LIBRARY

Michigan State
University

 

 

 

PLACE IN RETURN BOX to remove this checkout from your rocord.
TO AVOID FINES rotum on or More data duo.

DATE DUE DATE DUE DATE DUE

 

 

 

 

AUG 3 0 20

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

MSU I: An Affirmative ActioNEquai Oppoflunity Ila-mulch
Wan-9.1

THE FUNCTION AND EXPRESSION OF THE VER -1 GENE AND
LOCALIZATION OF THE VER—l PROTEIN INVOLVED IN AFLATOXIN Bl
BIOSYNTHESIS IN ASPERGHL US PARASITICUS
By

Shun-Hsin Liang

A DIS SERTATION

Submitted to
Michigan State University
in partial fulfillment Of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Food Science and Human Nutrition

1996

ABSTRACT
THE FUNCTION AND EXPRESSION OF THE VER -1 GENE AND
LOCALIZATION OF THE VER—l PROTEIN INVOLVED IN AFLATOXIN Bl
BIO SYNTHESIS IN ASPERGILL US PARASITICUS
By

Shun-Hsin Liang

The ultimate goal of the present research is the elimination of preharvest aflatoxin
contamination. One approach to accomplish this goal is to develop an understanding of the
molecular mechanisms regulating aflatoxin biosynthesis in toxigenic aspergilli. This study
focused on the ver-I gene associated with the aflatoxin biosynthetic pathway. This gene was
cloned and its nucleotide sequence was determined as part of a previous study on aflatoxin
B1 (AFBI) biosynthesis in Aspergillus parasiticus NRRL 5862. Two copies of the ver-I
gene, ver-IA and ver-IB, were tentatively identified in this fimgal Strain by Southern
hybridization analysis. Genetic complementation and nucleotide sequence data suggested that
the ver-I gene is involved in the conversion of versicolorin A (VA) to sterigmatocystin (ST).

To clearly establish the fiinction of the ver—I gene, I proposed to test three
hypotheses: (1) the ver-I gene encodes a protein which has enzymatic activity associated with
the conversion of VA to ST; (2) the pattern of Ver-l protein accumulation parallels AFB 1
synthesis; and (3) the Ver-l protein is closely associated with other proteins involved in AFBl
biosynthesis allowing the pathway to function in an efficient way. To address these
hypotheses, studies were designed to achieve the following specific aims : ( 1) identify ver-I

gene function by nucleotide sequence and gene disruption analyses; (2) identify the

accumulation pattern of the Ver-l protein in fungal cells grown in liquid or solid media; and
(3) identify the intracellular location of the Ver-l protein.

In specific aim 1, the methods consisted of gene disruption and nucleotide sequence
analysis. The methods used in specific aim 2 included generation of polyclonal antibodies
against the Ver-l protein, batch fermentation analysis, nutritional shift assay, and analysis of
ver-I promoter activity using the GUS reporter strain. In specific aim 3, the subcellular
localization of the Ver-l protein was performed by difi‘erential centrifugation and
immunofluorescence microscopy. The major findings of this research are : (1) ver-IA but not
ver-IB is directly involved in aflatoxin biosynthesis; (2) the timing of Ver-l protein
accumulation is positively correlated to aflatoxin accumulation; (3) expression of the ver-I
gene in a fungal colony is subject to temporal and Spatial regulation; (4) although the ver-I
gene is expressed in both vegetative hyphae and conidiophores, there is significantly more
Ver-l protein in the structures involved in asexual reproduction; and (5) the Ver-l protein

appears to be located in or tightly bound to a membrane-bound organelle in fimgal cells.

To my parents, brother, and lovely wife, Lie-Ken

iv

ACKNOWLEDGMENTS

First of all, my deep gratitude is to my advisor, Dr. John E. Linz. This dissertation
would not have been possible without his superior instruction and guidance. I feel very lucky
to be his student.

I also am extremely grateful to my guidance committee members, Dr. William G.
Helferich, Dr. James J. Pestka, and Dr. C. A. Reddy, whose expertise has broadened my
knowledge and makes this dissertation truly unique.

Finally, I would like to thank the following people in my laboratory for their generous
and professional help in my work: Dr. Frances Trail, Dr. Ninee Mahanti, Dr. Tzong-Shoon
Wu, Dr. Ludmila Rose, Matt Rarick, Michael Miller, David Wilson, and Renqing Zhou. I

love you all.

TABLE OF CONTENTS

LIST OF TABLES ................................................... ix

LIST OF FIGURES .................................................... x

INTRODUCTION AND RESEARCH RATIONALE .......................... 1
CHAPTER 1

LITERATURE REVIEW ............................................... 3

1. Natural occurrence of aflatoxins ........................................ 3

II. The aflatoxin problem ............................................... 6

A. Aflatoxins and public health ...................................... 6

B. Aflatoxins and economic losses .................................. 10

111. Control of aflatoxins ............................................... 12

A. Prevention .................................................. 12

B. Decontamination or detoxification ................................ 13

1. Removal of aflatoxins .................................... 13

(a) Physical separation ............................... 13

(b) Chemical separation .............................. 14

2. Degradation or detoxification .............................. 14

(a) Physical methods ................................. 14

(b) Chemical methods ................................ 15

(c) Biological methods ............................... 16

C. Reduction of aflatoxin bioavailability .............................. 16

1. Selective chemisorption .................................. 16

2. Chemoprotection against aflatoxin toxicity .................... 17

IV. Biosynthesis of aflatoxins ............................................ 17

A. Chemical and biochemical aspects ................................ 17

B. Genetic and molecular aspects ................................... 20

1. Classical genetic analysis ................................. 20

2. Molecular genetics ...................................... 21

V. Regulation of aflatoxin biosynthesis .................................... 24

vi

CHAPTER 2
STRUCTURAL AND FUNCTIONAL CHARACTERIZATION

OF THE PER-1 GENE ................................................ 27
I. Introduction ...................................................... 27
11. Materials and methods .............................................. 30
A. Strains and plasmids .......................................... 30
B. Bacterial cell transformation and plasmid purification .................. 31
C. Transformation of fungal protoplasts .............................. 34
D. Preparation and analysis of genomic DNA from fimgal cells ............. 34
E. Analysis of versicolorin A and aflatoxin production ................... 35
F. Nucleotide sequence analysis .................................... 35
III. Results .......................................................... 38
A. Restriction fragment length polymorphism (RFLP) analysis ............. 38
B. Identification of a duplicated chromosomal region containing
the ver-I gene ............................................... 38
C. Nucleotide sequence analysis of ver—IB ............................ 39
D. Functional analysis of the ver-I genes via recombinational
inactivation and complementation ................................ 39
IV. Discussion ....................................................... 51
CHAPTER 3
REGULATION OF VER-I GENE EXPRESSION IN FUNGAL CELLS .......... 58
I. Introduction ...................................................... 58
11. Materials and methods .............................................. 59
A. Strains and plasmids .......................................... 59
B. Generation of polyclonal antibodies against the Ver-l protein ........... 60
C. Batch fermentation analysis ..................................... 61
D. Nutritional shift assay ......................................... 61
E. Protein extraction and Western blot analysis ........................ 62
F. Analysis of mycelial dry weight and aflatoxin concentration ............. 62
G. Analysis of the accumulation of the Ver-l protein in a fungal colony ...... 62
H. Analysis of the expression of the ver-I gene in mycelia ................ 63
III. Results .......................................................... 64
A. Specificity of anti-Ver-l antibodies ............................... 64
B. Accumulation of the Ver-l protein in liquid cultures of A. parasiticus ..... 67
C. Accumulation of the Ver-l protein in a fungal colony ................. 72
D. Expression of the ver-I gene in mycelia ............................ 76
IV. Discussion ....................................................... 76
CHAPTER 4
INTRACELLULAR LOCALIZATION OF THE VER-l PROTEIN IN
FUNGAL CELLS .................................................... 87

vii

I. Introduction ...................................................... 87

11. Materials and methods .............................................. 88
A. Differential centrifugation of fungal cell extracts ..................... 88
1. Grinding in liquid nitrogen ................................ 88
2. Disruption of protoplasts by a homogenizer ................... 89
B. Calculation of the quantity of the Ver-l protein in cell fractions .......... 89
C. Analysis of marker enzyme activities .............................. 9O
1. Succinate dehydrogenase ................................. 9O
2. Catalase .............................................. 91
3. Acid phosphatase ....................................... 91
4. Glucose-6-phosphatase .................................. 92
D. Indirect immunofluorescence microscopy ........................... 92
III. Results .......................................................... 93
A. Subcellular localization of the Ver-l protein in A. parasiticus ........... 93
B. Intracellular localization of the Ver-l protein in A. parasiticus ........... 98
IV. Discussion ...................................................... 103
CONCLUSIONS .................................................... 1 07
APPENDIX
PHYSICAL AND TRANSCRIPTIONAL MAP OF AN AFLATOXIN GENE
CLUSTER IN ASPERGILLUS PARASITIC US ............................. 109
I. Introduction ..................................................... 109
11. Published paper .................................................. 110
111. Discussion ...................................................... 1 19
LIST OF REFERENCES ............................................. 120

viii

LIST OF TABLES

Table 1. Distribution of the Ver-l protein in cell fractions obtained from
A. parasiticus SU-l by differential centrifugation ..................... 96

Table 2. Distribution of marker enzymes in cell fractions obtained from
A. parasiticus SU-l by differential centrifugation ..................... 97

ix

LIST OF FIGURES

Figure 1. Chemical structures of naturally occurring aflatoxins ................... 4
Figure 2. Metabolic activation of aflatoxin B 1 and its harmfiJl effects ............... 9
Figure 3. The proposed biosynthetic pathway for aflatoxin B 1 ................... 19

Figure 4. Schematic representation of the location of ver-IA and ver-IB on cosmids
NorA and Ver2 .............................................. 28

Figure 5. Plasmids used for functional analysis of the ver-I gene in recombinational
inactivation and complementation experiments ....................... 32

Figure 6. Strategy for nucleotide sequence analysis of the 1.7-kb EcoRI-HindII
fragment which contains the ver-IB gene ........................... 36

Figure 7. Identification of a duplicated chromosomal region containing the ver-I
genes by Southern hybridization analysis ........................... 40

Figure 8. An alignment of the nucleotide sequences and the deduced amino acid
sequences of ver-IA and ver-IB .................................. 41

Figure 9. Schematic representation of the disruption of ver-IA by plasmid pDV—VA

(gene replacement) ........................................... 43
Figure 10.TLC analysis of the metabolites of A. parasiticus VAD-102 ............. 46
Figure 11.Southern hybridization analysis to confirm disruption of the ver-I gene . . . . 47

Figure 12.TLC analysis of the metabolites of transformants obtained by transformation
Of A. parasiticus VAD-102 with pVer—Ben ......................... 49

Figure 13. Southern hybridization analysis of A. parasiticus VAD-102 transformed
with pVer-Ben ............................................... 50

Figure l4.Proposed schemes for the enzymatic conversion of versicolorin A (VA) to

sterigmatocystin (ST) .........................................

Figure 15.A comparison of the deduced amino acid sequences of ver-IA, ver-IB,

Streptomyces coelicolor actIII , and Magnaporthe grisea ThnR. .........

Figure 16.Proposed Ver-l protein activity in a two-step dehydroxylation reaction

derived from analogous reactions in melanin and cynodontin biosynthesis . .

Figure 17.Westem blot analysis to measure the specificity of anti-Ver-l antibodies . . .

Figure 18.Batch fermentation analysis of Ver-l protein accumulation in

A. parasiticus SU-l ...........................................

Figure 19.Nutritional shift assay for the identification of Ver-l protein accumulation

in A. parasiticus SU-l .........................................

Figure 20.The morphology of a fungal colony grown on solid YES media ..........

Figure 21.Westem blot analysis of the proteins extracted from three concentric zones

in fungal colonies .............................................

Figure 22.Analysis of ver-I promoter activity in a fungal colony using the

ver-I/GUS reporter strain ......................................

Figure 23.The structure of a typical conidiophore ofAspergillus .................

Figure 24.Microscopic images from the analysis of ver-I promoter activity in mycelia

using the ver-I/GUS reporter strain ...............................

Figure 25.Westem blot analysis of cell fractions from A. parasiticus SU-l ..........

Figure 26.Localization of the Ver-l protein in fimgal cells by immunofluorescence

microscopy and laser confocal microscopy ..........................

53

54

57

65

68

7O

73

74

75

77

78

94

99

INTRODUCTION AND RESEARCH RATIONALE

Aflatoxin contamination of agricultural commodities and dietary staples such as corn,
peanuts, and cottonseed has caused worldwide economic and food safety problems (J elinek
et al., 1989). Its impact on the agricultural economy and potential impact on human health
has drawn scientists' efforts in an attempt to prevent aflatoxin contamination. The long term
goal of this research is the elimination of preharvest aflatoxin contamination from the food
chain. An elucidation of the molecular mechanisms which regulate aflatoxin biosynthesis in
toxigenic aspergilli may be the best approach to achieve the ultimate goal. For this purpose,
it is very important to study aflatoxin biosynthesis at the molecular level and to identify the
control points in the aflatoxin biosynthetic pathway. The information derived from these
analyses will help in development of strategies to inhibit aflatoxin production in the field.

In order to effectively understand the regulation of expression of genes involved in
aflatoxin biosynthesis, it is necessary to clone several of these genes and to clearly establish
their fiinction. Two pathway genes, nor-1 and ver-I, have been cloned in our laboratory by
genetic complementation of Aspergillus parasiticus mutants blocked at unique steps in
aflatoxin B1 (AFB,) biosynthesis (Chang et al., 1992; Skory et al., 1992). The present
research is mainly focused on the structural and functional characterization of the two copies
of the ver-1 gene (ver-IA and ver-IB) and of the Ver-l protein in A. parasiticus. The

objective of this study is to understand aflatoxin biosynthesis at the genetic and cellular levels

2

using the ver-I gene as the model system. To achieve this aim, an initial effort was made to
establish the function of two ver-I gene copies by recombinational inactivation and genetic
complementation experiments. The function of the ver-I gene was further examined by
studying the relationship between Ver-l protein accumulation and aflatoxin biosynthesis using
batch fermentation analysis and nutritional shift assay (liquid media systems). After the Per-1
gene was confirmed to be directly involved in AFBl biosynthesis, the regulation of its
expression was studied during growth on solid media. Finally, the location of the Ver-l
protein in fimgal cells was analyzed by subcellular fractionation and immunofluorescence
microscopy.

This research represents the first analysis to study the regulation of aflatoxin gene
expression and the location of aflatoxin-associated enzymes in the fiJngal cells using solid
growth media. The resulting data will help in understanding the function of aflatoxin
biosynthesis in fungi and may provide a potential approach to develop an efficient way to

prevent aflatoxin contamination.

CHAPTER 1

LITERATURE REVIEW

1. Natural occurrence of aflatoxins

Aflatoxins are biologically active mycotoxins produced by certain strains of the
imperfect fiingi Aspergillus parasiticus, A. flavus, and A. nomius (Bennett, 1979; Cotty et
al., 1994). A. nomius, however, is less important to the contamination of foods and feeds by
aflatoxins. The major aflatoxins of concern include aflatoxin B1, B2, G1, and G2 (Figure 1).
When resolved by thin-layer chromatography (TLC), these aflatoxins separate into individual
fluorescent compounds in the order given above. Aflatoxins are freely soluble in moderately
polar solvents such as chloroform, methanol, and dimethylsulphoxide, and also have some
water solubility (McLean and Dutton, 1995). These mycotoxins are usually found together
in contaminated foods and feeds. Aflatoxin Bl (AFBI), however, is the most abundant and
toxic. Aflatoxin M1 is the major metabolic product of AFB1 in animals and is usually found
in the milk and urine of diary cattle and other mammalian species that have consumed
aflatoxin-contaminated foods or feeds (Allcrofi et al., 1966).

Although both species can produce aflatoxins, A. parasiticus and A. flavus have
different abilities in toxin production. First, A. parasiticus produces the aflatoxin B and G
groups but A. flavus only produces the aflatoxin B group. Second, most A. parasiticus

isolates (> 90 %) produce aflatoxin whereas up to 35% of the A. flavus isolates may not

 

 

O o
o
l
\
1L /L ' /
O O OCH3
AFLATOXINBl

 

 

 

   

AF LATOXIN G 1 AF LATOXIN G 2

Figure 1. Chemical structures of naturally occurring aflatoxins.

5

produce aflatoxin (Bennett and Papa, 1988; Trail et al., 1995). The contamination of tree
nuts, peanuts, and other oilseeds such as corn and cottonseed by aflatoxins occurs under
certain environmental conditions when these crops are infected with toxigenic strains of the
fungus (Jelinek et al., 1989). Although the fungus may be killed or removed during
processing, aflatoxins ofien remain in the final product and thus contribute to the main point
of entry of aflatoxins into the food chain (Smith and Moss, 1985). For this reason, the
elimination of aflatoxin contamination at the preharvest stage is very important. An
alternative route of aflatoxin contamination is the postharvest infection of food or feed with
toxigenic fungi and the subsequent formation of toxin at some stage during processing,
transport, and storage.

A variety of factors which contribute to the contamination of aflatoxins in the food
chain include biological and environmental factors (Pestka and Casale, 1990). The biological
factors consist of substrate availability for toxigenic fimgi, competing microflora (Ellis et al.,
1991), and susceptibility of the crops. The environmental factors include temperature,
moisture, insect/bird damage, and mechanical injury of the crops. Warm temperatures and
drought conditions favor fungal growth and aflatoxin production in the field. Climatic
patterns thus could determine which regions are more prone to aflatoxin contamination. For
example, corn in the southeastern regions of the United States is frequently contaminated with
aflatoxins. Other areas of the country are occasionally susceptible. For example, due to
severe drought, widespread aflatoxin contamination in corn occurred in the midwestern
regions of the United States in 1983 and 1988 (Chu, 1991). In the postharvest stage,

aflatoxin contamination may result from warm temperatures and high humidity during storage.

II. The aflatoxin problem

Worldwide, aflatoxins are considered a public health and economic problem because
of their potent toxic effects on humans and animals (Chu, 1991; Eaton and Gallagher, 1994)
and the huge cost incurred by farmers or producers due to the loss of crops, animals and the
need for more careful agronomic practices (Shane, 1994).
A. Aflatoxins and public health

The threat of aflatoxins to human health was initially realized after they were directly
linked to acute hepatotoxicity in poultry (Turkey X disease) in 1960 (Blount, 1961) and later
their association with fatal human aflatoxicoses in India (Krishnamachari et al., 1975) and
West Africa (N gindu et al., 1982). In the outbreak of aflatoxicosis in northwest India, 108
persons died among 397 persons affected due to consumption of contaminated corn with
aflatoxin at levels of 0.25 to 15 mg/kg. The daily AFB, intake was about 55 jig/kg body
weight. The symptoms of affected individuals in this outbreak included vomiting, high fever,
rapid progressive jaundice, edema of the limbs, and swollen livers. Histopathological
examination showed extensive periportal fibrosis and bile duct proliferation of the liver and
gastrointestinal hemorrhages. In studies on animals, it was found that no species is resistant
to the acute toxic effects of aflatoxins (Newbeme and Butler, 1969). The first signs of
aflatoxicosis in animals are the lack of appetite and the loss of weight. Liver centrilobular
necrosis, fatty degeneration, and bile duct proliferation are the most common pathological
findings. For most of the animals tested, the LD50 for a single dose of AFB, is in the range
of 0.5 to 10 mg/kg of body weight. Although interspecific variation has been recognized for
acute effects, many factors, such as age, sex, nutritional status of diet, and mode of

application, affect the degree of toxicoses. In general, aflatoxin is more toxic to young

7

animals and males than females (Cullen and Newbeme, 1994). Besides the liver, many other
organs such as pulmonary (Wieder et al., 1968), gastrointestinal (Bulter, 1964; Deger, 1976),
renal (Epstein er al., 1969), nervous (Egbunike and Ikeguonu, 1984), reproductive (Ottinger
and Doerr, 1980), and immune systems (Kadian et al., 1988; Pestka and Bondy, 1990) are
more or less severely affected with high doses of AFB ,.

In well-developed countries, aflatoxin contamination in foods rarely reaches the level
that causes acute aflatoxicosis in humans. Hence, studies of aflatoxin toxicity on humans have
been focusing on its carcinogenic potential. The carcinogenic properties of aflatoxins have
been studied extensively, and much information has been produced concerning various aspects
of their mechanisms of action and their putative importance as risk factors for primary
hepatocellular carcinoma (PHC) in humans (Busby and Wogan, 1984). Primary liver tumors
have been induced by experimental administration of AFB, to animals of many species,
including fish (rainbow trout, salmon, and guppy), birds (duck), rodents (5 strains of rats, the
B6C3Fl mouse, tree shrew, and hamster), a carnivore (ferret), and subhuman primates
(rhesus, cynomolgus, and Afiican green monkeys) (W ogan, 1991). Each of these species has
been shown to be susceptible to induction of PHC by AFB, administration. However, in
relating this information to putative effects in humans, it is important to note that wide species
differences exist with respect to the carcinogenic potency of aflatoxins. The estimated
potency of AFB] in inducing liver tumors in animals showed that rats are highly susceptible,
mice are highly resistant, and primates are of intermediate susceptibility (Eaton and Gallagher,
1994). In utilizing information derived fi'om these experimental systems for assessing cancer
risks for humans resulting from aflatoxin exposures, it is of particular importance to compare

the animal data with estimates of human susceptibility derived from epidemiological

8

observations. The consolidated data have been collected from studies in Africa and Asia
(Hsieh, 1989; van Rensburg et al., 1985; Yu et al., 1898), where aflatoxin intake was
measured in populations in which PHC incidence was variable, as determined from cancer
registry information. In general, aflatoxin intake values increase in parallel with cancer
incidence. Although the causative role for aflatoxins in human liver cancer has not been
universally accepted because of the presence of endemic hepatitis B virus in high risk
populations (Stolofi‘, 1989), the International Agency for Research on Cancer has determined
that the combined experimental and epidemiological evidence was sufficient to designate
aflatoxins as human carcinogens (IARC, 1993).

Studies of the metabolism of AFB, have revealed that the compound is activated to
its toxic form predominantly by liver-specific, cytochrome P450 IIIA4 monooxygenases in
mammals (Coros etal., 1990). Metabolic activation of AFB, to AFB ,-8,9-epoxide (Figure
2) is believed to result in its toxicity, mutagenicity, and carcinogenicity (Campbell and Hayes,
1976; Gurtoo and Dave, 1975). The aflatoxins listed in order of terms of acute and chronic
toxicity are AFB, > AF G, > AFB, > AFG , suggesting that epoxidation of the 8,9-double
bond and also the presence of the cyclopentenone ring of the B compounds (when compared
with the six-membered lactone ring of the G compounds) may play a major role in the harmful
effects (McLean and Dutton, 1995). The AFB,-8,9-epoxide and its hydration product, the
dihydrodiol form of AFB, (Figure 2) are highly reactive and can covalently bind to cellular
macromolecules. The epoxide specifically binds to the N7 position of guanine of DNA and
RNA (Cory and Wogan, 1981), while the dihydrodiol links to proteins by the formation of
a Schiff base structure (Hsieh, 1987). The major DNA adduct formed in vivo and in vitro

afier depurination is 8,9-dihydro-8-(N7—guanyl)-9-hydroxy AFB, (AFB ,-guanine), which can

 

o 0
° I
0 0 OCH3
Aflatoxin Bl (AFBl)
wane P450 (3)
o o
o
l
O
AFBl-8,9-epoxrde O
Epoxide HO I
hydrolase
0 o o OCH
N +
o o HN 7|
~ N
O NH2 N
HO l l
DNA-dR—DNA
”° ° ° °°H3 8,9-Dihydro-8-[N7-guanyl]-9-hydroxyAFB1
AFBl-8,9-dihydrodiol
Protein binding Mutation
Toxicity Cancer

Figure 2. Metabolic activation of aflatoxin B1 and its harmful effects.
(Eaton and Gallagher, 1994)

10

thus be used as a biomarker in urine for aflatoxin exposure (Groopman, 1994). Depurination
at guanine residues could lead to a GC —> TA transversion during replication. Evidence has
been collected which suggests that AFB, induces a mutation of the p53 tumor-suppressor
gene in codon 249 (Hsu et al., 1991; Bressac et al., 1991) and may result in the development
of human hepatocellular carcinoma. It has also been reported that the activated form of AFB,
induces virus expression and tumor formation associated with ras (McMahon et al., 1980)
and myc (Larson et al., 1980) oncogenes.
B. Aflatoxins and economic losses

While considerable research has been directed to the prevalence, chemical
characterization, and biochemical action of aflatoxins, less attention has been paid toward
calculating the economic impact of aflatoxins on society. This is because hidden or indirect
factors may result in the significant financial losses which are difficult to evaluate. Instead of
considering aflatoxins only, people generally take all mycotoxins into account when
evaluating economic losses. Nevertheless, it is believed that aflatoxin-associated losses play
a major part because of the widespread contamination of aflatoxins in the world and their
extremely toxic effects. It was estimated that billions of dollars are lost annually in direct and
indirect costs which result from the fact that approximately 25% of the world’s crops are
affected by mycotoxins (CAST, 1989). This huge loss encompassed a broad category of crop
and animal industries, and extends through the food chain from producers to the consumer
(Shane, 1994).

Because of the threat of aflatoxins to public health, the US. Food and Drug
Administration has set action levels of 20 ppb for aflatoxins in human food, 0.5 ppb for AFM,

in milk and dairy products, and 20 to 300 ppb in most animal feeds (CAST, 1989). In some

11

European countries the action level is even more restrictive. The cost to the agronomic
sector is thus increased due to efforts to meet these guidelines.

The preharvest mycotoxin contamination of corn was recognized in the mid-19705
(Lisker and Lillehoj, 1991). Since then, significant costs have been associated with mainly
two approaches to reduce preharvest aflatoxin contamination, namely the development of
resistant cultivars and the improvement of farming practices. For example, significant costs
arise from implementation of breeding programs which are utilized to select cultivars resistant
to insect damage, drought, and fungal infection. Financial losses also may arise from the
reduced yield of the resistant crops. The costs associated with improved farming practices
include irrigation to prevent desiccation, the use of insecticides or firngicides to prevent
infection, the use of additional fertilizers to reduce environmental stress, and the use of
modified harvesting methods to avoid crop damage. Costs can also result fi'om improvements
in postharvest handling of grains and other ingredients. These include modified transport and
mechanization to reduce damage, intensified drying of grain to achieve desirable moisture
levels, improved storage control to prevent filngal contamination, and proper quality control
procedures to monitor toxin levels. If, at last, the contamination level is too high, it is
necessary to reduce the toxin content by physical or chemical treatment or the contaminated
grain must be either destroyed or downgraded and hence reduce the revenue for producers.

Aflatoxin contamination is also responsible for financial losses in domestic animal
production, including ruminant, monogastric, and aquatic species (Nelson and Christensen,
1978). The major economic impact consists of aflatoxin-induced death (Smith and Hamilton,
1970), a depression in growth rate and feed conversion efficiency (Dalvi, 1986), and an

increase in plant condemnation (Shane, 1991). In the poultry industry, ducklings are the most

12

susceptible species to aflatoxicosis. Turkey poults are more resistant than ducklings but are
ten times more sensitive than four-week-old chickens.

In summary, aflatoxin contamination causes huge financial losses to food processors,
and producers of commodities and domestic animals. The costs ultimately are borne by

consumers or the national economy.

111. Control of aflatoxins

Because of the public health and economic problems caused by aflatoxins, it is very
important and urgent to find ways to efficiently control their occurrence. Ideally, any method
should be technically and economically feasible if it is to be applied practically. Some other
factors should also be considered including safety, retention of nutritional elements, and the
lack of hannfirl effects to the environment. To date numerous methods have been tested
although they still do not meet all of the control criteria. In general, the control of aflatoxins
can be divided into three principal categories : prevention, decontamination or detoxification,
and reduction of aflatoxin bioavailability.
A. Prevention

The prevention of aflatoxin formation in agricultural products and other foodstuffs can
occur at the pre- and post-harvest level by regulating the environmental factors influencing
fungal growth and toxin production. Many environmental conditions have been identified
which can promote aflatoxin formation in growing crops including : insect infestation,
drought conditions, mechanical damage, nutritional deficiencies, and unseasonable
temperatures and rainfall (Smith and Moss, 1985a). Conventional on-farm preventative

techniques such as methods of cultivating to improve plant vigor, the use of insecticides and

13

fungicides to reduce insect and fungal infestation, irrigation to avoid drought conditions, and
the use of resistant varieties (Darrah and Barry, 1991; Scott and Zummo, 1988) have been
utilized to help overcome these environmental stress although these are ofien too costly or
are inefi‘ective. Post-harvest (during processing, storage, and shipment) control of aflatoxin
contamination prevents or delays toxigenic mold growth through manipulation of moisture
levels, temperature, aeration, and mold spore density (Darrah and Barry, 1991).

Biocontrol using nontoxigenic strains of A. parasiticus and A. flavus is another
potential approach for the preharvest prevention of aflatoxin contamination. It has been
demonstrated that this strategy can significantly reduce aflatoxin contamination in peanuts
and cottonseed (Cotty, 1990; Domer et al., 1992; Ehrlich, 1987). Biocompetitive control,
however, has to be evaluated for its feasibility regarding the stability of the non-aflatoxin
producing strains and the environmental impact. The potential for naturally occurring
nontoxigenic strains to produce aflatoxins may be a major concern (Rarick et al., 1994) of
this technology. Additional studies are needed to address the microbial ecological changes
after releasing the biocontrol agent in the field.
B. Decontamination or detoxification

Once a product is contaminated with aflatoxins, there are only two options if it is to
be used for human or animal consumption : to remove aflatoxins or to degrade them into non-
toxic compounds. Practical methods are being investigated for removal and detoxification
of aflatoxins from foods and feeds.
1. Removal of aflatoxins

(a) EMS—Cm. In large particle size agricultural products such as the peanut,

Brazil nut or almonds, aflatoxin contamination, when it occurs, is normally confined in any

14

batch to a small number of contaminated seeds or kernels (Smith and Moss, 1985a). It has
been shown that the level of aflatoxin in peanuts can be correlated with the proportion of
loose-shelled or discolored kernels. When these are discarded the remaining kernels are
relatively free of aflatoxins. Off-colored kernels can be separated either by hand or by passing
through color sorters (Cole, 1898). Density segregation by flotation in water or a salt
solution is another way to separate toxic kernels from sound, nontoxic kernels (Hagler, 1991;
Huff, 1980; Huff and Hagler, 1985; Kirksey et al., 1989). Physical separation, however,
causes some loss of raw materials.

(b) Wanting. Numerous solvent extraction systems have been developed
to remove aflatoxins from contaminated materials with minimal effects on protein content or
nutritional quality (Rayner, 1977). These systems include 95% ethanol, 90% aqueous
acetone, 80% isopropanol, hexane-methanol, and hexane-acetone-water mixtures. Although
solvent extraction can be highly successful in removing aflatoxins, the cost of the additional
processing and the need for special solvent-removing equipment, etc., have made these
processes of questionable economic value. Besides, residues of solvent in food and feed
would cause additional safety problems.

