Poster Abstracts

Poster number 1
Title: Proteome-wide effects of singlet oxygen produced by next generation Iridium anti-cancer metallodrugs
Presenting author: Kung Ching Cookson Chiu, University of Warwick
Other authors: Pingyu Zhang, Yuko P. Y. Lam, Huaiyi Huang, Christopher A. Wootton, Peter J. Sadler, Peter B. O'Connor

New metal-based anticancer drugs have been the focus of next-generation chemotherapy. Novel Ir(III) based drugs have been designed to minimise side effects on healthy cells, while increasing selectivity and efficacy towards cancer cells by inducing reactive oxygen species (ROS) when triggered with blue visible light. The catalytic production of singlet oxygen with Ir-complexes, and effects on target species within cells were investigated to uncover their mechanism of action on a proteome-wide scale. [1] FT-ICR MS analysis was conducted using a 12T Bruker SolariX (Bruker Daltonik GmbH, Bremen, Germany) FT-ICR mass spectrometer. nLC-MS/MS experiments were conducted via coupling an EASY nano-LC II system (Proxeon, Hemel Hempstead, UK) to the SolariX FT-ICR MS, separation was achieved using a water/acetonitrile gradient and a reverse phase C18 column. All data was analysed with Data Analysis v4.2 (Bruker Daltonik GmbH, Bremen, Germany) and the Mascot database search engine. Activation of the Ir metallodrugs caused production of reactive singlet oxygen species and extensive oxidation of nearby biomolecules. A whole range of oxidation products, up to 6 per peptide, were observed. The high resolution and mass accuracy and extensive MS/MS capabilities of the FT-ICR MS allowed unambiguous identification of the oxidation sites and the modifications induced. singlet oxygen was found to cause many methionine oxidation events (to sulfoxide and sulfone), but also oxidation of tryptophan residues, producing diagnostic kynurenine and 3-hydroxy kynurenine moieties. Histidine residues were oxidised to 2-oxo-histidine, which is commonly observed in singlet oxygen oxidation events. Bottom-up nLC-MS/MS experiments of drug-treated cancer cells revealed two key targets; aldose reductase (AR) and heat shock protein 70 (Hsp70). Quantitative nLC-MS/MS showed the increase in oxidised products in the drug-treated samples, calculated as 3.0-fold up-regulation with AR and 5.8-fold up-regulation with Hsp70, together with 9 proteins up-regulated along the glycolysis pathway. New Ir-based anticancer drug was designed to be highly and selectively cytotoxic to cancer cells when activated with visible blue light. Generation of singlet oxygen was confirmed by observing diagnostic oxidised species in peptides. Damage to specific cellular proteins, AR and Hsp70 was discovered in cells, induced by photodynamic therapy. 9 proteins along the glycolysis pathway were found up-regulated, which happened in the mitochondria of cells, disrupting the energy source of cancer cells.

References:
(1) Zhang, P.; Chiu, C. K. C.; Huang, H.; Lam, Y. P. Y.; Habtemariam, A.; Malcomson, T.; Paterson, M. J.; Clarkson, G. J.; O’Connor, P. B.; Chao, H.; Sadler, P. J. Organoiridium Photosensitizers Induce Specific Oxidative Attack on Proteins within Cancer Cells. Angew. Chemie Int. Ed. 2017, 14898–14902.

Poster number 2 
Title: Protein function prediction from amino acid sequence information by an ARMAP neural network 
Presenting author: Vassilis Cutsuridis, University of Lincoln 
Other authors: Patrick Ellison

Protein function prediction is of outmost importance in the development of new drugs. In this work we used machine learning classifiers including shallow and deep neural networks to classify a protein’s function based only on its amino-acid (AA) sequence. A Gene Ontology-based search against the UniProt/SwissProt database for “DNA sequence-specific binding proteins” and its complement set were retrieved. For training and testing, various size datasets were generated. The performances of the various classifiers was compared and contrasted. kNN outperformed SVM, feedforward and convolutional deep neural networks. The results of this study is of major importance in structural biology and biomedicine as it can provide a basis of the development of highly specific tools for genome modification and gene therapy. 

