ActoBioticsTM for antigen-specific type 1 diabetes intervention: mono- or combination therapy

In T1D patients, auto-Ags of pancreatic insulin-producing ß-cells are targeted by auto-reactive T cells. The underlying events triggering this excessive auto-reactivity remain unknown, but clearly involve genetic predisposition, allowing inappropriate interaction of environmental factors with the immune system. When diagnosed with T1D, patients are limited to exogenous insulin administrations to control their aberrant glucose metabolism. However, this does not fully prevent the occurrence of complications at later stages in the disease, making it difficult to pursue normal life quality and quantity. Moreover, this does not address the underlying Ag-specific autoimmunity. Various auto-Ags – a.o. PINS, GAD65 and IA-2 – are involved in the pathogenesis of T1D. Numerous studies in preclinical models have highlighted the potential of Ag-specific approaches to restore tolerance in a highly specific and safe manner. Oral auto-Ag administration uses the tolerogenic nature of the gut-associated immune system to induce Ag-specific tolerance, a phenomenon referred to as oral tolerance induction. However, due to gastric degradation, proper mucosal product delivery often imposes a challenge. Recombinant L. lactis have proven to be effective and safe biological carriers for gastrointestinal delivery of therapeutic products. Preclinical successes with systemic broad immune modulating or Ag-specific therapies have repeatedly failed to demonstrate prolonged efficacy in the clinic, urging the need to explore new therapeutic strategies, such as combinatorial immune approaches, targeting multiple molecular pathways. Previously, recombinant L. lactis secreting PINS and IL-10 have demonstrated to induce diabetes remission in a well-defined mouse model for T1D, when combined with low-doses anti-CD3 mAbs. Here, we wanted to explore (i) whether targeting other prominent auto-Ags – GAD65 and IA-2 – could enhance the therapeutic success of this combination strategy and (ii) whether co-administration of systemic immunomodulators are essential for therapeutic efficacy. First, we generated recombinant L. lactis secreting human GAD65 or IA-2, two major diabetes-related auto-Ags, by themselves and combined with the anti-inflammatory cytokine human IL-10. Prohibitive sequence obstacles hampering Ag secretion were resolved by trimming the full size proteins. GAD65370-575 and IA-2635-979 showed to be efficiently secreted by recombinant L. lactis while maintaining sufficient bacterial growth. Our data demonstrated that secretion efficiencies could be influenced by mere physico-chemical properties, such as hydrophobic regions, but emphasized that engineering strategies based on functional epitope-distribution could provide a more relevant secretion product in terms of epitope density and secretion efficiency. To explore the tolerogenic potential of mucosally delivered GAD65 and IA-2, recent-onset diabetic NOD mice received intragastric inoculations of recombinant L. lactis secreting GAD65370-575 or IA-2635-979 and IL-10. In combination with short-course low-dose anti-CD3, L. lactis secreting GAD65370-575 and IL-10 stabilized insulitis, preserved functional ß-cell mass and restored normoglycemia in recent-onset NOD mice, even when hyperglycemia was severe at diagnosis. Combination therapy did not eliminate pathogenic effector T cells, but significantly decreased the inflammatory profile of islet cell infiltrates locally in the pancreas. Moreover, this combi-GAD therapy increased the presence of functional CD4+Foxp3+CD25+ Tregs in the PLN. In addition, we show, for the first time, that also IA-2 enholds the potential to normalize hyperglycemia in recent-onset NOD mice. Regardless of IL-10 co-delivery, recombinant L. lactis secreting IA-2635-979 induced diabetes remission when combined with low-doses anti-CD3 mAbs. Moreover, efficacy for both auto-Ags was similar in mice with mild and severe hyperglycemia at time of diagnosis. The combined delivery of the three prominent auto-Ags – PINS, GAD65370-575 and IA-2635-979 – in our combination setting did not enhance efficacy. The comparable success rates suggest that, at least in part, Treg-mediated bystander suppression could account for the tolerogenic effects, as was partly demonstrated in cured combi-GAD-treated mice. These data further suggest that inherent features of the NOD mouse model, such as insufficient ß-cell mass or fullblown inflammation levels in a subgroup of diagnosed mice, could account for the limited efficacy. We further explored an innovative Ag-specific combination strategy that could open up new therapeutic possibilities for T1D. As a general conclusion, we can say that recombinant L. lactis are an exquisite and flexible biological delivery platform that could be applied for oral tolerance induction in the emerging field of autoimmune diseases. In the NOD mouse model, we have demonstrated therapeutic efficacy with a combinatory approach whereby systemic immunomodulation installs the tolerogenic conditions needed for Ag-specific immune regulation to stably restore tolerance. Small Ag-steered differences in therapeutic efficacy favor the exploration of a patient-tailored clinical validation. In order to ensure efficient clinical translation of combination therapies, a substantially validated choice of monotherapies, defined target groups and relevant endpoints are key and need to be carefully considered. Taken together, our data point to a clear opportunity of combinatory strategies, where recombinant L. lactis could make the difference in medicine.status: publishe