Impaired CD28-mediated co-stimulation in anergic T cells

We have investigated a CD28 co-stimulation in anergic T cellsin staphylococcal enterotoxin Binoculated mice by stimulatingthe cells with a plate-coated anti-TCR antibody in the presenceor absence of an anti-CD28 antibody. CD28 co-stimulation increasedthe levels of IL-2 and IL-4 mRNAs in nalve CD4⁺V_ß8⁺T cells. However, it did not increase the levels of IL-4 mRNAat all and only partially increased those of IL-2 mRNA in anergicT cells. It was demonstrated that CD28 co-stimulation was impairedso that it no longer stabilized cytoklne mRNAs in anergic cells.The levels of IL-4 mRNA in response to TCR stimulation werehigher in anergic T cells than those in nalve T cells in spiteof the defective CD28 co-stimulation in the former cells. Anergyinduction and generation of a T_h2-type immune response in vivoare discussed     

CD28 and T cell co-stimulation

Over the last decade the concept of T cell co-stimulation has emerged to take a central role in the process of T cell activation. However, the exact definition of co-stimulation is still unclear. In this review, we re-examine the concept of co-stimulation. We suggest that while co-stimulation is important, there is little evidence to link co-stimulation with T cell anergy. We then suggest a framework for studying co-stimulation. Focusing on recent advances in our understanding of CD28, we discuss four areas of T cell activation where co-stimulation may play a role.     

Association of CD40 ligand expression on HTLV-I-infected T cells and maturation of dendritic cells

Human T lymphotropic virus type I (HTLV-I) induces HTLV-I-associated myelopathy/tropical spastic paraparesis (HAM/TSP) and adult T-cell leukemia (ATL). The development of HAM/TSP is associated with rapid maturation of dendritic cells (DCs), while ATL is accomplished with their maturation defect. The DC maturation is induced by cell-to-cell contact with CD4+ T cells expressing CD40 ligand (L). We determined the influence of CD40L expressed on various HTLV-I-infected T cells on the DC maturation. Around 60% of CD4+ T cells infected with HTLV-I for 1 week, expressed CD40L molecules involved in DC maturation. DCs matured by the CD40L+ T cells activated autologous CD4+ and CD8+ T cells. HTLV-I-immortalized T-cell lines established from healthy donors consistently expressed CD40L molecules for 3 months, however, some lines lost the expression soon thereafter. Interleukin (IL)-2-independent and transformed lines lacked that expression. Furthermore, T cells obtained from HAM/TSP patients expressed CD40L molecules for at least 3 weeks, whereas T cells from ATL patients did not express that. The CD40L T cells did not induce DC maturation, and required exogenous CD40L molecules for maturation. The CD40L+ T-cell-induced maturation was blocked by anti-CD40L antibody. Therefore, the lack of CD40L expression on HTLV-I-infected T cells may be associated with the development of ATL.     

IL-10 directly acts on T cells by specifically altering the CD28 co-stimulation pathway

IL-10 induces T cell anergy in numerous mouse models and specific immunotherapy of allergy in humans. Here, we demonstrate that IL-10 directly acts on T cells which are stimulated via CD28 by efficiently blocking proliferation and cytokine production. T cells tolerized by IL-10 showed high viability and the unresponsive state was reversed by anti-CD3 monoclonal antibody (mAb) stimulation and IL-2, but not by anti-CD28 mAb stimulation. Signal transduction via CD28 requires CD28 tyrosine phosphorylation and binding of phosphatidylinositol 3-kinase. IL-10 inhibited tyrosine phosphorylation of CD28; thus, the phosphatidylinositol 3-kinase binding to CD28 was blocked. Consequently, IL-10 inhibited the antigen-induced secretion of both Th1 and Th2 cytokines, including IL-2, IFN-gamma, IL-4, IL-5 and IL-13. Furthermore, neutralization of endogenously produced IL-10 significantly increased T cell proliferation and both Th1 and Th2 cytokine production in vitro. Using superantigen stimuli, T cell suppression by IL-10 was merely induced at low doses when co-stimulation by CD28 was essential. Together, these data demonstrate that IL-10 directly acts on the CD28 signaling pathway and this represents an important T cell suppression mechanism leading to anergy.     

