Search

CD80 (B7-1) and CD86 (B7-2) on APC provide a major costimulatory signal through interactions with CD28 on T cells. Absent from resting human T cells, CD86 is up-regulated early upon T cell activation, whereas CD80 expression appears later. Whereas T cell expression of CD80 has been implicated in costimulation, the functional significance of CD86 expression on T cells is unclear. We now demonstrate that CD86 expressed on human CD4+ T cell clones does not provide a costimulatory signal for other CD4+ T cell clones. Binding studies using CD28-Ig and CTLA-4-Ig fusion proteins demonstrate that CD86 expressed on T cells has significantly reduced binding affinity for CTLA-4 and no detectable binding to CD28. Biochemical analysis demonstrates that post-translational modifications of CD86 in human T cells are different from those of CD86-transfected Chinese hamster ovary cells or EBV-transformed B cells, in that T cells express a hypoglycosylated form of CD86 on the surface membrane. Thus, our results suggest that while CD86 is expressed on a number of different cell types, its costimulatory function and affinity for its ligands may be regulated by cell type-specific post-translational modifications.

Myelin basic protein (MBP)-specific T cells were isolated directly from peripheral blood by stimulation either with native MBPp85-99 or altered peptide ligands (APLs) in which substitutions of the lysine were made at position 93, a TCR contact residue. We report here that the APL 93A could alter the cytokine profile of some autoreactive MBPp85-99 reactive T cell clones, switching them from a Th0 phenotype secreting high concentrations of IL-4, IL-5, and IFN-gamma into Th2 cells secreting significantly less IFN-gamma. However, in vitro stimulation with the 93A peptide, in some instances, induced T cells that responded better to the native MBPp85-99 peptide. Functionally, the APL 93A was shown to act as an antagonist for IFN-gamma secretion. Based on TCR sequencing and single cell cloning, this alteration in cytokine profile was determined to be the result of the differential activation of individual T cell clones. We discuss the implications of these data for the strength of signal model of T cell activation.

The influence of costimulation on the T cell response to altered peptide ligands that act as either partial or weak agonists for human CD4+ T cell clones was examined. Using stable Chinese hamster ovary (CHO) cell transfectants expressing DR2 (DRB1*1501) and human B7-1 or B7-2 as APC, presentation of native myelin basic protein (MBP) p85-99 peptide Ag or a partial agonist of MBP p85-99 induced equivalent T cell activation as measured by [3H]TdR incorporation and cytokine secretion. In marked contrast, presentation of cross-reactive peptides of MBP p85-99 that act as weak agonists with B7-1, but not B7-2, costimulation resulted in significant T cell activation as measured by [3H]TdR incorporation and cytokine secretion. These data suggest that decreasing the strength of the signal provided to the TCR allows differences in B7-1 and B7-2 signaling to be observed. Thus, the costimulatory environment during T cell activation may be a mechanism of regulating T cell cross-reactivity in the periphery.

The B7 family of costimulatory molecules provides the second signal necessary for activation of T cells. In the absence of the second signal, responding T cells become anergic. Although predominantly expressed on professional APCs, recent evidence shows that the B7 molecules are also expressed on T cells. To study the functions of B7 molecules on T cells, we transfected murine B7.1 (CD80) and B7.2 (CD86) cDNAs into the EL4 T cell thymoma cell line and examined the transfectants for their ability to costimulate T cell proliferation in vitro and to induce antitumor immunity in vivo. Here we show that although EL4-B7.1 cells costimulate T cells and induce tumor regression, EL4-B7.2 transfectants failed to costimulate T cell proliferation or induce tumor regression. To understand the cellular basis for this difference, we examined the binding of EL4-B7.1 and EL4-B7.2 to CTLA4 and CD28. Whereas EL4-B7.1 cells bound both CTLA4-Ig and CD28-Ig, EL4-B7.2 transfectants preferentially bound CTLA4-Ig, but not CD28-Ig. Similar binding data were obtained with freshly isolated murine T cells, which have been shown to constitutively express B7.2. Our data suggest, therefore, that B7.2 expressed on T cells may not costimulate but instead inhibit the T cell response by preferential binding to CTLA4.

