Cancerous cells reprogram their metabolism to accommodate their needs. A deep understanding of this metabolic reprograming can lead to novel and promising routes toward cancer treatment. Different human cancers differ with respect to aspects of their metabolic rewiring, however a profound knowledge in this field is lacking. The main goal of my thesis project is to characterize the ways in which one-carbon metabolism ? a metabolic pathway highly altered in cancer? is utilized by different human cancers and elucidate its consequences on downstream processes such as epigenetics and biosynthesis. Using information in genomic profiles of individual tumors, I build computational models to infer metabolic landscapes, assess their implications in patient survival, and predict response to chemotherapy. By studying one-carbon metabolites in the human serum, I confirm the relevance of theses findings in human contexts and suggest potentials for dietary intervention approaches.
In aim 1, I characterize the usage of serine through one-carbon metabolism across human cancers. I performed flux analysis using gene expression profiles of hundreds of human tumors, followed by experimental validation using metabolomics approaches.
In aim 2, I identify the determinants of human serum methionine. Diet records, serum metabolomics, and clinical data from a cohort of human subjects were incorporated into computational models. The determinants of variability in serum methionine were then quantified, suggesting a mechanism for regulation of cellular epigenetics by the diet.
In aim 3, I determine the sources of variation in DNA methylation across human cancers and the contribution of metabolism. I integrated molecular and clinical profiles of thousands of human tumors from the TCGA into machine-learning algorithms to identify their association with DNA methylation. A major contribution from one-carbon metabolism in regulating DNA methylation status in tumors was found.
In aim 4, I predict response to anti-metabolic chemotherapies based on tumor genomics. I plan to translate my previous findings into clinical discoveries. To this end, I will demonstrate how tumor profiles can be used to model patient survival, predict response to chemotherapy, and move toward precision medicine.
Alteration in cellular metabolism is one the hallmarks of cancer. This project aims to elucidate the consequences of these alterations in cancer and reveal how different human cancers differ with respect to their metabolism. Findings can lead to novel and promising routes toward cancer prevention and treatment such as precision medicine and dietary intervention.
|Salerno Jr, Stephen; Mehrmohamadi, Mahya; Liberti, Maria V et al. (2017) RRmix: A method for simultaneous batch effect correction and analysis of metabolomics data in the absence of internal standards. PLoS One 12:e0179530|