Role of CD28 co-stimulation in generation and maintenance of virus-specific T cells

Efficient induction of T cell responses is normally assumed to require both TCR-mediated signaling and engagement of co-stimulatory molecules, in particular CD28. However, the importance of CD28 co-stimulation in induction and maintenance of antiviral T cell responses is not clearly established. For this reason antiviral CD4(+) and CD8(+) T cell responses in CD28-deficient mice were studied using two different viruses [vesicular stomatitis virus and lymphocytic choriomeningitis virus (LCMV)]. Intracellular cytokine staining and/or MHC-peptide tetramers were used to enumerate antigen-specific T cells. In addition, we used DNA constructs encoding viral epitopes to probe the importance of the epitope itself. Our results reveal that while the context of antigen presentation (live virus versus DNA construct) is a critical factor in determining the requirement for CD28 co-stimulation, epitope and virus dose play little if any role. Direct visualization of antigen-specific cells also confirms the notion that CD28 is more critical for the generation of antiviral T(h)1 cells than for T(c)1 cells generated in response to the same virus (LCMV). Most importantly, the present study reveals that CD28 generally is essential for the host to respond optimally over a broad set of conditions, and our results may imply that the relatively CD28 independent activation of LCMV-specific CD8(+) T cells may represent an extreme situation related to the non-cytolytic nature of this virus allowing the delivery of a uniquely strong and prolonged signal 1.     

Positive and negative regulation of IL-12 receptor expression of naive CD4(+) T cells by CD28/CD152 co-stimulation

IL-12 is a critical cytokine for polarizing naive CD4(+) T cells toward Th1 subset. Therefore, it is important to elucidate the mechanism of IL-12R expression of naive CD4(+) T cells. In this report, we present evidence to show that expression of both IL-12Rbeta1 and beta2 mRNA is regulated by signals mediated through CD28 and CD152. Naive CD4(+) T cells stimulated with anti-CD3 alone neither expressed IL-12Rbeta2 mRNA nor bound detectable level of rIL-12, although they expressed a very low level of IL-12Rbeta1 mRNA when stimulated with a high dose of anti-CD3. Expression of IL-12Rbeta1 and beta2 mRNA was induced by the co-ligation of CD3 and CD28, and it was down-regulated by the ligation of CD152. CD28 ligation induced not only IL-12Rbeta1 and beta2 mRNA expression, but also enhanced IFN-gammaR to mediate up-regulation of IL-12R by IFN-gamma.     

B cells regulate expression of CD40 ligand on activated T cells by lowering the mRNA level and through the release of soluble CD40

The expression of CD40 ligand (CD40L) on activated T cells (CD4⁺ T cell clone MT9) is diminished when the T cells are cultured in the presence of B cells. This effect, observed both with normal tonsil B cells and with the B cell line JY, was detected after 6 h and sustained at least until 18 h of co-culture. Analysis of mRNA showed that CD40L mRNA levels were not modified after 6 h, but were significantly down-regulated after 18 h of co-culture with B cells. Although CD40L expression could not be detected by a CD40-Fc chimera, the molecule was still expressed at the membrane as shown with a polyclonal antiserum against CD40L (anti-TRAP). In addition, T cells activated in the presence of B cells were stained by a polyclonal antiserum against CD40, without the appearance of CD40 mRNA. These results indicated that a soluble form of CD40 (sCD40) bound to the expressed CD40L on T cells. The existence of sCD40 was confirmed by detection of sCD40 in B cell supernatants using a specific enzyme-linked immunosorbent assay. Collectively, these data show that B cells can regulate the expression of CD40L on activated T cells at least by two different mechanisms.     

Requirement for CD28 co-stimulation is lower in SHP-1-deficient T cells.

