Plasmacytoid dendritic cells (PDCs) are key effectors in host innate immunity and orchestrate adaptive immune responses. CpG oligodeoxynucleotides (ODN) have potent immunostimulatory effects on PDCs through TLR9 recognition and signaling. Little is known about the effects of CpG ODN on human PDC-mediated T cell priming. Here we show that type B CpG ODN effectively promotes PDCs to prime allogeneic naive CD4+CD25− T cells to differentiate into CD4+CD25+ regulatory T (Treg) cells. The CD4+CD25+ T cells induced by CpG ODN-activated PDCs express forkhead transcription factor 3 and produce IL-10, TGF-β, IFN-γ, and IL-6, but low IL-2 and IL-4. These CD4+CD25+ T cells are hyporesponsive to secondary alloantigen stimulation and strongly inhibit proliferation of autologous or allogeneic naive CD4+ T cells in an Ag-nonspecific manner. CpG ODN-activated PDCs require direct contact with T cells to induce CD4+CD25+ Treg cells. Interestingly, IL-10 and TGF-β were undetectable in the supernatants of CpG ODN-stimulated PDC cultures. Both CpG-A and CpG-C ODN-activated PDCs similarly induced the generation of CD4+CD25+ Treg cells with strong immune suppressive function. This study demonstrates that TLR9 stimulation can promote PDC-mediated generation of CD4+CD25+ Treg cells and suggests PDCs may play an important role in the maintenance of immunological tolerance.
Dendritic cells (DCs)3 are specialized APCs that play a unique role in orchestrating both innate and adaptive immunity. Multiple DC subsets have been identified (1). These subsets may have unique roles in regulating immunity and tolerance (2, 3, 4). One of these DC subsets is the lineage− (lin−) HLA-DR+CD11c−CD123+BDCA2+BDCA4+ plasmacytoid DCs (PDCs), also known as IFN-αβ-producing cells (5, 6, 7). Upon viral or bacterial stimulation, PDCs in human blood and peripheral lymphoid tissues rapidly produce large amounts of IFN-αβ, differentiate into mature CD11c− DCs and stimulate T cell-mediated adaptive immune responses. PDCs are known to play an important role in immunological tolerance. Blood PDCs matured with CD40 ligand (CD154) induce the generation of Th2 cells producing IL-4, IL-5, and IL-10 (8, 9). Naive CD8+ T cells primed with CD154-matured PDCs differentiated into IL-10-producing CD8+ regulatory T (Treg) cells that strongly inhibit allospecific proliferation of naive CD8+ T cells in a primary MLR (10). Studies in mice suggest that PDCs may have a role in the generation of CD4+ Treg cells (11, 12, 13). PDCs have also been reported to stimulate T cells and drive Th1 polarization (14). These findings suggest that the distinct capacity of PDCs to induce a Th1, Th2, or Treg response may largely depend upon the signals that induce their activation and maturation (2).
Recent studies of bacterial DNA containing unmethylated CpG motifs as immunostimulatory agents demonstrated that certain CpG DNA sequences can rapidly activate human PDCs isolated from blood, promoting their maturation and survival (15, 16, 17). Although the exact mechanism of CpG motif recognition and downstream signaling is not known, the receptor for CpG DNA has been identified as TLR9, a member of the Toll receptor family that comprises an elegant pathogen recognition system for host defense in innate immunity (18). Human TLR9 is found on PDCs but not on myeloid DCs (19, 20), thus, human PDCs but not myeloid DCs are directly responsive to CpG DNA stimulation (17, 19). To mimic the stimulatory capacity of bacterial CpG DNA, synthetic oligodeoxynucleotides (ODN) containing the signature CpG dinucleotides (CpG ODN) and various flanking sequences and backbones have been used in both human and mouse studies (21). At least three distinct classes of CpG ODNs with structural and functional differences have been identified (22, 23, 24). The CpG-A (also known as D type) ODN (2216) consists of a chimeric phosphorothioate and phosphodiester backbone, one or more central CpG dinucleotides arranged in a palindrome, and poly-G motifs (“G-tetrads”) at either or both the 5′ and 3′ ends (22, 25, 26). The CpG-B (also known as K type) ODN (2006) consists of one or more CpG motifs on a phosphorothioate backbone (27, 28). CpG-C ODN (2395) consists of a 5′ stimulatory hexameric CpG motif linked by a T spacer to 3′ palindromic sequences that are preferably GC-rich (23, 24). All three classes of CpG ODN are known to potently activate human PDCs. The major differences between these classes of CpG ODN include that CpG-A and CpG-C ODN, but not CpG-B ODN, can effectively induce high levels of IFN-α production from PDCs (22) and that CpG-B ODN and CpG-C ODN, but not CpG-A ODN, are strong stimulators of human B cells (23). However, little is known about the effects of these CpG ODN on human PDC-mediated T cell priming.
