|
The
Giovanni Armenise-Harvard Foundation Session 1: Signal Transduction, Oncogenes, Development Overview These presentations illustrate how signal transduction has snowballed over the past decade from a relatively narrow field into a broad discipline that touches nearly every facet of cell biology and medicine. In earlier times, a typical signal transduction experiment examined the effects of growth factors on cells in culture, Dr. Stephen C. Harrison said when he introduced this session. "Then, groups of people suddenly realized they were working on the same thing." Cancer cell biologists who originally explored the relay of signals within cells were joined by developmental biologists who studied the regulation of form and pattern and the differentiation of cells into structures such as skin, hair, or heart. More recently, clinical investigators jumped on the signal transduction bandwagon as they realized that this field might unlock an array of human diseases. The four papers in this session represented a wide spectrum: a classic look at oncogenic conversion, followed by two studies of cytoskeleton proteins, and finally an examination of signaling's role in the architecture of the heart. Each offered a detailed analysis of the function and structure of some of the players in some of the pathways that help determine larger phenomena. A common theme of these presentations, in Dr. Harrison's view, was that "signal transduction is taking over studies of the cytoskeleton." Without the microtubules, actin filaments, and other proteins of the cytoskeleton, eukaryotic cells would be shapeless and immobile. Cells require a whole library of elaborate computer codes to go about their business, and Dr. Harrison described signal transduction events as "microscopic subroutines within those codes."
Hepatocyte growth factor (HGF) is one of several scatter factors that play a pivotal role in normal embryonic development. These scatter factors stimulate branched morphogenesis, a process essential for formation of neural, epithelial, and some mesodermal-derived tissues, including muscles and bones. HGF's receptor is Met, encoded by a member of the MET/RON/SEA oncogene family. These receptors are connected to Ras, a signaling pathway that is an established contributor to malignancy and metastasis. Dr. Michieli and his colleagues have been studying scatter factors and their receptors for many years, and recently they have focused on mutations in the tyrosine kinase domain of Met. Such mutations have been identified in papillary renal carcinoma, a human kidney tumor, and the researchers wanted to pin down the mutation's contribution to disease. To do this, they analyzed the biochemical and biological properties of numerous Met mutants, and observed whether such mutations were sufficient to turn ordinary mouse fibroblasts into cancerous cells. Most of the mutants stepped up catalytic activity in the cells, and those with the greatest transforming potential had the highest kinase activity and the strongest link to signal transducers. The best transformers hyperactivated the Ras signaling pathway, while the less aggressive ones protected against apoptosis. But there was a catch. In epithelial cells, which don't make HGF on their own, even the mutations that raised catalytic activity to the highest levels could not turn cells malignant unless recombinant HGF was added to the mix. In mouse fibroblasts, which produce HGF, numerous mutants could turn cells malignant. Transformation was easily blocked, however, by adding HGF antagonists or by using site-directed mutagenesis to keep the Met receptor from binding HGF. These data suggest that although Met mutations may have the capacity to cause cancerous changes in cells, this won't happen in the absence of HGF. It is as though they know how to dance, but will only perform when they hear music. In vivo, the kidney and liver have abundant HGF, which makes it highly likely that the Met mutations found in human kidney tumors are indeed pathogenic.
Growth factors set off a series of chain reactions in cells, and while one regulates cell cycle a second may be molding the shape of the cell itself. Dr. Scita's laboratory has examined the cross-talk between the Ras pathway and Rac, a small guanine nucleotide (GTP)-binding protein that is a crucial organizer of the actin cytoskeleton. Rac has been identified as a key downstream target in Ras signaling, and the researchers performed a series of experiments aimed at identifying intermediaries that carry signals between the two. One player is a substrate of receptor tyrosine kinases called Eps8, which binds to a protein designated E3b1/Abi-1. Dr. Scita's team recently showed that Eps8 and E3b1/Abi-1 participate in the transduction of signals from Ras to Rac by regulating Rac-specific guanine nucleotide exchange (GEF) activities. The plot thickened when the researchers realized that in vivo, Eps8 and E3b1 form a tricomplex with a GEF protein called Sos-1. When all three work together, they enable Rac to organize actin filaments into "ruffles" of the cell membrane. But if either Sos-1 or E3b1 is blocked, the ruffles don't form. Further experiments indicated that although Sos-1 acts on Rac when it is part of this tricomplex, it functions quite differently if hooked to a different partner. When the Sos-1 protein is coupled with a Grb2, an adaptor protein, it is recruited to the plasma membrane where it activates Ras by catalyzing the exchange of guanosine diphosphate for guanosine triphosphate. Although Sos-1 is a versatile player that can function either upstream or downstream of Ras, it cannot play both roles at once. In vitro, it is clear that Grb2 and E3b1 compete for binding to Sos-1. In vivo, E3b1 overexpression kept Grb2 from associating with Sos-1 and favored the formation of the Eps8-E3b1-Sos-1 tricomplex. This complex has Rac-specific GEF. Additional experiments provided further evidence that Sos-1's specificity as a GEF depends entirely on how it is complexed: a receptor-Grb2-Sos-1 complex results in Ras activation, whereas an Eps8-E3b1-Sos-1 complex regulates Rac activation.
