Proteomic Pathways to Type 2 Diabetes in the Pancreatic Islet

Unveiling proteomic changes that occur through stages of pathogenesis can provide unparalleled insights into the mechanisms underpinning disease. Using fluorescence-activated cell sorting (FACS) and liquid chromatography mass spectrometry/mass spectrometry (LC-MS/MS), I aimed to characterise the deep proteome of both whole islets and individual islet cell types (alpha, beta, gamma, delta and epsilon) from mice and humans in variable states of health and disease. Acquisition of individual islet cell type proteomes will serve as a window of enquiry into islet cell regulation and intra-cellular communication providing insight into how these factors affect systemic glucose regulation. A panel of diverse mice were phenotypically characterised to observe their responses to diet or genetically induced metabolic stress. Characterisation of the islet proteomes of these mice revealed background strain as the major determinant of islet capacity to maintain function under increased metabolic stress. Isolation of the individual islet cell types using FACS requires the targeting of specific markers for each cell type. Currently, specific surface markers for several islet cell types are not known and/or antibodies are not available, and the alternative of using internal markers requires fixation. Standardised methods of protein fixation have a variable extent of modifications, for example paraformaldehyde, and hence are incompatible with peptide identification after MS analysis. I have overcome these obstacles by using a reversible crosslinker, dithiobis(succinimidyl propionate), which specifically targets primary amines in proteins. Upon reversal of the crosslinker, it leaves modifications of a standard size, which can be incorporated into the peptide identification analysis as a protein modification. In doing so, internal markers specific to each islet cell type can be utilized for isolation and subsequent proteomic analysis, whilst retaining greater than 96% protein identification compared to non-fixed cells. Preliminary MS analysis of isolated human β-cells revealed 4,561 proteins representing the largest human β-cell proteome to date