2. Degradation or detoxification

(a) Winds. Of these methods, irradiation and thermal inactivation are
considered here. Aflatoxins are sensitive to ultraviolet light. However, the degradation of
aflatoxin in contaminated products is dependent on the nature of the solvent, the toxin
concentration and the length of exposure to UV light (Shantha, 1986). Aflatoxins are
resistant to thermal inactivation and are not destroyed completely by boiling water,

autoclaving, or a variety of food and feed processing procedures (Phillips, 1994). With dry

15

heat such as roasting, temperatures approaching the melting point (250°C) of aflatoxin must
be used to effectively degrade the toxin. Increasing the moisture content and/or time of
heating increases the rate of aflatoxin degradation (Mann, 1967). However, the adverse
effects of heat treatment on the appearance and nutritional value of the product makes the
practical application of these methods highly doubtful.

(b) WM. Several chemicals have been tested for the destruction of
aflatoxins including acids, aldehydes, oxidizing agents and gases. The treatment of grain with
ammonia appears to be a valuable approach to the detoxification of aflatoxins. Ammonia
used as an anhydrous gas at elevated temperatures and pressures can cause a 95-98%
reduction in total aflatoxin concentration in peanuts, cottonseed meal and corn (Brekke et al.,
1977; Gardner et al., 1971; Park et al., 1984). This method is legal and is being used on a
commercial scale in certain states of the US (Alabama, Arizona, California, Georgia, North
Carolina and Texas) for animal feeds. Ammoniation is also used routinely in Mexico, France,
South Afiica, Senegal, Brazil, and India (Piva et al., 1995; Phillips et al., 1994). The
drawback of ammonia treatment is the need to build special plants because ammonia corrodes
metal and becomes explosive in the air. This would result in a substantial increase in costs
which cannot be afforded by most farmers. Besides, ammonia treatment may lead to an
undesirable brown color in the feed and the diminished content of specific amino acids such
as cystine, methionine and lysine (Piva et al., 1995).

Oxidizing agents such as sodium bisulfite have been shown to degrade aflatoxins in
naturally contaminated grain (Doyle and Marth, 1978). When compared with ammonia
treatment, bisulfite treatment is less efficient in detoxification of aflatoxins. Bisulfite

treatment, however, is much less costly than ammonia treatment. In addition, sodium bisulfite

16

is commonly added to food and drinks where it acts as antioxidant, enzyme inhibitor, and
bacteriostatic agent. It therefore may be competitive with the ammoniation process.

Calcium hydroxide is another chemical which has been shown to be able to reduce
aflatoxin levels (Codifer et al., 1976). The advantages of using sodium hydroxide are its low
cost and easy application because it is the cheapest alkali and can be readily mixed with the
feed to be detoxified. However, the low efficiency (less than 45 %) in the destruction of
AFB, is a big concern (Piva et al., 1995).

(clljiglggical methods. Many microorganisms including bacteria, actinomycetes, yeast,
molds, and algae show varying abilities to degrade aflatoxin. The most active organism so
far discovered is Flavobacterium aurantiacum (NRRL B-184) which in aqueous solution can
take up and metabolize aflatoxin B,, G, and M, (Ciegler et al., 1966). As yet no commercial
application has been developed because the safe and practical use of Flavobacterium
aurantiacum in feed or food has not been established.

C. Reduction of aflatoxin bioavailability
1. Selective chemisorption

Numerous studies have demonstrated that the use of clays in contaminated feeds can
reduce aflatoxin absorption in the intestine of animals. Tests in vitro showed that absorbents
such as aluminas, silicas, and aluminosilicates are capable of binding aflatoxin in solution
(Phillips et al., 1988). It was observed that hydrated sodium calcium aluminosilicates
(HSCAS) were the most efficient in binding aflatoxins. HSCAS are currently used as
anticaking agents for animal feeds and were found to prevent aflatoxicosis in domestic animals
(Davidson et al., 1987; Kubena er al., 1990) and to decrease the level of AFM, in the milk

of dairy cattle (Harvey er al., 1991).

l7
2. Chemoprotection against aflatoxin toxicity

Methods have been suggested for the protection of animals and humans against the
effects of aflatoxins by prior treatment with chemicals or drugs that induce protective
detoxifying liver enzymes (Kensler et al., 1991). A dithiolethione compound, oltipraz, has
been demonstrated to be a potent inhibitor of AFB ,-induced hepatocarcinogenesis in rats. It
is believed that this drug induces phase II enzymes, including the glutathione S-transferase,
and thus could enhance the detoxification of AFB, (Ansher et al., 1986; Bueding et al.,
1982)

In summary, since no practical method is currently available for preharvest prevention
of aflatoxin contamination of foods and feeds, all potential methods for postharvest aflatoxin
detoxification must be considered as important at this point in time. Nevertheless, the
prevention of aflatoxin contamination before harvest is the best long-term approach because
the technology would eliminate or reduce the need for handling aflatoxin-contaminated

commodities by growers or processors.

IV. Biosynthesis of aflatoxins
A. Chemical and biochemical aspects

Because of the extreme toxicity of aflatoxins, elucidation of the biosynthetic pathway
has become a very popular topic since their discovery in 1960. To date aflatoxin biosynthesis
is the best characterized biosynthetic pathway of fungal secondary metabolism. Studies using
blocked mutants, metabolic inhibitors, and radiolabeled precursors in bioconversion
experiments have lead to a relatively clear picture of the biochemical pathway utilized for

synthesis of AFB, (Bennett et al., 1980; Bhatnagar er al., 1987; Hsieh et al., 1973, 1976; Lee

18

etal., 1976; McCormick et al., 1987; Shroeder et al., 1974; Steyn et al., 1980). Data on the
biosynthetic intermediates and enzymes in the AFB, pathway have been collected (Bhatnagar
et al., 1992; Dutton, 1988) and the putative biosynthetic scheme is shown in Figure 3. AFB,
is a polyketide-derived secondary metabolite which is synthesized from acetate and malonate
in a process analogous to fatty acid synthesis. As described by Bennett and Christensen
(1983), the precursors in biosynthetic pathway include one acetyl CoA and 9 malonyl CoA.
The early steps of AFB, biosynthesis are the same as those in fatty acid biosynthesis with
condensation of acetyl COA and 2 Malonyl COA molecules in the presence of NADPH to
form a hexanoate starter unit (Towsend et al., 1984). Condensation of this starter unit with
7 malonyl COA molecules then proceeds to form a C20 polyketide chain without further
ketoreduction. Cyclization and oxidation then occur to form the first stable anthraquinone
intermediate, norsolorinic acid (NA), which is then converted to (sequentially) averantin
(AVN), averufanin (AVNN), averufin (AVF), versiconal hemiacetal acetate (VHA),
versiconal (VHOH), versicolorin B (VB), versicolorin A (VA), sterigmatocystin (ST), 0-
methylsterigmatocystin (OMST), and the final product AFB ,.

Mutants blocked in aflatoxin biosynthesis played an important role in elucidating this
pathway. In this regard, the pathway blocked mutants of A. parasiticus were more useful
than those of A. flavus. This is because mutants isolated from A. parasiticus accumulate
pigmented pathway intermediates. Most of the mutants of A. flavus, on the other hand, do
not accumulate pigments. For example, the pigments NA, AVN, AVF, and VA were
identified from A. parasiticus non-aflatoxin producing mutants. These four pigments were
then isolated fi'om fungal mycelia and their chemical structures identified. Radiolabelled

pigments were used in feeding studies to demonstrate that they could be converted to AFB,

l9

Acetyl COA
Malonyl COA
k, 0 OH o OH OH OH O OH
H3 CH3
——>
we“ We“
0 O
Norsolorinic Acid (NA) / Averantin (AVN)
OH O OH OH O OH
0 CH3 0
4——
... meme ..
O o
Averufin (AVF) \ Averufanin (AVNN)
OH O OH OH o OH
O
——->
M. .. . «a .. WIT“...
O O
Versiconal Herniacetal Acetate (VHA) / Versiconal (VHOH)
OH o OH OH O OH
I <—— O
at“, Fr“,
0 O
Versicolorin A (VA) Versicolorin B (VB)
@0013
O
o O OCH3 o o OCH3
Sterigmatocystin (ST) /0-methylsterigmatocystin (OMST)
O O
° 1
o o OCH3

Aflatoxin B 1 (AFB 1)

Figure 3. The proposed biosynthetic pathway for aflatoxin Bl

20

by toxigenic fungi. The sequential order of these intermediates was determined according to
their efficiencies of conversion to AFB, and their relative chemical structures. AVNN, a
metabolite of A. parasiticus was then demonstrated to be incorporated into AFB, and was
placed between AVN and AVF based on a logical organic conversion mechanism
(McCormick et al., 1987). VHA was identified because it accumulated after inhibition of an
esterase activity by the insecticide dichlorvos (Y ao and Hsieh, 1974). The intermediates after
VA are not pigmented. ST and OMST isolated from A. versicolor and A. parasiticus,
respectively, were placed downstream from VA by a bioconversion assay (Hsieh et al., 1973;
Bhatnagar et al., 1987).

It is estimated that at least 17 enzymatic activities are associated with this complex
pathway (Bhatnagar et al., 1992; Dutton, 1988). Several pathway enzymes have been
purified to homogeneity including two NA reductases (N A~ AVN; Bhatnagar and Cleveland,
1990; Chuturgoon et al., 1990), two VHA reductases (VHA-wersiconol acetate[VOAc];
Matsushima et al., 1994), one versiconal cyclase (VHOH~VB; Lin and Anderson, 1992;
Townsend et al., 1991), and two O-methyltransferases (ST-OMST; Bhatnagar et al., 1988;
Keller et al., 1992). Those enzymes which have been identified or partially purified are the
esterase (VHA-VHOH; Yabe and Hamaski, 1993), the desaturase (VB-'VA; Yabe et al.,
1991), and the oxidoreductase (OMST-AFB 1; Bhatnagar et al., 1989). Nevertheless, many
other enzymes involved in the biosynthetic pathway have not been identified.

B. Genetic and molecular aspects
1. Classical genetic analysis
Conventional genetic analysis of aflatoxin biosynthesis was hampered because neither

A. flavus nor A. parasiticus has a sexual reproductive stage. However, genetic linkage of

21

genes for aflatoxin biosynthesis in both fungi was studied using the parasexual cycle (Bennett,
1979; Papa, 1973). The genetic markers in genetic linkage analysis included auxotrophic
mutants, spore color mutants, and aflatoxin pathway mutants. The genetics of A. flavus,
however, is better understood than that of A. parasiticus. Over 30 genes including 11
aflatoxin genes have been mapped to 8 linkage groups (Bennett and Papa, 198 8). Of the 11
aflatoxin loci mapped, l is on linkage group II, 9 are on linkage group VII, and 1 is on
linkage group VIII. The aflatoxin loci are nonallelic and recessive in diploids, with the
exception of the mutant containing the afl-I allele.

The generation of physical mapping data of the aflatoxin genes was enhanced by the
development of the electrophoretic karyotype analysis. Using pulsed-field gel electrophoresis,
the genomes of A. flavus and A. parasiticus have been separated into 6 to 8 chromosomes
with sizes ranging from 3 to 7 Mb (Keller et al., 1992). In an attempt to assign the linkage
groups identified from genetic linkage studies to these separated chromosomes, Foutz et al.
(1995) have cloned seven previously mapped auxotrophic genes which hybridized to 7
individual chromosomes. These specific probes could help in future studies to determine the
karyotypic map of aflatoxin genes.

2. Molecular genetics

A logical approach to prevent aflatoxin contamination is to block their production in
the field at the preharvest stage. The development of a thorough understanding of aflatoxin
biosynthesis at the molecular level may aid in this approach. Cloning of genes associated with
aflatoxin biosynthesis is the first step to effectively understand the regulation of gene
expression. The cloned genes can be used as targets for gene disruption to generate

genetically stable nontoxigenic strains of Aspergillus spp. which can be utilized in the field

22

as biocontrol agents. On the other hand, cloning of aflatoxin pathway genes and subsequent
identification of regulatory genes will provide molecular probes for investigating the specific
molecular regulation of aflatoxin biosynthesis in fungal culture and in host plant tissue in
which aflatoxin contamination occurs. This could lead to the identification of candidate
compounds which are able to block aflatoxin biosynthesis in firngi. It may be possible to
directly use these compounds in the field or to construct resistant plants which naturally
produce these compounds.

Three different strategies have been used to isolate genes involved in aflatoxin
biosynthesis. Differential screening of a cDNA library, a relatively nonspecific approach, has
been used to identify genes in A. parasiticus (F eng et al., 1992) and A. flavus (Woloshuk and
Payne, 1994) that may be associated with aflatoxin biosynthesis. This method is based on the
principle that aflatoxin associated genes are transcribed under the conditions supporting
aflatoxin production. These genes, on the contrary, are not transcribed or are transcribed in
much lower amounts under non-aflatoxin supporting conditions. Therefore toxin specific
cDNA clones theoretically could be isolated by hybridization to RNA extracted from
aflatoxin-producing cultures but not to RNA from non-aflatoxin—producing cultures. The
major disadvantage of this approach is that many nonspecific cDNA clones may be obtained
and cause difficulties in the selection and study of their exact firnctions.

A second approach is reverse genetics. In this method, it is necessary to identify and
purify pathway proteins which can be used to generate antibodies for use as immunoscreening
probes. Alternatively, protein sequence data should allow the design of oligonucleotide
probes to isolate the specific genes. Using a reverse genetics approach, a 1.5-kb genomic

DNA clone (pF9-1) from A. flavus NRRL 3357 was identified with an oligonucleotide based

23

on the amino acid sequence of the N-terminus of the purified methyltransferase which
converts ST to OMST (Keller et al., 1992). This omrA gene also has been cloned from A.
parasiticus by using antibodies (raised to the purified methyltransferase) to screen a cDNA
expression library (Y u et al., 1993). Recently, an A. parasiticus dehydrogenase gene, norA,
was cloned by the same strategy using a monoclonal antibody raised against a purified
norsolorinic acid reductase involved in the conversion of NA to AVN.

A third approach involves genetic complementation of firngal mutants deficient in
aflatoxin biosynthesis followed by rescuing genes associated with aflatoxin production.
Genetic transformation systems have been developed for A. parasiticus (Horng et al., 1990;
Skory et al., 1990) and A. flavus (Woloshuk et al., 1989). Using these transformation
systems, the nor-1 gene, associated with the conversion of NA to AVN (Chang et al., 1992),
the ver-I gene, associated with the conversion of VA to ST (Skory et al., 1992), and the fas-
[A gene, encoding a putative fatty acid synthetase involved in polyketide backbone synthesis
(Mahanti et al., 1996), were cloned by genetic complementation of A. parasiticus mutants
blocked at unique steps in AFB, synthesis. Complementation was performed by
transformation of a cosmid DNA library constructed using genomic DNA from a wild-type
aflatoxin-producing strain A. parasiticus NRRL 5862 (SU-l). A regulatory gene, aflR, was
first isolated from A. flavus by genetic complementation (Payne et al., 1993). The homologue
of A. flavus aflR was later identified in A. parasiticus based on its ability to upregulate
aflatoxin biosynthesis when it was transformed into a wild-type toxigenic strain (Chang et al.,
1993).

The isolation of these pathway genes made it possible to elucidate a detailed physical

map of the genes involved in aflatoxin biosynthesis. In the process of cloning the nor-1 and

24

ver-I genes fiom A. parasiticus, it was found that these two genes are located within a 35-kb
genomic DNA fragment contained in cosmid NorA and on a single chromosome. The
aflatoxin gene cluster was firrther confirmed by mapping the position of fax-1A, aflR, and
omtA in the genomic DNA of both A. flavus and A. parasiticus. Transcriptional mapping
studies have shown that this cluster may extend up to 60-kb (Trail et al., 1995 ; Yu et al.,
1995). Taking advantage of transcripts identified in this cluster, it was possible to elucidate
the function of originally non-identified genes by gene disruption analysis. For example, a
pksA gene encoding a putative polyketide synthase was identified adjacent to nor-1 by gene
disruption analysis (Chang et al., 1995; Trail et al., 1995). The order of aflatoxin genes in the
clusters is similar in the two aflatoxin-producing Aspergillus. The spacing between these

genes, however, is different (Yu et al., 1995).

V. Regulation of aflatoxin biosynthesis

As a secondary metabolite, aflatoxins are produced when the fungus reaches the end
of active growth phase during batch fermentation. Because the precursor (i.e. acetate) of
aflatoxin biosynthesis is the product of the primary metabolism, the factors which regulate
primary metabolism could influence aflatoxin biosynthesis. Carbohydrate metabolism has
been shown to affect the production of aflatoxins. Several simple carbon sources, especially
glucose, sucrose, and maltose, have been shown to support and stimulate aflatoxin production
(Luchese and Harrigan, 1993). Glucose can be catabolized by aspergilli either by way of the
Embden-Meyerhoff or the hexose monophosphate pathways simultaneously (Zaika and
Buchanan, 1987). Aerobic conditions favor utilization of the hexose monophosphate pathway

and anaerobic conditions favor the Embden-Meyerhoff pathway. It has been reported that

25

the amount of aflatoxin production depends on the pathway by which glucose is catabolized
by the fimgi (Shih and Marth, 1974). The study showed that less aflatoxins were produced
in extensively aerated cultures of A. parasiticus and the efficiency of [l-“C]glucose
incorporation was less than that in stationary cultures. Shih and Marth (1974) concluded that
in less aerobic conditions oxidation of acetate (via the citric acid cycle) would be decreased
and more acetate would be available for synthesis of aflatoxins. It has also been proposed
that a low NADPH/NADP ratio favors aflatoxin production (Niehaus and Dilts, 1984). The
increased aflatoxin production observed in a less aerobic environment could therefore result
from the decreased formation of NADPH via the hexose monophosphate pathway. An
alcohol dehydrogenase gene, adh], has been cloned from A. flavus (Woloshuk and Payne,
1994). The transcription of this gene is induced during aflatoxin production. Although the
exact role of adh] in aflatoxin biosynthesis in not known, it is hypothesized that alcoholic
fermentation may be important for the use of glucose through the glycolytic pathway under
less aerobic conditions and may be associated with aflatoxin biosynthesis.

Instead of relying solely on the regulation of primary metabolism, aflatoxin
biosynthesis could also be regulated at the level of secondary metabolism. Skory et a1. (1993)
have reported that the regulated expression of the aflatoxin pathway genes, nor-1 and ver-I,
is in part at the level of transcription. The accumulation of the RNA transcripts from other
aflatoxin associated genes was shown to follow a similar pattern as nor-1 and ver-I (Trail et
al., 1995). This suggested that aflatoxin genes may be regulated by a common regulatory
factor or some specific factors activated at the same time for aflatoxin synthesis. The genetic
evidence suggests that aflR is a specific regulatory gene in aflatoxin biosynthesis. Metabolite

feeding studies demonstrated that aflR is required for the conversion of several different

26

pathway intermediates to aflatoxins (Payne et al., 1993). Moreover, the aflR gene is capable
of inducing transcription of several aflatoxin pathway genes and expression of several
aflatoxin pathway enzymatic activities simultaneously in cell extracts (Chang et al., 1995;
Payne er al., 1993). Analysis of the predicted amino acid sequence of aflR showed that there
is a cysteine—rich binuclear zinc cluster DNA-binding motif, Cys-Xaa2-Cys—Xaa6-Cys—Xaa6-
Cys-Xaa2-Cys-Xaa6-Cys, which has been found in several fungal transcriptional regulatory
proteins (W oloshuk et al., 1994).

In addition to aflR, a putative regulatory locus, afl-I, was identified in A. flavus
(Bennett and Papa, 1988). Recently, it was demonstrated that a heterozygous diploid strain
with both mutant (afl-I) and wild-type (afl-I’“) alleles lost its ability to produce aflatoxin and
to transcribe the nor-1, ver-I, and omtA genes. However, expression of the aflR gene was
not suppressed (Woloshuk et al., 1995) which suggests that aflR is not the only regulator of
AFB, biosynthesis. The real function ofafl-I in regulation of aflatoxin biosynthesis will not
be clear until the afl-I gene can be isolated in the future study.

Since several aflatoxin pathway genes have been cloned, it is now possible to elucidate
the cis-acting DNA elements and trans-acting proteins that regulate aflatoxin synthesis. One
approach to accomplish this task is to fuse the aflatoxin gene promoter to a reporter gene
such as the GUS gene (uidA gene; encodes B-glucuronidase). Fungal strains containing these
reporter constructs could be used to identify the cis-acting elements (Trail et al., 1994; Wu
and Linz, 1994). Afier identification of the cis-acting elements, promoter fragments can be
used in mobility shift assays to identify and thereafter purify the specific trans-acting proteins

which regulate aflatoxin biosynthesis.

CHAPTER 2

STRUCTURAL AND FUNCTIONAL CHARACTERIZATION
OF THE VER-I GENE

I. INTRODUCTION

In a previous study (Skory et al., 1992), the ver-I gene was cloned based on its ability
to complement the versicolorin A (VA) accumulating strain, A. parasiticus CSlO (ver-I, wh-
1, pyrG; derived from ATCC 36537; Lee et al., 1975), to produce aflatoxins. The nucleotide
sequence of a genomic DNA fragment and cDNA fragment covering the entire ver-I coding
region has been determined (Skory et al., 1992). The predicted amino acid sequence,
deduced fi'om the ver-I nucleotide sequence, was compared with the EMBL and GenBank
data bases. The search revealed striking similarity with Streptomyces ketoreductases involved
in polyketide biosynthesis. This observation resulted in the prediction that the ver-I gene
encodes a protein that has enzymatic activity associated with AFB, biosynthesis.

Southern hybridization analysis using the cloned ver-I gene as a probe indicated that
there are two copies of the ver-I gene in A. parasiticus SU—l (Skory, 1992). By in situ
colony hybridization of the genomic DNA library constructed by Skory et al (1992), two
cosmid vectors, NorA and Ver2, were found to contain the ver-I gene. The location of the
ver-I gene on the two cosmid vectors is shown in Figure 4. The gene copy located on a 2.1-

kb EcoRI fragment of the cosmid NorA was named ver-IA. The other ver-I gene copy

27

28

Figure 4. Schematic representation of the location of ver-IA and ver-IB on cosmids NorA
and Ver2, respectively. The number in parentheses is the size of the restriction fi'agment in

kilobase pairs. The arrow indicates the duplicated region. B, BamHI; E, EcoRI; S, SalI.

29

 

mm mm vm ON 3

N~ w v o 0.x
A 2.58 «as:
”NINA“
:3 i 5.3....
m m m m
3.8 ad 8.:

I

a. A 2580 <3: v

m w V7.8... ES: 5? V~.§\ 78: V23

 

 

 

m m mm m m m m m m
ad ad $8 5: 68?: ad 86

8.3 6.3

Figure 4.

30

located on a 5.1-kb EcoRI fragment of the cosmid Ver2 was designated ver-IB. The flanking
regions (3.2-kb EcoRI fragments; see Figure 4) Of ver-IA and ver-IB genes were found to
hybridize to each other in Southern hybridization analysis. This indicated that there is a
duplicated region around the ver-I gene in the chromosomal DNA.

Based on the complementation analysis and nucleotide sequence data, it was
hypothesized that the cloned ver-I gene encodes a protein involved in the conversion of the
AFB, pathway intermediate VA to sterigmatocystin (ST). To address this hypothesis, the
following experiments were designed and accomplished in this study. First, restriction
fiagment length polymorphism (RFLP) analysis of ver-IA, ver-IB, and the cloned ver-I gene
confirmed that the gene previously cloned is ver-IA. Second, a duplicated chromosomal
region (approximately 12-kb) was identified upstream from ver-IA and ver-IB by Southern
hybridization analysis. Third, the nucleotide sequence of ver-IB was determined. A
translational stop codon, found in the ver-IB coding region, indicated that it encodes a
truncated polypeptide with no firnction. Fourth, recombinational inactivation and genetic
complementation experiments confirmed that ver-IA is the only functional copy of ver-I in
A. parasiticus SU—l and that its gene product is directly involved in the conversion of VA to

ST.

11. MATERIALS AND METHODS
A. Strains and plasmids

Escherichia coli DHSa F' ° [F '/endAI hst17 (r; m; ) supE44 thi-I recA] gyrA
(Nal’) reIA1A(lacZYA-argF)u,69(m801acZAM15)](Gibco BRL, Life Technologies, Inc.

Gaithersburg, MD) was used to propagate plasmid DNA. A. parasiticus NRRLS 862 (SU-l;

31

ATCC 56775; Bennett, 1979), a wild-type aflatoxin-producing strain, was used as the control
strain. The nitrate reductase (niaD) deficient strain, NR-l , derived from A. parasiticus SU- 1,
was used as the recipient strain for ver-I gene disruption analysis. VA, used as a standard in
thin-layer chromatography (TLC) assays, was purified from A. parasiticus ATCC 36537 (ver-
1, wh-I) according to the procedure of Lee et al. (1975).

Plasmid pDV-VA (Figure 5A) was constructed for ver-I gene disruption. A 4.4-kb PstI
genomic DNA fragment containing ver-IA was inserted into pBluescriptII SK(-) to generate
pVer-AP. This plasmid was cut near the middle of the ver-IA coding region (StuI) and blunt-
end ligated to a 6.2-kb PvuII fragment containing the functional niaD gene. For ver-I
complementation experiments, plasmid pVer-Ben (Figure 5B) was constructed. A 4.1-kb
XbaI/XhoI DNA fragment containing ver-IA was subcloned into pBluescriptII SK (-) to
generate pVer-AX. Then, a 7.0-kb XbaI fragment containing the ben1r gene (confers
resistance to benomyl) fi'om pYTl (Wu et al., 1996) was inserted into pVer-AX to generate
pVer-Ben. pSL82 (Horng etal., 1990), a plasmid containing a complete copy ofniaD, and
pYTl, containing ben’, were used as positive controls in recombinational inactivation and
genetic complementation experiments, respectively.

B. Bacterial cell transformation and plasmid purification

The preparation and transformation of competent cells were conducted by a calcium
chloride method (Ausubel et al., 1993). Recombinant cells were screened on MacConkey
agar (Difco) or selected by standard colony hybridization techniques (Maniatis et al., 1989).
Minipreparation of plasmid DNA using alkaline lysis and large scale preparation of plasmid
DNA by CsCl/ethidium bromide equilibrium centrifugation were conducted using standard

methods described by Ausubel et al.(1993).

32

Figure 5. Plasmids used for functional analysis of the ver-I gene in recombinational
inactivation and complementation experiments. (A) The plasmid pDV-VA used for
recombinational inactivation of the ver-I gene. The white block region is a 6.2-kb PvuII
fragment containing the niaD gene. The black blocks represent the coding region of ver-IA
split by the niaD containing fragment. The hashed regions are the 5' and 3' flanking regions
Of ver-IA. The single black line represents pBluescriptII SK(-). (B) The plasmid pVer-Ben
used for complementation of A. parasiticus VAD-102. The white block is a 7.0-kb XbaI
fragment containing the ben’ gene. The black block is the coding region of ver-IA. The
hashed blocks are the 5' and 3' flanking regions of ver-IA. The single black line represents

pBluescriptII SK(-).

33

Figure 5.
A. ,
pDV-VA
13.5 kb
B.

14-0 lrb

 

34

C. Transformation of fungal protoplasts

Fungal protoplasts were transformed using a polyethylene glycol procedure (Oakley
et al., 1987) with minor modifications as described by Skory et al. (1990). Approximately
108 conidia of A. parasiticus NR-l were incubated in yeast extract-sucrose liquid medium
(YES; 2% yeast extract, 6% sucrose, pH 5.8) for 15 hr at 29°C with shaking (150 rpm) in the
dark. The mycelia were harvested and digested with Novozyme 234 (Novo Industries,
Danbury, Conn.) to generate protoplasts. To 100 pl protoplast suspension, 5 - 10 ug DNA
in 10 ul TE buffer (lOmM Tris-HCl, lmM EDTA, pH 8.0) and 50 pl PEG solution (25%
polyethylene glycol 3350, 50mM CaC12, lOmM Tris-HCl, pH 7.5) were added. The mixture
was incubated on ice for 20 min. 1 ml PEG solution was added and the mixture was firrther
incubated at room temperature for 30 min. Finally, the protoplasts were spread onto selective
agar media. Cells transformed with plasmid pSL82 and pDV-VA were screened for their
ability to utilize nitrate on Czapek-Dox agar (CZ agar, Difco), a defined medium containing
20% sucrose as the osmotic stabilizer. When transformed with pYTl or pVer-Ben, benomyl
resistant transformants were selected on CZ containing benomyl (2 ug/ml). Transforrnant
colonies were then transferred to coconut agar medium (CAL/I) for screening aflatoxin
production by visualization of blue fluorescence under UV light (Davis et al., 1987).
D. Preparation and analysis of genomic DNA from fungal cells

YES broth was used to grow fungal mycelia for preparation of genomic DNA. 100
ml of YES in a 250 ml Erlenmeyer flask was inoculated with 2x106 spores of individual fungal
isolates and incubated on a rotary shaker at 150 rpm at 30°C in the dark. Cultures were
grown for 48 h and a phenol-chloroform protocol previously described (Skory etal., 1990)

was used to isolate genomic DNA from mycelia. Restriction enzymes were purchased from

35

Boehringer Mannheim Biochemicals (Indianapolis, IN) and used according to the
manufacturer's instructions. Enzyme digestion, agarose gel electrophoresis, and Southern
hybridization analyses were performed according to standard procedures (Ausubel et al.,
1993). Radiolabeled DNA probes were generated with a Random-Primed DNA Labeling kit
(Boehringer Mannheim Biochemicals) by incorporation of [a-32P]dCTP (DuPont).
E. Analysis of versicolorin A and aflatoxin production

To qualitatively determine the metabolites which accumulated in transforrnant
colonies, TLC analysis was performed on activated silica TLC plates ( 10 by 10 cm) using
chloroform-acetone (95:5) as a solvent system. A mixture of aflatoxin B,, B,, G, and G,
(Sigma) and semipurified VA were resolved on the same plate as reference standards. To
identify the production of VA in ver-I gene disrupted transformants, TLC analysis was
performed using benzene-acetic acid (95:5) as a solvent system. The yellow pigment that
comigrated with the VA standard was scraped from the plates and dissolved in methanol or
ethanol and the absorbance spectrum from 200 to 600 nm was detemrined. The spectrum was
compared to previously published spectrum data for pure VA (Hamasaki et al., 1967).
F. Nucleotide sequence analysis

A 1.7-kb EcoRI/HindII DNA fragment containing the ver-IB gene was sequenced
using eight overlapping subclones which were inserted into pBluescriptII SK(-) (Figure 6).
DNA sequence analysis was performed on both strands with T3 and T7 primers by the
dideoxy-chain termination method (Sanger et al., 1977) with an automated nucleotide
sequencer (ABI robotic catalyst and 373A DNA sequencer) at the Plant Research Laboratory
at Michigan State University. Nucleotide sequence data were analyzed with the Wisconsin

Genetics Computer Group (GCG) package. A comparison between the predicted amino acid

36

Figure 6. Strategy for nucleotide sequence analysis of the 1.7-kb EcoRI-HindII fragment
which contains the ver-IB gene. Several subclones were cloned into the plasmid
pBluescriptII SK( -) and T3 and T7 primers were used for double-stranded sequencing. The
arrows indicate the direction and extent of sequencing. The black region represents the open

reading frame. B, BamHI; E, EcoRI; H, HindII; K, KpnI; M, MscI; S, SmaI.

