Improving Protein Delivery from Microparticles Using Blends of Poly(DL lactide co-glycolide) and Poly(ethylene oxide)-poly(propylene oxide) Copolymers

Purpose. Microparticles containing ovalbumin as a model for protein drugs were formulated from blends of poly(DL lactide-co-glycolide) and poly(ethylene oxide)-poly(propylene oxide) copolymers (Pluronic). The objectives were to achieve uniform release characteristics and improved protein delivery capacity. Methods. The water- in oil -in oil emulsion/solvent extraction technique was used for microparticle production. Results. A protein loading level of over 40% (w/w) was attained in microparticles having a mean diameter of approximately 5 µm. Linear protein release profiles over 25 days in vitro were exhibited by certain blend formulations incorporating hydrophilic Pluronic F127. The release profile tended to plateau after 10 days when the more hydrophobic Pluronic L121 copolymer was used to prepare microparticles. A delivery capacity of 3 µg OVA/mg particles/ day was achieved by formulation of microparticles using a 1:2 blend of PLG:Pluronic F127. Conclusions. The w/o/o formulation approach in combination with PLG:Pluronic blends shows potential for improving the delivery of therapeutic proteins and peptides from microparticulate systems. Novel vaccine formulations are also feasible by incorporation of Pluronic L121 in the microparticles as a co-adjuvant.     

Pharmaceutical Evaluation of Gas-Filled Microparticles as Gene Delivery System

Purpose. To produce and characterize a nonviral ultrasound-controlled release system of plasmid DNA (pDNA) encapsulated in gas-filled poly(D,L-lactide-co-glycolide) microparticles (PLGA-MPs). Methods. Different cationic polymers were used to form pDNA/polymer complexes to enhance the stability of pDNA during microparticle preparation. The physico-acoustical properties of the microparticles, particle size, pDNA integrity, encapsulation efficiency and pDNA release behavior were studied in vitro. Results. The microparticles had an average particle size of around 5 m. More than 50% of all microparticles contained a gas core, and when exposed to pulsed ultrasound as used for color Doppler imaging create a signal that yields typical color patterns (stimulated acoustic emission) as a result of the ultrasound-induced destruction of the microparticles. Thirty percent of the pDNA used was successfully encapsulated and approximately 10% of the encapsulated pDNA was released by ultrasound within 10 min. Conclusions. Plasmid DNA can be encapsulated in biodegradable gas-filled PLGA-MPs without hints for a structural disintegration. A pDNA release by ultrasound-induced microparticle-destruction could be shown in vitro.     

The characteristics of betamethasone-loaded chitosan microparticles by spray-drying method

Betamethasone (BTM)-loaded microparticles prepared by a spray drying method using chitosan (CTS) as raw material, type-A gelatin and ethylene oxide-propylene oxide block copolymer (Pluronic F68) as modifiers. The BTM-loaded in varied chitosan/Pluronic F68/gelatin microparticle formulations was investigated. By properly choosing excipient type and concentration a high degree of control was achieved over the physical properties of the BTM-loaded microparticles. Microparticle characteristics (zeta potential, tap density, particle size and yield), loading efficiencies, microparticle morphology and in-vitro release properties were examined. Surface morphological characteristics and surface charges of prepared microparticles were observed by using scanning electron microscopy (SEM) and microelectrophoresis. A SEM micrograph shows that the particle sizes of the varied chitosan composed microparticles ranged from 1.1-4.7 microm and the external surfaces appear smooth. The BTM-loaded microparticles entrapped in the chitosan/Pluronic F68/gelatin microparticles with trapping efficiencies up to 93%, collected yield rate 44%, and mean particle size varied between 1-3 microm, positive surface charge (20-40 mv), and tap densities (0.04-0.40 g/cm3) were obtained. The collected BTM yield and size of particle was increased with increasing BTM-loaded amount but both zeta potential and tap density of the particles decreased with increasing BTM-loaded amount. The in vitro release of BTM showed a dose-dependent burst followed by a slower release phase that was proportional to the drug concentration in the concentration range between 5-30%w/w. The in vitro drug release from the chitosan/Pluronic F68/gelatin 1/0.1/0.4 microspheres had a prolong release pattern. These formulation factors were correlated to particulate characteristics for optimizing BTM microspheres in pulmonary delivery.     

