SUPPRESSION MEDIATED BY ANERGIC CD4⁺ T CELLS REQUIRES STIMULATION BY MHC-PEPTIDE COMPLEXES AND CAN BE INDUCED IN THE PRESENCE OF COSTIMULATION¹

Abstract

Background. 

In this study, we have investigated the mechanisms involved in both the induction of suppressive anergy, the stability of the anergy induced, and the possible mechanisms by which the response of immunocompetent CD4⁺ T cells are suppressed.

Methods. 

We used immobilized anti-CD3 monoclonal antibody (mAb) to induce anergy in T helper (Th) 1 and Th0 cells reactive with MHC class II molecule H2 I-A^b.

Results. 

We observed that suppressive anergy was induced independently of costimulation in Th0 but not Th1 cells. Although the anergic and suppressive states of Th0 cells were stable in the presence of exogenous interleukin-2, this was not the case for Th1 cells. No evidence for linked epitope suppression was observed for any of the I-A^b reactive cells investigated. Neither anergy nor suppression was observed in Th0 cells upon restimulation with anti-CD3 in the presence of syngeneic antigen-presenting cells (APCs). However, anergy but not suppression was observed in co-cultures restimulated with anti–T-cell antigen receptor (TCR) mAbs/syngeneic APCs and suppression could be restored by the addition of I-A^b+ APCs.

Conclusions. 

Overall, these data suggested that the MHC-peptide complex recognized by the Th0 cells was required for suppression of the response of immunocompetent cells. We propose that suppression is mediated either by down-modulation of the MHC-peptide complex recognized by the anergic T cells or that a molecule specific to the MHC-peptide/TCR interaction facilitates negative regulation by APC:T or T:T interactions.

Potentially self-reactive T cells are subject to mechanisms of central and peripheral tolerance during the generation of the T-cell repertoire. T cells that escape deletion by negative selection in the thymus are further controlled in the periphery by four fundamental mechanisms that include clonal deletion ^(1–3), ignorance ⁽⁴⁾, clonal anergy ^(5,6), and regulation/suppression ^(7–9). Recently, in vitro and in vivo studies have confirmed a link between two of these principal mechanisms, regulation/suppression and anergy ^(10–13). Suppressive anergy in T cells can be generated in vitro using either anti-CD3 and clonotypic antibodies or after T-T cell stimulation ^{(10,12,14–16)}. These strategies are thought to induce anergy by stimulation of the T-cell antigen receptor (TCR) (signal 1) in the absence of costimulatory signals through CD28/B7 interactions (signal 2) ^{(14,15,17,18)}. However, the absence of costimulation is not an absolute requirement because in vitro anergy has been induced in the presence of costimulation in some systems ^(16,19,20). Alternative mechanisms have been proposed to account for the induction of the anergic state under costimulatory conditions including the generation of anergic cells in the presence of a productive immunological response ⁽²⁰⁾ and TCR occupancy in the presence of costimulation but in the absence of proliferation ⁽²¹⁾. Further evidence for the latter was provided by studies that investigated the effect of the immunosuppressive drug rapamycin on T-cell proliferation and anergy induction ⁽²²⁾.

Anergic T cells have been shown to suppress the response of immunocompetent T cells from the same clone in vitro, and there is some evidence that anergic T cells generated in vitro can mediate their suppressive effect after adoptive transfer in vivo ⁽¹²⁾. In the latter case, the survival of skin grafts, mismatched for a single amino acid in a donor MHC class II molecule, was prolonged. To use in vitro anergized T cells in vivo, the stability of the anergy induced and the potential of the anergized T cells to suppress T cells that respond to other alloantigens present in the graft (linked epitope suppression) are of critical importance ^(23–27).

How anergic T cells regulate the responses of immunocompetent cells is unknown. It has been suggested that anergic T helper (Th) 1 cells facilitate immunosuppression via competition with immunocompetent cells for locally produced growth factors such as interleukin (IL) 2 or ligands on the antigen-presenting cell (APC) surface, a phenomenon known as bystander suppression ^(10,23,28). It has also been proposed that anergic T cells exert their suppressive effects in an active and antigen-specific manner, which is dependent on cell-cell contact between the anergic T cell and the APC and is not mediated via soluble factors ⁽²⁷⁾.

