Aflatoxin-Producing Fungi and Contamination in Zambia
Item Type text; Electronic Dissertation
Authors Kachapulula, Paul W.
Publisher The University of Arizona.
Rights Copyright © is held by the author. Digital access to this material is made possible by the University Libraries, University of Arizona. Further transmission, reproduction or presentation (such as public display or performance) of protected items is prohibited except with permission of the author.
Download date 10/10/2021 19:50:47
Link to Item http://hdl.handle.net/10150/625642 1" "
" " AFLATOXIN-PRODUCING FUNGI AND CONTAMINATION IN ZAMBIA "
by
" Paul W. Kachapulula
______" Copyright © Paul W. Kachapulula 2017
A Dissertation Submitted to the Faculty of the "
SCHOOL OF PLANT SCIENCES "
In Partial Fulfillment of the Requirements " For the Degree of "
DOCTOR OF PHILOSOPHY " WITH A MAJOR IN PLANT PATHOLOGY
In the Graduate College " " THE UNIVERSITY OF"ARIZONA"
2017 "
" " 2" " THE UNIVERSITY OF ARIZONA GRADUATE COLLEGE
As members of the Dissertation Committee, we certify that we have read the dissertation prepared by Paul W. Kachapulula, titled “Aflatoxin-producing fungi and contamination in Zambia” and recommend that it be accepted as fulfilling the dissertation requirement for the Degree of Doctor of Philosophy. " " ______" Date: 3 August 2017" Peter J. Cotty " ______" Date: 3 August 2017" Barry M. Pryor """" ______" Date: 3 August 2017" Marc J. Orbach " ______" Date: 3 August 2017" Monica Schmidt " " " " ______" Date: 3 August 2017" Zhongguo Xiong " " " "
Final approval and acceptance of this dissertation is contingent upon the candidate’s submission of the final copies of the dissertation to the Graduate College.
I hereby certify that I have read this dissertation prepared under my direction and recommend that it be accepted as fulfilling the dissertation requirement.
______"Date: 3 August 2017" Dissertation Director: Peter J. Cotty
" " 3" " STATEMENT BY AUTHOR
This dissertation has been submitted in partial fulfillment of the requirements for an advanced degree at the University of Arizona and is deposited in the University Library to be made available to borrowers under rules of the Library.
Brief quotations from this dissertation are allowable without special permission, provided that an accurate acknowledgement of the source is made. Requests for permission for extended quotation from or reproduction of this manuscript in whole or in part may be granted by the head of the major department or the Dean of the Graduate College when in his or her judgment the proposed use of the material is in the interest of scholarship. In all other instances, however, permission must be obtained from the author.
SIGNED: Paul W. Kachapulula
" " 4" " ACKNOWLEDGEMENTS
I would like to thank my advisor Peter J. Cotty for his guidance, support and encouragement. Thank you for your wisdom and patience. Isaac Newton once said, “If I have seen further, it is by standing on the shoulders of giants”. Thank you for lifting me up.
I am indebted to my committee: Drs. Barry M. Pryor, Marc J. Orbach, Monica Schmidt and Zhongguo Xiong. I have learned a lot from you all both in class and in committee meetings. I appreciate the time you dedicated to me. Thank you.
To my friends and office mates Pummi, Lourena and Connel. Thank you for the laughs we shared. Thank you also for the debates we had on Aspergillus. To my friend Kennedy, thank you for accompanying me on those long drives collecting samples in Zambia. I thank God our paths crossed two decades ago. To Austin, thank you very much for all your technical assistance. Your hard work and outstanding work ethics helped me immensely.
I am grateful to my wife for her love and support. It has been a tough road, but you have made it bearable. Thank you for your encouragement and for the numerous sacrifices you have made to help me complete my training. Mulena mulimu akufuyole ahulu. Nkwagala nyo! To God be all the praise!
" " 5" " Paul W. Kachapulula is a fellow of the Norman E. Borlaug Leadership Enhancement in
Agriculture Program Funded by USAID.
" " 6" " DEDICATION
To Josephine (Babyner), Martha (Mama), Paul Jnr. (Chubby) and Josephine (Pinshe).
You four complete my life. I love you and thank God for you.
" " 7" "
TABLE OF CONTENTS! LIST OF FIGURES……………………………………………………………………...11
LIST OF TABLES………………………………………………………………………13
ABSTRACT……………………………………………………………………………..15
CHAPTER 1 - INTRODUCTION………………………………………………………17
CHAPTER 2 - AFLATOXIN CONTAMINATION OF GROUNDNUT AND MAIZE
IN ZAMBIA: OBSERVED AND POTENTIAL CONCENTRATIONS ...... 20
Introduction……………………………………………………………………………20 Materials and methods ...... 23 Study area ...... 23 Sampling ...... 23 Aflatoxin quantification in ground maize and groundnut ...... 26 Fungal isolation and identification ...... 29 Determining potential for aflatoxin formation after market ...... 29 Aflatoxin-producing ability of fungi from purchased crops ...... 30 Data analysis ...... 30 Results ...... 32 Influences of agroecology and crop host on crop aflatoxin content ...... 32 Aflatoxin formation after simulated poor storage ...... 38 Association of community composition and aflatoxigenicity with increases in crop aflatoxin content after simulated poor storage ...... 42 Aflatoxin-producing ability of fungi from purchased crops ...... 45 Discussion ...... 50 Influences of agroecology on aflatoxin concentration ...... 50 Exposure to aflatoxins through consumption of maize and groundnut ...... 52 Influences of fungal community structure on potential for crop contamination after market ...... 53
" " 8" " CHAPTER 3 - ASPERGILLUS SECTION FLAVI COMMUNITY STRUCTURE IN
ZAMBIA INFLUENCES AFLATOXIN CONTAMINATION OF MAIZE AND
GROUNDNUT ...... 56
Introduction ...... 56 Materials and methods ...... 59 Study area ...... 59 Sampling ...... 59 Isolation and identification of fungi from maize, groundnut and soils ...... 62 Community composition of Aspergillus section Flavi from soils of cultivated and non-cultivated areas ...... 62 Aflatoxin producing potential of A. parasiticus from cultivated fields and non- cultivated areas ...... 63 Data analysis ...... 64 Results ...... 66 Fungi in maize and groundnuts ...... 66 Association between quantity of Aspergillus section Flavi and aflatoxin concentration ...... 73 Aspergillus section Flavi from cultivated and non-cultivated soils ...... 76 Aflatoxin production by A. parasiticus from crops and soils ...... 79 Discussion ...... 83 Composition of Aspergillus section Flavi fungi in maize and groundnuts ...... 83 Association between quantity of Aspergillus section Flavi and aflatoxin concentration ...... 84 Aspergillus section Flavi from cultivated and non-cultivated soils ...... 85 Aflatoxin production by A. parasiticus from crops and soils ...... 86 CHAPTER 4 - AFLATOXIN CONTAMINATION OF DRIED INSECTS AND FISH
IN ZAMBIA...……………………………………………………………………89
Introduction ...... 89 Materials and methods ...... 92 Sampling ...... 92 Species assignment for caterpillars ...... 92
" " 9" " Aflatoxin quantification in ground insects and fish ...... 94 Isolation and identification of fungi from insects and fish ...... 95 Determining potential for aflatoxin formation after market ...... 96 Suitability of insects and fish as substrate for growth and aflatoxin-production by toxigenic Aspergillus section Flavi ...... 97 Data analysis ...... 97 Results ...... 99 Species assignment for caterpillars ...... 99 Aflatoxin in insects and fish ...... 99 Potential for aflatoxin formation after market ...... 100 Fungi from insects and fish ...... 100 Suitability of insects and fish as substrate for growth by toxigenic Aspergillus section Flavi ...... 101 Aflatoxin production on insects and fish ...... 102 Discussion ...... 103" Species assignment for caterpillars ...... 103 Aflatoxin in insects and fish ...... 103 Fungi from insects and fish ...... 104 Potential for aflatoxin formation after market ...... 106 Suitability of insects and fish as substrate for growth and aflatoxin-production by aflatoxigenic Aspergillus section Flavi ...... 107 CHAPTER 5" - AFLATOXIN CONTAMINATION OF DRIED FRUITS IN
ZAMBIA………………………………………………………………………..120"
Introduction ...... 120" Materials ...... 123" Sampling ...... 123 Aflatoxin quantification ...... 123 Isolation and identification of fungi from fruits ...... 124 Wild fruits as substrate for growth and aflatoxin production ...... 125 Data analysis ...... 126 Results ...... 127" Aflatoxin in fruit ...... 127
" " 10" " Fungi from ground fruit ...... 127 Wild fruits as substrate for growth by aflatoxin-producers ...... 128 Wild fruits as substrate for aflatoxin production ...... 128 Discussion ...... 130" Wild fruits contribute aflatoxins to the Zambian diet ...... 130 Aspergillus section Flavi in Zambia’s wild fruits ...... 131 Wild fruits as substrate for growth by aflatoxin-producing fungi ...... 132 Aflatoxin production on wild fruits ...... 133 CHAPTER 6" - FUNGI CLOSELY RELATED TO ASPERGILLUS PARASITICUS IN
ZAMBIA………………………………………………………………………..144"
Introduction ...... 144" Materials and methods ...... 147" Fungal isolates ...... 147 Morphology ...... 147 DNA isolation...... 148 Polymerase chain reaction amplification and phylogenetic analyses ...... 148 Aflatoxigenicity of lineages ...... 150 Growth rates of lineages ...... 150 Data analysis ...... 151 Results ...... 152" Clades of A. parasiticus in Zambia ...... 152 Frequencies of A. parasiticus-like fungi in single fields and forests………………..152 Aflatoxigenicity of A. parasiticus clades ...... 153 Growth rates of A. parasiticus clades ...... 153 Discussion ...... 154" New taxa closely related to A. parasiticus in Zambia ...... 154 Sympatric occurrence of A. parasiticus-like fungi in single fields and forests…….156 Aflatoxigenicity of A. parasiticus clades ...... 157 Growth rates of A. parasiticus clades ...... 157 WORKS CITED………………………………………………………………………..169"
"
" " 11" "
LIST OF FIGURES
Figure 1.1 Map of Zambia showing three agroecologies from which maize and groundnut samples were collected...... 25
Figure 1.2 Regression of crop aflatoxin contamination on rainfall...... 36
Figure 1.3 Regression of crop aflatoxin contamination on temperature…... ……………37"
Figure 2.1 Map of the three agroecologies of Zambia (I, II, and III)...... 61
Figure 3.1 Caterpillars used in the current study as collected from markets in
Zambia………………………………………………………………………………….113
Figure 3.2 Fishes used in the current study as collected from the markets in
Zambia……………………………………………………………………………….…114
Figure 3.3 Caterpillars exiting incubation during the SPSA test with visually evident fungal growth on un-inoculated, incubated caterpillars………………………….……..115
Figure 3.4 Fungal growth on caterpillars inoculated with A. parasiticus…………….………………………………………………………………...116
Figure 3.5 Phylogenetic relationships of caterpillar species from Zambia with known species………………………………………………………………………..…………117
Figure 3.6 Insects and fish in markets in Zambia………………………...……………119
Figure 4.1 Schinziophyton rautanenii and Parinari curatellifolia from the markets in
Zambia…………………………………………………………………………….……136"
Figure 4.2 Growth of Aspergillus section Flavi on inoculated fruits………….…….…137"
" " 12" " Figure 5.1 Phylogenetic relationships of putative A. parasiticus from
Zambia………………………………………………………………………………….159
Figure 5.2 Relationships between A. parasiticus from Zambia and close relatives…..……………………………………………………………………………..161
" "
" " 13" "
LIST OF TABLES
Table 1.1 Aflatoxin in maize and groundnut from 3 agroecologies and 23 districts in Zambia...... 27 Table 1.2 Aflatoxin distribution by category in agroecologies of Zambia ...... 34 Table 1.3 Association between proportions of safe groundnut or maize and agroecology ...... 35 Table 1.4 Aflatoxin increases in safe un-inoculated incubated maize and groundnuts in SPSA assays ...... 39 Table 1.5 Regression analyses of aflatoxin increase as explained by frequency of members of Aspergillus section Flavi community ...... 43 Table 1.6 Mean toxin-producing abilities of section Flavi fungi isolated from crops containing <10 µg kg-1 prior to incubation...... 47 Table 1.7 Comparing aflatoxigenicity of Aspergillus section Flavi isolates from maize and groundnuts ...... 49 Table 2.1 Distribution of fungi of Aspergillus section Flavi on maize ...... 68 Table 2.2 Distribution of fungi of Aspergillus section Flavi on groundnut ...... 70 Table 2.3 Incidence of A. flavus L strain morphotype, S strain morphotype fungi, and A. parasiticus on maize and groundnut in three agroecologies of Zambia…………………………………………………………………………………...72 Table 2.4 Coefficients of determination and other parameters for regression analyses of relationships between crop aflatoxin concentration and the quantity of propagules of A. parasiticus, the L morphotype of A. flavus, and S morphotype fungi……………………………………………………………………………………...74 Table 2.5 Distribution of fungi of Aspergillus section Flavi in non-cultivated and cultivated soils from three agroecologies…………………………………………………………………………….77 Table 2.6 Fungi of Aspergillus section Flavi in non-cultivated and cultivated soils…..…………………………………………………………………………………..78 Table 2.7 Aflatoxin-producing potential of A. parasiticus from crops and from cultivated and non-cultivated soils…………………………………………………………..…………………………..81 Table 3.1 Aflatoxin before and after incubation of insects and fish from Zambia………………………………………………………………………….………109 Table 3.2 Distribution of fungi of Aspergillus section Flavi on edible insects and fish ……….……………………………………………………………………………….…110 Table 3.3 Growth of four aflatoxin-producing fungi on edible caterpillars and fish from Zambia………………………………………………………………………………….111 Table 3.4 Aflatoxin production by 4 aflatoxin-producers on edible caterpillars from Zambia...... …………112
" " 14" " Table 4.1 Common wild fruits in Zambia and their uses………………………....……138 Table 4.2 Aflatoxin in wild fruits from Zambia………………………………………..139 Table 4.3 Distribution of fungi of Aspergillus section Flavi on wild fruits from Zambia…………………………………………………………………………….……140 Table 4.4 Regression analyses of aflatoxin concentration as explained by total quantity of Aspergillus section Flavi in fruits………………………………………………………141 Table 4.5 Propagules of five Aspergillus section Flavi fungi on inoculated wild fruits from Zambia…………………………………………………………………………….142 Table 4.6 Aflatoxin production by 5 aflatoxin-producers on six wild fruit species from Zambia……………………………………………………………………………….....143 Table 5.1 Aspergillus section Flavi isolates used in current study…………………….162 Table 5.2 Primers and annealing temperature for PCR amplfification………..……….165 Table 5.3 Aflatoxigenicity of A. parasiticus groups………..………………………….166 Table 5.4 Growth rates of A. parasiticus groups……………………………………….167 Table 5.5 Frequencies in individual fields and forests of A. parasiticus-like fungi belonging to the four predominant clades………………………………………………168
" " 15" "
ABSTRACT
Aflatoxins are cancer-causing, immuno-suppressive mycotoxins that frequently contaminate important staples in Zambia including maize and groundnut. Managing aflatoxins begins with understanding the distribution of aflatoxins across the target region. Seventeen percent of crops from markets contained aflatoxin concentrations above allowable levels in Zambia, with the frequency of contamination in groundnut and maize highest in warmest regions of the country.
Proper management of aflatoxin contamination requires a clear understanding of the etiologic agents of the observed contamination. Several species within Aspergillus section Flavi have been implicated as causal agents of aflatoxin contamination in Africa.
In Zambia, A. parasiticus was the main etiologic agent of aflatoxin contamination of maize and groundnut, although fungi with S morphology also caused contamination.
Aspergillus flavus L morphotype fungi were associated with reduced aflatoxins, suggesting natural biological control by atoxigenic strains may reduce aflatoxin contamination in Zambia.
In addition to maize and groundnut, wild insects, fruits and fish are important sources of food and incomes in Zambia. Unfortunately, both insects and wild plants are susceptible to aflatoxin contamination. To evaluate the safety of wild insects and fruit, concentrations of aflatoxins and presence of aflatoxin-producers were assessed. Some species of wild fruits and insects were found to have unsafe levels of aflatoxins suggesting mitigation efforts should target these important foods of Zambia in addition to crops such as groundnut and maize.
" " 16" " New lineages of aflatoxin-producing fungi have been described, and found associated with cases of aflatoxicoses in Kenya and elsewhere. Although A. parasiticus is highly frequent and an important etiologic agent of aflatoxin contamination, it is not known how this fungus is related to similar fungi elsewhere. A multigene phylogenetic analysis revealed at least two new groups divergent from known fungal species whose frequencies need to be modified if aflatoxin contamination of crops is to be reduced.
" " 17" "
CHAPTER 1 - INTRODUCTION
Aflatoxins are cancer-causing, immuno-suppressive mycotoxins that frequently contaminate maize, groundnut and several wild fruits and edible insects. These natural poisons degrade the health and prosperity of human populations in several warm regions of the world. Managing aflatoxins begins with understanding the distribution of aflatoxins across the target region. The work presented in appendix 1 sought to understand the distribution of aflatoxins in maize and groundnut across the three agroecologies of Zambia, and to investigate the potential for crops to become contaminated with aflatoxins. This was the first work to describe aflatoxin contamination in all the three agroecologies of Zambia and the first to develop a method for quantifying the risk of aflatoxin contamination as a factor of temperature, rainfall, and fungi present on the crop in storage. Seventeen percent of crops from markets contained aflatoxin concentrations above allowable levels in Zambia, with the frequency of contamination in groundnut and maize highest in warmest regions of the country.
The work presented in the second appendix sought to investigate the causal agents of aflatoxin contamination reported on maize and groundnut in appendix one. Proper management of aflatoxin contamination requires a clear understanding of the etiologic agents of the observed contamination. Several species within Aspergillus section Flavi have been implicated as causal agents of aflatoxin contamination in Africa. In Zambia,
A. parasiticus was the main etiologic agent of aflatoxin contamination of maize and groundnut, although fungi with S morphology also caused contamination. Aspergillus flavus L morphotype fungi were associated with reduced aflatoxins, suggesting natural biological control by atoxigenic strains may reduce aflatoxin contamination in Zambia.
" " 18" " The work presented in appendices three and four sought to understand aflatoxin contamination in fish, edible insects and wild fruits. In addition to maize and groundnut, wild insects, fruits and fish are important sources of food and incomes in Zambia.
Unfortunately, just as with maize and groundnut studied in appendices one and two, both insects and wild plants are susceptible to aflatoxin contamination. To evaluate the safety of wild insects and fruit, concentrations of aflatoxins and presence of aflatoxin-producers were assessed. Some species of wild fruits and insects were found to have unsafe levels of aflatoxins suggesting mitigation efforts should target these important foods of Zambia in addition to crops such as groundnut and maize.
During the course of the studies presented in appendices one through 4, A. parasiticus-like fungi were recovered. However, because the previous descriptions on A. parasiticus-like fungi in appendices one through 4 were based on morphological characteristics, there was need to investigate the said fungi using DNA-based techniques.
New lineages of aflatoxin-producing fungi have been described, and found associated with cases of aflatoxicoses in Kenya and elsewhere. Although A. parasiticus is a highly frequent and important etiologic agent of aflatoxin contamination, it is not known how this fungus is related to similar fungi elsewhere. A multigene phylogenetic analysis revealed at least two new groups of Aspergillus divergent from known fungal species whose frequencies need to be modified if aflatoxin contamination of crops is to be reduced.
Overall, this body of work creates a new context within which human exposure to aflatoxins in Zambia may be addressed. Exposure comes not just from crops, such as maize and groundnut, which attract efforts to mitigate exposure throughout warm regions
" " 19" " of the world, but also from foods that are harvested from nature and not cultivated. In rural Zambia, the extent of exposure is dictated by region of residence with more exposure in warmer, drier areas. The etiology of aflatoxin contamination in Zambia differs from many portions of the world. In Zambia, previously undescribed fungal taxa closely resembling A. parasiticus may be the most important causal agents of aflatoxin contamination.
" " 20" "
CHAPTER 2 - AFLATOXIN CONTAMINATION OF GROUNDNUT AND MAIZE IN ZAMBIA: OBSERVED AND POTENTIAL CONCENTRATIONS
Published in: Journal of Applied Microbiology (2017), Vol. 6 (122), p. 1471-1482
Introduction
Maize and groundnut are preferred crops for both commercial and small-holder farmers in Zambia. More than 80% of the farmers grow maize for self-consumption, sale or both in all three agroecologies (Tembo and Sitko 2013) with maize contributing up to
50% of daily calorie intake (FAO 2014). Groundnut, the second most widely cultivated crop, is also grown in all the agroecologies and international demand makes groundnut an important potential source of income. Groundnut and maize are susceptible to aflatoxin contamination. Heavy dependence on these two crops in Zambia may cause significant aflatoxin-associated health hazards. Liver cancer cases in both Africa and Asia are associated with aflatoxins (Liu et al. 2012). Aflatoxin contamination is caused by crop infection by one or more species of aflatoxin-producing fungi. These fungi disperse from soil, organic matter, and alternative hosts to developing crops. Crop infection and subsequent aflatoxin production are high when conditions are hot and dry during crop development and warm and humid after crop maturation and/or harvest (Cotty and Jaime-
Garcia 2007). Consumption of contaminated food may result in cirrhosis, liver cancer, reduced weight gains in livestock, stunted growth, and/or immune suppression (Turner et al. 2003; Gong et al. 2004; Williams et al. 2004). Severe acute aflatoxicoses that cause liver necrosis and death have been repeatedly documented in Kenya and India (Lewis et al. 2005; Probst et al. 2007; Reddy and Raghavender 2007). Enforcement of regulatory limits on aflatoxin concentrations in foods and feeds causes loss of markets for
" " 21" " agricultural products and reduced income (van Egmond et al. 2007; Wu 2014). Europe and South Africa, with regulatory limits of 4 and 10 µg kg-1 total aflatoxin, respectively, are important potential markets for agricultural commodities from Zambia. The country exported over 8000 metric tons of groundnut to Europe in the 1960s. However, this market collapsed due in part to crops found to be unacceptably contaminated in Europe
(Sitko et al. 2011).
The interplay of climate conditions with cropping systems and fungal community composition influences both the etiology of contamination and potential remedial measures (Cotty et al. 2008; Probst et al. 2010). The three agroecologies of Zambia differ in rainfall and temperature (Bunyolo et al. 1995). Variation among these agroecologies in aflatoxin incidence is underexplored. Risks posed by communities of aflatoxin-producing fungi are estimated in part by determining their average aflatoxin producing potential
(Cotty et al. 2008; Probst et al. 2010), information that is not available in Zambia. The most effective management strategy for aflatoxin is competitive exclusion of aflatoxin- producers by atoxigenic genotypes of A. flavus (Cotty and Bayman 1993). Frequencies of atoxigenic fungi may both contribute to explanations of contamination patterns and provide pools of germplasm from which to choose potential biological control fungi.
In order to expand data on aflatoxin incidences in maize and groundnut across agroecologies in Zambia and to identify causal agents of contamination in these regions, aflatoxin concentrations and infecting fungi were determined in crop samples collected from markets in 27 districts across three agroecologies. Weather variables were found to influence contamination and a method to assess the potential for aflatoxin levels to
" " 22" " increase in end user hands was developed. Continued safety of foods with low aflatoxins was found dependent on associated fungi and post-purchase storage conditions.
" " 23" " Materials and methods
Study area!!
Zambia lies between 80 and 180 South, and 220 and 340 East of the Greenwich meridian and is divided into three agroecologies (Bunyolo et al. 1995). Agroecology III covers northern areas 1100 to 1700 meters above sea level (masl) with annual rainfall
>1000 mm, and average temperature of 160C during the growing season (120 to 150 days between mid-November and the end of March; Bunyolo et al. 1995)." Agroecology II extends through central Zambia 900 to 1300 masl receiving between 800 and 1000 mm annual rain, and average temperature of 23 to 250C during the growing season (100 to
140 days between mid-November and the end of March; Bunyolo et al. 1995).
Agroecology I includes southern parts of Zambia and valleys below 900 masl with less than 800 mm average annual rainfall and 300C average temperature during the growing season (80 to 120 days between mid-November and the end of March; Bunyolo et al.
1995).
Sampling
In total 412 maize (250) and groundnut (162) grain samples were obtained from farm storage of subsistence farmers (22) and markets (390) and imported to the USDA,
ARS, Laboratory in the School of Plant Sciences, University of Arizona under permit number P526P-12-00853 awarded to Peter J. Cotty by the Animal Plant Health
Inspection Service of USDA. Samples originated from 27 districts spanning all three agroecologies (Tables 1.1 and Figure 1.1). Only samples for which retailers could verify local origin of crops were included. Average temperatures during the growing season and
" " 24" " annual rainfall data for the districts in the study were obtained from the Meteorological
Department of Zambia (Dr. K. Munyinda, personal communication).
" " 25" " Figure 1.1 Map of Zambia showing three agroecologies from which maize and groundnut samples were collected to determine distribution of aflatoxins. Dark green = agroecology III, Light green = agroecology II, and Grey = agroecology I. Dots represent sampling points. Scale bar = 0-500 km.
" " 26" " Aflatoxin quantification in ground maize and groundnut
Total aflatoxins were quantified with a GIPSA approved lateral flow immunochromatographic assay (Reveal Q+ for Aflatoxin, Neogen Corporation, Lansing,
MI) following modifications to the manufacturer’s instructions recommended by GIPSA.
Each entire crop (maize and groundnut) sample (350 g to 500 g) was ground with a knife mill (Retsch GM200, Retsch GmbH, Haan, Germany) to pass 75% of the ground material through a 20 mesh sieve, mixed thoroughly, and a 50 g sub-sample was blended with 250 ml of 65% ethanol and the aflatoxin content determined according to the manufacturer’s instructions.
" " 27" " Table 1.1 Aflatoxin in maize and groundnut from 3 agroecologies and 23 districts in Zambia.
Agroecology District No. of Aflatoxin concentration (µg kg-1) samples Maize Groundnut Range I Sesheke 32 22A 40.5A 5.3-621 Livingstone 11 1.4B 5.1B 3.9-6.4 Mean 12X 22X II Mazabuka 10 107.6A 23.4C 1.4-512 Nyimba 6 18B †NA *ND-101.3 Kaoma 51 8.4C 20.7C 3.8-125.1 Choma 15 5.2D 64.7C 1-130.3 Mkushi 3 4.9D NA 3.2-6.5 Senanga 20 4.8D 7C ND-16.4 Vumbwi 4 3.7D NA 1.8-6.2 Serenje 4 3.5D NA 1.6-5.2 Mongu 31 3.3D 285.4B ND-3420 Chadiza 3 2.6D NA 1.7-3.5 Monze 10 2.4D 361.2A 1.5-1192 Kalomo 11 2.3D 3.5C 1.3-6.2 Petauke 8 1.8D NA 1.1-2.9 Kabwe 12 1.6D 20.7C 1-122 Kapiri- 13 1.5D 26C 1.7-116 mposhi Chipata 17 1.3D NA 1-5.5 Chibombo 2 3.6 NA 2.4-4.8 Katete 2 2.6 NA 2.2-3 Mean 11X 90X III Mansa 25 60.5A 6.7A ND-1416 Isoka 4 13.8B NA 4.4-40.2
Mpongwe 5 2.1B 6.1A 2-2.1 Mean 25X(X) 6X(X) Overall 16(X) 39(Y) Means followed by the same letter within each agroecology for each crop are not
significantly different (P<0.05) by Tukey-Kramer’s HSD test. Letters x through y
(without parenthesis) indicate differences among agroecologies, and between maize and
" " 28" " groundnut (in parenthesis) by Tukey-Kramer’s HSD and Wilcoxon’s Signed-Rank tests, respectively.
