Non-CD28 costimulatory molecules present in T cell rafts induce T cell costimulation by enhancing the association of TCR with rafts.

While CD28 functions as the major T cell costimulatory receptor, a number of other T cell molecules have also been described to induce T cell costimulation. Here, we investigated the mechanisms by which costimulatory molecules other than CD28 contribute to T cell activation. Non-CD28 costimulatory molecules such as CD5, CD9, CD2, and CD44 were present in the detergent-insoluble glycolipid-enriched (DIG) fraction/raft of the T cell surface, which is rich in TCR signaling molecules and generates a TCR signal upon recruitment of the TCR complex. Compared with CD3 ligation, coligation of CD3 and CD5 as an example of DIG-resident costimulatory molecules led to an enhanced association of CD3 and DIG. Such a DIG redistribution markedly up-regulated TCR signaling as observed by ZAP-70/LAT activation and Ca2+ influx. Disruption of DIG structure using an agent capable of altering cholesterol organization potently diminished Ca2+ mobilization induced by the coligation of CD3 and CD5. This was associated with the inhibition of the redistribution of DIG although the association of CD3 and CD5 was not affected. Thus, the DIG-resident costimulatory molecules exert their costimulatory effects by contributing to an enhanced association of TCR/CD3 and DIG.

CD5 costimulation up-regulates the signaling to extracellular signal-regulated kinase activation in CD4+CD8+ thymocytes and supports their differentiation to the CD4 lineage.

CD5 positively costimulates TCR-stimulated mature T cells, whereas this molecule has been suggested to negatively regulate the activation of TCR-triggered thymocytes. We investigated the effect of CD5 costimulation on the differentiation of CD4+CD8+ thymocytes. Coligation of thymocytes with anti-CD3 and anti-CD5 induced enhanced tyrosine phosphorylation of LAT (linker for activation of T cells) and phospholipase C-gamma (PLC-gamma) compared with ligation with anti-CD3 alone. Despite increased phosphorylation of PLC-gamma, this treatment down-regulated Ca2+ influx. In contrast, the phosphorylation of LAT and enhanced association with Grb2 led to activation of extracellular signal-regulated kinase (ERK) mitogen-activated protein kinase. When CD3 and CD5 on CD4+CD8+ thymocytes in culture were coligated, they lost CD8, down-regulated CD4 expression, and induced CD69 expression, yielding a CD4+(dull)CD8-CD69+ population. An ERK inhibitor, PD98059, inhibited the generation of this population. The reduction of generation of CD4+CD8- cells resulted from decreased survival of these differentiating thymocytes. Consistent with this, PD98059 inhibited the anti-CD3/CD5-mediated Bcl-2 induction. These results indicate that CD5 down-regulates a branch of TCR signaling, whereas this molecule functions to support the differentiation of CD4+CD8+ thymocytes by up-regulating another branch of TCR signaling that leads to ERK activation.

Association of a tetraspanin CD9 with CD5 on the T cell surface: role of particular transmembrane domains in the association.

CD9 is a member of the tetraspanin superfamily which is characterized by four transmembrane (TM) domains and associates with other surface molecules. This tetraspanin was recently found to be expressed on mature T cells. Here, we investigated which molecules associate with CD9 on T cells and which CD9 domains are required for the association. Immunoprecipitation of T cell lysates with anti-CD9 mAb followed by immunoblotting with mAb against various T cell molecules showed the association of CD9 with CD3, CD4, CD5, CD2, CD29 and CD44. Because association with CD5 was most prominent, we determined the role of CD9 TM or extracellular (EC) domains in the association with CD5. CD9 mutant genes lacking each domain were constructed and introduced into EL4 thymoma cells deficient in CD9 but expressing CD5. Among various types of stable EL4 transfectants, EL4 transfected with the mutant gene lacking TM domains (TM2/TM3) between two EC domains expressed a small amount of the relevant protein without showing association with CD5. CD9(-)CD5(-) monkey COS-7 cells transfected with this mutant gene and the CD5 gene expressed both transfected gene products, but the association of these was not detected. EL4 cells transfected with a CD9/CD81 chimera gene (the CD9 gene containing TM2/TM3 of CD81) expressed the chimeric protein on the cell surface and showed association with CD5. These results suggest an essential role of particular CD9 TM domains in the surface expression of the CD9 molecule as well as the association with CD5.

