| The overarching goal of this dissertation is to understand how transmembrane receptors are activated by growth factors that play pivotal roles in cell-cell communications, growth, migration, and development. The first part concerns the study of Leucine-rich repeat-containing G-protein-coupled receptor members 4-6 (LGR 4-6), which are novel GPCRs essential in the regulation of Wnt signaling and maintenance of stem cell compartments in various organ systems. The stem cell growth factors R-Spondins (RSPOs) recruit LGR 4-6; together, the two proteins are thought to boost Wnt signaling by sequestering the Wnt negative regulator ZNRF3/RNF43 of the E3 ubiquitin ligase family. To date, the structures and functions of RSPO, LGR 4-6, and ZNRF3/RNF43 have not been characterized. The dissertation presents the first co-crystal structure of human RSPO1 in complex with the extracellular portion of human LGR5 and of RNF43. The elucidation of the RSPO1-LGR5-RNF43 ternary complex structure, together with biochemical and biophysical data, contributes to the field of Wnt signaling by providing insights into how the three components work together to potentiate Wnt signaling.;The second part seeks to understand how ligand-receptor recognition in the extracellular domain is conveyed to the intracellular side to affect receptor cytoplasmic domain activation. For this purpose, full-length, detergent-solubilized transmembrane PDGFRbeta receptor tyrosine kinase has been purified from mammalian cells and visualized under negative-stain electron microscopy. Three-dimensional reconstruction of a full-length PDGFRbeta receptor tyrosine kinase in complex with its physiological ligand PDGF-B has been achieved at ~27-A resolution. This initial low-resolution map extends our understanding towards receptor domains that have defied crystallization, namely, the membrane-proximal domains, the transmembrane domain, and the intracellular tyrosine kinase domains. Importantly, except the D1 Ig domain, all receptor domains are involved in receptor dimerization that is driven by ligand stimulation; in particular, we confirm the presence of transmembrane helix dimerization in a ligand-bound receptor. Furthermore, flexible fitting of atomic models of membrane-proximal D4-D5 domains leads to identification of a potential D5-D5 interface vulnerable for oncogenic mutations. Finally, an asymmetric arrangement observed for the tyrosine kinase domains suggest a mechanism for how different tyrosine residues in the kinase domains are trans-phosphorylated. |