" " 29" " Fungal isolation and identification
Maize and groundnut samples were weighed, dried to below 8% water content, ground to pass a #12 sieve in a laboratory mill described above, and homogenized. Fungi were recovered from ground crop material using dilution plate technique on modified rose Bengal agar (Cotty 1994). Ground crop material (0.1 to 10 g) was shaken in 50 ml of sterile distilled water for 20 minutes (100 rpm) on a reciprocal shaker. Aliquots (100 µl per plate) of the resulting suspension were spread on 3 plates of modified rose Bengal agar. Plates were incubated (3 days, 310 C, dark) and up to eight colonies of Aspergillus section Flavi were transferred to 5-2 agar (5% V8-juice; 2% agar, pH 5.2) and incubated
(7 days, 310C). Isolations were performed at least twice for each sample. Species and morphotypes were delineated into A. parasiticus, A. flavus L strain morphotype (average sclerotia diameter > 400 µm), and S strain morphotype (average sclerotia diameter < 400
µm) (Cotty 1989) using both macroscopic and microscopic characteristics. Fungi with S strain morphology were separated into SB and SBG based on production of either B or both B and G aflatoxins on maize (below).
Determining potential for aflatoxin formation after market
To determine the potential for aflatoxin concentrations to increase in market maize and groundnut during handling and storage, Simulated Poor Storage Assays
(SPSA) were conducted. Un-inoculated maize (n=80) and groundnut (n=67, Table 1.4) market samples with aflatoxin content below 10 µg kg-1 were thoroughly hand mixed and
10 g of each was placed onto metal sieves (10 cm diameter) in a sealed plastic box containing a moist sponge (4 cm × 4 cm × 4 cm) and incubated (310C, 7 days). After incubation, samples were ground in a blender (Waring 7012S, Waring, Torrington,
" " 30" " Connecticut) containing 50 ml 70% methanol at high speed for 20 seconds. The slurry was allowed to settle (5 min) and 4 µl of the supernatant was spotted directly onto thin- layer chromatography (TLC) plates (Silica gel 60, EMD, Darmstadt, Germany) adjacent to aflatoxin standards (Aflatoxin Mix Kit-M, Supelco) containing known quantities of aflatoxins B1, B2, G1 and G2. Plates were developed in ethyl ether-methanol-water,
96:3:1, air-dried and aflatoxins visualized under 365-nm UV light. Aflatoxins were quantified directly on TLC plates using a scanning densitometer (TLC Scanner 3, Camag
Scientific Inc, Wilmington, N.C.) running winCATS 1.4.2 (Camag Scientific Inc,
Wilmington, N.C.).
Aflatoxin-producing ability of fungi from purchased crops
Fungal isolates from maize and groundnut were assayed for aflatoxin-producing potential on sterile maize and groundnut. A randomly selected set of fungi consisting of
54 A. parasiticus, 36 S strain morphology fungi and 39 A. flavus L strain morphology fungi were inoculated onto undamaged maize and groundnut kernels (10g in 250 ml
Erlenmeyer flask) previously autoclaved for 60 minutes, cooled to room temperature and moisture adjusted to 30%. Each isolate was cultured (7 days, 100% RH, 310C) on both maize and groundnut after inoculation with 1,000,000 freshly-harvested spores from 7- day-old cultures. After incubation, sample cultures were blended in 50 ml of 70 % methanol and aflatoxins were quantified with TLC as previously described.
Data analysis
The total quantity of section Flavi fungi from each sample was calculated as
Colony Forming Units per gram (CFU g-1). Community composition of section Flavi was described as percent of A. flavus L strain morphotype (Cotty 1989) undelineated S strain
" " 31" " morphotype (Probst et al. 2007), and A. parasiticus recovered from each sample.
Quantities of section Flavi members were calculated as percent multiplied by total section Flavi CFU g-1. Aflatoxin-producing ability and aflatoxin content were measured in micrograms per kilogram (µg kg-1). Means were compared using Paired t-test and multiple comparisons were done using Analysis of Variance general linear models and
Tukey’s HSD Test as implemented in JMP 11. 1.1 (SAS Institute, Cary, NC). Association between proportion of crop having >10 µg kg-1 with crop type and agroecology were done using chi-square test of independence as implemented in JMP 11.1.1 (SAS Institute,
Cary, NC). Relationships between crop aflatoxin concentration with temperature and rainfall in 10 districts were investigated using regression analyses. Associations between aflatoxin increase and fungal proportions were investigated using regression analyses as implemented in JMP 11. 1.1 (SAS Institute, Cary, NC). Data were tested for normality and, if required, log transformed to normalize the distribution before analysis. However, actual means are presented for clarity. All tests were performed at α = 0.05. Where transformation did not achieve normality and equal variances, the non-parametric methods, Wilcoxon’s Rank Sum and Signed-Rank tests were applied.
" " 32" " Results
Influences of agroecology and crop host on crop aflatoxin content!!!
The highest average aflatoxin concentration (108 µg kg-1) in maize was detected in Mazabuka district while Chipata had the lowest (Table 1.1). Monze, the district next to
Mazabuka, registered the highest average aflatoxin concentration in groundnut (361 µg
-1 kg ). On average, there were no significant differences detected (F2,16, = 0.94, P = 0.40) in maize contamination among agroecologies (Table 1.1). Similarly, average aflatoxin in groundnuts did not differ significantly (F2,10 = 1.15, P = 0.36) among agroecologies
(Table 1). However, average aflatoxin concentrations were higher by a paired t-test (t13 =
2.45, P = 0.030) in groundnut (39 µg kg-1) than in maize (16 µg kg-1) when agroecologies were not considered (Table 1.1).
Percent samples exceeding the 4 µg kg-1 European regulatory limit for aflatoxin in food was 100% and 73% for groundnut and maize, respectively, in region I, while in region III it was below 30% for both crops (Table 1.2). The regulatory limits for total aflatoxin in crops intended for human consumption in Zambia is 10 µg kg-1. Proportions of maize and groundnut with >10 µg kg-1 total aflatoxins were compared in the three agroecologies. The hypotheses that the proportion of unsafe crop (i.e. >10 µg kg-1) is independent of agroecology and type of crop were tested. There was an association between groundnut safety and agroecology (P < 0.001, Table 1.3) while none was detected for maize (P = 0.1006, Table 1.3). The highest proportion of unsafe crop was in region I (58%) while region III had the least (7%). Proportions of unsafe crop depended
2 on crop type (χ (2, n=291)= 15.009, P < 0.001) and unsafe crop were higher in groundnut
(25%) than they were in maize (8%, Table 1.2).
" " 33" " Rainfall significantly (P<0.001) explained crop aflatoxin content (Figure 1.2) whereby increase in rainfall reduced aflatoxins fitting an exponential decay model
(y=10+232911*e(-0.0141*x) R2 = 0.89). Temperature significantly (P<0.03) explained crop aflatoxin content (Figure 1.3) whereby aflatoxins increased as a function of increase in temperature (y = -8.84 + 0.363x), R2 = 0.55).
" " 34" " Table 1.2 Aflatoxin distribution by category in agroecologies of Zambia
Proportion of samples in category Agroecology Total aflatoxin category (%) (µg kg-1) Groundnut Maize I >100 3.8 (1)* 0 (0) >20 3.8 (1) 20 (3)
>10 57.7 (15) 20 (3)
>4 100 (27) 73.3 (11)
<4 0 (0) 26.7 (4)
II >100 10.6 (10) 1.1 (1) >20 14.9 (14) 3.2 (3)
>10 21.3 (20) 5.3 (5)
>4 51 (48) 41.5 (39)
<4 48.9 (46) 58.5 (55)
III >100 3.3 (1) 3.1 (1) >20 6.7 (2) 9.4 (3)
>10 6.7 (2) 9.4 (3)
>4 26.7 (8) 21.9 (7)
<4 73.3 (22) 78.1 (25) Overall >10 25 8 *Values in parentheses refer to number of samples in category
" " 35" " Table 1.3 Association between proportions of safe groundnut or maize and agroecology
Agroecology Crop Aflatoxin safety category Total Safe* Unsafe* I Groundnut †11 (42%) 15 (58%) 26 Maize 12 (80%) 3 (20%) 15 II Groundnut 74 (79%) 20 (21%) 94 Maize 89 (95%) 5 (5%) 94 III Groundnut 28 (93%) 2 (7%) 30 Maize 29 (91%) 3 (9%) 32 *Samples below 10 µg kg-1 (µg kg-1) were considered safe, and those above as unsafe
(Regulatory limit for Zambia)
†Numbers inside and outside parenthesis refer to number of samples and proportion respectively in the category. Proportions were compared for each crop using the
Freeman-Halton test. P <0.001 for groundnut and 0.1006 for maize indicating presence of an association between the proportion of safe groundnut and agroecology but not maize.
" " 36" " Figure 1.2 Regression of crop aflatoxin contamination on rainfall in 10 districts of
Zambia. Y=10+232911*e(-0.0141*X); R2 = 0.89; P < 0.001
" " 37" " Figure 1.3 Regression of crop aflatoxin contamination on temperature in 10 districts of
Zambia. Y = -8.84 + 0.363X; R2 = 0.55; P < 0.05
" " 38" " Aflatoxin formation after simulated poor storage
Increases in aflatoxin content of several magnitudes were observed in both maize and groundnut purchased from markets and incubated at 310C and 100% RH (Table 1.4).
These increases occurred regardless of the agroecology from which the crops originated.
In most samples, all four aflatoxins were detected, with total aflatoxin increasing at least
-1 -1 1000-fold from 3 µg kg to 4,418 µg kg (t34 = 8.86, P < 0.001) in maize and 30,000-
-1 fold (from 3 to 100,302 µg kg ) in groundnuts (t39 = 12.19, P < 0.001). Most of the previously safe groundnut (87%) and maize (67%) exhibited toxin increases during incubation at high temperature and high humidity. Although both crops developed lethal levels of aflatoxins (Table 1.4), the increases were greater in groundnut than maize (t63, =
3.50, P < 0.001).
" " 39" " Table 1.4 Aflatoxin increases in safe un-inoculated incubated maize and groundnuts in
SPSA assays*
" " 40# #
% Agroecology District Average aflatoxin (µg kg-1) % Maize Average aflatoxin (µg kg-1) Total groundnut in incubated maize showing in incubated groundnuts showing crop
Before** After increase Before After increase aflatoxin (µg kg-1) I Sesheke B ND† 1,328 ND 17,593 1 B ND 29 ND 639 2 G ND 812 ND 8,001 1 G ND 21 ND 259 2
Total 5.9 50 (n=8) 7.8 26,492b(b) 83 (n=6) 28,682a 2,190a(a) II Kaoma B ND 604 ND 65,298 1 B ND 41 ND 9,710 2 G ND 682 ND 13,618 1 G ND 42 ND 3,636 2 Total 6.2 1,369(s)(a) 43 (n=21) 5.3 92,263(s)(b) 95 (n=21) 93,632
Mongu B ND 3,753 - ND 132,384 1 B ND 131 - ND 11,182 2 G ND 1,106 - ND 51,536 1 G ND 59 - ND 4,440 2 Total ND 5,050(s)(a) 76 (n=17) ND 199,541(s)(b) 95 (n=20) 204,591
Senanga B ND 6,731 ND 67,617 1 B ND 276 ND 16,732 2 G ND 3,468 ND 38,904 1 G ND 129 ND 8,202 2 Total 4.8 10,603(s)(a) 82 (n=11) ND 131,455s(b) 100 (n=5) 142,058
Mean 5,674a(a) ND 141,086a(b) 146,760b
III Mansa B ND 1,668 ND 25,212 1 # # 41# #
B ND 88 ND 998 2 G1 ND 1,053 ND 25,112
G2 ND 67 ND 435
Total ND 2,876a(a) 83 (n=23) ND 51,758b(b) 80 (n=15) 54,634a
Average across districts 3x 4,418y 100,302(y) 3(x) * Data is based on aflatoxin produced from uninoculated incubated maize (n=80) and groundnut (n=67) subsamples from the safe crop
(<10 µg kg-1). SPSA = Simulated Poor Storage Assay
**Before and after columns refer to aflatoxin concentration before and after incubation, respectively.
† ND is none detectable (Limit of detection is 2 µg kg-1). Aflatoxin chemotypes before incubation not included because quantities
were too low to detect.
Letters a, b and c separate means across agroecologies (without parentheses) and between maize and groundnut or in each row (in
parentheses). Letters x and y separate means before and after incubation in maize (without parentheses) and groundnut (in
parentheses). Means followed by the same letter are not significantly different (P<0.05) by Wilcoxon’s rank-sum and signed-rank
tests.
# # 42# # Association of community composition and aflatoxigenicity with increases in crop aflatoxin content after simulated poor storage
The association between community composition and aflatoxin increases under simulated poor storage and toxigenicities of associated fungi was investigated as previously described. Both the percent (arcsine transformed) and the quantity (log CFU g-
1) of the A. flavus community composed of the L strain morphology fungi inversely explained the percent increase in crop aflatoxin content in groundnut during incubation
(300C, 100% RH) (for proportion, log y = 11.527615-5.109288x, R2=0.55, P<0.001; for quantity, log y = 11.509575 - 1.135883x, R2=0.34, P<0.001). The quantity of S strain morphotype explained increases in aflatoxin in incubated groundnut (log y = 6.687114 +
1.0997904x, R2=0.31, P=0.0015) while that of A. parasiticus did not. The total quantity of fungi did not explain aflatoxin increases in incubated groundnut. Aflatoxin increases in incubated maize was not explained by either proportion or quantity of any of the section
Flavi fungi investigated (Table 1.5).
# # 43# #
Table 1.5 Regression analyses of aflatoxin increase as explained by frequency of members of Aspergillus section Flavi community*
Coefficient of Model significance Community component Intercept Rate of increase† determination (P) †† (R2) Groundnut % L** 11.527615 -5.109288 0.548 <0.0001 Quantity of L (CFU g-1) 9.6943513 -1.135883 0.338 0.0007 % P 8.3508565 2.3851836 0.064 0.1791 Quantity of P (CFU g-1) 7.2373955 0.8249238 0.121 0.06 % S 7.6824925 3.3907445 0.0143 0.196 Quantity of S (CFU g-1) 6.687114 1.0997904 0.308 0.0015 Total fungi (CFU g-1)*** -0.047534 0.0001 0.9489 Maize
% L 11.527615 -1.213759 0.023 0.3819 Quantity of L (CFU g-1) 4.5281037 0.3735226 0.04 0.2473 % P 5.9776435 -0.627793 0.001 0.8835 Quantity of P (CFU g-1) 5.8385679 0.13752 0.004 0.7224 % S 5.878224 0.7943146 0.004 0.7089 Quantity of S (CFU g-1) 5.5567231 0.6588243 0.08 0.0991 Total fungi (CFU g-1) 0.4543727 0.053 0.1831 * Data is based on 89 and 67 maize and groundnut samples, respectively, with aflatoxin concentration <10 µg kg-1
**L, P and S represent A. flavus L strain morphotype, A. parasiticus and S strain morphotype fungi, respectively. ***Total fungi refers to two morphotypes plus A. parasiticus combined. Percent occurrence data was arcsine-transformed while CFU g-1 was log- transformed prior to analyses.
# # 44# #
†This value represents the change in aflatoxin for a unit change in percentage or CFU g-1 of crop. Negative values reflect aflatoxin reduction.
††Significance set at P = 0.05.
# # 45# # Aflatoxin-producing ability of fungi from purchased crops
Quantification of the relative aflatoxin-producing potential of 51 A. flavus L strain morphotype (33 isolated from maize and 18 from groundnut), 54 A. parasiticus (28 isolated from maize and 26 from groundnut), and 38 S strain morphotype fungi (16 isolated from maize and 22 from groundnut) obtained from samples used in the incubation experiments was done on both maize and groundnut as previously described.
Ten (3 from maize and 7 from groundnut) of the S strain morphotype fungi produced only B aflatoxins (thus designated SB) and 28 (13 from maize and 15 from groundnut) produced both B and G aflatoxins (thus designated SBG) (Table 1.6). There were significant differences in aflatoxin B1 (F3,139 = 41.50, P<0.001) and total aflatoxin (F3,139
= 51.55, P<0.001) production among the section Flavi members. On groundnut the average total aflatoxin produced by isolates of A. parasiticus (237,000 µg kg-1) was significantly higher (P < 0.0126) than that produced by G aflatoxin producing S strain morphotype fungi (91,455 µg kg-1) by Student’s t-test. Quantities of aflatoxins produced on groundnut by S strain morphotype fungi that produced only B aflatoxins (4,157 µg kg-1) did not differ significantly (P=0.139) from that produced by A. flavus L strain morphotype isolates (4,168 µg kg-1); although each produced significantly less aflatoxins than either A. parasiticus (P < 0.001 for SB and for A. flavus L strain morphology) and S strain morphotype fungi that produced both B and G aflatoxins (P=0.0051 and P<0.001 for SB and A. flavus L strain morphotype, respectively) by Student’s t-test. Unlike on groundnut, the total aflatoxin produced by SBG (265,748 µg kg-1) and A. parasiticus
(192,398 µg kg-1) on maize did not differ significantly (P = 0.5187, Student’s t-test), but both taxa produced significantly more aflatoxins than the other taxa (Table 1.6).
# # 46# # Aflatoxin production by A. parasiticus did not significantly differ between maize and groundnut (t53 = 0.14, P = 0.8912) by paired t-test. However, significantly greater quantities of aflatoxins were produced on maize than groundnuts by both fungi with S strain morphology and the A. flavus L strain morphotype (P <0.001; Table 1.6). Fungi produced comparable amounts of aflatoxin irrespective of crop of origin (Table 1.7).
# # 47# #
Table 1.6 Mean toxin-producing abilities of section Flavi fungi isolated from crops containing <10 µg kg-1 prior to incubation.
No. of Crop average Taxon* Average Aflatoxin (µg kg-1) isolates† (µg kg-1) Maize Range Groundnut Range a(x) a(x) a P 33, 18 B1 80,104 5,527 - 219,006 136,098 2,166 - 3,048,587 108,101 B 2,232 ND - 14,485 2,317 ND - 9,042 2 G 106,917 123 - 761,700 96,397 349 - 356,995 1 G 3,146 ND - 29,233 2,272 ND - 13,046 2 Total 192,398 (a)(x) 7,408-497,384 237,085 (a)(x) 2,642-3,188,272 214,742(a) a(x) b(y) ab SBG** 13,15 B1 118,583 146-1,038,204 32,399 125-251,500 75,491 B 1,759 ND-9,715 547 ND -5,599 2 G 143,685 ND-1,415,343 57,252 ND -613,519 1 G 1,721 ND-11,398 1,257 ND -19,021 2 Total 265,748(a)(x) 248-2,453,547 91,455(b)(y) 125-814,764 178,602(a) b(x) c(y) bc SB** 3,7 B1 40,780 ND-120,298 4,008 ND-14,069 22,394 B 2,017 ND-4,409 149 ND-932 2 G ND†† ND ND ND 1 G ND ND ND ND 2 Total 42,798(b)(x) ND-124,489 4,157(c)(y) ND-15,001 23,477(b)
c(x) c(y) c L 26,18 B1 12,888 ND-153,433 4,011 ND-58,392 8,450 B 838 ND-10,850 157 ND-3,067 2 G ND ND ND ND 1 G ND ND ND ND 2 (c)(x) (c)(y) Total 13,727 ND-164,283 4,168 ND-61,460 8,948(b)
# # 48# #
*Taxon consists of P (A. parasiticus), L (A. flavus L strain morphotype) and S strain morphotype.
**Aspergillus section Flavi fungi with S strain morphology in Southern Africa may be either the S strain morphotype of A. flavus, A. minisclerotigenes, the unnamed taxon SBG from West Africa or the fungus associated with lethal Aflatoxicosis in Kenya
†Number before and after the comma represents number of isolates that originated from maize and groundnut, respectively.
†† “ND” means below the Limit of Detection, LOD (LOD = 20 µg kg-1).
Letters a, b and c separate means in each column for aflatoxin B1 (without parenthesis) and total aflatoxin (in parenthesis) among P (A. parasiticus), L (A. flavus L strain morphotype) and S strain morphotype. The letters x and y separate means in each row. Means followed by the same letter are not significantly different (P<0.05) by Tukey-Kramer’s HSD for between morphotype comparison and student’s t-test for within morphotype comparison.
# # 49# #
Table 1.7 Comparing aflatoxigenicity of Aspergillus section Flavi isolates from maize and groundnuts
Originating No. of Morpho-group aflatoxin on maize aflatoxin on groundnuts substrate isolates
B1 Total B1 Total P Maize 28 93,440a(x) 212,659ax 189,852a(x) 312,464ax Groundnut 26 65,742b(x) 170,579bx 78,210a(x) 155,908ax
f(x) fx f(y) fy SBG Maize 13 122,539 278,374 19,774 74,027 Groundnut 15 115,155f(x) 254,806fx 43,341f(y) 106,560fy j(x) jx j(x) jx SB Maize 2 42,234 45,054 8,298 8,764 j(x) jx j(y) jy Groundnut 6 53,889 56,311 3,914 4,007 L Maize 26 7,211q(x) 7,510qx 2,716q(y) 2,804qy Groundnut 13 36,143q(x) 38,834qx 10,305q(y) 10,746qy Letters a/b, f/g j/k and q/r separate means from the two crops within each morpho-group in the column while the letters x and y compare B1 (in parenthesis) and total aflatoxin (without parentheses) within each row. Means followed by the same letter are not significantly different (P<0.05) by Student’s paired t-test (within each row) or student’s t-test (within each morpho-group in each column).
# # 50# # Discussion
To determine the extent of the problem attributable to aflatoxin contamination of food, both detected concentrations and consumption habits must be taken into consideration (Marasas 1997). In Zambia, the majority of the population consumes maize daily with on average 50% of calories derived from maize based food (FAO 2014).
Groundnuts are an important source of energy in sauces and vegetables and as a snack and are both produced and consumed across the nation. Thus, unacceptable aflatoxin contents in 17 % of these primary staple crops from markets, as found in the current study, provides a greater risk to the population compared to regions with higher incidences and concentrations but with reduced rates of consumption and diets that are more diverse. In the current study, sufficient frequencies and concentrations of aflatoxins were detected to support development of aflatoxin management strategies for Zambia based on health concerns and not just the well-established impact of aflatoxins on access to international markets. Successful management strategies developed for Zambia will have to take into account the very high aflatoxin-producing potentials of the fungal communities detected in the current study (Table 1.6).
Influences of agroecology on aflatoxin concentration
Environmental events such as drought, temperature extremes, or rain on the mature crop have large impacts on crop aflatoxin content (Cotty and Jaime-Garcia 2007).
In a similar manner, perennial contamination is often characteristic of production areas with environmental conditions that favor both reproduction of the causative fungi and infection of susceptible crops. Contamination was most frequent and severe in the
# # 51# # warmest production areas of Zambia (Figure 1.3). Aflatoxin is widely distributed in maize and groundnut produced in Zambia (Table 1.1). Unsafe levels of aflatoxins occurred in all three agroecologies with average concentrations above the legal limit of
10 µg kg-1 in all agroecologies for maize (Table 1.1) and agroecologies I and II for groundnut (Table 1.1). Aflatoxin levels do not differ significantly among agroecologies
(Tables 1 and 2; Kankolongo et al. 2009). However, the frequency of unsafe groundnut
(> 10 µg kg-1) depended on agroecology (Table 1.3). The results for groundnut are consistent with climate being an important factor dictating the extent of contamination with the highest proportions of unsafe groundnut in agroecology I (warm and dry) and the lowest agroecology III (wetter and cooler).
The primary climatic differences among the agroecologies in Zambia are temperature and rainfall. Levels of aflatoxin were influenced by rainfall (Figure 1.2) and temperature (Figure 1.3). Aflatoxins increased when temperature increased, and decreased with higher annual quantity of rain resulting in the highest frequencies of unsafe crops in the warmest, driest regions. Low moisture combined with high temperature results in highly stressed plants with increased susceptibility to invasion by aflatoxin-producing fungi (Cotty et al. 1994; Cotty et al. 2008). Warm regions favor growth of aflatoxin-producing fungi (Cotty et al. 1994; Cotty et al. 2008) and stressed plants expend more energy maintaining crop development and less on defense activities such as phytoalexin production (Wotton and Strange 1987). Hot dry conditions cause reduced tissue integrity in developing plants (Odvody et al. 1997) and trigger early onset of developmental processes such as flowering (Doster and Michailides 1995; Hadavi
2005), which creates entry points that allow infection by aflatoxin-producing fungi.
# # 52# # However, rainfall and temperature alone do not adequately explain the observed variation in aflatoxin levels. For example, although Sesheke and Livingstone districts fall in the same agroecology and have comparable temperatures and rainfall, the two districts differed in aflatoxin levels in both maize and groundnut (Table 1.1).
Exposure to aflatoxins through consumption of maize and groundnut
Maize and groundnuts are both important food security crops in Zambia (Sitko et al. 2011; Tembo and Sitko 2013). In the current study, groundnuts had both higher average aflatoxin concentrations and a greater frequency of contamination than maize
(Tables 1.1 and 1.2). However, maize is consumed in higher quantities and at higher frequencies than groundnut, providing up to 50% of daily calorie intake (FAO 2014). As such, aflatoxin levels in maize, even though lower in concentration, pose a greater potential health burden than groundnut contamination. Average aflatoxin concentrations in maize are lower than those frequently reported in Kenya and much lower than those causing lethal acute aflatoxicoses in India and Kenya (Lewis et al. 2005; Reddy and
Raghavender 2007). However, a portion of the maize crop in Sesheke, Monze, Mongu and Mazabuka districts had aflatoxin concentrations sufficient to result in acute lethal aflatoxicosis if those crops served as the primary source of calories (Table 1.1). In the current study, crops were examined over both more diverse environments and greater expanses of Zambia than previously (Kannaiyan et al. 1987; Kankolongo et al. 2009;
Mukanga et al. 2010; Bumbangi et al. 2016) and greater quantities of aflatoxins were detected. These observations indicate a need for interventions to reduce aflatoxins, particularly in the warmer drier regions, where poor crop storage, common among small- scale farmers, may exacerbate contamination (Kankolongo et al. 2009).
# # 53# # Influences of fungal community structure on potential for crop contamination after market
The quantities of aflatoxins both at harvest and at markets may not fully represent the risk of aflatoxin exposure from the crop because crop-associated fungal communities remain with crops until consumption and may produce aflatoxins during handling, storage, and processing (Cotty et al. 1994; Cotty et al. 2008). Fungal communities on crops from each of Zambia’s agroecologies have high average aflatoxin-producing potentials (Table 1.6). Aflatoxins increase in poorly stored crops after harvest (Cotty et al. 1994; Cotty et al. 2008; Jaime et al. 2013). Biocontrol fungi retained on crops after harvest reduce aflatoxin increases in storage (Atehnkeng et al 2014). However, risks of aflatoxin increases attributable to crop-associated fungi after harvest previously have been difficult to quantify. Relative-risk of aflatoxin increases from crop-associated fungi was quantified in the current study with an SPSA. Risk quantified by SPSA varied among crops from 4,418 to 100,302 µg kg-1 (Table 1.4), with increases higher in groundnuts than maize. These aflatoxin risks, and mitigation options, need to be understood by farmers, processors, and end users. Some crops expressed no risk of increase in the SPSA assay (Table 1.4), possibly indicating fungal communities inadequate to support contamination (Cotty et al. 2008; Probst et al. 2010). Presence of atoxigenic A. flavus in fungal communities can prevent postharvest aflatoxin increases
(Atehnkeng et al. 2016).
Aspergillus section Flavi communities from crops subjected to SPSA consisted of the A. flavus L strain morphotype, A. parasiticus and fungi with S strain morphology that produced either only B aflatoxins (SB) or both B and G aflatoxins. Crops with high
# # 54# # frequencies of the L strain morphotype prior to incubation had little or no aflatoxins form during SPSA (Table 1.5). Most A. flavus L strain morphotypes from Zambia were capable of producing little or no aflatoxins (Table 1.6). Thus, the results from SPSA are similar to results from field trials where atoxigenic A. flavus biocontrol agents reduce crop aflatoxin content both prior to and after harvest (Atehnkeng et al. 2014; Atehnkeng et al. 2016). During SPSA groundnut aflatoxin content increases were greatest when high incidences of either S strain morphotype fungi or A. parasiticus were present (Table 1.5).
