Patients treated with high-dose intravenous immunoglobulin show selective activation of regulatory T cells

Abstract

Intravenous immunoglobulin (IVIg) is used to treat autoimmune and systemic inflammatory diseases caused by derailment of humoral and cellular immunity. In this study we investigated whether IVIg treatment can modulate regulatory T cells (T_regs) in humans in vivo. Blood was collected from IVIg-treated patients with immunodeficiency or autoimmune disease who were treated with low-dose (n = 12) or high-dose (n = 15) IVIg before, immediately after and at 7 days after treatment. Percentages and activation status of circulating CD4⁺CD25⁺forkhead box protein 3 (FoxP3⁺) T_regs and of conventional CD4⁺FoxP3⁻ T-helper cells (T_conv) were measured. The suppressive capacity of T_regs purified from blood collected at the time-points indicated was determined in an ex-vivo assay. High-dose, but not low-dose, IVIg treatment enhanced the activation status of circulating T_regs, as shown by increased FoxP3 and human leucocyte antigen D-related (HLA-DR) expression, while numbers of circulating T_regs remained unchanged. The enhanced activation was sustained for at least 7 days after infusion, and the suppressive capacity of purified T_regs was increased from 41 to 70% at day 7 after IVIg treatment. The activation status of T_conv was not affected by IVIg. We conclude that high-dose IVIg treatment activates T_regs selectively and enhances their suppressive function in humans in vivo. This effect may be one of the mechanisms by which IVIg restores imbalanced immune homeostasis in patients with autoimmune and systemic inflammatory disorders.

Introduction

Intravenous immunoglobulin (IVIg) was introduced initially as a replacement therapy for patients with immune deficiencies, but high-dose IVIg is now used widely for the treatment of autoimmune and systemic inflammatory diseases caused by autoantibodies and/or derailment of the cellular immune system 1,2. Moreover, IVIg has shown efficacy in prevention and treatment of organ allograft rejection and graft-versus-host disease (GVHD) 3–6. Several possible mechanisms of action that explain the beneficial effects of high-dose IVIg in autoantibody- and immune complex-mediated diseases have been unravelled during recent decades. These mechanisms include inhibition of the binding of immune complexes or cell-bound immunoglobulin (Ig)G to activating Fcγ receptors, saturation of the neonatal FcR resulting in an enhanced clearance of autoantibodies and up-regulation of the inhibitory FcγRIIb 1,2,7. However, the mechanisms by which high-dose IVIg is effective in diseases which are caused mainly by cell-mediated immune responses, such as Kawasaki disease, dermatomyositis, GVHD and acute transplant rejection, have not yet been elucidated fully. Recent in-vitro and mice studies suggest that one of the mechanisms by which IVIg suppresses cellular immunity is by activating CD4⁺CD25⁺forkhead box protein 3 (FoxP3⁺) regulatory T cells (T_regs) 8,9.

T_regs are critical regulators of cell-mediated immune responses and suppress pathogenic immune responses in autoimmune diseases, transplantation and GVHD 10. Current immunosuppressive drugs used to treat or to prevent these diseases exert generalized inhibition of the immune system, thereby disabling protective immunity against pathogens and malignancies. Therapeutic modalities that stimulate T_reg-mediated immune regulation without affecting global immune functions are attractive. Recently, we found that IVIg can activate both human and mouse CD4⁺CD25⁺FoxP3⁺ T_regs in vitro and increase their ability to suppress allogeneic T cell proliferation 11. By triggering functional activation of T_regs, IVIg prevented graft rejection in a fully mismatched skin transplant model. In line with our data, it has been shown that IVIg can prevent mice from developing experimental autoimmune encephalomyelitis 12 and herpes simplex virus-induced encephalitis 13 by enhancing the suppressive capacity and stimulating the expansion of T_regs. In addition, IVIg was found to enhance the suppressive capacity of human T_regs in vitro 14.

