Search

Inflammatory CD4(+) T cell responses to self or commensal bacteria underlie the pathogenesis of autoimmunity and inflammatory bowel disease (IBD), respectively. Although selection of self-specific T cells in the thymus limits responses to mammalian tissue antigens, the mechanisms that control selection of commensal bacteria-specific T cells remain poorly understood. Here, we demonstrate that group 3 innate lymphoid cell (ILC3)-intrinsic expression of major histocompatibility complex class II (MHCII) is regulated similarly to thymic epithelial cells and that MHCII(+) ILC3s directly induce cell death of activated commensal bacteria-specific T cells. Further, MHCII on colonic ILC3s was reduced in pediatric IBD patients. Collectively, these results define a selection pathway for commensal bacteria-specific CD4(+) T cells in the intestine and suggest that this process is dysregulated in human IBD.

Although strategies that block FOXP3-dependent regulatory T cell function (CTLA4 blockade) and the inhibitory receptor PD1 have shown great promise in promoting antitumor immune responses in humans, their widespread implementation for cancer immunotherapy has been hampered by significant off-target autoimmune side effects that can be lethal. Our work has shown that absence of OX40 and CD30 costimulatory signals prevents CD4 T cell-driven autoimmunity in Foxp3-deficient mice, suggesting a novel way to block these side effects. In this study, we show that excellent antitumor CD8 T cell responses can be achieved in Foxp3(KO) mice deficient in OX40 and CD30 signals, particularly in the presence of concurrent PD1 blockade. Furthermore, excellent antitumor immune responses can also be achieved using combinations of Abs that block CTLA4, PD1, OX40, and CD30 ligands, without CD4 T cell-driven autoimmunity. By dissociating autoimmune side effects from anticancer immune responses, this potentially shifts this antitumor approach to patients with far less advanced disease.

Although CD4(+) memory T cells reside within secondary lymphoid tissue, the major reservoir of these cells is in the lamina propria of the intestine. In this study, we demonstrate that, in the absence of signals through both OX40 and CD30, CD4(+) T cells are comprehensively depleted from the lamina propria. Deficiency in either CD30 or OX40 alone reduced CD4(+) T cell numbers, however, in mice deficient in both OX40 and CD30, CD4(+) T cell loss was greatly exacerbated. This loss of CD4(+) T cells was not due to a homing defect because CD30 x OX40-deficient OTII cells were not impaired in their ability to express CCR9 and alpha(4)beta(7) or traffic to the small intestine. There was also no difference in the priming of wild-type (WT) and CD30 x OX40-deficient OTII cells in the mesenteric lymph node after oral immunization. However, following oral immunization, CD30 x OX40-deficient OTII cells trafficked to the lamina propria but failed to persist compared with WT OTII cells. This was not due to reduced levels of Bcl-2 or Bcl-XL, because expression of these was comparable between WT and double knockout OTII cells. Collectively, these data demonstrate that signals through CD30 and OX40 are required for the survival of CD4(+) T cells within the small intestine lamina propria.