Cancer Cell Metabolism: Unique Features Inform New Therapeutic Opportunities.

Overview

Cancer cells become lethal when they form large tumors, metastasize, and colonize diverse tissue types. These functions depend on different metabolic pathways from those active in non-transformed cells. Rapidly growing and proliferating cells require constant biosynthesis, in addition to energy in the form of ATP that all cells need for normal functions. The glucose and glutamine that are exclusively catabolized to water and carbon dioxide in quiescent cells are partly diverted to macromolecule production in dividing cells. Cell metabolism has long been a focus of molecular biology; for the cancer field its study represents a return to original lines of inquiry after years of focus on the genetics of cell transformation and the oncogenes involved in signal transduction pathways. At the May 28, 2015, Hot Topics in Life Sciences symposium Cancer Cell Metabolism: Unique Features Inform New Therapeutic Opportunities, the speakers expressed hope that a synthesis of these two approaches will yield progress in cancer research.

Craig B. Thompson of the Memorial Sloan-Kettering Cancer Center reviewed the basics of cell metabolism. After the initial step in glucose metabolism—glycolysis, conversion of glucose to two molecules of pyruvate—mitochondrial oxidative phosphorylation usually proceeds to yield ATP. But even in the presence of oxygen, many cancer cells divert pyruvate to fermentation, producing lactate. This less rewarding mode of ATP production demands a relatively high rate of glycolysis. Otto Warburg described this shift toward "aerobic glycolysis" in cancer cells in 1924. The molecular and genetic basis of the Warburg effect, however, has only recently come to light. Contrary to Warburg's hypothesis that mitochondrial defects necessitate this shift, most cancer cells maintain the ability to execute oxidative phosphorylation and do fully catabolize a small amount of glucose.

Cancer cells are genetically differentiated from normal cells, but it is now clear that the metabolic shifts they exhibit are also partly required for division of normal cells. In a quiescent cell, maximum ATP production yields enough energy for cellular machinery, and at least 50% of free energy is used for ion transport across the membrane. When a cell divides, glycolytic intermediates are diverted from the tricarboxylic acid (TCA/Krebs) cycle to reserve carbon and nitrogen for fatty acid synthesis and for production of nonessential amino acids. DNA replication demands de novo nucleotide synthesis, beyond the supply garnered from recycling pathways in a non-dividing cell. Ribose, serine, and glycine (byproducts of glucose metabolism), as well as glutamine for pyrimidine production, are needed for nucleotide synthesis.

Cancer cells consume glucose and glutamine at higher levels than normal cells; many oncogenes implicated in signaling cascades also regulate metabolism of these nutrients. Research in cell signaling and metabolism may produce more effective combination therapies to treat cancer. (Image courtesy of Craig B. Thompson)

Thompson noted that previously overlooked clues from cell biology research are fundamental to current work to understand the unique metabolism of cancer cells. Harvey Eagle, who is largely credited with establishing a protocol and eponymous growth medium for culturing mammalian cells, did much of his foundational work with the HeLa cell line, derived from a pancreatic tumor. He observed a key feature of proliferating cells' metabolism when he found that these cells need to be cultured in a buffered solution supplemented with glucose and, to his surprise, high levels of the nonessential amino acid glutamine. In an anabolic cell where Kreb's intermediates are diverted to biosynthesis, glutamine metabolism provides a means of anaplerosis. Indeed, some tumors are said to be "glutamine addicted" and cannot survive without exogenous glutamine. The pharmaceutical company Calithera seeks to exploit this property with a new drug, CB-839, that inhibits glutaminase. It is in phase I clinical trials for treating cancer.

Christian Metallo of the University of California, San Diego, has used metabolic tracing to show that the glutamine contribution to lipogenesis shifts in mammalian cells in an oxygen-dependent manner. Cells cultured under hypoxic conditions, which might mimic those experienced by cancer cells in a poorly vascularized tumor, shift to reductive carboxylation of α-ketoglutarate. This glucose and glutamine derivative is also critical in stem cells. Lydia Finley of the Memorial Sloan-Kettering Cancer Center linked α-ketoglutarate levels in embryonic stem cells to the maintenance of demethylation and pluripotency.

