Deciphering the origin of Aspergillus flavus NRRL21882, the active biocontrol agent of Afla-Guard®

Single nucleotide polymorphisms (SNPs) of genome sequences of eight Aspergillus flavus and seven Aspergillus oryzae strains were extracted with Mauve, a multiple-genome alignment programme. A phylogenetic analysis with sequences comprised of concatenated total SNPs by the unweighted pair group method with arithmetic mean (UPGMA) of MAFFT adequately separated them into three groups, A. flavus S-morphotype, A. flavus L-morphotype and A. oryzae. Divergence time inferred for A. flavus NRRL21882, the active agent of the biocontrol product Afla-Guard^®, and S-morphotype was about 5·1 mya. Another biocontrol strain, A. flavus AF36, diverged from aflatoxigenic L-morphotype about 2·6–3·0 mya. Despite the close relatedness of A. oryzae to A. flavus, A. oryzae strains likely evolved from aflatoxigenic Aspergillus aflatoxiformans (=A. parvisclerotigenus). A survey of A. flavus populations implies that prior Afla-Guard^® applications are associated with prevalence of NRRL21882-type isolates in Mississippi fields. In addition, a few NRRL21882 relatives were identified. A. flavus Og0222, a biocontrol ingredient of Aflasafe™, was verified as a NRRL21882-type strain, having identical sequence breakpoints that led to deletion of aflatoxin and cyclopiazonic acid gene clusters. A similar UPGMA analysis suggests that the occurrence of NRRL21882-type strains is a more recent event.     

Global population structure and adaptive evolution of aflatoxin‐producing fungi

Aflatoxins produced by several species in Aspergillus section Flavi are a significant problem in agriculture and a continuous threat to human health. To provide insights into the biology and global population structure of species in section Flavi, a total of 1,304 isolates were sampled across six species (A. flavus, A. parasiticus, A. nomius, A. caelatus, A. tamarii, and A. alliaceus) from single fields in major peanut‐growing regions in Georgia (USA), Australia, Argentina, India, and Benin (Africa). We inferred maximum‐likelihood phylogenies for six loci, both combined and separately, including two aflatoxin cluster regions (aflM/alfN and aflW/aflX) and four noncluster regions (amdS, trpC, mfs and MAT), to examine population structure and history. We also employed principal component and STRUCTURE analysis to identify genetic clusters and their associations with six different categories (geography, species, precipitation, temperature, aflatoxin chemotype profile, and mating type). Overall, seven distinct genetic clusters were inferred, some of which were more strongly structured by G chemotype diversity than geography. Populations of A. flavus S in Benin were genetically distinct from all other section Flavi species for the loci examined, which suggests genetic isolation. Evidence of trans‐speciation within two noncluster regions, whereby A. flavus S_BG strains from Australia share haplotypes with either A. flavus or A. parasiticus, was observed. Finally, while clay soil and precipitation may influence species richness in Aspergillus section Flavi, other region‐specific environmental and genetic parameters must also be considered.     

Effect of nontoxigenic Aspergillus flavus and A. parasiticus on aflatoxin contamination of wounded peanut seeds inoculated with agricultural soil containing natural fungal populations

Peanuts and other seed and grain crops are commonly contaminated with carcinogenic aflatoxins, secondary metabolites produced by Aspergillus flavus and A. parasiticus. Aflatoxin contamination of peanuts in the field can be reduced by 77–98% with biological control through the application of nontoxigenic strains of these species, which competitively exclude native aflatoxin-producing strains from developing peanuts. In this study, viable peanut seeds were artificially wounded and inoculated with field soil containing natural fungal populations that were supplemented with conidia of nontoxigenic A. flavus NRRL 21882 (niaD nitrate-nonutilizing mutant) and A. parasiticus NRRL 21369 (conidial color mutant). Increasing soil densities of applied nontoxigenic strains generally resulted in an increase in the incidence of seed colonization by applied nontoxigenic strains, a decrease in seed colonization by native A. flavus and A. parasiticus, and a decrease in aflatoxin concentration in seeds. Reduction of aflatoxins in peanut seeds depended on both the density and the aflatoxin-producing potential of native populations and on the fungal strain used for biological control. Wild-type strain A. flavus NRRL 21882 and its niaD mutant were equally effective in reducing aflatoxins in peanuts, indicating that nitrate-nonutilizing mutants, which are easily monitored in the field, can be used for evaluating the efficacy of biocontrol strains.     