These data suggest that IVIg treatment can stimulate T_reg expansion and function, and may therefore be an attractive tool to re-establish T_reg-mediated immune regulation. However, the data on the effects of IVIg on T_regs are restricted to in-vitro and mice experiments and it is unknown whether IVIg treatment actually affects T_reg expansion and function in patients. Therefore, in the present study we analysed systematically the effects of either low- or high-dose IVIg treatment on the percentages, activation status and suppressive capacity of circulating T_regs in autoimmune and immunodeficient patients. We observed increased activation of T_regs after high-dose IVIg treatment, which was not observed for conventional T cells (T_conv). Additionally, T_regs isolated after high-dose IVIg treatment showed enhanced suppressive capacity ex vivo.

Material and methods

Patients

Twenty-seven patients (21 female/six male) who were treated with IVIg were included in this study. To assess whether IVIg had a dose-dependent effect on T_reg activation, we subdivided the patients into two groups: those who received ‘low-dose’ IVIg and those who received ‘high-dose’ IVIg. Because patients with immunodeficiency started initially with a supplementary dose of 0·4–0·6 g/kg, we defined ‘low-dose’ IVIg as a dose ≤0·6 g/kg and ‘high-dose’ IVIg as a dose >0·6 g/kg. The median dose of IVIg received by low-dose patients was 0·39 g/kg (range 0·25–0·59) and that of high-dose patients was 0·98 g/kg (range 0·65–1·71). High-dose IVIg-treated patients showed a significantly higher rise in plasma IgG levels (+17·3 mg/ml) compared to low-dose IVIg-treated patients (+8·2 mg/ml) immediately after treatment (P < 0·001), as quantified by immunoturbometric analysis (Roche Diagnostics GmbH, Mannheim, Germany). At day 7, plasma IgG levels dropped again in both groups (data not shown). The indications for IVIg treatment in these patients are depicted in Table 1a, b. In the high-dose treatment group (Table 2), five patients received IVIg for licensed indications, including common variable immunodeficiency (n = 2), hypogammaglobulinaemia (n = 1) and idiopathic thrombocytopenic purpura (n = 2). One patient with Good's syndrome received high-dose IVIg as supplementary therapy for hypogammaglobulinaemia. Other patients included in Table 2 received high-dose IVIg as anti-inflammatory therapy ‘off-label’, as they did not respond to conventional immunosuppressive treatment. The IVIg preparations used for treatment were Nanogam® (n = 13; Sanquin, Amsterdam, the Netherlands), Kiovig® (n = 9; Baxter, Deerfield, IL, USA), Flebogamma® (n = 4; Grifols, Barcolona, Spain) and Octagam® (n = 1; Octapharma, Lachen, Switzerland). With the exception of three patients, all patients had been treated previously with IVIg, with an average of 28 days (range 21–35 days) before this study. Twenty-one patients were receiving IVIg monotherapy and six patients received additional corticosteroid treatment. All patients showed clinical improvement after treatment.

Table 1a.

Patient characteristics in low-dose intravenous immunoglobulin (IVIg)-treated patients

IVIg treatment indication	No. of patients	Age in years, median (range)	IVIg dose in g/kg, median (range)
Common variable immunodeficiency	5	41 (20–57)	0·38 (0·30–0·48)
Hypogammaglobulinaemia	6	60 (45–77)	0·39 (0·25–0·59)
Agammaglobulinaemia	1	30	0·25

Table 1b.

Patient characteristics in high-dose intravenous immunoglobulin (IVIg)-treated patients

IVIg treatment indication	No. of patients	Age in years, median (range)	IVIg dose in g/kg, median (range)
Polymyositis†	3	41 (33–56)	1·00 (0·67–1·67)
Common variable immunodeficiency	2	40 (37–43)	0·68 (0·65–0·70)
Immune thrombocytopenic purpera	2	68 (65–71)	1·00
Acquired von Willebrand syndrome†	2	52 (40–63)	0·95 (0·93–0·98)
Hypogammaglobulinaemia	1	40	0·66
Good's syndrome/hypogammaglobulinaemia	1	69	0·86
Polyserositis e.c.i.	1	24	0·78
Polychondritis†	1	68	0·69
Dermatomyositis†	1	46	1·71
Systemic vasculitis e.c.i.†	1	64	1·4

^†IVIg treatment was indicated based on unresponsiveness to conventional treatment.

e.c.i., e causa ignota.