In this study, we investigated the mechanisms involved in both the induction of suppressive anergy, the stability of the anergy induced, and the possible mechanisms by which the response of untreated alloreactive CD4⁺ T cells are suppressed via the examination of the properties of Th1 and Th0 cells reactive with MHC class II molecule H2 I-A^b. We observed that suppressive anergy was induced independently of costimulation in Th0 but not Th1 cells. Moreover, although the anergic and suppressive states of Th0 cells were stable in the presence of exogenous IL-2, this was not the case for Th1 cells. No evidence for linked epitope suppression was observed for any of the I-A^b reactive cells investigated. Neither anergy nor suppression was observed in Th0 cells upon restimulation with anti-CD3 monoclonal antibody (mAb) in the presence of syngeneic APCs. However, anergy but not suppression was observed in co-cultures restimulated with anti-TCR mAbs/syngeneic APCs and suppression could be restored by the addition of I-A^b+ APCs. Overall, these data suggest that the MHC-peptide complex recognized by the Th0 cells was required for suppression of the response of immunocompetent cells from the same clone in vitro.

MATERIALS AND METHODS

Mice.

CBA.Ca mice (H2^K) and C57BL/10 mice (H2^b) were originally purchased from Harlan Olac, Ltd. (Bicester, U.K.) and were maintained in the Biomedical Services Unit at the John Radcliffe Hospital Oxford.

mAbs and reagents.

The anti-CD3 mAb 145–2C11 ⁽²⁹⁾, anti-TCR mAb H57–597, and anti-CD28 mAb 37.51 ⁽³⁰⁾ were purified from tissue culture supernatants with high-performance liquid chromatography. Eluted antibodies were dialyzed against phosphate-buffered saline (PBS). Phycoerythrin-labeled CD4 mAb was purchased from Pharmingen (Beckton Dickinson UK Ltd., Oxford). Tissue culture supernatant was grown from hybridomas that secreted antibodies against mouse TCR (H57–597), Vα2 (B20.1), Vβ6 (44.22.1), Vβ8 (F23.1), and Vβ10 (B21.5 and KT 10-b2), which were generously donated by Elisabeth Simpson (MRC Clinical Sciences Centre, Hammersmith Hospital, London, U.K.).

The establishment and maintenance of T-cell lines and clones.

CD4⁺ T cells (2×10⁷ in 4 ml) from CBA.Ca mice (H2^K) were initially cultured with irradiated (3000 rad) C57BL/10 (H2^b) splenocytes (2×10⁷, depleted of red blood cells). The bulk cell suspensions were restimulated with irradiated C57BL/10 splenocytes (1:1) every 12 days, and at the third stimulation, concanavalin A supernatant (ConA-SN, 20 U/ml IL-2) was added. ConA-SN was prepared from Lewis rats (Harlan Olac Ltd., Bicester, U.K.) as previously described. At the completion of the third cycle, a limiting dilution was performed at 10 cells/well for cell lines and 0.3 cells/well for T-cell clones in a flat-bottomed 96-well plate. Initially, 4×10⁵ C57BL/10 splenocytes were placed per well with ConA-SN and then restimulation was performed fortnightly with 2×10⁵ irradiated C57BL/10 splenocytes and ConA-SN. The increasing responder T-cell population was then expanded into larger volumes with a proportional increase in stimulator numbers.

Characterization of T-cell clones and lines.

