A wide range of cancers throughout the body are characterized by high nerve density and invasion of cancer cells within the nerves, a process called perineural invasion (PNI). Work in the field has shown that blocking nerves in organs with tumors leads to improved disease outcomes suggesting that finding ways to block tumor nerves could lead to new treatment approaches. Despite the importance of this, little is known about the molecular mechanisms underlying nerve-tumor interactions. An increasing number of studies have revealed that the genes associated with PNI are classical neurodevelopmental genes associated with neurodevelopmental processes. Therefore, the central hypothesis of this thesis was that nerve-tumor interactions result in part from an abnormal reactivation of the molecular pathways underlying the embryonic development of the nervous system. To test this hypothesis, public datasets from different types of cancer with high incidence of PNI were analyzed to identify molecular pathways common between these cancers. This analysis revealed that neurodevelopmental pathways accounted for 12 - 16% of the differentially expressed genes (DEGs), with axon guidance genes being markedly dysregulated. Overall, 17 different axon guidance gene families, including ephrin-Eph, semaphorin-neuropilin/plexin and slit-robo pathways were dysregulated. Further disruption of these pathways was a common feature across a number of cancers analyzed and their dysregulation had a significant impact on disease survival. Overall, this suggested that neurodevelopmental molecular pathways may contribute to tumor axonogenesis and PNI.
These findings suggested a significant role of neurodevelopmental pathways in cancer dysregulation. Thus, a comprehensive understanding of the pathways during the nervous system development is imperative. Therefore, in my thesis, the embryo was used as a tool to study the mechanisms by which these molecular pathways influence axonogenesis more broadly. First, the role of the axon guidance genes Slit/Robo was examined during mouse neurodevelopment. Our results showed that Robo2 enrichment influences the migration and axonal projections of spinal ipsilateral neurons. In parallel, we investigated the role of alternative splicing of transcription factors as a mechanism of increase neuronal diversity. In particular we examined the expression dynamics of Lhx9, a transcription factor that controls the expression of the axon guidance gene Robo3. Lhx9 splice variants showed a differential expression at key developmental points in the spinal cord, suggesting Lhx9 splice dynamics plays an important role in neural guidance choices.
In the third part of the thesis, I investigated the role of gap junctions, in nerve-tumor interactions, using pancreatic ductal adenocarcinoma (PDAC) cancer cells in vitro models. The connexin GJB2 emerged as the most overexpressed gap junction component in PDAC tumors. In vitro analysis, involving blocking gap junctions or connexin overexpression, revealed that gap junctions influenced PDAC cancer cell behaviors and properties. Further we developed a novel nerve-tumor assay and used it to examine the role of gap junction genes in PDAC cells neuronal growth.
Overall, this thesis postulated that several key molecular pathways crucial for normal nervous system embryonic development, could underly nerve- tumor interactions during cancer development and progression.