(B-59) Mechanistic computational modeling of sFLT1 secretion in endothelial cells

Introduction: The vascular endothelial growth factor (VEGF) family of ligands binds to cell-surface associated receptor tyrosine kinases (VEGFRs) on endothelial cells and regulates blood vessel growth. VEGF ligands can also bind to a soluble isoform of VEGFR1, sFLT1, and when bound cannot activate the VEGFRs. sFLT1 is secreted by endothelial cells and is thus a key extracellular VEGF sink and angiogenesis inhibitor which is essential for normal blood vessel development and dysregulated in vascular pathology. Despite the importance of sFLT1 for angiogenesis, little is known about the dynamics of sFLT1 secretion or relationships between intracellular and extracellular sFLT1 pools. Using data from a recent quantitative experimental study of sFLT1 secretion, as well as published sFLT1 secretion data, we developed a mechanistic computational model to quantitatively describe the kinetics of sFLT1 production, secretion, and degradation in endothelial cells.


Materials and Methods: We modeled sFLT1 secretion using a system of coupled deterministic delay differential equations (DDEs); we used these because regular non-delay ordinary differential equations (ODEs) could not capture the observations of pulse-chase experiments. The DDEs describe the concentrations of intracellular and extracellular sFLT1 pools as controlled by key mechanistic processes including production, secretion, intracellular degradation, and extracellular degradation. Parameter values for the model were derived from nonlinear least squares optimization of our model to experimental data from both novel and previously published time courses of sFLT1 secretion in human umbilical vein endothelial cells (HUVECs) under different conditions.1–3 We then used simulations to explore the impact of pharmacological or genetic inhibition of trafficking processes and compared our model predictions to cell culture experiments using trafficking inhibitors or RNAi.

Results, Conclusions, and Discussions: Using optimization methods applied to the mechanistic model and the available experimental data, we found multiple sets of parameter values that recapitulated observed sFLT1 trafficking dynamics across 3 independent studies. While at first it seemed that several parameters were unconstrained, close inspection identified that those parameters were collectively constrained to satisfy fixed algebraic relationships such that there was only one free parameter in the model.

Selecting a reasonable value for the level of intracellular sFLT1 was sufficient to enable these constraints to define specific values for the rate of production and rate constants of secretion and degradation. The resulting calculated secretion rate is comparable to the secretion rates of other key endothelial proteins in HUVECs4 and in excess of the predicted secretion rate required to create significant interstitial sFLT1 reservoirs in tissues.5



Our simulations of chemical inhibitors suggest that different inhibitors have different efficacies of sFLT1 trafficking inhibition. We calculated the fraction inhibition of trafficking rate constants required to reproduce changes in sFLT1 pools observed experimentally upon treatment with secretory pathway inhibitors. Our simulations suggest that at the tested doses, the Golgi inhibitor brefeldin inhibits the secretion rate constant by 90%, while both the exocytosis inhibitor TATNSF700 and clathrin-mediated trafficking inhibitor chlorpromazine inhibit the secretion rate constant by 60%.



This model can help us understand the levels and dynamics of intracellular and extracellular sFLT1, an essential prerequisite for understanding VEGF signaling network responses to physiological conditions, pathology, and pharmacological manipulation. We can add additional subcellular detail to the trafficking processes and integrate the model into larger mechanistic models of VEGFs and their signaling receptors.


Acknowledgements (Optional) :

References (Optional) : 1. Hornig, C. et al. Release and Complex Formation of Soluble VEGFR-1 from Endothelial Cells and Biological Fluids. Lab. Invest. 80, 443–454 (2000).

2. Jung, J.-J. et al. Secretion of Soluble Vascular Endothelial Growth Factor Receptor 1 (sVEGFR1/sFlt1) Requires Arf1, Arf6, and Rab11 GTPases. PLoS ONE 7, e44572 (2012).

3. Kinghorn, K. et al. A defined clathrin-mediated trafficking pathway regulates sFLT1/VEGFR1 secretion from endothelial cells. Preprint at bioRxiv doi:10.1101/2023.01.27.525517 (2023).

4. Hakkert, B. C., Rentenaar, J. M. & Van Mourik, J. A. Monocytes enhance endothelial von Willebrand factor release and prostacyclin production with different kinetics and dependency on intercellular contact between these two cell types. Br. J. Haematol. 80, 495–503 (1992).

5. Wu, F. T. H., Stefanini, M. O., Mac Gabhann, F. & Popel, A. S. A Compartment Model of VEGF Distribution in Humans in the Presence of Soluble VEGF Receptor-1 Acting as a Ligand Trap. PLoS ONE 4, e5108 (2009).

6. Beck, M. et al. The quantitative proteome of a human cell line. Mol. Syst. Biol. 7, 549 (2011).