Structural and functional characterization of extracellular domains of vascular endothelial growth factor receptor 1 and 2

Angiogenesis is the formation of new blood vessels from pre-existing vasculature and plays an essential role in normal organ development and in specific diseases in all higher organisms. Angiogenesis is therefore required already early in embryogenesis when the new blood and lymphatic systems develop. In adult organisms angiogenesis is required in numerous processes such as in vessel formation and remodeling in the female reproductive cycle, during wound healing, or in bone formation and remodeling. Aberrant excessive vessel formation, i.e. pathological angiogenesis, plays an important role in tumor progression, in diabetic retinopathy, rheumatoid arthritis or in psoriasis. The lack of angiogenesis leads to multiple vascular failure such as coronary artery disease. It is well established that the correct balance between pro- and anti-angiogenic growth factors, cytokines, and extracellular matrix components is essential for vascular homeostasis. One of the critical regulators of both physiological and pathological angiogenesis discovered more than 30 years ago is Vascular Endothelial Growth Factor (VEGF), regulating endothelial cell (EC) proliferation, migration, and survival but also vascular topology and permeability. VEGF is a family of cysteine linked dimeric growth factors consisting of five members, VEGF-A, -B, -C, -D and Placenta Growth Factor (PlGF). These soluble or matrix associated proteins bind to three type V receptor tyrosine kinases (RTKs), VEGF-receptor (VEGFR)-1 (also known as Flt1), VEGFR-2 (KDR/Flk1), and VEGFR-3 (Flt4). VEGFRs consist of an extracellular domain (ECD) built from seven immunoglobulin (Ig)-homology domains required for ligand binding and subsequent receptor dimerization. A single transmembrane (TM) helix connects the ECD to the cytoplasmic part containing a split tyrosine kinase domain. Ligand binding to VEGFR ectodomains promotes dimerization of receptor monomers, followed by receptor autophosphorylation and kinase activation. The activated receptor contains specific tyrosine residues in the kinase domain and the carboxy-terminal (C-terminal) domain acting as docking sites for a plethora of signaling proteins involved in multiple cellular signaling pathways.
Ig-homology domains 1-3 (VEGFR-3) or 2-3 (VEGFR-1 or -2) of the ECD form the ligand binding site, while domains 4-7 are involved in homotypic receptor contacts fulfilling a regulatory function, which was the subject of this thesis. I used isothermal titration calorimetry (ITC) in this study to determine the thermodynamic properties of ligand binding and dimer formation. The data show that the free energy of VEGF-A binding to domains 1-3 or the full-length ECD of VEGFR-2 is entropy driven and enthalpically unfavourable. Most importantly, the Gibbs free energy of VEGF-A binding to the full length ECD is 1.12 kcal/mol higher compared to the binding energy of domains 1-3. The endothermic component arising from the homotypic receptor contacts in domains 4-7 thus reduces the overall binding affinity of the full-length VEGFR-2 ECD by about 10 fold. This suggests that the homotypic interactions in domain 4-7 play a regulatory role in ligand binding and receptor activation, e.g. by promoting conformational rearrangements of receptor monomers required for active dimer formation. This mechanism might also prevent spontaneous activation of VEGFR-2 in the absence of ligand.
I also tried to crystallize the ECD of VEGFR-2 in complex with ligand. However, although I used a multitude of receptor ECD constructs, I did not obtain diffracting crystals. I therefore became involved in an accompanying project in the lab focusing on the crystal structure of the full-length VEGFR-1 ECD in complex with VEGF-A. This structure revealed distinct homotypic contacts in Ig-homology domains 5 and 7. To further characterize the contacts in domain 5 biochemically and to investigate their functional relevance in receptor activation I generated mutants disrupting specific hydrogen bonds and salt bridges involved in homotypic contact formation. The data showed a significant decrease in receptor phosphorylation activity upon stimulation with ligand. Similarly, I could show reduced receptor activity when the homologous residues were mutated in VEGFR-2. The biochemical characterization of these mutants thus document the regulatory role of domain 5 in VEGFR activation and identify domain 5 as a promising target for developing allosteric inhibitors of VEGFRs. The speciality of drugs proposed to target domain 5 lies in their ability to access the target receptor at a regulatory site in the extracellular receptor domain, which is easily accessible from the blood stream. In addition, the proposed drugs will be highly specific as compared with the currently used kinase inhibitors.