37

Figure 6.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

\\ l— — _
En... PPE ”Pl
\\ d _ _ _ _
”FE ”FE Eh...
— _ — A
I w v. 2.: m v. .2 m

9M Ho

38

sequences of ver-IA, ver-IB, the Streptomyces coelicolor actIII gene, and the Magnaporthe

grisea ThnR gene was made with Gap and aligned with Pileup software (GCG).

III. RESULTS
A. Restriction fragment length polymorphism (RFLP) analysis

During the isolation of the ver-I gene, several DNA fragments containing the ver-I
gene were obtained by marker rescue from an A. parasiticus CS-10 aflatoxin-producing
transforrnant (Skory et al., 1992). In the current study, several DNA fi'agments were
subcloned fiom cosmid NorA and Ver2. The DNA fragments subcloned from cosmid NorA
included a 2.1-kb EcoRI fragment (containing var-1A), a 3.2-kb EcoRI fragment (5' flanking
region of the ver-IA), and a 2.5-kb SalI fragment (3' flanking region of the ver-IA). The
DNA fragments subcloned from cosmid Ver2 included a 5.1-kb EcoRI fragment (containing
ver-IB), and a 3.2-kb EcoRI fragment (5' flanking region of the ver-IB) (see Figure 4 for
relative location of these DNA fragments). Restriction enzyme mapping was conducted on
these subcloned DNA fragments and the restriction patterns were compared with restriction
maps of the DNA fragments containing the cloned ver-I gene. The results indicated that the
cloned ver-I gene is ver-IA.
B. Identification of a duplicated chromosomal region containing the ver-I gene

The 5' flanking region of ver-IA (3.2-kb EcoRI fragment) hybridized to a similar sized
DNA fiagment upstream from ver-IB indicating that there is a duplicated region flanking the
ver-I gene in the chromosomal DNA. To determine the extent of this duplication, Southern
hybridization analysis of A. parasiticus genomic DNA was performed using DNA fragments

adjacent to ver-IA (isolated from cosmid NorA) as probes (see schematic in Figure 4). At

39

least two DNA fragments hybridized with the 1.4-kb, 3.2-kb, and 2.1-kb EcoRI probes in
genomic DNA digests but only one fragment was detected with the 1.9-kb EcoRI and 1.5-kb
BamHI/SalI probes (Figure 7). Since only one of the restriction enzymes utilized in the
analysis cut within any of the fragments used as probes (SmaI out within the 2.1-kb EcoRI
fragment), the data suggested that the region of duplication extended approximately 12-kb
upstream from the ver-IA and ver-IB genes.
C. Nucleotide sequence analysis of ver-IB

Nucleotide sequence analysis of a 1.7-kb EcoRI/HindII DNA fragment containing ver-
IB was conducted on both DNA strands. The alignment of the nucleotide and predicted
amino acid sequences of Per-1A and ver-IB is shown in Figure 8. The data demonstrated that
these two genes share 93% identity in nucleotide sequence. The deduced amino acid
sequences of the products of ver-IA and ver-IB share 97% similarity and 95% identity. A
translational stop codon was identified in the coding region of ver-IB (see asterisk in Figure
8) indicating that ver-IB may encode a truncated and nonfunctional polypeptide.
D. Functional analysis of the ver-I genes via recombinational inactivation and

complementation

To confirm gene firnction, plasmid pDV-VA (Figure 5A) was designed to disrupt
either copy of the ver-I gene by homologous recombination (gene replacement). pDV-VA
was linearized with XbaI and XhoI and transformed into an AFB, producing strain A.
parasiticus NR-1(afl*, niaD'), in order to disrupt the ver-I gene by gene replacement through
double cross-over recombination at homologous sites flanking the ver-I gene (Figure 9).
Because of the similarity between the flanking regions of ver-IA and ver-IB, this disruption

vector could in theory be used to disrupt both ver-I genes. In two separate experiments, a

40

 

Figure 7. Identification of a duplicated chromosomal region containing the ver-I genes by
Southern hybridization analysis. The genomic DNA from A. parasiticus SU—l was digested
with several restriction endonucleases. Digested DNA was electrophoresed in a 0.8%
agarose gel, transferred to Nytran membranes and probed with DNA fiagments subcloned
from cosrrrid NorA. The location of probes indicated above the blots is shown in Figure 4.

B, BamHI; Bg, BgIII; E, EcoRI; P, PstI; S, SaII; Sm, SmaI; X, XbaI; XII, X7101.

41

Figure 8. An alignment of the nucleotide sequences and the deduced amino acid sequences
of ver-1A and ver-IB. A stop codon within the coding region of ver-IB gene is indicated by

an asterisk.

42

Figure 8.

1

1

101

101

198

201

298

301

398

401

498

500

35

598

600
35

69
698

700
69

102
798

800
102

119
898

900
119

152
998

1000
152

185
1098

1100
185

218
1198

1200
218

232
1298

1300
232

1398

1400

1498

1500

GAAIICACIICIAAAIGAIACAAGCGCGAAIAICICCGAIIAAGCCCACGIIAAGAGIAIIIICCAAGACAIGCAGGGACAGAIACAGACIICCCICAAG

IIIIIII IIIIIIIIIIIIIIIIIIIIII IIIIIIIIIIIIIIIIIIIIIIII III IIIIIIIIIIIIIIIIIlllllltlllllllllllllll
mmmmmmmcmmmmmmmmrmmmrmm

mmmrmmmmrmnmmmmmm. . .cc
lilll IIIII Illllllllllll lllllllllil ItlllllllllllllllllIIIIIIIIIIIIIII llllltllllll Illlltlll I
WWWWIMITWWWMW

IIIICGIICAIIAIIIIGIIIIIGGIGIGAIIGGICCAGAGCCIGCICCIAIICICAGCIICCIAIGCIIICAGCCIGCCAIAAACAAGAIGIAIIACIG
llllllllllllllllllllllllllllllllllllllllllllllll Illlllllllllllllllllllllllll Illlllllllll Illlllll
IIIICGIICAIIAIIIIGIIIIIGGIGIGAIIGGICCAGAGCCIGCICIIAIICICAGCIICCIAIGCIIICAGCCIACCAIAAACAAGACGIAIIACIA

CAIAGAAGIIIIAGGCICGCCGCCCGAIGAGCIACIGGICIICAGAIAIICGGICICCGAGGAAAGAIIIGIIIGGIGGCCAACCAICCAIAGCIGCGIA

Illlllllllll Illlllllllll llllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllll
CAIAGAAGIIIIGGGCICGCCGCCCCAIGAGCIACIGGICIICAGAIAIICGGICICCGAGGAAAGAIIIGIIIGGIGGCCAACCAICCAIAGCIGCGIA

var—1A * M
IAIAIGIACIACAIGCCCGIICCCCIGGGTCACCGIIIICACAGAACIACACAICAITIIGCCTCCACAAAAICICIACCAIACACGAICCCGICAGCAI
lllllllll llll ll llll Illlllllllllll llllllllllllllllllllllllll Ill llllllll ll lllll I lllllll
IAIAIGIACCGCAIG.ACGTICCCAIGGGICACCGTITTAACAGAACIACACAICAITTIGCCICCCIAAAGTCTCTACCCIAGACGAIAICTTCAGCAI

var—18 ~ I

8 D N H R L D G K V A L V I G A G R G I G A A I A V' A L G E R G A
GICGGAIAAICACCGIIIAGAIGGCAAAGIGGCCIIGGIGACAGGCGCCGGCCGCGGCAICGGIGCIGCCAICGCCGICGCCCIIGGIGAGCGCGGAGCC
III II I lllllllllllllllllllllllllllll llllllllllll lllllllllllllllllllllllllllllllllllllllllllllll
GICCGACAGCCACCGIIIAGAIGGCAAAGIGGCCIIGGICACAGGCGCCGGCIGCGGCAICGGIGCIGCCAICGCCGICGCCCIIGGIGAGCGCGGAGCC

8 D 8 H R L D G K V A L V I G A G C G I G A A I A. V A L G E R G A

K V V V N I A H 8 R E A A E K V V E Q I K A X G I D A I A I Q A D V
AAAGICGIGGIIAACIAIGCCCACICCCGCGAGGCCGCGGAGAAAGIGGIIGAACAGAICAAGGCCAAIGGIACCGAIGCIAICGCAAICCAGGCCGAIG
Illllllllll lllll llllllllllllllllllllllllll llllll lllllllllllllllllllllllll llllllllllllllll llll
AAAGICGIGGIGAACIACGCCCACICCCGCGAGGCCGCGGAGAAGGIGGIICAACAGAICAAGGCCAAIGGIACCGACACIAICGCAAICCAGGCGGAIG
K V V V N Y A H 8 R E A A E K V V O Q I K A N G I D A I A. I Q A D V

G D P E A I A K L M A E I V R H F G Y L D I V 8 8 N A G I V 8 P G
ICGGGGAICCIGAGGCGACAGCGAAAIIAAIGGCGGAGACGGIGCGCCAIIIIGGCIACCIGGACAICGIGICAICGAACGCIGGAAIIGIAICGIICGG
Illllllllllllllllllllllll IlllllllllIlllllllllllllllllllll lllllllllllllllll lllllllllllllllllllllll
ICGGGGAICCIGAGGCGACAGCGAAGIIAAIGGCGGAGACGGIGCGCCAIIIIGGCIAACIGGACAICGIGICAICAAACGCIGGAAIIGIAICGIICGG

G D P E A I A K L M A E I V R H F G ' L D I V 8 8 N A G I V 8 F G

H L K D V I P E E P D R V F R V N
ICACCIGAAAGACGIGACCCCAGAAQtatgaaccacagataacgcattcaaggcatatgctaaagaaaacactagGAGIIIGACAGGGTCTTCCGGGTCA
Illllllllllll lllllll l llllllllllllllllllllll llllllllllllllll Illlll Illlllllll I lllllllll lllll
ICACCIGAAAGACAIGACCCCTGGAQtatgaaccacagataacgcactcaaggoatatgctaataaaaacattacGAGIIIGGCCGGGICIICCAGGICA

H L K D M I P G E F G R V P 0 V N

I R G 0 F F V A R E A Y R H M R E G G R I I L I 8 8 N I A C V K G
ACACICGIGGCCAGIICIICGIGGCGCGGGAGGCCIAICGCCAIAIGCGGGAAGGAGGCCGGAIIAICCIGACCAGCICIAACACCGCIIGCGICAAGGG
llIllllllllllllllllllllllllllllllIllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllllll
ACACICGIGGCCAGIICIICGTGGCGCGGGAGGCCIAICGCCAIAIGCGGGAAGGAGGCCGGAIIAICCIGACCAGCICIAACACCGCIIGCGICAAGGG

I R G Q P F V A R E A I R H N R E G G R I I L I 8 8 N I A C V K G

V P K H A. V I 8 G 8 K G A I D I F V R C M A I D C G D K K I I V N
GGIACCCAAACAIGCIGIAIACICCGGIICCAAGGGGGCIAIIGACACCIIIGIICGCIGCAIGGCCAIIGACIGCGGAGACAAGAAAAICACCGIGAAI
llIIIIIIlllllllllllllilllll llllllllllllll lllllllll illlllIIIlllllIlllllllllllllllllllllllll Illll
GGIACCCAAACAIGCIGIAIACICCGGGICCAAGGGGGCIAICGACACCIIICIICGCIGCAIGGCCAIIGACIGCGGAGACAAGAAAAICACIGIGAAC
V P K H A V Y 8 G 8 K G A I D I P L R C M A I D C G D K K I I V N

A V A P G A I K I D N F L A V 8 R E I I P N G E I P I D E 0 V D E
GCGGIGGCICCIGGAGCCAICAAGACIGAIAIGIIIIIGGCIGIGICGCGGGAGIAIAICCCCAAIGGIGAGACIIICACCGAIGAGCAGGIAGACGAGg
II Illlllll llllllll llllllllllllllllllll IlllllllllllllllllllllllllllllllllllIlllllllllllll lllllll
GCCGIGGCICCCGGAGCCAIIAAGACIGAIAIGIIIIIGGCAGIGICGCGGGAGIAIAICCCCAAIGGIGAGACIIICACCGAIGAGCAGGI

A V A P G A I K I D N F L A V 8 R E I I P N G E I F I D E 0 V D E

C A A H L 8 P L N R V G L P
tcagctttccccccataaactgcgtcttgttgggttcccgcttaacgaagtcttacctag!GTGCCGCTIGGCICICTCCTITGAACCGCGIGGGCCTCC
llll III I lllllllllllllllllllllll lllll Ill IlllIIIIlllllllllllllllllllllll lllll I lllllllill
tcagtgttctatctataaactgcgtcttgttcccttcatacttaaggaaatcttatctag?GTGCCGCTIGGCICTCTCCICTGAACAGGGIGGGCCICC

C A A w L 8 P L N R V G L P

V D V A R V V 8 F L A 8 D I A E H V 8 G K I I G V D G G A I R ’
CIGIGGAIGICGCCCGGGIAGIGAGCIICCIGGCAICIGACACAGCCGAAIGGGICCGIGGAAAGAICAIIGGGGIGGAIGGIGGCGCIIIICGAIAAAC
lllllll lllllllIllllllllllllllllllllI ll lllllllllllllllllllllllll lll lllllllllllllllllllll II III
CIGIGGACGICGCCCGGGIAGIGAGCIICCIGGCAICAGAIGCAGCCGAAIGGGICCGIGGAAAGAIIAIICGGGIGGAIGGIGGCGCIIIICCAIGAAC

V D V A R V V 8 F L A 8 D A A E H V 8 G K I I R V D G G A P P '

CIIIACCGCIAIAIACICGIGGGIGAAGIGIAIICICICGIAIIAIAAAGAGCIAGACGICGIAIIIGAIAGGAIIIGCIAGIIAAACIACAACGIAAIA
lllllll lllllll l lllllllllllllll Ill llll llllllllllllll I III llllll llllllllll lllllllll III I
AIIIACCGGIAIAIACACIIGGGIGAAGIGIAIIGICIIAIAIICIAAAGAGCIAGACGACAIAICAGAIAGGGIIIGCIAGIIGAACIACAACIIAACA

IAACGICIACIGCICCCAGGIAGCGGGGAAAAAGACCIIGIAIAIAIGCIIGAAAACCIIICACAIIACACIAAICACGGIAACIICAIAIAICCAAIGC

lllllllIllllllllllllllllllll llllllllllllllllllllllllllllllllll llllllllllllllllllllllllllllllllllll
IAACGICIACIGCICCCAGGIAGCGGGGGAAAAGACCIIGIAIAIAIGCIIGAAAACCIIICAIAIIACACIAAICACGGIAACIICAIAIAICCAAIGC

100

100

197

200

297

300

397

400

497

499

34

597

599
34

68
697

699
68

101
797

799
101

118
897

899
118

151
997

999
151

184
1097

1099
184

217
1197

1199
217

231
1297

1299
231

262
1397

1399
262

1497

1499

1597

1599

43

Figure 9. Schematic representation of the disruption of ver-IA by plasmid pDV-VA (gene
replacement). The black boxes located below the restriction map of a putative recombinant
clone are the predicted DNA fragments which would be observed using probe 1 and probe
2 in Southern hybridization analysis. B, BamHI; H, HindIII; P, PstI; Sa, SacI; S, SphI; Xb,

XbaI; X, X7101.

 

 

u a fix 0.3 3! 1.5 a» a.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

3x «.3 a a so. as x a
r 82.. l Q8: « sea I
1g “ u g
m n. m a 3.: x a m
<32. .832»... <39. 82.2%
.5. ad 1 a
3v. 1.! L L
3! ad. 0 n
<38.
<20 °_E°=°m 1— ./////V/////////7////////////////////////A — —
m I m x n. m
.<>.>o.:
28mm...
£233.55 I? ME
5?. ONE x

Figure 9.

45

total of 250 transformant colonies were generated. One clone, A. parasiticus VAD-102, was
identified which lost the ability to produce aflatoxin but did accumulate a yellow pigment
(presumptive VA) based on an initial screen on CAM. TLC analysis of cell extracts
confirmed that VAD-102 did not produce AFB, or G, but instead accumulated VA (Figure
10).

In an attempt to identify the genetic recombination events which occurred in VAD-
102, Southern blot analysis was conducted by cutting genomic DNA extracted from single
spore isolates of VAD-102 with four different combinations of restriction enzymes : (1)
BamHI, (2) P511, (3) P511 plus SacI, or (4) HindIII plus SphI. The Southern blots were
probed separately with radiolabelled probes 1 or 2 shown in Figure 9. Based on the schematic
(Figure 9) indicating the disruption of ver-IA, when genomic DNA was cut with BamHI and
probed with probe 1, the wild-type 2.9-kb BamHI fiagment would shifi to a 3.6-kb BamHI
fragment. IfPstI was used to cut the genomic DNA and probed with probe 1, the wild-type
4.4-kb PstI fragment would be replaced by a 6.2-kb PstI fragment. In the same way, if the
genomic DNA was digested with PstI plus SacI and probed with probe 2, the wild-type 4.4-
kb PstI fragment would be replaced by a 2.4-kb PstI-SacI fragment. The wild-type 3.3-kb
SphI-HindIII fragment would be replaced by a 2.8-kb fragment when genomic DNA was
digested with SphI and HindIII and probed with probe 2. The Southern blots demonstrated
that the wild type ver-IA gene fragment was replaced by the predicted disrupted DNA
fiagment with all restriction enzymes used in the analysis (Figure 11). These data indicated
that the ver-IA but not ver-IB gene was disrupted by gene replacement in VAD-102.
However, additional DNA fragments not predicted for a simple disruption event also

hybridized to the probes (Figure 11). The stronger hybridization signal (according to the

46

VA >
AFB, ,

l
l

 

Figure 10. TLC analysis of the metabolites of A. parasiticus VAD-102. Aflatoxin B and G
mixture (lane A) and VA purified from A. parasiticus ATCC 36537 (lane V) were used as
standards. The chloroform—extract of mycelia and growth media from A. parasiticus VAD-
102 (lane 2) was compared with that from the AFB, producing strain A. parasiticus NR-

1(lane 1). Solvent system : chlorofomr-acetone (95:5).

.‘8’ .

\III\ ((A1
{III/{‘11

47

 

Figure 11. Southern hybridization analysis to confirm disruption of the ver—I gene.
Genomic DNAS isolated fi'om A. parasiticus NR-l (lane N), A. parasiticus VAD-102 (lane
1), and a transforrnant which still produced aflatoxins (lane 2) were digested with BamHI
(blot a), PstI (blot b), PstI plus SacI (blot c), or HindIII plus SphI (blot d) and analyzed using
standard procedures. Blot a and b were hybridized with probe 1 as indicated in Figure 9.
Blot c and d were hybridized with probe 2. DNA fiagments designated A, the DNA fiagment
containing ver-IA; B, the DNA fragment containing ver-IB. Stars indicate DNA fragments
that are derived from gene replacement disruption of ver-IA. DNA size standards are

HindIII-digested lambda DNA in kilobases.

48

intensity) of these additional fragments suggested that multiple integration of the disruption
vector may have occurred in the genomic DNA. Southern hybridization analysis of these
same DNAs using the niaD gene as a probe showed that the illegitimate recombination did
not occur at the niaD gene locus (data not shown). Because multiple integrations of pDV-
VA occurred, we could not rule out the possibility that loss of AFB, synthesis in VAD-102
was due to an event other than disruption of ver-IA. To solve this problem, an alternative
hypothesis was proposed : if the accumulation of VA in strain VAD-102 is caused by
disruption of the ver-IA gene, aflatoxin production will be restored after the functional ver-IA
gene is put back into the genomic DNA.

In order to transform the ver-IA gene back into strain VAD-102, another selectable
marker, ben‘, was used in firngal transformation to avoid generation of auxotrophic mutations
in the VAD-102 strain. The ben’ gene is a mutated allele of the normal B-tubulin gene and
confers resistance to the fungicide benomyl. The gene has been cloned in our lab from a
benomyl resistant mutant of A. parasiticus (Wu et al., 1996). Strain VAD-102 was
transformed with pVer-Ben (Figure 5B) containing ver-IA and ben’. Nine benomyl resistant
transformants were obtained. TLC analysis of cell extracts of 5 transformants (A. parasiticus
VAD-BV l to 5) demonstrated that they were able to produce aflatoxin although they still
accumulated VA (Figure 12). Based on Southern hybridization analysis of their genomic
DNAs, the five aflatoxin-producing transformants all contained at least one copy of the wild-
type ver-IA gene (2.1-kb EcoRI fiagment; see Figure 13). It was difficult to identify the site
of integration of pVer-Ben in these transformants because the recipient strain already
harbored multiple truncated ver-IA sequences. The different location and copy number of

the wild-type ver-IA gene in the aflatoxin-producing transformants may explain different

49

VA»

AFB, >
AFG, >

 

AFVANV123456789

Figure 12. TLC analysis of the metabolites of transformants obtained by transformation of
A. parasiticus VAD-102 with pVer-Ben. The chloroform-extract from A. parasiticus NR-l
(lane 1‘0 and A. parasiticus VAD-102 (lane V) were resolved on the same plate with those
of the transformants A. parasiticus VAD-BV 1 to 9 (lanes 1 to 9). Aflatoxin B and G
mixture (lane AF) and VA purified fi'om A. parasiticus ATCC 36537 (lane VA) are used as

standards. Solvent system : chloroform-acetone (95:5).

 

Fr
D\!

t.. if" -‘I -\.€‘§" III .( If

50

 

Figure 13. Southern hybridization analysis of A. parasiticus VAD-102 transformed with
pVer-Ben. Genomic DNAs isolated from A. parasiticus NR-l (lane N), A. parasiticus VAD-
102 (lane V), and transformants (lanes 1 to 9) were digested with EcoRI and probed with a
0.8-kb SmaI/EcoRI DNA fragment containing ver-IA and analyzed by standard procedures.
DNA fragments designated A, the DNA fragment containing ver-IA; B, the DNA fragment

containing ver-IB. DNA size standards are HindIII-digested lambda DNA in kilobases.

51

levels of AFB, and VA observed. The remaining four benomyl resistant transformants (VAD-
BV 6 to 9) did not produce detectable AFB, (Figure 12) nor did they contain wild type copies
of ver-IA (Figure 13). These data suggested that disruption of ver-IA in VAD-102
accounted for the accumulation of the pathway intermediate VA and that a wild type ver-IA

gene could complement this genetic block in the disruptant strain.

IV. DISCUSSION

The data confirm that ver-IA is directly involved in the conversion of VA to ST in
AFB, biosynthesis in A. parasiticus SU—l. Of the two copies of ver-I present in this strain,
only ver-IA is functional. A nonsense mutation occurred in the coding region of ver-IB
which likely resulted in synthesis of a truncated polypeptide with no function. The direct
involvement of a homolog of ver-IA in ST biosynthesis in A. nidulans was also confirmed by
disruption of the .9th gene which was cloned based on sequence homology to ver-IA (Keller
et al., 1994). Disruption ofsth led to a block in ST production and accumulation of VA.

In this study, a linearized plasmid was used to disrupt ver-IA by gene replacement.
A study performed by Tatebayashi et al. (1994), which analyzed the DNA fragments
generated by illegitimate recombination in Schizosaccharomyces pombe, demonstrated that
linearized DNA can recircularize and integrate into multiple sites in genomic DNA through
homologous or nonhomologous recombination. These data may help explain the fact that not
only was ver-IA disrupted by gene replacement but also the dismption plasmid recircularized
and integrated (apparently multiple copies based on hybridization signal intensity) at one or
several loci.

Recently, transcript mapping together with gene complementation and inactivation

52

experiments in A. parasiticus showed that the aflatoxin pathway genes pksA,fas—1A, nor-1,
aflR, norA, ver-IA, and omtA are clustered in one linkage group (Trail et al., 1995; Yu et al.,
1995). Based on the results obtained in this study, it appears that part of the gene cluster is
duplicated in A. parasiticus SU-l. At least three previously identified genes, ver-IA, norA,
and cle, are located in this duplicated region. To date the presence of only one copy of any
of these genes has been demonstrated in A. flavus (Skory, 1992). The duplication of a
portion of the gene cluster in A. parasiticus (especially the duplication of one of the pathway
regulators) but not inA. flavus may help explain the observation that nearly all A. parasiticus
strains isolated produce high levels of aflatoxins, whereas many A. flavus isolates (up to 40%
or more) produce no aflatoxins (Bennett and Papa, 1988; Cleveland and Bhatnagar, 1991).

The conversion of VA to ST in Aspergillus is a complex reaction which contains at
least 5 enzymatic steps including deoxygenation, Baeyer-Villiger oxidation, lactone cleavage
and rearrangement, oxidative decarboxylation, and methylation (Bhatnagar et al., 1992;
Dutton, 1988; see Figure 14). The enzymes that catalyze the conversion of VA to ST have
not yet been identified. The lack of purified enzyme activities together with the absence of
identified intermediates between VA and ST make it difficult to elucidate the exact function
of the ver-I gene. Nucleotide sequence analysis of the ver-I gene suggested that it may
encode a ketoreductase (Skory et al., 1992). A comparison of the predicted amino acid
sequence of ver-I with the published polypeptide sequence for the Streptomyces coelicolor
actIII gene (Hallam et al., 1988), which encodes a ketoreductase associated with biosynthesis
of the polyketide actinorhodin, demonstrated a significant level of identity (~3 0%) between
these proteins (Figure 15). Based on this result, the Ver—l protein was proposed to be

responsible for a deoxygenation reaction. Unfortunately, the timing of deoxygenation in

53

CHOW

Wm
(10”),ng flfldation

0” ° 0“ OH o OH
Meow» m...
0

    
  
    

 

0
H20
B.V. oxidation 13:3;23283 311d
01 0 0'1
0 deoxygenation
H20 lactone cleavage and oxidative
rearrangement decarboxylation
C02

 

oxidative
decarboxylation C02

0 O

(DMST)

(ST)

Figure 14. Proposed schemes for the enzymatic conversion of versicolorin A (VA)
to sterigmatocystin (ST). 6—deoxy VA, 6-deoxyversicolorin A; DMST, demethyl-

sterigmatocystin.

54

Figure 15. A comparison of the deduced amino acid sequences of ver-IA, ver-IB,
Streptomyces coelicolor actIII (actIII; Hallam er al., 1988), and Magnaporthe grisea ThnR
(ThnR; Vidal-Cros et al., 1994). The in-frame stop codon within the ver-IB polypeptide is
indicated by an asterisk. Black squares represent amino acid identity. White squares represent

amino acid similarity.

55

Figure 15.

er-IB.
17]]HR
ide is

856111

 

56

conversion of VA to ST is still ambiguous because no putative intermediates have been
isolated from aflatoxin producing strains. 6-deoxyversicolorin A (6-deoxy VA), however, has
been identified to be a metabolite of Aspergillus versicolor which produces ST (Elsworthy
et al., 1970). This prompted us to propose that the Ver-l protein is involved in the
deoxygenation of VA to form 6-deoxy VA. In support of this proposed scheme, a
polyhydroxynaphthalene reductase (T4HN reductase) involved in melanin biosynthesis in
Magnaporthe grisea was recently purified to homogeneity (Vidal-Cros et al., 1994). This
reductase displays 66% identity and 82% similarity with the deduced arrrino acid sequence of
the ver-IA gene product (Figure 15). The dehydroxylation reaction in part catalyzed byM
grisea T4HN reductase is entirely analogous to the proposed deoxygenation of VA (Figure
16). The other analogous reaction is the reduction of emodin to chrysophanol at an early
stage of the biosynthesis of ergochromes, fungal pigments produced by Claviceps purpurea.
This conversion, mediated by NADPH, is believed to consist of two steps, reduction and
dehydration (Ichinose et al., 1993). Based on these observations, we hypothesize that' VA
is processed by two successively operating enzymes, the product of the ver-I gene and a
dehydratase, to form 6-deoxy VA (Figure 16). Before completely understanding the
conversion of VA to ST, however, it is necessary to clone several other genes involved in this
complex reaction. Keller et al. (1995) have isolated a second gene, stcS, involved in the
conversion of VA to ST in A. nidulans. stcS is located within 2-kb ofsth (ver-IA homolog)
in the ST gene cluster. The close spatial relationship between these VA associated genes
could lead to firture studies focused on the isolation and characterization of other genes

involved in the conversion of VA to ST in order to more clearly understand this process.

57

 

(1) MELANIN
OH OH OH OH
REDUCI'ASE DEHYDRATASE

0H,“.— 0,, OHJiIOOHl 3,,”
T4HN C)(‘l') SCYTALONE T3HN
(2) CYNODONTIN
OH O OH OH O
DEOXYGENASE
GOO — —-> WQOO
CH3 OH
O
EMODIN CHRYSOPHANOL
(3)AFLATOXIN
OH O VER-I OH O
OUCF W 000
O
VERSICOLORIN A GDEOXWERSICOLORIN A

Figure 16. Proposed Ver-l protein activity in a two-step dehydroxylation reaction
(reaction 3) derived from analogous reactions in melanin (reaction 1; Vidal-Cros er al.,
1994) and cynodontin (reaction 2; lchinose et al., 1993) biosynthesis. T4HN : 1,3,6,8-

tetrahydroxynathalene; T3HN 21,3,8-trihydroxynaphthalene.

CHAPTER 3

REGULATION OF VER-I GENE EXPRESSION IN
FUNGAL CELLS

I. INTRODUCTION

Understanding the process of aflatoxin gene expression is one essential step toward
elucidation of the molecular mechanisms which regulate aflatoxin biosynthesis. Two useful
indicators for gene expression are the pattern of accumulation of the transcripts (mRN A) and
proteins encoded by aflatoxin genes. Previous studies demonstrated that the appearance of
several aflatoxin metabolic enzymes (Anderson and Green, 1994; Cleveland et al., 1987; Lin
and Anderson, 1992) and the accumulation of nor-1, ver-I, and omtA transcripts (Skory et
al., 1993; Yu et al., 1993) coincide with the cessation of exponential growth of the fiingus
and the onset of aflatoxin production. In the current study, the expression of the ver-I gene
was monitored by detection of the gene product, the Ver-l protein. Since ver-I is directly
involved in aflatoxin biosynthesis, it was hypothesized that the accumulation of the Ver-l
protein should parallel AFB1 accumulation. Two complementary methods, batch
fermentation analysis and nutritional shift assay, were performed to address this hypothesis.
Because an enzyme activity assay of the Ver-l protein was unavailable, a polyclonal antibody
was generated for Western blot detection of the native Ver-l protein in A. parasiticus. To

generate the anti-Ver-l antibody, a ver-IA cDNA was expressed in Escherichia coli using

58

59

the vector pMAL-c2. The maltose-binding protein/Ver-l filSiOIl protein produced by E. coli
was used to generate polyclonal antibodies against the Ver-l protein. After immunoaflinity
purification, an anti-Ver-l antibody was obtained to specifically recognize the Ver-l protein
(~28-kDa) in fimgi by Western blot analysis.

