- NKT cells:
Natural killer T cells
Regulatory T cells
CD4+CD25+ and CD1d-restricted natural killer T (NKT) cells are thymus-derived self-reactive regulatory T cells that play a key role in the control of pathological immune responses. Little is known about functional cooperation between innate regulatory NKT cells and adaptive CD4+CD25+ regulatory cells. Here we show that human CD4+Vα24+Vβ11+ (CD4+ NKT) cells isolated from peripheral blood by flow cytometric cell sorting secrete substantial amounts of IL-2 after stimulation with dendritic cells (DC) and α-Galactosylceramide. When cocultured with CD4+CD25+ cells, CD4+ NKT cells promoted moderate proliferation of CD4+CD25+ cells. The proliferation of CD4+CD25+ T cells was due to soluble IL-2 produced by activated CD4+ NKT cells. The expanded CD4+CD25+ cells remained anergic and retained their potent suppressive properties. These findings indicate that unlike conventional CD4+ and CD8+ T cells, which are susceptible to CD4+CD25+ regulatory cell suppression, NKT cells promote CD4+CD25+ regulatory cell proliferation. These data raise the possibility that NKT cells can function as helper cells to CD4+CD25+ regulatory T cells, thereby providing a link between the two naturally occurring populations of regulatory T cells.
T cell-mediated regulation plays a key role in the prevention of autoimmunity, in transplantation tolerance and in the inhibition of collateral damage during immune responses to microbial infections. Recently, several populations of regulatory T cells (Treg) have been identified and characterized in both mice and humans, including naturally occurring self-reactive natural killer T (NKT) cells 1–3 and CD4+CD25+ cells 4, 5. NKT cells are a unique subset of T cells expressing a canonical αβ T cell receptor, Vα14Jα281 pairing with Vβ8 in the mouse and Vα24JαQ pairing with Vβ11 in man. They are restricted by the monomorphic MHC class I-like molecule CD1d, which presents self-antigen glycolipids. Isoglobotrihexosyl-ceramide (iGb3), a lysosomal glycosphingolipid, has been identified as the natural ligand recognized by both murine and human NKT cells 6. α-Galactosylceramide (α-GalCer) derived from marine sponges can also stimulate NKT cells 1–3. After activation NKT cells secrete Th1 and Th2 cytokines, which are thought to be responsible for their regulatory functions in vivo1–3. NKT cells are positively selected in the thymus 7–9 by CD1d-expressing CD4+CD8+ cortical thymocytes 10 in the presence of thymic stromal cell-derived IL-15 11 and IL-7 7, 12, and are exported to the periphery. In humans they represent less than 0.1% of peripheral blood lymphocytes 13, 14.
In contrast, CD4+CD25+ cells represent 5–15% of human peripheral CD4+ T cells 4, 5. CD4+CD25+ cells are functionally anergic after conventional T cell receptor (TCR) stimulation in vitro4, 5, but appear to be able to proliferate in vivo15–18. Several lines of evidence have suggested that thymic development and peripheral expansion of CD4+CD25+ cells seem to be dependent on stimulation by self-MHC class II-peptide complexes and local endogenous IL-2 19–21. In IL-2–/– knockout mice, CD4+CD25+ regulatory cells were not detected in the thymus or the periphery 19. Furthermore, transgene-encoded thymic expression of IL-2Rβ in IL-2Rβ–/– mice could prevent the autoimmune disease normally seen in these animals 20. IL-2 is also required for CD4+CD25+ cell function in vivo21; the source of IL-2 in this context, however, is unclear. Interestingly, one of the main effects of CD4+CD25+ regulatory cells is to inhibit IL-2 transcription by effector CD4+ and CD8+ T cells 4, 5.
One line of evidence has shown that human CD4+CD25+ Treg are able to suppress NKT cell lines in vitro22. It is not clear whether NKT cells could affect CD4+CD25+ regulatory cell function. Here we have investigated the interaction between freshly isolated human CD4+CD25+ and NKT regulatory cells ex vivo. The results indicate that NKT cells are rapid and efficient producers of IL-2 and that activated NKT cells promoted moderate proliferation of CD4+CD25+ regulatory cells. These findings imply a helper role of CD1d-restricted NKT cells for CD4+CD25+ regulatory T cells, thereby providing a link between innate and adaptive mechanisms of immune regulation.
