G. De Rosa et al., Influence of the co-encapsulation of different non-ionic surfactants on the properties of PLGA insulin-loaded microspheres, J CONTR REL, 69(2), 2000, pp. 283-295
The aim of this work was to produce insulin-loaded microspheres allowing th
e preservation of peptide stability during both particle processing and ins
ulin release. Our strategy was to combine the concepts of using surfactants
to improve insulin stability while optimising overall microsphere characte
ristics such as size, morphology, peptide loading and release. Bovine insul
in was encapsulated within poly(lactide-co-glycolide) (PLGA 50:50, Resomer
RG504H) microspheres by the multiple emulsion-solvent evaporation technique
. Microspheres were prepared by adding to the primary emulsion three non-io
nic surfactants, poloxamer 188, polysorbate 20 and sorbitan monooleate 80,
at different concentrations (1.5 and 3.0% w/v). The presence of surfactants
was found to decrease the mean diameter and to affect the morphology of th
e microspheres. Insulin encapsulation efficiency was reduced in the presenc
e of surfactants and especially for sorbitan monooleate 80, in a concentrat
ion-dependent mode. The influence of the surfactants on the interactions be
tween insulin and PLGA together with the primary emulsion stability were fo
und to be the major determinants of insulin encapsulation. The release of i
nsulin from microspheres was biphasic, showing an initial burst effect foll
owed by a near zero-order release for all the batches prepared. The initial
burst was related to the presence of insulin molecules located onto or nea
r to the microsphere surface. In the presence of surfactants, a faster insu
lin release with respect to microspheres encapsulating insulin alone was ob
served. Insulin stability within microspheres after processing, storage and
release was evaluated by reversed phase- and size-exclusion-HPLC. The anal
ysis of microsphere content after processing and 6 months of storage showed
that insulin did not undergo any chemical modification within microspheres
. On the contrary, during the period of sustained release insulin was trans
formed in a high-molecular weight product, the amount of which was related
to the surfactant used. In conclusion, polysorbate 20 at 3% w/v concentrati
on was the most effective in giving regular shaped particles with both good
insulin loading and slow release, and limiting insulin modification within
microspheres. (C) 2000 Elsevier Science B.V. All rights reserved.