The specific anti-Ver-l antibody was also used to study the regulation of Ver-l
protein accumulation in fiingal colonies grown on solid media. A temporal and spatial pattern
of Ver-l protein accumulation was observed using Western blot analysis of proteins extracted
from different areas of the fungal colony. This observation was confirmed using the ver-
1/GUS reporter strain (Wu, 1995) by monitoring GUS activity (B-glucuronidase). The ver-
1/GUS reporter strains were also applied to study the expression of the ver-I gene in fungal

mycelia using chromogenic and fluorescent substrates.

H. MATERIALS AND METHODS
A. Strains and plasmids

A. parasiticus NRRL 5862(SU-1; ATCC 56775) was used as a wild-type aflatoxin-
producing strain in the investigation of Ver-l protein accumulation. A. parasiticus VAD-102,
in which the ver-IA gene was disrupted, was used as a negative control for Ver-l protein
detection in Western blot analyses. A. parasiticus VAD-BV 1 to 5, five isolates of A.
parasiticus VAD-102 which were transformed with a wild-type ver-IA gene, served as
positive controls for Ver—l protein detection. Several ver-I/GUS reporter strains were used
in the analysis of Ver-l protein accumulation in fungal colonies and mycelia. A. parasiticus
le)6-6 No. l and No. 4, in which the functional ver-I promoter/GUS fusion construct was

integrated at the ver-I locus, were used as positive reporter strains. A. parasiticus pHD4-4

60

No. 3, in which a GUS construct without the ver-I promoter was integrated at the niaD gene
locus, was used as a negative reporter strain (Wu, 1995). The pMAL-c2 vector (New
England Biolabs) was used for expression of the maltose-binding protein (MBP)/Ver-1 fusion
protein in Escherichia coli BL21 (Novagen Inc. Madison, Wisconsin).
B. Generation of polyclonal antibodies against the Ver-l protein

The pMAL protein fusion and purification system (New England Biolabs) was used
for Ver-l protein production. The EcoRI/M101 ver-IA cDNA fragment (kindly supplied by
Dr. J efl' Cary, USDA-ARS, New Orleans, LA) was inserted between the EcoRI and SaII sites
in the polylinker of the pMAL-c2. This vector was then cut with EcoRI, treated with Klenow
enzyme, and religated to generate the correct reading frame fused with malE which encodes
the MBP. Expression of the MBP/Ver-l fusion protein in E. coli was induced by adding
IPTG (isopropyl B-D-thiogalactopyranoside) during active growth. The MBP/Ver-l fusion
protein was purified from the E. coli crude extracts using an amylose-resin column according
to the manufacturer's instructions. The purified MBP/Ver-l filSiOI‘l protein [400 ug in 200
pl phosphate-buffered saline (PBS), pH 7.2] was mixed with 200 pl TiterMax adjuvant
(Sigma) and used to immunize two rabbits to generate polyclonal antibodies. The rabbits
were boosted with the same preparation of antigen at 28 days after the first injection. Sera
were collected at 28, 58, and 65 days and the antibody titer was determined by ELISA using
preimmune sera as controls. Microtiter wells were coated with MBP/Ver-l proteins and then
blocked with 3% BSA in PBS. After washing the plates with PBS containing 0.2% (v/v)
tween 20 (PBS-Tween), the diluted serum was added and incubated for 60 min at 37°C. The
plate was washed 6 times with PBS-Tween and the alkaline phosphatase conjugated goat anti-

rabbit IgG was added to each well. After incubation for 60 min at 37°C, the plate was washed

61

6 times with PB S-Tween and the bound phosphatase activity was determined by adding 50
pl substrate solution consisting of 0.4 mM p-nitrophenyl phosphate, 10 mM diethanolamine,
and 0.5 mM MgCl2 in H20 (pH 9.5). The reaction was stopped by adding 50 pl stopping
solution containing 0.1 M EDTA in H20. Absorbance was read at 405 nm. Titers were
defined as the highest dilution of serum resulting in absorbance greater than the control.

The IgG antibody fraction was purified by precipitation of the antiserum with 33%
ammonium sulfate. After redissolving in PBS, the antibodies were passed through an affinity
column in which the MBP and E. coli proteins were conjugated to a CNBr-activated
Sepharose gel (Sigma). The antibodies which passed through the column were referred to
as anti-Ver—l antibodies.
C. Batch fermentation analysis

Conidia (2x106) of A. parasiticus SU-l were inoculated in 250 ml Erlenmeyer flasks
containing 100 ml YES broth (2% yeast extract, 6% sucrose; pH 5.8). The cultures were
incubated at 30°C (in the dark) in an orbital shaker (150 rpm). Flasks were removed at
appropriate time points after inoculation for the analysis of mycelial dry weight, aflatoxin
concentration, and protein extraction.
D. Nutritional shift assay

The nutritional shift assay was performed as previously described (Skory et al., 1993).
Glucose-mineral salts medium (GMS), which is able to induce aflatoxin production, and
peptone-mineral salts medium (PMS), a non-aflatoxin—supporting medium, were used to grow
mycelia in this assay. Conidia (2x107) of A. parasiticus SU—l were inoculated in 500 ml PMS
and grown for 65 h at 30°C (in the dark) in an orbital shaker (150 rpm). The mycelia were

harvested and equally distributed into fresh GMS or PMS media (30 ml for each sample).

62

The mycelia were collected at different time points (up to 48 h) after nutritional shift for
protein extraction and the aflatoxin concentration in the grth media was measured.
E. Protein extraction and Western blot analysis

Mycelia for protein extraction were ground in liquid nitrogen using a mortar and
pestle, suspended in TSA (2 mM Tris-CV40 mM NaCl/0.025% NaN3; pH 8.0) buffer
containing 1.0 mM phenylmethylsulfonyl fluoride (PMSF), and centrifirged at 20,000 x g for
15 min. The protein concentration in the supernatant was determined with the Bio-Rad
Protein Assay kit (Bio-Rad Laboratories, Hercules, CA) based on the Bradford method
(Bradford, 1976). 50-100 pg of protein from the supernatant were separated by
electrophoresis (12 % SDS-PAGE), transferred to PVDF membrane, and probed with anti-
Ver-l antibodies. ARad-Free chemiluminescent detection kit (Schleicher & Schuell, Keene,
NH) was utilized to detect antibodies bound to proteins.
F. Analysis of mycelial dry weight and aflatoxin concentration

Fungal mycelia were harvested by filtration through Miracloth and the dry weight was
measured after lyophilization or after complete drying at 70°C. The AFB1 concentration in the
filtrate was determined by an enzyme-linked immunosorbent assay (ELISA) described by
Pestka (1988) with AFBl monoclonal antibodies (kindly provided by Dr. J. Pestka, Michigan
State University).
G. Analysis of the accumulation of the Ver-l protein in a fungal colony

Plate cultures were grown by inoculating conidiospores of A. parasiticus SU-l onto
the center of Petri dishes which contained YES. When the colonies had grown up to a point
where well-defined sectors (different morphology in central, middle, and peripheral parts of

the colony) could be observed (3 to 6 days of incubation), they were cut in concentric zones

63

with a sterile scalpel. Mycelia from each concentric zone were used for protein extraction and
the same amount of protein was resolved by gel electrophoresis (12% SDS-PAGE), and the
Ver-l protein was analyzed by Western blot analysis.

In an alternative protocol, the ver-I/GUS reporter strains were grown the same way
on solid YES medium. Instead of cutting out the mycelia, the colony was overlaid with a 1.5
% agar solution containing 0.5 mg/ml X-gluc (5-bromo-4-chloro-3 -indolyl B-D-glucuronide)
and the plates were incubated at room temperature for 4 hr or until the blue color was well
developed.
H. Analysis of the expression of the ward gene in mycelia

A modified method for fungal slide culture (Harris, 1986) was used in this experiment.
The 1.5 % water agar was poured into Petri dishes and allowed to solidify. A 22-mm2 cover
glass was centered on the agar. One block of solid YES agar (5 to 8 mmz) was placed on the
center of the cover glass. The conidia of ver-I/GUS reporter strains were inoculated on each
side of YES agar followed by placement of the second cover slip on top of the nutrient block.
The slide cultures were incubated at 30°C from 24 to 65 hr (in the dark). After taking apart
the assemblage, the cover glass with mycelia attached to it was immersed in liquid YES
containing 0.5 mg/ml X-gluc and incubated for 4 hr. The preparation was then examined by
a Nikon Labophot microscope for distribution of the blue color. A FITC-conjugated GUS
substrate (CnFDGlcU; Molecular Probes, Inc.) was also used to replace the X-gluc in this

experiment, and the GUS activity was represented by the green fluorescence.

64

III.RESULTS
A. Specificity of anti-Ver—l antibodies

Previous data from nucleotide sequence analysis indicated that the ver-IA gene
encodes a polypeptide composed of 262 amino acids (Skory et al., 1992). The deduced
molecular mass of the Ver-l protein is 28-kDa. The specificity of the anti—Ver-l antibodies
raised against MBPNer-l fusion protein was tested by Western blot analyses. The ability of
this antibody preparation to detect the native Ver-l protein in protein extracts of A.
parasiticus NRRL 5862, a wild-type aflatoxin-producing strain, and A. parasiticus VAD-102,
a non-aflatoxin-producing strain which accumulates versicolorin A (VA) was measured.
Since the ver-IA gene was disrupted in strain VAD-102, it was predicted that the Ver-l
protein would not be present in the protein extract. A 28-kDa protein was detected in the
protein extract of wild-type strain but not in that of strain VAD-102 (Figure 17A).

In the study described in Chapter 2, the ver-IA gene was transformed into strain
VAD-102. Of the 9 transformants analyzed, 5 (transformants VAD-BV 1 to 5) restored ver-
1A gene function and aflatoxin production. Protein extracts of these 9 transformants were
employed in the Western blot analysis using anti-Ver-l antibodies as the probe (Figure 17B).
The data demonstrated that isolates which received the ver-IA gene and were restored in
AFBl production (transformants VAD-BV 1 to 5), regained the production of the 28-kDa
protein. These data confirmed that the anti-Ver-l antibodies specifically recognize the native
Ver-l protein in A. parasiticus. Because of this specificity, the anti-Ver-l antibodies raised

against MBP/Ver-l firsion protein were used to analyze Ver—l protein accumulation.

65

Figure 17. Western blot analysis to measure the specificity of anti-Ver-l antibodies. Proteins
were extracted from mycelia afler 60 - 72 hr growth in YES media. 50 - 100 pg of protein
were resolved by gel electrophoresis (12% SDS-PAGE), transferred to PVDF membrane, and
probed with anti-Ver-l antibodies raised against a MBP/Ver-l firsion protein. (A) Western
blot analysis of cell extracts from A. parasiticus NR1 (lane 1) and VAD-102 (lane 2). (B)
Western blot analysis of cell extracts fiom A. parasiticus NR1 (lane N), A. parasiticus VAD-
102 (lane V), and transformants (lane 1 to 9) obtained by transformation of VAD-102 with
a plasmid containing the wild type ver—IA gene. Transformants 1 to 5 (lane 1 to 5) are
aflatoxin producing strains. Transformants 6 to 9 (lane 6 to 9) are non-aflatoxin producing

strains. The molecular mass of marker proteins is indicated on the left of the blots A and B.

Figure 17.

66

 

67

B. Accumulation of the Ver-l protein in liquid cultures of A. parasiticus

Two complementary experiments, batch fermentation analysis and nutritional shift
assay, were performed to determine the relationship between Ver-l protein accumulation and
aflatoxin accumulation in A. parasiticus. In batch fermentation analysis, A. parasiticus NRRL
5862 was grown in YES medium for 5 days (conditions which induced aflatoxin production).
The growth of the mycelia and aflatoxin production were measured at appropriate time
intervals (Figure 18A). AFB, was first detected 36 hr after inoculation and accumulated
rapidly between 36 hr and 60 hr. The rate of AFB, accumulation decreased after 60 hr.
Western blot analysis (Figure 18B) revealed that the Ver-l protein was not detected until 36
hr after inoculation, at approximately the same time as AFB, was detected. The quantity of
the Ver-l protein reached its highest level at 60 hr and decreased thereafter. These data
indicated that the timing and rate of AFB, accumulation corresponded very well to the timing
and rate of accumulation of the Ver-l protein.

A nutritional shift assay was conducted to determine if the relationship between AFB,
accumulation and Ver-l protein accumulation was observed under a different set of
conditions which induce AFB, synthesis. AFB, and the Ver-l protein were not detected
during initial growth of the wild-type strain in PMS medium (non-AFB] inducing conditions)
(Figure 19). After transferring the mycelia to fresh PMS and GMS media, AFB, was first
detectable at 12 hr in GMS medium (AFB, inducing conditions) and increased up to 32 hr.
On the contrary, no detectable AFB, accumulation occurred up to 48 hr after the shift to fresh
PMS medium (Figure 19A). Western blot analysis (Figure 198) showed that the Ver-l
protein could be clearly detected at 12 hr and the quantity peaked at approximately 36 hr after

the shift to GMS. The Ver-l protein was not detected in mycelia shifted to fresh PMS.

68

Figure 18. Batch fermentation analysis of Ver—l protein accumulation in A. parasiticus SU-
1. (A) Mycelial growth (dry weight) and aflatoxin production (measured by ELISA) after
inoculation of conidiospores in YES media. (B) Western blot analysis of proteins extracted
from mycelia from 24 to 120 h after inoculation in YES media using similar methods as in

Figure 17. The molecular mass of marker proteins is indicated on the left of the blot.

69

€569: .833 cozmhcoocoo 5568.2
0
w 1 8 6 4 2 0

n u d - -

 

- 140

m m o O 0 0
4|
1

Time (hr)

 

 

 

 

+ Myeollal Dry Weight
35 - -o-Aflatoxln Concentration

 

h t- n - b Li 0
o. o. 5. o. 5. o 5. o.
4 3 2 2 1 1 0 0
a 68 mod 3925 to 6:85
n
w. .
n A

70

Figure 19. Nutritional shift assay for the identification of Ver-l protein accumulation in A.
parasiticus SU-l. (A) Aflatoxin accumulation in GMS and PMS media after nutritional shift
(measured by ELISA). (B) Western blot analysis of proteins extracted from mycelia grown
in GMS (G) and PMS (P) before (P at 0 h) and after nutritional shift. The molecular mass

of marker proteins is indicated on the right of the blot.

71

A.

 

6 5 4 3 2 1

=83... :ozfiucoocoo 5.83.2

 

0

TONI

Time ( hr)

111353

 

72

These data confirm that Ver-l protein accumulation correlates positively with AFB,
accumulation in the wild-type aflatoxin-producing strain.
C. Accumulation of the Ver-l protein in a fungal colony

Three concentric zones (center, middle, and periphery; see Figure 20), which differ
in their morphology, could be clearly observed in colonies grown for 72 to 144 hr on solid
YES medium. The central zone bears abundant pigmented conidia. The middle zone contains
more aerial hyphae with scattered conidia. The peripheral zone may be divided into two
parts, an outer, thin hyphal network border and a region where the hyphal network is piled
up, with the formation of aerial hyphae. The width of these three zones increased constantly
as grth proceeded. Colonies from 72 to 144 hr growth were cut in concentric zones and
the Ver-l protein was detemiined by Western blot analysis of proteins extracted from each
zone (Figure 21). The data indicated that (1) the central part of the colony contains less Ver-
1 protein than the middle and peripheral parts, and (2) the Ver-l protein concentration in each
concentric zone changed at different growth stages. In other words, there are spatial and
temporal expression patterns of the ver—I gene in a fungal colony.

A ver-I/GUS reporter strain, pHD6-6 transformant 1, showed the same morphology
as the wild-type aflatoxin-producing strain when grown on YES media. GUS expression in
pHD6-6 transformant 1 had previously been shown to be under the control of the ver-I
promoter (Wu, 1995) and thus could be used to monitor ver-I gene expression. By
overlaying a colony of the pHD6-6 transformant 1 with agar containing X-gluc, it was
observed that the blue staining was more intense in the peripheral part and the border between
the peripheral and middle parts of the colony (Figure 22). These data confirmed the

observation in Western blot analysis that the central part of the colony has a lower Ver-l

73

 

Figure 20. The morphology of a fungal colony grown on solid YES media. Conidiospores
of Aspergillus parasiticus SU-l were inoculated onto the center of the medium and grown
for 120 hr at 30°C in the dark. Three concentric zones with different morphologies were

designated C, central zone; M, middle zone; P, peripheral zone.

74

 

Figure 21. Western blot analysis of the proteins extracted from three concentric zones in
fungal colonies. Plate cultures were grown by inoculating conidiospores of A. parasiticus
SU-l onto the center of Petri dishes (solid YES) and incubated for 72 to 144 hr. Central (C),
middle (M), and peripheral (P) parts of the colonies were cut in concentric zones with a sterile
scalpel and used for protein extraction. 50 pg of protein were resolved by gel electrophoresis

(12% SDS-PAGE), transferred to PVDF membrane, and probed with anti-Verl antibodies.

 

75

 

Figure 22. Analysis of ver-I promoter activity in a fungal colony using the ver-I/GUS
reporter strain. Conidiospores of Aspergillus parasiticus pHD6-6 transformant 1 were
inoculated onto the center of the YES medium and incubated for 120 hr at 30°C in the dark.
The colony was overlaid with a 1.5 % agar solution containing 0.5 mg/ml X—gluc (5-bromo-4-
chloro-3-indolyl B-D-glucuronide) and the plates were incubated at room temperature for 4

hr. The blue staining indicates the location of GUS activity.

76

protein concentration.
D. Expression of the ver-I gene in mycelia

Using a slide culture method, the structure of fungi including vegetative mycelia,
conidiophores, and conidiospores could be easily observed by bright field microscopy. For
reference purposes, the structure of a typical conidiophore of Aspergillus is shown in Figure
23. Using pHD6-6 transformant 1 as the reporter strain, it was found that X-gluc staining
was very weak in the 24 hr culture and relatively strong in the 48 and 64 hr cultures. The
blue staining was well distributed in the vegetative mycelia (Figure 24A). In the
conidiophore, the blue staining seemed to concentrate in the vesicle or the area near it (Figure
24D and E). It was also observed that the older hyphae had weaker blue staining than the
younger hyphae. The same phenomenon was also observed when the fluorescent GUS

substrate was used (Figure 24F and G).

IV. DISCUSSION

We have successfully generated polyclonal antibodies to detect the native Ver-l
protein in A. parasiticus. The results suggested that the antibodies generated against proteins
expressed in E. coli possessed a high degree of specificity in detecting native Ver-l proteins
in A. parasiticus. The model we developed here for raising polyclonal antibodies to detect
the aflatoxin proteins is very usefiil especially when the enzymatic activity of a protein
involved in aflatoxin biosynthesis cannot be identified and hence cannot be purified. The
protein encoded by the ver-] gene provides a good example. The enzymatic activity of the
Ver-l protein has not been identified although the ver-I gene has been confirmed to be

directly involved in aflatoxin biosynthesis.

77

-ooacm». ~.

5 g g 8 g C

,, g ‘3 © ,.
i; 9 . . o 2 camera

9 . C 3

9 e

. .. .
PHAUOE

. ‘ A A.’ e

\ ' ‘I
. , “EMA
VE9CLE

 

\

STALK

FOOT CELL
VEGETA‘IWE CELL
J k /I

I ’I’

 

 

 

Figure 23. The structure of a typical conidiophore of Aspergillus.
(Timberlake, 1993)

78

Figure 24. Microscopic images from the analysis of ver-I promoter activity in mycelia using
the ver-I/GUS reporter strain. The fungi were grown on a cover glass for 48 or 65 hr using
the slide culture method. The mycelia were stained by immersing the cover glass in liquid
YES containing 0.5 mg/ml X-gluc (picture A - E) or C,,FDGlcU (picture F - H) and
incubated for 4 hr. The preparation was then examined by a Nikon Labophot microscope for
distribution of the blue color (bright field) or green fluorescence (UV). (A) A. parasiticus
pHD6-6 transformant 1 incubated on YES media for 48hr. x400. (B) A. parasiticus SU-l
incubated on YES media for 48 hr (negative control). x400. (C) A. parasiticus pHD6-6
transformant 1 incubated on YES media for 48 hr (without X-gluc staining). x400. (D) and
(F) A conidiophore (without conidia] head) of A. parasiticus pHD6-6 transformant 1
incubated on YES media for 48 hr. x400. (E) and (G) A conidiophore with conidia] head of
A. parasiticus pHD6-6 transformant 1 incubated on YES media for 65 hr. x400. (H) A.

parasiticus SU-l incubated on YES media for 65 hr (negative control). x100.

tic-n

Figure 24.

A.

 

 

Figure 24. (cont’d).

D.

 

80

 

 

 

 

I“‘{‘
{1’

I ‘fl‘l‘ll‘l ( I.

.(

1"!

(‘I

1“ ( ' I“

I f! (I. I] ll

82

Figure 24. (cont’d)

H.

 

83

It was not possible to investigate Ver-l protein accumulation in A. parasiticus until
specific antibodies against the Ver-l protein were generated in this study. The batch
fermentation analysis demonstrated that the accumulation of the Ver-l protein is consistent
with a pattern predicted for a protein involved in secondary metabolism. These data indicated
that the ver-I gene is expressed near the end of trophophase ( approximately 36 hr after
inoculation in YES medium) when aflatoxins begin to accumulate. The positive correlation
between ver-I gene expression and aflatoxin accumulation was fiirther confirmed in the
nutritional shift assay in which Ver-l protein accumulated to significant levels only in the
medium (GMS) which supported aflatoxin biosynthesis. These data were consistent with our
previous observation that the timing of ver-I RNA transcript accumulation corresponded to
the timing of aflatoxin accumulation (Skory et al., 1993).

To date, studies on the regulation of secondary metabolism are often based on
submerged, agitated, and batch liquid cultures which clearly separate the trophophase and
idiophase of fungal growth (Bu’Lock et al., 1965). This model system, however, could
potentially perturb the normal metabolic and physiological development of fungi and mask
functionally significant information (Campbell et al., 1981). In nature, the aspergilli grow on
solid substrates and grth is far fiom homogeneous. In search of firnctional information on
regulation of secondary metabolism, solid culture is a closer approximation to the natural
growth conditions of fungi. In this study, Ver-l protein accumulation was measured in a
fungal colony grown on solid media. The morphology of the firngal colony we described here
is very similar to that reported by Yanagita and Kogane (1962). In their study, the colonial
structures of Aspergillus niger and Penicillum urticae were divided into four zones. These

zones from outside to center are the extending zone, the productive zone, the fi'uiting zone,

84

and the aged zone. The extending zone is represented by the outermost periphery consisting
of a thin hyphal network extending centrifiigally. The productive zone is a piled-up hyphal
mat with aerial hyphae. In the fruiting zone, fi'uiting bodies (conidiophore, conidia-bearing
structure, and conidia) are being formed. The aged zone, the central part of the colony,
consists of abundant conidia. Yanagita and Kogane also measured the basophilic substances
such as RNA and DNA in each zone, and found that the productive zone contains more
basophilic substances than any other part of the colonial tissue.

Using Western blot (protein accumulation) and GUS activity analysis (ver-I promoter
function) in this study, it was determined that the quantity of Ver-l protein and ver-I
promoter activity are higher in the peripheral zone of the colony (Figure 21 and Figure 22).
The morphology of the peripheral zone with the most intense blue staining (Figure 22) is
identical to the productive zone described by Yanagita and Kogane. This means that the ver-
1 gene expression is more intense in the part of colony containing more basophilic substances.
This spatial pattern of ver-I gene expression is likely based on the physiological differentiation
in a mass of fungal cells deve10ping on a solid medium.

Fungal cell differentiation can be observed by growing the fungi on solid substrates.
When aspergilli grow on solid substrates, there are vegetative mycelia that grow in contact
with the substrate. From vegetative mycelia, the aerial mycelia, a population of developing
conidiophores, develop and eventually bear conidial heads (Bartman et al., 1981). A slide
culture method was applied in this study in an attempt to correlate aflatoxin biosynthesis and
cell differentiation. Using the ver-l/GUS reporter strain, the activity of the ver-I promoter,
which is directly involved in aflatoxin biosynthesis, was monitored indirectly by a GUS

activity assay. The GUS activity (blue staining or green fluorescence) was distributed in both

85

vegetative and conidiophore structures. Since the blue stain has been reported to be insoluble
and precipitates at the site of GUS activity (Jefferson, 1988), the distribution of GUS activity
is not likely the result of migration of the blue dye between cells. These data suggested that
aflatoxin gene expression and therefore aflatoxin biosynthesis can occur in both cell
structures. However, it remains possible that the GUS protein could be transported from cell
to cell.

To look more closely at this question, an alternative approach was utilized. The aerial
mycelia were separated from vegetative mycelia by scraping the top layer of a flingal colony
(pHD6-6 transformant 1) with a metal spatula after freezing the whole colony in liquid
nitrogen. Microscopic examination of a scraped preparation indicated that the bulk of
conidiophores and conidia had been removed. The remaining cell preparation (after scraping),
however, still contained a significant number of conidiophores on them. Despite this lack of
cleanliness, the scrape experiment is still capable of helping determine the distribution of
aflatoxin gene expression. The proteins were extracted from scraped and residual cells and
used for Western blot analysis for the Ver-l protein and the GUS activity assay. The results
indicated that accumulation of Ver-l protein and ver-I promoter activity occurred at
approximately equal levels in scraped and residual cells. These data confirm that the aflatoxin
gene expression occurs throughout the fungal colony. The above data, however, still do not
clearly establish the relationship between aflatoxin biosynthesis and cell differentiation because
the experimental design does not allow us to record the timing of the onset of aflatoxin
biosynthesis and cell development.

The role of secondary metabolism in the producing fungus remains unclear despite

decades of study. One proposal is that secondary metabolism may be involved in cell

86

differentiation (Campbell, 1983). By experimental design, any attempt to relate cell
differentiation to firngal secondary metabolism should show that the initiation of production
of secondary metabolites is accompanied by the development of one or more specific fimgal
cell phenotypes. By detailed microscopic observation of cell growth and the measurement
of the production of specific secondary metabolites, it has been reported that synthesis of
some secondary metabolites is correlated to cell differentiation. Peace et a1. (1981) have
shown that 6-methylsalicylic acid (6-MSA) is produced by P. patulum only after aerial
mycelia have begun to form. It was not clear whether 6-MSA biosynthesis occurs in
vegetative or aerial hyphae. Bartman et al. (1981) demonstrated that P. brevicompaclum
produces mycophenolic acid only after aerial mycelia have begun to form. The biosynthesis
likely takes place in vegetative hyphae and the bulk of the metabolite is excreted into the
grth medium. Bird et al. (1981) reported that brevianamides A and B are produced by
P. brevicompactum after conidiation had begun. These metabolites are located in the upper
part of conidiophores and conidial heads. An ergosterol formed by P. brevicompactum was
also shown to be produced after aerial mycelia began to form and the metabolite was
distributed in both vegetative and aerial hyphae (Bird and Campbell, 1982).

The involvement of aflatoxin biosynthesis in cell differentiation has been hypothesized
based on the observation that aflatoxin biosynthesis is associated with the sclerotia formation
(Trail et al., 1995). To address this hypothesis, future studies will analyze aflatoxin gene
expression together with the cell development during the growth of A. parasiticus on solid

media.

CHAPTER 4

INTRACELLULAR LOCALIZATION OF THE VER-l PROTEIN
IN F UNGAL CELLS

I. INTRODUCTION

While a lot of effort has been focused on genetic studies, little information has been
reported with regard to understanding AFB, biosynthesis at the cellular level. Specifically,
understanding the process of localization of proteins involved in AFB, biosynthesis may
generate sufficient information to allow the design of specific and efficient ways to block
aflatoxin production in fiingi. For those enzymes that have been identified to be involved in
aflatoxin biosynthesis, two activities, averantin monooxygenase (AVN-5'-hydroxyaverantin;
Yabe et al., 1993) and O-methylsterigmatocystin oxidoreductase (OMST~AFB,; Bhatnagar
et al., 1989; Yabe et al., 1989), are localized to the microsomal fraction. Several other
identified activities (Bhatnagar et al., 1989; Matsushima et al., 1994; Yabe and Hamaski,
1993), on the other hand, are found to be cytosolic or loosely bound to membranes. These
data suggest that the biosynthesis of aflatoxins is carried out by a combination of membrane-
associated and cytoplasmic enzyme activities. This could suggest that aflatoxins are
synthesized at specific sites within the fiingal cell because at least some of the enzymes may
localize to closely related sites to allow aflatoxin synthesis to be an efficient process. To test

this hypothesis, in this study we localized the Ver—l protein by the protocols of subcellular

87

88

fractionation and immunolocalization.

The subcellular localization of the Ver-l protein was determined by differential
centrifugation of crude cell extracts. The proteins in each fraction were analyzed by Western
blot analysis for the concentration of the Ver-l protein. Analysis of marker enzymes for
certain cell organelles was also employed to detemiine their distribution within the cell
fractions.

Indirect immunofluorescence (11F) microscopy was performed to localize the Ver-l
protein in hyphae and in individual cells of the fiingus. A slide culture method was used to
grow the fungi for IIF microscopy. The location of the Ver-l protein was observed by the
combination of anti-Verl antibodies and the goat-antirabbit polyclonal antibodies (secondary
antibody) labeled with fluorescein isothiocyanate (FITC) using a microscope equipped with
epifluorescent filters. Laser Scanning Confocal Microscopy (LSM) was also applied to

increase resolution and create a three-dimensional image.