Two distinct signals are required to activate T cells: an Ag-specific signal and a costimulatory signal mediated primarily by B7-1 (CD80) and B7-2 (CD86) through interactions with CD28. Costimulation appears to be critical in regulating autoreactive T cell responses. Here, we demonstrate that, in contrast to the parental uninfected T cell clone, a T cell clone infected by human T cell lymphotropic virus type I (HTLV-I) displays a remarkably enhanced response to Ag in the absence of B7 costimulation. Chinese hamster ovary cells either transfected with DRB1*1501 (t-DR2) alone or cotransfected with DR2 and either B7-1 or B7-2 were fixed, pulsed with myelin basic protein peptide 84-102 (MBPp84-102), and used as APCs. The MBPp84-102-reactive T cell clone Ob1A12.8 required costimulation with either B7-1 or B7-2 molecules, as the response to Ag was reduced by 90% in the absence of B7 costimulation. However, this requirement for B7 costimulation was abrogated after productive infection by HTLV-I. Stimulation of HTLV-I-infected T cells by MBPp84-102/t-DR2 induced the secretion of IL-5 and IFN-gamma, which approached the level induced in the presence of B7 costimulation, whereas IL-4 was induced to one third of its maximal level. Consistently, the secretion of IL-5 and IFN-gamma was not significantly inhibited by anti-B7-1 and B7-2 Abs, whereas IL-4 was inhibited by approximately 50%. In contrast, uninfected T cells required either B7-1 or B7-2 costimulation for significant cytokine secretion, and this response was inhibited by anti-B7-1 and B7-2 Abs. These findings suggest that HTLV-I-infected autoreactive T cells have the potential to induce an autoimmune response in the absence of B7 expression in the target organ. This may be of particular interest in the elucidation of HTLV-I pathogenicity given the association of HTLV-I infection with autoimmune-like diseases.

A clear differentiation of Th1 and Th2 cytokine-secreting subsets in humans has not yet been defined. To further examine cytokine-directed differentiation of human T cell responses to both exogenous and autoantigens, we generated 346 short term T cell lines at limiting dilutions from six normal individuals to tetanus toxoid and myelin basic protein in the presence of IL-2 with or without the addition of IL-12 and anti-IL-4 mAb. T cell lines were examined for [3H]thymidine incorporation and cytokine secretion of IFN-gamma, IL-4, and IL-10. After culture in the presence of IL-12 and anti-IL-4 mAb, the predominant T cell response to Ag stimulation was simultaneous secretion of IL-10 and IFN-gamma. The concomitant secretion of IL-10 and IFN-gamma by T cells was confirmed by stimulating lines in the absence of APCs with plate-bound anti-CD3 mAb after two rounds of Ag-specific stimulation. Moreover, IL-12 enhanced IL-10 and IFN-gamma production in a myelin basic protein-reactive T cell clone, demonstrating that a differentiated T cell clone could be induced to secrete both cytokines. The addition of a neutralizing anti-IFN-gamma Ab to cultures with IL-12 and anti-IL-4 mAb during the generation of tetanus toxoid-reactive lines had no effect on the induction of IL-10 and IFN-gamma secretion, indicating that IL-12 and not IFN-gamma was responsible for the induction of this subset of T cells. Thus, in human T cells, IL-12 induces concomitant secretion of IL-10 and IFN-gamma.