Ethics statement

The medical ethics committee of the Erasmus University Medical Centre approved the study, and written informed consent was obtained from all patients participating in the study.

Sample collection and preparation

Heparin-decoagulated blood samples were collected from healthy blood bank donors and from patients immediately before IVIg infusion immediately after IVIg infusion (4–24 h after the start of the infusion) and 7 days after infusion. Due to practical issues, we were unable to obtain blood samples from one patient immediately after IVIg treatment and from two patients 7 days after treatment. Plasma and peripheral blood mononuclear cells (PBMCs) were isolated from whole blood by density gradient sedimentation using Ficoll-Paque (GE Healthcare, Uppsala, Sweden). Plasma was collected from the gradient and centrifuged at 1157 g for 10 min at 4°C to remove thrombocytes and debris. For storage of PBMCs, the cells were cryopreserved in RPMI-1640 (Gibco BRL Life Technologies, Breda, the Netherlands) supplemented with 20% heat-inactivated fetal bovine serum (Hyclone, Logan, UT, USA) and 10% dimethylsulphoxide (DMSO; Sigma-Aldrich, St Louis, MO, USA). Until further analysis, the PBMC samples were stored at −135°C and plasma at −80°C. To minimize the possible effects of interassay variation, measurements in plasma and on PBMC obtained at different time-points from the same patient were performed on the same day.

Antibodies and flow cytometry

For the dentification of T cell subsets, PBMCs were stained with anti-CD3-AmCyan, anti-CD4-allophycocyanin (APC)-H7 and CD8-Pacific blue (all from BD Biosciences, San Jose, CA, USA). Percentages and activation status of T_regs were determined by surface labelling with anti-CD4-APC-H7, anti-CD25-phycoerythrin (PE)-cyanin 7 (Cy7) (both from BD Biosciences), anti-CD127-PE (Beckman Coulter, Immunotech, Marseille, France), anti-CD38-Pacific Blue (ExBio antibodies, Praha, Czech Republic), anti-human leucocyte antigen D-related (HLA-DR)-peridinin chlorophyll (PerCP) (eBioscience, San Diego, CA, USA) monoclonal antibody (mAb) and proper isotype controls. For the detection of memory (CD45RO⁺) T_regs, we additionally labelled the cell surface using CD45RO-fluorescein isothiocyanate (FITC) mAb (Beckman Coulter). Cells (1 × 10⁶) were incubated with the mAb in 50 μl phosphate-buffered saline (PBS) (Lonza, Verviers, Belgium) for 30 min on ice and protected from light. Subsequently, intracellular FoxP3⁺ staining was performed using anti- FoxP3–APC mAb (or rat IgG2a isotype control mAb) purchased from eBioscience, according to the manufacturer's instructions. To study proliferation, intracellular labelling with anti-Ki-67-FITC (or mIgG1 isotype control mAb) from BD Biosciences was performed. After staining, the cells were washed and resuspended in 100 μl PBS for measurement by flow cytometry (FACS Canto, BD Biosciences, San Jose, CA, USA). A minimum of 1·5 × 10⁵ mononuclear cells (MNC) were acquired. Analyses were performed by fluorescence activated cell sorter (FACS) Diva software (BD Bioscience). Viable MNC were gated based on forward-/side-scatter. For the calculation of percentages of cells expressing a certain marker, or for calculation of the geometric mean of fluorescence intensities (MFI), fluorescence of isotype-matched control mAb was subtracted from fluorescence of specific mAb.

Suppression assay

To determine the effect of IVIg treatment on the suppressive capacity of T_regs ex vivo, we collected additional blood from seven patients in the study who received high-dose IVIg. To purify T_regs and T responder cells (T_resp), we first purified non-touched CD3⁺ cells from thawed PBMCs from each time-point by performing immunomagnetic depletion of non-T cells using PE-conjugated antibodies against BDCA1 (Miltenyi Biotec, Bergisch Gladbach, Germany), CD14, CD56 (both from BD Bioscience) CD19, CD56, CD123 (all from Beckman Coulter) as well as CD15- and CD235-microbeads (both from Miltenyi Biotec). The purified CD3⁺ cells were then stained with anti-CD3-AmCyan, anti-CD4-APC-H7, anti-CD25-APC, 7-aminoactinomycin D (7-AAD) (all from BD Bioscience) and anti-CD127-PE mAb (Beckman Coulter Coulter). Subsequently, T_regs were obtained by flow cytometric sorting 7-AAD^–CD3⁺CD4⁺CD127^–CD25⁺ cells using FACS Aria (BD Biosciences). Purity of the T_regs (defined as CD4⁺CD25⁺CD127^–), as determined by flow cytometry, was 98±1%. T_resp were obtained by sorting both 7-AAD^-CD3⁺CD4^– and 7-AAD^–CD3⁺CD4⁺CD127⁺CD25^– cells, which resulted in a purity of 99±0·4% (defined as CD3⁺CD127⁺CD25^– cells), as determined by flow cytometry. Cells were recovered after cell sorting by incubation in culture medium [RPMI-1640 supplemented with penicillin (100 U/ml), streptomycin (100 μg/ml) and 10% fetal bovine serum (Hyclone, Logan, UT, USA)] at 37°C for at least 3 h prior to co-culture.

To compare their suppressive capacity, T_regs purified from the patient blood collected before IVIg treatment, immediately after treatment and at day 7 after treatment were co-cultured with T_resp (5 × 10⁴) obtained before treatment that were stimulated with 1 μg/ml phytohaemagglutinin (PHA) in the presence of 40 Gy-irradiated autologous PBMC (5 × 10⁴) at T_reg : T_resp ratios of 1:4 in culture medium in round-bottomed 96-well plates (Greiner, Alphen a/d Rijn, the Netherlands). After 5 days, supernatants were collected and the cumulative interferon (IFN)-γ secretion by T_resp was quantified by enzyme-linked immunosorbent assay (ELISA) (Ready-SET-Go®; eBioscience) according to the manufacturer's instructions. Each condition was tested at least in triplicate, from which means were calculated. These means were used to calculate the percentage of suppression using the formula: (IFN-γ level_Tresp–IFN-γ level_Tresp+Treg)/IFN-γ level_Tresp × 100%.

Statistical analyses

Differences in measured variables between time-points before and after IVIg treatment were analysed pairwise using the Wilcoxon signed-rank test. For analysis of the T_reg suppression assay we used the paired Student's t-test, as the differences in IFN-γ production by T_resp collected at different time-points before and after IVIg treatment were distributed normally. Two-sided P-values < 0·05 were considered significant. Statistical analysis was performed using GraphPad Prism version 5·0 (GraphPad Software, San Diego, CA, USA). In accordance with the statistical test used, all data presented in dot-plots show median values, but data on the suppressive capacity of T_reg (Fig. 2b) are depicted as means ± standard error of the mean (s.e.m.).

Fig. 2.