H. MATERIALS AND METHODS
A. Differential centrifugation of fungal cell extracts

The preparation of cell Iysates for subcellular fractionation was performed by two
different methods.
1. Grinding in liquid nitrogen

Conidia (2x10°) of A. parasiticus SU-l were inoculated into 100 ml of YES medium
and grown in the dark for 60 hr at 30°C with shaking (150 rpm). The mycelia were filtered
through Miracloth and frozen in liquid nitrogen. The frozen mycelia were ground using a

mortar and pestle, suspended in TSA (2 mM Tris-CV40 mM NaCl/0.025% NaN3; pH 8.0)

89

buffer containing the proteinase inhibitors phenylmethylsulfonyl fluoride (PMSF; 0.1 mM) and
aprotinin (50 pg/ml). The cell debris was removed by three successive centrifugation steps
at 1,000g for 10 min at 4°C. The resulting supernatant was fractionated into a pellet and a
supernatant by centrifugation at 6,000g for 20 min. The same procedure was repeated by
centrifugation of the subsequent supernatant at 20,000g for 30 min; 50,000g for 30 min;
100,000g for 60 min; and 140,000g for 60 min, all at 4°C. Each pellet was washed three
times and resuspended in TSA buffer. The protein concentration of the cell fractions was
determined with the Bio-Rad Protein Assay kit. 50-100 pg of protein from each pellet
fraction and the final supernatant were separated by electrophoresis ( 12 % SDS-PAGE),
transferred to PVDF membrane, and probed with anti-Verl antibodies.
2. Disruption of protoplasts by a homogenizer

The growth of A. parasiticus SU-l and the harvest of the mycelia was performed as
described above. The mycelia were digested with 1% Novozyme 234 in digestion buffer (0.6
M KCl, 0.1 M citric acid, pH 6.0) for 4 hr at 30°C. The protoplasts were collected by
passage through a 29 pm nylon mesh, and washed three times in 0.1 M MES buffer [2(N-
morpholino)ethane sulphonic acid, pH 7 .5] containing 0.6 M KCl. The resulting protoplasts
were resuspended in 0.1 M MES buffer containing 1.0 mM PMSF, 50 pg/ml aprotinin, 5 mM
DTT, and 0.6 M KCl. The cells were disrupted by a Potter homogenizer (Auther H. Thomas
Co.), and the resulting cell lysates were then treated as described above for differential
centrifiigation and Western blot analysis for the Ver-l protein.
B. Calculation of the quantity of the Ver-l protein in cell fractions

Western blot analysis of proteins of each cell fraction from differential centrifirgation

was used to determine the possible association of the Ver-l protein with organelles or the

90

cytoplasm. The relative but not exact quantity of the Ver-l protein could be determined by
this method. Two sets (duplicate) of samples were loaded onto the same SDS-PAGE gel.
After electrophoresis and transfer of the protein to PVDF membrane, one set of samples was
stained with Coomassie blue to visualize the total protein. The other set of samples was used
for Western blot analysis with anti-Verl antibodies. The Coomassie blue stained membrane
and the Western blot were scanned to record the images using a scanner (Epson ES-1000C),
and the intensity of the protein “bands” was analyzed by SigmaGel softwafe. The proportion
of Ver-l protein in each cell fraction was calculated by the following three steps.
(1) Specificguantiu (SQ) of the Ver—l protein (in each cell fraction)
= intensity of the Ver-l protein (on Western blots) / intensity of the total protein
loaded for analysis
(2) Total quantity (TQ) of the Ver-l protein (in each cell fraction)
= (SQ) x total protein in each cell fraction
(3) Bropgnign of the Yer-l protein in each cell fraction
= (TQ) / the sum of (TQ) of all cell fractions
C. Analysis of marker enzyme activities
Four marker enzymes were analyzed to determine the distribution of different
organelles in cell fiactions including succinate dehydrogenase, catalase, acid phosphatase, and
glucose-6-phosphatase.
1. Succinate dehydrogenase (marker for mitochondria)
A spectrophotometric measurement based on reduction of ferricyanide was used
according to Singer et al. (1965) with minor modifications. 0.6 ml Tris buffer (0.3 M, pH

7.6) was combined with 0.1 ml of 0.1 M potassium ferricyanide and 0.4 ml of 0.01 M

91

succinate. The final volume was adjusted to 3 ml by adding water. The optical density of this
mixture was measured at 450 run before adding the enzyme solution. 1~5 pl (20 to 30 pg
protein) of cell fraction preparation was then added to the reaction solution. The reaction
was incubated 30 to 60 min until the optical density no longer decreased and the final reading
was recorded. A control reaction was prepared by including 0.05 ml of 1.0 M malonate to
completely inhibit the succinate dehydrogenase. 2 pmoles of ferricyanide are reduced for
each pmole of succinate oxidized. The decrease in succinate was calculated using 2.06 as the
molar extinction coefficient of ferricyanide. 1 unit of succinate dehydrogenase was defined
as the amount of enzyme causing a decrease of 1 pmole succinate in 1 min.

2. Catalase (marker for microbodies)

The catalase activity was determined as described by Cohen et al. (1970). The
reaction mixture contained excess of H202 (1 ml of 2 % H202 solution) and 1~5 pl (20 to 30
pg protein) of cell fraction preparation. After incubation of the enzyme reaction for 10 to 30
min, the residual H202 was measured by reacting it with KMnQ and then measuring the
residual KMnO4 spectrophotometrically at 480 nm. One unit of catalase was defined as the
amount of enzyme required to liberate half the peroxide oxygen from a hydrogen peroxide
solution of any concentration in 100 sec (Luck, 1965).

3. Acid phosphatase (marker for lysosomes)

The activity of acid phosphatase was measured using an assay kit (Sigma) based on
the method described by Andersch and Szczypinski (1947). The assay and determination of
units of enzyme activity were performed according to manufacturer’s instnrctions. The Sigma
procedure depends on the hydrolysis of p-nitrophenyl phosphate by the acid phosphatase,

yielding p-nitrophenol and inorganic phosphate. Under alkaline conditions, p-nitrophenol is

92

converted to a yellow complex readily measured at 400-420 nm.
4. Glucose-6-phosphatase (marker for endoplasmic reticulum)

Glucose-6-phosphatase activity was determined by measuring the quantity of
inorganic phosphate formed after the reaction (Harper, 1965). The reaction mixture consisted
of 1~5 pl (20 to 30 pg protein) of cell fraction preparation, 0.1 ml citrate buffer (0.1 M, pH
6.5), and 0.1 ml of 0. 1M glucose-6-phosphate. A control was prepared by addition of the
same amount of sample to 0.2 M citrate buffer without the substrate. The mixture was
incubated at 37°C for 30 min, and 1 ml trichloroacetic acid (10 %) was added to stop the
enzyme reaction. After centrifiigation, 1 ml supernatant was mixed with 5 ml molybdate
solution (2x10‘3 M) and 1 ml reducing agent (4x10‘2 M 1-amino-2-naphthol-4-sulphonic acid).
A standard was prepared by replacing the sample supernatant with 1 ml of 5x10“1 M
phosphate standard solution. The mixture was stored at room temperature for 15 to 30 min
and the optical density was read at 660 nm. One unit of glucose-6-phosphatase was defined
as the quantity of enzyme required to liberate 1 pmole phosphate in 1 min.
D. Indirect immunofluorescence microscopy

A. parasiticus SU-l or VAD-102 were grown on cover slips as described in Chapter
3. The slide cultures were incubated in the dark at 30°C for 24 to 65 hr. The fiingal cells
were fixed for 4 hr at room temperature or 16 hr at 4°C in Histochoice Tissue Fixative
(Ameresco; Solon, OH). For partial digestion of the cell wall, the fimgal cells were subjected
to a cell wall digestive enzyme solution containing 1 % Novozyme 234 in water for 1 hr at
30°C. The cell membrane of the hyphae was then permeabilized with 0.1% saponin in Tris-
buffered saline (TBS, pH 7.3) for 30 min at room temperature. The hyphae were then

exposed to anti-Ver-l antibodies (10 pg/ml) in TBS containing 0.1 % saponin, and 1 %

93

bovine serum albumin (BSA) for 1 hr. This incubation was followed by thorough rinses with
TBS (3 times, 10 min each). The slide preparation was then incubated with labeled
secondary antibodies (100 fold dilution of FITC-conjugated anti-rabbit IgG; Sigma) for 1 hr.
For negative control experiments, the hyphae were labeled only with the secondary antibodies.
The hyphae were rinsed with TBS to remove the unbound secondary antibodies. After the
final wash with water, the preparation was air dried, mounted on slides (using Sigma
immunofluorescent mounting medium), and examined under a Nikon Labophot fluorescent

microscope or a laser scanning confocal microscope (Zeiss 10) for FITC fluorescence.

IH.RESULTS
A. Subcellular localization of the Ver—l protein in A. parasiticus

Western blot analysis (Figure 25) showed that the Ver—l protein was detected in all
pellets as well as supernatant fractions regardless of which method was used to prepare the
cell lysates. However, the distribution of Ver—l protein in cell fractions was different between
these two methods (Table 1). When mycelia were ground in liquid nitrogen, almost all the
Ver—l protein (96.9 %) was in the final supernatant fi'action. If the cell lysate was prepared
by disruption of protoplasts, the proportion of Ver—l protein in the supernatant fraction was
decreased to 80.95 %, and a significant increase of Ver-l protein was found in the 20,000g
pellet (Table 1). The analysis of marker enzyme activity (Table 2) indicated that a very high
percentage of organelle enzymes (e.g., 98 % of catalase activity and 98.24 % of acid
phosphatase activity) were distributed to the final supernatant fraction when the cell lysate
was prepared by grinding the mycelia in liquid nitrogen. The distribution of one marker

enzyme (acid phosphate) in the final supernatant was decreased from 98.24 % to 69.78 %,

94

Figure 25. Western blot analysis of cell fractions from A. parasiticus SU— l. The cell lysates
were prepared by grinding the mycelia in liquid nitrogen (panel A) or disruption of protoplasts
by a homogenizer (panel B). Cell pellets (P) and the final supernatant fi'action (S) were
obtained by differential centrifiigation of cell lysates at 6,000g (6P), 20,000g (20P), 50,000g
(SOP), 100,000g (100P), and 140,000g (140P). 50 pg of protein from each fraction were
resolved by gel electrophoresis (12% SDS-PAGE), transferred to PVDF membrane, and

probed with anti-Ver-l antibodies.

Figure 25.

 

96

 

 

3.8 3: ohm :3 8.3 7% .5553 5 555 75> 5 85855
8." m 2: :5. :5 3M: .555 75> 5 €555 2.555% 555?
5.? N3 53 mam 8.2 7% 8855 a 5555 5 55855
8? mm 3 SN 8% as 8553 a 555% .555
amuEmMofig \8 32572.53 5.23538an
855 Ed 52 and as 2.0 75V 2558 a 555 75> 5 5:555
8.02 38 5.8 $3 8.8 2 5m 555 75> 5 €55 2.555% 2622
8.3 m3 8N NS 93. 52 75V 5253 a 5555 .5 SE85
8.55 So 53. 8.5 was $5 8.5 5552.5. a 3:85 5555
acme“? 3%.: S 339% ESSA:
ESE—baa 8on 3:09 Bang 8on 3on
3:83 mass: @352 macaw $3.8 5:35

 

eaten—5.559 .utqouotmc .3
7.3m 9.392333% .V Set 3:33: 28.53.... :3 E 389:— 7...o> 2: he ecu—55mm: A 035,—.

97

cosh—.2... 9.95 5.... e: u 2.

3.3.2. 253.... 035082. e: u .1...

 

 

3.3 3... 2... S... 2.... .- 75. 3.2.2. 33:33.2.

3.... 2.... ma... .2 t... : 3. 5.2.2. .253... 332.22... 22..
uuumzuueiea .8 uuaasau 82.3895

2...... 2... S... S... 2... 2... ..\.. 3.3.2. 38.5.2.2. 523......

8.... 8... n3 3.... 2.... o... S. 3.2.3 5.82.235 92.222.555.320

3.: 8.... S... .6... .n... 2... TX. 3.2:. 38:93.2.

8...... 2.. z... a.» .3. 2. 5. £25.. 3:83;. 2.32.22... 22..

2...... ms... or... 2... .2... .n... TX. 3.2... 33:23:...

8...... 2... 2.... 3.. a... on... 3. 3.2.2. 3:22.22... 52......0

-. -- 2.... 3...... 2...... no..." .3. 3.2:. 1.3.3.2.:

...... .2. 2.... 3... on... m... 5. 5.2.2. «2.2.2.3.? 2.53.522. 3.58.5

5&2»... 3:... 5 353..» 3:83.:
2.32:1... 3:2— 3:2— .o=o.— 3:0.— .o=o._ .8 3.2.2 «Emu...—
uxac... 33...... 32...... 32.8 .33.... 33..

 

Stamp—.550 33:985.. .3
72m u:e.....§sem .v. 5...... 355.... 32.2....— =8 a. 85.3.... 31.5.. no 5.35.5.5 .N 035.

98

accompanied by a significant increase of enzyme distribution in the 20,000g and 50,000g
pellets when disruption of protoplasts by a homogenizer was employed to prepare the cell
lysate (Table 2). These data suggested that grinding the mycelia in liquid nitrogen
significantly damaged cell organelles and resulted in a higher level of distribution of marker
enzymes in the final supernatant fraction. Disruption of protoplasts using a homogenizer was
a better way to preserve intact cell organelles. Using this method the distribution of marker
enzymes and Ver-l protein increased in cell pellets fractions indicating that the Ver-l protein
may locate in an “organelle-like” structure. This “structure” was distributed mainly in the
20,000g (or less then 20,000g) pellet fiaction.
B. Intracellular localization of the Ver-l protein in A. parasiticus
Immunofluorescence microscopy using the anti-Ver—l antibodies demonstrated that
the Ver-l protein was tightly associated with particle-like fluorescent signals that were
distributed to the hyphae and conidiophores of an aflatoxin-producing strain (Figure 26A and
B). In contrast, a control strain, VAD-102 in which no Ver-l protein was synthesized,
showed no or very little fluorescent signal (Figure 26C). The negative control, when only
secondary antibodies were utilized on aflatoxin-producing strain (SU-l), showed no
fluorescent signal. Under low magnification (x100), it was observed that the overall
fluorescent signal was stronger in conidiophores then in vegetative hyphae (Figure 26D and
E). The fluorescent particles were then observed at higher magnification in 3 dimensions by
confocal microscope (Figure 26F), and the particles were determined to be inside the cell wall

(Figure 26G).

99

Figure 26. Localization of the Ver-l protein in fungal cells by immunofluorescence
microscopy and laser confocal microscopy. A. parasiticus SU—l and VAD-102 were grown
by the slide culture method for 65 hr on YES media. The cells were fixed, partially digested
by Novozyme 234, and incubated with anti-Verl antibodies. The bound antibodies were then
probed with the FITC-conjugated secondary antibodies. Preparations were examined in a
Nikon Labophot microscope equipped with epifluorescence filters for FITC fluorescence
(picture A ~ E) or a Zeiss 10 laser confocal microscope (picture F and G). The fluorescent
particles were observed in the hyphae and conidiophores of A. parasiticus SU-l (picture A,
B; x400) but very little in that of VAD-102 (picture C; x400). A stronger fluorescent
intensity was shown in conidiophores of A. parasitius SU-l under lower magnification
(picture D; x100). Picture E was a brightfield image of picture D. The fluorescent signals
were also observed by the laser confocal microscope (picture F; x4000). A fluorescent

particle was shown by confocal microscope to be inside the cell (picture G, top half).

100

Figure 26.

A.

edrz

SEEK?

 

.3.-

11
.u ‘

. . .,..
i)» 5“

 

 

101

Figure 26. (cont’d)

 

102

Figure 26. (cont’d)

 

1 03

IV. DISCUSSION

The location of one aflatoxin biosynthetic pathway protein, the Ver-l protein, in the
fimgal cell was investigated in this study. Subcellular localization was initially conducted by
differential centrifugation. Although Western blot analysis of cell fractions generated by
grinding cells under liquid nitrogen showed that the Ver-l protein was located in all pellet
fractions, a high percentage of the Ver-l protein was contained in the final supernatant
fi'action. It was thus not clear whether the Ver-l protein is located in the pellets or if it is just
carried-over contamination from the supernatant. The fact that a high percentage of marker
enzyme activities were also distributed to the supernatant fraction indicated that cell
organelles were damaged using the grinding procedure. Tolbert (1974) has reported that
most organelles are damaged by the grinding procedure, which therefore must be as brief as
possible. When the cell lysates were prepared by disruption of protoplasts by a homogenizer,
it was found that the distribution of marker enzymes and the Ver—l protein in the supernatant
was significantly decreased suggesting that this procedure is superior to the grinding method
for preparing samples for subcellular fractionation. The increase in the proportion of Ver—l
protein in pellet fractions, especially the 20,000g pellet, suggested that the Ver-l protein was
located in an organelle or membrane-bound structure and it was not due to contamination
from the proteins in the supernatant. The detection of the particle structures in the fungal cell
by immunofluorescent microscopy supported the observation resulting from the subcellular
fractionation. These Ver—l protein containing structures seemed to be more densely clustered
in the conidiophore and conidial head indicating a potential association of aflatoxin
biosynthesis with conidiophore development.

Differential centrifugation separates organelles on the basis of mass and shape. The

104

data here suggest that a relatively large organelle (in the pellet of 20,000g or less) may
contain the Ver-l protein inside, in the membrane, or tightly associated with the outside
surface. In general, the distribution of organelles from low to high gravitational forces are
nuclei (500 - 2,000g), mitochondria and microbodies (5,000 - 15,000g), lysosomes (20,000 -
50,000g), microsomes (90,000 - 120,000g), and ribosomes (140,000g) (Griffin, 1981).
However, the shape of the organelles may vary fi'om species to species (Markham, 1994).
Even within a single species, the developmental stage and nature of the carbon and energy
source can affect the number and the shape of organelles. Organelles such as microbodies and
lysosomes bounded by a single bilayer membrane are more vulnerable to the breakage
procedure of subcellular fractionation. The involvement of cell organelles in penicillin
biosynthesis was analyzed (Muller et al., 1991 and 1992) using immunological detection
methods in combination with subcellular fractionation and electron microscopy. The results
suggested that the three penicillin biosynthetic enzymes are located in different subcellular
compartments. The first step enzyme, 6-(L-or-aminoadipyl)-cysteinyl-D-valine synthetase
(ACVS) was found to be associated with membranes or small organelles. The next enzyme,
isopenicillin N-synthetase (IPNS), appeared to be a cytosolic enzyme. The enzyme involved
in the final step of penicillin biosynthesis, acyltransferase (AT), was located in microbodies.
This model implicates several transport steps for B-lactam intermediates and products.
However, it is still not clear why a specific enzymatic reaction is confined in an organelle.
The speculation is that the organelle may provide a specialized environment for the reaction,
for example, by providing a condition suitable for the existence of the substrate or the
reaction product, by the absence of competing enzymes, or by providing a different pH. It

is also possible that the organelle can protect cells from the toxicity of the products.

105

In eukaryotic cells, the sorting of proteins to their subcellular locations is often
mediated through N-terminal signal sequences that lack conservation and are cleaved upon
import (Blobel, 1980). This mode of protein segregation has been demonstrated for protein
import into the nucleus (Kalderon et al., 1984), mitochondria (Hase et al., 1984), chloroplasts
(Van den Broek et al., 1985), lysosomes (Johnson et al., 1987), and the endoplasmic
reticulum (secretion as well as retention of proteins; Walter and Lingappa, 1986). In contrast,
conserved tripeptide targeting signals (serine, alanine, or cysteine at the first position; lysine,
histidine, or arginine at the second position; leucine at the third position) at the extreme C-
terminus are necessary for directing proteins into microbodies (Gould et al., 1989). The
targeting signal, an alanine-arginine-leucine sequence, was found to be at the C-terminus of
the acyltransferase involved in the penicillin biosynthesis confirming its location in
microbodies (Muller et al., 1992).

By analysis of the cellular distribution of enzymatic activity, it has been proposed that
some enzymes involved in aflatoxin biosynthesis are membrane-bound while others appear
to be cytosolic. In this study, we presented the first physical observation that an aflatoxin
pathway enzyme, the Ver-l protein, might be located in a cell organelle. The Ver-l protein
may not be located in microbodies because the protein lacks a consensus C-terminal signal
sequence. However, there may be a unique system for import of proteins involved in
aflatoxin biosynthesis. It is unclear if there is a N-terminal signal sequence in the Ver-l
protein. This question cannot be solved until N-terminal sequence analysis is performed on
the Ver-l protein purified from the fiingus. To confirm and examine the possible structure
that is associated with the Ver-l protein, a fiJture study using immuno-electron microscopy

will be helpful. Also, using antibodies against other pathway enzymes for immuno-fluorescent

106

or immuno-electron microscopy will be useful in analyzing whether several membrane-

associated enzymes are confined in the same organelle.

CONCLUSIONS

In this dissertation, we have accomplished the specific aims proposed. (1) We have
established the structural and fiJnctional characteristics of two ver-I genes (ver-IA and ver-
IB) involved in aflatoxin biosynthesis. It was detemiined that the previously cloned ver-I
gene is ver-IA by restriction fragment length polymorphism (RFLP) analysis. A duplicated
chromosomal region (approximately 12-kb) was identified upstream fi'om ver-IA and ver-IB
by Southern hybridization analysis. The nucleotide sequence of ver-IB was determined, and
the predicted amino acid sequence of ver-IB shared 95% identity with ver-IA. A
translational stop codon, found in the ver-IB gene coding region, indicated that it encodes
a truncated polypeptide. To confirm the fimction of the ver-I genes in AFB, synthesis, a
plasmid (pDV-VA) was designed to disrupt ver-IA and/or ver-IB by transformation of the
AFB, producer A. parasiticus NR-l. One disruptant, VAD-102, was obtained which
accumulated the pathway intermediate VA. Southern hybridization analysis of VAD-102
revealed that ver-IA but not ver-IB was disrupted. A functional ver-IA gene was
transformed back into strain VAD-102. Transformants which received ver-IA produced
AFB, confirming that ver-IA is the only fianctional ver-I gene. (2) An anti—Ver-l antibody
which specifically recognized the fungal Ver-l protein was generated against the maltose-
binding protein/Ver-l fiision protein obtained by expression of the maIE/ver-IA cDNA

construct in E. coli. Batch fermentation and nutritional shift analyses indicated that the timing

107

108

of Ver-l protein accumulation is positively correlated to aflatoxin accumulation. The
expression of the ver-I gene in a fiingal colony grown on solid media was shown to exhibit
temporal and spatial patterns of regulation. Using the ver-I/GUS reporter strain, it was
demonstrated that fiJngal vegetative hyphae, conidiophores, and conidial heads all actively
transcribed the ver-I gene. (3) Localization of the Ver-l protein using subcellular
fractionation and immunofluorescence microscopy indicated that the Ver-l protein may be
located in a membrane-bound structure. Two lines of data suggest that AFB, biosynthesis
may occur at higher levels during early stages of conidiophore development. First, the
stronger or darker GUS staining (blue or fluorescent staining) was located to the younger
conidiophore, especially in the vesicle structure, when ver-I/GUS reporter strains were used
to examine ver—I promoter function. Second, a higher density of fluorescent signals in the
conidiOphore was observed when anti-Ver-l antibodies were used for immuno-detection of
the Ver-l protein.

This research represents one important approach toward achieving the long term goal
of identifying compounds which interfere with the expression of genes involved in AFB,
biosynthesis. Afier ver-I gene function was confirmed to be directly involved in AFB,
biosynthesis, the ver-I/GUS reporter strain can thus be utilized to monitor the effect of
various natural or synthetic compounds on the expression of the ver-I gene both in the
laboratory and the field. Using the ver-I gene and its product, the Ver-l protein, as a model
system, a preliminary understanding of aflatoxin biosynthesis at the cellular level has been
achieved. Further exploration in this field may lead to an understanding of the possible
function of aflatoxin biosynthesis in the fungus. The information presented here may

eventually help in solving the worldwide aflatoxin problem.

APPENDIX

APPENDIX

PHYSICAL AND TRANSCRIPTIONAL MAP OF AN AFLATOXIN GENE
CLUSTER IN ASPERGILL US PARASIT IC US

I. INTRODUCTION

A cooperative study in our lab was performed to analyze an aflatoxin gene cluster in
A. parasiticus NRRL 5862. During the genetic characterization of nor-1 and ver-I genes,
the cosmid NorA was identified to contain both genes. This was the first finding that the
aflatoxin genes may be physically linked together. A restriction enzyme map of the 35-kb
genomic DNA fragment insert in cosmid NorA was generated. Based on these initial studies,
a hypothesis was proposed : some other aflatoxin associated genes may also be located within
this chromosomal region. To address this hypothesis, Northern hybridization analyses were
performed on RNA isolated from a toxigenic strain of A. parasiticus using consecutive DNA
fragments from the 35-kb genomic DNA fragment of cosmid NorA as the probes. The
resulting transcriptional map of the genomic DNA insert in cosmid NorA revealed that 14
different RNA transcripts were localized to this region. Eight of these transcripts were
proposed to be derived from aflatoxin genes because the timing of their expression (transcript
accumulation) was similar to that of nor-1 and ver-I genes. Two of these eight hypothesized
aflatoxin genes, kaA (previously named gene-1; Trail et al., 1995) and fas-IA (previously
named uvm8; Mahanti et al., 1996) have been recently shown to be directly involved in

aflatoxin biosynthesis by gene disruption analysis.

109

110

II. PUBLISHED PAPER

Amen AND ENVIRONMENTAL MICROBIOLOGY. July 1995. p. 2665-2673 Vol. 61. No. 7
0099-2240/95/504.00+0

Copyright 0 1995. American Society for Microbiology

Physical and Transcriptional Map of an Aflatoxin Gene Cluster
in Aspergillus parasiticus and Functional Disruption of a
Gene Involved Early in the Aflatoxin Pathway

F. TRAIL. N. MAHANTIJ’ M. RARICK. R. MEHIGHA: S.-H. LIANG. R. ZHOU. AND J. E. LiNZ‘

Department of Food Science and Human Nutrition. Michigan State University.
East lansing. Michigan 48824

Received 6 January l995/Accepted 20 April I995

Two genes involved in aliatoxin B, (AFB!) biosynthesis in Mars, non-l and tier-l, were
localiaedtoaSS-kbregiononmeA.pamsirienschmmosomeandtothegenomk DNAfragmentearriedona
single cosmid. NorA. A physical and transcriptional map of the 354:!) genomic DNA insert in cosmid NorA was
prepared to help determine whether other genes located in the nor-14ers! region were involved in allatoxin
synthesis. Northern (RNA) analysis performed on RNA isolated from A. pamr'tr'cus SUI grown in allatoxin-
inducing medium localized I4 RNA transcripts encoded by this region. Eight of these transcripts. previously
unidentified, showed a pattern of accumulation similar to that of nor-l and rer-l. suggesting possible involve-
ment in AFBI synthesis. To directly test this hypothesis. gene-l. encoding one of the eight transcripts, was
disrupted in A. parasirieus CSIO. which accumulates the atlatoxin precursor versicolorin A. by insertion of
plasmid pAPNVES4. Thin-layer chromatography revealed that gene-I disruptant clones no longer accumulated
versicolorin A. Southern hybridization analysis of these clones indicated that gene-l had been disrupted by
insertion of the disruption vector. These data confirmed that gene-I is directly involved in AFB] synthesis. The
predicted amino acid sequence of two regions of gene-l showed a high degree of identity and similarity with the
B—ketoacyl-synthase and acyltransferase functional domains of polyketide synthases, consistent with a pro-

posedroleforgene-l in polyketidehaclrbonesynthesis.

 

Aflatoxins are potent teratogenic. mutagenic. and carcino-
genic secondary metabolites synthesized by certain strains of
Aspaxr'llus parasirr'cus and A. flams (25). Under the proper
environmental conditions. these ubiquitous fungi can produce
aflatoxin upon infection of many agricultural crops. including
peanuts. corn. cottonseed. and tree nuts (20). Because of the
difficulty in eflectivcly controlling aflatoxin contamination of
food and feed by traditional agricultural practices. recent re-
search efiorts have focused on developing an understanding of
the molecular biology of the aflaloxin biosynthetic pathway.
This knowledge may lead to novel methods for control of this
economically and agriculturally important problem.

Aflatoxins are polyketide-derived secondary metabolites.
The carbon backbone of aflatoxin B| (AFBI) is synthesized
from acetate and malonate in a process analogous to fatty acid
synthesis (9. 24. 53). A generally accepted pathway for the
synthesis of AFBI has been proposed (reviewed in references
9 and 10). The first stable intermediate identified in the path-
way is the decaketidc norsolorinic acid (NA), an anthraqui-
none. which is converted to averufin (AVF) by a multistep
series of reactions involving up to three alternative pathways
(9. 56). AVF is then converted to versiconal hemiacetal acc-
tate. versiconal. versicolorin B. versicolorin A (VA). demcth-
ylsterigmatoqstin, sterigmatocystin (ST). O-mcthylsterigmato-
cystin, and finally. to AFBl. As many as I? different enzyme
activities are proposed to be involved in aflatoxin synthesis (9.
24). Several of these enzymes have been purified to homoge-
neity (I. 8. ll. I6. 28. 34, 4t). 53).

‘Oorresponding author. Phone: (5|?) 353- 9624. Fax: (5|?) 353-
8963
t Present address: Department of Forensics. Michigan State Police

E Lansing. Mich.
it Present address: Sigma Chemical Co.. St. Louis. Mo.

2065

Aflatoximblockcd mutants (4. 32) and purified enzymes
have been used to done several genes involved in the aflatoxin
biosynthetic pathway. including nor-l (IS). encoding an activity
which converts NA to averantin; tier-l (48). encoding an activ-
ity associated with the conversion of VA to ST ; uvm8 (36).
encoding a putative fatty acid synthase involved in polyketide
backbone synthesis; amt-l, encoding a methyltransferase which
converts ST to O-mcthyl-ST (57): and (1le (I3. 45). apparently
involved in the regulation of pathway gene expression. The
recombinational inactivation (gene disruption) of nor-I (S4).
ver-l (33). uvmh‘ (36). and pluA (l4) in A. parasiticus and MA
(29) in A. rridulans (which synthesizes ST) firmly established
the functional role of these genes in the AFBI (or ST) biosyn-
thetic pathway.

Parasexual analyses of eight aflatoxin-bloclted mutants (in-
cluding an NA-accumulating strain) in A. [iavus suggested that
all loci were genetically linked on linkage group VII (44).
Attempts to demonstrate linkage of nor-l and tor-l genes in A.
[ramsiticus by parasexual analyses. however. gave conflicting
results (5. I2; reviewed in reference 7). The molecular genetic
analysis presented in the current study clearly demonstrates
the clustering (linkage) of nor-l. uvrn8. and tier-l within a 35okb
region on one chromosome in A. parasitic-us SUI. In addition.
restriction endonuclease analysis and transcript mapping of
this 35-kb region localized eight other transcripts that are ex-
pressed in a pattern similar to that of nor-l. ver-l. and uvm8.
suggesting that the genes encoding them are also involved in
aflatoxin production. To test this hypothesis. disruption of
gene-l (tentatively named because of its position at the far left
end of the cluster) encoding a 74th transcript within the gene
cluster (37) was accomplished in this study. Genetic and bio-
chemical analyses of disruptant clones and nucleotide se-
quence analysis of extensive regions within gene-l suggest that

Ill

2666 TRAIL ET AL

it encodes a polyketide synthase involved in AFB] biosynthe-
sis.

MATERIALS AND METHODS

Strains and culture conditions. Escherichia coli DHSo F‘ '[F'IendAI MR”
(r.' or") mp6“ rIu-I rrcAI gwA (Nal') reMleocZYA-argf)...” (mfllncZA
M [5)] was used for promgating plasmid DNA. A. parosr‘rktrs NRRL $862 (SUI).
a wild-type aflatorttn-produong strain. was used for preparation of RNA for
transcnpt mapping A. partisans GIG (yer-l wit-I per [48]) derived from A.
mm ATCC 36837 (rev-I wit-I [4]) was used as the host strum for the
d'uruption of gene-I. Strains CS") and ATCC 36537 are unable to convert VA
to ST as a result of a ml mutation. and neither produms dctectdtlc levels of
AFBI in liquid or on solid gmwth media. The following strains d A panorama
were used to analyze sclerotiunt development: AFBI-producirg drain SUI.
AVF-accumulating strain ATCC 245.“ (13). NA-accumulating strain ATCC
246% (32). and VA-accumulating strain ATCC 36537.

