Control of T cell hyperactivation in IL-2-deficient mice by CD4⁺CD25^– and CD4⁺CD25⁺ T cells: evidence for two distinct regulatory mechanisms

2.1 Amelioration of the IL-2-deficiency syndrome by restriction of the TCR repertoire

BALB/c mice are particularly sensitive to the immunopathology caused by a genetic deficiency in IL-2. As has been described in detail earlier 6, IL-2^– BALB/c mice do not survive beyond 5 weeks of age and die with the typical symptoms of the IL-2-deficiency syndrome. Since the ameliorating effect of a germ-free environment on the development of immune pathology had suggested that the accumulation of activated CD4 T cells in these animals is an antigen-driven event 5, we restricted the TCR repertoire of IL-2-deficient BALB/c mice by introducing the OVA plus IA^d-specific TCR DO.11.10 13. As expected, these TCR-transgenic IL-2-deficient BALB/c mice showed an increased life span with a mean survival time of 4 months (data not shown). This is, however, still well below the average life expectancy of wild-type BALB/c mice, indicating that the DO.11.10 transgene provided incomplete protection from T cell hyperactivation. Indeed, penetrance of the TCR transgene as assessed by expression of the KJ1-26 idiotypic determinant was incomplete (between 70 and 95%), and activated T cells accumulated with time, presumably as a result of endogenous TCR α-chain expression (data not shown). Nevertheless, young adult DO.11.10 IL-2^– BALB/c mice provided a reproducible source of antigen-responsive CD4 T cells 13 which were employed for the following experiments.

2.2 Clonal expansion and contraction of adoptively transferred DO.11.10 IL-2⁺ or IL-2^– CD4 T cells

Following the procedure established by Jenkins and colleagues for the evaluation of clonal dynamics of T lymphocytes in vivo 29, we transferred 2×10⁶ Id⁺ CD4 T cells from IL-2-deficient or IL-2 wild-type BALB/c mice into syngeneic BALB/c or athymic BALB/c nu/nu recipients. After subcutaneous immunization with the cognate OVA peptide 323–339 in IFA, absolute numbers of Id⁺ T cells in the draining lymph nodes were followed over time. As can be seen in Fig. 1, both IL-2 wild-type and IL-2-deficient Id⁺ CD4 T cells were increased in number over unstimulated controls 1 week after cell transfer into BALB/c mice (clonal expansion), as has recently been reported by others 14. Regardless of the presence of an intact IL-2 gene in the Id⁺ cells transferred, this was followed by a decrease to baseline levels over the next 2 weeks (clonal contraction). In contrast, when DO.11.10 transgenic CD4 T cells were transferred into athymic recipients, the kinetics of IL-2⁺ and IL-2^– Id⁺ cells recovered were dramatically different: While TCR transgenic cells with an intact IL-2 gene contracted normally, those from IL-2-deficient donors continued to expand throughout the observation period of 3 weeks. Thus this system recapitulates in a well-controlled fashion the accumulation of antigen-expanded T cells previously observed in IL-2 deficient animals 5, 8, 16. Moreover, the results show that T cells with an intact IL-2 gene, either present in the TCR-transgenic inoculum or in the recipient are able to counteract the accumulation of clonally expanded, activated CD4 T cells which is the hallmark of the IL-2-deficiency syndrome. This is in keeping with our earlier findings on mixed bone marrow chimeras in which a fraction of wild-type hematopoietic cells effectively prevented the IL-2-deficient T cells present in the same animal from causing immune pathology 11.

2.3 Restoration of clonal contraction of IL-2-deficient CD4 T cells by cotransferred T cells and T cell subsets

To test whether IL-2 sufficient T cells are able to control the abnormal accumulation of IL-2-deficient CD4 cells, T cells from normal BALB/c mice were cotransferred with IL-2-deficient DO.11.10 transgenic CD4 cells into athymic BALB/c recipients. As shown in Fig. 2A, the antigen-induced accumulation of Id⁺ IL-2-deficient CD4 cells observed three weeks later was completely abolished by the cotransfer of a fivefold excess of wild-type T cells. This suppression appeared to act mostly at the level of clonal contraction because only a small reduction in cell number was observed 1 week after cell transfer and antigenic stimulation but was complete 2 weeks later.

