Location: Food and Feed Safety Research
2025 Annual Report
Objectives
1. Identify differentially expressed genes in resistant [R] and susceptible [S] corn lines that can serve as targets in controlling Aspergillus flavus and aflatoxin contamination.
2. Identify and characterize corn seed metabolites for enhancing resistance to aflatoxin contamination.
3. Develop and evaluate transgenic corn lines by over-expressing resistance-associated protein genes, or gene editing and silencing of Aspergillus flavus genes critical to growth and aflatoxin production.
4. Develop detection methods for pre- and post- harvest contamination in food supply and predictive modeling for mycotoxin contamination.
Subobjective 4.A: Develop and evaluate different spectral-based imaging instruments for non-destructive detection of AF contamination in corn. Advance and commercialize hyperspectral-based, rapid and non-invasive, imaging technology.
Subobjective 4.B: Develop methods to detect and prevent contamination of alternative proteins from plant-based sources with fungal toxins.
Subobjective 4.C: Modeling mycotoxin contamination in the US and developing a forecast system to forewarn US stakeholders of mycotoxin contamination in corn.
Approach
Aflatoxin (AF) contamination in food and feed crops such as corn, peanut, cottonseed, and tree nuts, caused by Aspergillus flavus, is a global concern that compromises food safety and marketability. Aflatoxins are potent carcinogens and their contamination in food are one of the major causes of liver cancer worldwide. The most efficient and practical approach to reduce pre-harvest AF contamination in corn is the development of resistant lines, the overall goal of this project plan. We will delineate the molecular basis of A. flavus resistance in corn seeds through “systems biology” approach that will involve a combination of transcriptomic, proteomic, and metabolomic analyses. Identification of novel regulatory genes and gene networks that play key roles in host plant resistance against the fungus will contribute to the development of robust markers for use in marker-assisted breeding and/or to the identification of candidate genes for editing. We will apply functional genomics to identify key metabolic pathways (polyamines, carotenoids, flavonoids) that contribute to resistance against A. flavus and AF production. Transgenic corn expressing antifungal proteins/peptides that strongly inhibit A. flavus growth will be generated. In addition, “host induced gene silencing” approach will be used to target fungal genes that are critical for growth, pathogenesis, and production of AFs. In addition to food crops, safety of alternative proteins of plant origin from fungal toxins will also be appraised. Using artificial intelligence (AI) and machine learning (ML), algorithmic models will be constructed to predict in advance possible mycotoxin breakout so that timely mitigation efforts can be employed. Finally, non-destructive, hyperspectral-based imaging systems for several platforms will be refined and commercialized to detect AF contamination in stored kernels. The knowledge and products generated from this research will be invaluable for the consumers, stakeholder groups, scientific community, and regulatory agencies to protect and preserve food safety in the United States and abroad.
Progress Report
The research objectives, which fall under National Program 108 Food Safety, Component 1, Foodborne Contaminants, are designed to understand the preharvest aflatoxin (AF, a toxic and carcinogenic compound) contamination of important food and feed crops, such as corn, caused by the fungus, Aspergillus (A.) flavus, and develop effective mitigation strategies.
To accomplish these objectives, it is important first to identify and evaluate native corn genes (Objective 1) and metabolites (Objective 2) that contribute to resistance to AF contamination. We previously used molecular genetics approaches to identify several regions in corn chromosomes with roles in resistance against A. flavus infection and AF contamination have been identified previously. ARS researchers at New Orleans, Louisiana also conducted a comprehensive meta-analysis (examination of data from a number of independent studies plus our own analysis) of all genomic regions reported thus far and identified 58 regions of interest distributed over all 10 corn chromosomes and identified fourteen high-confidence genes that responded only to A. flavus infection and were not reported previously. Further evaluation of DNA markers linked to these genes suggested that three markers discriminated 14 A. flavus resistant cultivars from eight susceptible varieties. These genes are extremely useful for developing AF resistant corn varieties through marker-assisted breeding or targeted genome editing. We also identified 39 microRNAs (miRNAs) that were found to express differently in these varieties. These newly identified microRNAs and target genes, along with the ones identified from the meta-analysis, will be used as targets for future gene editing work to develop corn varieties with AF resistance, in collaboration with scientists at Louisiana State University, Baton Rouge, Louisiana (Agreement #6054-42000-027-004S).