A caspase inhibitor protects thymocytes from diverse signal-mediated apoptosis but not from clonal deletion in fetal thymus organ culture.

A family of caspases has been implicated as an effector in various forms of apoptosis. The present study investigated whether this family of proteases is involved in the induction of intrathymic clonal deletion in comparison with apoptosis induced in the thymus by various signals. Potent apoptosis of thymocytes was induced in fetal thymus organ cultures (FTOC) when FTOC were treated with glucocorticoid, radiation, and anti-CD3 monoclonal antibody (mAb). As a model of negative selection based on apoptotic clonal deletion, the elimination of Vbeta8-expressing thymocytes was induced by inoculating Staphylococcal enterotoxin B (SEB) into FTOC. Addition of a peptide-based caspase inhibitor resulted in the protection of thymocytes from apoptosis induced by glucocorticoid, radiation, and anti-CD3 mAb. In contrast, the same treatment failed to prevent clonal deletion of Vbeta8high thymocytes. These results suggest that different pathways of cell death operate in the thymus that may be distinguished depending on the caspase/protease utilized in each pathway.

Differential involvement of a Fas-CPP32-like protease pathway in apoptosis of TCR/CD9-costimulated, naive T cells and TCR-restimulated, activated T cells.

Our previous study showed that CD9 costimulation of TCR-triggered naive T cells elicits activation ([3H]TdR incorporation) that is similar to CD28 costimulation; however, unlike CD28 costimulation, CD9 costimulation results in apoptosis of these previously activated T cells. Here, we investigated whether the apoptosis occurring after TCR/CD9 stimulation is associated with a death pathway involving Fas stimulation and Fas-mediated caspase activation as observed in activation-induced cell death (AICD). In contrast to AICD, the apoptosis resulting from TCR/CD9 stimulation in C57BL/6 T cells was independent of Fas, because this form of apoptosis was not prevented by anti-Fas ligand mAb and was also induced in MRL/lpr T cells. AICD was observed at 12 h after the restimulation of activated T cells with anti-CD3 and reached a peak level at 24 h after this restimulation. CPP32-like protease activity was detected during AICD. Although TCR/CD9 stimulation-associated apoptosis was observed at 24 h after the stimulation of naive T cells and reached a peak level at 36 h after this stimulation, CPP32-like protease activity in these T cells was only marginal at all time points. Nevertheless, both forms of apoptosis were prevented similarly by two different peptide-based caspase inhibitors. These results indicate that the apoptosis that follows the T cell activation which is induced as a result of CD9 costimulation does not involve a Fas-CPP32-like protease pathway, but suggest that different caspase members are likely to be critical in this form of apoptosis.

A fundamental difference in the capacity to induce proliferation of naive T cells between CD28 and other co-stimulatory molecules.

T cell activation requires two signals: a signal from the TCR and a co-stimulatory signal provided by antigen-presenting cells (APC). In addition to CD28, multiple molecules on the T cell have been described to deliver co-stimulatory signals. Here, we investigated whether there exist quantitative or qualitative differences in the co-stimulatory capacity between CD28 and other molecules. Anti-CD28 monoclonal antibody (mAb) and mAb against CD5, CD9, CD2, CD44 or CD11a all induced activation of naive T cells in the absence of APC when co-immobilized with a submitogenic dose of anti-CD3 mAb. [3H]Thymidine incorporation determined 2 days after co-stimulation was all comparable. In contrast to progressive T cell proliferation induced by CD28 co-stimulation, co-stimulation by other T cell molecules led to a decrease in viable cell recovery along with the induction of apoptosis of once activated T cells. This was associated with a striking difference in IL-2 production; CD28 co-stimulation induced progressively increasing IL-2 production, whereas co-stimulation by other molecules produced limited amounts of IL-2. Addition of recombinant IL-2 to the latter cultures corrected the induction of apoptosis, resulting in levels of cellular proliferation comparable to those observed for CD28 co-stimulation. These results indicate that a fundamental difference exists in the nature of co-stimulation between CD28 and other molecules, which can be evaluated by the levels of IL-2 production, but not simply by [3H]thymidine incorporation.

