Inhibitory and Stimulatory Effects of IL-10 on Human CD8⁺ T Cells¹ Free

Interleukin-10, which is produced by a variety of cells, including T lymphocytes, B lymphocytes, and monocytes, has been identified as a cytokine with important antiinflammatory and immunosuppressive properties. In vitro studies have shown that IL-10 inhibits Ag-specific activation and proliferation of human peripheral blood T cells and T cell clones belonging to the Th0, Th1, or Th2 subsets (1, 2). This inhibition is associated with a reduction of the Ag-presenting capacity of monocytes. IL-10 down-regulates monokine production (3, 4) and the expression of CD54 (ICAM-1, the ligand for LFA-1), CD80, and CD86, which function as important costimulatory molecules for T cell activation (5, 6, 7). Furthermore, IL-10 down-regulates constitutive and IFN-γ- or IL-4-induced class II MHC expression on monocytes, dendritic cells, and Langerhans cells (2, 8, 9). It has also been shown that IL-10, in the absence of professional APC, has direct effects on CD4⁺ T by suppressing IL-2 and TNF-α secretion (10, 11). IL-10 also strongly reduces the proliferation and cytokine production of alloreactive T cells in MLR. Moreover, the generation of allospecific cytotoxic activity is inhibited by IL-10 (12). We have recently shown that in addition to inhibiting Ag-specific responses, IL-10 induces long lasting Ag-specific anergy in human CD4⁺ T cells (13).

On the other hand, IL-10 also has immunostimulatory effects on B and T cells. IL-10 augments the proliferation and the differentiation into Ab-secreting cells of splenic activated B cells (14, 15). IL-10 also rescues T cells from apoptotic cell death (16). In addition, it has been reported that IL-10 enhances the proliferative responses of murine thymocytes (17) and IL-2- and IL-4-driven proliferation of murine CD8⁺ T cells in vitro (18). IL-10 also has immunostimulatory effects in vivo. Injection of high doses of IL-10 (200 μg/mice/day) in mice with graft-vs-host disease, due to both major and minor MHC disparities, resulted in a exacerbation of the disease, which was probably mediated by IFN-γ production (19). In addition, in an IL-10 transgenic model, IL-10 enhances the accumulation of CD8⁺ T cells in the pancreata (20), leading to an earlier onset of diabetes in nonobese diabetic mice (21). These data suggest that IL-10 may have differential effects on CD4⁺ and CD8⁺ T cells.

In the present study, we demonstrate that IL-10 has a dual effect on human CD8⁺ T cells. IL-10 suppresses the proliferative responses and induces alloantigen-specific unresponsiveness in human CD8⁺ T cells activated in the presence of professional APC. These effects are indirect and are mediated through inhibition of the costimulatory functions of APC. In contrast, IL-10 has no direct inhibitory effect on the proliferation of CD8⁺ T cells activated via their TCR in the absence of APC and displays a growth-promoting activity on activated CD8⁺ T cells in combination with low doses of IL-2.

PBMC were isolated on Ficoll-Hypaque. CD8⁺ T cells were purified by negative selection using a mixture of Abs directed against non-CD8⁺ cells: CD4, CD14, CD16, CD19, CD20, CD56, and HLA-DR. Cells were incubated with saturating amounts of Abs for 20 min at 4°C. After washing, Dynabeads (Dynal, Oslo, Norway) were added at a ratio of 10 beads/target cell and incubated for 1 h at 4°C. Beads and contaminating cells were removed by magnetic field. The remaining cells were resuspended with the same amount of beads, and a second incubation period for 1 h at 4°C was performed. After removal of contaminating cells, CD8⁺ T cells were analyzed by cytofluorometry and showed >90 to 95% positive. In some experiments, CD8⁺ T cells were purified (>98% pure) by positive selection using magnetic beads coated with anti-CD8 mAb according to the manufacturer’s instructions (Dynal). Monocytes were purified by negative selection using the procedure described above with an Ab mixture containing CD2, CD3, CD8, CD16, CD19, CD20, and CD56. These monocytes were >95% CD14⁺ as shown by cytofluorometry.

