Abstract
T cell unresponsiveness to Ag stimulation can be induced by several means. The precise mechanism by which this process occurs remains poorly understood. Preincubating T cells with either EDCI-fixed APC or ionomycin is a proven means of inducing T cell anergy with reduced IL-2 production in response to Ag stimulation. Using T cells from mice expressing the TCR transgene DO11.10, which is specific for a peptide (323–339) derived from hen egg OVA, we demonstrate that naive cells obtained directly from the host are resistant to the anergy induction by either fixed APC or ionomycin. TCR transgenic mice also deficient in the recombination-activating gene-2 (RAG-2−/−), preventing the formation of T cells with endogenous TCRs, were immunized with OVA, and in vivo activated T cells with low expression of CD62 were isolated. These primed cells possess the same sensitivity to ionomycin-induced anergy as in vitro activated cell lines. This unresponsive state most profoundly affects Ag-induced IL-2 production, with IFN-γ and IL-3 affected to a lesser degree and no effect observed on IL-4 production. Thus, T cells in vivo can be distinguished phenotypically by their susceptibility to anergic stimuli. Anergy so induced affects selected T cell functions.
Ag stimulation of T cells using chemically fixed APC renders them unresponsive to subsequent rechallenge with functional APC (1). Stimulation by an alternative means, such as using Abs against the TCR complex or class II molecules in planar lipid membranes, generates similar anergic states (2, 3). Whether the same pathway is used in each of these modes is unclear, but such experimental models have indicated that a second signal in addition to TCR engagement is required for activation and avoidance of anergy induction (4). More recent studies have suggested that B7-1 or B7-2 binding to CD28 may account for this second signal, and that stimulation of the TCR and engagement of the CD28 molecule are sufficient for T cell activation and avoidance of the unresponsive state (5, 6, 7).
Subsequent studies analyzing the biochemical events in T cells exposed to anergic stimuli have suggested that these unresponsive cells maintain persistently high intracellular Ca2+ levels (8). Furthermore, these investigations have demonstrated that increasing the intracellular Ca2+ concentration with ionophores such as ionomycin is sufficient to establish the unresponsive state generated by other anergic stimuli. Other biochemical and molecular events that have been observed in these cells include altered tyrosine phosphorylation patterns of intracellular proteins (9, 10), defective activation of the Ras pathway (11, 12), and altered trans-activating activity of IL-2-specific proteins, such as activating protein-1 (13).
T cell clones have primarily been used in these cited studies and are thus cells that have been preactivated with Ag in vitro and maintained in culture. Little is known regarding the ability of generating this same anergic state in freshly isolated, naive T cells. Anergy induction has been shown to require an initial proliferative response that transforms a naive T cell to a preactivated in vitro memory cell (14, 15, 16). Previous studies have attempted to demonstrate that freshly isolated T cells can be rendered anergic. However, most studies have incubated these cells several days in culture, which may, in fact, change their phenotype (17, 18, 19). Indeed, such cells have been shown to possess a phenotype of an activated state, with low expression of CD62 (MEL-14) (20). Unfortunately, the analysis of Ag responses in naive, freshly isolated populations is limited by the low frequency of T cells specific for any one particular Ag. Ideally, such studies should be performed analyzing a cell population with one receptor specificity. This would allow for a specific Ag response to be evaluated after challenges with various stimuli.
We have proceeded with such studies in mice that possess a transgene-derived Ag-negative TCR. This allows for the identification and isolation of a specific T cell population in which the response to a specific Ag can be monitored. The DO11.10, mouse whose transgenic receptor is specific for a single polypeptide (323–339) contained within hen egg OVA presented by I-Ad class II MHC molecules, was used for these studies (21). To ensure that cells with alternative TCRs would not affect these results, the transgene was introduced into a mouse line deficient in the recombination-activating gene-2 (RAG-2)3 gene (22), preventing the formation of T cells with endogenous TCRs. Using these mice, we have demonstrated that 1) naive T cells are resistant to anergy induction; 2) the in vivo activated T cell is also susceptible to anergy induction, indicating that this phenotypic change is not due to the in vitro culture conditions; and 3) ionomycin-induced unresponsiveness affects the production of IL-2, but other cytokines expressed, such as Ag-induced IL-3 and IFN-γ, are affected to a lesser degree, while IL-4 production remains unaffected.