Suppression of allograft responses induced by interleukin-6, which selectively modulates interferon-gamma but not interleukin-2 production.

BACKGROUND: Interferon (IFN)-gamma produced by activated T cells represents an important effector cytokine in mediating an inflammatory response. METHODS: The present study investigated the modulation of allograft responses by inhibiting IFN-gamma production. C57BL/6 (B6) lymph node cells were stimulated with class II H2-disparate B6-C-H-2bm12 (bm12) spleen cells. RESULTS: Addition of interleukin (IL)-6 to the primary B6 anti-bm12 mixed lymphocyte reaction (MLR) inhibited neither proliferative responses nor IL-2 production. However, IL-6 induced a dose-dependent suppression of IFN-gamma production in the same MLR cultures. B6 mice were engrafted with bm12 skin grafts, and IL-6 was given to bm12 skin graft recipients every other day. T cells from these recipient mice produced significantly less IFN-gamma in secondary B6 anti-bm12 MLR than those from bm12 skin graft recipients that had not received IL-6 injections. IFN-gamma production by these T cells was suppressed more strongly when the secondary MLR was conducted in the presence of IL-6. In addition to suppression of IFN-gamma expression, IL-6 injections resulted in prolongation of bm12 skin graft survival. The critical involvement of IFN-gamma in anti-bm12 rejection responses was substantiated by evidence that administration of anti-IFN-gamma monoclonal antibody strikingly prolonged bm12 skin graft survival. The prolongation of graft survival by in vivo treatment with either IL-6 or anti-IFN-gamma monoclonal antibody was found to be induced without blocking cellular infiltration of the grafts. CONCLUSIONS: These results indicate that IFN-gamma acts as a key cytokine in a B6 anti-bm12 allograft response and that IL-6 may down-regulate this response by inhibiting IFN-gamma production of alloreactive T cells.

CD9-mediated costimulation of TCR-triggered naive T cells leads to activation followed by apoptosis.

The induction of full activation or death in TCR-triggered T cells depends largely on whether appropriate costimulatory signals are provided. In this study, we show that the costimulation of CD9 on naive T cells during TCR stimulation results in transient, albeit potent, activation followed by apoptosis, rather than full activation. Anti-CD9 mAb synergized with suboptimal doses of anti-CD3 mAb in inducing T cell activation. [3H]TdR incorporation determined 2 days after CD9 costimulation was as potent as that induced by CD28 costimulation. In contrast to progressive T cell proliferation induced by CD28 costimulation, CD9 costimulation led to the induction of apoptosis of once-activated T cells. Although IL-2R expression was induced significantly earlier and to a greater degree after CD9 costimulation than after CD28 costimulation, CD9 costimulation only transiently produced a small amount of IL-2 and induced apparently low levels of bcl-xL compared with those observed in CD28 costimulation. Addition of rIL-2 to cultures of CD9 costimulation induced strikingly enhanced expression of bcl proteins, especially of bcl-xL, and protected TCR-stimulated T cells from apoptosis. These data indicate that CD9-mediated costimulation of TCR-triggered naive T cells leads to activation followed by apoptosis as the result of failure to generate a positive signal for sufficient levels of IL-2 production.

Synergy between CD28 and CD9 costimulation for naive T-cell activation.

Our previous study demonstrated that CD9 is expressed on most mature naive T-cells and delivers a potent costimulatory signal that functions independently of CD28. Here, we investigated whether this CD9-mediated signal is different from the CD28-mediated signal in the mode of costimulation and whether both signals function synergistically for T-cell activation. Anti-CD9 or anti-CD28 monoclonal antibody (mAb) increased [3H]TdR incorporation of naive T-cells in the absence of antigen-presenting cells (APC) when coimmobilized with submitogenic doses of anti-CD3 mAb. The levels of costimulation induced by ligation of CD9 and CD28 were comparable. However, the costimulatory effect differed between soluble anti-CD9 and CD28 mAb. A soluble form of anti-CD28 mAb could costimulate anti-CD3-triggered T-cells, whereas soluble anti-CD9 mAb failed to costimulate. Although anti-CD28 costimulated naive T-cells treated with phorbol myristate acetate (PMA) instead of anti-CD3 mAb, a combination of PMA plus anti-CD9 mAb could not induce T-cell activation. The combined costimulation of anti-CD3-triggered T-cells with anti-CD9 and anti-CD28 mAbs resulted in strikingly enhanced [3H]TdR uptake and lymphokine (IL-2 and IFN-gamma) production when compared to those induced by each costimulation. These results suggest that CD9 and CD28 induce T-cell costimulation using different signaling pathways, thereby inducing synergy in T-cell activation.