Although studies suggest that PDCs play an important role in immunological tolerance induction (29), it is unknown whether human PDCs may play a role in the generation and function of CD4+CD25+ Treg cells. Recent studies have shown that CD4+CD25− T cells can be induced to differentiate into CD4+CD25+ T cells with regulatory cell function by cytokines (e.g., IL-10, TGF-β) (30, 31, 32), stimulation with immature myeloid DCs (33, 34, 35), or immunomodulatory drugs such as vitamin D3 (36). In this study, we assessed the effects of CpG ODN on purified human PDC-mediated priming of allogeneic naive CD4+ T cells. We demonstrate that CpG ODNs promote human PDC-mediated priming of naive CD4+CD25− T cells to differentiate into CD4+CD25+ Treg cells that potently suppress autologous and allogeneic T cell proliferation.
Materials and Methods
PBMC preparation, PDC, and T cell purifications
Human PBMC were isolated from apheresis products of healthy blood donors (Memorial Blood Centers of Minnesota, Minneapolis, MN) by Ficoll-Paque density gradient centrifugation. Blood PDCs were enriched from PBMC using blood DC Ag (BDCA)-4 cell isolation kits and the MACS system. The BDCA-4+ cell-enriched preparation was then stained with a mixture of FITC-conjugated mouse anti-human Abs directed against lineage (Lin) markers (CD3, CD14, CD16, CD19, CD20, and CD56), allophycocyanin-conjugated anti-CD11c, and PE-conjugated anti-CD123 (IL-3Rα) Abs. The labeled cells were sorted on a FACSVantage SE (BD Biosciences, San Jose, CA) to collect the Lin−CD11c−CD123+ PDCs. The purity of sorted PDCs was consistently higher than 98%. CD4+CD45RA+ naive T cells were isolated from PBMC by using CD4 T cell isolation kits followed by positive selection with CD45RA microbeads. The purity of naive CD4+ T cells was higher than 95% for CD4+CD45RA+ expression and <0.5% for CD25+ expression. Blood CD4+CD25+ natural-arising Treg cells were purified from CD4+ selected T cells by labeling cells with CD25 microbeads and positively selecting CD4+CD25+ T cells. B cells were purified from PBMC by labeling cells with CD19 microbeads and positively selecting CD19+ B cells. All cell isolations kits and microbeads were from Miltenyi Biotec (Bergisch Gladbach, Germany).
Phosphorothioate-modified CpG ODNs were obtained from Coley Pharmaceutical Group (Wellesley, MA). CpG-A ODN 2216: ggGGGACGATCGTCgggggG; CpG-B ODN 2006: tcgtcgttttgtcgttttgtcgtT; CpG-C ODN 2395: tcgtcgttttcggcgcgcgccg (sequences are shown 5′ to 3′; lower case letters, phosphorothioate linkage; capital letters, phosphodiester linkage 3′ of the base; bold, CpG dinucleotides). No endotoxin could be detected in CpG ODN preparations (<0.03 EU/ml, Limulus amebocyte lysate assay; BioWhittaker, Walkersville, MD). All three classes of CpG ODN were resuspended in TE buffer, diluted in PBS, and used at a final concentration of 3 μg/ml.