This study provides biochemical and functional evidence pertaining to the role of the intracellular protein tyrosine phosphatase, SHP-1, in influencing thresholds for TCR activation. Although the loss of SHP-1 in thymocytes from motheaten mice had minimal effects on the initial rise of cytosolic Ca(2+) concentration following TCR triggering, the post-stimulation equilibrium levels of Ca(2+) were consistently elevated. In keeping with a SHP-1 effect on PLCgamma function, IP3 generation was increased in SHP-1 deficient thymocytes. Importantly, we demonstrate that loss of SHP-1 results in a relaxation of the normally stringent co-stimulatory requirements for IL-2 production. SHP-1 deficient single-positive CD4(+) thymocytes revealed a significantly enhanced capacity to produce IL-2 in response to anti-CD3 stimulation alone. In contrast, the simultaneous triggering of CD3 and CD28 was required for equivalent IL-2 production in control single-positive CD4(+) thymocytes. Furthermore, SHP-1 deficient thymocytes generated an increased and prolonged proliferative response to anti-CD3 stimulation alone. In addition, the simultaneous triggering of CD28 and CD3 resulted in equivalent proliferative responses in SHP-1-deficient and control thymocytes, suggesting that a strong co-stimulatory signal is able to override the effect of SHP-1 loss on TCR hyperresponsiveness. Collectively, these results suggest that SHP-1, rather than acting directly on TCR signaling, may indirectly raise thresholds for TCR triggering by modulating co-stimulatory signals.     

CD40 expressed on human melanoma cells mediates T cell co-stimulation and tumor cell growth

CD40 is a 50 kDa molecule, a member of the tumor necrosis factor/nerve growth factor receptor family. It is expressed on B cells, monocytes, dendritic cells and various malignant cells. While the critical relevance of this molecule in T cell-dependent B cell activation is already established, the biological role of CD40-CD154 interaction in non-hematopoietic cells is still unknown. Here we show that CD40 is functionally expressed on human melanoma-derived cell lines. No correlation between surface CD40 expression and the origin of the cell line, primary versus metastatic, was observed. Melanoma cells were shown to be able to co-stimulate TCR-triggered human T cells; moreover, because they do not express CD80 or CD86 co-stimulatory structures, the involvement of additional pathways have to be postulated. We have identified CD40 as one of the molecules involved in melanoma cell-mediated co-stimulation of anti-CD3-triggered human CD4(+) T lymphocytes. In addition, a CD40-dependent pathway, able to enhance tumor cell proliferation at low serum concentrations, in vitro, has been shown to be functional in human melanoma cell lines.     

CD40/CD40L interaction is essential for the induction of EAE in the absence of CD28-mediated co-stimulation

CD28 provides a co-stimulatory signal critical for optimal T cell activation. We and others have shown that the B7/CD28 co-stimulatory pathway is a major regulatory pathway for the control of immune responses. Experimentally induced models of autoimmunity have been shown to be prevented or reduced in intensity in mice deficient for CD28. Here, we show that EAE and accompanying neuroantigen-specific immune responses are drastically reduced in the absence of CD28. However, we go on to show that EAE can be induced in CD28-deficient mice following two immunizations. After re-immunization, CD28-deficient mice develop severe EAE with myelin-specific responses equal to those of wildtype controls, and extensive demyelination in the spinal cord. Treatment of CD28-deficient mice with anti-CD40L at the time of immunization significantly reduced DTH responses and prevented the development of EAE following two immunizations, indicating a critical role for CD40/CD40L signaling in the absence of CD28. Taken together, our results indicate that CD28-mediated co-stimulation does not regulate immunological anergy. Instead, CD28 appears to adjust the threshold for activation and expansion of autoreactive cells.     

Requirements for the functional expression of OX40 ligand on human activated CD4+ and CD8+ T cells

Interaction between OX40 expressed on activated T cells and its ligand (OX40L) on antigen presenting cells (APC) provides a co-stimulatory signal for T cells to promote acquired immunity. In the present study, we have examined various culture conditions for optimum OX40L expression on T cells stimulated with immobilized anti-CD3/CD28 monoclonal antibodies (mAbs). Although the day 3 primed T cells expressed minimal OX40L, after repeated stimulations both the CD4+ and CD8+ T cells became OX40L positive as determined by flow cytometry. Interleukin (IL)-12 interfered with the OX40L expression. Among activated T cells, a higher frequency of CD8+ T cells expressed OX40L than CD4+ T cells. By blocking OX40L-OX40 interaction by an anti-OX40 mAb, the number of OX40L+ T cells significantly increased. Screening of various cytokines showed that transforming growth factor (TGF)-beta1 was capable of induction of OX40L on the activated T cells within 3 days. The OX40L expressed on T cells was functional, as they bound soluble OX40 and stimulated human immunodeficiency virus-1 (HIV-1) production from cell lines chronically infected with HIV-1 and expressing OX40. Altogether the present study findings indicate that functional OX40L is inducible on human activated CD4+ and CD8+ T cells, and that the expression is enhanced by TGF-beta1.     

