Removal and identification of external protein corona members from RBC-derived extracellular vesicles by surface manipulating antimicrobial peptides

Abstract

In the last years, extracellular vesicles (EVs), secreted by various cells and body fluids have shown extreme potential in biomedical applications. Increasing number of studies suggest that a protein corona could adhere to the surface of EVs which can have a fundamental effect on their function, targeting and therapeutical efficacy. However, removing and identifying these corona members is currently a challenging task to achieve. In this study we have employed red blood cell-derived extracellular vesicles (REVs) as a model system and three membrane active antimicrobial peptides (AMPs), LL-37, FK-16 and CM15, to test whether they can be used to remove protein corona members from the surface of vesicles. These AMPs were reported to preferentially exert their membrane-related activity via one of the common helical surface-covering models and do not significantly affect the interior of lipid bilayer bodies. The interaction between the peptides and the REVs was followed by biophysical techniques, such as flow-linear dichroism spectroscopy which provided the effective applicable peptide concentration for protein removal. REV samples were then subjected to subsequent size exclusion chromatography and to proteomics analysis. Based on the comparison of control REVs with the peptide treated samples, seventeen proteins were identified as external protein corona members. From the three investigated AMPs, FK-16 can be considered as the best candidate to further optimize EV-related applicability of AMPs. Our results on the REV model system envisage that membrane active peptides may become a useful set of tools in engineering and modifying surfaces of EVs and other lipid-based natural particles.

Keywords: extracellular vesicles; membrane active antimicrobial peptide; protein analysis; protein corona; spectroscopy.

SCHEME 1

Helical wheel diagrams of the peptides used in this study. Diagrams were drawn using HELIQUEST software (Gautier et al., 2008). The arrows indicate the direction of the hydrophobic face.

FIGURE 1

REV—AMP interactions studied by LD spectroscopy. (a) Signal intensity changes at the Soret band of REV samples upon addition of LL‐37 (blue), CM15 (orange) and FK‐16 (olive). For better visualization selected peptide concentrations are showed. (b) LD peak intensity at 420 nm as a function of AMPs. Data are normalized to the LD intensity of control REV.

FIGURE 2

Characterization of REV samples after AMP treatment. (a) Size exclusion HPLC chromatogram of vesicle samples at 280 nm. REVs were eluted at 0.38 V _E/V _T (V _E = elution volume, V _T = total volume). Chromatograms were collected after addition of 80 μM for LL‐37 and CM15 and 160 μM for FK‐16, respectively. (b) Concentration and size distribution of REV samples measured by microfluidic resistive pulse sensing (MRPS). Peptide concentrations were 80 μM for LL‐37 and CM15 and 160 μM for FK‐16, respectively. Both techniques indicate that significant amount of REVs are retained after treatment with AMPs.

FIGURE 3

Protein composition of the REV samples after AMP treatment. (a) Total and common proteins identified (common proteins are present in both control and AMP treated REV samples). (b) Classification of the identified proteins for each sample. (c) Venn diagram of overlapping proteins from control and AMP treated REV samples. (d) Comparison of proteins from Vesiclepedia Top 100EV with proteins of red blood cells membranes (RBC membrane*) and RBC‐MV* (*data from Prudent et al., 2018) as well as with REV proteins (REV_summ). The bar graphs demonstrate that vast majority of initial proteins are retained for AMP treated samples, which confirms that the vesicles remained largely intact. For all protein categories a slight decrease can be seen for the AMP‐treated samples suggesting that the peptides affect only a small subsection of the total protein content.

FIGURE 4

Proteomic differences of selected most abundant proteins in control and in AMP‐treated samples. (a) Characteristic proteins of the RBC membrane and of (b) subcellular origin were selected. The bar graphs demonstrate that almost all initial proteins are present after AMP treatment, in similar abundance as in control samples.

FIGURE 5

Comparison of selected protein corona members 2018; 2021) in control and in AMP‐treated samples.