Metabolism in tumour cells is adapted to the demands of a growing cell. The Warburg effect and increased use of glutamine are two well-known adaptations of tumour metabolism. Both require transporters to allow uptake of substrates and efflux of products. Differentiated cells are less reliant on these pathways and as a result are less vulnerable to drugs that curtail nutrient uptake. Thus drugs that reduce nutrient uptake are promising candidates for tumour therapy. Detailed understanding of tumour cell biology will allow the generation of new chemotherapeutic drugs with limited side effects. In this issue of the Biochemical Journal, Romero Rosales et al. have identified the mechanism by which the sphingolipid analogue FTY720 inhibits tumour growth.
- nutrient limitation
- sphingosine 1-phosphate
- Warburg effect
Tumour therapy has for many years relied on fairly non-specific drugs that are severely cytotoxic . With the increased understanding of tumour cell biology , more targeted drugs have been developed, such as Herceptin™  targeting a subset of breast cancers and Gleevec™  for chronic myeloid leukaemia. Herceptin is a recombinant antibody, binding to her 2 [EGFR (epidermal growth factor receptor) isoform 2], which is overexpressed in some types of breast cancer. Blocking the receptor in her 2-positive breast cancers reduces proliferative signals from the EGFR, thereby stopping cell growth . In chronic myeloid leukaemia, a translocation between chromosome 9 and 22 results in the generation of an artificial fusion protein comprising the c-ABL tyrosine kinase and the BCR (breakpoint cluster region) protein [5,6]. This BCR-ABL kinase is overactive, causing rapid division of a specific group of white blood cells. These examples illustrate that understanding of tumour cell biology can result in the generation of selective antitumour drugs.
Tumour cells not only have overly active signal transduction pathways, but also rely on an altered metabolism to support proliferation. Differentiated cells are largely catabolic, using nutrients to provide energy for their physiological tasks; net synthesis of cellular components is limited . A dividing cell, by contrast, requires production of DNA, RNA, membrane lipids and protein. Preventing nutrient supply to tumour cells or inhibiting essential anabolic pathways can thus be a powerful strategy to stop tumour growth.
One of the earliest strategies to specifically block metabolism of a growing cell has been the use of the folate analogue aminopterin to inhibit the recycling of dihydrofolate generated during the synthesis of thymidine . Methotrexate, which blocks the same step, is still in use as a cancer drug today . Another widely used chemotherapeutic drug is 5-fluorouracil, which blocks thymidilate synthase.
Tumour cells have additional metabolic adaptations to sustain growth, of which the Warburg effect is the best known . It refers to an increased use of glucose and production of lactate in tumour cells that forms the basis of tumour diagnosis by PET using 2-[18F]fluoro-2-deoxyglucose . Initially, production of lactate from glucose seems counterintuitive for a growing cell, as glycolysis provides less ATP than respiration. The argument has been put forward that the lack of oxygen in cancer cells causes the switch . Independent experiments, however, revealed that cancer cells produce large amounts of lactate, while still actively respiring [13,14]. The main advantage of the Warburg effect is an uncoupling of glycolysis from the TCA (tricarboxylic acid) cycle. This allows the TCA cycle to play a largely anabolic role by providing precursors for nucleotide biosynthesis (aspartate from oxaloactetate), protein biosynthesis (amino acids from various intermediates) and fatty acid biosynthesis (acetyl-CoA from citrate). The rate of glycolysis automatically increases to meet the demand of ATP for cell growth. More ATP is provided through glutaminolysis, another metabolic adaptation of tumour cells . This pathway provides energy from the oxidation of glutamine into lactate, aspartate or alanine using a partial TCA cycle. In addition, glutamine is used as an anaplerotic molecule for the TCA cycle to allow synthesis of precursors for anabolic reactions. Glutamine, which is a non-essential amino acid in most cells, thus becomes an essential amino acid for tumour cells . Similarly, asparaginase is used in cancer therapy to curtail the uptake of asparagine. Taken together, cancer cells have a high demand for glucose and amino acids for biosynthetic purposes and energy production. This suggests that blocking nutrient uptake at the plasma membrane would efficiently inhibit the growth of tumour cells. In addition, transporters have shown to provide substrates to the amino-acid-sensing mTOR (mammalian target of rapamycin) complex, the activation of which is permissive for cell growth [17,18].
