The calcineurin inhibitor tacrolimus allows the induction of functional CD4CD25 regulatory T cells by rabbit anti-thymocyte globulins

Abstract

Rabbit anti-thymocyte globulins (rATG) induce CD4(+)CD25(+)forkhead box P3 (FoxP3(+)) regulatory T cells that control alloreactivity. In the present study, we investigated whether rATG convert T cells into functional CD4(+)CD25(+)FoxP3(+)CD127(-/low) regulatory T cells in the presence of drugs that may hamper their induction and function, i.e. calcineurin inhibitors. CD25(neg) T cells were stimulated with rATG or control rabbit immunoglobulin G (rIgG) in the absence and presence of tacrolimus for 24 h. Flow cytometry was performed for CD4, CD25, FoxP3 and CD127 and the function of CD25(+) T cells was examined in suppression assays. MRNA expression profiles were composed to study the underlying mechanisms. After stimulation, the percentage CD4(+)CD25(+)FoxP3(+)CD127(-/low) increased (from 2% to 30%, mean, P < 0.01) and was higher in the rATG samples than in control rIgG samples (2%, P < 0.01). Interestingly, FoxP3(+)T cells were also induced when tacrolimus was present in the rATG cultures. Blockade of the interleukin (IL)-2 pathway did not affect the frequency of rATG-induced FoxP3(+) T cells. The rATG tacrolimus-induced CD25(+) T cells inhibited proliferative responses of alloantigen-stimulated effector T cells as vigorously as rATG-induced and natural CD4(+)CD25(+)FoxP3(+)CD127(-/low) T cells (67% +/- 18% versus 69% +/- 16% versus 45% +/- 20%, mean +/- standard error of the mean, respectively). At the mRNA-expression level, rATG-induced CD25(+) T cells abundantly expressed IL-10, IL-27, interferon (IFN)-gamma, perforin and granzyme B in contrast to natural CD25(+) T cells (all P = 0.03), while FoxP3 was expressed at a lower level (P = 0.03). These mRNA data were confirmed in regulatory T cells from kidney transplant patients. Our findings demonstrate that tacrolimus does not negatively affect the induction, phenotype and function of CD4(+)CD25(+) T cells, suggesting that rATG may induce regulatory T cells in patients who receive tacrolimus maintenance therapy.

Fig. 1

Induction of regulatory T cells. (a) Depletion of CD25⁺ T cells from peripheral blood mononuclear cells (PBMC). (b) Incubation of CD25^neg T cells with rabbit immunoglobulin G (rIgG), tacrolimus, rabbit anti-thymocyte globulins (rATG), rATG + tacrolimus, rATG + interleukin (IL)-2 receptor (R)-α and rATG + anti-IL-2. Gate shows the percentage of CD25⁺ of CD4⁺ T cells. (c) Percentage of CD25⁺ T cells of CD4⁺, n = 3. Differences were tested statistically by analysis of variance (anova), P < 0·0001. (d) Representative example of percentage of forkhead box P3 (FoxP3⁺)CD127^−/lowI within the induced CD25⁺ T cells of all cultures in the presence of rATG. CD25^neg cells from PBMC (top panels) were stimulated for 24 h with rATG, rATG + tacrolimus, rATG + anti-IL-2Rα and rATG + anti-IL-2. After 24 h, induced CD25⁺ T cells were gated and analysed for their FoxP3⁺CD127^−/low expression (middle panels). Gates for positivity were set on FoxP3⁺CD127^−/low cells within natural (nCD25⁺) T cells after 24 h of incubation with rATG. FoxP3⁺CD127^−/low cells were absent within CD25^neg cells incubated with control rIgG (lower panels). (e) Percentage of FoxP3⁺CD127^−/low of induced CD25⁺ T cells, three healthy individuals. Error bars represent mean ± standard error of the mean. Differences were tested statistically by anova. P < 0·0001; *P < 0·05; ***P < 0·001.

Fig. 1

Induction of regulatory T cells. (a) Depletion of CD25⁺ T cells from peripheral blood mononuclear cells (PBMC). (b) Incubation of CD25^neg T cells with rabbit immunoglobulin G (rIgG), tacrolimus, rabbit anti-thymocyte globulins (rATG), rATG + tacrolimus, rATG + interleukin (IL)-2 receptor (R)-α and rATG + anti-IL-2. Gate shows the percentage of CD25⁺ of CD4⁺ T cells. (c) Percentage of CD25⁺ T cells of CD4⁺, n = 3. Differences were tested statistically by analysis of variance (anova), P < 0·0001. (d) Representative example of percentage of forkhead box P3 (FoxP3⁺)CD127^−/lowI within the induced CD25⁺ T cells of all cultures in the presence of rATG. CD25^neg cells from PBMC (top panels) were stimulated for 24 h with rATG, rATG + tacrolimus, rATG + anti-IL-2Rα and rATG + anti-IL-2. After 24 h, induced CD25⁺ T cells were gated and analysed for their FoxP3⁺CD127^−/low expression (middle panels). Gates for positivity were set on FoxP3⁺CD127^−/low cells within natural (nCD25⁺) T cells after 24 h of incubation with rATG. FoxP3⁺CD127^−/low cells were absent within CD25^neg cells incubated with control rIgG (lower panels). (e) Percentage of FoxP3⁺CD127^−/low of induced CD25⁺ T cells, three healthy individuals. Error bars represent mean ± standard error of the mean. Differences were tested statistically by anova. P < 0·0001; *P < 0·05; ***P < 0·001.