Isolation of human NKT cells and CD4+CD25+ cells from PBMC
Human NKT cells have been defined by staining PBMC with CD1d-αGalCer tetramers 13, 14 or with antibodies specific for TCR Vα24+Vβ11+23. The frequency of NKT cells in healthy individuals has been estimated to be between 0.01% and 0.1% of fresh PBMC 13, 14, 23. To isolate this rare cell population, we first stained PBMC with anti-TCR Vα24 and anti-TCR Vβ11 antibodies and then enriched for Vα24+ cells using microbeads (Fig. 1A). Vα24+Vβ11+ cells were then selected from the enriched Vα24+ population using flow cytometric cell sorting. CD4+CD25+ cells were purified from the Vα24– population using magnetic beads as described 24. Fewer than 50% of Vα24+Vβ11+ cells expressed CD4, but none expressed CD8, indicating that they are CD4+ NKT cells or CD4– (so-called DN) NKT cells (Fig. 1A). To isolate CD4+ and CD4– NKT cells, the enriched Vα24+ cell population was stained with anti-CD4 antibody, and CD4+Vα24+Vβ11+ and CD4–Vα24+Vβ11+ cells were then sorted using flow cytometry. The purity of sorted NKT cells was more than 99%, and the purity of CD4+CD25+ cells was greater than 95% (data not shown). From 500 ml peripheral blood of nine healthy individuals, between 1.5×104 and 5.5×105 NKT cells were isolated compared with 2×106 to 12×106 CD4+CD25+ cells (Fig. 1B). The yields of selected NKT cell and CD4+CD25+ cell populations were 0.001–0.04% and 0.2–1.2% of total PBMC, respectively. Fig. 1C illustrates the numbers of isolated CD4+ and CD4– NKT cells from 500 ml peripheral blood of five healthy individuals. The numbers of CD4– NKT cells were higher than CD4+ NKT cells in all five individuals.
CD4+ NKT cells but not CD4– NKT cells secrete abundant amounts of IL-2 after activation
NKT cells have been extensively studied for their ability to produce either Th1 or Th2 cytokines (such as IL-4, IL-10, IL-13, TNF-α and IFN-γ) after activation, which are suggested to be responsible for their immunoregulatory functions in vivo1–3. It has been reported that when human peripheral blood lymphocytes are incubated overnight with α-GalCer, intracellular staining for IL-2 can be detected in the CD1d-αGalCer tetramer-positive population by flow cytometric analysis 14. To assess the capacity of freshly isolated peripheral blood NKT cells to produce IL-2, they were stimulated with irradiated allogeneic DC in the presence of α-GalCer. IL-2 secretion was measured analyzing the proliferation of CTLL-2 cells in response to culture supernatant removed at different time points after stimulation. As illustrated in Fig. 2A, NKT cells produced IL-2 promptly after stimulation, and production was sustained from 2 to 4 days. No IL-2 production was detected when NKT cells were stimulated with irradiated allogeneic DC only (data not shown). We next compared the IL-2-producing potential of CD4+ and CD4– NKT cells. As shown in Fig. 2B, when stimulated with irradiated allogeneic DC in the presence of α-GalCer, CD4+ NKT cells produced much more IL-2 than an equal number of CD4– NKT cells. As positive controls, 5×104 CD4+ T cells were stimulated under the same conditions, i.e. by irradiated allogeneic DC in the presence of α-GalCer. To compare the IL-2-producing capacity of CD4+ and CD4– NKT cells on a cell-to-cell basis, the cells were stimulated with anti-CD3 and anti-CD28 antibody-coated beads. Again, CD4– NKT cells produced less IL-2 than CD4+ NKT cells, even when double the cell number was present (Fig. 2C). Taken together, these data indicate that freshly isolated NKT cells, particularly the CD4+ subset, are capable of producing substantial amounts of IL-2 when activated by DC and α-GalCer or by anti-CD3/CD28 antibody-coated beads.