When the CD4 and CD8 subsets present in the wild-type T cell population were administered separately, both exhibited regulatory activity. However, since the CD4 fraction appeared slightly more effective than the CD8 fraction, and because suppression appeared to be in saturation, wild-type CD4 and CD8 T cells were titrated in another experiment at lower numbers to test for their suppressive potential. As shown in Fig. 2B, CD4 T cells inhibited the accumulation of IL-2-deficient TCR transgenic CD4 T cells much more efficiently than CD8 T cells, suggesting that CD4 cells are mainly responsible for the correction of IL-2-deficient CD4 T cell accumulation in an environment containing wild-type T cells. This result is in seeming contrast to the recent findings of Suzuki et al. 19 who found that CD8 cells are superior to CD4 cells in counteracting the accumulation of T cells lacking the IL-2Rβ chain. In this system, however, clonal contraction via paracrine provision of IL-2, expected to be primarily produced by CD4 T cells, cannot play a role. The ability of IL-2 to sensitize IL-2-deficient (but not IL-2Rβ-deficient) T cells for AICD 15, 16 thus provides an explanation for this discrepancy.

Interestingly, the residual IL-2-deficient Id⁺ cells recovered after cotransfer with wild-type CD4 T cells contained a higher fraction of cells expressing the activation marker CD69 than controls without regulatory cells (Fig. 2C). This supports the notion that these cells acted mainly at the level of clonal contraction rather than by inhibiting initial activation of the IL-2-deficient OVA-specific T cells.

2.4 CD4⁺CD25⁺ cells effectively counteract accumulation of antigen-activated IL-2-deficient CD4 cells

There are two obvious possible mechanisms through which wild-type CD4 T cells may correct the defective control of clone size in IL-2-deficient T cells: provision of paracrine IL-2 which may restore sensitivity to AICD, and counter-regulation of T cell activation by a regulatory T cell population. A good candidate for such a CD4 T cell subset are the CD4⁺CD25⁺ cells producing IL-10 and TGF-β but not IL-2 (reviewed in 21). These thymus-derived cells have previously been shown to suppress autoimmunity and inflammation in a variety of systems 23–26. To test whether CD4⁺CD25⁺ T cells from wild-type BALB/c mice were able to suppress the accumulation of IL-2-deficient antigen activated CD4 T cells, we initially sorted BALB/c lymph node CD4 T cells into the CD25⁺ and CD25^– subsets and tested both populations in the adoptive cotransfer system employed in the previous experiments. Analysis of recipients of CD25-depleted CD4 T cells was, however, confounded by the occurrence of diarrhea and wasting (data not shown). This is in line with the recent identification of CD4⁺CD25⁺ cells as suppressors of IBD and wasting induced by CD45RB^high CD4 T cells in response to microbial and dietary antigens of the gut 21, 23, 25, 26. Accordingly, unseparated CD4 T cells were compared with the CD4⁺CD25⁺ subset for their regulatory capacity upon cotransfer with IL-2-deficient TCR transgenic CD4 cells. As shown in Fig. 3, CD4⁺CD25⁺ cells were at least as potent as unseparated CD4 cells, demonstrating the efficacy of this subset in controlling hyperactivation of IL-2-deficient T cells.

2.5 Control of T cell expansion by CD4⁺CD25⁺ cells is independent of paracrine IL-2 delivery

CD4⁺25⁺ cells do not produce IL-2 when isolated ex vivo 22, suggesting that paracrine IL-2 delivery is not essential for preventing the accumulation of IL-2-deficient cells. To test whether the injected regulatory cells remain stable with regard to this phenotype, we tested the capacity of Id^– cells recovered 3 weeks after transfer to produce IL-2. As shown in Fig. 4A, there was a low background of CD4 cells capable of producing IL-2 in Id^– CD4 cells recovered from athymic mice that had not received regulatory cells. These cells are derived from the endogenous extrathymically matured T cell population found in athymic mice which is largely immuno-incompetent because of its oligoclonal repertoire 30. Upon transfer of unseparated CD4 T cells, the frequency of recovered Id^– CD4 cells able to produce IL-2 rose to about 30%. In contrast, no increase above background was detected upon cotransfer of CD4⁺CD25⁺ T cells, indicating that they had not acquired the capacity to produce IL-2 during the course of the adoptive transfer experiment.

To test for the persistence of CD4⁺CD25⁺ regulatory cells and for the impact of paracrine IL-2 on the TCR-transgenic IL-2 deficient indicator cells, the expression of CD25 itself was investigated on both populations. For comparison, mice that had received unseparated CD4 regulatory cells were included. As shown in Fig. 5B, CD25 expression was twice as frequent among Id^– cells recovered from mice which had received CD4⁺CD25⁺ regulators as compared to those that had received unseparated CD4 cells. In contrast, CD25 expression on the Id⁺ IL-2-deficient indicator cells followed an inverse pattern: cotransfer of unseparated CD4 T cells, but not of purified CD4⁺CD25⁺ cells had resulted in up-regulation of CD25 on the residual recovered TCR-transgenic cells. Since IL-2 is known to up-regulate CD25, the α-chain of its receptor 31, 32, this result suggests that unseparated CD4 T cells actually had provided IL-2 to the IL-2-deficient indicator cells whereas CD4⁺CD25⁺ cells had not.

Taken together, these results indicate that CD4⁺CD25⁺ cells do not down-regulate the accumulation of IL-2-deficient cells via paracrine provision of IL-2. In keeping with this finding, defective clonal contraction of T cells from IL-2Rβ-deficient mice that are unable to receive an IL-2 signal is also controlled by wild-type T cells 19. Although CD8 cells were found superior to CD4 cells in this system, the inhibitory component contained in the latter may have been mediated by their CD4⁺CD25⁺ subset.

2.6 Two mechanisms of clonal control?

Although the above results illustrate the efficacy of CD4⁺CD25⁺ cells as down-regulators of T cell hyperactivation caused by IL-2 deficiency, we also found that unseparated C4 cells of which only about 5–10% are CD25⁺ are similarly potent suppressors. This suggested that the dominant mechanisms of suppression mediated by unseparated versus purified CD4⁺CD25⁺ cells may have been different. Indeed, the increased expression of CD25 on residual IL-2 deficient TCR transgenic cells after stimulation in the presence of unseparated CD4 but not of CD4⁺CD25⁺ cells indicated that different signals had been received by the IL-2-deficient cells from the two populations of regulatory cells, and that sensitization for AICD by IL-2 may have played a role in the former but not in the latter setting. If so, initial activation would have proceeded unimpaired when the majority of cotransferred regulatory cells were CD25^– but able to produce IL-2 (whole CD4 cells). In contrast, our current knowledge of immune regulation by CD4⁺CD25⁺ cells indicates that they suppress clonal expansion of CD4⁺CD25^– T cells at the level of initial expansion 21.

To obtain further information on the activation status of IL-2-deficient TCR transgenic T cells recovered after stimulation in the presence of unseparated versus CD25⁺ wild-type CD4 T cells, the early activation marker CD69 was investigated. As shown in Fig. 2C and 7, the frequency of CD69 expressing cells within the Id⁺ population recovered 3 weeks after cell transfer was strongly increased when unseparated CD4 cells were used to prevent accumulation of clonally expanded cells, while cotransfer of CD4⁺CD25⁺ regulatory cells did not lead to increased CD69 expression. Thus, in spite of an equally effective containment of antigen-driven clonal expansion of IL-2-deficient CD4 cells in both settings, their activation status was actually enhanced in the presence of unseparated wild-type cells capable of producing IL-2 while this was not the case when CD4⁺CD25⁺ regulatory cells were employed which are incapable of producing IL-2.

4.1 Mice

BALB/c mice were bred at the Institute's barrier-contained facilities. BALB/c nu/nu mice were obtained from Harlan Winkelmann, Borchen, Germany, and Charles River, Sulzfeld Germany. Unirradiated male animals around 8 weeks of age were used as recipients for adoptive transfer. IL-2-deficient BALB/c mice expressing the DO11.10 TCR transgene were generated by backcrossing as described 13 from IL-2-deficient 2 and DO.11.10 transgenic 34 BALB/c mice. The colony was maintained with IL-2^+/– DO11.10 transgenic breeding pairs, and the IL-2 genotype offspring was identified by genomic PCR 2. IL-2-deficient (–/–) and -sufficient (+/+ or +/–) DO11.10 transgenic mice were used at 12 weeks or younger. These animals as well as all recipients of transferred cells were kept in sterilized individually ventilated cages throughout the experiment.

4.2 Cell purification and transfer

Lymph node cells were first enriched for T cells by nylon-wool passage followed by purification (>95%) of T cells or their CD4 or CD8 subsets using negative selection on Immunocolumns (Cedarlane Laboratories Ltd., Hornby, Ontario, Canada) according to the manufacturer's instructions. CD4⁺CD25⁺ cells were further purified (98%) by positive enrichment using a MiniMACS magnetic separator (Miltenyi Biotec, Bergisch-Gladbach, Germany) and electronic cell sorting (FACS Vantage^®, BD Lifesciences, Mountain View, CA). The adoptive transfer, immunization and monitoring of draining lymph nodes followed the procedure of Kearny et al. 29. Purified CD4⁺ lymph node T cells form DO11.10 TCR transgenic mice containing 2×10⁶ Id⁺ cells and, as given, wild-type T cells, were i.v. injected in.2 ml BSS. OVA_323–339 peptide (20 μg) were emulsified in IFA and injected s.c. at the scruff of the neck 1 day later. At the time points indicated, animals were sacrificed and axillary lymph nodes prepared for further analysis.

4.3 Antibodies

mAb to CD4 (clone GK1.5), CD8α (53–6.7), CD8β (53–5.8), CD16/32 (2.4G2), CD25 (7D4), CD69 (H1.2F3), IL-2 (JES5–16E3) were obtained unconjugated and FITC-, Cychrome- or PE-conjugated from PharMingen-BD Lifesciences. Anti-DO.11.10 idiotype (KJ1–26, 35) was prepared from a cell line kindly donated by Drs. Haskins, Kappler and Marrack, and FITC conjugated using standard procedures. Staining was performed on 1–3×10⁵ cells in PBS/ 0.1% BSA/0.02% NaN₃ on ice following standard procedures. For cell surface staining, Fc-receptors were blocked using anti-CD16/32, followed by incubation with fluorochrome-conjugated mAb. For intracellular cytokine staining, a modification of a previously described protocol 36 was used. Cells were stimulated for 4 h with PMA (Sigma, St. Louis, MO) at 2 ng/ml and ionomycin (Sigma) at 500 ng/ml. Brefeldin A (Sigma) was added at 1 μg/ml for the last 2 h. Cells were harvested, washed and stained for cell surface antigens. Fixation was performed with 4% formaldehyde for 20 min at room temperature, followed by permeabilization with 0.5% saponin in PBS/BSA/Azide for 10 min. Nonspecificbinding was blocked with normal rat Ig (10 μg/ml) and cells were stained with directly conjugated anti-IL-2 mAb or isotype control for 20 min at room temperature. Blocking by pretreatment with unlabeled anti-IL-2 mAb in a control sample was used to ensure specificity. Flow cytometric analyses was performed with a FACScan^® flow cytometer (Becton Dickinson). Living cells (10,000) were computed and analyzed using the Cell Quest software and are shown as dot plots using a logarithmic scale. Isotype-matched negative controls were used to set background levels to the first logarithmic decade. CD4⁺DO11.10 transgenic cells were identified by double staining with GK1.5 and CD4 and KJ1–26, and absolute numbers of recovered Id⁺ cells were calculated from their fraction among trypan-blue excluding total draining (axillary) lymph node cells.