Stability of proteins encapsulated in injectable and biodegradable poly(lactide-co-glycolide)-glucose millicylinders.

PURPOSE: To characterize protein stability in poly(lactide-co-glycolide) 50/50-glucose star (PLGA-Glu) injectable millicylinders and to compare results with linear PLGA 50/50. METHODS: Bovine serum albumin (BSA), a model protein, was encapsulated in PLGA-Glu and linear PLGA millicylinders by solvent-extrusion and incubated under physiological conditions. Important system properties were characterized, including: polymer molecular weight distribution, soluble acidic residues, polymer morphology, polymer water uptake, microclimate pH, protein content and release, and protein aggregation. The polymer microclimate late in the release incubation was simulated and protein recovery was analyzed by UV280, size exclusion chromatography, amino acid analysis, and a modified Bradford assay. RESULTS: PLGA-Glu contained higher levels of low molecular weight oligomers, more rapidly biodegraded, and exhibited a lower microclimate pH than the linear 50/50 PLGA, which is the most acidic type in the PLGA family. BSA, when encapsulated in PLGA-Glu millicylinders, underwent extensive noncovalent insoluble aggregation over 2 weeks in vitro release, which was almost completely inhibited upon co-encapsulation of Mg(OH)2. However, by 5 weeks release for base-containing formulations, although insoluble aggregation was still suppressed, the soluble fraction of protein in the polymer was unrecoverable by the modified Bradford assay. Polymer microclimate simulations with extensive protein analysis strongly suggested that the low recovery was mostly caused by base-catalyzed hydrolysis of the oligomeric fraction of BSA. CONCLUSIONS: In PLGA-Glu, the acidic microclimate was similarly responsible for insoluble aggregation of encapsulated BSA. BSA aggregation was inhibited in millicylinders by co-incorporation into the polymer an insoluble base, but over a shorter release interval than linear PLGA likely because of a more acidic microclimate in the star polymer.

BSA degradation under acidic conditions: a model for protein instability during release from PLGA delivery systems.

Acidification of the internal poly(lactide-co-glycolide) (PLGA) microenvironment is considered one of the major protein stresses during controlled release from such delivery systems. A model protein, bovine serum albumin (BSA), was incubated at 37 degrees C for 28 days to simulate the environment within the aqueous pores of PLGA during the release phase and to determine how acidic microclimate conditions affect BSA stability. Size-exclusion high performance liquid chromatography (SE-HPLC), SDS-PAGE, and infrared spectroscopy were used to monitor BSA degradation. BSA was most stable at pH 7, but rapidly degraded via aggregation and hydrolysis at pH 2. These simulated degradation products were nearly identical to that of unreleased BSA found entrapped within PLGA 50/50 millicylinders. At pH 2, changes in BSA conformation detected by various spectroscopic techniques were consistent with acid denaturation of the protein. By contrast, at pH 5 and above, damage to BSA was insufficient to explain the instability of the protein in the polymer. Thus, these data confirm the hypothesis that acid-induced unfolding is the basis of BSA aggregation in PLGA and the acidic microclimate within PLGA is indeed a dominant stress for encapsulated BSA. To increase the stability of proteins within PLGA systems, formulations must protect against potentially extreme acidification such that native structure is maintained.

Comparison of the effects of Mg(OH)2 and sucrose on the stability of bovine serum albumin encapsulated in injectable poly(D,L-lactide-co-glycolide) implants.

Incomplete release and poor stability of encapsulated proteins are common hurdles to overcome when developing poly(lactide-co-glycolide) (PLGA) controlled-release systems. Antacid excipients such as Mg(OH)2, which increase both microclimate pH and polymer water uptake, have been shown to prevent acid-induced instability of proteins encapsulated in PLGA. The purpose of this study was to delineate the effects of microclimate pH and polymer water content on the stability of encapsulated bovine serum albumin (BSA) by comparing the effects of Mg(OH)2 with those of another excipient, sucrose, which increases polymer water content without significantly affecting acid-base chemistry of the polymer. These two excipients, when encapsulated in PLGA at appropriate levels (3% Mg(OH)2 vs. 10% sucrose), were found to cause identical water sorption kinetics, thus allowing the effect of the two microclimate parameters to be determined. In contrast to their similar effects on polymer water sorption, Mg(OH)2 afforded a much greater stabilization effect on encapsulated BSA than did sucrose, with less than 7% aggregates for 3% Mg(OH)2 compared to 51% for 10% sucrose and 81% without either excipient after 4 weeks of incubation at 37 degrees C. When the protein stabilization rationale of neutralizing the acidic microenvironment by adding Mg(OH)2 was applied to the delivery of an important therapeutic protein, tissue plasminogen activator (t-PA), t-PA stability was also improved and the active protein was completely recovered during a one month period of in vitro release. These data demonstrated that although increased water uptake induced by antacid excipients may improve the stability of the encapsulated proteins, the homogeneous acid neutralization effect is unique to antacid excipients such as Mg(OH)2, which is necessary to maintain the stability of proteins in acidic PLGA specimens.

