Evaluating the biodistribution of osteosarcoma extracellular vesicles and their impact on lung macrophages

Abstract

Patients with osteosarcoma (OS), the most common primary cancer of the bone, often experience metastatic spread of their disease following their initial diagnosis. OS metastasis has a striking tropism for the lungs and is associated with extremely poor survival outcomes, with under 30% of these patients surviving an additional 5 years. Therapeutic options for patients with metastatic OS have not improved in over four decades, emphasizing the pressing need to better understand this disease process. Metastatic disease is preceded by a "pre-metastatic" phase in which tumors release factors into systemic circulation that can prime distant organs to create a favorable site for eventual circulating tumor cell seeding. One of the tumor-derived factors known to elicit the formation of these hospitable pre-metastatic niches in distant organs are extracellular vesicles (EVs), which are nano-sized membrane bound particles that carry biological information between cells. However, our understanding of how EVs released by primary OS tumor cells alter the lungs prior to metastasis is limited. We aimed to improve our scientific understanding of the impacts of OS EVs on the lungs. Our initial goal was to identify which cell types within the lungs are the primary targets of OS EVs. In order to accomplish this, we first optimized and validated a technique to fluorescently label small EVs (sEVs) in order to detect them within cells of the lungs. One of the widely used approaches for EV labeling employs lipophilic membrane dyes, which have been recently shown to have serious technical limitations due to the formation of non-EV fluorescent particles that can confound results. We hypothesized that by optimizing the sEV starting material to ensure high purity and concentration and by altering the buffer conditions to eliminate protein from the system, we could more efficiently label sEVs with lipophilic membrane dyes. We show that by doing so, we could achieve sEV membrane labeling while limiting false positive signal formation and detect these fluorescently-labeled sEVs within cells both in vitro and in vivo. This work improves upon a widely used technique to label EVs, revealing its limitations and how to overcome them to avoid confounding results. These results are described in Chapter 2 of this dissertation. After developing a highly validated technique to efficiently label sEVs, we determined the cellular targets of OS sEVs in the lungs. To do this, we designed a multi-parameter, spectral flow cytometry panel that could delineate many of the cell types that comprise the lungs including stromal and immune populations. Using multiple different OS sEV sources from three different species, we evaluated the biodistribution of OS sEVs in immunocompetent, syngeneic mouse models as well as immunocompromised models. We hypothesized that lung interstitial macrophages may be a target of OS sEVs as they are professional phagocytes that act as a gateway between the vasculature and lung interstitium. Interestingly, although we observed that all OS sEVs, regardless of cell source, were taken up by interstitial macrophages, there were additional cellular targets that appeared to be dependent on the OS EV source. OS sEVs sourced from the murine OS cell line K7M2 also uniquely targeted alveolar macrophages present within the airways whereas canine and human OS sEVs did not. These findings suggest an ability of K7M2 sEVs to access the airway spaces that could be in part related to the species-matched nature of these sEVs. Additionally, K7M2 OS sEVs led to subtle, but significant, alterations to both inflammatory monocyte and fibroblast populations in the lungs of immunocompetent mice. Because OS impacts both humans and dogs, we also leveraged dogs as a translationally relevant surrogate to study alterations to the pre-metastatic lung microenvironment. Using cytology, flow cytometry, and single cell RNA sequencing we documented alterations in specific canine alveolar macrophage populations across different stages of naturally occurring OS using bronchoalveolar lavage techniques. In summary, these results suggest unique, non-random, and potentially species-specific patterns of OS EV biodistribution within the lungs prior to tumor cell arrival. OS sEV priming additionally alters key lung cell populations that may influence metastatic progression. Finally, we show that interrogating the airways spaces of canine OS patients reveals shifts in alveolar macrophage populations that could inform mechanisms of metastasis or identify biomarkers of disease progression. These results are reflected in Chapter 3. Our results from Chapter 3 revealed that the cell types capable of taking up OS sEVs in the lungs were primarily macrophages. Thus, in Chapter 4 we aimed to directly evaluate the impact of OS sEVs on primary human lung resident and recruited macrophages. The existing literature evaluating OS sEV impact on macrophages has been performed in mouse cells or in immortalized, cultured human macrophages. We aimed to build on previous work by investigating the impact of OS sEVs on lung macrophages in a translationally relevant setting utilizing primary human donor-derived macrophages. We were able to analyze alveolar macrophages isolated from airway spaces as well as monocyte-derived macrophages from human blood (a proxy for interstitial or recruited lung macrophages) using RNA sequencing and cytokine analysis. Following treatment with OS sEVs we observed a strong inflammatory response in monocyte-derived macrophages, with canonical anti-viral pathway upregulation. On the other hand, alveolar macrophages displayed a more metered inflammatory response, with upregulation of wound healing pathways. The response of each cell type was largely unique, but there were several shared upregulated pathways related to innate immune stimulation. These data provide key directions for future investigation of the functional consequences of sEV-primed lung macrophages in the pre-metastatic lungs of OS patients. Collectively, this work has identified the cellular targets of OS sEVs and characterized their impact on these resident lung cells. With this, we have developed a strong foundation for future investigation of biomarkers and mechanisms of metastatic progression. Moving forward, we hope that our findings lead will help inform new therapeutic targets for the treatment of metastatic OS to improve the lives of both human and canine OS patients.