Teresa L. O'keeffe

MICROBIAL INTERACTIONS WITH MYCOTOXIGENIC FUNGI AND MYCOTOXINS

Mycotoxins such as aflatoxins, fumonisins, trichothecenes, and ochratoxins are contaminants of many agronomic crops worldwide, and cause both economic losses and health effects. The potential of antagonistic microorganisms to be developed into biological control agents has been investigated in several crop systems, as alternatives to chemical fungicides for control of mycotoxigenic fungi. Laboratory and greenhouse studies have identified a number of bacterial, yeast, and filamentous fungal isolates that reduce crop contamination of mycotoxigenic fungi, although investigations of field efficacy have been limited. These studies demonstrate that the diversity of ecological interactions between mycotoxigenic fungi and other resident microorganisms may provide tools for development of biocontrol methods to reduce mycotoxin contamination.

Isolation of maize soil and rhizosphere bacteria with antagonistic activity against Aspergillus flavus and Fusarium verticillioides.

Bacterial isolates from Mississippi maize field soil and maize rhizosphere samples were evaluated for their potential as biological control agents against Aspergillus flavus and Fusarium verticillioides. Isolated strains were screened for antagonistic activities in liquid coculture against A. flavus and on agar media against A. flavus and F. verticillioides. We identified 221 strains that inhibited growth of both fungi. These bacteria were further differentiated by their production of extracellular enzymes that hydrolyzed chitin and yeast cell walls and by production of antifungal metabolites. Based on molecular and nutritional identification of the bacterial strains, the most prevalent genera isolated from rhizosphere samples were Burkholderia and Pseudomonas, and the most prevalent genera isolated from nonrhizosphere soil were Pseudomonas and Bacillus. Less prevalent genera included Stenotrophomonas, Agrobacterium, Variovorax, Wautersia, and several genera of coryneform and enteric bacteria. In quantitative coculture assays, strains of P. chlororaphis and P. fluorescens consistently inhibited growth of A. flavus and F. verticillioides in different media. These results demonstrate the potential for developing individual biocontrol agents for simultaneous control of the mycotoxigenic A. flavus and F. verticillioides.

Incidence of Fumonisin B2 Production within Aspergillus Section Nigri Populations Isolated from California Raisins

Fungi belonging to Aspergillus section Nigri occur frequently and in high populations on grapes. Species within this section include Aspergillus niger, A. tubingensis, and A. carbonarius, and they are potential sources for mycotoxins including ochratoxin A and fumonisin B(2) (FB(2)) in grapes and grape products. Aspergillus section Nigri strains were isolated from California raisins to examine the frequency and extent of FB(2) production. Of 392 strains isolated, 197 strains were identified as A. niger, 131 of which produced FB(2). These strains produced from 1.2 to 27 μg/ml FB(2) in culture. PCR amplification of fum1 and fum19 gene fragments showed that all FB(2)-producing strains and nearly all nonproducing strains of A. niger contain these genes. An additional 175 strains were identified as A. tubingensis, none of which produced FB(2). PCR with fum1 and fum19 primers amplified gene fragments of 14 and 25% of A. tubingensis strains, respectively, suggesting that putative orthologs of A. niger fumonisin biosynthetic genes might occur in A. tubingensis. These results indicate that FB(2) production is common among field isolates of A. niger and suggest that the potential for FB(2) contamination of California raisins should be addressed further.