Topical delivery of urea encapsulated in biodegradable PLGA microparticles: O/W and W/O creams

This study describes the formulation and characterization of O/W and W/O creams containing urea-loaded microparticles prepared with poly (D, L-lactic-co-glycolic acid) (PLGA) in order to encapsulate and stabilize urea. The solvent evaporation method was used for preparing PLGA microparticles containing urea. The microparticles size was evaluated by laser light diffractometry. The resulting microparticles were then incorporated in O/W and W/O creams and stability and the release pattern from the creams was evaluated by UV-spectrophotometry. The particle size of PLGA microparticles was in the range of 1–5 µm and most microparticles had a particle size smaller than 3 µm. The encapsulation efficiency was calculated as 40.5%?±?3.4. This study also examined release pattern of urea which varied among different formulations. The results showed that the release from O/W creams followed Higuchi kinetics while the release from W/O creams showed the zero order kinetics and the creams containing microparticulated urea had slower release than free urea creams.     

Effect of structural relaxation on the preparation and drug release behavior of poly(lactic-co-glycolic)acid microparticle drug delivery systems

Control of burst release is a major challenge in the development of poly(lactide-co-glycolide) (PLGA) microparticle drug delivery systems. It has been well-documented in previous literature that formulation and processing variables determine particle morphology, which in turn, governs drug diffusivity and burst release. However, it is not generally appreciated that PLGA polymers used for microparticle systems are typically amorphous, and as such, undergo structural relaxation during processing and storage, characterized by enthalpy and volume reduction. Volume reduction due to structural relaxation can decrease drug diffusivity within microparticles and affect burst release. The magnitude of the driving force leading to structural relaxation is linked to the rate of particle hardening, and is affected by process parameters. Studies that directly address structural relaxation in PLGA microparticles indicate that the manufacturing process and residual solvent levels, as well as the nature of the interaction between drug and polymer affect the rate of structural relaxation. Therefore, the conditions chosen for particle fabrication may be a major source of variability in the burst release and may affect the stability of the drug release profile during storage. The potential effects of structural relaxation on drug release are likely to be formulation specific. Additional work is required to understand and control the relationship between microparticle processing, structural relaxation, and performance of PLGA microparticle drug delivery systems.     

Development of an Inhaled Sustained Release Dry Powder Formulation of Salbutamol Sulphate,an Antiasthmatic Drug

The present research was aimed to develop and characterize a sustained release dry powder inhalable formulation of salbutamol sulphate. The salbutamol sulphate microparticles were prepared by solvent evaporation method using biodegradable polymer poly (D,L-lactic-co-glycolic acid) to produce salbutamol sulphate microparticle mixed with carrier respirable grade lactose for oral inhalation of dry powder. The drug content were estimated to produce 1 mg sustained release salbutamol sulphate per dose. Total four formulations K1, K2, K3 and K4 were prepared with 1:1, 1:2, 1:3, 1:4 ratio of salbutamol sulphate:poly (D,L-lactic-co-glycolic acid). The developed formulations were studied for physicochemical properties, in vitro drug relase and Anderson cascade impaction studies. The prepared formulations effectively releases drug for 12 h in diffusion bag studies. Based on dissolution performance the 1:1 ratio of salbutamol sulphate:poly (D,L-lactic-co-glycolic acid) produces in vitro release 92.57% at 12 h and having particle size of microparticles (D_0.5μm) 5.02±0.6 and the pulmonary deposition of dry powder 34.5±3.21 (respiratory fraction in percentage).     

