A Role for TGF-β in the Generation and Expansion of CD4+CD25+ Regulatory T Cells from Human Peripheral Blood

Satoshi Yamagiwa, J. Dixon Gray, Shigeo Hashimoto and David A. Horwitz
J Immunol June 15, 2001, 166 (12) 7282-7289; DOI: https://doi.org/10.4049/jimmunol.166.12.7282

Abstract

An elusive goal in transplanting organs across histocompatibility barriers has been the induction of specific tolerance to avoid graft rejection. A considerable body of evidence exists that the thymus produces regulatory T cells that suppress the response of other T cells to antigenic stimulation. We report that TGF-β can induce certain CD4+ T cells in the naive (CD45RA+RO−) fraction in human peripheral blood to develop powerful, contact-dependent suppressive activity that is not antagonized by anti-TGF-β or anti-IL-10 mAbs. The costimulatory effects of TGF-β on naive CD4+ T cells up-regulated CD25 and CTLA-4 expression, increased their transition to the activated phenotype, but decreased activation-induced apoptosis. Suppressive activity was concentrated in the CD25+ fraction. These CD4+CD25+ regulatory cells prevented CD8+ T cells from proliferating in response to alloantigens and from becoming cytotoxic effector cells. Moreover, these regulatory cells exerted their suppressive activities in remarkably low numbers and maintained these effects even after they are expanded. Once activated, their suppressive properties were Ag nonspecific. Although <1% of naive CD4+ T cells expressed CD25, depletion of this subset before priming with TGF-β markedly decreased the generation of suppressive activity. This finding suggests that CD4+CD25+ regulatory T cells induced ex vivo are the progeny of thymus-derived regulatory T cells bearing a similar phenotype. The adoptive transfer of these regulatory T cells generated and expanded ex vivo has the potential to prevent rejection of allogeneic organ grafts.

Considerable evidence has accumulated for the existence of regulatory T cells that prevent autoimmunity and maintain transplantation tolerance (1, 2). One regulatory subset that has been well characterized expresses cell surface IL-2R α-chains (3, 4). These CD4+CD25+ T cells differentiate in the thymus and are exported to the periphery, where they suppress the activation of potentially self-reactive cells (3, 4, 5, 6, 7, 8). Injection of peripheral T cells from normal mice depleted of CD4+CD25+ T cells into athymic mice results in a high incidence of organ-specific autoimmune disease (3, 4). Moreover, certain strains of neonatally thymectomized mice develop multiorgan-specific autoimmunity (9, 10). These mice lack CD4+CD25+ cells because they are not produced until 1 wk after birth (11, 12).

CD4+CD25+ T cells are potent inhibitors of polyclonal T cell activation (12). After activation via the TCR, they inhibit IL-2 production by the responding T cells (13, 14, 15). Unlike other regulatory T cells, which produce inhibitory cytokines (16, 17), these cells suppress immune responses by a contact-dependent mechanism, at least in vitro (15). Others have described CD4+CD25+ cells that regulate anergy in neonatally tolerized mice (18).

Whether regulatory CD4+CD25+ T cells exist in humans and can be expanded in the periphery is not known. In addition to the well-described suppressive effects of TGF-β on T cell function (19, 20, 21, 22, 23, 24), this cytokine can also have positive effects, which are predominantly on naive T cells (25). TGF-β can synergize with IL-2 to prevent apoptosis and promote effector cell function (26, 27, 28), and it has a crucial role in the generation of CD8+ T cells that suppress Ab production (29, 30). Here we report that TGF-β also induces certain naive CD4+ T cells to become cells that suppress the generation of T cell cytotoxic activity. Using the allogeneic mixed lymphocyte reaction to stimulate naive CD4+ cells that express 1% IL-2R α-chains, we have learned that TGF-β enhances the number of CD25+ cells and that some of these T cells inhibit CD8+ cells from proliferating and developing cytolytic activity on restimulation. Surprisingly, depletion of the rare CD25+ cells in the CD4+CD45RA+RO− fraction markedly reduced the generation of suppressive activity; a result suggesting that TGF-β stimulates and expands the small number of thymus-derived CD4+CD25+-regulatory T cells that circulate in peripheral blood.

Materials and Methods

Reagents

Antibodies used were anti-CD3 (UCHT1; PharMingen, San Diego, CA); anti-CD4 (OKT4; American Type Culture Collection (ATCC), Rockville, MD); anti-CD8 (OKT8; ATCC); anti-CD11b (OKM1; ATCC); anti-CD16 (3G8); anti-CD25 (PharMingen); anti-CD45RA (PharMingen); anti-CD45RO (UCHL-1; ATCC); anti-CD69 (PharMingen); anti-CD74 (anti-HLA-DR) (L243, ATCC); anti-CD95 (Fas/APO-1) (PharMingen); anti-CD152 (CTLA-4; PharMingen). FITC-conjugated annexin V and anti-IL-10 were also purchased form PharMingen. Human rIL-2 was purchased from Chiron (Emeryville, CA). Human rTGF-β1, anti-TGF-β, and IL-10 were purchased from R&D Systems (Minneapolis, MN).

Cell isolation

PBMC were prepared from heparinized venous blood of healthy adult volunteers by Ficoll-Hypaque (Pharmacia, Piscataway, NJ) density gradient centrifugation as previously described (31). The PBL were immediately rosetted with 2-aminoethylisothiouronium bromide-treated SRBC. T cells were prepared from rosetting cells by negative selection following depletion of CD16+ and CD74+ cells using immunomagnetic beads (Dynal, Great Neck, NY). The percentage of CD3+ cells in this fraction was usually >95%. To obtain naive CD45RA+RO−CD4+ T cells, SRBC-rosetting cells were stained with mAbs to CD8, CD16, CD11b, CD74, and CD45RO and separated by negative selection using immunomagnetic beads. The same mAbs were used to obtain naive CD45RA+RO−CD8+ T cells except that CD4 was substituted for CD8. The purity of those T cells was usually>95%. In some experiments, CD25+ cells were also depleted by adding anti-CD25 to the panel of mAbs. In other experiments, after naive CD4+ cells had been cultured with irradiated allogeneic stimulators ± TGF-β1 (1 ng/ml) for 5 days, the CD4+CD25+ and CD4+CD25− subsets were obtained by sorting on a FACStarPlus flow cytometer (Becton Dickinson, San Jose, CA) using FITC-conjugated anti-CD25 and PE-conjugated anti-CD4. The resultant purity was >98%. The nonrosetted cells from each donor were frozen and thawed for use as stimulator cells in a subsequent experiment.

