x10** 3 0/3 140/3 Combinations are expressed as nmol FBi/nmol AFB|. Samples were prepared in triplicate.

The tubes were incubated in a water bath at 37°C with shaking (180-200

rpm) for 90 minutes. After incubation, the samples were placed in ice to stop the

reactions. Fumonisin and/or metabolites were extracted by adding 2.5 ml

methanol:water (3:1) to the culture tubes and vortexing. The resulting solution was

placed in a 20 ml scintillation vial. The reaction tubes were washed a total of 4

times with methanol:water (3:1) to obtain a total volume of 10 ml/sample. This

final solution was mixed thoroughly (vortex - 1 min) and the pH was adjusted to

5.8 -6 .5, if necessary. This solution was then applied to pre-conditioned anion

exchange (SAX) columns and FBi/metabolites were purified and quantified by

liquid chromatography according to AO AC official method of analysis 995.15

(1996). This method was described in detail in Chapter III.

161

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. d. Statistical analysis

Statistical analysis was performed using the SAS software (SAS. 1988).

One-way analysis of variance (ANOVA) was used to determine differences in toxin

recovery. Scheffe's test with an alpha value of 0.05 was used to compare

differences among the different treatments.

4. Effect of fumonisin Bi on 2-aminofluorene mutagenic potential

The mutagenic potential of 2-AF and 2-AF/FBi combinations was

determined following the methods for the Salmomllalmicxosovnal mutagenicity

assay reported by Maron and Ames (1983) using Salmonella tester strains TA-100

and TA-98 with and without metabolic activation. A detailed description of this

procedure was provided in Chapter III.

a. Mutagenic potential of 2-aminofluorene

A preliminary dose range finding assay was performed to determine an

ideal 2-AF dose to combine with FBi. 2-AF was tested in a range from lxlO 2 to

lxlO6 ng/plate using Salmonella tester strains TA-100 and TA-98 with and without

metabolic activation.

b. Effect of Fumonisin Bi on 2-aminofluorene mutagenicity

A fixed dose of lxlO 4 ng 2-AF/plate was selected to determine FBi’s effect

on 2-AF mutagenic potential. FBi was tested using the same concentration range

used to determine its effect on AFBi’s mutagenic potential (1-lxl0 5 ng/plate).

Samples were prepared in amber vials immediately before the assay using DMSO

as the carrier solvent. Dilutions were made from 10 jig FBj/pl DMSO and 100 pg

162

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 2-AF/jil DMSO slock solutions to yield the sample combinations presented in

Table 5.3.

c. Safety precautions

Since 2-AF is considered a mutagen and the Material Safety Data Sheet

information on its potential toxic effects was limited, the assay was carried out with

extreme caution using appropriate personal protection equipment that included

body suits, double nitrile gloves, half face negative pressure organic vapor

respirator and safety glasses. Sample size and number of assays performed were

kept to a minimum to limit waste generation. 2-AF was handled using disposable

glassware that was properly decontaminated and disposed of at conclusion of the

assay. Waste material generated from this experiment was autoclaved and disposed

through Louisiana State University Department of Occupational and Environmental

Safety once the assay was concluded.

Table 5.3 2-aminofluorene/fumonisin Bi combinations used to evaluate the effect of FBi on 2 AF mutagenic potential. 2-AF (ng/plate)

FBj (ng/plate) 0 1 x 1 0 “ 0 0 / 0 lx l 0 4/ 0 1 0 / 1 lxl 0 4/ 1 1 x 1 0 * 0 / 1 x 1 0 * lxl 0 4 / 1 x 1 0 1 lxlO2 0/1 xlO2 Ix l 0 4 /lx l 0 z lxlO 3 0 /lx l 0 J 1x 104/ 1 x 10j lxlO4 O/lxlO4 Ix l 0 4 /lx l 0 “ lxlO 3 O/lxlO3 lxlO’VlxlO3 All samples were tested in triplicate with and without metabolic activation.

163

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. C. Results and discussion

1. Effect of fumonisin Bi on aflatoxin Bi recovery after metabolic activation

When APBi was exposed to varying FBi concentrations, it was observed

that the results (Figure 5.2) emulated the mutagenicity behavioral pattern shown in

Figures 4.2 and 4.3. The amount of aflatoxin recovered decreased as FBi increased

until FBi reached lxlO 2 ng (1.4x1 O ' 1 nmol). At this point, AFBi recovery started

increasing. Although the only treatments that were significantly different (p<0.05)

from the control were observed when l.4xl0 4 and 1.4 nmol FBi were added, the

pattern was evident. This behavior suggested that direct chemical interaction was

not involved, at least not in all the FBj concentration range. If there had been a

chemical reaction, a single dose-response would have been observed. AFBi

recovery would have decreased as FBi increased.

Analysis of the controls indicated ca. 90% recovery from the extraction

procedure. This showed that the extraction method was suitable for this assay and

that the variations observed were due to chemical interaction. No peaks were

detected in the chromatograms of the controls, FBi without AFBi and S9.

Therefore, it was assumed that neither FBj nor S9 interfered with the results.

It is noteworthy that chromatograms showed a peak for the AFB i derivative

AFB2a. These results were not expected since AFBi had been exposed to S9

enzymatic activity. AFB 2 a is formed when AFBi is exposed to TFA. Hence, in

theory, AFBi in this study should have been transformed to other metabolites.

Since AFB 2 a can be obtained through biotransformation pathways, it was

164

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1.40E+01 1.40E+01 1.40E+02 1.4 FB, added (nmol) 1.40E-02 1.40E-02 1.40E-01 1.40E-03 1.5 3.5 2.5 Figure Figure 5.2 Aflatoxin B| mean±SEM ofthree recovery replicates. after reaction with fumonisin B| with metabolic activation. Values are c o> o o ca (U C3 £ O > o o u o § 0.5 '■J3

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. hypothesized that the exposure to FBi with metabolic activation yielded AFB^a- To

test this hypothesis, an underivatized sample was analyzed in the same HPLC

system. If AFB^a were generated from the S9 metabolic activity, a peak with a

similar retention time as AFB 2 a was expected from injection of underivatized

samples. However, there were no detectable peaks in the chromatograms of these

samples. Thus, it was assumed that AFB2a was not directly generated by S9

metabolic activity in this in vitro system.

It was also hypothesized that some other secondary compound was

generated during the incubation, and this compound was also derivatized to AFB 2a

or to a compound that co-eluted with AFB 2 a. Therefore, the peaks observed for

AFB2a could be a combination of one or more derivatized metabolites plus residual

AFBi that were co-extracted with CHCI3 . It would be assumed that the rest of the

AFBi was biotransformed and/or conjugated to more polar compounds that had a

higher affinity to the aqueous phase and were not extracted with CHCI 3 . To

qualitatively test this hypothesis, derivatized and underivatized pure AFBi and

CHCL-extracted samples were fractionated on a TLC plate. The results are shown

in Table 5.4.

Table 5.4 Rf values and description of TLC spots of derivatized and underivatized samples. Sample R f value Spot description AFBj+TFA 0.16 Bright blue fluorescence Sample+TFA 0.16 Bright blue fluorescence AFBi 0.63 Medium blue fluorescence Sample 0.71 Faint dark blue

166

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. The results (Table 5.4) suggest that an AFBi metabolite or byproduct was

recovered in the CHCI 3 fraction. This metabolite (Rf=0.7l) was transformed either

to AFB2a (Rf=0.16) or to a product that co-eluted with AFB2a (Rf=0.16) after

derivatization with TFA. It is interesting to note that Doyle and Marth (1978)

reported that when lactoperoxidase was used to degrade AFBi, derivatives that

co-chromatographed with AFB2a and derivatives that were water soluble were the

major degradation products of AFBi. This report, coupled with the results

presented in this study, provide evidence for the potential involvement of

peroxidase activity in AFBi mutagenicity. Further research is needed to

characterize the metabolites produced by this pathway and to determine the direct

effect of FBi on peroxidase-dependent AFBi oxidation.

2. Effect of aflatoxin Bi on fumonisin Bi recovery after metabolic activation

When FBj was exposed to AFBi with metabolic activation, no significant

(p<0.05) changes were observed (Figure 5.3). However, it was observed that the

control for the extraction method was ca. 8 8 %. While FBi recovery for all the

samples ranged from ca. 60-75%. Hence, some of the FBi was not recovered after

reaction with AFBi and metabolic activation. No dose-response relationship was

observed. This may be due to the fact that an excess FBi was used for this assay.

This amount of FBi (lxlO 3 ng) was selected to facilitate detection by HPLC. The

excess FBi in the system, may have masked its relationship with S9. However,

from under these experimental conditions, it can be assumed that direct chemical

interaction is not a major contributor to the shift of AFBi mutagenicity caused by

167

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 3.2 3.20E-01 3.20E-02 3.20E-02 AFB, AFB, added (nmol) 3.20E-03 0 80 60 20 40 100 140 120 Figure Figure 5.3 Fumonisin B| recovery after reaction with aflatoxin B| with metabolic activation. Values are mean+SEM mean+SEM ofthree replicates. " G > O u o cd

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. FB[. More research is needed to determine the role of FBi in xenobiotic co

oxidation mechanisms.

3. Effect of fumonisin Bt on 2-aminofluorene mutagenicity

Figures 5.4 and 5.5 show 2-AF mutagenic potential using Salmonella tester

strains TA-100 and TA-98 with and without metabolic activation. It was observed

that 2-AF exhibited the same behavior in both tester strains. When the activating

enzyme mixture (S9) was present, 2-AF mutagenic potential increased in a dose-

response fashion up to a concentration of 10 3 ng/plate. Plates containing 10 6

ng/plate with and without metabolic activation did not show any growth. Toxicity

testing on nutrient agar plates confirmed that 2-AF at this concentration was toxic

to both, Salmonella TA-100 and TA-98.

The results shown in Figures 5.4 and 5.5 implied that a concentration range

of 10 4 to 10 3 would be useful for testing the effect of FBi used to test AFBi

mutagenicity. It was determined that at this level, any potential variation caused by

FBj would be easily identified. This amount of 2-AF showed a marked mutagenic

response and was lxlO 2 fold lower that the amount that showed a marked cytotoxic

response. Therefore, 2-AF was fixed at 10 4 ng/plate and FBi (10°-10 3 ng/plate) was

added. These combinations did not show a cytotoxic response when evaluated on

nutrient agar plates.

When the effect of FBi (10°-105 ng/plate) on the mutagenic potential of 2-

AF (104 ng/plate) was evaluated, an interesting response was observed. FBi

changed 2-AF mutagenic potential in the same way as it affected the mutagenic

169

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. . Without Without S9 . +... Natural Revertants -»— With S9 1.00E+05 1.00E+06 2-AF (ng/plate) 2-AF 1.00E+03 1.00E+04 . 0 1.00E+02 800 200 600 400 Figure Figure 5.4 Mutagenic potential of 2-aminofluorene mean±SEM with ofthree replicates. and without metabolic activation (S9). Values are 1000 1800 1600 1400 1200

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. . Without Without S9 . — With S9 « ...Natural Revertants tester strain TA-98 with and without metabolic Salmonella 2-AF (ng/plate) 2-AF 1.00E+03 1.00E+04 1.00E+05 1.00E+06 . 1.00E+02 500 500 . Figure Figure 5.5 Mutagenic potential of 2-aminofluorene. activation (S9). activation (S9). Values are mean±SEM ofthree replicates. 2500 2000 2 o sj sj 1000 . f I 1500 . &

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. potential of AFBi (Figures 5.6 and 5.7). There was a marked decrease up to a point

(102 ng FBi/plate) where the mutagenic response started to return to its original

level. It was interesting to observe that the return occurred around the same FBi

concentrations that elicited a change in AFBi mutagenicity and chemical recovery.

These results suggest that FBi affects one or more of the common metabolic

pathways that are involved in 2-AF and AFBi activation/detoxification.

Leist et al. (1992) reported that enzymes from both, the microsomal and the

cytosolic fractions of hepatic S9 mix are involved in the activation of 2-AF to a

mutagen. Furthermore, there have been reports that show that the cytosolic fraction

of this mix can markedly enhance the mutagenic response of chemicals such as

AFBi (Saccone and Pariza, 1981), benzo[a]pyrene (Saccone and Pariza, 1981), 2-

acetylaminofluorene (Forster et al., 1981), 2-amino-3-methylimidazo-(4,5-

quinoline) (Abu-Shakra et al., 1986), benzidine (Smith and Ioannides, 1987) and

N-nitrosopiperidine (Ayrton et al.. 1987).

Leist et al. (1992) reported that the first and rate limiting step in the

bioactivation of aromatic amines is the N-hydroxylation catalyzed by microsomal

enzymes and that the hydroxylated amines act as the proximate mutagens. Phase II

conjugation enzymes produce electrophilic species such as the nitrenium ions that

can bind covalently to DNA. The involvement of enzymes, such as prostaglandin

endoperoxide synthetase, peroxidase and the peroxidase-like enzymes in the

bioactivation of 2-AF to a mutagenic species has been reported by several authors

(Isola et al., 1993; Corbett and Corbett, 1988). Leist et al. (1992) also reported that

172

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. tester strain With S9 With S9 — Without S9 Salmonella +.. .Natural Revertants FBI FBI (ng/plate) . ¥■

* -

0 0.00E+00 1.00E+00 1.00E+01 1.00E+02 1.00E+03 1.00E+04 1.00E+05 200 800 600 400 . Figure Figure 5.6 Mutagenic potential of 2-aminofluorene/Fumonisin B| combinations. TA-lOOwith withoutand metabolic activation (S9). Values are mean±SEM TA-lOOwith three ofreplicates. 1000 1400 1600 1800 2000 * 8 § § I - 1200 u>

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. tester strain Without Without S9 With With S9 Salmonella . .Natural. Revertants * ------* ------• FBI FBI (ng/plate) ------» * - * . 0 0.00E+00 1.00E+00 1.00E+05 1.00E+04 1.00E+01 1.00E+03 1.00E+02 500 Figure Figure 5.7 Mutagenic potential of 2-aminofluorene/Fumonisin Bi combinations. TA-98 TA-98 with and without metabolic activation (S9). Values are mean±SEM ofthree replicates. 1000 2000 2500 , 3000

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. cytosolic oxygenase activity is involved in the bioactivation of 2-AF. These authors

also reported that the cytosolic activation of 2-AF is dependent on the presence of

O2 , and that established oxygen radical scavengers such as DMSO and retinol

inhibit the reaction.

The oxygen radical scavenger potential of DMSO may have interfered with

the results presented in the current study. DMSO could have interfered with the

total response, but other factors may have also been involved. DMSO was used as a

negative control in this study. Therefore, the results can still be evaluated relative to

each other. Moreover, the chemical interaction study reported previously in this

chapter, used a different carrier system (methanol:water; 30:70) and the same dose-

response behavior was observed. If FBi is in fact involved in a free-radical

peroxidation mechanism that affects the biotransformation of chemicals such as

AFBi and 2-AF, its effect should be evaluated in a system that does not contain

added oxygen scavengers such as DMSO. This could give more marked responses

that would help elucidate the mechanism(s) involved.

The results from this study suggest that FBi affects AFBi and that 2-AF

biotransformation in a dose-response fashion through a similar mechanism. It has

been reported that both, 2-AF and AFBi are mainly activated by the cytochrome

P450 1A2 isozyme and by oxidative reactions mediated by peroxidase and

peroxidase-like enzymes.

From the fumonisin perspective, there is evidence that suggests that FBi

alters P450 1A2 activity (Martinez-Larranaga, 1996) and that it alters the levels and

175

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. fatty acid composition in rat liver in vitro and in vivo systems (Gelderblom et al.,

1996; 1997). The accumulation of polyunsaturated fatty acids coupled with

enhancement of peroxidase and peroxidase-like activities implies that FB i can alter

peroxidase-mediated reactions. Therefore, the evaluation of the involvement of FBi

on the activity of microsomal and cytosolic fractions and its specific effect on P450

1 A2 activity would be useful in determining its role in the activation/transformation

of xenobiotics, such as, AFBi. This information would help in assessing the

potential risk and food safety concerns associated with aflatoxin and fumonisin co

contamination in com.

176

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. VI. Effect of Fumonisin B| on gap-junction intercellular communication of clone-9 rat liver cells

A. Introduction

In the previous chapters, the possible interactions between aflatoxin B| and

fumonisin Bi were explored, mainly through the evaluation of the mutagenic

potential of different AFBjrFBi ratios. However, since it does not show a positive

response in the Salmonella/microsomal test and other mutagenicity assays FBi is

considered a non-genotoxic agent. FBi’s lack of mutagenicity has complicated the

development of short term assays that could help monitor its effect on an in vitro

system and facilitate the study of potential interactions. Therefore, the need for

rapid and reliable assays to monitor the effect of fumonisin was identified.

Molecular studies of human cancers have provided abundant evidence that

carcinogenesis requires an accumulation of mutational events before a normal cell

transforms into a malignant tumor (Fearon and Vogelstein, 1990). Therefore, the

use of short-term mutagenesis assays to evaluate chemical carcinogens is well

justified. However, it has been reported that standard short-term mutagenesis test

results identify no more than 65-89% of rodent carcinogens (Ashby and Tennant,

1991). Hence, Yamasaki (1996) has speculated that even if the concept of these

assays is on target, there is a misconception that genetic changes required for

carcinogenesis are always directly induced by exogenous carcinogens and thus,

detected by short term assays such as the Salmonella/microsomal mutagenicity

assay. The general concept that chemical carcinogens directly interact with DNA is

177

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. widely accepted because many environmental carcinogens, i.e., tobacco smoke,

aflatoxin, sunlight, etc.. have been implicated in certain types of cancer and because

the genetic basis of cancer has been supported with extensive evidence. However,

some of these concepts cannot be entirely translated into simple mutation assays. In

support of this idea, it has been reported that exposure to agents that induce cell

proliferation, block intercellular communication, or produce oxygen radicals may

eventually lead to genetic damage, but these mutational events are considered

indirect, and are usually not detected by mutation assays (Yamasaki. 1996). Several

authors have provided experimental evidence to consider that one possible non-

genotoxic mechanism of carcinogenesis is the inhibition of gap-junction

intercellular communication (GJIC) (Holden et al., 1997; Mesnil et al., 1995;

Yamasaki. 1990; Trosko and Chang, 1988; Loewenstein, 1979).

GJIC has been considered an important factor in cell growth and

differentiation. Furthermore, in recent years it has been widely believed that GJIC

plays a pivotal role in cell proliferation (Na et al., 1995). It is believed that

considerable levels of important biological materials, such as, growth control

factors are exchanged through gap junctions (GJs) (Na et al., 1995). GJs are

composed of protein channels that permit the exchange of nutrients, ionic signals

and regulatory molecules of up to 1500 Dalton between adjacent communication-

competent cells (El-fouly et al., 1987). It has been reported that certain non-

permeable molecules can be introduced into cells via a "scrape-loading” technique

(El-fouly et al., 1987). Scrape-loading has been effectively utilized to introduce

178

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. macromolecules into cells in culture by inducing a transient tear in the plasma

membrane without affecting cell viability or colony forming ability (McNeil et al.,

1984).

Lucifer yellow (3,6-Disulfonate-4amino-naphtalimide; MW=457.2) is a

water soluble, low molecular weight fluorescent dye that does not diffuse through

intact plasma membranes, but readily passes from one cell to another, potentially

revealing molecular relationships between cells (Hermanson, 1996). El-fouly et al.

(1987) determined that Lucifer Yellow transfers from cell to cell through gap

junctions. Therefore, these authors developed a Lucifer yellow scrape-loading and

dye transfer technique to study gap-junction intercellular communication. This

method can be used to determine if certain non-genotoxic chemical carcinogens are

involved in the carcinogenesis process by interfering with GJIC.

FB i non-genotoxic nature combined with the existing evidence regarding its

effect on lipid metabolism and membrane structure and function, led to the

hypothesis that it could affect GJIC. Thus, the objectives of this phase of the study

were (1) to determine if fumonisin inhibits GJIC in vitro, and (2) to determine if

APBi changed FBi effect on GJIC.

B. Materials and methods

1. Chemicals

Pure fumonisin B| was kindly provided by Dr. Robert M. Eppley (Center

for Food Safety and Applied Nutrition (CFSAN), U. S. Food and Drug

179

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Administration (FDA), Washington, D. C.). Pure aflatoxin B[ was provided by Dr.

Mary Trucksess (CFSAN-FDA, Washington, D. C.).

Ham’s F-12K cell culture media and fetal bovine serum were purchased

from American Type Culture Collection (ATCC) (Rockville, MD). Acetic acid,

Dulbecco’s phosphate buffer solution without calcium and magnesium (PBS-Ca,

Mg), Dulbecco’s phosphate buffer solution with calcium and magnesium (PBS+Ca,

Mg), ethanol, 0.25% trypsin, 0.03% versene solution. 30% hydrogen peroxide,

Lucifer yellow, methanol, neutral red, lOOx penicillin/streptomycin solution, trypan

blue, 25 cm 2 cell culture flasks and 35 mm cell culture dishes were acquired from

Sigma chemical company (St. Louis, MO).

2. Cell line

Clone-9 (CRL-1439) rat liver epithelial cell line isolated in 1968 from

normal liver taken from a young male rat (Gerschenson et al., 1975; Weinstein et

al., 1975) was purchased from ATCC (Rockville, MD). Cells were grown in Ham’s

F-12K medium with 10% fetal bovine serum, and 1% penicillin streptomycin

solution under an atmosphere of 5% CO 2 in humidified air at 37°C. Culture media

was renewed 1 to 2 times weekly. Subculturing was performed by removing old

media and adding fresh 0.25% trypsin, 0.03% versene solution for 2 to 3 minutes.

The trypsin solution was then removed and the culture was incubated at 37°C for

10-15 minutes to permit cell detachment. Fresh medium was then added, aspirated

and dispensed into new flasks in a 1 : 6 ratio. Bioassays were conducted with cells

grown in 35 mm culture dishes to confluency after 2-3 days growth.

180

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 3. Test solutions

Toxin dilutions were prepared immediately before the assay. Stock

solutions 10 pg/pl and lpg/pl of FBi and AFB[, respectively were prepared by

solubilizing pre-weighed pure toxins in a sterile 30% ethanol solution (0.22 p-

Millipore syringe filter (Millipore, New Bedford, MA)). Combination samples as

shown in Table 6.1 were prepared in sterile capped culture tubes using sterile 30%

ethanol as a solvent carrier. 30% ethanol solution was used as a negative control in

all assays.