Fungal strains were matntatncd as frozen spore stocks (approximately I0”
spores per ml) in 20% glycerol at —nrc. Coconut agar medium (CAM [2]). an
aflatoainvinducing medium. was used for rapid screening of fungal strains for
accumulation of AFBI and VA by visualization of blue and yellow fluoremencc.
respectively. under long-wave UV light. YES broth (2% yeast extract. 20%
sucrose; pll 55). a rich allatotrin—inducing medium. was used to grow mycelia for
DNA and RNA preparations and for thin-layer chromatography (TLC) assays.
Reddy's medium. a chemically defined aflatoain-tnducrng medium (47). also was
used to grow mycelia for RNA preparatioru

Isolation and analysis of RNA and DNA. Fungal cultures for DNA and RNA
preparations were grown in YES (DNA and RNA) or Reddy's medium (RNA)
at 29'thth shaking ( I75 rpm) and harvested at the times indicated in the figure
legends (48 h for DNA). DNA was purified from A. parasitic-ta by a published
modification (50) of a phenol-chloroform protocol developed for mammalian
DNA (3). Total RNA was purified from allatosindnduced cultures of A. porn-
sirmu SUI (IO‘ spores per ml). using a hot-phenol protocol previously described
(39). Restriction cndonucleases utilized in analysts of DNA were purchased from
Boehringer Mannheim Butchcmicals or New England BioLabs and were used
according to the manufacturer's instructions. Northern (RNA) and Southern
hybridization analyses were performed using published procedures (38). with a
modified hybridization bufler and conditions recommended by Stratagenc (Ton.
ing Systems. La lolla. Calif. ”P-radtolabcllcd DNA probur were prepared with
a Random Primed DNA Labelling Kit from Boehringer Mannhetrn Butcher!“-
cals After the final wash. nylon or nitrocellulose membranes were placed on
X-ray film (Kodak-XARS) at -—tlI'C. DNA probes unlined tn transcript mapping
(Northern analyses) and Southern hybridization analyses were generated frotn
DNA restriction fragments derived from amid NorA or crumd strbclttncs as
shown in Fig I The gene probes used ascontrols in transcript maturing amsrsted
of a 4.34m EcoRI-Sari DNA restriction fragment Ntlalcd from amid NorA
containing the pvt; gene from A. promotion (48) and a I.I-kb Arrl-Sarl frag-
ment containing part of the coding region of the gene encodmg B-tubulm in A.
pnrnsun‘us (S51).

Plasmid and can-id canstrndlan .d pnrlhotian. Cratmids NorA. Norll.
Ver2. Ver3. and Ver4 were isolated by in situ colony hybridization (3) of a
cosmid library containing A. pennants SUI genomic DNA cloned into the
cosmod vector p825 (4N). ‘zP-labellcd DNA restriction fragments containtng the
nor-I or trr-I genes (marl. LS-kb ”gill-Clef restriction fragment; oer-I. Illi-Itb
AtnI-Iiuml lI restriction fragment [49]) were used as probes to screen the library.

Plasmid pAI’NVI‘SM was constructed to disrupt gene-l by single-cutsstwcr
insertion till“ the homologous region of the chromosome (see Fig. 4). A 4.3-kb
licuRl-Sml A. [armour-ta genomic DNA restriction fragment subcloned from
cosmid NorA. encompassing the entire WC coding region (part of the original
leZS vector) plus a lb-ltb region within the coding region of gene-l. was
subcltmed into pUCI9 cut with (5(an and Serf.

Plasmid mrntpreparattons were performed by the boiling method (38); large-
scalc plasmrd preparations were performed amordtng to the alkaline lysis pro-
cedure of Mamatts ct al. (38).

Renrktknendoucleaxanalydahnserlptm‘phydcaluage
analysis of cosmid NorA. An 5'le restriction endonuclease digest of mid
NorA was prepared. The resulting fragments were subcloned into pUCl9 or
pBluescript SKIN -) except for a lilo-lib EcoRI fragment containtng the nor!
gene and a 4 64th EcoRI fragment immediately adjacent to the lilo-lib fragment.
The "lb-lib EcoRI fragment was cut into two fragments with Sad. and each
resulting fragment (4.3 and 6.3 kb: clones 3 and 4. respectively) was subcltmed
into pBluei-crrpt SKII( - ) cut with EcoRI and Surf. The 4.6-kb EcoRI fragment
was subcloned tnto pHIuescriptSKfM —) as three fragments: two ”ch EcoRI-
thdlll fragments flanking one IIb—kb Hindlll fragment (see Fig. 2). From these
subclones. a restriction map of the entire cosmid was prepared by mapping each
subclone wtth Itipnl. Apnl. Sinai. Sad. and Xhal (Fig. I). The genes nor-I. vet-I.
and mind were localized onto subclones by Southern hybridization analysis in
conjunction with the restriction endonuclease analysis The [Knitaut of aim was
provided as part of a separate study (58).

Measurement of atlatorin synthesis in A. parasiticns SUI. Measurements of
mycelial dry weights in Reddy's and YES growm media were performed csscn~

Ant. ENVIRON. Micaoatoe

tially as described previously (54). Direct competitive enzyme—linked immu-
nosorbent assay analyses of AFBI production were performed as desalted by
Pestka (46) uith AFB! monoclonal antibodies and AFN-horseradish peroxidase
conjugate (both kindly provided by .I. Pestka. Michigan State University).

Transformation of A. mass and genetic disruption of gene-I. Transfor-
mation of protoplasts of A parasmcus CSIO was conducted using minor modi-
fications of the polyethylene glycol method (43). as previously described (50).
[7va ‘ prototrophs were selected on Czapck Om medium (Difco). pAPNVES43
(see Fig. 4) was used as the disruption vector. pPGJJ. containing thcnvG gene
only (50). served as a control plasmid to measure the rate of successful trans-
formation Transformant clones were transferred to CAM to screen for VA
accumulation. Transformant clones were purified by single spore isolation three
succesive ttrnes

Genetic and biochemical analysis at gene-l m dance. Ehrlcntneyer
flasks (250 ml) containing Ill) ml of YES broth were inoatlated ( Ifln spores) with
gene-l disruptant chines or AT'I‘C 36537 (control; parental strain d transfor-
mation recipient strain CSIO) and incubated without agitation at 30‘C in the
dark. After 72 h of growth. mycelial mats were removed and blotted dry. Onc-
quarter (wet weight) of the mycelial mat was dned overnight at “PC to deter-
mine dry uetght. The remainder of each mycelial mat and the gromh medium
were extracted separately with acetone and then with chloroform as previously
described (54) TLC analyses of the solvent extracts were performed on activated
high-performance silica TLC plates ( It) by It) cm) in a chamber cquilibrated with
benzene-acetic and (955). Purified VA (gcncrrnasly provided by Decpak Bhat-
nagar) was resolved on each plate as a standard. DNA was purified from myce-
Iium grown separately in YES broth for 48 h as described above and was analyned
by Southern hybridization.

Sclerotiunt productian. Gene-I disruptant clones were tested for the ability to
produce sclerotta. Aflatoxtn-producrng stratn SUI. NA-accumulating strain
ATCC 2469". VA-accumulattng strain ATCC 36537. and AVF-accumulating
strain ATCC 245.” were grown under identical conditions for comparison.
Strains were center inoculated onto pctri plates containing approaimatcly 30 ml
of CAM medium and incubated for l4 days in the dark at 3tf'C. Sclcmtia were
harvested and counted by a wbltshcd modification (48) of a method [neviottsly
described by Cotty (l9).

Nucleotideseqnenceanalysis. Nucleotide sequence analysiswasconductedon
cosmid NorA subclones (clones 3 and 4. two SerfJicuRI restriction fragments
encoding a large patron of gencd) at the Plant Research Laboratory at Mich-
igan State Unnentty and by DNA Technologies Inc. Rochvillc. Md. Automated
nucleotide sequencers (Alli robotic catalyst and 373A DNA sequencer) and
fluorescent lalwlled T3 and T7 oligonucleotide printers were used to generate
and analyze dideoay sequence reactions. Nucleottdc sequence data were ana~
Iyu-d with the WN'INNO GCMIIO Computer (inmp Package. The ktcations of
introns and open reading frames were predicted with the software programs
Framcx Tcstcode. and ('odon Preference and the A. nrdulnns codtm usage hie
descrilxd previously (35. S4), ('ornparisims of predicted amino acid sequences to
EMIlI- and Gt-nllanlt database libraries were conducted with TFastA and Gap
and aligned with I‘rleup from the Wisconsin Genera Computer Group Package.

RESULTS

Restriction endonuclease analysis of cosmid NorA and
physical linkage of nor-l and tier-l. In screening an A. para-
siticus SUI genomic DNA cosmid library. four cosmid clones
hybridized to the wr-l probe (NorA. Ver2. Ver3. and Ver4)
and two clones hybridized to the nor-l probe (NorA and
NorB). Cosmid NorA was of panicular interest because it
hybridized to both the nor-I and vet-I gene probes. A restric-
tion endonuclease map of cosmid NorA was generated to allow
localization of nor-l. ver—l. and «mm? genes (an EcoRI and
Xbal restriction map is shown in Fig. I). Since cosmid NorA
hybridized to both nor-I and ver-l. it is suggested that either
the two genes are physically linked in the genome ofA. para-
siticus or nor-I and wr-l were brought together on cosmid
NorA due to recombination of normally unlinked chromo-
somal fragments. To distinguish between these possibilities.
Southern hybridization analyses were performed on cosmid
NorA and genomic DNA isolated directly from toxigenic A.
pamsr'ticus SUI (Fig. 2A). The nor-l probe hybridized to iden-
tical 224th Xhal DNA restriction fragments in cosmid NorA
and in genomic DNA. The vcr-l probe hybridized to a I9-kb
Xbal fragment in cosmid NorA and a 2l-kb Xbal fragment in
genomic DNA. A 3.2-kb Sad-BamHI subclone from cosmid
NorA which spanned the junction between the two large Xbal
fragments (22 and I9 kb) hybridized to the same 22- and ZI-kb

112

 

 

 

 

 

 

 

 

 

 

 

VOL 6!. I995 AFLATOXIN GENE CLUSTER IN A. PARASITICUS 2667
l 2 s
_ i ,
s
”'m 4.6-in iii-u 1.8-in

E X E E E
? I 4 I I I i

ace-l anal

L's-u.—
0.9-kb 1.5-in
1.0—kl: (gene-l) 6.5-kli (gene-1)
ritoars
r a 9 to
6 l u 51%;"
1:
ts»
1.9-kb 1.4-kb u-ui s x 3m 1.1-kb so-u

i: E i-: i x i II E i: r:
—'—-|=_.‘- 1 ' ' 4 ‘w

afll var-IA Alp

1.1-kb
tut. —L25-Ith— 1.15-in L'l-klr
2.1-kb ofii Les—i ins-tit.
Wins

FIG. I. Restriction endonuclease and transcript map of cosmid NorA. Sizes and locations of transcripts and 15le reflriction fragments are shown. nI’-Iabelled
probes used to locate transcripts (see Fig. 3B) are numbered I through 12. Locations of genes and directions of transcription are indicated when information is available.
Vector sequences are indicated by shaded blocks. The transcribed region of gene»! continues beyond the end of cosmrd NorA. All Xhal (X) and 154le (I5) sites are
shown. The locations of Surf (S) and liamlll (ll) sites are only included to mark the probe used in linkage analysis; other sites are not included.

Xbal fragments in genomic DNA as the nor-l and ver-l gcnc
probes (and to the 22- and I9-kb Xbaf fragments in cosmid
NorA). These results strongly suggest that the 22- and 2l-kb
DNA restriction fragments carrying nor-l and vet-l, respec-
tively. are directly linked in the genome of A. parasiticus SUI.
Since the vervl and the 3.2-kb Sad-BamHI probes lie within a
IZ-kb duplication of the region containing ver-l and aflR in the
genome of SUI (33. 48). additional bands of the predicted size
appeared in the genomic DNA analyzed with these probes
(ver-l probe. 8.9-kb fragment; 3.2okb Sad-BamHI probe. 8.9-
and 6.5-kb fragments; see Fig. 28 for schematic).

Transcript map of cosmid NorA. The appearance of nor-I
and vet-I transcripts occurs simultaneously in A. pamsiricus
SUI under difl'crent growth conditions, suggesting that they
are coordinately regulated in part at the transcriptional level
(49). Since the two genes were found to be linked on the
chromosome. a transcript mapping analysis of this 35-kb region
was initiated to determine the size. location. and pattern of
expression of other genes in the region. Genes with expression
patterns similar to those of nor-l and ver-l would be studied
further because of the potential for direct involvement in
AFB] synthesis. RNA was isolated at distinct time points from

mycelia of aflatoxin-producing A. parrm'ii'cus SUI grown in
YES or Reddy's medium (which induces AFBI synthesis). The
time courses of aflatoxin production and accumulation of my-
celial dry mass were qualitatively similar in the two media (Fig.
3C and D). The maximum rate of fungal growth occurred
between I8 and 36 h after inoculation. whereas the maximum
rate of aflatoxin synthesis occurred between 48 and 72 h. when
growth had slowed considerably in a transition between active
growth and stationary phase. Radiolabelled DNA probes
(numbered I to 12 in Fig. I) were used to analyze RNA
isolated at various times during fungal growth in YES or Red-
dy's medium. Northern analysis identified the size, location.
and pattern of accumulation of 14 transcripts in the region
encompassed by cosmid NorA (Fig. 38. Northern analyses;
and Fig. I. transcript map).

Transcript accumulation in Reddy's and YES media. The
pattern of expression observed for genes known to be involved
in AFBI biosynthesis (nor-l. vet-I. and M8) in Reddy's me-
dium (a chemically defined medium) showed very little tran-
script accumulation at the l8«h point and a high level of tran-
script accumulation betwecn 36 and 84 h (Fig. 3A). Eight
transcripts in the gene cluster (from left to right in Fig. I:

113

2668 TRAILEI'AL

 

 

A
B. ‘
”a i u ' ('qu
Co—idNo'A
_l—.21‘5——J L__W..
l I
cI—avm ......... ._uu_LLI._.I_ .......
I I
"m LJJ-kb‘
I g l l
l I I
'—-——zzn—J L—ztatII——'
I I l I
n . C . . I l

1_—l Lib
L654“ ‘—-l9~t¥-‘ .—
ll'hysial linkage between the nor- I and try-I genes. (A) Southern
hybriIliu-ii aha lysis-If Xhulcul gem-mu: DNA A. (FNIIIKI m SUI (lam hes
2 III 4) III XIII-cut eusmid NorA Ala( fit: I and S)' Thef wing radiolabelled
were used1nw-l (lane. land 2). Iw-l (lat: 4 and 5). and the

I'ILLCJ' nu“ L ‘ e guru”: .
ingthenur- I‘ Indtrv-I genes s(lane .1; sex parcel II). The four dasltalothe left of
la III (top to bottom III) the [Inilllns III 111 . v.4 . 6.033 44- III
IIurdlll rcl'ZrictiInI fragments of lambda DNA used as a sire standard (ll) \'Iul
cosmids- Nor :ln‘fld Ver2. and IheA Ill genomic cluster

I“ l

in panel A is shown The l9— lIh Mull restrhion lrlagrhenl Ill cosmid NurA Is
‘lhtn rcual ar nature of IhI
ANIn-vtalums

are given in the legend to Fig I Unlamlled restriciinrt Q“: are Iu'uRl Ult‘

genc~l. 0.9 kb; gene-2. 0.5, I.25. L65. 0.75. and L7 kh) showed
a pattern of accumulation similar to that of nor-I, wr-l. and
uvmfl. The data from Imr-l. tw-I. uvmli. and gene-l are shown
in Fig 3A to illustrate the transcript accumulation of these
eight genes. In contrast the accumulation of transcriptsh for
a gene associated with primary metabolism m(UTPb
synthesis) was observed to be high at the Ill-h time point.
Transcripts of benA (encoding B~tubulin) showed nearly uni-
form accumulation at all time points, including 18 h as would
be expected of a uselteeping gene. The I. l and 2.2-kb tran-
scripts(afl R)showed transcript accumulation patterns similar
othat of pyrG (data not shown). It Is not known whether the
gene encoding the l. l -ltb transcript is involved in secondary
meta ism; however uflR has been reported to be a posith
regulator of several genes in AFB] synthesis (55) so the ap-
of this transcript before t as: of nor-I and ver‘l Is not
unexpected The 4. 4-kb transcript was present in very small
quantities at all time points and encompassed the same region
of DNA as aflR. The nature of this transcript is not clear.
0 determine whether the type of growth medium influ~
enocd the relative expression of these two diflerent groups of
genes (i.e.. AFBl-relatcd genes versus primary metabolism or

Am. ENVIRON. MICROBIOL

housekeeping genes). the time course of expression in YES a
“ric IcAh"A-FB1 icindu etgn medium, was also analyzed (Fig SB)

The contrast between8 the pattern of expression of nor-I. wr-l
uvm8, and the eight AFBI- related genes and the expression of
pyrGan be is even more striking Transcripts of the AFB]-
related genes first appeared In the 40-h sample during a tran~
sition from active growth to “stationery phase" and decreased
significantly by 72 h. In contrast transcripts of pyrG and benA
were expressed at the highest levels during active growth (first
appearing in the Ill-h sample) and did not decrease until 24“ to
40 h after inoculation. as growth slowed. both YESa
Reddy s media. heappearance of transcripts "of AFB! -related
genes correlated well with the first appearance of AFB! In the
culture.

Recombinational inactivation of gene-l. 11m pattern of ex-
pression of eight transcripts in cosmid NorA (in addition to
nor-l. ver-l, and uvm8) was observed to correlate well with

FBI synthesis. suggesting that the genes encoding them are
involved in AFBI synthesis. To test this hypothesis. gene-l.
encoding a 7-kb transcript (whose function was unknown), was
disrupted by insertion of pAPNVES43 (schematic in Fig. 4)
which contained an internal fragment of the transcribed region

chromosome should result in insertion of the entire pAPN-
V843 vector into gene-l. inactivating its function.

pAPNVES43 was used to transform A. parasilicu: CSIO, a
VA-accumulating strain. In two separate experiments. l(l% of
the pyrG ‘ colonies did not accumulate a yellow pigment (in—
dicativc of loss of VA production) on CAM. No transfon'nants
were obtained when DNA was not present in the transforma-
tion mixture and no transformants lost their ability to produce
the yellow pigment (VA) when pPGIlJ carrying only the pyrG
selectable marker. was used as a control plasm

TIJC analysis of transformants Three transformants that no
longer appear ed to accumulate VA on CA and a known
VA-ziccumulating strain. ATCC 36537, were Mgrown in YES
medium (:Iilatoxin inducing) for further analysis. TLC analyses
of extracts of mycelial mats and the growth medium confirmed
a loss sfo VA production in all three transformed strains (Fig
5) whereas normal levels of VA were observed In the control
strain ATCC. 537 grown under identical conditions No af-
latoxin production was noted tn either the transformants or the
control strain. No new pigments appeared to accumulate in the
putative gcnc- l-disrupted transformants.

Genetic anal. vege-e-I disruptant clones. South-
ern hybridization analysis“ was performed on genomic DNA
isolated from the parental strain. C810. a five putative
gene~l disruptants (strains that no longer accumulated VA)
(Fig. 6). A lll.2~kh EcoRI genomic DNA fragment hybridized
to the pUC l9 probe (0.8-kb Seal-EcoRI). as expected, in four
of five transfortnants (lanes 2. 4. S. and 6) and to a 3—kb
fragment in the fifth transformant (Fig. 6. lane 3; see also
schematic in Fig. 4). The occurrence of the 3—kb DNA frag-
ment is likely due to genetic rearrangement during or after
integration of the disruption vector. The l0.2-kb fragment was
absent in the parental strain (lane I), as expected. An addi-
tional 8- kb fragment was present in two transformants (lanes 2
and 6) indicating that the disruption vector integrated at one
other site. Identical DNAsa samples were also hybn ridized to a
gene- l probe (0.6-kit Smal-Sacl fragment shown in Fig. 4)
located adjacent to the l.6-ltb gene-l fragment carried on
pAPNVES43. The expected Ill-kl) DNA fragment hybridized
to this gene-l probe in four of the trartsfonned strains (Fig. (1.
lanes 8. It). I l. and l2), indicating insertion of pAPNVES43 by
a single crossover at the homologous gene-l locus on the

VOL. 6|. 1995

A. Reddy's medium.

 

nor-l

114

AHATOXIN GENE CLUSTER IN A. PARASITTCUS

verJA
to 36 a so u as

1.1 “.O... 7.0 4‘

"16
l
103640600496

 

10364060 0496
:x““~

M ' 1.4
“all!!!

 

 

 

 

 

 

 

 

 

 

 

B.Vl-Suediun.
nor.1 ver.1A
1019244072 1016244072
. \".('~
1.3 '.$ 1.1 C 7.0
”'0 mun
l i
10 16 24 40 72 10 16 24 40 72
1.4 u - O Q an .yf
C. 0.
mm mm
3” an auto
-..° W"
250-1 »au
1‘ m‘
»ano
i” i * 2
a i g'“-‘ P‘. t
S m. 8 °
5 a i 5
‘ g 0 as F” g
m.
“-1 10~ 0
too
° 0 a: :4 1 . fl 1
° o t: at aa a an n 04

 

MM

FIG. 3. Accumulation oi transcripts oi genes in the AFBI cluster during lutch limitation/1. Mu NRRL W was inundated into Reddy's medium and
YES medium (time mm) and grown with shakingat arc. Sampleswre removed at the indicated timesloreatractinn oltotal RNA‘an-lysisdmycelial dryweight
and allatoxin. Northern analyses at the RNA extracted (rum samples groom in Rcddy‘s (A) and YES (8) media were done with the W min [or hyhridiuliun shim
in Fig l and descriied in Materials and Methods Hybridization tom-G and B-tuhulin genes (eontmls) is also shim. I'mductinn d M and mycelial dry weights
are shown in panels C (Reddy‘s medium) and 0 (YES medium). Vertical liars indiate standard errors oi the mean.

chromosome. The presence of a larger fragment in lane 9 and
the absence of a lO-ltb fragment support the genetic rearrange-
ment argument proposed for the same DNA sample probed
with pUCl9 (lane 2). The probe hybridized to the expected
112-lib EcoRI genomic DNA fragment in CSIO. DNA samples
were also hybridized to a pyrG gene probe. which confirmed
that the disruption vector was inserted by a single-crossover
event into gene-l (data not shown). The complete hybridiza-
tion analysis was repeated with genomic DNA cut with EcoRV
and Sacl (data not shown). These data confirmed the results
observed for the EcoRI digests.

Sclerotlum production. Two gene-l disruptant clones (Tfl
and TR) as well as strains SUI (aflatoxin accumulating).

ATCC 24690 (NA accumulating; small quantities of AFB! are
also produwd), ATCC 24551 (AVF accumulating). and ATCC
36537 (VA accumulating) were inoculated onto CAM and
grown for l4 days at 30°C. These strains could be divided
into three distinct groups on the basis of levels of sclerotia
produced (Table l). Gene-l disruptants (which do not ac-
cumulate AFBl or identifiable pathway intermediates) pro-
duced about three to six times the quantity of sclerotia pro.
duced by the wild type. SUI. The NA (an early pathway
intemediateyaccumulating strain produced quantities of scle-
rotia similar to those produwd by SUI. while the two strains
that accumulated the pathway intermediates VA and AVF
(intermediates near the middle of the pathway) and produced

H5

2670 TRAIL ET AL

 

 

 

t.
Sad
[all S-aI [call
M DNA '
_ Int-I
[all
Isol'
lul'
E4011
W n: I. ._:M I H hf."
I M WA. F _...

FIG. 4. Strategy for disruption of gene Apr-aim. ion
maps of thed eruption Inveet it pAP’NVES41I (top) the gene- I region est-0‘ the

 

eruption event (bottom) are (min Also indiatcld: are the pBlmwiInSKIk—m )
vector (with the Amp' gene). the region of gene
vector (Mad rs.) and flanking regions (open hail ).‘Die position of the Sad
milfragrnentusedlthotIther nana Iysisisindiatedhyamlidlinchclowlhi:
periotntc DNA

no detectable AFBI failed to produce sclerotia or produced
vry ew sc erot

DISCUSSION

Herc. evidence is presented that several genes involved in
AFBI biosynthesis (nor-I. ver-l. afiR. «Mill, and gene-l) are
physically linked on cosmid NorA and In the chromoso f A
pamsI'lI'cu: SUI. The 0011- I gene has also been linked to this
cluster of genes in A parui'imui and A flaw: (57). Nucleotide
sequence analysis of this entire gene cluster in A. [mmsiticux
and a structurally similar (but not identical) gene cluster in A.
nI’dIIIaIIi is progressing. DNA sequence analysis of the entire
region will allow identification of open reading frames. which
may provide clues about the possible function of the seven
other AFBI- related gcnc cs

It is not surprising that the aflatoxin genes wou uld be ar-

ranged in a cluster in the genome of AsptrgiIIILr organisms."

been found to be clustered. For example. genes
biosynthesis of penicillin (2| 42). lrichothecenes (26). and

 

 

Ant. ENVIRON. Micaootm.

1234‘07I0fi1112

—211
- .
— ~-- ".. —u
C 9 >1 ~' _
'4. 4‘
—M
FIG. 6. Southern hybridintion analysiaofgenomic DNA inflated front the

disrupted traiuforrnanta. DNA was cut with EcoRI and separated on a 01%
agarmegel.l.anesland7A.pIIt-sumllo517. lanaZtoPTAandlltull.

 

transform “unmixed with pAPNVFjIJ. A radiolabelledD Afrlgrnent of
POUCI - :1.“ l mt. smut .
‘)was toprrbe aniden

min Iatlc‘s7tol2. TheDNAaiu
markers indiated on the rightul are froma I Hiiidlll dips! of hacteriuphap
Iarnhda. Filrnwaseapusedforldaysal-

melanin (30) were all recently found to be clustered. What
advantage gene clustering affords the producing organism is
not clear. but one can imagine a selective advantage to having
genes of like function clustered together on a chic oniosomc f
clustering Is I1related to regulation of gene citpresion.

ational inactivation of gene- I provides the first
indication that this gene is directly involved In afiatoxin bio-
synt thesis and sets a precedent that other genes in the cluster.
which are expressed in a pattern similar to those of nor- -l and
yer-I (AFBI- re-lated genes) are aha) prime candidates to be

gene-l remains undescribe
research approaches provide clues about its function. Nucle-
otide sequence comparisons between the proposed amino acid
sequences in two distinct regions of gene- I and proteins in the
EMBL and GenBank database libraries were made. using
computer-assisted analyses (Wisconsin Genetics Computer
Group; TFastA and MOTIFS). High degrees of similarity
(XIV/6) and identity (64%) were observed between a IOU-ami-
no—acid domain in the gene-l protein sequence and the fl-kc~
toacyl- acyl carrier protcin- synthase (Fig. 7A) functional do—
main of the A. nidulam wA gene (4|). a polyketide synthase
gene (PKS) involved in conidial pigment mtbesis. The two
other proteins that showed high identity in the same region

spectivcly (Fig. 7A). A significant level of identity (20 to 32%)
was sobscrvcd In the acyltransferase functional domains of the

TABLE I. Sclerotium production In various strains of
icurgrown on CAM f I4days

 

 

g - Intermediate No. of sclerotia
. rain' _
mutating "I". t Plate 1
Tfl NU 5.020 1qu
m ”D 6.06 5.713
24690 NA 1.242 I.m2
24551 AVF 0 0
36537 VA 36 5
no 5 TLCanaIysisof, , . -—-— I “ ‘ ‘ ' J SUI AFBI 983 I.7Iti
ATCC 1607 It. - L I . . . ‘

 

.5 7.x
(lanes II to I9) are shown VA (Ianc I) was used as a standard fir plates were
ndphotographed under long wave UV light

 

 

' Strain 245.“ is derived from ATCC ISSI7 (24): all others are SUI deriva-

"ND. not detected.

A Beta-Ketoacyl ACP Synthase

116

zoo ooa
F.- DU:
“(2‘

«Ii-o arms m

 

as.
In» a

= . a 2 a a a

2 {tn g In 0 c r: t:
0< * at 3 * o¢ o< o<
33E“ 5: ‘ 3 E? E! E3
'._3 “a 3 a F; Pg ,2
I-cvfl 3': a L o o .
gas :it P :2“ =2“ :2“
05.135 .5 ‘ .. 03‘ 03‘ 03‘
=22: '2 E a “3%" 9%“ 9%“
dca- d‘ 0 ¢ 9.: 4 a n.:
((uivi dd «6 . <<u «in: do:

he GenBank and EMBL databases. Amino acids identical

. . k . h . , .
in at least two It! the presented sequences are shown as white letters on black backgrounds. When two dilerent pairs at identical amino acids occur at the same residue. the pairs which do not match the A. parasrrirm (A.p.)

sequence are shown In ilalia. Arrows in panels A and 8 Indicate acme-site cysteine: and serines. respectively. See text (or aria ysIs ut‘ comparisons.

lal‘nl

”mmn”

I

1

“up.
Ir

I

i

 

 

 

 

FIG. 7.

2672 TRAIL ET AL.

wA PKS (32%) and rat fatty acid synthase (PAS) (20% [51])
enzymes and a distinct region in the gene-l sequence (Fig. 7B).
The identity of gene-l in these two functional domains was
higher with PKS than with PAS. suggesting that gene-l en-
codes a PKS involved in AFBI synthesis. However. these data
do not rule out the possibility that gene-l encodes an FAS.

Limited nucleotide sequence analysis of umi8 (37) identified
a I80-amino-acid region with a high degree of sequence iden-
tity (48%) to a region of undefined function in FASI genes
encoding the beta subunit of FAS in the yeasts Sacdiammyres
cem'r'siae and Yamwia lipolyrica (3|). Townsend ct al. (53)
proposed that six-carbon hexanoate (two keto groups com-
pletely reduced to hydrocarbon) may serve as the starter unit
for AFBI synthesis and that hexanoate is extended by a PKS
without further ketoreduction to form a decaketide. NA. This
scheme would include at least one multifunctional enzyme. the
FAS. with the necessary activities to reduce keto groups to
hydrocarbons in order to synthesize the hexanoate starter.
Another set of activities. the PKS. without keto reduction ca-
pability (27). would then extend hexanoate to generate the
decaketide. NA. Our limited data are consistent with this
scheme. m'm8. which has a high degree of identity to yeast
FAS. could fill the hypothetical FAS role to produce hexano-
ate. which is extended by the product of gene-l (pks/I). a
putative PKS. In support of this theory. Chang et al. (l4) have
independently disrupted and sequenced more extensive re-
gions of gene-l. which they have called ples/t. Several func-
tional domains associated with polymerization of acetate (l3~
ketoacyl-synthase. acyltransferase. and acyl carrier protein)
were identified. but the analysis strikingly failed to find evi-
dence for a keto-reductasc. dehydratase, or enoyl reductase
involved in reduction of kcto (carbonyl) groups to hydrocar-
hon.

A second approach. gene disruption. clearly demonstrated
that gene-l activity occurs prior to the l’t'r-I gene in the path-
way. Further evidence for gene-l function is provided by the
studies on sclerotium production in gene-l-disruptcd transfor-
mants. The absence of production of sclerotia by strains that
accumulate pathway intermediates between NA and VA sug—
gests that gene-l is involved at a step prior to nor-I activity. In
a separate study it was shown that strains disrupted at iii-m8.
like strains disrupted in gene-l. produce sclerotia at levels
higher than those produced by SUI (36). Since it was demon-
strated that uwml’ activity occurs before nor-I in the pathway.
this would argue that gene-I. likc m'mli. is involved in some
stage of polyketide backbone synthesis.