Cell cultures

Purified naive CD4+ T cells (106 cells/ml) were added to 24-well microtiter plates (Falcon, Lincoln Park, NJ) in 2-ml volumes. Cells were suspended in AIM-V serum-free lymphocyte medium (Life Technologies, Gaithersburg, MD) because serum contains significant amounts of latent TGF-β. Cells were stimulated with irradiated (3000 cGy) allogeneic T cell-depleted PBMC (106 cells/ml) in the presence or absence of TGF-β (0.1–1 ng/ml) for 5 days. The phenotype of these alloactivated CD4+ cells compared with unstimulated CD4+ cells was assessed by flow cytometry. After washing to remove residual cytokines, the regulatory effects of these CD4+ subsets on syngeneic effector T cells was assessed in secondary cultures. These cultures consisted of freshly isolated effector T cells stimulated for 5 days with irradiated allogeneic T cell-depleted PBMC in the presence or absence of the indicated numbers of primed CD4+ T cells. In some experiments, various numbers of sorted (see above) CD4+CD25+ or CD4+CD25− were added to the secondary cultures. We assessed effector T cell phenotype, proliferative activity, and generation of CTL activity. In some secondary cultures, the medium contained 10% AB human serum (Omega Scientific, Tarzana, CA) instead of FCS.

Similar to the studies of Thornton and Shevach (13), we used the Transwell chambers (Corning Costar, Cambridge, MA) to assess whether surface contact was necessary for suppressive effects in the secondary cultures. These experiments were conducted in 24-well plates (0.8 ml) with effector T cells and irradiated (3000 cGy) allogeneic T cell-depleted PBMC in the presence or absence of conditioned CD4+ T cells mixed with irradiated stimulator cells in the Transwell. In other experiments, sorted CD4+CD25+ T cells were cultured with IL-2, 10 U/ml, for 5–7 days to determine their capacity for expansion. Various numbers of these cells were added to fresh T cells and irradiated stimulator cells, and their effect on the generation of CTL activity was assessed as described above.

Cytotoxicity assays

The cytotoxic activity was measured by a standard 4 h chromium release assay. In brief, 5 × 103 51Cr-labeled target cells (stimulator Con A blasts) were incubated with graded numbers of effector T cells in 200 μl medium in 96-well round-bottom microtiter plates (Falcon). The plates were centrifuged and incubated for 4 h at 37°C. Then supernatants were harvested and counted in a gamma counter. All test samples were measured in triplicate. The percentage of specific 51Cr release was calculated as: percent lysis = [(cpm released experimental − cpm spontaneous)/(cpm total lysis − cpm spontaneous)] × 100.

Proliferation assays

Purified T cells (105 cells/well) were cultured in 200 μl flat-bottom 96-well plates (Falcon) with irradiated (3000 cGy) allogeneic T cell-depleted PBMC (105 cells/well) and the indicated numbers of conditioned CD4+ T cell for 3 or 5 days at 37°C and 5% CO2. Cultures were pulsed with [3H]TdR for the last 18 h of culture. All tests were conducted in triplicate.

Immunofluorescence analysis

Cell surface Ag expressions on both conditioned CD4+ T cells and effector T cells were determined by immunofluorescence. Cells (105) were labeled with FITC-conjugated (anti-CD4, anti-CD25, anti-CD45RO, anti-CD69), PE-conjugated (anti-CD8, anti-CD25, anti-CD45RA, anti-CD95) or CyChrome-conjugated (anti-CD4, anti-CD8) mAbs. Cells were incubated with the appropriate mAbs for 30 min at 4°C in PBS with 0.1% BSA and 0.02 mM NaN3. After washing, the labeled cell samples were analyzed on a FACStarPlus flow cytometer using Cellquest software (Becton Dickinson).

To analyze cell divisions of effector T cells, effector T cells were labeled with CFSE (Molecular Probes, Eugene, OR). In brief, cells were washed and resuspended at a concentration of 107/ml in PBS. CFSE was added at a final concentration of 0.5 μM and incubated for 10 min at 37°C. The reaction was stopped by washing the cells with RPMI 1640 containing 10% FCS. After the culture of labeled effector T cells (106 cells/ml) with irradiated (3000 cGy) allogeneic T cell-depleted PBMC (106 cells/ml) in the presence or absence of the indicated numbers of conditioned CD4+ T cells, CFSE levels were analyzed by flow cytometry.

To learn whether CD4+ T cells primed with TGF-β could proliferate in response to the stimulator cells, these T cells and those primed without TGF-β were rested for 3 days, labeled with CFSE, and restimulated with irradiated E rosette-negative cells in medium containing 10% type AB human serum. Intensity of CFSE was determined after 3–7 days of culture.

Statistical analysis

The significance of the results was analyzed by Student’s t test performed using GraphPad software (GraphPad, San Diego, CA).