Materials and Methods
Mice
The DO11.10 TCR transgenic mice were described previously (21) and bred at the Washington University School of Medicine (St. Louis, MO) in a pathogen-free facility with monitored surveillance and serologic testing of sentinel mice every 3 mo. The RAG-2−/− mice were a gift from Dr. Y. Shinkai (22) and were used to establish DO11.10 TCR RAG-2−/−, H-2d mice. DO11.10 TCR RAG-2−/− mice were identified as those mice possessing T cells bearing the transgene-derived receptor using the Ab KJ1-26 and lacking cells expressing B cell surface markers as determined by FACS analysis of peripheral blood. Male and female mice used in the experiments discussed below were between 6 and 8 wk of age.
Flow cytometry
Cell surface immunofluorescence analysis of T cell populations was performed as described previously (10). In brief, 1 × 106 cells were incubated with the Ab at saturating concentrations at 4°C for 30 min, washed, and further incubated with FITC-, phycoerythrin-, or tricolor-conjugated appropriate secondary reagents for an additional 30 min. Control samples were prepared in the same manner, but without primary Ab or with an isotype-matched control. Samples were analyzed by a FACScan analyzer (Becton Dickinson, Mountain View, CA) using the CellQuest program. The mAbs used include KJ1-26, specific for an Id expressed on the transgene-derived TCR; DO11.10 (21); and MEL-14 (anti-CD62, PharMingen, San Diego, CA). Lymph node cells from OVA-primed mice were stained with KJ1-26 and biotinylated MEL-14, with a goat anti-mouse FITC secondary Ab (Caltag, San Francisco, CA) for KJ1-26, and with phycoerythrin-avidin secondary staining for MEL-14. CD62 low KL1-26-positive cells were separated by FACSVantge (Becton Dickinson). The resulting populations were >98% pure.
In vitro cultures
In vitro cultured T cell lines were established by culturing 2 × 107 spleen cells from transgenic mice in DMEM with 5% FCS and OVA at a final concentration of 1 mg/ml for 10 days.
Induction of anergy with fixed APC
Preparation of fixed APC was performed under previously described conditions (1). Briefly, BALB/c splenocytes were irradiated at a dose of 2600 rad and were incubated for 1 h on ice in 0.44 ml of 0.9% NaCl containing 75 nM EDCI (Calbiochem, La Jolla). The cells were washed extensively in serum-free medium to stop the coupling reaction. Normal APC consisted of BALB/c splenocytes irradiated at a dose of 2600 rad. T cells (2 × 106 cells) were incubated with EDCI-treated or normal APC (10 × 106 cells) in the presence of 50 μM cOVA323–339 peptide in 24-well plates (Corning, Corning, NY) in a total volume of 2 ml of DMEM with 5% FCS for 16 to 20 h. Cells were harvested and washed with medium three times, and 2 × 104 recovered cells were stimulated with normal APC (5 × 105) alone or with APC plus OVA (1 mg/ml) or PMA (10 ng/ml; Sigma Chemical Co., St. Louis, MO) and ionomycin (1 μM; Calbiochem, La Jolla, CA) in DMEM with 5% FCS a final volume of 200 μl in 96-well round-bottom plates (Costar, Cambridge, MA). After 24 h of culture, supernatants were harvested, and the presence of cytokine was measured.
Induction of anergy with ionomycin
Cells (1 × 106) were incubated with ionomycin (Calbiochem) at a final concentration of 1 μM for 20 h in DMEM with 5% FCS in a volume of 2 ml. Recovered cells were stimulated in the same manner as fixed APC-stimulated T cells for cytokine production.
Proliferation and lymphokine assays
IL-2-dependent T cell proliferation was measured by cultured T cells (2 × 104) in a final volume of 200 μl of DMEM with 5% FCS containing IL-2 at a final concentration of 10 U/ml in 96-well flat-bottom plates (Costar). The cells were incubated at 37°C for 24 h and harvested following a 6-h incubation with tritiated thymidine (New England Nuclear, Boston, MA) at 1 μCi/well. IL-2 activity was assayed using the IL-2-dependent CTLL line (23). IL-4 was measured using the IL-4-dependent cell line, 6I4, developed in this laboratory. IL-3 was assayed using the IL-3-dependent line FDC-P3 (24). IFN-γ activity was determined using the inhibition assay with the indicator line WEHI-279 (25). Units are defined as reciprocal dilution of culture supernatants that gave 50% of the maximum activity.