We investigated the immune response to proteolipid protein (PLP), the most abundant central nervous system myelin protein in humans. A total of 8207 short-term T cell lines were generated from 49 individuals, 39 patients with multiple sclerosis and 10 control subjects. As we have reported previously, the frequency of PLP-reactive T cells did not differ between the two groups. To determine immunodominant PLP epitopes, proliferative responses of 971 PLP-specific lines were tested with 27 overlapping 20-amino acid peptides encompassing the human PLP sequence and the binding affinities of the PLP peptides to DRB5*0101 and DRB1*1501, DR2 MHC class II isotypes associated with multiple sclerosis, were determined. The T cell response after primary PLP stimulation was focused on two immunodominant epitopes comprising residues p30-49 and p180-199. These two fragments were recognized after processing of native protein by APCs and were situated in hydrophilic regions of PLP exhibiting only moderate affinity to DR2 molecules. In contrast, when T cells from DR2+ subjects were stimulated initially by individual synthetic peptides with either high or low affinity to DRB5*0101 and DRB1*1501 isotypes, additional cryptic epitopes were recognized. MHC restriction of lines specific for the cryptic PLP epitopes were related to binding affinity to DR2 isotypes. Our results indicate that protein Ags are recognized in vivo as immunodominant epitopes after Ag processing by APCs and as cryptic epitopes after processing, presumably by extracellular proteolytic enzymes.

Infiltrating gamma delta T cells are potentially involved in the central nervous system demyelination in multiple sclerosis (MS). To further study this hypothesis, we analyzed the frequency and functional properties of gamma delta T cells in peripheral blood (PB) and paired cerebrospinal fluid (CSF) of patients with MS and control subjects, including patients with other neurologic diseases (OND) and healthy individuals. The frequency analysis was performed under limiting dilution condition using rIL-2 and PHA. After PHA stimulation, a significantly increased frequency of gamma delta T cells was observed in PB (14.7 x 10(-4)) and in CSF (15.8 x 10(-4)) of MS patients as compared with 4.3 x 10(-4) in PB and 3.9 x 10(-4) detected in CSF of patients with OND. The frequency was represented equally in OND patients and normal individuals. Similarly, the IL-2-responsive gamma delta T cells occurred at a higher frequency in PB of control subjects (1.1 x 10(-4)) in OND patients and 1.5 x 10(-4) in normal individuals). Forty-three percent (13 of 30) of the gamma delta T cell clones isolated from PB and CSF of MS patients responded to heat shock protein (HSP70) but not HSP65, whereas only 2 of 30 control gamma delta T cell clones reacted to the HSP. The majority of the gamma delta T cell clones were able to induce non-MHC-restricted cytolysis of Daudi cells. All clones displayed a substantial reactivity to bacterial superantigens staphylococcal enterotoxin B and toxic shock syndrome toxin-1, irrespective of their gamma delta V gene usage. Furthermore, the gamma delta T cell clones expressed predominantly TCRDV2 and GV2 genes (26 of 35 clones), whereas the clones derived from CSF of MS patients expressed either DV1 or DV2 genes. The obtained gamma delta clones, in general, represented rather heterogeneous clonal origins, even though a predominant clonal origin was found in a set of 10 gamma delta clones derived from one patient with MS. The present study provides new evidence supporting a possible role of gamma delta T cells in the secondary inflammatory processes in MS.

T cell proliferation is potently suppressed by TGF-beta during the G1 phase of the cell cycle. The mechanism appears to involve inhibition of cell cycle kinases that phosphorylate the retinoblastoma protein (pRb), a key regulator of cell cycle progression from G1 to S phase. Although productive infection with the human T cell lymphotropic virus type I (HTLV-I) induces T cell activation, it also, paradoxically, leads to increased production of TGF-beta. To investigate whether infection by HTLV-I conferred resistance to TGF-beta in non-immortalized T cells, we generated T cell clones from patients with HTLV-I myelopathy by direct single cell cloning. Here we report that HTLV-I-infected but not uninfected T cell clones have hyperphosphorylated pRb consistent with viral-induced T cell activation. Furthermore, the HTLV-I-infected T cells were resistant to growth suppression by rTGF-beta 1 and this correlated with the inability of TGF-beta 1 to prevent hyperphosphorylation of pRb. However, when spontaneously proliferating HTLV-I-infected T cell clones were further stimulated by cross-linking of the CD3/TCR complex, the superimposed proliferation was significantly inhibited by TGF-beta 1, suggesting that the TGF-beta 1 signaling pathway was intact. Together these findings suggest that HTLV-I induces T cell activation through a pathway that is insensitive to TGF-beta 1. This may have implications for the altered immune regulation in patients with HTLV-I myelopathy.