In vitro priming of naive CD4+ T cells
Purified CD4+CD45RA+ T cells were incubated with allogeneic PDCs, CD11c+ DCs, or irradiated B cells (30 Gy) at a 10:1 ratio in 24-well plates (e.g., 2 × 106 CD4+CD45RA+ T cells plus 2 × 105 PDCs per well) with or without the designated CpG ODN in complete medium (RPMI 1640 supplemented with 10% human AB serum, 2 mM glutamine, 0.1 mM nonessential amino acids, 1 mM sodium pyruvate, 50 μM 2-ME). As controls, naive CD4+ T cells were cultured in complete medium with or without CpG ODN. In Transwell experiments, naive CD4+ T cells were incubated with Transwells containing sorted allogeneic PDCs at a 10:1 ratio in 24-well plates (2 × 106 naive CD4+ T cells plus a Transwell containing 2 × 105 PDCs) in complete medium with or without CpG ODN. After 6–7 days, primed T cells were harvested, analyzed for their cell surface phenotype and cytokine production profile, and assessed for their function in MLR assays.
FITC-, PE-, allophycocyanin-, or CyChrome-conjugated mouse anti-human Abs directed against CD3, CD4, CD11c, CD25, CD40, CD45, CD45RA, CD45RO, CD80, CD86, CD123, HLA-DR, CTL-associated Ag-4 (CTLA-4: CD152), as well as isotype control Abs were from BD Pharmingen (San Diego, CA). FITC-conjugated mouse anti-human Abs directed against lineage (Lin) markers were from BD Immunocytometry Systems (San Jose, CA). Cells were stained with fluorescent Abs for 30 min on ice, washed twice with 0.2% FBS HBSS, and fixed with PBS containing 0.2% paraformaldehyde. Mean fluorescence intensity and positive cell percentages were determined by flow cytometry.
Cytokine production assays
The cytokine profile of PDC-primed CD4+ T cells were determined by incubating naive CD4+ T cells and allogeneic PDCs with or without CpG ODN. Naive CD4+ T cells cultured in complete medium alone were used as control. After 6–7 day, cells were harvested, washed, and restimulated at 1–2 × 106
T cell proliferation assays
Naive CD4+ T cells were primed with allogeneic PDCs or B cells for 6–7 days with or without CpG ODN. The function of primed CD4+ T cells were determined by plating these primed T cells at graded doses as responders to irradiated allogeneic PBMC in MLR cultures or as third-party T cells into MLR cultures in which freshly purified autologous or allogeneic naive CD4+ T cells were stimulated with irradiated allogeneic PBMC. In some experiments, CD25+ and CD25− T cell populations in the primed CD4+ T cells were separated using CD25 microbeads (Miltenyi Biotec) and used in functional assays. In all T cell proliferation assays, plates were incubated at 37°C for 5 day or the indicated time points and pulsed with 1 μCi of [3H]thymidine per well for the last 18 h before harvesting. All determinations were conducted in triplicate and [3H]thymidine incorporation was determined.
RT-PCR for TLR9 and Foxp3 expression
Total RNA was extracted from 2 × 106 freshly purified PDCs, CD4+CD25+ or CD4+CD25− T cells purified from PBMC or from CD4+ T cells primed with PDCs or PDCs plus CpG ODN. The RNA was reversely transcripted to cDNA, and analyzed for TLR9 and Foxp3 expression. PCR parameters: 94°C 5 min; 35 cycles: 94°C 30 s, 55°C 30 s, 72°C 1 min; and 72°C 7 min. The sequences of PCR primers for TLR9 (5′-TTATGGACTTCCTGCTGGAGGTGC-3′ and 5′-CTGCGTTTTGTCGAAGACCA-3′) and for Foxp3 (5′-ATGCCCAACCCCAGGCCTGGC-3′ and 5′-CTCCAGAGACTGTACCATCTC-3′) were as reported (19, 37
Data from experiments are expressed as the mean ± SD. Statistical analysis of the results between groups was performed by Student’s t test. Values of p < 0.05 were considered significant.