Fig. 5

Flow cytometric analyses of granzyme B expression (a) within natural (green curve), rabbit anti-thymocyte globulins (rATG) (I) (dark blue curve) and rATG–tacrolimus (I) (red curve) CD25⁺FoxP3⁺CD127^low T cells. The marker represents the percentage of granzyme B-positive cells. The table shows the median fluorescence of granzyme B. (b) Percentage of granzyme B-positive cells for three healthy individuals and (c) percentage of granzyme B-positive cells within CD4⁺CD25⁺ T cells of three healthy individuals. Mean ± standard error of the mean (s.e.m.) is shown. (d) One example of up-regulation of human leucocyte antigen (HLA) class II on CD4⁺ T cell blasts (gated on CD3⁺CD4⁺ T cells) in the presence of interferon (IFN)-γ. (e) Cytotoxicity assay with natural, rATG and rATG–tacrolimus-induced regulatory T cells (T_reg) as effector cells (E) and Europium-labelled target blasts (T). Natural and induced T_regs are co-cultured with allogeneic target blasts. Means ± s.e.m. are shown for six healthy controls.

Fig. 5

Flow cytometric analyses of granzyme B expression (a) within natural (green curve), rabbit anti-thymocyte globulins (rATG) (I) (dark blue curve) and rATG–tacrolimus (I) (red curve) CD25⁺FoxP3⁺CD127^low T cells. The marker represents the percentage of granzyme B-positive cells. The table shows the median fluorescence of granzyme B. (b) Percentage of granzyme B-positive cells for three healthy individuals and (c) percentage of granzyme B-positive cells within CD4⁺CD25⁺ T cells of three healthy individuals. Mean ± standard error of the mean (s.e.m.) is shown. (d) One example of up-regulation of human leucocyte antigen (HLA) class II on CD4⁺ T cell blasts (gated on CD3⁺CD4⁺ T cells) in the presence of interferon (IFN)-γ. (e) Cytotoxicity assay with natural, rATG and rATG–tacrolimus-induced regulatory T cells (T_reg) as effector cells (E) and Europium-labelled target blasts (T). Natural and induced T_regs are co-cultured with allogeneic target blasts. Means ± s.e.m. are shown for six healthy controls.

Fig. 2

P38 phospho-specific flow cytometry. (a) Representative example of P38 phosphorylation levels in unstimulated, phorbol myristate acetate (PMA)–ionomycin-stimulated, rabbit anti-thymocyte globulins (rATG)-stimulated CD4⁺CD25^neg cells. Median fluorescence intensities are depicted of phosphorylated P38 in CD4⁺ T cells after 24 h of culture. (b) P38 phosphorylation in unstimulated CD4⁺CD25^neg cells, PMA–ionomycin-stimulated CD4⁺CD25^neg cells, rATG-stimulated CD4⁺CD25^neg cells and rATG–tacrolimus-stimulated CD4⁺CD25^neg cells, n = 3. (c) phosphorylation of P38 in unstimulated and PMA–ionomycin-stimulated natural CD4⁺CD25⁺ (nCD4⁺CD25⁺) and CD4⁺CD25^neg cells from PBMC and rATG-induced CD4⁺CD25⁺ and CD25^neg cells.

Fig. 3

Suppression assays with natural (n) and rATG-induced (I) and rATG-tacrolimus-induced CD25⁺ T cells. (a) Proliferative responses after [³H]-thymidine incorporation are shown in counts per minute (cpm) of CD25^neg T cells in response to alloantigen (black bar) and of CD25^neg T cells in the presence of (n) CD25⁺ T cells (grey bar), rATG (I) (dotted bar) and rATG–tacrolimus-induced (striped bar) CD25⁺ T cells. (b) Percentage inhibition of the proliferation of CD25^neg T cells by (n) (grey bar), rATG-(I) (dotted bar) and rATG–tacrolimus-induced (striped bar) CD25⁺ T cells. Mean ± standard error of the mean are shown for five healthy individuals.

Fig. 4

Relative mRNA expression patterns of (n) (grey bar), rabbit anti-thymocyte globulins (rATG) (I) (dotted bar) and rATG–tacrolimus (I) (striped bar) CD25⁺ T cells for forkhead box P3 (FoxP3), interleukin (IL)-10, Epstein–Barr virus-induced gene 3 (EBI3), IL-27p28, interferon (IFN)-γ, perforin and granzyme B for four healthy individuals. Error bars represent mean ± standard error of the mean. Significance was tested by analysis of variance (anova); FoxP3; P < 0·0001, IL-10; P = 0·0007, EBI3; P < 0·0001, IL-27p28; P = 0·01, IFN-γ. P = 0·0008, perforin;P < 0·0001, granzyme B; P = 0·0014, *P < 0·05.

Fig. 6

Relative mRNA expression patterns of CD25, forkhead box P3 (FoxP3), interleukin (IL)-10, IL-2, interferon (IFN)-γ, perforin and granzyme B within CD25⁺ T cells that were isolated from peripheral blood mononuclear cells (PBMC) of rabbit anti-thymocyte globulins (rATG) (striped bars, n = 3) and control patients (grey bars, n = 3). Data are shown as mean ± standard error of the mean. Statistical significance was tested by Mann–Whitney U-test, *P < 0·05.