Poster number 3
Title: Machine learning algorithms for the predictions of protein function from primary structure 
Presenting author: Vassilis Cutsuridis, University of Lincoln 
Other authors: Michael Kokkinidis 

Protein function prediction is of outmost importance in the development of new drugs. In this work we used a default ARTMAP classifier to classify a protein’s function based only on its amino-acid (AA) sequence. A Gene Ontology-based search against the UniProt/SwissProt database for “DNA sequence-specific binding proteins” and its complement set were retrieved. For training and testing, various size datasets were generated . Datasets were generated either by random sampling from the existing categories or by classifying the proteins first into sub-groups based on a similarity measure and then randomly sampling from each sub-group. ARTMAP’s performance with the latter method performed better than with the former method in every size dataset and model. ARTMAP was successful in predicting the function of a protein from its AA sequence by extracting a shared sequence-specific feature that is linked to the specificity of DNA binding of proteins. This result is of major importance in structural biology and biomedicine as it can provide a basis of the development of highly specific tools for genome modification and gene therapy. 

Poster number 4
Title: The Next Dimension in Proteomics 
Presenting author: Pui Yiu Lam, University of Warwick 
Other authors: Christopher A. Wootton, Cookson K. C. Chiu, Tomos E. Morgan, Remy Gavard, Maria A. van Agthoven, M. Li, Mark P. Barrow, and Peter B. O’Connor 

Reversed phase liquid chromatography (RP-LC) is an essential tool for the proteomic studies due to its powerful separation of peptides and proteins prior to MS analysis; however, LC separation of extremely complex samples with widely varying properties suffers from the bias of the chromatography columns applied in LC separation and incompatibility with certain separation conditions such as solvent compatibility, pH, and column chemistry. An alternative analytical method which is able to handle complex proteomic samples and high flexibility in sample conditions is required. Herein, we apply two dimensional mass spectrometry (2DMS) to study proteomic samples ranging from a standard protein digest to a whole cell lysate digest. In the 2DMS experiments, we show that it is possible to obtain detailed sequence information from peptides without online LC separation. We then compared the results obtained from the 2DMS to the LC tandem MS data and show that 2DMS can achieve similar results to LC MS/MS for identified proteins in the proteomic samples shown. Furthermore, our data show that the 2DMS experiments tend to be more adept at analysing and identifying particularly hydrophilic, basic, or short, which is complementary to those being assigned in LC tandem MS. These results demonstrate that 2DMS provides an alternative technique for future proteomics. 

Poster number 5
Title: FXa Anticoagulant Bioactive Proteins Derived from Scorpion Venom 
Presenting author: Meng Li, University of Warwick 
Other authors: Yuko P. Y. Lam, Peng Chen, Remy Gavard, Cookson K. C. Chiu, Christopher A. Wootton, Tomos E. Morgan, Qiong Wu, Mark P. Barrow, Hongzheng Fu, Peter B. O’Connor 

Of the 1400 species of scorpions worldwide, the Chinese scorpion is used in Chinese medicines for cardiovascular problems, antimicrobial, and tumors. The venom is a complex mixture of proteins containing bioactive components, some showing inhibition of enzyme factor Xa, which prevents blood coagulation. Crude scorpion venom was fractionated using step elution on C18 cartridge, and in-vitro inhibition of factor Xa was tested. Nano-LC-FTICR MS/MS analysis of each fraction was acquired FT-ICR MS. 

Using direct infusion, more than hundred unique molecular species were observed in the crude venom, but with bottom-up proteomics, only a small number of proteins can be identified, even though many of these unidentified peptides have good fragmentation spectra. To accurately identify potential bioactive compounds, methodologies were developed to correlate the spectra with the bioactivity results. A list of compounds that strongly correlate to the bioactivity could be potential pharmaceutical targets. These potential targets are then sequenced using ECD, IRMPD and CAD. The potential factor Xa inhibitors were de novo sequenced using both top-down and bottom-up approaches. The high mass accuracy and resolving power of the FTICR-MS aids the sequencing study of these novel scorpion venom proteins. The method is applicable to identifying potential pharmaceutical leads from other natural products. " 

Poster number 6 
Title: LESA FAIMS imaging of intact proteins in tissue: The new frontier for in situ proteomics
Presenting author: Rian L Griffiths, University of Birmingham 
Other authors: Susan Abbatiello, Scott Kronewitter Iain B. Styles and Helen J. Cooper 