Expression of an inducible type of nitric oxide (NO) synthase in the thymus and involvement of NO in deletion of TCR-stimulated double-positive thymocytes.

The present study investigates the role of nitric oxide (NO) in the deletion of TCR-stimulated double-positive (DP) thymocytes. Fetal thymi expressed mRNA for an inducible type of NO synthase (iNOS). The levels of iNOS mRNA became maximal around gestation day 18 with a decline after birth. Administration of anti-CD3 mAb to fetal thymus organ culture (FTOC) or young mice resulted in enhanced expression of mRNAs for iNOS as well as IFN-gamma. Immunohistochemical analyses revealed that iNOS was produced in the corticomedullary junction and medulla. The effects of iNOS-induced NO on anti-CD3-unstimulated or anti-CD3-stimulated thymocytes were examined by culturing them in the presence or absence of a NO-generating compound. Stimulation of DP thymocytes with anti-CD3 alone induced the generation of CD4(low)CD8(low) thymocytes. The subsequent exposure of these anti-CD3-stimulated thymocytes to NO promoted down-regulation of CD4 and CD8 expression. The recovery of viable DP cells was considerably reduced compared with stimulation with anti-CD3 or NO alone. Even in a viable DP population, high incidences of DNA strand breaks were detected in the CD4(low)CD8(low) compartment. In contrast to DP cells, the recovery of viable single-positive cells was not decreased but rather slightly enhanced by treatment with anti-CD3 and/or NO. The recovery of anti-CD3-stimulated thymocytes were also reduced when cultured on the thymic stromal monolayer with the capacity to produce NO upon IFN-gamma stimulation. These results indicate that NO, which is generated in association with TCR stimulation in the thymus, functions to induce deletion of DP thymocytes, especially when their TCR is stimulated.

Inhibition of growth and metastasis of ovarian carcinoma by administering a drug capable of interfering with vascular endothelial growth factor activity.

The present study investigates the relationship between in vivo growth/metastasis of tumor cells and their capacity to produce the vascular endothelial growth factor (VEGF), as well as the regulation of tumor growth/metastasis using an angiogenesis-inhibitory drug. Two cloned tumor cell lines designated OV-LM and OV-HM were isolated from a murine ovarian carcinoma OV2944. OV-LM and OV-HM cells grew in cultures at comparable rates. However, when transplanted s.c. into syngeneic mice, OV-HM exhibited a faster growth rate and a much higher incidence of metastasis to lymph nodes and lung. Histologically, intense neovascularization was detected in sections of OV-HM but not of OV-LM tumor. OV-HM and OV-LM tumor cells obtained from in vitro cultures expressed high and low levels of VEGF mRNA, respectively. A difference in VEGF mRNA expression was much more clearly observed between RNAs prepared from fresh OV-HM and OV-LM tumor masses: RNA from OV-HM contained larger amounts of VEGF mRNA, whereas RNA from OV-LM exhibited only marginal levels of VEGF mRNA. An angiogenesis-inhibitory drug, FR118487 inhibited the VEGF-mediated in vitro growth of endothelial cells but did not affect the expression in vitro of VEGF mRNA by OV-HM tumor cells. Intraperitoneal injections of FR118487 into mice bearing OV-HM tumors resulted in: (i) a subsequent growth inhibition of primary tumors; (ii) a marked decrease in neovascularization inside tumor masses expressing comparable levels of VEGF mRNA to those detected in control OV-HM masses; and (iii) almost complete inhibition of metastasis to lymph nodes and lung. These results indicate that growth/metastasis of tumor cells correlates with their VEGF-producing capacity and that an angiogenesis inhibitor, FR118487, inhibits tumor growth and metastasis through mechanism(s) including the suppression of VEGF function in vivo.

