A molecular systems biomedicine approach to Diabetes: characterization, prioritization and network analysis of new candidate genes from the apoptotic machinery

Diabetes is a complex systemic disease, characterized by loss of pancreatic islet β cells and expansion of α cells. To gain a systems view on its etiology, we investigated the involvement of the apoptotic machinery (AM) through FACS, High-Throughput Real-Time RT-PCR, western analysis, and analyzed the modifications of AM transcriptome and AM protein nodes in two mouse pancreatic α and β cell lines (αTC1 and βTC1, respectively), after exposure to proinflammatory cytokines, for 24, 48, 72 h. AM genes markedly over or down expressed, and conserved between the mouse and Homo sapiens, were hypothesized to be causally involved in human Diabetes and ranked through functional (Fisher's inverse chi-square test) and protein–protein interaction (Wilcoxon matched-pairs signed-ranks test) prioritization methods. 34% of the 92 AM genes analyzed significantly varied their expression respect to controls. The increase in βTC1 cells of ATF3, BNIP3, NOS2, Ser20-p-P53, TNFRSF10B proapoptotic proteins and the decrease of antiapoptotic Tyr705-p-STAT3 suggest the involvement of both extrinsic and intrinsic apoptotic pathways. In αTC1 cells, neither the phosphorylated forms of P53 and STAT3 nor the death receptor TNFRSF10B changed their levels, whereas NOS2 was highly induced. Prioritization studies pinpointed DDIT3 and STAT3 as associated to Type 1 Diabetes, and MAP3K14, NFKB1, NFKBIA, NFKBIB, NFKB2, RELA, STAT3 as associated to Type 2 Diabetes. Analysis of α and β cells regulatory networks showed that activation of P53 pathway and inactivation of STAT3 pathway play a crucial role in β cell death, while members of the NFκB complex and their regulators act as a common master module in both cell phenotypes. Our studies demonstrate the critical role of AM genes DDIT3, MAP3K14, NFKB1, NFKBIA, NFKBIB, NFKB2, RELA, STAT3 in β cell death. We propose them as strong candidates for human Diabetes, as confirmed by the previous identification of DDIT3, NFKB1, NFKB2, RELA through epidemiologic and functional studies.