Elimination of lymphocytes, but not eosinophils, by Fas-mediated apoptosis in murine schistosomiasis.

Infection with the helminth parasite Schistosoma mansoni is associated with a pathogenic granulomatous response to parasite eggs. Multiple cell types constitute the granuloma with eosinophils achieving numerical dominance. We hypothesize that eosinophil dominance is achieved by selective apoptosis in lymphocytes. We report here that lymphocytes from both the spleens and granulomas of S. mansoni-infected mice undergo apoptosis. We also show that granuloma lymphocytes are more susceptible to Fas-FasL-mediated apoptosis than spleen lymphocytes and this apoptosis may be related to antigen concentration. Conversely, eosinophils from the granuloma and spleens of S. mansoni-infected mice are resistant to apoptosis in vivo and are protected in vitro from Fas-FasL-mediated apoptosis by the absence of FasL expression in the presence of Fas expression. Finally, the apoptotic regulatory molecules Bcl-2, Bcl-xL, and Bax, do not appear to play a significant role in the regulation of eosinophil apoptosis in the schistosome granuloma.

The development of a Th1-type response and resistance to Leishmania major infection in the absence of CD40-CD40L costimulation.

CD40-CD40L interactions have been shown to be essential for the production of IL-12 and IFN-gamma and control of L. major infection. In contrast, C57BL/6 mice deficient in CD28 develop a dominant Th1-type response and heal infection. In this study, we investigate the effects of a deficiency in both CD40L and CD28 molecules on the immune response and the course of L. major infection. We compared infection in mice genetically lacking CD40L (CD40L(-/-)), CD28 (CD28(-/-)), or both (CD40L(-/-)CD28(-/-)), and in C57BL/6 mice, all on a resistant background. Although CD40L(-/-) mice failed to control infection, CD28(-/-) and CD40L(-/-)CD28(-/-) mice, as well as C57BL/6 mice, spontaneously resolved their infections. Healing mice had reduced numbers of lesion parasites compared with nonhealing CD40L(-/-) mice. At wk 9 of infection, we detected similar levels of IL-4, IFN-gamma, IL-12p40, and IL-12Rbeta2 mRNA in draining lymph nodes of healing C57BL/6, CD28(-/-), and CD40L(-/-)CD28(-/-) mice, whereas CD40L(-/-) mice had increased mRNA levels for IL-4 but reduced levels for IFN-gamma, IL-12p40, and IL-12Rbeta2. In a separate experiment, blocking of the CD40-CD40L pathway using Ab to CD40L led to an exacerbation of infection in C57BL/6 mice, but had little or no effect on infection in CD28(-/-) mice. Together, these results demonstrate that in the absence of CD28 costimulation, CD40-CD40L interaction is not required for the development of a protective Th1-type response. The expression of IL-12p40, IL-12Rbeta2, and IFN-gamma in CD40L(-/-)CD28(-/-) mice further suggests the presence of an additional stimulus capable of regulating IL-12 and its receptors in absence of CD40-CD40L interactions.

Costimulation of memory T-cells by ICOS: a potential therapeutic target for autoimmunity?

Approaches that target costimulatory receptors are independent of T-cell receptor specificity and may be useful for T-cell-mediated diseases in which the antigens involved are not well defined. However, the proper costimulatory pathways need to be targeted. For example, therapies for human T-cell-mediated diseases need to be effective against previously activated memory cells. In this review, we use autoimmune demyelination as a paradigm for established immune-mediated pathogenesis. Studies with the human disease multiple sclerosis and the rodent model experimental autoimmune encephalomyelitis have suggested that the effectiveness of CD28 blockade, as a therapeutic strategy for established autoimmune demyelination, may be limited. ICOS, a receptor that appears to be involved in the costimulation of previously activated T-cells, may be an attractive alternative.

ICOS ligand costimulation is required for T-cell encephalitogenicity.

The interaction of ICOS with its ligand on APC provides a costimulatory signal to previously activated T-cells. In these studies, we blocked the ICOS:ICOS ligand interaction with ICOS-Ig during the in vitro activation of MBP-reactive transgenic CD4(+) T-cells. The presence of ICOS-Ig in these cultures inhibited the ability of the transgenic T-cells to transfer EAE, although they entered the brains of the recipient mice. ICOS-Ig increased apoptosis in the transgenic T-cells, especially in the memory population. This enhanced apoptosis was accompanied by an increase in the BAX/BCL-2 mRNA ratio. ICOS-Ig did not prevent IL2 production, demonstrating that IL-2 production is ICOS ligand independent. IFN-gamma and IL-10 production by the transgenic T-cells, however, was suppressed. Finally, ICOS-Ig injection into mice after the first signs of EAE ameliorated clinical disease. Therefore, ICOSL provides a signal distinct from CD28 costimulation that is required for the activation and viability of encephalitogenic T-cells.