A role for CD9 molecules in T cell activation.

Costimulation mediated by the CD28 molecule plays an important role in optimal activation of T cells. However, CD28-deficient mice can mount effective T cell-dependent immune responses, suggesting the existence of other costimulatory systems. In a search for other costimulatory molecules on T cells, we have developed a monoclonal antibody (mAb) that can costimulate T cells in the absence of antigen-presenting cells (APC). The molecule recognized by this mAb, 9D3, was found to be expressed on almost all mature T cells and to be a protein of approximately 24 kD molecular mass. By expression cloning, this molecule was identified as CD9, 9D3 (anti-CD9) synergized with suboptimal doses of anti-CD3 mAb in inducing proliferation by virgin T cells. Costimulation was induced by independent ligation of CD3 and CD9, suggesting that colocalization of these two molecules is not required for T cell activation. The costimulation by anti-CD9 was as potent as that by anti-CD28. Moreover, anti-CD9 costimulated in a CD28-independent way because anti-CD9 equally costimulated T cells from the CD28-deficient as well as wild-type mice. Thus, these results indicate that CD9 serves as a molecule on T cells that can deliver a potent CD28-independent costimulatory signal.

Molecular mechanisms underlying IFN-gamma-mediated tumor growth inhibition induced during tumor immunotherapy with rIL-12.

The present study investigates the molecular mechanisms by which IFN-gamma produced as a result of in vivo IL-12 administration exerts its anti-tumor effects. rIL-12 was administered three or five times into mice bearing CSA1M fibrosarcoma, OV-HM ovarian carcinoma or MCH-1-A1 fibrosarcoma. This regimen induced complete regression of CSA1M and OV-HM tumors but only transient growth inhibition of MCH-1-A1 tumors. The anti-tumor effects of IL-12 were associated with enhanced induction of IFN-gamma because these effects were abrogated by pretreatment of hosts with anti-IFN-gamma antibody. Exposure in vitro of the three types of tumor cells to rRFN-gamma resulted in moderate to potent inhibition of tumor cell growth. IFN-gamma stimulated the expression of mRNAs for an inducible type of NO synthase (iNOS) in CSA1M cells and indoleamine 2,3-dioxygenase (IDO), an enzyme capable of degrading tryptophan, in OH-HM cells, but induced only marginal levels of these mRNAs in MCH-1-A1 cells. In association with iNOS gene expression, IFN-gamma-stimulated CSA1M cells produced a large amount of NO which functioned to inhibit their own growth in vitro. Although OV-HM and MCH-1A1 cells did not produce NO, they also exhibited NO susceptibility. Whereas the tumor masses from IL-12-treated CSA1M-bearing or OV-HM-bearing mice induced higher levels of iNOS (for CSA1M) or IDO and iNOS (for OV-HM) mRNAs, the MCH-1-A1 tumor mass expressed lower levels of iNOS mRNA alone. Moreover, massive infiltration of CD4(+) and CD8(+) T cells and Mac-1(+) cells was seen only in the CSA1M and OV-HM tumors. Thus, these results indicate that IFN-gamma produced after IL-12 treatment induces the expression of various genes with potential to modulate tumor cell growth by acting directly on tumor cells or stimulating tumor-infiltrating lymphoid cells and that the effectiveness of IL-12 therapy is associated with the operation of these mechanisms.

IL-12 produced by antigen-presenting cells induces IL-2-independent proliferation of T helper cell clones.

