Understanding the Warburg Effect: The Metabolic Requirements of Cell Proliferation

W. H. Koppenol and P. L. Bounds highlight an often misunderstood feature of proliferative cell metabolism. There is substantial evidence demonstrating oxidative phosphorylation persists in the vast majority, if not all, tumors. We argued recently that continued oxidative phosphorylation was indeed associated with tumor cell metabolism (1) and depicted in Figure 2 of our Review that at least 5% of the glucose in proliferating cells is metabolized in the mitochondria ("Understanding the Warburg Effect: The metabolic requirements of cell proliferation," M. G. Vander Heiden et al., 22 May 2009, p. 1029). The observation that cancer cells continue to rely on oxidative phosphorylation, which requires oxygen, further underscores that the metabolic changes observed in proliferating cells are largely independent of oxygen availability. Nevertheless, the proliferative advantage of aerobic glycolysis with increased glucose uptake in cancer cells remains undefined and likely extends beyond the accounting for adenosine 5'-triphosphate (ATP) generation.

In most proliferating cells, oxidative phosphorylation continues to supply some of the ATP required for housekeeping functions and biosynthetic reactions. We agree that due to the substantial increase in glucose utilization, the ATP produced by glycolysis alone can contribute significantly to the total ATP production of proliferating cells. Attempts to quantitate ATP production experimentally in cancer cells have estimated that 80% of the ATP generated is derived from oxidative pathways while 20% comes from glycolysis (2), and others report that up to 50% of the ATP in cancer cells can be derived from glycolysis (3). These values are in agreement with Koppenol and Bounds' accounting of ATP production in cells which rely on aerobic glycolysis. However, it is also important to note that a significant percentage of the oxygen consumed by tumors' cells is not the result of glucose oxidation (2, 4). Alternate fuel sources also contribute to oxidative reactions in cancer cells as highlighted by work demonstrating the importance of glutamine for cancer cell proliferation (5).

While the combination of glycolytic and oxidative metabolism satisfies the ATP requirements of cell proliferation, there is evidence that the rate of aerobic glycolysis, including the accompanying increased rate of glucose utilization, is not elevated to satisfy ATP demand. The rate-limiting step in glycolysis of Ehrlich ascites tumors cells, the exact system used by Warburg in 1924 (6), was shown in 1973 to be limited by the consumption of ATP, not production (7). Efraim Racker, in his accounting of tumor cell aerobic glycolysis, stressed the importance of consuming ATP to support the rate of glycolysis in tumor cells (8). An important challenge to understanding proliferative cell metabolism is not determining how enough ATP is generated for growth, but rather how enough ATP is consumed to support the elevated flux through glycolysis (8). This consumption of ATP is needed to avoid allosteric inhibition of key rate limiting steps in glycolysis by a high ATP/AMP ratio (9).

In our Review we raise the possibility that a major advantage of aerobic glycolysis is to meet the metabolic requirements involved in duplicating biomass which extend beyond ATP generation. We propose that the increased uptake of glucose, despite much of it being converted to lactate, enables cells to meet the additional biosynthetic demands of cell proliferation including the generation of sufficient reducing equivalents as well as transformation of carbon into the appropriate scaffolds for biosynthesis. Most cancer cells do indeed have an intact ability to use oxidative phosphorylation. The characterization of fuel sources that support oxidative phosphorylation in cancer cells and their contribution to ATP production are important. However, understanding the metabolic state Warburg described years ago will require a careful analysis of metabolism that extends beyond an accounting of ATP production, lactate production and oxygen consumption.

Matthew G. Vander Heiden

Dana-Farber Cancer Institute, Beth-Israeil Deaconess Medical Center, and Harvard Medical School, Boston, MA 02115, USA.

Lewis C. Cantley

Beth-Israel Deaconess Medical Center and Harvard Medical School, Boston, MA 02115, USA.

Craig B. Thompson

Abrahamson Family Cancer Research Institute, University of Pennsylvania, Philadephia, PA 19104, USA.

