Hans Koll

Microencapsulation of rh-erythropoietin, using biodegradable poly(d,l-lactide-co-glycolide): protein stability and the effects of stabilizing excipients

Abstract Parenteral delivery systems allowing controlled drug release over one month are of particular interest for proteins and peptides. We investigated the microencapsulation of recombinant human erythropoietin (EPO), a stimulating factor of red blood cell production, into poly( d , l -lactide-co-glycolide) (PLG) microspheres, using a water-in-oil-in-water (W/O/W) double emulsion technique. The integrity and stability of EPO during microencapsulation and storage was characterized. Effects of various excipients on in vitro release properties and formation of EPO aggregates were investigated. The formation of EPO aggregates in the W/O/W double emulsion technique was mainly influenced by the first homogenizing step, when preparing the water-in-oil (W/O) emulsion, whereas the subsequent processing steps, including drying, proved to be noncritical. A rotor/stator homogenizer generated ca. 5% covalently bound EPO aggregates, ultrasonication and vortexing slightly increased aggregate-formation, as demonstrated by size-exclusion chromatography and native-polyacrylamide gel electrophoresis (PAGE). Using excipients, such as hydroxypropyl-β-cyclodextrin, l -arginine, or bovine serum albumin (BSA), a distinct reduction of the formation of EPO aggregates could be achieved. The discontinuous in vitro release behavior from PLG microspheres was not significantly modified by these additives, influencing predominantly the initial drug release phase. During the in vitro release, an accumulation of EPO aggregates in the residual microparticles was detected, which could not be suppressed by excipients. An accelerated stability test demonstrated no change in drug content, release behavior and aggregate profile over 56 days at −20, 8 °C or room temperature.

Erythropoietin loaded microspheres prepared from biodegradable LPLG–PEO–LPLG triblock copolymers: protein stabilization and in-vitro release properties

Biodegradable microspheres containing recombinant human Erythropoietin (EPO) were prepared from ABA triblock copolymers, consisting of hydrophobic poly(l-lactic-co-glycolic acid) A blocks and hydrophilic polyethylenoxide (PEO) B blocks. Different polymer compositions were studied for the microencapsulation of EPO using a modified double-emulsion process (W/O/W). The encapsulation efficiency for EPO, ranging from 72% to 99% was quite acceptable. The formation of high molecular weight EPO aggregates, however, was higher than in poly(d,l-lactide-co-glycolide) (PLG) microparticles. Using different excipients with known protein stabilizing properties, such as Bovine Serum Albumin (BSA), Poly-l-Histidine (PH), Poly-l-Arginine (PA) or a combination of PA with Dextran 40 (D40), the EPO aggregate content was significantly reduced to <5% of the encapsulated EPO. In contrast to PLG, ABA triblockcopolymers containing >7 mol % PEO, allowed a continuous release of EPO from microspheres for up to 2 weeks under in-vitro conditions. The release profile was comparable to FITC-Dextran 40 kDa (FD 40) loaded microspheres in the initial release phase, while EPO release was leveling off at later time points. BSA additionally prolonged the EPO release, while blends of PLG and PEO did not generate continuous EPO release profiles. LPLG-PEO-LPLG triblock-copolymers (35 mol % PEO; 30 kDa) in combination with 5% BSA yielded both an acceptable level of EPO aggregates and a continuous release profile under in-vitro conditions for up to 2 weeks. The formation of EPO aggregates at later time points is probably induced by acidic cleavage products of the biodegradable polymer and requires further optimization of the ABA polymer composition.

Recombinant human erythropoietin (rhEPO) loaded poly(lactide-co-glycolide) microspheres: influence of the encapsulation technique and polymer purity on microsphere characteristics

