. 21:43:.)14 radii; r... 4. .30..- . K .. A .1: £3- 9 A n 1‘.- ii}. ... . ‘ .. .zli,‘ ..I ..I:..: : ... ..J H'.Al 1K3: 1!(.J:; . .Vfl-oflu’ Olivia» Jig . 3.10.... » . o: . :l . . n . in... :1, ‘ . h.’llA-' ’ I:!:.¥$.IL.::.2 53w“... .4. .t 0.1!... ’OV‘J'I‘q . ( i 1.1.. :8. I 1.5. s :13: .1 3.9.3:. 1 4.41.. .1 . : .37 ’ i .. .. $21.33... .- O ‘ , I.'I‘I. o...( {tail-col P. .. . 5.. : Jug”. (I? '0': 7 ’.Of‘ ’nlot. n 1.: . a. 31:.» pram ‘5 l. 1"- If 'I""‘f - .53 I: O: of. S’dp-‘i Bambi... A . r.,)71nh.lv¢..fifl {.1 ; I!!!) z. (. pct. 13. I a}. l" . (PI-Ala; I. 00"}: vl.| {nu It‘d. It!!! ..r$‘$lllvoloa). Dalia [nit-yr 51 It?“ noiexl- 4319.. 10!..34 cud-dual} 012be v3 ‘ D n. I .0 Huang... as . kn: x ! 0|} 5 1'1- .hnr'yl 1001...?! 2...)»: 7...! 2:95 : D b o l :I: 3'55 ‘1: . on OH“. O grizignftt Itior‘nav’t‘i ('1’) I... . Lilli-8... '2 § .1.’.‘) I‘ll ‘, \I ‘ .bl..-I..lt1 I. r: .I a...) .58?ch In. '0 ll 'II‘D‘ l".'; ! . r .Jll‘fsl .l-Io. v1} l..>’ P..J. II. .I. : K... ;.t\q?'f‘1hnlin «‘0: 411105!) 1.0.... I! . A!.f¢-{IOO)IIAJDI|-b ‘7' n . ggnll'kfirlr 11.-.! s a. 1.58.... n . 5 5 lol. I trill Lott )‘lnvhflvlallzn-ae-vl a .1 fl captain! ”If 4.43.33. 4 iii. 3.. 11:15: "uu I, g '2 . nL.v. ‘Ob . A.‘ i i ‘1‘; 2;! A I’LIOAIE ‘13.} no“ .. u y it Inolfol ll.) , flu-'9‘! 4...?!)D. .rl‘::¢ taonlfur-‘v.‘a {..VA_.I..:I.1>§I.)I.II. ’i 1'; ; ‘. 1 {e g If! *5 f g FO‘AI'I '31.,” I'I'Iiv’.£. 1.0!. .onoi...‘. I'n’iv..'l~ll ..§O n! {:1}; n':‘l..l o.0A:... {fun . 140.00.. I . . P64. litu‘v‘a' . 11.13.71?! r. I... 15:. . ; :5 i; .>.w631. P!!f’.!v‘ VII." " I i . v1.1.1. Iza'v4.3¢.rvtseb£.:a vdftvorutvlutlvu - n, . (1...!) a.»u 3.3.3.1154 5 031.. " ¢jnhrll I} ‘ il.“- 3' ."C|'|.-d" - ‘ . NOV. 1 A if: . al.‘ A .»\...‘?1. . lit. 1‘“ Piaf. .11 3 . . 3 . . :..:...J.....:... 2‘ t:t¢.§flw1#.£.!. "O '3 ol- IOI) 01- ;}rro;,,1v¢l\ln\ r €000.00!)er ,‘nvgnvb.pi.notl:r!lo (10‘0“ ‘ ‘l 1‘ ’ i .3]; t'..‘ 5" .nl’ 0" {In 7;): ’ f! ‘ Us)? .4: :3 5...”... of». wow; u. . £‘ni’lfyzl (a u.|'0"ll‘" ozlfnilpou‘atu. iv a. ll .‘lg‘l‘lvitl‘:ln. n. hygdnflrg...’ . 151)); {gaf’:tf‘b‘" :‘oOlirzln I‘lybfllllrabtl" “lair“! IIi-I .l-l. v). I'vff‘a I‘ll!“ o 1):».1‘1. x v. to: 001v {all}!!! - XO‘IOII. . Ellv ,‘§O!‘ I . 9' O ‘ -” it uanIn,.-3..I. c» , rein”.'v:‘fl.b$u I ul. .5! Oil luai‘l v“ ll. .bbl. a 1):): .1: :toAIJIBVII‘b? . .' . uv'lll’nlftl: .: r .0... titliulclu - ‘11:? n. .3 .0]: Uzi. 1 .. )I ..I......4. {-zfllq th...llv-J I f..- ...I.|)J:ll to! .54!!! O. I . 7‘ I- ill ‘1! “. 'i It ‘ 1-. l u I I l . .l 5 n n! I . ‘I ." .rQHFJPJoNJ #11. I .3333}: A. l. . It. I. .r . 2‘5. . I a . .v 7- ... ‘ 1 , . . ‘1 £’ 0‘ {I‘P ‘ I .rtottuu Ivan I .1...I'.vu ' b I a 1 l r.... I: I -.|.I. pr} . o b" 1. WEEKS MI CHI IGAN STATEU I3 II III I IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII 300899 02 14 II This is to certify that the dissertation entitled Isolation and Characterization of a Gene from Aspergilius parasiticus associated with the Conversion of Versicoiorin A to Sterigmatocystin in Aflatoxin Biosynthesis presented by Christopher David Skory has been accepted towards fulfillment of the requirements for Doctoral Food Science degree in James J. Pestka, Ph.D. Major professor ‘JJ/zaxaa \/ ///~’;¢ Vita; 3Dam October 13, 1992 MS U is an Affirmative Action/Equal Opportunity Institution 0-12771 I, LIBRARY Michigan State University I It fi“ PLACE IN RETURN BOX to remove this checkout from your record. TO AVOID FINES return on or before date due. I—\_' DATE DUE DATE DUE DATE DUE fiI I fiI‘ I MSU is An Affirmative Action/Equal Opportunity Institution czbthpna-pd ISOLATION AND CHARACTERIZATION OF A GENE PROM ASPERGILLUS PERASITTCUS ASSOCIATED WITH THE CONVERSION OF VERSICOLORIN A TO STERIGMATOCYSTIN IN AFLATOXIN BIOSYNTHBSIS BY Christopher David Skory A DISSERTATION 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 1992 IXBSTTLACHF ISOLATION AND CHARACTERIZATION O? A GENE PROM ASPERGILLUS PARASITICUS ASSOCIATED WITH THE CONVERSION OF VERSICOLORIN A TO STERIGMATOCYSTIN IN AILATOXIN BIOSYNTEESIS BY Christopher David Skory A transformation system with efficiencies of 30 - 50 stable transformants per pg DNA was developed for Aspergillus parasiticus utilizing the homologous pyrG gene which encodes orotidine-monophosphate (0MP) decarboxylase. The pyrG gene from A. parasiticus was isolated by in situ plaque- hybridization of a lambda genomic DNA library using the heterologous A. nidulans pyrG gene as a probe. Uridine auxotrophs of A. parasiticus ATCC 36537 (var-1) , a mutant deficient in the conversion of the aflatoxin biosynthetic intermediate versicolorin A to sterigmatocystin, were isolated by selection on 5-f1uoro-orotic acid following nitrosoguanidine mutagenesis. Fungal isolates with mutations in the pyrG gene resulting in elimination of UMP decarboxylase activity were detected by assaying cell free extracts for their ability to convert “C labelled 0MP to [“C]- uridine- monophosphate (UMP). Transformation of A. parasiticus C810 (var-1, pyrG) protoplasts with the homologous pyrG gene restored the fungal cells to prototrophy. DNA isolated from the wild-type aflatoxin producing (Afl*) fungus Aspergillus parasiticus NRRL 5862 was then used to construct a cosmid genomic DNA library employing the homologous pyrG gene for selection of fungal transformants. The cosmid library was transformed into A. parasiticus C810 (var-1, pyrG). One pyrG*, Afl+ transformant was identified and used for marker rescue technique. Transformation of rescued DNA fragments into A. parasiticus C310 resulted in production of wild type levels of aflatoxin and abundant formation of sclerotia. The gene responsible for this complementation was identified by Northern RNA analyses and transformation with subcloned DNA fragments. The approximate location of transcription initiation and polyadenylation sites of ver-l were determined by RNase protection assay and cDNA sequence analysis. The predicted amino acid sequence revealed striking similarity'with.Streptomyces ketoreductases involved in polyketide biosynthesis. Appearance of ver-l RNA transcripts was not readily apparent until near the end of trophophase (growth phase) and this accumulation was followed approximately 8 hrs later by idiophase when aflatoxins are produced. Additionally, the presence of this transcript continued well beyond that which was observed with a gene such as pyrG that is involved in primary metabolism. The var-1 transcript did not accumulate when grown under non-aflatoxin supporting conditions, but was detectable 4-7 hrs after transfer to a gducose containing medium which does support aflatoxin biosynthesis. Dedicated to my parents and to all of my family. I owe my success to their love and support throughout my education. iv ACIWOWLEDGMENTS I thank Dr. Jyh-Song Horng who is primarily responsible for allowing the idea of isolating aflatoxin biosynthetic genes to become a reality. Thanks are also due to Dr. Perng- Kuang Chang who assisted Dr. Horng in the development of the niaD transformation system and for his help in the isolation of the nor-l gene. I am also grateful for the assistance of Reuvan Rasooly in sequencing the ver-l gene and thank Dr. Jeff Cary for sequencing the ver-l cDNA. I am especially indebted to the many people in the Food Microbiology lab for their assistance and support throughout my Doctoral project. Also, I appreciate the time and advice from Dr. Bill Helferich, Dr. Pat Oriel, and Dr. Pat Hart who have graciously served on my guidance committee. Most importantly, I want to express my extreme gratitude to Dr. James J. Pestka and Dr. John E. Linz. It was through their own success and their confidence in me that I found the motivation to succeed with this thesis. TABLE OF CONTENTS Page LIBTOPTABLBS................ .......... ................viii LI 8! or ’1 Guns O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O ix IMODUCTI on m ”T I om: O OOOOOO O O O O O O O O O O O O O O O O O O O O O O O 1 LITERATURE Rm 3' O O O O O O O O O O O O O O O O O O OOOOOOOO O O O O O O O O O O O O O 3 Natural occurence of aflatoxins .................... 3 Biosynthesis of aflatoxins ......................... 4 Regulation of aflatoxin biosynthesis ............... 23 Mutagenicity and carcinogenicity of aflatoxin and its intermediates .................. 26 “Tnlusmumons OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO 30 Strains and culture conditions ..................... 30 Chemicals .......................................... 31 Development of genetic transformation system ....... 31 Isolation of pyr mutant strains ............... 31 Enzymatic analysis of pyr mutants ............. 32 Transformation of fungal protoplasts .......... 36 Isolation and analysis of genomic DNA from fungal cells ............... 38 Isolation of var-1 gene ............................ 39 Preparation of cosmid library ................. 39 Preparation of cDNA library ................... 42 Screening cosmid genomic DNA library by genetic complementation of A. parasiticus C810 ...... 42 Confirmation of aflatoxin production by transformant colonies ....................... 43 Production of Sclerotia ....................... 44 Marker rescue of amp’ clones from an aflatoxin producing transformant .... ..... ............. 4S Isolation and analysis of RNA ................. 46 Analysis of RNA transcripts from var-1 gene ... 46 Sequence analysis ............................. 49 Computer analysis of sequence data ............ 50 Analysis of RNA expression for the var-1, nor-1, and pyrc genes . ..... .......... 51 Batch fermentation analysis ................... 51 vi nutritional Shift assay O O O O O O O O O O O O O O O O O O O O O O O 52 Analysis of dry weight, aflatoxin concentration, and RNA .......... ........ .... 53 RESULTS/DISCUSSION ...................................... 54 Isolation of the pyrG gene from A. parasiticus ..... 54 Isolation of uridine auxotrophs .................... 57 Enzymatic analysis of pyr mutants .................. 58 Transformation of pyrG mutants with plasmid pPGBJ .. 61 Southern analysis of prototrophic transformants from plasmid pPG3J ............................... 62 Enzymatic analysis of pyrG transformants ........... 66 Cosmid library preparation ......................... 67 Transformation of A. parasiticus C810 with pooled DNA library .......................... 68 Marker rescue of amp‘ gene 69 Transformation of A. parasiticus C810 with pVC1A-D . 70 Analysis of A. parasiticus Afl+ transformants ...... 73 Localization of mRNA transcripts encoded by the var-1 gene fragment .......................... 78 Location of the var-1 transcript ................... 81 Analysis of transcripts from the var-1 gene ........ 82 Nucleotide sequence analysis of the ver-l gene ..... 82 Linkage of nor-1 and ver-l genes ................... 91 Expression of aflatoxin associated genes during batch fermentation ........................ 92 Expression of aflatoxin associated genes following'nutritional shift ...................... 103 Southern analysis of A” parasiticus NRRL 5862 ...... 108 Southern analysis of related.Aspergilli sp ......... 114 conuunxuam00000000000000.ooooooooooooooooooooooo121 LlrmmcxrmOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO125 vii LIST OF TABLES Ilhll 239! 1. Production of aflatoxin by.AfI’mutants Of AO paraSIticus OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO 14 2. Production of sclerotia by various A. parasiticus strains grown on PDAmedium OOOOOOOOOOOOOOOOOOOOOO 75 3. Aflatoxin accumulation following nutritional Shift OOOOOOOOOOOOOOO_ OOOOOOO OOOOOOOVOOO 105 LIST OF FIGURES time use 1. Proposed pathway for the biosynthesis Of aflatOXin OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO 6 2. Proposed pathway for the production at “oraclorinic aCid OOOOOOOOOOOOOOOOOOOOOOOOOOOOO 8 3. Potential sources of malonyl CoA and acetyl CoA for incorporation into polyketide synthesis ...... 10 4. Proposed conversion of norsolorinic to aver-“fin OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO 15 5. Proposed conversion of versiconal hemiacetal acetate to aflatoxin B, and B, 20 6. SYDthBBIB Of uridine 5"m0n0ph08phate 000000000000 33 7o COsmid VBCtOI' szs coco-0000000000000...ooooooeooo 4O 8. Restriction endonuclease map and sequencing strategy of the var-1 gene ....................... 47 9. Restriction endonuclease maps of the pyrG gene ... 55 10. Restoration of orotidine-S'-monophosphate activity in pyrG mutants after transformation .... 59 11. Southern hybridization analysis of ten A. parasiticus pyrG transformants ................ 63 12. DNA fragments obtained from an A. parasiticus C810 aflatoxin producing transformant ................. 71 13. Summary of Northern analysis using subcloned fragments from pVC1A as probe .................... 79 14. RNase protection assay of 5' and 3' ends of the var-1 transcript .......................... 83 15. Nucleotide sequence of the var-1 gene ............ 86 ix 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. Dotplot analysis of deduced peptide sequences for the var-1 gene and Streptomyces actIII gene ...... Production of aflatoxin during batch fermentation ........................ Accumulation of var-1 transcript during batch fermentation ........................ Accumulation of nor-1 transcript during batch fermentation ........................ Accumulation of pyrG transcript during batch fermentation ........................ Accumulation of ver-l and nor-1 transcripts following nutrient shift ......................... Southern hybridization analysis of A. parasiticus genomic DNA digested with assorted restriction endonucleases and probed with var-1 .............. Southern hybridization analysis of A. parasiticus genomic DNA digested with assorted restriction endonucleases and probed with nor-1 .............. Southern hybridization analysis of genomic DNA from various Aspergillus spp. probed with var-1 .. Southern hybridization analysis of genomic DNA from various Aspergillus spp. probed with nor-1 .. 89 93 96 98 101 104 109 112 115 118 INTRODUCTION AND RATIONALE Aflatoxins are a group of low'molecular weight secondary metabolites produced by the fungi Aspergillus parasiticus and A. flavus. These compounds are known to be potent animal hepatocarcinogens and are suspected to be carcinogens of humans (Chu, 1991). Frequent. contamination. of‘ domestic agricultural commodities such as corn, peanuts, and cottonseed presents a potential threat to the health of animals and humans (Jelinek et a1, 1989). There has been considerable effort, since their discovery in the early 1960's, to study the biosynthesis of aflatoxins. This work has allowed the proposal of a generally accepted, but incomplete biosynthetic pathway scheme. Currently, there is a limited understanding of the details of the molecular mechanisms which regulate aflatoxin production. Investigation into the regulation of this biosynthetic pathway at the molecular level has proceeded primarily by three different strategies: (1) identification and purification of proteins involved in conversion steps leading to aflatoxins, (2) differential hybridization analysis for isolation of genes involved in aflatoxin biosynthesis, and, (3) complementation analysis using a genomic DNA library to ‘transform fungal mutants deficient in aflatoxin biosynthesis. 