The role of costimulation in autoimmune demyelination.

Experimental allergic encephalomyelitis (EAE) is a T cell-mediated, autoimmune disorder characterized by central nervous system (CNS) inflammation and demyelination, features reminiscent of the human disease, multiple sclerosis (MS). In addition to the signal the encephalitogenic T cell receives through the T cell receptor (TCR), a second signal, termed costimulation, is required for complete T cell activation. The B7 family of cell surface molecules expressed on antigen presenting cells (APC) is capable of providing this second signal to T cells via two receptors, CD28 and CTLA-4. Our studies have shown that costimulation provided by B7 molecules to its ligand CD28 is important in the initiation of the autoimmune response in EAE. Further, it appears the costimulation provided by B7-1 is important in disease development, while B7-2 may play an important regulatory role. We and others later showed that B7/CTLA-4 interaction plays a critical role in down-regulating the immune response. Previous work has shown that activated T cells and T cells of a memory phenotype are less dependent on costimulation than naive T cells. T cells reactive with myelin components that are involved in the pathogenesis of EAE and possibly MS would be expected to have been activated as part of the disease process. Building upon our prior work in the EAE model, we have tested the hypothesis that myelin-reactive T cells, which are relevant to the pathogenesis of CNS inflammatory demyelination, can be distinguished from naive myelin-reactive T cells by a lack of dependence upon costimulation for activation and that the costimulatory requirements of these myelin-reactive T cells change during the course of disease. Our studies in the EAE model have also addressed the mechanisms of extrathymic (peripheral) T cell tolerance following intravenous (i.v. ) administration of high dose antigen. It is believed that TCR signaling in the absence of costimulation is a vital component of peripheral tolerance mechanisms. However, recent evidence suggests that peripheral tolerance of antigen-specific T cells induced in vivo may require CTLA-4 engagement of the tolerized T cells. We have begun to examine the molecular mechanisms of tolerance induction following intravenous and intraperitoneal administration of myelin antigens in the EAE model and test the hypothesis that tolerance induction is dependent on the B7:CD28/CTLA-4 pathway. The results from our studies will enhance our understanding of the role that myelin-reactive T cells may play in the pathogenesis of MS. We have determined that MBP-reactive T cells in MS patients are less dependent upon CD28 costimulation than in normal controls, suggesting that these T cells were previously primed in vivo. Characterization of these CD28-independent myelin-specific T cells will have broad implications for a variety of immunologically based therapies in diseases such as MS.

Differential cytokine and chemokine production characterizes experimental autoimmune meningitis and experimental autoimmune encephalomyelitis.

After primary immunization with myelin/oligodendrocyte glycoprotein, CD28(-/-) mice developed experimental autoimmune meningitis (EAM) rather than experimental autoimmune encephalomyelitis (EAE). Cytokine and chemokine production in EAE and EAM were compared to understand the differences in disease phenotype. T cells from the central nervous system lesions of mice with either EAE or EAM expressed intracellular TNF-alpha. Splenic T cells from mice with EAM produced TNF-alpha and IL-6 but no IL-2. Conversely, EAE-derived splenic T cells produced TNF-alpha and IL-2 but no IL-6. Altered T cell differentiation in EAM was not due to a Th1 to Th2 shift, because equivalent amounts of T cell IFN-gamma mRNA were produced in both diseases. Neutrophils also produced inflammatory mediators such as TNF-alpha and IL-6 in EAM. Autocrine production of MIP-2 mRNA was observed in neutrophils from mice with EAM but not EAE. Therefore, distinct patterns of cytokines and chemokines distinguish EAE and EAM.

The role of CTLA-4 in tolerance induction and T cell differentiation in experimental autoimmune encephalomyelitis: i.p. antigen administration.