The Th1 cells used in this study were derived from CBA.Ca (H-2^k) mice and are I-A^b specific (present on C57BL/10 splenocytes) and TCRαβ⁺, CD3⁺, CD4⁺, and CD28⁺ positive. The clone TE5 is Vβ10⁺, and MD6 is a cell line with unknown Vα-Vβ status. Intracellular cytokine analysis of TE5 and MD6 have revealed a Th1 cytokine-secreting phenotype with production of interferon (IFN) γ predominantly but no detectable IL-4 after stimulation with pharmacological agents phorbol myristate acetate (PMA) (Sigma, St. Louis, MO) and ionomycin (Sigma, St. Louis, MO). Analysis of the cytokine pattern with reverse transcription–polymerase chain reaction (RT-PCR) for mRNA detection in cell lysates and standard sandwich ELISA for supernatants of tissue cultures after allogeneic stimulation confirms this phenotype by the detection of IFN-γ in the absence of IL-4 and IL-10. The Th0 clones used were derived from CBA.Ca (H-2^k) mice and are I-A^b specific and TCRαβ⁺, CD3⁺, CD4⁺, and CD28⁺ positive. The Th0 clones (AB, Vα2⁺ and Vβ6⁺; PW, Vβ8⁺) produce IFN-γ, IL-4 and IL-10 as seen by standard sandwich ELISA, RT-PCR, and intracellular cytokine analysis after PMA and ionomycin stimulation.

In vitro proliferation assays.

The medium comprised RPMI1640 supplemented with 10% fetal calf serum (PAA Laboratories, Linz, Austria), 2 mM l-glutamine (PAA Laboratories, Linz, Austria), 10 mM HEPES buffer (PAA Laboratories, Linz, Austria), 0.05 mM 2-mercaptoethanol (Sigma, St. Louis, MO), and 100 U/ml each of penicillin and streptomycin (Gibco, Paisley, Scotland). Clones (1–2×10⁴/well) were cultured with C57BL/10 splenocytes (2–5×10⁵/well) as standard control cultures in round-bottomed 96-well plates (Costar, Corning, NY). Suppression co-cultures contained the same number of untreated clones (1–2×10⁴/well) and differing numbers of anergic clones with irradiated C57BL/10 splenocytes (2–5×10⁵/well).

Induction of anergy with immobilized anti-CD3.

Purified anti-CD3 mAb 145–2C11 was diluted in PBS to a concentration of 10 μg/ml and then added to 6-well plates at room temperature for 2 hr. This was followed by washing three times with PBS and blocking with 1% bovine serum albumin in PBS for 2 hr at RT. The plates were used directly or stored at −20°C. Clones (2.5×10⁶) were incubated on plate-bound anti-CD3 for 2 days; this was followed by at least 2 days of rest before restimulation ⁽¹⁸⁾.

RESULTS

Suppressive anergy can be induced in Th0 and Th1 cells by immobilized anti-CD3.

First, we determined whether H2 I-A^b–specific CD4⁺ Th1 (TE5 and MD6) and Th0 (AB and PW) T-cell lines and clones rendered unresponsive by incubation with immobilized anti-CD3 could suppress the responses of untreated T-cell clones and lines. To investigate whether incubation with immobilized anti-CD3 alone (10 μg/ml) was sufficient to induce anergy, Th1 and Th0 cells were cultured with anti-CD3 for 2 days. Viable T cells were isolated and rested for 2 days in the absence of APCs to allow for re-expression of the TCRs. We confirmed by fluorescence-activated cell sorter analysis that the TCR-CD3 complex was re-expressed to levels similar to those expressed by untreated cells (data not shown). After the resting period, Th1 cells, TE5 and MD6, and Th0 cells, AB and PW, failed to proliferate to subsequent stimulation with I-A^b+ APCs (irradiated C57BL/10 splenocytes) indicative of the induction of anergy (Fig. 1). Moreover, in co-culture experiments, anergized cells suppressed the responses of untreated T cells to I-A^b+ APCs in a dose-dependent fashion (Fig. 1). Anergy and suppression were most profound in the H2 I-A^b–specific CD4⁺ Th0 clone PW with proliferative responses to I-A^b+ APCs reduced by as much as 90%. We concluded that incubation with immobilized anti-CD3 induced anergy and the ability to suppress the response of untreated cells in both Th1 and Th0 cells.