Table 6.1 Fumonisin Bi and aflatoxin Bi samples tested for cytotoxicity and effect on gap-j unction intercellular communication

AFBj (ng/plate)

FBi (ng/plate) 0 I 1 x 1 0 * lx l 0 z lx l 0 J 0 0 / 0 0 / 1 0 / 1 x 1 0 * 0/1 xlO2 0 /lxl0 3 1 1 / 0 * * **

1 x1 0 * lxl0 * / 0 * * * * lxlO3 lxl0 3 / 0 * * ** * lxlO3 lxl0 3 / 0 * * * * lxlO4 lxl0 4 / 0 * * ♦ lxlO5 lxl0 5/ 0 lxloVl Ixl0 3/lxl0 * Ixl 0 5/lxl0 2 Ixl0 5/lxl0 3 Combination values are expressed as ng FBi/pIate / ng AFBi/plate. *: Combination not tested All samples were prepared in triplicate

4. Growth curve

A growth curve to determine cell growth characteristics and standardize all

assays was obtained by growing cells in 25 cm 2 cell culture flask until they reached

a concentration of 5x10 4 cells/ml. Cells were counted by staining with trypan blue

181

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. and using an hemocyotmeter. Cell suspensions were added to 35 mm cell culture

dishes at four different concentration levels, 3xlOJ, 7xl03, 13x10J and 27xlOJ>

cells/ml. Volume was brought up to 2 ml/plate. These plates were incubated at

37°C in a 5% CO 2 humidified atmosphere. Cells were counted at 0, 55, 8 8 and 120

hours incubation.

5. Neutral red cytotoxicity assay

Cytotoxicity as determined by the neutral red uptake assay according to the

method reported by Borenfreund and Puemer (1985). Clone-9 cells were grown in

35 mm cell culture dishes at the same conditions and confluency (ca. 4x10 4

cells/ml) used for the cell communication bioassay (2-3 days). Old growth media

was discarded and 2 ml fresh media were added to each disc. 1 0 0 pi test solutions

(Table 6.1) were added to three plates/sample. Plates were incubated for 24 hours

and cytotoxicity was determined.

The neutral red solution was prepared by mixing it with Ham's F-12K

medium. The media containing neutral red was incubated for 2 hours in a water

bath at 37°C. The resulting precipitation was centrifuged at 2000 g and filtered

through a 0.22 p-Millipore syringe filter (Millipore, New Bedford, MA).

The cells were rinsed three times with PBS+Ca, Mg and then 2 ml fresh

growth media containing 0.033% neutral red was added to the cells. Plates were

incubated at 37°C for 1 hour. After incubation, the extracellular neutral red was

rinsed off with PBS+Ca, Mg and the cells were lysed with 2 ml of an aqueous

solution containing 1% acetic acid and 50% ethanol. The lysed cells were measured

182

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. for neutral red at a wavelength of 540 nm. Methanol (200 pi) was used as a positive

control and 30% ethanol was used as a negative control.

6. Gap-junction intercellular communication

Gap-junction intercellular communication in clone-9 rat liver cells was

measured using the scrape-loading dye transfer method reported by El-Fouly et al..

(1987) with a modification described by Na et al. (1994). Cells were grown to

confluency in 35 mm cell culture dishes (ca. 4xl04 cells/ml). Old media was

discarded and replaced with fresh media. Test samples (Table 6.1) were added (100

pi) 24 hours prior to scrape-loading. After 24 hours incubation, media was

aspirated and reserved in sterile capped culture tubes at 37°C. Cells were washed

twice with PBS-Ca, Mg. The plate was then flooded with PBS containing 1%

Lucifer yellow and then scrape loaded with a plastic scraper. The dye solution was

left on the cell monolayers for 1 min. and then the cultures were washed three times

with PBS with Ca and Mg to remove unloaded dye and background fluorescence.

Growth media was replaced in each dish and the dishes were incubated at 37°C in a

humidified atmosphere for 10 minutes to allow for dye transfer. Cultures were then

examined under an Olympus inverted microscope equipped with a reflected light

fluorescence attachment with HBO-100 W/2 mercury lamp (Olympus, Tokyo,

Japan). Gap-junction communication was measured by counting the number of

cells from the scrape-loaded line that were stained with the fluorescence dye. Cells

were counted on 20 different locations on each plate. Samples were analyzed in

triplicate.

183

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 7. Statistical analysis

Statistical analysis was performed using the SAS software (SAS, 1988).

One way analysis of variance (ANOVA) was used to determine differences in toxin

recovery. Scheffe's test with an alpha value of 0.05 was used to compare

differences among the treatments.

C. Results and discussion

1. Growth curve

A growth curve (Figure 6 .1) was constructed to optimize growth times and

standardize the experimental conditions for this assay. From the growth curve, it

was determined that plating a cell suspension containing ca. 1.3x10 4 cells/ml

yielded the adequate growth to confluency in the time required to perform this

experiment. This was crucial since it has been observed that FBi does not elicit any

effect to cells in culture until after an initial lag period of 24-48 hours (Yoo et al.,

1992). It was important to test cells at confluency, but that would stay in a log or

stationary phase throughout the assay. Once the growth parameters were

determined, they were used for the rest of the bioassays.

2. Neutral red (NR) cytotoxicity assay

The NR assay is based on the incorporation of a weakly cationic dye (NR)

into the lysosomes of viable cells after their incubation with toxic chemicals. NR

penetrates cell membranes by nonionic diffusion and binds intracellularly to sites of

184

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1.3E+04 cells/ml 1.3E+04 .. ■ . 6.6E+03 cells/ml _ ♦jjj _ 3.3E+03 cells/ml _ ♦jjj —*—2.6E+04 cells/ml Initial Initial Cell Concentrations 120 88 Time Time (hours) 55 0 * Figure Figure 6.1 Growth curves Clone-9 for cells. rat liver Values are mean+SEM of counts. thirty 1.00E+04 0.00E+00 , 3.00E+04 3.00E+04 . 2.00E+04 , 5.00E+04 5.00E+04 . 4.00E+04 . 6.00E+04 E u

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. the lysosomal matrix (Lullmann-Rauch, 1979). Xenobiotics that injure the plasma

or lysosomal membrane decrease NR uptake and subsequent retention. After the

cells are rinsed, only undamaged cells retain the dye. NR is then extracted from the

lysosomes and quantified spectrophotometrically. Methanol was used as a positive

control to optimize the neutral red cytotoxicity assay. Figure 6.2 shows the dose-

response relationship caused by the toxicity of methanol. It can be observed that as

methanol increased, the absorbance decreased indicating that less cells were able to

retain NR.

When cells exposed to FBi and/or AFBi were analyzed, an interesting

pattern was observed (Figure 6.3). All treatments had significantly higher (p<0.05)

absorbance than the control. This suggested that not only were the treatments not

acutely toxic, but they induced more intracellular NR retention. One plausible

explanation of this effect is the possibility that the toxins are indeed acting as

peroxisome proliferators as previously reported (Martinez-Larranaga, 1996). Since

the morphology of peroxisomes resembles that of lysosomes (Darnell et al., 1990),

it could be possible that exposure to the toxins caused an increase of these

organelles as an early toxic response. The peroxisomes may have retained the NR

causing the increase in absorbance. Although this hypothesis needs more research,

it provides further evidence that peroxisomes and oxidative metabolism may play

an important role in FBi toxicity and in the interaction between AFBi and FB|.

186

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

o u2 C /3 +1 G uca £

O4>

VO 1) -2 •a;Q. 3 -35O O ' c o ca e .c _ o

w U E o ^ Oc ca

■*-* < u

£ Cm o

! s c CN 4 > ■o M Oh u

'* O

H CN

v © M

187

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1.00E+03 AFBI 1.00E+03 1.00E+05 FBl(fixed)+0 1.00E+05 to —«—AFBI - + - - - •* FB1 £ £ -* Toxin concentration (ng/plate) 0 0 1 1.00E+01 1.00E+02 1.00E+03 1.00E+04 1.00E+05 Figure Figure 6.3 Toxic potential of fumonisin B| and/or aflatoxin B| Clone-9 to rat cells. liver Values are mean±SEM ofthree replicates. 1 0 0.2 0.1 0.6 0.8 0.3 0.9 0.7 3

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 3. Effect of fumonsin B| on gap-junction intercellular communication

The results of this study suggest that FBi inhibits gap-junction intercellular

communication of Clone-9 rat liver cells. Table 6.2 summarizes the results of this

assay and shows the significant differences between the different concentrations

tested.

Table 6.2 Effect of fumonisin Bi on gap-junction intercellular communication of Clone-9 rat liver cells

Sample Mean±SEM (n=60) *

Negative control 6 .2 ±0 . 1 A Positive control (H 2 O2 ) 400pM 1.3±0.2 F 1 ng FB [/plate 4.9±0.2 A,B lxlO1 ng FBi/plate 3.8±0.2 B,C lxlO2 ng FBi/plate 3.5±0.1 C lxlO3 ng FBi/plate 3.2±0.2 C lxlO4 ng FB,/plate 1.5±0.2 F lxlO3 ng FBi/plate 1.7±0.2 C,D,E,F lxlO5 ng FB i/plate + 1 ng AFB [/plate 1 .2 ±0 . 1 E,F 1X105 ng FB [/plate + lxlO1 ng AFB [/plate 2.6±0.3 C.D.E 1X105 ng FBi/plate + lxlO2 ng AFBi/plate 2 .8 ±0 . 2 C,D lXlO3 ng FB [/plate + lxlO3 ng AFB [/plate 2.9±0.3 C.D lxlO3 ng AFB [/plate 5.2±0.2 A GJIC is expressed as the mean number of fluorescent cells counted from the scrape- loaded line. Values with the same letter are not significantly different.

It was noted that FB| inhibits GJIC in a dose-response fashion (Figure 6.4).

Communication decreased as FBi concentration increased. These results show that

inhibition of GJIC may play a role in FBi carcinogenesis. This is supported by

other reports on FBi role on membrane structure and function. Yoo et al. (1992)

reported that after 24-48 hours of FB| exposure LLCP-PK1 cells lost cell-cell

contact.

189

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1.00E+04 1.00E+04 1.00E+05 FBI FBI (ng/plate) I.00E+01 I.00E+01 1.00E+02 1.00E+03 (400 pM)=0.21 ±0.029. ±0.029. are Values meaniSEM (400 pM)=0.21 ofsixty replicates. 2 O 2 0 k

: 0 Figure 6.4 Effect of fumonisin B| on gap junction intercellular communication (GJIC) of Clone-9 rat liver cells. cells. Positive control H 0.2 0.1 0.9 0.9 0.8 0-4 0.3 0.5 I o cu § § ^ ^ - 1 £ U 1 0.7 I «•« ’§ vO o

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Furthermore, Niedel et al. reported that TPA (12-O-tetradecanoylphorboI-

13-acetate), a known tumor promoter, inhibits GJIC. It has been speculated that

TPA mediates its tumor promoter effects through the activation of protein kinase C

(PK-C) and the subsequent phosphorylation of critical cellular proteins (Ashendel

et al., 1985; Blumberg et al., 1984; Niedel et al., 1983). There are also several

reports that suggest a role for TP A-induced production of active oxygen species in

the promotion of initiated cells (Kensler and Trush, 1984). Repine et a/. (1974) have

reported that TPA elicits a rapid burst of free radical scavenging enzymes. Several

authors have demonstrated that PK-C activation is involved in TPA-induced

inhibition of GJIC (Oh et al., 1988; Gainer and Murray, 1985; Enomoto and

Yamasaki, 1985; Davidson et al., 1985). Furthermore, PK-C has been shown to

phosphorylate the rat liver gap junction protein in a cell-free system (Takeda et al.,

1987). These reports could be related to some of the potential modes of action

suggested for FBi.

FBi has been shown to be a natural inhibitor of sphingolipid metabolism,

and sphingolipids are the natural inhibitors of the Ca2+ activated, phospholipid

dependent phosphorylating enzyme protein kinase C (PKC). Thus, it has been

speculated that FBi may also affect PKC-regulated functions (Kharmalov et al.,

1993; Hannun et al., 1986). Yeung et al. (1996) have reported that FBi induces

PKC translocation via direct interaction with the diacylglycerol binding site. PKC

is activated by diacylglycerol (DAG) generated via receptor-mediated hydrolysis of

inositol phospholipids to relay information derived from a variety of extracellular

191

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. signals to regulate many Ca2+-dependent processes. In addition, PKC also serves as

an intracellular receptor for tumor-promoting phorbol esters, which are stable DAG

analogues. The reports that FBi induces PKC translocation via a direct action on

the DAG site that also binds phorbol esters (Yeung et al., 1996) coupled with the

reports that PKC activity plays an important role in phorbol esters tumor promotion

activity (Ashendel et a l, 1985), suggest that FBi may inhibit GJIC through a

mechanism similar to that of the phorbol esters.

Na et al. (1995) reported that a dose of up to lxlO4 ng AFBi/ml did not

significantly inhibit GJIC. These results were confirmed in this study where lxlO3

ng/plate did not significantly (p<0.05) decrease GJIC (Table 6.2). Figure 6.5 shows

that when a range of 1 to lxlO3 ng AFB i/plate is combined with lxlO5 ng FBi/plate

there is relatively no change in GJIC inhibition. Therefore, these results imply that

there is no interaction between the two toxins in this system. FBi GJIC inhibitory

mode of action could be considered independent of AFBi. However, the metabolic

capability of Clone-9 rat liver cells is unknown. Hence, the lack of interaction

could be due to lack of enzyme systems capable of activating AFB). The possible

interaction between FBi/AFBt that could affect GJIC should be studied in

metabolically competent systems to determine if in reality AFBi does not affect

FBi inhibition of GJIC.

The results presented in this study suggest that the inhibition of GJIC could be a

non-genotoxic mechanism of FBi involvement in carcinogenesis. Furthermore, the

positive response observed with a short term in vitro assay such as the scrape-

192

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ng 3 lxlO (400 ±0.029; |iM)=0.21 2 O 2 AFBI AFBI (ng/plate) . 0 0 1 1.00E+01 1.00E+02 1.00E+03 0 0 1 Figure Figure 6.5 ofaflatoxin Effect Bi junction on the inhibition of gap intercellular communication (GJIC) ofClone- AFB |/plate=0.83+0.033. are mean+SEM AFB Values ofreplicates. sixty 9 9 rat liver cells caused by 1x10s ng FBj/plate. Positive control H 0.1 0.2 0.5 0.6 0.8 0.9

h 2 » » o o 2 g 0.7 o e* 03 O 1 0.4 U

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. loading technique provides a useful tool to study FB[. It was observed that

inhibition of GJIC in Clone-9 rat liver cells could be a useful end-point for future

FBi studies.

194

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. VII. Effect of hydrogen peroxide and sodium bicarbonate on the toxin content and the toxic/mutagenic potential of aflatoxin- fumonisin naturally co-contaminated corn

A. Introduction

Perhaps too much attention has been given to the carcinogenic potential of

mycotoxins, such as, aflatoxin. However, the possibility that low levels of

mycotoxins or combinations of several mycotoxins could cause secondary

mycotoxin-related syndromes in humans, in a way similar to those now being a

threat to animal populations cannot be overlooked. If this were true, then the real

implications of mycotoxin presence in human foods would be more threatening.

Until there is a better understanding on the mode of action and the interaction

between mycotoxins and the multiple components in diet, every effort should be

made to reduce or eliminate mycotoxins from the human dietary chain.

The previous chapters have shown some evidence that simultaneous

exposure to AFBi and FBi can elicit a wide variety of toxic responses. The exact

mechanisms by which this occurs are still poorly understood. Thus, it is impossible

to assess the real risk posed by co-exposure to these toxins. Furthermore, the

potential interactions presented in this study bring out the fact that exposure to

xenobiotics is a more complex issue. In reality, humans and animals get exposed to

a multitude of compounds that may have either toxic or protective effects.

Therefore, the ultimate toxic outcome relies on each individual’s capacity to

maintain a balance. Taking these concepts into consideration, it is clear that

195

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. limiting the exposure to agents with known toxic potential will help maintain the

equilibrium on the side of health.

Com is a dietary staple in many areas of the world. In some areas in Mexico

and Latin America, com can provide up to 70% of the daily caloric intake.

Therefore, potential exposure to AFBi and FBi is high. However, shifting

consumer preferences in areas of the world that are less poverty stricken has also

elevated the risk of exposure in populations that were traditionally not considered

at risk. A recent study has reported that in the United States of America, tortilla

sales between 1994 and 1995 showed that tortillas are the fastest growing segment

of the U.S. baking industry. This segment is expected to continue growing

approximately 54% in the next five years. According to the tortilla industry

association, tortillas are more popular today than all other ethnic breads, such as,

bagels, English muffins and pita bread. It has been estimated that industrial growth

in 1998 will continue to remain at about 10% to 14% for most tortilla producers.

Furthermore, the growth outlook for European markets is estimated to be 50%

(Cornell, 1998). Even if these trends include wheat-flour tortilla sales, they suggest

that dietary practices are rapidly changing; therefore, consumers are more liable to

get higher and more frequent exposure to mycotoxins, such as, AFBi and FBi in

com and com products. This, added to the lack of information regarding

fumonisins’ ultimate role in health and disease, and the evidence that relates

aflatoxins to mutagenic events and hepatic cancer, emphasize the need to develop

196

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. control mechanisms such as detoxification procedures which address multiple

mycotoxins.

In previous studies (Lopez-Garcia, 1995; Park et al., 1996), the use of

hydrogen peroxide and sodium bicarbonate was proposed as a specific alternative

for fumonisin detoxification during nixtamalization, a traditional process used in

tortilla manufacture which involves boiling and soaking of com in lime water. The

results of this study indicated the use of H 2 O2 and NAHCO 3 without

nixtamalization was effective in reducing fumonisins below detectable levels and

the reduction was accompanied by a decrease in toxicity to brine shrimp ( Artemia,

sp.). These studies also reported that no mutagenic by-products emerged as a result

of the reaction(s). Other studies (Valdivia, 1995) have reported that H 2O2 and

NaHCOs were effective in reducing aflatoxin levels in com. Furthermore, evidence

for the involvement of oxidative mechanisms in toxin degradation has been

presented in previous chapters. These reactions could be used to

detoxify/decontaminate com, before consumption.

H2 O2 and NaHCOj are readily available food grade ingredients and their

use could be adapted to any com processing operation. Thus, the objective of this

study was to determine if H 2O2 and/or NaHCC >3 were effective in

decontaminating/detoxifying aflatoxin- fumonisin-naturally co-contaminated com.

197

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. B. Materials and methods

1. Corn and bacterial cultures

Naturally contaminated com containing 375 pg/g total fumonisins and 20 ng/g

total aflatoxins was kindly provided by Dr. Jeff Ellis (Texas State Department of

Agricultural Chemistry).

Salmonella typhimurium tester strains TA-100 and TA-98 were kindly

provided by Dr. Bruce N. Ames (University of California Berkley, Berkley, CA).

Artemia sp. larvae were purchased from a local pet store (Baton Rouge, LA).

2. Chemicals

Acetic acid, ammonium sulfate, ampicillin, D-biotin, disodium

hydrogen phosphate, glucose- 6 -phosphate, L-histidine, magnesium sulfate, 2-

mercaptoethanol, ortho-phtaldehide, sodium ammonium sulfate, disodium

tetraborate, sodium chloride, sodium dihydrogen phosphate, 13-nicotinamide

adenine-dinucleotide phosphate (NADP) and trifluoroacetic acid were purchased

from Sigma Chemical Co. (St. Louis, MO). Chloroform (HPLC grade), citric acid

monohydrate, dimethyl sulfoxide, ethyl ether anhydrous, hexane (HPLC grade),

methanol (HPLC grade), potassium chloride, potassium phosphate dibasic

anhydrous and water (HPLC grade) were acquired from Mallinckrodt Baker, Inc.

(Paris, KT). Bacto agar and crystal violet were ordered from Difco Laboratories

(Detroit, MI); glucose from Fisher Scientific (Fair Lawn, NJ). Acetomitrile (HPLC

grade) was from Baxter Co. (Muskegon, MI). Ethanol was obtained from Aaper

Alcohol and Chemical Co. (Shelbyville, KT). Oxoid nutrient broth No. 2 was from

198

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Unipath LTD (Basingstoke, Hampshire, England). Post-mitochondrial supernatant

(S9 mix) was purchased from Molecular Toxicology, Inc. (Boone, NC).

Pure fumonisins B| and B 2 were kindly provided by Dr. Robert M. Eppley

(Center for Food Safety and Applied Nutrition - CFSAN, U. S. Food and Drug

Administration-FDA, Washington, D.C.). Pure aflatoxin Bi was generously

provided by Dr. Mary Trucksess (CFSAN-FDA, Washington, D. C.). Aflatoxin Bi,

B2 , Gi, and G 2 HPLC standard 50 jig/g mixture was purchased from Romer

Laboratories, Inc. (Union, MO).

3. Corn treatments

Naturally contaminated com containing 375 pg/g total fumonisins and 20

ng/g total aflatoxins was weighed and separated into 100 g portions. The different

treatments (Control-distilled water, 3% H 2 O2 , 2% NaHC0 3 , and 3% H 2 0 2 +2 %

NaHCOs) were randomly assigned to three samples. The com was exposed to the

treatments for one hour at room temperature with mixing every 2 0 min and then

frozen at 20°C to stop the reactions. After freezing, the samples were freeze-dried

for 72 hours. The dry samples were kept under refrigeration (4°C) until analysis.

4. Aflatoxin analysis

Extraction, clean up, and quantitative determination by liquid

chromatography (HPLC) were performed according to the method reported by

(Thean et al., 1980). This method was described in detail in Chapter III. The results

were compared to the control treatment (distilled water) and a percentage of

199

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. aflatoxin reduction was calculated. Aflatoxins Bi (AFBi) and B 2 (AFB2 ) were

quantified individually. Aflatoxins Gi and G 2 were not detected in the original com.

5. Fumonisin analysis

Fumonisin extraction, clean up and quantitative determination by HPLC

was performed according to the AOAC official method 995.15 for fumonisins Bi,

B2 , and B 3 (AOAC, 1995). The results were compared to the control treatment

(distilled water) and a percentage of fumonisin reduction was calculated.

Fumonisins Bi and B 2 were evaluated individually; fumonisin B 3 was not

determined due to lack of a pure toxin standard.

6. Toxicity/mutagenicity testing

A composite sample from each treatment was prepared by mixing the crude

extracts from the fumonisin and aflatoxin extractions. The fined solution contained

124 pg/ml total fumonisins and 6.7 ng/ml total aflatoxins.

7. Salmonella!microsomal Mutagenicity assay

Salmonella tester strain TA-100 was used to evaluate the mutagenic

potential of aflatoxin and the decontamination reaction products. The

Salmonella!microsomal mutagenicity assay was conducted essentially as described

by Maron and Ames (1983). This method was described in detail in Chapter III.

