NMR-BASED METABOLOMICS REVEALS DROUGHT RESPONSE IN GENETICALLY MODIFIED SUGARCANE

Sugarcane (Saccharum spp.) is an important economic crop in tropical and subtropical regions, widely used to produce sugar, ethanol, and bioenergy. However, water deficit is one of the main limiting factors for its productivity¹^,². Genetic improvement is one of the key agronomic strategies to enhance productivity³. In this context, the overexpression of the NPK1 gene—known for modulating responses to abiotic stress—may increase drought tolerance⁴. To understand the plant metabolic response under stress, NMR-based metabolomics and chemometrics are essential tools to investigate metabolic changes in conventional and transgenic plants under water deficit. For this purpose, three genotypes were evaluated: conventional plants (CONT) and genetically modified plants overexpressing the NPK1 gene by two transformation methods—Agrobacterium tumefaciens (AGRO) and biolistics (CV), after 200 days of planting. All samples were collected across six time points: T0, T1, T2, T3, T4 (progressive water deficit intervals), and T5 (rehydrated). Leaf samples were extracted using water and methanol. Extracts were dried and reconstituted in D₂O (aqueous extracts) or MeOD (methanolic extracts), for subsequent analysis on a Bruker AVANCE NEO 500 MHz NMR spectrometer (11.75 T), equipped with a 5 mm Smartprobe® detection probe. The ¹H NMR spectra were acquired at 20 °C using zg30, NOESYGPPR-1D, and CPMGpr pulse sequences, with 128 scans, a spectral width of 25 ppm, a total of 64k data points, totaling approximately 17 minutes per sample. Two-dimensional experiments (HSQC ¹H–¹³C, HMBC ¹³C–¹H, TOCSY ¹H–¹H, COSY ¹H–¹H, and JRes) were also employed to assist metabolite identification. ¹H NMR data were analyzed using supervised algorithms (Sparse Partial Least Squares – Discriminant Analysis, sPLS-DA), with AMIX® and MetaboAnalyst software. Among the tested sequences, NOESYGPPR1D demonstrated superior solvent signal suppression and signal-to-noise ratio. From the ¹H NMR data, 39 metabolites were identified in aqueous extracts and 18 in methanolic extracts, classified as organic acids, amino acids, carbohydrates, and other compounds. sPLS-DA results revealed that CONT plants utilize metabolites such as antioxidants (p-coumaric acid derivative (DAC) and glycosylated apigenin (APGG)), which act primarily against reactive oxygen species (ROS), in addition to amino acids and sugars that contribute to plasma membrane stabilization. Unlike conventional plants, CV transgenic plants mitigate ROS by activating a metabolic pathway featuring trigonelline and APGG, along with amino acids and sugars that contribute to membrane protection. In contrast, AGRO transgenic plants relied mainly on primary metabolites (triacylglycerols and sugars) to defend against drought stress, while also activating the metabolic pathway involving APGG as a key antioxidant against ROS accumulation. In conclusion, each genotype adopted distinct metabolic strategies in response to prolonged drought exposure. AGRO transgenic plants exhibited a more effective metabolic adaptation to water deficit, suggesting greater drought tolerance relative to other genotypes.