Cell division requires the action of key regulator proteins called cyclins and CDKs. It emerges that a cyclin–CDK complex can regulate cell metabolism, and targeting this metabolic regulation causes tumour regression in mice. See Letter p.426
Cellular metabolism is tightly coordinated with the needs of the existing cellular state. Dividing cells must duplicate their cellular components and synthesize large amounts of proteins, lipids and DNA. Yet how metabolic processes are regulated to efficiently generate this material needed for cell division is only beginning to be understood. On page 426, Wang et al.1 now reveal a direct link between the regulation of the cell cycle and that of cell metabolism.
D-type cyclin proteins and their catalytic binding-partner enzyme, (either one of the cyclin-dependent kinases CDK4 or CDK6), are required for cell division. They exhibit peak activity2 during the early cell-cycle stage known as the G1 phase, when the cell grows in size and synthesizes components needed for DNA replication and cell division. The protein retinoblastoma (Rb) is among the most extensively studied substrates of the cyclin D–CDK complex. Progression through G1 requires the action of E2F transcription factors; however, the activity of E2F proteins is blocked when they bind to Rb (ref. 3). Phosphorylation of Rb by the cyclin D–CDK complex releases E2F proteins from their inhibitory interaction with Rb, enabling cell-cycle progression. Inhibition of CDK4 and CDK6 commonly causes cell-cycle arrest in cancer cells, mainly because the Rb–E2F complex is stabilized4. Some cancer cells die when treated with inhibitors of CDK4 and CDK6 (ref. 4).
Investigating human tumour cells grown in vitro, Wang et al. found that CDK6 inhibition induces the death of cells that predominantly use the combination of cyclin D3 and CDK6. Unexpectedly, they discovered that this cell death did not require the presence of Rb. Wang and colleagues investigated how inhibition of CDK6 resulted in cell death that was independent of the role of Rb in cell-cycle regulation.
The authors searched for CDK targets that might be relevant to this process by looking for proteins that associate with the cyclin–CDK complex. This led to the identification of the enzymes phosphofructokinase 1 (PFK1) and pyruvate kinase M2 (PKM2). The authors demonstrated that these proteins are directly phosphorylated by a complex formed of the specific combination of cyclin D3 and CDK6. PFK1 and PKM2 each exist in both dimeric and tetrameric forms, with the tetrameric forms being more active5,6. Tests to investigate the effect of phosphorylation of these enzymes gave results consistent with a model in which phosphorylation inhibits the activities of PFK1 and PKM2 by reducing the formation of tetramers in favour of the less-active dimers.
PFK1 and PKM2 function in glycolysis, a key metabolic pathway that breaks down glucose through a series of intermediates to generate the molecule pyruvate. A reduction in the activities of PFK1 and PKM2 results in the accumulation of glycolytic intermediates5,6. If this occurs, rather than progressing through glycolysis to give pyruvate, these intermediates can feed into metabolic pathways known as the pentose phosphate pathway and the serine synthesis pathway (Fig. 1). The former yields the carbohydrate ribose, and the latter the amino acids serine and glycine, which are all important substrates for nucleotide synthesis. The two pathways also generate the cofactor molecule NADPH and the antioxidant peptide glutathione, both of which can neutralize reactive oxygen species (ROS)7.
Wang et al. demonstrated that phosphorylation of PFK1 and PKM2 by cyclin D3–CDK6 increased the flow of glycolytic intermediates into the pentose phosphate and serine synthesis pathways, increased NADPH and glutathione levels, and reduced the concentration of ROS. When the authors treated cells that express high levels of cyclin D3 and CDK6 with the drug palbociclib, which inhibits CDK4 and CDK6 activity, this resulted in cell death. However, cells treated with palbociclib and an antioxidant molecule survived. In addition, cells that expressed versions of PFK1 and PKM2 that cannot be phosphorylated by cyclin D3–CDK6 were substantially protected from ROS accumulation and cell death caused by palbociclib. These results indicate that an increase in ROS generated downstream of PFK1 and PKM2 activation is responsible for the cell death seen after palbociclib treatment.
By providing evidence for a direct link between the cell cycle and cell metabolism, the work by Wang and colleagues raises a fundamental question that has largely been overlooked by biologists: how does cell metabolism intersect with the cell cycle? The biosynthetic requirements of proliferating cells have become evident as a result of the current attention given to cancer metabolism8. Wang et al. show that a cell-cycle regulator directly influences metabolic processes; however, whether other cell-cycle regulators influence metabolism or whether metabolite levels are sensed by, and directly regulate, cell-cycle regulators remain open questions.
