Network approach reveals preferential T-cell and macrophage association with α-linked β-cells in early stage of insulitis in NOD mice.

Abstract

One of the challenges in studying islet inflammation - insulitis - is that it is a transient phenomenon. Traditional reporting of the insulitis progression is based on cumulative, donor-averaged values of leucocyte density in the vicinity of pancreatic islets, that hinders intra- and inter-islet heterogeneity of disease progression. Here, we aimed to understand why insulitis is non-uniform, often with peri-insulitis lesions formed on one side of an islet. To achieve this, we demonstrated applicability of network theory in detangling intra-islet multi-cellular interactions during insulitis. Specifically, we asked the question "what is unique about regions of the islet which interact with immune cells first". This study utilized the non-obese diabetic mouse model of type one diabetes and examined the interplay among -, {beta}-, T-cells, myeloid cells, and macrophages in pancreatic islets during the progression of insulitis. Disease evolution was tracked based on T/{beta} cell ratio in individual islets. In the early stage, we found that immune cells are preferentially interacting with -cell-rich regions of an islet. At the islet periphery -linked {beta}-cells were found to be targeted significantly more compared to those without -cell neighbors. Additionally, network analysis revealed increased T-myeloid, and T-macrophage interactions with all {beta}-cells.