CpG-B ODN-activated PDCs induce the generation of CD4+CD25+ T cells that are hyporesponsive and strongly suppress naive CD4+ T cell proliferation in MLR assays
Freshly isolated human PDCs from peripheral blood are weak APCs associated with low expression of costimulatory molecules such as CD40, CD80, and CD86. Triggering TLR9 by CpG-B ODN (2006) rapidly activated PDCs to up-regulate surface expression of CD40, CD80, and CD86 (Fig. 1⇓A) and to produce IFN-α, TNF-α, IL-6, but not IL-10, TGF-β, or IL-12 (Fig. 1⇓B). None of these cytokines were detected in the supernatants from PDCs cultured in medium alone (data not shown). Although freshly isolated PDCs induced a low proliferation of allogeneic naive CD4+ T cells, the presence of ODN 2006 increased PDC-induced proliferation of allogeneic naive CD4+ T cells by 2.0- to 6.4-fold (Fig. 2⇓A). The function of naive CD4+ T cells primed with ODN 2006-activated PDCs (ODN 2006-PDC) was assessed in secondary MLR cultures. CD4+ T cells primed with irradiated allogeneic PBMC or B cells exhibited a secondary proliferative response to the primed alloantigens during the 5-day MLR cultures. In contrast, ODN 2006-PDC primed CD4+ T cells fail to mount a secondary proliferative response to the primed alloantigens during the 5-day MLR cultures (Fig. 2⇓B and data not shown). Moreover, when the ODN 2006-PDC primed CD4+ T cells were added at graded doses into primary MLR cultures in which autologous naive CD4+ T cells were stimulated with irradiated alloantigens from the same priming donor, the proliferation of naive CD4+ T cells to alloantigens was suppressed in a primed T cell dose-dependent manner (Fig. 2⇓C). In contrast, CD4+ T cells primed with ODN 2006-activated B cells that proliferated in response to secondary alloantigen stimulation did not suppress the proliferation of autologous naive CD4+ T cells to the same alloantigens (Fig. 2⇓C). These findings indicate that CpG-B ODN-activated PDCs induce the generation of a CD4+ T cell population that is hyporesponsive to alloantigens and capable of suppressing naive T cell proliferation in MLR.
We next determined whether CpG-B ODN-PDCs induce the generation of CD4+CD25+ Treg cells from CD4+CD25− naive T cells. Freshly isolated naive CD4+CD45RA+ T cells did not express CD25, CTLA-4, or CD45RO Ags (Fig. 3⇓A). Naive CD4+ T cells cultured in media with or without ODN 2006 showed no phenotypic difference. Culturing naive CD4+ T cells with allogeneic PDCs induced CD4+CD25− T cells to differentiate into CD4+CD25+ T cells (Fig. 3⇓B). The frequency of CD4+CD25+ T cells generated in the PDC-primed T cells was significantly increased from 0.4% before culture to 9.5 ± 4.5% at day 7 of culture. The addition of ODN 2006 significantly promoted the PDC-induced generation of CD4+CD25+ T cells and increased the frequency of CD4+CD25+ T cells to 31.1 ± 9.7% at day 7 of culture (Fig. 3⇓, B and D). Separating PDCs and naive CD4+ T cells in culture by transwells abrogated the capability of PDCs to induce CD4+CD25+ T cells with or without ODN 2006 (Fig. 3⇓B). This finding suggests that PDC-induced generation of CD4+CD25+ T cells requires direct contact between PDCs and T cells. CD4+CD25+ T cells induced by ODN 2006-activated PDCs expressed CD45RO and CTLA-4 Ags (Fig. 3⇓C). Recent studies have shown that naturally arising mouse CD4+ Treg cells express certain TLRs, including TLR4, 5, 7, and 8 (38). It is not known whether human naturally arising or ex vivo induced CD4+CD25+ Treg cells express TLR9. RT-PCR experiments showed that PDCs expressed high levels of TLR9, whereas neither of the purified T cell populations (CD4+CD25+ and CD4+CD25− T cells purified from PBMC and PDC-primed CD4+ T cells with or without ODN 2006) expressed TLR9 (Fig. 4⇓A). This finding, together with the results that incubation of naive CD4+ T cells with ODN 2006 alone failed to induce CD4+CD25+ T cells excludes the possibility of direct stimulatory effect of CpG ODN on naive CD4+ T cells or PDC-primed CD4+ T cells, and suggests that ODN 2006 activates PDCs to promote the generation of CD4+CD25+ T cells.