2 The efforts of this project were initially directed at developing a genetic transformation system in A. parasiticus that would permit efficient manipulation of DNA in the fungus. This work was performed to provide a means of identifying and characterizing one of the genes (var-1) involved in the conversion of the aflatoxin precursor versicolorin A to sterigmatocystin. The isolation of the var-1 gene could then proceed by way of genetic complementation of A. parasiticus var-1 mutants that were unable to carry out this biotransformation step. The longterm goal of this research was to provide information that would ultimately contribute to the eventual elimination of aflatoxin contamination in food and feed. The var-1 gene is being used in studies of the expression of genes associated with aflatoxin biosynthesis. Understanding the control of aflatoxin biosynthesis should allow critical regulatory points in the pathway to be defined. These regions may then serve as inhibitory sites to be used for management of aflatoxin contamination. This information could facilitate the development of agents or genetically engineered plants which are capable of inhibiting toxin production. Another potential use for aflatoxin biosynthetic genes is the construction, by methods of gene disruption, of genetically stable atoxigenic fungal strains. These could then be used for competitive exclusion of aflatoxin producing strains in the field. Biocompetition of this type has proven to be quite successful in laboratory and field tests (Cole et a1, 1991; 3 Cotty, 1990; Cotty and Daigle, 1991; Ehrlich, 1987) LITERATURE REVIEW Natural Occurrence of aflatoxins Aflatoxins are produced primarily by the fungi A. parasiticus and A. flavus. The ubiquitous nature of these organisms presents a worldwide dilemma of preventing contamination by these toxin producers. However, the mere presence of these strains does not necessarily result in toxin contamination since not all strains of A. flavus produce aflatoxin. There are also numerous other factors such as substrate availability, temperature, humidity, and competing microflora which can affect the biosynthesis of aflatoxins (Heathcote, 1978). Many different types of food commodities throughout the world have increased susceptibility to aflatoxin contamination. These include corn, peanuts, brazil nuts, pistachio nuts, cottonseed, rice, and copra (Jelinek et a1, 1989) . Aflatoxin contamination is known to vary regionally as a result climate differences. Preharvest fungal growth in corn and peanuts is favored by warm temperatures and prolonged drought conditions which can lead to crop damage. Postharvest 4 aflatoxin production is typically seen with warm temperatures and high humidity. In the United States, corn is most frequently contaminated with aflatoxins in the southeastern regions of the United States as a result of these types of weather conditions. The midwestern regions typically have not encountered such problems with the exception of severe drought conditions in 1983 and 1988 which resulted in a widespread aflatoxin dilemma (CAST, 1989; Chu, 1989; Ellis et a1, 1991; Pestka and Casale, 1990). Aflatoxins can be potentially controlled by preventing fungal contamination or growth, precluding toxin production, or by detoxification of contaminated commodities. Ammoniation has been proposed as a means of lowering concentrations of aflatoxin in contaminated feeds (Park et a1, 1988). The effectiveness of this technique for ensuring the breakdown of aflatoxin and its mutagenic and toxic residues is still questioned. To date, the FDA has not approved ammoniation as a method of detoxifying foods and feeds. Thus, prevention of aflatoxin contamination by preventing fungal growth or toxin production appears to be the most feasible approach. Biosynthoais of aflatoxins. Aflatoxins are secondary metabolites which are produced by the imperfect fungi Aspergillus parasiticus and.A. flavus. Both species produce aflatoxin B1 and 8,, however, only A. S parasiticus is known to also produce aflatoxin GI and 6,. Biotransformation studies with blocked mutants using radiolabeled precursors and pathway intermediates have allowed an aflatoxin biosynthetic pathway to be proposed (Figure 1). Most efforts have focused on the production of aflatoxin 8‘ since this compound is the most abundant and most biologically active of the aflatoxins. The conversion of acetate to aflatoxin B1 (AFB,) through the polyketide pathway was first suggested by Biollaz et a1 (1970). It was determined using radiolabelled [1-“CJ-acetate and [2-“CJ-acetate that the carbon skeleton of AFB, was completely derived from acetate subunits. The condensation of these acetate molecules is believed to be catalyzed by the enzyme polyketide synthase (PKS) which is very similar in both structure and function to the eukaryotic enzyme fatty acid synthase (Dondadia et a1, 1991; Hopwood and Sherman, 1990) (Figure 2). Polyketide synthesis in aflatoxin biosynthesis typically consists of sequential condensation or chain elongation of malonyl subunits followed by modification of the growing chain. The source of the malonyl subunit is believed to be derived most often from the carboxylation of acetyl CoA at the expense of one ATP (Dutton, 1988; Bhatnagar et a1, 1992) (Figure 3). Condensation of malonyl CoA to the growing carbon chain, which is bound to the acyl carrier protein (ACP), is catalyzed by B-ketoacyl synthase. Acetyl transferase (or CoA- Figure 1. Proposed pathway for the biosynthesis of aflatoxin 8,. A. parasiticus mutants impaired in specific bioconversion steps are represented next to an arrow disrupted by a double line. The arrow disrupted by a single line represents inhibition of conversion by the insecticide dichlorvos. a) A. parasiticus . a ATCC 24690 mourn ” o a I I a. Nouolorinic Acid \ n o Averanun / A’“ A- 2 iii on o TTCSx-gznam Anrutanln O a (“/2 u A. arasi 'cus o a Tree 24551 Hydromraicolomna 0 \II 't as A. parasiticus \ HVN-l Versiconal HeminceLl 7...... on em) /\/ on \ Vanioolorln A Dichlorvos 0-mihybtcrlgmatocyafln \’ A. Eamiticua ’\ § 043 0 O Aflatoxin a, Figure 2. Proposed pathway for the production of norsolorinic acid through polyketide biosynthesis. The oval structure represents the polyketide synthase containing an acyl carrier protein (I) and a 6-ketoacyl synthase (0) . Thiol groups are represented as -8H. p ooooooooooooooo nnnnnnnn ============= ooooooooooooo ooooooooooooooooooooooooooooo 10 Figure 3. Potential sources of malonyl CoA and acetyl CoA for the incorporation into polyketide synthesis. TA, transamination; PCK, phosphoenolpyruvate carboxykinase; PC, pyruvate carboxylase; AC, acetyl CoA carboxylase; PD, pyruvate dehydrogenase; PK, pyruvate kinase; AS, asparaginase; OD, oxaloacetate dehydrogenase. 11 £6 33¢. <00-w-0fl0 nzo Noo ... 1:192 Iw< omcCOMOLvhnop enscomOLpaan ’ flau< Uaflth—Omhéz ‘ unseen—pus 17 solid column matrix and passing a crude cell free homogenate containing the enzyme through the affinity column (Chuturgoon et a1, 1990). This purified enzyme has been found to be capable of catalyzing the conversion of NA to averantin and has a molecular mass of approximately 140 KDa (Chuturgoon and Dutton, 1991). Bhatnagar et a1 (1990) were able to purify a NA reductase using a five step sequential purification scheme. It was determined with SDS-PAGE that two major protein bands of 38 KDa and 48 KDa were present in this purified sample. Polyclonal antibodies against this protein mix were capable of blocking the NA.acid.reductase activityu However, it is still questionable whether only one enzyme activity is present is this sample. Without knowing what bioconversions normally occur with NA, we cannot rule out the possibility that Bhatnagar et al are observing activity of two or more enzymes which are capable of converting NA to averantin by a different route than the proposed single step conversion. However, finding two supposedly different enzymes that apparently bind and catalyze the conversion of NA does seem to support the hypothesis of alternative pathway steps in the biotransformation of NA as suggested by Bhatnagar et a1 (1992) . Additionally, it is also possible that A. parasiticus ATCC 24690 (nor-1) may be leaky as a result of only one of the proposed bioconversion routes being impaired. Recently, Chang et a1 (1992) have isolated a gene that is capable of restoring aflatoxin biosynthesis to the described NA accumulating mutant through genetic complementation. 18 Sequence analysis of this gene suggests that it most likely encodes a dehydrogenase with a predicted molecular size of approximately 29 Kda (Chang et al, unpublished). It is believed that this dehydrogenase is involved in reduction of norsolorinic acid, since NA accumulates in the absence of this activity. However, the discrepancy in size from the 110 Kda and 38/48 kDa dehydrogenases previously isolated indicates either a flaw in the proposed pathway or an alternative catalytic activity for the enzyme encoded by the nor-1 gene. If the intermediate 5-hydroxyaverantin is included into Bhatnagar's proposed pathway, it is possible that the nor-1 dehydrogenase is instead involved in the conversion of 5- hydroxyaverantin to averantin. Successful gene disruption of the nor-1 gene in an aflatoxin producing A. parasiticus strain should help identify the function of this gene and determine whether alternative conversion steps exist. It is generally accepted that averufin.is converted to 1- hydroxyversicolorone and then eventually to vericonal hemiacetal acetate (VHA). Mutant strains are available that are impaired in each of these bioconversion steps (Donkersloot et a1, 1972; Townsend et a1, 1988). The pathway intermediate versiconal hemiacetal acetate was identified by Shroeder et a1 (1974) when it was noted that an orange pigment accumulated when the fungus was grown in the presence of the insecticide dichlorvos which is an acetylcholine esterase inhibitor. It was later determined that this accumulation was the result of enzymatic inhibition of the conversion of versiconal 19 hemiacetal acetate to versicolorin A (Yao and Hsieh, 1974; Bennett et a1, 1976) . Dichlorvos presumably prevents the removal of the esterified acetate on versiconal hemiacetal acetate which is required for the closure of the terminal furan ring (Bennett et a1, 1976). It has recently been proposed that the conversion of VHA to versicolorin A (VA) involves several steps that include a point of branching for the production of AFB, and AFB2 (Lin and Anderson, 1992; Yabe et a1, 1991; Yabe et a1, 1991; Bhatnagar et a1, 1991) (Figure 5). Hsieh et a1 (1989) first identified the enzymatic activities of aflatoxin producing strains that were capable of converting VHA to a versiconal hemiacetal alcohol (VI-I) and then eventually to versicolorin C (VC). These enzymatic steps were later confirmed to be catalyzed by an esterase and a cyclase respectively (Anderson and Chung, 1990). Lin and Anderson (1992) were able to purify the versiconal cyclase to homogeneity and determine that it contains two identical subunits of a molecular size of 72 KDa each. It should be noted that VC and versicolorin B (VB) are often used interchangeably because VC is a racemic mixture of VB (VB and VB') and can't be separated by conventional chromatographic techniques. versicolorin C (or VB) can be converted to either VA by way of a desaturase or to dihydrodemethylsterigmatocystin (DHDMST) by means not yet established. Because of substrate similarities, the enzymes involved in the conversion of VA to demethylsterigmatocystin (DMST) most likely also catalyze the conversion of VC to 20 Figure 5 . Proposed conversion of versiconal hemiacetal acetate to aflatoxin B, and aflatoxin B,. VHA, versiconal hemiacetal acetate; VI-I, versiconal hemiacetal alcohol; VA, versicolorin A; VC, versicolorin C; DMST, demethyl- sterigmatocystin; DHDMST, dihydrodemethylsterigmatocystin; 8T, sterigmatocystin ; DHST, dihydrosterigmatocystin; OMST, O- methylsterigmatocystin; DHOMST, dihydro-O- methylsterigmatocystin; AFB,, aflatoxin 8,; AFB,, aflatoxin 8,. 21 51mm... Esterase Tim- . O VHA ‘" VH O Cyclase m... Desa turase mu VA " vc (VB)°' ° °” DHDMST ° °‘ I o e .... ST '§ j“ DHST DHDMST ° AFB1 'E. C. AFBZ o— 0— : g <2 22 1 s gg‘qfi‘dg'og‘ . . ,3 .5 .' 22 DHDMST. A mutant deficient in the conversion of versicolorin A to sterigmatocystin was isolated by Bennett and Goldblatt (1973) using ultraviolet mutagenesis. This isolate was unable to produce aflatoxins and produced significant quantities of four bright yellow pigments with the predominant compound designated as versicolorin A (Lee et a1, 1975). Singh and Hsieh (1977) were able to confirm the position of VA in the biosynthetic pathway through the use of radiolabeled precursor studies“ Little is yet known regarding the conversion of this anthraquinone containing intermediate to the xanthone based compound, sterigmatocystin (ST). A gene has been isolated (described in this thesis) which is capable of restoring aflatoxin production to a strain derived from the VA accumulating mutant isolated by Bennett and Goldblatt (1973). Sequence analysis strongly suggests that the product of this gene is a reductase. However, it should be mentioned that multiple steps are most likely involved in the conversion of VA to ST and would not be expected to be catalyzed only by a single reductase. Sterigmatocystin was suggested to be a pathway intermediate by Hsieh.et a1 (1973) and later confirmed by Sing and Hsieh (1976). Bhatnagar et a1 (1988) demonstrated that sterigmatocystin was converted to O-methylsterigmatocystin (OMST) by a methyltransferase and finally to AFB, by an oxidoreductase (Bhatnagar et a1, 1991) . There is strong evidence that an immediate precursor to sterigmatocystin (8T) 23 is a compound, demethylsterigmatocystin (DMST) , that lacks methylation on both the C-7-OH and C-6-OH groups (Yabe et a1, 1989). Methylation of DMST and ST are believed to be catalyzed by different methylases which also appear to catalyze the methylation of dihydrodemethylsterigmatocystin (DHDMST) and dihydrosterigmatocystin (DHST) , respectively (Yabe et a1, 1989) . Additionally, the methoxy methyl group of AFB, was found to be indicative of methionine origin. A methyltransferase that appears to be involved in the methylation of ST and DHST has been purified to homogeneity and determined to contain subunits of a molecular mass of 110 XDa and 58 KDa (Bhatnagar et a1, 1989) . A second unique methyltransferase has recently been isolated by Keller et a1 (1991) and shown to also be involved in synthesis of aflatoxins. This methyltransferase is believed to be involved in methylation of DMST and DHDMST. Regulation of aflatoxin biosynthesis. Aflatoxins are considered to be secondary metabolites because they do not seem to be essential for growth and reproduction, and their synthesis seems to be unique from that of primary metabolism. However, regulation of secondary metabolism can not be examined without considering primary metabolism since the products of primary pathways serve as precursors for secondary metabolites (Drew and Demain, 1977) . 