Dissecting trichoderma antagonism: role of strain identity, volatiles, biomass, and morphology in suppressing cacao pathogens

Authors: Baek, Insuck; BHATT, JISHNU - Orise Fellow; Jang, Jae Hee; SEUNGHYUN, LIM - Orise Fellow; Lovelace, Amelia; MINHYEOK, CHA - Orise Fellow; LAKSHMAN, DILIP - Retired ARS Employee; Kim, Moon; Meinhardt, Lyndel; Park, Sunchung; Ahn, Ezekiel

Submitted to: Biological Control
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 5/29/2025
Publication Date: 6/2/2025
Citation: Baek, I., Bhatt, J., Jang, J., Seunghyun, L., Lovelace, A.H., Minhyeok, C., Lakshman, D., Kim, M.S., Meinhardt, L.W., Park, S., Ahn, E.J. 2025. Dissecting trichoderma antagonism: role of strain identity, volatiles, biomass, and morphology in suppressing cacao pathogens. Biological Control. https://doi.org/10.1016/j.biocontrol.2025.105807.
DOI: https://doi.org/10.1016/j.biocontrol.2025.105807

Interpretive Summary: Cacao, the source of chocolate, is a vital crop for millions of farmers, but its production is severely hampered by fungal diseases like anthracnose and leaf blight, caused by pathogens such as Colletotrichum and Pestalotiopsis. Current reliance on chemical fungicides raises concerns about cost, environmental impact, and the development of resistance. This study explored a more sustainable approach using beneficial fungi from the Trichoderma genus as natural enemies (biocontrol agents) against these cacao pathogens. We investigated the effectiveness of three different biocontrol isolates against six harmful fungal strains isolated from cacao in Ghana. Using laboratory tests and advanced image analysis, we measured how well each of the biocontrol isolates stopped the growth of these pathogens, observed how the fungi physically changed during their interaction, and used computer-based machine learning techniques to understand what factors predict a successful outcome. A key finding was that while all tested biocontrol strains could inhibit the pathogens, their effectiveness varied significantly. Our results suggest these beneficial fungi use multiple strategies, including releasing airborne chemicals (Volatile Organic Compounds) that slow pathogen growth and engaging in direct physical competition or attack, as evidenced by the dramatically increased inhibition when a larger amount of the biocontrol agents was present. Machine learning analysis revealed that the most critical factor determining the level of disease control was the specific combination of the biocontrol strain and the pathogen strain involved. Furthermore, the roundness (circularity) of the Trichoderma colony itself was identified as an important predictor of its inhibitory success. This research provides valuable insights for cacao researchers and breeders who can use this information to focus on developing biocontrol strategies using the most effective biocontrol strains. Plant pathologists will gain a better understanding of the complex interactions between beneficial and pathogenic fungi. Ultimately, cacao farmers stand to benefit from the potential development of new, sustainable tools to protect their crops, leading to more stable yields and income. The chocolate industry and consumers also benefit from efforts to secure a sustainable cacao supply.

Technical Abstract: This study investigated the in vitro antagonistic potential and interaction dynamics of three Trichoderma isolates (T. virens 11C-65-1, T. virens 29-8, Trichoderma spp. RC) against six Ghanaian isolates of cacao pathogens (Colletotrichum gloeosporioides and Pestalotiopsis sp.). We employed dual culture assays, quantitative morphological analysis using image processing (SmartGrain) to measure colony area, perimeter, length, width, length-to-width ratio (LWR), and circularity for both antagonist and pathogen. The role of Volatile Organic Compounds (VOCs) was assessed using sealed plate assays comparing standard plug inoculation versus pre-grown biomass, and the effect of UVC pretreatment (275 nm, 30 min) was examined. Multivariate statistics (ANOVA, Tukey's HSD, Pearson correlation, PCA, t-SNE, hierarchical clustering) and machine learning models (including Fit Stepwise, Generalized Regression Lasso, Boosted Tree, Bootstrap Forest, Neural Boosted via nested cross-validation) were utilized to analyze morphological data and predict pathogen colony area. Significant differences in antagonistic efficacy were observed (p < 0.05), establishing a clear hierarchy: T. virens 11C-65-1 > Trichoderma spp. RC > T. virens 29-8. Co-culture induced significant morphological alterations in both interacting fungi, varying with the specific Trichoderma-pathogen pair. Machine learning models predicted pathogen colony area with high accuracy (test set R² up to 0.94). Feature importance analysis identified the specific 'Pair' identity (Trichoderma-pathogen combination) as the dominant predictor (Total Effect = 0.659 in Neural Boosted model), followed by Trichoderma circularity (T-Cir, Total Effect = 0.212). Multivariate analyses confirmed the antagonistic relationship (e.g., negative correlation between T-Area and P-Area, r = -0.85) and highlighted the distinct variance components associated with size (PC1), T-Cir (PC2), and P-Cir (PC3). Hierarchical clustering linked Trichoderma isolate identity strongly with both T-Cir and P-Cir. VOC assays demonstrated significant inhibition by 11C-65-1, which was dramatically enhanced (p < 0.0001) when using pre-grown biomass, suggesting a combined effect of VOCs and direct physical interaction/competition. UVC pretreatment resulted in statistically significant (p < 0.05) but minimal morphological changes under the experimental conditions. These findings underscore the highly strain-specific nature of Trichoderma antagonism against cacao pathogens in vitro and identify T. virens 11C-65-1 as a particularly potent isolate. The results suggest multiple mechanisms contribute to inhibition and highlight the importance of the specific interaction context and antagonist morphology (circularity) in determining outcomes. This research provides a foundation for selecting promising biocontrol candidates for in vivo testing and mechanistic studies, and will be of interest to plant pathologists, mycologists, biocontrol researchers, and those working on sustainable cacao production.