Both S strain morphotype fungi and A. parasiticus consistently produce high concentrations of aflatoxins (Cotty and Cardwell 1999; Jaime!Garcia and Cotty 2006;
Cotty et al. 2008; Probst et al. 2010).
Aflatoxin increases in SPSA were higher in groundnut than in maize (Table 1.4), even though these crops originated from the same areas. However, more aflatoxins formed in maize inoculated with either SBG, SB or A. flavus fungi than groundnut (Table
1.6). The two crops became similarly contaminated when inoculated with A. parasiticus.
Fungi isolated from maize were just as toxigenic as those originating from groundnut
(Table 1.7). Differential performance of the two crops in SPSA is therefore not attributable to groundnut supporting greater aflatoxin production or containing isolates more toxigenic than maize. This reinforces the above observations that risk of aflatoxin contamination during SPSA, and presumably in the hands of the consumer, is most related to the mix of fungi on the crop. Associations between community composition and aflatoxin increases in the current study may be applied to aflatoxin management in
Zambia. By modifying fungal community composition to increase proportions of atoxigenic L strain morphotype fungi in the field and eventually on the crop, we could
# # 55# # achieve protection not only prior to harvest but also in storage (Atehnkeng et al. 2014;
Atehnkeng et al. 2016).
Aflatoxin contamination of maize and groundnut is common in Zambia and crops purchased with low aflatoxin content are frequently associated with fungi that may form aflatoxins in crops during handling and storage. Aflatoxins occurred in all agroecologies of Zambia with the highest contamination in warm, dry regions. A method for quantifying relative risk of crops to increases in aflatoxin content under poor storage was developed. The assay might be refined by simulating the range of conditions occurring during on-farm storage in regions of concern. Compositions of fungal communities associated with crops pre-storage dictated aflatoxin increases in storage with crops naturally containing atoxigenic A. flavus experiencing smaller increases. Consumers may purchase and keep groundnut and maize for long periods increasing vulnerability to aflatoxin increases. Modifying compositions of fungal communities associated with crops prior to harvest with biological control technology should reduce aflatoxin contamination incidences in warm dry agroecologies and reduce increases when proper handling and storage conditions are not practiced (Atehnkeng et al. 2014; Bandyopadhyay et al. 2016).
# # 56# # CHAPTER 3 - ASPERGILLUS SECTION FLAVI COMMUNITY STRUCTURE
IN ZAMBIA INFLUENCES AFLATOXIN CONTAMINATION OF
MAIZE AND GROUNDNUT
In press: International Journal of Food Microbiology
Introduction
Maize and groundnut are important crops for both commercial and smallholder farmers in Zambia. Maize is cultivated by more than 80% of the farmers in all agroecologies for self-consumption, sale or both (Tembo and Sitko 2013) and contributes up to 50% of daily calorie intake (FAO 2014). Groundnut is the second most widely cultivated crop and is grown in all the agroecologies of Zambia (Tembo and Sitko 2013).
International demand for groundnut provides an important potential source of income.
Groundnut and maize are susceptible to aflatoxin contamination and heavy dependence on these two crops in Zambia may result in significant aflatoxin associated hazards.
Consumption of aflatoxin-contaminated food may cause cirrhosis, liver cancer, stunting, reduced immunity, reduced weight-gain and/or rapid death (Gong et al. 2004;
Lewis et al. 2005; Liu et al. 2012; Probst et al. 2007; Reddy and Raghavender 2007;
Turner et al. 2003; Williams et al. 2004). Enforcement of regulatory limits on aflatoxin concentrations in foods and feeds causes loss of markets for agricultural products and reduced income (van Egmond et al. 2007; Wu 2014). Europe and South Africa, with regulatory limits of 4 and 10 ppb total aflatoxin, respectively, have been important markets for agricultural commodities from Zambia. The country exported over 8000 metric tons of groundnut to Europe in the 1960s. However, this market collapsed due in part to enforcement of aflatoxin regulations in Europe (Sitko et al. 2011). Improved
# # 57# # knowledge of the etiology of aflatoxin contamination in Zambia may reveal management options (Cotty et al. 2008).
Aflatoxin contamination is caused by crop infection by one or more species in
Aspergillus section Flavi. The fungi disperse from soil, organic matter, and alternative hosts to developing crops. Crop infection and subsequent aflatoxin production are high when conditions are hot and dry during crop development and warm and humid after crop maturation and/or harvest (Cotty and Jaime-Garcia 2007). The species most notorious for crop contamination are Aspergillus flavus (produces only B aflatoxins), A. parasiticus
(produces both B and G aflatoxins) and two unnamed taxa SB (only B aflatoxins) and SBG
(both B and G aflatoxins;#Cotty et al. 2008; Probst et al. 2010). Aflatoxin-producers are often sorted on the basis of sclerotial morphology (Cotty 1989). L morphotype fungi produce few large sclerotia (average diameter >400 µm) and S morphotype fungi produce numerous small sclerotia (average diameter <400 µm; (Cotty 1989). Fungi with S morphology frequently produce large quantities of aflatoxins. Molecular phylogenetic studies suggest S morphotype aflatoxin-producers are actually several species: a) A. flavus S strain; b) Lethal Aflatoxicosis Fungus (LAF) SB that severely contaminated maize and led to many deaths in Kenya (Probst et al. 2007); c) the un-named taxon SBG from West Africa (Cotty and Cardwell 1999); and d) A. minisclerotigenes (Pildain et al.
2008). Aspergillus parasiticus is also frequently described as an etiologic agent of groundnut aflatoxin contamination (Horn and Dorner 1998). Although all of these aflatoxin-producers may cause dangerous aflatoxin levels in crops when present in a conducive environment, genotypes vary in average aflatoxin-producing potential and the relative importance of specific etiologic agents may vary from one region to another
# # 58# # (Cotty et al. 2008). Frequencies of aflatoxin-producers on crops and relationships of fungal communities in non-cultivated soils to those resident in cultivated soils have not been characterized in Zambia. Non-cultivated areas, such as forests, may be reservoirs for aflatoxin-producers that may either move into cropping systems or cause contamination of non-cultivated fruits and grains (Boyd and Cotty 2001). Potential causal agents of aflatoxin contamination in cultivated and non-cultivated plants in Zambia need characterization, and the relationship of fungal community structure to aflatoxins in groundnut and maize needs investigation (Kachapulula et al. 2017).
In order to explore possibilities for limiting aflatoxin contamination in Zambia, compositions of Aspergillus section Flavi communities associated with aflatoxin contamination infecting maize and groundnut were explored and aflatoxin production by these communities was characterized and related to Aspergillus section Flavi resident in non-cultivated areas. Aspergillus parasiticus was found to be an important etiologic agent for both maize and groundnut and communities of Aspergillus section Flavi resident in native, non-cultivated areas appear to have influenced compositions of fungi infecting crops.
# # 59# # Materials and methods
Study area!
Zambia lies between 8o and 18o South, and 22o and 34o East of the Greenwich meridian and has three agroecologies designated I, II, and III (Bunyolo et al. 1995).
Agroecology III is the northern most with elevation 1100 to 1700 masl, annual rainfall
>1000 mm, and average annual temperature, 30-33oC (Bunyolo et al. 1995).
Agroecology II covers most of the land in agricultural production and all of central
Zambia. Elevation extends from 900 to 1300 masl with 800-1000 mm annual rain, and
30-32o C average annual temperature. Agroecology I extends across southern Zambia with elevations below 900 masl, <800 mm average annual rainfall, and 30-36o C average annual temperature (Bunyolo et al. 1995).
Sampling
Maize (n=250) and groundnut (n=162) samples from a previous study
(Kachapulula, in press) representing 27 districts and all three agroecologies of Zambia
(Tables 2.1 and 2.2) were included in the current study. In addition, 220 soils were sampled from cultivated fields (n=160) and from non-cultivated areas (n=60), in 16 districts covering all three agroecologies (Figure 2.1). Briefly, at least 4 locations or fields were sampled in each district. Three composite soil samples (100-175 g each) were obtained from each field by scooping soil subsamples at three random locations in each field to a depth of 2 cm (Cotty, 1997). In each agroecology, crop and soil samples were collected during the same trip with sampling occurring during January and May
(agroecologies 1, 2, & 3), and November (agroecologies 1 & 2). Soil and crop samples
# # 60# # were dried in a forced air oven (40°C) to 5-8% water content to prevent fungal growth after receipt and sealed in plastic bags to prevent rehydration. All crop and soil samples were imported to the USDA, ARS. Laboratory in the School of Plant Sciences,
University of Arizona under permit number P526P-12-00853 awarded to Peter J. Cotty by the Animal Plant Health Inspection Service of USDA.
# # 61# # Figure 2.1 Map of the three agroecologies of Zambia (I, II, and III). Filled circles indicate locations from which maize and groundnut samples were collected. Scale bar is in kilometers (Redrawn from Kachapulula et al. 2017).
# # 62# #
Isolation and identification of fungi from maize, groundnut and soils
Maize and groundnut samples were ground in a knife mill (Grindomix GM200,
Retsch GmbH, Haan, Germany) to pass a #12 sieve, and homogenized. Fungi were recovered from ground crop material and dry soil using dilution plate technique on modified rose Bengal agar (Cotty 1994). Briefly, ground crop material and soil (0.1 to 10 g) were shaken in 50 ml sterile distilled water (20 min, 100 rpm) on a reciprocal shaker
(KS-501, IKA Works Inc., Wilmington, NC, USA). Dilution plating was performed on modified rose Bengal agar in triplicate. Plates were incubated (3 days, 31oC, dark) and up to eight colonies of Aspergillus section Flavi were transferred to 5-2 agar (5% V8
Vegetable Juice (Campbell’s Soup Company, Camden, N.J., USA); 2% agar, pH 5.2,
Cotty, 1989). Fungi were stored in sterile water (2 ml) as plugs of sporulating culture after incubation for 7 days at 31oC (Cotty, 1988). Isolations were performed at least twice from each sample. Aspergillus species and strains were identified using both macroscopic and microscopic characteristics (Cotty 1989, 1994; Klich and Pitt 1988; Probst et al.
2007).
Community composition of Aspergillus section Flavi from soils of cultivated and non- cultivated areas
Quantities and community composition of Aspergillus section Flavi from cultivated and non-cultivated areas were compared. The total quantity of section Flavi fungi from each crop and soil sample was calculated as Colony Forming Units (CFU) per gram. Community composition of section Flavi was described as percent of A. flavus L- morphotype (Cotty 1989), undelineated S-morphotype species (Probst et al. 2007), A.
# # 63# # parasiticus, and A. tamarii recovered from each sample. Quantities of section Flavi members were calculated as the percent detected during isolation multiplied by total section Flavi CFU/g.
Aflatoxin producing potential of A. parasiticus from cultivated fields and non- cultivated areas
Aspergillus parasiticus isolates from maize (6), groundnut (6), and either cultivated (16) or non-cultivated soil (35) were assayed for aflatoxin-producing potential on sterile maize and groundnut. Fungi were inoculated onto undamaged, sterile maize and groundnut kernels (10g/250 ml Erlenmeyer flask) previously autoclaved for 60 minutes, cooled to room temperature, and adjusted to 30% water content. The following steps were performed to adjust moisture content of the maize and groundnut kernels post- autoclaving and to inoculate: (1) The initial moisture after autoclaving was measured using an HB43 Halogen Moisture Analyzer (Mettler Toledo, Columbus, OH) and the amount of water needed to raise the moisture of kernels to 30% determined; (2) Conidia of each isolate from 7-day-old cultures (grown on 5% V8-juice; 2% agar, pH 5.2, Cotty
1989) were harvested into sterile, deionized water (10 ml); (3) concentrations of conidia were estimated with turbidity (Orbeco-Hellige turbidimeter TB300IR; Orbeco Analytical
Systems, Farmingdale, NY) and using a nephelometric turbidity unit (NTU)-versus-CFU standard curve, where y is equal to 49,937x (x is NTU, and y is conidia/ml) (Probst et al
2010); (4) A spore suspension containing 1x106 conidia (usually about 500µl) was mixed with water to bring the final volume to that determined in step 1 above, added to 10 g of kernels in a flask and swirled to coat the kernels. Inoculated grains were incubated (7 days, 100% RH, 31oC). After incubation, sample cultures were blended in 50 ml of 70 %
# # 64# # methanol. The slurry was allowed to separate for 30 minutes and the supernatant was spotted directly onto thin-layer chromatography (TLC) plates (Silica gel 60, EMD,
Darmstadt, Germany) adjacent to aflatoxin standards (Aflatoxin Mix Kit-M, Supelco) containing known quantities of aflatoxins B1, B2, G1 and G2. Plates were developed in ethyl ether-methanol-water, 96:3:1, air-dried and aflatoxins visualized under 365-nm UV light. Aflatoxins were quantified directly on TLC plates using a scanning densitometer
(TLC Scanner 3, Camag Scientific Inc, Wilmington, N.C.).
Five market samples found in a previous study to contain >500 ppb total aflatoxins were subjected to aflatoxin analyses by TLC shortly after grinding to evaluate presence of B and G aflatoxins. Fifty grams of ground crop were extracted with 70% methanol (250 ml) and the methanol extract was directly spotted onto TLC plates and separated and quantified as above. The specific aflatoxins were identified by comparison with standards spotted on the same plate.
Data analysis
Statistical analyses were performed with JMP 11.1.1 (SAS Institute, Cary, NC).
Means for fungal frequencies in cultivated and adjacent non-cultivated soils in each district were compared using paired t-test were compared using the paired t-test and multiple comparisons were performed with Analysis of Variance (general linear models) followed by mean separation with Tukey’s HSD test. Relationships between crop aflatoxin concentration and quantities of each member of Aspergillus section Flavi were investigated with regression analyses. Data were tested for normality and, if required, log
(for aflatoxin concentrations and propagules per gram) or arcsine (for percentages)
# # 65# # transformed to normalize distributions. Actual means are presented for clarity. All tests were performed at α = 0.05.
# # 66# # Results
Fungi in maize and groundnuts !
Aspergillus section Flavi was recovered from all maize and groundnut samples. A total of 4,099 isolates were characterized from 412 samples (Tables 2.1 and 2.2). The frequencies of occurrence of Aspergillus section Flavi members in maize differed significantly (ANOVA, F3,48=18.6842, P<0.001) across agroecologies, with the A. flavus
L-morphotype dominating communities (60%), followed by A. parasiticus (24%) and S- morphotype fungi (16%, Table 2.1). In all agroecologies, the A. flavus L-morphotype was the most common member of Aspergillus section Flavi on maize making up 91%, 44%, and 45% of section Flavi in agroecologies III, II, and I, respectively. L morphotype frequencies in region III were higher than either region II or I (Tukey’s HSD, P<0.05).
Aspergillus parasiticus and fungi with S morphology were more common in agroecologies II and I (agroecology averages = 22 to 34% of section Flavi) than in agroecology III (average 4 to 6%).
In groundnut, there were also significant differences in Aspergillus section Flavi across agroecologies (ANOVA, F3,48=36.6726, P<0.001), with A. parasiticus dominating
(42%), then S-morphotype fungi (32%) and A. flavus L strain morphotype (26%, Table
2.2). Aspergillus flavus L strain morphotype frequencies on groundnut were higher in agroecologies III (35%) and I (35%) than II (7%, Tukey’s HSD, P=0.021) whereas the S- morphotype was more common in agroecology II (50%) than in I (31%) and III (17%,
Tukey’s HSD P=0.0048). Aspergillus parasiticus was equally prevalent on groundnut from all agroecologies (P=0.5512, Tukey’s HSD, Table 2.2).
# # 67# # Frequencies of section Flavi members differed between maize and groundnut
(Table 2.3) with the A. flavus L-strain morphotype higher (t103=8.468044, P<0.001) in maize (60%) than groundnut (26%, Table 2.3) and A. parasiticus higher (t103=3.97205,
P<0.001) in groundnut (42%) than maize (24%; Table 2.3). Fungi with S morphology followed the same trend as A. parasiticus (Table 2.3). In agroecology I, each section
Flavi member occurred at similar frequency on maize and groundnut (Table 2.3).
# # 68# #
Table 2.1 Distribution of fungi of Aspergillus section Flavi on maize†
# of Agroecology District % L* % S % P % T CFU/g isolates III Mansa 494 88 5 7 0 603 Mpongwe 33 93 7 0 0 27
Average†† 91a(x) 6a(y) 4b(y) 0a(y) 315
II Choma 111 20 24 56 0 12 Kabwe 125 31 22 47 0 13
Kalomo 95 52 6 42 0 2,080
Kaoma 244 61 18 15 6 843
Kapiri- 148 72 0 28 0 13 mposhi Mazabuka 70 6 70 24 0 41,167
Mongu 180 79 2 19 0 37
Monze 92 0 40 60 0 126
Senanga 152 73 14 13 0 626
Average†† 44b(x) 22a(x) 34a(x) 1a(y) 4,991
I Livingstone 68 30 22 48 0 146 Sesheke 150 59 21 20 0 686,602
Average†† 45b(x) 22a(xy) 34a(xy) 0a(y) 343,374
# # 69# #
Across 60(x) 16(y) 24(xy) 0(z) agroecology *L, S, P and T represent A. flavus L-morphotype, S-morphotype fungi, A. parasiticus and A. tamarii, respectively. †Percent data were arcsine transformed and CFU/g data were log transformed prior to analyses. Values followed by the same letter in each column (a,b,c) or row (x,y,z) do not differ by Tukey’s HSD test (α= 0.05). ††Average percentages for locations in each district were used for analyses and only district averages are presented.
# # 70# #
Table 2.2 Distribution of fungi of Aspergillus section Flavi on groundnut†
# of Agroecology District % L* % S % P % T CFU/g isolates III Mansa 359 50 19 31 0 113 Mpongwe 53 20 14 66 0 27
Average†† 35a(xy) 17b(y) 49a(x) 0a(z) 70
II Choma 98 0 53 47 0 12,572 Kabwe 126 20 44 36 0 576
Kalomo 88 0 44 56 0 590
Kaoma 374 8 43 48 1 521
Kapiri- 99 1 34 65 0 362 mposhi Mazabuka 81 0 64 36 0 7,806
Mongu 353 27 43 30 0 48,098
Monze 123 0 57 43 0 32,697
Senanga 124 4 65 31 0 110
Average†† 7b(y) 50a(x) 44a(x) 0a(y) 11,481
I Livingstone 101 51 43 6 0 7,926 Sesheke 158 19 19 60 2 70
Average†† 35a(x) 31b(x) 33a(x) 1a(y) 3,998
# # 71# #
Across 26(y) 32(x) 42(x) 0(z) agroecology *L, S, P and T represent A. flavus L-morphotype, S-morphotype fungi, A. parasiticus and A. tamarii, respectively. †Percent data were arcsine transformed prior to analyses. Values followed by the same letter in each column (a,b,c) or row (x,y,z) do not differ by Tukey’s HSD test (α= 0.05). ††Average percentages for locations in each district were used for analyses and only district averages are presented.
# # 72# # Table 2.3 Incidence of A. flavus L strain morphotype, S strain morphotype fungi, and A. parasiticus on maize and groundnut in three agroecologies of Zambia
Agroecology % L† % S % P Maize Groundnut Maize Groundnut Maize Groundnut III 91*x 35x 6*x 17y 4*y 49x II 44* y 7y 22*x 50x 34*x 44x I 45y 35x 22x 31y 34x 33x †L, S, and P represent A. flavus L morphotype, S morphotype fungi and A. parasiticus, respectively. * = Maize and groundnut values in the same fungus differ by Paired t-test (α= 0.05). Percent data were arcsine transformed before analyses. Values followed by the same letter (x/y) within the same column do not differ by Tukey’s HSD (α= 0.05).
# # 73# # Association between quantity of Aspergillus section Flavi and aflatoxin concentration
Quantities (CFU/g) of the A. flavus L strain morphotype were inversely related to aflatoxin concentration in groundnut from agroecology I (log y = 2.4990935 - 0.09966x,
R2=0.79, P=0.001), but were not related to either groundnut or maize aflatoxin concentrations in the other agroecologies (Table 2.4). Quantities of S-morphology fungi increased with aflatoxin concentrations in maize only from agroecology II (log y =
1.2273858 + 0.243253x, R2=0.37, P<0.001). Aspergillus parasiticus quantities were predictive of aflatoxin concentrations in groundnut in all three agroecologies
(agroecology I, log y = 1.9957586 + 0.1323517x, R2=0.63, P=0.018; agroecology II, log y = 0.4673417 + 0.3513556x, R2=0.30, P<0.001; agroecology III, log y = 0.25685 +
0.2277388x, R2=0.24, P=0.0491) and in maize from agroecology III (log y = 0.1956034
+ 0.510379x, R2=0.57, P<0.001).
# # 74# #
Table 2.4 Coefficients of determination and other parameters for regression analyses of relationships between crop aflatoxin concentration and the quantity of propagules of A. parasiticus, the L morphotype of A. flavus, and S morphotype fungi Community Coefficient of Model rate of Agroecology component* Intercept determination significance increaseX (Quantity of) (R2) (P) Y I Groundnut
Quantity of L (CFU/g) 2.50 -0.10 0.79 0.001
Quantity of P (CFU/g) 1.9957586 0.13 0.63 0.018
Quantity of S (CFU/g) NSZ NS NS NS
Maize
NS NS NS Quantity of L (CFU/g) NS # # #
NS NS NS NS Quantity of P (CFU/g) # # # #
Quantity of S (CFU/g) NS# NS# NS# NS# II Groundnut
NS NS NS NS Quantity of L (CFU/g) # # # #
Quantity of P (CFU/g) 0.4673417 0.35 0.3 <0.001
Quantity of S (CFU/g) NS# NS# NS# NS#
Maize
NS NS NS NS Quantity of L (CFU/g) # # # #
NS NS NS NS Quantity of P (CFU/g) # # # #
Quantity of S (CFU/g) 1.23 0.24 0.37 <0.001 III Groundnut
# # 75# #
NS NS NS NS Quantity of L (CFU/g) # # # #
Quantity of P (CFU/g) 0.26 0.23 0.24 0.049
Quantity of S (CFU/g) NS# NS# NS# NS#
Maize
Quantity of L (CFU/g) NS# NS# NS# NS#
Quantity of P (CFU/g) 0.20 0.51 0.57 <0.001
Quantity of S (CFU/g) NS# NS# NS# NS# *L, P and S represent A. flavus L-morphotype, A. parasiticus and S-morphotype fungi, respectively. CFU/g data was log-transformed prior to analyses. XThis value represents the change in aflatoxin for a unit change in CFU/g of crop. Negative values reflect lowering aflatoxin concentrations associated with increased quantities of fungus. YSignificance set at P = 0.05. ZNS = non-significant.
# # 76# # Aspergillus section Flavi from cultivated and non-cultivated soils
A total of 2,128 Aspergillus section Flavi isolates were obtained from 220 soil samples (Table 2.5). One or more Aspergillus section Flavi species was recovered from all soils. Species differed in overall frequency in both un-cultivated (F3,168=101.2705,
P<0.001) and cultivated (F3,324=113.0661, P<0.001) soils (Table 2.5). Aspergillus parasiticus was the most frequent Aspergillus section Flavi species in both soil types
(58% in cultivated and 69% in non-cultivated soils), and in all agroecologies, with the exception of cultivated soils in agroecology I where A. flavus L-strain morphotype was most frequent (Table 2.5). Aspergillus parasiticus was most frequent in the coolest agroecology, agroecology III, and, overall, least frequent in the agroecology with the highest average temperature and the least rainfall, agroecology I (Table 2.5). The frequency of Aspergillus section Flavi with S morphology did not differ among agroecologies in both cultivated (ANOVA, F2,79=2.7790 P=0.0682) and non-cultivated soils (ANOVA, F2,40=0.3660, P=0.6958, Table 2.5).
Frequencies of the A. flavus L-strain morphotype, S strain morphotype fungi, and
A. parasiticus did not differ significantly (P>0.05, paired t-test) between cultivated and non-cultivated soils regardless of agroecology. However, the overall quantity of
Aspergillus section flavi (CFU/g) was higher (P<0.001) in cultivated (175 CFU/g) than non-cultivated soils (25 CFU/g, Table 2.6).
# # 77" " Table 2.5 Distribution of fungi of Aspergillus section Flavi in non-cultivated and cultivated soils from three agroecologies§
† † Samples Isolates NC CV Temp.†† Rain†† Agroecology CFU/g (#) (#) (oC) (mm) % L* % S % P % T % L % S % P % T III 46 554 7b(y) 6a(y) 86a(x) 1a(y) 4b(y) 8a(y) 88a(x) 0a(y) 85 30-33 >1000
II 152 1280 32ab(y) 3a(y) 57b(x) 8a(y) 9b(yz) 22a(y) 66b(x) 3a(z) 107 30-33 800-1000
I 21 294 33a(y) 3a(y) 64ab(x) 0a(y) 68a(x) 12a(y) 19c(y) 1a(y) 108 30-36 <800
Across 73 709 24(y) 4(yz) 69(x) 3(z) 27(y) 14(y) 58(x) 1(z) 100 agroecologies
*L, S, P and T represent A. flavus L-morphotype, S-morphotype fungi, A. parasiticus and A. tamarii, respectively. †NC = Non- cultivated soil, CV = Cultivated soil. ††Temp. = Average annual temperature and rain = average annual rainfall. §Percent data were arcsine transformed prior to analyses. L, S, P, T indicate A. flavus L morphotype, S morphotype fungi, A. parasiticus and, A. tamarii, respectively. Values followed by the same letter in each column (a,b,c) or row (x,y,z) for non-cultivated and for cultivated soils, do not differ by Tukey’s HSD test (α= 0.05). Districts sampled include Mansa, Mpongwe and Kitwe in agroecology III; Chibombo, Chipata, Chongwe, Kabwe, Kaoma, Kapiri-mposhi, Lundazi, Mkushi, Petauke, Senanga and Serenje in agroecology II; Livingstone and Sesheke in agroecology I.
