Abstract
Dedifferentiation – the acquisition of an early developmental state - is a hallmark of cancer, however the underlying mechanisms remain elusive. The process of dedifferentiation is characterised by genetic and epigenetic changes that enable a phenotype of proliferation, self-renewal and a metabolism reminiscent of embryonic stem cells. Some cancers reacquire an epigenetic landscape that is reflective 0f early development, which corresponds with increased phenotypic plasticity and facilitates the activation of early developmental genes. To this end, the fundamental regulators of pluripotency are all potent oncogenes, suggesting that cancers are able to repurpose developmental genes to drive malignancy. There is compelling evidence to support that the pathway to dedifferentiation in cancer can contribute to invasion and metastasis. Understanding mechanisms that drive dedifferentiation in cancer will facilitate improvements in diagnosis and treatment.
Transposable elements (TEs) are increasingly becoming recognised as important regulators of development and disease. TEs are documented to function as tissue-specific genes and regulatory elements, during early human development. As such, TEs have driven regulatory innovation of developmental pathways and interact with important developmental genes, many of which also function as oncogenes. The recruitment of TE sequences has played an important role in placental evolution, TE-derived genes have been identified that play critical roles in placental development. The placenta shows striking similarities to tumours such as its ability to invade into the surrounding tissue and mediate the maternal immune response. TEs have also been implicated in cancer development, and are documented to contribute to oncogenesis through a range of different mechanisms.
Based on the known role of TEs in development and cancer, along with the functional and epigenetic similarities between early development and cancer, it was hypothesised that functional TEs that play an important role in early human development may also contribute to cancer progression. We propose that this occurs through the regulation of early developmental pathways that, when reactivated inappropriately in later-life, drive malignancy. As such, this project sought to identify novel, developmental TE-derived genes and regulatory elements and determine which of these become reactivated in cancer.
A novel bioinformatic pipeline was developed to quantify and characterise developmental-specific TE-expression in RNA-Sequencing data generated from early human developmental tissues (human embryonic stem cells and placenta). This generated a catalogue of TE-derived genes and putative regulatory elements that have been recruited to function in early human development. Expression of these genes (protein-coding and lncRNAs) was then investigated in RNA-Sequencing data from melanoma cell lines, and patient data of 33 different cancer types (TCGA). This identified TE-derived genes and regulatory elements that are specific to early development yet become reactivated in cancer.
To independently validate the hypothesis that developmental TEs become reactivated in cancers, expression of TE-derived promoters identified to be driving expression of oncogenes in human cancers were investigated in our early developmental RNA-Sequencing datasets. Remarkably, this revealed that some TE-oncogene regulatory relationships are derived from early development. Furthermore, analysis of prognostic enhancers previously identified in TCGA datasets confirms that some of these elements have developmental origins and are TE-derived. Expression of candidate developmental TE-derived genes was validated using RT-qPCR and/or the NanoString platform, which confirmed their specificity early developmental tissues and cancer. Finally, the methylation status of some candidate genes was interrogated using target-deep bisulfite-sequencing. This revealed that some of these genes show an inverse relationship between promoter methylation and expression, indicating that methylation may be a mechanism that permits their expression.
Taken together, these results support that epigenetic reactivation of early developmental TE-derived genes may contribute to the marked similarities between tumours and early developmental tissues. However, further work is needed to establish the specific functions of uncharacterised TE-derived genes in both development and cancer. The specificity of these genes to early development and cancer could make them promising therapeutic targets, warranting further investigation. Overall this project has highlighted the merit of using early development to further our understanding of cancer, particularly through investigating the largely untapped potential of TE sequences.