Abstract
The interactions of exenatide, a Trp-containing peptide used as a drug to treat diabetes, with liposomes were studied by isothermal titration calorimetry (ITC), tryptophan (Trp) fluorescence, and microscale thermophoresis measurements. The results are not only important for better understanding the release of this specific drug from vesicular phospholipid gel formulations but describe a general scenario as described before for various systems. This study introduces a model to fit these data on the basis of primary and secondary peptide-lipid interactions. Finally, resolving apparent inconsistencies between different methods aids the design and critical interpretation of binding experiments in general. Our results show that the net cationic exenatide adsorbs electrostatically to liposomes containing anionic diacyl phosphatidylglycerol lipids (PG); however, the ITC data could not properly be fitted by any established model. The combination of electrostatic adsorption of exenatide to the membrane surface and its self-association (K(d) = 46 μM) suggested the possibility of secondary binding of peptide to the first, primarily (i.e., lipid-) bound peptide layer. A global fit of the ITC data validated this model and suggested one peptide to bind primarily per five PG molecules with a K(d) ≈ 0.2 μM for PC/PG 1:1 and 0.6 μM for PC/PG 7:3 liposomes. Secondary binding shows a weaker affinity and a less exothermic or even endothermic enthalpy change. Depending on the concentration of liposomes, secondary binding may also lead to liposomal aggregation as detected by dynamic light-scattering measurements. ITC quantifies primary and secondary binding separately, whereas microscale thermophoresis and Trp fluorescence represent a summary or average of both effects, possibly with the fluorescence data showing somewhat greater weighting of primary binding. Systems with secondary peptide-peptide association within the membrane are mathematically analogous to the adsorption discussed here.