Purified recombinant IL-10 was provided by Schering-Plough Research Institute (Bloomfield, NJ), and rIL-2 was a gift from Dr. S. Menon (DNAX Research Institute, Palo Alto, CA). The anti-CD3 mAb SPV-T3 (22), anti-IL-2 (BG-5), anti-IL-2R (B-B10) (23) or anti-IL-10 (9D7) mAbs were previously described. Nonconjugated, phycoerythrin-conjugated, or FITC-conjugated CD2, CD3, CD4, CD8, CD14, CD16, CD19, CD20, CD28, CD54, CD56, CD80, HLA class I, HLA-DR, and controls mAbs of the appropriate isotypes were purchased from Becton Dickinson (Mountain View, CA) except for CD86 (PharMingen, San Diego, CA).

All proliferation assays were conducted in Yssel’s medium (24) supplemented with 10% FCS and 1% human serum. MLR were performed by stimulating purified CD8⁺ T cells (10⁵ cells/well) with purified irradiated (4000 rad) monocytes (10⁵ cells/well) in 200-μl round-bottom 96-well plates (Linbro, ICN, Aurora, OH). CD8⁺ T cell proliferation was measured after 5 days of incubation at 37°C and 5% CO₂ and a subsequent 12-h pulse with [³H]TdR, after which the cells were harvested as described (25).

For plate-bound anti-CD3 mAbs activation, 100 ng/ml of anti-CD3 mAbs diluted in 0.1 M Tris buffer, pH 9.5, were incubated for 1 week at 4°C in flat-bottom 96-well plates. After the plates were washed, CD8⁺ T cells were added at a concentration of 5 × 10⁴ cells/well. CD8⁺ T cell proliferation was measured after 3 days of incubation followed by a 12-h pulse with [³H]TdR. All tests were conducted in triplicate.

Ag-specific unresponsiveness was induced by culturing CD8⁺ T cells at 10⁶ cells/ml with purified irradiated allogeneic monocytes (10⁶ cells/ml), in the presence of IL-10 (100 U/ml) in 24-well plates. After 10 days, cells were collected, layered on a Ficoll gradient, and centrifuged for 30 min at 1200 rpm to remove dead cells. The viable cells were collected from the interface, washed twice, and restimulated at 5 × 10⁵ cells/ml with irradiated allogeneic monocytes (5 × 10⁵ cell/ml) from the same donor in 96-well plates.

Cell surface Ag expression on monocytes incubated with IL-10 for different period times was determined by immunofluorescence. Monocytes (10⁵) were labeled with phycoerythrin- or FITC-conjugated mAbs. Cells were incubated with the appropriate Ab for 30 min at 4°C in PBS with 0.1% BSA and 0.02 mM NaN₃. After three washes, the labeled cell samples were analyzed on a FACScan (Becton Dickinson, San Jose, CA).

The cytotoxic activity was measured as described previously (12). In brief, 2 × 10^{3 51}Cr-labeled target cells (PHA blasts) were incubated with graded numbers of CD8⁺ T effector cells in 200 μl of medium in round-bottom microtiter plates (Linbro, ICN). The plates were centrifuged and incubated for 4 h at 37°C. The supernatants were harvested with a Skatron (Sterling, VA) harvesting system and counted in a gamma counter. All test samples were measured in triplicate. The percentage of specific ⁵¹Cr release was calculated as described (12).

IL-10 inhibited in a dose-dependent manner the proliferation of purified CD8⁺ T cells in response to purified allogeneic irradiated monocytes in primary MLR. Strong inhibitory effects were observed at IL-10 concentrations as low as 10 U/ml, whereas maximal inhibition was observed at 100 U/ml (Fig. 1,A). These inhibitory effects of IL-10 on the proliferation of purified CD8⁺ T cells were completely restored by exogenous IL-2 (Fig. 1 A).

