Etiology of Aflatoxin Contamination of Dried Red Chilies and Diversity of Toxic Aspergilli in the United States

Etiology of Aflatoxin Contamination of Dried Red Chilies and Diversity of Toxic Aspergilli in The United States

Item Type text; Electronic Dissertation

Authors Singh, Pummi

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, presentation (such as public display or performance) of protected items is prohibited except with permission of the author.

Download date 30/09/2021 07:49:54

Link to Item http://hdl.handle.net/10150/632579 ETIOLOGY OF AFLATOXIN CONTAMINATION OF DRIED RED CHILIES AND DIVERSITY OF TOXIC ASPERGILLI IN THE UNITED STATES

Pummi Singh

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

2019

ACKNOWLEDGEMENTS

I am indebted to my PhD. advisor, Dr. Marc J. Orbach, a great teacher and philosopher, for his guidance, support and kindness. Thank you for believing in me and encouraging me to never give up. Thank you for being the amazing person you are. I will be always grateful to you.

My deepest heartfelt gratitude to Dr. Peter J. Cotty, my guide and mentor. Thank you for your wisdom, patience and for believing in me. Thank you for teaching me to be passionate about life and science.

My sincere thanks to all of my committee members: Dr. Zhongguo Xiong, Dr.

for the immense support and encouragement throughout my graduate career. I appreciate the time you dedicated to me. Thank you Dr. Hillary Mehl for always being there for me.

To the USDA-ARS Aflatoxin Lab which supported all of my graduate work. I am extremely thankful to Dr. Kenneth Callicott, Dr. Kenneth Shenge, Mrs. Connie

Graham and Mrs. Marina Wissotski for all the guidance, encouragement and help on a daily basis for innumerable things, also for several discussions and laughs.

To my labmates and friends, Dr. Paul Kachapulula, Lourena Arone Maxwell and

Connell Chinganda, you guys are amazing. Thank you for always being very supportive and making me laugh on some of those very stressful days.

I want to thank Eve Beauchemin, my sincere and dedicated undergrad who contributed to majority of my dissertation work. You are awesome Eve.

Thanks to all the friends because of whom I could survive graduate school, for all the laughs and support, Priya, Lipsa, Sundaresh, Sriram, Reena, Pradipta, Radha,

Neha and Bhushan.

No words can express my love and gratitude for my parents. Thank you for supporting me throughout and believing in me, for instilling the value of hard work and sincerity in me. I hope to live up to your expectations always. To my sister, who makes me feel special by looking up to me and for her immense faith in me.

The one person who is as passionate about my career is my better half, Abhishek

Pandey. We have come a long way and he has been a constant support. Thank you for your love, for bearing with my craziness and for always letting me be myself.

To the almighty, for bestowing in me the perseverance and dedication to achieve my goals. Har Har Mahadev.

DEDICATION

To mom, papa, Swikriti, Simran and Kishmish. You all make my life happy. To my beloved Abhishek, I love you and thank God for you.

Table of Contents

ABSTRACT ...... 8

INTRODUCTION ...... 11

APPENDIX A- Aflatoxin contamination of dried red chilies: Contrasts between the United States and Nigeria, two markets differing in regulation enforcement. 25 Abstract ...... 25 Introduction ...... 27 Materials and Methods ...... 29 Results ...... 34 Discussion ...... 36

APPENDIX B- Characterization of Aspergilli from dried red chilies (Capsicum spp.): Insights into the etiology of aflatoxin contamination ...... 47 Abstract ...... 47 Introduction ...... 49 Materials and Methods ...... 51 Results ...... 58 Discussion ...... 61

APPENDIX C- Phenotypic differentiation of two morphologically similar aflatoxin producers from West Africa ...... 77 Abstract ...... 77 Introduction ...... 79 Materials and Methods ...... 83 Results and Discussion ...... 87 Conclusion ...... 93

APPENDIX D - Aspergillus texensis: A novel aflatoxin producer with S morphology from the United States ...... 102 Abstract ...... 102

Introduction ...... 103 Materials and Methods ...... 106 Results ...... 112 Discussion ...... 116

APPENDIX E- Diversity among S morphology fungi in Aspergillus section Flavi from the United States ...... 136 Abstract ...... 136 Introduction ...... 138 Materials and Methods ...... 141 Results ...... 148 Discussion ...... 154

Literature Cited ...... 181

ABSTRACT

Aflatoxins are hepatocarcinogenic mycotoxins produced by fungi within

Aspergillus section Flavi. Aflatoxin-producing fungi frequently contaminate crops including maize, cottonseed, groundnuts, tree nuts, and spices. Contamination of food and feed with aflatoxins can have severe health consequences, including the death of humans and animals. Crops contaminated with aflatoxins may be excluded from markets, and, if in excess of regulatory limits, destroyed; this creates an immense economic burden for growers of crops susceptible to aflatoxin contamination.

The research presented in this dissertation focuses on two areas related to aflatoxin contamination of crops. The first one focusses on the prevalence of aflatoxin contamination in market-purchased chilies and etiology of contamination.

Dried red chili from the genus Capsicum is one of the most extensively consumed spices worldwide to enhance the flavor, color and aroma of different foods. Most chilies are produced in warm regions of the globe that where the climate favors growth and proliferation of filamentous fungi that produce aflatoxins. Appendix A presents a survey of aflatoxin contamination levels in chilies from markets in the

United States (US) and Nigeria, two markets differing in regulation enforcement.

Unacceptable concentrations of aflatoxins were detected in two percent of samples from US markets and seven percent of Nigerian chilies. Members of Aspergillus section Flavi were recovered from 40% of chilies from the US, although 64% of the samples were contaminated with aflatoxins, indicating sterilization of dried red chilies prior to sale in the US. Both average aflatoxin concentrations and fungal

8 quantities were significantly higher in Nigerian chilies compared to those purchased in the US. Incubation of un-inoculated chilies under conditions conducive for aflatoxin production led to a build-up of lethal concentrations of aflatoxins in Nigerian chilies, indicating higher risk associated with post-purchase practices of consumers in Nigeria compared to that in the US. Appendix B presents the identification of aflatoxin-producing Aspergilli recovered from chilies from markets in the US and Nigeria. A precise understanding of the etiology of aflatoxin contamination of crops is crucial for the development of mitigation strategies.

Investigation of Aspergillus communities associated with chilies and the etiology of aflatoxin contamination conducted using aflatoxin-based liquid culture and crop assays as well as multi-locus phylogenetics showed the occurrence of three distinct aflatoxin-producing species in US market chilies. Of these, A. aflatoxiformans is a non-native pathogen and was imported with chilies originating from Nigeria. Five distinct aflatoxin-producers, including a novel lineage, were discovered in Nigerian chilies. Aspergillus aflatoxiformans was identified as the most important etiologic agent of chili contamination in Nigeria based on its overall incidence and aflatoxin producing ability in chilies. small sclerotium (< 400 µm) morphology, and produces high concentrations of B and G aflatoxins in chilies. Several A. flavus L morphotype isolates were recovered from chilies that did not produce detectable concentrations of aflatoxins in vitro and can be a potential resource for the development of biological control agents for preventing aflatoxin contamination in chilies. In Appendix C, an in vitro assay has been developed for distinguishing isolates of A. aflatoxiformans and A.

9 minisclerotigenes, aflatoxin-producing species which co-occur in West Africa and contaminate major crops like maize, groundnuts, and chilies with unacceptable concentrations of aflatoxins. These fungi otherwise have largely overlapping morphologies and aflatoxin profiles and are indistinguishable in the absence of molecular tools.

The second part of this dissertation focuses on the genetic diversity of fungi with S morphology within Aspergillus section Flavi due to their highly aflatoxigenic potential. Fungal species with S morphology have been implicated as causal agents of crop contamination in both US and Africa. Appendix D describes a novel aflatoxin-producing species, Aspergillus texensis, from the US. Appendix E presents the characterization of S morphology fungi resident in US soils and crops using phylogenetics, aflatoxin profiles, and simple sequence repeat markers. Four phylogenetically distinct groups were identified, each of which produced dangerous concentrations of aflatoxins in maize and should be considered when devising management practices for aflatoxin mitigation.

INTRODUCTION

Aflatoxins (AFs), toxic secondary metabolites produced by species within

Aspergillus section Flavi, are a common contaminant in the global food supply.

The major aflatoxin producer, Aspergillus flavus Link, was described two centuries ago (Link, 1809). However, the toxicity of aflatoxins was only recognized in 1960, following the death of more than 100,000 turkeys in England because of consumption of contaminated peanut meal (Blount, 1961), following which, aflatoxins have received global attention to minimize negative impacts on health and economy. Aflatoxins contaminate a wide range of crops including cereals, cottonseed, groundnuts, tree nuts and spices (Allcroft and Carnaghan, 1962; Lee et al., 1989; Payne and Widstrom, 1992; Reddy et al., 2001; Schade et al., 1975).

Aflatoxins are categorized into four main types, aflatoxin B1, B2, G1 and G2, due to their blue (B toxins) or green fluorescence (G toxins) under ultraviolet light and differences in mobility during thin layer chromatography (Van der Zijden et al.,

1962; Hartley et al., 1963). Aflatoxins B1 and B2 can be biotransformed to aflatoxins

M1 and M2, which are secreted in the milk of lactating mammals leading to exposure of young offsprings to these metabolites (De longh et al., 1964).

Risk of aflatoxin exposure is the highest in developing regions of the world.

While the main route for exposure of humans and animals to aflatoxins is through ingestion of contaminated food, exposure can occur in utero (Wild et al., 1991), and by breastfeeding (Magoha et al., 2014). Transplacental exposure in utero is associated with lower birth weights and stunting in children. A study conducted in rural Kenya reported that mothers whose blood tested positive for aflatoxins gave

11 birth to babies with significantly lower birth weight compared to those with no detectable aflatoxins in the blood (De Vries et al., 1989). Sadeghi et al. (2009) reported an inverse correlation between AFM1 levels in breast milk samples from

Iranian mothers and length of infants at birth. Studies in West Africa investigating the relationship between aflatoxin exposure and growth in children demonstrated that aflatoxin-albumin adducts levels were higher in stunted children (height-for- age z- -2) compared to that with non-stunted children (Gong et al., 2002,

2003, 2004). Chronic consumption of sublethal concentrations of aflatoxins is associated with hepatocellular carcinomas (Liu and Wu, 2010). Liu et al. (2012) estimated about 172,000 liver cancer cases globally, attributable to aflatoxin exposure. Aflatoxin B1 (AFB1) is the only mycotoxin listed as a human carcinogen by the International Agency for Research on Cancer (IARC, 2002). Carcinogenicity of aflatoxin B1 is primarily attributed to its microsomal oxidation to the highly reactive AFB1-8,9-epoxide by the cytochrome P450 enzymes (Eaton and

Gallagher, 1994). Mutagenicity of the exo-epoxide is higher than that of the endo- epoxide due to its faster reactivity with DNA (Brown et al., 2009; Iyer et al., 1994).

The AFB1 exo-8,9-epoxide intercalates between DNA residues and reacts covalently with DNA to form guanine adducts at the N7 position, which are precursors to the genetic effects of AFB1 (Bailey et al., 1996). Formation of these adducts results in G to T transversions in codon 249 of the p53 tumor suppressor gene resulting in carcinomas (Smela et al., 2001).

Frequent consumption of food contaminated with high levels of aflatoxins may result in acute aflatoxicosis characterized by liver necrosis, bile duct

12 proliferation, edema, and rapid death (Williams et al., 2004). Episodes of acute aflatoxicosis have been reported from southeast Asia and Africa. More than 100 human deaths were recorded in parts of Western India due to an outbreak of hepatitis, which was traced to consumption of maize contaminated with aflatoxins; as much as 2 to 6 mg of aflatoxins were consumed (Krishnamachari et al., 1975).

Aflatoxins were detected during postmortem analysis of thirteen children in

Malaysia who reportedly died due to consumption of a Chinese noodle (Lye et al.,

1995). Human fatalities due to aflatoxicosis outbreaks have been repeatedly reported from Kenya since 1981 (Ngindu et al., 1982). The most severe outbreak occurred in 2004 and was caused by consumption of homegrown maize claiming

125 human lives out of a total of 317 reported cases of illness due to aflatoxicosis

(CDC, 2004). The most recent aflatoxicosis episodes have been reported from

Tanzania which recorded 20 and 4 deaths in 2016 and 2017, respectively (Africa

A, 2016; Outbreak News Today, 2017).

Aflatoxin concentrations are strictly regulated in food and feed in developed countries. The United States (US) regulates total aflatoxins in food intended for human consumption at 20 µg/kg and in milk at 0.5 µg/kg (USFDA, 2000). Aflatoxins are regulated at much lower concentrations in the European Union (0.05 µg/kg in milk, 4 µg/kg in maize and groundnuts, and 10 µg/kg in spices) (European

Commission, 2006). Although regulation enforcement in the US is largely able to minimize exposure to aflatoxins in the US food supply, severe liver damage and death of pets have been recorded in the country due to consumption of contaminated pet food (Stenske et al., 2006). Of several mycotoxins of concern,

13 aflatoxins cause the greatest economic losses in the US due to their toxicity and regulatory standards (Robens and Cardwell, 2003). Economic impacts due to aflatoxin contamination of food and feed in the US can include several aspects, such as control and monitoring costs, testing, insurance premiums, trading, and losses in the export market, to list a few (Mitchell et al., 2016, Robens and

Cardwell, 2003; Wu and Guclu, 2012). Mitchell et al. (2016) estimated losses of more than 50 million US dollars annually just in the maize industry in the US.

Contaminated crops/food are either destroyed or detoxified with ammoniation rendering it unsuitable for human consumption and thus of lower value (Park et al.,

1988). Maize lots exceeding the action levels for human food or feed are either rejected at grain elevators or are purchased at discounted prices, reducing the overall value of the crop (Mitchell et al., 2016). In addition to maize, groundnut, cottonseed and tree nuts (pistachio and almonds) industries suffer losses due to regulatory standards (Robens and Cardwell, 2003).

Currently, multiple aflatoxin-producing species are known with A. flavus recognized as the major aflatoxin-producer (Klich, 2007). Aspergillus flavus is a filamentous fungus in the order Eurotiales that is cosmopolitan and proliferates in the environment as a saprotroph (Klich, 2002) and as an opportunistic pathogen of plants and animals (Hedayati et al., 2007; St Leger et al., 2007). The saprotrophic phase of A. flavus primarily occurs in the soil where it colonizes organic debris and is present as mycelia or sclerotia (melanized survival structures) (Ashworth et al., 1969). Under conducive environmental conditions

(e.g. high temperatures), mycelia and sclerotia give rise to conidia that act as

14 airborne dispersal structures (Wicklow and Donahue, 1984), and infect susceptible plants, resulting in aflatoxin contamination. Conidia produced on leaves and plant surfaces serves as the secondary inoculum causing multiple infection cycles in a single growing season (Diener et al., 1987). Aflatoxin contamination of crops is a complex process starting in the field during crop development; physiological stress and insect damage can exacerbate infection, proliferation of the pathogen and aflatoxin production (Cotty et al., 2008; Russel et al., 1976). Crop contamination can often be detected by the presence of bright green-yellow florescence in infected crop components due to the production of kojic acid by aflatoxin-producers

(Zeringue et al., 1999). Crops can also be contaminated after maturation and during storage (Cotty et al., 2008). Warm temperatures (> 28°C) and high humidity provide conducive conditions for new infections and further contamination in already infected crops, until it is ultimately consumed (Bock and Cotty, 1999; Cotty et al., 2008; Kachapulula et al., 2017a).

Aspergillus flavus is delineated into the L and S morphotypes. Isolates with

L morphotype produce numerous conidia and few large (> 400 µm) sclerotia whereas those with the S morphotype produce abundant quantities of smaller (<

400 µm) sclerotia but sparse conidia (Cotty, 1989). The aflatoxin-producing ability of L morphotype isolates on susceptible crops is highly variable. Some isolates are atoxigenic (produce no aflatoxins), while others produce low to very high levels of aflatoxins in crops (Probst and Cotty, 2012). The atoxigenic L isolates of A. flavus have received considerable attention owing to their ability to act as biological control agents that reduce aflatoxin contamination in crops by outcompeting

15 aflatoxigenic isolates (Mehl and Cotty, 2010). In contrast, S morphotype isolates consistently produce high concentrations of aflatoxins and no atoxigenic S morphotype isolates have been reported till date (Mehl et al., 2012). Morphological differences between the two morphotypes of A. flavus may reflect differential niche adaptation. Production of large quantities of sclerotia and aflatoxins by the S isolates may confer fitness benefits, such as long-term survival and resistance to fungivory by insects in the soil (Drott et al., 2017; Mehl et al., 2012). In contrast, production of larger numbers of conidia by the L isolates may allow for dispersal to the phyllosphere. Comparative genomics of the L and S morphotype isolates of

A. flavus indicated morphotype unique proteins in the S isolates with putative antimicrobial activity, carbon and nitrogen metabolism, heavy metal tolerance and iron acquisition functions along with retention of intact secondary metabolite clusters, which can be advantageous in a competitive soil environment (Ohkura et al., 2018). Genomes of the L morphotype isolates are enriched in amino acid transporters and genes for nitrogen acquisition suggesting adaptive features for the nutrient-limited phyllosphere environment (Ohkura et al., 2018).

Aflatoxins are produced by enzymes and regulatory proteins encoded by more than 25 contiguous genes in a 70 kb gene cluster (Yu et al., 2005).

Aflatoxigenic Aspergilli produce either B aflatoxins (B1 and B2) or both B and G (G1 and G2) aflatoxins. The inability to produce G aflatoxins by certain species such as

A. flavus is due to deletions in portions of the cypA gene of the aflatoxin biosynthesis cluster that encodes for a cytochrome P450 monooxygenase (Ehrlich et al., 2004). Some A. flavus isolates are unable to produce aflatoxins.

Atoxigenicity has been shown to be due to either single nucleotide polymorphisms that prevent production of key enzymes of the aflatoxin biosynthesis pathway, deletions in biosynthetic genes or deletion of the entire aflatoxin gene cluster

(Adhikari et al., 2016).

Aflatoxin-producing fungi are prevalent in warm regions of the globe, especially between latitudes 35N and 35S (Cotty et al., 1994). Data from polyphasic descriptions of Aspergillus spp. indicate that section Flavi includes 34 taxa (Frisvad et al., 2019; Singh et al., 2018). Of these, A. flavus, A. parasiticus and species with an S morphology, including A. aflatoxiformans, A. minisclerotigenes and an unnamed lineage reported from Kenya (referred to as the Lethal Aflatoxicosis Fungus or LAF) have been repeatedly recovered from crops and are of concern due to their high aflatoxin-producing abilities (Cotty,

1996; Cardwell and Cotty, 2002; Horn et al., 1995; Kachapulula et al., 2017b;

Probst et al., 2012; Probst et al., 2014). The L morphotype of A. flavus is known from across the globe (Bayman and Cotty, 1991; Ehrlich et al., 2007; Mauro et al.,

2013), while distribution of isolates with S morphology, including the A. flavus S morphotype, A. aflatoxiformans, A. minisclerotigenes and LAF appears to be geographically restricted (Probst et al., 2014). For example, A. minisclerotigenes has been recovered from crops in South America, Australia, Europe and Africa, however, it has not been found in Asia and only one isolate has been recovered in the US till date (Pildain et al., 2008; Probst et al., 2014; Soares et al., 2012). The

A. flavus S morphotype has only been reported in the US and southeast Asia

(Cotty, 1989; Probst et al., 2012). Within Africa, A. aflatoxiformans has been

17 reported mainly from West Africa while LAF is known from Eastern and Southern

Africa (Cardwell and Cotty, 2002; Donner et al., 2009; Frisvad et al., 2019; Probst et al., 2014). The reason for these geographic restrictions is not known. However, a precise understanding of the etiology of aflatoxin contamination of crops and identification of causal agent(s) in affected regions is important for developing approaches for aflatoxin mitigation (Mehl et al., 2012). The etiology of crop contamination is often complicated by mixed infections and the occurrence of distinct fungal communities that can vary in virulence, competitive ability and aflatoxin production (Cotty et al., 2008; Mehl and Cotty, 2010). Therefore, incidence, and aflatoxin producing potential of individual species recovered from crops and agricultural soils are important factors for identification of significant causal agents of crops contamination.

Current management strategies for preventing aflatoxin contamination of crops include cultural practices, development of resistant cultivars, monitoring and destroying contaminated crops, and biological control using atoxigenic isolates of

A. flavus (Bandyopadhyay et al., 2016; Cotty et al., 2008; Klich, 2007). If management procedures fail to reduce aflatoxin levels to regulated concentrations in crops, other options may be utilized although they reduce the value of the crop.

These include feeding the contaminated crop to less sensitive animal species or chemical detoxification (Cotty et al., 2008; Klich, 2007). Feed containing less than

300 µg/kg aflatoxins can be fed to mature beef cattle and swine (USFDA, 2000).

Ammoniation of contaminated crops converts aflatoxins to less toxic products

18 following which crops can be fed to animals but remain unsuitable for human consumption (Cotty et al., 2008).

Management using cultural practices are targeted at lowering plant stress by reducing insect damage, growing regionally adapted cultivars, providing adequate fertilization and proper irrigation, and harvesting the crops soon after when maturation to prevent post-maturation build-up of aflatoxins in the field

(Cleveland et al., 1990; Klich, 2007). Unfortunately, these practices are not always effective and unacceptable contamination can still occur. Although transgenic Bt cotton resulted in reduced aflatoxin levels in the crop by reducing infection and insect damage by pink boll worm (Pectinophora gossypiella), infection through other modes of entry and subsequent contamination have rendered aflatoxin control in Bt cotton inconsistent (Cotty et al., 1997; Klich, 1997). Mature seeds of

Bt cultivars remain susceptible to fungal infection and aflatoxin contamination under high humidity and temperature conditions (Cotty et al., 1997). Aflatoxin mitigation by selection of resistant crop varieties has also been largely unsuccessful. Certain resistant varieties of maize and cotton have been identified but commercial cultivars totally resistant to aflatoxin contamination are still lacking

(Cleveland et al., 1990; Klich, 2007). Breeding efforts towards enhancement of host plant resistance were mostly unsuccessful due to failure of these resistant crops to provide protection against the diverse aflatoxigenic genotypes (Cleveland et al., 1990). Furthermore, host resistance was identified as multigenic and complex (Cleveland et al., 1990), and may not be effective in controlling post- harvest contamination of crops.

Biological control of aflatoxins using atoxigenic L morphotype isolates of A. flavus is the most promising method for aflatoxin management currently. Two atoxigenic A. flavus genotypes, AF36 and Afla-Guard®, are currently registered with the US Environmental Protection Agency (US-EPA) for commercial application (Dorner, 2004). Biological control products based on local atoxigenic

Aspergillus genotypes are also registered for use in Nigeria (Aflasafe), Kenya

(Aflasafe KE01) and Senegal/The Gambia (Aflasafe SN01) (Bandyopadhyay et al.,

2016). Aspergillus flavus AF36 was the first biocontrol product registered with US-

EPA for application on cottonseed, maize, and pistachio; Afla-Guard® is applied on maize and groundnut (Bandyopadhyay et al., 2016; Dorner, 2004). Application of biocontrol products displaces aflatoxigenic fungi and increases frequencies of the atoxigenic population in the hosts preventing accumulation of aflatoxins in treated crops (Cotty et al., 2006). Atoxigenic A. flavus isolates are naturally associated with crops and compete for crop substrates (Dorner, 2004). Selection of atoxigenic genotypes, native to the target area is favored because they are adapted to the target agroecosystem (Bandyopadhyay et al., 2016; Mehl et al.,

2012). The biocontrol formulation consists of a suitable carrier, such as wheat, sorghum or barley, which provides nutrients to the fungus (Cotty and Melon, 2006) when it is applied in the field. Atoxigenic isolate(s), which are active ingredients of the biocontrol product, produce conidia on the carrier grain that then disperse onto the target crop (Bandyopadhyay et al., 2016). Applications are made before resident aflatoxin-producers begin to increase, allowing for effective displacement by atoxigenic isolates (Bandyopadhyay et al., 2016; Mehl and Cotty, 2011). Once

20 established, atoxigenic genotypes increase in frequencies, creating their own founder populations and reshaping fungal community composition within the treated area, leading to an overall decrease in aflatoxin contamination of crops

(Bandyopadhyay et al., 2016). Applications of atoxigenic genotypes, and thus their presence in harvested crops does not affect the overall quantity of A. flavus on the crop or in the environment (Atehnkeng et al., 2014; Cotty, 1994b; Cotty et al., 2008;

Dorner, 2004). In addition to this, biocontrol products are effective in reducing aflatoxin accumulation post-harvest as atoxigenic genotypes move with the treated crop until consumed (Atehnkeng et al., 2014; Bandyopadhyay et al., 2016).

Aflatoxin contamination of chilies is a severe problem and red chilies available in markets often contain unacceptable concentrations of aflatoxins

(Golge et al., 2013; Iqbal et al., 2010; Reddy et al., 2001). Dried red chilies (Family:

Solanaceae; Genus: Capsicum) are primarily cultivated for their flavor and aroma and are one of the most consumed spices across the globe (Yogendrarajah et al.,

2014). Chili (dry) production exceeded 4 million tons in 2017 with India, Thailand,

China and Ethiopia being the top producers; major chili producing areas are in the zone where favorable conditions exist for infection by aflatoxigenic fungi

(FAOSTAT, 2017). In addition to the known health effects due to aflatoxin exposure, consumption of aflatoxin-contaminated chilies has been associated with the risk of gall bladder cancer in Chile, Bolivia and Peru (Asai et al., 2014; Nogueira et al., 2015). The European Union routinely rejects aflatoxin-contaminated chili shipments from various exporting nations rendering contaminated chilies an economic burden (RASFF, 2015). In this dissertation, appendices A, B and C focus

21 on aflatoxin contamination of dried red chilies, and the etiologic agents responsible for contamination.

The work presented in appendices D and E assesses the phylogenetic affiliations of highly aflatoxigenic S morphology fungi from US soils and maize.

Aspergilli within section Flavi with an S morphology are notorious for contaminating crops with dangerous concentrations of aflatoxins. Certain S morphology lineages have been associated with deadly aflatoxicosis events (Probst et al., 2012).

Knowledge of the genetic diversity of S morphology fungi in the US and their relationship with previously described aflatoxin producers remains limited.

Dissertation format

The work presented in Appendix A sought to understand the prevalence of aflatoxins in dried red chilies purchased from markets in the US compared to those from Nigeria, and to investigate the potential for post-harvest or post-purchase contamination of chilies. This work contrasted for the first time, aflatoxin concentrations and fungal quantities (Aspergillus section Flavi) in chilies originating from two markets that differ in their regulatory enforcement. Aflatoxin regulations are strictly enforced in the US whereas regulations remain poorly enforced in most developing nations, including Nigeria, leading to human exposure. The potential of contamination under post-harvest conditions such as drying, transportation or storage or post-purchase packaging and storage is reported.

The work presented in Appendix B characterized Aspergillus section Flavi fungi associated with chilies purchased from US and Nigerian markets. Aflatoxin-

22 producing species were identified and characterized using morphology, aflatoxin- producing ability, and multigene phylogenetics. Aflatoxigenicity of species recovered from chilies along with the incidence of individual species was used to determine the etiologic agents of aflatoxin contamination of chilies. Five distinct aflatoxin-producing species, including A. flavus, A. parasiticus, A. aflatoxiformans,

A. minisclerotigenes, and a novel section Flavi lineage were recovered from

Nigerian chilies. Aspergillus aflatoxiformans was identified as the primary causal agent of aflatoxin contamination in Nigerian chilies. More than 50% of chilies purchased from US markets were imported from various chili producing nations, rendering the etiology complex. Chilies purchased from US markets were infected by three aflatoxigenic Aspergilli, including the non-native pathogen, A. aflatoxiformans, indicating import of exotic pathogens into the US with the chilies.

Knowledge of fungal community composition in chilies and their aflatoxin- producing potentials can be utilized towards the development of mitigation strategies.

The work in Appendix C sought to examine the aflatoxin-producing ability of A. aflatoxiformans and A. minisclerotigenes in liquid culture fermentations with the aim to differentiate isolates of each species. Both species have an S morphology on various crops and growth media and overlapping secondary metabolite profiles. Since fungi can differ in incidence and aflatoxin-producing abilities in different crops, identification of the most important etiologic agent is crucial to determine the extent to which crops may be contaminated and for devising management strategies. Based on differences in aflatoxin production by

A. aflatoxiformans and A. minisclerotigenes in three different liquid culture fermentations, a microbiological assay is proposed for rapid and reliable identification of individuals of these two species.

Identification and polyphasic description of a novel taxon, A. texensis, is presented in Appendix D. Aspergillus texensis was recovered from maize and soils cropped to maize in the US and produces high concentrations of both B and G aflatoxins in maize. The work in Appendix E identified the occurrence of four distinct groups of S morphology isolates from the US using a multigene phylogenetic analysis, with each group capable of contaminating maize with high concentrations of aflatoxins. The occurrence of distinct S morphology fungi in the

US and implications on management are discussed.

APPENDIX A- Aflatoxin contamination of dried red chilies: Contrasts between the United States and Nigeria, two markets differing in regulation enforcement

Published in: FOOD CONTROL (2017), Vol. 80, p. 374-379

Abstract

Dried red chi to fork, chilies go through cropping, harvest, drying, processing and storage.

Chilies are susceptible to infection by aflatoxin producing fungi and subsequent contamination by aflatoxins at every stage. Aflatoxins are highly regulated, hepatotoxic carcinogens produced by fungi in Aspergillus section Flavi. The current study examined prevalence of aflatoxin B1 (AFB1) in chilies from markets across the United States (US) and Nigeria, and determined predisposition of chilies to aflatoxins post-harvest. Aflatoxin B1 was detected in 64% chilies from US markets (n = 169), and 93% of Nigerian chilies (n = 55) with a commercial lateral flow assay (Limit of Detection = 2 µg/kg). Two percent of US samples exceeded the aflatoxin regulatory limit of 20 µg/kg, while the highest concentration detected

Aspergillus spp. could be recovered only from 40% of samples from the US, and aflatoxin levels did not correlate with quantities of Aspergillus section Flavi (Colony Forming Units g-1), suggesting fungi associated with chilies in US markets were killed during processing. Both average AFB1 concentrations and fungal quantities were significantly higher (p < 0.01) in Nigerian chilies. The most contaminated sample contained 156 µg/kg AFB1. Aflatoxin concentrations in

Nigerian chilies increased as an exponential function of the quantities of

Aspergillus section Flavi (r 2 = 0.76). Results indicate that high rates of chili consumption may be associated with unacceptable aflatoxin exposure.

Introduction

Dried red chili (Capsicum spp.), a member of the nightshade family of

Solanacea, is used to enhance flavor, taste and aroma of foods. Chilies are native to the new world where they were domesticated about 6000 years ago (Perry et al., 2007). The economically notable species of Capsicum are C. annuum, C. baccatum, C. chinense, C. frutescens, and C. pubescens (Perry et al., 2007). Dried red chili is consumed across the globe in whole, crushed and ground forms. It stands second among spices in global consumption after black pepper (Piper nigrum) (Yogendrarajah et al., 2014). Over 80% of red chili is produced in tropical or sub-tropical regions (FAOSTAT, 2017). During the past decade, India, China

s dried chilies (FAOSTAT, 2017).

Fungal infection and subsequent mycotoxin contamination of chilies are affected by environmental conditions, with high temperatures and humidity favoring infection (El Mahgubi et al., 2013; Iamanaka et al., 2007; Iqbal et al.,

2010). Red chilies available in commercial markets are frequently contaminated with unacceptable concentrations of aflatoxins (Bircan, 2005; Paterson, 2007;

Reddy et al., 2001). This can be attributed to pre and/or post-harvest colonization of chilies by aflatoxin producing Aspergillus species. Inadequate conditions during drying, followed by poorly sheltered transport and storage can exacerbate contamination levels (Duman, 2010; Jalili and Jinap, 2012).

Aflatoxins in food and feed are of global concern because aflatoxins are highly toxic fungal metabolites that cause human cancer, immune suppression, and stunting (Khlangwiset et al., 2011; Liu and Wu, 2010; Mehl et al., 2012;

Schatzmayr and Streit, 2013; van Egmond and Jonker, 2004; Williams et al.,

2004). Several Aspergilli belonging to section Flavi cause aflatoxin contamination of a wide range of crops including maize, groundnuts, tree nuts, cottonseed and spices (Doster and Michailides, 1994; Jaime and Cotty, 2003; Probst et al., 2014;

Tansakul et al., 2013). Aflatoxin B1 is the only mycotoxin classified as group 1 human carcinogen by the International Agency for Research on Cancer

(International Agency for Research on Cancer, 2002). Aflatoxins are stringently regulated in the developed world, resulting in huge economic losses. Aflatoxin contamination of food and feed can result in annual losses of more than $500 million in the US alone (Robens and Cardwell, 2003; Vardon et al., 2003). In developing countries, aflatoxin is both an economic threat and a health concern.

Many developing countries do not enforce aflatoxin regulations in crops, resulting in exposure of humans to chronic and acute health risks (Shephard, 2003; Williams et al., 2004). Food and feed contaminated with aflatoxin concentrations above legislated limits can face border rejections, and loss of both markets, and product value. The European Union (EU) regulates AFB1 at 5 µg/kg and total aflatoxins at

10 µg/kg in spices (European Spice Association, 2004). Chili shipments to the EU from several nations are rejected every year due to aflatoxin contamination

(RASFF, 2015).

Most dry chilies in retail markets across the US are imported (FAOSTAT,

2017). Over 100,000 tonnes of dried red chili are imported annually into the US.

Aflatoxins are regulated at 20 µg/kg in the US in foods for human consumption.

This regulation can have a severe impact on the value of crops intended for US

28 markets. Nigeria accounts for about 50% of chili production on the continent of

Africa (FAOSTAT, 2017; Mohammed et al., 2015). Chilies are an integral part of

Nigerian cuisine. Although, Nigeria regulates AFB1 at 20 µg/kg (FAO, 2004), these regulations are less effectively enforced, leading to chronic exposure (Omojokun,

2013; Williams et al., 2004). The current study sought to: (i) contrast prevalence of aflatoxin contamination in dried red chili from markets in US with those from markets in Nigeria, (ii) determine the relationship between quantities of Aspergillus section Flavi and aflatoxin concentrations, and (iii) test potential for post-harvest contamination of market-purchased chilies.

Materials and Methods

Dried red chili samples

Dried red chili was purchased from retail markets in the United States (n = 169) and Nigeria (n = 55). Samples from US were collected during 2014-15 from retail markets in California (n = 68), Minnesota (n = 3), New York (n = 34), and Arizona

(n = 64), and consisted of whole (n = 60), ground (n = 78) and crushed (n = 12) chili, and paprika (n = 19). Fifty-eight percent of chili samples from US were labelled as imported from various countries (Table 1.1). All samples from Nigeria consisted of whole red chilies (n = 55), which were purchased from rural, small- scale markets in Kaduna (n = 50) and Lagos (n = 5) states during 2015-16. Chili samples averaged 200 g and ranged from 70 to 300 g. Nigerian samples were transferred to zippered plastic bags immediately after purchase, kept under ambient conditions, and shipped within a week of purchase. These samples were imported to the United States Department of Agriculture (USDA), Agricultural

Research Service (ARS) laboratory, at the University of Arizona, Tucson, under permits issued by the USDA Animal and Plant Health Inspection Service (APHIS).

Samples were dried in a forced air oven (40°C) to below 8% water content and sealed in plastic bags to prevent fungal growth after receipt. Whole and crushed chili samples were finely ground in a laboratory mill (Retsch Grindomix GM200,

Newtown, PA) for 30 s at 10,000 rpm for fungal isolation and aflatoxin B1 quantification. Chili and paprika, that were purchased ground, were analyzed for both fungi and aflatoxin B1 with no further processing.

Aflatoxin extraction and quantification

Aflatoxin B1 in chilies was detected and quantified using an immunochromatographic assay (Reveal Q+ for Aflatoxin testing, Neogen

Corporation, Lansing, MI) approved by the USDA, Grain Inspection, Packers and

Stockyards Administration (GIPSA) (Anonymous, 2015). Each ground chili sample was thoroughly mixed, and a 25 to 50 g sub-sample, depending on the quantity available, was analyzed for AFB1

Briefly, AFB1 was extracted with 65% ethanol by blending ground chili with either

125 ml or 250 ml of the solvent (for 25 g and 50 g of ground chili, respectively).