24 Additionally, it is believed that primary metabolites may often be involved in the induction of enzymes of secondary metabolism. Buchanan et a1 (1985) demonstrated that aflatoxin biosynthesis occurred during a period of repressed tricarboxylic1acid.cycle (TCA) activity. It was shown that the activities of glucose 6-phosphate dehydrogenase, mannitol dehydrogenase and malate dehydrogenase decreased shortly after the addition of glucose and just prior to the induction of aflatoxins (Buchanan et a1, 1984) . He suggested that aflatoxin biosynthesis may be regulated by a carbon catabolite repression of NADPH-generating and TCA cycle enzyme activities. The overall energy status of the fungal cell does not appear to significantly change as a result of these alterations in the glycolytic and TCA cycle activities. However, findings that concurrent inactivation of mitochondria occurs with glucose induced aflatoxin biosynthesis may suggest that control is partly regulated by the energy status of specific subcellular compartments (Buchanan et a1, 1987). Additional evidence of the relationship between aflatoxin biosynthesis and levels of NADPH was provided indirectly by the ability of zinc to induce aflatoxin biosynthesis (Coupland and Niehaus, 1987; Niehaus and Failla, 1984). Foreman and Niehaus (1985) revealed that zinc ions inhibit the enzyme activities of mannitol dehydrogenase, mannitol 1-phosphate dehydrogenase, glucose 6-phosphate dehydrogenase, and 6- phosphogluconate dehydrogenase. It is theorized that 25 inhibition of enzymes of the pentose phosphate pathway (also called.phosphogluconate pathway) and.NADPH generating enzymes of the mannitol cycle result in a lowering of the cellular NADPH/NADP ‘ratio .and. an increased. activity' of aflatoxin biosynthetic enzymes. Additional evidence to support this notion was provided by determining that nitrate induces synthesis of the enzymes in the pentose phosphate pathway and the mannitol cycle, thereby resulting in an inhibition of aflatoxin biosynthesis (Neihaus and Jiang, 1989). The precursors for aflatoxin biosynthesis rely on many metabolites of primary metabolism. A correlation between aflatoxigenic isolates and high pyruvate kinase activity was noted by Tyagi and Venkitasubramanian (1981). This relationship is believed to be a result of additional pyruvate being made available for conversion into malonyl CoA (Figure 3). Halonyl CoA, as previously described, can then be converted to acetate and incorporated into AFBV. ,Asparagine and aspartate can also stimulate aflatoxin biosynthesis presumably by providing malonyl CoA through transamination and oxidation (Dutton, 1988; Reddy et a1, 1971). It is generally believed that filamentous fungi usually control gene expression at the transcriptional level (Gurr et a1, 1989) . The work of Buchanan et a1 (1987) provided evidence that aflatoxin biosynthesis is at least partially regulated at the transcriptional level. This was demonstrated using transcription (actinomycin D) and translation (cycloheximide) inhibitors after culture 26 replacement from a medium unable to support aflatoxin production to an aflatoxin inducing and supporting medium. It was established that transcriptional and translational processes associated with aflatoxin biosynthesis occurred 3-6 hrs and 6-10 hrs respectively following transfer to the aflatoxin supporting medium. Cleveland and Bhatnagar (1990) found that the methyltransferase enzyme was produced de novo slightly before the onset of aflatoxin appearance. Hutegenicity end carcinogenicity of aflatoxin and its intermediates. All precursors prior to versicolorin A are believed to lack significant mutagenic potential. Wong et a1 (1977) concluded that the mmtagenicity, as determined by the Ames Test, of versicolorin A resides in the bisfuran ring rather than the anthraquinone moiety. Sterigmatocystin, which has a xanthone structure in place of the anthraquinone moiety demonstrated.a two-fold increase in.mutagenicity. Conversion of sterigmatocystin to the coumarin containing AFB, resulted in approximately a 10 fold increase in mutagenicity. Hendricks et a1 (1980) chose to use the embryo exposure technique as a means of predicting the carcinogenic potential of versicolorin A and sterigmatocystin. It was established that the relative hepatocarcinogenicity of AFB, to versicolorin A is between sixteen and fifty times more potent. 27 Sterigmatocystin was estimated to be approximately one-fourth the carcinogenic potency of AFB,. It is now believed that the biological activity of the intermediates and AFB, is generated by the epoxidation of the 2,3-vinyl ether double bond in the terminal furan ring and its subsequent covalent binding to nucleic acid (Essigman et a1, 1982). This bioactivation of AFB, in mammalian systems is believed to occur by way of the mixed function oxidase cytochrome P450 (Shimada and Guengerich, 1989; Kitada et a1, 1990) . The resultant epoxide is capable of forming adducts in DNA, primarily at nitrogen-7 of guanine, as well as reacting with RNA and protein (Stark, 1980) .‘ DNA adducts following exposure to AFB, are found primarily in the liver although they can also be found in other tissues (Wild et a1, 1990). AFB,-DNA adducts are believed to inhibit RNA synthesis (Yu et a1, 1990) and induce mutagenic effects (e.g. frameshifts) as well as several other genotoxic effects (Chu, 1991). The toxic effects of aflatoxins were first realized in 1960 after the death of approximately 100,000 young turkeys in England, which had consumed aflatoxin contaminated feed (Nesbitt et a1, 1962) . Since then, numerous studies have been performed to confirm the powerful toxic and carcinogenic potential of aflatoxins in animal models (Heathcote, 1978). Acute aflatoxicosis in humans has been observed in parts of Taiwan, Thailand, Uganda, and Kenya. Consumption of foods highly contaminated with aflatoxin has resulted in varied symptoms that include loss of appetite, vomiting, abdominal 28 pain, pulmonary edema, necrosis of the liver, signs which resemble Reye's syndrome, and death (CAST, 1989) . Aflatoxins have also been shown to have immunosuppressive effects in mammals. Recognizing aflatoxins as the root cause of an infectious disease is difficult since the symptoms of aflatoxicosis will likely be masked by the manifestation of the infection itself (Pestka and Bondy, 1990; Cusumano et a1, 1990). Identifying aflatoxins as a causative agent in human cancers becomes exceeding difficult because of the reliance on epidemiological studies. There does appear to be a geographical correlation between incidence of human liver cancer and prevalence of aflatoxins (Stoloff, 1976). Liver cancer is the most common malignancy in males in parts of South-East Asia and sub-Saharan Africa where dietary exposure to aflatoxins is high (Parkin et a1, 1980) . Wild et a1 (1990) .were able to detect aflatoxin-albumin adducts in the sera of 12-100% of subjects tested throughout parts of Africa. This contrasts with no detection of any such adducts from sera of those from France or Poland where exposure to aflatoxin is much lower. It can then be inferred that the high incidence of primary liver cancer could be a result of aflatoxin ingestion and metabolism to the mutagenic forms. However, studies are complicated by the fact that hepatitis B virus is usually endemic in many of regions where aflatoxins present a potential threat of inducing primary liver cancer. This interfering element of complexity with the lack of any other 29 strong epidemiological evidence has persuaded many researchers to argue that aflatoxins are not a probable human carcinogen (Stoloff, 1989; Campbell et a1, 1990; Campbell et a1, 1991). Whether aflatoxins act synergistically with hepatitis B virus to result in liver cancer is a strong possibility that is currently being investigated (Chu, 1991) . An interesting observation is the common association of primary liver carcinomas believed to be a result of aflatoxins and/or hepatitis B virus and mutations in the third position.of codon 249 of the p53 genem Studies with aflatoxin induced.tumors in nonhuman primate models suggest that mutations in the p53 gene are not prerequisite for primary liver cancer (Fuj imoto et a1, 1992). Further* studies ‘will be needed. to determine 'the relationship of this specific mutation and aflatoxin/hepatitis B exposure. While additional studies are needed to confirm the association between aflatoxins and primary liver cancer, there does appear to be ample evidence that aflatoxin pose a potential health risk. Acute toxicity in animals and humans should be reason enough to warrant aggressive research into the prevention of this common food contaminant. 30 MATERIALS AND IVIETHODS Strains and culture conditions. Escherichia coli H8101 [hsd820 (rang), recA13, ara-14, proA2, 1acY1, galK2, rpsL20, Sm', xyl-S mtl-l, supE44] was used for propagating plasmids. E. coli L8392 [Fe14‘(mcrA‘) hst514 (r,‘m,*)sup844, supF58, 1acY1,.ga1K2, galT22, metBl, tryR55] was the recipient strain for packaged Lambda phage. Aspergillus parasiticus NRRL 5862 (SU-l) served as the aflatoxin producing wild type strain. A. parasiticus C810 (ver-l, wh-l, pyrG) (Skory et a1 1990), derived from A. parasiticus ATCC 36537 (ver-l, wh-l) (Bennett and Goldblatt, 1973), was used in complementation studies for the isolation of the ver-l gene . A. parasiticus ATCC 36537 was obtained by ultraviolet mutagenesis of A. parasiticus NRRL 5862. This "blocked mutant" is unable to convert VA to ST. Although the exact nature of the ver-l mutation is not known, neither A. parasiticus ATCC 36537 (Bennett and Goldblatt, 1973; Bhatnagar et a1, 1992) nor A. parasiticus C810 produce detectable levels of aflatoxins in liquid or solid growth media. Additionally, neither strain has been observed or reported to revert back to Afl*. Fungal strains were maintained on potato dextrose agar (PDA) or Czapek-Dox agar (CZA) that served as a defined 31 medium. Coconut agar 'medium (CAM) was used for screening fungal strains for aflatoxin accumulation by visualization of blue fluorescence under ultraviolet light (Davis et al, 1987) . Cultures were incubated at 30°C unless otherwise indicated. Chemicals. Chemicals, unless specifically referenced in the text were purchased from Sigma Chemical Company. Agarose was purchased from Bethesda Research Laboratories (Gaithersburg, MD) . PDA and CZA were purchased from Difco Laboratories (Detroit, MI). Development of a genetic transformation system Isolation of pyr mutant strains. Conidiospores (2 x 10’) from the versicolorin A accumulating strain A. parasiticus ATCC 36537 were resuspended in 1 ml of Tris-maleate buffer (50 mM Tris-maleate [pH 7.3], 35 mM NaNO3, mH KCl, 4 mm 14980,, 65 11M FeSO,) and treated with N—methyl-N'-nitro-N-nitrosoguanidine (NTG) at a concentration of 170 ng/ml for 1 hr 45 min at room temperature. Conidia were washed and resuspended in PD broth (2.4 mg/ml uridine) 32 and grown in stationary culture at 30° C for two hrs before inoculating onto PDA containing 5-f1uoro-orotic acid (FOA) (2 mg/ml) and high concentrations of uridine (2.4 mg/ml) . FOA is lethal to uridine prototrophs of A. parasiticus presumably due to the enzymatic synthesis of the toxic intermediate 5-fluoro- UMP (Boeke et a1, 1984). The enzymes orotate phosphoribosyl transferase and orotidine monophosphate decarboxylase are involved in the conversion of orotate to orotidine monophosphate (OMP) and OMP to uridine monophosphate (UMP) respectively (Figure 6). Mutations in either the pyrF (orotate phosphoribosyl transferase) or the pyrG (orotidine monophosphate decarboxylase) gene resulting in loss of enzymatic activity will confer resistance to FOA by preventing the production of 5-fluoro-UMP. Colonies resistant to FOA were further tested for stable uridine auxotrophy by serial replica-plating onto CZ agar with and without uridine. Enzymatic analysis of pyr mutants. Since FOA selects for either pyrF or pyrG mutant strains, enzyme analysis with radiolabelled substrates was used to determine the ability of auxotrophic isolates to convert orotic acid to OMP (pyrF) and the ability to convert OH? to UMP (pyrG) , thereby differentiating between the two mutant types (Figure 6). The protocol employed was a modification of that used by Diez et al (1987) . Conidia (1 x 107) from 33 Figure 6. Synthesis of uridine 5'-monophosphate. Phosphoribosyl transferase is encoded by the gene pyrF, while »OMP-decarboxylase is encoded by the gene pyrG. PRPP, phosphoribosyl pyrophophate; P-P, pyrophosphate; OMP, orotidine 5'-monophosphate; UMP, uridine 5'-monophosphate. 34 and min a 0 ......M. .000 40 film—n— o_o< o_._.OmO _ z: 35 uridine auxotrophs were inoculated into 100 ml of PD broth containing uridine and grown 48 hrs with vigorous shaking at 30° C. Mycelium was harvested by filtration through cheesecloth, frozen in liquid nitrogen, and ground to a fine powder in a precooled mortar and pestle. Powdered mycelium 'was resuspended in enzyme buffer (25 mM Tris-Cl [pH 8.0], 8 mM MgCl,, 1 mM 2-mercaptoethanol) at a concentration of l g mycelium/ml buffer. This suspension was centrifuged at 4° C in a microcentrifuge for 2 min at 16,000 x g. The supernatant was transferred to a new tube and centrifuged under the same conditions for 15 min. The supernatant was again removed and 8.5 pl aliquots were dispensed into two prechilled microcentrifuge tubes. The remaining supernatant was used to confirm that protein concentration was between 0.5 - 1.0 mg protein/ml by the Bradford technique (1976). The standard curve for the Bradford method was performed with BSA dissolved in enzyme buffer (see above) since both Tris-Cl and 2- mercaptoethanol can affect the results of the protein determination. [6-“CJ-OMP was prepared as described by Diez et a1 (1987) utilizing the enzymatic conversion of [6-“C1-orotate (ICN, 50 mCi/mmol) with commercial yeast orotidine monophosphate pyrophosphorylase (orotate phosphoribosyl transferase) and 5-phosphoribosyl-l-pyrophosphate (PRPP). 'To the first tube containing the crude cell free enzyme extract, 3.5 pl of prepared 0MP was added followed by 0.5 pl 0.2 M sodium arsenate. To the second tube, 2.5 pl [6-“C1-orotate 36 (1.5 mM), 25 mM Tris-Cl [pH 8.0], 8 mM MgCl,), and 1 pl of PRPP (5 mg/ml, made fresh in chilled H,O) was added. Both tubes were incubated for 2 hrs at 30° C followed by precipitation of proteins by the addition of 6 p1 of methanol and storing overnight at -20° C. Tubes were centrifuged 3 min at 16k x g to remove precipitated proteins and 5 pl of the supernatant was spotted onto a 20 cm x 20 cm polyethylene- imine-cellulose thin-layer-chromatography (TLC) plate. Visual (observed under long ‘wave ‘U.V. light) and. l‘C-labelled standards of orotic acid, OMP, and UMP were used as controls for determining the location and identity of each compound in enzyme reactions. Samples were separated using 0.75 M Tris-Cl [pH 8.0] as the solvent and the dried TLC plates were then analyzed by autoradiography by exposing the TLC plate to Kodak XARS film for 48 hrs at -70° C. Transformation of fungal protoplasts. Plasmid DNA used for transformation was purified by CsCl density gradient centrifugation (Maniatis et al, 1982) and resuspended in TE (10 mM Tris-Cl [pH 8.0], 1 mM EDTA). The polyethylene glycol transformation procedure described by Oakley at al (1987) was performed using the following modifications. Instead of using swollen conidia as suggested by Oakley et al, conidiospores were grown 12-14 hrs in YES (24 yeast extract, 6% sucrose, pH 5.5) and supplemented with 100 37 pg/ml uridine for auxotrophic mutants. Harvested mycelium was digested with Novozyme 234 and B-glucuronidase in one half diluted YES osmotically stabilized with 0.6 M KCl as performed by Oakley. However, Driselase was eliminated from the protoplasting solution and enzymatic incubation time was decreased from 5 hrs to 3 hrs. Protoplasts were separated from mycelium by passing the cell suspension (using gravity) through a 60 um nylon mesh filter (Nytex, Tetko Corporation, Switzerland) which allowed only protoplasts and small mycelial fragments to pass through. Protoplasts were then harvested, washed with 0.6 M KCl, 0.5 M‘ CaCl, and treated with polyethylene glycol and 1-10 pg of DNA as described by Oakley et a1 and spread directly on C2 agar containing 0.6 M KCl. Protoplast preparations were always divided into two equal aliquots, one treated with DNA and the other untreated to serve as a control. Cotransformations were occasionally used to transform A. parasiticus C810 to pyrG”. Under these conditions, the pyrG selectable marker was included on a separate plasmid (pBP28) from the vector containing the transforming DNA of interest. Plasmid pBP28 was constructed by replacing the 0.28 kb figmHl/Sgll fragment from p8R322 with a 2.8 kb MllmI DNA restriction fragment containing the pyrG gene from pPG3J. The vector containing DNA of interest was added in a two fold molar ratio over pBP28 for a total of #3 pg DNA in the transformation mixture. 