(T1530-04-23) Constitution of a Helical Peptide on Exosome-Mimetic Vesicles

Purpose: Exosomes are gaining attention recently as drug carriers¹, as they can deliver materials such as microRNAs and proteins to targeted organs efficiently, of which destination is controlled by surface integrin molecules². However, utilizing real exosomes presents many difficulties in terms of safety, quality-control, and manufacturing. These difficulties can be overcome by their replacement with exosome-mimetic vesicles (EMVs). In this study, we elucidated how sphingomyelin (SM) and cholesterol (CH), which are main constituents of exosomes in addition to phospholipids, influenced physicochemical properties of the phospholipid vesicles. Then, influence of the membrane composition/property on the constitution process of a helical peptide that represents the transmembrane domain of integrin family was investigated. This information should help design and manufacture of EMVs capable of active targeting property.

Methods: Dipalmitoylphosphatidylcholine (DPPC), egg-sphingomyelin (SM) and cholesterol (CH) were used to prepare the vesicles by thin-film hydration method followed by high-pressure extrusion. The polycarbonate membrane with 100 nm pores was used to control the size of the vesicles. Composition of vesicles after the extrusion was confirmed with HPLC. Thermodynamic behavior and membrane polarity of the vesicles were evaluated with differential scanning calorimetry (DSC) and fluorescent measurement of polarity probe, laurdan, respectively. Phase separation of the lipid membrane was analyzed by fluorescent lifetime measurement of laurdan. Size distribution was analyzed by dynamic light scattering (DLS) and nanoparticle tracking analysis (NTA). Atomic force microscopy (AFM) was used for evaluating softness of the lipid membrane. A chemically synthesized 39 amino-acid peptide (ERAIPIWWVLVGVLGGLLLLTILVLAMWKVGFFKRNRPP) was used as a model helical peptide. Secondary structure and membrane insertion of the constituted peptide was evaluated with circular dichroism (CD) and zeta potential measurement.

Results: Loading of SM to the DPPC vesicle decreased the phase-transition temperature and increased the membrane polarity at 37˚C, whereas addition of CH exhibited the opposite trend (Fig.1 and Fig. 2). Those of the EMVs, which contained 40% of CH and 15% SM, were significantly lower and higher compared to the DPPC vesicles, respectively, suggesting stronger influence of CH. Fluorescent lifetime measurement at 37˚C revealed coexistence of three phases with different molecular mobilities. At 25˚C, the phase with the highest mobility appeared upon addition of SM, but not upon addition of CH only. Lipid composition affected the secondary structure of the constituted peptide and the zeta potential of the vesicles, which appeared to be influenced by location of the peptide. The addition of the peptide to the CH-bearing vesicles decreased the dispersion stability, whereas that to the SM-bearing vesicles changed the zeta potential from negative to positive, suggesting that the peptide was adsorbed/protruded on the surface. The successful insertion of the peptide was likely achieved only for the EMVs bearing 40% of CH and 15% SM, where an alpha-helix structure was successfully formed. This alpha-helix formation was more efficient when peptide-lipid suspension was left for a while prior to extrusion, suggesting easier constitution to planar membrane than to curved membrane (Fig. 3). Successful peptide constitution to 100 nm-diameter EMVs was likely tolerated because of the softness of the lipid membrane compared to other vesicles as revealed by the AFM study, which may be explained by presence of the highest mobility phase induced by the addition of SM. Thus, both CH and SM were essential for constitution of the helical peptide.

Conclusion: With regard to recent attention on exosome drug delivery systems, contributions of SM and CH on EMVs’ physicochemical properties were studied, and relevance of those factors to the secondary structure and insertion efficiency of integrin’s transmembrane helical peptide was revealed. Incorporation of the peptide was the most successful with EMVs bearing 40% of CH and 15% SM, and exclusion of SM or CH resulted in failure to form the helical structure or low dispersion stability. These results provided crucial insights that serve as a foundation for the development of EMVs as drug carriers.

References: [1] Kooijmans, S. A., Vader, P., van Dommelen, S. M., van Solinge, W. W., & Schiffelers, R. M.. Exosome mimetics: a novel class of drug delivery systems. International Journal of Nanomedicine, 7, 1525–1541 (2012). https://doi.org/10.2147/IJN.S29661
[2] Hoshino, A., Costa-Silva, B., Shen, TL. et al. Tumour exosome integrins determine organotropic metastasis. Nature, 527, 329–335 (2015). https://doi.org/10.1038/nature15756


Fig. 1 Phase transition temperatures (Tc) and enthalpy (ΔH) of lipid vesicles determined by DSC.
The numbers in the sample names indicate the mol% of SM or CH.


Fig. 2 Relative membrane polarity index calculated from the fluorescent spectra of laurdan.
Generalized Polarization (GP) values were calculated from the fluorescent intensities of laurdan at 440 nm and 490 nm. The numbers in the sample names indicate the mol% of SM or CH.


Fig. 3 CD spectra of the peptide constituted in EMVs.
An efficient alpha-helix formation of the peptide was observed for sample which was incubated at 25˚C for a day before extrusion. Immediately-extruded sample did not exhibit such spectrum, even when it was incubated after extrusion.