The Aflatoxin Biosynthesis Cluster Gene, aflX, Encodes an Oxidoreductase Involved in Conversion of Versicolorin A to Demethylsterigmatocystin

Biosynthesis of the toxic and carcinogenic aflatoxins by the fungus Aspergillus flavus is a complicated process involving more that 27 enzymes and regulatory factors encoded by a clustered group of genes. Previous studies found that three enzymes, encoded by verA, ver-1, and aflY, are required for conversion of versicolorin A (VA), to demethylsterigmatocystin. We now show that a fourth enzyme, encoded by the previously uncharacterized gene, aflX (ordB), is also required for this conversion. A homolog of this gene, stcQ, is present in the A. nidulans sterigmatocystin (ST) biosynthesis cluster. Disruption of aflX in Aspergillus flavus gave transformants that accumulated ~4-fold more VA and fourfold less aflatoxin than the untransformed strain. Southern and Northern blot analyses confirmed that aflX was the only gene disrupted in these transformants. Feeding ST or O-methylsterigmatocystin, but not VA or earlier precursor metabolites, restored normal levels of AF production. The protein encoded by aflX is predicted to have domains typical of an NADH-dependent oxidoreductase. It has 27% amino acid identity to a protein encoded by the aflatoxin cluster gene, aflO (avfA). Some of domains in the protein are similar to those of epoxide hydrolases.     

Effect of sexual recombination on population diversity in aflatoxin production by Aspergillus flavus and evidence for cryptic heterokaryosis

Aspergillus flavus is the major producer of carcinogenic aflatoxins (AFs) in crops worldwide. Natural populations of A. flavus show tremendous variation in AF production, some of which can be attributed to environmental conditions, differential regulation of the AF biosynthetic pathway and deletions or loss‐of‐function mutations in the AF gene cluster. Understanding the evolutionary processes that generate genetic diversity in A. flavus may also explain quantitative differences in aflatoxigenicity. Several population studies using multilocus genealogical approaches provide indirect evidence of recombination in the genome and specifically in the AF gene cluster. More recently, A. flavus has been shown to be functionally heterothallic and capable of sexual reproduction in laboratory crosses. In the present study, we characterize the progeny from nine A. flavus crosses using toxin phenotype assays, DNA sequence‐based markers and array comparative genome hybridization. We show high AF heritability linked to genetic variation in the AF gene cluster, as well as recombination through the independent assortment of chromosomes and through crossing over within the AF cluster that coincides with inferred recombination blocks and hotspots in natural populations. Moreover, the vertical transmission of cryptic alleles indicates that while an A. flavus deletion strain is predominantly homokaryotic, it may harbour AF cluster genes at a low copy number. Results from experimental matings indicate that sexual recombination is driving genetic and functional hyperdiversity in A. flavus. The results of this study have significant implications for managing AF contamination of crops and for improving biocontrol strategies using nonaflatoxigenic strains of A. flavus.     

Molecular characterization of toxigenic and atoxigenic Aspergillus flavus isolates,collected from peanut fields in China

Aims: The objectives of this study were to assess the genetic relationships between toxigenic and atoxigenic isolates of Aspergillus flavus collected from peanut fields in China, and to analyse deletions within the aflatoxin biosynthetic gene cluster for the atoxigenic isolates. Methods and Results: Analysis of random‐amplified polymorphic DNA and microsatellite‐primed PCR data showed that the toxigenic and atoxigenic isolates of A. flavus were not clustered based on their regions and their ability of aflatoxin and sclerotial production. These results were further supported by DNA sequence of ITS, pksA and omtA genes. PCR assays showed that 24 of 35 isolates containing no detectable aflatoxins had the entire aflatoxin gene cluster. Eleven atoxigenic isolates had five different deletion patterns in the cluster. Conclusions: Toxigenic and atoxigenic isolates of A. flavus are genetically similar, but some atoxigenic isolates having deletions within the aflatoxin gene cluster can be identified readily by PCR assays. Significance and Impact of the Study: Because the extensive deletions within the aflatoxin gene cluster are not rare in the atoxigenic isolates, analysis of deletion within the cluster would be an effective method for the rapid screening of atoxigenic isolates for developing biocontrol agents.     