The slurry was shaken on a rotary shaker (HS501digital, IKA LABORTECHNIK,

Germany) at full speed for 3 min and allowed to settle for an additional 3 min. The supernatant was filtered through Whatman No. 1 filter paper and AFB1 was immediately quantified using the immunoassay and the AccuScan Pro reader.

Analytical performance- Spike and recovery

Aflatoxin quantification was validated by spike and recovery experiments.

Ground red chili (5 g) with no detectable aflatoxin was spiked to either 50 or 100

µg/kg of aflatoxin B1 (AFB1 in methanol, Supelco, Bellefonte, PA). Aflatoxin was extracted and quantified as described above. Spike and recovery was performed in four replicates. Recovery rates were estimated using the following equation:

%Recovery = (Aflatoxin B1 concentration measured in spiked sample/ Spiked concentration) X 100

Precision of the analytical method was expressed as relative standard deviation

(RSD) of replicated results.

Quantification of Aspergillus section Flavi

Members of Aspergillus section Flavi were recovered and quantified through dilution plate technique on modified Rose Bengal agar (Cotty, 1994). Briefly, up to

10 g of ground chili was suspended in 50 ml sterile de-ionized water containing

0.01% Tween-80 by stirring for 20 minutes. A 200 µl aliquot of the resulting suspension from each sample was plated in triplicate on modified Rose Bengal agar and incubated for 3 days at 31°C in the dark. Adjustments were made to aliquot volume and/or crop quantity to allow no more than 10 Aspergillus section

Flavi colonies per plate for accurate enumeration. After incubation, Aspergillus section Flavi colonies were microscopically identified and enumerated (Colony forming units (CFU) g-1). Fungal isolations were performed at least twice for each chili sample. Up to five discrete colonies of section Flavi per isolation were sub- cultured onto 5-2 agar (5 % V-8 juice; 2% agar; pH 6.0) and incubated at 31° C for

5-7 days in dark. Isolates were saved and stored as 3 mm agar plugs of sporulating

31 culture in a vial containing sterile distilled water (2 ml). Limit of detection of fungi belonging to Aspergillus section Flavi was 1.1 CFU g-1 of chili.

Evaluation of contamination of chilies post-harvest

Red chilies may go through several steps of post-harvest processing before reaching various markets. These include drying, irradiation or fumigation to sterilize chilies, packaging, storage and transport. In order to evaluate the effect of poor conditions during any stage on aflatoxin concentration, un-inoculated whole chilies, without further processing after purchase were incubated at 31°C and

100% relative humidity (RH) for one week. Approximately 5 g of whole chilies purchased from markets that were contaminated with less than 20 µg/kg AFB1 but contained detectable quantities of Aspergillus section Flavi, were placed on a metal sieve (No. 12, 1.70 mm opening, Newark Wire Cloth Company, Clifton, NJ) contained within a sealed disinfected tub (5 liters), and incubated for 7 days at

31°C. Incubation was terminated by adding 50 ml of 85 % acetone to the chilies and grinding to homogeneity in a laboratory grade Waring Blender (seven-speed laboratory blender, Waring Laboratory, Torrington, CT) at full speed for 30 s. The ground chili slurry was allowed to sit for an hour in the dark, and the resulting supernatant was spotted directly onto thin-layer chromatography (TLC) plates

(Silica gel 60, EMD, Darmstadt, Germany) adjacent to an aflatoxin standard

(Aflatoxin Mix Kit-M, Supelco, Bellefonte, PA) containing a mixture of known concentrations of aflatoxins B1, B2, G1 and G2. Plates were developed in ethyl ether-methanol-water (96:3:1) and air dried. Aflatoxins were visualized under 365- nm UV light. Aflatoxin B1 was quantified directly on TLC plates using a scanning

32 densitometer (TLC Scanner 3, Camag Scientific Inc., Wilmington, NC). Samples that were initially negative for AFB1 were concentrated with a previously reported modification to the method of the Association of Official Analytical Chemists (Cotty and Lee, 1989; Mckinney, 1975; Stoloff and Scott, 1984). Briefly, a 20 ml solution of 1.1M (CH3COO)2Zn and 0.04M AlCl3 along with 80 ml of deionized water were added to the negative samples. After 5 min, 5 g of diatomaceous earth was added and the resulting mixture was shaken to allow mixing. These mixtures were allowed to sit for 1-2 hours in order to minimize interfering pigments, fatty acids and trace lipids, to obtain cleaner TLC extracts (Mckinney, 1975). Extracts were filtered through number 4 Whatman paper. Aflatoxin B1 was extracted from 100 ml filtrate by partitioning the filtrate twice with 25 ml of dichloromethane. The extracts were passed through a bed of anhydrous Na2SO4 and air dried. Residues were dissolved in dichloromethane, spotted on TLC plates, developed and quantified as above. The limit of detection was 17 µg/kg.

Data analysis

Aflatoxin B1 in chili samples was measured in µg/kg. Aflatoxin B1 concentration was log-transformed and subjected to Analysis of Variance

(ANOVA) as implemented in JMP 11.1.1(SAS Institute, Cary, NC). Average aflatoxin B1 concentrations were compared u - total quantity of fungi belonging to Aspergillus section Flavi from each chili sample

-1 was calculated as CFU g . Association between AFB1 and quantity of fungi belonging to Aspergillus section Flavi was assessed with regression (SAS version

9.2; SAS Institute, Cary, NC). Paired t-tests were used to compare aflatoxin

33 concentrations before and after incubation. Data were tested for normality and, if required, log transformed before analysis. True means are presented for clarity.

Results

Occurrence of aflatoxin B1 in dried red chili

Spike and recovery experiments resulted in mean recovery of (73±6.3)% and (72±10.4)% for 50 and 100 µg/kg spike of AFB1, respectively, indicating good precision. Most dried red chilies purchased in either US or Nigerian markets had

AFB1 concentrations detectable with a lateral flow immunochromatographic assay

(Limit of detection (LOD) = 2 µg/kg). Overall, 108 (64 %, n = 169) chili samples from US markets and 51 (93 %, n = 55) from Nigeria were contaminated with AFB1.

Nigerian chilies contained significantly higher concentrations of aflatoxin B1 than chilies from US markets (13.5 versus 5.1 µg/kg, Table 1. -test, p <

0.01). Thirty-eight percent of US chilies were contaminated with > 5 µg/kg AFB1

(mean = 11.1 µg/kg), and based on EU regulatory limits, all of these chilies would be rejected by the EU. Two percent of US chilies exceeded the US regulatory limit of 20 µg/kg (mean = 45.2 µg/kg). The highest concentration was detected in an imported whole chili sample (94.9 µg/kg AFB1). Average aflatoxin B1 concentrations among the different chili categories (whole, ground, crushed chili and paprika) were not significantly different (ANOVA, p > 0.05). In contrast to US chilies, 75% of chilies from Nigerian markets were contaminated with > 5 µg/kg

AFB1 (mean = 17.3 µg/kg, Table 1.2), and 7 % samples were contaminated with levels of AFB1 over 20 µg/kg (mean = 109 µg/kg, maximum 156 µg/kg).

Quantification of Aspergillus section Flavi

Fungi belonging to Aspergillus section Flavi were detected in only 67 (40%) chili samples from US markets, although 108 (64%) samples contained detectable levels of aflatoxins (Table 1.2, Table 1.3). No Aspergilli could be recovered from

102 (60%) samples (LOD for Aspergillus section Flavi = 1.1 CFU g-1), which included 65 (38%) samples contaminated with detectable levels of aflatoxin. The

67 chili samples from which Aspergillus section Flavi could be recovered included

-1 24 samples with no detectable AFB1.Overall, CFU g for section Flavi fungi in chilies purchased in the US ranged from < 1.1 to 2.4 x 105 (Mean = 1.87 x 103 CFU g-1). Six percent of the samples contained greater than 103 CFU g-1. In contrast, chilies from Nigerian markets contained significantly greater quantities of

Aspergillus section Flavi than chilies from US markets (Table 1. -test, p < 0.01). Aspergillus section Flavi was detected in all chili samples from Nigeria with populations ranging from 3.3 CFU to 7 x 105 CFU per gram of chili (Mean =

5.68 x 104 CFU g-1). Forty-five percent of Nigerian chilies contained more than 103

CFU of Aspergillus section Flavi per gram chili. The quantity of section Flavi was directly related to aflatoxin B1 content in Nigerian chilies (Aflatoxin B1 = 7.52 x

(0.004CFU/ g) 2 exp , r = 0.76, p < 0.01), but not in US chilies (Aflatoxin B1 = 5.17+0.06

CFU g-1, r2 = 0.01, p = 0.12) (Figure 1.1).

Post purchase risk in dried red chili

Incubation experiments were performed to quantify aflatoxin risk associated with the -purchase practices. Aflatoxin B1 concentrations increased significantly after un-inoculated chilies (USA, n = 11; Nigeria, n = 10) contaminated with less than 20 µg/kg AFB1 were incubated at 31°C and 100% RH

35 for 7 days, in both US and Nigerian samples (Paired t-test, p < 0.01, Table 1.4).

However, initial concentrations of AFB1 before incubation in both US and Nigerian chilies were not different (Table 1. -test, p > 0.05). Aflatoxin B1 concentrations reached dangerous levels in Nigerian chilies at the end of the incubation period, and mean AFB1 concentrations exceeded 700 µg/kg. Overall, increases in AFB1 concentrations in Nigerian chilies (133 fold) was greater than in

-test, p < 0.01).

Discussion

Aflatoxins frequently occur in chilies offered for sale in markets (Aydin et al.,

2007; Cho et al., 2008; Jalili and Jinap, 2012; Santos et al., 2010; Shamsuddin et al., 1995). Over three quarters of chilies are produced in regions with conditions favorable for crop infection by aflatoxin-producing fungi (i.e. high temperature and humidity (FAOSTAT, 2017)). Chili moisture at harvest is 65-80%, and in many chili producing countries, the spice is sun dried to reduce water content to below 10%

(Duman, 2010; Guide, 2011; Hossain and Bala, 2007; Paterson, 2007; Topuz et al., 2009; Yogendrarajah et al., 2014). Hence, chilies are rendered susceptible to aflatoxin contamination due to both climatic conditions during growth and post- harvest processing practices.

Unacceptable levels of aflatoxin in dried red chilies have been previously reported in chilies marketed in Ethiopia (250-525 µg/kg ) (Fufa and Urga, 1996),

India (2-969 µg/kg) (Reddy et al., 2001), Spain (1.4-64.4 µg/kg) (Santos et al.,

2011), Pakistan (6.8-96.2 µg/kg) (Paterson, 2007), and Turkey (0.24-165) (Golge et al., 2013).This study is the first to compare aflatoxin concentrations in chilies

36 from strictly regulated markets (US), with markets where regulations are largely unenforced (Nigeria). Out of the 169 chilies from retail markets in four states in the

US, 98 (58%) samples were labelled as imported from eleven countries (Table

1.1). Only 9 (5%) samples were labelled as Product of USA, and all of these contained acceptable levels of AFB1. Multiple samples originating from and packaged in major chili producing countries (including India, China, Thailand, and

Pakistan) were both contaminated with more than 5 µg/kg of AFB1, and contained

Aspergillus section Flavi propagules. The fungi causing the contamination were most likely associated with the crop prior to packaging in the country of origin. For some exporting countries, there were very few samples available from the sampled

US markets (Table 1.1). As a result, analyses for those countries are not a

Aflatoxin B1 was detected in 108 (64%) samples from markets in the US, with concentrations ranging from <2 to 94.9 µg/kg. Two percent of the samples had aflatoxin contamination levels that were above US regulatory limits (20 µg/kg) indicating an unanticipated level of aflatoxin exposure risk faced by consumers of chilies in the US. Compared to US samples, chilies originating from Nigerian markets contained significantly higher concentrations of AFB1 with 93% of chilies contaminated with detectable aflatoxin concentrations (highest detected 156 µg

AFB1/kg). Both the US and Nigeria regulate aflatoxin content of foods (FAO, 2004; van Egmond and Jonker, 2004). The US has the reputation of strictly enforcing aflatoxin regulations, and management of aflatoxin content is a serious concern of

37 the US food and spice industry (Dohlman, 2003). US chilies are also routinely subjected to antimicrobial processing. However, in rural markets of Nigeria, aflatoxin regulations are unenforced and antimicrobial processing is non-existent

(Omojokun, 2013). These results are similar to comparisons made between imported chilies purchased in either Sri Lanka or Belgium, with the Belgium chilies having lower aflatoxin levels (Yogendrarajah et al., 2014). However, in that study, the chilies in both countries were imported, whereas, in the current study, Nigerian chilies were produced and marketed locally.

Quantities of Aspergillus section Flavi were significantly higher in Nigerian chilies than in chilies from US markets. Although 108 samples from US markets contained detectable levels of AFB1, fungi belonging to Aspergillus section Flavi were recovered from only 64 samples (LOD = 1.1 CFU g-1). Notably, fungi could not be recovered from the two most contaminated samples containing 94.9 µg/kg and 29.0 µg/kg AFB1. Additionally, concentrations of AFB1 in chilies from US markets did not relate to quantities of Aspergillus section Flavi fungi in chilies (r2 =

0.01, p = 0.12; Figure 1.1). These results reflect processing practices that reduce or eliminate fungal propagules, including thermal processing (drying chilies with hot air or superheated steam), irradiation, and fumigation with ethylene oxide

(European Spice Association, 2004; Panda, 2010). Strict enforcement of aflatoxin regulations in the US may encourage exporting countries to sterilize chilies before export in order to prevent aflatoxin increases after packing. Aflatoxin-producers can associate with chilies during crop development, at maturation, and during post- harvest handling. If ambient conditions are conducive, aflatoxin contamination can

38 proceed during any of these stages (Cotty et al.,1994). Occurrence of AFB1 in chilies from which no fungi could be recovered (Tables 1.2 and 1.3) suggests that these chilies were subjected to anti-microbial processing only after the chilies were already contaminated with AFB1. This suggests that the anti-microbial processing techniques, that effectively eliminated aflatoxin-producing fungi associated with the chilies, were insufficient to reduce AFB1 levels to within acceptable limits.

Nevertheless, sterilization techniques are effective at reducing vulnerability of chilies to contamination during storage, as indicated by incubation of un-inoculated chilies contaminated with < 20 µg/kg AFB1. Although a significant increase in AFB1 concentration was observed from incubation at 100% RH in chilies purchased in both nations, the average increase in Nigerian samples was more than five-fold greater than in US chilies (Table 1.4). The different rates of increase may be attributed, at least in part, to different quantities of Aspergillus section Flavi resulting from post-harvest processing (Table 1.4). All chili samples included in the incubation experiment contained detectable quantities of Aspergillus section Flavi.

It is anticipated that chilies with no viable fungi would have even lower vulnerability to aflatoxin increases after purchase. Antimicrobial processing may provide an increased safety margin for US consumers by reducing sensitivity of chili aflatoxin content to post-purchase handling and storage.

Even though Aspergillus section Flavi was detected in all Nigerian chilies

-1 (LOD 1.1 CFU g ), concentrations of AFB1 were generally low (93% samples with

< 20 µg/kg AFB1). However, AFB1 concentrations in Nigerian chilies increased with the quantities of Aspergillus section Flavi (r2 = 0.76, p < 0.01; Figure 1.1). More

39 than 90% of Nigerian chilies were from markets in rural portions of Kaduna state.

Quantities of Aspergillus section Flavi (45% samples with >103 CFU g-1) in these chilies suggest they were not subjected to post-harvest antimicrobial processing.

In Nigerian chilies, the quantity of Aspergillus section Flavi in chilies was a good

2 indicator of aflatoxin B1 content in unprocessed chilies (r = 0.76). Furthermore, chilies purchased in Nigerian markets increased in aflatoxin B1 content 66 to 243 fold when incubated at high humidity and temperature without addition of aflatoxin producers (Table 1.4). Aflatoxin B1 content increased significantly less in similarly incubated chilies from US markets (Table 1.4). These results suggest that antimicrobial post-harvest processing reduces the risk of aflatoxin exposure.

However, post-harvest processing, a cost born by the exporting country, is not sufficient to reduce aflatoxin to within tolerable limits (Figure 1.1, Tables 1.2 and

1.3). Additionally, for techniques such as fumigation by ethylene oxide and irradiation, benefits must be balanced with the risk of low consumer acceptance

(Mostafavi et al., 2010; Schweiggert et al., 2007).

Results from the current study indicate that enforcement of US aflatoxin regulations is largely successful in protecting US consumers from unacceptable aflatoxin exposure through consumption of chilies because 98% of chilies from US markets had aflatoxins concentrations below the USFDA Action Level of 20 µg/kg.

However, although, mean aflatoxin concentration in Nigerian chilies was significantly higher compared to chilies from US markets, 93% of chilies from

Nigeria contained acceptable levels of AFB1 (< 20 µg/kg). Aflatoxin testing and management by industries and regulatory agencies in the US cost millions of

40 dollars (Robens and Cardwell, 2003). From the current results, it is clear that in spite of these costly measures to achieve regulatory standards for aflatoxin in human food, consumers are exposed to aflatoxins in both the US and Nigeria through consumption of chilies. Hence, stringent enforcement of regulations to achieve regulatory standards for aflatoxin levels in chilies is overall an expensive procedure that does not necessarily achieve cost effective consumer protection.

Development and use of mandatory HACCP procedures for prevention of aflatoxins in chilies may be a more cost-effective course of action (Marín et al.,

2009; Paterson, 2007; Torres et al., 2014). Currently, atoxigenic L strains of

Aspergillus flavus are used effectively as biological control to competitively exclude toxin producing strains and reduce aflatoxin contamination in several crops

(Atehnkeng et al., 2014; Chulze et al., 2015; Mauro et al., 2015; Mehl et al., 2012).

Similar technologies in chili producing regions may be a promising, inexpensive, low hazard, and consumer acceptable method to reduce aflatoxin contamination in chilies.

Table 1.1 Aflatoxin contamination and incidence of Aspergillus section Flavi in chili imported into US markets from various countries.

% Chili samples Country of Chili samples origin (n) >2 µg/kg >5 µg/kg Positive for Aflatoxin B Aflatoxin B section Flavi 1 1 Bangladesh 4 50 50 50 China 17 59 35 65 India 49 67 43 31 Korea 4 50 25 25 Mexico 10 30 20 20 Myanmar 1 100 100 100 Nigeria 1 100 0 100 Pakistan 2 100 100 50 Thailand 7 57 43 71 Turkey 2 50 0 0 USA 9 33 22 33 Vietnam 1 100 100 100 Unknown 62 73 35 39

Table 1.2 Aflatoxin B1 concentrations in chilies purchased from US and Nigerian markets.

% Samples with aflatoxin Aflatoxin B (µg/kg) 1 B Country Chili samples 1

Range Mean µg/kg µg/kg USA 169 ND-94.9 5.1* 36 38 2 Whole = 60 ND-94.9 5.2 48 32 3 Ground = 78 ND-25.5 4.7 29 44 1 Crushed = 12 ND-10.1 2.1 50 17 0 Paprika = 19 ND-28.6 7.0 21 53 5

Nigeria 55 (Whole only) ND-156 13.5* 7 75 7 *Aflatoxin B1 concentrations differed significantly between chilies from US and

Nigerian markets -test, p < 0.01).

Different forms of chilies from US markets did not differ significantly in AFB 1 content (ANOVA, p > 0.05).

Aflatoxin B1 concentrations were log transformed before statistical analyses. Limit of detection (LOD) = 2 µg/kg. ND- Not Detected.

Table 1.3 Incidence and quantity of Aspergillus section Flavi fungi in chili samples from US and Nigerian markets.

Aspergillus section Flavi Chilies positive for Chili (CFU g-1) Country Aspergillus section samples Flavi Range Mean

USA 169 67 (40%)* ND - 2.4 x 105 1.87 x 103* Nigeria 55 55 (100%)* 3.3 - 7 x 105 5.68 x 104* *Proportions or Colony Forming Units (CFU) of Aspergillus section Flavi per gram

-test, p < 0.01. CFU data were log transformed before analysis. LOD of Aspergillus section Flavi = 1.1 CFU g-1. ND- Not Detected.

Y X < 0.05). <0.05). p Fold increase Fold .-1 26 0.9-71 ag Mean Range 623 133 66-243 -test,

* * * O M After After (µg/kg) incubation incubation 1

83.2 769 R R Before Aflatoxin B Aflatoxin

45 incubation incubation

4.6 6.3 4A 4B

ofsection -1 beforeincubation CFUg concentrations in chilies as a result of incubation at 31°C, 100% RH. at 31°C, RH. a result of100% as chilies incubation in concentrations 1 Flavi < 0.01). Data were log transformed before analyses. Datatransformed log <0.01). were before p concentrations after 7 days incubation differ significantly from concentrations prior to incubation, from prior to significantly incubation, days differ 7 concentrations after incubation concentrations 1 Increase in aflatoxin B Increase aflatoxin in rp Country Crop Whole Chili USA (n = 11) 2.2 x 10x 2.2 10x 8.5 (n = USA 11) Chili Whole (n Nigeria = 10) Chili Whole Table 1.4 Table withincolumn a Values B *Aflatoxin t-test, Paired Figure 1.1 Relationship of aflatoxin B1 concentration to quantity of Aspergillus section Flavi (CFU g-1) for Nigerian chilies (Y = 7.52 x exp(0.004X), r2 = 0.76, p <

0.001) and US chilies (Y = 5.17+0.055X, r2 = 0.01, p = 0.12).

APPENDIX B- Characterization of Aspergilli from dried red chilies (Capsicum spp.): Insights into the etiology of aflatoxin contamination

Published in: INTERNATIONAL JOURNAL OF FOOD MICROBIOLOGY (2019),

Vol. 289, p. 145-153

Abstract

Aflatoxins are toxic carcinogens produced by several species of Aspergillus section Flavi, with some aflatoxin producers associated with specific crops. Red chilies (Capsicum spp.) are grown in warm regions that also favor aflatoxin- producers. Aflatoxins in red chilies may result in serious health concerns and severe economic losses. The current study sought to gain insight on causal agents of aflatoxin contamination in red chilies. Naturally contaminated chilies from markets in Nigeria (n = 55) and the United States (US) (n = 169) were examined.

The A. flavus L strain was the predominant member of Aspergillus section Flavi

(84%) in chilies. Highly toxigenic fungi with S strain morphology were also detected in chilies from both countries (11%), followed by A. tamarii (4.6%) and A. parasiticus (0.4%). Fungi with L morphology produced significantly lower quantities of aflatoxins (mean = 43 µg g-1) compared to S morphology fungi (mean = 667 µg g-1; p < 0.01) in liquid fermentation. Eighty-one percent of S morphology fungi from chilies in US markets produced only B aflatoxins, whereas 20%, all imported from

Nigeria, produced both B and G aflatoxins; all S morphology fungi from Nigerian chilies produced both B and G aflatoxins. Multi-gene phylogenetic analyses of partial gene sequences for nitrate reductase (niaD, 2.1 kb) and the aflatoxin pathway transcription factor (aflR, 1.9 kb) resolved Aspergilli recovered from chilies

47 into five highly supported distinct clades: 1) A. parasiticus; 2) A. flavus with either

L or S morphology; 3) A. minisclerotigenes; 4) A. aflatoxiformans, and 5) a new lineage. Aspergillus aflatoxiformans and the new lineage produced the highest concentrations of total aflatoxins in chilies, whereas A. flavus L strains produced the least. The results suggest etiology of aflatoxin contamination of chili is complex and may vary with region. Knowledge of causal agents of aflatoxin contamination of chilies will be helpful in developing mitigation strategies to prevent human exposure.

Introduction

Aflatoxins are hepatocarcinogenic metabolites produced by several

Aspergilli, which frequently contaminate food and feed crops including maize, groundnut, cottonseed, spices, and tree nuts (Doster et al., 2014; Kachapulula et al., 2017). Of the four major aflatoxins, B1, B2, G1 and G2, aflatoxin B1 is the most toxic, and is carcinogenic to both humans and animals (IARC, 2002). Most developed nations stringently enforce aflatoxin regulatory limits within food and feed (e.g. US regulates aflatoxins at 20 µg/kg total aflatoxins in human food) resulting in significant economic losses to growers (Robens and Cardwell, 2003; van Egmond et al., 2007). Across regions of the globe where regulations are either lacking or are not strictly enforced, aflatoxin contamination exposes humans and animals to severe health risks. Sub-lethal concentrations are associated with stunted development (Khlangwiset et al., 2011), immune suppression (Turner et al., 2003), and liver cancer (Liu and Wu, 2010), whereas high levels can cause liver cirrhosis followed by rapid death (CDC, 2004).

Aspergillus flavus is the most frequently implicated causal agent of aflatoxin contamination of crops (Klich, 2007). The species can be divided into two major morphotypes known as the L and S strains. The S strain isolates produce copious amounts of small sclerotia (< 400 µm) and fewer conidia, whereas the L strain isolates produce sparse amounts of large sclerotia (>400 µm) but abundant conidia

(Cotty, 1989). The L strain produces variable quantities of aflatoxins, and isolates can either be atoxigenic or produce moderate to high levels of aflatoxins; however, the S strain of A. flavus and other members of section Flavi with S morphology are

49 known to consistently produce high concentrations of aflatoxins (Cotty, 1989; Cotty and Cardwell, 1999; Probst et al., 2010). Aflatoxin contamination can start before harvest when crops are infected by resident Aspergilli dispersed by various mechanisms. Crop infection is facilitated by plant stress (physical damage, by insects, drought) and warm temperatures (above 28°C) (Cotty et al., 2008).

Contamination continues after crop maturation in the presence of conducive conditions, both in the field, and post-harvest during storage, processing or transportation (Cotty et al., 2008; Marín et al., 2009). Since conidia of Aspergilli are air-borne and ubiquitous, new infections may occur during post-harvest stages, exacerbating contamination levels (Cotty et al.,1994).

Red chili (Capsicum spp.), a member of the nightshade family of

Solanaceae, is a globally consumed spice. Capsicum was domesticated in

America about 6,000 years ago and has now spread into Asia, Africa and Europe

(Perry et al., 2007). Chilies are mainly cultivated in warm regions that provide suitable conditions for crop infection by Aspergillus propagules and subsequent contamination with aflatoxins. India, China and Thailand produced most of the

-harvest practices and conditions during growth render the spice susceptible to aflatoxin contamination. Contaminated chilies can result in loss of lucrative European and

US markets where aflatoxins are stringently regulated (European Spice

Association, 2004; Yu, 2012). Consumption of aflatoxin contaminated chilies has recently been associated with gall bladder cancer in Bolivia, Chile and Peru (Asai et al., 2014; Nogueira et al., 2015). Aflatoxin contamination is a problem in major

50 chili producing regions (Reddy et al., 2001; Shamsuddin et al., 1995; Singh and

Cotty, 2017), yet Aspergillus section Flavi associated with chilies and the etiology of chili contamination with aflatoxins have not been examined in detail. Structure of fungal communities associated with crops is an important determinant of the severity of aflatoxin contamination (Cotty et al., 2008; Probst et al., 2010), since higher incidences of aflatoxin-producers result in increased average aflatoxin- producing potentials of fungal communities, leading to overall high contamination

(Cotty et al., 2008).

A previous study evaluated aflatoxin contamination of dried red chilies between the US and Nigeria, two markets differing in regulation enforcement

(Singh and Cotty, 2017). Aflatoxin concentrations and fungal load were significantly higher in Nigerian chilies compared to those from US markets. The current study sought to understand communities of Aspergillus section Flavi associated with dried red chilies and to obtain insight into the etiology of contamination of dried red chilies. Our objectives were to (i) relate Aspergillus section Flavi in chilies from markets in the US and Nigeria to previously described members of section Flavi, and (ii) to determine the relative importance of each as etiologic agents of contamination. The acquired knowledge on aflatoxin-producing fungi associated with red chilies may be helpful in devising aflatoxin management strategies for the spice.

Materials and Methods

Sampling

Dried red chili was collected from markets in the US and Nigeria, as reported previously (Singh and Cotty, 2017). Briefly, 169 chili samples, sealed in airtight packets were purchased in the US from retail markets including supermarkets and ethnic groceries, where chilies were kept at ambient room temperature throughout storage. The retail stores were sampled in Arizona (n =

64), California (n = 68), Minnesota (n = 3), and New York (n = 34) during 2014-15 as representative locations for the US. US samples consisted of whole (n = 60, mean = 250 g), ground (n = 78, mean = 200 g) and crushed chili (n = 12, mean =

150 g), and paprika (n = 19, mean = 180 g). Fifty-eight percent of chili samples collected from US markets were imported (Singh and Cotty, 2017). Nations of origin for the imported chili samples are listed in Table 2.1. Only whole red chilies were collected in Nigeria (n = 55, mean = 70 g), primarily from rural, small-scale markets in Kaduna (n = 50) and Lagos (n = 5) states during 2015-16. Nigerian samples were imported to the USDA-ARS laboratory in the School of Plant

Sciences at the University of Arizona, Tucson, under permits issued by the USDA

Animal and Plant Health Inspection Service (APHIS) within a week of purchase.

Fungal isolation and characterization

Whole chili samples were dried (forced air oven, (40°C)) to below 8% moisture content and sealed in plastic bags; ground chili, crushed chili and paprika were sealed in plastic bags immediately after receipt. After bagging, samples were stored at room temperature. After drying, whole and crushed chili samples were finely ground in a laboratory mill (Retsch Grindomix GM200, Newtown, PA) for 30 s at 10,000 rpm prior to fungal isolation. In a previous study (Singh and Cotty,

2017), fungi belonging to Aspergillus section Flavi were recovered from the above samples by dilution plate technique on modified rose Bengal agar (Cotty, 1994a).

Fungi isolated in Singh and Cotty (2017) were assigned to morphological groups

(S morphology, A. flavus L morphotype, and A. parasiticus) using colony characteristics, sclerotia and spore morphology (Cotty, 1989; Klich and Pitt, 1988).

Fungal isolations were performed at least twice from each chili sample. All fungal isolates were subjected to dilution plating on malt agar (1% malt, 2% agar, 1000 ml of water) followed by incubation at 31°C for 48 h. At dilutions providing less than 10 colonies per plate, discrete colonies were transferred to 5-2 agar (5 % V-

8 juice; 2% agar; pH 6.0) and incubated at 31°C for 5-7 days in dark. Fungal isolates were stored as plugs of sporulating culture in sterile distilled water (2 ml) and used as working cultures for conducting aflatoxin and phylogenetic analyses.

Screen for aflatoxin producers

Aflatoxin production was evaluated for both L and S morphology isolates recovered from chilies. The L morphology fungi (n = 130) were randomly selected with at least one isolate from each chili sample positive for Aspergillus section

Flavi. Fungi with S morphology (n = 75) were randomly selected from 30 Nigerian and 5 US chili samples from which S morphology fungi were recovered. At least 1 isolate from each of the 35 samples positive for S morphology fungi was included.

Fungi were evaluated in a chemically-defined aflatoxin production liquid medium

(Mateles and Adye, 1965) supplemented with 22.5 mM urea as the sole nitrogen source (Cotty and Cardwell, 1999; Probst and Cotty, 2012). Fungal inoculum for each isolate was prepared as described previously (Probst and Cotty, 2012).

Erlenmeyer flasks containing 70 ml of the liquid medium were seeded with conidial suspensions (106 conidia ml-1), covered with stoppers that allow gas exchange, and incubated with agitation in dark for 5 days (31°C, 160 rpm). Fermentations were terminated by addition of acetone (70 ml acetone per 70 ml fermentation) and swirled to allow mixing. Cultures were allowed to sit for at least one hour to allow for lysis of fungal cells and release of aflatoxins contained in the mycelia.

Acetone extracts were directly spotted onto thin-layer chromatography (TLC) plates (Silica gel 60; EMD, Darmstadt, Germany) and separated adjacent to aflatoxin standards (Aflatoxin Mix Kit-M; Supelco, Bellefonte, PA). Plates were developed in a solution of ethyl ether-methanol-water (96:3:1), air-dried, and aflatoxins were visualized under 365-nm UV light. Total aflatoxins were quantified directly on TLC plates using a scanning densitometer (TLC Scanner 3, Camag

Scientific Inc, Wilmington, N.C.). Filtrates initially negative for aflatoxins were partitioned twice with dichloromethane and concentrated prior to quantification

(limit of detection 104 µg kg-1 mycelia) as previously described (Cardwell and

Cotty, 2002). Mycelial mass from the fermentation was captured during vacuum filtration on Whatman No. 1 filter paper and dried (40°C, 48h) in a forced air oven.

Aflatoxin concentrations were expressed as µg total aflatoxin per g mycelium.

DNA isolation and gene amplification

DNA extraction, PCR amplification and sequencing for fungi recovered from chilies and all reference isolates were done in the current study. Fungal cultures were grown and DNA was extracted as described previously (Callicott and Cotty,

2015). The reference isolates used in the current study were obtained from the

ARS Culture Collection, Peoria, IL, USA (indicated with NRRL in Table 2.5), the

American Type Culture Collection, Manassas, USA (indicated with ATCC in Table

2.5), or were present in the USDA, ARS, Tucson Laboratory Culture Collection.

Sequence analyses for chili and reference isolates were performed using two genomic regions; nitrate reductase (niaD) and aflatoxin pathway transcription factor (aflR). Partial gene sequences of both niaD (chromosome 4) and aflR

(chromosome 3) were amplified and sequenced in both directions, each with three sets of primers (Table 2.2) that covered approximately 2.1 kb of niaD and 1.9 kb of aflR genes. The primers aflR4F-4R and niaDCF-CR were designed in the current study based on genome sequence of A. flavus NRRL 3357 (GenBank accession no. AAIH02000041 and AAIH02000071) using Primer3 version 0.4.0

(Koressaar and Remm, 2007; Untergasser et al., 2012). PCR reactions were performed under the following conditions: 5 min at 94°C followed by 38 cycles of

94°C for 30 s, locus-specific annealing temperature for 30 s, 72°C for 30 s, and 5 min at 72°C. Amplicons were visualized on 1.0% agarose gels and sequenced by the University of Arizona Genetics core sequencing facility (UAGC, Tucson, AZ).

DNA sequence data and phylogenetics

Phylogenetic reconstruction was performed using bidirectional sequences of the genes aflR (1.9 kb) and niaD (2.1 kb) from 62 fungal isolates of Aspergillus section Flavi recovered from chilies and 17 reference isolates. Fungi were selected such that for each morphology (L, S or A. parasiticus) and aflatoxin producing ability (B or B and G), fungi from as many samples as possible were examined. All sequences utilized in phylogenetics were obtained during the current study.

Consensus sequences were created for each isolate by assembly of 6 reads per gene with visual inspection and alignment using the MUSCLE algorithm with the default settings in Geneious Pro Version 7.1.9. (Biomatters Ltd, Auckland, New

Zealand). DNA sequence alignments were refined manually. Phylogenetic analysis of sequence data for the two loci were performed for concatenated and individual sequences using Bayesian inference with 10 million generations

(MrBayes version 3.2.0; Huelsenbeck and Ronquist, 2001). Trees were drawn mid- point rooted using FigTree v.1.4.3 (Rambaut A, 2012). Maximum likelihood analysis for individual and concatenated sequences was performed in parallel to confirm tree topologies using PhyML (Phylogeny.fr (Dereeper et al., 2008;

Dereeper et al., 2010)). Data sets were bootstrapped with 500 replicates.

Crop inoculations

Results based on phylogenetic analyses were used to select fungi for inoculation onto sterile piquin chili (Capsicum annuum), maize (Zea mays) and groundnut (Arachis hypogaea). Four fungal isolates were randomly selected from each phylogenetically distinct clade and assessed for aflatoxin production on the three crops. In cases where clades contained fungi from both US and Nigerian chilies, isolates representative of both nations were randomly selected for aflatoxin analyses. For L morphology isolates, only aflatoxin producers, as determined from preliminary liquid culture aflatoxin assays, were included. Both maize and groundnut are hosts to members of section Flavi (Kachapulula et al., 2017).