38 Isolation and analysis of genomic DNA from fungal cells. Genomic DNA was isolated from fungal mycelium using a phenol/chloroform protocol for mammalian DNA isolation (Ausubel et al, 1987) IRestriction enzymes were purchased from Bethesda Research Laboratories, Inc (Gaithersburg, Md) and used according to the supplier's instructions. Southern hybridization analysis (Maniatis et a1, 1982) was performed with 32P labelled DNA probes generated by the random primer procedure (Random Primed DNA Labeling Kit, Boehringer Mannheim Biochemicals). Nitrocellulose filters (Schleicher & Schuell, Inc, Keene, NH) were hybridized to labelled probes overnight at 37° C in 6x SSC (1x SSC is: 0.15 M sodium chloride, 0.015 M sodium citrate)-5x Denhardt's solution (1% Ficoll, 1% polyvinylpyrrolidone, 1% BSA)-40% formamide-0.1% sodium dodecyl sulfate (808)-5 mM EDTA-100 pg denatured salmon sperm DNA/ml with probe (10‘ cpm/ml). Following hybridization, filters were washed twice in 2x SSC-0.1% SDS at 37°C for 20 min and then in 0.2x SSC-0.1% SDS at 65° for 20 min. Filters were exposed to Kodak XARS film for 20 hrs at -70° C with an intensifying screen (Lightning Plus-Dupont). 39 Isolation of the ver-l gene Preparation of cosmid library. The cosmid vector p825 (Figure 7) was constructed by ligation of the 2.8 kb BamHl/Sall pyrG containing restriction fragment from plasmid pPG3J, into the HindIII restriction endonuclease site of the cosmid vector pHC79 (Stratagene, LaJolla, CA). Ligation of the blunt ended pyrG fragment in p825 regenerated the fiinQIII site since end filling of the 8311 and.gingIII restriction endonuclease sites completed the recognition sequence for MIII. The ability to transform A. parasiticus C810 to prototrophy at efficiencies comparable to that obtained with pPG3J confirmed the presence of the pyrG gene in p825. The cosmid genomic DNA library was prepared according to the procedures of Maniatis et a1 (1982). DNA from the wild type Afl“ strain, A. parasiticus NRRL 5862, was used for library preparation. This DNA was partially digested by the restriction endonuclease ‘Sau3A and size fractionated by sucrose density gradient centrifugation. Fractions containing DNA fragments between 30 to 40va in size were ligated into the alkaline phosphatase-treated BamHl site of p825. Concatemers were then packaged into Lambda phage particles using a commercial packaging extract (Promega Corp., Madison, Wisconsin). 40 Figure 7. Cosmid vector p825 used in the preparation of an A. parasiticus NRRL 5862 cosmid genomic DNA library. This Vector ‘was constructed by blunt end ligation of a 2.8 figmHl/SQLI pyrG containing fragment from plasmid pPG3J (50) into the 1111351111 restriction endonuclease site of cosmid pHC79 (Stratagene, LaJolla, CA). DNA from a toxigenic A. parasiticus strain was digested with m3A, size fractionated, and then ligated into the alkaline phosphatase treated figmHl site of p825. 41 will 0.00 Clal 0.03 Clal 0.38 I Pstl 8.24 —“—‘_ E5 : amp-r y ori pBZS 9.23 Kb Ndel 6.93 411, 3qu 6.30 "Ha, {lm‘i‘lil ,.HaelimmwwMW BQII 4.87 Aval 4.83 coRV 0.59 Clal 0.98 ' tl 1.03 coRV 1.14 pyrG - val 1.82 val 2.20 ' tl 2.28 Hindlll 2.83 EcoRV 2.99 BamH1 3.18 Sphl 3.37 Sall 3.45 Nrul 3.78 Aval 4.23 42 The DNA library was completed by transfecting E. coli LE392 with recombinant phage particles and transferring these cells to Luria-Bertani (LB) agar medium containing ampicillin (50 pg/ml). Overnight incubation at 37° C yielded approximately 870 colony forming units per petri plate (11x150 mm) . Cells were harvested from ten plates and the suspension was used for cosmid isolation by alkaline lysis and CsCl purification (Maniatis et al, 1982). Preparation of cDNA library. A. parasiticus NRRL 5862 was inoculated into a glucose mineral salts medium (Adye and Mateles, 1964) and grown for a period of 48 hrs. RNA, isolated from the mycelium using a hot phenol purification protocol (Ausubel et al, 1987), was then used to prepare a cDNA library. This library was constructed by Stratagene (LaJolla, CA) using their Uni-Zap XR vector. Screening of the cosmid genomic DNA library by genetic complementation of A. parasiticus C810. DNA from the combined cosmid DNA library was used in transformation of A. parasiticus C810. Protoplasts were plated directly onto CAM supplemented with 20 percent sucrose as an osmotic stabilizer. CAM does not contain sufficient 43 uridine to support the growth of untransformed protoplasts. Only protoplasts receiving the pyrG marker, contained on the cosmid, were capable of growth on CAM. Transformant colonies displaying "blue fluorescence" under longHwave‘U.V. light were isolated and reinoculated onto fresh PDA for further analysis and verification of aflatoxin production. Confirmation of aflatoxin production by transformant clones. Conidia (1 x 105 from transformant colonies were inoculated into 20 ml of YES (2 % yeast extract, 20 % sucrose [pH 5.5]) medium contained in a 50 ml erlenmeyer flask. Each flask also contained four, 4.5 mm glass beads to aid in aeration of the culture and to facilitate dispersed mycelial growth (Heathcote and Hibbert, 1978; Reynolds and Pestka, 1991). Cultures were then grown for 3 days at 28° C in an orbital shaker (150 rpm). After incubation, cultures were filtered through miracloth (Behring Diagnostics, LaJolla, CA) and the filtrate was used for identification of aflatoxins. Only the filtrate was analyzed for aflatoxins since both aflatoxin 81 and.G1 are known to pass efficiently through the mycelial walls into the surrounding medium (Lin et al, 1992) . Filtrates from cultures of A. parasiticus C810 and A. parasiticus NRRL 5862 were included as aflatoxin production negative and positive controls respectively. Filtrates were tested using thin layer 44 chromatography (TLC) and enzyme linked immunoabsorbent assay (ELISA). TLC analyses were performed on activated 10 x 10 cm high performance silica TLC plates in an equilibrated chamber using chloroform/methanol (97:3) as a solvent system. ‘Versicolorin A (kindly supplied by L. Lee and D. Bhatnagar, USDA Agriculture Research Service, New Orleans, L.A.) and aflatoxin 81, 82, G1, and G2 were resolved on each plate as reference standards. Direct competitive ELISA analyses were performed according to Pestka (1988) using aflatoxin 81 monoclonal antibodies (kindly provided by J. Pestka, Michigan State University) and aflatoxin Bl-horseradish peroxidase conjugate, prepared using the methods of Chu et al (1977). Production of sclerotia. Transformant isolates were tested for their ability to produce sclerotia. A. parasiticus NRRL 5862 and A. parasiticus ATCC 36537 were included for comparisons. Approximately 103 previously frozen conidia (in 15% glycerol) of each strain were inoculated on the center of a 100x15 mm petri plate prepared with 25 ml of PDA.medium. Cultures were incubated at 30°C and 37°C for’a period of 10 days. Sclerotia were harvested according to Cotty (1988), counted, and then dried at 50°C for 48 hrs. Sclerotial diameters were measured using the aid of the.Java‘Video.Analysis System (Jandal Corp., 45 Corte Madera, CA). Between 32-100 sclerotia were chosen at random from each plate and measured from top to bottom near the estimated center of the vertical plane. Marker rescue of amp'gene from an aflatoxin producing (Afl‘) transformant. Four different restriction endonucleases (8311, 82111, figlII, and ngI) were used to digest genomic DNA from one A. parasiticus C810 Afl+ transformant clone. These enzymes were chosen because they out once within the cosmid vector p825, but do not cut within the pyrG or the amp‘ gene. The strategy was designed to allow recovery of fragments that contained the pyrG gene, the amp’ genes, and presumably a piece of the original A. parasiticus insert DNA because the second restriction endonuclease site must be situated somewhere within the flanking DNA. Each restriction enzyme was used separately to digest 50 pg of DNA from the Afl+ transformant. DNA was separated on a 1.2 % agarose gel. Regions of the gel corresponding to the desired fragments containing the amp' gene were removed and used for DNA isolation with Prep-A-Gene DNA isolation matrix (Biorad, Richmond, CA). Isolated DNA was diluted to 2 ng DNA/ul prior to T4 DNA ligase treatment to minimize intermolecular ligation. Circularized fragments were transformed into E. coli DH5c which was then grown on LB agar 46 medium containing ampicillin (50 pg/ml). Isolation and analysis of RNA from fungal cells. Conidia (1 x 10") of A. parasiticus were inoculated into 100 ml YES broth and grown at 28° C in an orbital shaker (150 rpm) for 3 days. The resulting mycelium was filtered through mira-cloth and then quickly frozen in liquid nitrogen. Ribonucleic acid was then isolated using a hot phenol purification method (Ausebel et al, 1987 ) . Polyadenylated RNA was isolated by affinity column chromatography using an oligo- dT matrix (Ausebel at al, 1987). Northern analysis of RNA samples was performed as described by Maniatis et a1 (1982). Analysis of RNA transcripts from the var-1 gene. The orientation of the ver-l transcript was first determined by Northern hybridization analysis with strand specific riboprobes made from plasmid pBSV1 (Figure 8), in which a 0.7 kb EggRl/BamHl ver-l fragment was ligated into the plasmid pBluescriptII KS- (Stratagene, LaJolla, CA). Bluescript plasmids contain T7 and T3 phage promoters on either side of the polylinker region that allow transcription of DNA inserts in both directions. Plasmid pBSV1 was linearized with restriction endonuclease enzymes m1 and $311 47 Figure 8. Restriction endonuclease map and sequencing strategy of the 1.8 kb EggRllflingIII fragment. The shaded . arrow indicates the direction of transcription for the Aspergillus parasiticus ver-l gene. Regions of DNA cloned into pBluescriptII plasmids for the construction of pBSV1, pBSV2, and p88V3 are illustrated directly beneath the corresponding area of the restriction map. Orientation of T7 and T3 primer regions, contained on pBluescriptII plasmids, are included with arrow designating the direction and extent of sequencing. Each arrow may represent data from more than one set of sequencing reactions. Oligonucleotide primers (P1, P2, and P3) were synthesized and used to sequence plasmid pBSV4 which contains the 1.8 kb EggRl/SQII fragment cloned into pBluescriptII KS-. E, MR1; 8, W1; N, 8531; A, 5231; S. 5311; H. HindIII- 48 ¢>mma «Ia ell E I o>mma 21 E Olllv ~>mma : ..llle 2. I Emma t . e 2. ._u _ _ _ _ _ Id < Z 2 m m Huh?» av. Nd 49 (located in the polylinker) to generate transcripts using T3 and T7, respectively. Radiolabelled riboprobes were made according to manufacturer's instructions with e-nP-rCTP. The 5' and 3' ends of the var-1 transcript were localized using a ribonuclease protection assay (Ausebel et al, 1987). Riboprobes for 5' end analysis were transcribed with T7 polymerase using gall linearized plasmid pBSV1 (Figure 8). Riboprobes for 3' end analysis were prepared using T7 polymerase with EcoRI-linearized plasmid pBSV3 (Figure 8) in which a 0.4 kb MI/mngIII var-1 fragment was blunt end ligated into the SmaI site of pBluescriptII SK-. Riboprobes were hybridized at 55° C to mRNA from A. parasiticus NRRL 5862 or tRNA (control). Samples were treated with RNase T1 and RNase A and separated by electrophoresis on a 5% polyacrylamide gel under denaturing conditions. Acrylamide gels were dried under vacuum and subjected to autoradiography using Kodak XARS film. Sequence analysis. Nucleotide sequence analysis was performed using the dideoxy chain termination method (Sanger et al, 1977)" employing Sequenase II (U88, Cleveland, OH) according to the manufacturer's instructions. The DNA inserts in plasmids pBSVl, pBSV2, and pBSV3, were sequenced on both strands using the T7 and T3 primers supplied in the kit (Figure 8). Plasmid 50 pBSV2 was constructed by ligation of a blunt end 0.6 kb BamHl/AvaI internal var-1 fragment into the SmaI site of pBluescriptII SK-. From this initial sequence information, oligonucleotide primers (15 nucleotides) were synthesized (MSU, Molecular Structure Facility) and used to reconfirm the sequence and to remove ambiguities. Plasmid pBSV4 (Figure 8) , which contains a 1.8 kb EggRl/SalI fragment, served as a template strand for sequencing with the synthesized oligonucleotide primers. Sequence analysis of the var-1 cDNA fragment was performed using the TAQuence sequencing kit (USB, Cleveland, OH) which utilized the thermostable ATaq DNA polymerase. The cDNA fragment was isolated by in situ colony hybridization of the cDNA library using the plasmid pBSV2 which contains an internal portion of the var-1 gene. Computer analysis of sequence data. Computer analyses of nucleotide data were performed using the Wisconsin Genetics Computer Group (GCG) package. The location of open reading frames, a translation start codon, and introns were predicted using the GCG software programs Frames, Testcode, and codon preference. Codon usage files were constructed using data from 45 different A. nidulans genes reported by LLoyd and Sharp (1991) . Amino acid comparisons were made using a window size of 30 and a stringency of 15. 51 Analysis of RNA expression for the ver-l and nor-l gene Batch fermentation analysis. Frozen (-80° C in 15% glycerol) conidia (2 x 10‘) of A. parasiticus NRRL 5862 were inoculated into 100 ml of defined Adye and Mateles (1964) minimal medium (AM) contained in a 250 ml silanized erlenmeyer flask. Five glass beads (3-4 mm dia) were included in each flask to aid in aeration of the culture and to facilitate dispersed mycelial growth (Reynolds and Pestka, 1991). Growth was visibly more uniform among duplicate flasks containing glass beads since they prevented the inconsistent aggregation or clumping of the mycelium.that is typically observed with Aspergillus shake cultures (without beads). All flasks were incubated in the absence of light at 29° C in an orbital shaker (220 rpm). Triplicate flasks were removed at each time point and analyzed individually for mycelial dry weight, pH of medium, and aflatoxin concentration in filtrate. The average of three values and the standard error were then calculated for measurements at each time point. One additional flask for each time point was included for RNA isolation. 