Research Program

Starving Cancer Cells to Death
We propose to use an age-old strategy of war, attacking the supply wagons, to block cancer cell growth. Different cancer cells are dependent on different nutrients: serine, alanine, glutamine, low density lipoprotein, and arginine have all been shown to be essential in tumor cells with particular oncogenic mutations. A classic example of the effectiveness of starving cells for nutrients is the clinical success of L-asparginase, an asparagine-degrading enzyme that has been used to treat leukemia patients for over 25 years. While blocking access to individual nutrients can be effective, cancer cells can develop resistance by activating parallel pathways or activating compensatory pathways. This is also true of small molecule inhibitors of the enzymes that metabolize nutrients that are conditionally essential in cancer cells: activation of parallel, compensatory pathways can lead to resistance to therapy. To overcome these therapeutic obstacles stemming from tumor heterogeneity, our strategy is to simultaneously block access to multiple nutrients using sphingolipids.

Sphingolipids are evolutionarily-conserved regulators of nutrient access. Heat-stressed yeast produce phytosphingosine, a sphingolipid that triggers adaptive quiescence by down-regulating amino acid transporter proteins. Mammalian cells do not make phytosphingosine, but do produce the related sphingolipid ceramide in response to many kinds of cellular stress. Ceramide has been termed a tumor suppressor lipid because it slows proliferation, induces differentiation, and call trigger cell death. My lab showed that a key mechanism by which ceramide slows growth is down-regulating transporters for both glucose and amino acids (Guenther et al., PNAS 2008). A key piece of evidence that ceramide starves cells to death is that cell permeant forms of pyruvate or alpha-ketoglutarate rescue cells from ceramide-induced death. Thus, sphingolipids might also be used as tools to starve mammalian cells to death.

Ceramide itself has poor drug properties: it is extremely hydrophobic and difficult to deliver and it is rapidly metabolized, especially in cancer cells. However, the FDA-approved sphingolipid drug FTY720 also induces transporter loss at higher doses than are required for its immunosuppressive actions at sphingosine-1-phosphate receptors. In fact, yeast starved for nutrients, exposed to phytosphingosine, or treated with FTY720 mount largely overlapping transcriptional responses. FTY720 has been shown by many groups to limit tumor growth and metastasis. However, FTY720 cannot be re-purposed for use in cancer patients because at the elevated anti-neoplastic dose it causes profound bradycardia by activating S1P receptors. Working with synthetic and medicinal chemist Stephen Hanessian, we have developed constrained analogs of FTY720 that no longer activate S1P receptors but retain the ability to down-regulate transporters. These analogs also share FTY720’s water solubility, oral bioavailability, and activity in multiple tumor models.

Current work in the lab is focused on two goals:

  1. Pre-clinical evaluation of FTY720 analogs. We are currently evaluating these compounds in patient-derived organoids and in patient-derived xenografts (PDX) to establish whether they will be active in heterogeneous humor tumors. The effect of these compounds on metastasis is also under study. Currently, prostate and breast cancer models are under evaluation. Cutting-edge techniques are being employed to identify the precise target of these molecules. Sphingolipids have long been known to activate protein phosphatase 2A, but which heterotrimers are affected and the mechanism for activation are currently unknown.
  2. Dissection of the mechanism of action of FTY720 analogs. We have found that these compounds starve cells through two mechanisms: triggering nutrient transporter down-regulation and blocking fusion of endosomes, autophagosomes, and macropinosomes with lysosomes. These effects on intracellular trafficking occur down-stream of PP2A activation. In addition to identifying the particular PP2A heterotrimers responsible for these effects, we are dissecting how protein dephosphorylation produces these alterations in trafficking in collaboration with Pierre Thibault’s laboratory at the Universite de Montreal.