The formation of blood vessels requires rapid proliferation of endothelial cells. The transcription factors FOXO1 and MYC have been found to regulate the metabolism and proliferation of vascular endothelial cells. See Letter p.216
Research during the past decade has yielded extensive knowledge of the cellular and molecular mechanisms of angiogenesis, the process through which new blood vessels form from existing ones. Although we have learnt much about how vessels sprout, elongate, branch, form lumens and regress1, one piece of the angiogenic puzzle has remained poorly explored: how the proliferation of the endothelial cells that line the vessels' interior is regulated. In this issue, Wilhelm et al.2 (page 216) show that the transcription factor FOXO1 couples growth-factor signalling to the metabolism, growth and division of endothelial cells. The authors also identify the protein MYC, a known driver of cancer development and the anabolic metabolism that constructs tissues3, as a key mediator of endothelial FOXO1 function.
Unicellular organisms have evolved to grow and divide whenever nutrients are abundant, and, conversely, to enter states of inactivity when nutrients are scarce. Multicellular organisms function differently. In our bodies, most cells are exposed to a constant, rich supply of nutrients, but proliferate only when stimulated by growth factors, such as during organ development and regeneration. For endothelial cells, the situation differs again — these cells proliferate and form new blood vessels when oxygen and nutrients are low, with the aim of increasing oxygen and nutrient delivery to other cells in the tissues. Once functional vessels have been established, the endothelial cells, now also exposed to high levels of oxygen and nutrients, cease to proliferate.
Signalling induced when vascular endothelial growth factor A (VEGFA) binds to VEGF receptor 2 (VEGFR2) is the principal driver of most of the fundamental morphogenetic events involved in angiogenesis, including endothelial cell proliferation. A key pathway downstream of VEGFR2 is the PI3K–AKT pathway4 — a powerful regulator of glucose metabolism and protein synthesis5. The protein AKT also inhibits the activity of FOXO transcription factors by phosphorylating them: this causes them to be redistributed from the cell nucleus to the cytoplasm6 (Fig. 1).
Wilhelm et al. set out to test the idea that FOXO1, the FOXO family member enriched in endothelial cells, might constitute a link between growth-factor signalling input, metabolism and cell proliferation (Fig. 1). The researchers inactivated the Foxo1 gene specifically in endothelial cells of newborn mice and found that this led to overgrowth of these cells and the formation of a hugely disorganized and dilated vascular network in the developing mouse retina. Conversely, endothelial-specific expression of a constitutively active FOXO1 protein resulted in a sparse retinal vasculature composed of fewer than normal endothelial cells.
Expression of constitutively active FOXO1 also led to decreased glucose uptake, glycolysis and lactate production in endothelial cells. The authors further observed decreases in oxygen consumption, the production of reactive oxygen species and levels of the energy-carrying molecule ATP — all features that correspond with reduced cellular metabolic activity. The cells survived, but entered a state of metabolic quiescence (a form of dormancy), accompanied by lower expression of genes that are targeted by the transcription factor MYC.
Because MYC is known to regulate all of the above-mentioned metabolic processes, and because inhibition of MYC by FOXO occurs in other cells6, the authors tested whether MYC might be the mediator of the proliferation-stimulating effect of Foxo1 inactivation. They found that constitutively active FOXO1 suppressed MYC expression and inactivation of Foxo1 had the opposite effect. FOXO1 also increased expression of negative regulators of MYC activity, including MXI1 and FBXW7, suggesting that FOXO1 inhibits MYC at several levels. Furthermore, the authors show that MYC overexpression restored metabolism and proliferation in endothelial cells with constitutively active FOXO, and repaired vascular defects induced by this treatment. Together, these data provide compelling evidence for MYC as an effector of Foxo1 deficiency in endothelial cells.
Wilhelm and colleagues' work reveals a central mechanism whereby the control of endothelial-cell proliferation is linked to the cells' metabolic state. But as with all good studies, it generates many questions. Does the amount of FOXO1 change during angiogenesis, or is its activity regulated solely by transport in and out of the nucleus, as the authors' results might suggest (see Figure 1a of the paper2)? Although the experimental methods used by the authors are the best available, they involved vast changes in FOXO1 levels (complete loss or several-fold increase), which would not occur in physiological settings. The striking normalization of the retinal vasculature observed in mice overexpressing both constitutively active FOXO1 and MYC might result from a new balance, achieved through similarly increased levels of the two. A more critical test of the role of FOXO1 as a 'rheostat' of vascular expansion, a term used by the authors, should ideally include manipulations of the activity of FOXO1 — such as of its nuclear–cytoplasmic shuttling — at normal levels.
Moreover, the question of how FOXO1 nuclear translocation is regulated in endothelial cells remains unresolved. In analogy with the regulation of other FOXO proteins by a growth-factor–PI3K–AKT axis, one would guess that signalling downstream of VEGFA–VEGFR2 binding plays a central part. However, FOXOs have several upstream inputs besides AKT6, and FOXO1 signalling and its role in endothelial cells might be multifaceted and context-dependent. For example, in adult mice, deletion of multiple FOXO proteins, including FOXO1, leads to the formation of benign endothelial-cell tumours known as haemangiomas in some organs, but not all7. Similarly, VEGF induces enhanced proliferation in cultures of non-FOXO-expressing endothelial cells from some organs, but not others7.
Although this association between FOXO dysregulation and endothelial tumour formation concurs with Wilhelm and colleagues' idea that FOXO1 is a major regulator of endothelial-cell proliferation, and extend their observations into adult animals, the non-uniform and tissue-type-dependent responses are intriguing. Further study of FOXOs is warranted, particularly in the regulation of endothelial metabolism and proliferation at different stages of development and different vascular sites.Footnote 1
Potente, M., Gerhardt, H. & Carmeliet, P. Cell 146, 873–887 (2011).
Wilhelm, K. et al. Nature 529, 216–220 (2016).
Stine, Z. E., Walton, Z. E., Altman, B. J., Hsieh, A. L. & Dang, C. V. Cancer Discov. 5, 1024–1039 (2015).
Olsson, A.-K., Dimberg, A., Kreuger, J. & Claesson-Welsh, L. Nature Rev. Mol. Cell Biol. 7, 359–371 (2006).
Vander Heiden, M. G., Cantley, L. C. & Thompson, C. B. Science 324, 1029–1033 (2009).
Eijkelenboom, A. & Burgering, B. M. T. Nature Rev. Mol. Cell Biol. 14, 83–97 (2013).
Paik, J.-H. et al. Cell 128, 309–323 (2007).
Related links in Nature Research
About this article
Cite this article
Betsholtz, C. Transcriptional control of endothelial energy. Nature 529, 160–161 (2016). https://doi.org/10.1038/nature16866
The FASEB Journal (2020)
Trends in Cell Biology (2018)
Frontiers in Oncology (2018)
NRF2 regulates endothelial glycolysis and proliferation with miR-93 and mediates the effects of oxidized phospholipids on endothelial activation
Nucleic Acids Research (2018)