CD80/CD28 co-stimulation in human brucellosis

Despite treatment, 10-30% of brucellosis patients develop chronic disease, characterized by atypical clinical picture and/or relapses. A defective T helper 1 (Th1) response and a low [corrected] percentage of CD4(+)/CD25(+) cells have been described in chronic brucellosis patients. CD80/CD28 co-stimulation is critical for an efficient Th1 response and has not been studied previously in human brucellosis. In order to investigate the role of CD80/CD28 co-stimulation, 13 acute brucellosis patients (AB), 22 chronic brucellosis patients (CB, 12/22 relapsing type-CB1 and 10/22 atypical type-CB2), 11 'cured' subjects and 15 healthy volunteers (controls) were studied. The percentage of CD4(+)/CD28(+) T lymphocytes and CD14(+)/CD80(+) monocytes were analysed by flow cytometry both ex vivo and after phytohaemagglutinin (PHA)-stimulation with or without heat-killed Brucella abortus (HkBA). Ex vivo analysis showed no differences between all groups studied. PHA stimulation up-regulated the percentage of CD80(+) monocytes in AB compared to 'cured' subjects and controls (P < 0.001), although the proportion of CD4(+)/CD28(+) cells did not alter. A higher percentage of CD80(+) monocytes was observed in the CB1 subgroup, compared to AB, 'cured' subjects and controls (P = 0.042, < 0.001 and < 0.001, respectively). CB2 was characterized by a lower percentage of CD80(+) monocytes in comparison to CB1 (P = 0.020). HkBA in PHA cultures down-regulated the percentage of CD80(+) monocytes compared to PHA alone in all groups, especially in AB and CB patients (P < 0.001 and P = 0.007, respectively). In conclusion, the diminished percentage of CD4(+)/CD25(+) T cells in CB is not associated with inadequate CD80/CD28 co-stimulation. We speculate that differential frequency of CD80(+) monocytes after PHA stimulation could serve as a qualitative parameter of disease status, related to the different clinical forms of chronic brucellosis.     

Regulation of CD28 expression on CD8+ T cells by CTLA-4

CD28 and CTLA-4 are the critical costimulatory receptors that predominantly determine the outcome of T cell stimulation, with CD28 promoting positive costimulation and CTLA-4 inducing inhibitory signals. Blockage of the B7-CD28/CTLA-4 pathway leads to transplantation tolerance. However, the exact mechanism of the inhibitory function of CTLA-4 remains elusive. Here, we investigated the influence of CTLA-4 expression on CD28 using CTLA-4-transfected Jurkat T cells as well as primary T cells. Up-regulation of CTLA-4 induced abrogation of IL-2 production, indicating an anergic phenotype of CTLA-4(high) T cells. Besides the negative signaling function of CTLA-4, we show for the first time that CTLA-4 expression promotes the down-regulation of CD28 on the T cell surface as a result of enhanced internalization and degradation of CD28. These data suggest that apart from the established competition for B7.1 and B7.2 by CTLA-4, inhibition of T cells by CTLA-4 might be additionally explained by reduction of CD28 on the cell surface, which might impede T cell response to stimulation. Our data provide a previously unrecognized mechanism for T cell regulation.     

Endothelial cells augment the expression of CD40 ligand on newly activated human CD4+ T cells through a CD2/LFA-3 signaling pathway