In previous research. an association between aflatoxin bio-
synthesis and sclerotium development has been observed (6.
l8. I9). Using a molecular genetia approach. Skory et al. (48)
observed that complementation of the l‘t'r-l mutation in strain
A. parasiticus CSIII. which accumulates VA and normally does
not produce sclerotia on potato dextrose agar. restored wild-
type levels of sclerotium production. The data presented here
for the AVF- and VA-accumulating strains suggest that accu-
mulation of pathway intermediates inhibits sclerotium devel-
opment. Strains which accumulate early pathway intermediates
(NA) or no pathway intermediates (gene-l and uvm8 dis-
ruptants) generate wild~type levels of sclerotia (or more). sug-
gesting that elimination of accumulation of intermediates in
the middle of the pathway allows sclerotial development to
occur. When no intermediates accumulate. sclerotial develop.
ment is apparently enhanced. Together. these observations
support the hypothesis that the biosynthetic pathway for alla-
toxin production strongly affects the development of sclerotia.
Since secondary metabolism has long been considered a form
of metabolic differentiation (7). it is not surprising that it may

117

Am. ENVIRON. MICROBIOL

also be linked to morphological dilerentiation. Just how this
link is structured remains to be elucidated.

ACKNOWLEDGMENTS

This work was supported by a cooperative research agreement with
the USDA and grant CA52tItl3-tllAl from the National Cancer Insti-
tute.

REFERENCES

I. Anderson. J. A. and L I). Green. I‘ll-l. Timing of appearance of versiconal
hemiacetal acetate esterase and irrational cyclase actnrny in cultures of
Aging-Illa; manna. Myalpatholugia Ikltfl—In.
Aneenlentne.S.N.LM.DeSiIvI.S. nndC.II.S.R.
Bundnnutha. WM. Cintnut as a medium for the emerimental production of
aflatmtin. Appl. Microbiol. “dill—94.

. Ausubel.F.M..IL BmLKEWnD-D.Mnnre.1.c.§eldinan.J.A.
Smith. and K. Struhl. I987, Current pmtociils in molecular biology. vol. I.
John Wiley A Sons Inc. New York.

. Bennett. J. W. and I. A. W. I973. The isolation of mutants of
AM”!!! Ilium and Ami-milk“ "mom with altered atlatotin produeing
ability. Sabouraudia ":235—2“.

. Bennett. J. W. L S. Lee. S. M. Sbnu. and (3. II. Bendmna. Will. Identi-
fication of averantin a an afIatoxin ll, int-cursor: placement in the biosyn-
thetic pathway. Appl- Environ. Microbiol. 39205409.

n. Iennett. J. W- P. M. Imng. S. Kruger. and I). Reyes. With. Sclerotial and

low aflatiuigenic mvflmlogicil variants from haploid and diploid Aynvyillm

fur-minor. Experientia dzawl-tfi l.

Bennett. J. W- and K. E. I’npn. I‘lltl. The afiatmigenic Ayn-rpm“ spp. Adv.

Plant Pathol. 6265—279.

It. Ilhutnagar. I)- and T. E. Cleveland. WWI. Purification and eh.iraeteriI.-iiion
of a redudase from alum-dim marinara SRRC 2M} invode in allatimn
biosynthesis. FASEB J. 412727.

9. lib-Intent. 0.. K. C. Ehdieh. nnd T. E. Garland. I992. ()xidatiim-redue-
lion reactants in bimynthesis of secondary metal-vines. p. l‘S—2Xh. In I).
Bhatnagar. Ii. 8. IJIIchoi. and I). K Arora (ed). Hamlin-A oI applied
mycologv. vol. 5. Marcel Dekker. Inc- New York.

ltt. Bhatnagar. I)- (3. Payne. 1. [-1. Lin. and T. E. Cleveland. INS. I‘otential
contributions of molecular genetics to atlatoxin elimination. Inform 6:2ti2—
272.

II. Bhatnagar. I)- A. II. J. UIIaIi. and T. E. (Jen-land. IW. Purification and
charactervatnm of a methyltransferase from Avirrxrllui purusmun SRRC
IM involved in allatoxin biosynthetic pathway. Prep. Iliochem. “LIN-349.

I2. Bradshaw. II. F... J. W. Bennett. and J. I". My. 1W0. Parasexual analysis
of AVln'tfltn (ruminant. J. (kn. Micriihiiil. I2912I I7—2Il‘.

I}. Chang. I’.-K.. J. W. (my. I). lib-Inna". T. E. (levelnnd. J. W. Bennett. J. E.
Uni. C. P. Wnlnshnls. and (3. A. Pnyne. I903. Cloning ol the Aquarium
[manna-m 4pm] gene ass-iterated with the regulation of allatoxin biosynthe-
sis Appl Environ. Mierobml. 59:3273—3279.

I4. Chang. PAL. J. W. Cary. J. Yn. I). Mr. and T. E. Cleveland. lh'
’mvllm lurmilii‘m pied. a homolog of Aqnmllm nululam n'A. is required
for aflatoiin II. biosynthesis, Submitted [or publication

I5 Chang. I’.-K.. C. I). Story. and J. E. Una. I992. (‘loning of: gene associated
with aflatosin Ill biosynthesis in Aspartllm pamunrm. ('urr. (ienet. “:2.“-
231.

In. Chum A. A. M. I". Dutton, nod II. K. Retry. WWI. "Ihe preparation of
an enzyme awieiated with alIatoxin hmsynthuis by alinity chromatography.
“"le Iliophys Re\ (‘ommuri “£23342.

I7. Cleveland. T. IL. and I). Bhatnagar. I‘Nlt. [iv-dents: for tie novu synthesis of
an allatmin pathway metliyltransleraw near the cesmtion of active growth
and the onset of allatoxin biosynthesis in Amery/Ian {uranium mycelia. ('an
J. Microbiol Mel—S.

I8. Cotty. P. J. I988. Allatoxin and sclerotial production by Ayn-rpm“ [lumiz
influence oI pll I’hytopathology 78:I25tl—|25‘l.

I9. Cotty. I’. J. I989. Virulence and cultural characteristics of two Amvllm
[Iai'ur strains pathogenic on cotton. Phytopatholog ”JUN-8N.

2". Council Inr Agricultnrnl Science and Techndnn (CAST). WtN. Mycotoiins:
cconomrc and health risks. Report ”6. Coum‘il for Agricultural Science and
Technology. Ames. Iowa.

2l. Diez.Il..S.(intierrea.J. Llnrredn. l'.vnnSnIingen.I.II.M.vanderVnr-rt.
and J. F. Martin. IWI, The cluster of penicillin bimynthetic genes. Identi-
fication and characteri/atiim of the pchAII gene encoding the alpha-ammo-
adtpyl-qsteinylwaline synthetase and linkage to the pch and pcnI)Ii genes.
J. Biol. Chem. MIMStl—lhfllfi.

22. Ibnadio. 5.. M. J. Stave-r. J. I. McAIpine. S. J. Swanson. and I. Kata. IWI
Modular organiration of genes required [or ample: polyketide biosynthesis.
Science 152:n75-n79.

23. Ihnlterslout. J. A. It. I. Mateleie and S. S. Yang. W72. Isolation of averulin
In im a mutant of Aslm-illm pumrmrm impaired in aflatoxin biosynthesis
Iliochem. Iliophys Res. (‘ommun 47:III5l-IIISS.

24. Iluttun. M. I". IN. Iinlymcs and allatoxin bnuynthesis. Microbiol, Rev.

,4

'd

b

‘Jt

.‘I

VOL 6|. I995

{2

3|.

32.

35.

37.

40.

4|.

42.

52:274-295.
W L l9”. Allatoxins and human health. C RC Press. Boea Raton.

Fla.

“that. T. M. S. P. McCormick. and A. E. Desjardins. I993. Evidence for a
gene cluster involving triclkithecene-pathway biosynthetic genes in Fuwrium
Miranda. Curr. Genet. 24:29l-295.

liq-nod. D. A. and D. II. Sherman. l99tl. Molecular genetic ol polyketides
and its mmparison to fatty and bnnynlheso. Annu. Rev. Genet. 14:39-66.
Keller. N. P. II. C. Dinning". Jr- D. Bhatnagar. T. E. Cleveland. and
A. II. J. Ulah. l992. Purification of a 40-kilodalton methyltranslenne active
in the allatoain biosynthetic pathway. Appl Enuron. Microbiol 59:479-484.

. w. N. P. N. J. Kanu. and 1'. ll. Ada-u. I994. Aiprrgdlui nodule": “TA

'3 required for production ol the mycotoxin sterigmatmystin. Appl. Environ.
Microbiol. 6.:I444-l45ll.

“hora. N. and 1'. Trace. I993. Gene cluster imohed in melanin biosyn-
thesis of the filamentous lungus Alremunu allmmlu. J. Bacteriol. I7$:4427-
4435.

Km II. C. “In". IL-F. Iledi. M. Schweieer. and E. Schwinn-r. I99I.
Tlic pentalunctional FASI genes ol Sacrhunmnrn mama: and Yam-mud
Iii-untied arc air-linear and comiderahly longer than previoust estimated.
Mol. Gen Genet. mutt-3M

lae. L S. J. W. Iennett. L A. (inldhlatt. and II. E landln. I970. Nor.
aolorinic acid lront a ntutant strain ol Aiptmllut puruuu‘r'tu. I. Am. Oil
Chem. Soc. 43:93—94.

. an. 5.4!. and J. E. “Ill. I994. Structural and functional characterization

ol the ter-l gem and proteim [mm Aymrrllm mmunrur amounted with
the amnntn til versicolorin A to sterngmatoqum tn aflatoun hnnynthenis.
In Proceedings. Current muex in lood utetv Toueology Center. Michigan
State Untvemty. ()ct. I994. Michigan State Unnemti. lust Liming.

Un. loll. and J. A. Anderson. I992. l’unlieation and properties ol WM)-
nal cyclase lrom Awllm [uruum‘ur Arch llnl‘hem. Biophy‘. 293267-70.
Unyd. A. T. and l'. M. Sharp. l99l. (‘odon usage in Amara/Int nidulam.
Mol. (ien. Genet. 230988-294.

Mahanti. N. l). Bhatnagar. and J. E. Una. I994. Structure and function of
«Will. a gene involved in [lilyltetide hackhone synthesix of the aflatoxtn
hanynthclrc pathway til AWN!“ (tamarind. American Phytoluttluilogical
Society Meeting, Albuquerque. NM.

Mnhantl. N. I". Trail. ll. Mehigll. T. S. Wu. ll. Zhou. ll. 5. Uang. M. Enrich.
and J. Una. I993. SIrueture/lundion analyxis of an allatmin gene cluster
imlated lrom Ayn-mam [armour-arr. p. 3|. In Proceedings. Alhltlxm elimi~
nation worhsliop. Little Rock. Art. Food Salety and Health. ARS. USDA.
Beltwille. Md.

Mania“; 1". E. II'. Frltnch. and J. San-haunt. I932 Molecular cloning: a
lahoratory manual. Cold Spring Ilzirhor Inhiratory. (‘ old Spring llarhor.
NY.

. Maraaatsn. M. I973. Preparation ol RNA lrom animal mllx Methods (‘ell

Biol 7:13-5l.

Mata-shin. K-I. Y. Andn. 1'. Hanna“. and K. Value. I994. Purification
and characterilation of tum Wtsiumal hemiacetal acetate reductasex tn-
volved tn allatoain biosynthesis. Appl. limrron. Microbiol. wlfitil—lfiti‘l.
Mama. M. E. and W. E. Tintherlalie. I992 The developmentally regu-
lated AWN!" ru'dulum iv/l gene encodes a polypeptide humming-nus to
[lilylnetide and latty acrd syntheses. Mol (ien. Genet. LIS:2ll5-2l2.
Montenegro. Ii. ll. Herve. I“. J. Fernanda. S. Gutierrez. and J. F. Martin.
I992. Rexolution ol cltrmntmimes Ill and VI ol «4‘I""R'”"‘ mdulum Ivy
pulsed-field gel electroplmresis shows that the penicalltn htmynthelic path-

118

A

43.

45.

47.

49.

St .

5 .

52.

53.

55.

55.1.
so.

5

\I

FLATOXIN GENE CLUSTER IN A. PARASITICUS 2673

way genes pelt/48. pchC. and prnDE Ie clustered on chromosome VI (3.0
megahases). J. Bacteriol. ”ems—um.
0nldey.B.R.J.Ellinehan.B.LMitchell.C.E.Oakley.C.aniona.
G. L Gray. and G. S. May. I987. Cloning. mapping. and molecular anal-
ysis ot the pyrG (orotidine-S’phosphale decarboxylasc) gene at mull!
Miriam. Gene “385-399.

Pena. K. E. “It“. Genetics olAsperp‘hnflmru: linkage ol allatmun mutants.
Can. J. Microhitil. W73.
PaynC.A.G.J.Nyurnn.D.w.T.ECIeveland.andC.l'.
Woloahnlt. I993. Cloning ol the all-2 pne involved in allatmin himyntheso
from Amflm flaws. Appl. Environ Microbial. StlSh—lnl.

Pestka. J. J. I988. Enhanced aurveillace ol loodhorne mycotoxins by im-
munochemical any. I. Aunt. OI. Anl. Chem. 7|:Itl75-ltlll.

Reddy. 1'. V. L Vlswanathan. and ‘I'. A. Ventitasnhra-anlan. I97l. High
aflatoxin production on a chemically defined medium. Appl. Microbiol. 12:
393-396.

Skory. C. 0.. Pa“. Chang. J. Cary. .d J. E. Una. I992. Itolation and
characterization of a gene from Agni-pillar mmv‘mwi animated with the
convemon ol versicolorin A to sterigmatocystin in allatoxin hunynthesis.
Appl. Environ. Mien-hail. 583527-3537.

Skory. C. D- P.-K. Chang. and J. E. Ihe. I993. Regulated expression ol the
nor-I and iw-l genes associated with aflatoxin himynthes'n‘. Appl. Environ
Microbiol. SthZ-IMli.

Skory. C. I)- J. S. Ilnrng. J. J. Mia. .d J. E. Una. I990. Transformation
of Aqrrrillm [lawsuit-us with a homolr‘ous gene (”1(1) involved in pyrim.
idine hunynthes'tx Appl. Environ. Mkud’tiol. 56:3.3l5-332ll

Smith. S. J. K. Naggert. B. Williams-AH. and C. M. Am. I992. Introneaon
organization ol the gene [or the multilunctional animal fatty acid \ynthase.
Proc, Natl Mad. SCI. USA 89H ltlS—Illli

Swan. 0.6-A. M. “MCVMC.MUMJ. A. Salas. I994.
Characterization ol a Suepkmra niobium" gene encoding a type I
polyketide synthase which has an unusual coding stupenee. Mol. Gen.
Genet. 241358-362.
TMCLSM.M¢G¢IR.S.W.MT.LCMILK.Mand
C. E. Inn-y III. I99l. Examination ol tetrahydro- and dihydrohisluran lor~
mation in aflatorun hnsynthesis: lrorn whole cells to purified enzymex p.
I4l—IS4. In R. J. Petrmltt and S. P. McCormick (ed). Scamdary-metaholite
Maynthesis and metaboliun. I’lenutn he». Inc. New York.

. Trail. I". I'.-K. Chang. J. Cary. and J. E. Una. I994. Structural and lune.

Iional analysts til the nor-I gene involved in the hithynthesia ol allatoxins hy
Ayn-callus paummau. Ami. lint/trill. Microbial. wand-MIG.

Wolosliuli. C. P. K. It. Foutz. J. F. Ire-er. 0. Wm. T. E Cleveland.
and (l. A. Payne. I994 Molecular characteriuitron ole/IR. a regulatory locus
[or allatoxin hnsynthesn‘. W. Environ. Microl'innl. 60:243-204.

Wu. 114$. and J. E Una. Unpublished data.

Yalie. K- Y. Mahayana. Y. Andn. ll. Natajina. and T. IIaniaaald. I993.
StereoeliemNry during allatmin huwynthes'n: oirwer‘ion of nonolortnic
nerd to averul'in. April. lznviron. Micndnol. 59:24ttn—2492.

.YnJ-J.W.er.I).IIhatnaaar.T.E(Jew-land.N.PJfieIlerJnd F.S.Chn.

I993 ('Iontng and characteriutnin ol a cDNA from Ammo [ruminants
encoding an U-methvltramlerase involved in allatoun hunynthesis. April.
Iznviron. Mieriihiol. 593504-35“.

VI. 1. I’.-I\'. Chang. J. W. Cary. M. With. I). Mm. T. E. Cleveland.
(3. A. Payne. and J. E. Una. I995. Comparative mapping ol allatotin pathway
gene clusten in Amery/Ito [mmmuur and AVI'INIUUI flaws. Appl. linvtron.
Mterohool. “2365—23”. A ‘

119

III.DISCUSSION

Gene clustering has been reported to be involved in numerous fiJngal secondary
metabolic pathways and fungal development including the penicillin biosynthetic genes in A.
nidulans and P. chrysogenum (Aharonowitz and Cohen, 1992; Montenegro et al., 1992), the
trichothecene biosynthetic genes in F usarium sporotrichioides (Hohn et al., 1993), and the
spore-specific genes inA. nidulans (Orr and Timberlake, 1982). Besides fungi, clustering of
genes involved in secondary metabolisms has been observed in several species of
Streptomyces producing polyketide-derived antibiotics such as actinorhodin, erythromycin,
tetracenomycin, streptomycin, griseusin, and granaticin (Ohnuki et al., 1985; Martin and
Liras, 1989).

The phenomenon of gene clustering was proposed to be a result of evolution
(Aharonowitz and Cohen, 1992). This process may facilitate gene regulation by
encompassing the positive or negative regulators within the gene cluster (Guilfoile and
Hutchinson, 1992; Sophianopoulou et al., 1993), or by providing a suitable chromatin
structure for coordinate regulatory control (Cavalli and Thoma, 1993; Wolfi‘e, 1994). The
role of gene clustering in aflatoxin biosynthesis has yet to be elucidated. Our preliminary data
(Wu, 1995) using the ver-I/GUS reporter strain demonstrated that an extremely high level
of GUS expression resulted from the insertion of the ver-I/GUS construct at the ver-I locus.
In contrast, a relatively low level of (500 to 1,000 fold less) GUS expression was obtained
if the ver-I/GUS construct was inserted at another chromosomal locus, the niaD gene. This
“local effect” may imply the importance of gene clustering in regulation of aflatoxin

biosynthesis.

LIST OF REFERENCES

LIST OF REFERENCES

Aharonowitz, Y., and G. Cohen. 1992. Penicillin and cephalosporin biosynthetic genes :
structure, organization, regulation, and evolution. Annu. Rev. Microbiol. 46 : 461-495.

Allcroft, R., S. Roger, G. Lewis, B. Nabney, and P. E. Best. 1966. Metabolism of aflatoxin
in sheep : excretion of the "milk-toxin". Nature 209 : 154-155.

Andersch, M. A., and A. J. Szczypinski. 1947. Use of p-nitrophenyl phosphate substrate
in determination of serum acid phosphatase. Am. J. Clin. Pathol. 17 : 571.

Anderson, J. A., and L. D. Green. 1994. Timing of appearance of versiconal hemiacetal
acetate esterase and versiconal cyclase activity in cultures of Aspergillus parasiticus.
Mycopathologia 126 : 169-172.

Ansher, S. S., P. Dolan, E. Bueding. 1986. Biochemical effects of dithiolthiones. Food
Chem Toxicol. 25 : 405-415.

Ausubel, F.M., R. Brent, R.E. Kingston, D.D. Moore, J.G. Seidman, J.A. Smith, and
K. Struhl. 1993. Current protocols in molecular biology, John Wiley & Sons, Inc., New
York.

Bartman, C. D., D. L. Doerfler, B. A. Bird, A. T. Remaley, J. N. Peace, and I. M.
Campbell. 1981. Mycophenolic acid production by Penicillum brevicompactum on solid

media. Appl. Environ. Microbiol. 41 : 729-736.

Bennett, J. W. 1979. Aflatoxins and anthraquinones from diploids of Aspergillus parasiticus.
J. Gen. Microbiol. 113 : 127-136.

Bennett, J. W., and S. B. Christensen. 1983. New perspectives on aflatoxin biosynthesis.
Adv. Appl. Microbiol. 29 : 53-92.

Bennett, J. W., L. S. Lee, S. M. Shoss, and G. H. Boudreaux. 1980. Identification of
averantin as an aflatoxin B1 precursor. Appl. Environ. Microbiol. 39 : 835-839.

Bennett, J. W., and K. E. Papa. 1988. The aflatoxigenic Aspergillus Spp, p. 265-279. In

120

121

G. S. Sidhu (ed), Advances in plant pathology, vol 6. Academic Press, New York.

Bhatnagar, D., and T. E. Cleveland. 1990. Purification and characterization of a reductase
from Aspergillus parasiticus SRRC 2043 involved in aflatoxin biosynthesis. FASEB J. 4 :
A2164.

Bhatnagar, D., T. E. Cleveland, and E. B. Lillehoj. 1989. Enzymes in aflatoxin B1
biosynthesis : strategies for identifying pertinent genes. Mycopathologia 107 : 75-83.

Bhatnagar, D., K. C. Erlich, and T. E. Cleveland. 1992. Oxidation-reduction reactions in
biosynthesis of secondary metabolites, p. 255-286. In D. Bhatnagar, E. B. Lillehoj, and D. K.
Arora (ed), Handbook of applied mycology, vol 5. Marcel Dekker, Inc., New York.

Bhatnagar, D., S. P. McCormick, L. S. Lee, and R. A. Hill. 1987. Identification of 0-
methylsten'gmatocystin as an aflatoxin B1 and G1 precursor in Aspergillus parasiticus. Appl.
Environ. Microbiol. 53 : 1028-1033.

Bhatnagar, D., A. H. J. Ullah, and T. E. Cleveland. 1988.Purification and characterization
of a methyltransferase from Aspergillus parasiticus SRRC 163 involved in aflatoxin
biosynthetic pathway. Prep. Biochem. 18 : 321-349.

Bird, B. A., and I. M. Campbell. 1982. Disposition of mycophenolic acid, brevianamide A,
asperphenamate, and ergosterol in solid cultures of Penicillium brevicompactum. Appl.
Environ. Microbiol. 43 : 345-348.

Bird, B. A., A. T. Remaley, and I. M. Campbell. 1981. Brevianamides A and B are formed
only after conidiation has begun in solid cultures of Penicillum brevicompactum. Appl.
Environ. Microbiol. 42 : 521-525.

Blobel, G. 1980. Intracellular protein topogenesis. Proc. Natl. Acad. Sci. USA. 77 : 1496-
1500.

Blount, W. P. 1961. Turkey "X" disease. J. Brit. Turkey Federation 9 : 55-58.

Bradford, M. M. 1976. A rapid and sensitive method for the quantification of microgram
quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 72 : 248-
254.

Brekke, O. L., R. O. Sinnhuber, G. B. Putnam, D. J. Lee, and A. Ciegler. 1977.
Aflatoxin in corn : ammonia inactivation and bioassay with rainbow trout. Appl. Environ.
Microbiol. 34 : 34-37.

Bressac, B., M. Kew, J. Wands, and M. Ozturk. 1991. Selective G to T mutations of p53
gene in hepatocellular carcinoma from southern Africa. Nature 350 : 429-431.

122

Bueding, E., P. Dolan, and J. P. LeRoy. 1982. The antischistosomal activity of oltipraz.
Res. Commun. Chem. Pathol. Pharmacol. 37 : 293.

Bu’Lock, J. D., D. Hamilton, M .A. Hulme, A. J. Powell, H. M. Smalley, D. Shepherd,
and G. N. Smith. 1965. Metabolic development and secondary biosynthesis in Penicillum
urticae. Can. J. Microbiol. 11 : 765-778.

Bulter, W. H. 1964. Acute toxicity of aflatoxin B1 in rats. Brit. J. Cancer 18 : 756-762.
Busby, W. F., and G. N. Wogan. 1984. Aflatoxins. Chem. Carcinogen 2 : 945-1136.

Campbell, I. M. 1983. Correlation of secondary metabolism and differentiation, p. 55-72.
In J. W. Bennett and A. Ciegler (ed), Secondary metabolism and differentiation in fungi.
Marcel Dekker, Inc., New York.

Campbell, I. M., C. D. Bartman, D. L. Doerfler, B. A. Bird, and A. T. Remaley. 1981.
The role of secondary metabolism in Penicillium brevicompactum. Adv. Biotechnol. 3 : 9-14.

Campbell, T. C., and J. R. Hayes. 1976. The role of aflatoxin metabolism in its toxic lesion.
Toxicol. Appl. Pharmacol. 35 : 199-222.

Cavalli, G., and F. Thoma. 1993. Chromatin transitions during activation and repression of
galactose-regulated genes in yeast. EMBO J. 12 : 4603-4613.

CAST, 1989. Council for Agricultural Science and Technology. Mycotoxins : economic and
health risks. Report 116. Ames, Iowa.

Chang, P.-K., J. W. Cary, D. Bhatnagar, T. E. Cleveland, J. W. Bennett, J. E. Linz, C.
P. Woloshuk, and G. A. Payne. 1993. Cloning of the Aspergillusparasiticus apa-Z gene
associated with the regulation of aflatoxin biosynthesis. Appl. Environ. Microbiol. 59 : 3273-
3279.

Chang, P.-K., J. W. Cary, J. Yu, D. Bhatnagar, and T. E. Thomas. 1995. The
Aspergillus parasiticus polyketide synthase gene pksA, a homolog of Aspergillus nidulans
wA, is required for aflatoxin B1 biosynthesis. Mo. Gen. Genet. 248 : 270-277.

Chang, P.-K., K. C. Ehrlich, J. Yu, D. Bhatnagar, and T. E. Cleveland. 1995. Increased
expression of Aspergillus parasiticus aflR, encoding a sequence-specific DNA binding protein
relieves nitrate inhibition of aflatoxin biosynthesis. Appl. Environ. Microbiol. 61 : 23 72-23 7 7.

Chang, P-K., C. D. Skory, and J. E. Linz. 1992. Cloning of a gene associated with
aflatoxin B1 biosynthesis in Aspergillusparasiticus. Curr. Genet. 21 : 231-233.

Chu, F. S. 1991. Mycotoxins : food contamination, mechanism, carcinogenic potential and

123

preventive measures. Mutat. Res. 259 : 291-306.

Chuturgoon, A. A., M. F. Dutton, and R. K. Berry. 1990. The preparation of an enzyme
associated with aflatoxin biosynthesis by affinity chromatography. Biochem. Biophys. Res.
Commun. 166 : 38-42.

Ciegler, A., E. B. Lillehoj, R. E. Peterson, and H. H. Hall. 1966. Microbial detoxification
of aflatoxin. Appl. Microbiol. 14 : 934-939.

Cleveland, T. E., and D. Bhatnagar. 1991. Molecular regulation of aflatoxin biosynthesis,
p.270-287. In GA. Bray and DH. Ryan (ed.), Mycotoxins, cancer, and health. Louisiana
State University Press, Baton Rouge.

Cleveland, T. E., A. R. Lax, L. S. Lee, and D. Bhatnagar. 1987 . Appearance of enzyme
activities catalyzing conversion of sterigmatocystin to aflatoxin B1 in late-growth phase
Aspergz’llus parasiticus cultures. Appl. Environ. Microbiol. 53 : 1711-1713.

Codifer, L. P., G. E. Mann, and F. G. Dollear. 1976. Aflatoxin inactivation : treatment of
peanut meal with formaldehyde and calcium hydroxide. J. Am. Oil Chem. Soc. 53 : 204-206.

Cohen, G., D. Dembiec, and J. Marcus. 1970. Measurement of catalase activity in tissue
extracts. Anal. Biochem. 34 : 30-38.

Cole, R. J. 1989. Technology of aflatoxin decontamination, p. 177-184. In S. Natori and Y.
Ueno (ed), Mycotoxins and phytotoxins. Elsevier Science, Amsterdam.

Coros, L., J. P. Rousset, F. Kiefer, F. J. Wiebel, and M. C. Weiss. 1990. Genetic
analysis of aflatoxin Bl activation in rat hepatoma cells. Mol. Gen. Genet. 222 : 291-296.

Cory, R. G., and G. N. Wogan. 1981. Temporal patterns of covalent DNA adducts in rat
liver after single and multiple doses of aflatoxin B 1. Cancer Res. 41 : 197-203.

Cotty, P. J. 1990. Effect of atoxigenic strains of Aspergillus flavus on aflatoxin
contamination of developing cottonseed. Plant Dis. 74 : 233-235.

Cotty, P. J., P. Bayman, D. S. Egel, and D. S. Elias. 1994. Agriculture, aflatoxins, and
Aspergillus, p. 1-27. In K.A. Powell, A. Fenwick, and IF. Peberdy (ed), The genus
Aspergillus. Plenum Press, New York.

Cullen, J. M., and P. M. Newbeme. 1994. Acute hepatoxicity of aflatoxins, p. 3-26. In D.
L. Eaton and J. D. Groopman (ed), The toxicology of aflatoxins : human health, veterinary
and agricultural significance. Academic Press, New York.

Dalvi, R. R. 1986. An overview of aflatoxicosis of poultry : its characteristics, prevention and

124

reduction. Vet. Res. Commun. 10 : 429-443.

Darrah, L., and D. Barry. 1991. Reduction of preharvest aflatoxin in corn, p. 288-310. In
G.A. Bray and DH. Ryan (ed.), Mycotoxins, cancer, and health. Louisiana State University
Press, Baton Rouge.

Davidson, J. N., J. G. Babish, K. A. Delaney, D. R. Taylor, and T. D. Phillips. 1987.
Hydrated sodium calcium aluminosilicate decreases the bioavailability of aflatoxin in the
chicken. Poult. Sci. 66 : 89.

Davis, N.D., S.K. Iyer, and U.L. Diener. 1987. Improved method of screening for aflatoxin
with a coconut agar medium. Appl. Environ. Microbiol. 53 : 1593-1595.

Deger, G. E. 1976. Aflatoxin-human colon carcinogenesis? Ann. Int. Med. 85 : 204-205.

Domer. J. W., R. J. Cole, and P. D. Blankenship. 1992. Use of a biocompetitive agent to
control preharvest aflatoxin in drought stressed peanuts. J. Food Protect. 55 : 888-892.

Doyle M. P., and E. H. Marth. 1978. Bisulfite degrades aflatoxins aflatoxin : effects of
temperature and concentration of bisulfite. J. Food Protect. 41 : 774-780.

Dutton, M. F. 1988. Enzymes and aflatoxin biosynthesis. Microbiol. Rev. 53 : 274-295.

Eaton, D. L., and E. P. Gallagher. 1994. Mechanisms of aflatoxin carcinogenesis. Annu.
Rev. Pharmacol. Toxicol. 34 : 135-172.

Egbunike, G. N., and F. I. Ikeguonu. 1984. Effect of aflatoxicosis on acetylcholinesterase
activity in the brain and adenohypophysis of the male rat. Neurosci. Lett. 52 : 171-174.

Ehrlich, K. 1987. Effect on aflatoxin production of competition between wild-type and
mutant strains of Aspergillus parasiticus. Mycopathologia 97 : 93 -96.