Results
Naive T cells are resistant to anergy induction by EDCI-fixed APC
Treatment of T cell clones with stimuli such as anti-CD3 Abs or Ag presented by chemically fixed APC results in the generation of an anergic state. T cells in this state are not capable of producing IL-2 upon challenge with Ag and functional APC. The effects of such treatments on freshly isolated naive T cells have not been clearly demonstrated. To facilitate this analysis, we used T cells from mice possessing a transgene-derived TCR (DO11.10) that is specific for OVA323–339 peptide presented by I-Ad. Most of this mouse line’s peripheral T cells possess the transgenic TCR, providing a naive T cell population whose function can be analyzed. Freshly isolated, naive T cells from DO11.10 transgenic mice and cell lines generated from in vitro activated T cells from this same mouse strain were incubated with fixed APC plus peptide for 16 to 20 h. The cells were washed and rested for 3 days. Surviving T cells were counted and stimulated with Ag and APC. The supernatants derived from these cultures were then assayed for IL-2 production. Surprisingly, naive T cells treated with fixed APC were capable of producing amounts of IL-2 comparable to those produced by untreated T cells (Fig. 1⇓A). In contrast, in vitro cultured cell lines became anergic with exposure to Ag presented by fixed APC, with a failure to produce IL-2 upon rechallenge with Ag and functional APC. This experiment was performed three times, and in each instance, IL-2 production, in response to Ag stimulation, was maintained in the naive T cell population, while IL-2 production was virtually undetectable in the in vitro activated cell line after ionomycin treatment. This unresponsive state induced in the T cell lines was not due to decreased cell viability, as they were still able to proliferate when the treated cells were incubated with exogenous IL-2 (Fig. 1⇓B). Naive, untreated T cells that did not express significant amounts of the IL-2R, as expected, did not proliferate in response to IL-2 exposure. However, naive T cells treated with Ag and fixed APC did respond to IL-2 exposure (Fig. 1⇓B). This indicates that the Ag presented by the fixed APC induced a change in the naive cells, making them responsive to the cytokine. Thus, naive T cells appear to be fundamentally different from T cell clones in their responses to Ag presented by fixed APC and their susceptibility to become anergic by such means.
Differential effects of fixed APC plus Ag on naive and in vitro cultured T cells. Naive T lymphocytes isolated from TCR transgenic mice and T cell lines were exposed to medium alone (open circles) or with fixed APC and Ag (solid circles) as described in Materials and Methods. A, Recovered cells were stimulated with BALB/c APC with the indicated dose of antigenic peptide for 24 h. The cultures supernatants (final concentration, 50% of the original culture) were assayed for IL-2 activity. The data shown represent thymidine incorporation of the IL-2-dependent cell line, CTLL-2. B, IL-2 responsiveness of T cells exposed to fixed APC was measured by culturing T cells recovered after initial treatment with exogenous IL-2 (10 U/ml) for 24 h. Thymidine incorporation per culture was measured as described in Materials and Methods.
Differences in response to ionomycin treatments of naive T cells and T cell lines
Previous studies have demonstrated that anergy induction via chemically fixed APCs involves a rise in intracellular Ca2+ (8). Furthermore, this anergic state can be reproduced by raising intracellular Ca2+ levels in susceptible T cell clones with ionophores such as ionomycin. However, used in the context of additional stimuli, Ca2+ influx can result in distinctly different responses in the cell. This is illustrated in Table I⇓. In vitro cultured cell lines derived from TCR transgenic mice were stimulated with PMA, ionomycin, and the combination of PMA and ionomycin. Alone, PMA and ionomycin were insufficient to elicit significant IL-2 production, while together they elicited a vigorous response. However, ionomycin and PMA had drastically different effects on the cell when the treated cells were subsequently challenged with Ag and appropriate APC. The ionomycin-treated T cell line failed to generate measurable IL-2 in response to Ag stimulation. PMA, in contrast, had no significant effect on the capacity of T cells to respond to the secondary stimulation. Thus, although both PMA and ionomycin failed to elicit any IL-2 production from the T cells when used alone, ionomycin, but not PMA, induced a unique change in the state of the cell to make it unresponsive to antigenic stimulation. This effect, however, was dependent upon the state of the cell, as naive freshly isolated T cells maintained their ability to produce IL-2 with Ag stimulation even after ionomycin treatment (Table I⇓). Naive cells, however, still appeared to be influenced by ionomycin, as the magnitude of the secondary response was greater than that of the response of the untreated population. This indicates that ionomycin induces a distinctly different state in the naive cell from that in the in vitro cultured T cell. Ionomycin was used in all subsequent experiments to further characterize the difference between naive and activated T cells. This method is cell density independent and thus excludes the possibility that the resistance of the naive cells to the fixed APC treatments is due to differences in engagement at the TCR level. The differences observed in this experimental system are more likely due to inherent differences in the T cell populations in their susceptibility to anergy induction.
