The nonobese diabetic (NOD) mouse develops spontaneous T-cell-dependent autoimmune diabetes. We tested here whether vaccination of NOD mice with a plasmid DNA encoding glutamic acid decarboxylase (GAD), an initial target islet antigen of autoimmune T cell repertoire, would modulate their diabetes. Our results showed that vaccination of young or old female NOD mice with the GAD-plasmid DNA, but not control-plasmid DNA, effectively prevented their diabetes, demonstrating that GAD-plasmid DNA vaccination is quite effective in abrogating diabetes even after the development of insulitis. The prevention of diabetes did not follow the induction of immunoregulatory Th2 cells but was dependent upon CD28/B7 costimulation. Our results suggest a potential for treating spontaneous autoimmune diabetes via DNA vaccination with plasmids encoding self-Ag.

The mechanisms controlling induction of anergy at the level of naive CD4+ T cells are poorly understood but thought to reflect limited contact with costimulatory molecules during T cell antigen receptor (TCR) ligation. To clarify this question, naive TCR transgenic CD4+ cells were exposed to specific peptide presented by transfected antigen-presenting cells (APC) expressing MHC class II molecules with defined accessory molecules. Significantly, culturing CD4(+) cells with APC expressing MHC II plus peptide alone elicited early TCR signaling but failed to induce either proliferation or anergy. Culture with APC expressing MHC II plus B7 molecules led to strong proliferation and T cell priming but no anergy. In marked contrast, conspicuous induction of anergy occurred after T cell culture with APC expressing MHC class II and intercellular adhesion molecule-1 (ICAM-1). Thus, at the level of naive CD4(+) cells, anergy induction appears to reflect selective contact with APC expressing ICAM-1 in the absence of B7.

During T-APC interactions in vivo, interfering with CD40-CD154 interactions leads to reduced T cell priming, defects in effector function, and, in some cases, T cell tolerance. As shown here, however, presentation of conventional peptide Ags by CD40-deficient spleen APC in vitro leads to normal CD4+ T cell proliferative responses. By contrast, responses to the same peptides presented by purified B cells were markedly reduced in the absence of CD40. Thus, the requirement for CD40-CD154 interactions appears to be strongly influenced by the type of APC involved. Analysis of responses to endogenous superantigens, which are known to be strongly dependent on B cells for presentation, indicated that CD4+ responses to strong Ags are less dependent on CD40 than are responses to weak Ags. Similar findings applied to negative selection in the thymus. Thus, deletion of potentially autoreactive cells depended on CD40 expression when B APC were involved, and this requirement was most pronounced when negative selection was directed to weak Ags.

The type of cytokines produced during T cell responses determines susceptibility or resistance to many pathogens and influences the development of autoimmunity and allergy. To define the role of individual accessory molecules in cytokine production during primary immune responses, Drosophila cell lines expressing murine major histocompatibility complex class II molecules with defined combinations of accessory molecules were used to present peptide antigen to naive T cell receptor transgenic T cells. Significantly, expression of B7.1 or B7.2 without additional accessory molecules led to very high production of interleukin (IL)-4, which contrasted with minimal IL-4 production elicited by conventional antigen presenting cells (APC). However, coexpression of ICAM-1 and B7 on Drosophila APC induced little IL-4, suggesting an inhibitory role for intercellular adhesion molecule-1 (ICAM-1). In support of this idea, stimulation of T cell receptor transgenic T cells with peptide presented by splenic APC devoid of ICAM-1 (from ICAM-1-deficient mice) led to high IL-4 production. Thus, the level of IL-4 production by naive CD4(+) T cells during typical primary responses appears to be controlled, at least in part, by T-APC interactions involving ICAM-1.