Liquid extraction surface analysis enables ambient sampling of intact proteins and other biomolecules directly from biological substrates. High field asymmetric waveform ion mobility spectrometry (FAIMS) enables molecular separation owing to different ion mobilities in high and low electric fields. Here we describe the benefits in sensitivity, separation and resolution achieved by incorporating a newly developed cylindrical FAIMS device into LESA MS and imaging workflows. Thin tissue sections of rat brain, testis and kidney were sampled with acetonitrile based solutions. One of the particular benefits of LESA is that after sampling, it is possible to electrospray for several minutes, allowing multiple FAIMS conditions to be considered within a single experiment. In this work various optimised voltages were considered for each tissue type. Lower and higher molecular weight intact proteins were detected at different voltages. Lipid species were separated from proteins, permitting improved spatial resolution in denaturing imaging experiments. From a single testis tissue location alone 1396 intact protein species were detected across 3 different FAIMS conditions. Furthermore, mass spectrometry images compiled from tissue sampling allows spatial profiling of an unprecedented number of intact protein species. This methodology presents new opportunities for in situ proteomics workflows. 

Poster number 7
Title: Proteomic analysis of targets of Ser65 phosphorylated ubiquitin. 
Presenting author: Julius Dongdem, School of Life Sciences, University of Nottingham 
Other authors: Simon Dawson, Robert Layfield 

Recent studies have highlighted additional levels of complexity in the post-translational modification of proteins by ubiquitin (Ub). For example, Ub itself can be modulated by phosphorylation to act as a second messenger in PINK1-Parkin mediated mitophagy. However, the full physiological significance of Ub phosphorylation is unclear. Thus, the project sought to catalogue mammalian target proteins modified by covalent attachment of phosphoSer65-(poly)Ub. 

Ub-WT and Ub-Ser65Asp (phosphomimetic mutant) sequences were engineered to allow overexpression of His/FLAG-tagged proteins in HEK293T cells. An additional Leu73Pro mutation was introduced to stabilise targets of Ub modification. Endogenous proteins modified by covalent attachment of transfected Ub sequences were purified by IMAC. Covalent modification was confirmed by western blotting and targets of modification were identified by LC-MS/MS. 

Western blotting affirmed greater target protein modification by Ub-Ser65Asp and Leu73Pro mutants, the latter an indication of higher resistance to deubiquitination. Proteomic analysis suggested differential modification of various target proteins by Ub compared to Ub-Ser65Asp, including endogenous SUMO2. By transfecting GFP-SUMO2 and its C-terminal–GG deletion mutant, along with phosphomimetic Ub, we confirm that Ub-Ser65Asp modifies SUMO2, rather than vice versa. Finally, we confirm that transfected His/FLAG-SUMO2 is modified by endogenous phosphoSer65-Ub. SUMO2 represents a novel target of Ser65 phosphorylated ubiquitin. 

Poster number 8
Title: A novel bioinformatic approach to identifying biomarkers of ageing and disease 
Presenting author: Matiss Ozols, University of Manchester 
Other authors: Sarah Hibbert, Alexander Eckersley, Jerico Revote, Jiangning Song, Christopher Griffiths, Rachel Watson and Michael Sherratt 

Age, disease and/or exposure to environmental factors can induce tissue remodelling as a consequence of alterations in either protein structure or relative protein composition and /abundance. The causative mechanisms may include post-translational modification by reactive oxygen species (ROS), sugars (generating advanced glycation end products AGEs), proteases or the accumulation of DNA damage. We have previously experimentally shown experimentally that proteins rich in UV-chromophore amino acids are differentially susceptible to UVR. Human skin is prone to ultraviolet radiation (UVR) damage leading to photoageing and attention to date has focused on abundant extracellular matrix (ECM) proteins in the dermis. However, it is evident that “minor” proteins also play an important role in maintaining tissue homeostasis. Here, we hypothesise that proteins and their genes can be stratified according to their susceptibility to modification. Our bioinformatic approaches characterise proteins defined by the Manchester Skin Proteome (http://www.manchesterproteome.manchester.ac.uk) according to their susceptibility to UV/ROS, glycation and protease cleavage (proteins) and photochemical modification by UVB (DNA). Using this approach collagenous proteins are predicted to be sensitive to glycation and proteolysis whilst their DNA is UV sensitive. These methods may be useful in identifying novel pathways and protein targets of ageing, inflammation and diabetes. 