Results

TGF-β can induce alloactivated naive CD4+ T cells to develop suppressive activity

Naive CD4+ cells developed potent suppressive activity when activated in the presence of TGF-β. These CD4+ T cells had the capacity to block the generation of CTL activity against allogeneic target cells. In a two-step coculture experiment, CD4+CD45RA+ RO− T cells were first incubated with irradiated allogeneic stimulator cells ± TGF-β1 for 5 days. Various numbers of these cells were then added to fresh T cells and stimulator cells, and the effect of these cells on the generation of CTL activity was assessed. On assay with chromium-labeled allogeneic target cells, CD4+ cells that had been primed in the presence of TGF-β almost completely blocked responder T cells from killing, whereas control CD4+ cells generally had minimal to modest suppressive activity (Fig. 1⇓A). Activation of CD4+ cells was needed for the development of inhibitory activity. CD4+ suppressor cells that had been primed with alloantigens in the presence of TGF-β will be called CD4reg. Unlike TGF-β, priming of the CD4+ cells with IL-10 did not induce suppressive activity (Fig. 1⇓B). TGF-β could also costimulate purified CD4+CD45RA−CD45RO+ cells to inhibit allo-CTL activity (Fig. 1⇓C). However, the effects of these cells on the development of T cell cytotoxicity were more variable. In some experiments, the effects of CD4 cells primed without TGF-β (CD4con) were also strongly suppressive, and the presence of TGF-β did not further increase this effect. Finally, naive CD8+ T cells primed with stimulator cells also developed moderate suppressive activity, but under these experimental conditions the presence of TGF-β with few exceptions, did not enhance this activity (Fig. 1⇓D),

FIGURE 1.
FIGURE 1.

Generation of potent suppressive activity from naive CD4+. Generation of in a two-step coculture procedure, purified T cell subsets from donor A were stimulated for 5 days with irradiated (3000 cGy) allogeneic stimulator cells from donor B ± TGF-β (1 ng/ml). The secondary cultures consisted of mixing the primed T cell subsets with fresh syngeneic T cells in the ratio indicated, and culturing these cells with irradiated stimulator cells from donor B for 5 days. CTL activity against Con A blasts from donor B was then determined in a standard 4-h chromium release assay at the E:T ratios indicated. A, Alloactivated CD4+ CD45RA+RO− cells primed with TGF-β markedly suppressed the generation of CTL activity in comparison with control (Con) CD4+ cells. Additional controls not shown were CD4+ cells cultured for 5 days with TGF-β, but without allogeneic MHC stimulation. These cells had no suppressive effects. B, Under similar experimental conditions, rIL-10 (10 pg/ml) was unable to induce suppressive activity. C, Suppressive effects of CD4+CD45RA−RO+ cells primed with allogeneic stimulator cells. D, Suppressive effects of CD8+CD45RA+RO− cells. TGF-β did not enhance the inhibitory effects of primed CD8+ cells. Resp., responder cells.

Similar to the murine CD4+-regulatory T cells described by Shevach et al. (6, 11, 12, 13), suppression by TGF-β-induced human CD4+ cells alloactivated in the presence of TGF-β appeared to require surface contact (Fig. 2⇓, A and B). Separation of these cells from responder T cells by a semipermeable membrane completely prevented their inhibitory effects. This was the case even if they were activated with stimulator cells. To learn whether surface contact induced the production of inhibitory cytokines, experiments were performed with mAbs that neutralized TGF-β and IL-10. These Abs, however, did not diminish the suppressive effects of CD4reg (Fig. 2⇓C). Irradiation of CD4reg with 3000 cGy after they had been primed with TGF-β, however, completely abolished suppressive activity (result not shown).

FIGURE 2.
FIGURE 2.

Suppressive effects of CD4reg require cell to cell contact. A, CD4reg and control CD4+ cells were generated by activating naive CD4+ cells with allogeneic stimulator cells (allo) ± TGF-β as described above. A, In secondary cultures, CD4reg or control CD4+ cells were directly mixed with the responder cells (Resp.); B, CD4+cells were separated from the responder T cells by a semipermeable membrane using Transwell chambers. The Transwells containing the CD4+ cells also included stimulator cells. Results are one of three independent experiments. C, Effects of anti (α)-TGF-β, anti-IL-10 and isotype-matched control IgG (all Abs were added at 10 mg/ml). One of two experiments is shown.

Costimulatory effects of TGF-β on alloactivated CD4+ T cells

Activation of naive CD4+ T cells in the presence of TGF-β enabled them to respond more vigorously to alloantigens, accelerated their conversion to the activated phenotype, and increased their viability. Fig. 3⇓ indicates properties of control and TGF-β-conditioned CD4+ cells after 5 days of culture. Allostimulated CD4+ cells displayed the typical features of activation and some were already undergoing activation-induced cell death as indicated by annexin V staining. Importantly, those alloactivated in the presence of TGF-β were even larger, some stained more intensely with CD4, and expression of CD25 was markedly increased. Table I⇓ shows the significant increase in mean values for expression of CD25 and for both intracellular and surface expression of CTLA-4 in 13 separate experiments. A greater percentage of CD4reg had also down-regulated surface expression of CD45RA (Fig. 3⇓). Significantly, annexin V staining was markedly decreased, and the total number of CD4reg recovered was 50–60% greater than control CD4+ cells. Thus, although CD4reg were more intensively activated that control CD4 cells, they were also more resistant to activation-induced apoptosis.

FIGURE 3.
FIGURE 3.

Activation of naive CD4+ T cells in the presence of TGF-β enabled them to respond more vigorously to alloantigens, accelerated their conversion to the activated phenotype, and increased their viability. Purified naive CD4+ T cells were cultured for 5 days without stimulator cells (gray shaded area), with irradiated allogeneic stimulator cells (thin dotted line), or with stimulator cells and TGF-β (1 ng/ml; thick line) and expression of cell surface or intracellular Ags were determined by flow cytometry. Cells (105) were labeled with FITC-conjugated (anti-CD4, anti-CD25), PE-conjugated (anti-CD45RA, anti-CTLA-4), or CyChrome-conjugated (anti-CD4) mAbs. CTLA-4 expression was analyzed by intracellular staining. To detect apoptotic cells, annexin V staining was performed according to the manufacturer’s instructions. This experiment is representative of at least five studies with each marker. FSC, Forward scatter.

View this table:
Table I.