Aflatoxin Bi ( 8 8 , 8 .8 , 0.88, and 0.088 ng/ml) and fumonisin B 1/B2 (125, 12.5, 1.25,

and 0.125 pg/ml) concentrations were run. Mutagenicity was evaluated with and

without metabolic activation (S-9). Composites from the com extracts were

prepared to obtain a representative sample for each treatment. Dilutions were made

200

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. from each sample to obtain ca. 12.4, 1.24, 0.124, and 0.0124 pg of

fumonisin/reaction products per plate and 0.67, 0.067, 0.0067, and 0.00067 ng of

aflatoxin/reaction products per plate.

8. Brine shrimp assay

The brine shrimp assay (Artemia sp.) was performed according to the

method reported by Park et al. (1986). Tubes containing 10 and 100 pi of the

composite sample extracts were evaporated to dryness under a N 2 stream in a water

bath at 60°C. Negative (methanol) and positive (fumonisin B[, 0.5 pg/tube)

controls and an extract of aflatoxin/fumonisin free com were included.

9. Statistical analysis

Statistical analysis was performed using the SAS software (SAS, 1988).

One-way analysis of variance (ANOVA) was used to determine differences in toxin

production and toxicity to S. typhimurium. Scheffe’s test with an alpha value of

0.05 was used to compare differences among the different treatments.

C. Results and discussion

The treatment involving the combination of H 2 O2 and NaHC 0 3 was the

most effective in significantly (p<0.05) reducing AFBi, FBi, and FB 2 (Figure 7.1).

However, less reduction was observed for AFB 2 . These results suggest that the

main oxidation reaction occurs on the double bond present in aflatoxin Bi (Figure

2.1). This effect becomes even more evident when observing the individual effect

of H 2 O2 and NaHCC >3 on AFB 2 (Figure 7.1). H 2 O2 showed no significant (p<0.05)

reduction of this toxin. However, treatment with NaHCC >3 resulted in higher

201

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ®H202/NaHC03 §|NaHC03 □H202 naturally contaminated corn 2 FB2 , , and fumonisin Bi and B 2 FBI Toxin AFB2 separate or in combination decontamination treatments.Values are mean±SEM 0 3 and NaHC 2 O 2 AFBI 02008948485353485323539048234853484848232323534848532323 i I I I I 0 10 90 80 70 60 50 40 30 20 100 Figure Figure 7.1 Chemical reduction of aflatoxin B) and B exposed to H ofthree replicates. G O 0> o 3 Pi 0 s- '■G "O to to o

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. reduction. In the case of FBi and FB 2 , an additive effect can be observed. The

treatment involving both, H 2 O2 and NaHC 0 3 reduced FB| and FB 2 twice as much

as the reduction observed with the individual chemicals. These results show that the

use of a combination of H 2 O2 and NaHC 0 3 is significantly effective in reducing

both, aflatoxins and fumonisins.

Hydrogen peroxide and sodium bicarbonate could be incorporated as an

additional processing step in any com processing set up. However, a potential

problem has been observed. The reaction with these chemicals is exothermic and

there is abundant foaming. This could pose a problem in an industrial operation.

Thus, new research on these treatments should include the addition of an anti-

foaming agent to reduce the foaming problem. Further evaluation of safety, and

effect on sensory characteristics and technological properties as well as pilot plant

escalation is recommended for these treatments.

With respect to brine shrimp toxic potential, positive toxic responses were

obtained by the untreated sample. However, the chemical treatments involving

H202/NaHCC>3 significantly (p<0.05) reduced the toxicity (Figure 7.2). When

comparing the chemical modification of fumonisins and aflatoxins based on HPLC

analysis and the toxic potential, a clear relationship was observed (Figure 7.3). The

higher percentage of reduction, the lower the brine shrimp mortality. The brine

shrimp assay is only a preliminary testing system. Therefore, further acute toxicity

testing of these treatments with more elaborate assays should be performed to

further evaluate their safety.

203

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. naturally 2 ■48 hours ■ hours 36 □ 24 hours □ hours 12 H202 NaHC03 H202 , , and fumonisin B) and B 2 NaHC03 Sample or combination separate decontaminationin treatments. (> 3 and and NaHC 2 O 2 (-) control (-) control (+) control Untreated 20 40 60 80 (+): (+): positive control=0.5pg FB|/tube; negative(-): control=methanol. Values are mean±SEM threeof replicates. Figure 7.2 Toxic potentials to brine shrimp of aflatoxin B| and B contaminated comexposed to H 100 to •£- o

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission ) ) as determined by 2 ®H202/NaHC03 □H202 □Untreated §|NaHC03 separate or in combination >3 hours) to Brine to Brine Mortality Mortality Shrimp (48 Shrimp (48 and NaHCC 2 O ; ; total fumonisin=FB|+FB 2 2 Total reduction fumonisin Total aflatoxin reduction 0 10 50 30 20 70 80 60 40 90 Values are mean±SEM three replicates. of Figure Figure 7.3 Total toxin reduction HPLC (total compared aflatoxin=AFB|+AFB to the toxic potential of the raw com extracts as determined by brine shrimp after 48 hour exposure. exposure. Extracts from decontamination treatments using H 100 xO 0s to 01 0

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. For the Salmonella!microsomal mutagenicity assay, sample revertant

colonies greater than or equal to two times the spontaneous reversion rate are

considered a positive mutagenic response. In samples analyzed without metabolic

activation (Figure 7.4). the untreated sample showed a positive mutagenic response.

Since AFBi is not mutagenic without metabolic activation, this suggests the

presence of mutagenic compounds other than AFBi. However, the treated samples

did not show a positive response. Therefore, these results imply that the treatments

might be effective in detoxifying other unidentified compounds.

Mutagenic responses were not apparent for any of the treatments with

metabolic activation (Figure 7.5). It is important to note that the untreated com

extracts did not contain a high enough AFBi concentration to elicit a mutagenic

response. However, the lack of mutagenicity presented by the treated samples

implies that no mutagenic by-products were generated in the oxidation reactions. It

is also important to consider that raw com extracts were used to determine the

samples’ toxic/mutagenic potential; therefore, the complex matrix may have

provided complex protective or antimutagenic reactions.

These treatments need more evaluation to be deemed safe. However, the

information presented in this study suggests that treatment with HiCh/NaHCOs

could be a good option for the com processing industry. Since these chemicals are

already in use as food grade additives with several applications in the food

industry, their addition during com processing operations could present a good

206

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TA- Salmonella Untreated H202/NaHC03 o_. _«_NaHC03 A-..H202 .. _ —*— Natural revertants separate or in combination) using tester strain 0 3 and NaHC 2 O 2 Sequential Sequential dilutions ofcorn extracts . 0 0 1.00E-03 1.00E-03 1.00E-02 1.00E-01 1 100 without metabolic activation. 100 are meaniSEMValues ofthree replicates. Figure Figure 7.4 Mutagenic potential of corn extracts from aflatoxin/fumonisin co-contaminated corn exposed to 100 decontamination treatments (H 300 200 400 03 o

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TA- Salmonella H202/NaHC03 . Untreated . . Natural revertants . o - . _«_NaHC03 ..*..H202 _ —#- separate or combination)in using tester strain >3 and NaHCC and 2 O 2 1.00E-02 1.00E-02 1.00E-01 Sequential Sequential dilutions ofcorn extracts . 0 0 1.00E-03 100 with metabolic activation. 100 are Values mean±SEM of three replicates. Figure 7.5 Mutagenic potential of corn extracts from aflatoxin/fumonisin co-contaminated corn exposed to decontamination treatments (H 100 200 400 300 3 "E, 13 Ni o 00

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. cost-effective option to reduce the risks associated with aflatoxin and fumonisin co

contaminated com. Since mycotoxin contamination is heterogeneous in nature, it is

difficult to estimate the real toxin content in the whole lot; even with the use of

good sampling programs. Therefore, the development of cost-effective

decontamination treatments will help establish manufacturing procedures that

include decontamination steps regardless of the results of mycotoxin analyses.

These processes would help reduce the overall potential threat posed by aflatoxin

and fumonisin co-contamination in com.

209

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. VIII. Reduction of risks associated with mycotoxin contamination*

A. Introduction

The unavoidable and unpredictable character of naturally occurring

toxicants, such as, mycotoxins poses a unique challenge to food safety. Surveys of

several agricultural commodities including products for human consumption, have

consistently revealed the presence of mycotoxins. It would be unthinkable to

destroy all contaminated products without seriously compromising the global

availability of food. Hence, risk management through processing and

decontamination/detoxification to avoid potential risks associated with mycotoxins,

and maintain the world's food supply has become a major issue in food safety. The

previous chapters have shown that there are many different issues involved in

mycotoxin contamination. Therefore, only a true interdisciplinary approach will

help in ultimately reducing the risks associated with dietary exposure to

mycotoxins.

B. Integrated mycotoxin management programs

A food safety program in a given country will have to consider factors such

as the public health significance of the contaminant, the effects on the availability

of the countries' food supply, economy, legal infrastructure, analytical resources,

producer/consumer education, the effectiveness of communication systems as well

as the availability of pre-harvest, harvest, and post- harvest technologies.

* Portions of this section has been reprinted with permission from Ref. 98-160), p. (407) by courtesy of Marcel Dekker Inc. (See Appendix C).

210

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Prevention through pre-harvest management is the best method for

controlling mycotoxin contamination; however, should the contamination occur,

the hazards associated with the toxin must be managed through post-harvest

procedures if the product is to be used for food or feed purposes. Ideally, there

should be an integrated management system where mycotoxin hazards would be

minimized in every phase of production.

Once a food commodity has been harvested, there should be routine steps to

ensure the safety of the product. Figure 8.1 shows a decision making process to

develop a food safety program that will ultimately prevent potential hazards

associated with exposure to mycotoxin contaminated food and feed. This process

involves several steps that include the identification of the toxicant, hazard

evaluation, exposure determination, risk determination, and risk management. A

hazard evaluation provides a description of the expected lesion and of the

dose/response relationship. The determination is made based on available data on

comparative toxicological properties of the compound of interest, usually using

laboratory animal and in vitro toxicological tests. Following the development of

suitable analytical and sampling procedures, a determination must be made of the

geographic and temporal distribution of the toxic agent in each major commodity

susceptible to contamination. Combined with the commodity consumption data,

this information can lead to a determination of exposure to the toxic agent by

various human and animal populations. A risk determination is made using

available information on hazard evaluation and exposure assessment (Park and

211

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Identification of the toxicant

Hazard evaluation

Exposure determ (nation

Risk determ ination

Risk m a n a g e m ent

Establishment ofa monitoring program Samp lin g P la n

A n a ly s is

Sam pie Sub-sam pling c o lie c tio n

Acceptable product V io la tive p ro d u c t

Oecontam ination p roced u res

D ive rs io n to less Safe.wholesom e risk uses food/feed supply

Product destru ctio n

Figure 8.1 Risk assessment analysis and food safety program for naturally occurring toxicants

212

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Stoloff, 1989). Finally, after the risk has been determined, risk management

decisions can be made. A good risk management program involves the

establishment of regulatory limits, a monitoring program, a sampling plan, and

development of alternative processing/ decontamination procedures, or diversion to

less risk uses. This program should also include the development of strategies to

destroy highly contaminated products.

Processing can play an important role in diminishing the potential risks of

mycotoxin-contaminated food commodities. Thus, it is important to evaluate the

effects of processing on individual toxins to determine the actual risk posed to

humans/animals. A large amount of data has been generated through surveys on

incidence and levels of mycotoxin contamination in agricultural products, such as

nuts, oilseeds and grains (Stoloff, 1982; 1976). However, published information on

mycotoxins in foods processed for human consumption is limited. More research is

needed to provide food manufacturers and regulatory authorities with information

on mycotoxin persistence and transformation during processing. This information

would be helpful in the development of effective food safety programs.

In the past, the establishment of a regulatory limit for the mycotoxin

considered to pose the highest risk, i.e., aflatoxin, was thought to be enough to

ensure a safe food supply. Any product that was found to be contaminated over the

regulatory limit had to be diverted to less risk uses or destroyed. At present, not

only is more information available, but also the quantity of food supply is more

limited throughout the world (FAO, 1996). Furthermore, there have been several

213

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. reports on the co-contamination of various toxins, i.e., aflatoxin Bj/fumonisin

B i (Park et al., 1996), ochratoxin A/aflatoxin Bj (Harvey et al., 1989; Brownie and

Brownie, 1988), ochratoxin A/citrinin (Brown et al., 1986; Glahn et al., 1988),

ochratoxin A/deoxynivalenol (Kubena et al., 1984), ochratoxin A/penicillic acid

(Kubena et al., 1984), ochratoxin A/T-2 toxin (Kubena et al., 1989),

aflatoxin/cyclopiazonic acid (Smith et al., 1992), aflatoxin/kojic acid (Giroir et al..

1991), aflatoxin Bj/deoxynivalenol (Huff et al., 1986), and aflatoxin Bj/T-2 toxin

(Harvey et al., 1990). More documentation on the levels and types of co-

contaminating mycotoxins occurring in different commodities is needed to more

clearly define the mycotoxin combinations and concentrations to which humans

and animals are likely to be exposed. A multi-toxin model should be considered

when evaluating a particular decontamination/detoxification process. This complex

environment poses new regulatory concerns. When the effect of processing and

modified processing to destroy/inactivate one or several toxins is considered, many

decision making paradoxes may rise. Should innovative processing designed to

detoxify a product be allowed, or should traditionally processed products

contaminated with low/medium doses of several mycotoxins stay in the market.

More research is needed in the area of alternative processing that will

decontaminate/detoxify mycotoxin-contaminated commodities, and render products

that will be approved by regulatory agencies and stay commercially competitive.

Ultimately, the best decontamination/detoxification process would be the one that

214

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. is approved by regulatory agencies, cost-effective, and that reduces the risks of

exposure to multiple-toxin contaminated commodities to acceptable levels.

When evaluating the effectiveness of a specific processing procedure on a

given mycotoxin or multi-toxin system, many factors need to be examined. These

factors include:

1. Chemical stability of the mycotoxin(s);

2. Nature of the process;

3. Type and interaction with the food matrix; and

4. Interaction between different mycotoxins present.

Some processes involve the separation of the commodity into different

fractions, i.e., milling or oil pressing. In these cases, it is important to determine

which fraction remains contaminated. If most of the toxin(s) remain with process

by-products and these are not used for human consumption, then the process may

be deemed effective in detoxifying the main product. However, caution should be

used when determining the end fate of the by-products, i.e., animal feed. When

traditional processing methods are not sufficient to decontaminate/detoxify the

commodity, then decontamination procedures and/or modified processing must be

considered (Figure 8.2). In these cases, careful evaluation of the procedure must be

performed before considering it as an alternative. It is well known that most

mycotoxins would not remain chemically stable in extreme conditions, i.e., extreme

pH, heat, etc. However, these procedures would not result in a usable food/feed

product. Thus, special considerations must be made when developing

215

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Laboratory Pilot plant Industrial conditions conditions scale processing

Evaluation

Chemical Technological modification Nutritional properties qualities

Toxic/mutagenic potential Sensory characteristics

Figure 8.2 Development of decontamination/detoxification procedures

216

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. decontamination/detoxification procedures. Park and co-workers (1993) have

proposed specific criteria for evaluating the acceptance of a given mycotoxin

reduction or decontamination procedure. In order to be effective, the process must:

1. Inactivate, destroy, or remove the toxin;

2. Not produce or leave toxic residues in the food/feed;

3. Retain nutritive value and food/feed acceptability of the product;

4. Not alter significantly the technological properties of the product;

5. If possible, destroy fungal spores.

Managing the risks associated with mycotoxin contamination should involve

an integrated system. Processed foods cannot be safe if prevention, control, good

manufacturing practices, and quality control are not used at all stages of

production. Figure 8.3 shows selected aspects of an integrated aflatoxin control

program. This can be used as a model system when developing food safety

programs for other mycotoxins or naturally occurring toxicants. Each phase would

have to be adapted according to the toxin and the commodity in question. A Hazard

Analysis Critical Control Point (HACCP) type of approach to processing

mycotoxin-contaminated commodities should be considered. Control parameters

would include timeliness of harvesting, temperature, and moisture during storage

and transportation, selection of agricultural products prior to processing;

processing/decontamination conditions, i.e., temperature, addition of chemicals and

final product storage and transportation. This approach should be developed not

only for specific single toxins, but also for multi - toxin models, i.e.,

217

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Insect m anagem ent

‘Management of crop residues P re -h a rv e s t * I r r Ig a tio n a n d m anagem ent mineral n u tritio n

Crop rotation

Resistant varieties

Tim e lin e s s

H a rv e s tin g Cleanup

D ry in g

Post-H arvest Storage: Moisture m anagem ent and insect control

Processing and manufacturing

Good manufacturing practices

S trie t q u a lity control

Decontamination strategies:

Physical separation Physical decontamination Biological decontamination Chemical inactivation Chem oadsorption

Figure 8.3 Integrated management for mycotoxin control

218

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. aflatoxin Bj/fumonisin Bj. As mentioned before, a food safety management

program for naturally occurring toxicants should involve different phases such as

the ones outlined below:

I. Setting of regulatory limits

II. Establishment of a monitoring program

A. Establishment of a sampling plan

1. Sample collection

2. Sample preparation

3. Analysis

B. Permitted uses of mycotoxin-contaminated products

C. Designation of use for violative products

III. Control through processing

A. Good manufacturing practices

B. Quality control

IV. Decontamination through specific treatments

A. Evaluation of the final product

B. Designation of use of treated product

The establishment of regulatory limits may vary in each country depending

on level of exposure, sociological, political and economic factors. Due to the

increased international trade, it has been suggested that an international standard be

set. However, at present, different regulations are in place. Table 8.1 shows current

regulatory status for aflatoxins in the United States. Currently, mycotoxin

219

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. regulatory data are available for ninety countries. Seventy seven countries are

known to have some regulations on mycotoxins. and thirteen countries are known

to have no regulations. No data are available for about forty countries with more

than five million inhabitants, most of them in Africa (FAO, 1998). Most of the

existing mycotoxin regulations are for aflatoxins (Table 8.2).

Table 8.1Current aflatoxin action levels (total) established by the Food and Drug Administration (Park, 1993) Food or feed Action level (pg/kg, ppb) Human foods (except milk) 20 Milk 0.5 Animal feeds (except as listed below) 20 Cottonseed meal (used for mature beef, swine and poultry rations) 300 Com and peanut meal for breeding beef cattle, breeding swine, or mature poultry too Com and peanut meal for finishing swine 200 Com and peanut meal for feedlot cattle 300

Table 8.2 Medians and ranges in 1995 of maximum tolerated levels (pg/kg) for aflatoxins and numbers of countries that have regulations for these (FAO, 1996). Median Range Number of Countries Bi in foodstuffs 4 0-30 33 B|, Ba, Gi, G2 in foodstuffs 8 0-50 48 Bi in foodstuffs for children 0.3 0-5 5 Mi in milk 0.05 0 - 1 17 Bi in feedstuffs 2 0 5-1000 19 Bi, B2 , Gi, G2 in feedstuffs 50 0 - 1 0 0 0 2 1

Once a regulatory limit has been set, sampling plays an important role in

determining the fate of a particular product. Adequate, random sampling techniques

should be used at each point of analysis. It is important to consider the existence of

220

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. "hot spots" or highly contaminated portions of the product. If these highly

contaminated portions are not detected, the toxin(s) would then be distributed

throughout the lot rendering a highly contaminated product. Once the levels of

contamination have been determined through a monitoring program, the final use

of the commodity can be determined. The product may be deemed either safe for

human consumption, safe for feed use through treatment, or totally unsuitable for

use at any point in the food chain.

A good example of management through a monitoring program is the

aflatoxin control program used by the State of Arizona (United States) (Park,

1993). During the year 1978, almost 910,000 pounds of milk were dumped with

contamination levels as high as 10 ppb aflatoxin Mj. As a result of this huge

commercial loss and the need to establish an effective aflatoxin control program,

the State instituted a program to monitor aflatoxin levels in whole cottonseed and

cottonseed products, the major source of aflatoxin contamination in feed. All

cottonseed produced in the state is tested for aflatoxin content. The maximum size

of the lots tested is 1 0 0 tons, and the testing is conducted in state-certified

laboratories. The end use of the product is dictated by aflatoxin levels found.

Cottonseed lots testing over 20 pg aflatoxin/kg are usually treated with ammonia to

reduce aflatoxin levels. Cottonseed products containing less than 20 ppb aflatoxin,

with or without ammonia decontamination treatment, can be used for dairy rations.

The key aspect of the aflatoxin decontamination program is that the product is

tested for aflatoxin levels before and after the ammonia treatment, as appropriate.

221

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. The use of this program has resulted in a significant benefit in keeping Arizona's

milk supply safe from aflatoxin contamination (Table 8.3).

Table 8.3 Aflatoxin residues in milk produced in Arizona (Park, 1993) ______No. of samples with an AFMi concentration of % of samples containing Year No. of samples 0.2-0.5 pg/1 >0.5 p.g/1 detectable levels of AFMi 1 9 7 9 * 535 140 0 26 1980 972 169 0 17 1981 940 109 0 1 1 1982 802 227 0 28 1989 900 1 2 1 0 13 a:a T_In 1978 i n 909,442 lb of milk with aflatoxin levels as high as 10 p.g/1 were destroyed.

In an integrated mycotoxin management program, each phase would help

prevent the risk of exposure to the toxins. After pre-harvest control, harvesting and

monitoring, most of the contaminated product would have been separated and

destined to different uses or decontamination treatments. However, industrial

processing is still an important phase of control. The final goal of processing would

not only be the decontamination of the commodity, but also the prevention of new

mold colonization with its subsequent mycotoxin production. Thus, when the

product reaches the consumer, the hazards associated with mycotoxin

contamination of food would have been minimized.

222

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. IX. Summary and conclusions

This study explored different aspects of aflatoxin B| and fumonisin Bi co

contamination in com. Special emphasis was placed on potential interactions

between toxin production and toxic/mutagenic potential.

Although no definite pattern to relate the ability of Aspergillus jlavus and

Fusarium moniliforme to produce different toxins and change the toxic/mutagenic

potential of contaminated com could be established, evidence that indicated that

many factors are simultaneously involved in these processes was presented. It was

shown that A. Jlavus and F. moniliforme can co-exist in the same environment and

produce different levels of aflatoxins and fumonisins, respectively. It was also

observed that old mycelia of a toxigenic strain of A. Jlavus can reduce aflatoxin

levels over time, and that F. moniliforme may either inhibit aflatoxin production or

enhance its degradation. Additionally, it was shown that F. moniliforme is capable

of producing unidentified compounds with mutagenic potential. There was

evidence that both mold strains are capable of producing a wide variety of

toxic/mutagenic compounds and that fungal interaction(s) can shift toxin

concentrations over storage time.

