Mass spectrometry is the primary technology of proteomics. For the analysis of complex proteomes, protein identities and quantities are inferred from their peptides that are generated by cleaving all proteins with the endopeptidase trypsin. But there is one major disadvantage that is due to biophysical differences, different peptides cause different intensities. Miscellaneous approaches have been developed to circumvent this problem based on the chemical or metabolic introduction of heavy stable isotopes. This enables to monitor protein abundance differences of two or more samples on the same tryptic peptides that differ in mass only. Absolute quantification can be achieved similar by spiking-in synthetic isotopical labeled counterparts of a sample’s tryptic peptides.
However, labeling technics suffer from high prices, introduced biases, need for extensive manual control, laborious implementation and implementation restrictions. Therefore, a multiplicity of label-free approaches have been developed that profit from instrumental improvements targeting reliability of identifications and reproducibility of quantitative values. No extensive systematic comparison of label-free quantitative parameters has been published so far presumably because of the laborious implementation. An analysis of primary label-free parameters and associated normalization methods is presented here that compares dynamic and linear ranges and accuracies in the estimation of protein amounts. This facilitated the establishment of label-free procedures addressing three fundamental questions in proteomics: what is a sample’s composition, are proteins that share a specific property enriched and what are the differences between two (or more) samples. A new mathematic model is presented that defines and elucidates enrichment.
The procedures were applied first to analyze and compare stem cell plasma membrane proteomes. This is an ambitious model for proteomics because of only small amounts of arduous to analyze, partial hydrophobic proteins in a complex proteomic and chemical background. It is of scientific relevance, as membrane proteins are the cell’s communication interface that enable cell type specific processes and hence can be used to define, isolate and quantify those. The success of cell surface proteome enrichment, the quantitative composition of the proteome and the proteomic difference between stem cells isolated from the dental pulp and cultivated in different media is shown.
Secondly, the procedures were applied to the analysis of transient protein networks that assemble onto proteo-liposomes in a newly designed recruitment assay that fully recapitulates membrane sorting as seen in vivo. All transmembrane proteins need to be trafficked to other organelles’ membranes by vesicular trafficking. Sorting signals within the cytosolic regions of the protein cargos trigger the formation of trafficking complexes around those. The transient membrane complexes additionally recognize organelle or organelle-domain specific membrane lipids, such as phosphatidylinositol phosphates. Different trafficking ways are characterized by different trafficking complexes. The elucidation of trafficking complexes that form around a transmembrane protein of interest discloses its trafficking routes and involved signaling processes. The synthetic proteo-liposomes were prepared from chemically defined lipids and heterologous expressed cytosolic domains of type I or type II membrane receptors.
The proteomic analyses of such samples are challenging because of huge proteomic backgrounds of proteins binding to the liposomes irrespective of the receptor and relatively small amounts and numbers of receptor-specific binders. Though the basic idea is to elucidate sorting machineries and study membrane trafficking processes, such experiments are untargeted and miscellaneous discoveries were achieved. We elucidated that the apical determinant crumbs 2 is a cargo of the retromer complex. This revealed a fameless level of control for the establishment of cell polarity. We found retromer along with the adapter complexes AP 4 and AP 5 trafficking the beta amyloid precursor protein APP. This confirmed recent publications and yielded new insights. Moreover, many more proteins and complexes appeared to associate with the cytosolic part of APP (AICD) in a membrane context-dependent or -independent manner. Among those, some were so far unknown to interact with AICD, like mTORC1 and the PIKFyve complex.
Identifer | oai:union.ndltd.org:DRESDEN/oai:qucosa:de:qucosa:27608 |
Date | 18 February 2014 |
Creators | Niehage, Christian |
Contributors | Hoflack, Bernard, Urlaub, Henning, Technische Universität Dresden |
Source Sets | Hochschulschriftenserver (HSSS) der SLUB Dresden |
Language | English |
Detected Language | English |
Type | doc-type:doctoralThesis, info:eu-repo/semantics/doctoralThesis, doc-type:Text |
Rights | info:eu-repo/semantics/openAccess |
Page generated in 0.0027 seconds