Events

Poster	Abstract
"Generation Of A Tissue And Organelle Specific Mouse Proteome Database" Thomas Kislinger, University of Toronto, thomas.kislinger@utoronto.ca	The laboratory mouse Mus musculus represents a key model organism for investigating the molecular basis of human physiology, development and disease. Nonetheless, many of the gene products encoded by the mouse genome are of unknown function. Furthermore, the degree of diversity in the proteome composition of mouse tissues and organs remains unclear. To this end, we have extensively analyzed the protein content of highly purified organelles (nuclei, mitochondria, microsomes & cytosol) isolated from six major mouse tissues (heart, brain, liver, lung, kidney & placenta) using multidimensional liquid chromatography - tandem mass spectrometry. Hundreds of thousands of SEQUEST spectral database search results were statistically validated by the use of the STATQUEST algorithm [Kislinger et al. Mol Cell Proteomics 2003; 2: 96-106] leading to the identification of thousands of high confidence proteins. Hierarchical clustering of the datasets revealed tissue- and organelle-specific protein expression patterns. Automated computer–based annotation of the clusters revealed significant organelle- and tissue- specific enrichment of proteins consistent with physiological and functional expectation for each tissue and organelle. Several hundred previously uncharacterized proteins were associated with these tissue and organelle clusters providing insight into their biological functions. This proteomics dataset will serve as a valuable resource for investigating and understanding the molecular effects of changes in physiological, developmental and disease state. Selected proteins demonstarting tissue and/or organelle specificity are being validated by Northern Blotting analysis, immunolocalization and possible by RNAi knock-out functional analysis.
"APTarray: An Acquired Pathogen Titre Microarray for the Determination of Serum Antibody Levels" Rob Shipman, NoAb BioDiscoveries Inc., rshipman@noabbiodiscoveries.com	Pathogenic bacterial and viral antigens were arrayed on activated, covalent- binding glass slides together with a serial dilution of a human IgG standard (APTarray). Anonymous human sera were incubated with the APTarray and the bound IgG detected by a second incubation with a fluorescent dye-labelled anti-IgG conjugate. Fluorescence images of the bound serum IgG were captured using a ScanArray Lite microarray slide scanner. Fluorescence intensity was compared to the IgG standard and the amount of IgG bound to each antigen determined. APTarray images revealed, at a glance, significant differences between individual serum IgG titres to the arrayed antigens. Absence of pathogen-specific IgG titres was indicated by the absence of fluorescence at specific antigen locations on the APTarray. Blood donation screening for pathogen exposure or infection, pre-admission screening for IgG titres against common hospital-acquired pathogens, confirmation of the vaccinal response to community-acquired pathogens and bio-warfare agents are potential applications of the APTarray.
Characterization of the N-Linked Glycans of Cationic Peanut Peroxidase Cunjie Zhang, University of Western Ontario, czhang2@uwo.ca	Cationic peanut peroxidase (CP) is N-glycosylated at Asn 60 (GPb), Asn 144 (GPc), and Asn 185 (GPa). Based on its behavior on ConA column, CP can be separated into two fractions: nonbinding (CP+) and binding (CP-). The difference between CP+ and CP- is believed to be due to heterogeneous glycosylation. Recently, mass spectrometry (MS) has become a key tool in the structural elucidation of glycans. In this study, we used a combination of off-line RP HPLC and nano ESI MS/MS on a Q-TOF 2 MS to determine the structure of glycans of the tryptic glycopeptides of CP+ and CP-. This approach reduces the sample processing time and increases the MS sensitivity when compared with the direct study of the glycans derived from PNGase digestion. In CP+, three major species of glycans namely, N₂M₃NFX, N₂M₃N₂FX, N₂M₃N₂HF₂X, were observed at all three sites. Other species were observed: N₂M₃FX at GPa and GPb; N₂M₂FX at GPb and GPc; N₂M₂NFX at GPc. In CP-, in addition to the glycan species observed in CP+, other larger species, were observed: N₂M₃N₂H₂F₃X at all three sites; N₂M₃NH₂FX, N₂M₃N₂H₂FX and N₂M₃N₃H₃F₂X at GPa; N₂M₃NHF₂X and N₂M₃N₂HF₂X at GPb; N₂M₃N₃H₃F₄X at GPc. The masses of of CP+ and CP- were determined by ESI MS and ranged from 34.5 to 35.7 kDa for CP+, and 34.5 to 37.7 kDa for CP-. The increased size of CP- vs CP+ is due to the presence of larger glycan chains at all three glycosylation sites. M=mannose, H=hexose, F=fucose, X=xylose and N=N-acetylglucosamine