Costimulatory effects of TGF-β on alloactivated naive T cellsa

CD4+ regulatory T cells are CD25+

Similar to the murine CD4+-regulatory T cells described by others (11, 12, 13, 14, 15), separation of CD4reg into CD25+ and CD25− fractions by cell sorting revealed that almost all of the suppressive activity was contained in the CD25+ fraction (Fig. 4⇓A). The CD4+CD25− fraction displayed only minimal suppressive activity. Moreover, the CD4+CD25+ cells were very potent as indicated by their ability to markedly inhibit the generation of CTL activity in very low numbers. Fig. 4⇓B shows that suppressive activity was only slightly diminished when the numbers of the CD4+ suppressor cells were reduced from 25% to 3% of total T cells. With this small number of added CD4+ cells, the minimal suppressive activity mediated by the CD25− subset had disappeared.

FIGURE 4.
FIGURE 4.

CD4reg display CD25 and require remarkably small numbers for potent suppressive activity. Naive CD4+ T cells primed with irradiated allogeneic stimulator cells ± TGF-β (1 ng/ml) and CD4reg were separated into CD25+ and CD25− fractions (Fr.) by cell sorting. A, Effect of primed CD4+ cells mixed with fresh T cells at a 1:4 ratio. The experimental protocol described in Fig. 1⇑ was used to assess regulation of the generation of CTL activity. B, Effect of various dilutions of these primed CD4+ T cells added to fresh responder cells (Resp.) on the generation of CTL activity at an E:T ratio of 100:1. Results shown are representative of three similar experiments.

CD4+CD25+ regulatory T cells can proliferate in response to alloantigens

Unlike CD4+-regulatory T cells generated by repeated stimulation with IL-10, which have low proliferative capacity (32), CD4reg induced in the presence of TGF-β proliferated in response to alloantigens and IL-2 and retained their suppressive capacity. Fig. 5⇓ shows the response of CD4reg and CD4con when cultured with stimulator cells. After priming, the cells were washed to remove residual TGF-β and cultured for 3 days; CFSE labeling revealed that they had returned to the resting state. After the addition of allogeneic stimulator cells to CD4+ cells freshly labeled with CFSE, a stronger proliferative response by CD4reg was apparent at day 3 and clearly evident by day 5.

FIGURE 5.
FIGURE 5.

CD4reg can proliferate in response to alloantigens. Naive CD4+ T cells and equal numbers of irradiated stimulator cells (1 × 106/ml) were cultured ± TGF-β for 5 days in serum-free medium. After a washing to remove residual TGF-β, the cells were rested for 3 days in culture medium containing 10% normal human serum. The cells were then labeled with CFSE and restimulated with irradiated allogeneic stimulator cells. The intensity of CFSE staining after 3 and 5 days of culture is shown. Heavy line, CD4reg; thin line, CD4con. Similar results have been obtained in four other experiments.

In other experiments, the CD25+ fraction of CD4reg was isolated by cell sorting and cultured for 5 days with IL-2. These cells expanded 7- to 14-fold during this time and retained their suppressive function in three separate experiments. Fig. 6⇓ shows strong inhibition of the generation of CTL activity when these CD4reg comprised only 5% of the total T cells. Moreover, this suppressive activity persisted when the number of these CD4+CD25+ cells was reduced to only 0.2% of total cells. Irradiation of these regulatory T cells, however, abolished their suppressive effects.

FIGURE 6.
FIGURE 6.

Expanded CD4reg retain their potent suppressive effects. The CD4+CD25+ fraction of naive CD4+ cells primed in the presence of TGF-β was obtained by cell sorting and cultured in the presence of IL-2 (10 U/ml) for 5 days. The percentages indicated were added to fresh T cells and examined for inhibition of the generation of CTL activity. Irradiated CD4+CD25+ cells served as controls. This is one of two such experiments.

Mechanism of action of CD4+CD25+ regulatory T cells

CD4reg blocked the proliferative response of responder T cells to alloantigens (Fig. 7⇓A). Gating on CD8+ cells by flow cytometry after a 5-day culture, these alloactivated T cells displayed the expected marked increase in CD25, increased expression of CD69 and Fas (CD95), and evidence of cell division by CFSE-labeled and propidium iodide-labeled effector T cells. Moreover, some CD8+ cells were undergoing activation-induced cell death as indicated by annexin V staining. In sharp contrast, in cultures containing CD4reg, up-regulation of CD25, CD69, and Fas by CD8+ cells was markedly inhibited, and almost none underwent apoptosis or cell division. Studies of the absolute numbers of CD8+ cells in culture revealed only a minimal change from the total count. Thus, CD4reg prevented CD8+ T cells from responding to alloantigens.

FIGURE 7.
FIGURE 7.

CD4reg block the proliferative response of responder T cells to alloantigens and CD8+ T cells are their principal target. A, CD4reg or primed control (Con) CD4+ cells were mixed with fresh T cells in a 1:4 ratio (105/well) and cultured in triplicate with irradiated stimulator cells (105/well) for 5 days. The cultures were pulsed with [3H]TdR for the last 18 h. B, Phenotype of the cultured CD8+ cells as determined by flow cytometry. Gating on CD8+ cells, the gray area indicates cells cultured for 5 days without stimulator cells. Thick line, CD8 cells cultured with stimulator cells and CD4reg; thin dotted line, alloactivated CD8 cells cultured with control CD4 cells. The lymphocytes (105) were labeled with FITC-conjugated (anti-CD69), PE-conjugated (anti-CD25, anti-CD95) or CyChrome-conjugated (anti-CD8) mAbs. Annexin V straining, CFSE levels, and DNA content by propidium iodide (PI) staining are also shown. These experiments were performed at least three times.

Other studies were conducted to learn the Ag specificity of CD4reg. After priming naive CD4+ T cells from donor A with stimulator cells from donor B, these activated T cells were mixed with autologous T cells and stimulator cells from donor B or from a third party, donor C. The effects of these CD4+ cells on the generation of CTL activity against chromium-labeled cells from the same or third party stimulator cells were assessed. In the experiment shown in Fig. 8⇓, CD4+ cells previously sensitized to the alloantigens of a given donor in the absence of TGF-β did not affect the generation of CTL activity by fresh T cells against donor B, as shown above. However, the addition of these control CD4+ cells to T cells responding to third party, donor C stimulator cells increased the CTL activity generated. By contrast, CD4reg specific for donor B also suppressed the development of CTL activity against donor C.

