Poly(L-lysine) as a model drug macromolecule with which to investigate secondary structure and microporous membrane transport, part 2: diffusion studies

Abstract

Peptide drugs are hydrophilic in nature and so their preferred pathway of membrane transport is by the paracellular route, which primarily involves passive diffusion across intercellular pores. The objective of the present study was to investigate the effect of secondary structure on the aqueous diffusion of a model polypeptide, poly(L-lysine), through a microporous membrane. The primary aim was to systematically evaluate the variables (e.g. viscosity and/or hydrodynamic radius) that may contribute to the difference, if any, in the calculated values of the aqueous diffusion coefficient (D(aq)) for each conformer of poly(L-lysine). Variations in pH and temperature of the medium were used to induce secondary structural changes in poly(L-lysine). Transport studies were conducted for 3 h at 25 or 37 degrees C using side-by-side diffusion cells. Hydrophilic microporous polyester membranes with a 1-microm pore diameter were used to measure the free diffusion of each conformer. The values for the apparent permeability (P(app)) and D(aq) were calculated using standard equations. The viscosity of each conformer solution was determined and the hydrodynamic radius of each conformer was then estimated. At 25 degrees C, both P(app) and D(aq) of the alpha-helix conformer were approximately the same as those of the random coil conformer. In contrast, at 37 degrees C, the P(app) and the D(aq) of the beta-sheet conformer were significantly (P < 0.05) less than those of the random coil conformer. At 25 degrees C, the solutions containing primarily either the random coil or the-helix conformers had approximately the same viscosity. On the other hand, at 37 degrees C, the solutions containing the beta-sheet conformer had a significantly (P < 0.05) higher viscosity than when this conformer was absent. The random coil and the alpha-helix conformers appeared to have comparable sizes, whereas the hydrodynamic radius estimated for the beta-sheet conformer was significantly (P < 0.05) larger than those for the other two conformers. In summary, changing the secondary structure of poly(L-lysine) from the random coil to the alpha-helix did not affect its P(app) and intrinsic D(aq). On the other hand, appearance of the beta-sheet conformer significantly decreased the values of P(app) and D(aq). The differences appeared to result from the significantly higher solution viscosity as well as the extended structure associated with the beta-sheet conformer of poly(L-lysine). This strategy may represent a potential mechanism to sustain the delivery of therapeutic peptide drugs from a controlled drug delivery device.