Transfection of a mouse dendritic cell line by plasmid DNA-loaded PLGA microparticles in vitro.

Targeting of DC for DNA vaccination may be achieved by DNA-loaded poly(lactide-co-glycolide) (PLGA) biodegradable microparticles, since DC efficiently capture these microparticles in vitro and in vivo. DNA was encapsulated in PLGA microparticles by spray-drying. Various additives were tested and process parameters adjusted in order to prevent degradation of the DNA during encapsulation. The highest degree of supercoiled DNA was maintained by adding a strong buffering agent, such as PBS or NaHCO(3), whereas the cryoprotective lactose did not show a significant protective effect. DNA-containing PLGA microparticles were administered to a mouse DC line. Transfection efficacy was compared with commonly employed cationic transfectants and was visually assessed by green fluorescent protein expression. Transfection rate was very low in DC for all microparticle formulations and was comparable with commonly used cationic transfectants. It is concluded that the transfection of DC using PLGA microparticles is feasible, but efforts need to be undertaken to improve transfection efficiency in vitro, which may in addition lead to improved immune responses in vivo.     

In vivo and in vitro characteristics for insulin-loaded PLA microparticles prepared by w/o/w solvent evaporation method with electrolytes in the continuous phase

Insulin-loaded poly(lactide) (PLA) microparticles were successfully prepared by 6% w/v PLA in the organic phase, 10% w/v PVP and varied types of 5%w/v electrolytes in the continuous phase, by using a water-in-oil-in-water emulsion/solvent extraction technique. Addition of electrolytes such as NaCl, CaCl₂ into the external phase significantly improved insulin entrapment efficiency compared to the case of no additives. NaCl was the most effective for obtaining high entrapment efficiency, with microparticle yield 81.2%, trapping efficiencies 49%, insulin-loading level 5.5% w/w and mean particle size 14.8?µm. The distribution (%) of insulin on the PLA microparticles surface, outer layer and core were 8, 37 and 43%, respectively. The cumulative release of insulin had an upper limit of ～24% of the insulin load at 24 days. A steady release rate was 0.5?µg insulin/mg microparticles/day of insulin release maintained for 24 days. Total protein-leaking amount was reduced after addition of electrolytes in the continuous aqueous phase. Rabbit glucose levels were evaluated after subcutaneous 20?mg insulin-loaded PLA microparticles or PLA blank microparticles. Study results show that the insulin-loaded PLA microparticles significantly reduced the glucose level than PLA blank microparticles. The insulin-loaded PLA microparticles, physicochemical characterization data and the animal result obtained in this study may be relevant in optimizing the PLA microparticle formulation incorporation and delivery insulin carriers.     

Modulation of Polyplex Release from Biodegradable Microparticles through Poly(ethylenimine) Modification and Varying Loading Concentration

Purpose

This work investigates the effects of hyaluronic acid (HA) conjugated onto branched poly(ethylenimine) (bPEI) and varying loading concentrations of these polymers complexed with DNA on their release from poly(DL-lactic-co-glycolic acid) (PLGA) microparticles and the transfection of target cells.

Methods

To examine the effect of alteration of the gene delivery polymer on the system, we observed the morphology, size, loading efficiency, polymer and DNA release, and the transfection efficiency for the microparticles formed with three internal phase loading concentrations during microparticle formation.

Results

Addition of HA to this vector allowed for increased loading concentration within these systems and significantly altered release kinetics without changing the morphology of the particles. The incorporation of HA onto the bPEI backbone significantly increased the transfection efficiency of the complexes released from the corresponding microparticle formulation.

Conclusions

The results show that the modification of bPEI with HA and the concentration of loaded polymer/DNA complexes can significantly alter the entrapment and release profiles from PLGA microparticles. This is significant in that it offers insight into the effects of modification of gene delivery vectors on a controlled release system designed to achieve a sustained therapeutic response.     