FIGURE 8.
FIGURE 8.

The suppressive effects of CD4reg are Ag nonspecific. The experimental protocol is similar to the studies described above. CD4reg cells from donor A displaying the activated phenotype shown in Fig. 3⇑ after 5 days of culture, were added to autologous T cells and allogeneic stimulator cells from donor B (A) or donor C (B) and cultured for an additional 5 days. CTL activity against Con A blasts from each of the allogeneic donors is shown. This experiment has been performed three times. Con, Control.

Unlike the suppressive effects of the CD4+CD25+ T cells described by others which could be overcome by exogenous IL-2 or anti-CD28 (13, 14), the addition of IL-2 to CD4reg only partially blocked their suppressive effects. As expected, IL-2 augmented the generation of CTL activity by responder T cells. The addition of IL-2 to control CD4+ cells restored CTL activity to baseline levels. However, the addition of IL-2 to cultures containing CD4reg did not completely restore CTL activity (Fig. 9⇓). Thus, at least some of the responder T cells remained nonresponsive even in the presence of IL-2. This experiment has been repeated three times with similar results.

FIGURE 9.
FIGURE 9.

The suppressive activity of CD4reg on CTL activity was only partially abrogated by the addition of IL-2. T cells were alloactivated for 5 days with irradiated stimulator cells in the presence or absence of IL-2 (10 or 50 U/ml). The effect of allo-primed CD4con or CD4reg on the generation of CTL activity is shown. Data are representative of four experiments. Resp., responder cells.

Because the naive fraction of CD4+ cells contains only rare CD25+ cells (Fig. 10⇓A), we reasoned that TGF-β had costimulated CD25− T cells to develop suppressive activity. However, we could not exclude the possibility that these CD4reg were derived from the CD25+ subset. To test this possibility, naive CD4+ cells were depleted of CD25+ cells and then primed in the presence of TGF-β. We found that, as before, TGF-β enhanced the expression of CD25 and the percentage of CD25+ cells. Moreover, the percentage of CD25+ cells after alloactivation was only modestly less in the subset that had initially been depleted of cells bearing this marker. However, after depletion of CD25+ naive CD4+ cells, the suppressive activity induced by TGF-β was markedly reduced (Fig. 10⇓B). This finding suggests that the regulatory T cells induced by TGF-β are the progeny of thymus-derived CD4+CD25+ T cells.

FIGURE 10.
FIGURE 10.

Depletion of CD25+ cells from the starting population impairs the generation of regulatory cells. Naive CD4+ cells were prepared with or without anti-CD25 Abs included in the staining cocktail to deplete the CD25+ subset which comprised <1%. These cells were cultured with irradiated allogeneic stimulator cells ± TGF-β (0.1 ng/ml). After 5 days, the cells were assayed for CD25 expression (A) or for regulatory activity (B) as described above at a CD4reg-T cell ratio of 1:4 and examined for inhibition of the generation of CTL activity.

Discussion

In this study, we have found that the costimulatory effects of TGF-β on naive CD4+ CD45RA+ T cells circulating in the periphery can result in the development of remarkably potent suppressive activity. TGF-β enhanced expression of CD25 and CTLA-4 and accelerated their differentiation to the CD45RA−-activated phenotype. The phenotype and suppressive effects of these CD4+ cells were similar, if not identical, with the thymus-derived CD4+CD25+-regulatory T cells described by others (3, 4, 5, 6, 7, 8, 33). Moreover, we have shown that CD4reg block CD8+ T cells from responding to alloantigens and that this effect is only partially reversed by exogenous IL-2. As expected, suppressive activity was also found in the CD45RA−RO+ fraction of CD4+cells, and TGF-β could also enhance this activity, but this effect was more variable.

Previously we had reported that TGF-β has a crucial role in the generation of CD8+ T cells that suppress Ab production (29, 30). In those studies, CD8+ cells were activated with Con A in the presence of TGF-β for 24 h, whereas in the present study alloantigen-activated CD4+ cells required a longer exposure with this cytokine for optimal up-regulation of CD25 and induction of potent suppressive activity. The more vigorous response of alloactivated naive CD4+ cells in the presence of TGF-β and the increased viability of these activated cells are consistent with the previously described positive effects of this cytokine on naive T cells (25, 26). We suggest that the costimulatory effects of TGF-β on naive T cells induce them to develop down-regulatory activity instead of becoming conventional T effector cells.

A surprising finding of this study was the principal source of CD4reg. Depletion studies revealed that these cells were derived from the rare CD25+ cells present in the CD45RA+RO− fraction of peripheral blood. As previously reported by others, TGF-β can up-regulate CD25 expression by activated naive T cells (25). However, those that were CD25− before activation generally had minimal suppressive activity at the time they were examined. It is not unlikely, however, that naive CD4+CD25− cells can also develop down-regulatory activity. We have recently observed that naive CD4+ T cells primed with superantigen in the presence of TGF-β produce inhibitory amounts of active TGF-β when restimulated one or more times (34).

Consistent with studies of murine CD4+CD25+ regulatory T cells, human CD4reg induced by TGF-β inhibit T cell responses by blocking responsiveness rather than by killing them. Although we considered the possibility that the failure of CD8+ cells to develop CTL activity could be due to cytolytic effects of CD4reg against the allogeneic stimulator cells, this was unlikely for the following reasons: 1) the numbers of CD4reg required for significant inhibition were too small for a significant effect on the Ag-presenting cells; 2) CD4reg had weak CTL activity, even after expansion in IL-2; and 3) other investigators have reported that TGF-β inhibits the development of CTL activity (21, 22).