" " 78" " Table 2.6 Fungi of Aspergillus section Flavi in non-cultivated and cultivated soils§
# of # of % S- % A. District % A. flavus-L % A. tamarii CFU/g samples isolates morphotype parasiticus
NC* CV** NC CV NC CV NC CV NC CV
Mansa 24 338 2 0 0 0 96 100 2 0 33 82 Mpongwe 23 312 12 7 13 12 75 81 0 0 11 179 Chibombo 23 275 0 11 0 46 100 43 0 0 30 511 Chongwe 21 208 23 28 0 1 59 67 18 4 5 101 Kaoma 30 277 76 2 2 8 7 65 15 25 19 20 Senanga 24 273 28 8 10 19 62 73 0 0 10 182 Sesheke 21 294 33 68 3 12 64 19 0 1 65 150 Average 24 281 25 18 4 14 66 64 5 4 25† 175† §Percent data were arcsine transformed and CFU/g data were log transformed prior to statistical comparisons. † = non-cultivated and cultivated values differ by Paired t-test (α= 0.05). *NC = Non-cultivated soil, CV = Cultivated soil
" " 79# # Aflatoxin production by A. parasiticus from crops and soils
On groundnut at 20oC, there were no significant differences (ANOVA,
F3,58=0.8027, P=0.4974, Table 2.7) in concentrations of aflatoxins produced by A. parasiticus isolates from maize (Mean = 123,810 µg/kg), groundnut (Mean = 196,997
µg/kg), cultivated (Mean = 145,613 µg/kg) and non-cultivated soil (Mean = 125,106
µg/kg). Similarly, the four groups of A. parasiticus did not differ in aflatoxin production at all other temperatures (25, 30 and 35oC) even when maize was used as the substrate
(Table 2.7). The highest concentrations of aflatoxin were produced on groundnut at 25oC by A. parasiticus from maize (Mean = 214,321 µg/kg), groundnut (Mean = 215,669
µg/kg), cultivated (Mean = 199,214 µg/kg) and non-cultivated soil (Mean = 196,632
µg/kg) and the least were at 35oC. A similar trend was observed when maize was used as substrate (Table 2.7). At 20oC, higher concentrations of aflatoxins were produced on groundnut than maize by A. parasiticus isolates from maize (Paired t-test, t5=4.120746,
P=0.009), groundnut (Paired t-test, t5=2.961985, P=0.0252), cultivated (Paired t-test, t15=2.838941, P=0.0124) and non-cultivated soil (Paired t-test, t34=2.06039, P=0.0473,
Table 2.7). However, at the other temperatures, aflatoxin production on the two substrates was similar. Aflatoxin production by all isolates combined differed on both groundnut
(ANOVA, F3,12 = 25.1034, P<0.001) and maize (ANOVA, F3,12 = 22.7206, P<0.001) at different temperatures (Table 2.7). Aflatoxin production on groundnut by all isolates was similar at 20oC, 25oC and 30oC and lower at 35oC (Tukeys’s HSD, P<0.001), whereas on maize, higher concentrations were produced at 25oC and 30oC than 20oC and 35oC
(Tukeys’s HSD, P<0.001, Table 2.7). All five highly contaminated maize and groundnut
# # 80# # market samples examined were found to contain both aflatoxins B and G (data not shown).
# # 81# #
Table 2.7 Aflatoxin-producing potential of A. parasiticus from crops and from cultivated and non-cultivated soils
Type Aflatoxin at 200C Aflatoxin at 250C Aflatoxin at 300C Aflatoxin at 350C Source of of isolate aflat Groundnut Maize Groundnut Maize Groundnut Maize Groundnut Maize oxin Maize B1 60,700 26,300 106,500 72,200 73,900 108,300 48,700 64,000
B2 700 1,200 700 4,200 2,800 2,900 16,200 300
G1 62,000 25,000 106,700 96,700 36,400 47,500 11,000 8,100
G2 400 700 400 2,800 1,800 2,000 8,000 200
Total 123,800* 53,200 214,300 175,900 114,900* 160,700 83,900 72,600
Groundnut B1 33,500 33,000 40,400 53,800 73,100 77,800 21,000 34,500
B2 0 700 0 800 0 100 200 0
G1 163,500 73,400 175,300 130,700 68,600 71,600 7,700 14,900
G2 0 400 0 500 0 100 100 0
Total 197,000* 107,500 215,700 185,800 131,700 149,600 29,000 49,400
Agricultural B1 40,600 39,200 53,200 66,500 73,500 63,200 26,200 37,400 Soil B2 900 1,100 3,300 4,000 8,900 2,100 1,200 2,300
G1 103,500 65,300 140,500 125,900 79,100 66,800 7,700 18,200
G2 600 800 2,200 2,800 5,900 1,400 900 1,600
Total 145,600* 106,400 199,200 199,200 167,400 133,500 36,000 59,500
Uncultivated B1 32,000 25,100 53,700 44,500 66,000 73,900 30,700 34,500 Soil B2 3,500 5,300 4,600 3,700 9,400 5,100 1,700 2,300
G1 87,400 48,300 135,200 82,000 58,900 57,700 10,500 15,900
G2 2,200 3,500 3,100 2,500 6,800 3,400 1,200 1,500
# # 82# #
Total 125,100* 82,200 196,600 132,700 141,100 140,100 44,100 54,200
All isolates Total 147,800A* 87,300Y 206,459A 173,415X 141,131A 145,985X 48,244B 58,931Y * = Maize and groundnut values (total aflatoxin) at the same temperature differ by Paired t-test (α= 0.05). Means in each column (total aflatoxin) are not significantly different by ANOVA. Means followed by the same letter (A/B for groundnut and X/Y for maize) for “All isolates” at the different temperatures do not differ by Tukey-Kramer’s HSD (α= 0.05). Data were log transformed prior to analyses.
# # 83# # Discussion
Dangers aflatoxins pose to human health, livestock productivity and trade are widely recognized (Gong et al. 2004; Lewis et al. 2005; Liu et al. 2012; Probst et al.
2007; Reddy and Raghavender 2007; Turner et al. 2003; van Egmond et al. 2007;
Williams et al. 2004; Wu 2014). Recent deaths from consumption of highly contaminated food in Kenya (Lewis et al. 2005; Probst et al. 2007) and Tanzania have increased recognition of the need to understand the etiologic agents of aflatoxin contamination in many parts of sub-Saharan Africa and the world. Although A. flavus, A. parasiticus, and the two unnamed taxa (SB and SBG) are associated with aflatoxin contamination of crops in warm parts of the world, the most important etiologic agents vary from one region to another (Cotty and Cardwell 1999; Horn and Dorner, 1998; Probst et al. 2007). In addition, some aflatoxin-producers have also been recovered from non-cultivated areas
(Boyd and Cotty, 2001) from which they may move into cropped areas, and provide region-specific influences on composition of communities of aflatoxin-producing fungi infecting and contaminating crops. The current study provides insights into compositions of communities of Aspergillus section Flavi in Zambia resident on maize, groundnut and soils. Compositions were found to influence observed aflatoxin concentrations and insights were developed into how communities of Aspergillus section Flavi from non- cultivated areas might shape those observed in cultivated areas and on crops.
Composition of Aspergillus section Flavi fungi in maize and groundnuts
Communities of Aspergillus section Flavi on maize consisted mostly of A. flavus
(Table 2.1), while those on groundnut were dominated by A. parasiticus (Table 2.2).
Aspergillus flavus L strain morphotype is reported to be the most prevalent member of
# # 84# # Aspergillus section Flavi in soils and crops from many areas including maize, groundnut, cottonseed, rice, sorghum and almonds ( Donner et al. 2015; Purcell et al. 1980;
Schroeder and Boller 1973). Aspergillus flavus is also a much more aggressive colonizer than A. parasiticus on groundnut (Horn 2005). The current study stands in contrast to the aforementioned reports in that A. parasiticus occurred in frequencies higher than A. flavus in groundnut and at levels higher than has been observed on maize in East (Probst et al. 2007; Probst et al. 2010) or West Africa (Atehnkeng et al. 2008). The agroecologies in Zambia apparently differ from previously examined systems. This may reflect in part the percent of land in Zambia not in crop production. Indeed in agroecology I, the A. flavus L strain morphotype has displaced A. parasiticus in cultivated but not in non-cultivated soils (Table 2.5). As agriculture becomes more intensive, this trend may become more widespread.
Association between quantity of Aspergillus section Flavi and aflatoxin concentration
To assess the risk that a given fungal group poses to aflatoxin contamination of a crop, the aflatoxin-producing potential and frequency of occurrence of the fungus in contaminated crops need to be considered (Mehl et al. 2012; Probst et al. 2007).
Regression analyses were conducted to examine agents associated with actual contamination events in market places previously reported in maize and groundnut in
Zambia (Kachapulula et al. 2017). Percentages of the Aspergillus section Flavi community composed of the L strain A. flavus were inversely related to aflatoxin concentrations in groundnut in agroecology I (Table 2.4). This suggests that other components of these communities are more important causal agents. Increases in S- morphology fungi positively explained aflatoxin concentrations only in maize from
# # 85# # agroecology II, indicating the potential of these fungi to exacerbate contamination.
However, this was observed in only one area. Prevalence of A. parasiticus positively explained aflatoxin concentrations in groundnut in all agroecologies and in maize from agroecology III (Table 2.4). Atoxigenic A. parasiticus are rare (Horn et al. 1996). All A. parasiticus isolates in the current study were highly toxigenic (Table 2.7). High toxigenicity combined with high prevalence in infected crops suggests that A. parasiticus is the most important etiologic agent of aflatoxin contamination in Zambia. This was further supported by presence of both B & G aflatoxins in 5 highly contaminated maize and groundnut samples from markets.
Aspergillus section Flavi from cultivated and non-cultivated soils
Incidence of A. parasiticus in soils of Zambia were much higher than incidences previously observed in soils from East or West Africa (Donner et al. 2009). Natural dominance of A. parasiticus in the soil could contribute to crop from Zambia having higher levels of this species than has been observed elsewhere (Donner et al. 2009; Horn and Dorner 1998). Horn and Dorner (1998) observed that A. parasiticus incidences were higher in fields cultivated to peanuts than in those where maize was grown. Most of the small scale farms sampled in the current study had mixed cropping, where maize or other cereals were grown in combination with groundnut (Tembo and Sitko 2013). Wide cultivation of groundnut in agroecologies of Zambia (Sitko et al. 2011) may contribute to high incidences of A. parasiticus in the soil and eventually on the crop. Within the agroecologies of Zambia, A. parasiticus incidences were highest in agroecology III and lowest in I (Table 2.5). Soils in the three regions differ in pH and temperature, with region III being the coolest and most acidic (Bunyolo et al. 1995). Low temperature
# # 86# # promotes crop colonization by A. parasiticus (Donner et al. 2015; Horn and Dorner
1998) and could contribute to the higher frequencies of this fungus observed in agroecology III of Zambia (Table 2.5). However, perhaps the most important factor influencing incidences of A. parasiticus in crops in Zambia is the natural distribution of this species as reflected in composition of the fungal community in non-cultivated soils.
Fungi are capable of dispersal under natural conditions and mixtures of different aflatoxigenic and atoxigenic fungi are found in both non-cultivated and cultivated fields
(Boyd and Cotty 2001). In Zambia, many small-scale farmers grow their crops adjacent to national forests. The current study reports similar patterns of community composition in non-cultivated and cultivated areas (Table 2.6) suggesting that Aspergillus species endemic to non-cultivated areas in Zambia influence compositions of fungal communities in cultivated areas to a greater extent than Aspergillus introduced with crops. Application of atoxigenic A. flavus based biocontrol products may prevent this movement and associated crop contamination.
Aflatoxin production by A. parasiticus from crops and soils
Assessing the risk a given fungal group poses to aflatoxin contamination requires knowledge of the aflatoxin-producing potential and frequency of occurrence of the fungus (Cotty et al. 2008; Probst et al. 2007). High frequencies of A. parasiticus were observed in crops and soils from Zambia (Tables 2.1-2.3, 2.5-2.6) and were associated with contaminated crops (Table 4). Aspergillus parasiticus from Zambia was highly toxigenic (Table 2.7) irrespective of the fungus source (i.e. maize, groundnut, or soil), substrate used for assaying aflatoxigenicity (maize or groundnut) or temperature at which aflatoxigenicity assessments were conducted. Aspergillus parasiticus from non-cultivated
# # 87# # areas was just as aflatoxigenic as those from crops or cultivated soil, suggesting that fungi endemic to non-cultivated areas in Zambia are potential reservoirs from which aflatoxigenic fungi disperse to crops. Similar to what has been reported before where aflatoxigenicity of A. parasiticus was the same on maize and groundnut at 31oC
(Kachapulula, in press), no differences were observed on the two crops at 25oC, 30oC or
35oC (Table 2.7). However, higher concentrations of aflatoxins were produced on groundnut than maize at 20oC. The mechanisms behind higher aflatoxigenicity on groundnut than maize at lower temperatures are currently unknown.
Aflatoxin-producing fungi are common in maize and groundnut, and in cultivated and non-cultivated soil in all agroecologies of Zambia. Aspergillus parasiticus and fungi with S strain morphology are important etiologic agents of crop contamination.
Aspergillus flavus L strain morphotype fungi were associated with lower aflatoxins in crops, possibly because this group of aflatoxin-producers are of lower average aflatoxin- production potential in Zambia (Kachapulula et al. 2017). L strain isolates with low aflatoxin-producing potential are also known to interfere with crop contamination by highly toxigenic fungi (Atehnkeng et al. 2014). Methods to increase incidences of atoxigenic L strain isolates, such as biocontrol (Atehnkeng et al. 2016), may lower aflatoxin contamination in Zambia. Cultivated and non-cultivated areas had comparable community structures of Aspergillus section Flavi. Moreover, fungi from non-cultivated areas were just as toxigenic as those from cultivated areas. Fungi endemic to non- cultivated areas in Zambia may influence the compositions of fungal communities in cultivated areas and are potential reservoirs from which toxigenic fungi disperse to crops.
# # 88# # Treatments with atoxigenic genotypes of the A. flavus L strain morphotype may reduce effects of fungal communities from non-cultivated areas on fungi infecting crops.
# # 89# #
CHAPTER 4 - AFLATOXIN CONTAMINATION OF DRIED INSECTS AND
FISH IN ZAMBIA
To be submitted to: Journal of Food Protection
Introduction
Insects are an important food and income source in Zambia, providing dietary protein and supplementing incomes in rural and urban areas (Silow 1976; Mbata 1999;
Mbata et al. 2002; Mbata and Chidumayo 2003; Mbata 1995). Edible insects are highly nutritious, comparable to or better than common sources of meat such as chicken and beef and, yet, are less expensive (Siulapwa et al. 2014). More than 60 insect species are consumed in Zambia, with the most popular being Lepidoptera larvae in the order
Saturniidae, grasshoppers and termites (Van Huis et al. 2013). The most common species of caterpillars include Gonimbrasia belina (L.) (Mopane worm, local name “Mumpa”),
G. zambesina (W.) (local name “Mumpa”), and Gynanisa maja (S.) (local name
“Chipumi”)(Mbata et al. 2002; Siulapwa et al. 2014; Cunningham 1996). Macrotermes falciger (local name “Inswa”) and Ruspolia differens (local name “Nshonkonono”) are the frequently consumed termite and grasshopper species, respectively (Siulapwa et al.
2014). In Zambia, insects are harvested in rural areas and sold in urban centers, making concerns over their safety relevant to the wider population.
Although the importance of insects in human diets worldwide is well known (Van
Huis 2013; DeFoliart 1989) and expected to rise due to demands from increased population (Van Huis 2013), concerns for safety of insects as human food have also risen
(Mpuchane et al. 1996). Insects could be contaminated with hazardous microbes and
# # 90# # mycotoxins such as aflatoxins (Mpuchane et al. 1996; Van Huis 2013). Aflatoxins are cancer-causing, immuno-suppressive mycotoxins that are associated with stunting, reduced weight-gain and/or rapid death (Gong et al. 2004; Lewis et al. 2005; Liu et al.
2012; Probst et al. 2007; Reddy and Raghavender 2007; Turner et al. 2003; Williams et al. 2004). Enforcement of aflatoxin regulatory limits in foods and feeds results in loss of markets and reduced income (Van Egmond et al. 2007; Wu 2014). Aflatoxins are produced by several species in Aspergillus section Flavi. The fungi disperse from soil, organic matter, and alternative hosts to crops, trees, animals, and foods. Species most notorious for contaminating foods with aflatoxins are Aspergillus flavus (produces only B aflatoxins) and A. parasiticus (produces both B and G aflatoxins) (Cotty et al. 2008;
Probst et al. 2010; Horn and Dorner 1998). However, recent work has revealed that the causal agents of aflatoxin contamination actually include several other species.
Aspergillus flavus is typically divided into morphotypes based on sclerotia size and habit.
The L morphotype produces few large (average diameter >400 µm) sclerotia and the S morphotype produces numerous small (average diameter <400 µm) sclerotia (Cotty
1989). Several species in addition to A. flavus have the S morphology. Both the S morphotype of A. flavus and the other S morphology Aspergilli consistently produce large quantities of aflatoxins. The phylogenetically delineated S-morphology taxa include: a) A. flavus S strain; b) Lethal Aflatoxicosis Fungus (LAF) SB that severely contaminated maize and led to many deaths in Kenya (Probst et al. 2007); c) the un- named taxon SBG from West Africa (Cotty and Cardwell 1999); and d) A. minisclerotigenes (Pildain et al. 2008). Aflatoxin-producing fungi have been isolated from insects and fish (Mpuchane et al. 1996; Jonsyn and Lahai 1992; Adebayo-Tayo et
# # 91# # al. 2008) and these fungi can infect, and sometimes kill live insects (Seye et al. 2014;
Drummond and Pinnock 1990). Aflatoxigenic fungi may also become associated with dried fish and insects through poor processing, such as sun-drying on the ground or in open environments, practices common in Zambia (Mbata and Chidumayo 2003). Soils in both cultivated and uncultivated areas of Zambia contain aflatoxin-producing fungi that cause crop contamination (Kachapulula et al. 2017). It is likely that these fungi also have the ability to produce aflatoxins in fish and insects (Adebayo-Tayo et al. 2008; Jonsyn and Lahai 1992; Mpuchane et al. 1996). Aspergillus species and genotypes vary in average aflatoxin-producing potential and the relative importance of specific etiologic agents may vary among regions (Cotty et al. 2008). To assess the extent to which mitigation may be required, it is important to characterize aflatoxin concentrations and frequencies of aflatoxin-producers in insects and fish from Zambia.
The current study sought to: 1) quantify aflatoxins in insects and fish from markets in Zambia, 2) characterize communities of Aspergillus section Flavi on insects and fish, and 3) assess the capacity of insects and fish from Zambia to support growth and aflatoxin production by the observed Aspergillus section Flavi. Results indicate that aflatoxins are common on marketed dried insects but not on marketed dried fish and that the A. flavus L morphotype is the most frequent member of Aspergillus section Flavi on both insects and fish. In addition, A. parasiticus and fungi with S morphology were also found to be common and shown to be capable of contaminating these foods with concentrations of aflatoxins many times legally allowed levels.
# # 92# # Materials and methods
Sampling
Dried caterpillar larvae (97), termites (4), and fish (66) were obtained from markets in 9 districts in Zambia including Mansa, Serenje, Lusaka, Kaoma, Kapiri,
Mposhi, Mazabuka, Choma, Livingstone and Sesheke. Three morphologically distinct caterpillars were collected and later identified as either Gonimbrasia zambesina (locally called “Mumpa”, Figure 3.1a), Gynanisa maja (locally called “Chipumi”, Figure 3.1c), respectively, or Nephele sp. in the current study (Figure 3.1b). A quarter of the
Gonimbrasia zambesina samples were later found to also contain up to 5 Gonimbrasia belina individuals per kilogram of sample (Figure 3.1d). The fishes consisted of the genera Oreochromis, Petrocephalus and Limnothrissa (Figures 3.2 and 3.6), whereas the termites were all Macrotermes falciger (Table 3.1). Where it was possible, five samples
(350 g to 500 g each) of each species were obtained from each market with at least 3 markets from each district. All samples were dried in a forced air oven (40°C) to 5-8% water content at the University of Zambia, to prevent fungal growth during transportation, and sealed in plastic bags to prevent rehydration. The insects and fishes were imported to the USDA, ARS Laboratory in the School of Plant Sciences, University of Arizona, under permit number P526P-12-00853 awarded to Peter J. Cotty by the
Animal and Plant Health Inspection Service of USDA.
Species assignment for caterpillars
In order to correctly assign species to caterpillars in the current study, cytochrome c oxidase Subunit 1 (COI), a sequence widely used in insect and Lepidoptera taxonomy
(Hebert et al. 2003; Rubinoff et al. 2017; Hajibabaei et al. 2006), was amplified from
# # 93# # caterpillar genomic DNA, sequenced, and compared with GenBank sequences of previously described species (Wilson et al. 2011), by phylogenetic analysis. DNA was isolated from individual dried caterpillars according an Aspergillus spore-extraction protocol (Callicott and Cotty 2015) with modifications. Briefly, four caterpillars from each of three species were washed in 80% ethanol with 0.1 % tween, rinsed in sterile distilled deionized water, and left to dry. Ground insect samples were placed into 500 µl lysis buffer (30 mmol l−1 Tris, 10 mmol l−1 EDTA, 1% SDS, pH 8·0), and incubated in a
Thermomixer 5436 (Eppendorf, Inc., Hamburg, Germany) for 1 h at 60°C and
800 rev min−1. After removing cell fragments by centrifugation, DNA was precipitated using ammonium acetate and ethanol (Sambrook et al. 1989), and re-suspended in 25 µl sterile water. Twenty microliters of Phenol:Chloroform:Isoamyl alcohol (25:24:1; 10 mM
Tris, pH 8.0, 1 mM EDTA) was added to purify the isolated DNA, shaken for 1 minute, and centrifuged. The supernatant was transferred to a fresh tube, and ammonium acetate and ethanol was used to precipitate DNA (Sambrook et al. 1989). DNA was quantified with a spectrophotometer (model ND-1000, NanoDrop Technologies, Wilmington, DE) and diluted to a final concentration of 5–10 ng/µl before PCR.
The 658 bp COI fragment was amplified using primer pair LCO1490 (5’-
GGTCAACAAATC-ATAAAGATATTGG-3’) and HCO2198 (5’-
TAAACTTCAGGGTGACCAAAAAATCA-3’) (Hebert et al. 2003; Folmer et al. 1994).
PCR reactions were performed in 20 µl reactions using 2 µl genomic DNA and a PCR
PreMix (AccuPower® HotStart, Bioneer Pacific, Australia) with one cycle of 1 min at 94 oC; five cycles of 1 min at 94 oC, 1.5 min at 45 oC and 1.5 min at 72 oC; 35 cycles of 1
# # 94# # min at 94 oC, 1.5 min at 50 oC and 1 min at 72 oC and a final cycle of 5 min at 72 oC
(Hebert et al. 2003).
In order to correctly assign caterpillars in the current study to species, the blast search in GenBank was used for COI sequence, and the three top matches were included in Bayesian analyses using MrBayes Version 3.2.6 (Huelsenbeck and Ronquist 2001).
Reference sequences obtained from GenBank were for the saturniidae Gynanisa maja spp. terrali (accession number KF491774), Gonimbrasia ertli (accession number
HQ574035), Gonimbrasia epimethea (accession number HQ574036), Gonimbrasia longicaudata (accession number HQ573883), Lobobunaea goodii (accession number
HQ573808), Nudaurelia jamesoni (accession number HQ574076), Bunaea alcinoe
(accession number HQ574067), Athletes albicans (accession number HQ574077), and sphingidae Nephele comma (accession numbers FJ485749 and JN678292), Nephele discifera (accession number JN678294), Nephele lannini (accession number JN678298),
Nephele monostigma (accession number JN678300), and Nephele subvaria (accession number JN678305) (Wilson et al. 2011). The pyralidae, Pyralis farinalis, which is sister to both saturniidae and sphingidae (Hebert et al. 2003) was used to root the tree.
Bayesian analyses were conducted with 10 million generations and branches with less than 95% posterior probability were collapsed. Trees were visualized with FigTree v.1.4.3 (http://tree.bio.ed.ac.uk/).
Aflatoxin quantification in ground insects and fish
Total aflatoxins were quantified with a lateral flow immunochromatographic assay (Reveal Q+ for Aflatoxin, Neogen Corporation, Lansing, MI) approved by Grain
Inspection, Parkers and Stockyards Administration (GIPSA). Modifications to the
# # 95# # manufacturer’s instructions recommended by GIPSA were followed. Briefly, each insect or fish sample (350 g to 500 g) was ground with a knife mill (Retsch GM200, Retsch
GmbH, Haan, Germany) to pass 75% of the ground material through a 20 mesh sieve, mixed thoroughly, and a 50 g sub-sample was blended with 250 ml of 65% ethanol.
Aflatoxin content was determined according to the manufacturer’s instructions. Because this aflatoxin quantification technique was not designed for insects and fish, the readings obtained were corrected using percentage recovery data from spike and recovery assays.
Briefly ground insect (5 g) with no detectable aflatoxin was spiked to 100 µg/kg of total aflatoxin using an aflatoxin standard (in methanol, Supelco, Bellefonte, PA). Total aflatoxin was extracted and quantified as described above. Spike and recovery was performed in five replicates. Recovery rates were estimated using the following equation:
% Recovery = (Aflatoxin concentration measured /Spiked concentration) x 100.
Precision of the analytical method was expressed as relative standard deviation (RSD) of replicated results.
Isolation and identification of fungi from insects and fish
Insect and fish samples were ground in a knife mill (Grindomix GM200, Retsch
GmbH, Haan, Germany) to pass a #12 sieve, and homogenized. Fungi were isolated from ground insect and fish material using dilution plate technique on modified rose
Bengal agar (Cotty 1994). Briefly, ground insect and fish material (0.1 to 10 g) were shaken (20 min, 100 rpm) in 50 ml sterile distilled water on a reciprocal shaker (KS-501,
IKA Works Inc., Wilmington, NC, USA). Dilution plating of the suspension was performed on modified rose Bengal agar in triplicate. Plates were incubated (3 days,
31oC, dark) and up to eight colonies of Aspergillus section Flavi were transferred to 5-2
# # 96# # agar (5% V8-juice; 2% agar, pH 5.2). Fungi were stored in sterile water (2 ml) as plugs of sporulating culture after incubation for 7 days at 31oC (Cotty 1988). Isolations were performed at least twice from each sample. Aspergillus species and strains were identified using both macroscopic and microscopic characteristics (Cotty 1989; Cotty
1994; Klich and Pitt 1988; Probst et al. 2007).
Determining potential for aflatoxin formation after market
To determine the potential for aflatoxin concentrations to increase in market insects and fish during handling and storage, a technique previously applied to maize and groundnut, Simulated Poor Storage Assays, (SPSA) (Kachapulula et al. 2017) was conducted. Briefly, un-inoculated caterpillar (n=10) and fish (n=10, Table 3.1) market samples were thoroughly hand mixed, 10 g was placed onto metal sieves (10 cm diameter), which were in sealed plastic boxes containing a moist sponge (4 cm × 4 cm ×
4 cm) and incubated (31oC, 7 days). After incubation, samples (Figure 3.3) were ground in a blender (Waring 7012S, Waring, Torrington, Connecticut) containing 50 ml 70% methanol at maximum speed for 20 seconds. The slurry was allowed to settle (20 min) and 4 µl of the supernatant was spotted directly onto thin-layer chromatography (TLC) plates (Silica gel 60, EMD, Darmstadt, Germany) adjacent to aflatoxin standards
(Aflatoxin Mix Kit-M, Supelco, St Louis) containing known quantities of aflatoxins B1,
B2, G1 and G2. Plates were developed in 96:3:1 ethyl ether-methanol-water, air-dried, after which aflatoxins were visualized under 365-nm UV light. Aflatoxins were quantified directly on TLC plates using a scanning densitometer (TLC Scanner 3, Camag
Scientific Inc, Wilmington, N.C.) running winCATS 1.4.2 (Camag Scientific Inc,
Wilmington, N.C.).
# # 97# # Suitability of insects and fish as substrate for growth and aflatoxin-production by toxigenic Aspergillus section Flavi
To evaluate the ability of caterpillars and fish from markets to support growth and aflatoxin contamination, inoculation tests were performed with known aflatoxigenic
Aspergillus section Flavi genotypes. Briefly, 4 isolates representing A. flavus L (AF13;
ATCC 96044; SRRC 1273) and S morphotype strains (AF70; ATCC MYA384), A. parasiticus (NRRL2999) and un-named taxon SBG (A-11612, Tables 3.3 and 3.4) were inoculated onto caterpillars and fish (10g in 250 ml Erlenmeyer flask) previously autoclaved for 20 minutes, cooled to room temperature and moisture adjusted to 30%.
One million freshly harvested spores from 7-day-old cultures of each isolate were used in the inoculations. The inoculated insects and fish were incubated (7 days, 100% RH,
31oC) to allow fungal growth and aflatoxin production. To quantify CFU/g after incubation, 100 ml of sterile distilled de-ionized water with 0.1% tween was added to each culture and shaken at 650 G on a mini orbital shaker (Troemner LLC, Thorofare NJ,
USA) for 10 minutes. CFU/g was determined by dilution plating the resulting suspension on rose Bengal agar in four replicates. To quantify aflatoxins, after incubation sample cultures (Figure 3.4) were blended in 50 ml of 70 % methanol and aflatoxins were quantified with TLC as described above.
Data analysis
Aflatoxin content in fish and insects directly from the markets, and aflatoxin produced in laboratory inoculation experiments were measured in micrograms per kilogram (µg kg-1). Total quantities of section Flavi in market samples and after inoculation and incubation were calculated as Colony Forming Unit per gram (CFU g-1)
# # 98# # of substrate. Community composition of section Flavi was described as percent of A. flavus L strain morphotype (Cotty 1989), un-delineated S strain morphotype (Probst et al.