References

1. H. R. Christofk et al., Nature 452, 230 (2008).

2. M. Guppy, P. Leedman, X. Zu, V. Russell, Biochem. J. 364, 309 (2002).

3. H. Schmidt, W. Siems, M. Muller, R. Dumdey, S. M. Rapoport, Exp. Cell Res. 194, 122 (1991).

4. R. J. DeBerardinis et al., Proc. Natl. Acad. Sci. U S A 104, 19345 (2007).

5. D. R. Wise et al., Proc. Natl. Acad. Sci. U S A 105, 18782 (2008).

6. O. Warburg, K. Posener, E. Negelein, Biochem. Z. 152, 319 (1924).

7. P. Scholnick, D. Lang, E. Racker, J. Biol. Chem. 248, 5175 (1973).

8. E. Racker, J. Cell. Physiol. 89, 697 (1976).

9. A. L. Lehninger, D. L. Nelson, M. M. Cox, Principles of Biochemistry (Worth Publishers, New York, ed. 2, 1993).

M. G. Vander Heiden et al. addressed the question of why a cell would "choose" glycolysis over oxidative phosphorylation, whereby only 2 instead of 36 adenosine 5'-triphosphate (ATP) are obtained per glucose molecule ("Understanding the Warburg Effect: The metabolic requirements of cell proliferation," Reviews, 22 May 2009, p. 1029). However, the question, as well as the proffered explanations, misses the point of Warburg's work. In 1923 and 1924, Warburg and co-workers reported that the rate of respiration in cancer cells is, within the error, identical to that of normal cells, and that the glucose uptake is approximately 10 times higher. Furthermore, for every 13 glucose molecules taken up, 1 is oxidized via respiration, while the remaining 12 are split to form lactic acid (1, 2). Thus, in the time that one glucose molecule produces 36 ATP via respiration, 24 additional ATP are generated via aerobic glycolysis. The question posed by Vander Heiden et al., "why do proliferating cells switch to a less efficient metabolism" is wrong on two counts: (i) cancer cells do not switch, but carry out both oxidative phosphorylation and aerobic glycolysis simultaneously, and (ii) glycolysis is not inefficient; even in the absence of dioxygen, cancer cells survive because lactic acid production via anaerobic glycolysis yields 2/3 of the ATP that a normal cell produces by respiration. Indeed, it is precisely this enormous uptake of glucose that allows cancerous tissue to be visualized by FDG-PET/CT, as shown in Fig. 4 of Vander Heiden et al.

The question of whether or not cancer cells respire normally has dogged the literature since the late 1920s. Experimental results of Warburg and co-workers and of Chance and co-workers (3, 4) clearly show that respiration is normal. Why then did Warburg himself in 1956 write that "…the respiration of all cancer cells is damaged..." (5)? Warburg later provided clarification that has largely gone unnoticed, writing that the respiration of cancer cells is small relative to the consumption of glucose, but not small relative to the respiration of normal cells (6). The confusion may have been prevented had Warburg formulated his definition of respiratory impairment 30 years earlier.

Willem H. Koppenol and Patricia L. Bounds

Institute of Inorganic Chemistry, Department of Chemistry and Applied Biosciences, ETH, Wolfgang-Pauli-Strasse 10, CH-8093 Zürich, Switzerland.

References

1. S. Minami, Biochem. Z. 142, 334 (1923).

2. O. Warburg, K. Posener, E. Negelein, Biochem. Z. 152, 309 (1924).

3. B. Chance, L. N. Castor, Science 116, 200 (1952).

4. B. Chance, B. Hess, Science 129, 700 (1959).

5. O. Warburg, Science 124, 269 (1956).

6. O. H. Warburg, New Methods of Cell Physiology Applied to Cancer, Photosynthesis, and Mechanism of X-Ray Action (Interscience Publishers, New York, 1962), pp. 631-632.