"Protein changes in brain endothelial cells during ischemia as analyzed by 2-D gel-based and ICAT™-based proteomic techniques" A.S. Haqqani, Institute for Biological Sciences, National Research Council arsalan.haqqani@nrc-cnrc.gc.ca	Brain ischemia is a localized deprivation of blood supply to brain tissue. Microvessels in brain are lined by brain endothelial cells (BEC) that display a tightly-sealed blood-brain barrier (BBB) phenotype. Brain ischemia rapidly initiates a series of biochemical and molecular processes that disrupt BBB leading to brain edema and inflammation. The exact molecular mechanism that leads to BBB disruption, including changes in gene and protein profiles occurring in BEC, is not known. In this study, we subjected BEC to in vitro ischemia and examined protein changes using both 2-D gel electrophoresis-based and isotope-coded affinity tag (ICAT™)-based proteomics methods. For 2-D gel analysis, both wide (3-10) and narrow (4-7) pH rages were used. Sypro Ruby-stained gels were quantitatively analysed using PDQuest™ 7.0. At least 30 unique proteins were identified using LC-MS/MS to be either up- or down- regulated by 2-fold. Induction of one of the protein was confirmed using a biochemical assay. For ICAT™ analysis, tryptic mixture of heavy-tagged ischemic and light-tagged control samples were examined by nanoLC-MS/MS (Q-TOF Ultima) and analyzed by MassLynx-3.5 and in-house softwares. More than 300 ICAT-labeled proteins have been identified and 20% of them are differentially regulated. Several of these proteins show similar profile change by 2D gel analysis. The observed changes in several proteins were consistent with some of the known phenotypic changes seen in BEC during ischemia. Preliminary results show that ICAT can also be used to analyze protein changes in laser-captured BEC in ischemic rat brain tissues.
"In vivo Proteomics of Francisella tularensis: a Preliminary Report" Susan Twine, National Research Council, susan.twine@nrc.ca	The aim of this work is to apply cell separation and proteomics techniques to examine a variety of bacterial pathogen-host interactions. Francisella tularensis is the causative agent of tularemia, a potentially fatal disease of humans and other mammals. The pathogen is considered a potential bioweapon because of its extreme infectivity, ease of dissemination and the capacity to cause illness or death in humans. Many studies of the organism have focused upon in vitro models, mimicking in vivo conditions. Here, we are developing immunomagnetic separation techniques for the rapid isolation of cells from the infected host in order that we may compare the protein expression of organisms in vivo and in vitro. Traditional methods of proteome analysis, such as Two-Dimensional Polyacrylamide Gel Electrophoresis (2D-PAGE) have been used concurrently with emerging techniques, such as isotope coded affinity tag (ICAT) to fully characterize differential protein expression. The ICAT approach confers the advantage of sensitivity, and the ability to carry out comparative quantitative analysis of proteins. We have demonstrated that Francisella may be harvested from culture medium using immunomagnetic separation in sufficient numbers to resolve the protein content using traditional 2D-PAGE. However, we have developed proteomics analysis protocols using Francisella Live Vaccine Strain exposed in vitro to hydrogen peroxide, which mimics intracellular conditions to which the pathogen would normally be exposed. A number of stress proteins are known to be upregulated under such conditions. We have detected these proteins using both 2D-PAGE and ICAT techniques