Evaluation of different genotypes of nontoxigenic Aspergillus flavus for their ability to reduce aflatoxin contamination in peanuts

Aflatoxins produced by the fungus Aspergillus flavus are potent carcinogens and account for large monetary losses worldwide in peanuts, maize, and cottonseed. Biological control in which a nontoxigenic strain of A. flavus is applied to crops at high concentrations effectively reduces aflatoxins through competition with native aflatoxigenic populations. In this study, eight nontoxigenic strains of A. flavus belonging to different vegetative compatibility groups and differing in deletion patterns within the aflatoxin gene cluster were evaluated for their ability to reduce aflatoxin B₁ when paired with eight aflatoxigenic strains on individual peanut seeds. Inoculation of wounded viable peanut seeds with conidia demonstrated that nontoxigenic strains differed in their ability to reduce aflatoxin B₁. Reductions in aflatoxin B₁ often exceeded expected reductions based on a 50:50 mixture of the two A. flavus strains, although one nontoxigenic strain significantly increased aflatoxin B₁ when paired with an aflatoxigenic strain. Therefore, nontoxigenicity alone is insufficient for selecting a biocontrol agent and it is also necessary to test the effectiveness of a nontoxigenic strain against a variety of aflatoxigenic strains.     

Sequence breakpoints in the aflatoxin biosynthesis gene cluster and flanking regions in nonaflatoxigenic Aspergillus flavus isolates

Aspergillus flavus populations are genetically diverse. Isolates that produce either, neither, or both aflatoxins and cyclopiazonic acid (CPA) are present in the field. We investigated defects in the aflatoxin gene cluster in 38 nonaflatoxigenic A. flavus isolates collected from southern United States. PCR assays using aflatoxin-gene-specific primers grouped these isolates into eight (A-H) deletion patterns. Patterns C, E, G, and H, which contain 40 kb deletions, were examined for their sequence breakpoints. Pattern C has one breakpoint in the cypA 3' untranslated region (UTR) and another in the verA coding region. Pattern E has a breakpoint in the amdA coding region and another in the ver1 5'UTR. Pattern G contains a deletion identical to the one found in pattern C and has another deletion that extends from the cypA coding region to one end of the chromosome as suggested by the presence of telomeric sequence repeats, CCCTAATGTTGA. Pattern H has a deletion of the entire aflatoxin gene cluster from the hexA coding region in the sugar utilization gene cluster to the telomeric region. Thus, deletions in the aflatoxin gene cluster among A. flavus isolates are not rare, and the patterns appear to be diverse. Genetic drift may be a driving force that is responsible for the loss of the entire aflatoxin gene cluster in nonaflatoxigenic A. flavus isolates when aflatoxins have lost their adaptive value in nature.     

β-Carotene inhibition of aflatoxin biosynthesis amongAspergillus flavus genotypes from Illinois corn

Thirty-nineAspergillus flavus genotypes (DNA fingerprinting) isolated from corn grown in a field near Kilbourne, Illinois were evaluated for their sensitivity to β-carotene (50 μg/ml) inhibition of aflatoxin B₁ biosynthesis. Inhibition of aflatoxin was greater than 90% for 28 of the genotypes and >70% for 38 of the 39 genotypes. FiveA. flavus strains (4 fingerprint groups) isolated from molded raw peanuts, NRRL 3239, NRRL 3357, NRRL 6514, NRRL 6515 and NRRL 13135, produced greater quantities of aflatoxin than all 39 genotypes isolated from corn, and were less sensitive to β-carotene inhibition.Aspergillus flavus NRRL 3357 is commonly used as inoculum in variety trials for aflatoxin resistance. Isolate identity and sensitivity to potential inhibitors in corn can be critical in assessing corn resistance to aflatoxin.     

Aflatoxin is degraded by mycelia from toxigenic and nontoxigenic strains of aspergilli grown on different substrates

The ability of 9-day-old mycelia of Aspergillus parasiticus NRRL 2999 to degrade aflatoxin varied depending on the substrate used to grow the mold. Substrates which allowed substantial mycelial growth yielded mycelia which actively degraded aflatoxin. Substrates which allowed minimal growth of mycelia yielded mycelia with little ability to degrade aflatoxin. Biodegradation of aflatoxin was also strain-dependent. A. parasiticus NRRL 2999 and NRRL 3000 actively degraded aflatoxin, A. flavus NRRL 3353 was less active, and A. flavus NRRL 482 and A. parasiticus NRRL 3315 degraded minimal amounts of aflatoxins. Those aspergilli producing greatest amounts of aflatoxin also degraded aflatoxins most rapidly, whereas those strains which produced minimal amounts of aflatoxin generally degraded aflatoxins less effectively. Substrates which allowed maximum aflatoxin production also yielded mycelia which actively degraded aflatoxins, whereas media which allowed limited production of aflatoxin generally yielded mycelia with minimal ability to degrade the toxin. Although exceptions exist, generally as aflatoxin production increased so did the ability of mycelia to degrade the toxin.     