Several years ago, Dr. Eck's laboratory was the first to determine the crystal structure of the protein encoded by Src, the first human oncogene to be discovered and the flagship of the tyrosine kinase class of cell-surface receptors. Src is like a switch, made of several different components, that can be flipped by a variety of stimuli. One part of the switch is the Src homology-3 (SH3) domain, a structure that Dr. Eck's team recently found not in cancer, where one might expect to see it, but in a hereditary disease of muscle. Dystrophin, the protein that is defective in Duchenne and Becker muscular dystrophies, serves as a scaffold for signaling molecules and forms a structural link between the actin cytoskeleton and the extracellular matrix. Dystrophin is linked to the cell membrane through a protein called beta-dystroglycan. The C-terminal region of dystrophin binds the cytoplasmic tail of beta-dystroglycan, in part through the interaction of a WW domain on dystrophin with a proline motif (PPxY) in the tail of beta-dystroglycan. This WW domain is homologous with SH3; in this setting, it is stabilized by an adjacent helical region that contains EF hand-like domains. The crystal structure of the dystrophin and beta-dystroglycan complex shows that beta-dystroglycan peptide binds a composite surface formed by the WW domain and one EF-hand. Embedded in a larger binding molecule, the WW-domain recognizes the PPxY motif much as SH3 would do. In a separate series of experiments, Dr. Eck's team found another Src-like structure, called SH2, in a negative regulator of signal transduction called CBL. This protein down-regulates tyrosine kinase receptors by marking them for proteolysis. The N-terminal of CBL contains an SH2 domain, again combined with an EF-hand structure, that grabs phosphorylated tyrosine kinases. Like Legos, standard modules crop up in different settings and their functions are at least partly determined by context. These observations show how efficient nature is at reusing the same hardware for different purposes, Dr. Eck said.
One of medicine's holy grails is to be able to repair damaged heart muscle. Dr. Mercola's research is predicated on the idea that if researchers knew exactly how cardiomyocytes develop in the embryo, it might eventually be possible to recreate myocardial tissue for therapeutic use. In order to actually form a heart, prospective heart tissue must receive signals from two adjacent tissues: endoderm that will form the floor of the pharynx and dorsal midline mesoderm that will form the notochord and head mesoderm. Once induced, the heart field is then subdivided into distinct myocardial and non-myocardial compartments, in part by interactions with neurogenic tissue. Recent experiments in his laboratory have focused on two systems that mediate these processes; one involves a growth factor and the other a receptor-ligand pair. Wnt is a diffusable growth factor, and Dr. Mercola's team has found that the dorsal midline mesoderm secretes several Wnt antagonists, such as Dkk1 and Frzb, that induce genes needed to begin turning non-cardiogenic mesoderm into myocardium. Although additional signals from the endoderm are required for progression to a heart tube, he believes that the location and extent of cardiogenic mesoderm in the embryo depends on the distribution of these endogenous Wnt antagonists. Tissue where Wnt is unopposed will not become part of the heart. Genes encoding the transmembrane receptor Notch1 and its ligand Serrate1 are expressed in a pattern that strongly suggests they subdivide the heart field into myocardium and non-muscular components such as valves. When the Notch pathway was activated through the downstream transcription factor Su(H), myocardial gene expression was inhibited and non-myocardial genetic markers increased. When Notch and Su(H) function was blocked, mesoderm differentiated into cardiomyocytes. Moreover, lineage analysis showed that cells where Notch signaling was activated did not contribute to myocardial tissue. Clearly, cells will only choose to become myocytes in the absence of Notch signaling.
|
Molecular
mechanisms underlying oncogenic conversion of scatter factor receptor
