SUMMARY Immunotherapies have shifted the paradigm for cancer treatment in the past decade. Among them, therapeutic cancer vaccines, consisting of tumor antigens and adjuvants for activating antigen presenting cells (e.g., dendritic cells (DCs)), have shown the promise to elicit persistent humoral and cytotoxic T lymphocyte (CTL) response, but are limited by the modest therapeutic benefit. Hurdles for developing potent cancer vaccines include the lack of available tumor antigens and sub-optimal modulation of DCs. Tumor-derived exosomes, the nano-sized extracellular vesicles secreted by tumor cells, are a good source of tumor antigens and have been widely explored as cancer vaccines in preclinical studies and clinical trials. However, the antitumor efficacy is far from satisfactory, likely as a result of poor CTL response. The incorporation of adjuvants that can bind to the toll like receptors on DC membrane and thus activate DCs has become standard practice for improving conventional vaccines, but simple mixing with adjuvants has failed to enhance the antitumor efficacy of tumor exosome vaccines. The primary goal of this project is to develop a strategy that can well integrate tumor exosomes and adjuvants for the development of potent tumor exosome vaccines. To achieve this, we utilize a metabolic glycan labeling approach to label glycoproteins and glycolipids on the membrane of cancer cells with chemical tags (e.g., azido groups), and hypothesize that exosomes secreted by these labeled cancer cells will carry chemical tags on the surface. The chemically tagged tumor exosomes can covalently capture adjuvants and other immunomodulatory agents via efficient and bioorthogonal click chemistries, potentially resulting in improved activation and antigen presentation of DCs and enhanced CTL response. This project will be organized around two specific aims. In Aim 1, metabolic glycan labeling of various types of cancer cells for the generation of azido- labeled exosomes will be explored. Two hypotheses will be tested in this aim: (1) azido-labeled tumor cells can secrete azido-labeled exosomes; (2) the azido groups on the surface of exosomes can efficiently conjugate dibenzocyclooctyne (DBCO)-molecules via click chemistry. In Aim 2, adjuvant-conjugated tumor exosomes will be explored as therapeutic cancer vaccines, with the resulting CTL response and antitumor efficacy evaluated in murine breast cancer and glioblastoma models. Tumor exosomes conjugated with both adjuvants and DC- targeting ligands (e.g., anti-DEC205) will also be developed and evaluated. Successful completion of the proposed studies not only will provide a generalizable technology to generate chemically tagged exosomes from tumor cells for tracking and targeting purposes, but will also result in the development of potent tumor exosome vaccines with great potential for clinical translation.