CD4+ NKT cells promote proliferation of CD4+CD25+ Treg
Freshly isolated CD4+CD25+ cells from peripheral blood are functionally anergic 4, 5, 24. However, they appear to be able to proliferate in vivo15–18. The peripheral expansion of CD4+CD25+ cells seems to require local endogenous IL-2 19–21. Interestingly, the key function of CD4+CD25+ regulatory cells is the inhibition of IL-2 transcription by effector CD4+ and CD8+ T cells 4, 5, 24. To search for the source of IL-2, we first tested the ability of DC to stimulate CD4+CD25+ cells. When CD4+CD25– or CD4+CD25+ cells were stimulated with irradiated allogeneic DC, CD4+CD25+ cells were much less proliferative than CD4+CD25– cells, indicating that they are anergic (Fig. 3A). The weak proliferation of CD4+CD25+ cells stimulated by allogeneic DC could not be observed by analyzing cell division of CFSE-labeled CD4+CD25+ cells (data not shown). CD4+CD25+ cells inhibited the proliferation of CD4+CD25– cells in a dose-dependent manner (Fig. 3A) and also inhibited IL-2 production by CD4+CD25– cells (data not shown) 24. When compared with CD4+CD25+ cells alone, allogeneic DC induced weak proliferation of CD4+CD25+ cells (Fig. 3B). The weak proliferation was enhanced when exogenous IL-2 was added (Fig. 3B). Addition of neutralizing anti-IL-2 antibody abolished the proliferation, suggesting that IL-2 derived from DC and/or CD4+CD25+ cells plays a role in the weak proliferation of CD4+CD25+ cells stimulated by allogeneic DC (Fig. 3B) 15.
To investigate the potential of NKT cells to promote CD4+CD25+ cell proliferation, we cocultured CD4+CD25+ cells with irradiated CD4+ or CD4– NKT cells at a ratio of 13:1 or 5:1, and the cell culture was stimulated with irradiated allogeneic DC. α-GalCer was added to all cultures. We used irradiated NKT cells because these cells can proliferate after TCR cross-linking (data not shown). As shown in panel C of Fig. 3, CD4+ NKT cells, but not CD4– NKT cells, enhanced the proliferation of CD4+CD25+ cells. The moderate increment of CD4+CD25+ cell proliferation was dependent on the numbers of CD4+ NKT cells added (Fig. 3D). The enhanced proliferation could not be observed when α-GalCer was not present in the culture (data not shown). To determine the role of IL-2 produced by NKT cells in the proliferation of CD4+CD25+ cells, neutralizing anti-IL-2 antibody was used. Indeed, addition of anti-IL-2 antibody abolished the proliferation of CD4+CD25+ cells stimulated by irradiated allogeneic DC in the presence of irradiated CD4+ NKT cells (Fig. 3C). To determine whether the enhanced proliferation of CD4+CD25+ cells in the presence of irradiated NKT cells was due to surface molecules expressed by activated NKT cells instead of secreted IL-2, CD4+CD25+ and irradiated NKT cells were separated by a semi-permeable membrane. The proliferation of CD4+CD25+ cells in the transwell system was doubled compared with that of CD4+CD25+ cells stimulated by allogeneic DC alone (Fig. 3E). Taken together, these results imply that NKT cells are able to produce IL-2 in the presence of CD4+CD25+ regulatory cells, and soluble IL-2 secreted by activated NKT cells is able to promote proliferation of CD4+CD25+ cells stimulated by allogeneic DC.
The CD4+CD25+ cells cocultured with NKT cells remain anergic and retain their potent suppressive properties
Constitutive expression of cytotoxic T lymphocyte-associated antigen (CTLA)-4 is one of the hallmarks of CD4+CD25+ Treg4, 5. We next analyzed the surface molecules expressed by the CD4+CD25+ cells cocultured with irradiated NKT cells. CTLA-4 expression was significantly higher on the CD4+CD25+ cells stimulated by allogeneic DC in the presence of α-GalCer with or without irradiated NKT cells than on freshly isolated CD4+CD25+ cells or CD4+CD25+ cells stimulated with IL-2 and α-GalCer only (Fig. 4A). The CD4+CD25+ cells retained high levels of CD25 and low levels of CD69 expression (data not shown). Surface expression of glucocorticoid-induced TNF receptor (GITR) was observed on neither fresh CD4+CD25+ cells nor those cocultured with NKT cells (data not shown). These data indicate that the presence of irradiated NKT cells has no effect on the phenotype of CD4+CD25+ cells stimulated by allogeneic DC.
To assess the suppressive functions of the CD4+CD25+ cells cocultured with irradiated NKT cells, fresh CD4+CD25– cells were mixed with the cocultured CD4+CD25+ cells and stimulated by irradiated allogeneic DC. The CD4+CD25+ cells inhibited proliferation of CD4+CD25– cells in a dose-dependent manner. More than 80% suppression was seen at a 1:8 ratio of suppressor cells to responder cells, and the percentage inhibition increased to 95% at a ratio of 1:1 (Fig. 4B). The CD4+CD25+ cells cocultured with irradiated NKT cells were more potent suppressor cells than freshly isolated CD4+CD25+ cells. Only 10% suppression was achieved when the ratio of fresh CD4+CD25+ to CD4+CD25– cells was 1:8, and when the ratio was increased to 1:1, only 70% suppression was observed (Fig. 4C). This is consistent with our recent observations that activated CD4+CD25+ cells or cell lines possess more potent suppressive properties than freshly isolated CD4+CD25+ cells (Jiang and Lechler, manuscript submitted). The CD4+CD25+ cells stimulated by DC in the presence or absence of irradiated NKT cells showed a similar degree of suppression, suggesting that NKT cells have no influence on the suppressive functions of CD4+CD25+ cells.
