(226h) Integrated Computational and Experimental Study to Dissect the Stress Response of Maize Root

Due to its widespread use as a food source and biofuel feedstock, maize (Zea mays) is considered as one of the most important crops. The growth of maize largely depends on its nutrient uptake through root from the soil. Hence, to study the growth, response, and associated metabolic reprogramming of maize-root to different stress conditions is becoming a seminal research direction. Although there are experimental studies on different aspects of maize-root metabolism under different stress conditions, still a holistic approach to study maize-root metabolism under different stress conditions is missing. To address that issue, a genome scale metabolic model (GSMM) for the maize-root of line B73 was reconstructed to study maize-root growth under nitrogen and phosphorus starvation conditions. The model was built based on the available information from public databases such as UniProt, KEGG, and MaizeCyc. This maize-root model included 6368 genes, 4002 reactions, and 4419 metabolites and a detailed gene-protein-reaction association. In parallel to the model reconstruction, transcriptomics and metabolomics data was generated from the roots of hydroponically grown maize plants. Upon incorporation of the transcriptomics data as regulatory constraints, the model was simulated under full-nutrient and nitrogen-starvation conditions. The model predicted an increased biomass growth in the nitrogen starvation condition and inferred that metabolite pool size of asparagine can be used as a key indicator of root biomass growth. Furthermore, the model identified several phosphatidyl components of biomass that are not coupled with the biomass growth and could play a key regulatory role in the biomass growth for nitrogen-starvation condition. To further examine the outcome of the model, non-overlapping flux-sum variability ratios of different metabolites were compared to the experimental measurements. This analysis revealed that the incorporation of transcriptomics data significantly improved the model’s capability to correctly predict fold changes of 2-oxoglutarate (2-OG), succinic acid, malic acid, carbohydrates, and different growth-associated amino acids from the metabolomics data. Overall, upon integration of transcriptomics data, the model prediction significantly improved and achieved 70% accuracy when comparing to the metabolite levels of these conditions with the experimentally measured ones. To study the dynamic behavior of the maize plants grown in the solid-medium, another set of root-specific transcriptomics data was generated over a three-week period under full-nutrient, nitrogen-starvation, and phosphorus-starvation conditions. Upon similar incorporation of this dynamic transcriptomics dataset, the model predicted nitrogen-starved root-biomass growth outpaced the full-nutrient root-biomass-growth in the third week, whereas phosphorus-starved root-biomass growth did the same in the second week. Fatty acid and amino acid metabolism played a key role in outpacing biomass growth in these stress conditions. Overall, the maize root model revealed important metabolic reprogramming under nitrogen and phosphorus starvation conditions which were eventually linked to the observed root phenotypes under these stress condition. Hence, an omics-integrated GSMM provides a promising tool to facilitate stress-condition analysis for maize root and can help engineer a better stress tolerant maize line in the future.