Thompson argued that metabolic shifts that define cancer cells are not a secondary consequence of transformation, as has long been thought. The most common genetic mutations that drive cancers are in proto-oncogenes and tumor suppressor genes, which regulate pathways that control cell division. The constitutive activation of the cell cycle and the loss of checkpoints, however, are not enough to drive unmitigated growth; the cell must also undergo metabolic transformation to meet the energetic and synthetic demands of growth. This hypothesis, although perhaps intuitive, has only gained prominence with observations linking metabolic genes to control by signal transduction pathways.

The most commonly mutated gene in cancers is KRAS. The KRAS protein, a GTPase, normally functions as a molecular switch, relaying signals received by receptor tyrosine kinases and other receptors of extracellular signals. Two of its main targets include the MAPK and PI3K signal transduction cascades. But many indirect targets of KRAS are involved in cellular metabolism, including glucose transporters that are positively regulated by the PI3K/Akt pathway. Glutamine-addicted tumors are often characterized by the oncogenic expression of Myc, a transcription factor that promotes the expression of glutamine transporters as well as metabolic enzymes needed for biosynthesis. Constitutively activated KRAS thus primes a cell to undergo aerobic glycolysis by ensuring a steady influx of glucose. Selina Chen-Kiang of the Weill Cornell Medical College showed that therapies targeting the cell cycle indirectly reduce PI3K activation in cancer cells. In collaboration with Pfizer, her research group has tested the Cdk4/6 inhibitor palbociclib in early-phase clinical trials in Mantle cell lymphoma (MCL) patients. Genetic and biochemical analysis of tumors from palbociclib-responsive MCL patients exhibited reduced glucose transporter expression. Palbociclib is currently being tested as a cancer therapy in five separate clinical trials.

Jon Blenis of the Weill Cornell Medical Center identified another common genetic hallmark of cancer linked to metabolism, activated mTORC1 (mTOR Complex 1) signaling. Pathways that activate this complex integrate metabolism by sensing levels of nutrients, such as amino acids leucine and glutamine. The complex then regulates glucose and glutamine metabolism, amino acid production, and lipid biosynthesis. Blenis's group has screened for targets in the pathways to identify combination therapies for mTORC1-activated cancers. One combines the glutamine metabolism inhibitor BAPTES with the HSP90 inhibitor 17-AAG to reduce transformed cells' ability to carry out aerobic glycolysis and confront oxidative stress. A novel downstream target of mTORC1, the kinase SPRK2, is another promising lead.

Each tumor arises from a unique sequence of genetic lesions. Does each also have a unique metabolic signature? Most conclusions on this topic are drawn from studying isolated cancer cell lines, but tumor cells may execute metabolism differently in vivo. Matthew G. Vander Heiden of the Massachusetts Institute of Technology described rodent models of non-small cell lung carcinoma and pancreatic cancer, both driven by KRAS-activation and p53-knockout mutations. The first model had tumors characterized by increased glucose uptake; surprisingly, these tumors had elevated glucose catabolism through oxidative phosphorylation, in addition to elevated glycolytic metabolism. Glutamine tracing, however, showed almost no glutamine anaplerosis in vivo. But when explanted to culture conditions, cells from these tumors acquired glutamine dependence. In the pancreatic cancer model, no single nutrient was clearly favored as a metabolite. Albumin labeling experiments, however, showed that most of the pancreatic cancer cell biomass was derived from this extracellular protein.

