Introduction
CD4+ T-cells are central players of adaptive immune responses, because they co-ordinate the activity of several other immune cells such as dendritic cells (DC), B cells and cytotoxic CD8+T-cells. CD4+T-cells can be subdivided into two broad subsets of helper and regulatory cells, which promote and suppress immune responses, respectively. Both regulatory and helper T-cells can produce Interleukin-10 (IL-10), which is a potent anti-inflammatory cytokine but also promotes cytotoxic T-cell responses and B-cell antibody production. Consequently, IL-10 has a dual role in autoimmunity and cancer: On the one hand, it is required to maintain intestinal immune homeostasis and to prevent colitis, but on the other hand it could promote the generation of pathogenic autoantibodies in systemic lupus erythematosus (SLE) or cytotoxic anti-tumor T-cell responses.

Role of IL-.10 producing regulatory T-cells in human immune-mediated diseases
Regulatory T-cells can be sub-divided into the well-defined CD25+Tregs, which express the lineage-defining transcription factor FOXP3, and the enigmatic Tr1-cells, which produce high levels of IL-10. It is unclear whether Tr1-cells represent an independent differentiation lineage or a transient and unstable activation stage. We identified a population of cytotoxic Tr1-like cells in human tissues that produce high levels of IL-10, are generated upon chronic antigenic activation. They stably express the lineage-defining transcription factor EOMES and are probably derived from pro-inflammatory EOMES+Th1-cells. EOMES+Tr1-like-cells are involved in several immune-mediated diseases. Thus, EOMES+Tr1-like cells, but not FOXP3+Tregs, were functionally impaired in systemic lupus erythematosus and in inflammatory bowel diseases (IBDs). Furthermore, they were enriched and clonally expanded in different types of human tumors. Tumour-infiltrating Tr1-like cells were associated with poor prognosis, but also with response to immunotherapy. In multiple sclerosis they were enriched and clonally expended in the cerebrospinal fluid, and they produced IL-10 in brain lesions. However, they were activated by EBV, but failed to respond to myelin-derived self-antigens, possibly explaining why they fail to prevent CNS damage.

Role of human Th17 subsets in immune-mediated diseases
We are also studying Th17-cells, since they drive pathogenic inflammation in several immune-mediated diseases. We found that different human Th17 subsets, which all express the chemokine receptor CCR6, are involved in individual pathologies.
We identified a population of non-polarised CCR6+ central memory T-cells that differentiate spontaneously to Th17-cells upon TCR stimulation, indicating that they represent pre-Th17-cells. They produce IL-10 and are auto-reactive, but possess a context-dependent helper/regulatory function: IL-10 produced by these CCR6+T-cells could inhibit auto-reactive T-cell proliferation induced by immature DC, but also induce IgG production by naive B-cells. They accumulated systemically in patients with systemic lupus erythematosus (SLE) and could promote extrafollicular autoantibody production, suggesting a pathogenic role. In STAT3-deficient patients with Job’s syndrome, Th17 differentiation was arrested at this pre-Th17 stage, and residual CCR6+ cells produced IL-10 instead of IL-17 in response to opportunistic pathogens that cause recurrent infections in these patients. In multiple sclerosis, auto-reactive CCR6+T-cells differentiated instead to Th1/17-cells, which were enriched in the cerebrospinal fluid after an attack and produced encephalitogenic cytokines like GM-CSF. Finally, in the gut we recently identified an IL-23-dependent subset of Th17 effector cells that produced very high levels of IL-17, together with some IFN-, and were associated with intestinal inflammation in Crohn’s Disease. They were activated by adherent-invasive E. Coli, but not by a commensal strain, providing thus a link between the dysbiosis of the intestinal microbiota and Th17-cells in IBDs.