More information is needed on the ecological relationships between A.

Jlavus and F. moniliforme and their potential interactive effects on mycotoxin

production and stability. It would be interesting to explore these relationships over

shorter storage periods (<15 days). Further understanding would also be gained by

performing a similar study at different moisture and temperature levels. These

223

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. levels should include storage conditions normally used for com. Results from such

studies would help assess the potential toxin production and degradation in real

conditions to better understand the risks associated with aflatoxin and fumonisin

co-contamination.

The potential involvement of fungal cytochrome P450 enzymes as well as

oxidation enzyme systems as key factors in toxin production and degradation was

exposed. Further research in this area would help understand toxin production and

degradation pathways to develop better mechanisms for biological control.

It was determined that different AFBi/FBi ratios can elicit divergent

responses in the Salmonella/microsomal mutagenicity assay using tester strains

TA-100 and TA-98. Evidence was provided to show that these reactions were

dependent on metabolic activation achieved by addition of S9 mix (rat liver

postmitochondrial supernatant). The results from this study suggested that FBj

affects AFBi mutagenic potential in a dose-dependent fashion. The mutagenic

potential of lxlO 3 ng AFBi/plate was reduced by addition of 1 to 10 2 ng FBj/plate.

However, higher FBi concentrations (lxlO 3 to lx l0 5 ng/plate) exerted an opposite

effect; and mutagenicity was returned to its original level in a dose-response type of

relationship. These complex interaction mechanisms may be attributed to several

factors that may include the direct chemical reaction between both toxins or toxin

byproducts, and/or a mechanistic interaction where FBi could be affecting one or

more of the enzyme systems involved in xenobiotic activation/detoxification

pathways.

224

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. When the effect of FBi on the mutagenic potential of 10 pg 2-

aminofluorene/plate was evaluated, the same response was observed. This provided

further evidence that FBi may affect some of the xenobiotic

activation/detoxification enzymes present in the S9 mix.

The study of the effect of FBi on individual microsomal enzymes, such as

P4501A2, and cytosolic enzymes prostaglandin synthetase, glutathione peroxidase

and glutathione-S-transferase among others, may help clarify the mechanisms

involved in interaction and understand the potential health hazards associated with

simultaneous exposure.

Addition of 1 to lxlO 5 ng FBi to lxlO 3 ng AFBi affected AFBi recovery in

the same fashion as it affected its mutagenic potential. Recovery of AFBi/AFBr

metabolites was decreased by addition of 1 to 10 2 ng FBj. In the lxlO 3 to lxlO 5 ng

FBi range, AFBi/AFBi-metabolites recovery was increased in a dose-dependent

way. These results indicated that the recovered metabolites may have been

responsible for the mutagenic response observed previously. Analysis by thin layer

chromatography showed that the recovered AFBi could have been a metabolite that

co-eluted with AFB 2a-

Further research in this area may help understand the mechanisms involved

in AFB i/FBi interactions. It would be interesting to characterize the metabolites

produced by these reactions. Additionally, a study of the kinetic parameters

involved in these reactions would also help determine the specific mechanisms

involved in these interactions.

225

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. This study also determined that FBi inhibits gap junctional intercellular

communication of Clone-9 rat liver cells in a dose-dependent fashion. These results

provide further information on the possible mechanisms involved in FBi’s

carcinogenicity. Furthermore, it was established that the use of the scrape-loading

dye transfer assay using Clone-9 rat liver cells could be used as a short-term assay

to study FB[. More research is needed to determine if this technique would be

useful for com extract analysis. This technique could be added to the battery of

short-term assays used to monitor the safety of com decontamination treatments.

When AFBi (1 to lxlO3 ng/plate) was added to FBi (1x103), no change in the

inhibition of gap junctional intercellular communication was observed. The

metabolic competency of Clone-9 rat liver cells is unknown. Therefore, although

there is some evidence, it cannot be concluded that that AFBi does not affect this

FBi toxic response.

It was also observed that exposure of clone-9 rat liver cells to FBi and/or

AFBi increased the retention of neutral red. Since this dye is mainly uptaken by

lysosome membranes, these results provided further evidence that peroxisome

proliferation and oxidative metabolism may be involved in the interaction between

these two toxins.

The use of hydrogen peroxide (3%) and sodium bicarbonate (2%) was

proposed as a decontamination treatment for both, aflatoxins and fumonisins. This

study showed that these treatments were effective not only in reducing the toxin

content of naturally contaminated com, but also in reducing toxicity to Artemia sp.

226

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. No mutagenic by-products were created by these treatments as tested by the

Salmonella!microsomal mutagenicity assay using tester strain TA-100 with and

without metabolic activation. Although further evaluation is needed, these

treatments could be a good cost-effective strategy to reduce the hazards associated

with aflatoxin and fumonisin co-contaminated com.

This study showed that simultaneous production of AFBi and FBi is a

dynamic process and exposure to both, AFBi and FBi can change toxic/mutagenic

responses. Furthermore, it was shown that testing two or more compounds changes

the perspective of their activity. Therefore, when risk assessment considerations are

made, potential interactions between co-contaminating agents should be taken into

account. Since aflatoxins and fumonisins are ubiquitous, unavoidable and

unpredictable contaminants of com, a multi-toxin interdisciplinary approach should

be taken not only by research groups, but also by regulatory agencies to better

understand and regulate the potential impact of co-contamination in food safety.

227

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. References

Abbas, H. K., Tanaka, T., Duke, S. O., Porter, J. K., Wray, E. M., Hodges, L., Sessions, A. E., Wang, E., Merrill, A. H. and Riley, R. T. 1994. Fumonisin and AAL-toxin-induced disruption of sphingolipid metabolism with accumulation of free sphingoid bases. Plant Physiol. 106:1085.

Abu-Shakra, A., Ioannides, C. and Walker, R. 1986. Metabolic activation of 2- amino-3-methylimidazo-(4,5-f)-quinoline by hepatic preparations- Contribution of the cytosolic fraction and its significance to strain differences. Mutagenesis. 1:367.

Achroder, H. W., Boiler, R. A. and Hein Jr., H. 1968. Reduction in aflatoxin contamination of rice by milling procedures. Cereal Chem. 45:574.

Alberts, J. F., Gelderblom, W. C. A., Thiel, P. G., Marasas, D. J., van Schalwyk, D. J. and Behrend, Y. 1990. Effects of temperature and incubation period on production of fumonisin Bi by Fusarium moniliforme. Appl. Environ. Microbiol. 56:1729.

Anderson, H. W., Nehrig, E. W. and Wichser, W. R. 1975. Aflatoxin contamination of com in the field. J. Agric. Food Chem. 23:775.

AOAC. 1996. Official method 995.15: Fumonisins Bi, B 2 , and B 3 in com: Liquid Chromatographic Method. AOAC International March Supplement. Chapter 49:49.

Aoyama, T., Yamano, S., Guzelian, P. S., Gelboin, H. V. and Gonzalez, F. J. 1990. 5 of 12 forms of vaccinia virus-expressed human hepatic cytochrome P450 metabolically activate aflatoxin Bj. Proc. Natl. Acad. Sci. USA 87:4790.

Applebaum , R. S. and Marth, E. H. Biogenesis of the C 20 polyketide, aflatoxin. 1981. Mycopathologia. 76:103.

Applebaum, R. S. and Marth, E. H. 1980. Inactivation of aflatoxin Mi by hydrogen peroxide. J. Food Prot. 43:820.

Appleton, B. S. and Campbell, T. C. 1983. Effect of high and low dietary protein on the dosing and postdosing periods of aflatoxin Bi-induced hepatic preneoplastic lesion development in the rat. Cancer Res. 43:2150.

Armolik, N. and Dickson, J. G. 1956. Minimum humidity requirements for germination of conidia associated with storage of grain. Phytopathology 46:462.

228

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Asao, T., Buchi, G., Abdel-Kader, M. M.. Chang, S. B., Wick, E. L. and Wogan, G. N. (1963). Aflatoxins B and G. J. Am. Chem. Soc. 85:1706.

Ashby, J. and Tennant, R. W. 1991. Definitive relationships among chemical structure, carcinogenicity and mutagenicity for 301 chemicals tested by the US NTP. Mutation Res. 257:259.

Ashendel, C. L. 1985. The phorbol ester receptor: A phospholipid-regulated protein kinase. Biochim. Biophys. Acta. 822:219.

Asplin, F. D. and Camaghan, R. B. A. 1961. The toxicity of certain groundnut meals for poultry with special reference to their effect on ducklings and chickens. Vet. Rec. 73:1215.

Ashworth, L. J., McMeans, J. L. and Brown, C. M. 1969. Infection of cotton by A. flavus; Regulatory aspects of the S and L problem. Mycol. Soc. Amer. Newslett. 42:5.

Ayerst, G. 1969. The effects of moisture and temperature on growth and spore germination in some fungi. J. Stored Prod. Res. 5:669.

Ayrton, A. D., Smith, J. N. and Ioannides, C. 1987. Bioactivation of N- nitrosopiperidine to mutagens: role of hepatic cytochrome P450 proteins and contribution of cytosolic fraction. Carcinogenesis. 8:1691.

Bacon, C. W. and Nelson, P. E. 1994. Fumonisin production in com by toxigenic strains of Fusarium moniliforme and Fusarium proliferatum. J. Food Prot. 57(6):514.

Bacon C. W., Bennett, R. M, Hinton, D. M. and Voss, K. A. 1992. Scanning electron microscopy of Fusarium moniliforme within asymptomatic com kernels associated with equine leukoencephalomalacia. Plant Disease. 76(2): 144.

Baertschi, S. W., Raney, K. D., Shimada, T., Harris, T. M. and Guengerich, F. P. 1989. Comparison of rates of enzymatic oxidation of aflatoxin Bj, G| and sterigmatocystin and activities of the epoxides in forming guanyl-N7adducts and inducing different genetic responses. Chem. Res. Toxicol. 2:114.

Bane, D. P., Newmann, E. J., Hall, W. F., Harlin, K. S. and Slite, R. L. N. 1992. Relationship between fumonisin contamination of feed and mystery swine disease. A case control study. Mycopathol. 117:121.

229

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Bane, D. P., Newmann, E. J. and Zuckermann, F. A. 1991. Effects of fumonisin contaminated screenings on reproduction and immune function in swine. Symposium. Raleigh, N. C. April 24-25.

Battista, J. R. and Mamett, L. J. 1985. Prostaglandin H syntethase-dependent epoxidation of aflatoxin Bj. Carcinogenesis. 6:1227.

Baur, F. J. 1975. Effect of storage upon aflatoxin levels in peanut materials. J. Am. Oil Chem. Soc. 52:263.

Begin, M. E., Ells, G. and Horribin, D. F. 1992. Polyunsaturated fatty acid-induced cytotoxicity against tumor cells and its relationship to lipid peroxidation. J. Natl. Cancer Inst. 80:188.

Bell, R. M., Hannun, Y. A. and Merrill, A. H. 1993. Advances in Lipid Research: Sphingolipids and Their Metabolites, Academic Press, Orlando, Florida, volumes 25-26.

Bennett, G. A. and Anderson, R. A. 1978. Distribution of aflatoxin and/or zearalenone in wet-milled com products: A review. J. Agr. Food Chem. 26:1055.

Bennett, G. A. and Richard, J. L. 1997. Influence of processing on Fusarium mycotoxins in contaminated grains. Food Technol. 50(5):235.

Bennett, G. A., Rottinhaus, G. and Nelson, T. L. 1992. Deoxynivalenol in wheat and flour from wheat cleaned by aspiration. Proceedings of Am. Bakers Assoc. ARS Workshop, New Orleans, Louisiana.

Bennett, G. A., Lagoda, A. A., Shotwell, O. L. and Hesseltine, C. W. 1981. Utilization of zearalenone-contaminated com for ethanol production. J. A. Oil Chem. Soc. 58:974.

Bennett, G. A., Vandegraft, E. E., Shotwell, O. L., Watson, S. D. and Bacon, B. J. 1978. Zearalenone: Distribution in wet-milling fractions from contaminated com. Cereal Chem. 55:455.

Bennett, G. A., Peplinski, A. J., Brekke, O. L. and Jackson L. K. 1976. Zearalenone: Distribuiton in dry-milled fractions of contaminated com. Cereal Chem. 53:299.

Bennett, M. V. L., Barrio, L. C., Bargiello, T. A. and Spray, D. C. 1991. Gap junctions: new tools, new answers, new questions. Nueron. 6:305.

230

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Benzuidenhout, S. C., Gelderblom, W. C. A.. Gorst-Allman. C. P., Horak, R. M., Marasas, W. F. O., Spiteller, G. and Vleggaar. R. 1988. Structure elucidation of the fumonisins, mycotoxins from Fusarium moniliforme. J. Chem. Soc. Chem. Commun. 743.

Beuchat, L. R. 1978. Microbial alterations of grains, legumes and oilseeds. Food Technol. 32(5): 193.

Beyer, E. C. 1993. Gap junctions. Int. Rev. Cytol. 137C:1.

Bhatnagar, D., Cleveland, T. E. and Cotty, P. E. 1994. Mycological aspects of Aflatoxin Formation. In: The Toxicology of Aflatoxins: Human Health, Veterinary and Agricultural Significance. Eaton, D. L. and Groopman, J. D. (Eds.). Academic Press, San Diego, California, USA.

Bhatnagar, D. Ehrilch, K. C. and Cleveland, T. E. 1991. Oxidation-reduction reactions in biosynthesis of secondary metabolites. In: Mycotoxins in Ecological Systems. Bhatnagar, D„ Lillehoj, E. B. and Arora, D. K. (Eds.). Marcel Dekker, New York, USA.

Blount, W. P. 1961. Turkey “X” disease. J. Brit. Turkey Fed. 9:52.

Blumberg, P. M., Jaken, S., Konig, B., Sharkey, N. A., Leach, K. L., Jeng, A. Y. and Yeh, E. 1984. Mechanism of action of the phorbol ester tumor promoters: Specific receptors for lipophilic ligands. Biochem, Pharmacol. 33:933.

Borenfreund, E. and Puemer, J. A. 1984. Toxicity determined in vitro by morphological alterations and neutral red absorption. Toxicol. Lett. 24:119.

Bottalico, A., Logrieco, A., and Visconti, A. 1989. Fusarium - mycotoxins, taxonomy and pathogenesis. Elsevier. Amsterdam, The Netherlands.

Brackett, R. E. and Marth, E. H. 1979. Ascorbic acid and ascorbate cause disappearance of patulin from buffer solutions and apple juice. J. Food Prot. 42:864.

Bradshaw, R. A. and Prentis, S. 1987. Oncogenes and Growth Factors. Elsevier, Amsterdam, The Netherlands.

Bressac, B., Kew, M., Wands, J. and Oztruk, M. 1991. Selective G to T mutations of the p53 gene in hepatocellular carcinoma from South Africa. Nature. 350:429.

231

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Brown, T. P., Manning, R. O., Fletcher, O. J. and Wyatt, R. D. 1986. The individual and combined effects of citrinin and ochratoxin A on renal ultrastructure in layer chicks. Avian Dis. 30:191.

Brownie, C. F. and Brownie, C. 1988. Preliminary study on serum enzyme changes in Long Evans rats given parenteral ochratoxin A, aflatoxin Bj and their combination. Vet. Human Toxicol. 30:211.

Bullerman, L. B., Draughon, F. A. 1994. Fusarium moniliforme and fumonisin symposium introduction. J. Food. Prot. 57:513.

Bullerman, L. B. and Tsai, W.-Y.J. 1994. Incidence and levels of Fusarium moniliforme, Fusarium proliferatum and fumonisins in com and com-based foods and feeds. J. Food Prot. 57: 541.

Bullerman, L. B. 1986. Mycotoxins and Food Safety. Food Technol. 40(5):59.

Burgess, L. W., Dodman, R. L., Pont, W. and Mayers, P. 1981. Fusarium diseases of wheat, maize and grain sorghum in Eastern Australia. In: Fusarium: Diseased Biology and Taxonomy. P. E. Nelson, T. A. Toussoun and R. J. Cook (Eds.). Pennsylvania State University Press. University Park, Pennsylvania.

Burgess, L. W., Summerell, B. A., Bullock, S., Gott, K. P. and Backhouse, D. 1994. Laboratory manual for Fusarium Research, 3 rd ed. University of Sydney, Sydney, Australia.

Burgos-Hemandez, A. 1998. Evaluation of chemical treatments and intrinsic factors on the mutagenic potential of aflatoxin-contaminated com. A Dissertation submitted to Louisiana State University for the Degree of Doctor of Philosophy.

Burroughs, L. F. 1977. Stability of patulin to sulfur dioxide and to yeast fermentation. J. AO AC 60:100.

Butler, T. 1902. Notes on a feeding experiment to produce leucoencephalitis in a horse with positive results. Amer. Vet. Rev. 26:748.

Busby, W. F. and Wogan, G. N. 1984. Aflatoxins. In: Chemical Carcinogens. Searle, C. E. (Ed.). Amer. Chem. Soc., Washington, D. C., USA.

Canfield, M. A., Annegers, J. F. and Brender, J. D. 1996. Hispanic origin and neural-tube defects in Houston/Harris County Texas. II. Risk factors. Am. J. Epidemiol, 143:12.

232

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Cassee, F. R., Groten, J. P, van Bladeren, J. and Feron, V. J. 1998. Toxicological evaluation and risk assessment of chemical mixtures. Critical Reviews in Toxicology. 28(1 ):73.

Cawood, N. E., Gelderblom, W. C. A., Alberts, J. E. and Snyman, S. D. 1994. Interaction of l4C-Iabelled fumonisin B mycotoxins with primary hepatocytes cultures. Fd. Chem. Toxicol. 32:627.

CAST. 1989. Mycotoxins: Economic and health risks. Report 116. Council for Agricultural Science and Technology. Ames, IA.

Chamberlain, W. J., Bacon, C. W., Norred, W. P. and Voss, K. A. 1993. Levels of fumonisin Bi in com naturally contaminated with aflatoxins. Food Chem. Toxic. 31(12):995.

Chen, J. P., Xie, W. P. and Mirocha, C. J. 1992. Production of the mycotoxin fumonisin B| by Alternaria alternata f sp. lycopersici. Phytopathol. 82.

Chinese Birth Defects Monitoring Program. 1990. Central nervous system congenital malformations, especially neural tube defects in 29 provinces, metropolitan cities and autonomous regions of China. Int. J. Epidemiol. 19:978.

Christensen, C. M., Mirocha, C. J. and Meronuck, R. A. 1977. Mold, mycotoxin and mycotoxicoses. Agricultural Experiment Station, Report No. 142. University of Minnesota, St. Paul, Minnesota, USA.

Chu, F. S. and Li, G. Y. 1994. Simultaneous occurrence of fumonisin Bi and other mycotoxins in moldy maize collected from People’s Republic of China in regions with high incidence of esophageal cancer. Appl. Environ. Microbiol. 60:847.

Chu, F. S., Chang, C. C., Ashoor, S. H. and Prentice, N. 1975. Stability of aflatoxin Bi and ochratoxin A in brewing. Appl. Microbiol. 29:313.

Cole, R. J. 1989. Technology of aflatoxin decontamination. In: Mycotoxins and Phycotoxins’88. Natori, S. Hashimoto, K., and Ueno, Y. (Eds.), Elsevier Scientific. Amsterdam. The Netherlands.

Cole, R. J., Hill, R. A., Blankenship, P. D., Sanders, T. H., and Garren, K. H. 1982. Influence of irrigation and drought stress on invasion by Aspergillus flavus of com kernels and peanut pods. Dev. Ind. Microbio. 23:229.

233

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Cole, R. J., Kirskey, J. W., Cutler, H. G., Doupnik, B. L. and Peckham, J. C. 1973. Toxin and Fusarium moniliforme: Effects on plants and animals. Science 179:1324.

Colvin, B. M. and Harrison, L. R. 1992. Fumonisin-induced pulmonary edema and hydrotorax in swine. Mycopathol. 117:79.

Conway, H. F., Anderson, R. A. and Bagley, E. B. 1978.. Detoxification of aflatoxin-contaminated com by roasting. Cereal Chem. 55:115.

Cook, R. J. 1981. Fusarium - Disease, biology and taxonomy. Pennsylvania State University Press. University Park, Pennsylvania.

Corbett, M. D. and Corbett, B. R. Nucleic acid binding of arylamines during the respiratory burst of human granulocytes. Chem. Res. Toxicol. 1:356.

Cornell, M. 1998. Talkin’ about tortillas. Baking and Snack.4:37.

Cotty, P. J. 1990. Effect of atoxigenic strains of Aspergillus flavus on aflatoxin contamination of developing cottonseed. Plant Dis. 74:233.

Cotty, P. J. 1989. Virulence and cultural characteristics of two Aspergillus flavus strains pathogenic on cotton. Phytopathol. 79:808.

Creizel, A. E. and Dudas, I. 1992. Prevention of the first occurrence of neural-tube defects by periconceptual vitamin supplementation. N. Engl. J. Med. 327:1832.

Cullen, J. M. and Newbeme, P. M. 1994. Acute hepatotoxicity of aflatoxins. In: The Toxicology of Aflatoxins: Human Health, Veterinary and Agricultural Significance. Eaton, D. L. and Groopman, J. D. (Eds.). Academic Press, Orlando, Florida, USA.

Dam, R. S., Tam, S. W. and Satterlee, L. D. 1977. Destruction of aflatoxins during fermentation and by-product isolation from artificially contaminated grains. Cereal Chem. 54:705.

D’Angelo, A. and Sehlub, J. 1997. Homocysteine and thrombotic disease. Blood. 90:1.

Dantzer, W. R., Lu, Z., Hopmans, E. C., Prisk, V., Cunnick, J. E., Murphy, P. A. and Hendrich, S. 1997. Reaction with fructose detoxifies fumonisin Bi while stimulating liver-associated natural killer cell activity in rats. J. Agric. Food Chem.45(3):803.

234

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Darnell. J., Lodish. H. and Baltimor, D. 1990. Molecular Cell Biology. Second Edition. Scientific American Books, New York, USA.

Davidson, J. S., Baumgarten, I. M. and Harley, E. H. 1985. Studies on the mechanism of phorbol ester-induced inhibition of intercellular junctional communication. Carcinogenesis. 6:1353.

Dawarkanath, C. T., Sreeniwasamurty, V. and Parpia, H. A. B. 1969. Aflatoxin in Indian peanut oil. J. Food Sci Technol. (Mysore)6:107.

