Poster Presentation 24th Annual Lorne Proteomics Symposium 2019

Multi-omic characterisation of bladder and lung carcinomas using a novel scanning quadrupole DIA acquisition method (#72)

Lee A Gethings 1 , Adam King 1 , Heather Patsiouras 2 , Robert Plumb 3
  1. Waters Corporation, Wilmslow, CHESHIRE, United Kingdom
  2. Waters Corporation, Sydney, NSW, Australia
  3. Waters Corporation, Milford, MA, USA

Cancer is one of the most complex, life threatening diseases, existing in many forms which have unknown pathogenesis. A combination of genetic and lifestyle factors are known to contribute towards increasing the probability of encountering cancer. Plasma samples were prepared for proteomic, metabolomic and lipidomic analyses. Label-free LC-MS data were acquired using a oa-QTof platform utilizing a broadband acquisition technique (SONAR) data independent acquisition workflow. Small molecule analysis consisted of using Progenesis QI for data processing to provide normalised values prior to statistical analysis. Unsupervised MVA of the data showed clear distinction between cohorts. OPLS-DA was used to filter for features of significant correlation and covariance prior to identification using HMDB (metabolites) and LipidMaps (lipids). Identifications matching the following criteria, mass accuracy <5 ppm, ANOVA p <0.05, %CV <30 and fold change >2 were considered for further interrogation. SONAR-based analysis indicates that the scanning quadrupole DIA enables over an order of magnitude more specificity than a static quadrupole operated with the same resolution and it was found that a quadrupole transmission window of approximately 10 Da provided optimum identifications. Proteomic data were processed using Progenesis QI for Proteomics and searched against a Uniprot Homo sapien database, containing reviewed entries and limited to 1% FDR. Additionally, we also searched the data against a spectral library using Spectronaut and comparatively analysed the results from both workflows. A number of significant proteins with differential regulation were exhibited for a number of protein groups involved in antigen and lipid binding. Proteins occurring in a minimum of two out of three replicates and with ANOVA p <0.05 were considered. Biological significance of the results was established by merging data from all three omic experiments and performing pathway analysis. A number of significant pathways including complement activation, B cell mediated immunity and receptor signalling were identified as key pathways.