Figure 1: Anergic alloreactive CD4⁺ Th0 and Th1 cells suppressed the responses of untreated Th0 and Th1 cells in vitro. CD4⁺ Th0/Th1 cells (1×10⁶/ml) were incubated with immobilized anti-CD3 mAb (10 μg/ml) for 48 hr and then rested in the absence of APCs for an additional 2 days. Untreated (U, grey bars) and anergic cells (A, black bars) (2×10⁴/well) were cultured alone (−APC) or in the presence of irradiated C57BL/10 splenocytes (+APC) (5×10⁵/well) for 3 days. To determine suppression, anergic cells were co-cultured with untreated cells (2×10⁴/well) at a ratio of 1:1 (1A:1U) and 2:1 (2A:1U) in the presence of irradiated I-A^b+ APCs (C57BL/10 splenocytes) (5×10⁵/well). Tritiated thymidine (0.5 μCi) was added for the final 16 to 18 hr. The results shown for the Th0 T-cell clones, AB (A) and PW (B), and Th1 T cells, MD6 (C) and TE5 (D), are representative of more than three independent experiments.

Suppression develops independently of costimulation in anergic Th0 but not Th1 cells.

Traditionally, it has been proposed that anergy is induced by stimulation through the T-cell receptor alone without accessory signals from costimulatory molecules and cytokines such as CD28 and IL-2 ^(31–35). More recently, it has been shown that anergy can develop by stimulation with altered peptide ligands or with superantigen expressed on the surface of live APCs ^(19,36,37). We wanted to investigate whether stimulation through the TCR complex in the absence of costimulation was essential for the generation of anergic cells capable of suppressing untreated cells from the same clone. Th1 and Th0 cells were incubated with plate-bound anti-CD3 in the presence or absence of costimulatory signals (soluble anti-CD28 or irradiated syngeneic APCs). These costimulatory conditions were sufficient to induce improved cell survival of Th1 and Th0 cells over T cells treated with anti-CD3 alone (data not shown). The anergic and suppressive properties of the Th0 cells (AB and PW) were not affected when anti-CD28 or irradiated syngeneic APCs were added to the induction cultures (Fig. 2, A, B, and E). In contrast, costimulation altered the threshold for suppression in Th1 cells. Additional anergized Th1 cells were needed to achieve the same level of suppression when soluble anti-CD28 was added to the induction culture (Fig. 2, C and D). In conclusion, the generation of suppressive anergy in Th0 cells by high doses of immobilized anti-CD3 was independent of costimulation.

Figure 2: Induction of anergy/suppression in Th0 clones is independent of costimulation. Th0 and Th1 cells were treated with immobilized anti-CD3 (as previously described) in the presence (white bars) or absence (black bars) of soluble anti-CD28 (10 μg/ml) (A–D) or syngeneic APCs (1×10⁶/ml) (E). Cells were harvested after 2 days and rested for a subsequent 2 days in the absence of any APCs or growth factors. They were then tested for their ability to respond to stimulation with irradiated I-A^b+ APCs alone (+APC) or their ability to suppress the response of untreated cells (2×10⁴/well) in co-culture at a range of ratios of anergic cells:untreated cells (from 1:1 to 4:1, 1A:1U to 4A:1U) and compared to the response of untreated cells (U+APC, grey bars). Tissue culture plates were harvested after 3 days, pulsing with tritiated thymidine for the final 16 to 18 hr. Data shown for anti-CD28 mAb treatment of Th0 clones, PW (A) and AB (B), are representative of five experiments and for Th1 cells, MD6 (C) and TE5 (D), of two experiments. Stimulation of PW with immobilized anti-CD3 and syngeneic APCs (E) has been repeated twice; the same results have also been seen with AB (data not shown).

IL-2 reversed suppressive anergy in Th1 but not in Th0 cells.

Previous studies have demonstrated that T cells anergized in vitro are anergic upon restimulation by professional APCs but hyperresponsive in the presence of APCs and IL-2 ^(18,27,38). The addition of exogenous recombinant IL-2 (50 U/ml) to cultures of anergic Th1 or Th0 cells cultured with I-A^b+ APCs induced proliferation in both populations (Fig. 3). To determine whether proliferation of anergic T cells on addition of IL-2 abrogated the ability of anergized Th0 and Th1 cells to suppress, we cultured the anergic T cells with a high dose of recombinant IL-2 (50 U/ml) for 3 days before restimulation. Subsequent restimulation cultures showed that the Th1 cell, MD6, was responsive to stimulation with I-A^b+ APCs and was incapable of suppressing an untreated Th1 response (Fig. 4 A). In contrast, the Th0 clone, PW, remained unresponsive and suppressive after stimulation with I-A^b+ APCs (Fig. 4 B). Therefore, although exogenous IL-2 was able to reverse the suppressive effect of anergic Th1 cells, Th0 cells maintained their ability to suppress the response of untreated T cells under the same conditions.