52 Nutritional shift assay. A modified nutritional shift protocol (Buchanan and Lewis, 1984; Buchanan et al, 1987) was employed to determine when the nor-1 and var-1 RNA transcripts accumulate relative to a shift from non-aflatoxin supporting conditions to those which support aflatoxin biosynthesis. Three silanized Fernbach flasks containing 500 ml of peptone mineral salts (PMS) medium (Buchanan et al, 1987) were each inoculated with 2.5 x 107 A. parasiticus NRRL 5862 conidia (from a frozen stock) and grown for 65 hrs at 29° C in an orbital shaker (200 rpm). PMS medium is similar to AM medium (Adye and Mateles, 1964) except that glucose is replaced by peptone which serves as the sole carbon source. This medium does not support or induce aflatoxin biosynthesis (Buchanan and Lewis, 1984) . The mycelium from all three flasks was combined, harvested by filtration through cheese cloth and distributed [5 g (wet ‘weight) mycelium/flask] to 250 ml silanized flasks containing 30 mls of either PMS or glucose mineral salts (GMS) medium (Buchanan et al, 1987) . GMS, which is capable of inducing and supporting aflatoxin biosynthesis is identical to PMS except that peptone (6%) is replaced by glucose (6%). Incubation of cultures after the shift was continued up to 48 hrs under the same conditions. One PMS and one GMS containing flask were removed at each time point so that mycelium could be harvested and used for RNA extraction. Filtrates from each time point were saved for quantitative analysis of aflatoxin B“ 53 Analysis of mycelial dry weight, aflatoxin concentration, and RNA for batch fermentation and nutritional shift analysis. Fungal cultures were removed from the incubator/shaker at specified times and the mycelium harvested by filtration through miracloth (Behring Diagnostics, LaJolla, CA). The collected mycelium from each culture flask was dried to completion at 70° C prior to weighing. A sample of each filtrate ‘was tested for pH and the remainder used for determination of aflatoxin production. Only the filtrate was tested since both aflatoxin B, and G, are known to efficiently pass through the mycelial wall at 29° C (Lin et al, 1980). Aflatoxin B, was analyzed using direct competitive enzyme linked immunoabsorbent assay (ELISA) according to Pestka (1988). Total RNA (nuclear and cytoplasmic) was isolated and purified using a hot phenol protocol (Ausebel et al, 1987). Northern analysis of RNA samples was performed as described by Maniatis (1982). Approximately 7 pg of total RNA per sample was separated by electrophoresis in a 1.2% denaturing formaldehyde agarose gel and then transfered by capillary action to a Nytran membrane (Schleicher and Scheull, INC, Keene, NH). A 0.6 kb AlaI/fiamHI restriction endonuclease fragment of an internal portion of the A. parasiticus ver-l gene (Skory et al, 1992) (Genbank accession # M91369), a 1.5 kb figlII/Qlal fragment containing the A. parasiticus nor-1 gene (Chang et al, 1992), and a 2.8 kb BamHl/Sall fragment 54 containing the A. parasiticus pyrG gene (Skory et al, 1990) were used as DNA probes for Northern analysis. RESULTS/DISCUSSION Isolation of the pyrG gene from A. parasiticus ATCC 36537. Transformation of A. parasiticus ATCC 36537 pyrG mutants using the heterologous gene from A. nidulans FGSC4 contained on circular pPL6 (Oakley, 1987) was unsuccessful on several attempts and thus prompted the isolation of the homologous pyrG gene. Southern analysis of EQQRI digested genomic DNA from A. parasiticus NRRL 5862 exhibited a single band when hybridized to a 1.5 kb 1111151111 restriction endonuclease fragment containing the pyrG gene from A. nidulans FGSC4 (Oakley et al, 1987). The pyrG gene (Skory et al, 1990) from A. parasiticus NRRL 5862 was then isolated by in situ plaque hybridization (Maniatis et al, 1987) of a lambda genomic DNA library (Horng et al, 1989) using the described pyrG gene from A. nidulans FGSC4 as a probe. A total of four isolates were obtained, at a frequency of 2 x 10‘ clones/Lambda plaque, that contained DNA inserts which hybridized strongly to the A. nidulans pyrG probe (Figure 9A) . An 8.1 kb MI fragment which hybridized to the A. nidulans probe was purified from one representative phage isolate (lambda pyr3) which appeared 55 Figure 9. (A) Restriction endonuclease analysis of inserted DNA fragments from four clones of an A. parasiticus NRRL 5862 Lambda genomic DNA library which hybridized to the A. nidulans FGSC4 pyrGr'probe. The 2.8 kb Sall/BamHI fragment represents the region of probe binding. (8) Restriction endonuclease analysis of plasmid pPG3J (10.8 Kb) which was used for transformation of uridine auxotrophs. The 8.1 kb Sagl fragment from Lambda isolate pyr3 was ligated into the 8391 (SstI) restriction endonuclease site of the polylinker region of pUC19. Only those enzymes listed in the restriction endonuclease map have tested. Additional sites in the in the polylinker region still remain. The darkened box represents the 2.8 Kb region from A. parasiticus which hybridized to the pyrG gene from A. nidulans FGSC4 while the open box is flanking DNA. Restriction endonuclease sites are as follows: SC. 5.8521; 3. 8.811131; 3. 5.8.11; Sa, £81133- 56 a. unmapped left lambda DNA arm 8c 8 S 8 Sa 1 kb pyrl E pyr4 ' estate — SC 3 S S 8 SC pyr3 1 1 1 1 L l.__ SC 3 S S 3 SB pyrS ‘ ‘ ' 1 ' ‘eeee I I 2.8 kb EcoRl SacI SalI BamHI “a pPGBJ 10.8 Kb 57 to contain the hybridizing region within the middle region of the A. parasiticus DNA insert. This .8353 (Sal) fragment was then subcloned into pUC19 to facilitate restriction endonuclease analysis. The resultant 10.8 kb plasmid was designated pPG3J (Figure 98). All phage isolates contained overlapping DNA fragments with the region hybridizing to the pyrG probe located on a common 2.8 Kb gall /B_amHI fragment (Figure 9A). Isolation of uridine auxotrophs from A. parasiticus ATCC 36537. NTG mutagenesis of A. parasiticus ATCC 36537 resulted in a 66% survival of conidiospores compared to non-mutagenized conidiospores when inoculated onto PDA. Inoculation of 1.3 x 10' viable mutagenized spores onto PDA supplemented with FOA and uridine resulted in FOA resistant (FOA’) colonies arising at a frequency of 3.5 x 10” colony forming units/viable conidiospore. Twenty-two percent of these FOA' isolates were determined to be stable uridine auxotrophs after serial inoculation onto CZ medium supplemented with uridine. All ten strains utilized in further analysis were morphologically indistinguishable from the original parent strain A. parasiticus ATCC 36537 and showed no detectable blue fluorescence when grown on coconut agar medium suggesting little or no aflatoxin production. 58 Enzymatic analysis of pyr mutants. Separation of OMP, orotate, and UMP on TLC plates yielded average Rf values of 0.4, 0.6, and 0.7 respectively. In cell free extracts of three A. parasiticus ATCC 36537 pyr isolates, C82, C83, and C810, OMP was converted to UMP in trace amounts suggesting that orotidine monophosphate decarboxylase activity was greatly impaired (Figure 10, panel A). Furthermore, [“C]-OMP accumulated when cell free extracts of A. parasiticus C82, C83, and C810 were incubated with [“C]- orotate plus PRPP indicating complete conversion of orotate to OMP with subsequent accumulation of OMP indicating active orotate phosphoribosyl transferase activity (data not shown). A. parasiticus ATCC 36537 which served as a positive control for functional pyrF and pyrG enzymatic activity was able to convert both [“CJ-OMP and [“CJ-orotate to [“CJ-UMP (data not shown). A. nidulans A722 which lacks orotidine monophosphate decarboxylase activity was used as a negative control. This strain did not accumulate [“C]-UMP when incubated with [“C]- OMP (data not shown). The remaining seven of the ten pyr mutants (CSl-C810) tested, converted ["CJ-OMP to [“C]-UMP at levels comparable to that in the A. parasiticus ATCC 36537 control suggesting a functional orotidine monophosphate decarboxylase enzyme (data not shown). The results strongly suggested that A. parasiticus uridine auxotrophs C82, C83, and C810 were pyrG mutants deficient in orotidine monophosphate decarboxylase activity. 59 Figure 10. Restoration of orotidine-5'-monophosphate activity in a pyrG mutant after transformation. Cell free extracts from A. parasiticus were assayed for the conversion of orotidine-5 ' -monophosphate (OMP) to uridine-5 ' - monophosphate (UMP). Extracts of .A. parasiticus C810 [panel A (uridine auxotrophic :mutant, not ‘transformed)] and .A. parasiticus UTS [panel 8(C810 transformed with pPG3J)] were incubated with [6-“C]-OMP and separated by thin layer chromatography (TLC). The TLC plate was then exposed 48 hrs to X-ray film. Oro, orotate. 60 61 Transformation of pyrG mutants with plasmid pP63J. Protoplast preparations of A. parasiticus typically yielded 1-3 x 10‘ protoplasts from 10' conidia with viability being approximately 25% when cells were grown under nonselective conditions (media supplemented with uridine). Transformation of a total of 10‘ viable protoplasts of A. parasiticus ATCC 36537 pyrG isolates C82 or C810 with 2 pg of pPG3J resulted in 20-50 stable transformants/pg DNA. Control protoplasts not treated with DNA, showed no evidence of germination on C2 (KCl) agar as determined by microscopic examination. Both circular and EQQRI linearized pPG3J were tested and yielded approximately equal efficiencies. However, linearized plasmids resulted in significantly fewer abortive transformants (colonies which initially grow on selective medium but are unable to do so upon further subculturing) . Transformant colonies typically appeared within 2 days but became readily distinguishable from abortive transformants at day 4 by their larger diameter (0.5-1.5 cm compared to abortive colonies which were typically less than 0.2 cm) and abundant conidiation. Transformant colonies transferred to C2 agar medium demonstrated colony morphology similar to A. parasiticus ATCC 36537. Efficiencies obtained with the homologous pyrG gene were slightly higher than those of Woloshuk g; a]. (1989) in which a heterologous pyrG transformation system in 5. mm was developed. 62 Southern analysis of prototrophic transformants from plasmid pPG3J. Southern hybridization analysis of genomic DNA purified from several transformant clones of A. parasiticus C810 suggested that both pyrG and pUC19 plasmid sequences integrated into the genome (Figure 11) . Transformant isolates 1-5 were transformed with linear pPG3J while isolates 6-10 were transformed with circular pPG3J. A single band was observed in MR1 digested genomic DNA from non-transformed A. parasiticus C810 when hybridized with the 2.8 kb SallifigmHl pyrG fragment. No hybridizing fragment was seen in the same DNA when probed with pUC19 confirming the lack of sequence similarity with this plasmid. Hybridization patterns for transformant isolates 4 and 8 were the same as the untransformed cells when probed with pyrG and did not show any hybridization to pUC19 sequences indicating that a double cross-over recombination event possibly occurred between the pyrG gene on the plasmid and the pyrG on the chromosome leading to replacement of the homologous DNA. Protoplasts not treated with DNA did not grow on CZ(KC1) medium implying that clones 4 and 8 were not revertants at the pyrG locus. It was not possible to conclusively determine from the data whether gene replacement also occurred in addition to integration of the plasmid sequences in the remaining transformant clones. A single cross-over and integration event of circular pPG3J into the pyrG locus was the most 63 Figure 11. Southern hybridization analysis of ten A. parasiticus pyrG transformants. Genomic DNA from untransformed A. parasiticus C810 (lane c), five isolates transformed.with linearized EQQRI pPG3J (lanes 1-5), and five additional isolates transformed with circular pPG3J (lanes 6- 10) were digested with EQQRI and electrophoresed in a 0.8% agarose gel (8 pg DNA per lane). DNA was blotted onto nitrocellulose and probed with 32P labelled (A) 2.8 kb Sall/Bgmfil pyrG fragment and (B) pUC19. 64 65 likely event for transformants 9 and 10. Such an event would result in two different MR1 fragments of approximately 23 kb and 11 kb each containing pyrG sequences, but only the larger of two would contain pUC19 sequences. The Southern data for the remaining transformants suggested integration of all, or part of the plasmid pPG3J into a site other than the pyrG locus. Additional hybridizing bands observed in genomic DNA from transformant isolates 1,2,3,5,6, and 7 when probed with the pyrG fragment were also noted in the same apparent locations when hybridized to pUC19, indicating that both pUC19 and pyrG sequences integrated into the genome. It should be mentioned that transformant isolates 3 and 5 may actually have resulted from integration of pPG3J sequences into the pyrG locus resulting in a larger MR1 band, although this is difficult to conclude since the intensity and size of this band prevents accurate interpretation. The multiple bands observed with transformant isolate 2 when probed with pUC19 may have resulted from integration and rearrangement of the pUC19 vector occurring independently of that from the pyrG DNA” The ‘magnified intensities of the additional 'pyrG hybridizing bands for transformant isolates 1,2,3,5, and 7 might suggest. multiple copies of the ‘plasmid. have been integrated since equal quantities of genomic DNA were added in each lane. 66 Enzymatic analysis of pyrG transformants. Enzymatic analysis with [“C]-OMP and [“CJ-orotate substrates was performed on cell extracts from five randomly isolated transformant colonies of A. parasiticus C810. All of the transformants tested were able to completely convert OMP to UMP and none accumulated OMP when incubated with orotic acid which demonstrated restored orotidine monophosphate decarboxylase activity (Figure 10, panel 8). In this study, enzymatic analysis was performed on cell free extracts from A. parasiticus pyr mutants for two reasons. First, it allowed differentiation between pyrG and pyrF mutant strains generated through mutagenesis and a positive selection protocol, thus eliminating the need to test all uridine auxotrophs isolated for the ability to be complemented by the cloned A. parasiticus pyrG gene. Secondly, it enabled confirmation of ablated orotidine monophosphate decarboxylase enzyme activity in A. parasiticus C810 and restored enzyme activity after complementation with the cloned pyrG gene. Although slight conversion of [“C]—OMP to [“C]-UMP seemed to be occurring in all three pyrG mutants C82, C83, and C810, suggesting the presence of leaky mutations, none of the mutant strains were able to grow on minimal medium. The reversion frequency in these mutants was shown to be less than 10", but it is possible that reversion did occur during cell growth in broth culture thus providing a small amount of functional orotidine monophosphate decarboxylase activity in the cell 67 free extracts. Results from enzymatic studies of transformants strongly suggest restored orotidine monophosphate decarboxylase activity as indicated by increased UMP formation from ["C] OMP and lack of OMP accumulation when incubated with [“C] orotate as compared to untransformed A. parasiticus C810 controls. Cosmid library preparation. DNA purified from A. parasiticus NRRL 5862 was confirmed by field inversion gel electrophoresis to be of high molecular weight (100-150 kb). The efficiency of packaging for the ligated concatamers was deduced by estimating the ability of phage particles to transfect E. coli LE392 (1.65 x 10’ transfectants per pg DNA). Cosmid preparations from twelve randomly isolated E. coli transfectant clones revealed the average size of the recombinant cosmid to be approximately 45 kb. Therefore, the average size of A. parasiticus DNA inserted into the original cosmid p825 (9.2 kb) was estimated to be 36 kb. Approximately 2,200 clones of this size would provide a 95% probability of obtaining a region of interest if one assumes that the total genome size of A. parasiticus is similar to A. nidulans (2.6 x 10‘ kb) (Timberlake, 1990). 