Recently, we conducted a meta-genetic analysis in collaboration with scientists at Louisiana State University to investigate 39,756 corn genes differentially expressed in response to pathogenic and environmental stress. We prioritized the genes for further study using machine learning (ML) models. Some of the genes we identified were previously shown to be involved in corn resistance to Aspergillus flavus, increasing our confidence in the new ML methods. Future studies aim to validate the roles of the 19 unique top-ranked genes, as predicted by the ML models, which greatly increases the potential for developing multiple stress-resistant corn varieties including resistance to AF contamination. All of this information has been shared with ARS and university breeders for advancing corn germplasm.
In Objective 2, we continued to evaluate the roles of native corn metabolites or endogenously produced chemicals such as, flavonoids, provitamin A (or carotenoids) and polyamines (ubiquitous nitrogenous molecules), in reducing A. flavus growth and AF contamination. The mechanism of action by flavonoids on the fungus has been elucidated. We examined the gene expression response of A. flavus after being treated with four different flavonoids and found many genes related to cell wall and cell membrane were upregulated. This supports our previous finding that flavonoids can modulate proliferation of the fungus as well as damage the cell wall. We have previously determined that the role of polyamines in plant disease resistance depends on the relative abundance of complex polyamines (such as, spermidine and spermine versus simpler diamines and breakdown products). Experiments designed to increase the levels of complex polyamines in corn by over-activation of the gene responsible for making the enzyme that produces the polyamine (S-adenosylmethionine decarboxylase or SAMDC) have been completed. Analysis of these corn lines with increased polyamine content will be analyzed for resistance to AF contamination after advancing the corn line gene uniformity (homozygosity) by self-pollination. We have also started evaluating AF resistance as influenced by different levels of carotenoids or provitamin A using the recently acquired corn germplasm. Once APHIS permits are obtained, field evaluations will be carried out and germplasm will be shared with breeders.
Utilizing the results from Objectives 1 and 2 and from the sister project 6054-41420-009-00D, we continued our molecular breeding of corn for resistance to AF contamination un-der the Objective 3 including corn transformation (introducing foreign genes into corn), work that inserts a known antifungal gene(s) in corn. For example, genes for two antifungal synthetic peptides (small proteins), which demonstrated less or no toxicity to animals and human beings, with broad spectrum antimicrobial activity, were introduced into corn. This work was performed in collaboration with Genvor LLC (Agreement #6054-42000-027-003C). We developed transgenic corn lines expressing a synthetic peptide - GV185 and transgenic lines are being evaluated for resistance to AF. We are also studying the mode of action of synthetic peptides at the molecular level by treating transgenic seeds expressing the peptides with Aspergillus and a non-treated control.
We continued to generate transgenic corn lines capable of turning off the expression of several key A. flavus genes responsible for growth, infection, and toxin production using RNA interference (RNAi) approaches. . Recently in collaboration with scientists in Louisiana State University, Baton Rouge, Louisiana, we evaluated several RNAi lines under field conditions and demonstrated significant reduction (72%) in AF contamination. The resistance trait was also successfully transferred to elite corn lines by crossing, resulting in a significant reduction in AF contamination as well. Best performing lines will be multiplied and shared with ARS and university breeders for field evaluations at multiple locations.
