Optimizing the therapeutic potential of extracellular vesicles in cancer

<p dir="ltr">Extracellular vesicles (EVs) are membrane-enclosed nanoparticles secreted by all cell types. They carry a variety of bioactive molecules, including lipids, proteins, and nucleic acids, which can act as signaling messengers to other cells. The two main types of EVs under investigation are microvesicles and exosomes. Microvesicles are formed by outward budding of the plasma membrane, while exosomes originate from inward budding within multivesicular bodies and are released upon fusion with the plasma membrane. EVs are found in various biological fluids such as blood, saliva, and urine, as well as in cell culture media. Their roles in intercellular communication have made them a focus of research in fundamental biology, biomarker discovery, and therapeutic development, including engineering EVs for clinical applications.</p><p dir="ltr">Studying EVs often requires their isolation, which presents several challenges. The physical and molecular similarities between EV subtypes make them difficult to separate, and common isolation protocols often co-isolate contaminants such as protein aggregates and lipoproteins. In Study I, we compared five fundamentally different EV isolation techniques, each based on distinct EV properties such as density, size, affinity, and hydrophobicity. These methods were applied to two biofluids, cell culture supernatant and plasma, at varying volumes to assess whether the optimal isolation method depends on the biofluid type. Plasma, for instance, contains high levels of lipoproteins that overlap with EVs in size and density, complicating isolation. EV isolates were characterized using transmission electron microscopy (TEM), nanoparticle tracking analysis (NTA), Bioanalyzer, flow cytometry, and proteomic profiling. All methods successfully enriched for EVs, but the composition and quality of the isolates varied depending on the biofluid, sample volume, and isolation technique. This study emphasized the importance of method selection, as each technique isolates different EV subpopulations with different contaminants, which can affect study outcomes and comparability. It also highlighted that some components initially considered as contaminants may be biologically relevant.</p><p dir="ltr">In Studies II and III, we explored the therapeutic potential of dendritic cell (DC)- derived EVs, particularly their use in cancer immunotherapy. Using EVs loaded with the model antigen ovalbumin (OVA), we investigated strategies to enhance their immunogenicity. In Study II, we focused on the immune checkpoint molecule PD-L1, which was present on our EVs. PD-L1 present on Tumor-derived EVs is known to suppress T cell activation, but its role on therapeutic DC-EVs was unclear. We generated PD-L1-deficient (PD-L1-/-) EVs and compared their immunogenicity to wild-type EVs in multiple mouse tumor models. While most models showed a trend toward stronger immune activation with PD-L1-/- EVs, a significant improvement was observed only in a prophylactic tumor model. These findings suggest that removing PD-L1 may enhance the efficacy of EV-based immunotherapies.</p><p dir="ltr">In Study III, we tested whether targeting EVs to B cells could further improve their immunogenicity. A soluble fusion protein (C1C2-D123) was developed to bind phosphatidylserine on EVs and CD21 on B cells. This protein successfully directed EVs to B cells in vitro and in vivo without altering their biodistribution. In an in vivo immunization model, EVs decorated with the fusion protein induced a higher frequency of antigen-specific CD8+ T cells compared to non-targeted EVs, supporting B cell targeting as a strategy to enhance immune responses.</p><p dir="ltr">In conclusion, this thesis highlights the importance of EV isolation strategy and cargo engineering in optimizing EV-based therapies. It demonstrates that both the method of isolation and the molecular composition of EVs significantly influence their biological and therapeutic potential.</p><h3>List of scientific papers</h3><p dir="ltr">I. Rosanne E. Veerman, <b>Loes Teeuwen</b>, Paulo Czarnewski, Gözde Güclüler Akpinar, AnnSofi Sandberg, Xiaofang Cao, Maria Pernemalm, Lukas M. Orre, Susanne Gabrielsson, Maria Eldh. Molecular evaluation of five different isolation methods for extracellular vesicles reveals different clinical applicability and subcellular origin. J Extracellular Vesicles. Volume 10, Epub e12128, (2021). <a href="https://doi.org/10.1002/jev2.12128" rel="noreferrer" target="_blank">https://doi.org/10.1002/jev2.12128</a></p><p dir="ltr">II. <b>Teeuwen*, L.</b>, Steiner*, L., Reinhardt, C., Offens, A., Kuipers, J.E., Martínez- Martínez, D., Mazouin, J., Chambers, B.J., Güçluler Akpinar, G., Gabrielsson, S. Removal of PD-L1 on extracellular vesicles for cancer vaccination modulates anti-tumor responses in a murine immunotherapy model. [Manuscript]</p><p dir="ltr">III. Annemarijn Offens, <b>Loes Teeuwen</b>, Gozde Gucluler Akpinar, Loïc Steiner, Sander Kooijmans, Doste Mamand, Hannah Weissinger, Alexander Kall, Maria Eldh, Oscar P.B. Wiklander, Samir El-Andaloussi, Mikael C.I. Karlsson, Pieter Vader, Susanne Gabrielsson. A fusion protein that targets antigen-loaded extracellular vesicles to B cells enhances antigen-specific T cell expansion. J Controlled Release, Volume 382, 113665 (2025). <a href="https://doi.org/10.1016/j.jconrel.2025.113665" rel="noreferrer" target="_blank">https://doi.org/10.1016/j.jconrel.2025.113665</a></p><p dir="ltr">* Shared first authorship.</p>