Healthy, undamaged chili pods were autoclaved for 20 min at 121°C in sealed paper bags. The intact, autoclaved paper bags containing the chili were dried in a

56 forced air oven (40°C) for 48 hours. Each chili bag was opened under asceptic conditions in a biological safety cabinet and allowed to sit for another 24 hours to let the volatiles formed during autoclaving to escape. Chilies were then weighed out into previously sterilized Erlenmeyer flasks (5 g per flask).

Healthy, undamaged kernels of maize and groundnut were autoclaved in

Erlenmeyer flasks (5 g per flask) for 20 min at 121°C. Each crop was inoculated with conidial suspensions adjusted to 106 conidia/ml. Inoculated crops adjusted to

30% moisture were incubated for 10 d at 31°C in the dark. Each crop was incubated in a different experiment. Treatments were replicated four times and each experiment was performed twice.

Aflatoxin quantification in crops

Crop cultures were ground in 85% acetone (50 ml) in a laboratory grade

Waring Blender (seven-speed laboratory blender, Waring Laboratory, Torrington,

CT) at full speed for 30 s. The ground crop-acetone slurry was allowed to sit for an hour in the dark, and the culture filtrate was spotted directly onto TLC plates (Silica gel 60, EMD, Darmstadt, Germany) adjacent to an aflatoxin standard (Aflatoxin

Mix Kit-M, Supelco, Bellefonte, PA) containing a mixture of known concentrations of aflatoxins B1, B2, G1 and G2. Plates were developed in ethyl ether-methanol- water (96:3:1), air-dried, and examined for aflatoxins under 365-nm UV light. Total aflatoxins were quantified directly on TLC plates with a scanning densitometer

(TLC Scanner 3, Camag Scientific Inc., Wilmington, NC, USA). Chili cultures initially negative for aflatoxins were concentrated with a previously reported modification to the method of the Association of Official Analytical Chemists (Cotty

57 and Lee, 1989; Mckinney, 1975; Stoloff and Scott, 1984). Briefly, a 20 ml solution of 1.1M (CH3COO)2Zn and 0.04M AlCl3 diluted with 80 ml of deionized water was added to the negative samples. After 5 min, 5 g of diatomaceous earth was added and the resulting mixture was shaken to allow mixing. These mixtures were allowed to sit for 1-2 hours in order to minimize interfering pigments, fatty acids, and trace lipids, and to obtain cleaner extracts for TLC (Mckinney, 1975). Extracts were filtered through number 4 Whatman paper. Aflatoxins were extracted from

100 ml filtrate by partitioning the filtrate twice with 25 ml of dichloromethane. The extracts were passed through a bed of anhydrous Na2SO4 and air-dried. Residues were dissolved in dichloromethane, spotted on TLC plates, developed and quantified as above. Both maize and groundnut cultures were positive for aflatoxins in the initial round of extraction and did not require concentration.

Data Analysis

Total aflatoxin was measured in µg/g. Aflatoxin concentrations were log transformed and subjected to Analysis of Variance (JMP 11.1.1, SAS Institute,

Cary, NC). In experiments with significant differences by ANOVA, mean

SD test (p = 0.05). True means are presented for clarity.

Results

Incidence of Aspergillus section Flavi in dried red chilies

A total of 530 Aspergillus section Flavi isolates were recovered from 169 samples purchased from US markets (Table 2.3). Aspergillus flavus L strain had the highest incidence (92.1%), followed by A. tamarii (4.3%), and fungi with S

58 morphology (3.0%). Aspergillus parasiticus occurred in low numbers (0.6%).

Aflatoxin assays and phylogenetic analyses further characterized the S morphology fungi into A. flavus S strain (2.4%) and A. aflatoxiformans (0.6%).

Fungal isolation from Nigerian samples resulted in a total of 565 Aspergillus section Flavi isolates from 55 samples (Table 2.3). Majority of these fungi were assigned to A. flavus L strain (76.7%). Incidences of fungi with S morphology were higher in Nigerian chilies than in chilies from US markets. Phylogenetics resolved the S morphology fungi into A. aflatoxiformans (8.3%), A. minisclerotigenes (8.0%), and a new lineage (2.8%) discovered in the current study. Aspergillus tamarii occurred at 4% whereas incidences of A. parasiticus were low (0.2%).

Aflatoxin producers from chilies: Occurrence of non-native aflatoxin producers in chilies from US markets

Fungi recovered from chilies were initially assayed for aflatoxin production in liquid fermentation. Out of the 205 isolates tested, over 70% of isolates produced aflatoxins (LOD = 104 µg kg-1 mycelia). All L morphology aflatoxin-producers recovered from chilies produced only B aflatoxins, with 71% from US (n = 65; mean

= 27 µg g-1 mycelium) and 40% from Nigerian (n = 65; mean = 58 µg g-1 mycelium) markets producing detectable quantities of aflatoxins (Table 2.4). The S morphology fungi produced higher concentrations (p < 0.01) of aflatoxins in A&M medium with urea than the Ls (Table 2.4). Three isolates from imported chili purchased at US markets had S morphology and produced both B and G aflatoxins. The chilies were imported from Nigeria. All S morphology fungi from

Nigerian chilies produced both B and G aflatoxins.

Phylogenetic analyses

Phylogenetic analyses of the concatenated niaD and aflR sequences (2.1 kb for niaD and 1.9 kb for aflR) resolved fungi from red chilies into five highly supported clades (Figure 2.1). The first taxon, A. parasiticus (e.g. reference isolates 2999 and BN009-E), was recovered in low numbers from both US and

Nigerian chilies. The second taxon, consisting of B aflatoxin-producers, included previously described L and S strain A. flavus isolates. Aspergillus flavus S strain isolates were only detected in US chilies, whereas L strain isolates occurred in both US and Nigerian chilies. The third taxon, A. minisclerotigenes (e.g. reference isolates A-11611, 4-2 and TAR3N43) was recovered only from Nigerian chilies.

The fourth taxon, A. aflatoxiformans, included fungi that grouped with reference isolates A-11612 and BN038-G, and consisted solely of fungi from chilies produced in Nigeria. The fifth and final clade was occupied by a novel lineage discovered in the current study. This new lineage with S morphology was detected only in

Nigerian chilies. No reference isolates grouped with this lineage.

The new lineage was sister to A. korhogoensis (Figure 2.1). In addition to the results from liquid fermentation, phylogenetic reconstruction validated the occurrence of A. aflatoxiformans isolates in chilies from US markets (Figure 2.1,

Table 2.5). Based on DNA sequence analysis and phylogenies, 44% of S morphology isolates from Nigeria were identified as A. aflatoxiformans, 42% as A. minisclerotigenes and 15% as the new lineage. Aspergillus korhogoensis was not detected in the current study.

Aflatoxin production in crops

Individual isolates of A. flavus S strain, A. parasiticus, the new lineage and

A. aflatoxiformans varied in aflatoxin producing ability on each of the hosts (p <

0.05) (Table 2.6). Both A. aflatoxiformans and the new lineage produced the highest concentrations of aflatoxins in chilies, maize and groundnuts. On chilies, these two species produced over 10 fold more total aflatoxins than the other S morphology taxa. The S strain of A. flavus and A. parasiticus produced similar concentrations of aflatoxins in chilies (Table 2.6). Among the S morphology fungi,

A. minisclerotigenes produced the lowest concentrations of aflatoxins in all the three crops. The L strain of A. flavus was the least toxic in chilies but produced quantities of aflatoxins comparable to A. minisclerotigenes in groundnuts and maize.

Discussion

Most aflatoxin-producers belong to Aspergillus section Flavi. Since aflatoxin-producing ability can be highly variable among species, strains and isolates, it is difficult to attribute specific etiologies to aflatoxin contamination of crops. To identify the most important causal agents of aflatoxin contamination of chilies, both, frequency of crop infection and aflatoxin-producing ability must be considered. Characterization of causal agents is critical for development of management procedures. Although occurrence of Aspergillus spp. in chilies has been reported in previous studies (Flannigan and Hui, 1976; Kiran et al., 2005;

Reddy et al., 2011), this study provides the first comprehensive report of identity and toxigenicity of species within section Flavi associated with dried red chilies.

The A. flavus L strain was the predominant member of section Flavi in chilies

(Table 2.3). Similar dominance of the L strain has been reported in other important hosts such as maize, cottonseed, legumes and tree nuts (Boyd and Cotty, 2001;

Doster et al., 2014; Bayman and Cotty, 1991; Probst et al., 2014). Fungi with S morphology were recovered from chilies purchased in both the US and Nigeria, with higher incidence in Nigerian chilies (Table 2.3). Aspergillus parasiticus, a highly toxigenic B and G aflatoxin-producing species, was rare in chilies, comprising of only 0.6% and 0.2% of Aspergillus section Flavi in chilies from US and Nigerian markets, respectively. Lower recoveries of fungal isolates from US samples were due to failure to recover section Flavi fungi from 102 (60%) samples; however, fungi were recovered from all Nigerian samples (Singh and Cotty, 2017).

Recently, several fungi with S morphology that are phylogenetically distinct but morphologically indistinguishable have been reported (Cotty, 1989; Pildain et al., 2008; Probst et al., 2012). The B aflatoxin-producing A.flavus S strain is only known to be common in crops and soil in the US (Cotty, 1989; Horn, 2003). This contrasts with A. flavus L strain, which has been reported in high frequencies across warm regions of the globe (Bayman and Cotty, 1991; Ehrlich et al., 2007;

Probst et al., 2014). The S strain of A. flavus has never been reported in West

Africa. However, a distinct species with S morphology and ability to produce both

B and G aflatoxins has been frequently isolated from maize and soil in Nigeria and

Benin and has been referred to as the unnamed taxon SBG (Atehnkeng et al., 2008;

Cardwell and Cotty, 2002; Donner et al., 2009). This taxon was recently described as A. aflatoxiformans (Frisvad et al., 2019). In the current study, all S morphology fungi from Nigerian chilies produced high concentrations of B and G aflatoxins in

62 liquid fermentation. This is in agreement with findings from previous studies on aflatoxin-producing fungi in West Africa (Atehnkeng et al., 2008; Cardwell and

Cotty, 2002; Donner et al., 2009). These distinct S morphology fungi that produce both B and G aflatoxins are resident in agroecosystems of Nigeria. In contrast, the only S morphology fungus known in the US is the A. flavus S strain, which produces only B aflatoxins (Cotty et al., 2008; Jaime-Garcia and Cotty, 2006a). In the current study, 3 isolates with S morphology, and B and G aflatoxin producing ability, were detected in chilies from US markets (Table 2.4). These isolates were traced to chili imported from Nigeria. Crop contamination with B and G aflatoxins in the US has been attributed to A. parasiticus, and S morphology fungi with the ability to produce both B and G aflatoxins have not been reported to date, indicating introduction of a highly toxigenic non-native pathogen to the US with chili. Although it is undesirable to allow non-native pathogens in a given region, such cases frequently occur because of trade and human movement. Similarly, both Dutch elm disease and chestnut blight are introduced pathogens that moved through commerce and caused death of billions of trees in the US (Agrios, 2005). If the highly toxigenic non-native aflatoxin producing fungi from Nigeria reach agroecosystems in the US, they may exacerbate aflatoxin contamination.

Phylogenetic reconstruction resolved Aspergilli recovered from chilies into five clades: a) A. parasiticus, b) the L and S strain of A. flavus, c) A. minisclerotigenes, d) A. aflatoxiformans, and e) a novel lineage. Fungi with S morphology from Nigerian chilies were indistinguishable based on morphology and types of aflatoxins produced. However, DNA sequence comparisons resolved

63 these fungi into three distinct taxa: A. minisclerotigenes, A. aflatoxiformans, and a new lineage. Aspergillus minisclerotigenes was described from Argentinian groundnuts (Pildain et al., 2008), and has been reported in central and southern

Africa, whereas A. aflatoxiformans predominates West Africa (Probst et al., 2014).

This is the first study to report occurrence of multiple lineages of genetically distinct

S morphology fungi with B and G aflatoxin producing ability in West Africa.

Furthermore, A. minisclerotigenes was detected at a higher frequency in Nigerian chilies (8% of all Aspergilli) compared to its sparse occurrence previously reported in maize from Central and Eastern Africa (Probst et al., 2014), maize and almonds from Portugal (Soares et al., 2012), and spices marketed in Morocco (El Mahgubi et al., 2013).

Aspergillus minisclerotigenes is indistinguishable from A. aflatoxiformans based on morphology and aflatoxin production. Both A. aflatoxiformans and A. minisclerotigenes produced similar concentrations of total aflatoxins in liquid fermentation (data not shown). Studies on distribution of Aspergillus section Flavi in West Africa (Cardwell and Cotty, 2002; Donner et al., 2009) based results on morphological and/or physiological characteristics of S morphology fungi, and as a result, were unable to resolve S morphology fungi with B and G aflatoxin producing ability into multiple taxa.

Members of A. aflatoxiformans and the new lineage produced the highest concentrations of aflatoxins in chili, groundnut and maize (Table 2.6). All of these are important crops in Nigeria based on both production and consumption

(FAOSTAT, 2017), and occurrence of these highly toxigenic fungi in Nigeria

64 suggests an unrealized basis for vulnerability to aflatoxin contamination. More than

90% of Nigerian chilies in the current study were collected from Kaduna state, which falls within the Northern Guinea Savannah (NGS) agroecosystem. Toxigenic

SBG fungi have previously been reported in the NGS region of Nigeria in high frequencies (Donner et al., 2009). Incidences within infecting fungal communities and aflatoxin-producing potential must both be considered when assessing the importance of specific taxa to determine etiology of aflatoxin contamination events.

Although the most toxic fungi, A. aflatoxiformans and members of the new lineage, had lower incidences than the A. flavus L strain, they produced at least 300 times more aflatoxins in chilies. On the other hand, isolates belonging to A. minisclerotigenes produced significantly lower concentrations of total aflatoxins in chilies compared to A. aflatoxiformans and the new lineage (Table 2.6); however, increased prevalence of A. minisclerotigenes can lead to contamination of the crop with unacceptable concentrations of B and G aflatoxins. Discovery of a novel lineage phylogenetically distinct from A. aflatoxiformans but very similar morphologically and physiologically, indicates that diverse communities of genetically distinct SBG-like fungi may be resident in agroecosystems across West

Africa.

Aflatoxin management strategies have received increased attention over the past two decades owing to serious health and economic concerns caused by aflatoxin contamination of crops. Biological control of aflatoxins using native, well- adapted atoxigenic L strains of A. flavus has been a successful strategy for mitigation of crop contamination (Atehnkeng et al., 2016; Cotty, 1994b; Cotty,

1999; Doster et al., 2014). Atoxigenic isolates are a subset of the L strain of A. flavus, which was the predominant member of Aspergillus section Flavi from chilies in the current study. Of the total L strain isolates tested for aflatoxin production,

45% did not produce detectable concentrations. The L strain isolates of A. flavus from the current study are a genetic resource for selection of active ingredients for biological control products directed at reducing aflatoxin contamination of chilies.

Representative isolates of the new lineage (NRRL 66829, NRRL 66830,

NRRL 66831) have been deposited at the ARS Culture Collection (NRRL) (Peoria,

IL, USA). DNA Sequences of representative isolates from each taxon were deposited at GenBank under accession numbers MH752557- MH752590 and

MH760519 - MH760551.

Table 2.1 Chili samples purchased from US markets and country of origin.

Chili samples Country of origin (n)

Bangladesh 4 China 17 India 49 Korea 4 Mexico 10 Myanmar 1 Nigeria 1 Pakistan 2 Thailand 7 Turkey 2 USA 9 Vietnam 1 Unknown 62

Reference Product Product size (bp) size ) used for PCR amplifications. for used ) a

(°C) 57 737 Probst et al., Probst 2012 737 et al., Probst 2012 745 Current Study 57 735 et al., Probst 2012 57 795 et al., Probst 2012 57 Current study 799 52 792 57 55 68 a eune T Sequence F-AGAGAGCCAACTGTCGGACCAA R-GGGTGACCAGAGAACTGCGTGAT F-GACTTCCGGCGCATAACACGTA R-ACGGTGGCGGGACTGTTGCTACA F-CGCCCATGACGGACTACGTT R-TGGTGGTTGATTCGATTGAGG F-CGGACGATAAGCAACAACAC R-GGATGAACACCCGTTAATCTGA F-ACGGCCGACAGAAGTGCTGA R-TGGGCGAAGAGACTCCCCGT F-GCAGCCCAATGGTCACTACGGC R- GGCTGCACGCCCAATGCTTC

Primers and locus specific (Tannealing Primers temperature and locus Primerpair aflR1F-1R aflR2F-2R aflR4F-4R niaDF-AR niaDBF-BR niaDCF-CR Table 2.2 2.2 Table A. A. parasiticus A. A. tamarii New Lineage

b A. A. aflatoxiformans Species/Taxon(%) A. A. minisclerotigenes 69

S strain S in chilies from US and Nigerian markets.fromNigerian chilies and US in A. flavus A.

Flavi Lstrain A. flavus A. section section No. of No. Isolates Aspergillus 55 565 76.7 0.0 8.0 8.3 2.8 4.0 0.2 0.2 4.0 2.8 8.3 8.0 0.0 76.7 565 55 6 50 21 . 00 . 00 . 0.6 4.3 0.0 0.6 0.0 2.4 92.1 530 169 Chili Chili

Samples

15 16 2014- 2015- Incidence ofIncidence

# USA Nigeria Year samples were collected. sampleswere Year onr Year Country Assignment of fungal recoveredisolates from red chilies to species/taxa based on morphological, physiological and countries and other imported. the samples) in US purchased in produced were(98 chilies Some Table 2.3 2.3 Table data. DNAsequence a b #

)

< 0.01). <0.01).

-1 A A B B

p p Mean (µg (µg g Total AF Total

) -1

(µg (µg g Mean AFG Mean from red chilies in liquid fermentation. liquid in chilies from red ) Flavi -1

(µg (µg g Mean AFBMean section section

70 Aflatoxin (AF) ProductionAflatoxin

(%) AFG AFG Aspergillus Aspergillus Producers Producers

(%) AFB AFB Producers Producers

# # Tested Isolates Isolates

6 7 0 7(D71 N 27 NA 58 (ND-761) 27 NA 0 (ND-2,450) 58 0 71 65 40 65 L L 1 10 0 4 3-1) 0 2.-5) 268 (25.5-255) 103 (33-614) 247 767 20 (2-2,765) 232 (3-4,960) 535 100 100 15 100 S 60 S Fungal Fungal Morphology

Aflatoxin production by members by Aflatoxinof production

USA Nigeria

Country Chili samples were purchased in USA but all did not originate from USA. from USA all but originate not purchased did sampleswere in USA. Chili Table 2.4 2.4 Table µg/kg =104 mycelium. LOD measured. aflatoxin concentrations of range parenthesis the in indicate Values Detected. Not ND- do not aflatoxins. produceG fungi These applicable. Not NA-

Species f Group

e Aflatoxins d 71 Morphology 904 L 2 AF 2 B L 96044* c ® MA8* B AF AF AF 2 2 2 B B B S S S MYA382* MYA383* MYA384* ® ® ® Isolate H09 B AF AP AF 2 AP AF AF 1 2 B 1 2 AP AP 2 BG AA AA B BG B L 1 1 B N/A 4 4 N/A BG L B L BG BG L CHL019 N/A CHL111 N/A CHL121 S CHL133 AA S CHL159 AM CHL187 AM CHL302 AM 4 CHL303 AA 3 CHL326 AM 3 BG CHL327 AM 3 BG AM 4 AF BG 3 BG 3 S BG 3 2 S BG S BG S BG B S S CHL568 S CHL583 S CHL603 L CHL627 CHL633 CHL639 CHL650 CHL674 CHL676 isolates used for phylogenetic reconstruction in thestudy. current in phylogeneticreconstruction usedfor isolates

b Flavi S NR 21 S 2 AF 2 B S ATCC AF12 = ATCC AF42 = ATCC AF70 = NRRL3251* USA ATCC AF13 = USA USA USA USA Origin Country of Country section section a

S ni H41SB2AF AF AF 2 2 2 B B B S S S Unknown Unknown Pakistan CHL441 India USA CHL358 India USA CHL446 USA Nigeria USA Nigeria USA Unknown India USA India USA USA USA USA iei Nigeria Nigeria Nigeria Nigeria Nigeria Nigeria Nigeria Nigeria Nigeria Nigeria Nigeria Nigeria Nigeria Nigeria Nigeria Nigeria Purchase Country of Country Aspergillus

S Unknown Unknown USA USA Soil hl UA Myanmar USA Chili Nigeria Nigeria Chili Walnuts Substrate Table 2.5 2.5 Table

S BG 4 AA AA 4 BG S

MYA380* ® H77 B 3 AM 3 BG AA NewLineage 4 AA S NewLineage 5 5 NewLineage BG AM 4 BG AA 5 AA BG CHL707 BG 3 S BG 4 S 4 NewLineage AM S BG 5 S BG BG NewLineage S AP AM 3 CHL813 AA 5 AF BG S CHL814 S 1 BG CHL816 3 S BG CHL819 4 2 CHL820 S BG BG S BG CHL845 B S CHL877 N/A CHL878 S S CHL888 L CHL895 CHL941 CHL946 CHL947 CHL962 CHL993 N0-* / B 1 AP 1 BG N/A BN009-E* A34* B 3 AM AM 3 3 BG BG S S TAR3N43* TAR4N30* BNO38G NRRL 2999* N/A BG 1 AP AP 1 BG N/A NRRL2999* RLA162 S G AA AM 4 3 BG BG S S NRRLA-11612* NRRLA-11611* ATCC H81=NR 62 S G NewLineage 5 BG NewLineage 5 S NewLineage BG CHL801NRRL 66829= 5 S BG CHL832NRRL 66830= S CHL884NRRL 66831=

Benin Benin Nigeria Nigeria Uganda Argentina Argentina

iei Ngra H73 B AF AA AF 2 AA 4 2 B 4 BG B BG L S L S CHL723 CHL740 CHL755 CHL798 Nigeria Nigeria Nigeria Nigeria Nigeria Nigeria Nigeria Nigeria Nigeria Nigeria Nigeria Nigeria Nigeria Nigeria Nigeria Nigeria Nigeria Nigeria Nigeria Nigeria Nigeria Nigeria Nigeria Nigeria Nigeria Nigeria Nigeria Nigeria Nigeria Nigeria Nigeria Nigeria Nigeria Nigeria Nigeria Nigeria Nigeria Nigeria Nigeria Nigeria Nigeria Nigeria Nigeria Nigeria Nigeria Nigeria

Soil Soil Groundnut Groundnut Groundnut , and AK- AK- and , A. aflatoxiformans A. , AA- , A. minisclerotigenes A. AM- 73 -* B 3 AM 3 BG S 4-2* A. flavus, flavus, A. RL678 S G / AK AK AK N/A AK N/A N/A BG N/A BG BG BG S S S S NRRL66708* NRRL66709* NRRL66710* NRRL66711* , AF- AF- ,

A. parasiticus A. Australia

Soil B refers to B aflatoxin production by fungal isolates; BG refers to production of both B and G aflatoxins. Aflatoxin Aflatoxin aflatoxins. G and B both of production to refers BG isolates; fungal by production aflatoxin B to refers B Species assignment; AP- AP- assignment; Species Country of purchase- Country from where chilies were purchased. Country purchased. were from chilies where purchase- of Country samplesoriginated. the fromactually Country where origin- of Country morphologyor applicable. group not is into classification N/A; Groundnut Groundnut Groundnut Groundnut Group indicates phylogenetic groups in Figure 2.1. groups Figure in phylogenetic Group indicates Fungal isolates recovered from chilies and reference isolates (indicated by *). Isolates with NRRL deposited at at deposited NRRL with Isolates *). by (indicated isolates reference and chilies from recovered isolates Fungal USDA ARS Culture Collection; with ATCC deposited at American Type Culture Collection. Isolates without NRRL NRRL without Isolates Collection. Culture Type American at deposited ATCC with Collection; Culture ARS USDA workwas performed. this laboratorywhere the USDA,ARS at ATCC Tucson or cultures reference of profiles aflatoxin study; current the in determined were fromchilies recovered fungi of profiles 2017).al., et al., 2012; et reported (Probst Carvajal-Campos werepreviously A.korhogoensis a b c d e f g

A A A y A B A A A x x A

B B y B B B B AB C y Total Total AF 41.4 44.3 104.0 101.0 72.8 111.0 355.0 221.0 284.0 243.0 296.0 428.0

A A A x x A

101.0 74.6 B B A A y B 31.6 33.7 70.0 31.5 71.0 37.7 51.5 287.0 NA 49.1 NA NA 139.0 NA 55.1 NA 80.7 NA 23.5 NA NA NA NA 75.0 263.0 158.0 161.0 164.0 222.0 347.0 137.0 395.0 235.0 264.0 193.0 Groundnut Groundnut 292.0

A x A A x B

B A A y B B B B AB C x B A AB x

B AFB AFG AFG AFB 9.8 10.6 21.3 31.5 37.7 101.0 139.0 23.5 74.6 63.0 123.0 78.2 85.0 112.0 103.0

34.0 30.0 287.0 49.1 55.1 80.7 35.0 92.0 219.0 137.0 82.0 131.0

136.0

A A A B A B yz A BC B yz C B A A y B C A A x B z

B C C ) -1 Total Total AF 48.7 126.0 201.0 161.0 134.0 100.0 254.0 374.0 345.0 239.0 462.0 354.0 618.0 708.0 519.0 from the current study in various study hosts. in various fromcurrent the

565.0

A A AB A B y B C A A x B

B A z C 212.0 166.0 Flavi Maize 37.4 95.0 133.0 55.7 111.0 184.0 94.0 512.0 NA 171.0 NA NA 336.0 NA 131.0 NA 182.0 NA 93.5 NA NA NA NA 73.0 195.0 268.0 192.0 182.0 334.0 257.0 440.0 496.0 382.0

399.0

B A B x A BC B x B A C B A x C

C B A AB z C C D C y D section section Aflatoxin (AF) (µg g (µg (AF)Aflatoxin 74 F AFG AFB 11.3 31.0 39.9 55.7 184.0 231.0 336.0 93.5 185.0 106.0 154.0 86.5 98.0 178.0 154.0

27.0 59.0 128.0 212.0

166.0

AB B AB A A w 512.0 171.0

A A A x A x

Aspergillus Aspergillus

y B C B B z Total Total AF 1.9 18.4 202.0 134.0 176.0 420.0 233.0

201.0

AB A x A 55.8

16.7 0.4 A B AB 1.7 8.4 7.7 C B B y z Chili 0.8 NA NA 9.5 101.0 199.0 116.0 89.8 AB A x A 13.8 30.4 22.3 NA NA NA 30.6 72.5 124.0 A A A y A B AB A 0.6 A 1.0 NA 4.2 2.6 z B C B B y z F AFG AFB 1.1 16.7 0.4 8.9 118.0

slts Isolates Average Average Aflatoxin production by members by Aflatoxinof production

a H85 07 . 12 68.0 50.0 1.2 1.2 3.3 0.7 0.5 1.2 0.5 0.7 CHL707 2.1 CHL845 131.0 CHL947 182.0 13.8 0.2 30.4 0.2 CHL303 22.3 0.3 CHL358 NA CHL441 NA NA 0.2 0.2 CHL404 0.3 CHL537 CHL755 4.2 5.1 25.2 CHL121 CHL302 2999 61.8 86.2 CHL832 221.0 CHL888 CHL941 L H81 101.0 CHL801 NL P H11 1.1 CHL111 AP 76.7 CHL326 AA AM CHL674 1.0 0.8 1.8 1.8 0.8 1.0 CHL674 AM FLCL8 . A0.8 NA 0.8 CHL280 L AF FS H18 1.0 CHL178 S AF Average Average Average Species Table 2.6 2.6 Table , , p p

A A x

B 89.8 935.0 398.0 463.0

A A x

B within each crop each within 59.0 523.0 232.0 276.0 A. minisclerotigenes Flavi

A AB x

C , , AM- 412.0 186.0 section section 30.8 167.0

C A B x A. parasiticus 204.0 782.0 487.0 510.0 Aspergillus

, AP- D A C x L L Sand strain morphotypes were excluded when 134.0 480.0 275.0 322.0 <0.01). >0.05). p p A. flavus p

A B x

D A. flavus 75 303.0 188.0 70.0 212.0

A A w

B 19.4 132.0 165.0 308.0 A

B A x 149.0 A 67.3 88.0 A x 10.9 B , and NL- New andLineage. , NL- 77.0

and total concentrations were compared by column between and within species, separately for chili, for separately species, within and between column by compared were concentrations total and

Average H63 8.5 159.0 CHL633 64.7 CHL819 CHL962 A.aflatoxiformans Not Not applicable. These fungi do not produce G aflatoxins. Species Species recovered from chilies used for crop inoculation; AF- AA- G B, Aflatoxin concentration 4 mean replicates. a is of toxin Each groundnut. maize and of species distinct by produced concentrations aflatoxin in Differences by are species indicated producedconcentrations within aflatoxin in Differences <0.01). not (ANOVA, ado within differ column a lacking letter Values species. andwithin between aflatoxin G comparing a based on concatenated concatenated on based Flavi section section Aspergillus 76 (2.1 kb) genes. Values above nodes are Bayesian posterior posterior abovenodesare Bayesian Values genes. kb) (2.1 niaD and and

(1.9 kb) aflR aflR Mid-point rooted Bayesian phylogeny of 62 of isolates 62 phylogenyof rootedBayesian Mid-point Figure 2.1 2.1 Figure of genesequences partial bootstrap replicates. values are from nodes below values and 500 probabilities APPENDIX C- Phenotypic differentiation of two morphologically similar aflatoxin producers from West Africa

To be submitted to: Applied and Environmental Microbiology

Abstract

Aflatoxins (AF) are hepatocarcinogenic metabolites produced by species within Aspergillus section Flavi. Aflatoxins contaminate cereals, groundnuts, tree nuts, and spices in warm regions of the world. Etiology of aflatoxin contamination may be complicated by mixed infections of multiple species with similar morphology and aflatoxin profiles. The current study investigates the physiology of

A. aflatoxiformans and A. minisclerotigenes in liquid fermentation. These species co-exist in West Africa and have overlapping phenotypes. Both species contaminate maize with large concentrations of aflatoxins under various environmental conditions and therefore to attribute etiologies of specific events, reliable differentiation of these fungi is needed. During liquid fermentation experiments, a physiological distinction between these two species was discovered, which is sufficient to differentiate the species and may be useful when

DNA technology is unavailable. The two species were compared in liquid fermentation in defined media containing either urea or ammonium as sole nitrogen sources, and a yeast-extract based medium with sucrose (YES). The species produced similar concentrations of AFB1 in media with urea or ammonium

(p > 0.05), but AFB1 production was inhibited for A. aflatoxiformans in YES medium

(p < 0.001). Although production of AFG1 by both species was similar in urea, A. minisclerotigenes produced higher concentrations of AFG1 in ammonium (p =

0.039). Similar to AFB1, AFG1 production by A. aflatoxiformans was inhibited in

YES medium. A reliable and convenient assay for differentiating the two species was designed. This assay may be useful for identifying specific etiologic agents of aflatoxin contamination episodes in West Africa.

Introduction

Aflatoxins (AF) are potent carcinogenic metabolites produced by fungi within Aspergillus section Flavi, which can infect and contaminate a wide range of crops, including cereals, groundnuts, cottonseed, tree nuts, and spices (Egal et al., 2005; Lee et al., 1989; Ortega-Beltran et al., 2018; Singh and Cotty, 2017;

Soares et al., 2012). Species within section Flavi produce four major aflatoxins, aflatoxin B1, B2, G1, and G2. The occurrence of aflatoxins in human food or animal feed is a serious health and economic threat worldwide. Aflatoxin B1 is the only mycotoxin listed as a human carcinogen by the International Agency for Research on Cancer (IARC, 2002). Chronic exposure to aflatoxins results in compromised immune system, growth impairment and hepatocellular carcinoma in humans and animals (Khlangwiset et al., 2011; Liu and Wu, 2010; Turner et al., 2003;).

Consumption of food contaminated with high concentrations of aflatoxins results in severe liver damage and rapid death (Williams et al., 2004). Epidemics of acute aflatoxicosis leading to more than 100 human deaths have been recorded in India,

Kenya, and recently in Tanzania (Africa, 2016; CDC, 2004; Krishnamachari et al.,

1975; Outbreak News Today, 2017). Stringent enforcement of aflatoxin regulations in developed nations (e.g. total aflatoxins are regulated at 20 µg/kg in the United

States) leads to rejection and/or destruction of contaminated food, which causes huge economic losses to growers (Robens and Cardwell, 2003).

Aflatoxin-producing fungi have a broad host range. Individual fields, regions, and agroecosystems harbor diverse communities of aflatoxin-producing

Aspergilli, and incidence of specific communities may directly influence aflatoxin

79 contamination of crops (Donner et al., 2015; Horn and Dorner, 1998; Probst et al.,

2010). Aspergillus flavus, which produces B aflatoxins, is the most frequently implicated causal agent of aflatoxin contamination of crops (Klich, 2007). It is delineated into the L and the S morphotypes, based on large and small sclerotium types, respectively. The L morphotype produces few, large sclerotia (> 400 µm), abundant quantities of conidia, and variable levels of aflatoxin ranging from atoxigenic isolates to others that produce medium to high levels of toxin (Bayman and Cotty, 1993; Cotty, 1989). The S morphotype of A. flavus produces copious quantities of small sclerotia (< 400 µm), sparse conidia, and consistently high levels of aflatoxins (Cotty, 1989). Atoxigenic isolates of A. flavus are active ingredients in biological control products used towards the management of aflatoxin contamination of crops worldwide (Bandyopadhyay et al., 2016; Cotty,

1994; Dorner and Lamb, 2006).

Aflatoxin contamination of food and feed in West Africa has been attributed to Aspergilli with S morphology (sclerotia < 400 µm) and the ability to produce both

B and G aflatoxins (SBG) (Agbetiameh et al., 2018; Cardwell and Cotty, 2002;

Donner et al., 2009). In the past decade, molecular phylogenetics has identified the occurrence of distinct species of SBG fungi in West Africa; A. minisclerotigenes,

A. aflatoxiformans, A. cerealis (previously A. korhogoensis), A. austwickii, and A. pipericola (Carvajal-Campos et al., 2017; Frisvad et al., 2019; Pildain et al., 2008;

Singh and Cotty, 2019). However, the significance of the more recently described species including A. cerealis, A. austwickii, and A. pipericola to crop contamination is unclear. On the other hand, A. aflatoxiformans, previously reported as an

80 unnamed taxon SBG or as A. parvisclerotigenus, has been frequently reported as an important etiologic agent of crop contamination in West Africa (Cardwell and

Cotty, 2002; Ezekiel et al., 2013; Perrone et al., 2014; Singh and Cotty, 2019). A. minisclerotigenes is widespread, and occurs in Sub-Saharan Africa, South

America, Australia, and Europe (Pildain et al., 2008; Probst et al., 2012; Soares et al., 2012). Although A. minisclerotigenes was previously only reported from the

Central and Eastern areas of Sub-Saharan Africa (Probst et al., 2014), Singh and

Cotty (2019) recovered this species in high frequencies from dried red chilies

(Capsicum spp.) originating from Nigeria, establishing the co-occurrence of A. minisclerotigenes and A. aflatoxiformans in Nigerian agroecosystems. Both A. aflatoxiformans and A. minisclerotigenes have an S morphology and produce both

B and G aflatoxins in various culture media and crops (Frisvad et al., 2019; Pildain et al., 2008; Singh and Cotty, 2019), rendering these species indistinguishable based on morphology and aflatoxin production. Additionally, the two species have largely overlapping morphological and secondary metabolite profiles. Currently, A. aflatoxiformans and A. minisclerotigenes are resolved using phylogenetics based on DNA sequences from multiple unlinked loci (Pildain et al., 2008; Singh and

Cotty, 2019). Both A. aflatoxiformans and A. minisclerotigenes can contaminate crops of significant importance in terms of consumption and production in West

Africa, including maize (Zea mays), groundnuts (Arachis hypogaea) and chilies

(Capsicum spp.) with high concentrations of aflatoxins (Singh and Cotty, 2019).

Compositions of fungal communities along with their aflatoxin-producing potentials are key determinants of aflatoxin contamination of crops. For instance, increased

81 frequencies of atoxigenic Aspergillus results in lower aflatoxin levels in crops, while occurrence of high aflatoxin-producers causes unacceptable crop contamination.