Recent evidence suggests that co-stimulation provided by B7 molecules through CTLA-4 is important in establishing peripheral tolerance. In the present study, we examined the kinetics of tolerance induction and T cell differentiation following i.p. administration of myelin basic protein (MBP) Ac1-11 in mice transgenic for a TCR V(beta)8.2 gene derived from an encephalitogenic T cell clone specific for MBP Ac1-11. Examination of the lymph node cell response after antigen administration demonstrated a dependence on CTLA-4 for i.p. tolerance induction. Examination of splenocyte responses suggested that i.p. antigen administration induced a T(h)2 response, which was potentiated by anti-CTLA-4 administration. Interestingly, i.p. tolerance was able to inhibit the induction of experimental autoimmune encephalomyelitis and anti-CTLA-4 administration did not alter this phenotype, suggesting that CTLA-4 blockade did not block tolerance induction. Thus, T cell differentiation and the dependence on CTLA-4 for tolerance induction following i.p. antigen administration differs between lymph node and spleen in a model of organ-specific autoimmunity.

The role of CTLA-4 in tolerance induction and ttigen administration cell differentiation in experimental autoimmune encephalomyelitis: i. v. antigen administration.

Interactions between B7 molecules on antigen-presenting cells and CTLA-4 on T cells have been shown to be important in establishing tolerance. In the present study, we examined the kinetics of tolerance induction following i.v. administration of myelin basic protein (MBP) Ac1-11 in mice transgenic for a TCR V(beta)8.2 gene derived from an encephalitogenic T cell clone specific for MBP Ac1-11. Examination of the lymph node cell (LNC) response 10 days after antigen administration demonstrated an accentuation of i.v. tolerance induction with anti-CTLA-4 blockade. Anergy was induced in splenocytes by i.v. antigen administration as shown by a decrease in MBP-specific proliferation and IL-2 production, and anti-CTLA-4 potentiated this effect. In addition, i.v. antigen plus anti-CTLA-4 and complete Freund's adjuvant was not encephalitogenic. Interestingly, i.v. tolerance (a single injection) did not inhibit experimental autoimmune encephalomyelitis (EAE) and anti-CTLA-4 administration did not alter this phenotype. These results suggest that while the majority of MBP-specific T cells are tolerized by i.v. antigen and that this process is potentiated by anti-CTLA-4 administration, a population of T cells remains that is quite efficient in mediating EAE.

Differential requirements of naïve and memory T cells for CD28 costimulation in autoimmune pathogenesis.

Experimental autoimmune encephalomyelitis (EAE) is the most extensively studied animal model of the human disease multiple sclerosis (MS). In EAE, CNS demyelination is induced by immunization with myelin proteins or adoptive transfer of myelin-reactive CD4+ T cells. Since the antigen specificity of the immune response believed to be responsible for the pathology of MS is not well defined, therapies that target aspects of T cell activation that are not antigen specific may be more applicable to the treatment of MS. As a result, understanding the role of costimulatory molecules in the activation of naïve and memory T cells has become an area of extensive investigation. Naïve T cells require two signals for activation. Signal one is provided by engagement of the T cell receptor (TCR) with MHC/peptide complexes and provides antigen specificity to the immune response. The second signal, termed costimulation, is usually provided by B7 molecules on APC to CD28 molecules expressed on T cells and is antigen-independent. This review will discuss our current understanding of costimulation in the induction and perpetuation of EAE, as well as the potential of costimulation blockade in the treatment of MS.

Blockade of CD28 during in vitro activation of encephalitogenic T cells or after disease onset ameliorates experimental autoimmune encephalomyelitis.

Previous studies have shown complex roles for the B7 receptors in providing both positive and negative regulation of experimental autoimmune encephalomyelitis (EAE). B7 blockade can ameliorate clinical EAE by indirectly interfering with CD28 signaling. However, B7 blockade can also result in disease exacerbation, presumably by interfering with regulatory B7:CTLA-4 interactions. Therefore, we have directly targeted T cell CD28 with specific mAbs both during initial Ag priming and after the onset of clinical signs of EAE. We found that CD28 blockade ameliorated EAE during the efferent and afferent limbs of the immune response. Disease amelioration at disease onset was associated with suppression of TNF-alpha production. Finally, Ab blockade of T cell CD28 during the first disease episode resulted in significant attenuation of the subsequent disease course, with no significant relapses. In contrast to previous studies targeting APC B7 with CTLA4-Ig, reagents targeting CD28 can block ongoing disease. Therefore, the present results suggest a clinically relevant therapeutic scenario for human diseases, such as multiple sclerosis.

