Ahmadi, Shiva: Application of BioID to in vitro organelle and in vivo cell-type-specific proteomics. - Bonn, 2020. - Dissertation, Rheinische Friedrich-Wilhelms-Universität Bonn.
Online-Ausgabe in bonndoc: https://nbn-resolving.org/urn:nbn:de:hbz:5-59196
@phdthesis{handle:20.500.11811/8598,
urn: https://nbn-resolving.org/urn:nbn:de:hbz:5-59196,
author = {{Shiva Ahmadi}},
title = {Application of BioID to in vitro organelle and in vivo cell-type-specific proteomics},
school = {Rheinische Friedrich-Wilhelms-Universität Bonn},
year = 2020,
month = sep,

note = {A genetically modified biotin ligase BirA (Arg118Gly), termed BioID (BirA*), can be induced to add a reactive form of biotin to proximate proteins in the cytosol and nucleus. The biotinylated proteins can be purified and identified by mass spectrometry (MS). In this project, we combined the BioID proximity biotinylation approach with MS-based quantitative proteomics for mapping organelle specific (in vitro) and cell-type-specific (in vivo) proteomes.
Using a stable isotope labeling by/with amino acids (SILAC) labeled Human embryonic kidney (HEK 293T) cell line transfected with BioID, we evaluated the effect of biotinylated proteins on the binding pattern of background/non-specific proteins to streptavidin beads. By combining biotinylated and two control samples in different ways and evaluating protein intensities we realized that non-specific proteins have a low chance for binding the beads when only one control sample exists and therefore their instensities are below MS detection limits. However as soon as a second control sample is added, the chance of non-specific proteins in binding the beads and being identified by MS increases. To distinguish between biotinylated and non-specifically binding proteins, using the two control samples we defined a statistical cut-off including 95% of the proteins in control samples. This cut-off was found to be almost identical with the cut-off determined by “slide approach” based on Gene Ontology (GO) analysis (Branon et al., 2018); this proved the reliability of our cut-off determination method. Organelle proteomics was performed using double control strategy (in a 3plex SILAC approach) and data validated by comparison to the literature. We found 97% of the Cytosol/nucleus proteins, 77% and 53% of nuclear and mitochondrial proteins previously linked to these organelles.
After successful implementation of BioID technique in cell culture and enrichment for organelle proteomes, we expanded the application of the technique to in vivo cell-type-specific analysis. It is reported that certain diseases only affect a subpopulation of a specific cell type in a tissue. The ability to examine cell-type-specific protein expression is critical for investigating the molecular pathology of such diseases. To this aim, under the control of Col1a1 and Rosa26 loci, two BioID mouse lines were generated. Universal expression of BioID was achieved by crossing these mice with lines constitutively expressing rtTA and Cre recombinase in all tissues. From these animals, twelve tissues were analyzed by Western blot confirming the expression of, and biotinylation by, BioID. Selected tissues were investigated by immunofluorescence and mass spectrometry. Using a dimethyl labeling-based strategy, on-bead digestion protocol was optimized improving the identification and enrichment rates by ~5-folds. The best method was applied on 3 tissue samples to enrich for biotinylated proteins. Biotinylation degrees were calculated as 46%, 87%, and 92% for lung, pancreas, and liver, respectively. Using Ins1Cre mouse strain, BioID was expressed exclusively in pancreatic beta cells, on-bead digestion performed and biotinylated and control samples labeled with different tandem mass tags (TMT). An in vivo beta-cell proteome was generated including a total number of 4133 proteins, 2992 significantly enriched (biotinylated); 96% of enriched proteins were previously identified by large scale islet or beta-cell omics studies. The dataset was depleted for markers of other pancreatic cell types. Using TMT strategy, the effect of pro- and anti-diabeticc drugs Streptozotocin (STZ) and Harmine, on the proteome of beta cells was evaluated. To validate the data, the regulated proteins were uploaded in GO platform and searched for enriched Reactome pathways and compared to previously published literature on mechanisms involved by Harmine and STZ. Harmine, known as an antidiabetic and anticancer drug, resulted in downregulation of 2518 proteins and upregulation of 97 proteins; more than 60 pathways enriched in this dataset were found previously connected to the effects of this drug. STZ, a diabetes-inducing drug, resulted in downregulation of 1274 proteins and upregulation of 6 proteins. Out of the enriched pathways, 86 were found previously linked to diabetes and STZ.},

url = {https://hdl.handle.net/20.500.11811/8598}
}

The following license files are associated with this item:

InCopyright