Composition and Surface Charge of DNA-Loaded Microparticles Determine Maturation and Cytokine Secretion in Human Dendritic Cells

PURPOSE: Biodegradable microparticles prepared from poly(lactide) (PLA) and poly(lactide-co-glycolide) (PLGA) have been shown to be promising carrier systems for vaccine delivery. Here, we have investigated the capacity of different PLA and PLGA microparticle formulations to induce stimulation of human blood monocyte-derived dendritic cells (DCs). METHODS: Stimulation of human derived dendritic cells by plain microparticles were compared with microparticles loaded with plasmid DNA or double-stranded salmon DNA either by encapsulation or adsorption to the surface of cationic microparticles. Stimulation of DCs was monitored by the up-regulation of surface maturation markers CD83 and CD86 and the secretion of IL-12 and TNF-alpha. RESULTS: Slowly degrading PLA microparticles did not induce any detectable stimulation or activation of DCs. In contrast, fast degrading PLGA microparticles were able to influence DC maturation and cytokine secretion dependent on their surface charge. Anionic PLGA microparticles induced an up-regulation of CD83 and high TNF-alpha secretion, which was further enhanced up to the level of the potent stimulator lipopolysaccharide (LPS) when plasmid DNA was encapsulated. Moreover, the secretion of significant amounts of IL-12 was observed. Cationic PLGA microparticles induced an up-regulation of CD86 and moderate TNF-alpha secretion, but no IL-12 secretion, with no additional effects in the presence of plasmid DNA. CONCLUSIONS: The data suggest that the composition and charge of biodegradable DNA-loaded microparticles profoundly influences maturation and cytokine secretion in DCs. Thus, the individual formulation of microparticles used as a vaccine carrier system might considerably influence the profile of the immune response.     

In vivo and in vitro characteristics for insulin-loaded PLA microparticles prepared by w/o/w solvent evaporation method with electrolytes in the continuous phase

Insulin-loaded poly(lactide) (PLA) microparticles were successfully prepared by 6% w/v PLA in the organic phase, 10% w/v PVP and varied types of 5%w/v electrolytes in the continuous phase, by using a water-in-oil-in-water emulsion/ solvent extraction technique. Addition of electrolytes such as NaCl, CaCl2 into the external phase significantly improved insulin entrapment efficiency compared to the case of no additives. NaCl was the most effective for obtaining high entrapment efficiency, with microparticle yield 81.2%, trapping efficiencies 49%, insulin-loading level 5.5% w/w and mean particle size 14.8 microm. The distribution (%) of insulin on the PLA microparticles surface, outer layer and core were 8, 37 and 43%, respectively. The cumulative release of insulin had an upper limit of approximately 24% of the insulin load at 24 days. A steady release rate was 0.5 microg insulin/mg microparticles/day of insulin release maintained for 24 days. Total protein-leaking amount was reduced after addition of electrolytes in the continuous aqueous phase. Rabbit glucose levels were evaluated after subcutaneous 20 mg insulin-loaded PLA microparticles or PLA blank microparticles. Study results show that the insulin-loaded PLA microparticles significantly reduced the glucose level than PLA blank microparticles. The insulin-loaded PLA microparticles, physicochemical characterization data and the animal result obtained in this study may be relevant in optimizing the PLA microparticle formulation incorporation and delivery insulin carriers.     

