EnglishCookie löschen - von nun an wird die Spracheinstellung Ihres Browsers verwendet.
D'souza, Rochelle (2013): Proteomics and phosphoproteomics applied to cell signaling and cancer. Dissertation, LMU München: Fakultät für Chemie und Pharmazie



Signaling networks control and regulate outcomes in cells and organisms in both normal physiology and pathophysiological states. Signaling is traditionally represented and studied as a series of stepwise enzymatic events constituting a cascade. However, it is increasingly apparent that such representations limit understanding of signal transduction since these linear cascades function in an interconnected network that includes extensive cross talk among receptors and pathways. Mass spectrometry (MS)-based proteomics is a useful tool that allows a system-wide investigation of signaling events at the levels of post-translational modifications (PTMs), protein-protein interactions and changes in protein expression on a large scale. This technology now allows accurate quantification of thousands of proteins and their modifications in response to any perturbation. This thesis work is dedicated to the optimization and employment of quantitative mass spectrometry to cellular signaling and an application to segregate two lymphoma subtypes at the levels of protein expression and phosphorylation, employing state of the art liquid chromatography (LC)-MS/MS technologies coupled with improved sample preparation techniques and data analysis algorithms. In the first project I investigated the feasibility of a new, high accuracy fragmentation method called higher energy collisional dissociation (HCD) for the analysis of phospho-peptides. Using this method we were able to measure the phospho-proteome of a single cell line in 24h of measurement time which was a great improvement to previous capabilities. This fragmentation method that was originally thought to be slower and less sensitive than the standard method of low resolution collision induced dissociation (CID) fragmentation. However, our work proves this not to be the case and we showed that HCD outperformed the existing low resolution strategy [1]. In the second project I employed this HCD fragmentation technique on the LTQ-Orbitrap Velos for addressing the clinical question of segregating two subtypes of diffuse B-cell lymphoma (DLBCL). These subtypes are histologically indistinguishable but had been segregated on the basis of a gene expression signature. I employed the recently developed ‘super-SILAC’ approach with a ‘super-SILAC mix’ of multiple labeled cell lines. This heavy reference mix was spiked into several cell lines derived from the two DLBCL subtypes and analyzed LC-MS, resulting in successful segregation based on a distinct proteomic signature [2]. The third project deals with the in-depth analysis of the phospho-proteome of a human cancer cell line on a quadrupole-Orbitrap mass spectrometer using a label-free quantification approach. Our analysis uncovered about 50,000 distinct phosphorylated peptides in a single cell type across a number of cellular conditions allowing assessment of global properties of this large dataset. Strikingly, we found that at least three-quarters of the proteome can be phosphorylated which is much higher than current estimates. We also analyzed phosphotyrosine events using enrichment with anti-phospho-tyrosine antibodies to identify more than 1,500 site specific phosphorylation events. Unexpectedly tyrosine phosphorylated proteins were enriched among higher abundance proteins. The observed difference in phospho-protein abundance correlated with the substrate Km values of tyrosine kinases. For the first time we calculated site specific occupancies using label- free quantification and observed widespread full phosphorylation site occupancy during mitosis. In the final and main project, I applied proteomics and phospho-proteomics to the study of signal transduction in response to transforming growth factor-beta (TGF-β), a multifunctional cytokine. TGF-β signaling regulates many biological outcomes including cell growth, differentiation, morphogenesis, tissue homeostasis and regeneration. The cellular responses to this multifunctional ligand are diverse and can even be opposed to each other, depending on the cell type and the conditions. To shed light on the reasons for the different outcomes, we analyzed the early phospho-proteome and ensuing proteome alterations in response to TGF-β treatment in a keratinocyte cell line. The early SILAC based phospho-proteome analysis uncovered over 20,000 phosphorylation events across five time points (0 to 20 min) of TGF-β treatment. Building on our recent advances in instrumentation, sample preparation, and data analysis algorithms we measured a deep TGF-β responsive proteome at six late time points (6h to 48h) with corresponding controls in only eight days of measurement time. Our label-free approach identified about 8,000 proteins and quantified more than 6,000 of them. This deep proteome covered well established pathways involved in TGF-β signaling, allowing global evaluation at the level of individual pathway members. Combining the TGF-β responsive proteome with an in-silico upstream regulator analysis, we correctly retrieved several known and predicted novel transcription factors driving TGF-β induced cytostasis, de-differentiation and epithelial to mesenchymal transition (EMT). The combined analysis of transcription factor regulation with early phosphorylation changes and proteome changes enabled visualization of the intricate interplay of key transcription factors, kinases and various pathways driving cytostatis, EMT and other processes induced by TGF-β. In summary, my thesis developed a highly efficient phospho-proteomic workflow, which was applied to the measurement of a very deep phospho-proteome of a single cancer cell line allowing analysis of its global features. The main achievement was the first in-depth and combined study of the phospho-proteome and resulting proteome changes following a defined signaling event, in this case leading to a time-resolved view of TGF- β signaling events relevant in cancer.