Previous studies showed that activation of CD4(+) T cells with mouse mammary tumor virus-encoded Mls(a) superantigens induces strong proliferative responses and interleukin-2 production but fails to elicit typical early T cell receptor (TCR)-mediated signal transduction events, such as hydrolysis of polyphosphoinositides (PI) or an increase in intracellular calcium. Here we show that the failure of Mls(a) antigen to activate PI hydrolysis applies when resting B cells are used as antigen-presenting cells (APC). By contrast, when Mls(a)-bearing B cells are activated for 24h by exposure to lipopolysaccharide or, more importantly, to Mls(a)-reactive T cells or anti-CD40 antibodies the cells develop the capacity to elicit easily detectable PI turnover. These studies demonstrate that, for B cells as APC, the initiation of certain TCR-associated signal transduction pathways can depend on activation of the APC. The data suggest that cross talk between T cells and resting B cells can suffice to generate competent B APC and lead to the delayed initiation of signaling pathways important in T cell responses.

It is well known that interactions between accessory molecules on T cells and their ligands on APC play a key role in regulating T cell effector activity. The factors controlling the expression of these molecules are thus important determinants in the outcome of T cell activation. We have examined the expression of the murine ligand for CD27, a costimulatory molecule on T cells. Evidence is shown that CD27L is expressed at a low level on resting B cells but not on T cells, and that activation of B cells by culture with LPS or anti-IgM Ab increases the expression of CD27L. Interestingly, coligation of CD40 down-regulates CD27L on LPS-activated B cells but not on anti-Ig-activated cells. These findings suggest that costimulation via the CD27-CD27L pathway may be limited to interactions involving Ag-specific B cells, i.e., B cells specifically activated via their Ig receptors. In addition, testing a spectrum of different cytokines indicated that IL-4 and TGF, but not IL-2, IL-10, or IFN-gamma, prevented up-regulation of CD27L expression on activated B cells even when activation was induced by Ig signaling. The capacity of IL-4 to prevent CD27L expression could thus serve to limit CD27-CD27L interactions to Th1-type T cell responses.

CD40 ligand (CD40L) expression on T cells is known to play a crucial role in B cell responses, Some evidence also supports a role for CD40L-CD40 interactions in T cell responses, at least in vivo. Whether the T cell requirement for these interactions is an invariable finding, however, is less clear. Here, we provide evidence that the Ag specificity of T cells influences the requirement for CD40L. T cell hybridomas with dual reactivity for two different Ags, allo-H2-A(p) and Mls(a) superantigens, display a differential requirement for CD40L expression. Whereas the response to splenic APC expressing Mls(a) Ags requires CD40L expression, the response to alloantigen-bearing APC does not. The requirement for CD40L expression for the Mls(a) response appears to reflect a strong dependence of this response on ICAM-1 (intercellular adhesion molecule-1) and the ability of CD40-mediated signals to regulate ICAM-1 expression. These findings demonstrate that CD40L-CD40-mediated cross-talk is important for some but not all T cell responses and is influenced by both the type of Ag recognized and the type of APC.

HLA-DM (DM) facilitates peptide loading of major histocompatibility complex class II molecules in human cell lines. Mice lacking functional H2-M, the mouse equivalent of DM, have normal amounts of class II molecules at the cell surface, but most of these are associated with invariant chain-derived CLIP peptides. These mice contain large numbers of CD4(+) T cells, which is indicative of positive selection in the thymus. Their CD4(+) cells were unresponsive to self H2-M-deficient antigen-presenting cells (APCs) but were hyperreactive to wild-type APCs. H2-M-deficient APCs failed to elicit proliferative responses from wild-type T cells.

Clonal elimination accounts for self-tolerance induction in the thymus and also affects mature T cells responding to exogenous antigens in the periphery. Recent evidence on the microenvironments, cell-cell interactions and signalling requirements for clonal deletion of immature and mature T cells is discussed.

Previous studies have shown that in thymectomized hosts exposure of mature T cells to Mlsa (mtv-7) Ag in vivo leads to specific tolerance and the disappearance of Mlsa-reactive V beta 6+ T cells after an initial phase of T cell expansion. To investigate the factors controlling postthymic elimination of mature T cells, we examined T cell responses to Mlsa and other endogenous superantigens in a number of different strain combinations. The results show that the extent of T cell expansion/deletion is influenced by various factors, including the H-2 haplotype of the host and the particular V beta studied. Collectively, the results suggest that the extent of elimination of mature T cells is variable and may be a function of high avidity interactions with APC.

