Teasing apart the required components of the immune system for diabetes in mouse models
by Adam Burrack, PhD
Today’s topic will cut to the nitty-gritty of figuring out how type 1 diabetes (T1D) happens. Since we cannot do reductionist experiments in primates or humans – removing one component at a time and seeing whether each in turn is required – this post will focus on studying T1D development utilizing genetics models and mouse immunology. In this post I will describe a recent exciting development using our frequent subject, the non-obese diabetic (NOD) mouse, which develops an eerily similar type 1 diabetes to human patients.
Today’s researcher-of-interest is Dr. Emil Unanue, at Washington University in St. Louis. Dr. Unanue has been working in research to determine the rate-limiting cells and processes of beta cell destruction for over 30 years. His lab has published extensively about the interactions between CD4+ T cells (so-called helper T cells) and dendritic cells, the cell type primarily responsible for “presenting”, or showing, peptides to T cells. In other words, dendritic cells are critical for promoting or preventing T cell responses in a variety of normal and disease settings. Importantly, dendritic cells can show peptides to T cells in the context of an entire alphabet of either “co-stimulating” or “co-inhibitory” signals. As one might expect from these terms, “co-stimulation” helps promote T cell responses, while “co-inhibitory” signals can prevent T cells from responding to a specific peptide. For these signals to work, the dendritic cell expresses one molecule of a paired set on its surface, and the T cell expresses the other member of the pair of its surface. Broadly speaking, these interactions promote “differentiation” of the CD4+ T cell to either an effector type (from co-stimulation signals) or a regulatory type (from co-inhibition).
For T1D development in the NOD mouse, the interaction between CD4+ T cells with dendritic cells in the pancreas-draining lymph node (PLN) is critically important for either “breaking tolerance” – and promoting diabetes – or maintaining tolerance. Specifically, early during the mouse’s life these interactions occur and establish a “clone” of T cells dedicated to either attacking or protecting the beta cells in the pancreas. In addition, a “third signal” affecting T cell activation in general are the cytokines present in the lymph node where T cell priming takes place. Cytokines are similar to hormones in that they can influence the behavior of a variety of cells and are transported via the blood. Some cytokines – pro-inflammatory signals associated with cell death – promote T cell responses. Other cytokines, associated with down-regulatory cells, prevent or inhibit T cell responses. To establish a “fate choice” for a set of T cells, the sum of these three signals are combined: (1) the strength of T cell receptor binding to the peptide (tighter binding is associated with effector T cell responses), (2) the co-stimulation/co-inhibition signals, and (3) local cytokines.
Dr. Unanue’s group has found one type of dendritic cell which is critical for activating T cells to attack beta cells in the pancreas. In a paper in the October 2014 issue of the journal Immunity – one of the high-tier journals for basic immunology research – his group used a genetic deletion strategy to remove dendritic cells expressing a particular transcription factor. Transcription factors are proteins that orchestrate the expression of other genes – turning them either on – called promoters – or off – called repressors – in response to very specific signals from the environment. Point being, transcription factor expression associates with cell “fate” choices.
In this paper the transcription factor BATF3 was removed from dendritic cells, and this prevented diabetes development in NOD mice. These dendritic cells – which express the CD103 molecule on their cell surface – appear to be required for the activation and infiltration of CD4+ T cells. Without this interaction between CD103-expressing dendritic cells and CD4+ T cells in the pancreas-draining lymph node, these genetically modified NOD mice do not develop diabetes. This is important because it establishes a rate-limiting interaction to develop autoreactive T cells which cause diabetes. This knowledge may inform future attempts at intervention strategies in the NOD mouse model to prevent presentation of specific targets of autoreactive T cells – such as insulin – in order to prevent development of autoreactive T cells. In other words, our goal will probably not become to deplete this group of CD103+ dendritic cells in the pancreas-draining lymph nodes of human patients susceptible to development T1D. Rather, one goal might be to change the outcome of this interaction. To accomplish the outcome of delayed or prevented disease, we may have to change the context in which these dendritic cells interact with CD4+ T cells, in particular the “co-stimulation/co-inhibition signals”, or the local cytokines. We now know a critical cell type – CD103+ DCs in the PLN – that could potentially be targeted in future, pre-clinical, therapeutic efforts.