To analyze whether the inhibition of proliferation by IL-10 reflected a direct effect on CD8⁺ T cells, highly purified CD8⁺ T cells were activated by cross-linked anti-CD3 mAbs, in the presence or absence of IL-10. No inhibitory effects of IL-10 on the proliferative responses of CD8⁺ T cells stimulated with different doses of anti-CD3 mAbs were observed (Fig. 1,B) even after addition of up to 10,000 U/ml of IL-10 (Fig. 1,A). These results suggest that the inhibitory effect of IL-10 on alloantigen-specific CD8⁺ T cell proliferation is indirect and is caused by inhibition of the APC function of monocytes. This was confirmed by the observation that IL-10 inhibited specifically the costimulatory functions of autologous monocytes on anti-CD3 mAb-induced proliferation of CD8⁺ T cells (Fig. 1,B). Moreover, the inhibitory effect of IL-10 was observed only after preincubating with IL-10 allogeneic monocytes before coculture (Fig. 1,C). No inhibition was observed after preincubation of CD8⁺ T cells with IL-10 (Fig. 1,C). Finally, transwell experiments where CD8⁺ T cells were stimulated with immobilized anti-CD3 mAb and cocultured with autologous monocytes either on close contact or separated by a semipermeable membrane have shown that IL-10 inhibited both the secretion of stimulatory monokines and costimulatory cell surface molecules (Fig. 1 D) and did not induced the secretion of a suppressive factor.

Recently, we showed that CD4⁺ T cells stimulated by alloantigens in the presence of IL-10 for prolonged periods of time failed to respond when restimulated with the same specific Ag (13). To analyze whether IL-10 had the same effect on CD8⁺ T cells, purified CD8⁺ T cells were stimulated by allogeneic monocytes in the presence or absence of IL-10 for 10 days. After this culture period, alloantigen-stimulated CD8⁺ T cells were restimulated with the same allogeneic irradiated monocytes in secondary MLR or with third-party allogeneic monocytes in primary MLR. CD8⁺ T cells that had been cultured in the absence of IL-10 proliferated strongly in response to the allogeneic stimuli (Fig. 2). In contrast, CD8⁺ T cells activated and incubated in the presence of IL-10 showed a strong reduction in their secondary proliferative responses to the same allogeneic monocytes (Fig. 2). However, they retained their capacity to proliferate to third-party alloantigens in primary MLR (Fig. 2). Comparable levels of cell death were observed in both culture conditions, and comparable cell numbers were harvested at termination of the assays (data not shown). These results indicate that IL-10 induces an Ag-specific unresponsive state in CD8⁺ T cells in a manner similar to that described for CD4⁺ T cells (13).

To determine whether the induction of the unresponsive state in CD8⁺ T cells following activation by allogeneic monocytes in the presence of IL-10 was due to a down-regulation of costimulatory signals, monocytes were incubated with IL-10 for different time periods, and the expression of costimulatory cell surface molecules was analyzed. Flow cytometric analysis by direct staining with a mAb against nonpolymorphic determinants of HLA class I indicated that IL-10 down-regulates HLA class I expression, but to a lesser extent than HLA class II expression (Fig. 3). In addition, IL-10 down-regulated the expression of the costimulatory molecules CD80 and CD86 and the adhesion molecule ICAM-1 (Fig. 3), as demonstrated previously (6, 26). However, no change in CD58 expression was observed (Fig. 3). Collectively, these results indicate that IL-10-induced alloantigen-specific unresponsiveness in CD8⁺ T cells is indirectly mediated by APC and is associated with down-regulation of costimulatory molecules and to a lesser extent of class I MHC Ags.

To test whether IL-10 also inhibits the generation of alloantigen specific cytotoxic T cells, purified CD8⁺ T cells were stimulated with purified allogeneic monocytes on primary MLR in the presence of different doses of IL-10. Five days later, the CD8⁺ T cells were collected and used as effector cells against autologous or allogeneic PHA blasts. CD8⁺ CTL killed the allogeneic T-cell blasts in a dose-dependent manner (Fig. 4), whereas no significant cytotoxic activities were observed against autologous T cell blasts (not shown). The alloantigen-specific cytotoxicity by CD8⁺ T cells that had been activated in the presence of IL-10 was strongly reduced (Fig. 4). Significant inhibition of specific cytotoxicity was already observed at 10 U/ml of IL-10, whereas >90% reduction in cytotoxic activity was obtained at an IL-10 concentration of 100 U/ml.

