To develop a generally applicable method to determine the encapsulation efficiency of proteins loaded in polymer matrices using a co-solvent system where one solvent is primarily chosen for its ability to dissolve the polymer while the other solvent is selected for its ability to precipitate the released protein.
Hemoglobin (Hb) and bovine serum albumin (BSA) were chosen as model proteins and ethylcellulose (EC, 100 cps, Dow Chemical Co.) and PLGA (65:35, molecular weight 40,000-75,000, Sigma-Aldrich) as model polymers. Microcapsules were manufactured by preparing a w/o emulsion of the aqueous solution of proteins in an organic solution of the polymers in dichloromethane (DCM). The emulsion was spray dried using a Büchi, model B-290 Mini Spray Dryer. The resultant microcapsules were collected and stored in a dessicator at 4oC.
Encapsulation efficiency was determined by initially soaking 30 mg of microcapsules in either 2 mL of the DCM : ethanol (1:1) solution (EC microcapsules) or DCM : ethyl acetate (1:1) solution (PLGA microcapsules) for 3 hours followed by vortexing for 5 mins in order to dissolve the polymer coat. The liberated protein was simultaneously precipitated by ethanol or the ethyl acetate and the precipitate was collected by centrifuging the suspension at 20,000 x g for 15 mins. The supernatant was decanted and the precipitate washed by repeating the process once more with another 2 mL aliquot of the cosolvent solution to remove any remaining polymer trapped in the precipitated protein. Finally, the precipitated protein was resolubilized in 1 mL of a solution containing 10 % w/v sodium dodecyl sulfate : 0.8 N sodium hydroxide: water (1:1:2). Protein content in the solution was determined with the Lowry assay.
Validation samples were prepared using blank microcapsules to determine if our process was able to completely recover encapsulated protein and if the magnitude of recovery was similar at low and high encapsulation efficiencies. Blank microcapsules were prepared as outlined above but by replacing the protein solutions with 0.1 M Tris buffer. For the validation experiments, we soaked 30 mg aliquots of blank microcapsules in either 2 mL of the DCM : ethanol solution (blank EC microcapsules) or DCM : ethyl acetate solution (blank PLGA microcapsules) for 3 hours followed by vortexing for 5 mins. Then we exogenously added 0.22 - 0.38 mg of Hb and BSA separately to one set of tubes and 0.66 – 1.05 mg of Hb and BSA separately to another set of tubes. These amounts represented the amount of proteins that would be present in the dissolved polymer solutions if the proteins were encapsulated at efficiencies of 30% and 90 %. The validation samples representing low and high encapsulation efficiencies were processed as outlined above. Simultaneously, we analyzed control samples that contained 0.22 - 0.38 mg and 0.66 – 1.05 mg of Hb and BSA that were directly dissolved in 1 mL of the resolubilizing solution. Since these control samples were not subject to protein precipitation and resolubilization of the precipitated protein, they represented 100 % recovery of proteins and served as a basis to calculate recoveries of the low and high encapsulation efficiency validation samples.
The recovery of Hb and BSA from the validation EC samples simulating low and high encapsulation efficiencies were 94 – 96 % while recoveries of Hb and BSA for the validation PLGA samples were 96 – 101 %. These results indicate that our process was able to quantitatively precipitate the proteins and then completely resolubilize them. The encapsulation efficiency of proteins encapsulated in EC and PLGA microcapsules were measured on three separate days to assess inter-day reproducibility. The mean (s.d.) entrapment efficiency for Hb in EC microcapsules was 52.4 (1.8) %, whereas that for BSA was 86.4 (1.8) %. The entrapment efficiency for Hb in PLGA microcapsules was 76.9 (4.6) % and that for BSA was 74.7 (6.4) %.
We have developed a method to measure the encapsulation efficiency of BSA and Hb in EC and PLGA microcapsules. The selected organic cosolvent systems will dissolve the polymer coat and simultaneously precipitate the released proteins. The precipitated proteins are isolated by centrifugation, completely dissolved in the sodium dodecyl sulfate / sodium hydroxide solution, and quantified using the Lowry Assay. Our method is highly reproducible, alleviates limitations of current methods, and is generally applicable to measure the encapsulation efficiency of proteins in polymer matrices.
Anil D'mello– Professor of Pharmaceutical Sciences, University of the Sciences in Philadelphia, Pennsylvania