The maintenance of transplantation tolerance induced in adult mice after short-term treatment with nonlytic monoclonal antibodies to CD4 and CD8 was investigated. CD4+ T cells from tolerant mice disabled naïve lymphocytes so that they too could not reject the graft. The naïve lymphocytes that had been so disabled also became tolerant and, in turn, developed the capacity to specifically disable other naïve lymphocytes. This process of "infectious" tolerance explains why no further immunosuppression was needed to maintain long-term transplantation tolerance.

TGF-beta1, expressed in the pancreatic islets, protects the nonobese diabetic (NOD) mouse from insulin-dependent diabetes mellitus (IDDM). The islet antigen-specific T cell response of ins-TGF-beta1 mice relied on different antigen-presenting cells (APC) from those used by NOD T cells. T cells from NOD mice utilized B cells to present islet antigen, whereas T cells from ins-TGF-beta1 mice utilized macrophages. In addition, the islet antigen-specific T cell repertoire of ins-TGF-beta1 mice was distinct and deviated toward an IL-4-producing Th2 phenotype. When ins-TGF-beta1 mice were treated with anti-iL-4 antibody, islet antigen-specific IFNGamma-producing Th1 cells were unleashed, and the incidence of diabetes increased to the level of NOD mice. This suggests active suppression of a diabetogenic T cell response. This study describes a novel mechanism in which expression of TGF-beta1 in the context of self-antigen shifts APC preference, deviating T cell responses to a Th2 phenotype, preventing IDDM.

Over the last five years it has become increasingly clear that the peripheral immune system can maintain tolerance to both self and non-self antigens through a variety of mechanisms. Although clonal deletion may play an important part in limiting rapidly expanding responses, there are many examples where antigen reactive T cells remain. It has been proposed that tolerance is maintained in this situation either by the induction of anergy or by ongoing suppression. The phenomenon known as immune deviation, where non-inflammatory Th2 responses could suppress Th1 and positively reinforce themselves provided an attractive explanation for infectious tolerance, where tolerant T cells could guide further naive T cells also to tolerance. However, experiments to test this hypothesis in the models of CD4 and CD8 antibody-induced tolerance have given conflicting data, with no clear evidence of Th2 responses in tolerant mice. In this paper we review recent data that IL-4 plays a role in suppression, but that the source of IL-4 may not be the tolerant/suppressor T cell. We also discuss how infectious tolerance can operate on third party antigens if they are linked on the same antigen presenting cell and how CD4+ T cells can suppress CD8+ T-cell responses. Finally, we suggest a model of infectious anergy that is compatible with the available data.

The induction of tolerance in a primed immune system would be valuable therapeutically, but has been difficult to achieve. Mice primed to multiple minor histoincompatible antigens (minors) are able to rapidly reject secondary grafts using either their CD4+ or CD8+ T-cell subpopulations. Short courses of treatment with nonlytic anti-CD4 and anti-CD8 antibodies targeted at both T-cell subsets can induce long-term peripheral T-cell tolerance in primed mice. We examine the mechanisms by which peripheral tolerance is maintained, and show that tolerant mice harbor CD4+ T cells capable of specifically suppressing rejection mediated by either subset of primed T cells. Remarkably, elimination of CD4+ T cells from tolerant mice resulted in graft rejection, suggesting that graft-reactive CD8+ T cells had not been eliminated, but had been under continuous regulation by "tolerant" CD4+ T cells. This result demonstrates that it may be possible to establish therapeutic operational tolerance without permanently inactivating all antigen-reactive cells.

Cachexia is the dramatic weight loss and muscle atrophy seen in chronic disease states, including autoimmunity, cancer, and infection, and is often associated with lymphopenia. We have previously shown that CD4+ T cells that express the lowest density of CD44 (CD4+CD44v.low) are significantly reduced in diabetic NOD mice that are cachexic compared with diabetic mice that are not cachexic. Using this model, and a model of cancer cachexia, we test the hypothesis that CD4+CD44v.low cells play an active role in protecting the host from cachexia. CD4+CD44v.low cells, but not CD4+ cells depleted of CD44v.low cells, delay the onset of wasting when infused into either diabetic or prediabetic NOD recipients. However, no significant effect on the severity of diabetes was detected. In a model of cancer cachexia, they significantly reduce muscle atrophy, and inhibit muscle protein loss and DNA loss, even when given after the onset of cachexia. Protection from wasting and muscle atrophy by CD4+CD44v.low cells is associated with protection from lymphopenia. These data suggest, for the first time, a role for an immune cell subset in protection from cachexia, and further suggest that the mechanism of protection is independent of protection from autoimmunity.