2007), and A. parasiticus recovered from each sample. Aflatoxins before and after incubation were compared using a paired t-test and multiple means (aflatoxins, CFU/g, and percent) were compared using Analysis of Variance general linear models and
Tukey’s HSD test as implemented in JMP 11.1.1 (SAS Institute, Cary, NC). Data were tested for normality and, if required, log transformed (aflatoxin and CFU data) to normalize distributions before analyses. Percent data were arcsine-transformed prior to analyses. However, actual means are presented for clarity. All tests were performed at α =
0.05.
# # 99# # Results
Species assignment for caterpillars
In order to verify species assignment, caterpillars from Zambia were compared with known species through phylogenetic analysis of 658 bp of cytochrome oxidase subunit 1. Three groups of caterpillars were resolved with all caterpillars morphologically assignable to Gynanisa maja grouping with reference G. maja (NCBI accession number
KF491774) in group 3 (Figure 3.5). Caterpillars assigned to Gonimbrasia zambesina based on morphology grouped with, but remain distinct from other Gonimbrasia (G. ertli,
NCBI accession number HQ574035; G. epimethea NCBI accession number HQ574036;
G. longicaudata NCBI accession number HQ573883; group 2, Figure 3.5). Reference sequences were not found for either G. zambesina or Gonimbrasia belina. The third morphological species purchased in Zambian markets was found only in Kaoma and grouped with members of the genus Nephele (N. comma, NCBI accessions FJ485749 and
JN678292; N. discifera, accession NCBI number JN678300; N. subvaria, NCBI accession number JN678305; group 1, Figure 3.5).
Aflatoxin in insects and fish
There were significant differences (ANOVA, F6,98 = 13.3965, P<0.001) in total aflatoxin concentration in market samples of various species of caterpillars and fish, with the highest average aflatoxin concentration (24 µg kg-1) in termites, M. falciger. Only the fish genus Oreochromis had no detectable aflatoxins (Table 3.1). Percent samples exceeding 10 µg kg-1 (Zambian regulatory limit for aflatoxin in food) was 100% for termites, 54.8% and 40.6% for the caterpillars Gonimbrasia zambesina and Gynanisa maja, respectively, and 15.8% for the fish genus Limnothrissa (Table 3.1). Although
# # 100# # none of the Nephele sp. (identified in current study, Figure 3.5) samples were above the
Zambian regulatory limit, all were above the European limit of 4 µg kg-1.
Potential for aflatoxin formation after market
Increases of several orders of magnitude were observed in aflatoxin content in some insects and fishes, even in samples that initially had acceptable levels of aflatoxins
(Table 1). Total aflatoxins increased at least 20-fold (on average from 11 to 229 µg/kg;
Paired t-test, P < 0.001), in Gynanisa maja, 600-fold (from 12 to 8102 µg/kg; Paired t- test, P < 0.001) in G. zambesina, 1,000-fold (from 5 to 6,187 µg/kg; Paired t-test, P <
0.001) in the Nephele sp, and from non-detectable levels to 23 µg/kg (Paired t-test, P <
0.05) in Oreochromis (Table 3.1). No significant increases were observed in the fish genera Petrocephalus and Limnothrissa.
Fungi from insects and fish
Aspergillus parasiticus, A. flavus L morphotype, fungi with S morphology and A. tamarii were recovered (Table 3.2). The Aspergillus flavus L morphotype was the most common Aspergillus section Flavi species associated with each of the insect and fish species, with the exception of Gynanisa maja for which A. parasiticus was the most frequent. There were significant differences (ANOVA, F6,43 = 5.2317, P<0.001, Table
3.2) in A. flavus L morphotype frequencies on insects and fishes, with the highest occurrence on Oreochromis (92%) and the least on Gynanisa maja (36%). Although frequency of Aspergillus section Flavi with S morphology was similar (8% +/- 8) among all the fishes and insects (ANOVA, F6,43 = 0.6835, P=0.6638, Table 3.2), the frequencies of A. parasiticus differed (ANOVA, F6,43 = 4.6609, P=0.001, Table 3.2), with the highest frequency (44%) on the caterpillar Gynanisa maja and none detected on Oreochromis
# # 101# # and M. falciger. Overall quantities (CFU/g) of Aspergillus section Flavi were similar (30
CFU/g ± 29) for all species (ANOVA, F6,43 = 1.4889, P=0.2056, Table 3.2).
Suitability of insects and fish as substrate for growth by toxigenic Aspergillus section
Flavi
All insects and fishes tested supported growth of all three Aspergillus species evaluated (Table 3, Figure 3.4). Species differed in ability to support growth by aflatoxin- producers (ANOVA, F5,17=17.7761, P<0.001), with the most growth (Tukey’s HSD,
P<0.05) on average on Nephele. (4.1x1015 CFU/g), and the least on Oreochromis
(4.2x109 CFU/g, Table 3.4). Average growth across all insects and fish for all fungi were also different (F5,17=17.6324, P<0.001). There were significant differences among species of insects and fish (ANOVA, F4,20=3541.340, P<0.001) in support of A. parasiticus growth with the most propagules produced on Nephele. (8.4x1012 CFU/g) and the least on
Oreochromis (5.2x107 CFU/g, Tukey’s HSD, P<0.05, Table 3.3). Differences (ANOVA,
F4,20=267.1297, P<0.001) were also observed in support of growth of the A. flavus S morphotype (AF70; ATCC MYA384) with the most growth on Gynanisa maja (7.8x108
CFU/g), Nephele sp. (7.4x108 CFU/g) and Limnothrissa (6.8x108 CFU/g), followed by G. zambesina (1.9x108 CFU/g) and Oreochromis (6.9x107 CFU/g, Table 3.3). The other A. flavus (AF 13; ATCC 96044; SRRC 1273; L morphotype) had much higher growth on
Nephele sp. (1.7x1016 CFU/g) than on any other insect or fish (Tukey’s HSD, P<0.05).
For the unnamed taxon SBG (A-11612) as with the other 3 fungi, Oreochromis supported the least growth.
# # 102# # Aflatoxin production on insects and fish
Each of the three caterpillar species supported production of over 28,000 µg/kg total aflatoxins (Table 3.4). The A. flavus S morphotype produced the highest concentrations of aflatoxin (126,700 µg/kg) with the most aflatoxins forming on Nephele
(114,800 µg/kg, Table 3.4). Production of aflatoxins differed among caterpillars for both the A. flavus L morphotype (ANOVA, F2,8=8.0041, P=0.0203) and the A. flavus S morphotype (ANOVA, F2,8=36.6697, P<0.001). No differences were observed in aflatoxin production by either A. parasiticus (ANOVA, F2,8=0.5353, P=0.6111) or the un-named SBG (ANOVA, F2,8=3.0788, P=0.1202) on fish and insects.
# # 103# # Discussion
Species assignment for caterpillars
Lepidoptera larvae are normally harvested for food after the third or fourth molt, gutted, dried, and cooked prior to consumption (Mbata and Chidumayo 2003). During these processing steps, features diagnostic of each species may be lost. In addition, caterpillars not yet described may be consumed in Zambia (Mbata 1995). To ensure correct species assignment, phylogenies based on 658 bp of cytochrome oxidase subunit
1 were reconstructed to compare the species sampled in markets to reference taxa for which sequences had been deposited at NCBI. Based on this analysis, at least three species of edible caterpillars were present in our collection: Gynanisa maja (group 3,
Figure 3.5), Gonimbrasia (group 2) and a Nephele species closely related to N. comma
(group 1). Due to paucity of Lepidopteran COI sequences in GenBank, larvae from the market initially identified based on morphology as Gonimbrasia zambesina and
Gonimbrasia belina could only be confirmed to the genus level (Figure 3.5).
!Aflatoxin in insects and fish
Aflatoxins are a danger to human health, livestock productivity and trade (Gong et al. 2004; Liu et al. 2012; Probst et al. 2007; Reddy and Raghavender 2007; Turner et al. 2003; Van Egmond et al. 2007; Williams et al. 2004; Wu 2014). Deaths from consumption of highly contaminated food in Kenya (Lewis et al. 2005; Liu et al. 2012;
Probst et al. 2007) and Tanzania and increasing mycotoxin safety concerns in edible insects (Mpuchane et al. 1996; Van Huis 2013) have increased the need to evaluate the safety of foods. Aflatoxins were detected in almost all fishes and insects, with the termites and two of the caterpillars (Gynanisa maja and Gonimbrasia zambesina) having
# # 104# # levels above those allowed for food in Zambia (Table 3.1). Aflatoxin levels in caterpillars in the current study were different from what has been previously reported, where average aflatoxins in many locations exceeded 20 µg/kg (Mpuchane et al. 1996). These differences may result from differences in species examined, differences in environmental conditions to which the insects were subjected during processing and storage (Cotty and Jaime-Garcia 2007; Kachapulula et al. 2017), and/or differences in
Aspergillus section Flavi community compositions (Probst et al 2010; Kachapulula et al. unpublished). On the contrary, average aflatoxins in fishes in the current study were similar to those previously reported on Gadus morhua, Katsuworus pelamis,
Pseudotolithus typhus Dasyatis margarita, Arius hendeloti, Ethalmosa fimbriota
Triuchurius trichurius Carchanas faunis Pentanemis qumquarius Cynoglossus browni and Drepane africana (Adebayo-Tayo et al. 2008; Jonsyn and Lahai 1992). However, the proportions of fish with aflatoxins ≥ 10 µg/kg in Petrocephalus (40%) and Limnothrissa
(15.8%) in the current study is still a reason for concern. None of the Oreochromis had detectable aflatoxins. Given that Oreochromis is processed using techniques and in environments similar to Petrocephalus and Limnothrissa, the absence of aflatoxins in
Oreochromis suggests it is not a suitable substrate for aflatoxin production or may not contain high enough proportions of aflatoxigenic fungi (Cotty and Jaime-Garcia 2007;
Kachapulula et al. 2017; Probst et al. 2010).
Fungi from insects and fish
To assess potential for food to become contaminated with aflatoxins, frequencies of aflatoxin-producing fungi must be considered (Mehl et al. 2012; Probst et al. 2007).
Aspergillus flavus L morphotype dominated most insects and fishes, although appreciable
# # 105# # amounts of the consistently aflatoxigenic A. parasiticus and S morphology fungi were also found on the three caterpillar species (Table 3.2). Previous studies of fungi on edible caterpillars and fish (Mpuchane et al. 1996; Jonsyn and Lahai 1992) have listed A. flavus
L as the only Aspergillus section Flavi on caterpillars. The current study reveals that additional Aspergillus section Flavi species can occur on caterpillars, and that the etiology of aflatoxin contamination of these valuable foods could be complex. The L morphotype of A. flavus is associated with high variability in aflatoxin production, with both highly aflatoxigenic and atoxigenic genotypes (Cotty 1989). In Zambia, high prevalence of the L morphotype of A. flavus has been associated with low aflatoxins in maize and groundnuts (Kachapulula et al. 2017). High L strain incidence my partially explain the low levels of aflatoxins observed in fishes (Table 3.1), particularly
Oreochromis, where the L morphotype was as high as 92% and no S morphotype or A. parasiticus were detected (Table 3.2). Similarly, Nephele sp. had lower amounts of aflatoxins compared with the other two caterpillar species (Table 3.1), possibly because the former had higher incidences of the L strain morphotype than other caterpillars. In addition, Nephele also had lower proportions of A. parasiticus and fungi with S morphology, than the other two caterpillar genera (Table 3.2). Aspergillus parasiticus and fungi with S morphology are almost always highly aflatoxigenic and frequencies of S morphotype fungi as low as 13 percent , as was observed with caterpillars in the current study, can cause high aflatoxins levels (Cotty et al. 2008). It is expected that under poor storage, aflatoxins might still increase in Nephele as both the average aflatoxin-producing potential of the fungal community and the extent of growth both contribute (Cotty and
Jaime-Garcia 2007; Kachapulula et al. 2017).
# # 106# # Potential for aflatoxin formation after market
When aflatoxin producers are present in food with low aflatoxins, as is the case in the current study, aflatoxin increases may occur during processing, transportation, or poor storage. A method for quantifying the risk of aflatoxin increase (Cotty and Jaime-Garcia
2007; Kachapulula et al. 2017) was applied in the current study. Aflatoxins increased by at least 20-fold in the caterpillars and 4-fold in Oreochromis after SPSA (Table 3.1). This suggests that even though aflatoxins in Nephele from markets in Zambia were present at permissible levels initially, poor storage could create environments conducive for aflatoxin production by the aflatoxin-producing fungi already colonizing the insects
(Table 3.2). This could result, as demonstrated in the current study (Table 3.1), in
Nephele with aflatoxin concentrations unsuitable for human consumption (Table 3.1).
Although aflatoxin levels in incubated Oreochromis rose during SPSA (Table 3.1), the increase was much lower than what would be expected where all environmental conditions are conducive for aflatoxin production and aflatoxigenic fungi are present.
This low increase in aflatoxins in Oreochromis suggests that the colonizing A. flavus L morphotype may have significant numbers of atoxigenic genotypes that interfere with the contamination process (Mehl et al. 2012). However, since the termites had both high aflatoxins in the market and high incidences of the A. flavus L strain, either the colonizing fungal populations differ in incidence of atoxigenic genotypes or other factors, such as nutritional composition of Oreochromis (Mehl and Cotty 2013), might play a role in reduced aflatoxin concentrations in Oreochromis both at the market and after incubation. #
# # 107# # Suitability of insects and fish as substrate for growth and aflatoxin-production by aflatoxigenic Aspergillus section Flavi
Growth and quantities of aflatoxins produced by A. flavus, A. parasiticus and the phylogenetically diverse fungi with S morphology can differ among fungal isolates and between different substrates (Mehl and Cotty 2013). To assess the suitability of fish and insects as hosts for growth and aflatoxin production, representative genotypes from the fungal groups above were inoculated onto sterile insects and fish. All species tested support growth by aflatoxigenic fungi (Table 3.3). However, caterpillars were better substrates for growth of aflatoxigenic fungi than fish (Table 3.3). In nature, members of the genus Aspergillus are known to be entomopathogenic (Drummond and Pinnock 1990;
Carver et al. 1987) and may have evolved to effectively utilize diverse insects as food sources. This may explain why more growth occurred on caterpillars than on fish (Table
3.3). Additionally, caterpillars supported accumulation of high concentrations of aflatoxins (more than 60,000 µg/kg). Aspergilli have been reported to infect and kill insect hosts by producing aflatoxins and it is therefore not surprising to see the aflatoxin- producers both growing well and producing significant quantities of aflatoxins on the food insects from Zambia in the current study.
Aflatoxins and aflatoxin-producing fungi are common on insects and fish in
Zambia. Presence of aflatoxigenic fungi in these foods poses risk of increased aflatoxin contamination when poorly stored. Fish and insects, particularly caterpillars, are suitable substrates for aflatoxin biosynthesis. Aflatoxin mitigation measures have targeted major agricultural products including cereal, peanut, and tree-nut crops and their byproducts.
However, it is evident from the current study that insects and fish could also be
# # 108# # problematic routes for exposure to aflatoxins. Aflatoxin mitigation measures should take into consideration exposure originating from beyond agricultural commodities and include dried edible insects and fish, so that foods with aflatoxins of 10 µg/kg and above are excluded from the food chain in Zambia.
# # 109$ $ Table 3.1 Aflatoxin before and after incubation (31°C, 100% RH, 7 d) of insects and fish from Zambia % Aflatoxin after incubation Samples Aflatoxin (µg/kg) Species samples (µg/kg) (#) Mean Range > 9.9 B1 G1 Total Gynanisa 49 11BC 2.9-24.4 40.6 214 15 229B* Gonimbrasia 44 12B 3.4-25.1 54.8 3,197 4,832 8102A* Nephele 4 5CD 4.3-6.1 0 3,315 2,370 6187A* Petrocephalus 25 9CD ND-20.4 40 9 ND 9C Limnothrissa 35 5D ND-17.2 15.8 10 ND 10C Oreochromis 6 †NDE ND 0 23 ND 23C* Macrotermes 4 24A 16-36.8 100 N/A N/A N/A Means followed by the same letter in each column are not significantly different (P<0.05) by Tukey-Kramer’s HSD test. * indicates significant differences in aflatoxin content before and after incubation by Paired T-test (P<0.05). All data were log transformed prior to analyses but actual means are presented. †ND = Non-detected (Limit of detection = 2 µg/kg).
$ $ 110$ $ Table 3.2 Distribution of fungi of Aspergillus section Flavi on edible insects and fish from Zambia. Genus Samples (#) %L %S %P %T CFU/g Gynanisa 49 36B(XY) 13A(Y) 44A(X) 7AB(Y) 31A Gonimbrasia 44 55AB(X) 13A(Y) 25AB(XY) 7AB(Y) 45A Nephele 4 66AB(X) 16A(Y) 18B(Y) 0B(Y) 22A Petrocephalus 25 88A(X) 4A(Y) 5B(Y) 3B(Y) 28A Limnothrissa 35 85A(X) 10A(Y) 3B(Y) 2B(Y) 19A Oreochromis 6 92A(X) 0A(Y) 0B(Y) 8AB(Y) 59A Macrotermes 4 68AB(X) 0A(Z) 0B(Z) 32A(Y) 6A *L, S, P, and T represent A. flavus L-morphotype, S-morphotype fungi, A. parasiticus and A. tamarii, respectively. †Percent data were arcsine transformed and CFU/g data were log transformed prior to analyses but actual means are presented here. Values followed by the same letter in each column (a,b,c) or row (x,y,z) do not differ by Tukey’s HSD test (α= 0.05).
$ $ 111$ $ Table 3.3 Growth of four aflatoxin-producing fungi on edible caterpillars and fish from Zambia. Growth of 4 aflatoxin-producers (CFU/g) Caterpillar or fish genus A. parasiticus A. flavus-S A. flavus-L Un-named SBG Average 10 B(Y) 8 A(Z) 11 C(X) 13 A(W) Gynanisa (7.8x10 ) (7.8x10 ) (5.0x10 ) (1.0x10 ) (2.7x1012)B 10 C(X) 8 B(Y) 15 B(W) 11 B(X) Gonibrasia (2.6x10 ) (1.9x10 ) (8.8x10 ) (2.2x10 ) (2.2x1015)A 12 A(Y) 8 A(Z) 16 A(W) 13 A(X) Nephele (8.4x10 ) (7.4x10 ) (1.7x10 ) (3.4x10 ) (4.1x1015)A 7 D(Y) 7 C(Y) 10 E(X) 6 E(Z) Oreochromis (5.2x10 ) (6.9x10 ) (1.7x10 ) (1.2x10 ) (4.2x109)C Petrocephalus n/a n/a n/a (5.4x106)D 7 D(Z) 8 A(X) 11 D(W) 8 C(Y) Limnothrissa (7.4x10 ) (6.8x10 ) (1.0x10 ) (2.1x10 ) (2.5x10 10)C 12 Y 8 Z 15 X 12 Y Average (1.7x10 ) (4.9x10 ) (5.1x10 ) (7.4x10 ) Means followed by the same letter in each column (A/B/C) or row (X/Y/Z) are not significantly different (P<0.05) by Tukey- Kramer’s HSD test. All data were log transformed prior to analyses but actual means are presented. N/A = Not tested.
$ $ 112$ $ Table 3.4 Aflatoxin production by 4 aflatoxin-producers on edible caterpillars from Zambia. Caterpillar Aflatoxin production by 4 toxigenic aspergilli (µg/kg)
genus A. flavus-L A. flavus-S A. parasiticus Un-named SBG Average Gynanisa 62,800X(AB) 48,000X(B) 14,800X(A) 65,500X(A) 47,800B Gonibrasia 21,300XY(B) 68,200X(B) 8,600Y(A) 15,200XY(A) 28,300B Nephele 106,800XY(A) 263,800W(A) 15,600Z(A) 72,900Y(A) 114,800A Average 63,700X 126,700W 13,000Y 51,200X 63,600 Means followed by the same letter in each column (A/B/C) or row (W/X/Y/Z) are not significantly different (P<0.05) by Tukey-Kramer’s HSD test. All data were log transformed prior to analyses but actual means are presented.
$ $ 113# # Figure 3.1 Caterpillars used in the current study as collected from markets in Zambia
a=Gonimbrasia zambesina, b= Nephele sp. (identified in current study), c=Gynanisa maja, d=Gonimbrasia belina. Identification based on morphology (a, c and d) and phylogenetics (a, b, c, and d).
# # 114# # Figure 3.2 Fishes used in the current study as collected from the markets in Zambia
a,c=Petrocephalus, b=Oreochromis, d,e=Limnothrissa. Identification based on morphology.
# # 115# # Figure 3.3 Caterpillars exiting incubation during the SPSA test with visually evident fungal growth on un-inoculated, incubated caterpillars
# # 116# # Figure 3.4 Fungal growth on caterpillars inoculated with A. parasiticus
# # 117# #
Figure 3.5 Phylogenetic relationships of caterpillar species from Zambia with known species
# # 118# #
Bayesian tree for caterpillar species obtained from Zambia (asterisks) and GenBank (North America and Costa Rica). Tree is based on cytochrome c oxidase subunit 1 (COI, 658 bp). Support values above nodes indicate posterior probabilities. Group assignments: Group 1 = Nephele; Group 2 = Gonimbrasia; Group 3 = Gynanisa. *Taxa from Zambia.
# # 119# # Figure 3.6 Insects and fish in markets in Zambia
Gynanisa (a), Gonimbrasia (b), Oreochromis (c) and Limnothrissa (d) in markets in Zambia.
# # 120# #
CHAPTER 5 - Aflatoxin contamination of dried fruits in Zambia
To be submitted to: Plant Pathology
Introduction
Wild fruits supplement diets and incomes of people in portions of rural Zambia
(Kalaba et al. 2009). Most people in rural areas gather fruits to supplement diets, especially during famines (Akinnifesi et al. 2008), and also for sale in urban centers
(Kalaba et al. 2009). More than 75 wild fruit species are consumed in Zambia and
Southern Africa (Akinnifesi et al. 2002; Kalaba et al. 2009), with some of the popular fruits being Parinari curatellifolia, Zizyphus mauritania, Schinziophyton rautanenii,
Uapaca kirkiana, Anisophyllea boehmii (Kalaba et al. 2009; Chadare et al. 2008; Zimba et al. 2005). Wild fruits have various uses (Table 4.1) including processing into juice and porridge. Fruit seeds (nuts) may be eaten as a snack or extracted for oil (Juliani et al.
2007; Rahul et al. 2015; Chadare et al. 2008; Vermaak et al. 2011; Benhura et al. 2015;
Maroyi 2011; Maruza et al. 2017; Njana et al. 2013). Consumption of wild fruits is expected to increase and efforts to domesticate wild fruit tree species are increasing
(Akinnifesi et al. 2002; Akinnifesi et al. 2006). With some of the fruits getting approvals from the European Commission and the Food and Drug Administration in the U.S.A as food ingredients, demand for wild fruits such as Adansonia digitata is also increasing in the western world, and is expected to outstrip supply (Buchmann et al. 2010; Buchwald-
Werner and Bischoff 2011). Food safety concerns associated with wild fruits affect consumers in both rural and urban areas.
Wild fruits are expected to remain a vital component of diets and, as such, it is important to ensure wild fruits are free of hazardous microbes and mycotoxins, such as
# # 121# # aflatoxins (Boyd and Cotty 2001). Consumption of food contaminated with aflatoxins can lead to liver cancer, immuno-suppression, stunting, reduced weight-gain, and rapid death in humans (Turner et al. 2003; Gong et al. 2004; Williams et al. 2004; Lewis et al. 2005;
Probst et al. 2007; Reddy and Raghavender 2007; Liu et al. 2012). Enforcement of aflatoxin regulatory limits result in loss of markets and reduced income (Van Egmond et al. 2007; Wu 2014). The most commonly reported aflatoxin producers are Aspergillus flavus (produces only B aflatoxins) and A. parasiticus (produces both B and G aflatoxins)(Horn and Dorner 1998). However, several closely related taxa are known to contaminate crops in Africa including A. minisclerotigenes and two unnamed taxa SB
(only B aflatoxins) and SBG (both B and G aflatoxins) (Cotty et al. 2008; Probst et al.
2010). Aspergillus flavus and other aflatoxin–producers are frequently placed into one of two morphotypes based on sclerotial morphology. These are either the L strain morphotype that produces few large sclerotia (average diameter >400 µm) or the S strain morphotype that produces numerous small sclerotia (average diameter <400 µm) (Cotty
1989). Fungi with S morphology frequently produce large quantities of aflatoxins and
DNA based phylogenetic evidence suggests these aflatoxin-producers belong to several species: a) A. flavus S strain; b) Lethal Aflatoxicosis Fungus (LAF) SB that led to many deaths in Kenya (Probst et al. 2007); c) the un-named taxon SBG from West Africa (Cotty and Cardwell 1999); and d) A. minisclerotigenes (Pildain et al. 2008).
Aflatoxin-producers infect and produce aflatoxins on both crops and wild plant species (Boyd and Cotty 2001). Domesticated plants differ in susceptibility to both aflatoxin-producers and aflatoxin contamination (Mehl and Cotty 2013; Kachapulula et al. 2017) and such variation is also expected among undomesticated plants. Knowledge
# # 122# # of plant species vulnerability to aflatoxin contamination may both inform aflatoxin mitigation efforts and facilitate the shaping of diets to limit aflatoxin exposure. In addition, genotypes of the aflatoxin-producers differ in aflatoxin-producing potential and the relative importance of specific etiologic agents may depend on region (Cotty et al.
2008; Probst et al. 2007). Aflatoxin levels and frequencies of aflatoxin-producers in wild fruits of Zambia have not been characterized.
In order to ascertain levels of aflatoxins and potential for contamination in wild fruits in Zambia, this study: 1) Quantified aflatoxins in the wild fruits Parinari curatellifolia, Zizyphus spp., Schinziophyton rautanenii, Tamarindus indica,
Vangueriopsis lanciflora, Thespecia garckeana and Adansonia digitate, 2) Characterized communities of Aspergillus section Flavi in the fruits, and 3) Assessed suitability of the fruits to support growth and aflatoxin production by aflatoxigenic Aspergillus section
Flavi. Communities on all fruits consisted mostly of the A. flavus L morphotype, although A. parasiticus and fungi with S morphology were also recovered and potential for aflatoxin varied with wild fruit species.
# # 123# #
Materials and methods
Sampling
Samples of dried fruits (114 total) consisting of Schinziophyton rautanenii (n=24),
Parinari curatellifolia (n=17) Vangueriopsis lanciflora (n=7), Zizyphus spp. (12),
Tamarindus indica (n=25), Adansonia digitata (n=9) and Thespecia garckeana (n=20,
Table 4.2) were collected from markets in 9 districts in Zambia including Lusaka,
Kaoma, Mongu, Senanga, Kapiri Mposhi, Mazabuka, Choma, Livingstone and Sesheke.
For each species, five samples (350 g to 500 g each) were obtained from each market with at least 3 markets from each district. The fruits were dried at the University of
Zambia in a forced air oven (40°C) to 5-8% water content, to prevent fungal growth, and sealed in plastic bags to prevent rehydration. The fruits were imported to the USDA,
ARS, Laboratory in the School of Plant Sciences, University of Arizona under permit number P526P-12-00853 awarded to Peter J. Cotty by the Animal Plant Health
Inspection Service (APHIS) of USDA.
Aflatoxin quantification
Total aflatoxins were quantified with a GIPSA approved lateral flow immunochromatographic assay (Reveal Q+ for Aflatoxin, Neogen Corporation, Lansing,
MI) following modifications to the manufacturer’s instructions recommended by GIPSA.
Briefly, each entire fruit sample (350 g to 500 g) was ground with either a knife mill
(Retsch GM200, Retsch GmbH, Haan, Germany) or cutting mill (Retsch SM100, Retsch
GmbH, Haan, Germany) to pass at least 75% of the ground material through a 20 mesh sieve, mixed thoroughly, and a 50 g sub-sample was blended with 250 ml of 65% ethanol. Aflatoxin content was determined according to the manufacturer’s instructions.