Ellis, W. 0., J. P. Smith, B. K. Simposon, and J. H. Oldham. 1991. Aflatoxins in food :
occurrence, biosynthesis, effects on organisms, detection, and methods of control. Crit. Rev.
Food Sci. Nutr. 30 : 403-439.

Elsworthy, G. C., J. S. E. Holker, J. M. McKeown, J. B. Robinson, and L. J. Mulheirn.
1970. The biosynthesis of the aflatoxins. Chem. Comm. 17 : 1069-1070.

Epstein, S. M., B. Bartus, and E. Farber. 1969. Renal epithelial neoplasms induced in male
Wistar rats by oral aflatoxin B1. Cancer Res. 29 : 1045-1050.

Feng, G. H., F. S. Chu, and T. Leonard. 1992. Molecular cloning of genes related to
aflatoxin biosynthesis by differential screening. Appl. Environ. Microbiol. 58 : 455-460.

125

Foutz, K. R., C. P. Woloshuk, and G. A. Payne. 1995. Cloning and assignment of linkage
group loci to karyotic map of the filamentous fungus Aspergillusflavus. Mycologia.

Gardner, H. K., S. P. Koltun, F. G. Dollear, and E. T. Rayner. 1971. Inactivation of
aflatoxins in peanut and cottonseed meals by ammoniation. J. Am. Oil Chem. Soc. 48 : 70-73.

Gould, S. J., G. A. Keller, N. Hosken, J. Wilkinson, and S. Subramani. 1989. A
conserved tripeptide sorts proteins to peroxisomes. J. Cell Biol. 108 : 1657-1664.

Griffin, D. H. 1981. Molecular architecture, p. 40-72. In Fungal physiology. John Wiley &
Son, New York.

Groopman, J. D. 1994. Molecular dosimetry methods for assessing human aflatoxin
exposure, p. 259-280. In D. L. Eaton and J. D. Groopman (ed.), The toxicology of aflatoxins
: human health, veterinary and agricultural significance. Academic Press, New York.

Guilfoile, P. G., and C. R. Hutchinson. 1992. The Slreptomyces glaucescens TcmR protein
represses transcription of the divergently oriented tcmR and tcmA genes by binding to an
intergenic operator region. J. Bacteriol. 174 : 3659-3666.

Gurtoo, H. L., and C. V. Dave. 1975. In vivo metabolic conversion of aflatoxins and
benzo(a)pyrene to nucleic acid-binding metabolites. Cancer Res. 35 : 382-3 89.

Hagler, W. M. Jr. 1991. Potential for detoxification of mycotoxin-contaminated
commodities, p. 253-269. In G. A. Bray and D. H. Ryan (ed), Mycotoxins, cancer, and
health. Louisiana State University Press, Baton Rouge.

Hallam, S., F. Malpartida, and D. Hopwood. 1988. Nucleotide sequence, transcription and
deduced function of a gene involved in polyketide antibiotic synthesis in Slreptomyces
coelicolor. Gene 74 : 305-320.

Hamasaki, T., Y. Hatsuda, N. Terashima, and M. Renbutsu. 1967. Studies on the
metabolites of Aspergillus versicolor, part V. - isolation and structures of three new
metabolites, versicolorin A, B and C. Agr. Biol. Chem. 31 : 11-17.

Harper, A. E. 1965. Glucose-6-phosphatase, p. 788-792. In H. U. Bergrneyer (ed.), Methods
of enzymatic analysis. Academic Press, New York.

Harris, J. L. 1986. Modified method for fungal slide culture. J. Clin. Microbiol. 24 : 460-
461.

Harvey, R. B., T. D. Phillips, J. A. Ellis, L. F. Kubena, W. E. Huff, and H. D. Petersen.
1991. Effect on aflatoxin Ml residues in milk by addition of hydrated sodium calcium
aluminosilicate to aflatoxin contaminated diets of dairy cows. Am. J. Vet. Res. 52 : 1556-

126

1559.

Hase, T., U. Muller, H. Reizman, and G. Schatz. 1984. A 70-kd protein of the yeast
mitochondrial outer membrane is targeted and anchored via its extreme amino terminus.
EMBO J. 3 : 3157-3164.

Hohn, T. M., S. P. McCormick, and A. E. Desjardins. 1993. Evidence for a gene cluster
involving trichothecene-pathway biosynthetic genes in F usarium sporotrichioides. Curr.
Genet. 24 : 291-295.

Horng, J. S., P-K. Chang, J. J Pestka, and J. E. Linz. 1990. Development of a
homologous transformation system for Aspergillus parasiticus with the gene encoding nitrate
reductase. Mol. Gen. Genet. 224 : 194-196.

Hsieh, D. P. H. 1987. Mode of action of mycotoxins, p. 149-176. In P. Krogh (ed),
Mycotoxins in food. Academic Press, Cambridge.

Hsieh, D. P. H. 1989. Potential human health hazards of mycotoxins, p. 69-80. In S. Natori,
K. Hashimoto and Y. Ueno (ed), Mycotoxins and phycotoxins ‘88. Elsevier, Amsterdam.

Hsieh, D. P. H., M. T. Lin, and R. C. Yao. 1973. Conversion of sterigmatocystin to
aflatoxin B1 by Aspergillusparasiticus. Biochem. Biophys. Res. Commun. 52 : 992-997.

Hsieh, D. P. H., M. T. Lin, R. C. Yao, and R. Singh. 1976. Biosynthesis of aflatoxins :
conversion of norsolorinic acid and other hypothetical intermediates into aflatoxin B1 by
Aspergillusparasiticus. J. Agric. Food Chem. 24 : 1167-1170.

Hsu, I. C., R. A. Metcalf, T. Sun, J. A. Welsh, N. J. Wang, and C. C. Harris. 1991.
Mutational hotspot in the p53 gene in human hepatocellular carcinomas. Nature 350 : 427-
428.

Huff, W. E. 1980. A physical method for the segregation of aflatoxin-contaminated corn.
Cereal Chem. 57 : 236-238.

Huff, W. E., and W. M. Jr. Hagler. 1985. Density segregation of corn and wheat naturally
contaminated with aflatoxin, deoxynivalenol, and zearalenone. J. Food Protect. 48 : 416-420.

IARC 1993. International Agency for Research on Cancer. IARC monographs on the
evaluation of carcinogenic risks to human. Some naturally occurring substances : food items
and constituents, heterocyclic aromatic amines and mycotoxins. 56 : 362.

lchinose, K., J. Kiyono, Y. Ebizuka, and U. Sankawa. 1993. Post-aromatic deoxygenation
in polyketide biosynthesis : reduction of aromatic rings in the biosynthesis of fungal melanin
and anthraquinone. Chem. Pharm. Bull. 41 : 2015-2021.

127

Jefferson, R. A. Plant reporter genes : the GUS gene fusion system, p. 247-263. In Genetic
engineering : principles and methods, vol 10. Plenum Press, New York.

Jelinek, C. F., A. E. Pohland, and G. E. Wood. 1989. Review of mycotoxin contamination
: worldwide occurrence of mycotoxins in foods and feeds-an update. J. Assoc. Off. Anal.
Chem. 72 : 223-230.

Johnson, L. M., V. A. Bankaitis, and S. D. Emr. 1987. Distinct sequence determinants
direct intracellular sorting and modification of a yeast vacuolar protease. Cell 48 : 875-885.

Kadian, S. K., D. P Monga, and M. C. Goel. 1988. Effect of aflatoxin B, on the delayed
type hypersensitivity and phagocytic activity of reticuloendothelial system in chickens.
Mycopathologia 104 : 33-36.

Kalderson, D., B. L. Robert, W. D. Richardson, and A. E. Smith. 1984. A short amino
acid sequence able to specify nuclear location. Cell 39 : 499-509.

Keller, N. P., J. W. Cary, T. E. Cleveland, D. Bhatnagar, and G. A. Payne. 1992.
Characterization of a DNA unique to Aspergillus section F lavi. FASEB J. 6 : A228.

Keller, N. P., T. E. Cleveland, and D. Bhatnagar. 1992. Variable electrophoretic
karyotypes of members of Aspergillus section Flavi. Curr. Genet. 21 : 371-375.

Keller, N. P., H. C. Dischinger, Jr., D. Bhatnagar, T. E. Cleveland, and A. H. J. Ullah.
1992. Purification of a 40-kilodalton methyltransferase active in the aflatoxin biosynthetic
pathway. Appl. Environ. Microbiol. 59 : 479-484.

Keller, N. P., N. J. Kantz, and T. H. Adams. 1994. Aspergillus nidulans verA is required
for production of the mycotoxin sterigmatocystin. Appl. Environ. Microbiol. 60: 1444-1450.

Keller, N. P., S. Segner, D. Bhatnagar, and T. H. Adams. 1995. stcS, a putative P-450
monooxygenase, is required for the conversion of versicolorin A to sterigmatocystin in
Aspergillus nidulans. Appl. Environ. Microbiol. 61 : 3628-3632.

Kensler, T. W., N. E. Davidson, P. A Egner, K. Z. Guyton, J. D. Groopman, T. J.
Curphey, Y-L. Liu, and B. D. Roebuck. 1991. Chemoprotection against aflatoxin-induced
heptocarcinogenesis by dithiolethiones, p. 238-252. In G.A. Bray and DH. Ryan (ed),
Mycotoxins, cancer, and health. Louisiana State University Press, Baton Rouge.

Kirksey, J. W., R. J. Cole, and J. W. Domer. 1989. Relationship between aflatoxin content
and buoyancy in Florunner peanuts. Peanut Sci. 16 : 48-51.

Krishnamachari, K.A.V.R., R. V. Bhat, V. Nagarajan, and T. B. G. Telak. 1975.
Hepatitis due to aflatoxicosis. An outbreak in Western India. Lancet 1 : 1061-1063.

128

Kubena, L. F., W. E. Huff, R. B. Harvey, A. G. Yersin, M. H. Elissalde, D. A. Witzel,
L. E. Giroir, and T. D. Phillips. 1990. Effects of hydrated sodium calcium aluminosilicate
on growing turkey poults during aflatoxicosis. Poult. Sci. 69 : 175.

Larson, P. S., G. McMahon, and G. N. Wogan. 1980. Modulation of c-myc gene expres-
sion in rat livers by aflatoxin B1 exposure and age. Fundam. Appl. Toxicol. 20 : 316-324.

Lee, L. S., J. W. Bennett, A. F. Cucullu, and R. L. Ory. 1976. Biosynthesis of aflatoxin
B1 - conversion of versicolorin A to aflatoxin B, by Aspergillus parasiticus. J. Agric. Food
Chem. 24 : 1167-1170.

Lee, L.S., J.W. Bennett, A.F. Cucullu, and J.B. Stanley. 1975. Synthesis of versicolorin
A by a strain of Aspergillus parasiticus deficient in aflatoxin production. J. Agri. Food
Chem. 23 : 1132-1134.

Lin, B-K., and J. A. Anderson. 1992. Purification and properties of versiconal cyclase from
Aspergillusparasilicus. Arch. Biochem. Biophys. 293 : 67-70.

Lisker, N., and E. B. Lillehoj. 1991. Prevention of mycotoxin contamination (principally
aflatoxins and Fusarium toxins) at the preharvest stage, p. 690-712. In J. E. Smith and R. S.
Henderson (ed), Mycotoxins and animals foods. CRC press, Boca Raton, Florida.

Luchese, R. H., and Harrigan, W. F. (1993). Biosynthesis of aflatoxin - the role of
nutritional factors. J. Appl. Bacteriol. 74 : 5-14.

Liick, H. 1965. Catalase, p. 885-894. In H. U. Bergmeyer (ed), Methods of enzymatic
analysis. Academic Press, New York.

Mahanti, N., D. Bhatnagar, J. W. Cary, L. Loubran, and J. E. Linz. 1996. Structure and
function of fas-IA, a gene encoding a putative fatty acid synthetase directly involved in
aflatoxin biosynthesis in Aspergillus parasiticus. Appl. Environ. Microbiol. 62 : 191-195.

Maniatis, T., E.F. Fritsch, and J. Sambrook. 1989. Molecular cloning : a laboratory
manual. Cold Spring Harbor Laboratory, Cold Spring Harbor, New York.

Mann, G. E., L. P. Jr. Codifer, and F. G. Dollear. 1967. Effect of heat on aflatoxins in
oilseed meals. J. Agric. Food Chem. 15 : 1090-1092.

Markham, P. 1994. Organelles of filamentous fungi, p. 75-98. In N. A. R. Gow and G. M.
Gadd (ed), The growing fungus. Chapman & Hall, London.

Martin, L. F., and P. Liras. 1989. Organization and expression of genes involved in the
biosynthesis of antibiotics and other secondary metabolites. Annu. Rev. Microbiol. 43 : 173-
206.

129

Matsushima, K-L, Y. Ando, T. Hamasaki, and K. Yabe. 1994. Purification and
characterization of two versiconal hemiacetal acetate reductases involved in aflatoxin
biosynthesis. Appl. Environ. Microbiol. 60 : 2561-2567.

McCormick, S. P., D. Bhatnagar, and L. S. Lee. 1987. Averufanin is an aflatoxin precursor
between averantin and averufin in the biosynthetic pathway. Appl. Environ. Microbiol. 53 :
14-16.

McLean, M., and M. F. Dutton. 1995. Cellular interactions and metabolism of aflatoxin :
an update. Pharmac. Ther. 65 : 163-192.

McMahon, G., E. F. Davis, L. J. Huber, W. Kim, and G. N. Wogan. 1980. Characteriza-
tion of c-ki-ras and N-ras in aflatoxin B,-induced rat liver tumors. Proc. Natl. Acad. Sci.
USA. 87: 1104-1108.

Montenegro, E., F. Fierro, F. J. Fernandez, S. Gutierrez, and J. F. Martin. 1992
Resolution of chromosomes III and IV by pulsed-field gel electrophoresis shows that the
penicillin biosynthetic pathway genes pchB, pch, and meE are clustered on chromosome
V1 (3.0 megabases). J. Bacteriol. 174 : 7063-7067.

Miiller, W. H., R. A. L. Bovenberg, M. H. Groothuis, F. Kattevilder, E. B. Smaal, L.
H. M. Van der Voort, and A. J. Verkleij. 1992. Involvement of microbodies in penicillin
biosynthesis. Biochim. Biophys. Acta 1116 : 210-213.

Miiller, W. H., T. P. Van der Krift, A. J. J. Krouwer, H. A. B. Wiisten, L. H. M. Van
der Voort, E. B. Smaal, and A. J. Verkleij. 1991. Localization of the pathway of the
penicillin biosynthesis in Penicillium chrysogenum. 10 : 489-495.

Nelson, G. H., and C. M. Christensen. 1978. Diagnosis of mycotoxicoses in animals and
poultry : an overview, p. 3-8. In T. D. Wyllie and L. G. Moorhouse (ed), Mycotoxic fiingi,
mycotoxins, and mycotoxicoses, Vol. 2. Marcel Dekker, New York.

Newbeme, P. M., and W. H. Butler. 1969. Acute and chronic effects of aflatoxin on the
liver of domestic and laboratory animals. Cancer Res. 29 : 236-250.

Ngindu, A., P. R. Kenya, D. M. Ocheng, T. N. Omondi, W. Ngure, D. Gatei, B. K.
Johnson, J. A. Ngira, H. Nandwa, A. J. Jansen, J. N. Kaviti, and T. A. Siongok. 1982.
Outbreak of acute hepatitis caused by aflatoxin poisoning in Kenya. Lancet 1346-1348.

N iehaus, W. G. Jr., and R. P. Dilts, Jr. 1984. Purification and characterization of glucose-
6-phosphate dehydrogenase from Aspergillus parasiticus. Arch. Biochem. Biophys. 228 :
113-119.

Oakley, B.R., J.E. Rinehart, B.L. Mitchell, C.E. Oakley, C. Carmona, G.L. Gray, and

130

G.S. May. 1987. Cloning, mapping and molecular analysis of the pyrG (orotidine-5'-
phosphate decarboxylase) gene of Aspergillus nidulans. Gene 61 : 385-399.

Ohnuki, T., T. Imanaka, and S. Aiba. 1985. Self-cloning of Streptomyces griseus of an str
gene cluster for streptomycin biosynthesis and streptomycin resistance. J. Bacteriol. 164 : 85-
94.

Orr, W. C., and W. E. Timber-lake. 1982. Clustering of spore-specific genes in Aspergillus
nidulans. Proc. Natl. Acad. Sci. USA. 79 : 5976-5979.

Ottinger, M. A., and J. A. Doerr. 1980. The early influence of aflatoxin upon sexual
maturation in the male Japanese quail. Poultry Sci. 59 : 1750-1754.

Papa, K. E. 1973. The parasexual cycle in Aspergillusflavus. Mycologia 65 : 1201-1205.

Park, D. L., L. S. Lee, and S. A. Kolton. 1984. Distribution of ammonia-related aflatoxin
reaction products in cottonseed meal. J. Am. Oil Chem. Soc. 61 : 1071-1074.

Payne, G. A., G. J. Nystrom, D. Bhatnagar, T. E. Cleveland, and C. P. Woloshuk. 1993.
Cloning of the afl-Z gene involved in aflatoxin biosynthesis from Aspergillusflavus. Appl.
Environ. Microbiol. 59 : 156-162.

Peace, J. N ., C. D. Bartman, D. L. Doerfler, and I. M. Campbell. 1981. 6-Methylsalicylic
acid production in solid cultures of Penicillum patulum occurs only when an aerial mycelium
is present. Appl. Environ. Microbiol. 41 : 1407-1412.

Pestka, J. J. 1988. Enhanced surveillance of foodbome mycotoxins by immunochemical
assay. J. Assoc. Off. Anal. Chem. 71 : 1075-1081.

Pestka, J. J., and G. S. Bondy. 1990. Alteration of immune firnction following dietary
mycotoxin exposure. Can. J. Physiol. Pharmac. 68 : 1009-1016.

Pestka, J. J., and W. L. Casale. 1990. Naturally occurring fimgal toxins, p. 613-638. In J.
O. Nriagu, and M. S. Simmons (ed), Food contamination from environmental sources. John
Wiley & Sons, New York.

Phillips, T. D., B. A. Clement, and D. L. Park. 1994. Approaches to reduction of aflatoxins
in food and feeds, p. 383-406. In D. L Eaton and J. D. Groopman (ed.), The toxicology of
aflatoxins : human health, veterinary and agricultural significance. Academic Press, New
York.

Phillips, T. D., L. F. Kubena, R. B. Harvey, D. R. Taylor, and N. D. Heidelbaugh. 1988.
Hydrated sodium calcium aluminosilicate : a high affinity sorbent for aflatoxin. Poult. Sci. 67
: 243 -247.

131

Piva, G., F. Galvano, A. Pietri, and A. Piva. 1995. Detoxification methods of aflatoxins.
A review. Nutri. Res. 15 : 767-776.

Rarick, M., R. Thomas, C. D. Skory, and J. E. Linz. 1994. Identification and analysis of
the expression of the aflatoxin biosynthesis genes nor-1 and ver-I in the commercial species
Aspergillus sojae and A. oryzae as well as toxigenic and nontoxigenic strains of A. flavus. In
Proceedings, Current issues in food safety. Toxicology Center, Michigan State University,
Oct. 1994. Michigan State University, East Lansing.

Rayner, E. T., S. P. Koltun, and F. G. Dollear. 1977. Solvent extraction of aflatoxins from
contaminated agricultural products. J. Am. Oil. Chem. Soc. 54 : 242-244.

Sanger, F., S. N icklen, and A.R. Coulson. 1977. DNA sequencing with chain-terminating
inhibitors. Proc. Natl. Acad. Sci. 74 : 5463-5467.

Scott, G. E., and N. Zummo. 1988. Sources of resistance in maize to kernel infection by
Aspergillusflavus in the field. Crop Sci. 28 : 504-507.

Shane, S. M. 1991. Economic significance of mycotoxicoses, p. 53-64. In G. A. Bray and
D. H. Ryan (ed), Mycotoxins, cancer, and health. Louisiana State University Press, Baton
Rouge.

Shane, S. M. 1994. Economic issues associated with aflatoxins, p. 513-527. In D. L. Eaton
and J. D. Groopman (ed.), The toxicology of aflatoxins : human health, veterinary and
agricultural significance. Academic Press, New York.

Shantha, T., M. Sreenivasa, E. R. Rati, and V. Prema. 1986. Detoxification of groundnut
seeds by urea and sunlight. J. Food safety 7 : 225-232.

Shih, C. N., and E. H. Marth. 1974. Aflatoxin formation, lipid synthesis and glucose
metabolism by Aspergillusparasiticus during incubation with and without agitation. Biochim.
Biophys. Acta. 338 : 286-296.

Shroeder, H. W., R. J. Cole, R. D. Grigsby, and H. Hein, Jr. 1974. Inhibition of aflatoxin
production and tentative identification of an aflatoxin intermediate 'versiconal acetate’ from

treatment with dichlorvus. Appl. Microbiol. 27 : 394-399.

Singer, T. P., P. Bernath, and C. J. Lusty. 1965. Succinate, p. 340-345. In H. U.
Bergmeyer (ed), Methods of enzymatic analysis. Academic Press, New York.

Skory, CD. 1992. Isolation and characterization of a gene from Aspergillus parasiticus
associated with the conversion of versicolorin A to sterigmatocystin in aflatoxin biosynthesis.
Ph.D. Thesis. Michigan State University, East Lansing, Michigan.

132

Skory, C. D., P-K. Chang, J. Cary, and J. E. Linz. 1992. Isolation and characterization
of a gene form Aspergillus parasiticus associated with the conversion of versicolorin A to
sterigmatocystin in aflatoxin biosynthesis. Appl. Environ. Microbiol. 58 : 3527-3537.

Skory, C. D., P-K. Chang, J. Cary, and J. E. Linz. 1993. Regulated expression of the nor-
1 and ver-I genes associated with aflatoxin biosynthesis. Appl. Environ. Microbiol. 59 : 1642-
1646.

Skory, C. D., P.-K. Chang, and J. E. Linz. 1993. Regulated expression of the nor-1 and
ver-I genes associated with aflatoxin biosynthesis. Appl. Environ. Microbiol. 59 : 1642-1646.

Skory, C. D., J. S. Horng, J. J. Pestka, and J. E. Linz. 1990. Transformation of
Aspergillusparasiticus with a homologous gene(pyrG) involved in pyrimidine biosynthesis.
Appl. Environ. Microbiol. 56 : 3315-3320.

Smith, J. W., and P. B. Hamilton. 1970. Aflatoxicosis in the broiler chicken. Poult. Sci. 49
: 207-215.

Smith, J. E., and M. 0. Moss. 1985. Mycotoxins : formation, analysis and significance, p.
83-103. John Wiley & Sons, New York.

Smith, J. E., and M. 0. Moss. 1985a. Mycotoxins : formation, analysis and significance,
p. 127-143. John Wiley & Sons, New York.

Sophianopoulou, V., T. Suarez, G. Diallinas, and C. Scazzocchio. 1993. Operator
derepressed mutations in the proline utilization gene cluster of Aspergillus nidulans. Mol.
Gen. Genet. 236 : 209-213.

Steyn, P. S., R. Vleggar, and P. L. Wessels. 1980. The biosynthesis of aflatoxin and its
congeners, p. 105-155. In P. S. Steyn (ed), The Biosynthesis of mycotoxins : a study in
secondary metabolism. Academic Press, New York.

Stoloff, L. 1989. Aflatoxin is not a probable human carcinogen : the published evidence is
sufficient. Regul. Toxicol. Pharmacol. 10 : 272-283.

Tatebayashi, K., J. Kato, and I. Hideo. 1994. Structural analyses of DNA fragments
integrated by illegitimate recombination in Schizosaccharomyces pombe. Mol. Gen. Genet.
244:114-119.

Timberlake, W. E. 1993. Translational triggering and feedback fixation in the control of
fungal development. Plant Cell 5 : 1453-1460.

Tolbert, N. E. 19 . Isolation of subcellular organelles of metabolism on isopycnic sucrose
gradients. Methods Enzymol. 31 : 734-746.

133

Townsend, C. A., S. B. Christensen, and K. Trautwein. 1984. Hexanoate as a starter unit
in polyketide biosynthesis. J. Am. Chem. Soc. 106 : 3868-3869.

Townsend, C. A., S. M. McGuire, S. W. Brobst, T. L. Graybill, K. Pal, and C. E. Ram.
1991. Examination of tetrahydro- and dihydrobisfiJran formation in aflatoxin biosynthesis :
from whole cells to purified enzymes, p. 141-154. In R. J. Petroski and S. P. McCormick
(ed.), Secondary metabolite biosynthesis and metabolism. Plenum Press, New York.

Trail, F., N. Mahanti, and J. E. Linz. 1995. Molecular biology of aflatoxin biosynthesis.
Microbiology 141 : 755-765.

Trail, F., N. Mahanti, M. Rarick, R. Mehigh, S.-H. Liang, R. Zhou, and J. E. Linz.
1995. A physical and transcriptional map of an aflatoxin gene cluster in Aspergillus
parasiticus and the fiinctional disruption of a gene involved early in the aflatoxin pathway.
Appl. Environ. Microbiol. 61 : 2665-2673.

Trail, F., T.-S. Wu, and J. E. Linz. 1994. Identification of regulatory elements in two genes
involved in aflatoxin biosynthesis. In proceedings : the American Phytopathology Society
National Meeting, Albuquerque, New Mexico.

Van den Broek, G., M. P. Timko, A. P. Kausch, A. R. Cashmore, M. Van Montagu,
and L. Kerrera-Estrella. 1985. Targeting of a foreign protein to chloroplasts by fusion to
the transit peptide from the small subunit of ribulose-l,5-bisphosphate carboxylase. Nature
313 : 358-363.

Van Rensburg, S. J., P. Cook-Mozaffari, and D. J. Van Schalkwyk, J. J. Van der Watt,
T. J. Vincent, and I. F. H. Purchase. 1985. Hepatocellular carcinoma and dietary aflatoxin
in Mozambique and Transkei. Brit. J. Cancer. 51 : 713-726.

Vidal-Cros, A., F. Viviani, G. Labesse, M. Boccara, and M. Gaudry. 1994.
Polyhydroxynaphthalene reductase involved in melanin biosynthesis in Magnaporthe grisea -
purification, cDNA cloning and sequencing. Eur. J. Biochem. 219 : 985-992.

Walter, P., and V. R. Lingappa. 1986. Mechanism of protein translocation across the
endoplasmic reticulum membrane. Annu. Rev. Cell Biol. 2 : 499-516.

Wieder, R., G. N. Wogan, and M. B. Shimkin. 1968. Pulmonary tumors in strain A mice
given injections of aflatoxin B,. J. Natl. Cancer Inst. 40 : 1195-1197.

Wogan, G. N. 1991. Aflatoxin as risk factors for primary hepatocellular carcinoma in human,
p. 3-17. In G. A Bray and D. H. Ryan (ed), Mycotoxins, cancer, and health. Louisiana State

University Press, Baton Rouge.

Wolffe, A. P. 1994. Nucleosome positioning and modification : chromatin structures that

134

potentiate transcription. Trends. Biol. Sci. 19 : 240-244.

Woloshuk, C. P., K. R. Foutz, J. F. Brewer, D. Bhatnagar, T. E. Cleveland, and G. A.
Payne. 1994. Molecular characterization of aflR, a regulatory locus for aflatoxin biosynthesis.
Appl. Environ. Microbiol. 60 : 2408-2414.

Woloshuk, C. P., and G. A. Payne. 1994. The alcohol dehydrogenase gene ath is induced
inAspergiIIusflavus grown on medium conducive to aflatoxin biosynthesis. Appl. Environ.
Microbiol. 60 : 670-676.

Woloshuk, C. P., G. L. Yousibova, J. A. Rollins, D. Bhatnagar, and G. A. Payne. 1995.
Molecular characterization of the afl-I locus in Aspergillusflavus. Appl. Environ. Microbiol.
62: 3016-3023.

Wu, T.-S. 1995. A genetic approach to develop a detailed understanding of the regulation
of aflatoxin biosynthesis in Aspergillus parasiticus. Ph.D. Thesis. Michigan State University,
East Lansing, Michigan.

Wu, T.-S., and J. E. Linz. 1994. Transcriptional analysis of genes involved in aflatoxin
production. In Proceedings : Current Issue in Food Safety. Toxicology Center, Michigan
State University, East Lansing, MI.

Wu, T.-S., C.D. Skory, J.S. Horng, and J.E. Linz. 1996. Cloning and functional analysis
of a B-tubulin gene from a benomyl resistant mutant of Aspergillus parasiticus. Gene, in
press.

Yabe, K., Y. Ando, and T. Hamaski. 1991. Desaturase activity in the branching step
between aflatoxin B, and G, and aflatoxin B2 and G2. Agric. Biol. Chem. 55 : 1907-1911.

Yabe, K., V. Ando, J. Hashimoto, and T. Hamasaki. 1989. Two distinct 0-
methyltransferases in aflatoxin biosynthesis. Appl. Environ. Microbiol. 55 : 2172-2177.

Yabe, K., and T. Hamaski. 1993. Stereochemistry during aflatoxin biosynthesis : cyclase
reaction in the conversion of versiconal to versicolorin B and racemization of versiconal
hemiacetal acetate. Appl. Environ. Microbiol. 59 : 2493-2500.

Yabe, K., Y. Matsuyama, V. Ando, H. Nakajimu, and T. Hamasaki. 1993.
Stereochemistry during aflatoxin biosynthesis : conversion of norsolorinic acid to averufin.

Appl. Environ. Microbiol. 59 : 2486-2492.

Yanagita, T., and F. Kogane. 1962. Growth and cytochemical differentiation of mold
colonies. J. Gen. Appl. Microbiol. 8 : 201-213.

Yao, R. C., and D. P. H. Hsieh. 1974. Step of dichlorvos inhibition in the pathway of

135

aflatoxin biosynthesis. Appl. Microbiol. 28 : 52-57.

Yu, J., J. W. Cary, D. Bhatnagar, T. E. Cleveland, N. P. Keller, and F. S. Chu. 1993.
Cloning and characterization of a cDNA from Aspergillus parasiticus encoding an 0-
methyltransferase involved in aflatoxin biosynthesis. Appl. Environ. Microbiol. 59 : 3564-
3571.

Yu, J., P.-K. Chang, M. Wright, D. Bhatnagar, T. E. Cleveland, G. A. Payne, and J.
E. Linz. 1995. Comparative mapping of aflatoxin pathway gene clusters in Aspergillus
parasiticus and Aspergillusflavus. Appl. Environ. Microbiol. 61 : 2365-2371.

Yu S. Z., Z. Q. Cheng, Y. K. Liu, Z. Y. Huang, and Y. F. Chao. 1989. The aflatoxins and
contaminated water in the etiological study of primary liver cancer, p. 37-44. In S. Natori, K.
Hashimoto and Y. Ueno (ed), Mycotoxins and phycotoxins ‘88. Elsevier, Amsterdam.

Zaika, L. L., and R. L. Buchanan. 1987 . Review of compounds affecting the biosynthesis
or bioregulation of aflatoxins. J. Food. Protect. 50 : 691-708.

"‘l111111«111111“