Recombinant human erythropoietin (EPO) and fluorescein isothiocyanate-labelled dextran (FITC-dextran) loaded biodegradable microspheres were prepared from poly(lactide-co-glycolide) (PLG) by a modified spray-drying technique. This microencapsulation method was compared with the water-in-oil-in-water (w/o/w) double-emulsion method. As expected, microsphere morphology, particle size and particle size distribution strongly depended on the production process. The spray-drying method was found to have a number of advantages compared to the w/o/w double-emulsion technique. The content of residual dichloromethane (DCM) in the final product was significantly lower in case of the microspheres prepared by spray-drying. Concerning EPO loaded microspheres, spray-drying yielded higher encapsulation efficiencies. Although the microspheres obtained by spray-drying are subjected to intensive mechanical and thermal stress during the preparation, the amount of aggregates of EPO in PLG microspheres were not increased compared to the w/o/w technique. Depending on the manufacturing method, addition of cyclic DL-lactide dimers (referred to as monomers in the following) affected the in vitro release profiles of EPO and FITC-dextran from PLG microspheres. Using differential scanning calorimetry it was shown that these low molecular weight substances only seem to be present inside the microspheres produced by spray-drying. DL-Lactide significantly reduced the initial burst release of both EPO and FITC-dextran. While the following release period of EPO was not affected by the DL-lactide content, a more linear FITC-dextran release pattern could be achieved. It can be concluded that the spray-drying technique provides a number of advantages compared to the w/o/w method. The modulation of protein release using low molecular weight additives is of particular interest for parenteral depot systems.

Biodegradable recombinant human erythropoietin loaded microspheres prepared from linear and star-branched block copolymers: influence of encapsulation technique and polymer composition on particle characteristics.

Recombinant human erythropoietin (EPO) and fluorescein isothiocyanate labeled dextran (FITC-dextran) loaded microspheres were prepared by a modified W/O/W double-emulsion technique. Biodegradable linear ABA block copolymers consisting of poly(L-lactide-co-glycolide) A blocks attached to central poly(ethyleneoxide) (PEO) B blocks and star-branched AB block copolymers containing A blocks of poly(L-lactide) or poly(L-lactide-co-glycolide) and star-branched poly(ethyleneoxide) B blocks were investigated for their potential as sustained release drug delivery systems. Microsphere characteristics were strongly influenced by the polymer composition. In the case of the linear block copolymers, a reduced lactic acid content in a linear block copolymer yielded smaller particles, a lower encapsulation efficiency, and a higher initial drug release both in the case of EPO and FITC-dextran. The investigation of the effects of several manufacturing parameters on microsphere formation showed that the process temperature plays an important role. Microsphere formation in a +1 degrees C environment resulted in higher drug loadings without increasing the amount of residual dichloromethane inside the particles. Other parameters such as the homogenization of the primary W/O emulsion and of the W/O/W double-emulsion have less impact on microsphere characteristics. Branched block copolymers containing star-shaped PEO also showed potential for the preparation of drug loaded microspheres. A certain amount of glycolic acid in the copolymer was necessary for the successful preparation of non-aggregating microspheres at room temperature. Again, the processing temperature strongly affected particle characteristics. Microsphere preparation at +1 degrees C allows the formation of microspheres from a polymer not containing glycolic acid, a result which could not be achieved at room temperature. Moreover, compared to microsphere formation at room temperature, the effective FITC-dextran loading was increased. Concerning the EPO loaded microspheres, the amount of EPO aggregated was comparable to that using the linear ABA polymers. A continuous release of the protein from these star-shaped polymers could not be achieved. In conclusion, apart from microsphere preparation in a +1 degrees C environment the choice of the polymer represents the main factor for a successful entrapment of proteins into biodegradable microspheres.

Glycopeptide profiling of human urinary erythropoietin by matrix-assisted laser desorption/ionization mass spectrometry.

The site-specific glycan heterogeneity of human urinary erythropoietin was investigated by matrix-assisted laser desorption/ionization mass spectrometry (MALDI-MS). Owing to the small amount of protein available, a strategy combining optimal sensitivity and specificity was used. Erythropoietin was reduced, S-alkylated and digested with endoproteinase Lys C. The peptides were separated by reversed-phase high-performance liquid chromatography and the molecular masses of the peptides determined by MALDI-MS. The peptides were identified by comparing the experimental masses with the masses predicted from the cDNA derived amino acid sequence. Glycopeptides were identified from the mass spectra based on the peak pattern caused by the glycan heterogeneity. They were further characterized after treatment with neuraminidase and endoproteases. All N-glycosylation sites exhibited fucose-containing complex-type glycans. The N-glycosylation sites at Asn38 and Asn83 are mainly occupied by tetraantennary glycans, whereas Asn24 is occupied by a mixture of bi-, tri- and tetraantennary glycans. A molecular mass glycoprofile for each glycosylation site was established based on the relative peak intensities observed in the MALDI mass spectra of the desialylated glycopeptides.