Superantigens stimulate powerful T-cell responses that can have marked effects in vivo, sometimes leading to shock or even death. The demonstration that strong T-cell responses to superantigens in vivo can be followed by tolerance, reflecting either clonal elimination or anergy, has provided important insights into how mature T cells can be regulated. Further progress in understanding the factors that control these responses relies heavily on defining the specific interactions between T-cell receptors, superantigens and major histocompatibility complex molecules which lead to T-cell activation as well as on the characterization of the specific signal transduction events and molecules involved in this activation. Significant progress has been made, during the past year, in the first area and these findings are summarized below; though less information is available in the latter area, recent observations relevant to this issue are discussed.

The thymus shapes the T cell repertoire by selecting T cells which recognize foreign antigens in association with self major histocompatibility complex (MHC) molecules and by tolerizing those cells which have the potential to react to self antigens.1,2 There is now convincing evidence that negative selection (tolerance induction) of potentially self-reactive cells can occur by a process of clonal deletion.3-5 Direct support for this viewpoint has come from studies with monoclonal antibodies (mAB) specific for particular T cell receptor (TCR) Vβ gene products. Thus, it has been found that mature T cells bearing Vβ11, Vβ17, and Vβ5-positive TCRs are selectively deleted in I-E-positive mouse strains, presumably because these TCRs have high reactivity for I-E molecules (perhaps complexed with self peptides).6–8 Similar results are found for Mls antigens, a set of poorly characterized self antigens which have the unusual property of stimulating very high primary mixed lymphocyte responses (MLR).9,10 Thus, Mlsa-positive mouse strains delete Vβ6 and Vβ8.1-positive T cells and Mlsc-positive strains delete T cells bearing Vβ3.11–14

Antigen-specific tolerance of T cells to minor lymphocyte stimulatory (Mls) antigens can be induced in mice by neonatal injection of foreign lymphohematopoietic cells. Although immune responses to Mls a antigens are controlled by B cells, CD8 + T cells were the most effective cell type for induction of Mls a tolerance. Tolerance was evident in both thymus and lymph nodes and could be induced by as few as 2 × 10 4 CD8 + T cells; these cells were 50 to 100 times as potent as CD4 + cells or B cells in causing functional tolerance and deletion of V β 6 + T cells. Thus, intrathymic contact with antigens expressed on CD8 + T cells may play an important role in controlling the normal development of tolerance.

To attempt to resolve the controversy on the role of thymic epithelial cells (TEC) in tolerance induction, athymic mice were grafted with allogeneic day‐14 fetal thymuses treated with deoxyguanosine in vitro . The data indicate that the tolerogenicity of TEC varies considerably according to the antigen and the subpopulation of T cells studied. For cytotoxic CD8 + cells responding to H‐2 class I antigens, TEC induce minimal tolerance. For proliferative responses of CD4 + cells, by contrast, TEC induce significant tolerance to H‐2 class II antigens but no detectable tolerance to Mls 3 antigens.

The specificity of mature CD8+ and CD4+ T lymphocytes is controlled by major histocompatibility complex (MHC) class I and class II molecules, respectively. The MHC class specificity of T cells is stringent in many assays, but is less evident when cells are supplemented with exogenous lymphokines. The repertoire of T cells is shaped through contact with MHC molecules in the thymus and involves a complex process of positive selection and negative selection (tolerance). Tolerance of immature T cells to MHC molecules can reflect either clonal deletion or anergy and results from intrathymic contact with several cell types, including epithelial cells and cells with antigen-presenting function. Unlike immature T cells, mature T cells are relatively resistant to tolerance induction. In certain situations partial unresponsiveness of mature T cells can be achieved by exposing T cells to foreign MHC molecules expressed on atypical antigen-presenting cells. Tolerance is rarely complete, however, and the precise requirements for tolerizing mature T cells are still unclear.