To determine whether IL-10 has direct effects on the proliferation of human CD8⁺ T cells, human purified CD8⁺ T cells were stimulated with cross-linked anti-CD3 mAb in the presence of anti-IL-2 or anti-IL-2R mAbs or both and IL-10. Anti-IL-2 mAbs strongly inhibited the proliferation of purified CD8⁺ T cells activated with anti-CD3 (Fig. 5,A). Similar results were obtained by adding anti-IL-2R α-chain (CD25) mAbs or both Abs, demonstrating that this proliferation is IL-2 dependent. Interestingly, addition of high concentrations of IL-10 (100 U/ml) restored CD8⁺ T cell proliferation, indicating that IL-10 may act as a growth factor for CD8⁺ T cells in the absence of IL-2. However, the proliferation of CD8⁺ T cells activated with immobilized anti-CD3 mAbs in the presence of saturating amounts of anti-IL-2 and/or anti-IL-2 R mAbs was never completely blocked even in the presence of a blocking anti-IL-10 mAb (Fig. 5,A), suggesting that IL-10 may act only as a cofactor for CD8⁺ T cell proliferation in the presence of low concentrations of IL-2. To test this possibility, IL-10 was added to CD8⁺ T cells preactivated by immobilized anti-CD3 mAb and cultured in the presence of IL-2. IL-10 enhances IL-2 driven proliferation of preactivated CD8⁺ T cells, especially when low IL-2 and high IL-10 concentrations were used (Fig. 5 B). IL-10 alone, in the absence of IL-2, consistently failed to induce significant levels of CD8⁺ T cell proliferation. This is because CD8⁺ T cells that have been preactivated for 5 days failed to produce detectable amounts of IL-2 (not shown). These data suggest that IL-10 synergizes with a low concentration of IL-2 in enhancing CD8⁺ T cell proliferation.

IL-10 inhibits monocyte-dependent, Ag-specific proliferation and cytokine production by human and murine CD4⁺ T cells (27, 28). Furthermore, IL-10 inhibits alloantigen-induced T-cell proliferation and allo-specific cytotoxic T cell generation (12). In addition to its suppressor activity on T cell proliferation, IL-10 has potent antiinflammatory effects (29) and inhibits the generation of reactive nitrogen intermediates by mouse macrophages (29). IL-10 also suppresses the production of proinflammatory cytokines (TNF-α, IL-1α, IL-1β, IL-6) and chemokines (IL-8, MIP-1α) by activated human monocytes (3, 29). These suppressive and antiinflammatory activities suggest a potential clinical use of IL-10 as a potent immunosuppressant in allogeneic transplantation and autoimmune diseases. The results of experimental (30) and clinical studies (31) support this hypothesis. High levels of endogenous IL-10 were observed in successfully transplanted SCID patients in whom tolerance was established (25). Furthermore, a correlation between high survival rates and high concentrations of endogenous IL-10 has been described in patients with malignancies transplanted with allogeneic bone marrow (32). These findings suggest that the concentrations of IL-10 may be important to prevent graft-vs-host disease and determine the outcome of the transplant. Moreover, in vivo injection of IL-10 in humans has been shown to have inhibitory effects on T cells and to suppress the production of the proinflammatory cytokines TNF-α and IL-1β (31).

However, other in vivo studies have shown that IL-10 was not able to suppress an immune response. IL-10 was not able to prevent graft-vs-host disease (33) and in some situations even had exacerbating effects (19) when administered either at the same time as or after bone marrow transplant. In addition, IL-10 transgenically expressed in β cells did not prevent or modify rejection of allogeneic fetal pancreata or adult β cells (34). The discrepancy between these in vivo effects of IL-10 in humans and mice may be due to the fact that under certain conditions IL-10 may act as an immunostimulant instead of as an immunosuppressant. In this study, we demonstrate that IL-10 has differential immunosuppressive or stimulatory effects on CD8⁺ T cells depending on their stage of activation.