Ag-presenting cells provide at least two distinct signals for T cell activation. T cell receptor-dependent stimulation is provided by presentation of a specific peptide Ag in association with MHC molecules. In addition, APC also supply costimulatory signals required for T cell activation that are neither Ag- nor MHC restricted. One such costimulatory signal is mediated via the interaction of B7 on APC with the CD28 receptor on T cells. Recently, CTLA-4 has been shown to be a second B7 receptor on T cells. In the present report, we have examined the expression of CD28 and CTLA-4 on a panel of resting and activated normal T cell subsets and T cell clones by RNA blot analysis in an attempt to determine whether their expression defines reciprocal or overlapping subsets. CD28 was detected in resting T cells, whereas CTLA-4 was not. After stimulation with PHA and PMA for 24 h, CTLA-4 mRNA was expressed in both the CD4+ and CD8+ subsets as well as in CD28+ T cells. We examined 37 human and six murine T cell clones that had been previously characterized for their cytokine production. After activation, CTLA-4 and CD28 mRNA were coexpressed in 36 of 37 human T cell clones and all six murine T cell clones. These included T cells of CD4+8-, CD4-8+, and CD4-8- phenotypes as well as clones with Th1 and Th2 cytokine profiles. In contrast, CD28 but not CTLA-4 mRNA was detected in leukemic T cell lines and myelomas. CTLA-4 and B7 mRNA but not CD28 mRNA was detected in two long term HTLV-I-transformed T cell lines. These data demonstrate that CD28 and CTLA-4 mRNA are coexpressed in most activated T cells and T cell clones, providing evidence that they do not define reciprocal subsets. Moreover, they are consistent with the hypothesis that B7 transmits its signal through a single receptor, CD28, on resting T cells, and multiple receptors, CD28 and CTLA-4, on activated T cells.

It has been difficult to demonstrate functionally distinct T cell populations in humans on the basis of cytokine secretion. As previous investigators have examined the T cell cytokine profile from immunized animals, we examined whether Th1 or Th2 type T cells could be identified in the peripheral blood or cerebrospinal fluid (CSF) immune compartments from subjects with or without inflammatory diseases. Using limiting dilution analysis and growth with PHA and IL-2/IL-4, we directly cloned a total of 177 T cells from the peripheral blood and CSF of seven subjects, four with inflammatory disease and three control subjects, and examined the cytokine message profile after stimulation with ionomycin and PMA. We found that most clones from both the peripheral blood and CSF express IL-1, IL-2, IL-4, IFN-gamma, or TNF-alpha cytokine mRNA after activation with ionomycin and PMA. All T cell clones tested produced TNF-alpha mRNA, and all but 14 produced IFN-gamma mRNA. As reported previously, Th0 cells, which produced IFN-gamma, IL-2, IL-4, and IL-5 mRNA, were found in most subjects. In striking contrast, Th1 cells, which expressed IL-2 and IFN-gamma but not IL-4 or IL-5 mRNA, were present in both peripheral blood and CSF of subjects with inflammatory disease but not found in peripheral blood or CSF of subjects without systemic inflammation. Th2 cells, expressing IL-4 and IL-5 but not IFN-gamma or IL-2 mRNA, were not found in any subject. These data present the first evidence for Th1 T cell clones in humans that may be associated with systemic inflammation.