Poly(epsilon-caprolactone) and Eudragit microparticles containing fludrocortisone acetate

Substitutive hormonal therapies have to be administered for long periods. Thus, the development of sustained-release forms, as microparticle suspensions, is interesting in order to improve patient compliance by reducing dosing frequencies and side effects. The aim of this work was to compare different formulations of fludrocortisone microparticles for the treatment of mineralocorticoid insufficiency. The study was done with different polymers (poly(epsilon-caprolactone), Eudragit RS and Eudragit RL) and different processes (O/W solvent evaporation methods and S/O/W evaporation methods). The use of a suspension of micronized drug in dichloromethane as dispersed phase (S/O/W method) significantly improved the process. Whereas low concentrations of FLU dissolved in the dispersed phase led to smooth-surface homogeneous microparticles and poor incorporation efficiency (5.8-7.3%); suspensions of FLU led to microparticles with numerous crystals on their surfaces (S/O/W microparticles) and high incorporation efficiency (about 79%). However, the best release profiles were obtained with microparticles prepared with 7.5 mg/ml of dichloromethane, near saturation. Moreover, the use of mixtures of poly(epsilon-caprolactone), Eudragit RS and RL did not improve the release profiles.     

Topical delivery of urea encapsulated in biodegradable PLGA microparticles: O/W and W/O creams

This study describes the formulation and characterization of O/W and W/O creams containing urea-loaded microparticles prepared with poly (D, L-lactic-co-glycolic acid) (PLGA) in order to encapsulate and stabilize urea. The solvent evaporation method was used for preparing PLGA microparticles containing urea. The microparticles size was evaluated by laser light diffractometry. The resulting microparticles were then incorporated in O/W and W/O creams and stability and the release pattern from the creams was evaluated by UV-spectrophotometry. The particle size of PLGA microparticles was in the range of 1-5 microm and most microparticles had a particle size smaller than 3 microm. The encapsulation efficiency was calculated as 40.5% +/- 3.4. This study also examined release pattern of urea which varied among different formulations. The results showed that the release from O/W creams followed Higuchi kinetics while the release from W/O creams showed the zero order kinetics and the creams containing microparticulated urea had slower release than free urea creams.     

Formulation and pharmacodynamic evaluation of captopril sustained release microparticles

Cellulose propionate (CP) microparticles containing captopril (CAP) were prepared by solvent evaporation technique. The effects of polymer molecular weight, polymer composition and drug:polymer ratios on the particle size, flow properties, morphology, surface properties and release characteristics of the prepared captopril microparticles were examined. The anti-hypertensive effect of the selected CAP formulation in comparison with aqueous drug solution was also evaluated in vivo using hypertensive rats. The formulation containing drug:polymer blend ratio 1:1.5 (1:1 low:high molecular weight CP), namely F7, was chosen as the selected formulation with regard to the encapsulation efficiency (75.1%), flow properties (theta=24 degrees, Carr index=5%, Hausner ratio=1.1, packing rate=0.535) and release characteristics. Initial burst effect was observed in the release profile of all examined formulations. DSC and SEM results indicated that the initial burst effect could be attributed to dissolution of CAP crystals present on the surface or embedded in the superficial layer of the matrix. The release kinetics of CAP from most microparticle formulations followed diffusion mechanism. After oral administration of the selected microparticle formulation (F7) to hypertensive rats, systolic blood pressure decreased gradually over 24 h compared to reference drug solution. These results may suggest the potential application of cellulose propionate microparticles as a suitable sustained release drug delivery system for captopril.     

Core–shell PLA–PVA porous microparticles as carriers for bacteriocin nisin

This work is focused on preparation of novel porous type of core–shell-structured microparticles based on polylactide (shell) and poly(vinyl alcohol) cross-linked with glutaric acid (GA) (core) prepared by water-in-oil-in-water solvent evaporation technique. The microparticle systems were used as delivery systems for immobilisation of model antibacterial agent – nisin. The effect of cross-linking and the initial amount of nisin on their morphology was investigated using scanning electron microscopy, BET analysis, zeta potential measurement and Fourier transform infra-red spectroscopy. Encapsulation efficiency and release profile of nisin from the microparticles were studied by high performance liquid chromatography. Antibacterial activity of the prepared systems was tested by dilution and spread plate technique. Results showed the microparticles in the size range of 9–16?μm in diameter with spherical multi-hollow core–shell structure. The presence of cross-linking agent GA influences the release profile of the peptide and has synergistic effect on Listeria monocytogenes growth reduction.     

