Resumen de Comprehensive landscape of dna methylation in colorectal cancer
Colorectal cancer (CRC) is already the second leading cause of cancer deaths worldwide, being the third most commonly diagnosed malignancy in developed countries. CRC typically arises as a consequence of the accumulation of genetic and epigenetic alterations in colonic epithelial cells. Particularly, aberrant DNA methylation, are recognized as common molecular feature in human colorectal tumors. However, nowadays, we know relatively little related to the cause and consequence of DNA methylation in CRC. Therefore, we hypothesized that DNA methylation landscape could provide crucial information in the management of CRC, which may be affected by potential epigenetic drivers, such as vitamin D, since several epidemiological studies have suggested a protective role. This approach could also elucidate what potential pathways are involved in colorectal carcinogenesis, and identify key CRC risk factors. In addition, the reversible nature of epigenetic changes may enable us to identify potential DNA methylation drivers, including some bioactive food components, to reverse them and further decrease the risk of CRC.
Therefore, the aims of this thesis were i) to evaluate the association between global DNA methylation (through long interspersed nuclear element 1 (LINE1)) and survival rates in patients with CRC, ii) to investigate the association between serum 25-hydroxyvitamin D (25(OH)D) and DNA methylation in the CRC context (both epigenome-wide association study and specific promoter methylation), and finally, iii) to analyze the global DNA methylation landscape of visceral adipose tissue from patients with CRC, and its relationship with the risk of CRC.
Our results showed that i) CRC patients treated with neoadjuvant therapy, and had low LINE1 methylation levels in tumor area, had worse disease-free and overall survival rates than those with normal or high LINE1 methylation levels (p=0.004 and p=0.0049, respectively). Indeed, LINE1 was hypermethylated in those CRC patients treated with neoadjuvant therapy, when compared to those patients with CRC and without therapy (p<0.05) ii). In addition, 25(OH)D was associated with low methylation levels at the promoter of secreted frizzled related protein 2 (SFRP2) gene (an extracellular Wnt antagonists and tumor suppressor gene), which both were also related to neoadjuvant treatment under linear regression model. Based on these results, we conducted an epigenome-wide association study, in which we found that 25(OH)D was positively associated with global DNA methylation in tumor tissue from patients with CRC. We identified protein kinase inhibitor alpha (PKIA) gene as an epigenetic signature linked to CRC. In addition, many of epigenetically influenced genes were related to cell adhesion and extracellular matrix, as well as signaling transduction Finally, iii) we characterized for the first time the epigenetic status of adipose tissue, as a potential factor risk for CRC. Accordingly, we identified key pathways related to metabolism and thermogenesis, insulin resistance and adipocytokine signaling, as well as tumoral transformation processes. In addition, the specific epigenetic characterization of adipogenic, inflammatory and vitamin D related-metabolism genes showed that they are found upregulated and hypomethylated in patients with CRC, when compared to the healthy subjects (p<0.05). Interestingly, a decrease in interleukin 6 (IL6) methylation was associated with increased risk of CRC, in which 25(OH)D partially mediated this association (p<0.05). Finally, high IL6 gene expression was associated with poor survival in patients with CRC (p<0.05).
This thesis contributes to our understanding of CRC epigenetic status through the characterization of potential prognostic specific-tissue biomarker, to predict survival of the treated CRC patients, and help to personalize treatment in the CRC context. In addition, we identified 25(OH)D as epigenetic modifier in CRC, which could be an interesting target at adipose tissue levels for CRC prevention, through vitamin D interventions. Finally, we characterized for the first time the epigenetic landscape of adipose tissue and CRC, purposed several candidate genes, and identified the epigenetic contribution of adipose tissue on systemic inflammation to increase the risk of CRC. Loci identified in this thesis can accurately discriminate CRC tissue from non-cancerous/normal tissue and could be developed into a simple and cost-effective test for the detection and the management of CRC.