Cultured human endothelial cells (EC) increase CD40 ligand expression on polyclonally activated human peripheral blood CD4⁺ helper T cells compared to T cells activated in the absence of accessory cells or in the presence of peripheral blood adherent cells or B cells. Induction of CD40 ligand expression appears to be biphasic with early induction observable at 6 h and later induction observable at 24 h. EC cause T cells to increase CD40 ligand expression during the early phase at 6 h after activation. CD40 ligand expression is restricted to the CD4⁺ helper T cell subset of the peripheral blood T cells, even when EC are present. Blocking monoclonal antibodies to co-stimulatory molecules on EC and T cells indicate that the CD2/LFA-3 pathway, which also contributes to induction of augmented interleukin-2 (IL-2) secretion, is involved in EC-induced upregulation of CD40 ligand. Exogenous IL-2 can also increase CD40 ligand expression. However, increased IL-2 secretion in the presence of EC can not fully account for endothelial-induced CD40 ligand up-regulation as (1) the effect of exogenous IL-2 is greater at 24 h than at 6 h, whereas the opposite is true for EC; (2) the effect of saturating levels of IL-2 is considerably smaller than that of EC; and (3) blocking of IL-2 receptors does not fully inhibit endothelial effects on CD40 ligand expression. We conclude that EC provide unique co-stimulatory signals that affect the phenotype of activated CD4⁺ T cells.     

Requirement of CD28-CD86 co-stimulation in the interaction between antigen-primed T helper type 2 and B cells

The interaction between CD28 and its ligands, CD80 and CD86, is crucial for an optimal activation of antigen-specific T cells. However, the requirement of CD80 or CD86 co-stimulation in Th2 cell differentiation and activation is controversial. Freshly isolated murine CD4+ and CD8+ T cells were incubated with P815 transfectants expressing a similar level of either CD80 or CD86 in the presence of anti-CD3 mAb. Both CD80 and CD86 co-stimulated the proliferation of CD4+ and CD8+ T cells at comparable time-kinetics and magnitude, but CD86 alone was able to co- stimulate IL-4 and especially IL-10 production in CD4+ T cells. In typical Th2-dependent immune responses elicited by Nippostrongylus brasillensis infection, the anti-CD86 mAb treatment but not the anti- CD80 mAb treatment efficiently inhibited antigen-specific IgE and IgG1 production, which was accompanied with the reduced IL-4 production. Our results suggest that CD86 co-stimulation plays a dominant role not only in the primary activation of Th2 cells but also in the secondary interaction between antigen-primed Th2 cells and B cells.      

Impaired expression of CD28 on T cells in hairy cell leukemia

T cells from patients with active hairy cell leukemia (HCL), a chronic B cell malignancy, show poor proliferation in response to allogeneic peripheral blood mononuclear cells (PBMC). In order to study the T cell dysfunction, the expression of several adhesion and costimulatory molecules was analyzed by flow cytometry. Circulating T cells from HCL patients showed increased percentages of CD28(-) in all T cell subsets. In some patients the percentage of CD28(-) T cells within the CD4(+) subset was increased up to 80%. These CD4(+)CD28(-) T cells did not proliferate in a mixed lymphocyte culture (MLC) against allogeneic PBMC. After enrichment for CD4(+)CD28(+) T cells, the proliferative response in the MLC was recovered, but this response was still lower than the proliferative response from control T cells. In conclusion, lack of CD28 on T cells and a restricted T cell repertoire may contribute to immune deficiency in patients with HCL.     

Ligation of B7 with CD28/CTLA-4 on T cells results in CD40 ligand expression,interleukin-4 secretion and efficient help for antibody production by B cells

It has been extensively shown that when T cells are co-stimulated with B7-CD28 interaction, a strong proliferative as well as cytolytic T cell response can be induced. In contrast, there exists only indirect evidence that the B7-CD28 interaction is of importance for the induction of T cell helper functions in B cell responses. Here we have used mouse fibroblasts transfected with the human Fcγ receptor type II and human B7 to address this issue. We found that T cells, when activated through the T cell receptor (TcR)/CD3 complex with monoclonal antibodies and co-stimulated by B7-CD28 interaction, can provide efficient help for the induction of both IgM and IgG production by resting B cells. This helper activity is, at least in part, mediated by the interaction between the CD40 ligand on the T cells and CD40 on the B cells. We also demonstrate that more than one signal to the T cell is required for the induction of the CD40 ligand, one being delivered through the TcR/CD3 complex and the second by ligation of CD28 with the B7 molecule. In addition to the induction of cognate T helper function, we provide evidence that co-stimulation of T cells with B7-CD28 interaction can result in the secretion of both Th1- and Th2-type lymphokines.