Microparticle-mediated transfection of non-phagocytic cells in vitro

DNA-loaded microparticles represent an attractive delivery system to target professional antigen presenting cells (APC) for the delivery of DNA vaccines. Microparticles exhibiting a positively charged surface were prepared by the incorporation of two selected cationic polymers into a poly(D, L-lactide-co-glycolide) polymer (PLGA) core. The toxicity of the different formulations was checked in two cell lines and was found to be comparable to plain PLGA particles. Increased toxicity of some formulations was observed in primary macrophages (Mphi) with high phagocytosis activity. Plasmid DNA was efficiently adsorbed to the microparticle surfaces, and the different formulations were checked for their transfection efficiency in phagocytic and non-phagocytic cells. Interestingly, the most pronounced gene transfer efficiency was observed in a non-phagocytic 293 cell line when compared to a macrophage cell line and primary Mphi. Possible mechanisms include the dissociation of DNA-polymer complex and subsequent transfection of the cells. Microscopic observation of fluorescent-labeled DNA in primary Mphi revealed large amounts of DNA entering the cells, but no detectable DNA inside the nuclei. We conclude that phagocytic professional APC represent a group of cells, which is especially difficult to transfect when compared to other cell types. The administration of DNA in vivo is likely to predominantly result in the transfection of non-lymphoid cells unless there is a possibility to provide efficient targeting and trafficking of the DNA to the nucleus of professional APC. Although DNA-loaded PEI and DAEM microparticles resulted in significant transfection of cells, toxicity and transfection efficiency was not superior to that of DNA complexed with soluble PEI and DAEM.     

Microparticle-mediated Transfection of Non-phagocytic Cells In Vitro

DNA-loaded microparticles represent an attractive delivery system to target professional antigen presenting cells (APC) for the delivery of DNA vaccines. Microparticles exhibiting a positively charged surface were prepared by the incorporation of two selected cationic polymers into a poly (d, l -lactide-co-glycolide) polymer (PLGA) core. The toxicity of the different formulations was checked in two cell lines and was found to be comparable to plain PLGA particles. Increased toxicity of some formulations was observed in primary macrophages (M Φ) with high phagocytosis activity. Plasmid DNA was efficiently adsorbed to the microparticle surfaces, and the different formulations were checked for their transfection efficiency in phagocytic and non-phagocytic cells. Interestingly, the most pronounced gene transfer efficiency was observed in a non-phagocytic 293 cell line when compared to a macrophage cell line and primary M Φ. Possible mechanisms include the dissociation of DNA-polymer complex and subsequent transfection of the cells. Microscopic observation of fluorescent-labeled DNA in primary M Φ revealed large amounts of DNA entering the cells, but no detectable DNA inside the nuclei. We conclude that phagocytic professional APC represent a group of cells, which is especially difficult to transfect when compared to other cell types. The administration of DNA in vivo is likely to predominantly result in the transfection of non-lymphoid cells unless there is a possibility to provide efficient targeting and trafficking of the DNA to the nucleus of professional APC. Although DNA-loaded PEI and DAEM microparticles resulted in significant transfection of cells, toxicity and transfection efficiency was not superior to that of DNA complexed with soluble PEI and DAEM.     

Manipulation of beclomethasone-hydrofluoroalkane interactions using biocompatible macromolecules