# # 124# # The aflatoxin quantification technique used was not designed for wild fruits, and as such, results were corrected by spike and recovery experiments done for each fruit. Briefly, ground fruit (5 g) with no detectable aflatoxin was spiked to 100 µg/kg of total aflatoxin using an aflatoxin standard (in methanol, Supelco, Bellefonte, PA). Total aflatoxin was extracted and quantified as described above. Spike and recovery was performed in five replicates. Recovery rates were estimated using the following equation: % Recovery =
Total aflatoxin concentration measured in spiked sample/Spiked concentration x 100.
Precision of the analytical method was expressed as relative standard deviation (RSD) of replicated results.
Isolation and identification of fungi from fruits
Dried fruit samples were ground in either a knife mill (Tamarindus indica,
Zizyphus spp., Adansonia digitata, Vangueriopsis lanciflora, Grindomix GM200, Retsch
GmbH, Haan, Germany) or cutting mill (Schinziophyton rautanenii, Thespecia garckeana and Parinari curatellifolia; Retsch SM100, Retsch GmbH, Haan, Germany) to pass a #12 sieve, and homogenized. Fungi were isolated from ground fruit material using dilution plate technique on modified rose Bengal agar (Cotty 1994). Briefly, ground fruit material
(0.1 to 10 g) was shaken in 50 ml sterile distilled water (20 min, 100 rpm) on a reciprocal shaker (KS-501, IKA Works Inc., Wilmington, NC, USA). Dilution plating was performed on modified rose Bengal agar in triplicate. Plates were incubated (3 days,
31oC, dark) and up to eight colonies of Aspergillus section Flavi were transferred to 5-2 agar (5% V8-juice; 2% agar, pH 5.2). Fungi were stored in sterile water (2 ml) as plugs of sporulating culture after incubation for 7 days at 31oC (Cotty 1988). Isolations were performed a minimum of twice to obtain a total of at least 15 isolates from each sample.
# # 125# # Aspergillus species and strains were identified using both macroscopic and microscopic characteristics (Cotty 1989; Cotty 1994; Probst et al. 2007; Klich and Pitt 1988).
Wild fruits as substrate for growth and aflatoxin production
To evaluate ability of wild fruit to support growth and aflatoxin contamination, colony forming units (CFU) and aflatoxin concentrations were measured on fruit previously inoculated with aflatoxigenic Aspergillus section Flavi. Briefly, 5 isolates representing the A. flavus L strain morphotype (AF13; ATCC 96044; SRRC 1273) and the A. flavus S strain morphotype (AF70; ATCC MYA384), A. parasiticus (NRRL 2999),
A. minisclerotigenes (A-11611) and the un-named taxon SBG (A-11612, Tables 4.5 and
4.6) were inoculated onto sterile whole fruits (10 g in 250 ml Erlenmeyer flask) previously autoclaved for 20 minutes, cooled to room temperature and moisture adjusted to 30%. Spore suspensions containing one million freshly harvested spores from 7-day- old cultures were used as inoculum. After incubation (7 days, 100% RH, 310C), 100 ml of sterile distilled de-ionized water with 0.1% tween 80 was added and flasks were shaken (650 G, mini orbital shaker, Troemner LLC, Thorofare NJ, USA) for 10 minutes.
CFU was determined by dilution plating the resulting suspension on rose Bengal agar in five replicates. To quantify aflatoxin production, cultures were blended in 50 ml of 70% methanol (20 seconds, maximum speed, Waring 7012S, Waring, Torrington,
Connecticut). The slurry was allowed to settle (20 min) and 4 µl of the supernatant was spotted directly onto thin-layer chromatography (TLC) plates (Silica gel 60, EMD,
Darmstadt, Germany) adjacent to aflatoxin standards (Aflatoxin Mix Kit-M, Supelco, St
Louis) containing known quantities of aflatoxins B1, B2, G1 and G2. Plates were developed in 96:3:1 ethyl ether-methanol-water, air-dried, after which aflatoxins were
# # 126# # visualized under 365-nm UV light. Aflatoxins were quantified directly on TLC plates using a scanning densitometer (TLC Scanner 3, Camag Scientific Inc, Wilmington, N.C.) running winCATS 1.4.2 (Camag Scientific Inc, Wilmington, N.C.).
Data analysis
Aflatoxin concentrations in market samples and aflatoxins produced on fruits in laboratory assays were quantified in micrograms per kilogram (µg kg-1). Total quantity of section Flavi fungi from each assay was calculated as CFU per gram (CFU g-1).
Compositions of section Flavi communities were described as the percent composed of the A. flavus L strain morphotype (Cotty 1989), un-delineated S strain morphotype
(Probst et al. 2007), A. parasiticus and A. tamarii . Comparisons of both aflatoxin concentrations and fungal populations were performed by Analysis of Variance using general linear models (GLM) and Tukey’s HSD mean comparison test as implemented in
JMP 11.1.1 (SAS Institute, Cary, NC). Data were tested for normality and, if required, log transformed (aflatoxin and CFU data) to normalize distributions before analyses.
Percent data were arcsine-transformed prior to analyses. However, actual means are presented for clarity. All tests were performed at α = 0.05.
# # 127# #
Results
Aflatoxin in fruit
There were significant differences among market-collected fruits in aflatoxin concentrations (ANOVA, F6,98=25.3786, P<0.001), with the highest average aflatoxin (57
µg/kg) in S. rautanenii and the lowest in T. indica (3 µg/kg, Table 4.2). More than eighty percent of S. rautanenii samples had aflatoxin concentrations above the regulatory limit for food in Zambia (10 µg/kg), with fruits containing as much as 128 µg/kg (Table 4.2).
Vangueriopsis lanciflora and Thespesia garckeana also had average aflatoxins above the regulatory limit for Zambia (12 µg/kg and 11 µg/kg, respectively), with 53% and 71% of fruits having aflatoxins concentrations higher than 10 µg/kg (Table 4.2). The average aflatoxin concentrations in Parinari curatellifolia (6 µg/kg), Zizyphus spp. (6 µg/kg) and
Adansonia digitata (4 µg/kg) were all below maximum allowable levels in food in
Zambian, although 10% of the Zizyphus spp. were above the regulatory limit (Table 4.2).
Fungi from ground fruit
Aspergillus section Flavi was recovered from all wild fruits and consisted of A. parasiticus, A. flavus L strain, S morphotype fungi and A. tamarii (Table 4.3). The overall quantities (CFU/g) of Aspergillus section Flavi in the fruits varied (ANOVA, F6,74
= 4.7008, P<0.001, Table 4.3) with the highest in Parinari curatellifolia (56 CFU/g) and the lowest in Tamarindus indica (3 CFU/g). Overall frequencies of Aspergillus section
Flavi on fruits differed among the fungi (ANOVA, F3,24 = 35.0131, P<0.001), with the A. flavus L morphotype the most frequent (77.5%; Tukey’s HSD, P<0.05). When each fruit species was considered separately, the A. flavus L strain occurred in the greatest concentrations (Tukey’s HSD, P<0.05) on all fruits except for V. lanciflora, where
# # 128# # amounts of A. parasiticus and A. flavus L morphotype were recovered (51.2% and 44.8
%, respectively). Although frequencies on fruits were similar for S morphotype fungi
(ANOVA, F6,100 = 1.0444, P=0.4169) and A. tamarii (ANOVA, F6,100 = 0.8472,
P=0.5440), higher frequencies of A. parasiticus were found on V. lanciflora (44.8%) and
T. garckeana (38.2%) than all other fruits (Tukey’s HSD, P<0.05, Table 4.3).
Wild fruits as substrate for growth by aflatoxin-producers
Wild fruits differed in ability to support growth by aflatoxin-producers (ANOVA,
F5,23=176.2224, P<0.001), with the highest average growth occurring on V. lanciflora
(1.1x109 CFU/g) and no growth detected on Tamarindus indica (Table 4.5). Average growth across all fruits did not differ among fungi (F4,24=0.0045, P>0.05). There were significant differences (ANOVA, F5,24=1740.5680, P<0.001) in A. parasiticus NRRL
2999 growth among fruit, with the most propagules produced on V. lanciflora (6.1x108
CFU/g) (Table 4.5). Similarly, the A. flavus S morphotype (AF70; ATCC MYA384) and
9 9 the un-named taxon SBG (A-11612) grew most on V. lanciflora (1.7x10 and 1.9x10
CFU/g, respectively) (Table 4.5). On the other hand, A. minisclerotigenes (A-11612) grew the most on Thespesia garckeana (6.5x108 CFU/g). None of the fungi grew on
Tamarindus indica (Table 4.5).
Wild fruits as substrate for aflatoxin production
With the exception of Tamarindus indica, all fruit species supported production of aflatoxin concentrations greater than 10 mg/kg (Table 4.6). On average, there were significant differences in concentrations of aflatoxins produced among fruits (ANOVA,
F5,24 = 61.9592, P<0.001), with the highest concentrations produced on T. garckeana
(average = 73,500 µg/kg) and none detected on T. indica (Table 4.6). Overall, there were
# # 129# # no significant differences (ANOVA, F4,25 = 0.0215, P = 0.9990) in concentrations of aflatoxins produced among the six Aspergillus section Flavi (Table 4.6). Aspergillus parasiticus produced higher concentrations of aflatoxins on S. rautanenii (182,200
µg/kg) and T. garckeana (131,100 µg/kg) than on Zizyphus spp. (16,100 µg/kg), P. curatellifolia (5,081 µg/kg), V. lanciflora (2,729 µg/kg) or T. indica (Tukey’s HSD, P <
0.05). Aspergillus flavus L (AF13; ATCC 96044; SRRC 1273) and S (AF70; ATCC
MYA384) morphotypes each produced similar concentrations on fruits except
Tamarindus (Table 4.6). On the other hand, A. minisclerotigenes was the most aflatoxigenic species on Vangueriopsis (76,604 µg/kg). The highest concentrations of aflatoxin produced by any isolate-fruit combination was by the un-named taxon SBG on
Thespesia garckeana (212,596 µg/kg).
# # 130# #
Discussion
Wild fruits contribute aflatoxins to the Zambian diet!
Aflatoxin contamination of crops and foods is detrimental to human health, livestock productivity and trade (Gong et al. 2004; Liu et al. 2012; Probst et al. 2007;
Reddy and Raghavender 2007; Turner et al. 2003; Van Egmond et al. 2007; Wu 2014).
Consumption of highly contaminated food resulted in rapid deaths in Kenya (Lewis et al.
2005; Probst et al. 2007), Tanzania and India (Reddy and Raghavender 2007). Given that wild plants may become contaminated with aflatoxins (Boyd and Cotty 2001) and consumption of wild fruits and products containing wild fruits is expected to rise
(Akinnifesi et al. 2006; Akinnifesi et al. 2002; Buchmann et al. 2010; Chadare et al.
2008), it is important to ensure these foods are free from aflatoxins. In the current study, aflatoxins were detected in all fruits (Table 4.2) similar to what was previously reported for wild fruits in North America (Boyd and Cotty 2001). However, to the best of our knowledge, this is the first report of aflatoxins in wild tropical fruits including S. rautanenii, V. lanciflora, Thespesia garckeana, P. curatellifolia, Zizyphus spp. and
Adansonia digitata. As is the case with crop plants, susceptibility to aflatoxin contamination in wild fruits varies among species (Boyd and Cotty, 2001; Table 4.2), with some S. rautanenii samples having aflatoxin concentrations exceeding 125µg/kg, compared with no aflatoxins in Tamarindus indica. The differences among fruits in concentrations of aflatoxins may be attributable to chemical composition (Mehl and
Cotty 2013), variation in structures of communities of Aspergillus section Flavi , and/or fruit specific processing and storage practices. Although it is not known whether the aflatoxins detected in S. rautanenii were mostly in the nut or pulp, both components of
# # 131# # the fruit are important in human diets and would contribute to aflatoxin exposure through the Zambian diet.
Aspergillus section Flavi in Zambia’s wild fruits
Aspergillus section Flavi community structure and aflatoxin-producing potential of fungi present in crops are important factors influencing contamination (Mehl et al.
2012; Probst et al. 2007). In wild fruits of Zambia, the A. flavus L morphotype was the most frequent (77.5%) Aspergillus section Flavi (Table 4.3). High frequencies of A. flavus L morphotype on fruit are not surprising because even though soils in forests and cultivated areas of Zambia are dominated by A. parasiticus (Kachapulula unpublished),
A. flavus L morphotype is a much more competitive colonizer of crops (Horn 2005), and potentially also wild plants. However, it was unexpected that high amounts of aflatoxins were found on S. rautanenii given that A. flavus L morphotype from Zambia are mostly either atoxigenic or have low aflatoxin-producing potential (Kachapulula et al. 2017). In addition, the high aflatoxin-producing taxa, A. parasiticus and S morphotype fungi, were either absent or occurred at inconsequential frequencies (Cotty et al. 2008) on S. rautanenii (Table 4.3). It is possible however, that S. rautanenii selects for aflatoxigenic
A. flavus L morphotype (Sweany et al. 2011). Aflatoxin-production is highly variable among members of the L morphotype of A. flavus and both high aflatoxin-producers and atoxigenic isolates may be found (Cotty 1989). On the other hand, S. rautanenii, V. lanciflora and Thespecia garckeana had appreciable levels of S morphotype fungi and/or
A. parasiticus (Table 3) and unsurprisingly (Cotty et al. 2008), high concentrations of aflatoxins (Table 4.2). The low levels of aflatoxins in Parinari curatellifolia, a plant with close to 13 % S morphology fungi (Table 4.3), could imply that environmental conditions
# # 132# # in Zambia for the sampling years were not favorable for fruit infection and subsequent aflatoxin-production.
Wild fruits as substrate for growth by aflatoxin-producing fungi
The suitability of wild fruits as substrate for growth of aflatoxigenic fungi was evaluated by quantifying the propagules produced by reference aflatoxigenic fungi. All tested fruits supported growth by aflatoxigenic fungi except Tamarindus indica (Table
4.5). Both crops (Mehl and Cotty 2013) and wild plants (Boyd and Cotty 2001) are susceptible to infection by aflatoxin-producing fungi. Invasion of host tissues requires plant-degrading enzymes, ability to evade or suppress host defenses, and capacity to overcome plant antimicrobial compounds (Agrios 2005). In the current study,
Vangueriopsis lanciflora supported the greatest growth by all aflatoxigenic fungi. On the other hand, low amounts of Aspergillus section Flavi were found in Tamarindus indica market samples (Table 4.3) and no growth by aflatoxin-producers was detected even after inoculation (Table 4.5). Phenolic antioxidants from Tamarindus indica have been shown to have antifungal properties (Luengthanaphol et al. 2004; Sudjaroen et al. 2005). These antioxidants may be responsible for the absence of aflatoxin contamination and inhibition fungal growth even when inoculated with high spore levels, on Tamarindus indica. Wild fruits are important sources of income and food in Zambia. Results of the current study suggest that Tamarindus indica is a food free from aflatoxin risk. However, all the other examined wild fruits supported growth by aflatoxin-producing fungi and therefore potential measures for aflatoxin mitigation might include strategies for reducing growth of aflatoxigenic fungi on wild fruits.
# # 133# # Aflatoxin production on wild fruits
Aflatoxin-producers may grow on substrates without producing aflatoxins.
Aflatoxin production is dependent on environmental conditions (Cotty and Jaime-Garcia
2007; Kachapulula et al. 2017) and the nutritional content of the substrate (Mehl and
Cotty 2013). Fruit suitability for growth of aflatoxin-producers contributes to the fruit vulnerability. However, the quantity of Aspergillus section Flavi on market samples was not related to the quantity of aflatoxins in those fruits (Table 4.4). This suggests factors other than just support of growth dictate the extent of contamination.
Inoculation of wild fruits with known aflatoxin-producers representing several fungal species, under an environment conducive for aflatoxin production, provide insight into the innate capacity of the examined fruits to become contaminated. The fruits varied widely in ability to support aflatoxin biosynthesis and the fruit most vulnerable to contamination was dependent on the infecting fungus. One fruit (Tamarindus indica) supported no aflatoxin production, but the rest became contaminated with very high concentrations of aflatoxins (Table 4.6). Even fruits for which market samples were safe for consumption (Parinari curatellifolia and Zizyphus spp., Table 4.2) developed high
(>6,000 and >18,000 µg/kg for Parinari curatellifolia and Zizyphus spp., respectively) concentrations of aflatoxins after inoculation with aflatoxin-producers (Table 4.5). The second largest concentrations of aflatoxin observed (over 180,000 µg/kg total aflatoxins) resulted from inoculation of the most frequently contaminated fruit from the market
(Schinziophyton rautanenii) with A. parasiticus, the most common aflatoxin producer in
Zambia (Kachapulula unpublished). However, in the current study, the fungus inoculated dictated which fruit was most susceptible to contamination. When fruits were inoculated
# # 134# # with A. parasiticus, Schinziophyton rautanenii supported higher aflatoxin concentrations than Parinari curatellifolia or Zizyphus spp. These results suggest that when assessing fruit vulnerability to contamination both the frequencies of specific aflatoxin-producers in the habitat and the suitability of the fruit for contamination by the most common fungi present must be considered. The current study was the first to quantify both the infecting fungi and fruit suitability for aflatoxin production in Zambia’s diverse forest to market system.
Examples where the aflatoxin-producing potential of the infecting fungal community may have modulated the extent to which wild fruits became contaminated may exist in the current study with Parinari curatellifolia and Zizyphus spp. Parinari curatellifolia and Zizyphus spp supported high aflatoxin concentrations when inoculated in the lab but were also most frequently infected by the A. flavus-L morphotype (Table
4.3), which for Zambia, is mostly either atoxigenic or of low aflatoxigenicity
(Kachapulula et al. 2017). These results suggest that the low aflatoxin content observed in Parinari curatellifolia and Zizyphus spp. market samples is not a result of innate resistance of the fruits, but a combination of safe handling (Kachapulula et al. 2017) and safe fungal community composition (Cotty et al. 2008; Probst et al. 2007; Probst et al.
2010).
Aflatoxin contamination and aflatoxin-producing fungi are common on wild fruits from Zambia. With the exception of Tamarindus indica, all wild fruits either had aflatoxin levels above regulatory limits for Zambia, or were suitable substrates for growth and aflatoxin production by aflatoxigenic fungi. Aflatoxin preventative measures, the most effective of which is biological control, have traditionally targeted susceptible
# # 135# # agricultural products including cereal and tree-nut crops and their byproducts. The current study indicates that wild fruits could also be routes for exposure of the Zambian population to aflatoxins. Aflatoxin mitigation measures should be expanded to include wild fruits in Zambia.
# # 136# # Figure 4.1 Schinziophyton rautanenii (top row) and Parinari curatellifolia (bottom row) from the markets in Zambia.
# Top row: (Schinziophyton rautanenii), first two images is nut and last two are fruit without and with pulp. Bottom row (Parinari curatellifolia) first image on left is nut, second is cross-section of fruit showing characteristic two nuts and last two are fruit without and with pulp.
# # 137# # Figure 4.2 Growth of Aspergillus section Flavi on inoculated fruits
Left to right: A. flavus L morphotype on Parinari curatellifolia, A. parasiticus on
Thespesia garckeana and A. minisclerotigenes on Vangueriopsis lanciflora. All other fungi had similar growth on all fruits except on Tamarindus indica.
# # 138$ $ Table 4.1. Common wild fruits in Zambia and their uses
Fruit species Uses Reference Fruit pulp used to make beverages, porridges. Leaves, bark Chadare et al. 2008; Rahul et Adansonia digitata and seeds medicinal al. 2015;Vermaak et al. 2011 Fruit pulp used to make beverages, porridges. Leaf extracts Akinnifesi et al. 2006; Parinari curatellifolia medicinal. Nuts eaten as snacks and used to extract oil Benhura et al. 2015 Fruit pulp used to make beverages, porridges. Nuts eaten as Zimba et al. 2005; Vermaak et Schinziophyton rautanenii snacks and used to extract oil al. 2011 Fruit pulp used to make beverages, porridges. Leaves are used Tamarindus indica Ebifa-Othieno et al. 2017; in medicines and used in feed Thespesia garckeana Fruit pulp used to make beverages, porridges. Medicinal Maroyi 2011 Personal communication (Ms. Vangueriopsis lanciflora Fruit pulp used to make beverages, porridges. Lubinda) Zizyphus spp. Beverages, jams, cakes, medicinal Maruza et al. 2017
$ $ 139$ $ Table 4.2 Aflatoxin in wild fruits from Zambia. Samples Aflatoxin (µg/kg) Samples in categories (%) Species (#) Mean Range* < 4 µg/kg 4-9.9 µg/kg 10-19.9 µg/kg > 20 µg/kg Adansonia digitata 9 4BC ND-5 66.7 33.3 0 0
Parinari curatellifolia 17 6BC ND-8.6 29 71 0 0
Schinziophyton rautanenii 24 57A 3.4-128.6 4.5 13.6 9.1 72.8
Tamarindus indica 25 3C ND-9 78.3 21.7 0 0
Thespesia garckeana 20 11B 3.9-23.2 5.9 41.2 47 5.9
Vangueriopsis lanciflora 7 12B 6.6-18.9 0 28.6 71.4 0
Zizyphus spp. 12 $$ 6BC ND-24.4 $$ 70 20 10 0 Samples were composed of multiple fruits and weighted 300 to 500 g. Means followed by the same letter are not significantly different (P<0.05) by Tukey-Kramer’s HSD test. Limit of detection = 2 µg/kg. *ND= non-detected
$ $ 140$ $ Table 4.3 Distribution of fungi of Aspergillus section Flavi on wild fruits from Zambia.
Samples Wild fruit species %L*† %S %P %T CFU/g (#) Adansonia digitata 9 81.9AB(X) 9.8A(Y) 8.3BC(Y) 0A(Y) 8AB Parinari curatellifolia 17 87.5AB(X) 12.5A(Y) 0C(Y) 0A(Y) 56A Schinziophyton 24 97.1A(X) 2.9A(Y) 0C(Y) 0A(Y) 13AB rautanenii Tamarindus indica 25 88.4AB(X) 3.6A(Y) 8BC(Y) 0A(Y) 3B Thespecia garckeana 20 45.3C(X) 13.9A(Y) 38.2A(X) 2.7A(Y) 10A Vangueriopsis lanciflora 7 51.2BC(X) 0A(Y) 44.8A(X) 0A(Y) 5AB Zizyphus spp. 12 91.3AB(X) 8.7A(Y) 0C(Y) 0A(Y) 38A Average 77.5X 7.3Y 14.2Y 0.4Y 19.0 *L, S, P and T represent A. flavus L-morphotype, S-morphotype fungi, A. parasiticus and A. tamarii, respectively. †Percent data were arcsine transformed and CFU/g data were log transformed prior to analyses but actual means are presented here. Means followed by the same letter (A/B/C) among wild fruits (columns) or (X/Y/Z) among fungal species or morphotypes (rows) indicate means do not significantly differ by Tukey’s HSD test (α= 0.05).
$ $ 141# # Table 4.4 Regression analyses of aflatoxin concentrations as explained by total quantity of Aspergillus section Flavi in fruits
Coefficient of Model significance Fruit genera Intercept Rate of increase† determination (P) †† (R2) Quantity in Adansonia (CFU/g) 0.5090815 0.3679195 0.226 0.1958 Quantity in Parinari (CFU/g) 2.214135 -0.26716 0.221 0.0771 Quantity in Schinziophyton (CFU/g) 3.6218153 0.2103414 0.093 0.25 Quantity in Thespesia (CFU/g) 2.5777619 -0.144557 0.028 0.5183 Quantity in Vangueriopsis (CFU/g) 2.5981314 -0.112625 0.066 0.6762 Quantity in Zizyphus spp. (CFU/g) 0.7441127 0.2640896 0.34797 0.6005 CFU g-1 was log-transformed prior to analyses. †This value represents the change in aflatoxin for a unit change in CFU g-1 of fruit. Negative values reflect aflatoxin reduction. ††Significance set at P = 0.05.
# # 142# # Table 4.5 Propagules of five Aspergillus section Flavi fungi on inoculated wild fruits from Zambia.
Growth of aflatoxin-producers on fruits (CFU/g) Wild fruit Species A. Un-named A. parasiticus† A. flavus-S A. flavus-L Average minisclerotigenes SBG Parinari curatellifolia 1.0x108B(X) 8.1x107B(X) 4.6x107B(Y) 1.5x108B(X) 9.1x107B(X) 9.4x107C Schinziophyton rautanenii 1.2x108B(X) 2.3x106C(Z) 2.7x108A(X) 5.7x107C(Y) 3.1x107C(Y) 9.5x107C Tamarindus indica ND* NDD NDC NDE NDE NDD 2.2 Thespesia garckeana 1.6x107C(Z) 8.7x107B(Y) 6.5x108A(W) 1.3x108B(Y) 2.2x108B x108A(X) Vangueriopsis lanciflora 6.1x108A(Y) 1.7x109A(X) N/A 1.5x108B(Z) 1.9x109A(X) 1.1x109A 7.3 7.3 Zizyphus spp. 2.0x106D(Y) 4.4x106D(Y) 7.3x106D(Y) 3.2x107C x107B(X) x107B(X) Average 1.7x108X 4.0x108X 1.5x108X 2.0x108X 4.4x108X †A. parasiticus = Isolate NRRL 2999, A. flavus-S = Isolate AF70/ATCC MYA384, A. flavus-L = Isolate AF13/ATCC
96044/SRRC 1273, A. minisclerotigenes = Isolate A-11611, Un-named SBG = Isolate A-11612. Means followed by the same (A/B/C) among fruits (columns) or row (X/Y/Z) among fungal species or morphotype (rows) were not significantly different (P<0.05) by Tukey-Kramer’s HSD test. *ND= non-detected.
# # 143# # Table 4.6 Aflatoxin production by 5 aflatoxin-producers on six wild fruit species from Zambia.
Aflatoxin production (µg/kg) Wild fruit genus †A. parasiticus A. flavus-S A. flavus-L A. minisclerotigenes Un-named SBG Average Parinari 5,081BC(X) 1,517A(X) 14,655A(X) 7,950AB(X) 5,573B(X) 6,955A Schinziophyton 182,214A(X) 17,256A(Y) 6,356A(Y) 31,407AB(XY) 19,109B(Y) 51,268A Tamarindus 0D 0B 0B 0C 0C 0B Thespesia 131,065A(X) 9,843A(Y) 9,395A(Y) 4,730B(Y) 212,596A(X) 73,526A
Vangueriopsis 2,729C(X) 15,687A(XY) 33,336A(XY) 76,604A(X) 5,594B(XY) 26,790A
Zizyphus spp. 16,060B(X) 39,222A(X) 11,946A(X) 4,338B(X) 20,113B(X) 18,336A Average 56,192X 13,921X 12,615X 20,839X 43,831X †A. parasiticus = Isolate NRRL 2999, A. flavus-S = Isolate AF70/ATCC MYA384, A. flavus-L = Isolate AF13/ATCC
96044/SRRC 1273, A. minisclerotigenes = Isolate A-11611, Un-named SBG = Isolate A-11612. Means followed by the same letter (A/B/C) among fruits (columns) or (X/Y/Z) among fungal species or morphotypes were not significantly different
(P<0.05) by Tukey-Kramer’s HSD test.
# # 144# #
CHAPTER 6 - FUNGI CLOSELY RELATED TO ASPERGILLUS PARASITICUS
IN ZAMBIA
To be submitted to: Applied and Environmental Microbiology
Introduction
Aflatoxins are carcinogenic fungal metabolites that contaminate maize, groundnuts, wild fruits and insects in Zambia (Kankolongo et al. 2009; Kachapulula et al.
2017; Bumbangi et al. 2016). Consumption of food contaminated with aflatoxins may lead to liver cancer, suppression of the immune system, stunting, reduced weight-gain, and rapid death (Gong et al. 2004; Lewis et al. 2005; Liu et al. 2012; Probst et al. 2007;
Reddy and Raghavender 2007; Turner et al. 2003; Van Egmond et al. 2007; Williams et al. 2004). Crops, food, and feed contaminated with aflatoxins are excluded from high value markets and may have to be destroyed, resulting in economic losses (Van Egmond et al. 2007). Aflatoxin mitigation measures requires a clear understanding of the causal agents of contamination (Cotty et al. 2008).
