Defects in Antigen-Presenting Cells in the BB-DP Rat Model of Diabetes

Type-1 diabetes is the result of a T cell mediated immune response against the insulin-producing β cells in the islet of Langerhans. In humans, until now, the disease is only clearly detectable at the onset of the disease. Therefore studies to identify initial factors involved in the etio-pathogenesis are impossible in humans prone to develop diabetes. In order to study the early, prodromal phases of type-1 diabetes we used a spontaneous rodent animal model of the disease, the Biobreeding-Diabetes Prone (BB-DP) rat. This rat develops diabetes, because it is in particular defective for a population of regulatory T cells, the ART2+ regulatory T cells and because it possesses the disease-prone MHC haplotype RT1u (iddm1). We investigated (Chapter 2) the myeloid dendritic cells (DC) in this animal model, since DC, the antigen-presenting cells par excellence, are able to elicit immune responses from naïve T cells and are known to be involved in autoimmune responses because they are capable of modulating immunity versus tolerance. We studied bone-marrow precursor derived myeloid DC of three BB-DP rat sub-lines (Worcester, Groningen, Seattle) to identify defects in these DC, which could be responsible for the defective tolerance induction towards diabetes-associated islet autoantigens in this rat model. We found that the myeloid DC generated from bone-marrow precursors were defective in these three BB- DP rat sub-lines, showing an immature, more macrophage-like phenotype, a low MHC class II expression on their surface, a reduced T cell stimulatory capacity, a reduced capability to differentiate into fully mature DC and a reduced production of the immunosuppressive cytokine IL-10 as compared to two control rat strains (Wistar, F344). We assume that such DC defects contribute to the decreased tolerance towards islet autoantigens in the autoimmune diabetes of the BB-DP rat, since such defective DC are in particular defective to stimulate ART2+ regulatory T cells sufficiently. We went on to study the gene linkage of the DC defects in the BB rat model (Chapter 2 and Chapter 3) and firstly we studied the DC development from bone marrow precursors in BB-diabetes resistant (DR) rats of the Seattle (S) sub-line. The BB-DP/S rat develops diabetes, the BB-DR/S rat not. The BB-DP/S rat is lymphopenic (particularly for the ART2+ regulatory T cells), the BB-DR/S rat is not. Genetically the BB-DP/S and DR/S rat differ for the Ian5 gene at the lyp gene locus (iddm2). This gene regulates the apoptosis of recent thymic emigrants and particularly induces the lymphopenia of the ART2+ Treg cells in the rat. In this model we found almost similar defects as in the generation of DC from bone-marrow precursors, but there was a difference with the BB-DP rat in that the DC had a normal MHC class II expression. Secondly we used F344.lyp rats congenic for a large stretch of chromosome 4 of the BBDP/S rat which harbors the lyp gene (iddm2), but also other diabetes-linked g! enes such asiddm4. This F344.lyp rat resembles the BB-DP rat in that it has a strong lymphopenia, but in contrast to the BB-DP rat does not develop diabetes, due to the lack of other important diabetes susceptibility genes (such as iddm1). The DC of these rats showed almost similar defects as found in the BB-DP rat model, but in contrast had a normal capability to differentiate into fully mature DC. From these observations in the various BB-related models we were able to construct a relatively detailed linkage of lyp and other BB genes with the DC differentiation defectspresent in the BB-DP rat: 1. the low MHC class II expression is linked to lyp (iddm2), 2. the reduced T cell stimulatory capacity is linked to other genes on chromosome 4 (candidate gene: iddm4) and 3. the reduced IL-10 production has a complex linkage pattern, including iddm2. In Chapter 4 we set out to isolate the thymus DC of the BB-DP rats and isolated the low-density non-adherent cells from rat thymuses. For other tissues (lymph nodes, spleen, thyroid) this procedure results in a (relatively) pure population of DC. To our surprise the thymus isolated low-density cells were not DC, but appeared to be a relatively unknown population of accessory cells in the thymus, the branched cortical macrophages. These cortical macrophages are involved in positive selection in the thymus cortex. The branched cortical macrophages of the BB-DP and the F344.lyp rat appeared to be defective in function and less capable of rescuing double-positive thymocytes from apoptosis as compared to BB-DR and F344.+/+ rats. This defect was thus linked at least in part to the lyp gene. Interestingly the defective branched cortical macrophages of the BB-DP and F344.lyp rat were also poor rescuers of the thymus generated ART2+ Treg cells. Hence defects in the branched cortical ! macrophages, as described here in the BB-DP rat model, most likely contribute both to the T cell lymphopenia of the rat and its deficient tolerogenic state. In experiments described in Chapter 5 we simulated the rendezvous between antigen and endocytic proteases in human monocyte-derived DC. The most important endocytic proteases are the cysteine cathepsins, which we can identify with a chemical tool, called activity-based probe (ABP), which identifies all the active cysteine cathepsins. By the action of these cysteine cathepsins exogenous proteins/peptides are digested into small peptides, which are eligible for loading onto MHC class II molecules followed by presentation to T cells. After internalization of ABP by macropinocytosis only cathepsin S is targeted. Blocking of the vacuolar ATPase abolishes this CatS- selective targeting, while LPS-induced maturation of DC results in degradation of active CatS, in keeping with the concept that immature DC change from antigen-uptake cells to antigen-presenting cells during maturation. Conjugation of the ABP to a protein facilitated the delivery to endocytic proteases and resulted in l! abelling of sizable amounts of CatB and CatX, although CatS still remained the major protease reachedby this construct. Conjugation of the probe to a cell-penetrating peptide (CPP) routed the tracer to the entire panel of intracellular cathepsins, independently from endocytosis or LPSstimulation. Thus, different means of internalization result in differential targeting of active cathepsins in live monocyte-derived DC. We also concluded that CPP may serve as vehicles to target antigen more efficiently into cells. Taken together, when our DC data obtained in the BB rat model of type-1 diabetes are compared to DC data previously obtained in a mouse model of type-1 diabetes (the NOD mouse) and to those in type-1 diabetic patients, a blueprint regarding the aberrant function of professional myeloid APC (DC and accessory macrophages) becomes evident: The T cell stimulatory function of such cells is defective, while the cells are more macrophage-like and in a pro-inflammatory state. The pro-inflammatory state of the APC skews naïve T effector cells into predominantly “dangerous” Th1 cells. Due to their defective T cell stimulatory capability the APC support/induce Treg cells insufficiently and are less capable of inducing tolerance via activation induced T cell death (AITCD). We therefore consider the here described defects in myeloid DC as cornerstones in the etio- pathogenesis of T1D contributing to the breakdown of tolerance towards islet autoantigens finally leading to the destruct! ion of the insulin-producing β cells.

• View at Worldcat

Free Full Text ( Final Version , 4mb )