Poster number 9
Title: Automatic identification of metal-bound biomolecules using SNAP-LC, imaging, and 2D-MS 
Presenting author: Christopher Wootton, University of Warwick 
Other authors: Yuko P. Y. Lam, Matthew Willetts, Maria A. van Agthoven, Mark P. Barrow, Peter J. Sadler and Peter B. O’Connor 

The majority of chemotherapy utilises metal-based compounds to preferentially attack cancer cells. New metallodrugs are currently in development based on various precious metals such as platinum, ruthenium, iridium, and osmium. Proteomic investigations of these compounds and their effects on cellular components is widely sought after to understand the varying mechanisms of action for these new potential drugs, but very difficult to interpret. To accurately identify and assign these species we present a new method to reveal metallodrug-modified species within complex data sets such as LCMS runs. The SNAP algorithm, currently used for peak picking in single spectra, has been modified to accurately identify metal modified components using the metal’s characteristic isotope pattern combined with averagine. 

SNAP-LC analyses each and every scan from large LCMS acquisitions and produces a “metal chromatogram” of the specified metal searched. This method was successfully applied to tryptic digest samples containing Pt, Os, and Ir metallodrugs. SNAP-LC was then applied to a proteomic investigation of an osmium metallodrug administered to a human ovarian cancer cell line, successfully identifying four modified species within ~10,000 observed, automatically and accurately. 

Despite isotope expansion issues, SNAP-LC is also shown to be suited to high-resolution top-down fragmentation of larger, intact protein species by modification to search through protein MS/MS spectra and careful tuning of parameters to analyse metal-containing fragments. 

Finally the prospect of applying SNAP to MALDI imaging runs and 2D-MS data was investigated to reveal metallodrug behaviour on larger scales. SNAP analysis of 2D data sets was of particular interest due to their complex nature and diagonal dependence of isotopes with respect to precursor and data point location. SNAP-2D allows accurate determination of fragment and precursor charge states for both metal-containing and standard species. 

Poster number 10
Title: Buffer Concentration Effects on the Quantitation of the Isomerization Products of Deamidation in Peptides using Electron Capture Dissociation Mass Spectrometry. 
Presenting author: Anisha Haris, University of Warwick 
Other authors: Yuko P. Y. Lam, Tomos E. Morgan, Cookson K. C. Chiu, Christopher A. Wootton, Mark P. Barrow, and Peter B. O’Connor 

Deamidation is a spontaneous non-enzymatic post translational modification (PTM), whereby asparagine (Asn) is converted to aspartic acid (Asp) and isoaspartic acid (isoAsp). The isoforms of Asp (L-Asp, D-Asp, L-isoAsp and D-isoAsp) can inactivate or contribute to unwanted isomers in proteins. Understanding the balance between these forms as well as being able to distinguish between the isomeric pairs is important. 

In short, unstructured peptides, the ratio of formation of isoAsp to Asp is typically 3:1. However, in proteins the ratio could be different and may be influenced by the structure. 

The ability to distinguish between Asp and isoAsp and quantitation of the ratio of these isomers will help us to gain a deeper understanding of the effect that they have on protein structure, which is known to alter the activity and function of the proteins. 

There is also an ionic strength dependence of deamidation. However, the significance of this effect on deamidation and the ratio of isoAsp to Asp is still under consideration. 

It can be difficult to separate isoAsp from Asp using conventional reverse-phase HPLC methods due to their slight structural differences. Baseline resolution is required and customized separation for each peptide mixture is often necessary with shallow gradients and long run times. 

Early work using tandem mass spectrometry focused on fragment ion abundances to differentiate Asp from isoAsp. However, peak intensities may not be reliable or reproducible due to varying experimental conditions. Ultra-high resolution FT-ICR MS and electron based fragmentation methods are more beneficial in distinguishing Asp and isoAsp and quantifying the ratio of these isomers using electron capture dissociation. 