We investigated the role of IL-12 in proliferation of various Th cell clones (class II-alloreactive (4-86 and 4-55) and keyhole limpet hemocyanin + self I-Ek-reactive (9-16)) following stimulation with Ag on APCs. These clones proliferated in response to stimulation with rIL-2, rIL-12, or Ag/APC. The proliferation induced by Ag/APC stimulation was not affected by anti-IL-2 Ab but was markedly inhibited by anti-IL-12 Abs. Consistent with this finding was the absence of detectable IL-2 activity in culture supernatants 12 to 48 h after Ag/APC stimulation, and the detection of significant levels of IL-12 in an Ab-capture bioassay. IL-12 was produced within 12 h after Ag/APC stimulation, reaching a peak after 18 to 24 h. The production of IL-12 in cultures of Th clones and APC contrasted with the production of IL-2 but not IL-12 upon allostimulation of primary T cells and the inhibition of their proliferation exclusively by anti-IL-2 Abs. Analysis of the expression of IL-12-binding sites on Th cells revealed low levels of IL-12 receptors in resting Th clones but high IL-12R levels 2 to 3 days after Ag/APC stimulation, declining gradually thereafter. The changes in IL-12R expression levels correlated closely with the IL-12 responsiveness of Th populations at various times after Ag/APC stimulation; Th populations obtained 3 and 10 days after Ag/APC stimulation exhibited very high and weak or marginal responsiveness to rIL-12, respectively, whereas the responses to rIL-2 were comparable in both Th populations. These results indicate that the Ag/APC-stimulated proliferation of terminally differentiated Th clones, in contrast to naive T cells, depends on the production of IL-12 by APC and on the simultaneous up-regulation of IL-12R on Th cells rather than on an IL-2 autocrine mechanism.

CD28 co-stimulatory signals induce IL-2 receptor expression on antigen-stimulated virgin T cells by an IL-2-independent mechanism.

Intravenous sensitization of C57BL/6 (B6) mice with class II H-2-disparate B6-C-H-2bm12(bm12) resting B cells induced anti-bm12 CD4+ T cell tolerance as shown by hyporesponsiveness in the anti-bm12 mixed lymphocyte reaction (MLR). The present study investigated the mechanism(s) of the failure of bm12 B cells to stimulate the proliferation of B6 anti-bm12 CD4+ T cells. While stimulation in vitro to B6 splenic T cells with bm12 antigen-presenting cells (APC) induced IL-2 mRNA expression and IL-2 production, T cells stimulated with bm12 B cells expressed much less IL-2 mRNA and secreted very low but detectable levels of IL-2. Moreover, the T cells stimulated with the bm12 B cells did not proliferate and this was not corrected by the addition of rIL-2 responsiveness. Further, whereas IL-2 receptor (IL-2R) alpha chain expression was significantly induced on B6 T cells stimulated with bm12 APC; stimulation with bm12 B cells did not induce IL-2R expression over background levels. However, virgin T cells stimulated with both bm12 B cells and anti-CD28 mAb proliferated and displayed a dramatic increase in IL-2 production as well as IL-2R expression to levels commensurate with those resulting from bm12 B cells plus anti-CD28 mAb even in the presence of sufficient amounts of anti-IL-2 mAb for neutralizing produced IL-2; while levels of IL-2R were significantly lower compared to those induced in the absence of anti-IL-2 mAb, increased frequencies of IL-2R+ cells were comparable. Conversely, IL-2R was not induced by bm12 B cell stimulation in the presence of IL-2. Moreover, IL-2R expression and proliferation induced by stimulation with bm12 APC was inhibited by CTLA-4-Ig, a soluble recombinant fusion protein capable of blocking the CD28 co-stimulatory signals not only stimulate IL-2 production but also induce IL-2R expression by an IL-2-independent mechanism.

Administration of recombinant interleukin 12 prevents outgrowth of tumor cells metastasizing spontaneously to lung and lymph nodes.