Figure 3: Exogenous IL-2 restores proliferation in anergic Th1 and Th0 T cells. Th0 and Th1 T cells (2×10⁴/well) were anergized by immobilized anti-CD3 mAbs and then cultured with irradiated I-A^b+ APCs (A+APC) (5×10⁵/well) in the presence or absence of exogenous recombinant rat IL-2 (A+APC+IL-2) (50 U/ml) and compared to the response of untreated T cells from the same clone (U+APC, grey bar). Cultures were maintained for 3 days, including the addition of tritiated thymidine for the final 16 to 18 hr. The data shown are representative of seven independent experiments.

Figure 4: Exogenous IL-2 reverses anergy and suppression in Th1 cells but not Th0 clones. Th0/Th1 T cells were anergized as previously discussed. Anergic cells (1×10⁶/ml) were then cultured with exogenous recombinant rat IL-2 (50 U/ml) in the absence of any APCs for 3 days and then rested for 24 hr. Th1 cell line MD6 (A) and Th0 T cell clone PW (B) were then cultured alone (2×10⁴/well) (−APC) with irradiated I-A^b+ APCs (5×10⁵/well) alone (+APC, white bar) or in co-culture with untreated T cells from the same clone at a ratio of 1:1 (1A:1U, black bar) and compared with the response of untreated T cells (+APC, grey bar). The data are representative of two experiments with Th1 cell line MD6 (A) and five experiments with Th0 clone PW (B).

Anergic Th1 and Th0 cells failed to exhibit linked suppression.

It has been suggested previously that suppressive anergic T cells are capable of inhibiting responses to other antigens presented on the same APC (linked suppression). This phenomenon has been reported after exposure of human T-cell clones to immobilized anti-CD3 ⁽²⁵⁾ and after anergy induced by T-T cell antigen presentation in rat T cells ⁽²⁷⁾. The anergic Th0 clone, PW, was unable to suppress the response of untreated Th0 or Th1 cells that express different TCRs but recognize the same MHC class II molecule I-A^b presented on the surface of irradiated I-A^b+ APCs (Fig. 5, A and B). Anergic Th1 cells were equally incapable of suppressing untreated cells from different T-cell clones (data not shown). Under the same conditions, anergic Th1 and Th0 cells were able to suppress Th1 and Th0 cells from the same clone that expressed the same TCR (Fig. 5 C). These data show that these murine alloreactive CD4⁺ Th1 and Th0 T cells do not exhibit linked suppression.

Figure 5: Linked epitope suppression is absent in Th0 T-cell clones. A, Untreated MD6 (2×10⁴/well) was cultured alone (−APC) or with irradiated I-A^b+ APCs in the presence (1A:1U) or absence (U+APC, grey bar) of anergic PW; suppression co-cultures contained equal numbers of anergic PW and untreated MD6. B, Untreated AB (2×10⁴/well) was cultured with irradiated I-A^b+ APCs in the presence (1A:1U) or absence (U+APC, grey bar) of anergic PW; suppression co-cultures contained equal numbers of anergic PW and untreated AB. C, Anergic PW (2×10⁴/well) was restimulated with irradiated I-A^b+ APCs (5×10⁵/well) alone (A+APC) or in co-culture with untreated T cells from the same clone at a ratio of 1:1 (1A+1U) and compared with the response of untreated PW (2×10⁴/well) to I-A^b+ APCs (U+APC, grey bar). These results are representative of three independent experiments.

Interaction between the TCR and peptide-MHC class II complex is essential for suppression but not anergy in Th0 cells.