Fungi frequently implicated in aflatoxin contamination of crops in Africa belong to Aspergillus section Flavi. These include Aspergillus flavus (produces only B aflatoxins), A. parasiticus (produces both B and G aflatoxins) and two unnamed taxa SB
(only B aflatoxins) and SBG (both B and G aflatoxins)(Cotty et al. 2008; Probst et al.
2010). In Zambia, fungi morphologically assignable to A. parasiticus both dominate
Aspergillus section Flavi populations in cultivated and non-cultivated soils of Zambia, and infect maize and groundnut at frequencies higher than observed elsewhere
(Kachapulula et al. 2017). Aspergillus parasiticus-like fungi are the most important etiologic agents of maize and groundnut aflatoxin contamination across the agroecologies
# 145# # of Zambia (Kachapulula unpublished). However, relationships of the Aspergillus parasiticus-like fungi in Zambia to typical A. parasiticus and previously described taxa from other regions (Garber and Cotty 2014) is unknown. Improved characterization of
Zambia’s A. parasiticus-like fungi may be key to development of effective management for that nation. Aspergillus parasiticus-like fungi in Zambia were identified based on morphology alone (Kachapulula et al. 2017). Morphological differences between A. parasiticus and close relatives such as A. sojae are not always evident (Chang et al.
2007), and relationships among these closely related species is best determined with
DNA sequence information. Previously unrecognized lineages of fungi may play important roles as causal agents of aflatoxin contamination. In Kenya, contamination events causing severe aflatoxicoses were attributed to a previously unknown fungi morphologically similar to the S strain morphotype of A. flavus (Probst et al. 2007).
Further work detected multiple lineages of fungi with similar S strain morphology causing contamination in various locations across East and West Africa. Similarly, divergent lineages of Aspergillus parasiticus-like fungi may contribute to crop aflatoxin contamination in Zambia. Improved knowledge on these important etiologic agents of maize and groundnut aflatoxin contamination in Zambia may reveal potential management options (Cotty et al. 2008).
Because aflatoxin mitigation requires a clear understanding of etiology (Mehl et al. 2012; Cotty et al. 2008), the current study sought to determine relationships of A. parasiticus-like fungi from Zambia to previously characterized taxa from other parts of the world. The results suggest that the prevalent Aspergillus parasiticus-like causative agents of aflatoxin contamination in Zambia compose 4 distinct lineages, two of which
# 146# # represent distinct previously undescribed taxa. All four lineages can contaminate maize and groundnut under controlled conditions with concentrations of aflatoxins above 1 mg/kg.
# 147# # Materials and methods
Fungal isolates
Fourty eight fungal isolates belonging to Aspergillus section Flavi were used for morphological and phylogenetic comparisons (Table 5.1). The isolates consisted of 37 putative A. parasiticus from Zambia, 1 sugarcane-associated A. parasiticus isolate
(Garber and Cotty 2014), A. parasiticus ex-type (NRRL 502), 2 A. parasiticus from previous studies (NRRL 2999 and BN009E)(Kumeda et al. 2003), 2 putative A. flavus isolates from Zambia, 1 isolate of A. minisclerotigenes (A-11611), 1 isolate of un-named
SBG (A-11612), A. flavus ex-type (NRRL 1957) and 4 A. flavus isolates previously reported by this laboratory (Table 5.1). Fungi previously reported as close relatives of A. parasiticus were also included for comparison. These were A. sojae (NRRL 1988), A. toxicarius (NRRL 13075), A. terricola var americanus type isolate (NRRL 424), and A. parvisclerotigenus (NRRL 62796) (Gonçalves et al. 2012; Varga et al. 2011). Chipata district (n=16) and Mpongwe district (n=13), and from non-cultivated areas (n=13) were also analyzed in a similar manner. Cultures were grown in the dark at 31°C on 5-2 medium (Cotty 1989), and isolates were stored as plugs of sporulating cultures in sterile distilled water at 4°C.
Morphology
Isolates from 7-day-old cultures (5-2 agar, 31°C, darkness) were identified by macroscopic colony and microscopic (×400) conidia characteristics (Klich and Pitt 1988;
Kurtzman et al. 1987). Isolates were identified as A. parasiticus if they had dark green to olive colonies with rough-surfaced conidia, A. flavus L strains if they had light green, grayish green or moss green colony appearance with smooth-surfaced conidia or A.
# 148# # tamarii if colonies were olive-brown to near-coffee colonies with roughened thick-walled conidia (Kornerup and Wanscher 1978; Klich and Pitt 1988). Development of a bright orange color on the reverse side of Aspergillus flavus and parasiticus agar (AFPA) medium (31°C, 5 days) was required for confirmation of either A. parasiticus or A. flavus
(Cotty 1994; Pitt et al. 1983).
DNA isolation.
Genomic DNA was isolated from isolates previously subjected to single conidium transfer as described (Callicott and Cotty 2015). Briefly, spores were dislodged from cultures (31°C, 7 days, darkness) on 5-2 agar (Cotty 1989) into 1.5 ml of 0.1% Tween-80
10 mM EDTA with sterile cotton swabs. The spore suspension was centrifuged at 8,000 x g in a 2 ml eppendorf tube for 5 minutes. The supernatant was discarded and lysis buffer (450 µl, 30 mmol l−1 Tris, 10 mmol l−1 EDTA, 1% SDS, pH 8.0) was added and incubated in a Thermomixer 5436 (Eppendorf, Inc., Hamburg, Germany) for 1 h, 60°C and 800 rev min−1. After removing cell fragments by centrifugation, DNA was precipitated with ammonium acetate and ethanol (Sambrook et al. 1989) and re- suspended in 25 µl sterile water. DNA was quantified with a spectrophotometer (model
ND-1000, NanoDrop) and diluted to a final concentration of 5–10 ng/µl prior to PCR.
Polymerase chain reaction amplification and phylogenetic analyses
Sequences from three genes; aflatoxin biosynthesis transcription factor (aflR, 1.8 kb), nitrate reductase (niaD, 2.2 kb), and calmodulin (cmd, 1.9 kb), were amplified, sequenced, and used in phylogenetic analyses. Three overlapping pieces of each gene were amplified using the primers and conditions listed in Table 5.2. PCR reactions were carried out in 20 µl of AccuPower HotStart PCR PreMix polypropropylene tube strips
# 149# # (Bioneer Corporation, Daejeon, Korea) using 18 µl of autoclaved, distilled, deionized water and 2 µl of genomic DNA (15 ng/µl final DNA concentration). All PCR reactions included a denaturation and hot start step of 5 min at 94°C. For the aflR primers conditions included 35 cycles of 94°C for 20 s, 57°C for 20 s, and 72°C for 30 s; and final extension at 72°C for 10 min. For niaD hot start was followed by 38 cycles of 94°C for 20 s, 58°C for 20 s, and 72°C for 30 s; and final extension at 72°C for 10 min. For cmD 2F-2R primers hot start was followed by 38 cycles of 94°C for 20 s, 48°C for 20 s, and 72°C for 30 s; and final extension at 72°C for 5 min. For cmd 3F-3R primers, conditions were similar to that of cmD 2F-2R, except the annealing temperature was
52oC. For cmD 42-627 primers hot start was followed by 38 cycles of 94°C for 20 s,
56°C for 20 s, and 72°C for 30 s; and final extension at 72°C for 5 min. PCR products were separated by electrophoresis on 1% agarose gels to verify singularity. Excess primers and unincorporated nucleotides were degraded with ExoSAP-IT (USB
Corporation, Cleveland) (1 µl of ExoSAP-IT in 16 µl of PCR product at 37°C for 1 h, followed by 85°C for 15 min). Bidirectional sequencing of all products was performed by the University of Arizona core sequencing facility, UAGC, with a 3730XL DNA
Analyzer (Applied Biosystems, Foster City, CA).
Phylogenies were created by Maximum likelihood (ML) using PhyML version
3.0 (Guindon et al. 2010) and by Bayesian analysis using Mr Bayes version 3.2.1
(Ronquist and Huelsenbeck 2003). Support for inferred topologies was evaluated with
1000 bootstrap replicates and branches with less than 80% support or less than 95% posterior probability were collapsed. Gaps were treated as missing data. Trees were
# 150# # rooted with A. nomius (NRRL 13137) and visualized with FigTree v.1.3.1
(http://tree.bio.ed.ac.uk/).
Aflatoxigenicity of lineages
To characterize aflatoxigenicity of identified lineages, at least three isolates from each clade were inoculated onto undamaged, sterile maize kernels (10g/250 ml
Erlenmeyer flask) previously autoclaved for 60 minutes, cooled to room temperature, and adjusted to 30% water content. Freshly-harvested conidia of each isolate from 7-day-old monoconidial cultures (grown on 5% V8-juice; 2% agar, pH 5.2) were cultured (7 days,
100% RH, 31oC) on maize (1x106 conidia per flask). After incubation, sample cultures were blended in 50 ml, 70 % methanol. The slurry was allowed to separate for 30 minutes and the supernatant was spotted directly onto thin-layer chromatography (TLC) plates
(Silica gel 60, EMD, Darmstadt, Germany) adjacent to aflatoxin standards (Aflatoxin
Mix Kit-M, Supelco) containing known quantities of aflatoxins B1, B2, G1 and G2. Plates were developed in ethyl ether-methanol-water, 96:3:1, air-dried, and aflatoxins visualized under 365-nm UV light. Aflatoxins were quantified directly on TLC plates using a scanning densitometer (TLC Scanner 3, Camag Scientific Inc, Wilmington, N.C.).
Growth rates of lineages
To characterize lineages further, growth on 0.1X Czapeck’s agar (Mehl and Cotty
2013) was determined. Each liter of 0.1X Czapek’s agar contained 8.8 mM sucrose and
3.5 mM NaNO3, while quantities for the rest of the ingredients remained at the same levels as in full-strength Czapek’s media (i.e each liter contains 0.5 g each of K2HPO4,
. KH2PO4, MgSO4 7H2O, and KCl, and 1 ml micronutrient solution containing 0.11 g
# 151# #
. . . MnSO4, 0.5 g (NH4)6Mo7O24 4H2O, 10.0 g Fe2(SO4)3 6H2O, 17.6 g ZnSO4 7H2O, 0.7 g
. . Na2B4O7 10H2O, and 0.3 g CuSO4 5H2O and 20 g Bacto agar).
Briefly, single-spore isolates were cultivated on 5-2 agar (5% V8 juice, 2% agar, pH 5.2; Cotty 1989), and plugs of sporulating cultures were suspended in 4-ml vials containing sterile distilled water. Conidial suspensions were used to centrally seed 5-2 agar and incubated (7 days, 31oC). Conidia were dislodged from plates with sterile cotton swabs and suspended in sterile distilled water. Conidia were quantified by measuring turbidity of suspensions (Turbidimeter; Orbeco Analytical Systems, Farmingdale, NY) and using a nephalometric turbidity unit (NTU)-versus-CFU standard curve, where y is equal to 49,937x (x is NTU, and y is conidia/ml) (Mehl and Cotty 2013). Conidia were adjusted to a concentration of 1x106 conidia per microliter. One hundred microliters of a suspension containing 1 million conidia was centrally seeded onto 0.1X Czapek’s agar plates and incubated (31oC). After 24 hrs, a circle was drawn on the underside of the plate to trace growth. Tracing was done again after 96 hrs, and growth was measured as the average distance between the first and second trace for each of 4 quadrants on a petri- dish. Growth rate was determined as average growth per hour over the 72-hr period.
Data analysis
Comparisons of both aflatoxin concentrations (µg/kg) and growth rates were performed by Analysis of Variance using general linear models (GLM) and Tukey’s HSD mean comparison test as implemented in JMP 11.1.1 (SAS Institute, Cary, NC). Paired t- test was used to compare aflatoxin B1 versus G1 produced by each phylogenetic group.
Aflatoxin concentrations were log-transformed prior to analyses. However, actual means are presented for clarity. All tests were performed at α = 0.05.
# 152# # Results
Clades of A. parasiticus in Zambia
All A. parasiticus isolates fell in one clade, sister to the A. flavus group (Figure
5.1). The A. flavus isolates resolved into a single clade, regardless of origin from either
North America or Africa. Reference A. parasiticus and the A. parasiticus-like fungi from
Zambia exhibited considerable diversity, resolving into four well supported clades. The group (Clade 2) with all the reference A. parasiticus contained 5% of the isolates resembling A. parasiticus. Clade 1, 3 and 4 contained 8 to 32 percent of the A. parasitiucs-like fungi, but 24% of these fungi did not form a distinct lineage instead forming a polytomy basal to groups 3 and 4. Additional phylogenetic analyses were performed (Figure 5.2) to assess relationships of the Aspergillus parasiticus-like fungi to previously described fungi with similar morphology including A. toxicarius, A. novoparasiticus, A. sojae and A. terricola var americanus. Only A. toxicarious was closely related to one of the clades of Zambian fungi (Clade 3, Figure 5.2).
Frequencies of A. parasiticus-like fungi in single fields and forests
Of 16 isolates from maize, groundnut and soil planted to both crops from a single field in Chipata, isolates belonging to clades 2 (19% of isolates), 3 (37% of isolates), and
4 (44% of isolates) were found (Figure 5.2; Table 5.5). No fungi belonging to clade 1 were detected in the Chipata field. Isolates from maize, groundnut and soil planted to both crops in a single field from Mpongwe consisted of fungi from the clades 3 (69% of isolates) and 4 (31% of isolates). Clades 1 and 2 were not detected in the Mpongwe field.
Both fields had mixed cropping. Aspergillus parasiticus-like fungi from non-cultivated soils in forests consisted of isolates from clades 1 (12% of isolates), 3 (44% of isolates)
# 153# # and 4 (44% of isolates). No fungi belonging to clade 2 were detected from forest soils
(Table 5.5).
Aflatoxigenicity of A. parasiticus clades
There were no significant differences (F3,53=0.5834, P=0.6351) in average total aflatoxin production among fungi in clades 1 (86,900 µg/kg), 2 (142,200 µg/kg), 3
(102,000 µg/kg) and 4 (90,000 µg/kg, Table 5.3). Although aflatoxins B1 and G1 were produced in similar concetrations in each of the clades 1, 3 and 4, production of G1
(62,000 µg/kg) was higher (Paired t-test, P=0.0014) than that of B1 (23,100 µg/kg) by fungi in clade 2 (Table 5.3).
Growth rates of A. parasiticus clades
There were no significant differences in average growth rates among clades at
o o 20 C (F3,52=1.7479, P=0.1687) or 30 C (F3,52=0.8412, P=0.4775, Table 5.4). Growth
o rates among clades varied at 25 C (F3,52=13.7264, P<0.001), with rates higher in clades 2
(0.0241 mm/hr) and 3 (0.0240 mm/hr) and lower in clades 1 (0.0192 mm/hr) and 4
(0.0201 mm/hr, Table 5.4). For all clades, the highest growth rates were at 30oC and 35 oC, followed by 25oC, with 20 oC supporting the slowest growth (Table 5.4).
# 154# # Discussion
The impact of aflatoxin-contaminated crops on liver cancer, stunting, immune suppression, and reduced incomes for farmers are well documented (Gong et al. 2004;
Liu et al. 2012; Turner et al. 2003; Van Egmond et al. 2007; Williams et al. 2004; Wu
2014). Aflatoxin contamination of crops may be attributed several distinct taxa including
A. flavus, A. parasiticus, and two unnamed taxa (SB and SBG). However, etiologic agents of aflatoxin contamination of crops may vary by region (Cotty and Cardwell 1999; Probst et al. 2007; Horn and Dorner 1998). In Zambia, A. parasiticus-like fungi are the most frequent members of Aspergillus section Flavi in soils and the chief etiologic agents of aflatoxins in maize and groundnuts in all agroecologies (Kachapulula et al. 2017). The current study provides insights into the diversity among A. parasiticus-like fungi in
Zambia resident on maize, groundnut and soils. The results suggest there are four distinct, common, and well supported lineages (Figure 5.1) of Aspergillus parasiticus-like fungi with similar aflatoxin-producing potentials (Table 5.3) and similar temperature requirements for growth (Table 5.4).
New taxa closely related to A. parasiticus in Zambia
A multi-gene phylogeny constructed with a total of 5.9 kb of sequence resolved four common highly supported lineages of A. parasiticus-like fungi from Zambia. One of the lineages contains all the reference isolates of A. parasiticus and the reference isolate of A. sojae. However, the four lineages are divergent form each other, and consideration should be given to erecting a new taxon for each of the three unnamed taxa. All four clades consisted of fungi from multiple agroecologies, which may reflect movement among regions and/or adaptation to a wide range of environments. For example, group 1
# # 155# # was detected in all three agroecologies of Zambia. Given that agroecologies differ in average annual rainfall, temperature and soil characteristics (Bunyolo et al. 1995), the presence of isolates from multiple locations might indicate high adaptability to environments and to multiple niches.
New fungal taxa closely related to A. parasiticus have been reported in the recent past (Gonçalves et al. 2012; Garber and Cotty 2014; Pildain et al. 2008). Garber and
Cotty (2014) reported a group of A. parasiticus associated with sugarcane production in
USA and Japan. Although the sugarcane-associated group was sister to group 1 (Figure
5.2), the group was not detected in the current study either because it does not occur in high enough frequencies in Zambia to be detected or because no sugarcane fields were sampled in the current study. Similarly, no A. novoparasiticus was found. Fungi from group 2 were closely related to A. sojae, A. terricola var americanus and previously characterized A. parasiticus while those from group 3 formed a clade with A. toxicarius
(Figure 5.2). Aspergillus toxicarius has been described as a synonym of A. parasiticus
(Christensen 1981; Wicklow 1983). The close association between A. parasiticus, A. sojae, A. terricola var americanus and A. toxicarius has led to suggestions that A. parasiticus should be viewed not as a single species but a species complex (Gonçalves et al. 2012). However, fungi in both groups 2 and 3 from Zambia produced aflatoxins, while both A. sojae and A. terricola var americanus are atoxigenic (Chang et al. 2007). Another close relative of A. parasiticus, A. arachidicola was not included in the phylogenetic analyses in the current study because it’s colony lacks the characteristic color for A. parasiticus and has a negative reaction on AFPA (Pildain et al. 2008). Groups 1 and 4 are distinct from previously described A. parasiticus and may be unique to Zambia. Eight of
# # 156# # the A. parasiticus-like fungi did not fall into any of the four clades, suggesting they might represent rare, yet to be identified clades. Fungi closely resembling A. parasiticus have also been reported on almonds in Portugal (Rodrigues et al. 2011). Without access to sequences or cultures, it remains unknown how the Portugal groups relate the new groups reported in Zambia.
Previous reports have used less sequence to reconstruct phylogenies among A. parasiticus-like fungi. In the current study 5.9 kb of sequence was used. The improved resolution associated with a large amount of sequence may be the reason new groups were discovered in the current study. Furthermore, the large amount of sequence used here lends credence to the identified groups.
Sympatric occurrence of A. parasiticus-like fungi in single fields and forests
On average, a household in Zambia cultivates up to1 hectare of land (Tembo and
Sitko 2013). Cropping is generally mixed with multiple crops in close proximity to each other. The diversity of A. parasiticus-like fungi in a single field was fairly high, with most clades found across the country. Fungi from each of several clades were found to occur in individual fields in Chipata and Mpongwe districts. Although sexual reproduction in A. parasiticus has been reported before (Horn et al. 2009), the sympatric occurrence of clades observed in the current study (Table 5.5) suggests that clonal reproduction is the main mode of propagation of A. parasiticus clades in Zambia.
Previously (Kachapulula unpublished), no differences in aflatoxin-producing potential were observed among A. parasiticus-like fungi from non-cultivated forests and from agricultural fields in Zambia, suggesting that fungi from forests may be influencing aflatoxin contamination of crops. In the current study, all but one clade from agricultural
# # 157# # fields were also found in forests, and the most frequent clades in cultivated soil and crops
(clades 3 and 4, Figure 5.1) were also the predominant clades in non-cultivated areas/forests (Table 5.5). These data are consistent with movement of A. parasiticus-like fungi from forests to agricultural fields, and further reinforces the importance of A. parasiticus-like fungi from forests in etiology of aflatoxin contamination of crops.
Aflatoxigenicity of A. parasiticus clades
Importance of each A. parasiticus-like lineage described in the current paper to crop contamination can be suggested by taking into consideration both frequencies of crop infection by and average aflatoxin-producing potential of that lineage (Mehl et al.
2012; Probst et al. 2007). Aflatoxin-producing potentials of the four lineages (groups 1,
2, 3 and 4, Figure 5.1) were high but similar (Table 5.3) and all four groups are widely distributed in Zambia (Table 5.1). This group of aflatoxin-producers are together important causal agents of contamination. However, distribution of these taxa does not appear to be the determining factor leading to increased incidences of contamination in
Zambia’s agroecology I (Kachapulula et al. 2017). Rather, increased levels in agroecology 1 is more likely attributable to the warm, dry conditions.
Although the aflatoxin-producing potentials were similar among lineages, more aflatoxin G1 than B1 were produced by A. parasiticus in lineage 2 (Table 5.3). Alterations in the ratio of G1 over B1 might be a useful indicator of crop infection by this lineage.
Growth rates of A. parasiticus clades
Growth, sclerotia production and sporulation are important indicators of niche adaptation by Aspergillus spp. All lineages grew equally well at 30oC and 35oC (Table
# # 158# # 5.4), suggesting all may grow, proliferate, and potentially contribute to aflatoxin contamination of crops in both agroecologies I and II where these temperatures are common (Bunyolo et al. 1995). Lineages 2 and 3 grow faster than lineages 1 and 4 at
20OC, suggesting that at low temperatures seen in in agroecology III (Bunyolo et al.
1995) the former (2 and 3) are likely to be the most important etiologic agents of aflatoxin contamination. The use of biological control to reduce aflatoxin contamination in agroecology III will require selection and use of atoxigenic A. flavus genotypes adapted to growth at low temperatures. Such a strategy will ensure competitive exclusion of the fungi from lineages 2 and 3 in agroecology III.
# # 159# # Figure 5.1 Phylogenetic relationships of putative A. parasiticus from Zambia
# # 160# #
Maximum likelihood tree for 48 Aspergillus section Flavi from Zambia, Uganda and
America. Tree is based on concatenated sequences of the aflatoxin biosynthesis transcription factor gene (aflR, 1.8 kb) nitrate reductase gene (niaD, 2.2 kb) and calmodulin gene (cmd, 1.9 kb). Values above nodes indicate bootstrap values from 1000 replicates derived from maximum likelihood analyses (before comma) and posterior probabilities from Bayesian analyses (after comma). All clades, even those left unlabeled, have support values ≥ 80 % for maximum likelihood and ≥ 95 % posterior probabilities.
Asterisks indicate previously described fungi. Group assignments: A. flavus group
(isolates from Zambia and USA); Group 2 = A. parasiticus group (isolates from Zambia,
Uganda and Benin); Groups 1, 3 and 4, A. parasiticus groups from Zambia. The tree is rooted with A. nomius (NRRL 13137) as the outgroup. Isolates and group assignment are listed in Table 5.1.
# # 161# # Figure 5.2. Relationships between A. parasiticus from Zambia and close relatives
Maximum likelihood tree for representative isolates from groups 1-4 from figure 5.1 with known fungi. Tree is based on concatenated sequences of sequences of the aflatoxin biosynthesis transcription factor gene (aflR, 1.8 kb) nitrate reductase gene (niaD, 2.2 kb) and calmodulin gene (cmd, 1.3 kb). Values above nodes indicate bootstrap values from
1000 replicates derived from maximum likelihood analyses (before comma) and posterior probabilities from Bayesian analyses (after comma). All clades have support values ≥ 80
% for maximum likelihood and ≥ 95 % posterior probabilities. Taxa with binomial names refer to previously characterized species and included in the analysis to serve as reference points. The tree is rooted with A. nomius (NRRL 13137) as the outgroup. All isolates and group assignment are listed in Table 5.1.
# # 162$ $ Table 5.1 Aspergillus section Flavi isolates used in the current study
Isolate Other names Species Group Location Substrate Citation Aflatoxin NRRL M93, ATCC Kurtzman et al nomius Outgroup Illinois, USA Wheat BG 13137 15546 1987 AF13 ATCC 96044 flavus A. flavus Arizona, USA Citrus soil Cotty 1989 B Cotton field AF12 ATCC MYA-382 flavus A. flavus Arizona, USA Cotty 1989 B soil Cotton field AF42 ATCC MYA-383 flavus A. flavus Arizona, USA Cotty 1989 B soil Cotton field AF70 ATCC MYA-384 flavus A. flavus Arizona, USA Cotty 1989 B soil Kumeda et al NRRL 2999 ATCC 26691 parasiticus 2 Uganda Peanut BG 2003 Mixed crop BN009E parasiticus 2 Benin Pitt et al 1983 BG soil SU1 ATCC 56775 parasiticus 2 Uganda Peanut 56 BG Sugarcane Garber and Ray4-C parasiticus Texas, USA BG field soil Cotty 2014 Pildain et al A-11611 minisclerotigenes Nigeria Soil BG 2003 A-11612 Strain S Nigeria Soil BG BG 13MSA flavus A. flavus Kabwe, Zambia Maize Current study B
S76MSA flavus A. flavus Choma, Zambia Maize Current study B
S92MSA parasiticus 1 Kitwe, Zambia Maize Current study BG
Maize field 40MFA parasiticus 1 Mkushi, Zambia Current study BG soil Livingstone, S84MSH parasiticus 1 Maize Current study BG Zambia 105MSA parasiticus 2 Chipata, Zambia Maize Current study BG
S5V parasiticus 2 Lusaka, Zambia Maize Current study BG
S121WDA parasiticus 3 Chibombo, Forest soil Current study BG
$ $ 163$ $ Zambia Chibombo, S121WDE parasiticus 3 Forest soil Current study BG Zambia Maize field 8MFA parasiticus 3 Kabwe, Zambia Current study BG soil Maize field 79MFS parasiticus 3 Petauke, Zambia Current study BG soil Chibombo, Groundnut 3GFV parasiticus 3 Current study BG Zambia field soil Groundnut 13GFL parasiticus 3 Kabwe, Zambia Current study BG field soil S24L parasiticus 3 Lusaka, Zambia Maize Current study BG
Maize field 13MFN parasiticus 3 Kbwe, Zambia Current study BG soil Maize field 13MFA parasiticus 3 Kabwe, Zambia Current study BG soil S28I parasiticus 3 Lusaka, Zambia Maize Current study BG
Chibombo, Groundnut 3GFN parasiticus 3 Current study BG Zambia field soil Groundnut 105GFA parasiticus Chipata, Zambia Current study BG field soil Groundnut 105GFB parasiticus Chipata, Zambia Current study BG field soil Maize field 105MFA parasiticus Chipata, Zambia Current study BG soil S27C parasiticus Lusaka, Zambia Maize Current study BG
Chibombo, S121WDC parasiticus Forest soil Current study BG Zambia Chibombo, S121WDD parasiticus Forest soil Current study BG Zambia S4G parasiticus Lusaka, Zambia Maize Current study BG
Kapiri Mposhi, 25MSX parasiticus Maize Current study BG Zambia
$ $ 164$ $ Livingstone, S61MSC parasiticus 4 Maize Current study BG Zambia Livingstone, S68MSC parasiticus 4 Maize Current study BG Zambia S70MSG parasiticus 4 Monze, Zambia Maize Current study BG
S21T parasiticus 4 Lusaka, Zambia Maize Current study BG
S1C parasiticus 4 Lusaka, Zambia Maize Current study BG
Groundnut 40GFV parasiticus 4 Mkushi, Zambia Current study BG field soil Maize field 79MFC parasiticus 4 Petauke, Zambia Current study BG soil Maize field 70MSF parasiticus 4 Nyimba, Zambia Current study BG soil 134MSO parasiticus 4 Vumbwi, Zambia Maize Current study BG
S29H parasiticus 4 Lusaka, Zambia Maize Current study BG
Kapiri Mposhi, 30MSV parasiticus 4 Maize Current study BG Zambia Groundnut 105GFT parasiticus 4 Chipata, Zambia Current study BG field soil S24G parasiticus 4 Lusaka, Zambia Maize Current study BG
$
$ $ 165$ $ Table 5.2 Primers and annealing temperature for PCR amplification
Product size Forward Reverse T (°C) a (bp) aflR1F AGAGAGCCAACTGTCGGACCAA aflR1R GGGTGACCAGAGAACTGCGTGAT 57 600 aflR2F GACTTCCGGCGCATAACACGT A aflR2R ACGGTGGCGGGACTGTTGCTACA 57 600 aflR4F CGCCCATGACGCACTACGTT aflR4R TGGTGGTTGATTCGATTGAGG 57 600 niaDF CGGACGATAAGCAACAACAC niaDAR GGATGAACACCCGTTAATCTG 58 733 niaD1F ACTCCAGATAGCCATGTTCC niaD1R GGTCCAGGGCCCAGTTCAAT 58 720 niaD3F GTCACTACGGCACATCTA niaD3R CATCCATCCTGTAGGCAT 58 747 cmd2F GGCTGGATGTGTGTAAATC cmd2R ATTGGTCGCATTTGAAGGG 48 630 cmd3F GTTAGTGGTTAGTCGCAG cmd3R CTTCAGCTCTCTGGAATC 52 621 cmd42 GGCCTTCTCCCTATTCGTAA cmd637 CTCGCGGATCATCTCATC 56 649
$ $ 166# # Table 5.3 Aflatoxigenicity of A. parasiticus groups
Group Total aflatoxin (µg/kg) B1 G1 G1/B1 1 86,944A 41,330A 42,864A 1.037 2 142,176A 23,113A 61,976A 2.681 3 102,026A 47,219A 50,249A 1.064 4 90,179A 40,729A 48,145A 1.182 Means followed by the same letter in each column (A/B/C) or row (X/Y/Z) are not significantly different (P<0.05) by Tukey-Kramer’s HSD test. Means were log- transformed prior to analysis.