Comparison of anergy induction between naive T cells and in vitro cultured T cell lines, with ionomycin treatment a
In vivo primed T cells are analogous to T cell lines in their susceptibility to anergy induction
Most studies examining the effects of anergy-inducing stimuli have used T cell lines maintained in culture for at least several days. Since freshly isolated T cells are resistant to anergy induction, one explanation is that the maintenance of cells in culture renders them susceptible. Alternatively, if ionomycin sensitivity is a property acquired with T cell activation, primed T cells isolated directly from the immunized host should possess the same sensitivity to ionomycin as that observed in the T cell line. To address the susceptibility of in vivo primed T cells to anergy induction, it was necessary to acquire a means by which uniform cell populations could be isolated from the host in both primed and unprimed states. Ideally, this would entail the identification of a cell population with a single TCR and would exclude those cells that may possess alternative receptors. To accommodate these needs, DO11.10 mice were bred with mice who were homozygous for the disruption of the RAG-2 gene, (RAG-2−/−) (22). This prevents the rearrangement of endogenous receptor genes and assures that all the T cells in the periphery possess a single, uniform TCR. These mice were used in the subsequent analysis.
DO 11.10 TCR RAG-2−/− transgenic mice were immunized in the footpad with OVA (100 μg) in CFA. The T cells from the draining lymph nodes were isolated 7 days after immunization. The in vivo primed T cell population was analyzed by surface immunofluorescence and compared with cells from nonimmunized mice. Lymph node cells from naive mice were stained with an anti-Id Ab (KJ1-26) and counterstained with an anti-CD62 Ab (MEL-14). As shown in Figure 2⇓, the vast majority of TCR-positive cells in the naive mouse express high levels of CD62 (L-selectin). Upon priming with Ag, the number of cells in draining lymph node as well as in spleen increased significantly (Table II⇓), and the expansion of Ag-reactive T cells accounted for this increase as the cells expressed the transgene-derived receptor, as measured by the Ab KJ1-26 (Fig. 2⇓). A significant fraction of T cells from immunized mice were now low for CD62 expression, consistent with the findings of other studies characterizing the surface phenotype of activated T cells (26, 27).
Surface CD62 expression in naive, in vivo activated, and in vitro activated T cells. Lymph node cells from nonimmunized TCR transgenic mice on a RAG-2-deficient background (naive) and from Ag-primed DO 11.10 RAG-2−/− mice (primed lymph node) and in vitro cultured T cells from the same mouse strain (cell line) were stained with anti-idiotypic Ab to TCR (KJ1-26) and anti-CD62 Ab (MEL 14). Stained cells were analyzed by FACScan. The control consisted of lymph node cells stained with secondary reagents alone. The percentage of the total cells is indicated for each quadrant when more than one population exists.
