News & Views | Published:


Endothelial metabolism—more complex (III) than previously thought

Nature Metabolismvolume 1pages1415 (2019) | Download Citation

The growth of new blood vessels (angiogenesis) is a bio-energetically demanding process. Surprisingly, the specific role of mitochondria in angiogenesis remains unclear. A study in this issue of Nature Metabolism now demonstrates that mitochondrial respiration is essential for angiogenic growth by controlling endothelial proliferation.

The emergence of mitochondria is considered to be a fundamental step in the evolution of life. By maximizing energy production, this double-membrane-bound organelle helped meet the higher energy demands of multicellular organisms and thus became an essential part of most eukaryotes1. Mitochondria generate energy in the form of adenosine triphosphate (ATP) in a process termed mitochondrial oxidative phosphorylation (OXPHOS). This involves the oxidation of nutrients (for example, glucose) in the tricarboxylic acid (TCA) cycle, whose chemical energy is harnessed to reduce the electron carriers nicotinamide adenine dinucleotide (NAD+) and flavin adenine nicotinamide (FAD). Their high-energy electrons are transferred to the electron transport chain (ETC) to power the synthesis of ATP at the inner mitochondrial membrane. Because oxygen (O2) is required as the final electron acceptor in the ETC, O2 is essential for OXPHOS2.

Most differentiated cells (for example, neurons and cardiomyocytes) rely on this type of mitochondrial respiration to provide sufficient energy for their cellular functions3. An exception is the vascular endothelium, the single layer of cells that lines the interior of all blood vessels. Its endothelial cells (ECs) have immediate access to ample O2 in the bloodstream, yet they generate 85% of their ATP demand via aerobic glycolysis4. Although aerobic glycolysis is an inefficient way to generate ATP, it confers metabolic advantages to the endothelium. For instance, glycolysis does not require O2 for energy generation, thus allowing ECs to thrive in O2-depleted (hypoxic) environments. This property is particularly relevant for the growth of new blood vessels (angiogenesis), during which ECs must migrate and proliferate in hypoxic tissues (Fig. 1a)5. These considerations have led to the conception that ECs do not utilise mitochondria during angiogenesis and that mitochondrial respiration is dispensable for endothelial functions. A study by Diebold et al.6 in this issue of Nature Metabolism now shows that this view needs to be revised. The authors provide evidence that mitochondrial respiration is necessary for endothelial proliferation and that its inhibition blocks angiogenesis under physiological and pathological conditions.

Fig. 1: Mitochondrial respiration supports endothelial proliferation.
Fig. 1

a, Schematic of angiogenic blood vessels, highlighting the proliferative activity of ECs in the nascent vasculature. Proliferating ECs are indicated in red, while non-proliferating ECs are shown in yellow. b, Diebold et al.6 show that ECs use mitochondrial respiratory chain metabolism primarily to recycle NAD+, which is needed for biosynthetic reactions (for example, synthesis of aspartate) during cell proliferation. c, When complex III of the ETC is blocked, ECs have lower levels of NAD+ and aspartate and thus are unable to proliferate. The thinner arrows in c indicate reduced mitochondrial flux relative to b. ADP, adenosine diphosphate; α-KG, alpha-ketoglutarate; H+, proton; e, electron; H2O, water; CI–CV, complex I–complex V, respectively.

The authors set out to investigate the role of oxidative mitochondrial metabolism during angiogenesis by inhibiting the complexes of the respiratory chain. They first cultured ECs with the complex III inhibitor antimycin A, which led to the expected loss of mitochondrial respiration. This drug also caused a profound inhibition of endothelial proliferation, while migration and invasion into a three-dimensional matrix were unaffected. The researchers repeated these experiments with a complex I inhibitor called piericidin, which yielded similar results. Because endothelial survival was not compromised under these conditions, the authors argue that mitochondrial respiration has functions beyond (homeostatic) ATP production that are necessary for endothelial proliferation.

Mitochondrial respiration is also coupled to oxidative TCA cycle flux, which generates metabolic intermediates for macromolecular biosynthesis and proliferation. Therefore, Diebold et al.6 performed metabolic profiling in respiration-inhibited ECs and observed a reduction in TCA cycle metabolites as well as a depletion of the non-essential amino acid aspartate. However, supplementing respiration-deficient ECs with aspartate did not normalize their proliferative activity. Instead, restoring the NAD+/NADH ratio, which was also decreased in antimycin A–treated ECs, rescued aspartate levels and proliferation. These data suggest that a primary function of the endothelial respiratory chain is to recycle NAD+, which is needed for biosynthetic processes such as the generation of aspartate (Fig. 1b,c).