The inhibitory effect of Bacillus megaterium on aflatoxin and cyclopiazonic acid biosynthetic pathway gene expression in Aspergillus flavus

Aspergillus flavus is one of the major moulds that colonize peanut in the field and during storage. The impact to human and animal health, and to the economy in agriculture and commerce, is significant since this mold produces the most potent known natural toxins, aflatoxins, which are carcinogenic, mutagenic, immunosuppressive, and teratogenic. A strain of marine Bacillus megaterium isolated from the Yellow Sea of East China was evaluated for its effect in inhibiting aflatoxin formation in A. flavus through down-regulating aflatoxin pathway gene expression as demonstrated by gene chip analysis. Aflatoxin accumulation in potato dextrose broth liquid medium and liquid minimal medium was almost totally (more than 98 %) inhibited by co-cultivation with B. megaterium. Growth was also reduced. Using expression studies, we identified the fungal genes down-regulated by co-cultivation with B. megaterium across the entire fungal genome and specifically within the aflatoxin pathway gene cluster (aflF, aflT, aflS, aflJ, aflL, aflX). Modulating the expression of these genes could be used for controlling aflatoxin contamination in crops such as corn, cotton, and peanut. Importantly, the expression of the regulatory gene aflS was significantly down-regulated during co-cultivation. We present a model showing a hypothesis of the regulatory mechanism of aflatoxin production suppression by AflS and AflR through B. megaterium co-cultivation.     

Isolation of Bacterial Antagonists of Aspergillus flavus from Almonds

Bacteria were isolated from California almond orchard samples to evaluate their potential antifungal activity against aflatoxin-producing Aspergillus flavus. Fungal populations from the same samples were examined to determine the incidence of aflatoxigenic Aspergillus species. Antagonistic activities of the isolated bacterial strains were screened against a nonaflatoxigenic nor mutant of A. flavus, which accumulates the pigmented aflatoxin precursor norsolorinic acid (NOR) under conditions conducive to aflatoxin production. Using solid and liquid media in coculture assays, 171 bacteria isolated from almond flowers, immature nut fruits, and mature nut fruits showed inhibition of A. flavus growth and/or inhibition of NOR accumulation. Bacterial isolates were further characterized for production of extracellular enzymes capable of hydrolyzing chitin or yeast cell walls. Molecular and physiological identification of the bacterial strains indicated that the predominant genera isolated were Bacillus, Pseudomonas, Ralstonia, and Burkholderia, as well as several plant-associated enteric and nonenteric bacteria. A set of 20 isolates was selected for further study based on their species identification, antifungal phenotypes, and extracellular enzyme production. Quantitative assays using these isolates in liquid coculture with a wild-type, aflatoxin-producing A. flavus strain showed that a number of strains completely inhibited fungal growth in three different media. These results indicate the potential for development of bacterial antagonists as biological control agents against aflatoxigenic aspergilli on almonds.     

Method for monitoring deletions in the aflatoxin biosynthesis gene cluster of Aspergillus flavus with multiplex PCR

The report presents a rapid, inexpensive and simple method for monitoring indels with influence on aflatoxin biosynthesis within Aspergillus flavus populations. PCR primers were developed for 32 markers spaced approximately every 5 kb from 20 kb proximal to the aflatoxin biosynthesis gene cluster to the telomere repeat. This region includes gene clusters required for biosynthesis of aflatoxins and cyclopiazonic acid; the resulting data were named cluster amplification patterns (CAPs). CAP markers are amplified in four multiplex PCRs, greatly reducing the cost and time to monitor indels within this region across populations. The method also provides a practical tool for characterizing intraspecific variability in A. flavus not captured with other methods.