To investigate whether anergy and suppression in alloreactive CD4⁺ Th0 cells required specific interaction with the I-A^b class II MHC-peptide complex, anergized Th0 cells were restimulated either alone or in co-cultures with untreated T cells using mAbs specific for the αβ chains of the TCR (H57–597 or F23.1) or ε chain of CD3 (145–2C11) in the presence of irradiated syngeneic APCs. This system has been used previously to show anergy in naïve CD4⁺ T cells ⁽²⁰⁾. However, anergic Th0 cells, AB and PW, proliferated in restimulation cultures that contained soluble anti-CD3 (Fig. 6 A). Stimulation of anergic Th0 cells, PW and AB, with mAbs specific for the αβ TCR did, however, facilitate anergy (Fig. 6 B). The anergic Th0 clone, PW, also remained unresponsive to the anti-TCR antibody F23.1, specific for the Vβ8 TCR (data not shown). The proliferation of anergic cells in response to anti-CD3 was similar to that seen with untreated cells (Fig. 6 A). Furthermore, anergic Th0 clones were unable to suppress the response of untreated Th0 clones to either soluble anti-CD3 or anti-TCR mAbs (Fig. 6, A and B). Suppression was only observed when irradiated I-A^b+ APCs were present in the suppression co-cultures (Fig. 6 C). The prevention of suppression by mAb stimulation did not appear to be related to the strength of signal through the TCR complex because reduction of the concentration of either anti-CD3 or anti-TCR mAb did not reinstate suppression in the Th0 clone PW (Fig. 7, A and B). Interestingly, even at a higher concentration of anti-TCR antibody, the anergic state was still stable (Fig. 7 B). The importance of the MHC-peptide complex in facilitating suppression was confirmed by the addition of I-A^b+ APCs into suppression co-cultures restimulated with anti-TCR mAbs and syngeneic APCs. Suppression was restored in these cultures to levels previously seen in suppression cultures that contained I-A^b+ APCs in the absence of anti-TCR mAbs (Fig. 8). These data suggest that the MHC-peptide complex plays a pivotal role in facilitating suppression by anergic Th0 cells.

Figure 6: The MHC-peptide complex plays an essential role in suppression mediated by anergic CD4⁺ Th0 T-cell clones. A, Anergic AB was cultured alone (−145–2C11), in the presence of soluble 145–2C11 (2 μg/ml) and irradiated syngeneic APC (5×10⁵/well) alone (A+145–2C11), or in co-culture with untreated AB (1A:1U) and compared with the response of untreated AB to stimulation with soluble 145–2C11 and irradiated syngeneic APCs (U+145–2C11). B, Anergic PW was cultured alone (−H57), in the presence of soluble H57–597 (2 μg/ml) and irradiated syngeneic APC (5×10⁵/well) alone (A+H57), or in co-culture with untreated PW (1A:1U) and compared with the response of untreated PW to stimulation with soluble H57–597 and irradiated syngeneic APCs (U+H57). C, Anergic PW was cultured alone (2×10⁴/well) (−APC), with irradiated I-A^b+ APCs alone (5×10⁵/well) (A+APC), or in co-culture with untreated PW (2×10⁴/well) (1A:1U), and the response was compared with untreated PW to irradiated I-A^b+ APCs (U+APC) under the same conditions. Cells were cultured for 48 (black bars) and 72 hr (white bars), pulsing with tritiated thymidine for the final 16 to 18 hr in each case. Data shown are representative of three independent experiments.

Figure 7: Anti-TCR and anti-CD3 mAbs prevented suppression mediated by anergic Th0 cells at a range of concentrations. Anergic PW (2×10⁴/well) were stimulated with either 145–2C11 (A) (0.002–2 μg/ml) or H57–597 (B) (20–0.2 μg/ml) and syngeneic APCs in the presence (filled triangles) or absence (filled diamonds) of untreated PW at a ratio of 1:1 and compared with the response of untreated PW (filled squares) under the same conditions. The dark circles on the Y-axes represent suppression co-cultures that comprise anergic PW (2×10⁴/well), untreated PW (2×10⁴/well), and I-A^b+ APCs (5×10⁵/well). These results are representative of three independent experiments performed with the Th0 clone PW.

