Analysis of endothelial cells, which are involved in blood-vessel formation, unexpectedly reveals that proliferation in this cell type depends on fatty-acid oxidation to support DNA synthesis. See Article p.192
The formation of blood vessels requires the migration and proliferation of a vascular cell type called endothelial cells. Most cells assimilate the biomass and energy required for proliferation by converting abundant carbon sources such as sugars and amino acids into macromolecular building blocks. But on page 192 of this issue, Schoors et al.1 report that the oxidation of fatty acids to acetyl-CoA molecules generates an unexpected carbon source that is required in endothelial cells to produce nucleotides for DNA synthesis. Blocking fatty-acid oxidation suppresses endothelial-cell proliferation and protects mice from a common form of vision loss called retinopathy of prematurity, which is caused by uncontrolled blood-vessel formation.
Intermediary metabolism — the process by which cells produce and consume energy — is akin to a network of roads that facilitate traffic flow. The overall function of the network relies on a finite set of major intersections at which pathways meet, before diverging to feed sub-networks. Acetyl-CoA is positioned at one of the most complex of these intersections. This metabolite is produced from the degradation of carbohydrates, lipids and proteins, and feeds carbon into the tricarboxylic acid cycle, a major hub for both energy formation and macromolecular synthesis. Furthermore, because acetyl-CoA is the substrate for acetylation, a molecular modification to DNA-associated proteins that is typically associated with active transcription, it is intimately linked with gene expression and the decision to enter a proliferative state2,3, thereby tethering intermediary metabolism to cell function.
During angiogenesis (the formation of blood vessels), endothelial cells respond to extracellular signals by activating a complex program of proliferation and migration. The energy sources in angiogenic endothelial cells have been characterized4, but the sources of acetyl-CoA, and their role in angiogenesis, have remained unknown. In some energetically demanding tissues, such as the heart and skeletal muscle, oxidation of long-chain fatty acids provides a rich source of acetyl-CoA. This pathway involves a repeating sequence of reactions in organelles called mitochondria, with each round generating reducing equivalents to produce energy and liberating two carbons from the fatty acid as acetyl-CoA.
To study the role of fatty-acid oxidation (FAO) in the endothelial cells of mice, Schoors et al. inactivated carnitine palmitoyltransferase 1A (CPT1A), an enzyme required for mitochondria to import long-chain fatty acids. This rendered endothelial cells unable to proliferate, impairing vessel sprouting and angiogenesis both in cultured cells and in the mouse retina.
Surprisingly, despite the well-known role of FAO in energy production, the authors found that the amount of energy generated by endothelial cells was not altered when CPT1A was absent. Instead, several metabolites produced from acetyl-CoA became depleted, particularly deoxyribonucleotide triphosphates (dNTPs), the building blocks for DNA. Consistent with a specific role for FAO in supporting DNA synthesis, CPT1A loss did not impair the production of either proteins or RNA. Furthermore, providing endothelial cells with dNTPs or acetate, an acetyl-CoA precursor, completely reversed the effects of CPT1A loss on proliferation. Endothelial cells were unusual in this respect; most other proliferating cells, including cancer cell lines, did not use fatty acids as a major carbon source for DNA synthesis.
These findings are of interest for several reasons. First, the requirement for FAO in supporting DNA synthesis would have been difficult to predict on the basis of existing work in proliferative-cell metabolism. Research in cancer-cell metabolism5 has emphasized the role of glucose and glutamine in feeding the acetyl-CoA pool during growth, and recent work6 suggests that acetate is also a source of acetyl-CoA in tumours. But even though some tumours require FAO to survive7, the fact that FAO provides carbon for DNA synthesis was unexpected.
Second, the work strikingly demonstrates divergent metabolic requirements for the two major activities of endothelial cells during angiogenesis: migration and proliferation (Fig. 1). Endothelial cells that lack CPT1A fail to proliferate normally, but their migration is unperturbed. Thus, FAO has a more specialized role in endothelial cells than does glucose metabolism, which these authors previously demonstrated was required for both migration and proliferation4.
It is to be hoped that Schoors and colleagues' study will stimulate more work to understand this unusual form of metabolic specialization. Given the availability of multiple carbon sources for acetyl-CoA and the tricarboxylic acid cycle, it is intriguing that endothelial cells cannot compensate for CPT1A loss. Does FAO establish a metabolically distinct acetyl-CoA pool that is preferentially channelled towards dNTP synthesis? Additional sophisticated metabolic flux studies, similar to those the authors perform in the paper, might help to answer this question.
An alternative, but not mutually exclusive, explanation is that FAO provides metabolic benefits beyond supplying the cell with carbon. A striking finding from the paper is that CPT1A loss reduces dNTP levels without affecting the precursors of RNA, ribonucleotide triphosphates (rNTPs), perhaps providing a clue to the exquisite requirement of these cells for FAO. Converting rNTPs to dNTPs requires reducing equivalents, which are produced in abundance by FAO. In other models of cell proliferation, conditions that stimulate FAO at the expense of fatty-acid synthesis improve the overall availability of reducing equivalents8. It is therefore possible that subtle redox changes contribute to the selective depletion of dNTPs in endothelial cells lacking CPT1A.
The metabolic dependency unveiled by this study has therapeutic potential for diseases that are associated with abnormal endothelial-cell proliferation. The authors demonstrate that systemic FAO blockade with a chemical inhibitor alleviates excessive angiogenesis in a mouse model of retinopathy of prematurity. This disease affects more than 50% of newborn babies with extremely low birth weights worldwide, and is a major source of long-term impairment of visual function. It will be interesting and important to determine whether other forms of abnormal angiogenesis, including the neovascularization of tumours, also require FAO.
Schoors, S. et al. Nature 520, 192–197 (2015).
Cai, L., Sutter, B. M., Li, B. & Tu, B. P. Mol. Cell 42, 426–437 (2011).
Wellen, K. E. et al. Science 324, 1076–1080 (2009).
De Bock, K., Georgiadou, M. & Carmeliet, P. Cell Metab. 18, 634–647 (2013).
Hensley, C. T., Wasti, A. T. & DeBerardinis, R. J. J. Clin. Invest. 123, 3678–3684 (2013).
Comerford, S. A. et al. Cell 159, 1591–1602 (2014).
Caro, P. et al. Cancer Cell 22, 547–560 (2012).
Jeon, S.-M., Chandel, N. S. & Hay, N. Nature 485, 661–665 (2012).
About this article