Significance and Impact of the Study

Aflatoxins, potent naturally‐occurring carcinogens, cause significant agricultural problems. The most effective method for preventing contamination of crops with aflatoxins is through use of atoxigenic strains of Aspergillus flavus to alter the population structure of this species and reduce incidences of aflatoxin producers. Cluster amplification pattern (CAP) is a rapid multiplex PCR method for identifying and monitoring indels associated with atoxigenicity in A. flavus. Compared to previous techniques, the reported method allows for increased resolution, reduced cost, and greater speed in monitoring the stability of atoxigenic strains, incidences of indel mediated atoxigenicity and the structure of A. flavus populations.     

Studies on the biocontrol of aflatoxin in maize in Thailand

Aflatoxins in maize and peanuts remain a major cause of liver cancer and other human and animal health issues. The principal causal fungi are Aspergillus flavus and A. parasiticus. Relatively little attention has been paid to reducing aflatoxin formation before harvest. The most promising approach is biocontrol by competitive exclusion. This project aimed to demonstrate the efficacy of locally isolated strains of A. flavus for biocontrol of aflatoxin in maize in Thailand. After a rigorous process utilising molecular methods was used to select non-toxigenic A. flavus strains, field inoculum was produced by using hulled rice coated with A. flavus spores in molasses. Field experiments were conducted over two years in two districts, one of light sandy soil (Chokchai), the other a heavy, close textured, soil (Pakchong). Postharvest treatments representative of local practice were also undertaken. Crops 1 and 2 were not significantly contaminated with aflatoxin at the time of harvest, so any impact of biocontrol could not be assessed. However, wet shelling plus storage before drying resulted in increased aflatoxin contamination; biocontrol had no impact on this increase. In crops 3 and 4, biocontrol had a beneficial impact in some freshly harvested maize. Biocontrol treatments also significantly reduced aflatoxin contamination in samples from some treatments stored for two or four days after shelling, but had minimal effect in others. These experiments demonstrated that biocontrol can be highly effective in reducing aflatoxin contamination in maize in Thailand, both at harvest and during poor postharvest crop handling. However, results were inconsistent.     

Selection of non-toxigenic strains of Aspergillus flavus for biocontrol of aflatoxins in maize in Thailand

Aflatoxin production in maize and peanuts remains a major public health problem, especially in developing countries. One promising method for combating aflatoxin formation is biocontrol using competitive exclusion, a management strategy currently being studied in maize crops in Thailand. It is important that the strains of Aspergillus flavus used in biocontrol be non-toxigenic and be incapable of reversion to toxigenicity. In the current study, 80 non-toxigenic strains of A. flavus, randomly selected from commercially produced dried maize samples from several sources in Thailand, were screened for their potential as biocontrol strains by examining the 24 aflatoxin biosynthesis genes, using a PCR assay. Assessment of the presence or absence of PCR products provides an indication of the function of pathway genes. Of the 80 strains, 78 showed no PCR products from one or more genes in the aflatoxin biosynthesis pathway. Twenty-three isolates showed only one failure, in the aflD (nor-1) gene, but most isolates failed to produce a PCR product for two or more genes. Nineteen isolates (24%) failed to show a PCR product in 10 or more genes. Altogether, 45 PCR product patterns were observed, usually common to only one or two isolates, indicating great diversity in the aflatoxin biosynthesis pathway in A. flavus isolates taken from dried Thai maize. Although the absence of a particular PCR product is not conclusive evidence that the particular gene is non-functional, the absence of several such PCR products provides reasonable evidence that the isolate in question will not revert to toxigenicity in the field.     

Microbe‐mediated control of Aspergillus flavus in stored rice grains with a focus on aflatoxin inhibition and biodegradation

Biological control of mycotoxigenic fungi using antagonistic microbes is a promising alternative to agricultural chemicals for postharvest storage. In this study, we evaluated rice‐derived bacterial strains to identify biocontrol agents to inhibit Aspergillus flavus in stored rice grains. Consequently, we obtained three potential biocontrol strains (Microbacterium testaceum KU313, Bacillus megaterium KU143 and Pseudomonas protegens AS15) from 26 tested strains that were prescreened from the 460 strains isolated from rice grains. The three selected strains proved to be effective biocontrol agents showing antifungal activity against A. flavus and good colonisation ability on rice grains, along with inhibition of the fungal growth and aflatoxin production. In particular, P. protegens AS15 greatly inhibited the aflatoxins produced by A. flavus on rice grains to 8.68 (percent aflatoxin reduction relative to control = 82.9%) and 18.05 (68.3 %) ng g^?1 dry weight of rice grains, compared with the 50.89 and 56.97 ng g^?1 dry weight of rice grains of the MgSO₄ control at 1 and 2 weeks after inoculation, respectively. In addition, strain AS15 had a significant ability to not only degrade aflatoxin B₁ (the most harmful aflatoxin), but also utilise the toxin for bacterial growth in a nutrient‐deficient medium. Therefore, the selected bacterial strains could be environmentally sound alternatives for the management of A. flavus and aflatoxin production by reducing the fungal damage to stored rice grains. This would also reduce the human and animal health hazards associated with the consumption of fungus‐contaminated rice grains. To our knowledge, this is the first report of the potential of the bacterial species M. testaceum and P. protegens as biocontrol agents for controlling aflatoxigenic A. flavus on stored rice grains.     