CD4+CD25+ T cells induced by ODN 2006-activated PDCs express Foxp3 and produce a Treg cell cytokine profile
Foxp3, a forkhead transcription factor, has been identified as a gene preferentially expressed in CD4+CD25+ Treg cells. The expression of Foxp3 is associated with the development and function of CD4+CD25+ Treg cells in mice and humans (39, 40). RT-PCR experiments were performed to determine Foxp3 expression in CD4+CD25+ and CD4+CD25− T cells isolated from PBMC or from ODN 2006-PDC primed naive CD4+ T cell cultures. Strong Foxp3 expression was detected in naturally occurring CD4+CD25+ Treg cells purified from PBMC (Fig. 4⇑A). Negligible Foxp3 expression was detected in PDCs and naive CD4+ T cells purified from PBMC. Intriguingly, CD4+CD25+ T cells isolated from either PDC-primed or ODN 2006-PDC primed naive CD4+CD25− T cell cultures showed strong Foxp3 expression, whereas the CD4+CD25− T cells purified from both PDC-primed or ODN 2006-PDC primed naive CD4+ T cell cultures remained negative for Foxp3 expression (Fig. 4⇑A). Real-time quantitative RT-PCR results showed that Foxp3 expression in natural occurring CD4+CD25+ Treg cells and the CD4+CD25+ T cells induced by ODN 2006-activated PDCs were 27.9- and 20.4-fold higher than Foxp3 expression in the naive CD4+CD25− T cells, respectively (Fig. 4⇑B). Analysis of cytokine production showed that PDC-primed CD4+ T cells produced higher amounts of IL-10 and IFN-γ than those produced by naive CD4+ T cells cultured in media alone. The addition of ODN 2006 to PDC-T cell priming cultures enhanced T cell production of IL-10, TGF-β, IFN-γ, IL-6 but not IL-2 or IL-4 (Fig. 5⇓). These findings demonstrate that type B CpG ODN enhances PDC-induced differentiation of CD4+CD25− T cells to CD4+CD25+ T cells that express Foxp3 and produce a Treg cytokine profile.
CD4+CD25+ T cells induced by ODN 2006-activated PDCs suppress autologous and allogeneic T cell proliferation in an Ag-nonspecific manner
Functional analysis of naive CD4+ T cells (donor A) primed with allogeneic PDCs (donor C), with or without the presence of ODN 2006, showed that CD4+ T cells primed under either culture condition were hyporesponsive to secondary alloantigen stimulation and failed to mount a secondary proliferative response to the primed alloantigens (data not shown). When PDC-primed or ODN 2006-PDC primed CD4+ T cells (donor A vs C) were added to a primary MLR in which purified autologous (donor A) naive CD4+ T cells were stimulated with the primed alloantigens (donor C), they strongly suppressed T cell proliferation in a primed T cell dose-dependent manner (Fig. 6⇓A). ODN 2006-PDC primed CD4+ T cells were more effective than PDC-primed CD4+ T cells in suppressing naive CD4+ T cell proliferation in MLR. To determine whether this difference is attributable to the higher frequency of CD4+CD25+ Treg cells present in the ODN 2006-PDC primed CD4+ T cell population, the CD4+CD25+ T cells were isolated. Functional analysis showed that purified CD4+CD25+ T cells were not only immunosuppressive but also were more efficient than the total ODN 2006-PDC primed T cell population in suppressing autologous T cell proliferation to alloantigens in MLR (Fig. 6⇓B). In subsequent experiments, naive CD4+ T cells (donor A) primed by ODN 2006-activated PDCs (donor C) were separated into CD4+CD25+ and CD4+CD25− T cell populations. These CD4+CD25+ T cells (donor A vs C) were hyporesponsive to secondary alloantigen stimulation and failed to mount a secondary proliferative response to either the primed alloantigens (donor C) or third-party alloantigens (donor D) (Fig. 7⇓A). When the purified CD4+CD25+ T cells were added at graded doses into primary MLR cultures in which purified autologous (donor A) or allogeneic (donor B) naive CD4+ T cells were stimulated