Comparison of total cell number in the lymph node and spleen in naive and primed DO11.10 Rag-2 −/− micea
To determine whether the in vivo activated population had acquired sensitivity to ionomycin-induced anergy, the low CD62 population was purified using FACS, so that >98% of the cells were low for expression of CD62 yet still expressed the transgene-derived TCR. This purified T cell population was compared with freshly isolated naive cells, which were stained with the same Abs as the in vivo primed cells, and the in vitro cultured T cell line, all derived from the DO11.10 TCR RAG-2−/− mouse line for their sensitivity to ionomycin. The three cell populations were incubated with ionomycin for 16 to 20 h, washed, and rechallenged with Ag and APC. Consistent with the previous studies, the in vitro cultured T cell line was rendered unresponsive with ionomycin treatment, while the naive T cell remained reactive after the same treatment (Table III⇓). These cells were still viable and capable of IL-2 production upon stimulation with PMA and ionomycin. Thus, the unresponsive state elicited by ionomycin was not due a toxic effect on the cells. When the in vivo primed, low CD62 T cells were exposed to ionomycin and tested for their Ag reactivity, their capacity to produce IL-2 in response to Ag was greatly reduced compared with the that of the same T cell population not exposed to ionomycin. This experiment was performed three times, and in each case, ionomycin treatments of the in vivo primed, low CD62 population reduced the amount of IL-2 production in response to Ag stimulation by at least 1 order of magnitude. This effect was simply not due to the effects of the Abs used to isolate the cells, since similar treatments of the naive population failed to affect their resistance to ionomycin. Thus, in vivo activated T cells are sensitive to ionomycin-induced anergy induction in a fashion analogous to that of the in vitro activated T cell line.
Similar sensitivities of in vivo-primed and in vitro-stimulated cell lines to ionomycin-induced anergy a
The susceptibility to anergy induction is confined to IL-2 expression
Further analysis was performed to determine whether other cytokine production pathways were affected by ionomycin in the same fashion as IL-2. In vitro cultured T cells were incubated with ionomycin for 16 h and stimulated with Ag and APC. Culture supernatants were tested for IL-2, IL-3, IL-4, and IFN-γ. Similar to the previous assay, ionomycin-treated cells produced no IL-2 upon restimulation with Ag and APC. However, IL-4 production remained comparable to that of untreated T cells. In contrast, the amounts of IFN-γ and IL-3 produced by ionomycin-treated cells were significantly lower than those produced by nontreated cells, but were still readily detectable (Table IV⇓). This experiment was performed three times. In each instance, IL-4 production after Ag stimulation did not decrease in response to ionomycin treatment, while the effects of ionomycin on IFN-γ and IL-3 expression failed to reduce the response more than 3- to 10-fold, with detectable activity remaining. This is in contrast to IL-2 production, for which ionomycin treatments resulted in no measurable biologic activity in the in vitro cell line after Ag stimulation. Thus, each of these cytokines must be regulated in a different fashion than IL-2, and anergy induction seems to be specific for IL-2 and IL-2-dependent proliferation of T cells in vitro.
Ionomycin’s anergy induction primarily affects IL-2 expression in in vitro-cultured T cell lines with little influence on other cytokinesa
Discussion
We have demonstrated that the naive T cells are distinctly different from preactivated T cells in their susceptibility to anergy induction. In vitro cultured T cells can be rendered anergic by either Ag presented by fixed APC or by exposure to ionomycin. These T cells fail to produce IL-2 or proliferate upon rechallenge with Ag presented by functional APC. The treated cells are still viable, as they proliferate in response to exogenous IL-2 and produce IL-2 in response to PMA and ionomycin. In contrast, naive T cells are resistant to such anergy induction and maintain their ability to generate IL-2. This resistance is not due to a lack of signaling of the T cells with these stimuli, since naive T cells exposed to such stimuli exhibit an increased reactivity to IL-2 and greater cytokine-producing capacity than nontreated naive T cells. This functional change indicates that events elicited in the naive cell by the anergy-inducing stimuli are distinct from those induced by in vitro activated T cells.
The data presented above indicate that naive cells can be converted into ionomycin-susceptible cells with conventional immunization. This was demonstrated in TCR transgenic mice on a RAG-2−/− background in which all the T cells expressed only one receptor. After priming, this uniform T cell population expanded in significant numbers in response to the Ag. A large fraction of cells expressed low levels of CD62 (L-selectin), but a significant number of T cells remained high in CD62 expression. Since the low CD62 population alone cannot account for the entire increase in the number of T cells, some T cells were stimulated by Ag and maintained high CD62 expression. We observed similar heterogeneity for IL-2R and CD44 (pgp-1) expression (data not shown). It is not clear at present whether this heterogeneity represents a distinct T cell activation state. However, it is likely that this heterogeneity represents different stages of the T cell as it undergoes Ag-driven proliferation. Alternatively, the increase in the total number of high CD62 cells could be due to the migration of T cells into the lymphoid tissues that have not yet encountered Ag and lowered their expression of CD62. Thus, to restrict our analysis to an Ag-primed population, we have used exclusively low CD62 T cells and have demonstrated that the conversion from a naive to a primed T cell renders the cell susceptible to ionomycin-induced anergy. The relationship among surface phenotypic changes, cell cycle regulation, and susceptibility to anergy induction in primed mice requires further investigation to help elucidate a biochemical basis for the anergy susceptibility.