68 Transformation of A. parasiticus C810 with "pooled" DNA library. A. parasiticus C810 was transformed with the combined cosmid DNA library at an efficiency of 29 pyrG+ transformants/pg DNA“ Approximately 400 colony forming units (cfu) were visible after 2 of days incubation at 32° C. No growth was observed on the control plates inoculated with protoplasts untreated with DNA. Almost all of the transformant colonies were observed to have the same morphology as the parent strain, A. parasiticus ATCC 36537, grown under similar conditions. One colony fluoresced blue under long wave U.V. light on the fourth day of incubation. Reinoculation of this putative Afl+ isolate onto CAM demonstrated that it still retained its ability to produce a blue fluorescing compound which was released into the surrounding medium. Additionally, it no longer accumulated significant quantities of the bright yellow fluorescent pigment, as seen in the recipient strain, which contains versicolorin A as a major component (Lee et al, 1975). A green spored transformant was also isolated following transformation with the pooled library. When viewed from the bottom of the petri plate, this isolate appeared visibly identical in both mycelial color and growth patterns to A. ,parasiticus.ATTC 36537 after six days of growth (30° C) on PDA medium. When viewed from the top, its morphology was indistinguishable from A. parasiticus ATCC 36537, except that 69 spores appeared green instead of white. Microscopic examination revealed that the majority of conidia were actually yellow to yellow/green. Some areas of entirely green conidia, similar to wild type A. parasiticus NRRL 5862, were observed near the center of the colony. The interest in this particular transformant is the possibility of isolating one or more of the genes responsible for green pigmentation of A. parasiticus spores. It has been suggested that spore pigmentation and aflatoxin production are somehow associated, although there is no direct evidence to substantiate this notion. Each of the A. parasiticus mutants blocked in aflatoxin biosynthesis possess a spore color very unlike the parent strain. This isolate could prove useful in later research to study such a relationship. Marker rescue of amp’ gene. Southern analysis of DNA purified from the putative Afl+ transformant using the cosmid vector pHC79 as a probe suggested that 12.5 kb gall, 9.8 kb 83111, 11.0 kb gphl, and 7.2 kb Edgl restriction fragments contained the amp' and pyrG genes plus a piece of the original A. parasiticus DNA insert. Ampicillin resistant clones of E. coli DHSa containing plasmids made by circularization of each of these restriction endonuclease fragments were isolated. Restriction endonuclease analysis of each rescued plasmid with the original restriction 70 enzyme used for its isolation, confirmed that plasmid sizes were consistent with predictions based on Southern analysis of DNA from the Afl+ transformant (Figure 12) . Isolated plasmids were designated pVC1A (83111) , pVC18 (Niel) , pVC1C (gall) , and pVC1D (gphl) for the restriction endonuclease enzymes which were used in their recovery. All plasmids appeared to have overlapping restriction endonuclease sites and most of the DNA from cosmid p825 was apparently present except for a deletion of approximately 2 kb in the region of the Lambda cohesive site. It may be possible that only part of the cosmid integrated or that rearrangement of the cosmid.occurred.during transformation. Deletions are quite common in cosmids of this size (Gibson et al, 1987) and may have occurred during replication in E. coli. Transformation of A. parasiticus C810 with pVC1A-D. Plasmids pVC1A, pVC18, pVC1C, and pVC1D were used to transform A. parasiticus C810 to test for the presence of the pyrG and ver-l genes. Pyrimidine prototrophic transformants were obtained using all plasmids except pVC18. The percentage of Afl“ isolates in the total number of pyrG“ transformants were 29%, 20%, and, 18% for pVC1A, pVC1C, and pVC1D respectively. 'The fortuitous complementation of the var-l gene using the rescued fragments eliminated the necessity to conduct in situ colony hybridization of the 71 Figure 12. DNA fragments obtained from an A. parasiticus C810 aflatoxin producing transformant using marker rescue. The region corresponding to the cosmid vector p825 is represented at the bottom of the figure. Plasmid pVC18 is not shown. B, 83111; N, ml; 8a, gall; Sp, gpnl; Cos, cohesive site in Lambda. 72 soboo> Eamoo J I 0.3m as; moo ax o.—. |_lm.m .92 .n. m OW ...... 90>“. cw z..z .0 m b mm ...... 05%.. m a U) z. Z z- m- 50>“. 73 cosmid genomic library using the rescued fragments as probes. Analysis of A. parasiticus Afl+ transformants. Five clones transformed with pVC1A (designated P1-P5) which demonstrated blue fluorescence on CAM and five transformants also transformed with pVC1A (designated N1-N5) that lacked blue fluorescence were subcultured onto CAM for further analysis. Upon visual inspection of the cultures, Afl+ transformants (Pl-P5) were easily distinguishable from Afl' strains by the second day of incubation. All Afl‘ transformants except P2 accumulated little detectable yellow pigmentation in the mycelium, whereas Afl'isolates (N1-N5) accumulated abundant quantities of yellow pigment. This yellow pigment has been confirmed by both TLC and ELISA to contain predominantly VA (Reynolds and Pestka, 1991). Under long wave U.V. light, all of the Afl+ transformants produced a blue fluorescent pigment by the third day of incubation at 30° C; No blue fluorescent pigment was detectable in clones N1-N5. One Afl+ transformant, P2, produced a small quantity of a blue fluorescent compound and accumulated less yellow pigment than the.AfI'isolates. By using thin layer chromatography and direct competitive ELISA analysis, it was confirmed that. A” parasiticus P1, P3, P4, and P5 ‘were producing aflatoxin B, in amounts (approximately 20 pg (1 10) aflatoxin/ml filtrate) comparable to A. parasiticus NRRL 5862 74 grown under identical conditions. A. parasiticus P2 produced aflatoxin B, at an approximately 10 fold lower level. No aflatoxins were detected (1 ng/ml, minimum detectable level for monoclonal antibody employed) in filtrates from A. parasiticus N1, N2, N3, N4, N5, or A. parasiticus ATCC 36537. TLC analysis demonstrated VA was present in small amounts in the filtrates [from all of the non-aflatoxin producing strains, A. parasiticus ATCC 36537, and also from A. parasiticus P2. VA is not normally found in the filtrate since it does not traverse the cell wall efficiently (Lee et al, 1975). However, it is likely that some of this pigment is released into the broth medium when the integrity of the cell wall is lost during cell death or as cells are physically disrupted by the glass beads. Each of theAfl+ and.Afl'transformants were reinoculated onto PDA and grown for 10 days at 30°C and 37°C. Sclerotia, which are composed of a highly pigmented dense mass of hyphae, were abundant at 30° C in the wild type A. parasiticus NRRL 5862 and in all of the Afl+ transformants except A. parasiticus P2 which was not producing any of these dormant structures (Table II). No sclerotia were observed in any of the Afl'transformants or in the mutant strain A. parasiticus ATCC 36537 when grown on PDA under identical conditions. However, A. parasiticus ATCC 36537 did produce sclerotia at 37°C which is inhibitory to aflatoxin biosynthesis (and VA biosynthesis) (Niehaus, 1989); no yellow pigmentation of the mycelium was evident at this temperature. Transformants could 75 Table II. Sclerotial production of various A. parasiticus strains grown on PDA medium for a period of 10 days. 30' I sclerotia/ avg. dia I sclerotia] avg. dia. strain 10 cm? (pm)‘ 10 cm’ (pm) 91 130 317153 --' -- P2 0 --- -- -- P3 180 360156 -- ~- P4 7 332151 -- -- P5 120 329150 -- '- Nl-NS 0 --- -- - NRRL 5862 69 338155 '32 341155 ATCC 36537 0 --- 33 372154 'Between 32-100 sclerotia were choosen at random for each strain. 'Growth was negligible (colony diameter <0.5 cm) at 37°C for all pyrimidine transformants tested. 76 not be analyzed at 37°C because of lack of growth presumably because the pyrG gene used for transformation was not effectively expressed at this temperature. These intriguing observations strongly support previous hypotheses (Bennett et al, 1986; Cotty, 1988, 1989) that sclerotia development and aflatoxin biosynthesis in A. ,parasiticus are associated. However, it is not clear whether regulation of these two processes are somehow interrelated or if there may also be a direct functional relationship (i.e. aflatoxins or pathway intermediates affect sclerotia development). The appearance of sclerotia in A. parasiticus ATCC 36537 grown at 37°C (an inhibitory temperature for aflatoxin biosynthesis) might suggest that the restored sclerotia production is a result of the disappearance of versicolorin A. If VA is inhibitory to sclerotia production, than preventing VA accumulation by either increasing its conversion to aflatoxin B, or by suppressing its synthesis should restore sclerotia. Secondary metabolism in Streptomyces is also believed to be linked with.morphologica1 development (Horinouchi et al, 1986). This hypothesis was supported by the isolation of a pleiotropic gene (afsB) which positively regulates the biosynthesis of A-factor, the polyketide derived antibiotic actinorhodin, and undecylprodigiosinm However, our data suggest that the var-1 gene is a structural gene encoding a ketoreductase. It seems improbable that a gene product with enzymatic activity would also have direct regulatory functions. However, aflatoxin B, 77 or pathway intermediates which are produced in pyrG+ Afl+ A. parasiticus C810 transformants could have a direct or indirect effect on regulatory proteins involved in aflatoxin biosynthesis and/or sclerotia development. Southern analysis was performed on BamHl digested genomic DNA from A. parasiticus P1-P5, and C810 using the 2.6 kb Pan/gphl var-1 containing restriction endonuclease fragment as probe (data not shown). One additional hybridization band that did not appear in the A. parasiticus C810 control was observed in approximately the same location for A. parasiticus Pl-PS. The relative intensities of these bands suggested that only one or at most two additional copies of transforming ver- 1 DNA had integrated into the genome. Additional studies will be performed to determine if integration of ver-l at a specific locus is required for full complementation activity. Differences in the percentage of Afl“ isolates among the total number of transformants may be a function of the various var-1 containing fragments having preferred sites of integration which may or may not complement the var-1 mutation. Since selection of transformants was conducted based on a functional pyrG gene, it is probable that integration of the transforming plasmid into the chromosome may have resulted in nonfunctional copies of the var-1 gene and therefore less than 100 % restoration of aflatoxin biosynthesis among the total number of transformants. 78 Localization of mRNA transcripts encoded on the Afl" complementing DNA fragment. We were able to predict the approximate location of the region responsible for the ver-l complementation activity by identifying the smallest common restriction endonuclease fragment present in the var-1 complementing plasmids pVC1A, pVC1C, and pVC1D. Since we knew from control transformations that Afl+ complementing activity was not contained on the cosmid vector p825, we could determine that it must lie between the pyrG gene in the vector and the first gall restriction endonuclease site within the insert (Figure 12). Northern analysis was used to determine the size and location of transcription units contained on the Afl+ complementing DNA fragments (Figure 13) . Overlapping restriction fragments were used as DNA probes for hybridization to RNA purified from A. parasiticus NRRL 5862 or A. parasiticus ATTC 36537. No significant differences in hybridization patterns were noted between the two strains. Three different RNAs (0.6, 1.0, and 1.5 kb) were detected by Northern analyses using radiolabeled probes made from the 3.4 kb ngI/gglll restriction fragment. Based on hybridization patterns observed with the DNA probes, it was predicted that the 0.6 kb RNA was encoded primarily on the 0.5 Hindlll restriction fragment with a small portion of the transcription coding region extending into the 0.4 kb AxglfianIII fragment. The region encoding the 1.0 kb RNA was localized to the 1.8 kb 79 Figure 13. Summary of Northern analyses using probes made from DNA fragments subcloned from the 3.4 kb QLaJ/gglll region of pVC1A. Poly-adenylated RNA was used for hybridization to the labelled fragments. The approximate location of transcripts is represented by solid bars at the top of the diagram. The arrow designates direction of transcription. Asterisks on the ends of the bars reflect approximate locations of transcript ends. Shaded area beneath the 1.8 kb EggRl/fiinfilll shows the smallest fragment to functionally complement ver-l. ND, none detected. 80 o; o._. o: 09.0.0 m. To; 0. fled 9x me .o._. 66 j mew. 23 kb), A. nidulans SRRC 1079 (11.2 kb), and A. nidulans ATCC 32610 (11.2 kb) might suggest that only A. parasiticus contains two copies of ver-l like sequences. However, Southern hybridization analysis with additional restriction endonucleases would be required before this could be concluded since restrictions endonuclease maps most likely differ between these species. The same membrane was stripped and reprobed with the nor- 1 containing 1.5 EgoRl/Qlal fragment (Figure 25). A 5.8 kb DNA fragment hybridizing to the nor-1 was observed for all species except for A. nidulans, A. tamarii, and A. versicolor which did not have any detectable hybridization signals. Intensities were of approximately equal values for all samples that hybridized to the nor-1 gene. These data suggest that nucleotide sequences similar to the var-1 and nor-1 aflatoxin biosynthetic genes are not unique to A. parasiticus. Both genes were found to hybridize very strongly to A. sojae sp. which shares a high degree of overall DNA relatedness with A. parasiticus. Likewise, A. oryzae sp. which shares a great degree of DNA relatedness with A. flavus also hybridized to these genes (Kurtzman et a1, 1987; Klich and Pitt, 1988). Aflatoxin production in A. 118 Figure 25. Southern hybridization analysis of genomic DNA from various Aspergillus spp. digested with BamHl. Digested .DNA was electrophoresed in a 0.8% agarose gel (0.8 pg per lane), transferred to Nytran membranes (Schleicher and Schuell, INC, Keene, NH) and probed with a nor-1 containing 1.5 Kb EggRl/glal DNA fragment. 