The present study investigates the ability of recombinant interleukin 12 (rIL-12) to modulate the growth of a primary tumor as well as the outgrowth of metastatic tumor cells in an ovarian carcinoma (OV-HM) model. This aggressive tumor displayed rapid growth of the primary tumor mass, high incidence of metastases to lung and lymph nodes, and invasion from the primary s.c. site to the peritoneal cavity. Starting 12 days after s.c. tumor cell implantation, several i.p. injections of rIL-12 at 2-3 day intervals resulted in regression of growing tumors. These treated mice did not show signs of metastases or tumor recurrence at the original site. One month after tumor implantation, untreated mice did not have visible lung metastasis, but some did have palpable lymph nodes. At this stage, the primary tumors of animals without palpable lymph nodes were surgically resected. When examined 2 months later, most animals had developed lymph node and lung metastases. In contrast, rIL-12 injections after tumor resection inhibited the development of metastases in both lung and lymph nodes. This contrasted with the failure of IL-2 to prevent metastases. Even for mice already showing signs of lymph node metastases or invasion of the abdominal wall, rIL-12 administration after tumor resection prevented further invasion to the peritoneal cavity and growth of metastatic tumor cells in lung. It was somewhat surprising that the IL-12 treatment of animals after 1 month of tumor growth without resection also resulted in complete tumor regression, as well as eradication of micrometastasis that would have occurred before the treatment. Moreover, they exhibited resistance to a rechallenge with the same tumor but not with a second tumor. Thus, this tumor system provides a relevant model to clinical situations in terms of treatment of advanced tumors and metastases. These results also indicate that IL-12 can induce a curative immune response, even in the face of an aggressive micrometastasizing tumor.

Thymic stromal cells eliminate T cells stimulated with antigen plus stromal Ia molecules through their cross-talk involving the production of interferon-gamma and nitric oxide.

We previously established a thymic stromal cell clone capable of inducing differentiation of immature thymocytes and described a clonal elimination model in which T cell clones are killed on the monolayer of this stromal clone by stimulation of their T cell receptors (TCR) with antigen plus stromal Ia molecules. This study investigated molecular mechanisms underlying this phenomenon. Antigenic stimulation on thymic stromal cells produced large amounts of interferon-gamma (IFN-gamma) and small amounts of tumor necrosis factor-alpha (TNF-alpha). Addition of anti-IFN-gamma monoclonal antibody (mAb) to these cultures largely prevented death of TCR-stimulated T cells. T cell death was also induced when cultures were treated with recombinant IFN-gamma (rIFN-gamma) or rTNF-alpha instead of the relevant antigen, showing that these lymphokines are involved in the process of T cell death. It was further demonstrated that these lymphokines, especially IFN-gamma, induced the expression of mRNA for the inducible type of nitric oxide (NO) synthase in thymic stromal cells and that enhanced levels of NO were produced by stromal cells cultures with T cells plus antigen or stimulated with rIFN-gamma or rTNF-alpha. NO was found to be critically responsible for inducing T cell death on the stromal cell monolayer following stimulation of T cells with antigen or of stromal cells with rIFN-gamma or rTNF-alpha, because T cells death was completely prevented by addition of NG-monomethyl-L-arginine (L-NMMA), which is capable of inhibiting NO production. These results indicate that elimination of TCR-stimulated T cells on thymic stromal monolayers with the capacity to support thymocyte differentiation is induced by the cross-talk between IFN-gamma/TNF-alpha-producing T cells and stromal cells capable of producing NO in response to these lymphokines.

Thymic stroma-derived T-cell inhibitory factor (TSTIF). 2: TSTIF acts on the antigen-presenting cell to inhibit antigen-stimulated T-cell proliferation.

Culture supernatant (SN) was obtained from the monolayer of the MRL104.8a thymic stromal cell clone. This SN alone induced proliferation of helper T-cell (Th) clones because it contained IL-7. However, addition of the SN to cultures of Th stimulated with antigen plus antigen-presenting cells (APC) resulted in potent inhibition of their proliferation. This suppression was ascribed to a factor (designated thymic stroma-derived T-cell inhibitory factor, TSTIF) that is contained in the MRL104.8a SN and distinct from IL-7. TSTIF affected antigen-stimulated proliferation of both type 1 helper (Th1) and type 2 helper (Th2) T-cell clones. The TSTIF effect was also observed by the presence of the MRL104.8a SN only in the initial 24 hr pre-culture during the entire course (48-72 hr) of antigenic stimulation. Pre-exposure of Th cells to the SN in the absence of Ag/APC induced their proliferation upon stimulation with Ag/APC in the next 48 hr cultures. However, pre-cultures of Th cells with the SN in the presence of APC alone (without antigen) resulted in potent inhibition of the subsequent Ag/APC-stimulated proliferation. Interaction of TSTIF with APC but not with responding Th cells was further demonstrated in the following experiment: APC alone were exposed to the MRL104.8a SN and used for stimulation of Th that had not been exposed to the SN. Such an APC population exhibited a remarkably reduced capacity to induce antigen-stimulated Th proliferation when compared to that induced by freshly prepared APC or APC cultured in the absence of the MRL104.8a SN. These results indicate that TSTIF exerts its inhibitory effect on the antigen-stimulated T-cell proliferation by acting on APC.