In addition to developing crops resistant to AF contamination, ARS researchers in New Orleans, Louisiana, in collaboration with the Department of Agricultural and Biological Engineering at Mississippi State University (MSU) (Agreement # 6054-42000-027-006S) developed a non-invasive, inexpensive and rapid hyperspectral and multispectral imaging methodology that collects and processes information across the light-spectrum. This imaging technique aims to detect and quantify AFs in corn kernels produced under Objective 4A. We conducted co-inoculation experiments in the field, where randomly selected corn ears were inoculated with A. flavus or Fusarium moniliforme separately, or co-inoculated with both fungi concurrently for eight weeks. We also conducted a parallel experiment in the lab with kernels for two weeks-long infection. Based on visible fluorescence under black light illumination we randomly selected infected ears from the field. We isolated the kernels from each treatment group and imaged with the Visual Near Infra-Red instrument under UV and halogen illumination. Currently, we are analyzing the image data for spectral difference and measuring mycotoxin (both aflatoxin and fumonisin) levels We are mirroring the field experiment in a lab-based co-inoculation experiment to study the feasibility of using image-based detection of co-existing mycotoxins, such as AF and fumonisin.
Under Objective 4B, we evaluated the safety of protein-rich legumes (or pulses) from mycotoxin contamination. These pulses are becoming popular with health-conscious individuals as alternative sources of proteins. Inoculation of chickpeas, lentils, and peas with toxin-producing A. flavus showed that legume seeds are resistant to AF contamination, even when they showed limited fungal growth on the seedcoats. Inoculation of corn kernels, on the other hand, produced higher levels of AFs compared to pulses. While pulses are less susceptible to AF contamination than corn, we are also determining whether other AF and other toxins are produced by the fungus when seeds were inoculated for two weeks. We recently completed several greenhouse experiments with inoculated pulse pods (soybean as an infection resistant positive control, lentils, chickpea and peas plus corn as infection sensitive negative control) at various developmental growth and development of pods. Further analyses of fungal growth and toxin levels will shed light on the vulnerability or resistance of pulses to mycotoxin contamination.
In support of Objective 4C, we developed an online decision support tool that uses predictive data analytics, machine learning (ML), and artificial intelligence (AI) to describe the potential for aflatoxin contamination. Region-specific mycotoxin contamination risks can be assessed using big data sets such as historical and forecasted weather sets, satellite collected parameters, and other sources of data. The overall objective is to develop computational predictive models that can be used to develop a web-based US-centric alert system for possible contamination of corn with mycotoxins. This online resource would be beneficial to U.S. farmers, producers, and other stakeholders. Initial collaboration with scientists at Wageningen University in the Netherlands (Agreement #6054-41420-009-006S) was very helpful in developing the prediction models. We have developed and published effective models for three corn-growing states – Illinois, Iowa and Texas. In collaboration with scientists at University of Texas at Arlington, Texas (Agreement # 6054-42000-027-007S), data collected from corn field sites on soil depth, calcium levels, water availability and soil organic matter on mycotoxin contamination have helped to refine and improve model prediction accuracy. Data collection and sampling from Ohio-grown corn is currently underway.
Accomplishments
1. Early detection of harmful fungi can save corn crops in Texas. A team of ARS scientists in New Orleans, Louisiana, in collaboration with scientists from the University of Texas at Arlington, developed a science-based model to help corn farmers and producers accurately predict early outbreaks of aflatoxins, which are produced by fungi that contaminate corn and can cause cancer and harm to animals and humans. U.S. corn farmers and producers can suffer severe economic losses up to $1.6 billion each year across the United States due to aflatoxin outbreaks. This makes early risk prediction especially important to allow for targeted prevention and mitigation strategies so farmers and producers can save their crops. The research team developed a model to predict aflatoxin outbreaks in Texas-grown corn, using data from remote sensing satellites, soil properties, and meteorological data. The model covers the entire state and its four distinctly different agro-ecoregions in Texas. The prediction accuracy currently stands at 73% and will be improved with additional input from Texas corn producers.