Decker, W. J. 1980. Activated charcoal absorbs aflatoxin Bj. Vet. Human Toxicol. 22:388.

Desjardins, A. E., Plattner, R. D. and Nelson. P. E. 1997. Production of fumonisin Bi and moniliformin by Giberella fujikori from rice from various geographic areas. Appl. and Env. Microbiol. 63(5): 1838.

Dickens, J. W. and Whitaker, T. B. 1975. Efficacy of electronic color sorting and hand picking to remove aflatoxin contaminated kernels from commercial lots of shelled peanuts. Peanut Sci. 2:45.

Doko, M. B. and Visconti, A. 1994. Occurrence of fumonisins Bi and B? in corn- based products from the Swiss market. J. Agric. Food Chem. 40:1352.

Domsch, K. H., Gams, W., and Anderson, T.-H. 1980. Compendium of Soil Fungi, 2 vols. Academic Press. London.

Domer, J. W., Cole, R. J., Lomax, L. G., Gosser, H. S. and Diener, U. L. 1983. Cyclopiazonic acid production by Aspergillus flavus and its effects on broiler chickens. Appl. Environ. Microbiol. 46:698.

Doyle, M. P., Applebaum, R. S., Brackett, R. E. and Marth, E. H. 1982. Physical, chemical and biological degradation of mycotoxins in foods and agricultural commodities. J. Food Prot. 45(10):964.

Doyle, M. P. and Marth, E. H. 1979. Peroxidase activity in mycelia of Aspergillus parasiticus that degrade aflatoxin, Europ. J. Appl. Microbiol. Biotechnol. 7:211.

Doyle, M. P. and Marth, E. H. 1978a. Aflatoxin at several initial concentrations is degraded by different amounts of mycelium of Aspergillus parasiticus. Z. Lebens. Unters.-Forsch. 166:359.

235

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Doyle, M. P. and Marth, E. H. 1978b. Aflatoxin is degraded at different temperatures and pH values by mycelia of Aspergillus parasiticus grown 5 to 18 days with and without agitation. J. Food Prot. 41:549.

Doyle, M. P. and Marth, E. H. 1978c. Aflatoxin is degraded by mycelia from toxigenic and nontoxigenic strains of aspergilli grown on different substrates. Mycopathologia. 63:145.

Doyle, M. P. and Marth, E. H. 1978d. Bisulfite degrades aflatoxins: Effect of citric acid and methanol and possible mechanism of degradation. J. Food Prot. 41:774.

Doyle M. P. and Marth, E. H. 1978e. Bisulfite degrades aflatoxins: Effect of temperature and concentration of bisulfite. J. Food Prot. 41:891.

Doyle, M. P. and Marth, E. H. 1978f. Degradation of aflatoxin by Iactoperoxidase. Z. Lebensm. Unters.-Forsch. 166:271.

Doyle M. P. and Marth, E. H. 1978g. Peroxidase activity in mycelia of Aspergillus parasiticus that degrade aflatoxin. Europ. J. Appl. Microbiol. Biotechnol. 7:211.

Drew, S. W. and Demain, A. L. 1977. Effect of primary metabolites on secondary metabolism. Ann. Rev. Microbiol. 31:343.

Dupuy, J., LeBars, P., Boudra, H., and LeBars, J. 1993. Thermostability of fumonisin Bi, a mycotoxin from Fusarium moniliforme in com. Appl. Environ. Microbiol. 59:2864.

El-Banna, A. A., Lau, P.-Y., and Scott, P. M. 1983. Fate of mycotoxins during processing of foodstuffs. II. Deoxynivalenol (vomitoxin) during making of Egyptian bread. J. Food Prot. 46:484.

El-Fouly, M. H., Trosko, J. E. and Chang, C. C. 1987. Scrape-loading and dye transfer: A rapid and simple technique to study gap junctional intercellular communication. Exp. Cell Res. 168:422.

Enomoto, T. and Yamasaki, H. 1985. Rapid inhibition of intercellular communication between BALB c3T3 cells by diacylglycerol, a possible endogenous functional analogue of phorbol esters. Cancer Res. 45:3706. Enomoto, T., Sasaki, Y., Shiba, Y., Kanno, Y., and Yamasaki, H. 1981. Tumor promoters can cause a rapid and reversible inhibition of the formation and maintenance of electrical cell coupling in culture. Proc. Natl. Acad. Sci. USA. 78:5628.

236

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Essigmann. J. M., Croy, R. G., Nazdan, A. M., Busby, W. F., Reinhold, V. N., Buchi, G., and Wogan, G. N. 1977. Structural identification of the major DNA adduct formed by aflatoxin Bi in vitro. Proc. Natl. Acad. Sci. USA. 74:1870.

Fearon, E. R. and Vogelstein, B. 1990 A genetic model for colorectal tumorigenesis. Cell. 61(5):759.

Ferguson, S. A., St. Omer, V. E. V., Kwon, O. S., Holson, R. R., Houston, R. J., Rottinghaus, G. E. and Slikker, W. 1997. Prenatal fumonisin (FBt) treatment in rats results in minimal maternal or offspring toxicity. Neurotoxicology 18(2):561.

Feuell, A. J. 1966. Aflatoxin in groundnuts. IX. Problems of detoxification. Trop. Sci. 8:61.

Fincham, J. E. 1992. Atherogenic effects in a nonhuman primate of Fusarium moniliforme cultures added to a carbohydrate-diet. Atheroescl. 94(1): 13.

Firozi, P. F., Aboobaker, V. S. and Bhattacharya, R. K. 1986. Modulation by certain factors of metabolic activation of aflatoxin Bi as detected in vitro in a simple fluorimetric assay. Chemical Biological Interactions 59:173.

Floss, J. L., Casteel, S. W., Johnson, G. C., Rottinhaus, G. E. and Krause, G. F. 1994a. Developmental toxicity in hamsters of an aqueous extract of Fusarium moniliforme culture material containing known quantities of fumonisin Bj. Vet. Hum. Toxicol. 36:5.

Floss, J. L., Casteel, S. W., Johnson, G. C., Rottinghaus, G. E. and Krause, G. F. 1994b. Developmental toxicity of fumonisin in Syrian hamsters. Mycopathologia. 128:33.

Flynn, T. J., Pritchard, D., Bradlaw, J. A., Eppley, R. and Page, S. 1996. In vitro embryotoxicity of fumonisin Bi evaluated with cultured postimplantation staged rat embryos. In vitro Toxicol. 9:271.

Food and Agriculture Organization of the United Nations. 1998. Worldwide regulations for mycotoxins. A compendium. FAO food and Nutrition Paper. In Press. Food and Agriculture Organization of the United Nations. 1996. Basic facts of the world cereal situation. Food Outlook. No. 5/6.

237

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Forrester. L. M., Neal, G. E., Judah, D. J.. Glancey, M. J. and Wolf, C. R. 1990. Evidence for involvement of multiple forms of cytochrome P-450 in aflatoxin Bi metabolism in human liver. Proc. Natl. Acad.Sci. USA. 87:8306.

Forster, R., Green, M. H. L. and Priestley, A. 1981. Enhancement of S9 activation by SI05 cytosolic fraction. Carcinogenesis. 2:1081.

Fortnum, B. A. and Manwiller, A. 1985. Effects of irrigation and kernel injury on aflatoxin Bi production in selected maize hybrids. Plant Dis. 69:262.

Foster, P. L., Eisenstadt, E. and Miller, J. H. 1983. Base substitution induced by metabolically activated aflatoxin Bj. Proc. Natl. Acad. Sci. USA. 80:2695.

Fraumeni, J. F. and Blot, W. J. 1977. Geographic variation in esophageal cancer mortality in the United States. J. Chron. Dis. 30:759.

Furtado, R. M., Pearson, A. M., Gray, J. I., Hogberg, M. G., and Miller, E. R. 1981. Effects of cooking and/or processing upon levels of aflatoxins in meat from pigs fed a contaminated diet. J. Food Sci. 46:1306.

Gainer, H. St. C. and Murray, A. W. 1985. Diacylglycerol inhibits gap junctional communication in cultured epidermal cells: Evidence for a role of protein kinase C. Biochem. Biophys. Res. Commun. 126:1109.

Gallagher, E. P., Kunze, K. L., Stapleton, P. L. and Eaton, D. L. 1996. The kinetics of aflatoxin Bi oxidation by human cDNA-expressed and human liver microsomal cytochromes P450 1A2 and 3A4. Toxicol. Appl. Pharmacol. 141:595.

Gallagher, E. P., Wienkers, L. C., Kunze, K. L. Stapleton, P. J. and Eaton, D. L. 1994. Role of CYP1A2 and 3A4 in the bioactivation of aflatoxin Bi in human liver microsomes. Cancer Res. 54:101.

Gallagher, R. T., Richard, J. L., Stahr, H. M., and Cole, R. J. 1978. Cyclopiazonic acid production by aflatoxigenic and nonaflatoxigenic strains of Aspergillus flavus. Mycopathologia 66:31.

Gamer, R. C., Miller, E. C. and Miller, J. A. 1972. Liver microsomal metabolism of aflatoxin Bi to a reactive derivative toxic to Salmonella typhimiurium TA 1530. Cancer Res. 32:2058.

238

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Gavino, V. C., Miller, J. S., Ikharebha, S. O., Milo, G. E. and Maizewell, D. G. 1981. Effect of polyunsaturated fatty acids and antioxidants on lipid peroxidation in tissue cultures. J. Lipid Res. 22:763.

Gelderblom, W. C. A., Smuts, C. M., Abel, S., Snyman, S. D., van der Westhuizen, L., Huber, W. W. and Swanevelder, S. 1997. Effect of fumonisin Bi on the levels and fatty acid composition of selected lipids in rat liver in vivo. Food and Chem. Toxicol. 35:647.

Gelderblom, W. C. A., Snyman, S. D., Abel, S., Lebepe-Mazur, S., Smuts, C. M., Van der Westhuizen, L., Marasas, W. F. O., Victor, T. C., Knasmuller, S. and Huber, W. 1996. Hepatotoxicity and carcinogenicity of the fumonisins in rats: A review regarding mechanistic implications for establishing risk in humans. In: Fumonisins in Food. Jackson, L. S., DeVries, J. W. and Bullerman, L. B. (Eds.). Plenum Press, New York, USA.

Gelderblom, W. C. A., Snyman, S. D., van der Westhuizen, L. and Marasas, W. F. O. 1995. Mitoinhibitory effect of fumonisin Bi on rat hepatocytes in primary culture. Carcinogenesis. 16:625.

Gelderblom, W. C. A., Cawood, M. E., Snyman, S. D. and Marasas, W. F. O. 1994. Fumonisin B| dosimetry in relation to cancer initiation in rat liver. Carcinogenesis. 15:209.

Gelderblom, W. C. A., Cawood, M. E., Snyman, D., Vleggaar, R. and Marasas, W. F. O. 1993. Structure-activity relationships of fumonisins in short-term carcinogenesis and cytotoxicity assays. Food Chem. Toxicol. 31:407.

Gelderblom, W. C. A., Semple, E. and Farver, E. 1992. The cancer initiating potential of the fumonisin mycotoxins produced by Fusarium moniliforme. Carcinogenesis. 13:433.

Gelderblom, W. C. A., Kriek, N. P. J., Marasas, W. F.O. and Thiel, P. G. 1991. Toxicity and carcinogenicity of the Fusarium moniliforme metabolite, fumonisin Bi, in rats. Carcinogenesis. 12:1247.

Gelderblom, W. C. A., Jaskiewicz, K., Marasas, W. F. O., Thiel, P. G., Horak, R. M., Vleggaar, R. and Kriek, N. P. J. 1989. Fumonsins-novel mycotoxins with cancer-promoting activity produced by Fusarium moniliforme. Appl. and Environ. Microbiol. 54(7): 1806.

239

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Gelderblom, W. C. A., Jaskewicz, K., Marasas. W. F. O.. Thiel, P. G., Horak, R. M., Vleggaar, R. and Kriek, N. P. J. 1988. Cancer promotion potential of different strains of Fusarium moniliforme in a short-term cancer initiation/promotion assay. Carcinogenesis. 9:1405.

Gerschenson, L. E. and Thompson, E. B. 1974. Gene Expression and carcinogenesis in cultured liver. Academic Press, New York, USA.

Giroir, L. E., Huff, W. E., Kubena, R. B., Harvey, R. B., Elissalde, M. H., Witzel, D. A., Yersin. A. G. and Ivie, G. W. 1991. The individual and combined toxicity of kojic acid and aflatoxin in broiler chickens. Poult. Sci. 70:1351.

Glahn, R. P., Wideman, R. F., Envangelisti, J. W. and Huff, W. E. 1988. Effects of ochratoxin A alone and in combination with citrinin on kidney function of single comb white Leghorn pullets. Poult. Sci. 67:1034.

Goodenough, D. A., Paul, D. L. and Jesaitis, L. 1988. Topological distribution of two connexin 32 antigenic sites in intact and split rodent hepatocyte gap junctions. J. Cell Biol. 107:1817.

Gracian, J. and Arevalo G. 1980. Presencia de aflatoxinas en los productos del olivar. Grasas y Aceites. 31:167.

Gross, S. M., Reddy, R. V., Rottinhaus, G. E., Johnson, G. and Reddy, C. S. 1994. Developmental effects of fumonisin B| containing Fusarium moniliforme culture extractin CD 1 mice. Mycopathologia. 123:171.

Guenguerich, F. P., Ueng, Y.-F., Kim, B.-R., Langouet, S., Coles, B., Iyerm R. S., Their, R., Harris, T. M., Shimada, T., Yamazaki, H., Ketterer, B. and Guillouzo, A. 1996. Activation of toxic chemicals by cytochrome P450 enzymes. /^. Biological Reactive Intermediates V. Snyder R. (Ed.) Plenum Press, New York, USA.

Gumprecht, L. 1994. Fumonisin Bi linked to kidney damage in rabbits. Food Chem. News. 36(5):26.

Hagler, Jr., W. M., Hutchins, J. E. and Hamilton, P. B. 1982. Destruction of aflatoxin in com and sodium bisulfite. J. Food Protect. 45:1287.

Hagmann, W., Steffan, A. M., Kim, A. and Keppler, D. 1987. Leukotrienes as mediators in frog vims 3-induced hepatitis in rats. Hepatology. 7:732.

Haliburton, J. C. and Buck, W. B. 1986. Equine Ieucoencephalomalacia: An historical review. Vet. Med. Anim. Sc. 33:75.

240

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Hamid, A. B. and Smith. J. E. 1987. Degradation of aflatoxin by Aspergillus flavus. J. General Microbiol. 133:2023.

Hammons, G. J., Guenguerich, F. P., Weiss, C. C., Beland, F. A. and Kadlubar, F. F. 1985. Metabolic oxidation of carcinogenic arylamines by rat, dog and human hepatic microsomes and by purified flavin-containing and cytochrome P450 monooxygenases. Cancer Res. 45:3578.

Hannun, Y., Loomis, C. R., Merrill, A. H. and Bell, R. M. 1986. Sphingosine inhibition of protein kinase C activity and of phorbol dibutyrate binding in vitro and in human platelets. J. Biol. Chem. 261:12604.

Harrison, L. R., Colvin, B. M., Greene, J. T., Newman, L. E. and Cole, J. R. 1990. Pulmonary edema and hydro thorax in swine produced by fumonisin B|, a toxic metabolite of Fusarium moniliforme. J. Vet. Diagn. Invest. 2:217.

Harvey, R. B., Kubena, L. F., Huff, W. E., Corrier, D. E. and Rottinghaus, G. E. 1990. Effects of treatment of growing swine with aflatoxin and T-2 toxin. Am. J. Vet. Res. 51:1688.

Harvey, R. B., Huff, W. E., Kubena, L. F. and T. D. Phillips. 1989. Evaluation of diets co-contaminated with aflatoxin and ochratoxin fed to growing pigs. Am. J. Vet. Res. 50:1400.

Harwig, J., Chen, Y. K. and Collins-Thompson, D. L. 1974. Stability of ochratoxin A in beans during canning. Can. Inst. Food Sci. Technol. J. 7:288.

Harwig, J., Scott, P. M., Kennedu, B. P. C. and Chen, Y.-K. 1973. Disappearance of patulin from apple juice fermented by Saccharomyces sp. Can. Inst. Food Sci. Technol. J. 6:45.

Haschek, W. M., Motelin, G., Ness, D. K., Harlin, K. S., Hall, W. F., Vesonder, R. F., Peterson, R. E. and Beasley, V. R. 1992. Characterization of fumonisin toxicity in orally and intravenously dosed. Swine. Mycopathol. 117:83.

Haschek, W. M. and Haliburton, J. C. 1986. Diagnosis of mycotoxicoses. Symposium. Ames, Iowa. October 11-13.

Henrich, S., Miller, K. A., Wilson, T. A., and Murphy, P. A. 1993. Toxicity of Fusarium proliferatum - fermented nixtamalized corn-based diets fed to rats: Effects on nutritional status. J. Agric. Food Chem. 41:1649.

Hermanson, G. T. 1996. Bioconjugate Techniques. Academic Press, San Diego, California, USA.

241

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Ho, T. A., Coutts, T. M., Rowland, I. R. and Allrick, A. J. 1992. Inhibition of the metabolism of mutagens occurring in food by arachidonic acid. Mutation Research. 269:279.

Holden, P. R., McGuire, B. Stoler, A., Balmain, A. and Pitts, J. D. 1997. Changes in gap junctional intercellular communication in mouse skin carcinogenesis. Carcinogenesis. 18(1): 15.

Hsieh, D. P. H. and Wong, J. J. 1994. Pharmacokinetics and Excretion of Aflatoxins. In: The Toxicology of Aflatoxins: Human Health, Veterinary and Agricultural Significance. Eaton, D. L. and Groopman J. D. (Eds.) Academic Press, San Diego, California, USA.

Huff, W. E., Kubena, L. F., Harvey, R. B., Hagler, W. M., Swanson, S. P., Phillips, T. D. and Creger, C. R. 1986. Individual and combined effects of aflatoxin and deoxynivalenol (DON, vomitoxin) in broiler chickens. Poult. Sci. 66:2351.

Hughes, H., Farhood, A. and Jaeschke, H. 1992. Role of leukotriene B 4 in the pathogenesis of hepatic ischemia-reperfusion injury in the rat. Prostaglandins, Leukotrienes and Essential Fatty Acids. 45:113.

Hyun, V. L. and Lloyd, A. B. 1984. Synthesis and degradation of aflatoxins by Aspergillus parasiticus. I. Synthesis of aflatoxin Bi by young mycelium and its subsequent degradation in aging mycelium. Aust. J. Biol. Sci. 37:1

Hyun, V. L., Gerdes, R. G. and Lloyd, A. B. 1984. Synthesis an degradation of aflatoxin by Aspergillus parasiticus. II. Comparative toxicity and mutagenicity and its breakdown products. Aust. J. Biol. Sci. 37(3): 123.

IARC (International Agency for Research in Cancer). 1987. IARC monograph on the evaluation of carcinogenic risk to humans. IARC, Lyon, France. Suppl. 7:83.

ICMSF (International Commission on Microbiological Specifications for Foods). 1996. Toxigenic fungi: Aspergillus. In: Microorganisms in Foods. 5. Characteristics of Food Pathogens. Blackie Academic and Professional. London.

Isola, V. J., Trumble, S. J. Ruzek, M. and Gentile, J. M. 1990. Activation of 2- aminofluorene into a Salmonella mutagen by human leukocytes. Environ. Mol. Mutagen. 15(Suppl.-17):27.

242

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Jackson, L. S., Hlywka, J. J, Kannaki, K. R. and Bullerman, L. B. 1996. Effects of thermal processing on the stability of fumonisins. In: Fumonisins in Food. Jacson, L. S., DeVries, J. W., and Bullerman, L. B. (Eds.). Plenum Press, New York, USA.

Jackson, L. S., Hlywka, J. J., Senthil, K. R., Bullerman, L. B. and Musser, S. M. 1995. The effects of time, temperature and pH in the stability of fumonisin Bi in an aqueous model system. J. Agric. Food Chem. 1995.

Jardine, D. J., Leslie, J. F. 1992. Aggressiveness of Giberella fujikori (Fusarium moniliforme) isolates to grain sorghum under green-house conditions. Phytopathol. 76:897.

Jaskiewicz, K., van Rensberg, S. J., Marasas, W. F. O. and Gelderblom, W. C. A. 1987. Carcinogenicity of Fusarium moniliforme culture material in rats. JNCI. 78:321.

Javed, T., Bennett, G. A., Richard, J. L., Dombrink-Kurtzman, M. A., Cote, L. M. and Buck, W. B. 1993. Mortality in broiler chicks on feed amended with Fusarium proliferatum culture material or with purified fumonisin Bi and moniliformin.

Johnson, P.. Smith, E. E. and Phillips, T. D. 1993. The effects of fumonisin Bi on rat embryos in culture. Toxicologist. 13:257.

Jones, R. H., Duncan, H. E., Payne, G. A. and Leonard, J. L. 1980. Factors influencing infection by Aspergillus flavus in silk-inoculated com. Plant Dis. 64:858.

Jossefsson, B. G. E. and Moeller, T. E. 1980, Heat stability of ochratoxin A in pig products. J. Sci Food Agric. 31:1313.

Kawada, N., Mizoguchi, Y., Kobayashi, K., Yamamoto, S. and Morisawa, S. 1990. Arachidonic acid metabolites in carbon tetrachloride induced liver injury. Gastroenterologica Japonica. 25:363.

Kellerman, T. S., Marasas, W. F. O., Thiel, P. G. , Gelderblom, W. C. A. and Cawood, M. 1990. Leucoencephalomalacia in two horses induced by oral dosing of fumonisin Bj. Ondersterpoort J. Vet. Res. 57:269.

Kellerman, T. S., Marasas, W. F. O., Pienaar, J. G. and Naude, T. W. 1972. A mycotoxicosis of Equidae caused by Fusarium moniliforme Sheldon. Onderstepoort J. Vet. Res. 39:205.

243

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Kensler, T. W., Groopman, J. D., Eaton, D. L., Curphey, T. J. and Roebuck, B. D. 1992. Potent inhibition of aflatoxin-induced hepatic tumorigenesis by monofunctional enzyme inducer l,2-dithiol-3-thiome. Carcinogenesis 13:95.

Kensler, T. M. and Trush, M. A. 1984. Role of oxygen radicals in tumor promotion. Environ. Mutagen. 6:593.

Kharmalov, A. Guidotti, A., Costa, E., Hayes, R. and Armstrong, D. 1993. Semisynthetic sphingolipids prevent protein kinase C translocation and neuronal damage in the perifocal area following a photochemically induced thrombotic brain cortical lesion. J. Neurosci. 13:2483.