The aim of this work was to physically stabilise beclomethasone dipropionate (BDP) microparticles within a hydrofluoroalkane (HFA) propellant using biocompatible polymers in order to allow the efficient delivery of the steroid to the airways from a pressurised metered dose inhaler (pMDI). BDP microparticles were coated with a number of different "amphiphilic" macromolecular excipients by spray-drying an aqueous BDP suspension in which the excipients were dissolved. The physical stability of the coated BDP microparticles was assessed both indirectly using a twin-stage impinger (TSI) and directly using "in-situ" laser diffraction particle size analysis in a range of nonpolar solvents. The solubility of the formulation excipients within a number of the nonpolar vehicles was determined using an internally manufactured filtration rig and the influence of zeta potential within the microparticle suspensions measured in a series surrogate nonpolar systems. The size of the pure BDP microparticles increased significantly (p < 0.05, ANOVA) from 3.13 +/- 0.15 microm to 9.86 +/- 0.50 microm upon suspension within a nonpolar HFA solvent. However, the addition of poly(vinyl alcohol) (PVA) and poly(vinyl pyrrolidone) (PVP) to the BDP microparticles dramatically reduce this aggregation leading to the production of physically stable suspensions with excellent aerosolisation properties (Stage 2 deposition >40% in the twin-stage impinger). It is postulated that the enhanced physical stability observed when PVA and PVP are coated onto BDP microparticles is partially as a result of steric stabilisation in HFA solvents. However, the large zeta potential associated with the nonpolar microparticle suspensions suggest that charge stabilisation may also influence the physical stability within these systems.     

Chitosan-based macrophage-mediated drug targeting for the treatment of experimental visceral leishmaniasis

The potential of chitosan microparticles as a carrier of doxorubicin for the treatment of visceral leishmaniasis was evaluated by macrophage-mediated drug targeting approach. Cationic charge of doxorubicin was masked by complexing it with dextran sulphate (a poly anion) in order to facilitate its incorporation into cationic chitosan microparticles. Prior to in vitro and in vivo studies, characterization studies were carried out systematically: particle size (～1.049?μm), surface morphology (fluorescence microscopy - spherical structured microparticles), Fourier transform infrared spectroscopy (to characterize effective cross-linking) and differential scanning calorimetry. In vitro studies were carried out in J774.1 in order to check the effective endocytotic uptake of microparticles by macrophages. In vivo studies were conducted in Syrian golden hamsters as per well-established protocols and the results drawn from in vivo studies displayed substantial reduction in leishmanial parasite load for doxorubicin-encapsulated chitosan microparticles: ～78.2?±?10.4%, when compared to the control (free doxorubicin): 33.3?±?2.4%.     

Comparison of protein loaded poly(epsilon-caprolactone) microparticles prepared by the hot-melt technique

This study was systematically designed to compare bovine serum albumin (BSA) loaded poly(epsilon-caprolactone) (PCL) microparticles based on a 2(3) factorial experiment. The microparticles were prepared by the holt-melt technique without using an organic solvent for polymer solubilization. The influence of the particle size of protein, protein/polymer ratio, and hydrophilic PEG on the surface morphology, particle size as well as the yield of PCL microparticles, encapsulation efficiency of BSA, and in vitro release properties were investigated. The structure of BSA remained its integrity using this technique. The mean particle size of BSA-loaded microparticles were in the range of 12.7 +/- 0.1-16.9 +/- 0.8 microm, and all of the particles were smooth on the surface. The production yield of microparticles and the encapsulation efficiencies were high, and the values were in the range of 94.8 +/- 1.6%-98.1 +/- 1.0% and 94.9 +/- 9.6%-98.6 + 0.3%, respectively. The burst release of BSA was in the range of 8.2 +/- 0.4%-61.0 +/- 0.8%, which strongly depended on the formulation. None of three variables affected the yield of microparticles prepared from eight formulations (p > 0.05). However, the particle size of BSA significantly affected the size and the burst release as well as the cumulative release of protein in these microparticles (p < 0.05). The initial loading of BSA in terms of BSA/PCL ratio and the amount of PEG blended with PCL significantly affected all of the properties, except the yield (p < 0.05). The smaller the particle size of the BSA, the smaller the size of the resulting microparticles. Since the total surface area of the small particles was larger than that of the large particles, this accounted for the high burst release of protein from the microparticles encapsulating triturated-BSA.