Preparation and characterization of L – Dopa loaded chitosan – based dry powder: Rescue / continuous supplement in Parkinson’s disease via inhalation

Abstract

L – Dopa (LD) oral administration for controlling Parkinson's disease (PD) motor symptoms at the later stage will become inefficient pharmacodynamically due to the motor deteriorations result from LD fluctuated pharmacokinetic parameters especially gastrointestinal absorption and short plasma half-life. The goal from this work was preparing LD loaded chitosan-based dry powder nanoparticles (LD-CSNPs) for pulmonary delivery and hence reducing the dose required and frequency of administration. By employing Box-Behnken experimental design (BBD) for optimizing the dependent variables that include; the particle size (PS) [Y₁] and encapsulation efficiency (EE) [Y₂] of LD NPs prepared from chitosan (CS) by ionotropic gelation method, at different combinations of independent variables that include; CS concentration [X₁], chitosan/tripolyphosphate (CS/TPP) mass ratio [X₂] and LD/CS mass ratio [X₃]. Ultracentrifuge, dynamic light scattering and spectrophotometer were used to measure NPs EE, size and LD quantification respectively. The optimum conditions were determined by subsequent regression and multicriteria decision analyses of the output data. The independent variables had interactive effects and greatly affect both responses. The optimum conditions for NPs production are the CS concentration [X₁] of 1.2-1.6 mg/ml, CS/TPP mass ratio [X₂] of 6-7 and LD/CS mass ratio [X₃] of 1.8-2, which yielded NPs with PS range between 243-266 nm and EE% range between 53-58%. The x-ray powder diffraction (XRPD) shows that LD low intense peak reveal it's dispersion in a homogenous pattern and hence its nature becomes amorphous within the carrier and the Fourier transform infrared (FTIR) spectra shows that no physicochemical interaction occurs between the LD and the carrier. Korsmeyer-Peppas model was the best to fit the in vitro release data, the non-Fickian (anomalous) diffusion was the mechanism of release and the Weibull 4 model was the best for dissolution data fitting where the curve of release was parabolic (b<1, case 3). This procedure has optimized the LD-CSNPs PS and EE.