with irradiated PBMC from donor C or donor D, they strongly suppressed the proliferation of both autologous and allogeneic naive CD4+ T cells in a CD4+CD25+ T cell dose-dependent manner (Fig. 7⇓A). The CD4+CD25+ T cells effectively inhibited the naive CD4+ T cell proliferation to alloantigens at a suppressor to responder ratio lower than 1:33. Depletion of CD4+CD25+ T cells abrogated the immunosuppressive effect of the ODN 2006-PDC primed CD4+ T cells (Fig. 7⇓A). Kinetic analysis of primary MLR cultures containing the CD4+CD25+ or CD4+CD25− T cells purified from ODN 2006-PDC primed CD4+ T cell cultures (donor A vs C) showed that CD4+CD25+ T cells consistently suppressed the proliferation of autologous naive CD4+ T cells (donor A) to alloantigen stimulation during the 5-day culture (Fig. 7⇓B). The suppression mediated by ODN 2006-PDC induced CD4+CD25+ T cells in the MLR system was CD4+CD25+ T cell dose-dependent. The addition of the CD4+CD25− T cells into the primary MLR cultures did not alter the proliferation curves of freshly purified autologous naive CD4+ T cells to alloantigens during the same time course. CD4+CD25+ T cells, but not CD4+CD25− T cells, from naive CD4+ T cells primed with allogeneic PDC alone, similarly suppress the proliferation of autologous and allogeneic naive CD4+ T cell to allogeneic stimulation in an Ag-nonspecific manner (data not shown). These results demonstrate that CD4+CD25+ T cells induced by PDCs or CpG ODN-activated PDCs suppress autologous and allogeneic T cell proliferation in an Ag-nonspecific manner.
CpG-A and CpG-C ODN activated PDCs also induce the generation of CD4+CD25+ Treg cells that suppress autologous and allogeneic T cell proliferation in MLR
CpG-A ODN (2216) and CpG-C ODN (2395) are also known to potently activate human PDCs. Unlike CpG-B ODN, both CpG-A and CpG-C ODN can effectively induce high levels of IFN-α production from PDC precursors. To determine whether CpG-A or CpG-C ODN-activated PDCs similarly induce the generation of CD4+CD25+ Treg cells, naive CD4+ T cells were cultured with allogeneic PDCs in the presence or absence of CpG-A, CpG-B, or CpG-C ODN. Interestingly, PDC activation by all three classes of CpG ODN (class A, B, and C) induced the generation of CD4+CD25+ T cells that strongly suppress autologous and allogeneic T cell proliferation in MLR (Fig. 8⇓ and data not shown). The phenotypic change and cytokine production profile of ODN 2216-activated PDCs was similar to that shown with CpG-B ODN in Fig. 1⇑, except for IFN-α production. IFN-α accumulation in 24–48 h culture supernatants from ODN 2216-activated PDCs was 1532- to 1940-fold (24213–35110 pg/ml vs 13.9–22.4 pg/ml) higher than that induced by ODN 2006-activated PDCs. Similar to the results obtained using ODN 2006 activated PDCs, CD4+25+ T cells from naive CD4+ T cells primed with ODN 2216 activated PDCs were also hyporesponsive and failed to mount a secondary proliferative response to either primed (donor C) or third-party alloantigens (donor D) (Fig. 9⇓A). In experiments mirroring those done with ODN 2006 (Fig. 7⇑), ODN 2216-PDC primed CD4+25+ T cells suppressed the proliferation of autologous and allogeneic naive CD4+ T cells to alloantigenic stimulation. Kinetic analysis again revealed that the naive CD4+ T cell proliferation to alloantigen was consistently suppressed throughout the 5-day culture, and that suppression was CD4+25+ T cell dose dependent. Addition of ODN 2216-PDC primed CD4+CD25− T cells failed to suppress naive T cell proliferation to alloantigen at any point during the 5-day culture (Fig. 9⇓B). These results demonstrate that PDCs activated by all three classes of CpG ODN induce the generation of CD4+CD25+ T cells that strongly inhibit proliferation of naive CD4+ T cells in an Ag-nonspecific manner.
