Optimization of Ultrasound-Assisted Extraction of (-)-Stepholidine from Onychopetalum amazonicum Leaves Using Response Surface Methodology.

(-)-Stepholidine, a naturally occurring alkaloid found in the leaves of Onychopetalum amazonicum (Annonaceae), has shown significant pharmacological potential and serves as an important precursor in the synthesis of bioactive compounds. Currently, ultrasound-assisted extraction (UAE), combined with optimization through response surface methodology (RSM), has emerged as a promising strategy to enhance the efficiency of extraction processes. This study aimed to develop an efficient method for extracting (-)-stepholidine from the leaves of O. amazonicum. Powdered leaves were extracted using methanol-based systems assisted by ultrasound, following a central composite rotatable design (CCRD) with three factors: plant-to-solvent ratio (X1), methanol concentration (X2), and extraction time (X3). The samples were quantified by NMR using the PULCON method. The model was validated by ANOVA, and the optimal conditions were determined using RSM. The factors X2 2 (methanol percentage), X3 (extraction time), and the interaction X1·X3 (plant-to-solvent ratio × extraction time) significantly influenced (-)-stepholidine concentration (p < 0.05). The model showed a good fit (R2 = 0.728) and was statistically significant (p = 0.0005). The optimal conditions identified were as follows: a plant-to-solvent ratio of 1:10, 100% methanol, and 20 min of extraction. Experimental validation yielded an average (-)-stepholidine concentration of 82.8 ± 1.3 mg per g of extract, close to the predicted value of 81.2 mg/g. This study demonstrates that optimizing extraction parameters, such as the plant-to-solvent ratio, methanol percentage, and extraction time, is crucial for maximizing (-)-stepholidine recovery. These findings also support the potential of UAE combined with qNMR as a reliable and reproducible approach for the extraction and quantification of bioactive compounds from plant matrices.