RESUMO
The process of amygdalin extraction from plum seeds was optimized using central composite design (CCD) and multilayer perceptron (MLP). The effect of time, ethanol concentration, solid-to-liquid ratio, and temperature on the amygdalin content in the extracts was estimated using both mathematical models. The MLP 4-3-1 with exponential function in hidden layer and linear function in output layer was used for describing the extraction process. MLP model was more superior compared with CCD model due to better prediction ability. According to MLP model, the suggested optimal conditions are: time of 120 min, 100% (v/v) ethanol, solid-to liquid ratio of 1:25 (m/v) and temperature of 34.4°C. The predicted value of amygdalin content in the dried extract (25.42 g per 100 g) at these conditions was experimentally confirmed (25.30 g per 100 g of dried extract). Amygdalin (>90%) was isolated from the complex extraction mixture and structurally characterized by FT-IR, UV, and MS methods.
RESUMO
OBJECTIVE: The suitability of the rabbit as an animal model for the primary screening and selection of the pilot scale batches during the early stages of the formulation development was studied. MATERIALS AND METHODS: Three modified-release formulations of aminophylline consisted of Carbopol® 971P/HPMC K4M (F-I), and HPMC K100M (F-II) or HPMC K4M (F-III) were used. Commercial products were Aminofilin retard 350 mg tablets, Srbolek, Serbia (R-I) and Phyllocontin(®) 350, tablets Purdue Frederic, Canada (R-II). RESULTS: Calculated release rate constants and the ƒ2 values between R-I/F-I (84.1) and R-II/F-III (83.4) indicated similar in vitro release while the coefficient n showed presence of different mechanisms of release from Anomalous transport, Fickian diffusion to Case-II transport. Higher Tmax, was found in the rabbits, dosed with F-II (12.00 h), F-III (10.50 h), and R-II (15.00 h) formulation. The highest Cmax (9.22 mg/L) was obtained with F-II, similar lower values was seen for F-I and F-III, while commercial products showed the lowest values R-I (5.58 mg/L) and R-II (4.18 mg/L). Higher AUC values were detected for all three formulations (from 115.90 to 204.06 mgh/L) in relation to commercial products (105.33 and 113.25 mgh/L). DISCUSSION AND CONCLUSION: The results demonstrated a good correlation of Level A (r(2) = 0.97) for the two formulations (F-I, F-III) and commercial product (R-I) indicates that there is a reasonable assumption that the rabbit might be use as a model for the preliminary comparison of scale up formulations in the early stages of the product development.