"Challenges in Determining the Subcellular Location of Proteins"
Professor Kathryn Lilley, University of Cambridge
Proteins adopt multiple functions controlled by their sub-cellular location, binding partners and post transcriptional and post translational modification status, and structure. Such differential control significantly increases the functionality of the proteome over what is encoded by the genome. The processes governing these features are highly dynamic. The ability to chart changes in the dynamic proteome upon perturbation such as cell stress is of paramount importance to our understanding of cellular mechanisms. Methods designed to map these changes are very reliant on technologies that result in reproducible data and excellent subcellular resolution.
We have applied the hyperLOPIT method (Christoforou Nat. Comm. 2016) to gain insight into the steady state location of proteins in yeast cells and different mammalian cell lines with high subcellular resolution, and reproducibility. We use a combination of different fractionation methods based on detergent solubilisation, differential and equilibrium density centrifugation to fractionate cells into distinct subcellular fractions. We map the distribution of proteins through fractions using quantitative proteomics approaches, and apply bespoke machine learning tools to further analyse data and classify proteins into distinct subcellular niches (Breckels PLoS Comp Biol. 2016 and F1000Res 2017, Gatto Bioinformatics, 2014.).
We show that high quality data can be achieved using different spatial proteomics approaches, but demonstrate that the choice of analytical workflows impacts the number of false discoveries incurred and thus conclusions which can be drawn. Furthermore, comparison with high content data obtained by microscopy methods shows some overlap between spatial assignments, but highlight inherent issues of poor quality immune reagents in the case of immunohistochemistry and mis-localisation artefacts which may result from fluorescent fusion proteins.
We also demonstrate that over half of the proteome is located in multiple places giving insight into spatially dependent functionality of proteins, and link localization to structural elements within the proteome, especially those involving intrinsic disorder. Finally, we show the spatial partitioning of functional units.
"Identification of a Novel Aflotoxin-Amino Acid Adduct and its Potential as a Detoxification Product using High Resolution and Tandem Mass Spectrometry"
Blake Rushing, Laboratory of Professor Mustafa Selim, East Carolina University, Brody School of Medicine, Department of Pharmacology and Toxicology
Aflatoxin B1 (AFB1) is a class 1 carcinogen and a common food contaminant worldwide. It is also a major cause of the development of hepatocellular carcinoma (HCC), and is estimated to play a causative role in up to 28% of all HCC cases. Existing strategies to reduce AFB1 exposure are limited and thus, many people are exposed to uncontrollable levels of this toxin. Our study aims to 1) develop a new chemical treatment process to modify AFB1 into a non-carcinogenic form using benign reagents found in human diets and 2) verify that these new products have reduced genotoxicity. Our treatment strategy targets the mutagenic site of the AFB1 molecule, the 8,9-double bond, by adducting it to selected amino acids.
Identification and quantification of aflatoxins was carried out using high performance liquid chromatography-electrospray ionization-time of flight mass spectrometry (HPLC-ESI-TOFMS). The detoxification strategy involved the hydration of AFB1 to AFB2a using organic acids which was then adducted to amino acids. Accurate mass measurements, MS/MS fragmentation patterns, and nuclear magnetic resonance (NMR) data of AFB2a-amino acid adducts support a novel adduct structure. AFB2a-arginine was chosen as the top detoxification candidate due to no significant breakdown in simulated digestive fluids, its high polarity, and a 10,000-fold decrease in its octanol-water partition coefficient compared to AFB1 (determined using HPLC-ESI-TOFMS). Genotoxicity was evaluated using the SOS-Chromotest and Ames’ test in which AFB2a-arginine has shown a dramatic decrease in genotoxicity. Additionally, interactions between these toxins and DNA were compared by identifying the formation of base pair adducts using TOFMS. Results show clear DNA-adduct formation with AFB1 while AFB2a-Arg was completely unreactive. Together the data support that when adducted to arginine, AFB1loses its mutagenicity.