Figure 8: Suppression mediated by anergic Th0 cells is only observed in the presence of I-A^b+ APCs. The Th0 clone, PW, was rendered anergic with immobilized anti-CD3. Anergic cells were co-cultured with untreated cells at a ratio of 1:1 with irradiated I-A^b+ APCs (5×10⁵/well), H57–597 (2 μg/ml)/syngeneic APCs (5×10⁵/well), or H57–597/syngeneic APCs/I-A^b+ APCs. Cultures were incubated for 3 days and pulsed with tritiated thymidine for the final 16 to 18 hr. These data are representative of two independent experiments.

DISCUSSION

Classic in vitro anergy is thought to develop if T cells receive a signal through the TCR in the absence of a second signal delivered by costimulatory molecules ^(17,39,40). Two methods used to generate anergic cells, stimulation by high doses of immobilized anti-CD3 and T-T cell antigen presentation, are thought to operate by this principle ^(14,15). Furthermore, these systems allow the development of anergic T cells that suppress the activation of untreated T cells from the same clone-suppressive anergy ^{(10,12,16,27)}. Both the mechanism for the induction of suppressive anergy in T cells and the mechanism through which anergic T cells mediate suppression are far from clear.

In this study, we have shown that the requirements for the induction of anergy by high doses of immobilized anti-CD3 in murine CD4⁺ T-cell clones are different for CD4⁺ I-A^b–reactive T cells of the Th1 and Th0 phenotypes. Th0 cells were anergic and capable of suppressing the response of immunocompetent Th0 cells from the same clone when induction cultures were conducted in the presence of full costimulation, either immobilized or soluble stimulatory anti-CD28 or syngeneic T-cell–depleted APCs (Fig. 2). The development of suppressive anergy was prevented in induction cultures with Th1 cells when full costimulation was provided in addition to immobilized anti-CD3 (Fig. 2). These data for Th1 cells are supported by previous reports that suggest that anergy could be prevented by the addition of accessory cells ⁽¹⁴⁾ or stimulation through CD28, both of which provide an essential second signal for the production of IL-2 ⁽³⁸⁾. IL-2 is thought to prevent anergy by causing proliferation of the CD4⁺ Th1 clones, thereby diluting out negative regulatory factors ⁽³⁸⁾ or leading to direct inhibition of negative factors produced by TCR ligation ⁽²²⁾. In our system, the addition of cyclosporine, which blocks events proximal to TCR ligation, to the induction phase of anergy prevented both the development of anergy and the suppression in Th1 cells, which suggests that negative regulatory factors may also be produced by stimulation with high doses of immobilized anti-CD3 mAbs (P.R.W. and K.J.W., unpublished observations, 2000). The essential role of IL-2 in preventing anergy has also been demonstrated by incubation of T-cell clones with IL-2 receptor blocking and IL-2 neutralizing antibodies in the presence of live APCs and peptides ⁽²¹⁾. A similar observation was made when CD4⁺ T-cell clones were stimulated with anti-TCR and anti-CD28 in the presence of rapamycin ⁽²²⁾.

Unresponsiveness induced by anti-CD3 could be reversed by the addition of exogenous IL-2 to restimulation cultures in both Th1 and Th0 cell populations in our system. Many groups have demonstrated that this leads to a reversal of anergy in Th1 clones ^(25,38); however, anergic suppressive CD4⁺ and CD25⁺ T cells proliferate in response to IL-2. After washing to remove IL-2, subsequent restimulation shows that these cells are still anergic and suppress the response of CD4⁺ and CD25⁻ thymocytes ⁽¹³⁾. Previous studies have demonstrated that the addition of IL-2 to anergic cells caused them to proliferate; however, it was not shown whether the anergic state was reversed by reculturing anergic cells treated with IL-2 in subsequent restimulation cultures. We have observed that anergy is reversed in CD4⁺ Th1 cells after treatment with IL-2; however, anergic Th0 cells remained anergic and retained their ability to suppress immunocompetent T cells from the same clone. These findings indicate that suppressive anergy in Th0 cells, in contrast to Th1 cells, may be a more stable condition, and this might explain why a deviated Th2 cytokine profile has been shown to be beneficial but not mandatory in various protocols that induce peripheral T-cell tolerance ^(41–44).