1) A. parasiticus NRRL 5862, 2) A. nidulans ATCC 26451, 3) A. nidulans SRRC 1076, 4) A. nidulans ATCC 32610, 5) A. flavus SRRC 1273, 6) A. tamarii SRRC 1244, 7) A. sojae SRRC 1120, 8) A. sojae ATCC 42251, 9) A. sojae ATCC 9362, 10) A. versicolor ATCC 26939, 11) A. oryzae ATCC 14895, 12) A. tamarii (var flavus) SRRC 2045, 13) A. parasiticus NRRL 5862. 119 120 oryzae and A. sojae has always been a concern because these strains are commonly used in food fermentations. However, extensive research with these fungi have failed to demonstrate their ability to produce aflatoxins (Barbesgaard et al, 1992) . A. nidulans sp. and A. versicolor sp. produce sterigmatocystin (Chung et al, 1989; Hajjar et al, 1990; Hamasaki et al, 1967) and thus might be expected to contain similar aflatoxin biosynthetic genes. However, we found that the var-1 only weakly. hybridized to the A. nidulans DNA (Figure 24) and that no hybridization could be detected with the nor-1 gene. Likewise, no hybridization signal was detected for A. versicolor when probed with the ver-l or the .nor-l genes (Figures 24 and 25). The lack of homology between these genes could suggest that the aflatoxin biosynthetic pathway has been present for a very long period of time. It is possible that the ability to produce sterigmatocystin was present in a common ancestor and evolved to eventually produce aflatoxin B, in the related species A. parasiticus, A. flavus, and A. nomius. Many of these evolved fungi retained their ability to produce sterigmatocystin while some such as A. oryzae, and A. sojae lost that ability. We would expect genetic changes in these aflatoxin biosynthetic genes since secondary metabolites are not essential for growth and therefore are not under immense selective pressure to retain the same genetic code. Organisms such as A. parasiticus and A. sojae have a high degree of DNA relatedness, suggesting that they have only recently evolved from a common ancestor, 121 might be expected to share much more homology between aflatoxin biosynthetic genes than for A. parasiticus and A. versicolor which are believed to have diverged much earlier. While these ideas are purely speculative, they do help to explain the commonness of sterigmatocystin production and the uniqueness of aflatoxin 8, production. Further analysis using additional aflatoxin biosynthetic genes should help identify the diversity of these genes and allow the proposal of divergent or possibly convergent evolutionary models. It will also be interesting to compare the RNA expression of aflatoxin biosynthetic genes found in both aflatoxin B, producers and nonproducers. Studies comparing the DNA relatedness and rRNA will also help define the genetic variation between related fungi. This will eventually allow the proposal of a workable model which helps to explain the genetic evolution of aflatoxin biosynthesis. CONCLUDING REMARKS The A. parasiticus transformation system based on restoration of uridine prototrophy (pyrG) was used successfully in isolating a gene (var-1) associated with aflatoxin biosynthesis. The data strongly suggest that this var-1 gene is associated with the conversion of VA - ST because it complements Afl' mutants which are unable to carry 122 out this biotransformation step. Since this complex conversion of an anthraquinone moiety to a xanthone structure is believed to occur through a number of enzymatic steps, it is probable that the ver-l polypeptide must cooperate with the products of other genes. Nucleotide sequence analysis proved to be extremely useful in predicting that the var-1 gene encodes a protein which has enzymatic activity associated with this bioconversion. The predicted var-1 protein displays significant conservation of size and amino acid sequence with reductases and dehydrogenases involved in modification of ring structures. Further support of this predicted enzymatic function is the hypothesis of a NADPH dependent reductase involved in the conversion of VA to ST (Bhatnagar’et al, 1992; Dutton, 1988). Following complementation with the var-1 gene in Afl' mutants, biosynthesis of aflatoxins 8, and G, were also restored in addition to aflatoxins B, and G, (Bhatnagar, personal communication). This suggests that the product of var-1 gene is also associated with the conversion of versicolorin B‘to dihydrosterigmatocystin (Lin and Anderson, 1992; Yabe et al, 1991). This proposal is consistent with theories that speculate that one set of enzymes carries out parallel functions on alternative aflatoxin intermediates, at a branch in the pathway, to allow for the synthesis of aflatoxins B, then G, at one branch of the pathway and B, then (B at the other branch (Figure 5) (Bhatnagar et a1, 1992). Although we cannot conclusively rule out the possibility that ver-l encodes a suppressor activity, the data strongly 123 suggest that the restoration of aflatoxin biosynthesis in ver- 1 transformants is due to a direct, complementation of a missing enzyme activity. 1) Nucleotide sequence analysis predicts a reductase function for the product of the var-1 gene. 2) No Afl+ transformants have ever been observed when A. parasiticus C810 is transformed with only the pyrG gene, suggesting that the var-1 mutation is quite stable (i.e. it is improbable that we are seeing reversion of a mutation). 3) Versicolorin A is converted to aflatoxin in Afl“ transformants at an efficiency comparable to the wild type strain. This efficiency would not be anticipated in transformants with only one additional copy of a gene encoding an enzyme which fortuitously converts VA to St. 4) RNA transcripts from the ver-l gene accumulate only during idiophase suggesting that the gene functions in secondary metabolism. The results of this study also imply an unknown but useful purpose for aflatoxin biosynthesis. Possible functions or benefits of aflatoxin production have been studied with ambiguous results (Heathcote, 1978) . To put such a tremendous amount of cellular energy into biosynthesis of a secondary metabolite without any function might be considered wasteful for the organism. However, it has been proposed that the lack of discernable biological activity for a secondary metabolite simply reflects that the correct screening assays have not been employed (Maplestone et al, 1992; Vinning, 1992). Finding sequences, in a number of related Aspergillus species, that hybridize very strongly to the aflatoxin biosynthetic 124 genes var-1 and .nor-1. might suggest ‘that aflatoxin (or sterigmatocystin) biosynthesis has been around for quite sometime. The linkage of aflatoxin biosynthetic genes also provides evidence that this secondary metabolite provides a beneficial function. Genes for secondary metabolism are often clustered together presumably because of selective pressure. If production of a secondary metabolite is useful to the organism, then it might be favorable if the genes were closer together to facilitate signaling between genes and to increase the probability that they would be passed on as a unit to further generations (Maplestone et al, 1992). It would make little sense that such a pathway would evolve and remain as a clustered unit if there were not selective pressure to keep it together. The information and tools obtained from this research project should allow for the continued success in the.study'of molecular control of aflatoxin biosynthesis. Preliminary studies suggest that the aflatoxin gene cluster is able to complement at least two additional mutants blocked at bioconversions steps. Additionally, it appears that disruption of aflatoxin biosynthetic genes is feasible and could possibly be used in field studies for the prevention of aflatoxin contamination by methods of competitive exclusion. Methods of aflatoxin regulation should soon begin to be understood as studies with gene expression continue. It is hoped that this research will eventually benefit efforts to facilitate the elimination of aflatoxin contamination. 125 LITERATURE CITED Adye, J. and R.I. Mateleles. 1964. Incorporation of labelled compounds into aflatoxin. Bichim. Biophys. Acta. 86:418-420. Anderson, J.A. and C.H. Chung. 1990. Conversion of versiconal acetate to versiconal and versicolorin C in extracts from Aspergillus parasiticus. Mycopathologia 110:31-35. Anderson, J.A., C.H. Chung, S.U. Cho. 1990. Versicolorin A hemiacetal, hydroxydihydrosterigmatocystin, and aflatoxin Ga reductase activity in extracts from Aspergillus parasiticus. Mycopathologia 111:39-45. Ausubel, F. M., R. Brent, R. 8. Kingston, D. D. Moore, J. G. Seidman, J. A. Smith, and K. Struhl' (eds). 1987. Current protocols in molecular biology, vol I. John Wiley 5 Sons. New York. Barbesgaard, P., H.P. Heldt-Hansen, and 8. Diderichsen. 1992. On the safety of Aspergillus oryzae: a review. Appl. Microbiol. Biotechnol. 36:569-572. Beck, J., S.Ripka, A. Siegner, E. Schiltz, and E. Schweizer. 1990. The multifunctional 6-methylsalicylic acid synthase gene of Penicillium patulum. Eur. J. Biochem. 192:487-498. Bennett, J.W., and L.A. Goldblatt. 1973. The isolation of mutants of Aspergillus flavus and Aspergillus parasiticus with altered aflatoxin producing ability. Sabouraudia 11:235-241. Bennett, J.W., L.S. Lee, and A.F. Cucullu. 1976. Effect of dichlorvos on aflatoxin and versicolorin A production in Aspergillus parasiticus. Bot. Gaz. 137(4):318-324. Bennett, J.W., L.S. Lee, S.M. Shoss, and.G.H. Boudreaux. 1980. Identification of averantin as an aflatoxin B, precursor: placement in the biosynthetic pathway. Appl. Environ. Microbiol. 39:835-839. Bennett, J.W., C.H. Vinnett, and W.R. Goynes,JR. 1980. Aspects of parasexual analysis in Aspergillus parasiticus. Can. J. Microbiol. 26:706-713. . Bennett, J.W., P.M. Leong, S. Kruger, and D. Keyes. 1986. Sclerotial and low aflatoxigenic morphological variants from haploid and diploid .Aspergillus ,parasiticus. Experientia 42:848-851. Bennett, J.W., and K.E. Papa. 1987. The aflatoxigenic Aspergillus spp. Adv. Plant Pathol. 6:263-280. 126 Berry, Colin. 1988. The pathology of mycotoxins. J. Path. 154:301-311. . Bhatnagar, D., A.H.J. Ullah, and T. 8. Cleveland. 1988. Purification and characterization of a methyltransferase from Aspergillus parasiticus SRRC 163 involved in aflatoxin biosynthetic pathway. Prep. Biochem. 18(3):321-349. Bhatnagar, D., T.E. Cleveland, and. 8.8. Lillehoj. 1989. Enzymes in aflatoxin B, biosynthes is: Strategies for identifying pertinent genes. Mycopathologia 107:75-83. 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 D.G. Kingston. 1991. Enzymological Evidence for Separate Pathways for Aflatoxin B, and B, Biosynthesis. Biochem. 30(17) ;4343-4350. Bhatnagar, D., K.C. Erlich, and T.E. Cleveland. 1992. Oxidation-reduction reactions in biosynthesis of secondary metabolites and their control by environmental factors, p. 255-286. In D. Bhatnagar, E.8. Lillehoj, and D.K. Arora (ed.) Handbook of applied mycology, vol V. Marcel Dekker, Inc, New York. Bhatnagar, R.K., S. Ahmad, K.G. Mukerji, and T.A. Venkitasubramanian. 1986. Nitrogen metabolism in Aspergillus parasiticus NRRL 3240 and A. flavus NRRL 3537 in relation to aflatoxin production. J. Appl. Bact. 60:203-211. Biollazz, M., G. Buchi, and G. Milne. 1970. The biosynthesis of aflatoxins. J. Am. Chem. Soc. 92:1035-1043. Boeke, J.D., F. LaCroute, and G.R. Fink. 1984. A positive selection for mutants lacking orotidine-5'-phosphate decarboxylase activity in yeast: 5-fluoro-orotic acid resistance. Mol. Gen. Genet. 197: 345-346. Bradford, M.M.. 1976. A rapid and sensitive method for the quantitation of:microgram.quantities of protein utilizing the principle of protein dye binding. Anal. Biochem. 72:248-254. Bradshaw, R.E., J.W. Bennett, and J.F. Peberdy. 1983. Parasexual analysis of Aspergillus ]parasiticus. J. Gen. Microbiol. 129:2117-2123. Buchanan, R.L., and D.F. Lewis. 1984. Regulation of aflatoxin biosynthesis: Effect of glucose on activities of various glycolytic enzymes. Appl. Environ. Micro. 48(2):306-310. 127 Buchanan, R.L., D. Federowicz, and H.G. Stahl. 1985. Activities of tricarboxylic acid cycle enzymes in an aflatoxigenic strain of.Aspergillus parasiticus after peptone to glucose carbon source shift. Trans. Br. Mycol. Soc. 84(2):267-275. Buchanan, R.L., S.B. Jones, W;V. Gerasimowicz, L.L. Zaika, H.G. Stahl, and L.A. Ocker. 1987. Regulation of aflatoxin biosynthesis: Assesment of the role of cellular energy status as a regulator of the induction of aflatoxin production. Appl. Environ. Micro. 53(6):1224-1231. Caballero, J.L., E. Martinez, F. Malpartida, and D.A. Hopwood. 1991. Organisation and function of the actVA region of the actinorhodin biosynthetic gene cluster of Streptomyces coelicolor. Mol. Gen. Genet. 230:401-412. Campbell, T.C., J. Chen, C. Liu, J. Li, and 8. Parpia. 1990. Nonassociation of aflatoxin with primary liver cancer in a cross-sectional ecological survey in the People's Republic of China. Cancer Res. 50:6882-6893. Campbell, T.C., J. Chen, T.A. Brun, C. Liu, and.C.A, Geissler. 1991. Aflatoxin and liver cancer. The Lancet. 335:1165. CAST (Council for Agicultural Science and Technology). 1989. Mycotoxins: Economic and health risks. Library of Congress Task force report. 29-31. Chang, P.K., C.S. Skory, and J.E. Linz. 1992. Cloning of a gene associated with aflatoxin 8, biosynthesis in Aspergillus parasiticus. Curr. Gen. 21(3): Chu, F.S., M.S. Hsia, and P.S. Sun. 1977. Preparation and characterization of aflatoxin 8,-1-(O-carboxymethyl) oxime. J. Chu, F.S. 1991. Mycotoxins:food contamination, mechanisms, carcinogenic potential, and preventative measures. Mutation Res. 259:291-306. Chung, D.H., M.M. Abouzied, and J.J. Pestka. 1989. Immunochemical assay applied to mycotoxin biosynthesis:ELlSA comparisons of sterigmatocystin production by Aspergillus versicolor and Aspergillus nidulans. Mycopath. 107:93-100. 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(1):38-42. Chuturgoon, A.A. and M.F. Dutton. 1991. The affinity purification and characterization of. a dehydrogenase from 128 Aspergillus parasiticus involved in aflatoxin 8, biosynthesis. Prep. Biochm. 21(2&3):125-140. Cleveland, T.E. 1987. Appearance of enzyme activities catalyzing conversion of sterigmatocystin to aflatoxin 8,, in late-growth phase Aspergillus parasiticus cultures. Appl. Environ. Microbiol. 53:1711-1713. Cleveland, T.E. 1989. Conversion of dihydro-O-methyl- sterigmatocystin to aflatoxin B, by Aspergillus parasiticus. Arch. Environ. Contam. Toxicol. 18:429-433. Cleveland, T.E. and D. Bhatnagar. 1990. Evidence for de novo synthesis of an aflatoxin pathway methyltransferase near the cessation of active growth and the onset of aflatoxin biosynthesis in Aspergillus parasiticus mycelia. Can. J. Micro. 36:1-5. Cole, R.J., J.W. Dorner, P.D. Blankenship, and 8. Horn. 1991. The use of biocompetitive agents for preharvest aflatoxin control in peanuts. 1991 Aflatoxin elimination workshop. Atlanta, Georgia. Cotty, P.J. 1988. Aflatoxin and sclerotial production by Aspergillus flavus: Influence of pH. Phytopath. 78(9). 1250- 1253. Cotty, P.J. 1989. Virulence and cultural characteristics of two Aspergillus flavus strains pathogenic on cotton. Phytopathology 79(7):808-814. Cotty, P.J. 1990. Effect of atoxigenic strains of Aspergillus flavus on aflatoxin contamination of developing cottonseed. Plant Disease 74(3):233-235. Cotty, P.J. and D.J. Daigle. 