Susceptibility or resistance to Leishmania infection is dictated by the macrophages evolved under the influence of IL-3 or GM-CSF.

Although enhanced monocytopoiesis is a hallmark of leishmaniasis, its significance in determining the course of the disease has not been addressed. While the number of granulocyte-macrophage colony-stimulating factor (GM-CSF)-secreting cells increases in the draining lymph nodes in a resistant mouse strain (C57BL/6) during disease, in a susceptible strain (BALB/c) the number of interleukin-3 (IL-3)-secreting cells increases. Treatment of BALB/c mice with anti-IL-3 antibody significantly reduces the disease score. Bone marrow macrophages derived under stimulation with IL-3 (IL-3-Mphi) or GM-CSF (GM-Mphi) differ functionally. GM-Mphi are significantly more responsive to IFN-gamma-induced augmentation and more refractory to IL-4-mediated suppression of anti-leishmanial activity than IL-3-Mphi. LPS-induced IL-12 and TNF-alpha secretion by both the susceptible and resistant strain-derived macrophage subsets are down-regulated. Despite down-regulation of IL-12 secretion, GM-Mphi favor expansion of IFN-gamma-secreting cells and IL-3-Mphi favor IL-6-dependent expansion of the IL-4-secreting Th subset. Adoptive transfer of leishmanial antigen-pulsed IL-3-Mphi and GM-Mphi prior to infection either aggravated or reduced the disease score, respectively, in BALB/c mice. Anti-IL-6 treatment reverted the Th subset profile not only in vitro but also in vivo, resulting in a reduced disease score in both infected BALB/c mice and IL-3-Mphi recipients. The disease score in IL-3-Mphi recipients is also reduced significantly after anti-IL-4 treatment.

Experimental autoimmune meningitis: a novel neurological disease in CD28-deficient mice.

C57BL/6 mice develop T-cell-mediated experimental autoimmune encephalomyelitis (EAE) after immunization with the neuroantigen myelin oligodendrocyte glycoprotein. (MOG). We immunized CD28-deficient C57BL/6 mice to determine the role of T cell costimulation in the immune response to MOG. CD28-/- mice developed experimental autoimmune meningitis (EAM). EAM is a fatal, acute disease characterized by simultaneous weakness in all limbs, photophobia, irritability, and spatial disorientation. Histologically, EAM consisted of an infiltrate of myeloid, monocytic, and lymphocytic leukocytes within the leptomeninges. In contrast, the brain parenchyma was unaffected. EAM was mediated by CD4+ T cells since CD4 depletion prevented the disease. Upon rechallenge, mice in which EAM was prevented by CD4+ cell depletion developed EAE not EAM. Therefore, the presence or absence of CD28 determines the initial phenotype of the immune response to MOG. EAM, which develops in the absence of CD28, is a unique experimental model for immune-mediated aseptic meningitis.

Activated eosinophils are the major source of Th2-associated cytokines in the schistosome granuloma.

Eosinophils are a numerically dominant cell population within the schistosome granuloma. These granuloma eosinophils can produce a variety of cytokines, including IL-2, IL-4, IL-5, and IFN-gamma. Therefore, eosinophils may play a key role in the determination of the unique cytokine microenvironment within the granuloma milieu. These studies investigated the potential role of eosinophils in the regulation of granuloma immunopathology. We have characterized spleen- and granuloma-derived eosinophils based on cellular activation and cytokine production during the development of murine schistosomiasis. Based on the criteria of hypodensity and CD69 expression, granuloma eosinophils were highly activated and very homogeneous at 7 and 11 wk postinfection. Splenic eosinophils were also activated at 7 wk postinfection, but were much more heterogeneous than their granuloma counterparts. By 11 wk postinfection, few hypodense splenic eosinophils were observed. Eosinophils represented the majority of cytokine-producing cells in the granuloma and were a dominant source of IL-4. Eosinophils also produced IL-2, IL-5, and IFN-gamma, using the criteria of mRNA in situ hybridization and intracellular cytokine staining by FACS. Granuloma eosinophil activation and cytokine production were greatest at the time of maximum granuloma formation, i.e., 10-12 wk after initial cercarial exposure. Therefore, locally activated eosinophils, not Th2 lymphocytes, produce the majority of Th2 cytokines in the granuloma milieu and may be important determinators of immunopathology in schistosomiasis.

CTLA4 (CD152) modulates the Th subset response and alters the course of experimental Leishmania major infection.