# # 167# # Table 5.4 Growth rates of A. parasiticus groups
Growth rate (mm/hr) Clade 20OC 25OC 30OC 35OC 1 0.0164A(Z) 0.0192B(Y) 0.0316A(X) 0.0314A(X) 2 0.0141A(Z) 0.0241A(Y) 0.0307A(X) 0.0304AB(X) 3 0.0172A(Z) 0.0240A(Y) 0.0304A(X) 0.0293B(X) 4 0.0159A(Z) 0.0201B(Y) 0.0306A(X) 0.0284B(X) Means followed by the same letter in each column (A/B/C) or row (X/Y/Z) are not significantly different (P<0.05) by Tukey-Kramer’s HSD test. Means were log- transformed prior to analyses.
# # 168# # Table 5.5 Frequencies in individual fields and forests of Aspergillus parasiticus-like fungi belonging to the four predominant clades
Frequencies of clades in each field and the wild Location (%) # of isolates Clade 1 Clade 2 Clade 3 Clade 4 Chipata farm A 0 19 37 44 16 Mpongwe farm A 0 0 69 31 13 Wild/forests 12 0 44 44 16 Clades are defined in Figure 5.2.
# # 169# # WORKS CITED
Adebayo-Tayo., Bukola, C., Onilude, A.A. and Ukpe, G.P. (2008) Mycofloral of smoke- dried fishes sold in Uyo, Eastern Nigeria. World J Agri Sci 4, 346-350. Agrios, G.N. (2005) Plant Pathology (5th Ed). Academic Press, San Diego, CA, pp 177- 202. Akinnifesi, F.K., Kwesiga, F.R., Mhango, J., Mkonda, A., Chilanga, T. and Swai, R. (2004) Domesticating priority miombo indigenous fruit trees as a promising livelihood option for small-holder farmers in Southern Africa. Acta Hortic 632, 15-30. Akinnifesi, F.K., Kwesiga, F., Mhango, J., Chilanga, T., Mkonda, A., Kadu, C.A.C., Kadzere, I., Mithofer, D., Saka, J.D.K. and Sileshi, G. (2006) Towards the development of miombo fruit trees as commercial tree crops in southern Africa. For, Trees and Livelihoods 16, 103-121. Akinnifesi, F.K., Sileshi, G., Ajayi, O.C., Chirwa, P.W., Kwesiga, F.R. and Harawa, R. (2008) Contributions of agroforestry research and development to livelihood of smallholder farmers in Southern Africa: 2. Fruit, medicinal, fuelwood and fodder tree systems. Agric J 3, 76-88. Atehnkeng, J., Ojiambo, P.S., Cotty P.J. and Bandyopadhyay, R. (2014) Field efficacy of a mixture of atoxigenic Aspergillus flavus Link:Fr vegetative compatibility groups in preventing aflatoxin contamination in maize (Zea mays L.). Biol Control 72, 62-70. Atehnkeng, J., Donner, M., Ojiambo, P.S., Ikotun, B., Augusto, J., Cotty, P.J. and Bandyopadhyay, J. (2016) Environmental distribution and genetic diversity of vegetative compatibility groups determine biocontrol strategies to mitigate aflatoxin contamination of maize by Aspergillus flavus. Microb Biotechnol 9, 75- 88. Bandyopadhyay, R., Ortega-Beltran, A., Akande, A., Mutegi, C., Atehnkeng, J., Kaptoge, L., Senghor, A.L., Adhikari, B.N. and Cotty, P.J. (2016) Biological control of aflatoxins in Africa: current status and potential challenges in the face of climate change. World Mycotoxin J 9, 771-789. Benhura, C., Kugara, J., Muchuweti, M., Nyagura, S.F., Matarise, F., Gombiro, P.E. and Nyandoro, G. (2015) Drying kinetics of syrup of Parinari curatellifolia fruit and cereal based product, zvambwa. J Food Sci Technol 52, 4965-4974. Boyd, M.L. and Cotty, P.J. (2001) Aspergillus flavus and aflatoxin contamination of leguminous trees of the Sonoran Desert in Arizona. Phytopathology 91, 913-919. Buchmann, C., Prehsler, S., Hartl, A. and Vogl, C.R. (2010) The importance of baobab (Adansonia digitata L.) in rural West African subsistence—suggestion of a cautionary approach to international market export of baobab fruits. Ecol Food Nutr 49, 145-172. Buchwald-Werner, S. and Bischoff, F. (2011) Natural Products–Market Development and Potentials. In Industrial Scale Natural Products Extraction ed. Bart H.J. and Pilz, S. pp 247-267. Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim, Germany. Bumbangi, N.F., Muma, J.B., Choongo, K., Mukanga, M., Velu, M.R., Veldman, F., Hatloy, A. and Mapatano, M.A. (2016) Occurrence and factors associated with
# # 170# # aflatoxin contamination of raw peanuts from Lusaka district's markets, Zambia. Food Control 68, 291-296. Bunyolo, A., Chirwa, B. and Muchinda, M. (1995) Agro-ecological and climatic conditions. In Zambia Seed Technology Handbook ed. Muliokela, S.W. pp 19-23. Ministry of Agriculture Food and Fisheries. Lusaka. Zambia. Callicott, K.A. and Cotty, P.J. (2015) Method for monitoring deletions in the aflatoxin biosynthesis gene cluster of Aspergillus flavus with multiplex PCR. Lett Appl Microbiol 60, 60-65. Carver, M., Inkerman, P.A. and Ashbolt N.J (1987) Anagyrus saccharicola Timberlake (Hymenoptera: Encyrtidae) and other biota associated with Saccharicoccus sacchari (Cockerell)(Homoptera: Pseudococcidae) in Australia. Austral Entomology 26, 367-368. Chadare, F.J., Linnemann, A.R., Hounhouigan, J.D., Nout, M.J.R. and Van Boekel, M.A.J.S. (2008) Baobab food products: a review on their composition and nutritional value. Critic Rev Food Sci Nutr 49, 254-274. Chang, P., Matsushima, K., Takahashi, T., Yu, J., Abe, K., Bhatnagar, D., Yuan, G., Koyama, Y. and Cleveland, T.E. (2007) Understanding nonaflatoxigenicity of Aspergillus sojae: a windfall of aflatoxin biosynthesis research. Appl Microbiol Biotechnol 76, 977-984. Christensen, M. (1981) A synoptic key and evaluation of species in the Aspergillus flavus group. Mycologia 73, 1056-1084. Cotty, P.J. (1994) Comparison of four media for the isolation of Aspergillus flavus group fungi. Mycopathologia, 125, 157-62. Cotty, P.J. (1997) Aflatoxin-producing potential of communities of Aspergillus section Flavi from cotton-producing areas in the United States. Mycol Res 101, 698-704. Cotty, P.J. and Cardwell, K.F. (1999) Divergence of West African and North American Communities of Aspergillus Section Flavi. Appl Environ Microbiol 65, 2264- 2266. Cotty, P.J., Probst, C. and Jaime-Garcia, R. (2008) Etiology and management of aflatoxin contamination. In Mycotoxins: Detection Methods, Management, Public Health, and Agricultural Trade ed. Leslie, J. F., Bandyopadhyay, R. and Visconti, A., pp 287-299. Oxfordshire, UK: CAB International. Cotty, P.J. (1988) Aflatoxin and sclerotial production by Aspergillus flavus: influence of pH. Phytopathology 78, 1250-1253. Cotty, P.J. (1989) Virulence and cultural characteristics of two Aspergillus flavus strains pathogenic on cotton. Phytopathology 79, 808-814. Cotty, P.J., Bayman, D.S., Egel, D.S. and Elias, K.S. (1994) Agriculture, aflatoxins and Aspergillus. In The genus Aspergillus: From taxonomy and genetics to Industrial applications ed. Powell, K.A., Renwick, A. and Peberdy, J.F. pp 285. New York, NY: Plenum Press. Cotty, P.J. and Bayman, P. (1993) Competitive exclusion of a toxigenic strain of Aspergillus flavus by an atoxigenic strain. Phytopathology 83, 1283-1287. Cotty, P.J., and Jaime-Garcia, R. (2007) Influences of climate on aflatoxin-producing fungi and aflatoxin contamination. Int J Food Microbiol 119, 109. Cunningham, T. (1996) Saturniid Subsidy: Cash and Protein from Edible Caterpillars of Zambesian Woodlands. The Miombo in transition: Woodland and Welfare in
# # 171# # Africa. Centre for International Forestry Research, Bangor, Indonesia, pp 107- 108. DeFoliart, G.R. (1989) The human use of insects as food and as animal feed. Bulletin of the Entomological Society of America 35, 22-36. Donner, M., Atehnkeng, J., Sikora, R.A., Bandyopadhyay, R., Cotty, P.J. (2009) Distribution of Aspergillus section Flavi in soils of maize fields in three agroecological zones of Nigeria. Soil Biol Biochem 41, 37-44. Donner, M., Lichtemberg, P.S.F., Doster, M., Picot, A., Cotty, P.J., Puckett, R.D., and Michailides, T.J. (2015) Community structure of Aspergillus flavus and A. parasiticus in major almond-producing areas of California, United States. Plant Dis 99, 1161-1169. Doster, M.A., and Michailides, T.J. (1995) The relationship between date of hull splitting and decay of pistachio nuts by Aspergillus species. Plant Dis 79, 766-769. Drummond, J., and Pinnock, D.E. (1990) Aflatoxin production by entomopathogenic isolates of Aspergillus parasiticus and Aspergillus flavus. J invertebr pathol 55, 332-336. FAO (2014) http://faostat3.fao.org/home/E. Folmer, O., Black, M., Hoeh, W., Lutz, R., and Vrijenhoek, R. (1994) DNA primers for amplification of mitochondrial cytochrome c oxidase subunit I from diverse metazoan invertebrates. Mol Mar Biol Biotechnol 3, 294-299. Garber, N.P. and Cotty, P.J. (2014) Aspergillus parasiticus communities associated with sugarcane in the Rio Grande Valley of Texas: implications of global transport and host association within Aspergillus section Flavi. Phytopathology 104, 462-471. Gonçalves, S.S., Stchigel, A.M., Cano, J.F., Godoy-Martinez, P.C., Colombo, A.L. and Guarro, J. (2012) Aspergillus novoparasiticus: a new clinical species of the section Flavi. Med Mycol 50, 152-160. Gong, Y., Hounsa, A., Egal, E., Turner, P.C., Sutcliffe, A.E., Hall, A.J., Cardwell, K. and Wild, C.P. (2004) Postweaning exposure to aflatoxin results in impaired child growth: a longitudinal study in Benin, West Africa. Environ Health Perspect 112, 1334-1338. Guindon, S., Dufayard, J., Lefort, V., Anisimova, M., Hordijk, W. and Gascuel, O. (2010) New algorithms and methods to estimate maximum-likelihood phylogenies: assessing the performance of PhyML 3.0. Syst Biol 59, 307-321. Hadavi, E. (2005) Several physical properties of aflatoxin-contaminated pistachio nuts: Application of BGY fluorescence for separation of aflatoxin-contaminated nuts. Food Addit Contam 22, 1144-1153. Hajibabaei, M., Janzen, D.H., Burns, J.M., Hallwachs, W. and Paul DN Hebert. 2006. DNA barcodes distinguish species of tropical Lepidoptera. Proceedings of the National Academy of Sciences of the United States of America 103, 968-971. Hebert, P.D.N., Cywinska, A. and Ball, S.L. 2003. Biological identifications through DNA barcodes. Proceedings of the Royal Society of London B: Biological Sciences 270,313-321. Horn, B.W. (2005) Colonization of wounded peanut seeds by soil fungi: selectivity for species from Aspergillus section Flavi. Mycologia 97, 202-217.
# # 172# # Horn, B.W., Dorner, J.W. (1998) Soil populations of Aspergillus species from section Flavi along a transect through peanut-growing regions of the United States. Mycologia 90, 767-776. Horn, B.W., Greene, R.L., Sobolev, V.S., Dorner, J.W., Powell, J.H., Layton, R.C. (1996) Association of morphology and mycotoxin production with vegetative compatibility groups in Aspergillus flavus, A. parasiticus, and A. tamarii. Mycologia 88, 574-587. Horn, B.W. and Dorner, J.W. (1998) Soil populations of Aspergillus species from section Flavi along a transect through peanut-growing regions of the United States. Mycologia 90, 767-776. Horn, B.W., Ramirez-Prado, J.H. and Carbone, I. (2009) Sexual reproduction and recombination in the aflatoxin-producing fungus Aspergillus parasiticus. Fungal Genetics and Biolo 46, 169-175. Huelsenbeck, J.P. and Ronquist, F. (2001) MRBAYES: Bayesian inference of phylogenetic trees. Bioinformatics 17, 754-755. Jaime!Garcia, R. and Cotty, P.J. (2006) Spatial distribution of Aspergillus flavus and its toxigenic strains on commercial cottonseed from south Texas and its relationship to aflatoxin contamination. Plant Pathol 55, 358-366. Jaime, R., McKamey, J. and Cotty, P.J. (2013) Module storage time, leaf grade and seed moisture influence fiber quality and aflatoxin contamination of cotton in South Texas. J Cotton Sci 17, 60-68. Jonsyn, F.E. and Lahai, G.P. (1992) Mycotoxic flora and mycotoxins in smoke dried fish from Sierra Leone. Mol Nutr Food Res 36, 485-489. Juliani, H.R., Koroch, A.R., Simon, J.E. and Wamulwange, C. (2007) Mungongo cold pressed oil (Schinziophyton rautanenii): A new natural product with potential cosmetic applications. International Symposium on Medicinal and Nutraceutical Plants 756, 407-412. Kachapulula, P.W., Akello, J., Bandyopadhyay, R. and Cotty, P.J. (2017) Aflatoxin contamination of groundnut and maize in Zambia: Observed and Potential Concentrations. J Appl Microbiol 6, 1471-1482 Kalaba, F.K., Chirwa, P.W. and Prozesky, H. (2009) The contribution of indigenous fruit trees in sustaining rural livelihoods and conservation of natural resources. J Hort Forestry 1, 1-6. Kankolongo, M.A., Hell, K. and Nawa, I.N. (2009) Assessment for fungal, mycotoxin and insect spoilage in maize stored for human consumption in Zambia. J Sci Food Agric 89, 1366-1375. Kannaiyan, J., Sandhu, R.S. and Phiri, A.L. (1987) Aflatoxin and Aspergillus flavus Contamination Problems of Groundnuts in Zambia. In Aflatoxin Contamination of Groundnut: proceedings of the International Crops Research Institute for Semi- Arid Tropics (ICRISAT). pp 65-70. Klich, M.A. and Pitt, J.I. (1988) Differentiation of Aspergillus flavus from A. parasiticus and other closely related species. Trans Br Mycol Soc 91, 99-108. Kornerup, A. and Wanscher, J.H. (1978) Methuen handbook of colour (3rd ed). Methuen, London, England. Kumeda, Y., Asao, T., Takahashi, H. and Ichinoe, M. (2003) High prevalence of B and G aflatoxin-producing fungi in sugarcane field soil in Japan: Heteroduplex panel
# # 173# # analysis identifies a new genotype within Aspergillus section Flavi and Aspergillus nomius. FEMS Microbiol Ecol 45, 229-238. Kurtzman, C.P., Horn, B.W. and Hesseltine, C.W. (1987) Aspergillus nomius, a new aflatoxin-producing species related to Aspergillus flavus and Aspergillus tamarii. Antonie van Leeuwenhoek 53, 147-158. Lewis, L., Onsongo, M., Njapau, H., Schurz-Rogers, H., Luber, G., Kieszak, S., Nyamongo, J., Backer, L., Dahiye, A.M. and Misore, A. (2005) Aflatoxin contamination of commercial maize products during an outbreak of acute aflatoxicosis in eastern and central Kenya. Environ Health Perspect 113, 1763- 1767. Liu, Y., Chang, C.C., Marsh, G.M. and Wu, F. (2012) Population attributable risk of aflatoxin-related liver cancer: systematic review and meta-analysis. Eur J Cancer 48, 2125-2136. Luengthanaphol, S., Mongkholkhajornsilp, D., Douglas, S., Douglas, P.L., Pengsopa, L. and Pongamphai, S. (2004) Extraction of antioxidants from sweet Thai tamarind seed coat––preliminary experiments. J Food Eng 63, 247-252. Marasas, W.F.O. (1997) Risk assessment of fumonisins produced by Fusarium moniliforme in corn. Cereal Res Commun 25, 399-406. Maroyi, A. (2011) An ethnobotanical survey of medicinal plants used by the people in Nhema communal area, Zimbabwe. J Ethnopharmacol 136, 347-354. Maruza, IM, L Musemwa, S Mapurazi, P Matsika, VT Munyati, and S Ndhleve. 2017. 'Future prospects of Ziziphus mauritiana in alleviating household food insecurity and illnesses in arid and semi-arid areas: A review', World Development Perspectives, 5, 1-6. Matumba, L., Sulyok, M., Njoroge, S.M.C., Ediage, E.J., Van Poucke, C., De Saeger, S. and Krska, R. (2015) Uncommon occurrence ratios of aflatoxin B1, B2, G1, and G2 in maize and groundnuts from Malawi. Mycotoxin research 31, 57-62. Mbata, K.J. (1995) Traditional use of arthropods in Zambia. Food Insects Newsletters 8, 5-7. Mbata, K.J. (1999) Traditional uses of arthropods in Zambia: II. Medicinal and miscellaneous uses. The Food Insects Newsletter 12, 1-7. Mbata, K.J., Chidumayo, E.N. and Lwatula, C.M. (2002) Traditional regulation of edible caterpillar exploitation in the Kopa area of Mpika district in northern Zambia. J Insect Conservation 6, 115-130. Mbata, K.J. and Chidumayo, E.N. (2003) Traditional values of caterpillars (Insecta: Lepidoptera) among the Bisa people of Zambia. Int J Tropic Insect Scie 23, 341- 354. Mehl, H.L., Jaime, R., Callicott, K.A., Probst, C., Garber, N.P., Ortega!Beltran, A., Grubisha, L.C., Cotty, P.J. (2012) Aspergillus flavus diversity on crops and in the environment can be exploited to reduce aflatoxin exposure and improve health. Ann. N. Y. Acad. Sci. 1273, 7-17. Mehl, H.L. and Cotty, J.P. (2013a) Nutrient environments influence competition among Aspergillus flavus genotypes. Appl Environ Microbiol 79, 1473-1480. Mehl, H.L. and Cotty, P.J. (2013b) Influence of plant host species on intraspecific competition during infection by Aspergillus flavus. Plant Pathology 62, 1310- 1318.
# # 174# # Mukanga, M., Derera, J., Tongoona, P. and Laing, M.D. (2010) A survey of pre-harvest ear rot diseases of maize and associated mycotoxins in south and central Zambia. Int J Food Microbiol 141, 213-221. Mpuchane, S., Taligoola, H.K. and Gashe, B.A. (1996) Fungi associated with Imbrasia belina, an edible caterpillar. Botswana Notes and Records 28, 193-197. Njana, M.A., Kajembe, G.C. and Malimbwi, R.E. (2013) Are miombo woodlands vital to livelihoods of rural households? Evidence from Urumwa and surrounding communities, Tabora, Tanzania. For Trees Livelihoods 22, 124-140. Odvody, G.N., Spencer, N. and Remmers, J. (1997) A description of silk cut, a stress- related loss of kernel integrity in preharvest maize. Plant Dis 81, 439-444. Pildain, M.B., Frisvad, J.C., Vaamonde, G., Cabral, D., Varga, J., Samson, R.A. (2008) Two novel aflatoxin-producing Aspergillus species from Argentinean peanuts. Int J Syst Evol Microbiol 58, 725-35. Pitt, J.I., Hocking, A.D and Glenn, D.R. (1983) An improved medium for the detection of Aspergillus flavus and A. parasiticus. J Appl Bacteriol 54, 109-114. Probst, C., Njapau, H. and Cotty, P.J. (2007) Outbreak of an Acute Aflatoxicosis in Kenya in 2004: Identification of the Causal Agent. Appl Environ Microbiol 73, 2762-2764. Probst, C., Schulthess, F. and Cotty, P.J. (2010) Impact of Aspergillus section Flavi community structure on the development of lethal levels of aflatoxins in Kenyan maize (Zea mays). J Appl Microbiol 108, 600-610. Purcell, S.L., Phillips, D.J., Mackey, B.E. (1980) Distribution of Aspergillus flavus and other fungi in several almond-growing areas of California. Phytopathology 70, 926-929. Rahul, J., Jain, M.K., Singh, S.P., Kamal, R.K., Naz, A., Gupta, A.K. and Mrityunjay, S.K. (2015) Adansonia digitata L.(baobab): a review of traditional information and taxonomic description. Asian Pac J Trop Biomed 5, 79-84. Reddy, B.N. and Raghavender, C.R. (2007) Outbreaks of aflatoxicoses in India. Afr J Food Agric Nutri Dev 7, 5. Rodrigues, P., Santos, C., Venâncio, A. and Lima, N. (2011) Species identification of Aspergillus section Flavi isolates from Portuguese almonds using phenotypic, including MALDI TOF ICMS, and molecular approaches. J Appl Microbiol 111, 877-892. Ronquist, F. and Huelsenbeck, J.P. (2003) MrBayes 3: Bayesian phylogenetic inference under mixed models. Bioinformatics 19, 1572-1574. Rubinoff, D., Jose, M.S. and Peigler, R.S. (2017) Multi gene phylogeny of the Hemileuca maia complex (Saturniidae) across North America suggests complex phylogeography and rapid ecological diversification. Systematic Entomology doi:10.1111/syen.12235. Sambrook, J., Fritsch, E.F. and Maniatis, T. (1989) Molecular cloning: a laboratory manual (Cold spring harbor laboratory press). Schroeder, H.W., Boller, R.A. (1973) Aflatoxin production of species and strains of the Aspergillus flavus group isolated from field crops. Appl. Microbiol. 25, 885-889. Seye, F., Bawin, T., Boukraa, S., Zimmer, J., Ndiaye, M., Delvigne, F. and Francis, F. (2014) Effect of entomopathogenic Aspergillus strains against the pea aphid,
# # 175# # Acyrthosiphon pisum (Hemiptera: Aphididae). Applied entomology and zoology 49, 453-458. Silow, C.A. (1976) Edible and other insects of mid-western Zambia: studies in ethno- entomology Vol 2. Almqvist & Wiksell. Sitko, N.J., Chapoto, A., Kabwe, S., Tembo, S., Hichaambwa, M., Lubinda, R., Chiwawa, H., Mataa, M., Heck, S. and Nthani, D. (2011) Technical compendium: Descriptive agricultural statistics and analysis for Zambia in support of the USAID mission’s Feed the Future Strategic Review (No. 104016). Michigan State University, Department of Agricultural, Food, and Resource Economics. Siulapwa, N., Mwambungu, A., Lungu, E. and Sichilima, W. (2014) Nutritional value of four common edible insects in Zambia. Int J Scie Res 3, 876-884. Sudjaroen, Y., Haubner, R., Würtele, G., Hull, W.E., Erben, G., Spiegelhalder, B., Changbumrung, S., Bartsch, H. and Owen, R.W. (2005) Isolation and structure elucidation of phenolic antioxidants from Tamarind (Tamarindus indica L.) seeds and pericarp. Food Chem Toxicol 43, 1673-1682. Sweany, R.R., Damann Jr, K.E. and Kaller, D.M. (2011) Comparison of soil and corn kernel Aspergillus flavus populations: evidence for niche specialization. Phytopathology 101, 952-959. Tembo, S. and Sitko, N. (2013) Technical Compendium: Descriptive Agricultural Statistics and Analysis for Zambia. Michigan State University, Department of Agricultural, Food, and Resource Economics. Turner, P.C., Moore, S.E., Hall, A.J., Prentice, A.M. and Wild C.P. (2003) Modification of immune function through exposure to dietary aflatoxin in Gambian children. Environ Health Perspect 111, 217-220. van Egmond, H.P., Schothorst, R.C. and Jonker, M.A. (2007) Regulations relating to mycotoxins in food: perspectives in a global and European context. Anal Bioanal Chem 389,147-157. Van Huis, A. (2013) Potential of insects as food and feed in assuring food security. Annu Rev Entomol 58, 563-583. Van Huis, A., Van Itterbeeck, J., Klunder, H., Mertens, E., Halloran, A., Muir, G. and Vantomme, P. (2013) Edible insects: future prospects for food and feed security (no. 171). Food and agriculture organization of the United nations (FAO). Varga, J, Frisvad, J.C. and Samson, R.A. (2011) Two new aflatoxin producing species, and an overview of Aspergillus section Flavi. Stud Mycol 69, 57-80. Vermaak, I., Kamatou, G.P.P., Komane-Mofokeng, B., Viljoen, A.M. and Beckett, K. (2011) African seed oils of commercial importance—Cosmetic applications. S Afr J Bot 77, 920-933. Wicklow, D.T. (1983) Taxonomic features and ecological significance of sclerotia. South. Coop. Ser. Bull 279, 6-12. Williams, J.H., Phillips, T.D., Jolly, P.E., Stiles, J.K., Jolly, C.M. and Aggarwal, D. (2004) Human aflatoxicosis in developing countries: a review of toxicology, exposure, potential health consequences, and interventions. Am j Clin Nutr 80, 1106-1122. Wilson, J.J., Rougerie, R., Schonfeld, J., Janzen, D.H., Hallwachs, W., Hajibabaei, M., Kitching, I.J., Haxaire, J. and Hebert, P.D.N. (2011) When species matches are
# # 176# # unavailable are DNA barcodes correctly assigned to higher taxa? An assessment using sphingid moths. BMC Ecol 11, 18. Wotton, H.R. and Strange, R.N. (1987) Increased susceptibility and reduced phytoalexin accumulation in drought-stressed peanut kernels challenged with Aspergillus flavus. Appl Environ Microbiol 53, 270-273. Wu, F. (2014) Global impacts of aflatoxin in maize: trade and human health. World Mycotoxin J 8, 137-142. Zimba, N., Wren, S. and Stucki, A. (2005) Three major tree nut oils of southern central Africa: their uses and future as commercial base oils. Int J Aromather 15, 177- 182.
#
# #