Murine CD4+CD25+ cells inhibit IL-2 production by Ag-stimulated T cells and thereby block T cell proliferation (14, 15), a result in agreement with the present study. Another group has reported that CD4+CD25+ cells render CD8+ cells anergic. These workers injected semiallogeneic splenocytes into neonatal mice and observed long term Ag-specific CD8 cell nonresponsiveness. Interestingly, although IL-2 could not restore CD8 CTL activity, this result was achieved after removal of regulatory CD4+CD25+ T cells from the cell suspensions (18). Thus, as long as regulatory T cells were present, even IL-2 could not reverse anergy. In in vitro studies, the anergy induced by mouse CD4+CD25+ cells can be overcome by IL-2 or anti-CD28 (14, 15). In the present experiments, the addition of exogenous IL-2 to cultures containing human CD4reg restored CD8 CTL activity partially but not completely. This result is consistent with the role of CD4+CD25+ cells in sustaining transplantation tolerance (18).

The suppressive effects of human CD4reg, like mouse CD4+CD25+ T cells, are Ag nonspecific. Once activated, these regulatory T cells have broad suppressive effects (14, 15). In the present experiments, besides inhibiting T cells responding to the same stimulator cells used to induce them, human CD4reg equally inhibited T cells responding to third party stimulator cells.

CTLA-4 is a potent inhibitor of T cell responses (35, 36). Two groups have recently reported that CTLA-4 plays a key role in the T cell-mediated dominant immunological self tolerance. Blockade of CTLA-4 abolished the suppressive activities of CD4+CD25+ T cells (37, 38). In this study, the costimulatory effects of TGF-β markedly up-regulated CTLA-4. In preliminary studies, Fab anti-CTLA-4 Abs obtained from Dr. B. M. Carreno (Genetics Institute, Cambridge, MA) did not inhibit the suppressive effects of CD4reg.

The mechanism of action of CD4reg remains to be determined but is clearly different from Tr1 CD4+ cells repeatedly stimulated with IL-10 or activated with immature dendritic cells (32, 39). Tr1 cells are anergic, and their immunosuppressive effects are mediated by IL-10 and TGF-β. Anergic T cells suppress other T cell responses by targeting APC (40, 41). Under the experimental conditions described here, IL-10 could not be substituted for TGF-β to induce CD4reg (Fig. 1⇑B). Moreover, CD4reg had the characteristics of activated rather than anergic cells in primary cultures (Fig. 3⇑), and on restimulation with alloantigen these T cells proliferated even more vigorously than control CD4+ cells (Fig. 5⇑). In agreement with the studies of Thornton and Shevach (15), the suppressive activities were abolished by separation of the regulatory cells and effector cells by a semipermeable membrane, whereas neutralizing anti-TGF-β and anti-IL-10 Abs did not have this effect (Fig. 2⇑C).

To explain how the addition of <1 regulatory T cell to 100 T cells can have potent inhibitory effects, it is likely that these cells expand rapidly. Consistent with this possibility, the suppressive effects of CD4reg were radiosensitive even after they were generated (Fig. 6⇑). The effects of irradiation on suppressor T cells are well known (42, 43, 44, 45). It is also possible that CD4reg are part of a network that facilitates other cells to produce inhibitory cytokines (46). This would explain why CD4reg have contact-dependent inhibitory effects in vitro but appear to depend on the production of TGF-β and IL-10 in vivo (2, 47, 48).

The ability to generate powerful regulatory T cells ex vivo has immediate clinical relevance. Allogeneic stem cell transplantation has great potential for the treatment of certain individuals with neoplastic or heritable diseases, but is limited by graft vs host disease. Similarly, although there has been great progress in managing acute rejection after transplantation of solid organs, survival of kidney grafts after 1 year has not increased significantly (49). The use of regulatory T cells generated ex vivo for the prevention of graft rejection as well as a treatment for autoimmunity has great potential. Unlike Tr1-like cells, both murine CD4+CD25+ T cells (15) and the comparable human T cell subset described in this report can be propagated ex vivo and retain their powerful suppressive effects after expansion. The positive effects of TGF-β on the viability of these regulatory cells are also encouraging. These findings add to the potential of adoptive regulatory T cell immunotherapy as a novel treatment strategy that will lack the toxic side effects of the agents currently in use.

Acknowledgments

We thank Harold Soucier for his skilled support in the flow cytometry analysis and Dr. Gunther Dennert for his constructive comments on the manuscript.

Footnotes

  • 1 This work was supported by grants from the National Institutes of Health (AI-41768), the Nora Eccles Treadwell Foundation, and the Southern California Chapter of the Arthritis Foundation.

  • 2 Current address: Niigata University, School of Medicine, Third Department of Internal Medicine, 1-757 Asahimachi-dori, Niigata, Niigata 951-8510, Japan.

  • 3 Address correspondence and reprint requests to Dr. David A. Horwitz, Division of Rheumatology and Immunology, University of Southern California, Keck School of Medicine, 2011 Zonal Avenue, HMR 711, Los Angeles, CA 90033. E-mail address: dhorwitz{at}hsc.usc.edu

  • Received November 15, 2000.
  • Accepted April 16, 2001.