It has long been assumed that extracellular protein is not a major fuel for most cells, but as Thompson observed, there is evidence that mammalian cells can draw on this energy source. Most cell culture media are supplemented with serum containing albumin, and the 0.74 mM concentration of albumin found in human plasma is equivalent to a 400 mM source of amino acids. The finding that cancer cells and cultured normal cells ingest extracellular protein through micropinocytosis has invigorated the field. The normal pathway to nutrient acquisition in slime molds, this process involves forming a vesicle around the extracellular contents and ingesting whatever is present in the surrounding media. The tumor cells in Vander Heiden's pancreatic cancer model fed biosynthesis and anaplerosis through this unusual process of macropinocytosis and protein catabolism.

Using stable isotope tracers and mass spectrometry to quantify label incorporation, Metallo made the surprising observation that branched-chain amino acids are significant contributors to fatty acid synthesis in proliferating adipocytes. Cancer cells also catabolize protein through a lysosomal pathway to fuel biosynthesis. Inhibiting autophagy and protein catabolism in tumor cells could thus be new strategies for cancer therapies. (Image courtesy of Christian Metallo)

Thompson's data show that mouse embryonic fibroblasts grown in leucine-depleted media cease proliferating. But their growth is rescued by the addition of albumin in excess of 3%. The addition of chloroquine to inhibit lysosomal proteolysis confirmed that albumin is metabolized via a macropinocytic mechanism; blocking the cell's ability to form lysosomal vesicles abrogated the albumin rescue. Alec C. Kimmelman of Harvard Medical School explained that autophagy, another lysosome-dependent pathway, could be a therapeutic target in cancers such as pancreatic adenocarcinoma. Activated KRAS–driven pancreatic cancers have historically been among the most difficult to treat, with a 5-year survival rate of 6%. Kimmelman's data show that genetic or pharmacologic inhibition of autophagy slows growth of pancreatic cancer cell lines and of tumors in genetically engineered mouse models. This work also demonstrates that autophagy has key roles in the metabolism of these tumors.

Altered metabolism facilitates the rapid proliferation of transformed cells; it is also implicated in metastasis and in the maintenance of pluripotency. Elena Piskounova of the University of Texas Southwestern Medical Center explained that successful metastasis, which is associated with worse cancer outcomes, is a relatively rare event; for every thousand cells that dissociate from a tumor, only one or two will successfully colonize a new site. Using a xenograft model with human melanoma cells introduced to immunocompromised mice, Piskounova showed that the limiting step in metastasis was cell survival in the circulatory system. Metabolic profiling revealed that these cells have high levels of oxidized glutathione and reactive oxygen species. Treatment of the xenografted mice with antioxidants increased the rate of metastasis, suggesting that, to survive, circulating tumor cells (CTCs) must overcome oxidative stress. Piskounova explained that cancer cells' need for NADPH, a reducing molecule that helps cells combat oxidative stress, might be met through folate metabolism.

If it survives oxidative stress, a CTC must adapt metabolically to the target tissue it colonizes. Sohail Tavazoie of the Rockefeller University studies how colon cancer–derived cells colonize the liver. A screen for microRNAs that suppress colon-to-liver metastasis identified mir-483 and mir-551a, which targeted the brain-type creatine kinase (CKB). Further study revealed that this kinase is secreted from cells and then converts creatine to phosphocreatine in an ATP-dependent reaction. Transport of the phosphocreatine back into the cells provides a catabolic substrate to fuel cellular energy and biosynthetic needs. Drugs that promote oxidative stress or block folate metabolism, NADPH production, or creatine utilization hold promise as therapies to prevent metastasis.

Use the tabs above to find multimedia from this event.

Presentations available from:
John Blenis, PhD (Weill Cornell Medical College)
Lydia Finley, PhD (Memorial Sloan-Kettering Cancer Center)
Elena Piskounova, PhD (University of Texas Southwestern Medical Center)
Sohail Tavazoie, MD, PhD (The Rockefeller University)
Craig B. Thompson, MD (Memorial Sloan-Kettering Cancer Center)

How to cite this eBriefing

The New York Academy of Sciences. Cancer Cell Metabolism: Unique Features Inform New Therapeutic Opportunities. Academy eBriefings. 2015. Available at: www.nyas.org/TumorMetabolism-eB