Khmet, J. and Mahboubi, E. 1972. Esophageal cancer in the Caspian litoral of Iran: Initial strudies. Sci. 175:846.

Kim, H. J., Choi, E. S. and Wade, A. E. 1990. Effect of dietary fat on the induction of hepatic microsomlal cytochrome P450 isozymes by phenobarbital. Biochem. Pharm. 39:1423.

Kirskey, J. W., Cole, R. J. and Domer, J. W. 1989. Relationship between aflatoxin content and buoyancy in Florunner peanuts. Peanut Sci. 16:47.

Klich, M. A. and Pitt, J. I. 1988. Differentiation of Aspergillus flavus from A. parasiticus and other closely relates species. Trans. Br. Mycol. Soc. 91:99.

Klich, M. A., Thomas, S. H., and Mellon, J. E. 1984. Field studies on the mode of entry of Aspergillus flavus into cotton seeds. Mycologia 76:665.

Koehler, B. 1942. Natural mode of entrance of fungi into com ears and some symptoms that indicate infection. J. Agric. Res. 64:421.

Kraus, G. A., Applegate, J. M. and Reynolds, D. 1992. Synthesis of analogs of fumonisin B|. J. Agric. Food Chem. 40:2331.

Kriek, E. 1992. Fifty years of research in N-Acetyl-2-aminofluorene, one of the most versatile compounds in experimental cancer research. Res. Clin. Oncol. 118:481.

Krishnan, K. and Brodeur, J. 1991. Toxicological consequences of combined exposure to environmental pollutants. Arch. Complex Environ. Studies. 3:1.

244

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Kubena, L. F. and Harvey, R. B., Huff, W. E., Corrier, D. E., Phillips. T. D. and Rottinghaus, G. E. 1989. Influence of ochratoxin A and T-2 toxin singly and in combination on broiler chickens. Poult. Sci. 68:867.

Kubena, L. F., Phillips, T. D., Witzel, D. A. and Heidelbaugh, N. D. 1984. Toxicity of ochratoxin a and penicillic acid to chicks. Bull. Environ, Contain. Toxicol. 32:717.

Kulkhami, A. P. 1996. Role of xenobiotic metabolism in developmental toxicity. In: Handbook of Developmental Toxicology. Hood, R. D. (Ed.). CRC Press, Boca Raton, Florida, USA.

Kumagai, S. 1989. Intestinal absorption and excretion of aflatoxin in rats. Toxicol. Appl. Pharmacol. 97:88.

Kurtzman C. P., Smiley, M. J., Robnett, C. J. and Wicklow, D. T. 1986. DNA relatedness among wild and domesticated species in the Aspergillus flavus group. Mycologia. 78:95.

Kwon, O.-S., Schmued, L. C. and Slikker, W. 1997. Fumonisin Bi in developing rats alters brain sphinganine levels and myelination. Neurotox. 18(2):571.

LaBorde, J. B., Terry, K. K., Howard, P. C., Chen, J. J. M., Collins, T. F. X., Shackelford, M. E. and Hansen, D. K. 1997. Lack of Embryotoxicity of Fumonisin Bi in New Zealand White Rabbits. Fundam. Appl. Toxicol. 40:120.

Lake, B. G. 1995. Peroxisome proliferation: Current mechanisms relating to non- genotoxic carcinogenesis. Toxicology letters 82/83:673.

Lancaster, C. M., Jenkins, F. P. and Phillip. P. 1961. Toxicity associated with certain samples of groundnuts. Nature 192:1095.

Langouet, S., Coles, B. Morel, F., Maheo, K., Ketterer, B. and Guillozo, A. 1996. Metabolism of Aflatoxin B i by Human Hepatocytes in Primary Culture. In: Biological Reactive Intermediates V. Snyder, R. (Ed.). Plenum Press, New York, USA.

Laurent, D., Platzer, N., Kohler, F., Sauviat, M. P. and Pellegrin, F. 1989. Macrofusine et mictomoniline: Deux nouvelles mycotoxines isolees de mais infeste par Fusarium moniliforme Sheldon, Microbiol. Alim. Nutr. 7:9. LeBars, J., LeBars, P., Dupuy, J., Boudra, H. and Cassini, R. 1992. Biotic and abiotic factors in fumonisin production and accumulation. 106th Annu. Assoc. Off. Anal. Chem. Meet., Cincinnati, Ohio, (Abstract).

245

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. LeBars, J.; LeBars, P.; Dupuy, J.; Boudra, H. and Cassini, R. 1994. Biotic and abiotic factors in fumonisin Bi production and stability. J. AO AC Int. 77:517.

Lebepe-Mazur, S., Bal, H., Hopmans, E., Murphy, P. and Hendrich, S. 1995. Fumonisin Bi is fetotoxic in rats. Vet. Human Toxicol. 37:126.

Ledoux, D. R., Brown, T. P., Weibking, T. S. and Rottinghaus, G. E. 1992. Fumonisin toxicity in broiler chicks. J. Vet. Diagn. Invest. 4:330.

Leist, M., Ayrton, A. D. and Ioannides, C. 1992. A cytosolic oxygenase activity involved in the bioactivation of 2-amino fluorene. Toxicology. 71:7.

Leslie, J. F., Doe, F. J., Plattner, R. D., Shackelford, D. D. and Jonz, J. 1992a. Fumonisin BI production and vegetative compatibility of strains from Gibberella fujikori mating population “A” ( Fusarium moniliforme). Mycopathologia 117:37.

Leslie, J. F., Plattner, R. D„ Desjardins, A. E., and Klittich, C. J. R. 1992b. Fumonisin B| productions from different mating populations of Giberella fujikori (Fusarium section Liseola). Phytopathology 82(3):341.

Levi, C. P. 1980. Mycotoxins in coffee. J. AO AC 63:1282.

Levi, C. P., Temk, H. L. and Yeransian, J. A, 1975. Investigations of mycotoxins relative to coffee. Colloq. Int. Chim. Cafes. (C. R.) 7:287.

Lidell, C. M. and Burgess, L. W. 1985. Survival of Fusarium moniliforme at controlled temperature and relative humidity. Trans. Br. Mycol. Soc. 84:121.

Lieu, F. Y. and Bullerman, L. B. 1977. Production and stability of aflatoxins, penicillic acid and patulin in several substrates. J. Food. Sci. 46:1306.

Lillehoj, E. B., Kwolek, W. F., Homer, E. S., Widstrom, N. W., Josephson, L. M., Franz, A. O. and Catalano, E. A. 1980. Aflatoxin contamination of preharvest com: Role of Aspergillus flavus inoculum and insect damage. Cereal chem. 57:255.

Lin, J.-K., Kennan, K. A., Miller, E. C. and Miller, J. A. 1978. Reduced nicotinamide adenine dinucleotide phosphate dependent formation of 2,3- dihydro-2,3-dihydroxyaflatoxin Bi from aflatoxin Bi by hepatic microsomes. Cancer Res. 38:2424.

246

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Lindroth, S. 1980. Thermal destruction of patulin in berries and berry jams. J. Food Safety. 2:165.

Liu, L. and Massey, T. E. 1992. Bioactivation of aflatoxin Bi by lipooxygenases, prostaglandin H synthase and cytochorme P450 monooxygenase in guinea- pig tissues. Carcinogenesis. 13:533.

Loewenstein, W. R. 1979. Junctional intercellular communication and the control of cell growth. Biochim. Biophys. Acta. 560:1.

Lomax, L. G., Cole, R. J. and Domer. J. W. 1984. The toxicity of cyclopiazonic acid in weaned pigs. Vet. Pathol. 21:418.

Lopez-Garcia, R. and Park, D. L. 1998. Effectiveness of Post-Harves Procedures in Management of Mycotoxin Hazards. In: Mycotoxins in Agriculture and Food Safety. Bhatnagar, D. and Sinha, S. (Eds.). Marcel Dekker, New York, USA.

Lopez-Garcia, R. 1995. Efecto de diversos tratamientos quimicos sobre la estructura y el potencial toxico/mutagenico de la fumonisina B|. Tesis de Licenciatura. Escuela de Ciencias Quimicas. Universidad La Salle, Mexico, D. F., Mexico.

Loveland, P. M., Nixon, J. E. and Bailey, G. S. 1984. Glucoronides in bile of rainbow trout (Salmo gairdneri). Carcinogenesis. 8:1065.

Lovett, J., Thompson, Jr., R. G. and Boutin, B. K. 1975. Trimming as a means of removing patulin from fungus-rotted apples. J. AO AC 58:909.

Lullmann-Rauch, R. 1979. Drug-induced lysosomal storage disorders. In: Lysosomes in applied biology and therapeutics. Dingle, J. T., Jacques, P. J. and Shaw, I. H. (Eds.). North-Holland Publishing Company. Amsterdam, The Netherlands.

Luter, L., Wyslouzil, W. and Kashyap, S. C. 1982. The destruction of aflatoxins in poeanuts by microwave roasting. Can. Inst. Food Sci. Technol. J. 15:236.

Machen, M. D., Clement, B. A., Shepherd, E. C., Sarr, A. B., Pettit, R. E., and Phillips, T. D. 1988. Sorption of aflatoxins from peanut oil by alluminosilicates. Toxicologist 8:265.

Macias, P., Pinto, M. C. and Campillo, J. E. 1987. Purification and partial characterization of rat liver lipooxygenase. Verlag der Zeitschrift fur Naturforschung. 42:1343.

247

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Mansimanco, N. Remancle, J. and Ramaut, J. 1973. Elimination of aflatoxin B| by absorbent clays in contaminated substrates. Ann. Nutr. Alim. 23:137.

Marasas, W. F. O., Kellerman, T. S., Gelderblom. W. C. A., Coetzer, J. A. W., Thiel, P. G., Van der Lugt, J. J. 1988. Leukoencephalomalacia in a horse induced by fumonisin Bj, isolated from Fusarium moniliforme. Onderstepoort J. Vet. Res. 55:197.

Marasas, W. F. O., Nelson, P. E. and Tousson, T. A. 1984. Toxigenic Fusarium Species. Pennsylvania State University Press. University Park, Pennsylvania.

Marasas, W. F. O., Wehner, F. C., Van Rensburg, S. J. and Van Schalkwyk, D. J. 1981. Mycoflora of com produced in human esophageal cancer areas in Transkei, South Africa. Phytopathology. 71:792.

Marasas, W. F. O., Van Resburg, S. J. and Mirocha, C. J. 1979. Incidence of Fusarium species and the mycotoxins deoxynivalenol and zearalenone, in com produced in esophageal cancer areas in Transkei. J. Agric. Food Chem. 27:1108.

Marasas, W. F. O., Kriek, N. P. J., Steyn, M., van Resburg, S. J. and van Schlkwyk, D. J. 1978. Mycotoxicological investigations on Zambian maize. Food Cosmet. Toxicol. 16:39.

Marasas, W. F. O., Kellerman, T. S., Pienaar, J. G., Naude, T. W. 1976. Leukoencephalomalacia: A mycotoxicosis of Equidae caused by Fusarium moniliforme Sheldon. Onderstepoort J. Vet. Res. 43:113.

Matossian, M. K. 1982. Ergot and the Salem witchcraft affair. Am. Sci. 70:355.

Marin, S., Sancis, V., Vinas, I., Canela, R. and Magan, N. 1995. Effect of water activity and temperature on growth and fumonisin B i and B 2 production by Fusarium proliferatum and F. moniliforme on maize grain Lett. Appl. Microbiol. 21:298.

Maron, D. M. and Ames, B. N. 1983. Revised methods for the Salmonella mutagenicity test. Mutation Research. 113:173.

Martinez-Larranaga, M. R., Anadon, A., Diaz, M. J., Fernandez, R., Sevil, B., Femandez-Cruz, M. L., Fernandez, M. C., Martinez, M. A. and Anton, R. 1996. Induction of cytochrome P4501A1 and P4504A1 activities and peroxisomal proliferation by fumonisin Bi. Toxicol. Appl. Pharmacol. 141:185.

248

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Martinova, E. A. and Merrill, A. H. 1995. Fumonisin Bi alters sphingolipid metabolism and immune function in BALB/c mice:immunological responses to fumonisin Bi. Mycopathologia. 130:163.

Maruyama, T., Ueda, N., Yoshimoto, T., Yamamoto, S., Komatsu, N. and Watanabe, K. 1989. Immunohistochemical study of arachidonate 12- lipoxygenase in porcine tissues. J. of Histochemistry and Cytochemistry. 37:1125.

McDonald D. and Harkness, C. 1967. Aflatoxin in the groundnut at harvest in northern Nigeria. Trop. Sci. 9:148.

McNeil, P. E., Murphy, R. F., Lanni F. and Taylor, D. L. 1984. A method for incorporating macromolecules into adherent cells. J. Cell. Biol. 98:1556.

Merrill, A. H., Schmelz. E.-M., Wang, E., Dillehay, D. L., Rice, L. G., Meredith, F. and Riley, R. T. 1997. Importance o f sphingolipids and inhibitors of sphingolipid metabolism as components of animal diets. J. Nutr. 127:830S.

Merrill, A. H., Wang, E., Vales, T. R., Smith, E. R., Schorede, J. J., Menaldino, D. S., Alexander, C., Crane, H. M., Xia, J., Liotta, D. C., Meredith, F. I. and Riley, R. T. 1996. Fumonisin toxicity and sphingolipid biosynthesis. In: Fumonisins in Food. Jackson, L. S., DeVries, J. W. and Bullerman, L. B. (Eds.) Plenum Press, New York, USA.

Merrill, A. H and Schroeder, J. J. 1993. Lipid modulation of cell function. Annu. Rev. Nutr. 13:539.

Merrill, A. H., van Echten, G., Wang, E. and Sandhoff, K. 1993a. Fumonisin Bi inhibits sphingosine (sphinganine) N-acetyltransferase and de novo sphingolipid biosynthesis in cultured neurons in situ.. J. Biol. Chem. 208(36):27299.

Merril, A. H., Wang, E., Gilchrist, D. G., Riley, R. T. 1993b. Fumonisins and other inhibitors of de novo sphingolipid biosynthesis. In: Advances in Lipid Research: Sphingolipids and Their Metabolites. Bell, R. M., Hannun, Y. A. and Merrill, A. H. (Eds.). Academic Press, San Diego, California, USA.

Merrill, A. H., Lingrell, S., Wang, E., Nikolova-Karakashian, M. Vales, T. R. and Vance, D. E. 1995. Sphingolipid biosynthesis de novo by rat hepatocytes in culture. J. Biol. Chem. 270(23): 13834.

249

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Merrill, A. H., Scmelz, E.-M., Wang, E., Dillehay, D. L., Rice, L. G., Meredith, F. and Riley, R. T. 1997. Importance of sphingolipids and inhibitors of sphingolipid metabolism as components of animal diets. J. Nutr. 127:830S.

Mesnil, M. Krutovskikh, V., Omoru, Y. and Yamasaki, H. 1995. Role of blocked gap junctional intercellular communication in non-genotoxic carcinogenesis. Toxicology letters. 82/83:701.

Metcalfe, S. A. and Neal, G. E. 1983. The metabolism of aflatoxin Bi by hepatocytes isolated from rats following the in vivo administration of some xenobiotics. Carcinogenesis. 4:1007.

Meydahl, M., Blumberg, J. B. and Hayes, K. C. 1985. Dietary fat unsaturation enhances drug metabolism in cebus but not in squirrel monkeys. J. Nutr. 115. 573.

Milks, L. C., Mukar, N. M., Houghten, R. Unwin, N. and Gilula, N. B. 1988. Topology of the 32 kd liver gap junction protein determined by site-directed antibody localizations. EMBO J. 7:2967.

Miller, S. A. 1991. Food additives and contaminats. In: Casarett and Doull’s Toxicology: The Basic Science of Poisons. Amdur, M. O., Doull. J. and Klaassen, C. D. (Eds.). 4lhed., McGraw Hill, New York, USA.

Moreno, A. P., Fishman, G. I. and Spray, D. C. 1992. Phosphorylation shifts unitary conductance and modifies voltage dependent kinetics of human connexin43 gap junction channels. Biophys. J. 62( 1 ):51.

Mumtaz, M. 1995. Risk assessment of chemical mixtures from a public health perspective. Toxicology Letters. 82/83:527.

Murphy, P. A., Henrich,, S., Hopmans, E. C., Hauck, C. C., Lu, Z. Buseman, G. and Munkvold, G. 1996. Effect of processing on fumonisin content of com. In: Fumonisins in Food. Jackson, L. S., DeVries, J. W. and Bullerman, L. B. (Eds.) Plenum Press, New York, USA.

Murphy, P. A., Hopmans, E. C., Miller, K. and Henrich, S. 1995. Can fumonisins be detoxified? In: Natural Protectants and Natural Toxicants in Foods. Bidlack, W. R. and Omaye, T. (Eds.). Technomic Publishing Company, Lancaster, Pennsylvania, USA.

Murphy, P. A., Rice, L. G. and Ross, P. F. 1993. Fumonisins Bi, B 2, and B 3 content of Iowa, Wisconsin, and Illinois com and com screenings. J. Agr. Food Chem. 41:263.

250

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Murray. A. W. and Fitzgerald, D. J. 1979. Tumor promoters inhibit metabolic cooperation in co-cultures of epidermal and 3T3 cells. Biochem. Biophys. Res. Commun. 91:365.

Na, M. R., Koo, S. K., Kim, D. Y., Park, S. D., Rhee, S. K., Kang, K. W. and Joe C. O. 1995. In vitro inhibition of gap junctional intercellular communication by chemical carcinogens. Toxicology 98:199.

Ncyiyana, D. J. 1986. Neural tube defects among rural blacks in a Transkei district. S. Afr. Med. J. 69:618.

Nelson, P. E., Toussoun, T. A. and Marasas, W. F. O. 1983. Fusarium Species. An illustrated Manual for Identification. Pennsylvania State University Press. University Park, Pennsylvania.

Nelson, P. E., Plattner, R. D., Shackelford, D. D. and Desjardins, A. E. 1992. Fumonisin Bi production by Fusarium species other than F. moniliforme in section Liseola and by some related species. Appl. Environ. Microbio. 58:984.

Nemoto, N. and Takayama, S. 1984. Arachidionic acid-dependent activation of benzo[a]pyrene to bind to proteins with cytosolic and microsomal fractions from rat liver and lung. Carcinogenesis. 5:961.

Nemoto, N. and Takayama, S. 1983. Effects of unsaturated fatty acids on metabolism of benzo[a]pyrene in an NADPH-fortified rat liver microsomal system. Carcinogenesis. 4( 10): 1253.

Newbeme, P. M. and Butler, W. H. 1969. Acute and chronic effects of aflatoxin on the liver of domestic and laboratory animals. A review. Cancer Res. 29:236.

Niedel, J. E., Kuhn, L. J. and Vanderbark, G. R. 1983. Phorbol diester receptor co- purifies with protein kinase C. Proc. Natl. Acad. Sci. USA. 80:36.

Norred, W. P., Voss, K. A., Riley, R. T. and Plattner, R. D. 1996. Fumonisin toxicity and metabolism studies at the USDA. In: Fumonisins in Food. Jackson, L. S., DeVries, J. W. and Bullerman, L. B. (Eds.). Plenum Press, New York, USA.

Norred, W. P., Plattner, R. D. and Chamberlain, W. J. 1993. Distribution and excretion of [ 14C]fumonisin Bi in male Sprague-Dawley rats. Nat. Toxins, 1:341.

251

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Norred, W. P., Wang, E., Hwansoo, Y., Riley, R. T. and Merril. A. H. 1992. In vitro toxicology of fumonisins and the mechanistic implications. Mycopathologia. 117:73.

Norred, W. P., Voss, K. A., Bacon, C. W., and Riley, R. T. 1991. Effectiveness of ammonia treatment in detoxification of fumonisin-contaminated com. Food Che. Toxic. 29(12):819.

Nuehring, L. P., Rowland, G. N., Harrison, L. R., Cole, R. J. and Domer, J. W. 1985. Cyclopiazonic acid mycotoxicosis in the dog. Am. J. Vet. Res. 46:1670.

Nyvall, R. F., Kommendahl, T. 1970. Saprophytism and survival of Fusarium moniliforme in com stalks. Phytopathol. 60:1233.

Oh, S. Y., Madhukar, B. V. and Trosko, J. E. 1988. Inhibition of gap junctional blockage by palmitoyl carnitine and TMB - 8 in a rat liver epithelial cell line. Carcinogenesis. 9:135.

Okoye, Z. S. C. 1987. Stability of zearalenone in naturally contaminated com during Nigerian traditional brewing. Food Add. Contam. 4:59.

Olutiola, P. O. 1976. Some environmental and nutritional factors affecting growth and sporulation of Aspergillus flavus. Trans. Br. Mycol. Soc. 66:131.

Osborne, B. G. 1979. Reverse-phase high performance liquid chromatography determination of ochratoxin A in flour and bakery products. J. Sci. Food Agric. 30:1065.

Osweiler, G. D., Ross, P. F., Wilson, T. M., Nelson, P. E., Witte, S. T. 1992. Characterization of an epizootic of pulmonary edema in swine associated with fumonisin in com screenings. J. Vet. Diag. Invest. 4:53.

Park, D. L., Lopez-Garcia, R., Trujillo-Preciado, S., and Price, R. L. 1996. Reduction of risks associated with fumonisin contamination in com. In: Fumonisins in Food. Jackson, L. S., DeVries, J. W., and Bullerman, L. B. (Eds.). Plenum Press, New York, USA.

Park, D. L. 1993. Controlling aflatoxin in food and feed. Food Tech. 47(10):92.

Park D. L., Rua, S. M., Mirocha, C. J., Abd-alla, E. A. M., and Weng, C. Y. 1992. Mutagenic potentials of fumonisin contaminated com following ammonia decontamination procedures. Mycopahtol. 117:105.

252

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Park, D. L. and Stoloff, L. 1989. Aflatoxin control-how a regulatory agency managed risk from an unavoidable natural toxicant in food and feed. Regulatory Toxicology and Pharmacology 9:109.

Park, D. L., Lee, L. S., Price, R. L., and Pohland, A. E. 1988. Review of decontamination of aflatoxin by ammoniation: Current status and regulation. J. AOAC 71:685.

Park, D. L., Aguirre-Flores, I., Scott, W. F. and Alterman, E.1986. Evaluation of chicken embryo, brine shrimp and bacterial bioassays for saxitoxin. J. of Tox. and Env. Health. 18:589.

Parker, W. A. and Melnick, D. 1966. Absence of aflatoxin from refined vegetable oils. J. A. Oil Chem. Soc. 43:635.

