Microfluidic Preparation and Evaluation of Multivesicular Liposomes Containing Gastrodin for Oral Delivery across the Blood–Brain Barrier

In this study, multivesicular liposomes (MVLs) were prepared by microfluidic technology and used for delivering gastrodin (GAS), a water-soluble drug, across the blood-brain barrier (BBB). The formulations and preparation parameters in preparing gastrodin multivesicular liposomes (GAS-MVLs) were both optimized. Some properties of GAS-MVLs including morphology, particle size, encapsulation efficiency, and <i>in vitro</i> release were evaluated. An <i>in vitro</i> BBB model was established by coculturing mouse brain endothelial cells (bEnd.3) and astrocytes (C8-D1A). The permeability of GAS-MVLs across the BBB model was evaluated. Finally, the permeability of GAS-MVLs across BBB was evaluated by <i>in vivo</i> pharmacokinetics in mice. The concentrations of GAS in the blood and brain were determined by high-performance liquid chromatography (HPLC), and then brain-targeting efficiency (BTE), relative uptake rate (Re), and peak concentration ratio (Ce) were calculated. The results showed that, using a Y-type microfluidic chip and setting the flow rate ratio of the second aqueous phase to the W/O emulsion phase at 23, with a total flow rate of 0.184 m/s, the prepared GAS-MVLs showed an obvious multivesicular structure and a relatively narrow distribution of particle sizes. The prepared GAS-MVLs were spherical with a dense structure. The average particle size was 2.09 ± 0.17 μm. The average encapsulation rate was (34.47 ± 0.39)%. The particle size of MVLs prepared by the microfluidic method was much smaller than that prepared by the traditional method, which was usually larger than 10 μm. After 6 h from the beginning of the administration, the apparent transmittance of GAS-MVLs in the <i>in vitro</i> BBB model was 67.71%, which was 1.92 times higher than that of the GAS solution. <i>In vivo</i> pharmacokinetic study showed that the intracerebral area under curve (AUC) of GAS-MVLs was 5.68 times higher than that of the GAS solution, and the e peak concentration (<i>C</i>_max) was 2.036 times higher than that of the GAS solution. BTE was 1.945, intracerebral Re was 5.688, and Ce was 2.036. Both <i>in vitro</i> and <i>in vivo</i> experiment results showed that GAS-MVLs prepared by microfluidic technology in this study significantly delivered GAS across BBB and enriched GAS in the brain. It provides a possibility for brain-targeting delivery of GAS in the prevention and treatment of central nervous system diseases by oral administration and lays the foundation for further development of oral brain-targeted preparations of GAS.