IL-10 decreases proliferation and cytotoxic functions of CD8⁺ T cells when added before or at the time of activation. Moreover, human CD8⁺ T cells activated by allogeneic monocytes in the presence of IL-10 for 10 days were rendered unresponsive to a secondary stimulation with the same Ag. However, these T cells proliferated normally in response to third-party alloantigens. Similar results have been obtained with CD4⁺ T cells (13). In contrast to the results obtained with CD4⁺ T cells, IL-10 failed to decrease the proliferative responses of CD8⁺ T cells induced by activation with cross-linked anti-CD3 mAbs. Furthermore, no unresponsive state was achieved after incubation of CD8⁺ T cells with IL-10 for 10 days following stimulation with cross-linked CD3 mAb. Taken together, these data indicate that T cell unresponsiveness induced by IL-10 is not related to a direct effect on CD8⁺ T cells but to inhibition of Ag-presenting and accessory function of monocytes, which is associated with down-regulation of monokine secretion and expression of class I MHC and accessory molecules on these cells (Figs. 1 and 3).

It is well established that mouse CD4⁺ and CD8⁺ T cell clones lose their ability to proliferate and to secrete IL-2 and that they become anergic when stimulated with fixed APC (35, 36). Additional experiments have shown that these activation conditions allow TCR engagement in the absence of the costimulatory signals mediated by APC. Indeed, it has been shown, mostly for CD4⁺ T cell clones, that anergy is induced when T cells are stimulated by APC in which the costimulatory signals are blocked by Abs directed against key costimulatory molecules such as CD80, CD86, CD54, or CD58 (37, 38, 39). However, CD4⁺ and CD8⁺ T cells may have different requirements for costimulation and induction of anergy, which could explain the differences in their susceptibility to be rendered unresponsive by IL-10. In CD4⁺ T cells expressing CD80, costimulatory signals provided by T-T cell interaction can be sufficient to reach the threshold required to induce cell proliferation after CD3 cross-linking (40). Il-10-mediated down-regulation of these costimulatory signals provided by CD4⁺ T cells may explain the observed anergic state of CD4⁺ T cells. CD8⁺ T cells seem to be less dependent on costimulatory signals since these cells could be induced to proliferate in response to stimulation by fibroblasts, whereas CD4⁺ T cells failed to do so. Therefore, it may be speculated that because of this relative independence of costimulatory signals, IL-10 is not effective in inhibiting proliferation and inducing unresponsiveness of purified CD8⁺ T cells activated in the absence of professional APC.

The failure of CD8⁺ T cells to proliferate in response to allogeneic monocytes in the presence of IL-10 correlated with a reduced capability of these CD8⁺ T cells to lyse allogeneic target cells. These results may reflect a decrease in the frequency of effector cells due to IL-10 inhibition of T cell activation in primary MLR. Alternatively, it is possible that a functional impairment of the CD8⁺ T cells is induced by in vitro priming in the presence of IL-10, resulting in lack of both proliferation and cytotoxic activity. This latter hypothesis is supported by a recent study in which it is shown that IL-10 is responsible for the lack of cytotoxic activity by specific CD8⁺ T cell clones and that blocking of endogenous IL-10 production restored the cytotoxic activity (41). In contrast to the suppressive effects of IL-10 on the proliferative and cytotoxic responses of CD8⁺ T cells activated in the presence of APC, IL-10 enhances the proliferation of preactivated human CD8⁺ T cells induced by IL-2. Therefore, IL-10 has a stimulatory function on activated CD8⁺ T cells by acting as a cofactor for IL-2. The mechanism by which IL-10 acts as a cofactor for IL-2 in promoting the proliferation of activated CD8⁺ T cells is unclear. However, in contrast to a previous report on CD4⁺ T cells (42), IL-10 does not enhance the IL-2R α-chain on activated CD8⁺ T cells (not shown). Moreover, the observation that IL-10 sustains CD8⁺ T cell proliferation in the presence of an anti-IL-2R mAb suggests that the synergy between IL-10 and IL-2 is not mediated through IL-2R expression.

Overall, these data suggest that IL-10 may inhibit or stimulate CD8⁺ T cells, depending on the time at which the cells are exposed to IL-10. These differential effects may explain some discrepancies observed in vivo in different experimental models. It can be hypothesized that IL-10 is effective in vivo in preventing and inhibiting undesirable immune responses mediated by CD8⁺ T cells, if present at the same time as Ag exposure, but that it fails to suppress ongoing responses mediated by activated CD8⁺ T cells. This should be taken into account for the future potential clinical use of IL-10 in autoimmune diseases and organ transplantation.

We thank Lewis Lanier for critical review of the manuscript and JoAnn Katheiser for secretarial assistance.