To gain further insights into the physiological relevance of these findings, Diebold et al.6 inactivated the mitochondrial respiratory chain in ECs of newborn mice and analysed angiogenesis in the retina. To this end, the researchers deleted the ubiquinol-cytochrome c reductase, complex III subunit VII (Uqcrq) gene, which is a critical component of the ETC. Loss of this gene had dramatic consequences for retinal angiogenesis and resulted in a sparse vascular network with fewer ECs, which were also less proliferative. Consistent with the authors’ in vitro studies, these effects were independent of cell death. Strikingly, most of the endothelial-specific Uqcrq knockout mice died between 3 and 4 weeks of age, underscoring the importance of the mitochondrial respiratory chain for the expanding endothelium. Searching for an underlying molecular mechanism, the researchers purified ECs from mouse lung, a tissue that is rich in ECs. Gene expression analysis did not reveal mechanistic insights into the UQCRQ-mutant phenotype. Therefore, the authors further analysed metabolite levels and found that Uqcrq deficiency caused a global reduction in amino acid levels—possibly explaining the proliferation defects in these cells. Finally, Diebold et al.6 turned to adult mice to investigate the functional consequences of endothelial UQCRQ deficiency on aberrant angiogenic responses in tumours. Here too, genetic inactivation of complex III reduced endothelial proliferation and vascular growth, giving rise to tumours that were smaller than the ones grown in wild-type mice.

Taken together, the Diebold et al.6 experiments reveal that angiogenic ECs not only are capable of mitochondrial respiration, but also require respiration for vascular growth. Their work further illustrates that endothelial mitochondria function as biosynthetic organelles that support cell proliferation—a finding that aligns with previous observations in cancer cells7,8,9. However, as with all unexpected results, many open questions remain. For instance, does the extent of mitochondrial respiration change during angiogenesis, or is its activity solely determined by metabolic flux and oxygen availability? Similarly, how do ECs coordinate aerobic glycolysis and respiratory-chain-linked metabolism during the angiogenic process? The authors’ studies suggest that both processes run simultaneously, rather than replacing each other, but exactly how this works remains unknown. Another question of interest is whether all ECs respond similarly to changes in mitochondrial respiration. The authors analysed retinal and lung ECs and obtained similar results, but ECs in other vascular beds might react differently. Indeed, it is becoming increasingly clear that ECs are a heterogeneous cell population with distinct molecular and functional properties and that these cells are confronted with different metabolic environments10. For instance, ECs at the blood–air barrier in the lung face high oxygen levels, whereas ECs in the bone marrow reside in a relatively hypoxic milieu. It is thus conceivable that organ-specific ECs have distinct rates of mitochondrial respiration, which might co-determine sensitivity or outcome to respiratory chain inhibition. These questions notwithstanding, Diebold et al.6 provide exciting insights into how ECs utilise mitochondrial metabolism and suggest that mitochondria have underappreciated potential for groundbreaking findings in vascular biology and beyond.


  1. 1.

    Roger, A. J., Muñoz-Gómez, S. A. & Kamikawa, R. Curr. Biol. 27, R1177–R1192 (2017).

  2. 2.

    Palm, W. & Thompson, C. B. Nature 546, 234–242 (2017).

  3. 3.

    Vander Heiden, M. G., Cantley, L. C. & Thompson, C. B. Science 324, 1029–1033 (2009).

  4. 4.

    De Bock, K. et al. Cell 154, 651–663 (2013).

  5. 5.

    Potente, M. & Carmeliet, P. Annu. Rev. Physiol. 79, 43–66 (2017).

  6. 6.

    Diebold, L. P. et al. Nat. Metab. (2019).

  7. 7.

    Sullivan, L. B. et al. Cell 162, 552–563 (2015).

  8. 8.

    Birsoy, K. et al. Cell 162, 540–551 (2015).

  9. 9.

    Titov, D. V. et al. Science 352, 231–235 (2016).

  10. 10.

    Potente, M. & Mäkinen, T. Nat. Rev. Mol. Cell Biol. 18, 477–494 (2017).

Download references

Author information


  1. Angiogenesis & Metabolism Laboratory, Max Planck Institute for Heart and Lung Research, Bad Nauheim, Germany

    • Jorge Andrade
    •  & Michael Potente
  2. International Institute of Molecular and Cell Biology, Warsaw, Poland

    • Michael Potente
  3. DZHK (German Center for Cardiovascular Research), Frankfurt Rhine-Main, Berlin, Germany

    • Michael Potente


  1. Search for Jorge Andrade in:

  2. Search for Michael Potente in:

Competing interests

The authors declare no competing interests.

Corresponding author

Correspondence to Michael Potente.

About this article

Publication history


Issue Date


Newsletter Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing