Interleukin-2 is essential for CD4+CD25+ regulatory T cell function



Constitutive expression of CD25, the IL-2 receptor α-chain, defines a distinct population of CD4+ T cells (Treg) with suppressive activity in vitro and in vivo. IL-2 has been implicated in the generation and maintenance of Treg, however, a functional contribution of the IL-2 receptor during suppression is thus far unknown. We show that IL-2 is required for Treg function in vitro, since suppression is completely abrogated by selective blocking of the IL-2 receptor on Treg during co-culture with responder T cells. We demonstrate that Treg, which do not produce IL-2, compete for IL-2 secreted by responder T cells. In accordance with the idea of competition being part of the suppressive mechanism, in vitro neutralization of IL-2 mimics all effects of Treg. Conversely, recombinant IL-2 abrogates inhibition of IL-2 production in responder T cells, the hallmark of Treg suppression. Finally, activation in the presence of IL-2 primes Treg to produce IL-10 upon secondary stimulation, indicating that IL-2 uptake is also required to induce additional suppressive factors that might be more relevant for suppression in vivo. We propose the parakrine uptake of soluble mediators as a flexible mechanism to adapt Treg activity to the strength of the responder T cell reaction.


Carboxy-fluorescein diacetate-succinimidyl ester


Regulatory T cells

1 Introduction

Distinct T cell populations with suppressive or regulatory activity contribute to the control of auto-reactive T cells and the maintenance of natural self-tolerance in vivo 13. Among those the small population of CD4+ T cells expressing CD25 in the steady state (Treg), is particularly well-characterized for its suppressive function both in vitro and in vivo 47. Although CD25, the α-chain of the high-affinity IL-2 receptor, is the marker commonly used for the isolation of these cells, its functional importance has not been clarified. It is generally assumed that IL-2 is necessary for the development and homeostasis of Treg 8, 9. Mice deficient for IL-2 or the IL-2 receptor have reduced numbers of regulatory T cells both in the thymus and in the periphery 10, 11, and develop a lethal phenotype characterized by T cell hyperproliferation 1214. It has also been shown that IL-2 signaling, probably via STAT5, is required to maintain or activate Treg in the periphery 11, 1517.

Since Treg do not produce IL-2 10, 18, 19, the constitutive expression of the high-affinity IL-2 receptor may be required to take up IL-2 secreted by other cells. Recently, it has been demonstrated that wild-type Treg show decreased survival after transfer into CD28-deficient mice 20, which could be rescued by IL-2, indicating that Treg depend on exogenous survival factors. Whether Treg also require IL-2 uptake or signaling via the IL-2 receptor to exert their suppressive function has not been addressed in detail. IL-2 consumption could enable them to efficiently deprive responder cells of this essential growth factor, implying competition for IL-2 as part of the suppressive mechanism in vitro. Pre-activation of Treg in the presence of IL-2 up-regulates IL-2 receptor expression and increases their regulatory capacity 6, 21. However, the mechanisms by which suppression is mediated in vitro or in vivo are largely unknown.

Suppressive cytokines like IL-10 and TGF-β seem to play a role in vivo2225 but not in vitro6, 26, 27. In line with this, under physiologic conditions Treg appear not to produce detectable amounts of IL-10 upon stimulation ex vivo, whereas IL-10 can be detected after in vivo priming of Treg 28, 29. The factors that act in vivo to induce IL-10 production are unknown.

Currently, a cell contact-dependent mechanism for in vitro suppression is favored by many in the field. This is based mainly on findings that culture supernatants from activated CD4+CD25+ T cells show no inherent suppressive activity and that suppression is abrogated in transwell cultures 6, 26. However, short distance interactions cannot be excluded by this assay and the nature of a potential suppressive ligand/receptor interaction remains elusive.

Here, we have directly addressed the question whether IL-2 is required by Treg to exert their suppressive function. Using in vitro co-culture systems, we found that IL-2 is not only necessary to efficiently activate Treg but is also needed during suppression of responder cells, as specific blocking of the IL-2 receptor on Treg leads to a complete loss of regulatory activity. We further show that Treg compensate their lack of IL-2 production by efficient competition for IL-2 secreted by responder T cells. In addition, IL-2 primes Treg to produce IL-10 upon re-stimulation. Our findings identify IL-2 as an important regulator of Treg activity and more generally imply cellular cross-talk via soluble mediators as a mechanism for the modulation of Treg function.

2 Results

2.1 IL-2 is required for suppressive activity of CD4+CD25+ Treg in vitro

Recently, it has been reported that pre-activation of Treg by anti-CD3 antibody plus IL-2 enhances their suppressive capacity 21. To generate active suppressors, we pre-activated Treg for 24 h using plate-bound anti-CD3 plus IL-2 or anti-IL-2 and assayed their ability to suppress the proliferation of responder T cells at various ratios. As shown in Fig. 1A, IL-2 enhances the suppressive activity of Treg ∼20-fold as compared to pre-activation in the presence of anti-IL-2.

Figure 1.

IL-2 is required for CD4+CD25+ Treg-mediated suppression. (A) Purified CD4+CD25+ Treg were pre-activated for 24 h with immobilized anti-CD3 in the presence of either rIL-2, neutralizing anti-IL-2 antibody or an isotype control. CD4+CD25 responder T cells labeled with CFDA-SE were cultured alone or together with pre-activated Treg at the indicated ratios in the presence of anti-CD3 and irradiated APC for 3 days. Proliferation was measured and quantitated as described in Sect. 4 and responder T cells cultured alone were set to 100%. (B) Murine CD4+CD25+ Treg were pre-activated for 24 h in the presence of IL-2. Murine pre-activated Treg or CD4+CD25 T cells as controls were co-cultured with CFDA-SE labeled human CD4+ responder T cells at a ratio of 1:2. Proliferation of human responder T cells on day 3 was quantitated as described in Sect. 4. Proliferation of human responder cells without murine T cells was set to 100%. To specifically block the IL-2 receptor on Treg, 50 µg/ml anti-CD25 plus anti-CD122 or an isotype control were added. (C) Proliferation of CFDA-SE labeled human CD4+ T cells in the presence of murine Treg or control cells is depicted (experimental setting as in B). The percentage of proliferated cells is indicated.

To address the question whether IL-2, apart from being a potent initial activator of Treg, is functionally involved in suppression during co-culture with responder T cells, we established a human-mouse chimeric suppression assay. The use of murine Treg in combination with human responder T cells allowed us to selectively block CD25 and CD122, the α- and β-chains of the IL-2 receptor, on Treg without influencing the responder T cells. When highly-purified human CD4+CD25 T cells were cultured at a 2:1 ratio with pre-activated murine CD4+CD25+ Treg, proliferation of human responder T cells was inhibited by about 50%, whereas murine CD25 control cells had no influence (Fig. 1B). To determine the functional relevance of IL-2 for Treg activity, we selectively blocked the uptake of the cytokine by Treg using antibodies against the murine IL-2 receptor α- (CD25) and β-chain (CD122). Interestingly, this led to a complete loss of suppressive activity (Fig. 1B). Antibodies against either CD25 or CD122 alone were partially effective in blocking suppression (data not shown). This clearly demonstrates that the uptake of IL-2 by CD4+CD25+ Treg is absolutely required during suppression. Since Treg had already been pre-activated for 24 h with IL-2, this abrogation of suppressive activity cannot be explained by inappropriate activation due to a lack of IL-2 but rather shows the direct involvement of the IL-2 receptor in the suppressive mechanism.

2.2 Treg constitutively possess the capacity to capture IL-2

CD25+ Treg do not produce IL-2 by themselves 10, 18, 19. After we had shown that IL-2 uptake is indeed required for suppression, we hypothesized that Treg compete for IL-2 produced by responder T cells. To test whether the high expression of the IL-2 receptor α-chain on Treg directly correlates with the capacity to efficiently bind IL-2, we used an IL-2-IgG fusion protein for surface staining of ex vivo isolated CD4+ T cells. As shown in Fig. 2A, binding of the fusion protein was detectable selectively on Treg and the staining intensity directly correlated with CD25 expression. The binding of the fusion protein was completely inhibited by antibodies against CD25 and CD122, demonstrating the specificity of the interaction. Isolated CD25+ Treg showed a further increased IL-2 binding capacity upon pre-activation in the presence of IL-2, whereas the binding capacity of responder T cells did not improve compared to pre-activation with anti-CD3 alone (Fig. 2B). Thus, Treg but not responder T cells constitutively possess the capacity to bind IL-2 without prior activation. In contrast to responder T cells, Treg further increase this IL-2 binding potential upon exposure to exogenous IL-2.

Figure 2.

CD4+CD25+ Treg have a high capacity to capture IL-2. (A) Spleen cells were stained with an IL-2-hIgG2b fusion protein, anti-CD4 and a non-blocking anti-CD25. The staining was blocked with neutralizing antibodies against CD25 and CD122. Data are gated on CD4+ cells. (B) CD4+CD25+ Treg or CD4+CD25 responder T cells were taken ex vivo or pre-activated for 24 h with anti-CD3 ± IL-2 and stained for IL-2 receptor expression using the IL-2 fusion protein. The geometric mean fluorescence intensity (MFI) of the staining is given.

2.3 Treg take up IL-2 from responder T cells during co-culture

It is well known that CD25 expression is positively regulated by IL-2 3032. Therefore, we measured the expression of CD25 on both Treg and responder T cells in an attempt to prove competition for IL-2 during co-culture. Highly-purified, CD4+CD25+ Treg and CD4+CD25 responder T cells, taken freshly ex vivo, were cultured either alone or at a ratio of 1:2. As shown in Fig. 3A, CD25 responder T cells strongly proliferated and up-regulated CD25 when activated alone. As expected, co-culture with CD25+ Treg resulted in a strong inhibition of proliferation of CD25 responder T cells (Fig. 3B). Interestingly, the induction of CD25 expression on responder T cells was almost completely inhibited, even on the small number of cells which had still proliferated under these conditions. At the same time, CD25+ Treg up-regulated CD25 expression about sixfold in the presence of responder cells as compared to CD25+ cells cultured alone (Fig. 3H). Since IL-2 is known to regulate CD25 expression, we analyzed whether the addition of IL-2 or anti-IL-2 could mimic the effect that the co-culture had on CD25 expression. As shown in Fig. 3E, addition of 30 ng/ml of IL-2 to the co-culture abrogated the inhibition of proliferation and restored CD25 expression on responder T cells, whereas addition of anti-IL-2 blocked the up-regulation of CD25 on Treg in the co-culture system (Fig. 3F and H). The addition of IL-2 was also sufficient to up-regulate CD25 expression on Treg cells cultured alone (Fig. 3G), to the same extent as in the presence of responder T cells (Fig. 3B). Conversely, the blocking of IL-2 in the monoculture of responder T cells resulted in inhibition of both proliferation and CD25 expression (Fig. 3D); which is comparable to the effects observed for co-culture with Treg (Fig. 3B). Taken together, these observations show that IL-2 is not completely suppressed but is produced in low amounts by responder T cells and is selectively taken up by Treg during co-culture.

Figure 3.

CD4+CD25+ Treg take up IL-2 produced by responder T cells. Purified CD4+CD25+ Treg (C, G) and CD4+CD25- responder T cells (A, D) were CFDA-SE labeled and cultured alone or Treg were co-cultured with CFDA-SE labeled responder cells at a ratio of 1:2 (B, E, F) and anti-CD3 and irradiated APC. After 3 days CD4+ cells were analyzed for CD25 expression. Recombinant IL-2 (E, G) or neutralizing anti-IL-2 antibody (D, F) were added as indicated. (H) The CD25 geometric mean fluorescence intensity (MFI) of the Treg populations (A, B, F) is indicated.

2.4 No evidence for de novo induction of a suppressive molecule by pre-activation of Treg

We have shown that IL-2 strongly enhances the suppressive potential of Treg and is required for in vitro inhibition. This could either be due to the induction of a suppressive molecule or an increased capacity to compete for IL-2 as a result of high affinity IL-2 receptor up-regulation. If the expression of a suppressive molecule is induced, in vitro suppression by pre-activated Treg should occur faster as compared to Treg taken freshly ex vivo, since the latter require additional time for activation during co-culture. Alternatively, if Treg simply compete for IL-2, the same kinetics of suppression by Treg would be expected for pre-activated and ex vivo Treg. Furthermore, the kinetics of inhibition in this case should be mimicked by adding a neutralizing anti-IL-2 antibody to responder cells cultured alone.

Since the inhibition of IL-2 production is the hallmark of Treg function in vitro 6, 26, we compared the kinetics of IL-2 suppression induced by Treg with the effect of an anti-IL-2 antibody. Responders were cultured with anti-CD3 and irradiated APC either alone or in the presence of blocking anti-IL-2 or together with CD4+CD25+ Treg at a 2 : 1 ratio. Treg were either taken directly ex vivo or had been pre-activated for 24 h with anti-CD3 plus IL-2. The frequency of IL-2 producing responder T cells was monitored over time using the affinity matrix technology 33, 34. In responder T cells cultured alone, IL-2 production steadily increased from 2–3% positive cells after 1 h to 8% after 9 h and about 25% after 24 h; thereafter, it dropped to 15% at 48 h and 1% at 72 h (Fig. 4). Antibody neutralization of IL-2 did not influence IL-2 secretion during the first 6 h, but blocked its increase observed after 9 h. Interestingly, the suppression of IL-2 production by Treg followed exactly the same kinetics and was first observable after about 9 h irrespective of whether Treg had been pre-activated or not (Fig. 4).

Figure 4.

Kinetics of suppression of IL-2 production. Purified CD4+CD25 T cells from OVA-TCR-transgenic DO.11.10 mice were stimulated with anti-CD3 either alone, in the presence of a neutralizing anti-IL-2 antibody or at a 2:1 ratio with CD4+CD25+ Treg from BALB/c mice. The Treg were either taken freshly ex vivo or had been pre-activated for 24 h with anti-CD3 and IL-2. At the indicated time points the frequency of IL-2-producing cells within the OVA-TCR-transgenic T cell population was determined by IL-2 cellular secretion assay.

These data show that anti-IL-2 mimics Treg-induced suppression and that Treg inhibit IL-2 up-regulation with the same kinetics irrespective of their state of activation at the onset of culture, arguing against a suppressive factor induced by pre-activation. Furthermore, initial IL-2 production by responder T cells, which is independent of IL-2 co-stimulation, cannot be suppressed even by pre-activated Treg. Altogether these results suggest that Treg switch off IL-2 synthesis mainly via competition for IL-2 although an additional suppressive mechanism cannot be ruled out completely.

2.5 T cells become resistant to suppression by Treg in the presence of high doses of IL-2

If the reduction of available IL-2 significantly contributes to suppressive activity of Treg, high doses of IL-2 should abrogate all aspects of suppression. It is known that exogenous IL-2 restores the proliferative capacity of responder T cells. However, this could simply be a secondary effect compensating for the lack of IL-2 production in suppressed cells. Therefore, we analyzed whether the capacity of responder T cells to produce IL-2 is also restored by exogenous IL-2. As shown in Fig. 5A, Treg almost completely suppress IL-2 secretion of responder T cells (26 vs. 5% after 24 h). The addition of IL-2 to the co-culture completely restored the capacity to produce IL-2. In contrast IL-4, IL-7 and IL-15 had no effect on IL-2 production although they partially restored T cell proliferation (Fig. 5B). These data show that responder T cells are completely resistant to Treg suppression in the presence of high doses of IL-2.

Figure 5.

In the presence of IL-2 responder T cells are resistant to Treg suppression. (A) Pre-activated Treg from BALB/c mice were cultured at a ratio of 1:2 with responder T cells from OVA-TCR-transgenic DO.11.10 mice in the presence of anti-CD3 and irradiated APC. Where indicated, either 50 ng/ml of IL-2, IL-4, IL-7 or IL-15 were added to the culture. The frequency of IL-2-producing target cells was determined after 24 h using the IL-2 cellular secretion assay. (B) Pre-activated Treg were cultured at a 1:2 ratio with CFDA-SE-labeled responder T cells for 3 days in the presence of either IL-2, IL-4, IL-7 or IL-15 and proliferation was analyzed after 3 days.

2.6 IL-2 primes Treg for IL-10 production

IL-10 has been described to be involved in Treg function in vivo but not in vitro. In line with published data, blocking anti-IL-10 antibodies did not abrogate suppression and IL-10 secreting cells were not detectable during in vitro culture (data not shown). Treg have been reported to produce IL-10 upon re-stimulation when primed by immunization in vivo28 or when isolated from inflamed tissues 29. We analyzed if IL-2 could be responsible for this effect. CD25+ Treg or CD25 naïve T cells were cultured with APC and anti-CD3 alone or in the presence of IL-2. IL-4 was also added to some cultures since it is a potent inducer of IL-10 in naïve T cells 35 and is required to maintain IL-10 memory in Th2 cells 36. On day 3, the cells were re-stimulated for 5 h using PMA/ionomycin and stained for intracellular IL-10. As shown in Fig. 6, IL-2 induced a small but significant increase in IL-10-producing cells among Treg which was boosted by the addition of IL-4. In contrast, IL-4 had no effect on Treg in the absence of IL-2 but induced IL-10 production in naïve T cells. Conversely, IL-2 did not induce IL-10 in CD25 control cells. This further excludes the possibility that IL-10 producing Treg are derived from a small contaminating population of non-regulatory T cells. No IL-10 production was observed without re-stimulation. These data show that IL-2 has the potency to prime Treg for IL-10 production upon re-stimulation, indicating an additional role for IL-2 in the regulation of Treg function in vivo.

Figure 6.

Treg are primed by IL-2 to produce IL-10 upon restimulation. CD4+CD25+ Treg or CD4+CD62L+ naïve T cells were cultured for 72 h with irradiated APC and stimulated with anti-CD3. Where indicated, 50 ng/ml IL-2 and 20µg/ml of neutralizing anti-IL-4, 50 ng/ml IL-4 and 20 µg/ml anti-IL-2, 50 ng/ml IL-2 and 50 ng/ml IL-4, or an isotype control were added to the culture. After 72 h cells were re-stimulated with PMA/ionomycin and stained intracellularly for IL-10.

3 Discussion

So far, the functional contribution of the IL-2 receptor to the suppressive activity of CD4+CD25+ Treg has not been clarified 7. IL-2 has mainly been described to play an important role for thymic development or the peripheral maintenance of the CD4+CD25+ subset 10, 11, 15, 19, 37. However, Furtado et al. 15 showed that regulatory T cells can be generated in the thymus in the absence of IL-2, whereas IL-2 signals are required at a later developmental stage to generate active suppressors in vivo. Recently, it has also been reported that IL-2 can enhance the suppressive activity of Treg in vitro 21. We corroborate this finding by showing that IL-2 during a short pre-activation period is required to increase the suppressive activity about 20-fold as compared to pre-activation in the absence of IL-2. More importantly, by selectively blocking the IL-2 receptor of Treg we found that a continued uptake of IL-2 is required to exert suppressor function even for previously activated Treg.

This raises the question whether IL-2 signals are needed to maintain the expression of a suppressive molecule or whether competition for IL-2 is already a mechanism of suppression in itself. As we show here, competition for IL-2 is in fact an important regulatory mechanism, inasmuch as Treg do not produce IL-2 although it is essential for their suppressor function. Our data on CD25 expression of Treg vs. responder T cells (Fig. 3) and the kinetics of IL-2 secretion during co-culture (Fig. 4) clearly demonstrate competition by showing that low levels of IL-2 are indeed present during suppression and that this IL-2 is selectively consumed by Treg. The fact that Treg but not responder T cells possess the capacity to constitutively capture IL-2, as observed by binding of an IL-2-IgG fusion protein, provides the molecular basis for this selective uptake. In addition, it is known that IL-2 bound to its high-affinity receptor is rapidly internalized (t½ =10–15 min) and degraded 38, 39, which multiplies the capacity of a single cell to deplete IL-2 from the culture. Similar to our in vitro data, Klein et al. 28 observed in an in vivo system that responder T cells express less CD25 in the presence of Treg suggesting that competition for IL-2 can take place in vivo. Our data also confirm the observation that the level of CD25 expression of human Treg clones positively correlates with their suppressive activity 40.

Our results imply competition for IL-2 as a suppressive mechanism in vitro. All Treg-mediated suppressive effects on IL-2 production, proliferation, and kinetics of responder T cells suppression were exactly mimicked by anti-IL-2 antibodies and were abrogated by exogenous IL-2, but not by other cytokines tested. In particular, we show here for the first time that only IL-2 abrogates suppression of IL-2 production, which has been claimed to be the hallmark of Treg function 6, 26, whereas IL-4, IL-7 and IL-15 restore proliferation but not IL-2 production. These effects are easily explained by a model of competition for IL-2.

However, it is difficult to distinguish suppression by deprivation of IL-2 from a regulatory signal, the expression of which is induced by IL-2. Our results argue against the induction of a suppressive molecule. If the expression of a gene for such a factor would be induced de novo by IL-2, one would expect an earlier onset of suppression by pre-activated Treg as compared to ex vivo Treg. This was not the case in our experiments. Pre-activated Treg, although having a strongly increased suppressive capacity on a per cell basis, reveal similar kinetics of suppression as Treg isolated freshly ex vivo. Both Treg populations are unable to block IL-2 production during the first 6 h, which we have shown to be independent of autokrine IL-2. Instead, they block late (>6 h) up-regulation of IL-2, which depends on autokrine IL-2 and is similarly blocked by adding an anti-IL-2 antibody to responder T cells alone. These results argue again in favor of in vitro suppression by Treg which is based on competition for IL-2 with responder T cells.

Such a model of competition for secreted cytokines is in contrast to the idea of cell contact-dependent suppression mediated by a membrane-bound ligand. Several groups have shown in transwell in vitro culture assays that spatial separation of Treg and responder T cells abrogates suppression 6, 18, 26. This result, however, although often interpreted as prove of a cell contact-dependent suppression, is also compatible with a competition model in view of the local accumulation of cytokines in the immediate vicinity of the secreting cells. Efficient competition for IL-2 requires colocalization of the two cell types, since IL-2 secreted by responder T cells has to be removed immediately from the microenvironment by Treg in order to prevent the autokrine positive feed-back loop of IL-2 production, which we have demonstrated here (Fig. 4). However, the existence of an as yet unidentified suppressive factor relevant also for in vitro suppression cannot be excluded by our experiments.

In vivo IL-2 is probably not required for T cell proliferation, as demonstrated by IL-2- and IL-2R-deficient mice, which even develop a T cell hyperproliferative phenotype 1214, 41. Interestingly, the adoptive transfer of wild-type CD25+ Treg can prevent the pathology of IL-2R-deficient mice but not that of IL-2-deficient mice 8, 11. Although these data argue against competition for IL-2 as the only mechanism of suppression in vivo, they support the idea that IL-2 is required to induce the regulatory function of Treg. Here, we provide evidence that IL-10 might be such a regulatory effector molecule, by showing that IL-2 primes Treg to produce IL-10 upon secondary stimulation. However, it is unlikely that IL-10 contributes to the enhanced in vitro suppressive activity of Treg stimulated by IL-2, since neutralizing anti-IL-10 did not abrogate suppression. Moreover, we failed to detect IL-10 producing Treg upon primary stimulation in vitro (data not shown). This is in accordance with previously published data on the role of IL-10 for in vitro suppression 6, 26.

However, the observation that IL-2 mediates the priming of Treg to express IL-10 upon secondary stimulation might resolve the apparent discrepancies reported for the role of IL-10 in immune regulation in vitro and in vivo22, 28. In vivo, one could envision a biphasic response of Treg (see Fig. 7). Primary activation takes place in lymphoid organs in the presence of responder T cells providing IL-2. If the IL-2 levels are low, i.e. induced by continuous TCR-triggering with self-peptide/MHC complexes, competition for IL-2 would mainly be required to maintain the normal Treg pool. Simultaneously, the removal of IL-2 from the microenvironment would help to suppress unwanted immune reactions by increasing the activation threshold, e.g. of auto-reactive responder T cells. At high local IL-2 levels, i.e. following immunogenic signals, competition between Treg and responder T cells is overcome. Treg will be activated and expand as it has been indicated recently 28. Most importantly, after such primary activation in the presence of IL-2, Treg would now be primed to secrete IL-10 upon re-stimulation with antigen, either in the secondary lymphoid organs or after entering inflamed tissues. This model is supported by recent reports showing that Treg activated by antigen in vivo for a few days or isolated from inflamed tissues produce IL-10 upon re-stimulation ex vivo28, 29.

Figure 7.

Model of Treg regulation by IL-2. Treg activity is regulated by IL-2 and adapted to the strength of the antigenic stimulus of responder T cells (see text for details).

In summary, our data reveal competition for IL-2 as an important mechanism for Treg function. We propose a network of mutual regulation of Treg and responder T cells. By uptake of IL-2, Treg can adjust their suppressive capacity to the strength of a certain immune reaction providing the immune system with a flexible auto-regulatory loop. Furthermore, our model is able to integrate the apparently opposing concepts on the suppressive mechanism of Treg in vitro and in vivo.

4 Materials and methods

4.1 Mice

BALB/c and OVA-TCRtg/tg DO.11.10 mice were purchased from the BgVV (Berlin, Germany). All mice were housed in a specific pathogen-free (SPF) environment and were used at 8 to 10 weeks of age.

4.2 Antibodies

The following anti-mouse antibodies were either conjugated in house or purchased as indicated: anti-CD3 (145-2C11, BD-PharMingen, San Diego, CA), FITC-, PE- or peridinin chlorophyll (PerCP)-conjugated anti-CD4 (GK1.5, BD-PharMingen), allophycocyanin-conjugated anti-CD25 (PC61, BD-PharMingen), biotinylated anti-CD25 (7D4, BD-PharMingen), anti-CD25-PE (7D4, Miltenyi Biotec, Bergisch Gladbach, Germany), anti-CD122 (TM-β1, BD-PharMingen), Cy5-conjugated anti-DO.11.10 OVA-TCR (KJ1.26) and anti-IL-10-PE (JES5-16E3, BD-PharMingen). For neutralization of IL-2 or IL-4, anti-mouse IL-2 (S4B6) or anti-mouse IL-4 (11B11) were added (20–50 µg/ml). The anti-human antibodies anti-CD3 (UCHT-1) and anti-CD28 (CD28.2) were purchased from BD-PharMingen. Cy5- or FITC-conjugated anti-CD4 (TT1) was used for staining. Goat anti-mouse IgG (Dianova, Hamburg, Germany) was used to cross-link anti-CD3 and anti-CD28 in mouse/human co-culture experiment. All antibodies used in cell culture were of NA/LE quality.

4.3 Cell staining and purification

For isolation of CD4+CD25+ Treg lymph node and spleen cell suspensions were stained with anti-CD4-FITC and biotinylated anti-CD25 followed by incubation with anti-biotin microbeads and sorted by AutoMACS™ (Miltenyi Biotec). Subsequently, CD25 cells were labeled with anti-FITC microbeads and sorted for CD4 expression. For isolation of naïve CD4+CD25 T cells, CD25 depleted spleen and lymph node cells were double sorted using CD4-FITC/anti-FITC multisort microbeads and anti-CD62L microbeads. Antigen presenting cells were sorted using anti-MHC class II microbeads.

For isolation of human CD4+ cells, PBMC were labeled with anti-human CD4 microbeads and sorted via AutoMACS™. The purity of the various sorted cell populations was higher than 93%.

4.4 CFDA-SE labeling

CD4+CD25 or CD4+CD25+ T cells were washed with PBS, resuspended in a 1 µM solution of carboxy-fluorescein diacetate-succinimidyl ester (CFDA-SE) (Sigma, St Louis, MO) at a density of 1×107/ml and incubated for 3 min at room temperature. The labeling reaction was stopped by washing with RPMI 1640 culture medium (BioWhittaker, Walkersville, MD) containing 10% fetal calf serum (FCS).

4.5 Pre-activation

Murine CD4+CD25+ or CD4+CD25 T cells were stimulated with plate-bound anti-CD3 alone or in the presence of 50 ng/ml recombinant murine IL-2 (R&D-Systems, GB) or neutralizing anti-IL-2 for 24 h. For coating, the culture dishes were incubated for 3 h at 37°C with 10 µg/ml anti-CD3 in PBS.

4.6 Proliferation assays

The in vitro suppressive activity was analyzed as described previously 42. Briefly, 0.33×106 irradiated APC and 0.18×106 T cells in total were incubated for 72 h with 1 µg/ml anti-CD3 in a 96-well U-bottom plate. CD25+ Treg and CFDA-SE-labeled CD25 responder T cells were mixed as indicated. RPMI-1640 supplemented with 10% heat inactivated FCS, 100 U/ml penicillin plus 100 U/ml streptomycin, 2 mM L-glutamine and 50 µM 2-ME (Sigma) was used for cell cultures. When indicated 50 ng/ml of recombinant cytokines murine rIL-2, rIL-4, rIL-7 or human rIL-15 were added.

For the mouse/human co-culture assay, 0.16×106 pre-activated (see above) murine CD4+CD25+ and 0.33×106 human CD4+ T cells were cultured with 1 µg/ml anti-human CD3 and anti-CD28, and 10 µg/ml goat anti-mouse IgG for 72 h in a 96-well U-bottom plate. The murine IL-2 receptor was blocked with 50 µg/ml anti-CD25 (PC61) and 50 µg/ml anti-CD122 (TM-β1). Proliferation was quantitated by calculating the total number of cell divisions per precursor cell from the number of cells in each generation as described previously.

4.7 Identification of IL-2-secreting cells

IL-2-secreting T cells were identified using the IL-2 secretion assay (Miltenyi Biotec) following the instructions of the manufacturer. Briefly, T cells were activated under the indicated conditions. To distinguish between Treg and responder T cells, we used Treg from BALB/c mice and CD25 responder T cells from DO.11.10 OVA-TCRtg/tg animals. At various time points cells were washed, labeled with the anti-IL-2 affinity matrix and cultured at 37 °C for 45 min at a cell density of about 1×105 cells/ml to prevent cross-feeding of IL-2. Captured IL-2 was then detected by anti-IL-2-PE on CD4+ OVA-TCR+ cells.

4.8 Restimulation and intracellular staining of IL-10

CD4+CD25+ Treg or CD4+ CD62L+ naïve T cells were cultured under the indicated conditions (anti-IL-2 and anti-IL-4 were used at 20 µg/ml; rIL-2 and rIL-4 were added at 50 ng/ml) or taken directly ex vivo, washed with RPMI-1640, 10% FCS and stimulated for 5 h with 10 ng/ml PMA and 1 µg/ml ionomycin (both Sigma). Brefeldin A (Sigma) was added at 5 µg/ml for the last 3 h of stimulation. Following fixation with 2% formaldehyde the cells were permeabilized with 0.5% saponin (Sigma) in PBS/BSA/Azide and stained with PE-conjugated anti-IL-10 and FITC-conjugated anti-CD4.

4.9 Staining with IL-2-IgG2b fusion protein

Freshly isolated spleen and lymph node cells or isolated and pre-activated CD4+CD25+ or CD4+CD25 T cells were incubated with 1 µg/ml IL-2-IgG2b fusion protein (43; gift from Dr. Bulfone-Paus) for 15 min followed by biotinylated anti−mIgG2b antibody (Dianova, Hamburg, Germany) and SA-APC (BD-PharMingen). In addition, the cells were stained for CD4 and CD25 expression. To block binding of the fusion protein, cells were incubated with anti-CD25 and anti-CD122 (25 µg/ml) for 10 min prior to incubation with the fusion protein.

All flow-cytometric analyses were performed on a FACSCaliburTM using CELLQuestTM research software (BD-Biosciences).


This work was supported by the BMBF Kompetenznetz Rheuma (Grant 01 GI 0344). Sascha Rutz was supported by a grant from the Boehringer-Ingelheim Fonds. We thank Alf Hamann, Andreas Radbruch, Andreas Thiel, and Farah Hatam for helpful discussions and critical reading of the manuscript.


  1. 1


  2. 2


  3. 3


  4. 4


  5. 5


  6. 6


  7. 7