A novel trap for dynamic protein interactions in cytokine signaling

Abstract

As we move towards integrating the elements of the ‘part lists’ that genomics has delivered, the study of protein interactions becomes a central task. Several tools have already been applied successfully to ‘fish’ for protein interactions in the sea of tens of thousands of proteins in the cell. Such techniques include the yeast-two hybrid system and its mammalian equivalents, and more recently, methods based on protein fragment complementation. Current techniques assume that there is a static interaction between a stable complex of ‘bait’ and ‘prey’ peptides. However, many protein interactions, especially in signal transduction, are transient. They act like a ‘one-touch button’ on electronic devices: after the button is briefly pressed and released, the effect persists (e.g. the device is on). Continuous pressing of the button–that is, binding and activation of the protein–is often not necessary and might not even be physiological. This dynamic aspect is lost in two-hybrid methods that rely on fixed, constitutive protein complexes.Now, Eyckerman et al. [1xDesign and application of a cytokine-receptor-based interaction trap. Eyckerman, S. et al. Nat. Cell Biol. 2001; 3: 1114–1119Crossref | PubMed | Scopus (122)See all References][1] have devised a new two-hybrid technique that can be used to capture the inducible and transient interactions that act as switches in signaling cascades. They have developed a mammalian protein–protein interaction trap (MAPPIT) based on complementation of the STAT-dependent signal transduction pathway. STATs are proteins that bind to ligand-activated cytokine receptor complexes. Thus activated, they move to the nucleus to act as transcription factor. Mutating all three conserved tyrosine residues in the leptin receptor cytosolic domain eliminates STAT binding and phosphorylation. Bait polypeptides are fused to the C-terminal cytosolic domain of the leptin receptor, and prey polypeptides are fused to a fragment of gp130, an IL-6 receptor subunit that contains four STAT-binding sites. Thus, binding of the ligand to the extracellular receptor domain induces the activation of intracellular receptor-associated kinases, but STAT recruitment and activation cannot proceed unless complementation of the bait–prey interaction restores the STAT-activation capability of the mutated receptor. Complementation is assayed by a luciferase reporter under the control of a STAT-activated promoter. Chimeric receptors with different extracellular receptor domains fused to the mutated leptin receptor cytosolic domain can then be used to examine ligand-specific interactions.In this study, the authors used MAPPIT to screen a retroviral cDNA library for proteins that interact with a chimeric erythropoeitin receptor (EpoR). They identified a new interaction between the erythropoietin receptor and the ‘suppressor of cytokine-signaling’ protein SOCS-2. Furthermore, the use of mutant bait constructs allowed them to identify essential phosphorylation sites on the bait. For example, the authors were able to determine that Tyr402 in EpoR is required for interaction with SOCS-2 because no activation was obtained with Tyr402-mutant EpoR.The MAPPIT strategy is well suited for the study of signal transduction cascades. A major advantage is that the readout exploits the nuclear translocation capability of activated STATs, rather than relying on nuclear shuttling of bait and prey protein. Therefore, interactions are not forced into the artificial context of nuclear transactivation but are examined in the natural context of post-receptor signal transduction events. The MAPPIT method provides a new tool for the analysis of physiologically relevant, inducible protein interactions and can provide a new insight into the dynamic aspects of cytokine signaling.