Cyclodextrin complexation: influence on the solubility, stability, and cytotoxicity of camptothecin, an antineoplastic agent.

The solubility of camptothecin (CPT), a highly potent antineoplastic agent, as a function of different concentrations of cyclodextrins (alpha-cyclodextrin, alpha-CD; beta-cyclodextrin, beta-CD; and gamma-cyclodextrin, gamma-CD; hydroxypropyl-beta-cyclodextrin, HP-beta-CD; and randomly substituted dimethyl-beta-cyclodextrin, RDM-beta-CD, and dimethyl-gamma-cyclodextrin, RDM-beta-CD) in 0.02 N HCl solution at 25 degrees C was investigated. The results showed a linear increase in the solubility of CPT with increasing concentration of CDs. The apparent stability constants (K(c)) for the CPT complexes with alpha-CD, beta-CD, gamma-CD, HP-beta-CD, RDM-beta-CD, and RDM-gamma-CD were 188, 266, 73, 160, 910, and 40.6 M(-1), respectively, suggesting that RDM-beta-CD afforded the most stable complex. At a 25% w/v concentration of RDM-beta-CD, the solubility of CPT was 228.45 +/- 8.45 microg/ml, about 171 times higher than that in 0.02 N HCl. The stability of CPT in pH 7.4 buffer at 25 degrees C also increased linearly with an increase in the concentration of RDM-beta-CD. The observed pseudo-first-order hydrolysis rate constants (k(obs)) for the free and complexed CPT were 11.8 x 10(-3) and 1.18 x 10(-3) min(-1), corresponding to an increase in half-life of CPT from 58.7 to 587.3 min, respectively. The preliminary cytotoxicity study against the human-derived myeloid THP-1 leukemia cell line showed RDM-beta-CD/CPT and HP-beta-CD/CPT complexes to be about two-fold more active than free CPT. In conclusion, the results showed that CDs, in general, and RDM-beta-CD, in particular, are effective complexing agents and can be used to improve the solubility and stability of CPT. The increase in cytotoxicity of CPT in the presence of CD is likely due to an increase in its stability.

Pore closing and opening in biodegradable polymers and their effect on the controlled release of proteins.

The purpose of this paper was to investigate the phenomena of pore closing and opening in microspheres of poly(lactic-co-glycolic acid) (PLGA) and PLGA-glucose star copolymer (PLGA-Glu) and their effects on protein release. We used scanning electron microscopy (SEM) and laser scanning confocal microscopy (LSCM) to visually characterize the pore state and the uptake of dextran labeled with pH-insensitive probes by microspheres, as an indicator of pore connectivity. The effect of temperature on initial protein release from microspheres was also investigated. It was found that (1) pore closing occurs in both PLGA and PLGA-Glu; (2) pore closing can take place at later time during incubation at physiological condition (37 degrees C) as well as during the initial stage; (3) pore closing is much more significant at elevated temperatures; (4) previously isolated pores can become open by, for example, osmotic-mediated events; and (5) pore closing/opening correlates with the release rate of biomacromolecules from PLGA or PLGA-Glu microspheres. The pore closing/opening appeared potentially a universal event throughout the release period dictating the kinetics of protein release from PLGA microspheres. Hence, these results strongly suggest that open and isolated pores are able to toggle back-and-forth periodically during PLGA degradation while controlling protein release; these observations imply a novel new hypothesis concerning erosion-controlled release of biomacromolecules from PLGA and related polymers.