Recent studies demonstrate that DCs not only play a key role in the induction of immune responses, but also participate in the induction and maintenance of immune tolerance (4). In this study, we demonstrate that human PDCs prime naive CD4+CD25− T cells to differentiate into CD4+CD25+ Treg cells characterized as Foxp3+ IL-10-producing immunosuppressive T cells. Treatment of human PDCs with all three classes (type A, B, and C) of CpG ODN-induced the generation of CD4+CD25+ Treg cells with strong Ag-nonspecific immunosuppressive effects on naive CD4+ T cell proliferation. These findings provide new insights to understanding the immunostimulatory effects of CpG ODN on human PDCs and their potential therapeutic application in the induction and maintenance of immune tolerance.
It is known that blood PDCs can differentiate into mature CD11c− DCs following immunostimulatory signals from microbial pathogenic stimuli. These maturation stimuli are recognized by the engagement of specific receptors (e.g., TLR9) on DCs upon microbial infection or by cytokines produced during inflammation or infection (16, 19). PDCs have been shown to induce either a Th2 or a Th1 response in naive CD4+ T cells depending upon the stimulatory signals (8, 9, 14). A recent study also showed that CD154-matured PDCs prime naive CD8+ T cells to differentiate into IL-10-producing CD8+ Treg cells that display poor secondary proliferative and cytolytic responses to alloantigen but strongly inhibit allospecific proliferation of naive CD8+ T cells in a primary MLR (10). The cytokine profile of HSV-infected PDCs primed T cells is characterized by production of IFN-γ and IL-10, low IL-4 and no IL-5. CD154-matured PDCs induced a Th2 type response that resulted in significant production of IL-4, IL-5, IL-10 but some IFN-γ in PDC-primed naive CD4+ T cells. Unlike the cytokine profile reported in these prior studies, our results show that CpG ODN-activated PDCs prime allogeneic naive CD4+ T cells to produce IL-10, TGF-β, IFN-γ but negligible levels of IL-2 and IL-4, a typical Treg cell type cytokine profile (41, 42, 43, 44). Our findings demonstrate that CpG ODN preferentially polarize PDC-primed CD4+ naive T cells to differentiate into CD4+CD25+ Treg cells.
Little is known about human CD4+CD25+ Treg cell interactions with other immune-regulatory cells such as subsets of DCs. Studies by Jonuleit et al. (33) showed that in vitro repetitive weekly stimulations of human naive CD4+ T cells with allogeneic monocyte-derived immature DCs induced an IL-10-producing, nonproliferating CD4+CD25+ Treg cell population. It has been suggested that immature DCs play an important role in maintaining immune tolerance and prime T cells to differentiate into regulatory/suppressor T cells whereas mature DCs prime T cells to induce strong immune responses. In mice, Treg cell development and/or function controlled by immature DCs can be reversed by CD154 ligation or TLR stimulation (45, 46, 47). However, this paradigm does not apply to the previous finding that CD154-activated human PDCs are tolerogenic (10). Our results further demonstrate that human PDCs activated by CpG ODN not only induce but enhance the generation of CD4+CD25+ Treg cells with strong immunosuppressive function. We show that blood PDC precursors can prime allogeneic naive CD4+ T cells to generate CD4+CD25+ Treg cells in a 7-day culture. The addition of CpG ODN effectively promoted PDC-induced generation of CD4+CD25+ Treg cells with phenotypic and functional properties similar to the reported CD4+CD25+ Treg cells isolated from human peripheral blood (42, 48, 49, 50). These findings suggest that the capacity of DCs in maintaining immune tolerance may not be simply attributed to their immature stage (low levels of expression of MHC molecules, CD80, and CD86), it may well depend upon the subtypes of DCs and the signals that induce their activation. The induction of CD4+CD25+ Treg cells by CpG ODN-activated PDCs requires PDC-T cell direct contact. There is a possibility that although CpG ODN-treated PDCs up-regulate costimulatory molecules, there may be a concomitant up-regulation of suppressive surface molecules such as Ig-like