1991. Update on Aspergillus flavus populations and prevention of aflatoxin through intraspecific competition. 1991 Aflatoxin elimination workshop. Atlanta, Georgia. Coupland, K. and W.G. Niehaus. 1987. Effect of nitrogen supply, Zn“fi and salt concentration on Kojic acid and versicolorin biosyntheis by Aspergillus parasiticus. Exp. Mycol. 11:206-213. Cusumano, V., 6.8. Costa, and S. Seminara. 1990. Effect of aflatoxins on rat peritoneal macrophages. Appl. Environ. Micro. 56(11):3482-3484. 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. 129 Davis, N.K. and K.E. Chater. 1990. Spore colour in Streptomyces coelicolor A3 (2) involves the developmentally regulated synthesis of a compound biosynthetically related to polyketide antibiotics. Molec. Micro. 4(10):1679-1691. Delledonne, M., R. Porcari, and C. Fogher. 1990. Nucleotide sequence of the nodG gene of Azospirillum brasilense. Nucleic Acids Res. 18(21):6435. Diez, 8., E. Alvarez, J.M. Cantoral, J.L. Barredo, and J.F. Martin. 1987. Selection and characterization of pyrG mutants of Penicilliuml chrysogenum lacking' orotidine-5'-phosphate decarboxylase and complementation by the pyr4 gene of Neurospora crassa. Curr. Genet. 12:277-282. Donadio, 8., M.J. Staver, J.B. McAlpine, S.J. Swanson, and L. Katz. 1991. Modular organization of genes required for complex polyketide biosynthesis. Biochem. Biophy. Res. Comm. 47:1051- 1055. Donkersloot, J.A., R. I. Mateles, and 8.8. Yang. 1972. Isolation of averufin from a mutant of Aspergillus parasiticus impaired in aflatoxin biosynthesis. Biochem. Biophys. Res. Commun. 47:1051-1055. Dorner, J.W., R.J. Cole, and U.L. Diener. 1984. The relationship of Aspergillus flavus and Aspergillus parasiticus with reference to production of aflatoxins and cyclopiazonic acid. Mycopathologia 87:13-15. Drew, S.W., and A.L. Demain. 1977. Effects of primary metabolites on secondary metabolism . Ann . Rev. Microbiol . 3 1 : 34 3-56 . Dutton, M.F. 1988. Enzymes and aflatoxin biosynthesis. Microbiol. Rev. 53:274-95. Ehrlich, K., 1987. Effect on aflatoxin production of competetion between wild-type and mutant strains of Aspergillus parasiticus. Mycopathologia 97:93-96. Ellis, W.O., J.P. Smith, and.B.K. Simpson. 1991. Aflatoxins in food: Occurence, biosynthesis, effects on organisms, detection, and methods of control. Crit. Rev. Food Sci. Nutr. 30(3):403-439. Essigmann, J.M., R.G. Croy, R.A. Bennett, and G.N. Wogan. 1982. Metabolic activation of aflatoxin 8,: patterns of DNA adduct formation, removal, and. excretion in :relation ‘to carcinogenesis. Drug. Metab. Rev. 13:581-602. Feng, G.H., F.S. Chu, and T.J. Leonard. 1992. Molecular cloning of genes related to aflatoxin biosynthesis by 130 differential screening. Appl. Environ. Micro. 58(2):455-460. Fincham, J.R. 1989. Transformation in fungi. Microbiol. Rev. 53:148-170. Foreman, J.E. and W.G. Niehaus. 1985. 2n+2 induced cooperativity of mannitol-l-phosphate dehydrogenase from Aspergillus parasiticus. J. Biol. Chem. 260:10019-10022. Freifelder, D. 1987. Molecular Biology, 2nd ed. Jones and Bartlett Publishers, Inc. Boston:510. Fujimoto, Y., L.L. Hampton, L.D. Luo, P.J. Wirth, and 8.8. Thorgeirsson. 1992. Low frequency of p53 gene mutation in tumors induced by aflatoxin B, in nonhuman primates. Canc. Res. 52:1044-1046. Gibson, T.J., A.R. Coulson, J.E. Sulston, and P.R. Little. 1987. Lorist 2, a cosmid with transcriptional terminators insulating vector genes from interference by promoter within the insert: effect on DNA yeild and cloned insert frequency. Gene. 53:275-281. Gurr, S.J., S.E. Unkles, and J.R. Kinghorn. 1987. The structure and organization of nuclear genes of filamentous fungi. In:Kinghorn JR (ed) Gene structure in eukaryotic microbes. IRL Press, Oxford, 95-138. Hallam, S., F. Malpartida, and D. Hopwood. 1988. Nucleotide sequence, transcription and deduced. function of a gene involved in polyketide antibiotic synthesis in Streptomyces coelicolor. Gene 74:305-320. Hamasaki, T., Y. Hatsuda, N. Terashima, and M. Renbutsu. 1967. Studies on the metabolites of Aspergillus versicolor. Agr. Biol. Chem. 31(1):11-17. Hayes, A.W. 1980. Mycotoxins: A review of biological effects and their role in human diseases. Clin. Toxicol. 17:45-63. Heathcote, J.G. and J.R. Hibbert, Aflatoxins: Chemical and .biological aspects, (New“York: Elsevier'Scientific, 1978), pg. 1-201 e Heilmann, J.J., H.J. Maegert, and H.G. Gassen. 1988. Identification and isolation of glucose dehydrogenase genes of Bacillus megarterium M1286 and their expression in Escherichia coli. Eur. J. Biochem. 174:485-490. Hendricks, J.D., R.O. Sinnhuber, J.H. Wales, M.E. Stack, and D.P.H. Hsieh. 1980. Hepatocarcinogenicity of sterigmatocystin and versicolorin A to rainbow trout (Salmo gairdneri) embryos. JNCI 64(6):1503-1506. I 131 Hopwood, D.A. and D.H. Sherman. 1990. Molecular genetics of polyketides and its comparison to fatty acid biosynthesis. Ann. Rev. Gen. 24:37-66. Horng, J.S., J.E. Linz, and J.J. Pestka. 1989. Cloning and characterization of the trpC gene from an aflatoxigenic strain of Aspergillus parasiticus. App. Environ. Micro. 55:2561-2568. Horinouchi, 8., H. Suzuki, and T. Beppu. 1986. Nucleotide sequence of afSB, a pleiotropic gene involved in secondary metabolism in Streptomyces coelicolor A3(2) and Streptomyces lividans. J. Bact. 168(1):257-269. Hsieh, D.P.H., c.c. Wan, and 3.11. Billington. 1989. A versiconal hemiacetal acetate converting enzyme in aflatoxin biosynthesis. Mycopathologia 107:121-126. Jelinek, C.F., A.E. Pohland, and 6.8. Wood. 1989. Worldwide occurrence of mycotoxins in foods and feeds- an update. J. Assoc. Anal. Chem. 72:223-230. Kamps, M.P., 8.8. Taylor, and 8.M. Sefton. 1984. Direct evidence that oncogenic tyrosine kinases and cyclic AMP- dependent protein kinase have homologous ATP-binding sites. Nature 310(16):589-591. Keller, N., H.C. Dischinger, D. Bhatnagar, T.E. Cleveland, and A.H.J. Ullah. 1991. Purification of a second methyltransferase active in the aflatoxin biosynthetic pathway. FASEB J. A2461. Kitada, M., M. Taneda, K. Ohta, K. Nagashima, K. Itahashi, and T. Kamataki. 1990. Metabolic activation of aflatoxin B, and 2- amino-3-methylimidazo[4,5,-f]-quinoline by human adult and fetal livers. Cancer Res. 50:2641-2645. Klich, M.A. and J.I. Pitt. 1988. Differentiation of Aspergillus flavus from A. parasiticus and other closely related species. Trans. Br. Mycol. Soc. 91(1):99-108. Kurtzman, C.P., 8.W. Horn, and C.W. Hesseltine. 1987. Aspergillus.nomius, a new aflatoxin-producing species related to Aspergillus flavus and A. tamarii. Antonie Van Leeuwenhook 53:147-158. Lee, L.S., J. W. Bennett, L.A. Goldblatt, and RJE. Lundin. 1970. Norsolorinic acid from a mutant strain of Aspergillus parasiticus. J. Am. Oil Chem. Soc. 48:93-94. Lee, L.S., J.Wu Bennett,.A.F. Cucullu, and.J.8. Stanley. 1975. Synthesis of versicolorin A by a mutant strain of Aspergillus parasiticus deficient in aflatoxin production. J. Agr. Food Chem. 23:1132-1134. 132 Lin, B.K. and J.A. Anderson. 1992. Purification and properties of versiconal cyclase from Aspergillus parasiticus. Arch. Biochm. Biophys. 293(1):67-70. Lin, Y.C., J.C. Ayres, and P.E. Koehler. 1980. Influence of temperature cycling on the production of aflatoxins 81 and G1 by Aspergillus parasiticus. App. Environ. Micro. 40(2):333- 336. Lloyd, A.T. and P.M. Sharp. 1991. Codon usage in Aspergillus nidulans. Mol. Gen. Genet. 230:288-294. Maniatis, T., E.F. Fritsch, and .J.Sambrook (eds). 1982. Molecular Cloning: a laboratory manual. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY. Maplestone, R.A., M.J. Stone, and D.H. Williams. 1992. The evolutionary role of secondary metabolites - a review. Gene. 115: 151-157 . McCormick, S.P., D. Bhatnagar, and L.S. Lee. 1987. Averufanin is an aflatoxin B, precursor between averantin and averufin in the biosynthetic pathway. Appl. Environ. Micro. 53(1):14-16. Nesbitt, B.F., J. O'Kelly, K. Sargeant, andMA. Sheridan. 1962. Toxic metabolites of Aspergillus flavus. Nature 195:1060-1062. Niehaus, W.G. and L.J. Failla. 1984. Effect of zinc on versicolorin production by a mutant strain of Aspergillus parasiticus. Exp. Mycol. 8:80-84. Niehaus, W.G. and W.P. Jiang. 1989. Nitrate induces enzymes of the mannitol cycle and suppresses versicolorin synthesis in Aspergillus parasiticus. Mycopath. 107:131-137. Niehaus, W.G. 1989. Versicolorin synthesis by Aspergillus parasiticus:regulation by temperature and zinc. Exp. Mycol. 13:20-26. Novak, Lon and LHM. Kohn. 1990. Developmental proteins in Aspergillus. Exp. Mycol. 14:339-350. Oakley, B. R., J. E. Rinehart, B. L. Mitchell, C. E. Oakley, C. Carmona, G. L. Gray, and G. S. May. 1987. Cloning, mapping and molecular analysis of the pyrG (orotidine-5'-phosphate decarboxylase) gene of Aapagglllaa 11.118.198.5- Gene. 61:385-399. Papa, K.E. 1984. Genetics of Aspergillus f1avus:linkage of aflatoxin mutants. Can. J. Microbiol. 30:68-73. Park, D.L., L.S. Lee, R.L. Price, and A.E. Pohland. 1988. Review of the decontamination of aflatoxins by ammoniation: Current status and regulation. J. Assoc. Off. Anal. Chem. 133 71(4):685-703. Parkin, D.M., E. Laara, and C.S. Muir. 1988. Estimates of the worldwide frequency of sixteen.major cancers in 1980. Int. J. Cancer 41:184-197. Payne, G.A., and C.P. Woloshuk. 1989. Transformation of Aspergillus flavus to study aflatoxin biosynthesis. Mycopathologia 107:139-144. Payne, G.A. 1991. Report to the aflatoxin technical advisory group. 1991 Aflatoxin elimination workshop. Atlanta, Georgia. Pestka, J.J. 1988. Enhanced surveillance of foodborne mycotoxins by immunochemical assay. J. Assoc. Off. Anal. Chem. 71(6) :1075-1081. Pestka , J. J. and G. S . Bondy. 1990 . Alteration of immune function following dietary mycotoxin exposure. Can. J. Physiol. 68: 1009-1016. Pestka, J.J. and W.L. Casale. 1990. Naturally occuring fungal toxins. In J.O. Ariam and M.S. Simmons (ed). Food contamination from environmental sources. John Wiley 3 Sons. Reddy, T.V. , L. Viswanathan, and T.A. Venkitasubramanian. 1971. High aflatoxin production on a chemically defined medium. Appl. Micro. 22(3):393-396. Reddy, T.V. , L. Viswanathan, and T.A. Venkitasubramanian. 1979. Factors affecting aflatoxin production by Aspergillus parasiticus in a chemically defined medium. J. Gen. Micro. 114:409-413. Reynolds”, G.N. and J.J. Pestka. 1991. Enzyme-linked immunosorbent assay of versicolorin A and related aflatoxin biosynthetic precursors. J. Food Protect. 54(2):105-108. Sanger, F., S. Nicklen, and A.R. Coulson. 1977 . DNA sequencing with chain terminating inhibitors. Proc. Natl. Acad. Sci. 74:5463-5467. Schroeder, H. W., and W.R. Kelton. 1975. Production of sterigmatocystin by some species of the genus Aspergillus and its toxicity to chicken embryos. Appl. Micro. 30:589-591. Sharma, A., A.G. Behere, S.R. Padwal-Desai, and G.B. Nadkarni. 1980. Influence of inoculum size of Aspergillus parasiticus spores on aflatoxin production. Appl. Environ. Micro. 40(6):989-993. Sherman, D.H., r. Malpartida, M.J. Bibb, M.M. Kisser, M.J. Bibb, and D.A. Hopwood. 1989. Structure and deduced function 134 of the granaticin-producing polyketide synthase gene cluster of Streptomyces violaceoruber Tu22. EMBO J. 8(9):2717-2725. Shimada, T. and F.P. Guengerich. 1989. Evidence for cytochrom P-450, the nifedipine oxidase, being the principle enzyme involved in the bioactivation of aflatoxins in the human liver. Proc. Natl. Acad. Sci. U.S.A. 86:462-465. Singh, R., and D.P. Hsieh. 1976. Enzymatic conversion of sterigmatocystin into aflatoxin B, by cell-free extracts of Aspergillus parasiticus. Appl. Environ. Micro. 31(5):743-745. Skory, C.D., J.S. Horng, J.J. Pestka, and J.E. Linz. 1990. Transformation of Aspergillus parasiticus with the homologous gene (pyrG) involved in pyrimidine biosynthesis. App. Environ. Micro. 56(11):3315-3320. Skory, C.D., P.K. Chang, and J.E. Linz. 1991. Isolation of a gene involved in the aflatoxin biosynthetic pathway of Aspergillus parasiticus. ASM Annual Meeting. Dallas, Texas. Skory, C.D., P.K. Chang, and J.E. Linz. 1992. Isolation of a gene from Aspergillus parasiticus associated with the conversion of versicolorin to sterigmatocystin in aflatoxin biosynthesis. Appl. Environ. Micro. In press. Stark, A.A. 1980. Mutagenicity and carcinogenicity of mycotoxins: DNA binding as a possible mode of action. Ann. Rev. Microbiol. 34:235-62. Stoloff, L. 1976. Occurrence of mycotoxins in food and feed. In Rodricks, J.V. (ed.) Mycotoxins and other fungal related food problems. American Chemical Society, Washington, DC, pp. 23-50. Stoloff, L. 1989. Aflatoxin is not a probable human carcinogen: The published evidence is sufficient. Regulatory Tox. and Pharm. 10:272-283. Timberlake, W.E. 1990. Molecular genetics of Aspergillus development. Annu. Rev. Genet. 24:5-36. Townsend, C.A., K.A. Plavcan, K. Pal, and 8.W. Brobst. 1988. Hydroxyversicolorone: isolation and characterization of a potential intermediate in aflatoxin biosythesis. J} Am. Chem. SOC. 53:2472-2477. Tyagi, J.S. and T.A. venkitasubramanian. 1981. The role of glycolosis in aflatoxin biosynthesis. Can. J. Micro. 27:1276- 1282. Vining, L.C. 1992. Secondary metabolism, inventive evolution and biochemical diversity - a review. Gene 115:135-140. 135 Wild, C.P., Y.z. Jiang, S.J. Allen, L.A.M. Jansen, A.J. Hall, and R. Montesano. 1990. Aflatoxin-albumin adducts in human sera from different regions of the world. 11(12):2271-2274. Wild, C.P., R. Montesano, J. Van Benthem, E. Scherer, and L. DenEngelse. 1990. Intercellular variation in levels of adducts of aflatoxin B, and G, in DNA from rat tissues: a quantitative immunocytochemical study. J. Cancer Res. Clin. Oncol. 116:134- 140. Woloshuk, C.P., E.R. Seip, G.A. Payne, and C.R. Adkins. 1989. Genetic transformation system for the aflatoxin-producing fungus W flaflg. Appl. Environ. Micro. 55(1):86-90. Wong, J.J., R. Singh, and D.P. Hsieh. 1977. Mutagenicity of fungal metabolites related to aflatoxin biosynthesis. Mutat. Res. 44:447-450. Yabe, K., Y. Ando, and T. Hamasaki.1988. Biosynthetic relationship among aflatoxins 8,, 8,, G,, and G,. Appl. Environ. Microbiol. 54: 2101-2106. Yabe, K., Y. Ando, J. Hashimoto, and T. Hamasaki. 1989. Two distinct O-methyltransferases in aflatoxin biosynthesis. Appl. Environ. Micro. 55(9):2172-2177. Yabe, K., Y. Nakamura, H. Nakajima, Y. Ando, and T. Hamasaki. 1991. Enzymatic conversion of norsolorinic acid to averufin in aflatoxin biosynthesis. Appl. Environ. Micro 57(5) :1340-1345. Yabe, K., Y. Ando, and T. Hamasaki. 1991..A metabolic grid among versiconal hemiacetal acetate, versiconol acetate, versiconol and versiconal during aflatoxin biosynthesis. J. Gen. Micro. 137:22469-2475. Yabe, K., Y. Ando, and T. Hamasaki. 1991. Desaturase activity in the branching step between aflatoxins B, and G, and aflatoxins B, and G,. Agric. Biol. Chem. 55(7):1907-1911. Yao, R.G., and D.P. Hsieh. 1974. Step of dichlorvos inhibition in the pathway of aflatoxin biosynthesis. Appl. Micro. 28(1):52-57. Yu, F.L., W. Bender, and LN. Geronimo. 1990. Base and sequence specificities of aflatoxin B, binding to single and double-stranded DNAs. Carcinogenesis 11(3):475-478. Zaika, L.L. and R.L. Buchanan. 1987. Review of compound affecting the biosynthesis or bioregulation of aflatoxins. J. Food Protect. 50(8):691-708.