Other investigators have noted previously that naive cells appeared to be uninfluenced by the effects of fixed APC treatments (28). These studies were unable to clearly demonstrate differences between primed and naive cells in IL-2 production following exposure to fixed APCs and concluded that the means by which anergy is induced may only affect certain T cell subtypes. Our assay systems, however, were able to detect clear differences between naive and primed T cells in their susceptibility to anergy induction. By using cell sorting for a low CD62 population to isolate in vivo primed T cells and by allowing the in vitro activated cell line to incubate for a sufficient amount of time in culture, we were able to analyze more uniform populations that were devoid of cells that may be at different stages of activation and thus may possess different susceptibilities to anergic stimuli. In addition, the use of ionomycin as the means of inducing anergy ensured that the T cells assayed were uniformly treated so that the response assessed was not dependent upon the degree of engagement of the cell’s receptor with a fixed APC, which may fail to induce anergy. Our data suggest that selecting for cells expressing low CD62 expression is a means of identifying a T cell population that is sensitive to the effects of anergic stimuli.
Ionomycin treatments have different effects on different cytokine production pathways. IL-2 production is dramatically reduced with subsequent Ag stimulation, while IL-4, in contrast, appears to be unaffected. Ionomycin reduced the responses of IFN-γ and IL-3 production to Ag, but its effects on these two cytokines were incomplete. The T cell lines used in these experiments were only stimulated for 7 days and were thus heterogeneous. Cells within this mixture predominantly consist of CD4 Th cells expressing Th1 (IFN-γ) or Th2 (IL-4) cytokines at various stages of differentiation. Since IL-2 can be generated by cells that produce either Th1 or Th2 (29) cytokines, ionomycin must have influences on more than one population, since no significant IL-2 activity can be measured following such treatments.
One possibility accounting for the different effects of ionomycin on the expression of different cytokines is that ionomycin may render activated T cells unresponsive to Ag-induced protein kinase C (PKC) activation. Previous investigations have demonstrated the requirement for PKC activation for IL-2 expression, while other cytokines, such as IL-4, lack this dependence (30, 31, 32) This lack of kinase activation is not irreversible, since PMA and ionomycin can still induce IL-2 from anergized T cells. It is unclear whether a cytokine’s dependence on PKC activation directly determines its sensitivity to anergic stimuli. The specific biochemical events required for anergy induction remain to be defined and thus may involve additional pathways distinct from PKC activation and Ca2+ influx (33) Despite this possibility, our investigations have demonstrated that T cells can possess different susceptibilities to anergy induction, and the use of T cells that resist anergy induction would provide a useful tool for further investigation of the biochemical nature of this process.
Our findings presented in this report do not address the critical steps of tolerance induction in vivo or whether prior activation is required to induce tolerance in vivo. Furthermore, the physiologic role of anergy defined by a lack of IL-2-producing capacity in vivo is not clearly understood. The observations identify two distinct states for the T cell: the naive, anergy-resistant state and the primed, anergy-susceptible state. Further characterization of the two states and the molecular event that maintains each state will provide a better understanding of immune regulation.
Footnotes
-
↵1 This work was supported by National Institute of Health Grants K11AI01108 and RO1-AI30803.
-
↵2 Address correspondence and reprint requests to Dr. Robert J. Hayashi, Department of Pediatrics, Washington University School of Medicine, Box 8116, 660 South Euclid, St. Louis, MO 63110.
-
↵3 Abbreviations used in this paper: RAG-2, recombination-activating gene-2; PKC, protein kinase C.
- Received June 17, 1997.
- Accepted September 12, 1997.
- Copyright © 1998 by The American Association of Immunologists
References
In this issue
