I. Background

Interactions between biopolymers play critical roles in cellular processes. In particular, protein-protein interactions provide a widespread mechanism by which the flow of inter- and intracellular information is governed. Such interactions play a critical role in many biological processes, including signal transduction, gene regulation, viral assembly and replication, and antibody-antigen complexes. It is crucial for cell survival to maintain proper connectivity of protein-interaction interactions and prevent abnormal interactions. I plan to focus my future research on the study of the protein-protein interactions in cellular signaling networks, especially artificial manipulation and systematic disruption for desired signaling behaviors.

Eukaryotic cells are bombarded with a vast array of external signals, but they have the remarkable ability to correctly and specifically process this information to yield precise cellular responses. In many cases this information is transmitted by networks of protein-protein interactions. Common examples are the mitogen-activated protein kinase (MAPK) cascades, modules of three kinases that sequentially activate one another. In mammals there are three MAPK families: the extracellular signal-regulated protein kinases (ERK's), p38 MAPK's and the c-Jun N-terminal kinases (JNK's). These pathways have significant roles in mediating signals triggered by cytokines, growth factors and environmental stress, and are involved in immune responses, cell proliferation, cell differentiation and cell death.1 Despite the presence of multiple kinases, cells maintain exquisite signaling specificity. How then is the proper response generated and the inappropriate responses avoided- This problem is particularly acute in the case of MAPK cascades that share common components. For example in mammalian cells, the MAPK kinase MKK4 can phosphorylate two possible downstream MAPK’s, JNK or p38, each of which lead to a physiologically distinct output. How is the choice between multiple possible outputs made?

An emerging paradigm is that protein-protein interactions between individual signaling proteins play a critical role in achieving signaling specificity. Interesting examples of this are scaffold proteins-proteins with multiple binding sites for proteins in a given pathway. Such scaffold proteins appear to non-covalently tether sets of signaling proteins together so that they can interact with one another in a specific and efficient manner. Scaffolds are postulated to contribute to proper pathway input/output control, as well as spatial and temporal regulation.2 The best characterized role of MAPK scaffolds is in the budding yeast, Saccharomyces cerevisiae. The mating MAPK cascade employs the scaffold proteins, Ste5.3 The Ste5 scaffold protein contains separate binding domains for a series of kinases involved in a cascade3,4 (Figure 1A). The importance of Ste5 in maintaining signaling specificity is emphasized by the fact that although the mating cascade shares a common signaling kinase, Ste11, with another MAPK cascade they do not cross talk under normal condition.5 Putative MAPK scaffold proteins have also been identified in mammalian cells. JNK interacting protein-1 (JIP-1) is thought to function as a scaffold in the JNK MAPK pathway6 and KSR-1 has been implicated as a scaffold in the ERK MAPK pathway7. The cytoplasmic protein JIP-1 binds selectively to the MAPK JNK but not to other related MAPK's, including p38 and ERK. JIP-1 appears to play an analogous role in structural organization of kinases to that of Ste5 scaffold in the mating response MAPK pathway in yeast8 (Figure 1B).

II. Specific Aims

1) Pathway Engineering - to design synthetic scaffolds to artificially change and/or redirect the flow of signaling information in diseased cells.

2) Pathway Disruption - to perform systematic disruption of specific protein-protein interactions in signaling networks that are related in diseases.

III. Rationale & Strategy

1) Pathway Engineering. Signaling specificity can be achieved by the connectivity of a stimulus (A) and a response (A’) that generates a specific input/output relationship (A → A’) (Figure 2A). Wiring a correct set of signaling components between input and response is crucial for signaling specificity. Scaffold proteins have been postulated to provide a platform onto which non-covalent interactions between pathway components take place. Under normal condition, mechanisms of specific signaling prevent a cross talk between two unrelated pathways (A and B). However, it would be possible to rewire the signaling information if one can engineer synthetic scaffolds that recruit an non-natural set of components from two unrelated pathways (A → B’). In my previous study, I have shown that a primitive, chimeric scaffold can trigger a signal diversion between two distinct MAPK pathways in yeast.9

For signal rewiring, I will attempt to design non-natural, synthetic scaffold proteins that can specifically bind to and reinforce the interactions between signaling components from unrelated pathways to create a novel signaling pathway. I propose to apply this strategy of pathway engineering to diseases caused by signaling defects such as cancer and immune disorders. Specifically, I will target the ERK MAPK pathway (proliferation) and the Bad/Bax pathway (apoptosis) in tumor cells. Using the pathway engineering strategy, I will attempt to shunt proliferation stimuli to apoptotic responses for simultaneous promotion of cell death and suppression of proliferation in a cancer cell.

To design a synthetic scaffold, I will utilize well-studied, modular protein interaction domains such as PDZ and SH3 domains.10 These heterologous interaction domains will be subject to a molecular evolution11 for specificity toward individual signaling components from proliferation and apoptotic pathways. The resulting mutant PDZ domains will be fused together via rational or combinatorial fashion to create a series of synthetic scaffolds that can recruit non-natural sets of signaling proteins from the two pathways (Figure 2B). The putative synthetic scaffolds composed of heterologous interaction modules will be expressed in a cancer cell and be tested for the ability to trigger a apoptotic response upon stimulation by growth factors.