References

  1. Shevach, E. A.. 2000. Regulatory T cells in autoimmunity. Annu. Rev. Immunol. 18: 423
  2. Seddon, B., D. Mason. 2000. The third function of the thymus. Immunol. Today 21: 95
  3. Sakaguchi, S., N. Sakaguchi, M. Asano, M. Itoh, M. Toda. 1995. Immunologic self-tolerance maintained by activated T cells expressing IL-2 receptor α-chains (CD25): breakdown of a single mechanism of self-tolerance causes various autoimmune diseases. J. Immunol. 155: 1151
  4. Asano, M., M. Toda., N. Sakaguchi, S. Sakaguchi. 1996. Autoimmune disease as a consequence of developmental abnormality of a T cell subpopulation. J. Exp. Med. 184: 387
  5. Papiernik, M., L. do Carmo, C. Pontoux, A. Joret, B. Rocha, C. Penit, M. Dy. 1997. T Cell deletion induced by chronic infection with mouse mammary tumor virus spares CD25-positive, IL-10-producing T cell population with infectious capacity. J. Immunol. 158: 4642
  6. Suri-Payer, E., A. Z. Amar, A. M. Thornton, E. M. Shevach. 1998. CD4+CD25+ T cells inhibit both the induction and effector function of autoreactive T cells and represent a unique lineage of immunoregulatory cells. J. Immunol. 160: 1212
  7. Jackson, A. L., H. Matsumoto, M. Janszen, V. Maino, A. Blidy, S. Sheye. 1990. Restricted expression of p55 interleukin 2 receptor (CD25) on human T cells. Clin. Immunol. Immunopathol. 54: 126
  8. Kanegane, H., T. Miyawaki, K. Kato, T. Yokoi, T. Uehara, A. Yachie, N. Taniguchi. 1991. A novel subpopulation of CD45RA+CD4+ T cells expressing IL-2 receptor α-chain (CD25) and having a functionally transitional nature into memory cells. Int. Immunnol. 3: 1349
  9. Kojima, A., R. T. Prehn. 1981. Genetic susceptibility to post-thymectomy autoimmune diseases in mice. Immunogenetics 14: 15
  10. Sakaguchi, S., K. Fukuma, K. Kuribayashi, T. Masuda. 1985. Organ-specific autoimmune diseases induced in mice by elimination of T cell subset. I. Evidence for the active participation of T cells in natural self-tolerance: deficit of a T cell subset as a possible cause of autoimmune disease. J. Exp. Med. 161: 72
  11. Shevach, E. M., A. Thornton, E. Suri-Payer. 1998. T lymphocyte-mediated control of autoimmunity. Novartis Found. Symp. 215: 200
  12. Suri-Payer, E., A. Z. Amar., R. McHugh, K. Natarajan, D. H. Margulies, E. M. Shevach. 1999. Post-thymectomy autoimmune gastritis: fine specificity and pathogenicity of anti-H/K ATPase-reactive T cells. Eur. J. Immunol. 29: 669
  13. Thornton, A. M., E. M. Shevach. 1998. CD4+ immunoregulatory T cells suppress polyclonal T cell activation in vitro by inhibiting interleukin 2 production. J. Exp. Med. 188: 287
  14. Takahashi, T., Y. Kuniyasu, M. Toda, N. Sakaguchi, M. Itoh, M. Iwata, J. Shimizu, S. Sakaguchi. 1998. Immunologic self tolerance maintained by CD25+CD4+ naturally anergic and suppressive T cells: induction of autoimmune disease by breaking their anergic/suppressive state. Int. Immunol. 10: 1969
  15. Thornton, A. M., E. M. Shevach. 1999. Suppressor effector function of CD4+CD25+ immunoregulatory T cells is antigen non-specific. J. Immunol. 164: 183
  16. Weiner, H. L., A. Friedman, A. Miller, S. J. Khoury, A. Al-Sabbagh, L. Santos, M. Sayegh, R. B. Nussenblatt, D. E. Trentham, D. A. Hafler. 1994. Oral tolerance: immunologic mechanisms and treatment of animal and human organ-specific autoimmune diseases by oral administration of autoantigens. Annu. Rev. Immunol. 12: 809
  17. Powrie, F., J. Carlino, M. W. Leach, S. Mauze, R. L. Coffman. 1996. A critical role for transforming growth factor-β but not interleukin 4 in the suppression of T helper type 1-mediated colitis by CD45RBlow CD4+ T cells. J. Exp. Med. 183: 2669
  18. Gao, Q., T. M. Rouse, K. Kazmerzak, E. H. Field. 1999. CD4+CD25+ cells regulate CD8 cell anergy in neonatal tolerant mice. Transplantation 68: 1891
  19. Letterio, J. J., A. B. Roberts. 1998. Regulation of immune responses by TGF-β. Annu. Rev. Immunol. 16: 137
  20. Fox, F. E., H. C. Ford, R. Douglas, S. Cherian, P. C. Nowell. 1993. Evidence that TGFβ can inhibit human T lymphocyte proliferation through paracrine and autocrine mechanisms. Cell. Immunol. 150: 45
  21. Ranges, G. E., I. S. Figari, T. Espevik, M. A. Palladino. 1987. Inhibition of cytotoxic T cell development by transforming growth factor β and reversal by tumor necrosis factor α. J. Exp. Med. 166: 991
  22. Smyth, M. J., S. L. Strobl, H. A. Young, J. R. Ortaldo, A. C. Ochoa. 1991. Regulation of lymphokine-activated killer activity and pore-forming protein gene expression in human peripheral blood CD8+ T lymphocytes: inhibition by transforming growth factor-β. J. Immunol. 146: 3289
  23. Espevik, T., A. Waage, A. Faxvang, M. R. Shalaby. 1990. Regulation of interleukin-2 and interleukin-6 production from T-cells: involvement of interleukin-1β and transforming growth factor-β. Cell. Immunol. 126: 47
  24. Seder, R. A., T. Marth, M. C. Sieve, W. Strober, J. J. Letterio, A. B. Roberts, B. Kelsall. 1998. Factors involved in the differentiation of TGF-β-producing cells from naive CD4+ T cells: IL-4 and IFN-γ have opposing effects, while TGF-β positively regulates its own production. J. Immunol. 160: 5719