The regulation of metabolism by cell-cycle mediators makes intuitive sense. It has been argued9 that temporal compartmentalization, in which different cellular processes occur at different times, enables the efficient coordination of metabolic activity and minimizes futile reactions. Temporal control of metabolic activity could also allow efficient channelling of products between the enzymes that synthesize the building blocks of macromolecules and those that use them. Consistent with this model, yeast grown in nutrient-limited conditions exhibits regular cycling of metabolic processes10.
The cell cycle has metabolic needs linked to its specific stages, and so components of the cell-cycle process itself would be fitting drivers of metabolic cycling. For example, the amplification of nucleotide synthesis in the G1 phase, which immediately precedes the S phase of the cell cycle in which DNA is replicated, and the downregulation of nucleotide synthesis at other stages of the cell cycle, would limit unnecessary nutrient usage and energy expenditure. How PFK1 and PKM2 phosphorylation levels vary during the cell cycle was not addressed by Wang and colleagues.
Given that the cyclical activity of CDK6 is at its peak during the G1 phase2, it could be predicted that PFK1 and PKM2 phosphorylation also cycle and exhibit peak levels during G1, resulting in maximum production of nucleotide building blocks just before DNA replication. In yeast, DNA replication is restricted to the phase of their metabolic cycle when oxygen consumption is minimal10. This is thought11 to preserve genome integrity by protecting replicating DNA from oxidative damage, which can potentially cause mutations. Wang and colleagues' observation that human cells can increase their production of both NADPH and glutathione as a consequence of the action of a G1-phase-associated kinase might be an alternative mechanism to reduce the concentration of ROS just before DNA replication.
Drugs targeting cancer metabolism are beginning to enter the clinic12. The transformation of healthy cells into cancer cells is accompanied by extensive metabolic changes, but which metabolic reactions are essential for proliferation is unclear. It seems reasonable to speculate that the metabolic processes directly regulated by the cell cycle are those most essential for cell division.
Wang et al. transplanted 33 samples of human tumour cells into mice and treated them with ribociclib, another inhibitor of CDK4 and CDK6. This reduced the growth rate for most of the tumours, probably because of Rb sequestration of E2F transcription factors and cell-cycle arrest, but did not cause the tumours to shrink. However, the three tumours that had high levels of cyclin D3–CDK6 complexes decreased in size. Further insight into connection points between the cell cycle and metabolism might pave the way for the development of successful cancer therapies that can target such metabolic vulnerabilities of tumour cells.Footnote 1
Wang, H. et al. Nature 546, 426–430 (2017).
Matsushime, H. et al. Mol. Cell. Biol. 14, 2066–2076 (1994).
Deshpande, A., Sicinski, P. & Hinds, P. W. Oncogene 24, 2909–2915 (2005).
O'Leary, B., Finn, R. S. & Turner, N. C. Nature Rev. Clin. Oncol. 13, 417–430 (2016).
Yi, W. et al. Science 337, 975–980 (2012).
Anastasiou, D. et al. Science 334, 1278–1283 (2011).
Cairns, R. A., Harris, I. S. & Mak, T. W. Nature Rev. Cancer 11, 85–95 (2011).
Vander Heiden, M. G., Cantley, L. C. & Thompson, C. B. Science 324, 1029–1033 (2009).
Tu, B. P. & McKnight, S. L. Nature Rev. Mol. Cell. Biol. 7, 696–701 (2006).
Tu, B. P., Kudlicki, A., Rowicka, M. & McKnight, S. L. Science 310, 1152–1158 (2005).
Chen, Z., Odstrcil, E. A., Tu, B. P. & McKnight, S. L. Science 316, 1916–1919 (2007).
Gross, M. I. et al. Mol. Cancer Ther. 13, 890–901 (2014).
Related links in Nature Research
About this article
Precise Regulation of Enzyme Cascade Catalytic Efficiency with DNA Tetrahedron as Scaffold for Ultrasensitive Electrochemical Detection of DNA
Analytical Chemistry (2019)
Isolation and identification of phosphorylated lysine peptides by retention time difference combining dimethyl labeling strategy
Science China Chemistry (2019)
Synthesis, anticancer activity and mechanism of iron chelator derived from 2,6-diacetylpyridine bis(acylhydrazones)
Journal of Inorganic Biochemistry (2019)
Electrocatalytic Efficiency Regulation between Target-Induced HRP-Mimicking DNAzyme and GOx with Low Background for Ultrasensitive Detection of Thrombin
Analytical Chemistry (2019)
International Journal of Molecular Sciences (2017)