Regulation of aflatoxin biosynthesis. 1 Comparative study of mycelial composition and glycolysis in aflatoxigen and nonaflatoxigenic strains.

A comparative biochemical study of an aflatoxigenic strain Aspergillus parasiticus NRRL 3240 and a nonaflatoxigenic strain A. flavus NRRL 3237 was carried out in order to have a better idea of regulation of aflatoxin biosynthesis. The results obtained revealed continuous primary metabolic activity (protein synthesis) in the nonaflatoxigenic strain while the aflatoxigenic stain showed inhibition of protein and nucleic acid synthesis. The aflatoxigenic strain showed higher levels of oxygen uptake, RNA, NAD, FMN and activities of glycolytic enzymes. Furthermore, it had lower of lipids and reduced activity of glucose-6-phosphate dehydrogenase, which is a source for NADPH. The differences observed have been discussed in relation to aflatoxin biosynthesis and its regulation.     

Sexuality Generates Diversity in the Aflatoxin Gene Cluster: Evidence on a Global Scale

Aflatoxins are produced by Aspergillus flavus and A. parasiticus in oil-rich seed and grain crops and are a serious problem in agriculture, with aflatoxin B₁ being the most carcinogenic natural compound known. Sexual reproduction in these species occurs between individuals belonging to different vegetative compatibility groups (VCGs). We examined natural genetic variation in 758 isolates of A. flavus, A. parasiticus and A. minisclerotigenes sampled from single peanut fields in the United States (Georgia), Africa (Benin), Argentina (Córdoba), Australia (Queensland) and India (Karnataka). Analysis of DNA sequence variation across multiple intergenic regions in the aflatoxin gene clusters of A. flavus, A. parasiticus and A. minisclerotigenes revealed significant linkage disequilibrium (LD) organized into distinct blocks that are conserved across different localities, suggesting that genetic recombination is nonrandom and a global occurrence. To assess the contributions of asexual and sexual reproduction to fixation and maintenance of toxin chemotype diversity in populations from each locality/species, we tested the null hypothesis of an equal number of MAT1-1 and MAT1-2 mating-type individuals, which is indicative of a sexually recombining population. All samples were clone-corrected using multi-locus sequence typing which associates closely with VCG. For both A. flavus and A. parasiticus, when the proportions of MAT1-1 and MAT1-2 were significantly different, there was more extensive LD in the aflatoxin cluster and populations were fixed for specific toxin chemotype classes, either the non-aflatoxigenic class in A. flavus or the B₁-dominant and G₁-dominant classes in A. parasiticus. A mating type ratio close to 1∶1 in A. flavus, A. parasiticus and A. minisclerotigenes was associated with higher recombination rates in the aflatoxin cluster and less pronounced chemotype differences in populations. This work shows that the reproductive nature of the population (more sexual versus more asexual) is predictive of aflatoxin chemotype diversity in these agriculturally important fungi.     

Morphological changes in strains of Aspergillus flavus link ex fries and Aspergillus parasiticus speare related with aflatoxin production

Two strains ofAspergillus flavus Linkex Fr. and two strains ofA. parasiticus Speare were cultured on crushed moist wheat (Triticum durum var. Pané no. 247) for aflatoxin production studies in correlation with morphological changes. The toxicogenic strains were adapted to the substratum by means of successive transfers at regular intervals (72 h.)The amount aflatoxins synthesized by the toxicogenic strains decreased gradually after succesive subculturing. The decrease was accompanied by marked morphological changes. One of the strains studied,A. flavus NRRL 3251, lost completly the capacity of aflatoxin synthesis after several subcultures, presenting at the same time strong morphological variations.A. flavus CBS 120.62 also lost its toxicogenicity after six subcultures.