Enhanced suppressive activity of regulatory T cells (T_regs) purified from blood after intravenous immunoglobulin (IVIg) treatment. (a) Regulatory T cells (T_regs) were obtained by flow cytometric sorting on 7-aminoactinomycin D (7-AAD)^–CD3⁺CD4⁺CD127^–CD25⁺ cells (T_regs) from peripheral blood mononuclear cells (PBMC) collected from seven high-dose IVIg-treated patients before treatment, immediately after treatment and at day 7 after treatment, and T responder cells (T_resp) were obtained by sorting both 7-AAD^–CD3⁺CD4^– (T_resp I) and 7-AAD^–CD3⁺CD4⁺CD127⁺CD25^– (T_resp II) cells from autologous PBMCs collected before IVIg-treatment. (b) T_regs purified from the patients before, after and at day 7 after treatment were co-cultured with T_resp (5 × 10⁴) obtained before treatment that were stimulated with 1 μg/ml phytohaemagglutinin (PHA) in the presence of 40 Gy-irradiated autologous PBMCs (5 × 10⁴) at a T_reg : T_resp ratio of 1:4 in culture medium in round-bottomed 96-well plates. After 5 days, supernatants were collected and the cumulative interferon (IFN)-γ production by T_resp was quantified by enzyme-linked immunosorbent assay. Each condition was tested at least in triplicate from which means were calculated. From each patient, these means were used to calculate the percentage of suppression using the formula: (IFN-γ level_Tresp–IFN-γ level_Tresp–Treg)/IFN-γ level_Tresp × 100%. Bar graphs represent mean ± standard error of the mean. **P < 0·01, compared to T_resp alone.

Results

IVIg enhances selectively the activation status of T_regs in vivo

IVIg treatment had no effect on the numbers of circulating CD4⁺ and CD8⁺ T cells (Fig. 1a). To study the effects of IVIg treatment on T_regs, we measured the percentages and immunophenotype of circulating T_regs. The gating strategy used to determine CD4⁺CD25⁺FoxP3⁺ cells is shown in Fig. 1b. At baseline, we observed comparable percentages of circulating T_regs in patients who were treated with low-dose IVIg (n = 12) or high-dose IVIg (n = 15) and in healthy controls (n = 10) (Fig. 1c). In addition, baseline activation status of the T_regs in the low- and high-dose patients were comparable to healthy controls, as assessed by expression levels of FoxP3 (Fig. 1d) and HLA-DR (Fig. 1e).

Fig. 1.

Intravenous immunoglobulin (IVIg) treatment selectively activates circulating regulatory T cells (T_regs), but not conventional T helper cells (T_conv). (a) Percentages of CD3⁺CD4⁺ and CD3⁺CD8⁺ in peripheral blood mononuclear cells (PBMCs) did not alter after IVIg treatment. (b) Density plot shows the gating strategy of T_regs and T_conv after gating on CD4⁺ cells. Histograms show intracellular forkhead box protein 3 (FoxP3) expression in T_regs and surface expression of human leucocyte antigen D-related (HLA-DR) on T_regs and T_conv obtained before, immediately after and at day 7 after IVIg treatment of a representative patient. (c) Baseline percentages of CD25⁺FoxP3⁺ T_regs within circulating CD4⁺ T cells, (d) mean fluorescence intensity (MFI) of FoxP3 in T_regs and (e) percentages of T_regs expressing HLA-DR from healthy blood donors (n = 10) and patients treated with low-dose IVIg (n = 12) or high-dose IVIg (n = 15). (f) Percentages of CD25⁺FoxP3⁺ T_regs within the CD4⁺ population before, immediately after and 7 days after IVIg treatment in patients who received low-dose (n = 12) or high-dose IVIg (n = 15). (g) MFI of FoxP3 in T_regs, (h) percentages of T_regs and (i) percentages of T_conv expressing HLA-DR before, immediately after and 7 days after IVIg treatment in patients who received low-dose (n = 12) or high-dose IVIg treatment (n = 15). Horizontal lines represent median. *P < 0·05.

IVIg treatment had no effect on the percentages of circulating T_regs in both low- and high-dose IVIg-treated patients (Fig. 1f). Although low-dose patients showed a slight increase of T_reg percentages, this increase was not significant. Additionally, Ki-67 staining in T_regs showed no increased proliferation of T_regs after IVIg treatment in both groups (data not shown). However, in patients treated with high-dose IVIg, we observed an increased expression level of FoxP3 in CD4⁺CD25⁺FoxP3⁺ T_regs immediately after IVIg treatment, with an average increase of 29% (Fig. 1b,g, P = 0·04). This increased FoxP3 expression level was sustained on day 7 after infusion (+35%, P = 0·02).

