Modeling of interactions between vascular endothelial growth factor family members and their receptors KDR and Flt-1
The vascular endothelial growth factor (VEGF) family is prominently involved in angiogenesis, the formation of new vasculature. Overexpression has been linked to pathological conditions including solid tumor growth. Responses to VEGF are mediated by Flt-1 and KDR, two cell-surface receptors selectively expressed by vascular endothelial cells. A secreted form of Flt-1 (sFlt1) retains high affinity for VEGF and can compete for VEGF binding, thereby inhibiting angiogenic responses. In this study, a computational model of this ligand-receptor binding system is used to quantitatively describe the activation of KDR signaling by VEGF and its inhibition by sFlt-1. The model consists of a system of 12 ordinary differential equations, based on mass action kinetics, which describe, based on mass-action kinetics, the formation and disappearance of receptor-ligand complexes competent to transduce a VEGF signal. Cell-based assays were used to test predictions of the in silico model by measuring dynamics of 1) KDR tyrosine hosphorylation and 2) generation of intracellular second messengers in response to VEGF stimulation in the presence and absence of sFlt-1. Results indicate that the model predicts well the maximal activation in the presence of VEGF-A or the KDR-selective agonist VEGF-E. Subsequent characteristics of binding, such as time-to-peak activations, were less realistically modeled, suggesting that the mathematical descriptions of certain receptor fates (receptor internalization, sequestration or externalization) may need refinement. Also presented are the results of KDR activation experiments in the simultaneous presence of VEGF-A, sFlt1, and the Flt1-selective agonist placental growth factor (PlGF). Here, the model predicts accurately the percentage of inhibition of maximal activation when sFlt-1 is present in the system. Finally, a revised model that includes receptor heterodimerization is presented that suggests sFlt-1 can more effectively inhibit VEGF binding and activation when acting in a dominant negative fashion. Overall, the findings support the idea that the early events in VEGF signaling can be modeled successfully. A natural extension of the model, refined based on experimental testing, will be its application to normal physiological systems.