A precise understanding of the etiology of aflatoxin contamination of crops and accurate identification of causal agents are therefore critical for predicting risk of contamination and the development of aflatoxin mitigation strategies. Although

DNA sequencing and phylogenetics can reliably distinguish A. aflatoxiformans from A. minisclerotigenes, this approach requires technical expertise and instrumentation that may not be easily accessible in many developing countries.

Liquid fermentations have been traditionally used to measure aflatoxigenicity of isolates along with fungal biomass and their ability to modify the pH of the fermentation medium (Mateles and Adye, 1965; Cotty and Cardwell,

1999; Probst and Cotty, 2012). Previous studies have explored the effects of one or more factors (varying concentrations of NaCl or sucrose, initial pH of the medium, different N sources, fermentation with or without agitation etc.) on aflatoxin production by isolates of A. flavus and A. parasiticus (Davis et al., 1966;

Reddy et al., 1979; Probst and Cotty, 2012; Shih and Marth, 1972; Shih and Marth,

1974). The current study sought to assess effects of different fermentation media

(Yeast Extract and Sucrose (YES), Adye and Mateles medium supplemented with either urea or (NH4)2SO4) on aflatoxin production by the morphologically similar species A. aflatoxiformans and A. minisclerotigenes. We discovered that A. aflatoxiformans produced significantly lower concentrations of aflatoxins in YES medium compared to that by A. minisclerotigenes. We utilized this phenotype to

82 design a microbiological assay utilizing YES medium as a potential tool to reliably distinguish the two species.

Materials and Methods

Fungal isolates and inoculum preparation

Isolates of A. aflatoxiformans (n = 45) and A. minisclerotigenes (n = 44) previously recovered from dried red chilies produced in Nigeria (Singh and Cotty,

2019) were included in this study. The fungal inoculum was prepared for each isolate as described previously (Mehl and Cotty, 2011; Probst and Cotty, 2012).

Briefly, conidial suspensions from water vials were centrally seeded onto 5/2 agar

(5% V8 juice, 2% agar, pH = 6.0) and isolates were grown in the dark for 5-7 d at

31°C. Conidia were swabbed with sterile cotton swabs and transferred into glass vials with Teflon septa containing 10 ml sterile ultrapure water. Conidial concentrations were estimated according to Mehl and Cotty (2012), and the final concentration of each suspension was adjusted to 106 conidia/ml.

Liquid fermentation assays and assessment of aflatoxin production

Three different liquid media were used to evaluate aflatoxin production by fungal isolates in the current study: Yeast extract and sucrose (YES) medium

(Davis et al., 1966) and Adye and Mateles (A&M) medium (Mateles and Adye,

1965) amended with either 22.5 mM ammonium sulfate ((NH4)2SO4) or 22.5 mM urea as sole nitrogen source (Table 3.1). Urea was filter sterilized and added aseptically to autoclaved media, while ammonium sulfate was added prior to autoclaving the medium (Probst and Cotty, 2012). Aflatoxin production by A. aflatoxiformans and A. minisclerotigenes was initially evaluated in YES medium

(pH = 6.5). Erlenmeyer flasks containing 70 ml of the medium were aseptically seeded with conidial suspensions (106 conidia/ml of the suspension; 100µl/flask) and incubated in the dark for 7 d at 31°C with agitation. Fungal cells were lysed with acetone (70 ml acetone per 70 ml fermentation) at the end of the fermentation period by swirling the flasks to allow mixing, after which cultures were allowed to sit for at least an hour. Acetone extracts were directly spotted onto thin layer chromatography (TLC) plates (Silica gel 60, EMD, Darmstadt, Germany) and separated adjacent to aflatoxin standards (Aflatoxin Mix Kit-M Supelco, Bellefonte,

PA) with Ether:Methanol:Water (96:3:1). Aflatoxins were directly quantified on TLC plates by scanning florescence densitometry under 365 nm UV light (TLC Scanner

3, Camag Scientific Inc., Wilmington, NC). Fungal mycelia from the fermentation were filtered through Whatman no.4 filter paper using vacuum filtration, dried

(40°C, 48 h) and weighed. Filtrates free of mycelia and initially negative for aflatoxins were partitioned twice with dichloromethane and concentrated prior to quantification as previously described (Cardwell and Cotty, 2002).

Based on results from initial screening of isolates in the YES medium, four representative isolates of A. aflatoxiformans and A. minisclerotigenes were chosen for evaluation of aflatoxin production in YES, A&M with ammonium sulfate, and

A&M with urea media. Fungal isolates NRRL A-11612 and NRRL A-11611 from

Nigerian groundnuts (Hesseltine et al., 1970) were used as reference isolates of

A. aflatoxiformans and A. minisclerotigenes, respectively. The remaining six isolates were chosen to represent isolates recovered from chilies sampled from different locations in Nigeria. Inoculum for each of the eight isolates was prepared

84 as described above. Isolates were inoculated aseptically into each of the three media (100 µl of 106 conidia/ml suspension per 70 ml). All media were adjusted to pH = 4.75 before autoclaving. Fermentations were carried out for 7 d in the dark at

31°C. At the end of the incubation period, pH was measured and the experiment was terminated by the addition of 70 ml acetone (50% acetone vol/vol). Aflatoxins were extracted, concentrated and quantified as described above. Each treatment was replicated four times and the experiment was performed twice.

In order to test the effect of sucrose concentration in the YES medium on aflatoxin production by members of A. aflatoxiformans and A. minisclerotigenes, isolates were inoculated into YES medium (pH = 4.75) containing 5%, 10%, 15%, and 20% sucrose. Isolates were replicated thrice and incubated with and without agitation for 3 d at 31°C in the dark. Total aflatoxins, mycelial mass, and pH were measured at the end of the incubation period as described above.

A microbiological assay utilizing YES and A&M with urea media was designed to differentiate A. aflatoxiformans and A. minisclerotigenes. Eleven isolates of A. aflatoxiformans and A. minisclerotigenes were evaluated for aflatoxin production under shaking and stationary conditions for 3 d at 31°C. Fungal isolates were selected such that isolates were representative of location and year of sampling. Aflatoxin concentrations, biomass production and pH were estimated at the end of the incubation period. Ratios of aflatoxin concentrations produced in

A&M medium with urea to that of YES were calculated.

Aflatoxin production in maize

Fungal isolates evaluated for aflatoxin production in liquid fermentations were also assessed for aflatoxin production in maize (Zea mays). Healthy, undamaged maize kernels adjusted to 25% moisture were autoclaved in

Erlenmeyer flasks (10 g per flask) for 20 min at 121°C. Aflatoxin production on autoclaved maize is a good predictor of aflatoxin production in viable maize kernels

(Probst and Cotty, 2012). Moisture content of maize before and after autoclaving was measured with an HB43 Halogen Moisture Analyzer (Mettler Toledo,

Columbus, Ohio, USA). Maize was inoculated with 100 µl of conidial suspensions

(106 condia/ml), adjusted to 30% moisture with sterile, ultrapure water and incubated for 7 d at 25°C, 30°C, 35°C and 40°C in the dark. Each treatment was replicated thrice. At the end of the incubation period, maize-fungal cultures were ground in 50 ml of 85% acetone in a laboratory grade Waring Blender (seven- speed laboratory blender, Waring Laboratory, Torrington, CT) at full speed for 30 s. Aflatoxins were extracted and quantified as previously reported (Singh et al.,

2018).

Data analysis

Aflatoxin concentrations and fungal biomass were measured in µg/g and g, respectively. Aflatoxins produced by individual isolates, pH at the end of incubation and fungal biomass were analyzed using Analysis of Variance as implemented in

JMP 11.1.1 (SAS Institute, Cary, NC, USA, 2013). Means were separated using

p = 0.05). Differences in mean aflatoxin concentrations, pH and

-test (p = 0.05).

Data were tested for normality and, if required, log transformed before analysis.

True means are presented for clarity.

Results and Discussion

Aspergillus aflatoxiformans and A. minisclerotigenes differ in aflatoxin production in Yeast Extract and Sucrose medium

Aflatoxin production by isolates of A. aflatoxiformans (n = 45) and A. minisclerotigenes (n = 44) was initially assessed by growth in YES medium (pH =

6.5) at 31 °C for 7 days. This medium supported production of low levels of aflatoxins by A. aflatoxiformans (Mean = 1.9 total aflatoxin g-1 mycelia; range =

0.02 to 5.65 µg/g), whereas A. minisclerotigenes produced high concentrations of total aflatoxins (Mean = 89 µg/g; range = 32 to 460 µg/g). Fourteen isolates from each species were re-evaluated in Adye & Mateles (A&M) medium supplemented with urea as the sole nitrogen source. In contrast to the above results, fungal isolates from both species produced high and similar concentrations of total aflatoxins (p = 0.64) in A&M medium with urea (Mean = 507 µg/g). These results clearly indicated that aflatoxin production by A. aflatoxiformans is inhibited in YES medium, a phenotype that can be utilized to differentiate the two species that are otherwise phenotypically similar.

Aflatoxin production by A. aflatoxiformans and A. minisclerotigenes in liquid fermentations

To assess variation in aflatoxin producing ability of A. aflatoxiformans and

A. minisclerotigenes in liquid media, four isolates representative of each species were assayed in three separate media, (i) YES, (ii) A&M supplemented with urea,

87 and (iii) A&M containing (NH4)2SO4 (Table 3.1). Each of these media supports aflatoxin production by fungi with L or S morphology within section Flavi (Davis et al., 1966; Cotty and Cardwell, 1999; Probst and Cotty, 2012). The A&M medium with different modifications used in the current study has been utilized previously to assess aflatoxigenicity of isolates from soils and crops and to study the etiology of aflatoxin contamination (Cotty, 1997; Cardwell and Cotty, 2002; Atehnkeng et al., 2008). Aspergillus aflatoxiformans and A. minisclerotigenes produced similar concentrations of aflatoxins B1 and G1 in the A&M medium with urea (p > 0.05), and individual isolates did not differ (Table 3.2; p > 0.05). Production of aflatoxins

B1 and G1 in A&M media containing (NH4)2SO4 as the sole nitrogen source differed among isolates (Table 3.2; p < 0.001). Aspergillus minisclerotigenes produced significantly higher concentrations of aflatoxin G1 (p = 0.039) in A&M medium supplemented with (NH4)2SO4, but average aflatoxin B1 concentrations were similar for both species (Table 3.2). Aspergillus aflatoxiformans produced 7.5 times more aflatoxin B1 and 31 times more aflatoxin G1 on average in A&M medium containing urea compared to that with (NH4)2SO4 (Table 3.2). Aspergillus minisclerotigenes produced 3.7 times more aflatoxin B1 and nearly 8 times more aflatoxin G1 in the medium supplemented with urea versus (NH4)2SO4 (Table 3.2).

These results are in agreement with Cotty and Cardwell (1999) who reported increased production of aflatoxins B1 and G1 by African SBG isolates in A&M medium containing urea compared to the medium supplemented with (NH 4)2SO4 as the sole nitrogen source.

Production of both aflatoxins B1 and G1 by A. aflatoxiformans was significantly lower in the YES medium (pH = 4.75) compared to that of A. minisclerotigenes (Table 3.2; p < 0.001), as also observed during initial evaluation of these fungi in the YES medium (pH = 6.5). Isolates of A. minisclerotigenes produced at least 50 times more aflatoxin B1 and 25 times more aflatoxin G1 in the

YES medium compared to that by isolates of A. aflatoxiformans (Table 3.2). Since the YES medium contained higher concentrations of sucrose (15%) compared to either of the A&M media (5%), the effect of sucrose concentration on aflatoxin formation was tested in the YES medium under shaking and stationary conditions.

Total aflatoxin production by A. aflatoxiformans was inhibited to a non-detectable level in the YES medium irrespective of sucrose concentration from 5-20% under shaking and stationary conditions at 31°C (Table 3.3). Aflatoxin production, pH and fungal biomass were higher when cultures were stationary versus shaking. pH modification by A. aflatoxiformans and A. minisclerotigenes

All eight fungal isolates modified the pH during growth in each medium tested. Medium composition influenced the extent to which pH changed, and compared to an initial pH of 4.75, pH was modified by the end of fermentation in the decreasing order of YES < A&M with urea < A&M with ammonium. Although the final pH of A&M medium with urea among individual isolates was significantly different (Table 3.2; p < 0.001), the average final pH for A. aflatoxiformans and A. minisclerotigenes did not differ (Table 3.2; p = 0.83). The final pH of YES medium differed both among individual isolates (p < 0.001) and between the two species

(p < 0.0024). It is noteworthy that aflatoxin production did not depend on the initial

89 pH of the YES medium because aflatoxin production in YES fermentations at either pH = 6.5 (initial aflatoxin screen assay) or at pH = 4.75 (Table 3.2) was inhibited for A. aflatoxiformans whereas A. minisclerotigenes produced high concentrations of aflatoxins in YES medium at either pH. A&M medium containing (NH4)2SO4 was the most acidic at the end of the fermentation but the final pH did not differ among isolates (p = 0.498) or between species (p = 0.41). Similar results were reported by Cotty and Cardwell (1999) and Probst and Cotty (2012) in terms of pH modification of the A&M media during evaluation of fungal isolates for aflatoxin production; the A&M medium containing (NH4)2SO4 was more acidic by the end of the fermentation period in comparison to the A&M medium supplemented with urea or the YES medium.

Growth of A. aflatoxiformans and A. minisclerotigenes in liquid fermentations

All isolates, irrespective of species, produced the highest biomass in YES medium, and mycelial mass did not differ among isolates (Table 3.2; p = 0.89) or between species (p = 0.33). Differences were detected in biomass production in

A&M medium with (NH4)2SO4 among isolates (p < 0.05) but not between species.

Fungal growth was significantly different among isolates (p < 0.0001) and between species in A&M medium containing urea; A. minisclerotigenes produced higher biomass in this medium (Table 3.2; p < 0.0001). Concentrations of aflatoxins B1 and G1 produced in each medium did not depend on fungal growth and biomass production. Although A. aflatoxiformans produced the lowest concentrations of aflatoxins in YES medium, it produced the greatest mycelial mass in this medium,

90 and its mycelial mass was comparable to that of A. minisclerotigenes indicating that decreased production of aflatoxins by A. aflatoxiformans was not due to a reduction in biomass (Table 3.2).

Susceptibility of maize to high levels of aflatoxin contamination by A. aflatoxiformans and A. minisclerotigenes

In order to assess the toxic potential of A. aflatoxiformans and A. minisclerotigenes under different environmental conditions, aflatoxin producing ability of the two species was further evaluated on maize at 25°C, 30°C, 35°C, and

40°C (Table 3.4). Maize is an important staple in Nigeria and annual production exceeds 10 million metric tons every year (FAOSTAT, 2017). It is estimated that

Nigerians may be exposed to ~5.0 mg aflatoxin person-1 year-1 through maize consumption (Bandyopadhyay et al., 2007). Temperatures were chosen based on climate data from maize growing regions in Nigeria (Climates to travel, 2019; World data atlas-Knoema, 2019), and temperatures typically encountered during storage.

Overall, aflatoxin production by members of both species was high at 25°C, 30°C and 35°C, and A. aflatoxiformans produced significantly higher concentrations of total aflatoxins at each of these temperatures (Table 3.4; p < 0.001). Furthermore, isolates differed in aflatoxin producing potential at 25°C, 30°C and 35°C (p <

0.001), and certain individuals of A. minisclerotigenes produced similar concentrations of aflatoxins to those of A. aflatoxiformans (Table 3.4). Highest concentrations of aflatoxins were observed at 30°C followed by 35°C and 25°C by each species. Notably, although fungal isolates produced the lowest concentrations of aflatoxins at 25°C (Range: 13.0-154 µg/g), these are

91 unacceptable and dangerous concentrations for human and animal consumption.

Neither A. aflatoxiformans nor A. minisclerotigenes produced detectable concentrations of aflatoxins at 40°C (LOD = 0.42 µg/g of maize grain). Due to the ability of A. aflatoxiformans to produce higher concentrations of aflatoxins in maize compared to A. minisclerotigenes, occurrence of this species even at low frequencies can result in unacceptable crop contamination. Nevertheless, the toxic potential of A. minisclerotigenes can render this species dangerous in terms of crop contamination. It is therefore important to develop methods/techniques that can identify the most important etiologic agent of crop contamination when mixed infections by many fungal species occur.

Assay to distinguish A. aflatoxiformans and A. minisclerotigenes

We tested the utility of YES medium to reliably differentiate A. aflatoxiformans from A. minisclerotigenes by evaluating total aflatoxin production in this medium by 11 representative isolates of each of the two species with and without agitation for 3 d at 31°C. Different sets of isolates were used in this assay compared to what were used in the previous experiment in order to validate the observed phenotype of inhibited aflatoxin production in YES medium by multiple members of A. aflatoxiformans. Reference isolates of both species were included.

In addition to shaking, stationary conditions were included to ensure reliable identification of A. aflatoxiformans and A. minisclerotigenes in laboratories where incubators with shakers may not be available. Results were similar to what was previously observed with isolates of A. minisclerotigenes producing higher concentrations of total aflatoxins in YES medium than A. aflatoxiformans under

92 shaking (17.0 to 300 µg/g) or stationary (25.4 to 1,052 µg/g) conditions (Table 3.5).

These same isolates were further tested for aflatoxin production in A&M medium with urea as the sole nitrogen source, and ratios of total aflatoxins produced by each isolate in the two media were estimated. Overall, aflatoxin concentrations were higher for all cultures when grown under stationary conditions irrespective of species and medium (Table 3.5). Isolates belonging to A. aflatoxiformans produced at least 122 and 124 times more total aflatoxins in A&M medium with urea versus YES medium, with and without agitation, respectively. However, ratios of aflatoxin production by A. minisclerotigenes were in the range of 0.86-13.3 with agitation and 0.23-11.9 when stationary. Based on these results, fungal isolates with an SBG morphology can be identified as A. aflatoxiformans when ratios of aflatoxin concentrations in A&M medium with urea versus YES medium are greater than 80, and as A. minisclerotigenes when ratios are less than 80, with or without agitation. This ratio is expected to include any potential outlier isolates from each species based on results in Table 3.2 & 3.5.

Conclusion

Aspergillus aflatoxiformans and A. minisclerotigenes have highly similar morphologies on various culture media, overlapping secondary metabolite profiles, and production of high concentrations of aflatoxins in synthetic media and on crops

(Cotty and Cardwell, 1999; Pildain et al., 2008; Singh and Cotty, 2019). Aflatoxin contamination of crops in West Africa has often been attributed to A. aflatoxiformans due to the aflatoxin producing ability of this species and its frequency of occurrence in agricultural soils and crops (Agbetiameh et al., 2018;

Cardwell and Cotty, 2002; Donner et al., 2009). Only a single isolate of A. minisclerotigenes was known from West Africa until recently (Hesseltine et al.,

1970; Pildain et al., 2008), and the species was thought to be more common in

Eastern and Southern Africa (Probst et al., 2014). A recent study investigating etiology of aflatoxin contamination of dried red chilies produced in Nigeria reported

A. minisclerotigenes in Nigerian chilies in high numbers (8% of all Aspergillus section Flavi) (Singh and Cotty, 2019), confirming the co-occurrence of these two genetically distinct species in West Africa. Aspergillus minisclerotigenes was first described a decade ago by Pildain et al., (2008) who reported that only DNA sequences could separate A. minisclerotigenes and A. aflatoxiformans (referred to as A. parvisclerotigenus in Pildain et al., 2008). Both A. minisclerotigenes and A. aflatoxiformans can contaminate important crops such as maize, groundnuts, and chilies with unacceptable concentrations of aflatoxins (Singh and Cotty, 2019).

Furthermore, both species produced high concentrations of aflatoxins at 25°C, 30

°C and 35°C on maize (Table 3.4), indicating their toxic potential at different temperatures which are prevalent both pre- and post-harvest. More than 99% of the human population in several areas of West Africa suffers chronic exposure to aflatoxins due to consumption of contaminated cereals (Gong et al., 2002;

Bandyopadhyay et al., 2007). These facts clearly indicate the need for reliable detection and identification of etiologic agents of aflatoxin contamination of crops in West Africa, which may be complicated by co-infection with species such as A. minisclerotigenes and A. aflatoxiformans that share several phenotypic characteristics. This can be especially challenging in regions of West Africa where

DNA-based technologies and the necessary technical expertise may not be easily available or accessible. The microbiological assay in the current study based on inhibition of aflatoxin production by A. aflatoxiformans in YES medium is a simple and reliable tool to distinguish the two species that are otherwise indistinguishable in the absence of DNA-based methods. Both stationary incubation and fermentation with agitation can be utilized to differentiate isolates of A. aflatoxiformans from that of A. minisclerotigenes recovered from soils or crops within 72 hours.

TABLE 3.1 Composition of liquid media.

Medium Composition (liter-1) 20 g Yeast extract, 150 g Sucrose, 1 ml micronutrient YES solution, pH 6.5 or 4.75 depending on the experiment 50 g Sucrose, 10 g KH PO , 2 g MgSO 7H O, 1 ml A&M 2 4 4 2 micronutrient solution, pH 4.75

0.11 g MnSO H O, 0.5 g (NH ) Mo O 4H O, 17.6 g Micronutrient solution 4 2 4 6 7 24 2 ZnSO4 7H2O, 0.7 g Na2B4O7 10H2O, 0.3 g CuSO4 5H2O

AB A B AB B AB AB B

4 0.74 0.83 0.67 0.74 0.71 0.76 0.73 0.69

0.75 0.72 CD BC D BCD AB BC AB A y x 0.66 0.74 0.63 0.70 0.77 0.75

0.96 0.93 0.81 1.02 0.76 0.86

E Ue NH Urea YES 4 in liquid fermentations in liquid 2.22 2.23 2.23 2.22 2.21 0.78 2.19 0.87 0.85 2.18 0.85 2.25

ABC ABC BCD BCD CD AB A D 3.74 3.75 3.62 3.63 3.59 3.78 3.51 3.83 .9 .2 .3 0.69 0.93 0.79 2.22 0.83 3.69 2.21 3.68 B C C C A B B C y x 4.71 4.22 4.34 4.43 5.05 4.74 . . 4.74 4.34 A.minisclerotigenes

4.43 4.72

BCD BC D AB CD BCD y x E Ue NH Urea YES A and 4 BCD 9.9 15.1 21.0 3.32 33.3 5.78 (µg/g) Final pH Mycelia (g) Mycelia pH Final (µg/g) 1

803 357 280 104 129 370 97

4 7.91 344 8 12.3 386 A.minisclerotigenes 6 121 469 2 41.9 328 C C C C AB AB B y A x 1.81 1.13 0.67 1.06 46.7 48.7 AM-

A. aflatoxiformans

> 0.05). >0.05). p and by by

45.3 1

363 BC BC C BC E Ue NH Urea YES ABC AB AB A 4 52.9 132 119 76.2 104 24.8 1.17 126 94.8 121 µ/) AflatoxinG (µg/g) and G

1 1 1

, final pH , offinal medium, the liquid massmycelial and by individual produced isolates 1 1,265 710 383 496 292 288

, , G

20 27.3 250 1 713 91 327 951 445 C C C C B y B B A x A. aflatoxiformans, A.aflatoxiformans, 1.50 2.07 1.64 1.85 103 108 1.77 232 AflatoxinB

11612 11611 - - Isolate Isolate CHL568 CHL740 CHL877 CHL603 A A

Production B of aflatoxins E Ue NH Urea YES

AA AA AM AM AM Average Average Species 95.2 CHL707 623 CHL845 Species assignment; AA- AA- assignment; Species TABLE 3.2 TABLE ammonium or urea A&Mmediumwith either YESsulfate. or supplemented containing B Concentrations of aflatoxins same by followed Means replicates. four of means are Values tested. medium each for column by compared were upper- # and and A. aflatoxiformans A.aflatoxiformans 98 >0.05). p concentrations, final pH and mycelia produced and finalpH by concentrations, produced mycelia 1 and G 1 < 0.01). <0.01). p p -test, -test, were compared for each medium tested. Means followed by different lower-case letters in bold letters in lower-case different by followed Means each mediumtested. for werecompared Differences in average aflatoxin B average aflatoxin in Differences A.minisclerotigenes not ( ado within differ column a lacking letter Values

(AM)

1.07 0.77 1.22 20%

(g) c 1.05 0.76 1.19 15%

0.89 0.76 0.99 Mycelia A.minisclerotigenes

% 10% 5% and and 0.78 0.72 0.85

(AA) 20% 99

15% b

pH .6 .7 4.51 4.37 4.85 4.46 4.82 4.68 5.23 5.00 5.05

A. aflatoxiformans A.aflatoxiformans

% 10% 5% 4.68 5.02 5.24 With Agitation With

Agitation Without

20%

(µg/g) a 15%

42 67 20.1 36.7 14.2

Total aflatoxins Total D D D ND ND 10% 5% ND ND ND ND ND ND 12.7

Production of total aflatoxins by of Production total

Values are averages of three replicates. three replicates. are averages of Values AA AA M3. 507. 2451 .649 .408 .909 0.98 0.97 0.89 0.82 5.04 4.91 4.86 5.14 42.4 AM 73.8 45.0 36.0 AM Species in the Yeast Extract and Sucrose medium with different sucrose concentrations during fermentation. during concentrations sucrose Sucrose medium with different and Extract the Yeast in Table 3.3 Table mycelia. µg per aflatoxin gramLOD-1.7 of Detectable; Not ND- a,b,c Table 3.4 Aflatoxin production by A. aflatoxiformans and A. minisclerotigenes in maize at different temperatures.

Total aflatoxin (µg/g) Species Isolate 25°C 30°C 35°C A. aflatoxiformans A-11612 128a 573abc 283a CHL568 154a 655ab 544a CHL740 146a 612ab 541a CHL877 84a 756a 584a A. minisclerotigenes A-11611 14.4b 34.6e 17.9c CHL707 13.0b 67.2de 27.9c CHL845 26.9b 165cd 90.0b CHL603 88.8a 196bcd 30.7c Average A. aflatoxiformans 128A 649A 488A

Average A. minisclerotigenes 35.8B 116B 41.6B Concentrations of total aflatoxins produced by individual isolates were compared by column for each temperature. Values are means of three replicates. Means followed by different lower- p < 0.001)

Differences in aflatoxin concentrations produced by A. aflatoxiformans and A. minisclerotigenes are indicated by bold upper-case letters ( -test, p <

0.001).

100

TABLE 3.5 Total aflatoxin (AF) production by representative isolates of A. aflatoxiformans (AA) and A. minisclerotigenes (AM) in A&M medium with urea and

YES liquid fermentations, with and without agitation, and their ratios.

With Agitation Without Agitation

Species Isolate Total AF (µg/g) Total AF (µg/g) Ratioa Ratiob

Urea YES Urea YES

AA A-11612 3,929 1.14 3,447 2,423 3.99 608 CHL514 831 1.75 475 1,556 8.15 191 CHL562 749 1.42 527 2,636 5.42 486 CHL596 278 1.00 278 188 0.26 732 CHL633 490 0.63 777 1,783 12.6 142 CHL675 2,008 16.5 122 8,902 9.09 980 CHL731 907 1.38 659 2,237 2.99 748 CHL812 2,634 2.40 1,099 3,382 27.3 124 CHL819 1,600 1.46 1,100 1,563 0.68 2296 CHL856 1,554 2.36 658 1,710 4.19 408 CHL878 923 0.79 1,170 937 3.32 282

AM A-11611 93.4 17.0 5.48 59.3 87.2 0.68 CHL583 208 20.6 10.1 152 31.7 4.78 CHL621 218 73.9 2.94 81.0 346 0.23 CHL636 260 300 0.86 347 1,052 0.33 CHL644 45.2 46.8 0.97 71.1 56.4 1.26 CHL661 440 33.2 13.3 453 37.9 12.0 CHL690 581 76.5 7.59 811 74.5 10.9 CHL799 147 19.6 7.47 226 25.4 8.89 CHL847 504 67.0 7.52 697 100 6.96 CHL895 139 30.3 4.58 190 98.5 1.93 CHL947 98.3 83.6 1.18 231 76.2 3.03

Ratios of total aflatoxin produced in the A&M medium with urea to that in the YES medium with agitationa and under stationary conditionsb.

101

APPENDIX D - Aspergillus texensis: A novel aflatoxin producer with S morphology from the United States

Published in: TOXINS (2018), Vol. 10(12), p. 513

Abstract

Aflatoxins are carcinogenic metabolites produced primarily by fungi within

Aspergillus section Flavi. These fungi infect a wide range of crops in warm regions.

Molecular phylogenetic analyses of fungi with S morphology (average sclerotium size < 400 µm) within section Flavi collected from across the United States (US) resulted in the discovery of a novel aflatoxin-producing species, Aspergillus texensis. Aspergillus texensis was isolated from maize grown in Arkansas,

Louisiana, and Texas, and from soils cropped to maize in Texas. Aspergillus texensis produces sparse conidia and abundant sclerotia on various culture media, and on maize. Physiological studies have revealed optimal growth on culture media at 35°C. All isolates of A. texensis produced B and G aflatoxins, cyclopiazonic acid and aspergillic acid. Aspergillus texensis and A. flavus S strain morphotypes produced similar concentrations of total aflatoxins on maize (p >

0.05). Phylogenetic analyses of aflatoxin-producers based on partial gene

-tubulin (0.9 kb), calmodulin (1.2 kb), and nitrate reductase (2.1 kb) genes placed A. texensis in a highly supported monophyletic clade closely related to A. minisclerotigenes and a previously reported unnamed lineage designated Lethal Aflatoxicosis Fungus.

102

Introduction

Aflatoxins (AF) are extremely potent naturally-occurring hepatocarcinogenic mycotoxins produced by several members of the genus Aspergillus on important commodities such as maize, groundnut, tree nuts, spices, and cottonseed (Cotty,

1997; Donner et al., 2015; Kachapulula et al., 2017; Probst et al., 2014; Singh and

Cotty, 2017). Naturally occurring aflatoxin-producers contaminate food and feed with four major aflatoxins, i.e., aflatoxins B1, B2, G1 and G2. Bio-transformation of aflatoxins from consumption of contaminated food results in formation of aflatoxins

M1 and M2, which are secreted into milk (De Iongh et al., 1964; Holzapfel et al.,

1966). Although certain species within Aspergillus section Ochraceorosei and

Aspergillus section Nidulantes also produce aflatoxins (Frisvad et al., 2004; Klich et al., 2000), the most economically important aflatoxin-producers belong to

Aspergillus section Flavi (Cotty, 1996; Ehrlich et al., 2007; Singh and Cotty, 2019).

Aspergillus flavus, A. parasiticus, A. aflatoxiformans from West Africa, and an unnamed lineage referred to as the Lethal Aflatoxicosis Fungus (LAF) from Kenya are particularly notorious members of Aspergillus section Flavi responsible for contamination of crops with high levels of aflatoxins (Cardwell and Cotty, 2002;

Cotty et al., 2008; Donner et al., 2009; Frisvad et al., 2019; Kachapulula et al.,

2017; Probst et al., 2012). These fungi colonize a wide range of host crops resulting in dangerous concentrations of aflatoxins under conducive environmental conditions (high temperature, i.e., >28°C, humidity and plant stress) (Cotty and

Jaime-Garcia, 2007).

103

Aflatoxins are both health and economic threats. Aflatoxin B1 has been categorized as a Group 1 human carcinogen by IARC (IARC, 2002). In developing nations, aflatoxin regulations largely remain unenforced and crops are consumed without monitoring, resulting in frequent exposure of humans and animals

(Shephard, 2003; Williams et al., 2004). Chronic dietary exposure to sub-lethal concentrations can cause immune suppression (Turner et al., 2003), impaired growth (Gong et al., 2004; Khlangwiset et al., 2011), and liver cancer (Liu and Wu,

2010; Wang and Tang, 2004). Ingestion of high concentrations of aflatoxins may result in liver necrosis followed by rapid death (Shephard, 2003; Williams et al.,

2004). Acute cases of aflatoxin poisoning have resulted in human deaths in Kenya and Tanzania (Africa A, 2016; CDC, 2004; Outbreak News Today, 2017).

Aflatoxins are a severe economic burden in the developed world where regulations are stringently enforced leading to heavy economic losses incurred by growers

(Robens and Cardwell, 2003). These regulatory enforcements result in rejection of contaminated food/feed followed by loss of markets and expense to the exporter.

Since aflatoxins remain a global concern, it becomes crucial to identify and characterize aflatoxin-producing species in order to design management strategies for aflatoxin contamination of crops.

Aspergillus flavus and A. parasiticus are the most commonly implicated causal agents of aflatoxin contamination of crops (Cotty et al., 1994). The filamentous fungus A. flavus produces only B aflatoxins and has two morphotypes. The L strain morphotype is characterized by the production of copious conidia, few large

104 sclerotia (> 400 µm), and variable concentrations of aflatoxins, such that many L strain morphotype fungi are atoxigenic (do not produce aflatoxins); the S strain morphotype produces large quantities of small sclerotia (< 400 µm) and sparse conidia (Cotty, 1989). The A. flavus S strain morphotype consistently produces high concentrations of aflatoxins. Aspergillus parasiticus produces both B and G aflatoxins. The A. flavus S strain morphotype produces only B aflatoxins and is the more commonly occurring S morphology fungus in North America (Cotty, 1989;

Donner et al., 2015; Jaime-Garcia and Cotty, 2010). However, several highly aflatoxigenic S morphology fungi belonging to genetically distinct taxa, i.e., A. minisclerotigenes, A. aflatoxiformans, A. cerealis (previously A. korhogoensis), and LAF, are known from Sub-Saharan Africa (Cardwell and Cotty, 2002; Carvajal-

Campos et al., 2017, Donner et al., 2009; Frisvad et al., 2019; Probst et al., 2014).

During phylogenetic analysis of fungi with S morphology belonging to Aspergillus section Flavi isolated from soils and maize collected from across the US, we discovered fungal isolates that produced both B and G aflatoxins but were morphologically indistinguishable from the A. flavus S strain morphotype.

Phylogenetic reconstruction using multi locus gene sequences with previously described members of Aspergillus section Flavi indicated that these B and G aflatoxin producers represent a novel, undescribed species.

Aspergillus section Flavi contains several phylogenetically distinct species with

S morphology characterized by production of small sclerotia (< 400 µm) (Cotty,

1989; Frisvad et al., 2019; Probst et al., 2012). Description of species within

105 section Flavi solely based on phenotypic characteristics can be erroneous due to overlapping character states (Peterson et al., 2001; Ito et al., 2001). The current study used a polyphasic approach to compare newly discovered species to those previously described. Phenotypic description included macro- and micromorphology, growth at different temperatures, and production of aflatoxins, aspergillic acid, and cyclopiazonic acid. Phylogenetic reconstructions based on multiple unlinked loci were utilized to determine relationships of the novel species to those previously described with S morphology.

Materials and Methods

Fungal isolates and morphology

Eleven fungal isolates were chosen from a survey of S morphology fungi, collected from across the US, based on their ability to produce both B and G aflatoxins. These isolates were recovered from soil and maize samples at the

USDA-ARS laboratory in Tucson, Arizona. Fungal isolates with known phylogenetic placement were obtained from the ARS Culture Collection, Peoria,

IL, USA (NRRL in Table 4.1), the American Type Culture Collection, Manassas,

USA (ATCC in Table 4.1), the Fungal Genetics Stock Center, Manhattan, KS, USA

(A in Table 4.1), or the USDA-ARS Laboratory Culture Collection in Tucson,

Arizona (Table 4.1).

In order to compare morphological features and growth, fungal isolates were grown for 7 d in the dark at 5°C, 10°C, 15°C, 20°C, 25°C, 30°C, 35°C, 37°C, 40°C, and 42°C using center point inoculation on Czapek agar (20 g Bacto agar, 30 g

106 sucrose, 3 g NaNO3, 0.5 g KH2PO4, 0.5 g K2HPO4, 0.5 g MgSO4 * 7H2O, 0.5 g KCl, and 1 ml micronutrients per liter of deionized distilled water, pH = 6.0), Czapek with

2% NaCl, Czapek yeast extract agar (CYA) (Czapek agar with 0.5% yeast extract),

Malt agar and V8 agar (5% V8 juice and 2% agar, pH = 6.0). Micronutrients were prepared according to Adye and Mateles (1964). Colony diameters were recorded

(four replicates per isolate) at each temperature. For micromorphological observations, 3 d old cultures grown on Czapek agar were viewed and captured with a differential interference contrast microscope (Model BH-2, Olympus,

Shinjuku, Tokyo, Japan) equipped with an OMAX 5.0MP USB2.0 digital camera and the software package ToupView (v 3.7, ToupTek Photonics, Hangzhou, China,

2014).

