Mechanisms of Olfactory Pathobiology – Post-Viral Damage and Neoplasia

Abstract

The olfactory epithelium (OE), a neuro-epithelium in the nasal cavity responsible for detecting odorants, is unique in its ability to constantly regenerate neurons throughout life. In addition to neurons, the OE is composed of other diverse cell types, including sustentacular support cells, Bowman’s glands, and microvillar cells, all of which can be derived from a common basal progenitor. Disruptions to this dynamic stem cell niche can alter olfactory function and come in a variety of forms, owing to the OE’s exposure to the outside world. Notably, SARS-CoV-2 infection commonly causes olfactory loss in the acute phase resulting from infection of sustentacular cells. However, millions of individuals have gone on to develop persistent olfactory dysfunction months to years following COVID-19, due to incompletely understood biology. Here, we used a biopsy-based approach and performed single cell RNA-sequencing and immunohistochemistry on cohorts of post-acute sequalae of COVID-19 (PASC) hyposmic patients paired with controls to investigate the pathobiological mechanism within the OE of these patients. We found a significantly enriched pro-inflammatory T cell infiltrate composed of interferon-γ producing γδ T cells in PASC hyposmic patients. Interestingly, while this did not impair the overall steps of neurogenesis, we observed significantly fewer olfactory sensory neurons in PASC hyposmic patients, suggesting that persistent immune infiltrates may bias olfactory basal cell differentiation away from neuronal lineages or that chronic interferon-γ may impair full maturation or survival of olfactory sensory neurons. This work suggests that modulating the immune infiltrate in the OE in this subset of patients may help to restore normal olfactory function. An additional area focus of this dissertation is on olfactory neuroblastoma (ONB), a tumor believed to arise from the OE, causing significant morbidity and mortality in patients, even in cases of low grade tumors. ONB accounts for roughly 3% of sinonasal malignancy, and owing to its rarity and lack of model systems, mechanisms of tumorigenesis and proliferation have historically not been well understood. Here, utilizing new advances in spatial transcriptomics and deconvolution of existing bulk RNA-Seq ONB datasets, we discover that ONB tends to follow similar differentiation trajectories to the normal OE, including neuronal and non-neuronal cell fates. A pan-cancer atlas comparison reveals that ONB resembles small cell lung cancer and neuroendocrine prostate cancers significantly more than neuroblastoma and other neural tumors. Finally, using patient-derived tissue we developed an in vitro organoid model of ONB that reliably maintains tumor heterogeneity. These results may guide future clinical approaches to ONB, including grouping ONB patients into drug trials for other neuroendocrine cancers, and also provide important tools for in vitro drug screening.