Parkinson, A. 1996. Biotransformation of xenobiotics. /w.Casaret & Doull’s Toxicology. The Basic Science of Poisons. Klaassen, C. D., Amdur, M. O. and Doull, J. (Eds.). Fifth Edition. McGraw-Hill, New York, USA.

Pavao, A. C., Soares-Nieto, L. A., Ferreira-Neto, J. and Leao, M. B. C. 1995. Structure and activity of aflatoxins B and G. Journal of Molecular Structure (Theochem). 337:57.

Payne, G. A., Kamprath, E. J. and Adkins, C. R. 1989. Increased aflatoxin contamination in nitrogen stressed com. Plant Dis. 72:422.

Payne, G. A., Cassel, D. K. and Adkins, C. R. 1986. Reduction of aflatoxin contamination by irrigation and tillage. Phytopathology. 76:679.

Peers, F. G. and Linsell, C. A. 1975. Aflatoxin contamination and its heat stability in Indian cooking oils. Trap. Sci. 17:229.

Pensala, O., Niskanan, A. and Lindroth, S. 1981. Aflatoxin production in black currant, blueberry and strawberry jams. J. Food Prot, 41:344.

Pereira, T. A. and Das, N. P. 1991. Assay of liver cytosol lipoxygenase by differential pulse polarography. Annalytical Biochemistry. 197:96.

Perez, H. D., Roll, F. J., Bissell, D. M., Shak, S. and Goldstein, I. M. 1984. Production of chemotactic activity for polymorphonuclear leukocytes by cultured rat hepatocytes exposed to ethanol. J. of clinical Investigation. 74:1350.

253

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Pestka, J. J., Azcona Olivera. J. I., Plattner, R. D., Minervini, F., Doko, M. B. and Visconti, A. 1994. Comparative assessment of fumonisin in grain-based foods by ELISA, GC-MS, and HPLC. J. Food Prot. 57:169.

Pettit, R. E., Taber, R. A., Shroeder, H. W., and Harrison, A. L. 1971. Influence of fungicides and irrigation practice on aflatoxin in peanuts before digging. Appl. Microbiol. 22:629.

Phillips, T. D., Clement, B. A. and Park, D. L. 1994. Approaches to reduction of aflatoxins in foods and feeds. In: The Toxicology of Aflatoxins. Human Health, Veterinary and Agricultural Significance. Eaton, D. L and Groopman, J. D. (Eds.), Academic Press, San Diego, USA.

Pitt, J. I. 1989. Field studies on Aspergillus flavus and aflatoxins in Australian groundnuts. In: Aflatoxin contamination of Groundnuts. McDonald, D. and Mehan, V. K. (Eds.) ICRISAT, Patancheru, India.

Pitt, J. I. and Hocking, A. D. 1977. Influence of solute and hydrogen ion concentration on the water relations of some xerophilic fungi. J. Gen, Microbiol. 101:35.

Pitt, J. I. and Hocking, A. D. 1997. Fungi and Food Spoilage. Second Edition. Blackie Academic and Professional. London.

Pittet, A., Parisod, V. and Schellenberg, M. 1992. Occurrence of fumonisins Bi and B2 in com-based products from the Swiss market. J. Agric. Food Chem. 40:1352.

Pohland, A. E. and Allen, R. 1970. Stability studies with patulin. J. AO AC. 53:688.

Prelusky, D. B., Trenholm, H. L. and Savard, M. E. 1994. Pharmacokinetic fate of l4C-labeIled fumonisin Bi in swine. Nat. Toxins. 2:73.

Price, R. L. and Jorgensen, K. V. 1985. Effects of processing on aflatoxin levels and on mutagenic potential of tortillas made from naturally contaminated com. J. Food Sci. 50:347.

Purchase, I. F. H. 1971. The acute toxicity of the mycotoxin ciclopiazonic acid to rats. Toxicol. Appl. Pharmacol. 18:114.

Ramsdell, H. S. and Eaton, D. L. 1990. Species susceptibility to aflatoxin Bi carcinogenesis: Comparative kinetics of microsomal biotransformation. Cancer Res. 50:615.

254

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reiss, J. 1978. Mycotoxins in foodstuff. XI. Fate of aflatoxin Bi during preparation and baking of whole-meal wheat bread. Cereal Chem. 55:421.

Repine, J. E., White, J. G., Clawson, C. C. and Holmes, B. M. 1974. The influence of phorbol myristate acetate on oxygen consumption by polymorphonuclear leukocytes. J. Lab. Clin. Med. 83:911.

Rheeder, J. P., Marasas, W. F. O., Thiel, P. G., Sydenham, E. W. and Shephard, G. S. 1992. Fusarium moniliforme and fumonisins in com in relation to human esophageal cancer in Transkei. Phytopathol. 82:353.

Ricci, P. F. and Melton, L. S. 1985. Regulating cancer risks. Environ. Sci Technol. 19:473.

Rice, L. G. and Ross, F. P. 1994. Methods for detection and quantitation of fumonisins in com, cereal products and animal excreta. J. Food Prot. 57: 536.

Richard, J. L., Peden, W. M. and Thurston, J. R. 1989. Combined cyclopiazonic acid and aflatoxin effects on serum bacteriostasis, glycocholic acid, complement activity, and enzymes and histopatholofic changes in guinea pigs. In: Cellular and Molecular Mode of Action of selected Microbial Toxins. Pohland, A. E., Dowell, V. R. and Richard, J. L. (Eds.) Plenum Press, New York, USA.

Riley, R. T., Hinton, D. M., Chamberlain, W. J., Bacon, C. W., Wang, E., Merrill, A. H. and Voss, K.A. 1994. Dietary fumonisin Bi induces disruption of sphingolipid metabolism in Sprague-Dawley rats: A new mechanism of nephrotoxicity. J. Nutr. 124:594.

Riley, R. T., Norred, W. P. and Bacon, C. W. 1993. Fungal toxins in foods: Recent concerns. Annu. Rev. Nutr. 13:167.

Roebuck, B. D., Liu, Y.-L., Rogers, A. E., Groopman, J. D. and Kensler, T. W. 1991. Protection against aflatoxin Bi-induced hepatocarcinogenesis in F344 rats by 5-(2-pyranizinyl)-4-mthyl-l,2-dithiol-3-thione (oltipraz): Predictive role for short term molecular dosimetry. Cancer Res. 51:5501.

Rogers, A. 1994. Nutritional modulation of aflatoxin carcinogenesis. In; The Toxicology of Aflatoxins: Human Health, Veterinary and Agricultural Significance. Academic Press, San Diego, California, USA.

255

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Roll, F. J., Bissel, D. M. and Perez, H. D. 1986. Human hepatocytes metabolizing ethanol generate a nonpolar chemotactic factor for human neutrophils. Biochemical and Biophysical Research Communications. 137:688.

Rosello-Catafau, J., Hotter, G., Closa, D., Ortiz, M. A., Pou-Torello, J. M., Gimeno, M., Broque, G. and Gelpi, E. 1994. Liver lipooxygenase arachidonic acid metabolites in streptomzotocin-induced diabetes in rats. Prostaglandins, Leukotrienes and Essential Fatty Acids. 51:411.

Ross, P. F. 1994. What are we going to do with this dead horse? J. AO AC Int., 77:2.

Ross, F. P., Rice, L. G., Osweiler, G. D., Nelson, P. E., Richard, J. L., Wilson, T. M. 1992. A review and update of animal toxicoses associated with fumonisin-contaminated feeds and production of fumonisins by Fusarium isolates. Mycopathol. 117:109.

Ross, P. F., Rice, L. G., Plattner, R. D., Osweiler, G. D., Wilson, T. D., Owenas, D. L., Nelson, H. A., and Richard, J. L. 1991. Concentrations of fumonisin Bi in feeds associated with animal health problems, Mycopathol. 114:129.

Ross, P. F ., Nelson, P. E., Richard, J. L., Osweiler, G. D., Rice, L. G., Plattner, R. D. and Wilson, T. M. 1990. Production of fumonisins by Fusarium moniliforme and Fusarium proliferaturm isolates associated with equine leukoencephalomalacia and pulmonary edema syndrome in swine, Appl. Environ. Microbiol. 56:3225.

Roy, S. K. and Kulkami, A. P. 1997. Aflatoxin Bj epoxidation catalyzed by partially purified human liver lipoxygenase. Xenobiotica. 27(2):231.

Roy, S. K. and Kulkami, A. P. 1994. Dioxygenase and co-oxidase activities of rat hepatic cytosolic lipoxygenease. Journal of Biochemical Toxicology. 10:111.

Saccardo, P. A. 1881. Fungi Italici Autographice Delinati, Fig. 789. Patavii.

Saccone, G. T. P. and Pariza, M. W. 1981. Enhancement of hepatic microsome- mediated bacterial mutagenesis by the rat liver souluble protein fraction. Mutat. Res. 88:135.

Saez, J. C., Connor, J. A., Spray, D. C. and Bennett, M. V. L. 1989. Hepatocyte gap junctions are permeable to the second messenger inositol 1,4,5-triphosphate, and to calcium ions. Pore. Natl. Acad. Sci. USA. 86:2708.

256

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Salama, A. M. and Mishricky, A. G., 1973. Seed transmission of maize wilt fungi with special reference to Fusarium moniliforme Sheld. Phytopathol. 77:356.

Sandhu, P., Guo, Z., Baba, T., Martin, M. V., Tukey, R. H. and Guengerich, F. P. 1994. Expression of modified human cytochrome P4501A2 in Escherichia coll. Stabilization, purification, spectral characterization, and catalytic activities of the enzyme. Arch. Biochem. Biophys. 309:168.

Sands, D. C., McIntyre, J. L. and Walton, G. S. 1976. Use of activated charcoal for the removal of patulin from cider. Appl. Env. Microbiol. 32:388.

SAS Institute, Inc. 1988. SAS/STAT guide for Personal Computer. Version 6.0 SAS Institute, Cary, NC.

Schwartz, A., Rapaport, E., Hirschberg, K., and Futerman, A. H. 1995. A Regulatory role for sphingolipids in neuronal growth: Inhibition of sphingolipid synthesis and degradation have opposite effects on axonal branching. J. Biol. Chem. 270(18): 10990.

Scott, P. M. 1984. Effects of food processing on mycotoxins. J. Food Protec. 47(6):489.

Scott, P. M and Lawrence, G. A. 1994. Stability and problems in recovery of fumonisins added to corn-based foods. J. AO AC Int. 77:541.

Scott, P. M., Kanhere, S. R., and Weber, D. 1993. Analysis of Canadian and imported beers for Fusarium mycotoxins by gas chromatography- mass spectroscopy. Food Add. Contam. 10:381.

Scott, P. M., Kanhere, S. R., Dailey, E. F., and Farber, J. M. 1992. Fermentation of wort containing deoxynivelenol and zearalenone. Mycotoxin Res. 8:58.

Scott, P. M., Fuleki, T. and Harwig, J. P. 1977. Patulin content of juice and wine produced by moldy grapes. J. Agr. Food Chem. 25:434.

Seitz, L. M., Eustace, W. D., Mohr, H. E., Shogren, M. D. and Yamasaki, W. T. 1986. Cleaning, milling and baking tests with hard red winter wheat containing deoxynivalenol. Cereal Chem. 63:146.

Shank, R. C. 1981. Environmental toxicoses in humans. In: Mycotoxins and N- Nitroso-compounds:Environmental Risks. Shank, R. C. (Ed.) vol 1. CRC Press, Boca Raton, Florida, USA.

257

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Shantha. T. and Sreenivasamurthy, M. 1977. Photo-destruction of aflatoxin in groundnut oil. Indian J. Technol. 15:453.

Sheffield, V. C., Beck, J. S., Kwitek, A. E., Sanstrom, D. W. and Stone, E. M. 1993. The sensitivity of single-strand conformation polymorphism analyses for the detection o f single base substitutions. Genomics 16:325.

Sheldon, L. 1904. A com mold ( Fusarium moniliforme n. sp.) 17th Ann. Rep. Agric. Exp. Stn.: Nebtraska, USA. 23.

Shephard, G. S., Thiel, P. G., Sydenham, E. W., Alberts, J. F. and Gelderblom, W. C. A. 1992. Fate of single dose of the 14C-labelled mycotoxin, fumonisin B| in rats. Toxicon. 30:768.

Shier, W. T., Abbas, H. K. and Mirocha, C. J. 1991. Toxicity of mycotoxins fumonisin Bi and B 2 and Alternaria alternata f. sp. lycopersici toxin (AAL) in cultured mamamlian cells. Mycopathol. 116:97.

Shiratori, Y., Hai, K., Takada, H., Kiriyama, H., Nagura. T., Matsumoto, K., Kamai, K., Okano, K. and Tanaka, M. 1992. Mechanism of accumulation of macrophages in galactosamine-induced liver injury: Effect of lipooxygenases inhibitors on chemotaxis of spleen cells. Pathobiology, 60:316.

Shroeder, J. J., Crane, H. M., Xia, J., Liotta, D. C. and Merril, A. H. 1994. Disruption of sphingolipid metabolism and stimulation of DNA synthesis by fumonisin Bj: A molecular mechanism for carcinogenesis associated withFusarium moniliforme. J. Biol. Chem. 269(5):3475.

Shroeder, H. W. and Boiler R. A. 1973. Aflatoxin production of species and strains of the Aspergillus flavus group isolated from field crops. Appl. and Environ. Microbiol. 25:885.

Simpson, I., Rose, B. and Loewenstein, W. R. 1977. Size limit of molecules permeating the junctional membrane channels. Science. 195:294.

Smith, E. E., Kubena, L. F., Braithwaite, C. E., Harvey, R. B., Phillips, T. D. and Reine, A. H. 1992. Toxicological evaluation of aflatoxin and cyclopiazonic acid in broiler chickens. Poult. Sci. 71:1136.

Smith, J. N. and Ioannides, C. 1987. Cytosolic potentiation of the rat hepatic microsome-mediated mutagenicity of benzidine. Mutagenesis 1:367.

258

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Smith, J. E. and Moss, M. O. 1985. Mycotoxins: Formation, analysis and significance. John Wiley, New York.

Smith, E. R. and Merrill, A. H. 1995. Differential roles of de novo sphingolipid biosynthesis and turnover in the “burst of free sphingosine and sphinganine, and their 1-phosphates and N-acyl-derivatives, that occurs upon changing the medium of cells in culture. J. Biol. Chem. 270(32): 18749.

Solanki, V., Rana, R. S. and Slaga T. 1981. Diminution of mouse epidermal superoxide dismutase and catalase activities by tumor promoters. Carcinogenesis. 2:1141.

Stack, M. E. and Eppley, R. M. 1992. Liquid chromatographic determination of fumonisins Bi and B 2 in com and com-based products. J. AO AC Int. 75:834.

Stephenson, L. W. and Russell, T. E. 1974. The association of Aspergillus flavus with hemipterous and other insects infesting cotton bracts and foliage. Phytopahology 64:1502.

Stevens, V. L. and Tang, J. 1997. Fumonisin Bi -induced sphingolipid depletion inhibits vitamin uptake via the glycosylphosphatidylinositol-anchored folate receptor. J. Biol. Chem. 272:18020.

Stohs, S. J. and Wu, L. J. 1981. Effects of prostaglandins Ei, Ei, F?a and arachidonic acid on aryl hydrocarbon hydroxylase activity of intestinal and hepatic microsomes from male rats. Res. Commun. Chem. Pathol. Pharmacol. 34:461.

Stoloff, L. 1982. Mycotoxins as potential environmental carcinogens. In: Carcinogens and Mutagens in the Environment. Stich, H. F. (Ed.). CRC Press, Boca Raton, Florida, USA.

Stoloff, L., Trucksess, M. W., Anderson, P. W., Glabe, E. F., and Aldridge, J. G. 1978. Determination of the potential for mycotoxin contamination of pasta products. J. Food Sci. 43:228.

Stoloff, L. 1976. Occurrence of mycotoxins in foods and feeds. 1976. In: Mycotoxins and other fungal realted food problems. Rodricks, J. V. (Ed.). Adv. Chem. Ser. No. 149. American Chemical Society, Washington, D. C.

Stott, W. T. and Bullerman, L. B. 1976. Instability of patulin in cheddar cheese. J. Food Sci. 41:201.

259

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Stryer, L. Biochemistry. Third Edition. W. H. Freeman and company. New York, USA.

Sweeley, C. C. 1991. Sphingolipids. In: Biochemistry of Lipids, Lipoproteins and Membranes. Vance, D. E. and Vance, J. E. (Eds.). Elsevier Science, Amsterdam, The Netherlands.

Swenson. D. H., Lin, J.-K., Miller, E. C. and Miller, J. A. 1977. Aflatoxin B 1-2 ,3 - oxide as a probable intermediate in the covalent binding of aflatoxins B| and B 2 to rat liver DNA and ribosomal RNA in vivo. Cancer Res. 37:172.

Sydenham, E. C. 1994. Fumonisins: chromatographic methodology and their role in human and animal health. Thesis presented for the degree of Doctor of Phylosophy, University of Cape Town, South Africa.

Sydenham, E. W., Shephard, G. S., Thiel, P. G., Marasas W. F. 0. and Stodkenstrom. 1991. Fumonisin contamination of commercial corn-based human foodstuffs. J. Agric. Food Chem. 39(11 ):2014.

Sydenham, E. W., Gelderblom, W. C. A., Thiel, P. G. and Marasas, W. F. O. 1990 Evidence for the natural occurrence of fumonisin B| in com. J. Agric. Food Chem. 38:285.

Takeda, A., Hashimoto, E., Yamamura, H. and Shimazy, T. 1987. Phosphorylation of liver gap junction protein by protein kinase C. FEBS Lett. 210:169.

Tanaka, T., Abbas, H. and Duke, S. 1993. Structure-dependent phytotoxicity of fumonisins and related compounds in duckweed bioassay. Phytochem. 33(4):779.

Terao, K.. 1967. The effect of aflatoxin on chick-embryo liver cells. Exp. Cell. Res. 48:151.

Texas Department of Health. 1992. An Investigation of a Cluster of Neural Tube Defects in Cameron County, Texas. Texas Department of Health, Austin, Texas.

Thean, J. E., Lorenz, D. R., Wilson, D. M., Rodgers, K. and Gueldner, R. C. 1980. Extraction, cleanup and quantitative determination of aflatoxins in com. J. Assoc. Offic. Anal. Chem. 63(3):631.

Thiel, P. G., Marasas, W. F. O., Sydenham, E. W., Shephard, G. S., Gelderblom, W. C. A. and Nieuwenhuis, J. J. 1991. Survey of fumonisin production by Fusarium species. Appl. and Env. Microbiol. 57(4): 1089.

260

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Tiegs. G. and Wendel, A. 1988. Leukotriene-mediated liver injury. Biochemical Pharmacology. 37:2569.

Townsend, C. A., Christensen, S. B. and Trautwein, K. 1984. Hexanoate as a starter unit in polyketide synthesis. J. Am. Chem. Soc. 106:3868.

Trenholm, H. L., Charmley, L. L., Prelusky, P. B. M. and Warner, R. M. 1991. Two physical methods for the decontamination of four cereals contaminated with deoxynivalenol and zearalenone. J. Agr. Food Chem. 39:356.

Trosko, J. E. and Chang, C. C. 1988. Nongenotoxic mechanisms in carcinogenesis. Role of inhibited intercellular communication. In: Carcinogen Risk Assessment: New directions in the qualitative and quantitative aspects. Hart, R. W. and Hoerger, F. G. (Eds.). Branbury report 31. Cold Spring Harbor Laboratory, New York, USA.

Trosko, J. E. and Chang, C. C. 1983. Potential role of intercellular communication in the rate-limiting step in carcinogenesis. J. Am. Coll. Toxicol., 2:5.

Trucksess, M. W., Mislevic, P. B., Young, K. Bruce, V. R. and Page, S. W. 1987. Cyclopiazonic acid production by cultures of Aspergillus and Penicillium species isolated from dry beans, com meal, macademia nuts and pecans. J. Assoc. Off. Anal. Chem. 70:123.

Tsai, S.-Y. and Pence, B. C. 1992. Effects of dietary fat on hepatic microsomal and cytosolic mutagenic activation of 2-aminofluorene. Cancer Letters 67:87.

Ueng, Y.-F., Shimada, T., Yamazaki, H. and Guengerich, F. P. 1995. Oxidation of aflatoxin Bi by bacterial recombinant human cytochrome P450 enzymes. Chem. Res. Toxicol. 8:218.

Ueno, Y. 1987. Mycotoxins. In: Toxicological Aspects of Food. Miller, K. (Ed.). Elsevier, Ney York.

Ueno, Y. 1983. Tricothecenes: Chemical, biological and toxicological aspects. Elsevier, New York.

Ulloa, M. and Herrera, T. 1970. Persitencia de las aflatoxinas durante la fermentation del pozol. Rev. Lat-Amer. Microbiol. 12:19.

Ulloa Sosa, M. and Shroeder, H. W. 1969. Note on aflatoxin decomposition in the process of making tortillas from com, Cereal Chem. 46:397.

261

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Valdivia Retes, M. T.. Efecto del peroxido de hidrogeno y el bicarbonato de sodio sobre el contenido de aflatoxinas Bi y Bi en maiz. Tesis de Licenciatura. Escuela de Ciencias Quimicas. Universidad La Salle, Mexico, D. F„ Mexico.

van der Brink, H. J. M., van Gourcom, R. F. M., van den Hondel, C. A. M. J. J. and Punt, P. J. 1998. Cytochrome P450 enzyme systems in fungi. Fungal Genetics and Biology. 23:1.

van Egmond, H. P. 1983. Mycotoxins in dairy products. Food Chem. 11:289.

van Meer, G. and Burger, K. N. J. 1992. Sphingolipid trafficking sorted out? Trends Cell Biol. 2:332.

van Rensburg, S. J. 1985. Recent studies on the etiology of oesophageal cancer. S. Afr. Cancer Bull. 29:22.

Viljoen, D. L., Buccimazza, S., Dunne, T. and Molteno, C. 1995. The prevalence and prevention of neural tube defects in Cape Town. S. Aff. Med. J. 85:630.

Voss, K. 1994. Fumonisin Bi linked to kidney damage in rabbits. Food Chem. News. 36(5):26.

Voss, K. A., Norred, W. P., Plattner, R. D. and Bacon, C. W. 1989. Hepatotoxicity and renal toxicity in rats of com samples associated with field cases of leukoencephalomalacia. Food Chem. Toxicol. 27:89.

Visconti, A. and Doko, M. B. 1994. Survey of fumonisin production by Fusarium isolated from cereals in Europe. J. AO AC Int. 77:546.