Production of aspergillic acid, aflatoxins, and cyclopiazonic acid

Aspergillic acid production was analyzed by inoculating fungal isolates onto

Aspergillus flavus and parasiticus agar (AFPA) (Pitt et al., 1983), which was incubated for 7 d in the dark at 25°C, 30°C, and 35°C. Isolates were replicated four times at each temperature.

Aflatoxins were analyzed and quantified for fungi with S morphology belonging to five phylogenetically distinct taxa. Each taxon was represented by three isolates.

Conidial suspensions were inoculated onto 10 g autoclaved maize (Pioneer hybrid

N82VGT), as previously described (Probst and Cotty, 2012). Briefly, single spored isolates were seeded at the center of V8 agar plates and incubated for 7 d at 31°C.

Conidia were swabbed from plate s with sterile cotton swabs into sterile distilled

107 water (10 ml). The quantity of conidia was measured using a turbidity meter

(Turbidimeter, Orbeco Analytical Systems) and maize was inoculated with 10 6 conidia/ml of each fungal isolate. Water content of maize was adjusted to 30%.

Maize cultures were incubated at 31°C for 7 d in the dark to allow fungal growth and aflatoxin formation. The experiment was terminated by addition of 50 ml 85% acetone and crop cultures were ground to homogeneity in a laboratory grade

Waring Blender (seven-speed laboratory blender, Waring Laboratory, Torrington,

CT, USA) at full speed for 30 s. The culture filtrate was spotted directly onto thin- layer chromatography (TLC) plates (Silica gel 60, EMD, Darmstadt, Germany) adjacent to aflatoxin standards (Aflatoxin Mix Kit-M, Supelco, Bellefonte, PA, USA) containing a mixture of known concentrations of aflatoxins B1, B2, G1, and G2.

Plates were developed in ethyl ether-methanol-water (96:3:1), air-dried, and aflatoxins were visualized under 365-nm UV light. Total aflatoxins were quantified directly on TLC plates with a scanning densitometer (TLC Scanner 3, Camag

Scientific Inc., Wilmington, NC, USA). Treatments were replicated four times and each experiment was performed twice.

Cyclopiazonic acid analyses were performed according to (Lansden, 1986).

Conidial suspensions were inoculated onto 20 g of autoclaved maize using the method described above. For CPA analyses, inoculated maize was incubated for

7 d at 31°C in the dark, and the experiment was terminated by addition of 100 ml of extraction solvent (20% methanol, 80% chloroform and 0.2% of 85% phosphoric acid). Maize-fungal cultures were ground at full speed to homogeneity in a

108 laboratory grade Waring Blender. The slurry was transferred to polypropylene containers and shaken for 60 min at 200 rpm on a rotary shaker (HS501digital, IKA

Works Inc., Wilmington, NC, USA). Extracts were filtered through Whatman No. 4 filter paper into 100 ml cylinders and the volume of extract was recorded. Each extract was transferred into a separatory funnel followed by addition of 100 ml of

0.5 N NaHCO3 with 3% (w/v) NaCl and gently shaken. The mixture was allowed to settle for 30 min and the bottom layer was discarded. Seven ml of concentrated

HCl was added dropwise to each sample and gently shaken for 30 s. Once the bubbling subsided, samples were extracted twice with 25 ml of chloroform.

Extracts were combined, dried, and re-suspended in 4 ml of chloroform. Extracts were spotted directly on TLC plates alongside a CPA standard of known concentration. Plates were dried at 50°C for 15 min and then cooled for 1 min.

Appearance of purple bands alongside the CPA standard indicated that samples were positive for CPA. Aspergillus flavus L strain isolate AF36, which is a registered biological control in the US, was used as the positive control (Chang et al., 2005). CPA was quantified on TLC plates by scanning fluorescence densitometry at 546 nm (TLC Scanner 3, Camag Scientific Inc., Wilmington, NC).

DNA extraction and PCR amplifications

Fungal isolates were grown and DNA extraction was performed as described previously (Callicott and Cotty, 2015). DNA concentration was adjusted to 5 ng/µl for PCR reactions. Partial gene fragments of -tubulin (benA) (0.9 kb), calmodulin

109

(cmdA) (1.2 kb), and nitrate reductase (niaD) (2.1 kb) genes were amplified (Table

4.2). The primer pair Bt3a-

-tubulin. Primers were designed based on genome sequence of A. nomius NRRL 13137 (GenBank accession no. JNOM01000216) using Primer3 version 0.4.0 (Koressaar and Remm, 2007; Untergasser et al.,

2012). PCR reactions were performed in 20 µl using 5 ng genomic DNA with a

PCR premix (AccuPower® HotStart, Bioneer, Alameda, CA, USA) on a MyCycler thermocycler (Bio-Rad Laboratories, Richmond, CA, USA) under the following

-tubulin, 5 min at 94°C followed by 35 cycles of 96°C for 30 s, locus-specific annealing temperature for 1 min (Table 4.2), 72°C for 1 min, and 5 min at 72°C; for cmdA and niaD genes, 5 min at 94°C followed by 38 cycles of

94°C for 20 s, locus-specific annealing temperature for 30 s (Table 4.2), 72°C for

1 min, and 5 min at 72°C. Amplicons were visualized with SYBR Gold after 1.0% agarose gel electrophoresis and sequenced at the University of Arizona Genetics

Core facility (UAGC, Tucson, AZ, USA) using primers mentioned in Table 4.2. For mating type analyses, portions of MAT1-1 and MAT1-2 were amplified and characterized from all 11 A. texensis isolates following [Ramirez-Prado et al., 2008] with slight modifications: 2 min at 95°C followed by 30 cycles of 94°C for 30 s,

54°C for 30 s, 72°C for 45 s, and 2 min at 72°C. Amplicons were visualized using a 1.5% agarose gel. DNA sequences of representative isolates are deposited at

Genbank (Table 4.3).

Molecular analysis and phylogenetics

110

Bidirectional sequences were used to create a consensus sequence of each amplicon. Gene segments were assembled with either two (benA and cmdA) or six (niaD) amplicons per gene, corrected manually and gene alignments were generated using the MUSCLE algorithm within Geneious Pro Version 7.1.9

(Biomatters Ltd., Auckland, New Zealand). Phylogenetic trees were generated for concatenated and individual gene sequences following Bayesian inference using

MrBayes version 3.2.6 (Huelsenbeck and Ronquist, 2001) and maximum likelihood (ML) analysis with PhyML at Phylogeny.fr (Dereeper et al., 2008;

Dereeper et al., 2010) to confirm tree topologies. Bayesian inference was conducted by running Markov Chain Monte Carlo analysis for up to 10 million generations. For ML analysis, datasets were bootstrapped with 500 replicates.

Trees were mid-point rooted using FigTree v.1.4.3 (Rambaut A, 2012). The presence of a complete norB-cypA sequence in the A. texensis aflatoxin biosynthesis gene cluster was confirmed with previously described primer sets

(Ehrlich et al., 2004).

Data analysis

Total aflatoxin was measured in µg/g. Aflatoxin concentrations produced by each isolate and each taxon were analyzed using Analysis of Variance as implemented in JMP 11.1.1 (SAS Institute, Cary, NC, USA, 2013). Means were

p = 0.05). Fungal colony diameters were measured in millimeters (mm). Diameters for each medium and temperature were

p = 0.05). Data were tested for

111 normality prior to statistical analyses and, if required, log-transformed. True means are presented for clarity.

Results

Molecular analyses and phylogenetics

Homologous DNA sequences of reference isolates of described species within section Flavi with S morphology (e.g., A. flavus S strain morphotype, A. minisclerotigenes, A. cerealis, A. aflatoxiformans), representatives of the fungi associated with lethal aflatoxicosis in Kenya (LAF), and other aflatoxin-producers

(A. nomius, A. pseudotamarii, A. parasiticus) were aligned with sequences of S morphology fungi recovered from the US (Table 4.1). Phylogenetic reconstruction using Bayesian Inference (BI) and Maximum Likelihood (ML) analyses from partial

-tubulin (benA, 0.9kb, chromosome 6), calmodulin (cmdA,

1.2kb, chromosome 2), and nitrate reductase (niaD, 2.1kb, chromosome 4) genes for individual and concatenated sequences yielded similar topologies. The new species, named Aspergillus texensis, occupied a highly supported monophyletic clade in concatenated benA, cmdA, and niaD gene phylogenies (Figure 4.1), and was sister to a clade containing A. minisclerotigenes and LAF fungi. Phylogenetic reconstruction based individually upon cmdA and niaD sequences showed that each of these genes was sufficient to resolve A. texensis into a monophyletic clade with high Bayesian posterior probability and bootstrap support (Figure 4.2 and 4.3).

The A. flavus S strain morphotype, which is a frequently reported S morphology fungus in the US, is phylogenetically distinct from A. texensis (Figure 4.1).

112

Additionally, molecular analysis of the norB-cypA region did not detect any deletions in the cypA gene of A. texensis, whereas the A. flavus S strain morphotype isolates contained either a 0.9 or 1.5kb deletion in this region of the aflatoxin biosynthesis cluster required for G aflatoxin production.

All isolates of A. texensis, regardless of state of origin, contained only the

MAT1-1 idiomorph at the mating-type locus, suggesting that A. texensis is heterothallic. Each isolate amplified only a single amplicon of approximately 390 bp in the MAT locus PCR assay. This is characteristic of the MAT1-1 idiomorph.

The MAT1-2 idiomorph that is highly conserved in Aspergillus section Flavi should produce a 270 bp amplicon, but was not detected in any A. texensis isolate.

Aspergillic acid

All isolates of A. texensis produced aspergillic acid, which was detected by the bright-orange reaction on the reverse side of AFPA medium (Figure 4.2; Table

4.4), similar to A. flavus and A. minisclerotigenes. Colony texture of A. texensis on

AFPA after 7 d at 25, 30 and 35°C was floccose with abundant white mycelia and immature white sclerotia.

Cyclopiazonic acid

Aspergillus texensis produced cyclopiazonic acid (CPA) (Table 4.4); CPA concentrations ranged from 7.5 26.6 µg/g with a mean of 17.4 µg/g. Similarly, all

S strain morphotype isolates of A. flavus (n = 3) tested in the current study produced CPA (Range: 16.0 30.6 µg/g; Mean = 25.0 µg/g).

Aflatoxins

113

Aspergillus texensis produced B1, B2, G1, and G2 aflatoxins in maize. Aflatoxin concentrations ranged from 11 71 µg/g AFB1, 0.6 2.6 µg/g AFB2, 66 225 µg/g

AFG1, and 2.0 7.7 µg/g AFG2. The S strain morphotype of A. flavus, which is the most commonly reported S morphology fungus in the US, produces only B aflatoxins. In an experiment comparing total aflatoxin production by genetically distinct S morphology fungi, mean aflatoxin concentrations produced by A. texensis and A. flavus S morphotype fungi were similar at 31°C (Table 4.5; p >

0.05). Aspergillus texensis produced at least two-fold more G aflatoxins than B aflatoxins (Table 4.5). Fungi with S morphology from the African continent, including isolates of A. aflatoxiformans and LAF, produced the highest concentrations of total aflatoxins in maize, while A. minisclerotigenes produced the lowest quantities. However, all fungi with S morphology evaluated in the current study overall produced lethal concentrations of total aflatoxins in maize at 31°C

(Range: 33 568 µg/g).

Taxonomy

Aspergillus texensis P.Singh, M.J.Orbach, and P.J.Cotty sp. nov.

MycoBank MB828668. Figure 4.2.

Etymology: The species epithet texensis first isolates of the new species were collected.

Diagnosis: Aspergillus texensis is closely related to A. minisclerotigenes and the unnamed lineage LAF. Aspergillus minisclerotigenes grows faster than A. texensis on Czapek agar at 20°C and 37°C, and on V8 agar at 37°C; however, A.

114 texensis grows faster on V8 agar at 40°C (Table 4.6). The unnamed lineage LAF lacks the ability to produce G aflatoxins unlike A. texensis, which produces both B and G aflatoxins (Table 4.5).

Typus: United States of America, Texas, Waxahachi, soil cropped to maize

(Zea mays), collected by P.J.Cotty (holotype NRRL 66855, ex-type: WXMXP1R3-

B R = A2292).

Colony characteristics: Aspergillus texensis colonies attained an average diameter of 53 mm (51 58 mm) at 25°C, 55 mm (53 59 mm) at 37°C, and 7 mm

(4 9 mm) at 42°C on Czapek agar after 7 d. Colony diameters were greater on

Czapek agar with 2% NaCl and Czapek agar with yeast extract (CYA) media at these temperatures (Table 4.6). Maximum radial growth of A. texensis occurred at

35°C on all media. Aspergillus texensis did not germinate at 5°C or 10°C. Colony surface on Malt agar was floccose with dominant white mycelia, velvety on CYA and V8 (Figure 4.2). Colony reverse buff on CZ, CYA and Malt.

Micromorphology: Abundant production of dark black sclerotia (150 300 µm) on the agar surface was observed on Czapek, CYA, Malt, and V8 agar (Figure

4.2). Fungal isolates produced sparse yellow-green conidia on all media tested; conidia were circular and smooth walled (3-6 µm diameter). Vesicle globose, 30

60 µm in diameter. Conidiophores with stippled stipes, hyaline, 400 800 × 10 14

µm, phialides 6 11 × 3 5. Aspergillus texensis produced copious quantities of sclerotia but fewer conidia on maize after incubation at 31°C for 7 d (Figure 4.2).

115

The type and other representative isolates of A. texensis have been submitted to the ARS Culture Collection (NRRL) (Peoria, IL, USA) and the Fungal Genetics

Stock Center, Manhattan, KS (Table 4.1).

Discussion

Most aflatoxin producing fungi belong to Aspergillus section Flavi with several members recognized as economically important agents of aflatoxin contamination of crops (Cotty et al., 2008). Although many species within section Flavi produce aflatoxins, fungi that have repeatedly been recovered at high frequencies from crops and agricultural soils, such as A. flavus L and S strain morphotypes, A. parasiticus, A. minisclerotigenes, A. aflatoxiformans from West Africa, and the fungi associated with lethal aflatoxicoses in Kenya (LAF), are the primary concern as causal agents of aflatoxin contamination of crops (Cardwell and Cotty, 2002;

Cotty, 1997; Donner et al., 2009; Probst et al., 2012; Singh and Cotty, 2019).

Morphology (sclerotia, conidia, colony characteristics) and physiology (growth rates at different temperatures), and mycotoxin production have been conventional tools in identification and characterization of species within section Flavi

(Hesseltine et al., 1970; Saito and Tsuruta, 1993). However, in the past decade, fungi within section Flavi with overlapping morphological and physiological characteristics have been assigned to genetically distinct taxa based on DNA sequences (Carvajal-Campos et al., 2017; Frisvad et al., 2019; Pildain et al., 2008;

Probst et al., 2012). For instance, the A. flavus S Strain morphotype and LAF fungi produce small sclerotia (< 400 µm) and only B aflatoxins; however, multi locus

116 gene genealogies, and differences in deletions in the norB-cypA region of the aflatoxin biosynthesis gene cluster, clearly show that A. flavus and LAF are distinct species (Probst et al., 2007; Probst et al., 2012). Likewise, A. minisclerotigenes,

A. aflatoxiformans, and A. cerealis show high morphological similarity (numerous small sclerotia) and produce B and G aflatoxins. DNA sequence data from multiple unlinked loci strongly support placement of each of these into a distinct monophyletic taxon (Carvajal-Campos et al., 2017; Frisvad et al., 2019; Pildain et al., 2008). The utilization of a polyphasic approach, which combines morphology, physiology, and DNA-based analyses for recognition of novel species, provides a powerful tool for researchers to approach cryptic diversity within Aspergillus section Flavi. However, multi-locus DNA sequence-based phylogenetics are frequently sufficient for delineation of new taxa (Frisvad et al., 2019; Samson and

Varga, 2009; Soares et al., 2012).

Aspergillus texensis forms a highly supported distinct terminal group in phylogenies of sequence data from three unlinked loci (benA, cmdA, niaD). The branch containing A. texensis is congruent, and the same isolates occur as a terminal group in trees constructed from individual or concatenated genes with strong statistical support, both by Bayesian posterior probability and bootstrap analyses (Figure 4.1, 4.2 and 4.3). In light of the aforementioned results, A. texensis fulfills requirements of the phylogenetic species concept, which limits species boundaries to a monophyletic group within which a pattern of ancestry and descent exists (Ito et al., 2001; Peterson, 2008; Samson and Varga, 2009). This,

117 along with physiological data and types of secondary metabolites produced, distinguishes A. texensis as a new species.

DNA sequences resulting fro -tubulin gene amplified using primers from (Glass and Donaldson, 1995) were highly conserved among closely- related aflatoxin producing species. The current study therefore incorporated

-tubulin using primer pair Bt3a-3b (Table

4.2) to include more variable characters to seek better resolution. However, 0.9 kb

453 intron positions) still did not contain enough variable characters to resolve multiple aflatoxin producing species. Nevertheless, individually, calmodulin and nitrate reductase sequences could clearly separate previously reported aflatoxin- producers and A. texensis in both Bayesian and Maximum Likelihood topologies with high Bayesian posterior probabilities and Bootstrap support (Figure 4.1 and

4.2 -tubulin encodes for proteins, which along with alpha tubulin, polymerize into microtubules, cellular structures that are crucial in multiple cellular processes. Owing to this highly conserved function, DNA

-tubulin may not be sufficient to reveal the cryptic diversity among more closely related species within Aspergillus section Flavi, while calmodulin and nitrate reductase could be superior choices to identify novel species within section

Flavi. This is in agreement with previous studies describing novel species within section Flavi that have often reported phylogenies based on concatenated gene

-tubulin, calmodulin, RNA polymerase, internal transcribed

118 spacers (ITS), etc. with high support for the clade representing a new species

(Carvajal-Capos et al., 2017; Ito et al., 2001; Soares et al., 2012). However, gene trees based on individual sequences of RNA p -tubulin either did not resolve members of section Flavi or did so with low branch support (Ito et al., 2001; Soares et al., 2012). On the other hand, the calmodulin gene provides a useful tool to identify and resolve closely-related aflatoxin producers of section

Flavi (Frisvad et al., 2019; Ito et al., 2001; Peterson et al., 2001; Soares et al.,

2012). The nitrate reductase gene, which encodes for an enzyme essential for nitrate assimilation in fungi, is also a powerful tool to identify and characterize species within section Flavi, as demonstrated in the current study, and by Probst et al., 2014 and Singh and Cotty, 2019.

The fungal isolates representing A. texensis sp. nov. produce abundant quantities of small sclerotia (< 400 µm diameter) on several growth media, and on maize kernels (Figure 4.2). Conidia are smooth walled and appear light green to yellowish green, similar to the S strain morphotype of A. flavus. Although microscopic characters and growth at different temperatures in various media largely overlap (Table 4.6), A. texensis is characterized by production of both B and G aflatoxins, unlike the A. flavus S strain morphotype, which produces only B aflatoxins. This is due to either a 0.9 or 1.5 kb deletion in the norB-cypA region of the aflatoxin biosynthesis gene cluster in A. flavus, which has resulted in loss of G aflatoxin production by this species (Ehrlich et al., 2004); however, A. texensis has an intact norB-cypA region. Total aflatoxin production on maize by A. flavus S

119 strain morphotype did not differ from that of A. texensis (Table 4.5; p > 0.05); however, A. texensis produced significantly lower concentrations of B aflatoxins

(Table 4.5; p < 0.05). Additionally, phylogenetic reconstruction based on 4.1kb sequence data from three unlinked genes (benA, cmdA and niaD) clearly demonstrate that A. texensis is a genetically distinct, distant relative of A. flavus.

Aspergillus texensis is closely related to A. minisclerotigenes (first described from Argentinian groundnuts) (Pildain et al., 2008) and LAF (the lineage with S morphology responsible for severe aflatoxin contamination that led to hundreds of deaths in Kenya) (Probst et al., 2007; Probst et al., 2012) (Figure 4.1). Both A. minisclerotigenes and LAF have been reported in very low frequencies in the US

(one isolate of A. minisclerotigenes and three isolates of LAF) (Pildain et al., 2008;

Probst et al., 2012;). However, higher incidences of A. minisclerotigenes are known from groundnuts in Argentina (Pildain et al., 2008), almonds and maize in

Portugal (Soares et al., 2012), maize in Eastern and Central Africa (Probst et al.,

2014), and chilies in Nigeria (Singh and Cotty, 2019). Similarly, LAF has been recovered in high frequencies from maize in Kenya (Probst et al., 2012).

Production of G aflatoxins by A. texensis differentiates it from LAF, which produces only B aflatoxins due to a characteristic 2.2 kb deletion in the norB-cypA region of the aflatoxin biosynthesis gene cluster (Probst et al., 2012). Aspergillus texensis is distinguished in growth from A. minisclerotigenes on Czapek agar at 20°C and

37°C, and on V8 agar at 37°C and 40°C (Table 4.6). Furthermore, A. texensis produces higher concentrations of aflatoxins than A. minisclerotigenes on maize

120

(Table 4.5; p < 0.05). Also, phylogenetic analyses of DNA sequence data strongly support the delineation of these two species (Figure 4.1, 4.2 and 4.3).

Fungal isolates of most species within section Flavi contain either MAT1-1 or

MAT1-2 idiomorphs at the mating type locus, and are therefore heterothallic

(Carvajal-Campos et al., 2017). All isolates of A. texensis examined to date have the MAT1-1 idiomorph suggesting clonal evolution in the absence of meiotic recombination, which within heterothallic species requires the presence of MAT1-

2 idiomorph strains for sexual reproduction. However, further sampling is needed to determine whether isolates of A. texensis with the MAT1-2 idiomorph are present, as only 11 isolates have been collected to date.

Aflatoxigenic fungi frequently reported across the US include A. flavus L strain morphotype (isolates can be toxigenic or atoxigenic), A. flavus S strain morphotype, and A. parasiticus (Bayman and Cotty, 1991; Boyd and Cotty, 2001;

Jaime-Garcia and Cotty, 2004; Horn et al., 1995). Accurate identification and characterization of aflatoxin producing fungi allows clarification of the etiology of crop contamination and development of procedures for aflatoxin mitigation.

Aspergillus texensis is known from 11 isolates currently, and was recovered from soils cropped to maize, and maize grain produced in Arkansas, Louisiana, and

Texas. Aflatoxin contamination of crops is severe in these regions (Jaime-Garcia and Cotty, 2003; King et al., 2000; Tubajika et al., 1999). The majority of A. texensis isolates (82%) discovered in the current study are from Texas, where contamination is a perennial issue. The highly toxic A. flavus S strain morphotype

121 has been reported in high frequencies from hot and dry regions of Texas, and increased proportions of these fungi have been associated with increased soil temperature (Jaime-Garcia and Cotty, 2006b; Jaime-Garcia and Cotty, 2010). The morphology of Aspergillus texensis is highly similar to that of the S strain morphotype of A. flavus with numerous small (< 400 µm) sclerotia. Furthermore, both A. texensis and the A. flavus S strain morphotype produce high and comparable concentrations of aflatoxins in maize. These data, along with the aforementioned observations, suggest that co-occurrence of these S morphology aflatoxin producers in the US may lead to crop contamination with high aflatoxin concentrations. The current study also indicates that G aflatoxins in maize are not solely attributable to A. parasiticus.

122

Kurtzman Kurtzman et 1987 al., B.W., National Horn Peanut Dawson, NRRLGA(in Lab, database) B.W., National Horn Peanut Dawson, NRRLGA(in Lab, database) F.M. NRRLScales (in database) Hesseltine 1970; et al., 2008etal., Pildain et al., Probst 2012 1989 Cotty, Cotty, Singh and 2019

Argentina Ito et et al.,2001 Ito Jong, and Wei 1986 Horn, 1997 USA Wheat, Argentina USA Soil, USA Georgia, Soil, USA None et 1974al., Rambo Georgia, Soil, USA Georgia, Soil, USA et al., Geiser 1998 Uganda Peanut, Soil, Benin Cotty, Singh and 2019 Nigeria Soil, 1989 Cotty, Soil, Australia Peanut, Argentina USA Soil, Arizona, 1989 Cotty, Hesseltine 1970et al., USA Soil, Arizona, Cotty, Singh and 2019 California, Walnut, USA USA Soil, Arizona, Study Current Chili Chili PakistanStudyChili, Current Texas, USA Soil, Study Current USAMaize, Texas, Maize,Louisiana, USA 123 123

S Morphotype Soil, Arizona, USA Cotty, 1989 Cotty, USA Arizona, Soil, Morphotype S Morphotype S Morphotype S Morphotype S Morphotype L Morphotype L Morphotype L Morphotype L A. nomius A. pseudotamarii A. caelatus A. parasiticus A. parasiticus A. parasiticus A. parasiticus A. parasiticus A. parasiticus A. minisclerotigenes A. minisclerotigenes A. minisclerotigenes A. flavus A. flavus A. flavus A. flavus A. flavus A. flavus A. flavus A. flavus A. texensis A. texensis A. texensis A.

382 - 96044

Origin of fungal isolates examined in the current study. the current in examined Origin of fungal isolates

pce/Txn ore Citation Source Taxon Species/ # Isolate Isolate 13137NRRL 443NRRL 25528NRRL 465NRRL 29538NRRL 29590NRRL 424NRRL 2999NRRL BN009-E A-11611 NRRL 4-2 TAR3N43 ATCC MYA-384AF70/ ATCC MYA-383AF42/ ATCC MYAAF12/ 3251NRRL ATCC AF13/ CHL019 CHL159 CHL187 NRRL R/ 66855/W11P1R3-B A2292 NRRL EC42-I/ 66856/A2295 NRRL 66857/ BRG3224-E/ A2296 Table 4.1 Table

Current Study Current Frisvad et al., 2019, Hesseltine 1970et al., Frisvad Frisvadet al., 2019, 2019 et al., Frisvad Frisvadet al., 2019, 2019et al., Frisvad Frisvadet al., 2019, 2019et al.,

Soil, Texas, USA Current Study Current Study Current Texas, USA Soil, Study Current Study Current Maize,Arkansas,USA Study Current Texas, USA Soil, Study Current USAMaize, Texas, Study Current et al., Probst 2007 Texas, USA Soil, Texas, USA Soil, et al., Probst 2007 Texas, USA Soil, Texas, USA Soil, Maize,Kenya Maize,Kenya Soil, Benin Soil, Benin 124 124 Groundnut, Nigeria Groundnut, Soil, Benin

A. texensis A. texensis A. texensis A. texensis A. texensis A. texensis A. texensis A. texensis A. AflatoxicosisLethal Fungus AflatoxicosisLethal Fungus aflatoxiformans A. aflatoxiformans A. aflatoxiformans A. aflatoxiformans A. MYA-380

I/NRRL 66859/ -

A - Isolates were obtained from the ARS Culture Collection, Peoria, IL, USA (NRRL), the American Type Type the American USA Peoria,(NRRL), IL, fromCollection, wereobtained ARS Culture the Isolates WXMX1-1G/ NRRL 66858/WXMX1-1G/ A2294 VCSSBC1 A2293 J35-E VC16 CTL-1I Q P2R2-A 1-1-O 1-1L A1170 K805-E/ K784-D/ A1168 A-11612NRRL MYA-381 ATCC BN040-B/ ATCC BN038-G/ ATCC MYA-379BN008-R/ Culture Collection, Manassas, USA (ATCC), Fungal Genetics Stock Center, Manhattan, KS (A) or were were or (A) KS Manhattan, Center, GeneticsStock USAFungal Manassas, (ATCC), Collection, Culture Collection. Culture ARS, Laboratory Tucson USDA- the at present #

Reference

(°C) a 51 Glass Donaldson, and Glass 1995 study Current 51 et al., Probst 2012 56 et al., Probst 2012 56 et al., Probst 2012 48 et al., Probst 2012 52 Cotty, and Singh 2019 57 55 ) for PCR amplifications. amplifications. PCR for )

a 125 125 GCAGCCCAATGGTCACTACGGC - eune T Sequence F-GGTAACCAAATCCGTGCTGCTTTC R-ACCCTCAGTGTAGTGACCCTTGGC F-CGTCGTTCATTCGAGGTGTA R-CCGCTCAACTTCAAGTCCAT F-GGCCTTCTCCCTATTCGTAA R-CTCGCGGATCATCTCATC F-GGCTGGATGTGTGTAAATC R-ATTGGTCGCATTTGAAGGG F-CGGACGATAAGCAACAACAC R-GGATGAACACCCGTTAATCTGA F-ACGGCCGACAGAAGTGCTGA R-TGGGCGAAGAGACTCCCCGT F R-GGCTGCACGCCCAATGCTTC -tubulin -tubulin Target Target Gene Calmodulin Nitrate reductase

Primers and locus specific (Tannealing Primers temperatures and locus PrimerPair Bt2a-2b Bt3a-3b cmd42-637 cmd2F-2R niaDF-AR niaDBF-BR niaDCF-CR Table 4.2 Table

MK119671 MK119672 MK119673 MK119674 MK119675 MK119676 MK119677 MK119678 MK119679 MK119680 MK119681 MK119682 MK119683 MK119684 MK119685 MK119686

K162 MH760519 MH760520 MK119692 MH760522 MK119693 MH760525 MK119694 MH760528 MK119695 MH760529 MK119696 MH760530 MK119697 MH760531 MK119698 MH760532 MK119699 MH760533 MK119700 MH760534 MK119701 MH760537 MK119702 MH760538 MK119703 MK119704 MK119705 MK119706 MK119707 MK119708 MK119709 MK119710 MK119711 MK119712 MK119713 MK119714 MK119715 MK119716 MK119717 MK119718 MK119719 MK119720

126 126 eA mA niaD cmdA NumberGenBank Accession benA MK119726 MK119727 MK119728 MK119729 MK119730 MK119731 MK119732 MK119733 MK119734 MK119735 MK119736 MK119737 MK119738 MK119739 MK119740 MK119741 MK119742 MK119743 MK119744 MK119745 MK119746 MK119747 MK119748 MK119749 MK119750 MK119751 MK119752 MK119753 MK119754

G - NRRL 13137NRRL TAR3N43 4-2 A-11611NRRL CHL159 CHL187 AF13 AF12 AF42 AF70 3251NRRL 2999NRRL BN009-E 424 NRRL 465NRRL 29538NRRL 29590NRRL 66708NRRL 66709NRRL 66710NRRL A-11612NRRL BN038 BN040-B BN008-R A1168 A1170 66855NRRL 66856NRRL 66857NRRL

Isolates used in the current study with GenBank accession numbers. studyGenBank usedwith current Isolates the in pce Isolate Species Aspergillus nomius minisclerotigenes Aspergillus minisclerotigenes Aspergillus minisclerotigenes Aspergillus flavus Aspergillus flavus Aspergillus flavus Aspergillus flavus Aspergillus flavus Aspergillus flavus Aspergillus flavus Aspergillus Aspergillus parasiticus Aspergillus parasiticus Aspergillus parasiticus Aspergillus parasiticus Aspergillus parasiticus Aspergillus parasiticus cerealis Aspergillus cerealis Aspergillus cerealis Aspergillus Aspergillus aflatoxiformans Aspergillus aflatoxiformans Aspergillus aflatoxiformans Aspergillus aflatoxiformans Unnamedlineage Lethal AflatoxicosisFungus Unnamedlineage Lethal AflatoxicosisFungus texensis Aspergillus texensis Aspergillus texensis Aspergillus Table 4.3 Table

MK119687 MK119688 MK119689 MK119690 MK119691

MK119721 MK119721 MK119722 MK119723 MK119724 MK119725

127 127 MK119755 MK119756 MK119757 MK119758 MK119759

NRRL 66858NRRL 66859NRRL 443NRRL 20818NRRL 25528NRRL

Aspergillus texensis Aspergillus texensis Aspergillus Aspergillus pseudotamarii tamarii Aspergillus caelatus Aspergillus Table 4.4 Production of major mycotoxinsa by certain species within

Aspergillus section Flavi.

Aflatoxin Aflatoxin Aflatoxin Aflatoxin Aspergillic Species CPA B1 B2 G1 G2 Acid A. texensis + + + + + + A. flavus + + ± + A. + + + + + + minisclerotigenes A. parasiticus + + + + + a Data from Peterson et al., 2001, Pildain et al., 2008 and the current study.

128

Table 4.5 Aflatoxin production by fungi with S morphology within Aspergillus section Flavi.

Aflatoxin (µg/g) Source/ Species Isolate Total Location AFB AFG AF

BRG3224 E Maize/USA 42 111 153 EC42-I Maize/USA 40 99 139 A. texensis W11P1R3-B R Soil/USA 47 111 158 Average 43C 107B 150B

AF12 Soil/USA 190 NA 190 A. flavus S AF42 Soil/USA 181 NA 181 Morphotype AF70 Soil/USA 191 NA 191 Average 187AB NA* 187B

Groundnut/ A-11611 31x 72x 103x Nigeria A. minisclerotigenes 4-2 Soil/Australia 9y 24y 33y TAR3N43 Soil/Argentina 9y 25y 34y Average 17D 40C 57C

Groundnut/ A-11612 92y 202y 294y Nigeria A. aflatoxiformans BN038-G Soil/Benin 105y 129z 234y BN040-B Soil/Benin 207x 361x 568x Average 134B 231A 365A

K805-E Maize/Kenya 304 NA 304xy Lethal Aflatoxicosis K784-D Maize/Kenya 361 NA 361x Fungus K849-B Maize/Kenya 215 NA 215y Average 294A NA* 294A

B aflatoxin, G aflatoxin and total aflatoxin concentrations were compared between and within species. Each toxin concentration is a mean of four replicates. Means followed by different upper case letters in bold (A,B,C) or lower case letters (x,y)

p < 0.01). Values within a column lacking a letter do not differ (ANOVA, p > 0.05). * Fungi which do not produce G aflatoxins were excluded when comparing G aflatoxin production between and within species.

129

Flavi section section Aspergillus 9 11 42°C 7 8 8 13 12 15 10 11 7 8 7 7 9

a b b 54 56 40°C 46 37 39 68 69 73 49 41 30 32 42 28 31

b b a b a a 65 57 37°C 55 54 63 72 65 72 64 55 55 58 52 61 67 78 71 35°C 73 65 75 77 71 77 78 69 67 68 72 68 73 130 130 77 72 30°C 67 66 75 72 69 73 77 68 57 67 68 66 74

ab b a ab b a Colony DiameterColony(mm) 71 64 25°C 53 50 62 60 57 66 68 46 44 69 60 53 59

b b a 20°C 31 35 41 44 44 44 46 42 44 30 34 43 33 32 31

b a b a ab b a ab b 9 11 15°C 3 7 3 11 12 10 9 7 5 3 4 3 2

Colony diameter 7 d post inoculation. Statistical comparisons among species comparisons among species Statistical post d diameter 7 inoculation. Colony Species Species A. flavus A. flavus A. flavus A. flavus A. flavus A. A. texensis A. texensis A. texensis A. texensis A. texensis A. A. minisclerotigenes A. minisclerotigenes A. minisclerotigenes A. minisclerotigenes A. minisclerotigenes A. Influence of temperature on radial growth of three S morphology species in in species morphology S three of growth radial on temperature of Influence

V8 CZ Malt Malt CYA NaCl CZ with with CZ Media compositions in Materials and Methods; CZ-Czapek solution agar, CZ with NaCl-Czapek with CZ solution agar, and CZ-Czapek Methods; Materials in compositions Media Medium

Table 4.6 Table media.agar on solution agar, agar, extract Malt-Malt yeast solutionwith CYA-Czapek 2% NaCl, with agar solution juice agar. V8-V8 aremeans Values combination. medium/temperature for each independently were performed

>0.05). p p 131 131 of four replicate four of Figure 4.1 Mid-point rooted Bayesian phylogeny of A. texensis and closely related

S morphology fungi with several additional species within section Flavi for reference. Phylogeny is based on concatenated benA (0.9kb), cmdA (1.2kb), and niaD (2.1 kb) genes of Aspergillus section Flavi. Values above nodes or before commas are Bayesian posterior probabilities and values below nodes or after commas are bootstrap values from 500 replicates. Aspergillus nomius NRRL

13137 was used as the outgroup. *Aflatoxin producers with S morphology.