Occurrence of Ochratoxin A Contamination and Detection of Ochratoxigenic Aspergillus Species in Retail Samples of Dried Fruits and Nuts

Ochratoxin A (OTA) is a mycotoxin produced by several species of Aspergillus and Penicillium and is a potential contaminant of a wide variety of food products. To determine the incidence of OTA contamination in dried fruits and tree nuts, retail packaged and bulk raisins, dates, figs, prunes, almonds, pistachios, and walnuts were collected from small and large supermarkets in seven areas of the United States between 2012 and 2014. Of the 665 samples analyzed, OTA was detected in 48 raisin samples, 4 fig samples, 4 pistachio samples, and 1 date sample. OTA contamination levels ranged from 0.28 to 15.34 ng/g in dried fruits and 1.87 to 890 ng/g in pistachios; two raisin samples and one pistachio sample exceeded the European Union regulatory limit of 10 ng/g. PCR detection of potential OTA-producing Aspergillus species revealed the presence of A. niger, A. welwitschiae, and A. carbonarius in 20, 7, and 7 of the 57 OTA-contaminated samples, respectively. However, OTA-producing A. carbonarius was isolated from only one raisin sample, and no other OTA-producing Aspergillus species were found. These results suggest that raisins are more frequently contaminated with low levels of OTA than are other dried fruits and nuts and that Aspergillus species are the likely source of that contamination.

Detection and discrimination of four Aspergillus section Nigri species by PCR

Species of Aspergillus section Nigri are not easily distinguished by traditional morphological techniques, and typically are identified by DNA sequencing methods. We developed four PCR primers to distinguish between Aspergillus niger, Aspergillus welwitschiae, Aspergillus carbonarius and Aspergillus tubingensis, based on species‐conserved differences in the calmodulin gene sequence. PCR amplification from total DNA using these primers was species specific; no amplification occurred from nontarget species DNA for each primer pair. Species‐specific PCR could distinguish between species in mixed DNA templates, indicating a utility in determining culture uniformity of isolated Aspergillus strains. In addition, with these primer sets, each species could be detected in soil following mixed‐species inoculation with Aspergillus spores. This indicates that PCR with these species‐specific primers may be useful in determining the distribution of Aspergillus species in environmental samples without the need for species identification from isolated strains, as well as detecting species that may be infrequently isolated by culture‐based methods.

Strain of Fusarium oxysporum Isolated from Almond Hulls Produces Styrene and 7-Methyl-1,3,5-cyclooctatriene as the Principal Volatile Components

An isolated strain of Fusarium oxysporum from the hulls of Prunus dulcis (sweet almond) was found to produce relatively large quantities of the hydrocarbons styrene and two isomers of 7-methyl-1,3,5- cyclooctatriene (MCOT). Production of styrene and MCOT was reproduced on a small scale using potato dextrose agar as a growth medium and scaled up using 1 L of inoculated potato dextrose broth. The compounds were trapped as volatile organic compounds (VOCs) onto solid-phase microextraction (SPME) for small scale and Tenax for large scale and then isolated using standard high-performance liquid chromatography (HPLC) methods. Styrene was authenticated by a comparison to the retention times, fragmentation patterns, and calculated retention indices of a commercially available sample. The identity of MCOT was verified by a short chemical synthesis and a comparison of spectroscopic data to the isolated sample. A biosynthetic scheme of styrene is proposed on the basis of a (13)C-labeling study. This is the first report of MCOT isolated as a natural product.

Inhibition of Aspergillus flavus in Soil by Antagonistic Pseudomonas Strains Reduces the Potential for Airborne Spore Dispersal

Pseudomonas chlororaphis strain JP1015 and P. fluorescens strain JP2175 were previously isolated from Mississippi cornfield soil samples and selected for their growth inhibition of Aspergillus flavus in laboratory culture. In this study, the antifungal activity of these bacterial strains against A. flavus in soil coculture was determined. Growth of A. flavus was inhibited up to 100-fold by P. chlororaphis strain JP1015 and up to 58-fold by P. fluorescens strain JP2175 within 3 days following soil coinoculation. A. flavus propagule densities after 16 days remained 7- to 20-fold lower in soil treated with either bacterial strain. Using a bench-scale wind chamber, we demonstrated that treatments of soil with P. chlororaphis strain JP1015 and P. fluorescens strain JP2175 reduced airborne spores dispersed across a 1 m distance by 75- to 1,000-fold and 10- to 50-fold, respectively, depending on soil type and inoculum level. These results suggest that application of these bacterial strains may be effective in reducing soil populations of mycotoxigenic fungi, thereby reducing fungal spore formation, and ultimately reducing the potential for crop plant infection via airborne transmission.