The functional effects resulting from CD4 and CD8 perturbation were analyzed by using a CD4+CD8+ clone and anti-CD4 and anti-CD8 monoclonal antibodies. Perturbation of CD8, but not CD4, by soluble antibody resulted in the inhibition of CD3-T cell receptor (CD3-Ti) triggering as determined by flow cytometric measurements of intracellular free Ca2+ concentrations. In addition, the CD3-T cell receptor-mediated cytotoxic function of the CD4+CD8+ clone was inhibited by anti-CD8, but not by anti-CD4. These results suggest that CD8, but not CD4, was functionally associated with CD3-Ti on the CD4+CD8+ clone. Although CD4 perturbation did not affect CD3-Ti-mediated activities, it resulted in the inhibition of the interleukin 2-dependent proliferation of this clone. Perturbation of CD8 did not affect the interleukin 2 dependent proliferation of the CD4+CD8+ clone. On the other hand, CD4 molecules of another CD4+CD8- clone unlike those of the CD4+CD8+ clone, were clearly linked to T cell receptor function. These results indicate that CD4 perturbation can result in two distinct regulatory activities; one involves the regulation of CD3-T cell receptor function, whereas the other is not directly associated with CD3-T cell antigen receptor function. The data are also consistent with the notion that CD4 and CD8 do not merely function as recognition and adhesion elements for accessory cell major histocompatibility complex molecules, but have a direct role in the regulation of T cell activation.

To analyze expression of the 2H4 (CD45R) Ag on inflammatory cells in the central nervous system immune response, immunohistochemical staining with a panel of anti-T cell mAb was performed on central nervous system tissues from 12 patients with multiple sclerosis (MS) and 8 patients with viral encephalitis. Only rare cells were stained with anti-2H4 in MS plaques, plaque edges, and adjacent white matter, whereas 2H4+ cells were more numerous in viral encephalitis (p less than 0.001). By contrast, no quantitative differences were found between MS plaque edges and viral encephalitis with anti-4B4 (helper-inducer function associated), anti-CD4, anti-CD3, and anti-IL-2R mAb, although there were fewer CD8+ cells in MS (p less than 0.01). These data indicate that the 2H4 Ag is selectively decreased and, because it is associated with suppressor-inducer function of CD4+ cells, that there may be a defect in the down-regulation of the in situ immune response in MS.

The immunologic responses to anti-T4 and anti-T11 mAb infusions were assessed in subjects with chronic progressive multiple sclerosis as part of phase I clinical studies. Eight subjects received five daily infusions (0.2 mg/kg/day) of either anti-T11 (CD2) or anti-T4 (CD4) murine mAb. It was found that in vivo anti-T cell mAb infusions specifically suppress in vitro measurements of the human immune response; anti-T11 mAb blocked T cell activation via the CD2 SRBC-binding protein, and anti-T4 mAb infusions abolished PWM-induced Ig synthesis without lysis of the CD4+ T cell subpopulations. With repeated infusions of the anti-T11 mAb in three subjects, anti-mouse antibodies were found in the circulation. Although the majority of human anti-mouse antibody was not isotype specific, significant anti-idiotypic-like activity was observed after repeated infusions in two of three subjects. The human anti-mouse antibodies were almost exclusively of the IgG isotype. The magnitude of the human anti-mouse response was significantly less after administration of either anti-T11 or anti-T4 as compared with anti-T12 mAb. Murine anti-T cell mAb can provide a specific, benign form of acute immunosuppression in humans. However, repeated administration of these reagents in more chronic diseases can result in anti-Id-like antibodies that block binding of the anti-T cell mAb to the T cell surface.

T cells were directly cloned from autopsied MS brain plaque tissue and reactivity was measured with the major encephalitogenic neuroantigens, myelin basic protein (MBP), and proteolipid protein (PLP). Control clones were simultaneously derived from the blood. The proportion of T4+ and T8+ T cell clones from the brain tissue differed from that of peripheral blood T cell clones derived at the same time, suggesting that the clones were not derived from the peripheral blood. None of 57 brain-derived T cell clones proliferated to either MBP or PLP, although they responded well to PHA and IL 2. An additional 235 clones derived from the cerebrospinal fluid and 126 clones from the peripheral blood of other subjects with multiple sclerosis also did not proliferate to MBP or PLP. In contrast, five of nine T4+ clones from the CSF of a subject with postinfectious encephalomyelitis exhibited low but clear reactivity to human MBP, supporting the possible role of MBP as the target antigen in this disease. These studies, the first to clone T cells directly from MS plaque tissue, suggest that the lack of consistent T cell reactivity to MBP or PLP in the peripheral blood of MS patients does not appear to be secondary to the sequestration of a large number of these cells in the brain.