Since both the nature and the amplitude of an antigen-specific T cell response are dependent on co-stimulatory signals, we have investigated the role of CD28/CD152-mediated T cell co-stimulation in the regulation of experimental cutaneous leishmaniasis. CD28-deficient mice and their wild-type littermates are equally susceptible to Leishmania major infection. Whole anti-CD152 antibody significantly exacerbates the disease while anti-CD152 Fab ameliorates the disease in genetically susceptible BALB/c mice but not in C57BL/6, a resistant strain. The anti-CD152-induced exacerbation of the disease is accompanied by increased IL-4-secreting cell number, diminished parasite-specific delayed-type hypersensitivity (DTH) response and augmented anti-2,4,6-trinitrophenyl (TNP) IgG1 in response to TNP-leishmanial antigen crude soluble antigen (CSA), suggesting an exaggerated Th2 type of response. Anti-CD152 Fab-mediated amelioration of the disease is associated with increased IFN-gamma-secreting cell number, increased parasite-specific DTH response and enhanced IgG2a isotype in response to TNP-CSA suggesting a Th1 type of response. Unlike TNP-CSA, TNP-keyhole limpet hemocyanin does not induce the change in Ig isotype, indicating that the immunomodulatory effect of anti-CD152 is antigen specific. Anti-CD152 antibody-induced early change in Th subsets suggests an important role for CD152 in determining the course of L. major infection, perhaps by alteration of Th subset differentiation.

The schistosome granuloma: characterization of lymphocyte migration, activation, and cytokine production.

Granuloma formation and its regulation are dependent on lymphocytes. Therefore, we compared the characteristics of lymphocytes derived from the spleens and granulomas of Schistosoma mansoni-infected mice during the course of their disease. We examined lymphocyte cell cycle kinetics, migration, expression of activation Ags (CD69 and IL-2R), cytokine production (IL-2, IL-4, IFN-gamma), and apoptosis. Lymphocytes in the G2/M phase of the cell cycle and high levels of lymphocyte intracellular IL-2 were found in the spleen but not in the granuloma. Cell trafficking experiments showed Ag-specific recruitment of schistosomal egg Ag (SEA)-reactive lymphoblasts into granulomas in vivo, as well as recruitment to, residence within, and egress from granulomas in vitro. Granuloma-derived lymphocytes were more highly activated than splenic lymphocytes based on higher levels of CD69 and IL-2R expression. While the granuloma microenvironment was rich in Th2 cytokines, during peak granuloma formation, the lymphocytes per se from the spleen and granuloma did not exhibit a dominant Th1 or Th2 cytokine profile, producing low but similar levels of IL-4 and IFN-gamma. The discrepancy between high IL-2R expression and low levels of IL-2 protein production by granuloma lymphocytes was associated with increased apoptosis in the granuloma compared with the spleen. These findings support the hypothesis that granulomas may play a role in the regulation of systemic pathology in schistosomiasis by adversely affecting the survival of SEA-reactive, immunopathogenic T lymphocytes.

Decreased dependence of myelin basic protein-reactive T cells on CD28-mediated costimulation in multiple sclerosis patients. A marker of activated/memory T cells.

Although multiple sclerosis (MS) patients and healthy individuals have similar frequencies of myelin basic protein (MBP)-specific T cells, the activation state of these cells has not been well characterized. Therefore, we investigated the dependence of MBP-reactive T cells on CD28-mediated costimulation in MS patients, healthy controls, and stroke patients. MBP-reactive T cells from healthy controls and stroke patients failed to proliferate efficiently when costimulation was blocked using anti-CD28, consistent with a naive T cell response. In contrast, MBP-specific T cell proliferation was not inhibited, or was only partially inhibited when CD28-mediated costimulation was blocked in MS patients. Blockade of CD28 failed to inhibit tetanus toxoid-specific T cell proliferation in both the controls and MS patients, demonstrating that memory cells are not dependent on CD28-mediated costimulation. Limiting dilution analysis indicated that the frequency of MBP-reactive T cells was significantly decreased in healthy controls compared with MS patients when CD28-mediated costimulation was blocked. These data suggest that MBP-reactive T cells are more likely to have been activated in vivo and/or differentiated into memory T cells in MS patients compared with controls, indicating that these cells may be participating in the pathogenesis of MS.

Nucleic acid vaccine-induced immune responses require CD28 costimulation and are regulated by CTLA4.