132

Figure 4.2 Mid-point rooted Bayesian phylogeny of A. texensis and closely related S morphology fungi with several additional species for reference based on partial calmodulin gene, cmdA (1.2 kb). Values above nodes are Bayesian posterior probabilities and values below nodes are bootstrap values from 500 replicates.

133

Figure 4.3 Mid-point rooted Bayesian phylogeny of A. texensis and closely related S morphology fungi with several additional species for reference based on a portion of nitrate reductase gene, niaD (2.1 kb). Values above nodes or before commas are Bayesian posterior probabilities and values below nodes or after commas are bootstrap values from 500 replicates.

134

Figure 4.4 (a-e) Colonies of A. texensis grown at 25°C for 7 days on (a) CZ,

(b) CYA, (c) Malt agar, (d) reverse on AFPA, and (e) V8 agar. (f) A. texensis on maize grown at 31°C for 7 days; (g-h) Conidiophores. Bars- 10 µm.

135

APPENDIX E- Diversity among S morphology fungi in Aspergillus section

Flavi from the United States

To be submitted to: Fungal Diversity

Abstract

Aflatoxins are carcinogenic metabolites produced by several species in Aspergillus section Flavi. These mycotoxins can contaminate a wide range of crops in warm regions of the globe. Several phylogenetically distinct fungi within section Flavi have an S morphology (average sclerotium size < 400 µm) and produce high concentrations of aflatoxins (B or B and G). Fungi with S morphology are of serious concern in terms of crop contamination in the United States (US) and Africa because of their consistent production of high levels of aflatoxins. A previously reported unnamed lineage referred to as the Lethal Aflatoxicosis

Fungus (LAF) was identified as the causal agent of maize contamination in Kenya which caused more than 100 human deaths. Aflatoxins remain a serious economic threat in the US. Although S morphology Aspergilli have been associated with severe crop contamination in the US, little is known about their genetic diversity.

The current study characterized Aspergilli with S morphology (n = 494) collected between 2002 and 2017 from soil and maize samples across regions of the US where aflatoxin contamination is a perennial problem. Phylogenetic analyses based on partial gene sequences of the calmodulin (1.9 kb) and nitrate reductase

(2.1 kb) genes resolved S morphology isolates from the US into four distinct clades: 1) Aspergillus flavus S morphotype (89.7%); 2) A. agricola sp. nov. (2.4%);

136

3) A. texensis (2.2%), and 4) a new SB lineage (5.7%). While all four groups of the

S morphology fungi produced high concentrations of aflatoxins in maize at 25°C,

30°C and 35°C, A. flavus S morphotype isolates was the only group that produced unacceptable concentrations of aflatoxins even at 40°C. Further genotyping of the

A. flavus S isolates using 17 simple sequence repeat markers revealed high genetic diversity with 202 haplotypes from 443 isolates and 3 to 20 alleles per locus. Knowledge of the occurrence of distinct species and haplotypes of S morphology fungi that are highly aflatoxigenic under a range of environmental conditions can be utilized towards aflatoxin management.

137

Introduction

The filamentous fungus Aspergillus flavus Link (Ascomycota, Eurotiales) is ubiquitous in the environment and thrives on living and dead organic matter.

Aspergillus flavus and other closely related species within Aspergillus section Flavi pose a serious threat to agriculture, as well as human and animal health due to their production of aflatoxins on a variety of substrates. Aflatoxins are highly toxic, carcinogenic polyketides that cause severe health and economic hazards (Park and Liang, 1993). Aflatoxin-producers within Aspergillus section Flavi can infect a wide range of food and feed crops, such as maize, sorghum, groundnuts, cottonseed, chilies and tree nuts, along with edible insects, and contaminate them with aflatoxins (Kachapulula et al., 2017; Kachapulula et al., 2018; Ortega-beltran et al., 2018; Rodrigues et al., 2009; Shetty and Bhatt, 1997; Shotwell et al., 1969;

Singh and Cotty, 2017). Chronic exposure to sub-lethal concentrations of aflatoxins through consumption of contaminated food or feed results in immune suppression (Jiang et al., 2005; Owaga et al., 2011), stunting (Cardeilhac et al.,

1970; Gong et al., 2004), and hepatocellular carcinomas (Henry et al., 1999; Liu and Wu, 2010). Acute poisoning can cause severe liver damage followed by rapid death (Africa A, 2016; CDC, 2004; Krishnamachari et al., 1975; Outbreak News

Today, 2017). Strict monitoring and enforcement of regulations to limit aflatoxin concentrations in food and feed limit exposure to this mycotoxin in developed nations. Losses from aflatoxins in the US are largely associated with market loss in the form of reduced prices for crops and disposal of large quantities of

138 contaminated food rather than health effects (Cotty et al., 2008; Wu and Guclu,

2012). Economic repercussions due to these regulations are huge with annual losses that may exceed $500 million in the United States (US) alone (Mitchell et al., 2016; Robens and Cardwell, 2003). Monetary losses ranging from $52 million to nearly $2 billion annually were estimated only from aflatoxin contamination of maize in the US (Mitchell et al., 2016).

Aspergillus flavus and A. parasiticus have been frequently implicated as primary causal agents of aflatoxin contamination of crops (Klich MA, 2007).

Aspergillus flavus was delineated into the L and S morphotypes three decades ago

(Cotty, 1989). The L morphotype produces few large size sclerotia (> 400 µm) and abundant conidia, whereas the S morphotype produces copious smaller sclerotia

(< 400 µm) but relatively few conidia (Cotty, 1989). The ability of the L morphotype of A. flavus to produce aflatoxins varies such that fungal isolates with the L morphotype may produce no aflatoxins (atoxigenic isolates), or low to very high concentrations of aflatoxins (Bayman and Cotty, 1993; Mehl et al., 2012). Fungi with the S morphotype consistently produce high concentrations of aflatoxins

(Cotty, 1989; Probst et al., 2012; Singh and Cotty, 2018; Singh and Cotty, 2019).

Aspergillus section Flavi contains several described species that are morphologically similar to but phylogenetically distinct from the S morphotype of

A. flavus. While the significance of many of the newly described S morphology species to crop contamination is unclear, the S morphotype of A. flavus has been linked to severe contamination of maize and cottonseed in the US (Cotty, 1996;

139

Jaime-Garcia and Cotty, 2004; Jaime-Garcia and Cotty, 2006a), and A. aflatoxiformans (previously reported as an unnamed taxon SBG or A. parvisclerotigenus), A. minisclerotigenes and an unnamed lineage from Kenya, referred to as the Lethal Aflatoxicosis Fungus (LAF) have been associated with crop contamination in Sub-Saharan Africa (Cardwell and Cotty, 2002; Probst et al.,

2012; Probst et al., 2014; Singh and Cotty, 2019). Crop infection by S morphology fungi is of concern because these fungi are notorious for the production of high concentrations of aflatoxins and can cause unacceptable contamination even at low incidences (Cardwell and Cotty, 2002; Cotty et al., 2008; Probst et al., 2010;

Singh and Cotty, 2019). The unnamed lineage LAF was responsible for the deadly aflatoxicosis outbreak in Kenya in 2004 that claimed the lives of more than 100 people (Probst et al., 2007; Probst et al., 2012). Despite the toxic potential of S morphology fungi, studies on their occurrence and diversity across regions of the globe that favor aflatoxin contamination of crops are sporadic. Probst et al. (2014) reported region-specific occurrence of distinct S morphology fungi across Sub-

Saharan Africa; A. aflatoxiformans (reported as Strain SBG) was only detected in

West Africa, while LAF (reported as New lineage SB) and A. minisclerotigenes were confined to Eastern and Southern Africa. However, A. minisclerotigenes has recently been reported from dried red chilies originating from Nigeria, establishing its occurrence in West Africa (Singh and Cotty, 2019).

In the US, aflatoxin contamination of susceptible crops, including cottonseed and maize in Arizona, Texas and the southeastern US is of significant

140 economic concern (Jaime-Garcia and Cotty, 2003; Mitchell et al., 2016).

Environmental conditions (high temperatures and sub-arid to arid conditions) in these states/regions favor growth and proliferation of aflatoxigenic fungi and crop contamination (Cotty and Jaime-Garcia, 2007) rendering aflatoxin contamination of crops a perennial concern. Even though fungi with S morphology are associated with severe crop contamination, knowledge of the incidence and genetic diversity of these highly aflatoxigenic fungi in the US is limited and unexplored.

The objectives of the current study were to (i) characterize fungi with S morphology within Aspergillus section Flavi recovered from soil and maize samples collected from across the susceptible regions of the US, (ii) assess phylogenetic relationships of US S morphology isolates to previously described species of aflatoxin-producing fungi, (iii) determine the relative significance of S morphology fungi to aflatoxin contamination events, and (iv) test the utility of simple sequence repeat (SSR) markers for assessment of diversity among closely related

S morphology genotypes. A novel aflatoxin producing species discovered in the current study is also described. The acquired knowledge on the occurrence of distinct communities of highly aflatoxigenic S morphology fungi can be utilized for better aflatoxin management in the US.

Materials and Methods

Fungal isolates

Fungal isolates (n = 494) used in this study were recovered from soil and maize samples obtained as part of a large-scale study of the diversity of

141

Aspergillus spp. in Arizona, Texas, and the southeastern US (Figure 5.1). Isolates are available in the culture collection at the USDA-ARS Aflatoxin Laboratory in

Tucson, Arizona. Isolates originated from soils collected in Arizona from 1989 to

2005, maize and soil samples collected in Texas from 2004 to 2014, and maize collected in the southeastern US from 2015 to 2017 (Table 5.1). Several reference isolates with known affiliations to already described species/lineages within

Aspergillus section Flavi were obtained from the ARS Culture Collection, Peoria,

IL, USA, the American Type Culture Collection, Manassas, USA, or the USDA-

ARS Aflatoxin Laboratory culture collection in Tucson, Arizona (Table 5.2).

DNA extraction and gene amplification

Isolates were grown on V8 agar (5% V-8 juice; 2% agar; pH 6.0) with 2%

NaCl for 7 d at 31°C in the dark, and DNA was extracted as described previously

(Callicott and Cotty, 2015). Concentration of stock DNA was quantified and diluted to 5 ng/µl for PCR. Deletions in the norB-cypA genes of the aflatoxin biosynthesis gene cluster resulting in loss of G aflatoxin production by isolates were determined using primer sets AP1729-3551 (Ehrlich et al., 2004) and CP5F-R (Probst et al.,

2012) as described previously. Amplicons were visualized on 1% agarose gels against 1kb Plus ladder (Thermo Scientific, Waltham, MA, USA) for sizing and positive controls including A. parasiticus (no deletion), A. flavus L or S strain morphotypes (0.9 or 1.5 kb deletions) and the previously reported LAF (2.2 kb deletion).

142

Partial gene fragments of the calmodulin (cmdA on chromosome 2) and the nitrate reductase (niaD on chromosome 4) genes were amplified with three sets of primers covering approximately 1.9 kb and 2.1 kb of cmdA and niaD genes, respectively (Table 5.3). The primer pair cmd3F-3R was designed based on genome sequence of A. flavus NRRL 3357 (GenBank accession no.

AAIH02000003) using Primer3 version 0.4.0 (Koressaar and Remm, 2007;

Untergasser et al., 2012). PCR was performed in 20 µl volumes using 5 ng of genomic DNA with a PCR premix (AccuPower® HotStart, Bioneer, Alameda, CA,

USA) on a MyCycler thermocycler (Bio-Rad Laboratories, Richmond, CA, USA) under the following conditions: for cmdA and niaD genes, 5 min at 94°C followed by 38 cycles of 94°C for 20 s, locus-specific annealing temperatures for 30 s (Table

5.3), and 72°C for 1 min, and a final extension set for 5 min at 72°C. Amplicons

(three for each gene) were visualized on 1% agarose gels followed by bidirectional sequencing of each amplicon at the University of Arizona Genetics Core Facility

(UAGC, Tucson, AZ, USA) using the amplification primers (Table 5.3).

DNA sequence data and phylogenetics

Fungi with S morphology collected from across the US and isolates from previously described species were used for phylogenetic comparisons.

Bidirectional sequences of cmdA (1.9 kb) and niaD (2.1 kb) were used to create a consensus sequence for each amplicon by assembling six reads per gene with visual inspection and alignment using the MUSCLE algorithm within Geneious Pro

Version 7.1.9 (Biomatters Ltd., Auckland, New Zealand). DNA sequence

143 alignments were refined manually. Individual and concatenated phylogenies were constructed for both loci using Bayesian inference with 10 million generations

(MrBayes version 3.2.6; Huelsenbeck and Ronquist, 2001) and maximum likelihood (ML) analyses with PhyML at Phylogeny.fr (Dereeper et al., 2008;

Dereeper et al., 2010) to confirm tree topologies. Data sets were bootstrapped with

500 replicates for ML analysis. Trees were drawn mid-point rooted using FigTree v.1.4.3 (Rambaut, 2012).

SSR genotyping and genetic diversity

Aspergillus flavus S morphotype isolates (n = 443) were further genotyped using 17 SSR loci markers from eight chromosomes of A. flavus according to

Grubisha and Cotty (2009) and Islam et al. (2018). Simple sequence repeat multiplex PCR and genotyping were conducted as previously described (Grubisha and Cotty, 2009; Islam et al., 2018). At least 20% of the isolates representative of distinct haplotypes were subjected to three independent amplification and size determination runs for all loci for verification of results.

Multilocus SSR haplotypes (genotypes) were identified using HAPLOTYPE-

ANALYSIS V 1.05 (Eliades and Eliades, 2009). Sample correction was performed to include each haplotype only once for each sample prior to genetic analysis.

Locus and region-wide allelic and genetic diversity, including the number of alleles, number of private alleles, and haploid genetic diversity (H) were calculated using

GenAlEx version 6.51b2 (Peakall and Smouse, 2006; Smouse et al., 2017). In order to compare allelic diversity between A. flavus S and L morphotypes, SSR

144 data of 391 L morphotype isolates, available as part of an ongoing SSR database project at the USDA-ARS Aflatoxin Laboratory, was used. These isolates originated from soils in Arizona (113 isolates, collected from 1997 to 2000), maize in southeastern USA (162 isolates, collected 2015 to 2016), and maize and soils in Texas (116 isolates, collected from 2017 to 2018).

Aflatoxin production in maize

Fungi were selected for aflatoxin assays on maize based on results from phylogenetic reconstruction. Four fungal isolates were selected from each phylogenetically distinct clade of S morphology fungi detected within the US, with isolates originating from different states (when possible) or different regions from the same state. Isolates were inoculated on sterile maize as described previously

(Probst and Cotty, 2012; Singh et al., 2018). Briefly, healthy and undamaged maize kernels (Pioneer hybrid N82VGT) were autoclaved in Erlenmeyer flasks (5 g per flask) for 20 min at 121°C. Each flask was inoculated with 100 µl of conidial suspensions adjusted to 106 conidia ml 1. Water content of maize was adjusted to

30%. Maize cultures were incubated for 7 d at four temperatures, 25°C, 30°C, 35°C and 40°C, in the dark to allow fungal growth and aflatoxin formation. Treatments were represented by S morphology communities from the US incubated at the above-mentioned temperatures and each treatment consisted of four isolates.

Each experiment was performed twice.

Fungal growth was stopped at the end of the incubation period by addition of 50 ml of 85% acetone. Maize-fungus cultures were ground to homogeneity in a

145 laboratory grade Waring Blender (seven-speed laboratory blender, Waring

Laboratory, Torrington, CT) at full speed for 30 s and left in the dark for at least 1 h to allow for fungal cell lysis and release of aflatoxins. The culture filtrates were separated on thin-layer chromatography (TLC) plates (Silica gel 60, EMD,

Darmstadt, Germany) alongside aflatoxin standards (Aflatoxin Mix Kit-M, Supelco,

Bellefonte, PA, USA) containing a mixture of known concentrations of aflatoxins

B1, B2, G1, and G2. TLC plates were developed in ethyl ether-methanol-water

(96:3:1), air-dried, and aflatoxins were visualized under 365-nm UV light. Samples initially negative for aflatoxins were diluted with an equal volume of water and extracted twice with 25 mL of dichloromethane. The extracts were passed through a bed of anhydrous Na2SO4, dried and resuspended in a volume of dichloromethane that allowed accurate quantification. Extracts were spotted on

TLC plates as mentioned above, visualized and quantified with a scanning densitometer (TLC Scanner 3, Camag Scientific Inc., Wilmington, NC, USA).

Taxonomy

Based on the phylogenetic analysis described above, morphology and growth of fungal isolates belonging to a distinct, previously undescribed taxon was characterized by incubating isolates for 7 d in the dark at 25°C, 30°C, 37°C and

42°C following center point inoculation on Czapek agar, Czapek agar with 0.5% yeast extract (CYA), Malt agar and V8 agar (5% V8 juice and 2% agar, pH = 6.0) according to Singh et al. (2018). Growth was evaluated by measuring colony diameters (four replicates per isolate) at each temperature.

146

All isolates were analyzed for production of aspergillic acid by inoculation onto Aspergillus flavus and parasiticus agar (AFPA) (Pitt et al., 1983) and incubated for 7 d in the dark at 25°C, 30°C, and 37°C. Isolates were replicated four times at each temperature.

For micromorphological observations, mounts of 3 d old cultures grown on

Czapek agar were made in lactophenol cotton blue with a drop of ethanol to wash excess conidia. Fungal structures were viewed and captured with a differential interference contrast Leica DMI6000B microscope (Leica Microsystems, Buffalo

Grove, IL, USA) equipped with a Hamamatsu Flash 4.0 (Hamamatsu Corporation,

Bridgewater, NJ, USA) digital camera and the software package Leica Application

Suite LAS v 3.3.

Aflatoxin production by all isolates of the new taxon was evaluated by inoculating 100 µl of conidial suspensions (106 conidia/ml) on sterile maize (5 g flask-1) followed by incubation at 31°C for 7 d. Aflatoxins were extracted and quantified at the end of incubation period as described above.

Data analysis

Total aflatoxins were measured in µg/kg. Aflatoxin concentrations produced by each phylogenetic group at individual temperatures were subjected to analysis of variance and p = 0.05), as implemented in JMP 11.1.1 (SAS

Institute, Cary, NC, USA, 2013). Fungal colony diameters were measured in millimeters (mm). Aflatoxin concentrations were tested for normality prior to

147 statistical analyses and log-transformed if required. True means are presented for clarity.

Results

Species of S morphology fungi in the US

Phylogenetic reconstruction using individual and concatenated cmdA and niaD sequences resolved US S morphology isolates into four distinct clades

(Figure 5.2). Both Bayesian inference and maximum likelihood analyses yielded similar topologies for individual and concatenated trees with high Bayesian posterior probabilities and bootstrap support for each of the four clades. The first clade (Group 1) consisted of B aflatoxin producers and was identified as A. flavus based on reference isolates (e.g. NRRL 3251, AF70, AF42 and AF12). Aspergillus flavus S morphotype was the predominant S morphology fungus detected in the

US (Table 5.4). The A. flavus S morphotype was detected in Arizona, southeastern

US, and Texas and was the only species with S morphology detected in Arizona

(Table 5.4).

The second clade (Group 2) was comprised of isolates that also produced only B aflatoxins and was sister to A. flavus. Isolates in this clade originated from soils and maize in Texas. This clade also included six isolates previously reported as A. flavus S morphotype, two of which were from Texas (TXA35-K and TX06CB

9-G) and four others that were from Thailand (Sukhothai19, Sanpatong22, Ubon3 and Yuin20). However, this clade was phylogenetically distinct from A. flavus in both individual and concatenated phylogenies of partial sequences of cmdA and

148 niaD and was therefore identified as a novel taxon, Aspergillus agricola (see

Taxonomy).

The third clade (Group 3), previously described as A. texensis (e.g. reference isolate NRRL 66855), consisted of B and G aflatoxin producers. Isolates belonging to A. texensis originated from Texas and the southeastern US. The fourth and final clade (Group 4) consisted of B aflatoxin-producers that were recovered from maize and soil samples from the southeastern US and Texas.

Reference isolates of LAF from Probst et al. (2012) were included in the phylogenetic analysis and these LAF isolates resolved into multiple lineages in

Bayesian and ML phylogenies (Figure 5.2). Five of these isolates previously reported as LAF by Probst et al. (2012) (two Kenyan isolates K44-K and K849-B, and three isolates from Texas TX07CB73-I, TXLaFeria 2-F and TX04A5-B) grouped with isolates in clade four. The remaining Kenyan LAF isolates resolved into multiple lineages, which were distinct from the fourth clade (Figure 5.2). It was therefore concluded that group four may represent a new lineage. Aspergillus texensis was sister to this new SB lineage (Group 4). Both A. texensis and isolates from the new SB lineage were closely related to A. minisclerotigenes (e.g. NRRL

A-11611) and to Kenyan S isolates (e.g. K784-D, K805-E, K108-H and K771-B).

Each of the four clades of S morphology fungi detected within the US were distantly related to S morphology fungi that have primarily been reported from Africa (A. aflatoxiformans, A. cerealis).

Deletions in the norB-cypA region of aflatoxin cluster

149

Polymerase chain reaction amplification of the norB-cypA region of the aflatoxin biosynthesis cluster using primers from Ehrlich et al. (2004) and alignment with reference isolate A. parasiticus SU-1 revealed a 1.5 kb deletion in

A. flavus S morphotype isolates, a 0.9 kb deletion in A. agricola and an intact norB- cypA region in A. texensis. Four isolates in clade four along with both the Kenyan isolates did not yield an amplification product with primers from Ehrlich et al. (2004) but did amplify with primers reported by Probst et al. (2012). These amplicons revealed a 2.2 kb deletion in these isolates. However, several isolates in clade four failed to yield amplification products with either sets of primers suggesting a distinct structural variation (indel, inversion or translocation) from other B aflatoxin producers. This discovery further validates that clade four represents a new lineage of S morphology fungi with B aflatoxin producing ability. Based on the phylogenetic analyses and deletions in the norB-cypA region of the aflatoxin biosynthesis cluster, 89.7% of the 494 US S morphology isolates were identified as A. flavus S morphotype, 2.4% as A. agricola, 2.2% as A. texensis and 5.7% as the new SB lineage (clade four).

Aflatoxigenicity of S morphology species

S morphology fungi from each of the four clades produced high concentrations of aflatoxins at 25°C, 30°C and 35°C on maize (Table 5.5).

Aspergillus texensis, A. agricola and the new SB lineage produced the highest concentrations of aflatoxins at 30°C whereas maximum aflatoxin production by A. flavus S morphotype occurred at 35°C (Table 5.5). Aflatoxin production by S

150 morphology fungi from each clade was at least 90 times higher at 30°C and 70 times higher at 35°C compared to that at 25°C. The total concentrations of aflatoxins produced did not differ among species at 25°C (p = 0.302) or 30°C (p =

0.106). The A. flavus S morphotype produced significantly higher concentrations of aflatoxins than A. texensis, A. agricola and the new lineage at 35°C (p < 0.01).

Aflatoxin production was low for A. texensis, A. agricola and the new SB lineage at

40°C (less than 100 µg/kg total aflatoxins), although A. flavus S morphotype isolates still produced significantly high concentrations of aflatoxins at 40°C (Mean

= 6,519 µg/kg; range = 3,558 to 14,118 µg/kg total aflatoxins; p < 0.001).

Genetic diversity within A. flavus S morphotype

Simple sequence repeat primers previously designed for the L morphotype of A. flavus successfully amplified all the 17 loci of A. flavus S morphotype isolates

(n = 443). Amplifications were free of PCR artifacts, and final primer combinations in multiplex PCRs generated only a single peak in the expected size range for each locus. There were no missing data or null alleles. Aspergillus flavus S morphotype isolates were diverse at 17 SSR loci with 3-20 alleles per locus, and 0.089 to 0.811 haploid gene diversity (Table 5.6), although the L morphotype of A. flavus displayed greater diversity with higher number of alleles per locus (7-38) and haploid gene diversity of 0.525 to 0.908 at each SSR locus.

Among 443 isolates, 202 haplotypes were identified. Sixteen haplotypes were detected in more than one location, with three haplotypes, including the most frequent one, present in all locations sampled (Arizona, southeastern US and

151

Texas). Five haplotypes were shared between Texas and Arizona, six between

Texas and the southeastern US, and two between Arizona and the southeastern

US. Seventy-four haplotypes were unique to Arizona (Table 5.6) of which 20 were displayed by two or more isolates. Out of 15 and 98 private haplotypes in the southeastern US and Texas, respectively, two haplotypes in the southeastern US and 28 in Texas were detected more than once (Table 5.6). Haploid genetic diversity varied among locations in an area-wide comparison, with the highest diversity in Texas, followed by the southeastern US and Arizona (Table 5.6).

Taxonomy Aspergillus agricola P.Singh, M.J.Orbach, K.A.Callicott and P.J.Cotty sp. nov.

Figure 5.3.

Etymology: the type was isolated.

Diagnosis: Aspergillus agricola is closely related to A. flavus. Aspergillus flavus grows faster than A. agricola at 42°C on Czapek agar, CYA, Malt and V8 agar.

However, A. agricola grows faster than A. flavus at 25°C on Malt and V8.

Aspergillus Agricola produces visibly more conidia on CZ, CYA and V8 than A. flavus S morphotype strains. The A. flavus S strains contains a 1.5 kb deletion in the norB-cypA region of the aflatoxin biosynthesis gene cluster, whereas A. agricola contains a 0.9 kb deletion.

Typus: United States of America, Texas, Coastal Bend, soil cropped to maize

(Zea mays), collected by P.J.Cotty (Type: CR9-G).

152

Colony characteristics: Aspergillus agricola colonies grown for 7 d, attained an average diameter of 59 mm (54 62 mm) at 25°C, 71mm (59-74) at 30°C, 63 mm

(60 67 mm) at 37°C, and 9 mm (6-10 mm) at 42°C on Czapek agar. Colony diameters were greater on Czapek agar with yeast extract (CYA) medium at these temperatures. Maximum radial growth of A. agricola occurred at 30°C on all media.

Colony surface was velvety on CYA and V8 (Figure 5.3) while the colony reverse ranged from buff to brown on CZ, CYA and V8.

Micromorphology: Abundant production of dark black sclerotia (150 350 µm) on the agar surface was observed on Czapek, CYA, Malt, and V8 agar (Figure

5.3). Fungal isolates produced light green conidia on all media tested; conidia were circular and smooth walled (2.5-5 µm diameter). Vesicle globose, 30 70 µm in diameter. Conidiophores with smooth stipes, hyaline, 300 600 × 4 7 µm, phialides

6 8 × 2 4 µm. Aspergillus agricola produced abundant dark sclerotia and green conidia on maize after incubation at 31°C for 7 d (Figure 5.3g).

All isolates of A. agricola produced a bright-orange reaction on the reverse side of AFPA medium as a result of production of aspergillic acid (Figure 5.3b).

Colony texture on AFPA after 7 d at 25, 30 and 37°C was floccose with abundant white mycelia, dark sclerotia around the inoculation point but mostly immature white sclerotia on the colony surface.

Aspergillus agricola produced B1 and B2 aflatoxins in maize. Aflatoxin concentrations ranged from 96,434 648,171 µg/kg AFB1 (mean = 269,982 µg/kg) and 1,860 23,951 µg/kg AFB2 (mean = 11,287 µg/kg).

153

Discussion

With the utilization of molecular (norB-cypA deletions), phylogenetic (cmdA and niaD genes) and aflatoxin analyses, the current study surveyed a large collection of S morphology fungi within Aspergillus section Flavi in the US and detected the occurrence of four genetically distinct groups (Figure 5.2). Each of these groups was monophyletic with high Bayesian posterior probability and bootstrap support in a multigene phylogeny constructed with a total of 4.0 kb of sequence data from two unlinked genes, cmdA (1.9 kb) and niaD (2.1 kb). The A. flavus S morphotype was the dominant S morphology species in the US, followed by a new SB lineage reported in the current study, A. agricola sp. nov. and A. texensis (Table 5.4). Aspergillus flavus S morphotype strains were detected in all three geographic areas analyzed, Arizona, Texas and the southeastern US, indicating superior adaptability to diverse environments and niches. The newly described species, A. agricola was only detected in Texas while A. texensis and the new SB lineage occurred in both Texas and the southeastern US. The underlying mechanisms behind this distribution remain unclear but may be attributed to differences in environmental conditions (e.g. average daily temperature, humidity), host crop and/or soil characteristics (Cotty and Jaime-

Garcia, 2007; Horn and Donner, 1998, 1999; Orum et al., 1997; Schroeder and

Boller, 1973).

Aspergillus agricola is sister to A. flavus, a species which contains isolates with both L and S morphologies (Figure 5.2). Aspergillus agricola, like A. flavus,

154 produces only B aflatoxins. In A. agricola, this is due to a 0.9 kb deletion in the norB-cypA region of the aflatoxin gene cluster. Deletions in the cypA gene, which is located near the distal terminus of aflatoxin biosynthesis cluster, result in the inability to produce G aflatoxins (Ehrlich et al., 2004). Previous studies have reported deletions of 0.9 and/or 1.5 kb in both L and S morphotype strains of A. flavus in the norB-cypA region resulting in only B aflatoxin production. Although A. flavus L morphotype isolates exist that contain either one or the other deletion, all

S morphotype isolates that contain a 0.9 kb deletion, previously identified as A. flavus, are actually A. agricola, based on results of the current study. This included isolates from Texas (TXA35-K and TX06CB 9-G) and Thailand (Sukhothai19,

Sanpatong 22, Ubon3 and Yuin20) (Probst et al., 2012). Additionally, the current study detected a 1.5 kb deletion in the norB-cypA region of the S morphotype isolates from Philippines, which grouped with reference isolates of A. flavus in concatenated phylogenies of partial cmdA and niaD genes (Table 5.2).

Furthermore, screening of A. flavus S morphotype isolates from all of the 202 haplotypes in the current study revealed a 1.5 kb deletion in each haplotype, confirming the 1.5 kb deletion in the norB-cypA region of all A. flavus S morphotype isolates.

Aspergillus texensis, the only B and G aflatoxin producing S morphology species in the US, was sister to the new unnamed SB lineage, both of which are closely related to A. minisclerotigenes (Figure 5.2). All members of the new SB lineage produce only B aflatoxins and included some Kenyan isolates (K44-K and

155

K805-E) previously reported from the lethal aflatoxicosis episodes in Kenya (Figure

5.2 and Table 5.2) (Probst et al., 2012). These Kenyan isolates produce B aflatoxins due to a 2.2 kb deletion in the norB-cypA region (Probst et al., 2012). In addition to the Kenyan isolates, three isolates from Texas (TX07CB73-I,

TXLaFeria 2-F and TX04A5-B) were also reported to contain the 2.2 kb deletion

(Probst et al., 2012). However, in the current study, this deletion was not detected in these Texas isolates nor in 24 out of 28 other isolates in the new SB lineage

(clade four) (Table 5.2). The mutation resulting in loss of G aflatoxin production in these isolates is currently unknown and under investigation. However, in addition to the two Kenyan isolates, four other isolates from the new SB lineage contained the 2.2 kb deletion (Table 5.2), indicating that more than one genetic mutation is likely responsible for loss of G aflatoxin production in this lineage. A complete taxonomic description of this new lineage is recommended.

Both Bayesian and ML phylogenies based on concatenated cmdA and niaD genes showed that the previously reported Kenyan isolates belong to multiple lineages and that the 2.2 kb deletion is a paraphyletic character (Figure 5.2). While some of these lineages containing the Kenyan fungi are closely related to A. minisclerotigenes, isolates K44-K and K-805E belong to the new SB lineage discovered in the current study and are more closely related to A. texensis.

Although A. minisclerotigenes is known from South America, Sub-Saharan Africa,

Australia and Europe, it was not detected in the current study of US isolates.

156

Aflatoxigenicity of S morphology species from the US was similar at 25°C and 30°C whereas the A. flavus S morphotype was the most toxic at warmer temperatures (35°C and 40°C). Aflatoxin concentrations produced on maize at

25°C were high enough to be fatal to humans and animals, and each species produced more than 100,000 µg/kg total aflatoxins in maize at 30°C, with A. texensis producing the highest levels of aflatoxins. Previous studies have indicated a drastic decrease in aflatoxin production by Aspergillus isolates at temperatures above 32°C and higher (Joffe and Lisker, 1969; OBrian et al., 2007; Schindler et al., 1967). In contrast in the current study, aflatoxin production at 35°C was still high and unacceptable, even though aflatoxin concentrations produced by A. texensis, A. agricola and the new SB lineage at 35°C were lower than that at 30°C

(Table 5.5). In fact, the A. flavus S morphotype produced the highest concentrations of aflatoxins at 35°C and even at 40°C, A. flavus S isolates produced more than 3000 µg/kg total aflatoxins. This is in contrast with results from previous studies that were unable to detect aflatoxins at or above 37°C (Joffe and

Lisker, 1969; OBrian et al., 2007; Schindler et al., 1967). Failure to detect aflatoxins in these studies may be due to methods to assay aflatoxin production, including production in submerged liquid cultures, rather than on a susceptible host. Probst and Cotty (2012) have demonstrated that aflatoxin-production in various fermentation media does not correlate with toxigenicity of isolates in a viable host.

Unlike the A. flavus S strains, aflatoxin production by each of the other three S morphology species was inhibited at 40°C suggesting that these species may not

157 be of significant concern in terms of crop contamination under very hot environmental conditions. The A. flavus S morphotype has been held responsible for severe contamination of cottonseed and maize in the US (Cotty, 1996; Jaime-

Garcia and Cotty, 2003). In the current study, members of this species comprised almost 90% of all S morphology fungi in the US and were detected across all the regions sampled. This combined with its aflatoxin-producing potential from 25 to

40°C clearly indicates that the S morphotype of A. flavus should be the main target of aflatoxin mitigation strategies in the US.

SSR markers previously designed for the L morphotype of A. flavus were applied for the first time to the A. flavus S morphotype isolates to assess genetic diversity within the S morphotype individuals in comparison with that of the L morphotype fungi. Genetic diversity among A. flavus L isolates resident in soils and crops has primarily been assessed using vegetative compatibility analyses and previous studies have reported high VCG diversity of A. flavus L morphotype populations examined from warm-agroecologies (Bayman and Cotty, 1991; Novas and Cabral, 2002; Picot et al.,2018; Probst et al., 2011). A recent study reported high genetic diversity among 2,744 A. flavus L morphotype isolates from Kenyan soils based on the SSR markers, with 9 to 72 alleles detected per locus (Islam et al., 2018). The current study detected 7 to 38 alleles per locus in 391 L morphotype fungi, but only 3 to 20 alleles in 443 S morphotype isolates. Haploid gene diversity was also higher in A. flavus L than S isolates. Differences in genetic diversity between the L and S morphotype isolates of A. flavus may be attributed to higher

158 population size of the L isolates when isolated from crop and soil samples; several studies investigating diversity of fungi within section Flavi from crops or soils in a region have reported predominance of A. flavus L isolates, whereas incidence of fungi with S morphology has often been low (Atehnkeng et al., 2008; Cardwell and

Cotty, 2002; Donner et al., 2009; Probst et al., 2014). Region-wide haploid genetic diversity among A. flavus S isolates was highest in Texas, where the largest number of isolates originated, followed by the southeastern US and Arizona. Out of the 202 haplotypes detected, most were private to a given region, with only three haplotypes shared across all regions. While all Texas A. flavus S isolates were recovered from maize and soils cropped to maize and all the southeastern US isolates were recovered from maize, all isolates from Arizona were from soils cropped to cottonseed. Differences in temperature, soil properties, average precipitation and crops grown in a region may support certain fungal genotypes versus others (Donner et al., 2009; Jaime-Garcia and Cotty, 2010; Mehl et al.,

2012; Mehl and Cotty, 2013). This is further strengthened by our discovery that only A. flavus S morphotype isolates were detected in Arizona, while sympatric occurrence of distinct aflatoxigenic species was detected in Texas (four species) and the southeastern US (three species), regions differing in environmental conditions and crops grown.

