Engagement of multicomponent immunoreceptors such as the T cell antigen receptor results in rapid recruitment and activation of multiple protein tyrosine kinases (PTKs) including Lck, Fyn, ZAP-70 and Itk. These PTKs then phosphorylate a number of enzymes and adapter molecules involved in complex signaling cascades. Our studies have focused on a critical substrate of the PTKs, LAT (linker for activation of T cells), a 36-38kD integral membrane protein. LAT is a critical transmembrane adapter protein. We have performed studies to characterize how LAT is phosphorylated and binds a number of critical signaling molecules, thus bringing other adapter molecules and enzymes in multimolecular complexes to the plasma membrane in the vicinity of the activated TCR. Biochemical, biophysical, microscopic and genetic techniques are currently employed to study the characteristics of LAT-based signaling complexes and the enzyme pathways that are coupled to and activated at LAT complexes. In previous years we have published a number of studies focused on protein-protein interactions and the structure of signaling complexes. In one study, published in 2006 (Houtman et. al. Nature Struc. &Molec. Biol.) we proposed that LAT-based complexes can oligomerize and thus acquire a higher order structure. In that study we addressed the binding of LAT to the adapter molecule Grb2 and, in turn, Grb2 binding to the Ras activator protein, SOS. We demonstrated that after tyrosine phosphorylation, LAT could bind as many as three Grb2 adapter molecules via the Grb2 SH2 domain. We also showed that Grb2, via its SH3 domains, could bind two sites on SOS. We hypothesized that these binding events might mediate oligomerization between the LAT, Grb2 and SOS molecules. We showed by analytical ultracentrifugation that these molecules formed complexes with a size consistent with oligomerization. Additional studies in which we expressed versions of these molecules in a T cell line supported the idea that oligomerization could occur in cells and could have functional importance. Two published studies this year returned to this theme. To test the significance of oligomerization of LAT, Grb2 and SOS in vivo, we started with mice, which we had generated, whose T cells lack SOS. Thymocytes from these mice tested by activation of the TCR showed a decrease in Erk activation, which might have been expected in SOS-deficient mice. Remarkably calcium flux was also depressed. This finding would be unlikely due to a defect in SOS enzymatic activity. We reconstituted SOS expression by crossing the SOS-deficient mice to mice in which wild-type or mutant SOS molecules were expressed from transgenes. The mutant SOS molecules lacked either SOS enzymatic activity, lacked the potential to oligomerize with Grb2 and LAT, or lacked both functions. We found in the offspring that reconstitution with SOS molecules with intact enzymatic function was required for restoration of Erk activity. However calcium flux responses required SOS molecules that could oligomerize. A far more complex biological response, thymocyte proliferation, was also tested. SOS-deficient mice show a marked deficiency in thymocyte expansion. Generation of normal numbers of thymocytes was inhibited either by SOS molecules deficient in enzyme activity or oligmerization potential, that is both of these functions were required. In a final interesting experiment we showed that these two functions could be reconstituted in trans by adding back separate SOS molecules, one with the enzymatic defect and the other with the oligomerization defect. Our studies in mice confirmed and expanded previous work from the lab to show that the higher order structure of signaling complexes was necessary for optimal signaling and biological function. Presumably oligomerization is required to bring multiple molecules together to enhance signaling. These results are probably not unique to T cells as several growth factor receptorssystems are likely to also utilize this oligomerization mechanism. Additional support for this model was obtained from our study of two other molecules that can be found in LAT-based complexes. The adapter molecule SLP-76 is also recruited to LAT via a Grb2-like molecule, Gads. SLP-76 itself has multiple binding partners and one of these, the adapter molecule ADAP, can bind SLP-76 via the SLP-76 SH2 domain. Multiple phosphorylated tyrosine residues on ADAP were proposed to be the sites of interaction with the SLP-76 SH2 domain. We confirmed that three of these phosphorylated ADAP tyrosines were capable of high affinity binding to the SH2 domain. That observation again raised the potential that oligomerization of complexes mediated by multi-point binding could occur. We again used purified proteins, this time phosphorylated ADAP polypeptides and SLP-76, to show that oligomerization of this complex can take place in vitro. Additionally we showed that optimal TCR-mediated signaling required all three phosphorylatable ADAP tyrosines. Our conclusion from this and the other study, reviewed above, is that higher order formation of signaling complexes occurs and is required for optimal signaling. We assume that these complexes are highly dynamic and may be only transient. The means by which these complexes are regulated in vivo will be important to determine. Moreover targeting these structures might prove therapeutically useful in situations such as cancer in which hyperactive signaling may contribute to disease.
Showing the most recent 10 out of 32 publications