Immunization with plasmids expressing specific genes (DNA or nucleic acid vaccination (NAV)) elicits robust humoral and cell-mediated immune responses. The mechanisms involved in T cell activation by NAV are incompletely characterized. We have examined the costimulatory requirements of NAV. CD28-deficient mice did not mount Ab or CTL responses following i.m. immunization with eukaryotic expression plasmids encoding the bacterial gene beta-galactosidase (beta gal). Because these mice retained their ability to up-regulate the CTLA4 receptor (a negative regulator of T cell activation), we examined CTLA4's role in the response of wild-type BALB/c mice to NAV. Intact anti-CTLA4 mAb but not Fab fragments suppressed the primary humoral response to pCIA/beta gal without affecting recall responses, indicating CTLA4 activation inhibited Ab production but not T cell priming. Blockade of the ligands for CD28 and CTLA4, CD80 (B7-1) and CD86 (B7-2), revealed distinct and nonoverlapping function. Blockade of CD80 at initial immunization completely abrogated primary and secondary Ab responses, whereas blockade of CD86 suppressed primary but not secondary responses. Simultaneous blockade of CD80 + CD86 was less effective at suppressing Ab responses than either alone. Enhancement of costimulation via coinjection of B7-expressing plasmids augmented CTL responses but not Ab responses, and without evidence of Th1 to Th2 skewing. These findings suggest complex and distinct roles for CD28, CTLA4, CD80, and CD86 in T cell costimulation following nucleic acid vaccination.

Effects of blocking B7-1 and B7-2 interactions during a type 2 in vivo immune response.

The costimulatory signal provided to T cells through CD28/CTLA-4 interactions is required for in vivo Th cell effector function associated with cytokine production. However, it is uncertain whether the two well-characterized ligands for these molecules, B7-1 and B7-2, differentially influence the consequent development of a type 1 or a type 2 primary response. We have examined the in vivo effects of blocking B7-1 and/or B7-2 ligand interactions on the type 2 mucosal immune response that follows oral infection of mice with the nematode parasite, Heligmosomoides polygyrus. Administration of the combination of anti-B7-1 and anti-B7-2 Abs inhibited H. polygyrus-induced increases in serum IgG1 and IgE levels, the expansion of mesenteric lymph node (MLN) germinal centers, in situ CD4+ T cell expansion, elevated blood eosinophils, and increased intestinal mucosal mast cells. Similarly, both Abs blocked MLN and Peyer's patch cytokine gene expression and elevations in MLN T cell-derived IL-4 protein secretion. However, in the same experiments, administration of either anti-B7-1 or anti-B7-2 Abs alone had little effect on any of these parameters. T cell and B cell activation was also blocked by the combination of anti-B7-2 and a B7-1-specific mutant Y100F CTLA-4Ig construct. These results suggest that to the extent that anti-B7-1 and anti-B7-2 mAbs block B7 interactions, either B7-1 or B7-2 ligand interactions can provide the required costimulatory signals that lead to T cell effector function during a type 2 in vivo immune response.

Mitogenic stimulation of T cells reveals differing contributions for B7-1 (CD80) and B7-2 (CD86) costimulation.

The requirement of accessory cells for concanavalin A (Con A) activation of T cells suggests delivery of a separate costimulatory signal. However, the costimulatory pathways involved have not been identified. These studies assess the role of CD28-B7-mediated costimulation during T-cell activation by Con A. The B7-1/B7-2 binding protein CTLA4-Ig inhibited the proliferative response of primary lymph node cells to either Con A or soluble anti-CD3 mAb. This suppression was dose dependent and could be reversed by CD28 cross-linking. CTLA4-Ig also completely suppressed induction of interleukin-2 (IL-2) mRNA by Con A. CTLA4-Ig-mediated suppression was not due to blockade of the Con A 'receptor(s)' or of the primary activation signal (as measured by the intracellular calcium response). Although both B7-1 and B7-2 were up-regulated following Con A activation, each played a different role in proliferation and cytokine production. Individually, anti-B7-2 Fab partially inhibited the Con A response whereas anti-B7-1 Fab had no effect. However, the combination of anti-B7-1 and anti-B7-2 Fab completely suppressed proliferation and IL-2 production. Therefore, while a part of the Con A response requires B7-2, the remainder of the response can utilize either B7-1 or B7-2. Together, these results demonstrate that Con A activation of T cells requires the delivery of a separate costimulatory signal that is mediated almost entirely by the B7 receptors.