Danish Medical Bulletin    |    Contact Front page  

 
Proteomic Analysis of Protein Kinase A Substrates and Polo-like kinase 1 Interactors using Bioinformatics and Mass Spectrometry

Majbrit Hjerrild  


  Download file   (3424 KB)

Accepted by: Faculty of Health Sciences University of Copenhagen
Defended on: April 25, 2006
Official opponents: Forest M. White , John Mundy , Jørgen Olsen
Tutors: Steen Gammeltoft

Published in the PhD Database: March 21, 2006


English abstract
Protein phosphorylation catalyzed by protein kinases is important for regulation of most biological functions. Up to 50% of all proteins are thought to be modified by phosphorylation. Despite the importance and widespread occurrence of phosphorylation, identification of phosphoproteins and localization of the phosphorylation sites is still a major challenge in proteomics research.
The aim of this Ph.d.-project is to integrate molecular biology, mass spectrometry and bioinformatics in phosphoproteomics. The first challenge in phosphoproteome analysis is identification of new kinase substrates. Computational sequence analysis of proteins can predict candidate substrates for the kinase of interest. In this project, kinase-specific artificial neural networks are presented for 12 protein kinases and the performance of the protein kinase A (PKA)-specific neural network is experimentally validated. The second challenge in phosphoproteome analysis is mapping of phosphorylation sites. Mass spectrometry is often the method of choice for this analysis. We demonstrate that LC-MS/MS analysis is an efficient method for mapping of phosphorylation sites. The last challenge in phosphoproteomics is detailed characterization of the physiological function of the phosphorylation site. In this Ph.d.-project, bioinformatics, mass spectrometry and molecular biology were used to identify and characterize four novel PKA substrates: Necdin, RFX5, En-2 and Wee1. In addition, a mass spectrometry-based analysis of Polo-box Domain interacting proteins was performed.



Danish abstract
Protein fosforylering, katalyseret af protein kinaser, er en vigtig biologisk reguleringsmekanisme. Op til 50% af alle proteiner forventes at være modificeret ved hjælp af fosforylering. På trods af vigtigheden og hyppigheden af fosforylering er identificering af fosfoproteiner og lokalisering af fosforyleringssites stadig en kæmpe udfordring i proteom analyse.
Formålet med dette Ph.d.-projekt er at integrere molekylær biologi, massespektrometri og bioinformatik i fosfoproteom analyse. Den første udfordring i fosfoproteom analyse er identificering af nye kinase substrater. Computer-baseret sekvensanalyse kan forudsige mulige substrater for en given kinase. I dette projekt beskrives udvikling af kinase-specifikke kunstige neurale netværk for 12 protein kinaser og det protein kinase A (PKA)-specifikke neurale netværk blev valideret eksperimentelt. Den anden udfordring i fosfoproteom analyse er identificering af fosforyleringssites. Massespektrometri er ofte den foretrukne metode til mapping af fosforyleringssites. Vi demonstrerer at LC-MS/MS er en meget effektiv metode til identificering af fosforyleringssites. Den sidste udfordring i fosfoproteom analyse er detaljeret karakterisering af den fysiologiske betydning af fosforyleringssitet. I dette Ph.d.-projekt karakteriseres fire nye PKA substrater: Necdin, RFX5, En-2 and Wee1 ved hjælp af bioinformatik, massespektrometri og molekylær biologi. Desuden, identificeres Polo-Box domæne interaktionspartnere ved hjælp af massespektrometri.