Poster number 11
Title: Development of Liquid Extraction Surface Analysis Mass Spectrometry for Identification of ESKAPE Pathogens 
Presenting author: Jana Havlikova, University of Birmingham 
Other authors: Klaudia I. Kocurek, Willem van Schaik, Robin C. May, Iain B. Styles, Helen J. Cooper 

The ESKAPE pathogens are a group of microorganisms which account for most hospital-acquired infections, while levels of antibiotic resistance within this group is rising. Therefore fast and accurate analysis is very important. Liquid extraction surface analysis mass spectrometry (LESA MS) is a tool suitable for direct analysis of bacterial samples and identification of bacterial proteins without any specific sample preparation. Bacterial strains S. aureus MSSA476, P. aeruginosa PS1054, K. pneumoniae KP257 were grown on Lysogeny broth agar (LBA) and blood agar, E. faecium E745 and E. faecalis V583 were cultured on brain heart infusion medium. A strain of S. aureus NCTC13435 was cultured on a 3D skin model (Labskin). 

A number of proteins have been identified from the KP257, E745 and V583 colonies by LESA MS. The molecular masses of the 50S ribosomal protein L29 differed between the two Enterococci strains, suggesting the protein may be a useful biomarker for identification of Enterococcus strains. For bacteria cultured on the blood agar and the Labskin model, peaks corresponding to substrate proteins were identified by top-down protein analysis what demonstrates the ability of identification of substrate protein peaks together with bacterial protein peaks in the mass spectra by LESA MS. 

Poster number 12
Title: Tagging and Enrichment of Oxidative damage 
Presenting author: Georgina Charlton, University of Warwick 
Other authors: Cleidiane Zampronio, John Sinclair, Peter Kilby & Alex Jones 

Carbonylation modifications are used as key biomarkers of oxidative stress in animals and plants. However, due to their low abundance, carbonylation modifications have been difficult to analyse and most previous enrichment work has been done at the protein level, meaning most peptides analysed were not carbonylated. The abundance of carbonylation of proteins is typically measured using colourimetry after derivatisation with 2,4-Dinitrophenylhydrazine (DNPH). However, colourimetry does not reveal the identification of the protein nor the site of modification. We have explored using DNPH derivatisation as a basis for enrichment of carbonylated peptides before analysis by mass spectrometry, but with limited success probably due to the reversible nature of the hydrazone bond formed. As an alternative, we are developing an enrichment strategy at the peptide level using aminoxyTMT and Biotin-LC-Hydrazide. A suitable enrichment strategy will allow analysis of complex mixtures giving us a better understanding of proteins and sites which are more susceptible to carbonylation. 

Poster number 13
Title: Zinc-alpha-2-glycoprotein and glucose control: A cross-sectional untargeted proteomic analysis, targeted assay analysis and meta-analysis of observational studies 
Presenting author: Hattie Pearsey, Leicester-Loughborough Biomedical Research Centre and Leicester Precision Medicine Institute 
Other authors: Joseph Henson, Jack A. Sargeant, David Webb, Toru Suzuki, Helen Waller, Kamlesh Khunti, Leong Ng, Daniel J. Cuthbertson, Melanie J. Davies and Thomas E. Yates 

Aims: 1) To explore differences in the plasma proteome of white European (WE) and South Asian (SA) adults with normal glucose control (NGC) or non-diabetic hyperglycaemia (NDH), 

Main Findings: 72 men who were either WE or SA and with NGC or NDH, were selected from a diabetes screening trail database. Analysis of baseline samples, using liquid-chromatography Mass Spectrometry, identified 155 proteins, of which zinc-alpha-2-glycoprotein (ZAG) was the only protein that was significantly different between both ethnicities and glycaemic control statuses. Verification of findings determined that ZAG was lower in SA vs. WE individuals (36.43 mg. L-1 vs. 41.96 mg. L-1; p <0.05) but there was no difference between glycaemic categories (38.57 mg. L-1 for NGC vs. 39.62 mg. L-1 for NDH). A systematic search of five databases for observational studies on circulating ZAG in adult populations identified 1,347 papers. 12 studies were considered eligible for inclusion for pooling with our targeted assay analyses. Meta-analysis of findings suggested that circulating ZAG is significantly lower in individuals with metabolic dysfunction compared to healthy controls (-4.53 [95% CI: -7.67, -1.39] mg. L-1). Adjustment for BMI resulted in reduced heterogeneity and with the pooled mean difference remaining statistically significant (-2.17 [-3.44, -0.90] mg. L-1). 