Our results demonstrate that peripheral blood CD4+ NKT cells secrete substantial amounts of IL-2 after activation, even at very low cell numbers. When cocultured with CD4+CD25+ Treg, CD4+ NKT cells were able to promote CD4+CD25+ cell proliferation. However, the expanded CD4+CD25+ cells retained their potent suppressive functions. These data demonstrate that unlike other cells such as conventional CD4+ and CD8+ T cells or B cells, which are susceptible to CD4+CD25+ regulatory cell suppression, NKT cells are capable of promoting CD4+CD25+ regulatory cell proliferation. The moderate increment of CD4+CD25+ regulatory cell proliferation by NKT cells may be due to the fact that in vitro expansion of human CD4+CD25+ cells requires very strong TCR stimulation, i.e. anti-CD3 and anti-CD28 antibody-coated beads and the presence of very high doses of IL-2 (Jiang and Lechler, manuscript submitted). This implies a role for CD1d-restricted NKT cells as innate helper cells for CD4+CD25+ Treg.
Several studies have demonstrated that human NKT cell lines or clones are able to provide help to various cell types. One study showed that NKT cell clones can induce DC maturation in vitro25. NKT cell clones were also shown to induce proliferation of and immunoglobulin production by autologous memory and naive B cells 26. IL-2 production by human NKT cell lines could activate NK cells, thereby mediating their anti-tumor activity indirectly 27. Our data indicate that freshly isolated CD4+ NKT cells from human peripheral blood can function as helper cells to CD4+CD25+ Treg. A recent study showed that human CD4+CD25+ regulatory cells can suppress proliferation of and cytokine production by Vα24+ NKT cell lines 22. It is possible that freshly isolated CD4+ NKT cells and in vitro-cultured Vα24+ NKT cell lines have different susceptibilities to CD4+CD25+ cell-mediated suppression under different stimulation conditions 28.
A direct link between NKT cells and CD4+CD25+ Tregin vivo has not been demonstrated. One line of evidence showed that after repetitive injection of α-GalCer into mice, the expression of CD25 on splenocytes was increased twofold 29. Interestingly, in a peripheral tolerance induction model involving anterior chamber-associated immune deviation, CD1d-restricted CD4+ NKT cells were able to induce IL-10-producing regulatory CD8+ T cells in vivo30. In a murine study of oral tolerance to nickel, CD4+ NKT cells were also required for the induction of specific Treg31. Although the involvement of CD4+CD25+ Treg in these models has yet to be defined, this study supports the concept that activation of CD1d-restricted NKT cells may also foster the generation of other populations of Tregin vivo. Our findings suggest that further investigation of CD1d-restricted NKT cell help to CD4+CD25+ Treg could shed further light on the interaction between innate and adaptive mechanisms of immune regulation and facilitate the development of more effective therapeutic strategies for the control of autoimmune disease, transplant rejection and allergy.
Materials and methods
The murine CTLL-2 cell line proliferates in a dose-dependent manner in response to murine IL-2 and IL-4, but only to human IL-2 or IL-15. Cells were maintained in RPMI 1640 medium supplemented with 10% FCS and 20 μg/ml gentamicin (Sigma, UK) in the presence of 10 U/ml recombinant human IL-2 (Roche, Switzerland).