Characterization of Aspergillus section Nigri species populations in vineyard soil using droplet digital PCR.

Identification of populations of Aspergillus section Nigri species in environmental samples using traditional methods is laborious and impractical for large numbers of samples. We developed species‐specific primers and probes for quantitative droplet digital PCR (ddPCR) to improve sample throughput and simultaneously detect multiple species in each sample. The ddPCR method was used to distinguish Aspergillus niger, Aspergillus welwitschiae, Aspergillus tubingensis and Aspergillus carbonarius in mixed samples of total DNA. Relative abundance of each species measured by ddPCR agreed with input ratios of template DNAs. Soil samples were collected at six time points over two growing seasons from two raisin vineyards in Fresno County, California. Aspergillus section Nigri strains were detected in these soils in the range of 102–105 CFU g−1. Relative abundance of each species varied widely among samples, but in 52 of 60 samples, A. niger was the most abundant species, ranging from 38 to 88% of the total population. In combination with total plate counts, this ddPCR method provides a high‐throughput method for describing population dynamics of important potential mycotoxin‐producing species in environmental samples.

Distribution and mycotoxigenic potential of Aspergillus section Nigri species in naturally contaminated almonds.

In a previous study, inedible almond pick-out samples were assayed for aflatoxin and aflatoxigenic Aspergillus species. These samples contained high populations of black-spored Aspergillus section Nigri species. To investigate whether these species may contribute to the total potential mycotoxin content of almonds, Aspergillus section Nigri strains were isolated from these samples and assayed for ochratoxin A (OTA) and fumonisin B2 (FB2). The majority of isolates (117 strains, 68%) were identified as Aspergillus tubingensis, which do not produce either mycotoxin. Of the 47 Aspergillus niger and Aspergillus awamori isolates, 34 strains (72%) produced FB2 on CY20S agar, and representative strains produced lower but measurable amounts of FB2 on almond meal agar. No OTA-producing strains of Aspergillus section Nigri were detected. Almond pick-out samples contained no measurable FB2, suggesting that properly dried and stored almonds are not conducive for FB2 production by resident A. niger and A. awamori populations. However, 3 of 21 samples contained low levels (<1.5 ng/g) of OTA, indicating that sporadic OTA contamination may occur but may be caused by OTA-producing strains of other Aspergillus species.

Population Dynamics of Aspergillus Section Nigri Species on Vineyard Samples of Grapes and Raisins

Several species of Aspergillus section Nigri, including potential mycotoxin producers, are common residents of grape vineyards, but the relative population size of individual species throughout the growing season is difficult to determine using traditional isolation and identification methods. Using a quantitative droplet digital PCR (ddPCR) method in combination with dilution plating, total Aspergillus section Nigri populations and relative proportions of A. niger, A. welwitschiae, A. carbonarius, and A. tubingensis were measured from vineyard samples without the need for identifying individual fungal isolates. Grapes were sampled from two raisin vineyards (vineyards A and B) at berry set, veraison, harvest, and raisin stages in two consecutive years. Plate counts showed that the total population of Aspergillus section Nigri present on the fruit increased from berry set to raisin and became a larger component of the total recovered fungal population in both vineyards in both years. Results from ddPCR analysis showed that the relative proportion of A. carbonarius among the four species assayed increased later in the season (harvest and raisin) in comparison to earlier in the season (berry set and veraison). Total fungal and Aspergillus section Nigri plate counts were not significantly different between vineyards in either year. However, vineyard A generally showed higher proportions of A. carbonarius in harvest and raisin samples than vineyard B. This coincided with higher incidence and levels of ochratoxin A in vineyard A harvest and raisin fruit than in vineyard B fruit. This work demonstrates that this ddPCR method is a useful tool for culture-independent monitoring of populations of mycotoxigenic Aspergillus species during grape and raisin production.