Poster number 14
Title: Taking the Phosphate 
Presenting author: Jess Watts, University of Warwick 
Other authors: Georgina Charlton, John Hammond (University of Reading), Liz Wellington, Alex Jones 

Phosphate is an essential macronutrient required for plant growth, however its availability is often the limiting factor affecting crop growth. The majority of phosphate within soil is organic phosphate (Po) which is inaccessible for plants to take up, phosphate needs to be in its orthophosphate (Pi) form to be accessible by plants. Many plants have adapted ways to try and improve the availability of phosphate, one such way is through the secretion of certain compounds, such as acid phosphatases (APase), RNase’s, carbolic acids directly into the surrounding soil. While working with two different strains of Brassica rapa plants, R500 and IMB211; which are efficient and inefficient at phosphate uptake respectively. We found that there is a significant difference in the abundance of proteins secreted from R500 plants into a hydroponic growth media than the IMB211 plants secrete, particularly when the plants were grown in low phosphate conditions. R500 exudate proteins were found to have APase activity while the IMB211 exudates did not. Our findings could indicate that the reason R500 plants are more efficient at phosphate uptake compared to IMB211, is due to the differences in secreted proteins mobilising Pi availability from the soil. 

Poster number 15
Title: Enhancing primary production of microbial phototrophic organisms 
Presenting author: James Kerr, University of Warwick 
Other authors: Dr. Joseph Christie-Oleza 

Marine primary production is mainly driven by microscopic phytoplankton in the vast photic zones of the oligotrophic open ocean. Primary production is ultimately converted into dissolved organic matter (DOM), the main source of carbon and energy for heterotrophic bacteria and also the larger marine food web. Phytoplankton, through cell death, active efflux, and membrane leakage, thus drive bacterial community dynamics as the main source of DOM. It has recently been observed by this group that nutrient exchange is important in phototroph-heterotroph interactions, and that different phytoplankton produce different metabolites affecting the behaviour of co-occurring heterotrophic bacteria. This project aims to examine the key metabolites produced by various phototrophs to determine how they exert an influence on a model heterotroph. So far, mass spectrometric analysis of the exoproteome of defined combinations of different phototrophs and the model Ruegeria pomeroyi has been compared to Ruegeria pomeroyi grown axenically. This research is vital in understanding the underpinning system driving marine primary production. Such an understanding would inform the engineering of microbial communities to provide industrially valuable catalytic activities, e.g. such as is needed in bioremediation, and in enhancing biofuel production. 

Poster number 16
Title: Investigating Protein Arginine Methylation in Glioblastoma 
Presenting author: Sabrina Samuel, University of Hull 
Other authors: Dr Pedro Beltran-Alvarez, Prof John Greenman 

Despite intense research, the prognosis of Glioblastoma (GBM) patients remains poor, with a median survival of 1.5 years from diagnosis. The need for new therapeutic options is therefore imperative and evidence has suggested a role for PRMTs, a group of enzymes responsible for the transfer of methyl groups to arginine residues. Such evidence includes their increased expression in GBM and the loss of cell viability following their depletion in GBM cells. 

Here, we aimed to determine the anti-proliferative effect of PRMT inhibiting drugs on the GBM cell line U87-MG. MTS assays were used to determine changes in cell viability and western blot was used to confirm drug specificity. Inhibition of PRMTs has shown variable losses in cell viability in 2D and 3D cultures, with the greatest by the PRMT1 inhibitor Furamidine. Due to the dynamic and interactive nature of post-translational modifications, we also investigated possible cross-talk events involving PRMTs in GBM cells. By inhibition of PRMT1, we were able to show changes in symmetrical dimethylation and lysine acetylation, suggesting cross-talk between the groups of enzymes. Our experiments help elucidate cellular and biochemical mechanisms in GBM cell lines, contributing to the understanding of this devastating brain tumour. 

Poster number 17
Title: 2D FT-ICR MS/MS analysis of IgG1 
Presenting author: Johanna Paris, University of Warwick 
Other authors: Tomos E. Morgan, Chistopher A. Wootton, Frederik Lermyte, Yuko P. Y. Lam, Mark P. Barrow, John O'Hara, Peter B. O'connor 

Recombinant monoclonal antibodies and derivatives are widely used as therapeutic drugs. They are susceptible to posttranslational modifications (PTMs) that could occur during the manufacturing process and storage, resulting in product-related impurities. PTMs can change the efficacy, toxicity and the clearance of the antibody; therefore, they need to be well monitored and characterised. 