Nutrient supply to tumours can be targeted at the tissue level by inhibiting the growth of blood vessels . In the absence of vascularization, tumours can only grow to very small sizes because metabolism is limited by non-specific diffusion of nutrients through the extracellular space. On a cellular basis, this is mirrored by the need for transporters to bring nutrients into the cell. It is in this context that Romero Rosales et al.  have identified the mechanism by which the sphingolipid analogue FTY720 inhibits tumour growth. The study shows that FTY720 down-regulates a number of nutrient transporters with high efficacy. FTY720 inhibits both S1P (sphingosine 1-phosphate) lyase and sphingosine kinase. S1P is a lipid second messenger that acts through four different G-protein-coupled receptors . It promotes cell proliferation and is involved in chemotaxis and cytoskeletal rearrangements. As a result, FTY720 has been evaluated for cancer therapy , but the mechanism by which it kills tumour cells has remained unknown. Rosales et al.  show that FTY720 decreases the surface levels of the 4F2hc (4F2 heavy chain), which is a trafficking subunit for a variety of neutral amino acid transporters, such as LAT1 (L-type amino acid transporter 1), LAT2 and asc . In addition, it also down-regulates surface expression of GLUT1 (glucose transporter 1) and the cationic amino acid transporter CAT-1. Interestingly, even a partial down-regulation of these transporters (>30%) caused cell death. Autophagy was induced in these cells, owing to lack of nutrients, and mTOR activity was also reduced. Addition of membrane-permeant nutrients such as dimethyl succinate and oleic acid counteracted the action of FTY720. FTY720 was also active in a mouse model of BCR-ABL-driven leukaemia. The mechanism by which FTY720 induces transporter withdrawal from membranes still needs to be established. It appears to be independent of PKC (protein kinase C) signalling, which also down-regulates CAT-1  and 4F2hc . One of the reasons why FTY720 is currently not used for tumour therapy is that phosphorylated FTY720 is an agonist on S1P receptors (S1P3), activating a G-protein-controlled potassium channel in the heart, thereby slowing down the heart rate. However, the dose required to trigger nutrient transporter withdrawal is several orders of magnitude above the EC50 for S1P receptors, suggesting that it acts via two different pathways. Accordingly, phosphorylated FTY720 does not affect surface expression of transporters. Moreover, another sphingosine analogue AAL-149, which cannot activate S1P receptors, was still able to down-regulate transporters. This provides proof of concept that both functions can be separated and that drugs can be identified that specifically down-regulate transporter expression.
Overall, inhibition of nutrient uptake is a promising strategy to inhibit tumour growth. It is fairly selective due to the metabolic demands of tumour cells. The potential of this strategy is illustrated by a differential drug screen carried out on renal carcinoma cells with a mutation in the VHL (von Hippel–Lindau) tumour suppressor gene compared with cells with a re-introduced wild-type VHL gene. The screen identified a group of compounds that inhibit expression of GLUT1, which is highly up-regulated in renal carcinoma cells owing to the Warburg effect . The drugs effectively kill VHL-deficient cells, whereas there is little effect on wild-type cells. In mice, the drugs reduced glucose uptake into the tumour and reduced tumour growth. GLUT1 is the major GLUT in the blood–brain barrier and it remains to be shown whether these drugs have side effects in humans on brain metabolism, however, a ketogenic diet might be used to overcome reduced GLUT1 activity at the blood–brain barrier. The biggest challenge of targeting metabolism for tumour therapy is the selectivity against stem cells. Stem cells are likely to use the same metabolism as cancer cells. As a result, many drugs that block tumour metabolism are likely to affect stem cell metabolism, particularly in haemopoietic stem cells.
Abbreviations: 4F2hc, 4F2 heavy chain; BCR, breakpoint cluster region; EGFR, epidermal growth factor receptor; GLUT, glucose transporter; her 2, EGFR isoform 2; LAT, L-type amino acid transporter; mTOR, mammalian target of rapamycin; S1P, sphingosine 1-phosphate; TCA, tricarboxylic acid; VHL, von Hippel–Lindau
- © The Authors Journal compilation © 2011 Biochemical Society