2. Biotech corn controls aflatoxin contamination. Aflatoxin contamination in corn, caused by the fungus Aspergillus flavus, is a global concern that compromises food safety and marketability. ARS scientists in New Orleans, Louisiana, in collaboration with scientists from Louisiana State University, Baton Rouge, have produced several transgenic corn lines which are capable of stopping the growth of Aspergillus flavus and/or aflatoxin production. RNA interference techniques were used to silence fungal genes responsible for growth and/or genes involved in aflatoxin production. This inhibition of gene expression is done at the RNA level; as such there is no transfer of DNA to the progeny plants. ARS scientists have silenced nine growth-oriented or toxin producing genes in A. flavus, one at a time. Significant reduction in aflatoxin levels has been observed in each case. We also demonstrated the inheritance of resistance when crossed with commercial corn lines. After field evaluation with APHIS permits, best performing varieties will be released to breeders.
3. Safety of protein-rich pulses from aflatoxin contamination. Toxin producing fungi, Aspergillus and Fusarium, pose significant threats to food safety and marketability. Crop-specific differences in susceptibility to these fungi can influence contamination levels. Pulses or legumes, including chickpeas, lentils, and peas, are highly nutritious and protein-dense food products that are gaining popularity due to global demand. Despite increased usage in human diets, relevant information on the safety of these pulses from aflatoxin contamination is not available. ARS scientists in New Orleans, Louisiana assayed pulse seeds in the laboratory for aflatoxin contamination and determined that they did not produce higher amounts of aflatoxins compared to corn and other grains, and thus the risks of aflatoxin contamination are low.
Review Publications
Castano-Duque, L.M., Avila, A., Mack, B.M., Winzeler, H.E., Blackstock, J.M., Lebar, M.D., Moore, G.G., Owens, P.R., Mehl, H.L., Su, J., Lindsay, J.A., Rajasekaran, K. 2025. Prediction of aflatoxin contamination outbreaks in Texas corn using mechanistic and machine learning models. Frontiers in Microbiology. 16. Article 1528997. https://doi.org/10.3389/fmicb.2025.1528997.
Tao, F., Yao, H., Hruska, Z., Rajasekaran, K., Qin, J., Kim, M., Chao, K. 2024. Raman hyperspectral imaging as a potential tool for rapid and nondestructive identification of aflatoxin contamination in corn kernels. Journal of Food Protection. 87(9). 100335. https://doi.org/10.1016/j.jfp.2024.100335.
Omolehin, O., Raruang, Y., Hu, D., Han, Z., Promyou, S., Brown, R.L., Wei, Q., Rajasekaran, K., Cary, J.W., Wang, K., Jeffers, D., Chen, Z. 2024. Host-Induced Gene Silencing of the Aspergillus flavus O-methyl Transferase Gene Enhanced Maize Aflatoxin Resistance. Toxins. 17(1). https://doi.org/10.3390/toxins17010008.
Branstad-Spates, E., Castano-Duque, L.M., Mosher, G., Hurburgh, Jr, C., Rajasekaran, K., Owens, P.R., Winzeler, H.E., Bowers, E. 2024. Predicting fumonisins in Iowa corn: gradient boosting machine learning. Cereal Chemistry. https://doi.org/10.1002/cche.10824.
Castano-Duque, L.M., Lebar, M.D., Mack, B.M., Lohmar, J.M., Carter Wientjes, C.H. 2024. Investigating the impact of flavonoids on aspergillus flavus: Insights into cell wall damage and biofilms. The Journal of Fungi. 10(9). 665. https://doi.org/10.3390/jof10090665.
Branstad-Spates, E.H., Sickler, C.M., Lebar, M.D., Carter Wientjes, C.H., Rajasekaran, K. 2025. Evaluation of aflatoxin contamination in protein-rich pulses using a GFP-expressing Aspergillus flavus strain. Frontiers in Microbiology. 16. Article 1587035. https://doi.org/10.3389/fmicb.2025.1587035.
Pradhan, A., Gandham, P., Rajasekaran, K., Baisakh, N. 2025. Predictive prioritization of genes significantly associated with biotic and abiotic stresses in maize using machine learning algorithms. Frontiers in Plant Science. 16 Article 1611784. https://doi.org/10.3389/fpls.2025.1611784.