The poster will investigate the potential of 2D MS for the analysis of the antibody. 2D MS is a powerful technique, which enables the correlation between the precursor and its fragments in a single experiment without prior isolation. A tryptic-digested antibody was analysed by 2D IRMPD MS/MS with a 12T FT-ICR mass spectrometer. 

Poster number 18
Title: DESI imaging of intact proteins in human non-alcoholic steatohepatitis tissue: A large cohort study. 
Presenting author: James Hughes, University of Birmingham 
Other authors: Rian Griffiths, Mark Towers, Emmanuelle Claude, Patricia Lalor, Iain Styles, Helen Cooper 

Desorption electrospray ionisation (DESI) has been used previously to analyse protein species directly deposited on to surfaces and to probe the distribution of small molecules, including lipids, within biological tissue samples. Herein, we report the first use of DESI to probe the distribution of proteins in diseased human liver samples at a lateral resolution of 150 μm. DESI MSI of proteins was achieved using a combination of optimised solvent conditions, modified DESI spray head, and travelling wave ion mobility spectrometry. The tissue under analysis was derived from human patients suffering from non-alcoholic steatohepatitis (NASH), a severe presentation of non-alcoholic fatty liver disease. NASH is characterised by presence of both inflammation and extensive formation of fibrotic tissue within the liver. MSI sampling was complimented by H&E staining of serial sections to allow correlation between MS data and tissue features. Several proteins including haemoglobin subunits α & β, and liver fatty acid binding protein (and natural T94A variant) were observed in multiple high charge states. Importantly, the distribution of proteins within the tissue was not homogenous and correlated with H&E tissue features (e.g. scar tissue). Based on this success, a larger cohort (n=20) is currently under investigation and results will be discussed 

Poster number 19
Title: An integrative multi-omics approach for improved discovery of biomarkers related to prostate cancer progression 
Presenting author: Sarah Wagner, John van Geest Cancer Research Centre, Nottingham Trent University, Clifton Lane, Nottingham NG11 8NS, UK 
Other authors: Graham R. Ball, Clare Coveney, Naomi Dunning-Foreman, Sibylle Heidelberger, Nick Morrice, A. Graham Pockley, Robert Rees, Stephen Tate, Jayakumar Vadakekolathu, Amanda Miles, David Boocock 

Prostate cancer (PCa) is the second most common cause of cancer-related death in the UK. This is primarily due to the development of metastasis, which decreases the 5-year survival rate to 30%. The process highly implicated in the spread of cancer is called epithelial to mesenchymal transition (EMT). Here, we present the use of two inducible EMT models for an improved understanding of metastasis in PCa and the discovery of potential biomarkers for disease progression. 

For this, two prostate cancer cell lines were stimulated with transforming growth factor β, which induced morphological changes. The EMT induction was confirmed using quantitative real-time PCR, immunofluorescence staining and western blot analysis. Proteomic and transcriptomic profiles of both models in their induced and non-induced states were generated and their subsequent integration highlighted 13 potential biomarkers indicative of aggressive disease and metastasis development. Four of these were taken forward and further validated using tissue microarrays and the in silico analysis of public available datasets. The generated results have shown a significant association of all 4 markers with disease progression. This highlights the successful application of an integrated omics approach for the discovery of novel disease-associated biomarkers for prostate cancer progression. 

Poster number 20 
Title: Detection of Bevacizumab in 3D Osteosarcoma Model using Mass Spectrometry Imaging 
Presenting author: Lucy Flint, Sheffield Hallam University 
Other authors: Neil Cross, Laura Cole, Malcolm Clench 

Bevacizumab, a biopharmaceutical, acts as a tumour angiogenesis inhibitor that selectively binds to the vascular endothelial growth factor (VEGF). Matrix-assisted laser desorption ionisation mass spectrometry imaging (MALDI-MSI) is a powerful technique that can determine the spatial distribution of biopharmaceuticals and biological responses simultaneously in situ, without the use of labels. The multiplex nature of MALDI-MSI has further been expanded to the analysis of drug absorption and responses in 3D cell culture models, in particular tumour spheroids. In this study, a spheroid model term "aggregoid" was developed from osteosarcoma cell lines to analyse the spatial distribution of Bevacizumab by MALDI-MSI. Lipid and protein-originated peptides species were identified within the aggregoid by optimisation of the MALDI-MSI approach. A "bottom-up" proteomics strategy identified unique CDR peptides of Bevacizumab for the initial development of detection within the 3D model. Optimisation of the detection of Bevacizumab in the aggregoid model is further ongoing. 