Identification and characterization of aflatoxigenic fungi facilitates development of management procedures for prevention of aflatoxin contamination of crops. Relative frequencies of aflatoxigenic genotypes can vary among crops,

159 regions, seasons and years such that a continuously fluctuating assembly of genetically diverse aflatoxin-producers may exist in fields/regions. The current study provides insights into the diversity and incidence of highly toxigenic S morphology fungi from regions of the US that suffer from perennial risk of aflatoxin contamination. Four phylogenetically distinct species with differences in aflatoxin profiles and norB-cypA deletions in the aflatoxin cluster were discovered.

Aflatoxigenicity of each of the S morphology groups detected within the US even at 35°C clearly suggests their potential to contaminate crops under warm conditions, while the A. flavus S morphotype can be a serious concern even at

40°C. Fungi with S morphology are favored by warm environments (Bock et al.,

2004; Orum et al., 1997). Higher incidence of the A. flavus S morphotype at elevated soil temperatures has been reported (Jaime-Garcia and Cotty, 2010).

Similarly, A. aflatoxiformans, A. minisclerotigenes and the fungi responsible for deadly aflatoxicosis in Kenya have S morphology and are associated with aflatoxin contamination of crops in semi-arid and sub-humid regions of Sub-Saharan Africa

(Agbetiameh et al., 2018; Cardwell and Cotty, 2002; Donner et al., 2009; Probst et al., 2014; Singh and Cotty, 2019). Although incidence of A. texensis, A. agricola and the new SB lineage were overall low, higher proportions may be favored by prolonged periods of hot and dry conditions, exacerbating crop contamination

(Jaime-Gracia and Cotty, 2010). The occurrence and incidence of the S morphology fungi in temperate regions of the US remains unexplored. However, these fungi may become a serious threat to food safety and food security in

160 currently un-affected regions of the US with the predicted warmer environment under climate change (Cotty and Jaime-Garcia, 2007; Jaime-Gracia and Cotty,

2010). Aflatoxin management using atoxigenic isolates of A. flavus must consider the occurrence of these genetically distinct S morphology fungi in US soils and crops, and their aflatoxigenicity under a range of environmental conditions. Active ingredients of the biological control agents should be selected for adaptive traits, such as long-term persistence and survival under harsh conditions (elevated temperatures) in target regions to achieve long-term reductions in the aflatoxin- producing potential of Aspergillus communities and genotypes associated with crop contamination.

161

Table 5.1 Isolate information for 494 S morphology isolates from Aspergillus section Flavi used in the current study.

County/ Total State Source Year # Isolates Region Isolates

Alabama Blount Maize 2015 1 27 Boaz 1 Elmore 1 Lawrence 3 Morgan 10 Jackson 2017 1 Unknown 11

Arizona Yuma Soil 1989 3 145 La Paz 2002 12 Maricopa 24 Mohave 4 Pinal 19 Unknown 1 Yuma 12 La Paz 2003 5 Maricopa 10 Mohave 2 Pinal 5 Yuma 5 Buckeye 2004 1 Mohawk Valley 1 Unknown 3 Yuma 1 Aztec 2005 1 Unknown 30 Maricopa Unknown 1 Mohawk Valley Unknown 5

Arkansas Greene Maize 2015 6 7 Pulaski 1

Georgia Bryan Maize 2015 2 5 Bulloch 2 Washington 1

Louisiana Madison Maize 2015 16 25 Morehouse 6 Unknown 2017 3

162

Mississippi Perry Maize 2015 3 4 Warren 1

Texas Coastal Bend Soil 2004 4 281 Dacosta 2 Edroy 4 Gregory 9 Hidalgo 4 Placeto 3 RGV 3 St Paul 5 Unknown 3 Coastal Bend Soil 2006 3 Dacosta Soil 2007 3 Driscoll 1 Edroy 4 Ganado 1 Gregory 2 Placeto 1 St Paul 2 Taylor 3 Unknown 1 Bee Maize 2008 1 Calhoun 1 Ellis 1 Fort Bend 1 Grayson 1 Hidalgo 1 Jackson 1 Matagorda 1 Victoria 1 Wharton 1 Unknown 1 Bee Maize 2009 1 Colorado 1 Dorchester 2 Ellis 1 Fannin 1 Fort Bend 1 San Patricio 1 Wharton 1 Williamson 1 Bee Maize 2010 1 Cameron 1 Hidalgo 1 Live Oak 1 Ellis Soil 2012 114 Ellis Soil 2013 54

163

Ellis Soil 2014 28 Edroy Soil Unknown 2

164

etal., 2012,

Citation et al., Probst 2012, StudyCurrent et al., Probst 2012, Study Current et al., Probst 2012, StudyCurrent et al., Probst 2012, StudyCurrent Probst StudyCurrent Study Current Study Current

norB-cypA norB-cypA (kb) Deletion 1.5 1.5

B 1.5 Current Study Current B 1.5 B B

165 165 Arizona, USAArizona, Georgia,USA Mississippi, USA

Soil Maize isolates used for phylogenetic reconstruction in the current study. the study. current phylogeneticreconstruction in usedfor isolates Flavi ore rgn AF Origin Source b

1 Soil Arizona, USA B 1.5 Cotty, 1989 Cotty, 1989 Cotty, 1989 Cotty, 1.5 B USA Arizona, 1.5 1.5 B B Soil USA Arizona, USA Arizona, Cottonseed Soil 1 1 1 1 1 section section

K - 5 -

® ® ®

E Aspergillus Group 1803SW a - Hesseltine 1970et al., 1.5 1.5 B 1.5 USA California, B 1.5 B Walnut Philippines B Philippines 1.5 Philippines 1.5 Soil B 1 Study Current Soil Isolate B Study Current Philippines Soil Study Current Philippines Study Current 3251NRRL 1 = ATCC AF12 1.5 MYA382 Soil Study Current 1 = ATCC AF42 1.5 Study Current Study Current Soil B MYA383 Study Current 1.5 1 = ATCC AF70 B USA Arizona, Study Current 1.5 MYA384 USA Alabama, B L1A3 B Arkansas,USA 1 1.5 1.5 1.5 Soil 1 Maize L1D2 USA Texas, 1.5 Maize B B B 1.5 L2E1 B Maize Maize USA Texas, B USA Texas, USA Texas, MINIC4 USA Texas, 1 1 Maize USA Texas, V2D2 Maize Maize 1 Maize 31520-2405-2-H 30609 1 Maize 1 C GNFHP4 H VCSPWS AGSTNW 1 1 1 GVPE01D 1 1 BA12-J C1-G WX13 MX1-4B-J WX13 A10-A-S E28-L J12-E Table 5.2 5.2 Table

Current Study Current Cotty, Singh and 2019 Study Current et al., Probst 2012 Singh et 2018 al., Singh et 20 al.,

1.5 0.9 0.9 0.0 0.0

B B B B B,G B,G

166 166 Texas, USA Texas, Pakistan USA Texas, Thailand USA Louisiana, USA Texas,

Maize Chili Maize Soil Maize Soil

1 Soil USA B 0.9 Cotty, 1989 Cotty, 0.9 B USA Soil 1 1 1 2 2 3 3

F K

- - E - BC31-E 1 Maize Texas, USA B 1.5 Current Study Current Cotty, Singh and 2019 et al., Probst 2012 1.5 Study Current Study Current B Study Current et al., Probst 2012 Study Current USA Texas, B Study Current 0.9 Study Current Study Current 0.9 Maize 0.9 Study B Current Unknown 0.9 Study Current B 0.9 0.9 B USA Texas, B 0.9 Study Current B USA B Texas, Chili 0.9 USA Texas, Study Current USA Texas, 0.9 B 1 USA Texas, USA Texas, 0.9 B Soil et al., Probst 2012 et al., Probst 2012 0.9 B USA Texas, Soil Soil B USA Texas, Soil Maize B 0.9 USA Texas, 1 Soil USA Texas, 0.9 Soil BC31-E B USA Texas, 2 Soil AT52 0.9 Singh et 2018 al., 0.9 B Soil= ATCC AF13 2 USA Texas, 2 Soil Singh et 2018 al., 96044 B 2 B 2 Soil CHL159 Singh et 2018 al., USA Texas, 2 CHL187 Singh et 2018 al., Maize 2 Singh et 2018 al., TXA35-K Thailand 2 0.0 Thailand TX06CB 9-G Singh et 2018 al., Singh et 2018 al., Maize 2 2018 al., Singh et B,G Singh et CR9-G 2018 al., 2 0.0 E27-J 2 B,G 0.0 C3-J USA Texas, Soil Soil 0.0 B,G 2 A2-A 0.0 et al., Probst B,G 2007 USA Texas, J15-H B,G et 0.0 al., Probst 2007 0.0 J11-B USA Texas, 2 0.0 0.0 Soil B,G B,G J11-C USA Texas, B,G USA Texas, Maize B,G Arkansas,USA J12-F 2 E13-L 2 USA Texas, Soil USA Texas, USA Texas, BC7-C 2.2 Maize Soil BC09 Soil 2.2 EC37-C 3 B Soil Sukhothai19 3 Soil B Soil Sanpatong22 Ubon3 3 66855NRRL 3 3 Kenya 66856 3NRRL Kenya 66857NRRL 3 66858NRRL 3 3 Maize 66859NRRL Maize J35 VC16-A CTL-1I Q P2R2-A LAF 1-1-O LAF 1-1L A1170 K805-E/ K784-D/ A1168

Probst et al., Probst 2012 etal., Probst 2012, Study Current et al., Probst 2007, Study Current et al., Probst 2012, Study Current et al., Probst 2012, Study Current et al., Probst 2012, Study Current Study Current Study Current Study Current Study Current Cardwell, and Cotty 1999 Cardwell, and Cotty 1999

2.2 Unknown Unknown 2.2 Unknown

B B B B B

167 Kenya USA Texas, USA Texas, USA Texas, USA Texas,

Maize Soil Soil Soil Maize

LAF 4 0.0 B,G 0.0 4 B,G Benin 4 Benin Soil 4 Soil AA AA

MYA379 MYA380 D G C

® ® B - - - - B - K108-H LAF Maize Kenya B 2.2 Probst et al., Probst 2012 2.2 2.2 B B 2.2 Kenya Kenya B Maize Unknown Maize B Unknown Kenya Study Current USA Texas, LAF B Study Current Study Study Current Current Unknown Study Current Maize Study 4 Current USA Texas, Unknown B Maize Study Current Unknown Unknown Study Current Unknown Study Current B Unknown Study Current USA Texas, Unknown USA Louisiana, B B Cottonseed K108-H B Soil Unknown USA Louisiana, Study Current B K771 Unknown B USA Louisiana, 4 USA Texas, Unknown Maize USA Louisiana, K44-K B USA Texas, 4 Study Current B Maize USA Louisiana, Unknown 2.2 B Maize A1171 K849-B/ Study Current Study Current Maize USA Texas, 4 Soil B 4 Study Current B Maize Unknown USA Texas, Soil TX07CB73-I 4 USA Texas, Unknown USA Texas, B TXLaFeria 2-F 4 Soil 4 Soil B TX04A5-B USA Texas, 2.2 4 4 Hesseltine 1970 et al., 2.2 Soil 4 Soil A BRG3458 USA Texas, 4 B E BRG3458 B H BRG3458 Soil 4 USA Texas, J BRG3458 Maize 4 USA Texas, JBRG5138 A5-B-S 0.0 4 Maize 4 CR20 B,G Maize D16-J 4 D25-A-S 4 E21 A34-N Nigeria CR24-F Groundnut 4 J15-B 4 CR10 T01-M VC8-L EC24 AA EC49-L BG14-F A-11612NRRL = BN008R ATCC =BN038G ATCC Cotty Cardwell, and Cotty 1999 Dawson, Peanut Lab, (NRRL GA database) Dawson, Peanut Lab, (NRRL GA database) Carvajal-Campos et Frisvad2017; al., et 2019 al., Carvajal-Campos et Frisvad2017; al., et 2019 al., Carvajal-Campos et Frisvad2017; al., et 2019 al., , 0.0 B,G B,G 0.0 0.0 B,G 0.0 B,G 168 168 USA B,G 0.0 Kurtzman Kurtzman 1987 et al., etal., Peterson 2001 0.0 B,G 0.0 B,G USA Japan Frass, Frass, Silkworm AA Soil Benin B,G 0.0 0.0 B,G Benin Soil AA MYA381 ® BN009-E AP Soil Benin B,G 0.0 None None Jong,and Wei 1986 et1974 Ramboal., 0.0 0.0 B,G B,G Geiser 1998et al., 0.0 Cotty, Singh and 2019 0.0 B,G Benin Cotty, Singh and 2019 0.0 B,G USA Cotty, et Singh and 2019 al., Probst 2012 Hesseltine 1970 et al., B,G Cotty, Singh and 2019 0.0 2019 Cotty, Singh and B,G Soil Uganda 0.0 USA Groundnut 0.0 B,G 0.0 0.0 Australia B,G 0.0 USA 0.0 B,G B,G B,G AP Soil 0.0 B,G B,G Nigeria Soil Soil AP Nigeria Argentina AP =BN040B Nigeria Groundnut Nigeria ATCC Nigeria Groundnut Chili AP BN009-E Nigeria Chili 2999NRRL AM Chili AP 465NRRL Chili AM 29538NRRL Chili AM AM 29590NRRL AM AM Groundnut A-11611NRRL AM TAR3N43 AM 4-2 CHL583 Groundnut CHL663 CHL707 AC CHL845 Groundnut CHL895 AC Wheat 66708 NRRL AC 66709NRRL AN AB 66710NRRL 13137NRRL 26010NRRL Fungal Fungal isolates with S morphology from the current study and reference fromisolates previously described aflatoxin- producing species. species. producing a A. A. bombycis. A.bombycis. AB- A. nomius, A.nomius, AN- A.parasiticus, AP- 169 169 A. minisclerotigenes, A.minisclerotigenes, , AM- , A.cerealis AC- AC- Group indicates phylogenetic groups identified in Figure 5.2 and other aflatoxin-producing species. AA- AA- species. aflatoxin-producing other and 5.2 Figure in identified groups phylogenetic indicates Group Aflatoxin Aflatoxin profiles of isolates. B refers to B aflatoxin production and BG refers to production of both B and G aflatoxins. aflatoxiformans, aflatoxiformans, b c Reference

(°C) a 56 Probst et al., Probst 2012 et al., Probst 2012 56 Study Current 48 et al., Probst 2012 50 et al., Probst 2012 52 Cotty, Singh and 2019 57 Ehrlich et 2004 al., 55 et al., Probst 2012 58 62 ) for PCR amplifications of target genesfrom of target amplifications PCR for ) a

170 170 GGGACCCTTTTCCGGTGCGG GGCTGCACGCCCAATGCTTC

- - F-GGCCTTCTCCCTATTCGTAA R-CTCGCGGATCATCTCATC F-GGCTGGATGTGTGTAAATC R-ATTGGTCGCATTTGAAGGG F-GTTAGTGGTTAGTCGCAG R-CTTCAGCTCTCTGGAATC F-CGGACGATAAGCAACAACAC R-GGATGAACACCCGTTAATCTGA F-ACGGCCGACAGAAGTGCTGA R-TGGGCGAAGAGACTCCCCGT F-GCAGCCCAATGGTCACTACGGC R CCACC GTGCCCAGCATCTTGGT F- R-AAGGACTTGATGATTCCTC F R-GGCGGCCCCTCAGCAAACAT . . Flavi agtGn Sqec T Sequence Gene Target cmdA niaD norB-cypA norB-cypA section section

Primers and locus specific (Tannealing Primers temperatures and locus PrimerPair cmd42-637 cmd2F-2R cmd3F-3R niaDF-AR niaDBF-BR niaDCF-CR AP1729-3551 CP5F-R Table 5.3 Table Aspergillus

A. agricola A.

B (%) b New S New Lineage Lineage in United States. the in Species/Lineage Flavi A. texensis A. section section A. flavus A. 171 171 # of # Aspergillus Isolates ore Year Source a Incidence of S morphology fungi within within fungi ofSIncidence morphology

Texas Maize & soil 2004-14 281 88 3.2 5.3 4.3 4.3 5.3 3.2 88 281 2004-14 soil Maize& Texas rzn Si 20-5 4 10 0 0 0 0 100 145 2002-05 Soil Arizona Location otesenUA az 21-7 8 8 1 0 19 3 78 68 2015-17 Maize SoutheasternUSA Assignment of fungal isolates to species/lineage based on molecular and phylogenetic analyses in the the phylogenetic in molecularto analyses on and species/lineage based isolates fungal of Assignment Location of origin of isolates in the United States. Southeastern USA includes Alabama, Arkansas, Alabama, USAincludes Arkansas, States.Southeastern the United of origin isolates of in Location Table 5.4 5.4 Table Mississippi. and Louisiana Georgia, study. current a b

6,519A 21 72 99B 16

26,735 67,155 96,187B 74,453

54,657 141,992 128,318 191,641 Total (µg/kg) Aflatoxin Total

172 172 5C 0C 5C 40°C 35°C 30°C 25°C 356,751A 162,981 1,152 648 1,021 1,329 26B 67,162B 1,225 231,740 943

Calhoun, mMaizeTX, Coastal soil Bend, TX, Fort Bend, maize TX,

G L - - AF70 AZ, Yuma, soil 1,849 313,371 530,896 4,119 4,119 530,896 313,371 4,282 24B 1,849 189,690 14,118 97,296B 3,558 19 131,868 306,738 164,807 399,681 103,266 80,555 103,421 Yuma, soil AZ, 751 790 167,163 1,211 San Patricio, TX, Soil 797 33 24 Morgan, AL, maize 596 186,866 AF70 95,029 maize Bryan, GA, C GNFHP4 45 190,850 246,556 E2-H Gregory, 38 soil TX, 31,649 JBABJB E 1,375 119,737 Coastal soil Bend, TX, Mean 542 37,092 159,572 241 J15-B 166,208 Franklin, maize LA, 1,860 CR20-D 41 944 174,615 H BRG3458 72,274 VC8 Gregory, soil TX, 1,492 Mean 262,663 Bee,maize TX, Bexar, maize TX, CR9 28 664 J55-H 63,677 BC7-C BC09-F 170,988 Mean Ellis, soil TX, 1,041 66855 NRRL Greene, AR, maize 66856NRRL 66859 NRRL Mean

Aflatoxin production by distinct groups of S morphology fungi in maize. maize. morphologyfungi bygroups in S distinct of production Aflatoxin S S

B region, source State, Isolate Species flavus A. morphotype 21 58,245 S New lineage 301,667 843 San Patricio, TX, soil agricola A. J35-E texensis A. Total aflatoxin concentrations were compared by column between species at each temperature. Mean aflatoxin aflatoxin Mean temperature. each at species between column by compared were concentrations differ aflatoxin column Total a within letters upper-case different by followed Means isolates. four of average is concentration Table 5.5 5.5 Table > > p 173 173 < 0.01). Means within a column lacking a letter do not differ do (ANOVA, differ a letter column a not lacking within Means <0.01). p p 0.05)

H 0.647

180 -

Size Size (bp) 157

# # Alleles 7

H 0.239

165 - S morphotype isolates recovered from across from across recovered morphotype isolates S

Size Size (bp) 157 S Morphotype L Morphotype L Morphotype S 174 174

# # Alleles 0.838 144-236 26 0.818 0.811 344-388 147-218 16 19 0.624 347-385 3 10

/1918

21 9-1 .5 420430.780 290-403 14 0.657 296-314 5 /2911 10 Aspergillus flavus flavus Aspergillus

8 1918 / 23 1 3748 .9 2 3548 0.864 365-438 20 0.795 367-408 12 /2634 /2911 4 144-188 0.544 11 144-222 0.736 0.736 144-222 11 0.544 144-188 4 /2911 20 1 1628 .8 2 1725 0.848 117-215 27 0.686 126-228 /2911 10 /2504 20 1 1612 .1 1 1318 0.525 113-148 12 0.516 116-152 11 0.650 165-428 /2504 13 0.482 169-206 5 /2541 8 /2634 12 299-358 0.736 26 299-391 0.759 0.759 299-391 26 0.865 0.736 139-250 299-358 28 12 0.539 143-296 /2634 19 /2634 8 (AGG) (GTT) /1866 18 122-182 0.757 19 122-200 0.856 0.856 122-200 0.634 19 128-163 0.757 122-182 11 18 0.527 131-154 /1866 5 /1918 13 16 4 12 4 25 2 1927 .7 3 1929 0.908 159-269 0.840 162-203 38 0.671 159-227 19 0.753 162-195 20 13 /2856 /1739 11 10 16 7 5-7 061 3 5-8 0.508 253-286 13 0.691 259-275 7 /1569 9 31 8 16 /2856 5 125-135 0.089 7 125-137 0.576 0.576 125-137 7 0.089 125-135 5 /2856 16 10 12 7

Repeat motif Repeat Scaffold* and (ACAT)

Chromosome 2L

Haploid diversity for 17 SSR loci of SSR17loci diversityHaploid for

F3 U (CTT) (GAG) SSR (TTTA) Locus (TTC) (GTC) 4U 7U AF13 3L AF43 6U AF22 AF31 3L AF34 AF54

(TTG) PCR (TCT) 1L Panel AF28 (TTC) (AAG) A (TTG) 2L (AGA) AF53 6U (AAG) 7L AF42 B AF8 (AT) (AC) 5L (AT) AF16 2L (GT) 1L AF17 D AF11 4L 2U AF66 2U AF64 8U AF63 E AF55 Table 5.6 5.6 Table the morphotype. L of diversity compared haploid to 2002-17 during States United the 2009 Cotty, and from*Data Grubisha each SSRlocus. at Number alleles of Alleles- # SSRbaselocus. pairs each of(bp) sizes at Range allele in Size- 175 175 H- Haploid gene diversity calculated using GenAlEx 6.51b2 (Peakall and Smouse, 2006; Smouse et al., 2017). al., Smouseand et Smouse,2006; GenAlEx (Peakall geneusing calculated diversity 6.51b2 Haploid H- Table 5.7 Genetic diversity of Aspergillus flavus S morphotype isolates recovered from soil and maize samples across the United States from 2002-

2017.

Location Source Year N NS NH NPH NA (range) HA

Arizona Soil 2002-05 145 145 84 74 5.118 (2-10) 0.505

Southeastern Maize 2015-17 54 38 26 15 5.412 (1-13) 0.577 US# Maize Texas 2004-16 245 206 112 98 8.588 (3-17) 0.612 & soil

#Southeastern US includes Alabama, Arkansas, Georgia, Louisiana and

Mississippi.

N- Total number of A. flavus S morphotype isolates from each location.

NS- Number of isolates after sample correction.

NH- Number of haplotypes.

NPH- Number of private haplotypes.

NA- Average number of alleles across 17 polymorphic SSR loci.

HA- Haploid gene diversity calculated using GenAlex 6.51b2 (Smouse et al.,

2017).

176

Figure 5.1 Locations where soil and maize samples were collected from 2002 to

2017. Filled circles indicate counties in each state.

177

Figure 5.2 Mid-point rooted Bayesian phylogeny of S morphology fungi and other aflatoxin-producers within Aspergillus section Flavi based on concatenated partial sequence of cmdA (1.9 kb) and niaD (2.1kb) genes. K1, K2 and K3 are lineages consisting of Kenyan fungi; K1: Isolates K805-E and K784-D, K2: Isolate K108-H and K3: Isolate K771-B (Probst et al., 2012). Values above nodes or before commas are Bayesian posterior probabilities and values below nodes or after commas are bootstrap support from 500 replicates.

178

Figure 5.3 (a-e) Colonies of Aspergillus agricola grown at 25°C for 7 d (top) and 3 d (bottom) on (a) czapek, (b) reverse on AFPA, (c) V8 agar, (d) czapek with yeast extract, and (e) malt agar. (f-g) A. agricola sclerotia on Czapek agar and maize grown at 31°C for 7 d; (h-i) conidiophores.

179

180

Literature Cited

Adhikari, B. N., Bandyopadhyay, R., & Cotty, P. J. (2016). Degeneration of aflatoxin gene clusters in Aspergillus flavus from Africa and North America. AMB Express, 6(1), 62.

Adye, J., & Mateles, R. I. (1964). Incorporation of labelled compounds into aflatoxins. Biochimica et Biophysica Acta (BBA)-General Subjects, 86(2), 418- 420.

Aflatoxin Prevention. Arkansas Corn and Grain Sorghum Board. Available online: http://www.corn-sorghum.org/aflatoxin-prevention-0.html.

Africa, A. (2016). Tanzania: Food poisoning linked to 14 deaths in 2 regions.

Agbetiameh, D., Ortega-Beltran, A., Awuah, R. T., Atehnkeng, J., Cotty, P. J., & Bandyopadhyay, R. (2018). Prevalence of aflatoxin contamination in maize and groundnut in Ghana: population structure, distribution, and toxigenicity of the causal agents. Plant disease, 102(4), 764-772.

Agrios, G.N. (2005). Plant Pathology, fifth ed. Elsevier Academic Press. USA.

Allcroft, R., & Carnaghan, R. B. A. (1962). Groundnut toxicity. Aspergillus flavus toxin (aflatoxin) in animal products: preliminary communication. Veterinary Record, 74, 863-864.

Anonymous. (2015). NEOGEN REVEAL Q+ FOR AFLATOXIN USING ACCUSCAN III and ACCUSCAN PRO READERS. https://www.gipsa.usda.gov/fgis/metheqp/instructions/8085%20AS%20III%20and %20Pro%20Revision%200.pdf.

Asai, T., Tsuchiya, Y., Okano, K., Piscoya, A., Yoshito Nishi, C., Ikoma, T., Oyama T., Ikegami K., & Yamamoto, M. (2014). Aflatoxin contamination of red chili pepper from Bolivia and Peru, countries with high gallbladder cancer incidence rates. Asian Pacific Journal of Cancer Prevention, 13(10), 5167-5170.

Ashworth Jr, L. J., McMeans, J. L., & Brown, C. M. (1969). Infection of cotton by Aspergillus flavus: epidemiology of the disease. Journal of Stored Products Research, 5(3), 193-202.

Atehnkeng, J., Ojiambo, P. S., Donner, M., Ikotun, T., Sikora, R. A., Cotty, P. J., & Bandyopadhyay, R. (2008). Distribution and toxigenicity of Aspergillus species isolated from maize kernels from three agro-ecological zones in Nigeria. International Journal of Food Microbiology, 122(1-2), 74-84.

181

Atehnkeng, J., Ojiambo, P. S., Cotty, P. J., & 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.). Biological Control, 72, 62-70.

Atehnkeng, J., Donner, M., Ojiambo, P. S., Ikotun, B., Augusto, J., Cotty, P. J., & Bandyopadhyay, R. (2016). Environmental distribution and genetic diversity of vegetative compatibility groups determine biocontrol strategies to mitigate aflatoxin contamination of maize by Aspergillus flavus. Microbial biotechnology, 9(1), 75-88.

aflatoxin B1 levels in powdered red pepper. Food Control, 18(9), 1015-1018.

Bailey, E. A., Iyer, R. S., Stone, M. P., Harris, T. M., & Essigmann, J. M. (1996). Mutational properties of the primary aflatoxin B1-DNA adduct. Proceedings of the National Academy of Sciences, 93(4), 1535-1539.

Bandyopadhyay, R., Kumar, M., & Leslie, J. F. (2007). Relative severity of aflatoxin contamination of cereal crops in West Africa. Food additives and contaminants, 24(10), 1109-1114.

Bandyopadhyay, R., Ortega-Beltran, A., Akande, A., Mutegi, C., Atehnkeng, J., Kaptoge, L., ... & Cotty, P. J. (2016). Biological control of aflatoxins in Africa: current status and potential challenges in the face of climate change. World Mycotoxin Journal, 9(5), 771-789.

Bayman, P., & Cotty, P. J. (1991). Vegetative compatibility and genetic diversity in the Aspergillus flavus population of a single field. Canadian Journal of Botany, 69(8), 1707-1711.

Bayman, P., & Cotty, P. J. (1993). Genetic diversity in Aspergillus flavus: association with aflatoxin production and morphology. Canadian Journal of Botany, 71(1), 23-31.

Bircan, C. (2005). The determination of aflatoxins in spices by immunoaffinity column extraction using HPLC. International journal of food science & technology, 40, 929-934.

Turkeys, 9(2), 52-55.

Bock, C. H., & Cotty, P. J. (1999). The relationship of gin date to aflatoxin contamination of cottonseed in Arizona. Plant Disease, 83(3), 279-285.

Bock, C. H., Mackey, B., & Cotty, P. J. (2004). Population dynamics of Aspergillus flavus in the air of an intensively cultivated region of south west Arizona. Plant Pathology, 53(4), 422-433.

182

Boyd, M. L., & Cotty, P. J. (2001). Aspergillus flavus and aflatoxin contamination of leguminous trees of the Sonoran Desert in Arizona. Phytopathology, 91(9), 913- 919.

Brown, K. L., Bren, U., Stone, M. P., & Guengerich, F. P. (2009). Inherent stereospecificity in the reaction of aflatoxin B1 8, 9-epoxide with deoxyguanosine and efficiency of DNA catalysis. Chemical research in toxicology, 22(5), 913-917.

Callicott, K. A., & Cotty, P. J. (2015). Method for monitoring deletions in the aflatoxin biosynthesis gene cluster of Aspergillus flavus with multiplex PCR. Letters in applied microbiology, 60(1), 60-65.

Cardeilhac, P. T., Schroeder, E. C., Perdomo, J. T., Combs, G. E., & Edds, G. T. (1970). Stunted pigs from sows fed crude aflatoxins. Toxicology and applied pharmacology, 17(2), 548-550.

Cardwell, K. F., & Cotty, P. J. (2002). Distribution of Aspergillus section Flavi among field soils from the four agroecological zones of the Republic of Benin, West Africa. Plant Disease, 86(4), 434-439.

Carvajal-Campos, A., Manizan, A., Tadrist, S., Akaki, D., Koffi-Nevry, R., Moore, G., Fapohunda, S., Bailly, S., Montet, D., Oswald, I., & Lorber, S. (2017). Aspergillus korhogoensis, a novel aflatoxin producing species from the Côte Toxins, 9(11), 353.

Centers for Disease Control and Prevention (CDC). (2004). Outbreak of aflatoxin poisoning--eastern and central provinces, Kenya, January-July 2004. MMWR. Morbidity and mortality weekly report, 53(34), 790.

Chang, P. K., Horn, B. W., & Dorner, J. W. (2005). Sequence breakpoints in the aflatoxin biosynthesis gene cluster and flanking regions in nonaflatoxigenic Aspergillus flavus isolates. Fungal Genetics and Biology, 42(11), 914-923.

Cho, S. H., Lee, C. H., Jang, M. R., Son, Y. W., Lee, S. M., Choi, I. S., Kim, S. H. & Kim, D. B. (2008). Aflatoxins contamination in spices and processed spice products commercialized in Korea. Food Chemistry, 107(3), 1283-1288.

Chulze, S. N., Palazzini, J. M., Torres, A. M., Barros, G., Ponsone, M. L., Geisen, R., Schmidt-Heydt, M. & Köhl, J. (2015). Biological control as a strategy to reduce the impact of mycotoxins in peanuts, grapes and cereals in Argentina. Food Additives & Contaminants: Part A, 32(4), 471-479.

Cleveland, T. E., Bhatnagar, D., & Cotty, P. J. (1990). Control of biosynthesis of aflatoxin in strains of Aspergillus flavus. A perspective on aflatoxin in field crops

183 and animal food products in the United States. ARS-83. Agricultural Research Service, US Department of Agriculture, Washington, DC, 67-73. Climates to travel. https://www.climatestotravel.com/climate/nigeria.

Cotty, P. J. (1989). Virulence and cultural characteristics of two Aspergillus flavus strains pathogenic on cotton. Phytopathology, 79(7), 808-814.

Cotty, P. J. (1994). Comparison of four media for the isolation of Aspergillus flavus group fungi. Mycopathologia, 125, 157-162.

Cotty, P. J. (1994). Influence of field application of an atoxigenic strain of Aspergillus flavus on the populations of A.flavus infecting cotton bolls and on the aflatoxin content of cottonseed. Phytopathology, 84(11), 1270-1277.

Cotty, P. J. (1996). Aflatoxin contamination of commercial cottonseed caused by the S strain of Aspergillus flavus. Phytopathology, 86, S71.

Cotty, P. J. (1997). Aflatoxin-producing potential of communities of Aspergillus section Flavi from cotton producing areas in the United States. Mycological Research, 101(6), 698-704.

Cotty, P. J. (1999). Atoxigenic strains of Aspergillus flavus have been applied to commercial cotton fields for three years. In Proc Beltwide Cotton Conf (Vol. 1, p. 108).

Cotty, P.J., 2006. Biocompetitive exclusion of toxigenic fungi. In: Barug, D., Bhatnagar, D., Van Egdmond, H.P., Van der Kamp, J.W., Van Osenbruggen, W.A. and Visconti, A. (eds.) The mycotoxin factbook. Wageningen Academic Publishers, Wageningen, the Netherlands, pp. 179-197.

Cotty, P.J., Bayman, P., Egel, D.S., & Elias, K.S. (1994). Agriculture, aflatoxins, and Aspergillus. In K. A. Powell, A. Renwick & J. Perberdy (Eds.), The Genus Aspergillus: From Taxonomy and Genetics to Industrial Application (pp. 1 27). Plenum Press, New York.

Cotty, P. J., & Cardwell, K. F. (1999). Divergence of West African and North American Communities of Aspergillus Section Flavi. Applied and Environmental Microbiology, 65(5), 2264-2266.

Cotty, P. J., Howell, D. R., Bock, C., & Tellez, A. (1997, January). Aflatoxin contamination of commercially grown transgenic Bt cottonseed. In Proc. Beltwide Cotton Prod. Res. Conf. National Cotton Council of America, Memphis, TN (pp. 108-110).

184

Cotty, P. J., & Jaime-Garcia, R. (2007). Influences of climate on aflatoxin producing fungi and aflatoxin contamination. International journal of food microbiology, 119(1-2), 109-115.

Cotty, P.J., & Lee, L. (1989). Aflatoxin contamination of cottonseed: comparison of pink bollworm damaged and undamaged bolls. Tropical Science, 29, 273-277.

Cotty, P. J., & Mellon, J. E. (2006). Ecology of aflatoxin producing fungi and biocontrol of aflatoxin contamination. Mycotoxin Research, 22(2), 110-117.

Cotty, P. J., Probst, C., & Jaime-Garcia, R. (2008). Etiology and management of aflatoxin contamination. Mycotoxins: detection methods, management, public health, and agricultural trade. CABI, Wallingford, United Kingdom, 287-299.

Davis, N. D., Diener, U. L., & Eldridge, D. W. (1966). Production of aflatoxins B1 and G1 by Aspergillus flavus in a semisynthetic medium. Appl. Environ. Microbiol., 14(3), 378-380.

De Iongh, H., Vles, R. O., & Van Pelt, J. G. (1964). Milk of mammals fed an aflatoxin-containing diet. Nature, 202, 466-467.

De Vries, H. R., Maxwell, S. M., & Hendrickse, R. G. (1989). Foetal and neonatal exposure to aflatoxins. Acta paediatrica Scandinavica, 78(3), 373.

Dereeper, A., Guignon, V., Blanc, G., Audic, S., Buffet, S., Chevenet, F., Dufayard, J.F., Guindon, S., Lefort, V., Lescot, M., & Claverie, J. M. (2008). Phylogeny. fr: robust phylogenetic analysis for the non-specialist. Nucleic Acids Research (36), W465-W469.

Dereeper, A., Audic, S., Claverie, J. M., & Blanc, G. (2010). BLAST-EXPLORER helps you building datasets for phylogenetic analysis. BMC evolutionary biology, 10(1), 8.

Diener, U. L., Cole, R. J., Sanders, T. H., Payne, G. A., Lee, L. S., & Klich, M. A. (1987). Epidemiology of aflatoxin formation by Aspergillus flavus. Annual review of phytopathology, 25(1), 249-270.

Dohlman, E. (2003). Mycotoxin hazards and regulations: impacts on food and animal feed crop trade. In J. Buzby (Ed.), International trade and food safety: Economic theory and case studies, agricultural economic report (pp. 97-108). Washington DC. USA: USDA.