There is some evidence that anergic T cells generated in vitro can mediate their suppressive effect after adoptive transfer in vivo ⁽¹²⁾. If in vitro anergized T-cell clones are to be used in vivo as a cell-based therapy to prevent graft rejection, the stability of the anergy induced and the potential of the anergized T cells to suppress other related T cells (linked suppression) are crucial ^(25,27). Linked suppression has been observed previously when the distinct MHC-peptide complexes for both the anergic and untreated clones in the co-culture were presented by the same APC ^(10,23,27). Linked epitope suppression was not observed in this study (Fig. 5). The anergic Th1 and Th0 T cells were unable to suppress the response of each other or the response of a polyclonal naïve CD4⁺ T-cell population responding to I-A^b+ APCs (Fig. 5; O.W., P.R.W., and K.J.W., unpublished observations, 2000). Although we have not shown formally that the MHC-peptide complexes specific for individual Th0/Th1 T cell clones are expressed on the same APCs, it is very likely that individual I-A^b+ APCs express the MHC-peptide complexes recognized by both T-cell populations in the co-culture experiments, because the same APCs will stimulate the clones under the appropriate conditions. Because suppression with these CD4⁺ I-A^b specific T cells was only mediated when both the untreated and anergic cells were specific for the same MHC-peptide complex, it seems possible that the mechanism for suppression may be competition for their specific MHC-peptide complexes. Recently, it has been shown that peptide-specific CD8⁺ T cells internalize peptide MHC complexes through T-cell receptor–mediated endocytosis ⁽⁴⁵⁾. If CD4⁺ T cells are capable of mediating internalization of MHC class II complexes, it is possible that anergic CD4⁺ T cells would compete with untreated T cells for the limited amounts of MHC peptide complexes. This mechanism has been suggested to cause the exhaustion/down-regulation of T-cell immune responses and would offer one explanation as to why linked suppression did not appear to operate in our system ⁽⁴⁵⁾.

The absence of linked suppression suggested that a more specific mechanism was involved and that it was not as simple as regulation of the antigen-presenting ability of the APC ⁽²⁷⁾. The most specific part of this co-culture system is the interaction between the TCR and the specific MHC-peptide complexes recognized by each of the clones. This report shows that stimulation through the TCR with a panel of anti-TCR mAbs can not facilitate suppression mediated by anergic Th0 cells (Fig. 6). This not only suggests that the APC is essential in mediating suppression but that it is the specific interaction between the MHC-peptide complex of the APC and the TCR that facilitates suppression by anergic Th0 cells. The role of the MHC molecule in the control of T-cell responses has been suggested previously but never formally shown ⁽⁴⁶⁾. Interestingly, stimulation through the CD3ε subunit of the TCR/CD3 complex prevented Th0 cells from developing an anergic phenotype whereas stimulation though the αβ subunit allowed the cells to retain the anergic state (Fig. 6). This difference was not due to the strength of signal through the TCR, because both anti-CD3 and anti-TCR mAbs caused the same level of proliferation in the Th1 T-cell line MD6 (data not shown). Moreover, titration of anti-CD3 and anti-TCR mAbs did not reinstate suppression mediated by anergic Th0 cells (Fig. 7). The lack of suppression but the retention of the anergic state seen in response to stimulation with anti-TCR antibodies suggested that anergy and suppression were independent events that could only be linked in the presence of their specific MHC-peptide complex.

In this report, we have demonstrated that the requirements for costimulation during the induction of suppressive anergy are distinct in mouse CD4⁺ I-A^b–specific Th0 and Th1 cells. The stability of the anergic suppressive state was maintained in anergic Th0 clones after treatment with IL-2, suggesting that they may play a role in maintaining the state of tolerance in vivo. This report further substantiates the role of the APC in mediating suppression by anergic CD4⁺ T cells and suggests an essential role for the specific interaction of the MHC-peptide complex and the T-cell receptor. We propose that suppression is mediated either by down-modulation of the MHC-peptide complex required by the anergic T cells or that a molecule specific to the MHC-peptide/TCR interaction facilitates negative regulation by APC:T or T:T interactions.