Abstract Emerging evidence suggests that tumor-derived extracellular vesicles (tEVs) can influence immune cell behavior both locally within the tumor microenvironment and remotely by accessing the lymphatic and systemic circulation. The in vivo effects of such interactions before and after immunotherapy remain largely unappreciated. In this study, we aimed to develop a sensitive but stringent approach to identify host cells that interact with tEVs, with the ultimate goal of studying how tEV influence immune cells in their native microenvironment. Investigation of tEVs’ roles in cancer has been hampered by the need to isolate them before intravenous reinfusion in animals, which can introduce biases such as homogenization of tEV diversity, judgement of amount to reinfuse and assumptions on blood vs lymph biodistribution. To circumvent these issues, we and others pioneered the use of genetically engineered tumor cells to express membrane-bound reporters, which allow to label untouched, endogenously released tEVs and follow their interactions with immune cells. Still, the amount of reporter proteins that tEVs can carry is relatively small and thus the sensitivity of the approach is suboptimal. Here we addressed this issue by engineering tEVs to display a membrane-bound form of Sortase A, a bacterial transpeptidase that catalyzes the transfer of reporter proteins on the much bigger surface of tEV-binding cells. SrtA catalyzes the formation of a peptide bond between a consensus peptide and an N-terminal glycine of a nearby cell surface proteins (e.g. MHC-I, MHC-II, VE-Cadherin, CD19, integrins). We tested 4 different SrtA protein designs and selected the best performing construct for tEV studies. We genetically encoded SrtA in parental tumor cell lines, which then release SrtA as a tEV-bound transmembrane protein. Once a SrtA+ tEV interacts with a target cell in the presence of the consensus peptide (conjugated to a fluorescent reporter), SrtA catalyzes the formation of a covalent bond between the reporter and surface proteins of the tEV-binding cell. As expected, we observed significant increase in labeling when target cells were engineered to express a surface protein carrying 5 N-terminal glycine residues, which confirms that labeling of tEV-binding cells is due to actual transfer of fluorescent reporter – and not due to acyl intermediate formation. As compared to indirect labeling of EV-binding cells (e.g. using the EV marker CD63 fused to GFP), SrtA-based approach shows 1-2 log increase in sensitivity, depending on the EV-producing cell type. Overall, the catalytic nature of our EV reporter system lowers the amount of tEVs a host cell needs to bind before being detectable, thereby increasing sensitivity. We are currently testing our SrtA-based approach in vivo, where endogenous tEVs mainly accumulate in sentinel lymph nodes. We expect to unearth the full set of lymph node immune cells interacting with native tEVs and to identify whether immunecheckpoint blockade therapy affects tEV tropism toward immune cells. Citation Format: Ferdinando Pucci, Nicklas Hamilton, Natalie Claudio, Randall Armstrong. Tumor-immune communication via extracellular vesicles [abstract]. In: Abstracts: AACR Virtual Special Conference: Tumor Immunology and Immunotherapy; 2020 Oct 19-20. Philadelphia (PA): AACR; Cancer Immunol Res 2021;9(2 Suppl):Abstract nr PO007.
Read full abstract