Thymic stroma-derived T-cell inhibitory factor (TSTIF) 1. TSTIF induces inhibition of antigen-stimulated T-cell proliferation.

The present study investigates the capacity of the MRL104.8a thymic stromal cell clone to modulate T-cell growth. The culture supernatant (SN) from the MRL104.8a stromal cell monolayer was added to cultures of Th-clones with or without T-cell receptor (TCR) stimulation as provided by antigen (Ag) plus splenic antigen-presenting cells (APC). The results demonstrated that the MRL104.8a SN containing IL-7 activity induced dose-dependent proliferation of Th cells when they were not stimulated with Ag/APC. In contrast, addition of the same SN to cultures of Th cells during stimulation with Ag/APC resulted in potent dose-dependent inhibition of their proliferation. IL-7 contained in the SN was neither responsible for, nor involved in the inhibition event, because the inhibition was not observed with rIL-7 and was not neutralized by anti-IL-7 antibody. The growth inhibition of the Th clone in the presence of Ag plus APC was also induced by IL-10 or TGF-beta. However, the MRL104.8a SN-induced growth inhibition was mediated by a factor distinct from these cytokines, because (1) IL-10 cDNA was not amplified in polymerase chain reaction (PCR) products derived from MRL104.8a cells; (2) TGF-beta cDNA was detected in the PCR products, but only marginal levels of TGF-beta activity in an active form were found in the MRL104.8a SN and the SN-induced inhibition was not prevented by anti-TGF-beta antibody; and (3) addition of rIL-7 to antigen-stimulated cultures containing rTGF-beta or rIL-10 induced IL-7 mediated Th proliferation, whereas the MRL104.8a SN-induced inhibition was still observed in the presence of excess rIL-7. Moreover, this factor, designated thymic stroma-derived T-cell inhibitory factor, was found to have a m.w. of 20-25 x 10(3) and to exhibit heparin-binding property. Thus, these results indicate that the MRL104.8a thymic stromal cell clone produces a potentially novel factor that induces inhibition of antigen-stimulated T-cell proliferation.

Identification of IL-7-dependent bone marrow-derived Thy-1-B220- lymphoid cell clones that rearrange and express both Ig and T cell receptor genes.

Bone marrow stromal cell lines and lymphoid cell lines were co-established from the Whitlock-Witte type of long term liquid cultures of MRL/1 and C57BL/10 (B10) (Thy-1.1) bone marrow cells. The present study investigates the immunologic nature of parental and cloned lymphoid cell lines. Both strains of parental lines and their clones did not grow alone but proliferated on the monolayers of co-established parental stromal cell lines from a syngeneic or alternative strain. When various lymphokines or cytokines were tested for their capacity to support the growth of these lymphoid cell clones, only IL-7 could substitute for the growth-promoting function of stromal cells. These IL-7-dependent clones expressed neither Thy-1 nor B220 Ag. However, all of them from two strains were found to rearrange synchronously H chain of Ig as well as gamma chain of TCR genes. Some of the clones transcribed a mature size of IgH mRNA. Co-expression of mRNA for lambda 5 but not for IgL chain (kappa, lambda) genes resulted in the generation of cell surface mu chain in these clones. Other clones expressed a smaller size of IgH mRNA without exhibiting surface mu chain. Irrespective of the differences in IgH rearrangements and its mRNA expression, a mature size TCR gamma mRNA was detected in all of the clones. Thus, these results demonstrate the existence of untransformed (IL-7-dependent) immature lymphoid cells rearranging both Ig and TCR genes. Their unique features concerning cell surface markers (B220- mu+), specific growth factor requirement, and various modes of Ig/TCR gene rearrangements are discussed in the context of early lymphoid development.