Structure-function relationships of growth factors and their receptors

The primary amino acid sequences of several receptor tyrosine kinases have recently made it possible to deduce similarities in the molecular organization of these large multidomain proteins. This has allowed a classification of these receptors into three groups (see Waterfield this Issue and for review in Ref.1). Class I includes the EGF receptor and the neu proto-oncogene, Class II includes the insulin and insulin-like growth factor 1 (IGF-1) receptors, and Class III the platelet derived growth factor (PDGF) and the colony stimulating factor 1 (CSF-1) receptors. The conformation of the ligands for the Classes I and II receptors have been defined using X-ray diffraction, 2-D nuclear magnetic resonance (NMR) and knowledge based modelling procedures. It seems that the ligands are more diverse in sequence than the receptor tyrosine kinases so they cannot be classified as rigorously. However, certain features are common to all growth factors (so far defined) which form compact, globular structures and this allows a discussion of possible interactions between the ligand and receptor; but in the absence of a molecular structure for any of the receptors, we can only review biochemical evidence and deductions from predictive and modelling studies. Various models for the signal transduction process are discussed in the light of current work on receptor interactions.