Vudathala, D. K., Prelusky, D. B., Ayroud, M., Trenholm, H. L. and Miller, J. D. 1994. Pharmacokinetic fate and pathological effects of l 4C-fumonisin Bj in laying hens. Nat. toxins. 2:81

Walton, G. S., Sands, D. C., and McIntyre, J. L. 1976. Patulin removal from cider using activated charcoal. Proc. Am. Phytopathol Soc. 3:255.

Wang, E., Ross, P. F., Wilson, T. M., Riley, R. T. and Merrill, A. H. 1992. Increases in serum sphingosine and sphinganine and decreases in complex sphingolipids in ponies given feed containing fumonisins, mycotoxins produced by Fusarium moniliforme. J. Nutr. 122:1706.

262

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Wang, E., Ross, P. F., Wilson, T. M., Riley. R. T. and Merrill, A. H.1991. Inhibition of sphingolipid biosynthesis by fumonisins: Implications for diseases associated with Fusarium moniliforme. J. Biol. Chem. 266:14486.

Weinstein, B., Orenstein, J. M., Gebert R., Kaign, E. and Stadler, U. C. 1975. Growth and structural properties of epithelial cell cultures established from normal rat liver and chemically induced hepatomas. Cancer Research. 35:253.

Wheeler, K. A., Hurdman, B. F. and Pitt, J. I. 1991. Influence of pH on the growth of some toxigenic species of Aspergillus, Penicillium and Fusarium. Int. J. Food Microbiol. 12:141.

Wicklow, D. T. 1983. Taxonomic features and ecological significance of sclerotia. In: Aflatoxin and Aspergillus flavus in Com. Dienwer, U. L., Asquith, R. L., and Dickens, J. W. (Eds.). Alabama Agricultural Experimental Station, Auburn University, Auburn, USA.

Wilson, D. M. and Payne, G. A. 1994. Factors Affecting Aspergillus flavus group infection and aflatoxin contamination of crops. In: The Toxicology of Aflatoxins: Human Health, Veterinary and Agricultural Significance, Eaton, D. L. and Groopman J. D. (Eds.). Academic Press, San Diego, California, USA.

Wilson, T. M., Ross, P. F., Owens, L. G. Rice. S. A., Green. S. J. and Nelson, H. A. 1992. Experimental reproduction of ELEM: A study to determine the minimum toxic dose in ponies. Mycopathologia, 117:115.

Wilson, T. J. and Romer, T. R. 1991. Use of mycosep multifunctional cleanup column for liquid chromatographic determination of aflatoxin in agricultural products. J. Assoc. Off. Anal. Chem. 74(6):951.

Wilson, T. M., Ross, P. F., Owens, L. G., Rice, S. A., Green, S. J. and Nelson, H. A. 1991. ELEM and its relationship to fumonisins in feed. Symposium. Raleigh, N. C. April 24-25.

Wilson, T. M., Ross, P. F., Rice, L. G., Osweiler, G. D., Nelson, H. A., Owens, D. L., Plattner, R. D., Reggiardo, C., Noon, T. M. and Pickrell, J. W. 1990. Fumonisin Bj levels associated with an epizootic of equine leukoencephalomalacia. J. Vet, Diagn. Invest. 2:213.

Wilson, R., Zirpin, R., Ragsdale, S. and Busbee, D. 1985. Uptake and vascular transport of ingested aflatoxin. Toxicol. Lett. 29:169.

263

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Wiseman, D. W.. Applebaum, R. S., Brackett, R. E., and Marth, E. H. 1983. Distribution and resistance to pasteurization of aflatoxin Mi in naturally contaminated whole milk, cream and skim milk. J. Food Prot. 46:530.

Wogan, G. N. and Newbeme, P. M. 1967. Dose-response characteristics of aflatoxin Bi carcinogenesis in the rat. Cancer Res. 27:2370.

Wogan, G. N. 1966. Chemical nature and biological effects of aflatoxins. Bacteriol. Rev. 30:460.

Wolf, D. P. 1998. Inhibition of cellular uptake of folate by blocking sunthesis of the membrane folate receptor. Nutr. Rev. 56(3):86.

Wolf, G. 1994. Mechanism of the mitogenic and carcinogenic action of fumonisin Bi, a mycotoxin. Nutr. Rev. 52:246.

Wolfel, C., Platt, K.-L., Dogra, S., Glatt, H. Wachter, F. and Doehmer, J. 1991. Stable expression of rat cytochrome P4501A2 cDNA and hydroxylation of 17P-estradiol and 2-aminofluorene in V79 Chinese hamster cells. Molecular Carcinogenesis. 4:489.

Wood, G. M. 1982. Effects of processing on mycotoxins in maize. Chem. Ind. 972.

Wood, G. M., Cooper, S. J. and Chapman, W. B. 1982. Problems associated with laboratory simulation of effects of food processes on mycotoxins. Proceedings of V International Symposium on Mycotoxins, Vienna, Austria.

Wu, W. W., McDonough, V. M., Nickels, J. T.. Ko, J., Fischi, A. S., Vales. T. R., Merrill, A. H. and Carman, G. M. 1995. Regulation of lipid biosynthesis in Sacharomyces cervisiae by fumonisin Bj. J. of Biol. Chem. 270(22): 13171.

Yamaski, H. 1996. Role of disrupted gap junctional intercellular communication in detection and characterization of carcinogens. Mutation Research. 365:91.

Yamasaki, H. 1996. Role of disrupted gap junctional intercellular comunication in detection and characterization of carcinogens. Mutation Research. 365:91.

Yamasaki, H. and Naus, C. C. G. 1996. Role of connexin genes in growth control. Carcinogenesis 17:1199.

Yamasaki, H. 1990. Role of disrupted gap junctional intercellular communication in detection and characterization of carcinogens. Mutation Research. 365:91.

264

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Yamasaki, H. 1990. Gap junctional intercellular communication and carcinogenesis. Carcinogenesis. 11(7): 11051.

Yamasaki, H. and Katoh, H. 1988. Further evidence for the involvement of gap- junctional intercellular communication in induction and maintenance of transformed foci in BALB/c 3T3 cells. Cancer Res. 48:3490.

Yamasaki, H.. Hollestein, M., Mesnil, M., Martel, N. and Aguelon, A. M. 1987. Selective lack of intercellular communication between transformed and nontransformed cells as a common property of chemical and oncogene transformation of BALB/c 3T3 cells. Cancer Res. 47:5658.

Yamasaki, H. and Enomoto, T. 1985. Role of intercellular communication in BALB/c3T3 cell transformation. In: Carcinogenesis, vol. 9, A comprehensive Sruvey-Mammalian Cell Transformation: Mechanisms of Carcinogenesis and Assays for Carcinogens. Rave Press. New York, New York, USA.

Yang, R. S. H. 1994. Introduction to the toxicology of chemical mixtures. In: Toxicology of Chemical Mixtures, Yang, R. S. H. (Ed.). Academic Press, San Diego, California, USA.

Yeung, J. M., Wang, H.-Y and Prelusky, D. 1996. Fumonsin Bi induces protein kinase C translocation via direct interaction with diacylglycerol binding site. Toxicol. Appl. Pharmacol. 141:178.

Yoo, H.-S., Norred, W. P., Showker, J. and Riley, R. T. 1996. Elevated sphingoid bases and complex sphingolipid depletion as contributing factors in fumonisin cytotoxicity. Toxicol. Appl. Pharmacol. 138:211.

Yoo, H., Norred, W. P., Wang, E., Merrill, A. H. and Riley, R. T. 1992. Sphingosine inhibition of de novo sphingolipid biosynthesis and cytotoxicity are correlated in LLC-PK1 cells. Toxicol, and Appl. Pharmacol. 114:9.

Yoshizawa, T., Yamashita, A. and Chokerhawom, N. 1996. Occurrence of fumonisins and aflatoxins in com from Thailand. Food Addit. Contam. 13:163.

Yotti, L. P., Chang, C. C. and Trosko, J. E. 1979. Elimination of metabolic cooperation in Chinese hamster cells by a tumor promoter. Science. 206:1089.

265

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Zacharias, C., Van Echten-Deckert, G., Wang, E., Merrill, A. H., and Sandhoff. K.. 1996. The effect of fumonisin Bi on developing chick embryos: Correlation between de novo sphingolipid biosynthesis and gross morphological changes. Glycoconj. J. 13:167.

Zampighi, G. and Unwin, P. N. T. 1979. Two forms of isolated gap junctions. J. Mo. Biol. 135:451.

Zhang, J. L. and Nicholson, B. J. 1989. Sequence and tissue distribution of a second protein of hepatic gap junctions, Cx26, as deduced from its cDNA. J. Cell Biol. 109:3391.

Zuckerman, A. J., Tsiquaye, K. N. and Fulton, F. 1967. Tissue culture of human embryo liver cells and the cytotoxicity of aflatoxins B. Br. J. Exp. Pathol. 48:20.

Zummo N. and Scott, G. E. 1992. Interaction of Fusarium moniliforme and Aspergillus flavus on kernel infection and aflatoxin contamination in maize ears. Plant disease 76:771.

266

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Appendix A: List of Abbreviations

2-AF 2-aminofluorene

AA Arachidonic acid

AFB| Aflatoxin Bi

AFBi Aflatoxin B 2

AFBaa Aflatoxin B 2a

AFGi Aflatoxin Gi

AFG2 Aflatoxin G 2

AFMi Aflatoxin Mi

AFPi Aflatoxin Pi

AFQi Aflatoxin Qi

AP Aminopentol

ATCC American type culture collection

CFSAN Center for Food Safety and Applied Nutrition

DAG Diacylglycerol

DMSO Dimethyl sulfoxide

EC Human esophageal cancer

ELEM Equine leukoencephalomalacia

ER Endoplasmic reticulum

FAO Food and Agriculture Organization of the United Nations

FBi Fumonisin Bi

FB2 Fumonisin B 2

267

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. f b 3 Fumonisin B 3

FDA United States Food and Drug Administration

FMO Flavin containing monooxygenase

G-6 -P Glucose- 6 -phosphate

GJ Gap junction

GJIC Gap junction intercellular communication

GPI Glycosylphosphatidylinositol anchor

GST Glutathione-S-transferase

HPLC High pressure liquid chromatography

Ksa 3 -ketosphinganine

LC Liquid chromatography

LO Lipoxygenase

LY Lucifer yellow

MFR Membrane folate receptor

MSD Mystery swine disease

MW Molecular weight

NADP P-nicotinamide-adenine-dinucleotide-phosphate

N-Acyl-So Ceramides

NTD Neural tube defects

NR Neutral red

OPA orto-phtaldehyde

P450 Cytochrome P450

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Pal-CoA Palmitoyl-CoA

PBS Phosphate buffer solution

PBS-Ca, Mg Phosphate buffer solution without calcium or magnesium

PBS+Ca, Mg Phosphate buffer solution with calcium or magnesium

PCR Polymerase chain reaction

PDA Potato dextrose agar

PKC Protein kinase C

PHF Partially hydrolyzed fumonisin

PHS Prostaglandin H synthetase

PPE Porcine pulmonary edema

PUFA Polyunsaturated fatty acids

S9 Sprague-Dawley rat liver post-mitochondrial supernatant

Sa Sphinganine

Sa-P Sphinganine-phosphate

SEM Standard error mean

So Sphingosine

So-P Sphingosine-phosphate

TCA Tricarballilic acid

TFA Trifluoroacetic acid

TLC Thin layer chromatography

269

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Appendix B: Raw data for interaction study

Experiment 1 AFB1/FB1 Combinations Salmonella tester strain TA-100 with S9 Sample information is presented on Table 4.1 NA: Not available

Sample plate a plate b plate c Mean SEM Nat Rev. 152 154 154 153 1

Fumonisin A 1 ng 138 150 148 145 4 B lO ng 155 260 328 248 50 C 100 ng 276 200 182 219 29 DIOOOng 168 176 NA 172 57 E l 0000 ng 179 183 168 177 4 F 100,000 ng 169 98 196 154 29

Aflatoxin + Fumonisin l(lngAFBl) 116 136 141 131 8 1A 224 180 93 166 38 IB 187 132 145 155 17 1C 184 198 182 188 5 ID 179 169 187 178 5 IE 135 155 144 145 6 IF 180 126 168 158 16

2 (10 ng AFB1) 214 207 190 204 7 2 A 174 181 208 188 10 2B 214 195 204 204 5 2C 428 266 348 347 47 2D 224 360 296 293 39 2E 236 218 206 220 9 2F 216 184 198 199 9

3 (100 ng AFB1) 536 528 516 527 6 3A 620 468 417 502 61 3B 508 548 468 508 23 3C 556 432 504 497 36 3D 1704 1512 1500 1572 66 3E 1356 1308 992 1219 114 3F 536 470 512 506 19

4 (lOOOng AFB1) 2988 2216 2616 2607 223 4 A 1824 1776 1764 1788 18 4B 1128 1332 1332 1264 68 4C 1220 924 952 1032 94 4D 1488 1752 1632 1624 76 4E 1380 1392 1500 1424 38 4F 2268 1788 NA 2028 240

270

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. AFBi/FBi Combinations Salmonella tester strain TA-98 with S9 Sample information is presented on Table 4.1 NA: Not available

Sample plate a plate b plate c Mean SEM

Nat. Rev. 28 24 26 26 1

1 30 32 31 31 1 1A 36 30 27 31 3 IB 35 39 42 39 2 1C 31 31 NA 31 0 ID NANA 43 43 14 IE 45 42 52 46 3 IF 41 46 36 41 3

2 39 44 37 40 2 2A 53 12 27 31 12 2B 21 NA NA 21 0 2C 35 52 45 44 5 2D 29 52 32 38 7 2E 43 47 43 44 1 2F 38 44 44 42 2

3 248 158 376 261 63 3A 132 236 208 192 31 3B 120 NA NA 120 0 3C 78 82 96 85 5 3D 152 228 168 183 23 3E 200 118 216 178 30 3F 200 146 244 197 28

4 972 1092 1020 1028 35 4A 864 1212 1394 1157 155 4B 960 720 936 872 76 4C 468 672 408 516 80 4D NA NA 588 588 196 4E 744 900 924 856 56 4F 1080 1296 972 1116 95

A 31 27 36 31 3 B 22 10 24 19 4 C 39 35 41 38 2 D 35 56 40 44 6 E 27 47 40 38 6 F 31 50 39 40 6

271

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Experiment 2 AFB1/FB 1 Combinations Salmonella tester strain TA-100 with S9 Sample information is presented on Table 4.1 NA: Not available

Sample plate a plate b plate c Mean SEM

Nat. Rev. 352 388 348 363 13

1 336 356 382 358 13 1A 460 392 420 424 2 0 IB 415 512 392 440 37 1C 496 400 404 433 31 ID 480 456 448 461 1 0 IE 428 556 480 488 37 IF 432 488 464 461 16

2 440 508 476 475 2 0 2A 388 392 468 416 26 2B 472 664 NC 568 197 2C 432 284 428 381 49 2D 432 364 360 385 23 2E 512 520 392 475 41 2F 616 584 548 583 2 0

3 664 720 788 724 36 3A 416 500 720 545 91 3C 472 612 504 529 42 3D 572 636 856 6 8 8 8 6 3E 684 720 716 707 11 3F 1420 736 972 1043 2 0 1

A 456 440 428 441 8 B 2976 3648 3504 3376 204 C 528 408 484 473 35 D 436 508 476 473 2 1 E 628 520 480 543 44 F 512 556 560 543 15

272

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. AFB1/FB1 Combinations Salmonella tester strain TA-100 without S9 Sample information is presented on Table 4.1 NA: Not available

Sample plate a plate b plate c Mean SEM

Nat. Rev. 154 144 172 154 8

1 146 148 177 157 1 0 1A 139 135 137 137 1 IB 146 135 166 149 9 1C 189 143 169 167 13 ID 97 126 143 1 2 2 13 IE 126 117 165 136 15 IF 126 142 107 125 1 0

2 118 134 149 134 9 2A 106 105 143 118 13 2B 157 1 2 1 141 140 1 0 2C 151 154 173 159 7 2D 163 158 153 158 3 2E 151 156 171 159 6 2F 181 161 163 168 6

3 61 1 2 0 128 103 2 1 3A 125 130 1 0 1 119 9 3B 176 147 169 164 9 3C 214 158 142 171 2 2 3D 147 199 181 176 15 3E 125 119 127 124 2 3F 127 152 162 147 1 0

A 127 125 1 2 1 124 2 B 143 173 157 158 9 C 95 105 1 0 2 1 0 1 3 D 143 126 174 148 14 E 170 157 169 165 4 F 153 116 158 142 13 273

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Experiment 3 AFB1/FB1 Combinations Salmonella tester strain TA-100 with S9 Sample information is presented on Table 4.1 NA: Not available

Sample plate a plate b plate c Mean SEM

Nat.Rev. 189 199 173 187 8

1 126 117 292 178 57 IA 127 146 133 135 6 IB 121 145 104 123 12 1C 178 131 168 159 14 ID 160 163 176 166 5 IE 36 32 34 34 1 IF 119 132 117 123 5

2 133 163 159 152 9 2A 142 133 116 130 8 2B 141 161 165 156 7 2C 153 152 163 156 4 2D 172 168 143 161 9 2E 632 199 215 349 142 2F 126 130 155 137 9

3 236 226 228 230 3 3A 129 132 129 130 1 3B 178 181 144 168 12 3C 152 137 151 147 5 3D 324 198 228 250 00 3E 158 171 182 170 7 3F 213 285 167 222 34

4 311 461 552 441 70 4A 392 336 202 310 56 4B 271 468 324 354 59 4C 396 456 396 416 20 4D 544 536 528 536 5 4E 720 672 968 787 92 4F 816 928 580 775 103

A 106 123 204 144 30 B 148 116 145 136 10 C 145 125 130 133 6 D 45 131 123 100 27 E 148 135 157 147 6 F 166 184 169 173 6

274

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. AFBj/FBi Combinations Salmonella tester strain TA-98 with S9 Sample information is presented on Table 4.1 NA: Not available

Sample plate a plate b plate c Mean SEM

Nat .Rev. 27 36 34 32 3

1 28 21 31 27 3 1A 33 30 24 29 3 IB 14 23 21 19 3 1C 23 20 27 23 2 ID 26 36 24 29 4 IE 35 30 32 32 1

2 29 27 28 28 1 2A 28 25 19 24 3 2B 162 185 169 172 7 2C 29 18 33 27 4 2D 28 35 23 29 3 2E 26 23 29 26 2

3 22 40 31 31 5 3A 29 23 36 29 4 3B 35 31 22 29 4 3C 25 21 33 26 4 3D 31 31 37 33 2 3E 37 28 30 32 3

4 71 66 54 64 5 4A 77 59 80 72 7 4B 67 53 48 56 6 4C 67 54 65 62 4 4D 127 113 158 133 13 4E 203 118 119 147 28

A 32 22 21 25 4 B 31 25 27 28 2 C 17 24 29 23 3 D 20 30 21 24 3 E 22 30 24 25 2 275

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Appendix C: Letter of permission from Marcel Dekker, Inc.

I L o u i s i a n a S t a t e U n i v e r s i t y LSU Department of food Science

May 1,1998

Ms. Julia Mulligan Marcel Dekker 270 Madison Ave. New York, NY 10016

Dear Ms. Mulliean

I am writing to you regarding our contribution of the chapter “Effectiveness o f Post-Harvest Procedures in Management o f Mycotoxin Hazards" that will be printed in the book Mycotoxins in Agriculture and Food Safety (Sinha/Bhatnagar, Eds.). I would like to request your permission to include this chapter as part of my doctoral dissertation at Louisiana State University. In order to print this chapter in my dissertation, I need a written letter o f permission from your company. This letter would be included as an appendix in the final document The dissertation will be submitted to Louisiana State University Graduate School in June and will be copyrighted.

Please let me know if you need more information regarding this matter. I appreciate your help, and look forward to hearing from you soon. C £ € ^ ^

Sincerely yours.

Rebeca Lopez-Garcia Ph.D. Candidate PERMISSION GRANTED with Vm imoonuuemd th«t CKoo*' credit M (non to tiercel M w Inc. TV'-Cl A. j «*—..> 0 Lot mould mdude: • > nunc 1*1. TITLE O f BOO* OR iCHRNAL «______Number _ . I t o m l O m c me. * T

1W»£ vi, rtf* Tear 0< PubhCitnn Gcct -tun to be leprinted ihmAd ti* ime* Rcpnntad Irom Rd I L P *>» V Mated O iU arhe.______

276

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Vita

The author was bom on January 1, 1971, being second child to Dr. Raul B.

Lopez-Garcia and Mrs. Rebeca Gomez-Deses de Lopez-Garcia and sister to Mr.

Raul B. Lopez-Garcia. In 1995. she graduated with high honors from La

Universidad La Salle in Mexico City, Mexico, with a major in Chemistry and a

concentration in Food Science. Immediately after, she enrolled in Louisiana State

University to pursue a doctoral degree in Food Science with the Interdepartamental

Concentration in Toxicology. Her research interests focused on the study of

possible interactions in multi-toxin systems, in particular, aflatoxin Bi and

fumonisin Bi in com; and the development of industrial decontamination

procedures. During her doctoral program, the author became highly involved with

the Institute of Food Technologists where she held several local and national

leadership positions. In 1996, she was conferred membership to the Omicron Delta

Kappa National Leadership Honor Society. Throughout her program, she received

fellowship awards from The Institute of Food Technologists and The American

Association of Cereal Chemists. Her studies were supported by the Mexican

government through El Consejo Nccional de Ciencia y Tecnologia (CoNaCyT) and

by Louisiana State University through a graduate tuition award. Additionally she

held research and teaching assistant positions in the Department of Food Science,

Louisiana State University. Currently, she is a candidate for the degree of Doctor of

Philosophy, which will be conferred in August, 1998.

277

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. DOCTORAL EXAMINATION AND DISSERTATION REPORT

Candidate: Rebeca Lopez-Garcia

Major Field: Food Science

Title of Dissertation: Aflatoxin and Fumonisin Co-contamination: Interactive Effects, Possible Mechanisms of Toxicity, and Decontamination Procedures.

Approved:

EXAMINING COMMITTEE:

C "'C,

r D c > \

Date of Examination:

June 29, 1998______

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. IMAGE EVALUATION TEST TARGET (Q A -3 ) ✓

v ^

7 ,

L i HZ8 m 1.0 Uh lii 1 »«l u ££ m i±° lllll 2.0 i . i 1.8

1.25 1.4 1.6

150mm

IIN/WGE. Inc 1653 East Main Street Rochester, NY 14609 USA Phone: 716/482-0300 Fax: 716/288-5989

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.