Anti-angiogenic effects of VEGF stimulation on endothelium deficient in phosphoinositide recycling

Amber N Stratman, Olivia M Farrelly, Constantinos M Mikelis, Mayumi F Miller, Zhiyong Wang, Van N Pham, Andrew E Davis, Margaret C Burns, Sofia A Pezoa, Daniel Castranova, Tina M Kilts, George E Davis, J. Silvio Gutkind, Weijun Pan, Brant M Weinstein

Preprint posted on August 28, 2018

When not recycling cleans a mess; halting tumor-associated vascular overgrowth by blocking lipid recycling

Selected by Coert Margadant

When not recycling cleans a mess; halting tumor-associated vascular overgrowth by blocking lipid recycling



The formation of new blood vessels through sprouting angiogenesis is essential for many physiological processes, but excessive angiogenesis is a hallmark of cancer. Vascular endothelial growth factor (VEGF) signaling through the receptor tyrosine kinase VEGFR2 is a key player in sprouting angiogenesis, but strategies to combat cancer by targeting the VEGF-VEGFR2 system have largely failed because tumor cells can overcome such inhibition, for example by excess production of VEGF and other pro-angiogenic cytokines. VEGF-VEGFR2 interaction activates a number of signaling events in endothelial cells, including the generation of phosphoinositide lipid secondary messengers by phosphoinositide 3-kinase and phospholipase C-gamma (PLCγ), which both use phosphatidylinositol-(4,5)-bisphosphate (PIP2) as a substrate. PLCγ converts PIP2 to IP3 and diacylglycerol (DAG). The latter is ‘recycled’ to Cytidine-Diphosphate (CDP)-DAG by CDP-Diacylglycerol Synthases or CDS enzymes, which leads to replenishment of PIP2. A previous publication from the Weinstein lab had shown that this CDS enzyme-dependent phospholipid recycling loop is required for VEGF-induced angiogenesis, because loss or knockdown of CDS enzymes reduces angiogenesis, and this can be rescued by experimental elevation of PIP2 levels (1).


Key findings

In this preprint, the authors further explore the role of PIP2 recycling in VEGF signaling and angiogenesis. Strikingly, they find in human endothelial cells in vitro and zebrafish embryos that VEGF treatment does not counteract the angiogenic defects induced by loss of CDS enzymes, but in fact exacerbates them. The VEGF effect is dose-dependent and can be rescued by PIP2 administration, suggesting that in the absence of PIP2 recycling, high VEGF concentrations deplete PIP2 levels faster and more strongly, thereby ‘exhausting’ the pool of PIP2 and halting VEGF signaling. Thus, inhibition of PIP2 recycling can actually switch VEGF from a pro-angiogenic to an anti-angiogenic stimulus.

So, does this also work in tumors? To answer this question the authors employed a number of different approaches including intravascular administration of either morpholinos against CDS2, or pharmacological agents that block phosphoinositide recycling, in two different murine tumor allograft models (melanoma and lung carcinoma). All of these treatments reduced tumor growth, which was not due to direct effects on the tumor cells, and lead to a strongly diminished vascular density. Importantly, the reduction in angiogenesis could be reversed by co-administration of PIP2 liposomes. In addition to the mentioned strategies, which systemically inhibit PIP2 recycling, the authors made similar observations in an inducible endothelial-specific CDS2 knockout mouse model. Thus, inhibition of PIP2 recycling blocks growth and expansion of tumor-associated vessels, but interestingly without effects on the endogenous vasculature. Tumor vessels are likely more prone to this treatment because they are hyperactivated and consume (and recycle) much more PIP2 than normal vessels, which are mainly quiescent. Therefore, these findings open up the intriguing possibility that blockade of phosphoinositide recycling may improve anti-angiogenic treatment in tumors.


What I like

This paper shows how a ‘classic’ pro-angiogenic signal such as VEGF can be turned into an anti-angiogenic signal by changing the balance between consumption and recycling of essential downstream mediators. Since side-effects on the normal vasculature seem minimal, probably because quiescent vessels do not consume such excess amounts of phosphoinositides, the results may guide new options to target tumor angiogenesis. Importantly, overproduction of VEGF in the tumor micro-environment will in this scenario not overcome the treatment, but may in fact enhance its effectiveness.


Future directions

Obviously, it remains to be determined how the data presented here will translate to human patients. Will vessels in different tumors respond differently? Will only vessels in VEGF-overproducing tumors be sensitive to a block in PIP2 recycling? Does the sensitivity depend on the presence of other cytokines and growth factors? While further studies will be required, the current work provides an exciting novel concept in anti-angiogenic treatment.




1 Pan W, et al. (2012) CDP-diacylglycerol synthetase-controlled phosphoinositide availability limits VEGFA signaling and vascular morphogenesis. Blood 120:489-98





Posted on: 6th September 2018

Read preprint (2 votes)

  • Have your say

    Your email address will not be published. Required fields are marked *

    This site uses Akismet to reduce spam. Learn how your comment data is processed.

    Sign up to customise the site to your preferences and to receive alerts

    Register here