Wendell A. Lim
Protein-Protein Interactions in Signal Transduction
Cells have a remarkable capacity to respond to different environmental conditions and signals. These responses can range from the induction of latent biochemical activities, to the launching of specific patterns of growth, differentiation, or death. Over the last several years, it has become clear that these responses are controlled by biochemical circuits . These circuits are composed of sequentially interacting proteins whose sole function is to process and transmit regulatory information. We are only begining to phenomenologically map out these circuits, but it is already apparent that they are extraordinarily complex. Circuits are not simply linear, rather they have many branch points, allowing for cross-talk between pathways, the detection multiple inputs, and the coordinated and interdependent formulation of appropriate responses. Thus it is increasingly appropriate to compare biological regulatory networks to computational or electrical circuits.
We would like to understand from a biophysical perspective, how such regulatory circuits are constructed and how they function. Given the complexity of these circuits, we have chosen to take the following reductionist approach -- we focus on the isolated interactions between individual protein components . By doing so, we hope to gain a detailed understanding of the following issues:
SPECIFICITY - What determines the correct connectivity of regulatory circuits, given that the components are diffusable proteins? How do two communicating proteins specifically recognize one another and discriminate against the myriad of other proteins encountered in the cellular enviroment?
MECHANISM - How do interactions result in a transmission of the signal?
REGULATION - How are interactions modulated/regulated by multiple signals and inputs?
In the long-term we hope that such studies will lead not only to a deeper understanding of how biological signaling pathways work, but also to the development of ways to block and modulate pathways for therapeutic purposes, as well as to design and engineer protein interactions and protein-based circuits for biotechnological applications.
To answer such questions in a complete way, we utilize an array of complementary experimental approaches: x-ray crystallography to define the structure of interacting proteins; biophysical analysis to quantitate the thermodynamics and kinetics of interactions; combinatorial chemistry and mutagenesis to probe the mechanism of interactions; and biochemical and molecular genetic approaches to test how the interaction functions in the biological context of a signaling pathway. We specifically chose to work on systems which relatively simple, so that we can study them in a more comprehensive and detailed way than is possible with more complex systems. We also try to study systems that are genetically amenable so that we can test function in vivo.
To date we have been focusing our efforts on the Src-homology 3 (SH3) domain, a small, commonly found signaling component which mediates specific interactions between appropriate partner signaling proteins. This domain functions by recognizing certain proline-rich sequences found in target proteins. We have been working with an SH3 domain from the C. elegans protein, Sem-5, a protein which plays an important role in postion dependent cell-fate decisions during development of the organism. Through crystallographic, thermodynamic, and mutagenic analyses, we have made significant progress in elucidating the structural and energetic rules by which this SH3 domain recognizes its specific molecular partners . Most remarkably, we have found that SH3 domains are conformationally reversible -- their recognition surface can be used to bind one subset of peptide ligands in one orientation, and another subset in the opposite orientation. This reversibility may allow SH3 interactions to function as a switch point in signaling. We are continuing to explore how SH3 domains are used as building blocks in signaling networks and how SH3 interactions can be regulated. We are using our knowledge of SH3 interactions to semi-rationally design inhibitors. We have also begun work on several other novel protein-protein interaction systems.