transcripts 3 and 4 (51, 52) and Notch receptor ligands (50, 53, 54) as well as down-regulation of surface markers such as glucocorticoid-induced TNF receptor ligand (55, 56, 57). Interestingly, we did not detect IL-10 and TGF-β, known to induce CD4+25− T cells to acquire regulatory function (30, 31, 32), in supernatants derived from CpG ODN-stimulated PDC cultures. There is also a possibility that PDCs may produce other cytokines that serve to polarize naive CD4+ T cells to differentiate into CD4+CD25+ Treg cells. Although the molecular mechanisms by which PDCs induce CD4+CD25+ Treg cells remain to be investigated, the findings from this study demonstrate that human PDCs play a role in CD4+CD25+ Treg cell generation and that CpG ODN can efficiently promote the PDC-induced generation of CD4+CD25+ Treg cells. In vivo, the effects of CpG ODN on PDC-mediated T cell priming will be complicated by the presence of mixed cell populations. Proinflammatory cytokines produced by CpG ODN-activated PDCs will lead to subsequent activation of CD11c+ immature DCs and NK cells, which produce other cytokines that may affect the microenvironment and outcome of naive T cell priming. The role of CpG ODN in promoting PDC-mediated generation of Treg cells is likely balanced by other immune effects in vivo.
In experimental bone marrow transplantation models, donor CD4+CD25+ Treg cells have been shown to play a pivotal role in preventing graft-versus-host disease (GVHD) and in tolerance induction to allogeneic hemopoietic cell transplants (HCT) (58, 59, 60, 61, 62). CD4+CD25+ T cells isolated from donor mice are potent inhibitors of alloresponses in vitro and induce marked protection from lethal GVHD in vivo (58, 60, 61, 62). Removal of CD4+CD25+ T cells present in the graft dramatically accelerates GVHD, whereas the addition of either freshly isolated or in vitro expanded CD4+CD25+ donor Treg cells significantly delays or prevents GVHD (58, 60, 61, 62). Adoptive transfer of ex vivo activated, cultured CD4+CD25+ T cells resulted in significant inhibition of rapidly lethal GVHD (58, 61, 62). These findings show the great potential of Treg cells as new therapeutics for controlling GVHD in allogeneic HCT (63). Several clinical studies indicated that PDCs may play an important role in modulating immune responses after HCT to facilitate engraftment and prevent GVHD reactions (64, 65). Our finding that human PDCs prime allogeneic naive CD4+ T cells to generate CD4+CD25+ Treg cells with strong immunosuppressive effects of alloresponses in vitro represent potential therapeutic uses of PDCs as cellular therapies to modulate immune responses post–HCT or the use of PDC-induced CD4+CD25+ Treg cells to control GVHD in allogeneic HCT. We have recently developed a novel hemopoietic progenitor cell culture system that allows in vitro generation of large numbers of human PDCs from CD34+ hemopoietic progenitor cells to facilitate future studies of PDC development, their immune function and potential clinical application in immune-based therapies (66). The use of CpG ODN to modulate PDC function and to specifically regulate immune responses in the recipient may provide novel immune-based therapies to control GVHD and viral diseases in post–HCT patients.
We thank Melinda Berthold for technical support, and Janet Peller and Greg Veltri at the Cancer Center Flow Cytometry Core for cell sorting.
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
↵1 This work was supported in part by research grants from Leukemia Research Foundation and Ginkgo Biomedical Research Institute, Japan (to W.C.), and by National Institutes of Health Grant RO1 CA72669 (to B.R.B.).
↵2 Address correspondence and reprint requests to Dr. Wei Chen, University of Minnesota Cancer Center, Mayo Mail Code 806, 420 Delaware Street SE, Minneapolis, MN 55455. E-mail address:
↵3 Abbreviations used in this paper: DC, dendritic cell; PDC, plasmacytoid DC; Treg, T regulatory cell; GVHD, graft-versus-host disease; ODN, oligodeoxynucleotide; BDCA, blood DC Ag; HCT, hemopoietic cell transplant.
- Received March 23, 2004.
- Accepted July 22, 2004.
- Copyright © 2004 by The American Association of Immunologists