Poster number 21
Title: Are amino-acid hydroxylation reversible post-translational modifications? 
Presenting author: alex von kriegsheim, University of Edinburgh 

Amino acid hydroxylations are common post-translational modifications (PTMs) which can regulate protein functions in multiple ways. The modification can introduce a novel group which can be functionalised by other modification, such as glycosylation. Hydroxylations can additionally generate novel protein-protein interactions that can stabilise protein super structures, such as cross-linking collagen fibres, as well as generating functional signalling complexes. The most prominent example of a hydroxylation-regulated signalling protein is HIF1a, the master regulator of the cellular hypoxic response. Aside from HIF, many more signalling regulators have been recently identified as being regulated by hydroxylations. In contrast to most enzymatically regulated signalling PTMs, hydroxylations are generally regarded as irreversible modification, more akin to protein damage than to signalling regulators. In light of this conceptual discrepancy, we wanted to determine if the cellular machinery could reverse protein hydroxylations. We applied a molecular modelling approach and validated it by monitoring hydroxylation changes in live cells by a SILAC-based pulse-chase spike-in experiment. Using this approach, we were able to determine that asparaginyl hydroxylations are reversible on three individual proteins, including HIF1a. 

Poster number 22
Title: Molecularly Imprinted Polymers (MIPs) for the detection of low abundance proteins as biomarkers for lung cancer 
Presenting author: Rachel Norman, Leicester Cancer Research Centre, RKCSB, University of Leicester, University Road, Leicester, LE1 7RH. 
Other authors: Sergey Piletsky, Leong L Ng, Antonio Guerreiro, Francesco Canfarotta, Donald JL Jones 

Lung cancer is the most common cause of cancer-related death. The poor prognosis is principally due to patients being diagnosed with advanced disease which is more difficult to treat. Single amino acid substitutions in KRAS are driver mutations occurring early in carcinogenesis. Detecting these mutations could enable diagnosis of early stage cancer, improving survival. MIPs are like plastic antibodies produced by polymerising a monomer such as acrylamide in the presence of a template. Once the template has been removed, the polymer retains its shape and is capable of rebinding the template. MIPs were produced using the KRAS C-terminal peptide as the template. Plasma samples from non-small cell lung cancer (NSCLC) patients were incubated with the MIPs on Spin X centrifugal filters. After washing to remove other proteins, KRAS was digested on the MIPs before selected reaction monitoring (SRM) LC-MS/MS analysis on a Waters nanoACQUITY UPLC coupled to a Waters Xevo triple quadrupole mass spectrometer. A multiplexed SRM method was developed to identify wild-type KRAS and the five most common G12 mutations. G12C mutant peptide was detected in a patient plasma sample, corroborating with the DNA mutation data for that sample. 

Poster number 23
Title: Probing protein-ligand interactions by native LESA mass spectrometry 
Presenting author: Eva Illes-Toth, University of Birmingham, School of Biosciences, Birmingham, B15 2TT 
Other authors: Helen J. Cooper 

LESA (Liquid Extraction Surface Analysis) is a method that enables direct sampling of proteins from a solid surface e.g., glass or tissue sections under near native conditions. Our aim is to utilise this tool to characterise protein-ligand interactions with respect to stoichiometry and binding affinities. 

Here, we explored the binding of bovine carbonic anhydrase with sulfanilamide and chlorothiazide, and cytochrome c with cisplatin using LESA in combination with high resolution mass spectrometry. Solutions containing protein and ligand were dried onto glass. The dried sample spots were subsequently sampled by LESA with native like solvents. Preliminary data indicate that binding of these ligands can be maintained under these conditions. Carbonic anhydrase shows adduct formation with both of its respective ligands at micromolar concentrations, in particular with chlorothiazide. Cytochrome c forms a series of adducts with cisplatin. This work will form the basis of protein-drug binding studies performed without extensive sample preparation, under physiological conditions. 

Our data demonstrate the suitability of LESA for studying protein-ligand interactions of model proteins under native conditions. Subsequent studies will focus on biophysical characterisation of pathologically relevant proteins with their drug candidates.