What are they? Fibroblast growth factors (FGFs) are a family of small (20–30kDa) intercellular signalling proteins which are frequently found to be glycosylated. The first draft of the human genome identified 29 potential FGF genes, and so far transcripts have been identified for 23 of them. Two FGFs have been identified in Caenorhabditis elegans (Let-756 and Egl-17) while Drosophila appears only to have a single FGF gene (branchless). The identification of six FGF genes in the sea squirt, Ciona, a basal chordate, suggests that the process of gene duplication and divergence was well underway early during chordate evolution.How do they signal across the cell membrane? Signal transduction occurs via a family of four transmembrane receptor tyrosine kinases; presentation of FGF to the receptor by a sulphated glycosaminoglycan (GAG) co-factor is obligatory for receptor activation. As for other receptor tyrosine kinases, activation of FGF receptors appears to be achieved by dimerisation, although the precise receptor:ligand:co-factor stoichiometry remains controversial.What signalling cascades do they activate? A lot. Currently, it seems that activation of the Ras/MAP kinase pathway underlies many cellular responses to FGF. This is achieved via interaction of FRS2 adaptor proteins with the receptor, and direct and indirect recruitment and activation of up to six Grb2–Sos complexes. Further Grb2–Sos complexes are recruited to FGF-receptor-bound Shp-2. Grb2 also recruits Gab1, thereby initiating the PI 3-kinase pathway. The inositol lipid/calcium pathway is also activated following interaction of phospholipase C-γ with the FGF receptor and there is evidence for FGF receptor interaction with downstream components Crk, Shc, Shb, Src and p85.Is there feedback regulation? Receptor activity appears to be antagonised by transmembrane proteins, tyrosine phosphatases and Sef proteins, while the Ras/MAP kinase pathway is regulated at multiple levels. MAP kinase phosphorylates FRS2, preventing Grb2 binding; it also induces MAP kinase phosphatase, a MAP kinase antagonist, and the Sprouty proteins that inhibit Grb2–FRS2 interactions. Sprouty proteins can also bind the ubiquitin ligase Cbl, and through this interaction they may regulate FRS2–FGF receptor ubiquitination and degradation.This all sounds very complex. Yes, and that's not all. The RNAs for FGF receptors 1–3 are subject to alternative splicing which alters their specificity for the different ligands, and it is also clear that ligand–receptor specificities can be modulated according to which GAGs present the ligands. Alternative splicing also can generate secreted FGF receptor extracellular domains, which likely function as competitive inhibitors. Indeed, a recently identified fifth human FGF receptor lacks all cytoplasmic and transmembrane sequences.View Large Image | View Hi-Res Image | Download PowerPoint SlideI hope that the biology of the ligands is simpler. I'm afraid not. The RNAs for many of these also undergo alternative splicing that sometimes affects receptor specificity, and antisense transcripts may regulate their expression. Also, while most are constitutively secreted, secretion of some FGFs, including the prototypic members FGF1 and FGF2, appears to occur through regulated pathways. Several FGFs have also been found to enter the cell nucleus, either of cells that synthesise them or following interaction with surface FGFRs. Indeed, nuclear complexes including active FGF receptor kinase domains have been reported but nuclear functions remain unclear.Okay, okay, but what do they do? They have been implicated in the regulation of virtually all key cellular responses: proliferation, apoptosis, differentiation, regional identity, metabolic control, movement and chemotaxis. Clinically, they are potent oncogenic and angiogenic agents, while mutations in FGF receptors underlie at least 11 syndromes. Developmentally, FGFs are required for many processes including mesoderm and neural induction, limb outgrowth, branching morphogenesis in the lung, regional patterning of the developing brain, induction of the otic placode and development of cranial cartilage.So do they make fibroblasts grow and divide? Although FGFs are mitogens for many transformed fibroblast cell lines, such as the ubiquitous NIH 3T3 cell, many primary fibroblasts are unresponsive to FGFs. But then what's in a name?