Computational models of protein-phosphorylation protein interfaces and their applications
| Autor: | Lee, Li-Yu, 李力渝 |
|---|---|
| Rok vydání: | 2019 |
| Druh dokumentu: | 學位論文 ; thesis |
| Popis: | 107 Protein phosphorylation is one of the important post-translational modifications, plays a key role in the phosphorylation of substrate proteins in cellular signaling transduction. Phosphoprotein binding domains (PPBDs) have specific binding affinity to phosphorylated sites in proteins playing a pivotal role in connecting the kinases and the effector molecules, with important roles in human diseases, like cancer. Currently, the approach of kinase inhibitor design lacks selectivity due to non-specific binding to the target protein leading to serious side-effects. Understanding how the phosphoprotein binding domains recognize phosphorylation proteins can help us have a more detailed understanding of signaling transduction pathways. Thus to study phosphoprotein binding domains and their possible mechanisms of common or specific interactions, we propose a systematic computational model of phosphorylation protein-protein interfaces (PPIs) using evolution and mutual information (MI) based on the protein sequence. Mutual information (MI) is useful for quantifying cross-correlations between amino acid substitutions in proteins. For example, Pin1-WW domain will bind phosphoserine to activate Pin1 to regulate the cell cycle in neurodegenerative diseases and cancer. Also, TIFA-FHA domain is also one of the PPBD involving immune responses, playing a role in human diseases, like cancer. From the model, we identify important sites in WW domains like in the Pin1, R14, S16, and F25 sites; FHA domains like in TIFA, K93, Q64 sites (showing high MI). These sites are important interaction mechanisms with phosphopeptide. We evaluated our model for finding inhibitors for the FHA domain involved in DNA damage and cancers. For this, we docked NCI compound library using lab’s virtual screen tool GEMDOCK and screened out 10 potential compounds to test for binding affinity using surface plasmon resonance (SPR) assay and observed 3 compounds to have a high affinity for the FHA domain. This highlights our model can find important protein-compounds interactions and apply for inhibitor discovery. |
| Databáze: | Networked Digital Library of Theses & Dissertations |
| Externí odkaz: |
|