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 Biology and Biochemistry, 41(1), 37-44.

185

Donner, M., Lichtemberg, P. S., Doster, M., Picot, A., Cotty, P. J., Puckett, R. D., & Michailides, T. J. (2015). Community structure of Aspergillus flavus and A. parasiticus in major almond-producing areas of California, United States. Plant disease, 99(8), 1161-1169.

Dorner, J. W. (2004). Biological control of aflatoxin contamination of crops. Journal of Toxicology: Toxin Reviews, 23(2-3), 425-450.

Dorner, J. W., & Lamb, M. C. (2006). Development and commercial use of afla- guard®, an aflatoxin biocontrol agent. Mycotoxin Research, 22(1), 33-38.

Doster, M. A., Cotty, P. J., & Michailides, T. J. (2014). Evaluation of the atoxigenic Aspergillus flavus strain AF36 in pistachio orchards. Plant disease, 98(7), 948- 956.

Doster, M., & Michailides, T. (1994). Aspergillus molds and aflatoxins in pistachio nuts in California. Phytopathology, 84, 583-590.

Drott, M. T., Lazzaro, B. P., Brown, D. L., Carbone, I., & Milgroom, M. G. (2017). Balancing selection for aflatoxin in Aspergillus flavus is maintained through interference competition with, and fungivory by insects. Proceedings of the Royal Society B: Biological Sciences, 284(1869), 20172408.

Duman, A.D. (2010). Storage of red chili pepper under hermetically sealed or vacuum conditions for preservation of its quality and prevention of mycotoxin occurrence. Journal of stored products research, 46, 155-160.

Eaton, D. L., & Gallagher, E. P. (1994). Mechanisms of aflatoxin carcinogenesis. Annual review of pharmacology and toxicology, 34(1), 135-172.

Egal, S., Hounsa, A., Gong, Y. Y., Turner, P. C., Wild, C. P., Hall, A. J., ... & Cardwell, K. F. (2005). Dietary exposure to aflatoxin from maize and groundnut in young children from Benin and Togo, West Africa. International Journal of Food Microbiology, 104(2), 215-224.

Ehrlich, K. C., Chang, P. K., Yu, J., & Cotty, P. J. (2004). Aflatoxin biosynthesis cluster gene cypA is required for G aflatoxin formation. Applied and Environmental Microbiology, 70(11), 6518-6524.

Ehrlich, K. C., Kobbeman, K., Montalbano, B. G., & Cotty, P. J. (2007). Aflatoxin- producing Aspergillus species from Thailand. International Journal of food microbiology, 114(2), 153-159.

Eliades, N. G., & Eliades, D. G. (2009). Haplotype Analysis: software for analysis of haplotype data. Forest Goettingen (Germany): Genetics and Forest Tree Breeding, Georg-August University Goettingen.

186

El Mahgubi, A., Puel, O., Bailly, S., Tadrist, S., Querin, A., Ouadia, A., Oswald, I.P. & Bailly, J. D. (2013). Distribution and toxigenicity of Aspergillus section Flavi in spices marketed in Morocco. Food Control, 32(1), 143-148.

European Commission. (2006). Commission Regulation (EC) No 1881/2006 of 19 December 2006 setting maximum levels for certain contaminants in foodstuffs. Official Journal of the European Union, 364(365 324).

European Spice Association. (2004). European Spice Association Quality Minima Document. Bonn, Germany.

Ezekiel, C. N., Sulyok, M., Frisvad, J. C., Somorin, Y. M., Warth, B., Houbraken, J., ... & Odebode, A. C. (2013). Fungal and mycotoxin assessment of dried edible mushroom in Nigeria. International Journal of Food Microbiology, 162(3), 231-236.

FAOSTAT. 2017. http://www.fao.org/faostat/en/#data/QC/visualize.

Flannigan, B., & Hui, S. C. (1976). The occurrence of aflatoxin producing strains of Aspergillus flavus in the mould floras of ground spices. Journal of Applied Bacteriology, 41(3), 411-418.

Food and Agriculture Organization. (2004). Worldwide regulations for mycotoxins in food and feed in 2003. FAO Food and Nutrition Paper No. 81. Food and Agriculture Organization of the United Nations, Rome, Italy.

Frisvad, J. C., Hubka, V., Ezekiel, C. N., Hong, S. B., Nováková, A., Chen, A. J.,

(2019). Taxonomy of Aspergillus section Flavi and their production of aflatoxins, ochratoxins and other mycotoxins. Studies in mycology, 93, 1-63.

Frisvad, J. C., Samson, R. A., & Smedsgaard, J. (2004). Emericella astellata, a new producer of aflatoxin B1, B2 and sterigmatocystin. Letters in applied microbiology, 38(5), 440-445.

Fufa, H., & Urga, K. (1996). Screening of aflatoxins in Shiro and ground red pepper in Addis Ababa. Ethiopian medical journal, 34(4), 243-249.

Geiser, D. M., Pitt, J. I., & Taylor, J. W. (1998). Cryptic speciation and recombination in the aflatoxin-producing fungus Aspergillus flavus. Proceedings of the National Academy of Sciences, 95(1), 388-393.

Glass, N. L., & Donaldson, G. C. (1995). Development of primer sets designed for use with the PCR to amplify conserved genes from filamentous ascomycetes. Appl. Environ. Microbiol., 61(4), 1323-1330.

187

Golge, O., Hepsag, F., & Kabak, B. (2013). Incidence and level of aflatoxin contamination in chilli commercialised in Turkey. Food Control, 33(2), 514-520.

Gong, Y. Y., Cardwell, K., Hounsa, A., Egal, S., Turner, P. C., Hall, A. J., & Wild, C. P. (2002). Dietary aflatoxin exposure and impaired growth in young children from Benin and Togo: cross sectional study. Bmj, 325(7354), 20-21.

Gong, Y., Egal, S., Hounsa, A., Turner, P. C., Hall, A. J., Cardwell, K. F., & Wild, C. P. (2003). Determinants of aflatoxin exposure in young children from Benin and Togo, West Africa: the critical role of weaning. International journal of epidemiology, 32(4), 556-562.

Gong, Y., Hounsa, A., Egal, S., Turner, P. C., Sutcliffe, A. E., Hall, A. J., ... & Wild, C. P. (2004). Postweaning exposure to aflatoxin results in impaired child growth: a longitudinal study in Benin, West Africa. Environmental health perspectives, 112(13), 1334-1338.

Grubisha, L. C., & Cotty, P. J. (2009). Twenty four microsatellite markers for the aflatoxin producing fungus Aspergillus flavus. Molecular Ecology Resources, 9(1), 264-267.

Guide, H. (2011). Postharvest Handling of Dehydrated Chiles. http://aces.nmsu.edu/pubs/_h/H236/welcome.html.

Hartley, R. D., NESBITT, B. F., & O'kelly, J. (1963). Toxic metabolites of Aspergillus flavus. Nature, London, 198(4885).

Hedayati, M. T., Pasqualotto, A. C., Warn, P. A., Bowyer, P., & Denning, D. W. (2007). Aspergillus flavus: human pathogen, allergen and mycotoxin producer. Microbiology, 153(6), 1677-1692.

Henry, S. H., Bosch, F. X., Troxell, T. C., & Bolger, P. M. (1999). Reducing liver cancer--global control of aflatoxin. Science, 286(5449), 2453-2454.

Hesseltine, C. W., Shotwell, O. L., Smith, M., Ellis, J. J., Vandegraft, E., & Shannon, G. (1970). Production of various aflatoxins by strains of the Aspergillus flavus series. In Proceedings of the first US Japan Conference on toxic microorganisms (pp. 202-210). Washington, DC: US Government Printing Office.

Holzapfel, C. W., Steyn, P. S., & Purchase, I. F. H. (1966). Isolation and structure of aflatoxins M1 and M2. Tetrahedron Letters, 7(25), 2799-2803.

Horn, B. W. (1997). Aspergillus caelatus, a new species in section Flavi. Mycotaxon, 61, 185-191.

188

Horn, B. W. (2003). Ecology and population biology of aflatoxigenic fungi in soil. Journal of Toxicology: Toxin Reviews, 22(2-3), 351-379.

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(5), 767-776.

Horn, B. W., & Dorner, J. W. (1999). Regional differences in production of aflatoxin B1 and cyclopiazonic acid by soil isolates of Aspergillus flavus along a transect within the United States. Appl. Environ. Microbiol., 65(4), 1444-1449.

Horn, B. W., Greene, R. L., & Dorner, J. W. (1995). Effect of corn and peanut cultivation on soil populations of Aspergillus flavus and A. parasiticus in southwestern Georgia. Applied and Environmental Microbiology, 61(7), 2472- 2475.

Hossain, M., & Bala, B. (2007). Drying of hot chilli using solar tunnel drier. Solar Energy, 81, 85-92.

Huelsenbeck, J. P., & Ronquist, F. (2001). MRBAYES: Bayesian inference of phylogenetic trees. Bioinformatics, 17(8), 754-755.

Iamanaka, B. T., de Menezes, H. C., Vicente, E., Leite, R. S., & Taniwaki, M. H. (2007). Aflatoxigenic fungi and aflatoxins occurrence in sultanas and dried figs commercialized in Brazil. Food Control, 18(5), 454-457.

IARC Working Group on the Evaluation of Carcinogenic Risks to Humans, World Health Organization, & International Agency for Research on Cancer. (2002). Some traditional herbal medicines, some mycotoxins, naphthalene and styrene (No. 82). World Health Organization.

Iqbal, S. Z., Paterson, R. R. M., Bhatti, I. A., & Asi, M. R. (2010). Survey of aflatoxins in chillies from Pakistan produced in rural, semi-rural and urban environments. Food Additives and Contaminants, 3(4), 268-274.

Islam, M. S., Callicott, K. A., Mutegi, C., Bandyopadhyay, R., & Cotty, P. J. (2018). Aspergillus flavus resident in Kenya: High genetic diversity in an ancient population primarily shaped by clonal reproduction and mutation-driven evolution. Fungal ecology, 35, 20-33.

Iyer, R. S., Coles, B. F., Raney, K. D., Thier, R., Guengerich, F. P., & Harris, T. M. (1994). DNA adduction by the potent carcinogen aflatoxin B1: mechanistic studies. Journal of the American Chemical Society, 116(5), 1603-1609.

Jaime-Garcia, R., & Cotty, P. J. (2003). Aflatoxin contamination of commercial cottonseed in south Texas. Phytopathology, 93(9), 1190-1200.

189

Jaime-Garcia, R., & Cotty, P. J. (2004). Aspergillus flavus in soils and corncobs in south Texas: implications for management of aflatoxins in corn-cotton rotations. Plant Disease, 88(12), 1366-1371.

Jaime Garcia, R., & Cotty, P. J. (2006a). Spatial distribution of Aspergillus flavus and its toxigenic strains on commercial cottonseed from south Texas and its relationship to aflatoxin contamination. Plant Pathology, 55(3), 358-366.

Jaime-Garcia, R., & Cotty, P. J. (2006b). Spatial relationships of soil texture and crop rotation to Aspergillus flavus community structure in South Texas. Phytopathology, 96(6), 599-607.

Jaime-Garcia, R., & Cotty, P. J. (2010). Crop rotation and soil temperature influence the community structure of Aspergillus flavus in soil. Soil Biology and Biochemistry, 42(10), 1842-1847.

Jalili, M., & Jinap, S. (2012). Natural occurrence of aflatoxins and ochratoxin A in commercial dried chili. Food Control, 24, 160-164.

Jiang, Y. I., Jolly, P. E., Ellis, W. O., Wang, J. S., Phillips, T. D., & Williams, J. H. (2005). Aflatoxin B1 albumin adduct levels and cellular immune status in Ghanaians. International immunology, 17(6), 807-814.

Joffe, A. Z., & Lisker, N. (1969). Effects of light, temperature, and pH value on aflatoxin production in vitro. Appl. Environ. Microbiol., 18(3), 517-518.

Kachapulula, P. W., Akello, J., Bandyopadhyay, R., & Cotty, P. J. (2017a). Aflatoxin contamination of groundnut and maize in Zambia: observed and potential concentrations. Journal of applied microbiology, 122(6), 1471-1482.

Kachapulula, P. W., Akello, J., Bandyopadhyay, R., & Cotty, P. J. (2017b). Aspergillus section Flavi community structure in Zambia influences aflatoxin contamination of maize and groundnut. International journal of food microbiology, 261, 49-56.

Kachapulula, P. W., Akello, J., Bandyopadhyay, R., & Cotty, P. J. (2018). Aflatoxin Contamination of Dried Insects and Fish in Zambia. Journal of food protection, 81(9), 1508-1518.

Khlangwiset, P., Shephard, G.S., & Wu, F. (2011). Aflatoxins and growth impairment: a review. Critical reviews in toxicology, 41, 740-755.

King, J. M., Walker, T., Njapau, H., Park, D. L., & Damann, K. E. (2000). Managing aflatoxin contamination in corn: Scientists use integrated approach to solution. LOUISIANA AGRICULTURE, 43(2), 20-21.

190

Kiran, D. R., Narayana, K. J., & Vijayalakshmi, M. (2005). Aflatoxin B 1 production in chillies (Capsicum annuum L.) kept in cold stores. African Journal of Biotechnology, 4(8), 791-795.

Klich, M. A. (2002). Biogeography of Aspergillus species in soil and litter. Mycologia, 94(1), 21-27.

Klich, M. A. (2007). Aspergillus flavus: the major producer of aflatoxin. Molecular plant pathology, 8(6), 713-722.

Klich, M. A., Mullaney, E. J., Daly, C. B., & Cary, J. W. (2000). Molecular and physiological aspects of aflatoxin and sterigmatocystin biosynthesis by Aspergillus tamarii and A. ochraceoroseus. Applied Microbiology and Biotechnology, 53(5), 605-609.

Klich, M. A., & Pitt, J. I. (1988). Differentiation of Aspergillus flavus from A. parasiticus and other closely related species. Transactions of the British Mycological Society, 91(1), 99-108.

Koressaar, T., & Remm, M. (2007). Enhancements and modifications of primer design program Primer3. Bioinformatics, 23(10), 1289-1291.

Krishnamachari, K. A. V. R., Nagarajan, V., Bhat, R., & Tilak, T. B. G. (1975). Hepatitis due to aflatoxicosis: an outbreak in western India. The Lancet, 305(7915), 1061-1063.

Kurtzman, C. P., Horn, B. W., & Hesseltine, C. W. (1987). Aspergillus nomius, a new aflatoxin-producing species related to Aspergillus flavus and Aspergillus tamarii. Antonie van leeuwenhoek, 53(3), 147-158.

Lansden, J. A. (1986). Determination of cyclopiazonic acid in peanuts and corn by thin layer chromatography. Journal-Association of Official Analytical Chemists, 69(6), 964-966.

Lee, L. S., Klich, M. A., Cotty, P. J., & Zeringue, H. J. (1989). Aflatoxin in Arizona cottonseed: increase in toxin formation during field drying of bolls. Archives of Environmental Contamination and Toxicology, 18(3), 416-420.

Link, H.F. (1809). Observationes in Ordines plantarum naturales. Dissertatio prima, complectens Anandrarum ordines Epiphytas, Mucedines Gastomycos et Fungos Der Gesellschaft Naturforschender Freunde zu Berlin. Magazin fu¨r die neuesten Entdeckungen in der gesamten Naturkunde 3, 1 42.

191

Liu, Y., Chang, C. C. H., Marsh, G. M., & Wu, F. (2012). Population attributable risk of aflatoxin-related liver cancer: systematic review and meta- analysis. European Journal of Cancer, 48(14), 2125-2136.

Liu, Y., & Wu, F. (2010). Global burden of aflatoxin-induced hepatocellular carcinoma: a risk assessment. Environmental health perspectives, 118, 818-824.

Lye, M. S., Ghazali, A. A., Mohan, J., Alwin, N., & Nair, R. C. (1995). An outbreak of acute hepatic encephalopathy due to severe aflatoxicosis in Malaysia. The American journal of tropical medicine and hygiene, 53(1), 68-72.

Magoha, H., Kimanya, M., De Meulenaer, B., Roberfroid, D., Lachat, C., & Kolsteren, P. (2014). Association between aflatoxin M1 exposure through breast milk and growth impairment in infants from Northern Tanzania. World Mycotoxin Journal, 7(3), 277-284.

Marín, S., Colom, C., Sanchis, V., & Ramos, A. J. (2009). Modelling of growth of aflatoxigenic A. flavus isolates from red chilli powder as a function of water availability. International Journal of Food Microbiology, 128(3), 491-496.

Mateles, R. I., & Adye, J. C. (1965). Production of aflatoxins in submerged culture. Appl. Environ. Microbiol., 13(2), 208-211.

Mauro, A., Battilani, P., Callicott, K. A., Giorni, P., Pietri, A., & Cotty, P. J. (2013). Structure of an Aspergillus flavus population from maize kernels in northern Italy. International journal of food microbiology, 162(1), 1-7.

Mauro, A., Battilani, P., & Cotty, P. J. (2015). Atoxigenic Aspergillus flavus endemic to Italy for biocontrol of aflatoxins in maize. Biocontrol, 60(1), 125-134.

Mckinney, J. D. (1975). Use of zinc acetate in extract purification for aflatoxin assay of cottonseed products. , 52(6), 213- 213.

Mehl, H. L., & Cotty, P. J. (2010). Variation in competitive ability among isolates of Aspergillus flavus from different vegetative compatibility groups during maize infection. Phytopathology, 100(2), 150-159.

Mehl, H. L., & Cotty, P. J. (2011). Influence of the host contact sequence on the outcome of competition among Aspergillus flavus isolates during host tissue invasion. Appl. Environ. Microbiol., 77(5), 1691-1697.

Mehl, H. L., & Cotty, P. J. (2013). Influence of plant host species on intraspecific competition during infection by Aspergillus flavus. Plant Pathology, 62(6), 1310- 1318.

192

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. Annals of the New York Academy of Sciences, 1273(1), 7-17.

Mitchell, N. J., Bowers, E., Hurburgh, C., & Wu, F. (2016). Potential economic losses to the US corn industry from aflatoxin contamination. Food Additives & Contaminants: Part A, 33(3), 540-550.

Mohammed, B., Abdulsalam, Z., & Ahmed, B. (2015). Profitability in Chilli Pepper Production in Kaduna State, Nigeria. British Journal of Applied Science and Technology, 12(3), 1-9.

Mostafavi, H. A., Fathollahi, H., & Motamedi, F. (2010). Food irradiation: Applications, public acceptance and global trade. African Journal of Biotechnology, 9(20), 2826-2833.

Ngindu, A., Kenya, P., Ocheng, D., Omondi, T., Ngare, W., Gatei, D., ... & Kaviti, J. (1982). Outbreak of acute hepatitis caused by aflatoxin poisoning in Kenya. The Lancet, 319(8285), 1346-1348.

Nogueira, L., Foerster, C., Groopman, J., Egner, P., Koshiol, J., & Ferreccio, C. (2015). Association of aflatoxin with gallbladder cancer in Chile. Jama, 313(20), 2075-2077.

Novas, M. V., & Cabral, D. (2002). Association of mycotoxin and sclerotia production with compatibility groups in Aspergillus flavus from peanut in Argentina. Plant Disease, 86(3), 215-219.

OBrian, G. R., Georgianna, D. R., Wilkinson, J. R., Yu, J., Abbas, H. K., Bhatnagar, D., ... & Payne, G. A. (2007). The effect of elevated temperature on gene transcription and aflatoxin biosynthesis. Mycologia, 99(2), 232-239.

Ohkura, M., Cotty, P. J., & Orbach, M. J. (2018). Comparative genomics of Aspergillus flavus S and L morphotypes yield Insights into niche adaptation. G3: Genes, Genomes, Genetics, 8(12), 3915-3930.

Omojokun, J. (2013). Regulation and enforcement of legislation on food safety in Nigeria. In: H.A. Makun (Ed.), Mycotoxin and Food Safety in Developing Countries (pp. 251-268). InTech Publishers, Rijeka, Croatia.

Ortega-Beltran, A., Moral, J., Puckett, R. D., Morgan, D. P., Cotty, P. J., & Michailides, T. J. (2018). Fungal communities associated with almond throughout crop development: Implications for aflatoxin biocontrol management in California. PloS one, 13(6), e0199127.

193

Orum, T. V., Bigelow, D. M., Nelson, M. R., Howell, D. R., & Cotty, P. J. (1997). Spatial and temporal patterns of Aspergillus flavus strain composition and propagule density in Yuma County, Arizona, soils. Plant Disease, 81(8), 911-916.

Owaga, E., Muga, R., Mumbo, H., & Aila, F. (2011). Chronic dietary aflatoxins exposure in Kenya and emerging public health concerns of impaired growth and immune suppression in children. International Journal of Biological and Chemical Sciences, 5(3).

Outbreak News Today. (2017). Aflatoxin kills 4 children in Tanzania, linked to consumption of maize.

Panda, H. (2010). Handbook on Spices and Condiments (Cultivation, Processing and Extraction). ASIA PACIFIC BUSINESS PRESS Inc.

Park, D. L., Lee, L. S., Price, R. L., & Pohland, A. E. (1988). Review of the decontamination of aflatoxins by ammoniation: Current status and regulation. Journal-Association of Official Analytical Chemists, 71(4), 685-703.

Park, D. L., & Liang, B. (1993). Perspectives on aflatoxin control for human food and animal feed. Trends in Food Science & Technology, 4(10), 334-342.

Paterson, R.R.M. (2007). Aflatoxins contamination in chilli samples from Pakistan. Food Control, 18, 817-820.

Payne, G. A., & Widstrom, N. W. (1992). Aflatoxin in maize. Critical Reviews in Plant Sciences, 10(5), 423-440.

Peakall, R. O. D., & Smouse, P. E. (2006). GENALEX 6: genetic analysis in Excel. Population genetic software for teaching and research. Molecular ecology notes, 6(1), 288-295.

Perrone, G., Haidukowski, M., Stea, G., Epifani, F., Bandyopadhyay, R., Leslie, J. F., & Logrieco, A. (2014). Population structure and Aflatoxin production by Aspergillus Sect. Flavi from maize in Nigeria and Ghana. Food Microbiology, 41, 52-59.

Perry, L., Dickau, R., Zarrillo, S., Holst, I., Pearsall, D.M., Piperno, D.R., Berman, M.J., Cooke, R.G., Rademaker, K., & Ranere, A.J. (2007). Starch fossils and the domestication and dispersal of chili peppers (Capsicum spp. L.) in the Americas. Science, 315, 986-988.

Peterson, S. W. (2008). Phylogenetic analysis of Aspergillus species using DNA sequences from four loci. Mycologia, 100(2), 205-226.

194

Peterson, S. W., Ito, Y., Horn, B. W., & Goto, T. (2001). Aspergillus bombycis, a new aflatoxigenic species and genetic variation in its sibling species, A. nomius. Mycologia, 689-703.

Picot, A., Doster, M., Islam, M. S., Callicott, K., Ortega-Beltran, A., Cotty, P., & Michailides, T. (2018). Distribution and incidence of atoxigenic Aspergillus flavus VCG in tree crop orchards in California: A strategy for identifying potential antagonists, the example of almonds. International journal of food microbiology, 265, 55-64.

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. International Journal of Systematic and Evolutionary Microbiology, 58(3), 725-735.

Pitt, J. I., Hocking, A. D., & Glenn, D. R. (1983). An improved medium for the detection of Aspergillus flavus and A. parasiticus. Journal of Applied Bacteriology, 54(1), 109-114.

Probst, C., Bandyopadhyay, R., & Cotty, P.J. (2014). Diversity of aflatoxin- producing fungi and their impact on food safety in sub-Saharan Africa. International journal of food microbiology, 174, 113-122.

Probst, C., Callicott, K. A., & Cotty, P. J. (2012). Deadly strains of Kenyan Aspergillus are distinct from other aflatoxin producers. European Journal of Plant Pathology, 132(3), 419-429.

Probst, C., & Cotty, P. J. (2012). Relationships between in vivo and in vitro aflatoxin production: reliable prediction of fungal ability to contaminate maize with aflatoxins. Fungal biology, 116(4), 503-510.

Probst, C., Njapau, H., & Cotty, P. J. (2007). Outbreak of an acute aflatoxicosis in Kenya in 2004: identification of the causal agent. Applied and Environmental Microbiology, 73(8), 2762-2764.

Probst, C., Schulthess, F., & Cotty, P. J. (2010). Impact of Aspergillus section Flavi community structure on the development of lethal levels of aflatoxins in Kenyan maize (Zea mays). Journal of Applied Microbiology, 108(2), 600-610.

Rambaut, A., 2012. FigTree v1. 4. http://tree.bio.ed.ac.uk/.

Rambo, G. W., Tuite, J., & Crane, P. (1974). Preharvest inoculation and infection of dent corn ears with Aspergillus flavus and Aspergillus parasiticus. Phytopathology, 64, 797 800.

195

Ramirez-Prado, J. H., Moore, G. G., Horn, B. W., & Carbone, I. (2008). Characterization and population analysis of the mating-type genes in Aspergillus flavus and Aspergillus parasiticus. Fungal Genetics and Biology, 45(9), 1292- 1299.

RASFF. (2015). Rapid Alert System for Food and Feed (RASFF). Directorate general for health and consumer protection. Brussels: European Commission.

Reddy, T. V., Viswanathan, L., & Venkitasubramanian, T. A. (1979). Factors affecting aflatoxin production by Aspergillus parasiticus in a chemically defined medium. Microbiology, 114(2), 409-413.

Reddy, S. V., Kiran Mayi, D., Uma Reddy, M., Thirumala-Devi, K., & Reddy, D. V. R. (2001). Aflatoxins B1 in different grades of chillies (Capsicum annum L.) in India as determined by indirect competitive-ELISA. Food additives & contaminants, 18(6), 553-558.

Robens, J., & Cardwell, K. (2003). The costs of mycotoxin management to the USA: management of aflatoxins in the United States. Journal of Toxicology: Toxin Reviews, 22(2-3), 139-152.

Rodrigues, P., Venâncio, A., Kozakiewicz, Z., & Lima, N. (2009). A polyphasic approach to the identification of aflatoxigenic and non-aflatoxigenic strains of Aspergillus section Flavi isolated from Portuguese almonds. International journal of food microbiology, 129(2), 187-193.

Russell, T. E., Watson, T. F., & Ryan, G. F. (1976). Field accumulation of aflatoxin in cottonseed as influenced by irrigation termination dates and pink bollworm infestation. Appl. Environ. Microbiol., 31(5), 711-713.

Sadeghi, N., Oveisi, M. R., Jannat, B., Hajimahmoodi, M., Bonyani, H., & Jannat, F. (2009). Incidence of aflatoxin M1 in human breast milk in Tehran, Iran. Food control, 20(1), 75-78.

Saito, M., & Tsuruta, O. (1993). A new variety of Aspergillus flavus from tropical soil in Thailand and its aflatoxin productivity. JSM Mycotoxins, 1993(37), 31-36.

Samson, R. A., & Varga, J. (2009). What is a species in Aspergillus?. Medical mycology, 47(sup1), S13-S20.

Santos, L., Marín, S., Mateo, E. M., Gil-Serna, J., Valle-Algarra, F. M., Patiño, B., & Ramos, A. J. (2011). Mycobiota and co-occurrence of mycotoxins in Capsicum powder. International journal of food microbiology, 151(3), 270-276.

196

Santos, L., Marín, S., Sanchis, V., & Ramos, A. J. (2010). Co-occurrence of aflatoxins, ochratoxin A and zearalenone in Capsicum powder samples available on the Spanish market. Food Chemistry, 122(3), 826-830.

Schade, J. E., McGreevy, K., King, A. D., Mackey, B., & Fuller, G. (1975). Incidence of aflatoxin in California almonds. Appl. Environ. Microbiol., 29(1), 48- 53.

Schatzmayr, G., & Streit, E. (2013). Global occurrence of mycotoxins in the food and feed chain: facts and figures. World Mycotoxin Journal, 6, 213-222.

Schindler, A. F., Palmer, J. G., & Eisenberg, W. V. (1967). Aflatoxin production by Aspergillus flavus as related to various temperatures. Appl. Environ. Microbiol., 15(5), 1006-1009.

Schroeder, H. W., & Boller, R. A. (1973). Aflatoxin production of species and strains of the Aspergillus flavus group isolated from field crops. Appl. Environ. Microbiol., 25(6), 885-889.

Schweiggert, U., Carle, R., & Schieber, A. (2007). Conventional and alternative processes for spice production a review. Trends in Food Science & Technology, 18(5), 260-268.

Shamsuddin, Z. A., Khan, M. A., Khan, B. A., Ahmad, M. A., & Ahmed, A. (1995). Contamination of red chilli with aflatoxin B1 in Pakistan. Mycotoxin research, 11(1), 21-24.

Shephard, G.S. (2003). Aflatoxin and food safety: recent African perspectives. Journal of Toxicology: Toxin Reviews, 22, 267-286.

Shetty, P. H., & Bhat, R. V. (1997). Natural occurrence of fumonisin B1 and its co- occurrence with aflatoxin B1 in Indian sorghum, maize, and poultry feeds. Journal of agricultural and food chemistry, 45(6), 2170-2173.

Shih, C. N., & Marth, E. H. (1972). Production of aflatoxin in a medium fortified with sodium chloride. Journal of dairy science, 55(10), 1415-1419.

Shih, C. N., & Marth, E. H. (1974). Aflatoxin formation, lipid synthesis, and glucose metabolism by Aspergillus parasiticus during incubation with and without agitation. Biochimica et Biophysica Acta (BBA)-General Subjects, 338(1), 286- 296.

Shotwell, O. L., Hesseltine, C. W., Burmeister, H. R., Kwolek, W. F., Shannon, G. M., & Hall, H. H. (1969). Survey of cereal grains and soybeans for the presence of aflatoxin: I. Wheat, grain sorghum and oats. Cereal Chem, 46(446).

197

Singh, P., & Cotty, P. J. (2017). Aflatoxin contamination of dried red chilies: Contrasts between the United States and Nigeria, two markets differing in regulation enforcement. Food control, 80, 374-379.

Singh, P., & Cotty, P. J. (2019). Characterization of Aspergilli from dried red chilies (Capsicum spp.): Insights into the etiology of aflatoxin contamination. International journal of food microbiology, 289, 145-153.

Singh, P., Orbach, M., & Cotty, P. (2018). Aspergillus texensis: A Novel Aflatoxin Producer with S Morphology from the United States. Toxins, 10(12), 513.

Smela, M. E., Currier, S. S., Bailey, E. A., & Essigmann, J. M. (2001). The chemistry and biology of aflatoxin B1: from mutational spectrometry to carcinogenesis. Carcinogenesis, 22(4), 535-545.

Smouse, P. E., Banks, S. C., & Peakall, R. (2017). Converting quadratic entropy to diversity: Both animals and alleles are diverse, but some are more diverse than others. PloS one, 12(10), e0185499.

Soares, C., Rodrigues, P., Peterson, S. W., Lima, N., & Venâncio, A. (2012). Three new species of Aspergillus section Flavi isolated from almonds and maize in Portugal. Mycologia, 104(3), 682-697.

Stenske, K. A., Smith, J. R., Newman, S. J., Newman, L. B., & Kirk, C. A. (2006). Aflatoxicosis in dogs and dealing with suspected contaminated commercial foods. Journal of the American Veterinary Medical Association, 228(11), 1686- 1691.

St Leger, R. J., Joshi, L., & Roberts, D. W. (1997). Adaptation of proteases and carbohydrases of saprophytic, phytopathogenic and entomopathogenic fungi to the requirements of their ecological niches. Microbiology, 143(6), 1983-1992.

Official Methods of Analysis of the Association of Official Analytical Chemists (pp. 477

Tansakul, N., Limsuwan, S., Böhm, J., Hollmann, M., & Razzazi-Fazeli, E. (2013). Aflatoxins in selected Thai commodities. Food Additives & Contaminants: Part B, 6, 254-259.

Topuz, A., Feng, H., & Kushad, M. (2009). The effect of drying method and storage on color characteristics of paprika. LWT-Food Science and Technology, 42, 1667- 1673.

198

Torres, A. M., Barros, G. G., Palacios, S. A., Chulze, S. N., & Battilani, P. (2014). Review on pre-and post-harvest management of peanuts to minimize aflatoxin contamination. Food Research International, 62, 11-19.

Tubajika, K. M., Mascagni, H. J., Damann, K. E., & Russin, J. S. (1999). Nitrogen fertilizer influence on aflatoxin contamination of corn in Louisiana. Journal of agricultural and food chemistry, 47(12), 5257-5260.

Turner, P. C., Moore, S. E., Hall, A. J., Prentice, A. M., & Wild, C. P. (2003). Modification of immune function through exposure to dietary aflatoxin in Gambian children. Environmental health perspectives, 111(2), 217-220.

Untergasser, A., Cutcutache, I., Koressaar, T., Ye, J., Faircloth, B. C., Remm, M., & Rozen, S. G. (2012). Primer3 new capabilities and interfaces. Nucleic acids research, 40(15), e115.

US Food and Drug Administration. (2000). Guidance for industry: action levels for poisonous or deleterious substances in human food and animal feed. USFDA, Washington, DC.

Van der Zijden, A. S. M., Koelensmid, W. B., BOLDIXGH, J., Barrett, C. B., Ord, W. O., & Philp, J. (1962). Isolation in crystalline form of a toxin responsible for Turkey X disease. Nature, London, 195(4846). van Egmond, H.P., & Jonker, M.A. (2004). Worldwide regulations on aflatoxins the situation in 2002. Journal of Toxicology: Toxin Reviews, 23(2-3), 273-293.

Vardon, P., McLaughlin, C., & Nardinelli, C. (2003). Potential economic costs of mycotoxins in the United States. In: J.L. Richard, & G.A. Payne (Eds.), Mycotoxins: Risk in Plant, Animal and Human Systems (pp. 136-142); CAST Task Force Report No. 139. Ames, IA.

Wei, D. L., & Jong, S. C. (1986). Production of aflatoxins by strains of the Aspergillus flavus group maintained in ATCC. Mycopathologia, 93(1), 19-24.

Wicklow, D. T., & Donahue, J. E. (1984). Sporogenic germination of sclerotia in Aspergillus flavus and A. parasiticus. Transactions of the British Mycological Society, 82(4), 621-624.

Wild, C. P., Rasheed, F. N., Jawla, M. F. B., Hall, A. J., Jansen, L. A. M., & Montesano, R. (1991). In-utero exposure to aflatoxin in West Africa. The Lancet, 337(8757), 1602.

Williams, J.H., Phillips, T.D., Jolly, P.E., Stiles, J.K., Jolly, C.M., & Aggarwal, D. (2004). Human aflatoxicosis in developing countries: a review of toxicology,

199 exposure, potential health consequences, and interventions. The American journal of clinical nutrition, 80, 1106-1122.

World data atlas-Knoema. https://knoema.com/atlas/Nigeria/maps/Production-of- Maize Wu, F., & Guclu, H. (2012). Aflatoxin regulations in a network of global maize trade. PloS one, 7(9), e45151.

Yogendrarajah, P., Jacxsens, L., De Saeger, S., & De Meulenaer, B. (2014). Co- occurrence of multiple mycotoxins in dry chilli (Capsicum annum L.) samples from the markets of Sri Lanka and Belgium. Food Control, 46, 26-34. Yoko, I. T. O., Peterson, S. W., Donald, T., & Tetsuhisa, G. O. T. O. (2001). Aspergillus pseudotamarii, a new aflatoxin producing species in Aspergillus section Flavi. Mycological Research, 105(2), 233-239. Yu, J. (2012). Current understanding on aflatoxin biosynthesis and future perspective in reducing aflatoxin contamination. Toxins, 4(11), 1024-1057.

Yu, J., Chang, P. K., Ehrlich, K. C., Cary, J. W., Bhatnagar, D., Cleveland, T. E., ... & Bennett, J. W. (2004). Clustered pathway genes in aflatoxin biosynthesis. Appl. Environ. Microbiol., 70(3), 1253-1262. Zeringue Jr, H. J., Shih, B. Y., Maskos, K., & Grimm, D. (1999). Identification of the bright-greenish-yellow-fluorescence (BGY-F) compound on cotton lint associated with aflatoxin contamination in cottonseed. Phytochemistry, 52(8), 1391-1397.

200