Jonathan L. Coloff, Ph.D. - US grants

The metabolic phenotypes associated with cancer offer great promise for therapeutic intervention, as tumor cells are believed to display dramatically different metabolic programs than non9transformed cells. Years of work have led to the development of a model in which oncogenic signaling rewires metabolic pathways in order to meet the biosynthetic demands of cell replication. It is important to note, however, that proliferation rates vary widely in tumors, and cells in most tumors are proliferating far slower than cells growing in culture. Nevertheless, over 80% of tumors display the elevated glucose uptake that is characteristic of unique tumor metabolism. This suggests that there may be proliferation9dependent and proliferation9independent aspects of tumor metabolism that we do not fully understand. My work has begun to address this question by investigating how metabolism changes as normal epithelial cells transition between proliferative and quiescent cell states. I have then applied this knowledge to cancer by performing a comprehensive bioinformatic analysis of how proliferation contributes to metabolic gene expression patterns in human tumors. This work has demonstrated the importance of proliferation in determining the cancer metabolic phenotype, and has also suggested that oncogenic signaling can also impact metabolism independent of proliferation. The experiments described in this proposal will further investigate how pro9growth signaling and proliferation interact to regulate cellular metabolism in normal and cancer cells, and will identify metabolic vulnerabilities in non9proliferating tumor cells?an area of significant clinical need due to their resistance to many traditional therapies. Aim 1 will elucidate proliferation9dependent and signaling9dependent aspects of cellular metabolism and will explore the importance of biosynthetic demands as a driver of the cancer metabolic phenotype. Aim 2 will identify unique metabolic programs in non9proliferating tumor cells?such as therapy resistant cells and cancer stem cells?that can be targeted to simultaneously kill proliferating and non9proliferating cancer cells. And Aim 3 will explore glucose metabolism as a proliferation9independent aspect of tumor metabolism that forces metabolic adaptations in non9proliferating cancer cells, thereby creating unique metabolic vulnerabilities. Together, these studies will advance our understanding of how and why cancer cells are metabolically different from normal cells, and will hasten progress towards successfully targeting cancer metabolism in the clinic by identifying truly unique aspects of the metabolism of cancer cells.

PROJECT SUMMARY/ABSTRACT A major challenge of targeting metabolism for cancer therapy is pathway redundancy, where multiple sources of critical nutrients can diminish the effects of metabolic therapies. An example of this can be found in recent attempts to target the serine synthesis pathway for cancer therapy, where the abundance of serine available to be taken up from the circulation has hampered the success of inhibitors of serine biosynthesis. This places a premium on pursuing strategies of limiting pathway redundancy if we wish to successfully target serine and other critical metabolic pathways for cancer therapy. We have taken the approach of analyzing human tumor gene expression data to identify scenarios where pathway redundancy is limited due to lineage-dependent gene expression, thereby creating potential vulnerabilities. Using this approach, we have found that the two major lineages of breast tumors?luminal and basal?express vastly different levels of PSAT1 (phosphoserine aminotransferase 1), the gene encoding the second enzyme of the serine synthesis pathway. Luminal breast cancer cells, which express extremely low levels of PSAT1, are unable to activate the serine synthesis pathway even when extracellular serine is completely absent. As a result, they are entirely dependent on exogenous serine for proliferation and survival. This is in contrast to basal breast cancer cells, which are able to synthesize serine and proliferate in the absence of extracellular serine. Mechanistically, this serine auxotrophy appears to be due to luminal-specific methylation of the PSAT1 gene. Based on this data, we have developed the hypothesis that lineage-specific epigenetic silencing of the PSAT1 gene induces serine auxotrophy in luminal breast tumors and makes them vulnerable to inhibition of serine uptake. In this proposal, we will 1) determine whether luminal breast tumors are sensitive to dietary serine starvation in vivo, 2) define the mechanism of PSAT1 suppression in luminal tumors, and 3) identify and characterize serine transporters as potential pharmacological targets of this vulnerability. While luminal breast cancer patients initially have a favorable prognosis due to the utility of endocrine therapies, over half of all patients eventually develop resistance to these therapies and undergo relapse. As a result, over half of all breast cancer fatalities are due to luminal breast cancer, making this an area of significant unmet clinical need. The experiments described in this proposal have the potential to identify new therapeutic options for patients with advanced luminal breast cancer.