CD4 and CD8 molecules have been implicated in the regulation of T cell activation. In the present study, CD4 and CD8 were modified by increased phosphorylation when T cell clones or T cells were either exposed to phorbol-12-myristate- 13-acetate or were triggered via the CD3-T cell receptor complex. Activation of T cells through the CD2 sheep erythrocyte binding protein, using anti-T11(2) and -T11(3) antibodies, also resulted in CD4 and CD8 phosphorylation. These findings suggest that signals derived from two different receptor pathways can converge and result in similar molecular modifications of CD4 and CD8. Furthermore, phorbol myristate acetate treatment or activation via the CD2 pathway induced phosphorylation of the CD4 and CD8 molecules of thymocytes, suggesting that these molecules may be functional in thymus. Together, our findings indicate that CD4 and CD8 phosphorylation is a consequence of T cell triggering, and suggest that CD4 and CD8 phosphorylation may represent a molecular signaling mechanism among the CD3-T cell receptor complex, CD2, CD4, and CD8.

Ta1 is a 105,000 dalton protein that is weakly expressed on a small fraction of resting human peripheral blood T cells but strongly expressed in vitro on T cell clones and a substantial proportion of activated T cells. Unlike receptors for growth factors such as IL 2, the Ta1 antigen is present on T cell lines and clones irrespective of cell cycle. The function of Ta1 was investigated after separation of T lymphocytes into Ta1-enriched and Ta1-depleted subpopulations that were obtained from normal human subjects. Although Ta1-enriched T cells constitute only 10 to 15% of the E rosette-positive lymphocyte population, most, if not all, of the anamnestic response to the recall antigens tetanus toxoid and mumps reside in the Ta1+ population. Both Ta1-enriched and -depleted cells responded equally well to the mitogen PHA. The autologous mixed lymphocyte response was also greater in the Ta1-enriched subpopulation but not to the degree seen with soluble antigen. Increased proliferation was not due simple to increased inducer cell function within the Ta1+ subpopulations because both Ta1- and Ta1+ cells induced similar amounts of Ig synthesis in the presence of PWM. Additionally, increasing numbers of Ta1- cells did not suppress the enhanced proliferative responses of Ta1+ cells, and thus Ta1- cells do not appear to be functioning as suppressor cells. The Ta1 antigen appears to be a marker for previously activated T cells in peripheral blood, and this subpopulation appears to include T memory cells.

The Ta1 (CDw26) Ag distinguishes a subset of circulating T lymphocytes that is the major population proliferating to recall Ag challenge. Unlike receptors for growth factors such as IL-2 and transferrin, the Ta1 Ag is present on T cell lines and clones irrespective of cell cycle. The appearance of Ta1 on T cells that respond to recall Ag allowed us to investigate activation requirements that may be associated with T cell immune memory. Ta1+ peripheral blood T cells were induced to proliferate by mAb recognizing either the invariant chains of the TCR, or by pairs of mitogenic antibodies directed to the CD2 molecule. In contrast, Ta1- cells were not stimulated by these antibodies. In addition, Ta1-cells did not proliferate maximally after addition of the phorbol ester PMA in combination with the calcium ionophore Ionomycin, suggesting that the intracellular targets of these agents may not be fully active. Anti-CD3-induced elevation of intracellular calcium levels was equivalent in the two subpopulations, suggesting that calcium mobilization mechanisms were intact. In contrast, PMA-induced phosphorylation of TCR CD3 chains was significantly greater in Ta1+ cells as compared to Ta1- T cells. Taken together, our results indicate that Ta1 expression, which is associated with T cell activation and memory, may be causally related to TCR and CD2-mediated activation mechanisms. The PMA inducible TCR phosphorylation in Ta1+ memory cells associated with their increased ability to proliferate after CD3/TCR or CD2 stimulation suggests that intracellular phosphorylation events may be causally associated with T cell immune memory.