2) Pathway Disruption. In mammalian cells, a given input (ligand) usually triggers many different signaling responses in different cell types. The decision as to which response to trigger in a certain cell type is made by a subtle and precise signaling mechanism yet to be understood. Traditionally, strategies to treat diseases caused by abnormal signaling have relied on blocking of ligand binding to corresponding receptors. However, this approach has a drawback of significant side effects due to a system-wide shutdown of the receptor-dependent pathways. For efficient treatment of diseases, therefore, it is desired to develop a strategy to surgically block the disease-related branch(es) of signaling cascades by identifying and disrupting specific protein-protein interactions.

Recent efforts on proteomics analyses have accumulated a vast data set of protein networks inside a cell and begun to elucidate precise information about linkages of protein-protein interactions in various disease states. This information allows to pinpoint the specific protein interactions downstream of receptors that are critical for the onset and progress of diseases of interest. For example, an unusual activation of JNK1 MAPK in JIP-1-dependent manner has been implicated in the oncogenic transformation precursor B-cells caused by the Bcr-Abl oncogene.12 Because of their important role in disease processes including cancer, inflammation and immune disorders, MAPK signaling components are important targets for therapeutic intervention. Especially, scaffold-mediated interactions can be a novel therapeutic target for surgical disruption of a specific subset of MAPK signaling.

I will primarily focus on the mammalian MAPK scaffolds, JIP-1 and KSR-1 from the JNK and the ERK MAPK pathways, respectively. I propose to systematically disrupt the scaffold-kinase interactions in such diseased cells. In my previous study, I developed a novel genetic selection to screen an encoded combinatorial peptide library for dissociative inhibitors of a protein-protein interaction of interest.13 This selection strategy is amenable to ELISA-base assays for a high-throughput screen. I propose to apply this selection strategy to the systematic disruption of the JIP-1 and KSR-1 scaffold interactions with their component kinases. The positive isolates from selection will be tested for their ability to disrupt the target interactions in vitro and to block the target signaling pathways in vivo. The blocking of signaling by the dissociative inhibitors will be evaluated by monitoring its impact on the progress of disease states of interest. This strategy of pathway disruption by specific inhibition of critical protein interactions will be a valuable tool for evaluation of certain interactions for their role in disease development as well as the rapid generation of therapeutic agents for intervention of the disease itself.

IV. Conclusion

Here, I propose the strategies to artificially manipulate cellular signaling network by rewiring and disrupting specific protein-protein interactions, pathway engineering and pathway disruption. Combined together, these approaches will allow us to precisely alter signaling responses, to bypass the point of defect in signaling, and to surgically block a subset of signaling cascades. The ability to artificially manipulate signaling behavior will provide a powerful means for treatment of diseases caused by abnormal signaling and thus, for the development of therapeutic agents.

References

1. Johnson, G. L. & Lapadat, R. Mitogen-activated protein kinase pathways mediated by ERK, JNK, and p38 protein kinases. Science 298, 1911-2 (2002).

2. Burack, W. R. & Shaw, A. S. Signal transduction: hanging on a scaffold. Curr Opin Cell Biol 12, 211-6. (2000).

3. Choi, K. Y., Satterberg, B., Lyons, D. M. & Elion, E. A. Ste5 tethers multiple protein kinases in the MAP kinase cascade required for mating in S. cerevisiae. Cell 78, 499-512 (1994).

4. Inouye, C., Dhillon, N., Durfee, T., Zambryski, P. C. & Thorner, J. Mutational analysis of STE5 in the yeast Saccharomyces cerevisiae: application of a differential interaction trap assay for examining protein-protein interactions. Genetics 147, 479-92. (1997).

5. O'Rourke, S. M. & Herskowitz, I. The Hog1 MAPK prevents cross talk between the HOG and pheromone response MAPK pathways in Saccharomyces cerevisiae. Genes Dev 12, 2874-86. (1998).

6. Whitmarsh, A. J., Cavanagh, J., Tournier, C., Yasuda, J. & Davis, R. J. A mammalian scaffold complex that selectively mediates MAP kinase activation. Science 281, 1671-1674 (1998).

7. Yu, W., Fantl, W. J., Harrowe, G. & Williams, L. T. Regulation of the MAP kinase pathway by mammalian Ksr through direct interaction with MEK and ERK. Curr Biol 8, 56-64 (1998).

8. Whitmarsh, A. J. & Davis, R. J. Structural organization of MAP-kinase signaling modules by scaffold proteins in yeast and mammals. Trends Biochem Sci 23, 481-5 (1998).

9. Park, S. H., Zarrinpar, A. & Lim, W. A. Rewiring MAP kinase pathways using alternative scaffold assembly mechanisms. Science 299, 1061-4 (2003).

10. Pawson, T. Protein modules and signalling networks. Nature 373, 573-580 (1995).

11. Schneider, S. et al. Mutagenesis and selection of PDZ domains that bind new protein targets. Nat Biotechnol 17, 170-5 (1999).

12. Dickens, M. et al. A cytoplasmic inhibitor of the JNK signal transduction pathway. Science 277, 693-696 (1997).

13. Park, S. H. & Raines, R. T. Genetic selection for dissociative inhibitors of designated protein-protein interactions. Nature Biotechnol 18, 847-51 (2000).