  25. De Jong, R., R. A. van Lier, F. W. Ruscetti, C. Schmitt, P. Debre, and M. D. Mossalayi. 1994. Differential effect of transforming growth factor-βl on the activation of human naive and memory CD4+ T lymphocytes. Int. Immunol. 6:63.l.
  26. Zhang, X., L. Giangreco, H. E. Broome, C. M. Dargan, S. L. Swain. 1995. Control of CD4+ effector fate: transforming growth factor β1 and interleukin 2 synergize to prevent apoptosis and promote effector function. J. Exp. Med. 182: 699
  27. Cerwenka, A., D. Bevec, O. Majdic, W. Holter. 1996. Fas- and activation-induced apoptosis are reduced in human T cells preactivated in the presence of TGF-β 1. J. Immunol. 156: 459
  28. Cerwenka, A., D. Bevec, O. Majdic, W. Knapp, W. Holter. 1994. TGF-β is a potent inducer of human effector T cells. J. Immunol. 153: 4367
  29. Gray, J. D., M. Hirokawa, D. A. Horwitz. 1994. The role of transforming growth factor β in the generation of suppression: an interaction between CD8+ T cells and NK cells. J. Exp. Med. 180: 1937
  30. Gray, J. D., M. Hirokawa, K. Ohtsuka, M. M. Stimmler, D. A. Horwitz. 1998. Generation of an inhibitory circuit involving CD8+ T cells, interleukin 2, and NK cell-derived TGF-β: contrasting effects of anti-CD2 and anti-CD3. J. Immunol. 160: 2248
  31. Horwitz, D. A., J. D. Gray, S. C. Behrendsen, M. Kubin, M. Rengaraju, K. Ohtsuka, G. Trinchieri. 1998. Decreased production of interleukin-12 and other Th-1-type cytokines in patients with recent-onset systemic lupus erythematosus. Arthritis Rheum. 41: 838
  32. Groux, H., A. O’Garra, M. Bigler, M. Rouleau, S. Antojejko, J. E. De Vries, M. G. Roncarolo. 1997. A CD4+ T-cell subset inhibits antigen-specific T-cell responses and prevents colitis. Nature 389: 737
  33. Itoh, M., T. Takahashi, N. Sakaguchi, Y. Kuniyasu, J. Shimizu, F. Otsuka, S. Sakaguchi. 1999. Thymus and autoimmunity: production of CD25+CD4+ naturally anergic and suppressive T cells as a key function of the thymus in maintaining immunologic self-tolerance. J. Immunol. 162: 5317
  34. Lee, W., S. Yamagiwa, J. D. Gray, and D. A. Horwitz. Repeated stimulation of CD4+ T cells conditioned with TGF-β results in the production of the active form of this cytokine. FASEB 14:A989.
  35. Walunas, T. L., D. J. Lenschaow, C. Y. Bakker, P. S. Linsley, G. J. Freeman, J. M. Green, C. B. Thompson, J. A. Bluestone. 1994. CTLA-4 can function as a negative regulator of T cell activation. Immunity 1: 405
  36. Bluestone, J. A.. 1997. Is CTLA-4 a master switch for peripheral T cell tolerance?. J. Immunol. 158: 1989
  37. Read, S., V. Malmstrom, F. Powrie. 2000. Cytotoxic T lymphocyte associated antigen 4 plays an essential role in the function of CD25+CD4+ regulatory cells that control intestinal inflammation. J. Exp. Med. 192: 29
  38. Takahashi, T., T. Tagami, S. Yamazaki, T. Uede, J. Shimizu, N. Sakaguchi, T. Mak, S. Sakaguchi. 2000. Immunoregulatory self tolerance maintained by CD25+CD4+ regulatory T cells constitutively expressing cytotoxic T lymphocyte associated antigen 4. J. Exp. Med. 192: 303
  39. Jonuleit, H., E. Schmitt, G. Schuler, K. Jurgen, and A. H. Enk. Induction of interleukin 10-producing, nonproliferating CD4+ T cells with regulatory properties by repetitive stimulation with allogeneic immature human dendritic cells. J. Exp. Med. 192:1213.
  40. Taams, L. S., A. J. M. L. van Rensen, M. C. M. Poelen, C. A. C. M. van Els, A. C. Besseling, J. P. A. Wagenaar, W. van Eden, M. H. M. Wauben. 1998. Anergic T cells actively suppress T cell responses via the antigen-presenting cell. Eur. J. Immunol. 28: 2902
  41. Vendetti, S., J. G. Chai, J. Dyson, E. Simpson, G. Lombardi, R. Lechler. 2000. Anergic T cells inhibit the antigen-presenting function of dendritic cells. J. Immunol. 165: 1175
  42. Macphail, S., O. Stutman. 1982. Suppressor T cells activated in a primary in vitro response to non-major histocompatibility. J. Exp. Med. 156: 1398
  43. Thomas, Y., L. Rogozinski, O. H. Irigoyen, H. H. Shen, M. A. Talle, G. Goldstein, L. Chess. 1982. Functional analysis of human T cell subsets defined by monoclonal antibodies. V. Suppressor cells within the activated OKT4+ population belong to a distinct subset. J. Immunol. 128: 1386
  44. Fox, E. J., R. G. Cook, D. E. Lewis, R. R. Rich. 1986. Proliferation signals for suppressor T cells. Helper cells stimulated with pokeweed mitogen in vitro produce a suppressor cell growth factor. J. Clin. Invest. 78: 214
  45. Kanof, M. E., W. Strober, S. P. James. 1987. Induction of CD4 suppressor T cells with anti-Leu-8 antibody. J. Immunol. 139: 49
  46. Bloom, B. R., P. Salgame, B. Diamond. 1992. Revisiting and revising suppressor T cells. Immunol. Today 13: 131
  47. Powrie, F., J. Carlino, M. W. Leach, S. Mauze, R. L. Coffman. 1996. A critical role for transforming growth factor-beta but not interleukin 4 in the suppression of T helper type 1-mediated colitis by CD45RB(low) CD4+ T cells. J. Exp. Med. 183: 2669
  48. Asseman, C., S. Mauze, M. W. Leach, R. L. Coffman, F. Powrie. 1999. An essential role for interleukin 10 in the function of regulatory T cells that inhibit intestinal inflammation. J. Exp. Med. 190: 995
  49. Auchincloss, H., M. Sykes, D. H. Sachs. 1998. Transplantation immunology. W. E. Paul, ed. Fundamental Immunology 4th Ed.1175 Lippincott-Raven, Philadelphia.