Purification of CD4+ NKT, CD4– NKT, CD4+CD25– and CD4+CD25+ T cell populations
PBMC from healthy blood donors were separated by density gradient centrifugation over Lymphoprep (Nycomed, Birmingham, UK). For isolation of NKT cells, PBMC were stained with anti-TCR Vα24-Phycoerythrin (PE) (clone C15, Immunotech, France) and anti-TCR Vβ11-Fluorescein isothiocyanate (FITC) (clone C21, Immunotech) antibodies and then enriched for Vα24+ cells using anti-PE Miltenyi beads (Miltenyi, UK). Vα24+Vβ11+ cells were then selected from the enriched Vα24+ population using flow cytometric cell sorting. For isolation of CD4+ and CD4– NKT cells, the enriched Vα24+ cell population was stained with anti-CD4-Tricolor (TC) antibody, and CD4+ and CD4– NKT cells were then sorted by flow cytometry. CD4+CD25+ and CD4+CD25– cells were purified from the Vα24– population using Dynabeads® (Dynal, Norway). PBMC were first incubated in medium supplemented with 2% FCS at 37°C in tissue culture flasks twice for 45 min to remove adherent cells. The non-adherent cells were then collected and washed. Non-CD4+ cells were depleted by incubation with a cocktail of antibodies: anti-CD8 (clone OKT8), anti-CD14 (BA8, Diaclone, UK), anti-CD16 (B-E16, Diaclone), anti-CD19 (BC3, Diaclone), anti-CD33 (WM53, Diaclone), anti-CD56 (BA19, Diaclone) and anti-TCR-γ/δ-1 (Becton Dickinson, UK) followed by magnetic bead separation using anti-mouse IgG-coated beads (Dynal). CD4+CD25– cells were obtained by incubation with anti-CD25 magnetic beads (Dynal), and CD4+CD25+ cells were obtained by using Detachbeads (Dynal). The purities of CD4+CD25– and CD4+CD25+ populations were determined by flow cytometry and were routinely >95%.
DC were prepared from adherent cells derived from allogeneic PBMC in RPMI 1640 medium supplemented with 5% human serum (Sigma) and 20 μg/ml gentamicin. Recombinant human GM-CSF (1,000 U/ml; Glaxo, UK) and IL-4 (1,000 U/ml; R&D Systems, UK) were added on the initial day of culture. Cytokines were replenished every other day (days 2 and 4) by removing 0.3 ml medium and adding back 0.5 ml fresh medium containing cytokines. On day 5 or 6, DC were harvested and used for stimulation of T cell cultures.
T cell proliferation assay
Freshly isolated CD4+CD25+ cells (2×104 per well) were cultured with irradiated (60 Gy) allogeneic DC (4,000 per well) in the presence or absence of irradiated (60 Gy) NKT cells in a 96-well round-bottomed microplate. α-GalCer (100 ng/ml; Kirin, Japan) was added directly at the initiation of the culture. The final volume per well was 200 μl. The cultures were labeled with [3H]thymidine (0.5 μCi per well) after 72 h incubation and harvested 18 h later. [3H]thymidine incorporation was then measured in an LK Betaplate liquid scintillation counter (Wallac, Inc., UK). Mean counts per minute (cpm) of triplicate cultures and the standard deviation of the mean were calculated.
Freshly isolated CD4+CD25– T cells (2×104 per well) were stimulated with irradiated (60 Gy) allogeneic DC (4,000 per well) in the presence of different numbers of freshly isolated CD4+CD25+ T cells or CD4+CD25+ T cells that had been cocultured with irradiated NKT cells in the presence of α-GalCer (100 ng/ml). The cultures were labeled with [3H]thymidine (0.5 μCi per well) after 5 days of incubation and harvested 18 h later.
Transwells of pore size 0.4 μm were used (Costar, UK). Freshly isolated CD4+CD25+ T cells (5×105 cells) were stimulated with irradiated allogeneic DC (1×105 cells) in the lower chamber of a transwell plate in the presence of α-GalCer (100 ng/ml). Irradiated allogeneic DC (1×105 cells) in the presence of α-GalCer (100 ng/ml) with or without irradiated NKT cells (1×105 cells) were placed in the upper or lower chamber of the transwell plate. Cells from the lower chamber (CD4+CD25+ T cells and irradiated DC) were transferred to a 96-well plate 4 days later, labeled with [3H]thymidine (0.5 μCi per well) and harvested after 18 h incubation.
CTLL-2 cell proliferation assay
Supernatant (50 μl) was added to CTLL-2 cells (1×104 per well) in 150 μl medium. The cultures were labeled with [3H]thymidine (0.5 μCi per well) after 24 h incubation and were harvested 18 h later. In each experiment different concentrations of standard recombinant IL-2 were used as positive controls.
Statistical analyses were performed using the Student's t-test. Differences between values were considered statistically significant at p<0.05.
We are grateful to Prof. Andrew George for invaluable discussions and comments. We thank Drs. Gary Warnes, Kirsty Allen, Ayad Eddaoudi and Cathy Simpson of the FACS Laboratory at the Cancer Research Institute UK for their excellent expertise in flow cytometric cell sorting.