"DomPred: Domain Linker Prediction By Protein Sequence" Michel Dumontier, University of Toronto	The identification and annotation of protein domains provides a critical step in the accurate prediction of molecular function. Experimental methods of protein structure determination such as X-Ray Crystallography and Nuclear Magnetic Resonance are complicated if not deterred by flexible linker regions or hampered by large protein sequences. We present here a computational method to identify the boundaries of protein domains from sequence information alone. A domain linker index (DSF) was derived from the ratio of the linker regions to compact domain amino acid composition from a non-redundant structure dataset. These linker regions exhibited a propensity for Pro and Gly but not for hydrophobic residues. DomPred converts a protein sequence to a smoothed numeric profile using any amino acid index and Discrete Fourier transform. Domain linkers are then predicted from Z score values generated from the distribution of scores over the full dataset. The DSF obtains 40% in sensitivity and 34% in specificity in a single-linker dataset (within ±20 residues from linker). However, the combination of DSF and scores from an entropy-based amino acid index increase the overall sensitivity to 64% and specificity to 35%. These results have been put into context in relation to other methods, datasets and results obtained from other prediction methods. DomPred provides a basic means by which sequence-based prediction of protein linker domains can be evaluated using combinations of simple amino acid indexes. In the absence of other information, DomPred should prove to be a valuable tool for delineating protein domains.
"Species-Specific Protein Sequence and Fold Optimizations" Michel Dumontier, University of Toronto	An organism's ability to adapt to its particular environmental niche is of fundamental importance to its survival and proliferation. In the largest study of its kind, we sought to identify and exploit the amino-acid signatures that make species-specific protein adaptation possible across 100 complete genomes. Environmental niche was determined to be a significant factor in variability from correspondence analysis using the amino acid composition of over 360,000 predicted open reading frames (ORFs) from 17 archae, 76 bacteria and 7 eukaryote complete genomes. Additionally, we found clusters of phylogenetically unrelated archae and bacteria that share similar environments by amino acid composition clustering. Composition analyses of conservative, domain-based homology modeling suggested an enrichment of small hydrophobic residues Ala, Gly, Val and charged residues Asp, Glu, His and Arg across all genomes. However, larger aromatic residues Phe, Trp and Tyr are reduced in folds, and these results were not affected by low complexity biases. We derived two simple log-odds scoring functions from ORFs (CG) and folds (CF) for each of the complete genomes. CF achieved an average cross-validation success rate of 85 ± 8% whereas the CG detected 73 ± 9% species-specific sequences when competing against all other non-redundant CG. Continuously updated results are available at http://genome.mshri.on.ca. Our analysis of amino acid compositions from the complete genomes provides stronger evidence for species-specific and environmental residue preferences in genomic sequences as well as in folds. Scoring functions derived from this work will be useful in future protein engineering experiments and possibly in identifying horizontal transfer events.
"Human Serum Proteome by LC/LC-MS/MS" John Marshall, Ryerson University, 4marshal@ryerson.ca	Mankind aims to enumerate the proteins of the human body and elucidate their functional and structural relationship. The first step is to identify the proteins with respect to a nucleic acid data base such as the refC database. A variety of protein preparation techniques were compared to produce the list presented here. The resulting peptides were separated by chromatography prior to tandems mass spectrometry (LC-MS/MS. The mass spectrometers employed included Paul ESI -ion trapand MALDI-Qq-TOF. Here we present a list of some ~3000 protiens from human serum. Most of the peptides on this list are shown with a unified score > 2400. Where there is no unified score shown the peptides were matched against nucleic acid databases using SEQUEST (X-CORR:1+/1.9, 2+/2.5, 3+/3.75 DELTA-CORR> 0.1) or MACOT (p<0.05) . Together these methods show coverage of all of the commonly-known serum proteins, for example, the complement factors or apo-lipoproteins and many proteins not previously observed in blood.