Next, we analysed the expression of the activation marker HLA-DR on CD4⁺CD25⁺FoxP3⁺ T_reg, gated as shown in Fig. 1b. We found that the proportions of HLA-DR expressing T_reg increased gradually after IVIg treatment, again only in high-dose-treated patients, and was 37% higher immediately after (P = 0·02) and 39% at day 7 after treatment (P = 0·1) (Fig. 1b,h). Prednisone treatment did not influence the expression of FoxP3 and HLA-DR, as all six patients who were co-treated with prednisone (low-dose n = 3, high-dose n = 3) showed the same patterns as patients without prednisone treatment (data not shown). In addition, the same trends of T_reg activation were found in patients with autoimmune disease and immunodeficient patients who were treated with high-dose IVIg (Supplementary Fig. S1).

To determine whether IVIg also affects the activation status of T_conv, we measured HLA-DR expression on CD4⁺FoxP3^– T_conv (Fig. 1b,i). Compellingly, HLA-DR expression remained low on T_conv after IVIg treatment, even in high-dose IVIg-treated patients, which may suggest that IVIg treatment selectively activates CD4⁺CD25⁺FoxP3⁺ T_regs, but not CD4⁺FoxP3^– T_conv in humans in vivo.

To study the effect of IVIg on the differentiation stage of T_regs, we measured CD45RO expression on T_regs. We found no effect of IVIg treatment on numbers of CD45RO⁺ T_regs in peripheral blood in both groups, suggesting that IVIg did not drive conversion from naive into memory T_regs within the time-frame of the study (data not shown).

IVIg increases the suppressive capacity of T_regs ex vivo

To investigate whether or not IVIg treatment stimulates the suppressive capacity of T_regs, we isolated CD3⁺CD4⁺CD25⁺CD127^– T_regs and autologous CD3⁺CD25^–CD127⁺ T_resp from PBMCs of seven patients who received high-dose IVIg that were collected before, after and at day 7 after IVIg treatment. The gating strategy for flow cytometric purification of these cells is depicted in Fig. 2a. T_resp isolated from blood collected before IVIg infusion were stimulated with PHA and co-cultured with T_regs purified from blood samples collected before, after and at day 7 after treatment in a T_reg : T_resp ratio of 1:4. After 5 days of culture, cumulative IFN-γ production by T_resp was determined. T_regs that were isolated before treatment showed suppression of IFN-γproduction of 41%, while T_regs isolated at day 7 demonstrated a suppression of 70% (P = 0·001, Fig. 2b). Thus, in accordance with the observed increased T_reg activation status at day 7 after IVIg treatment, T_regs isolated at 7 days also showed a higher suppressive capacity. These data suggest that IVIg treatment may stimulate the suppressive function of circulating T_regs for up to at least 1 week after treatment.

Discussion

Multiple ongoing studies are being performed to induce T_reg expansion or stimulate T_reg function in order to control T cell-mediated disorders. Interestingly, the current study suggests that high-dose IVIg treatment may be useful to enhance T_reg activation and suppressive function in vivo. We show that IVIg treatment enhances both the activation status and suppressive function of T_regs in patients. Upon high-dose IVIg treatment, a gradual rise in FoxP3 and HLA-DR expression in circulating CD4⁺CD25⁺FoxP3⁺ T_regs was found, with the highest expression on day 7 after infusion. This finding indicates that IVIg treatment activates T_regs for at least 1 week, when IVIg levels already declined. T_reg activation was clearly dependent upon the IVIg dose, as patients who received low-dose IVIg (≤0·6 g/kg) did not show significant effects. Activation of T_reg was not restricted to patients with autoimmune disease, as immunodeficient patients who received high-dose IVIg also showed a rise in FoxP3 and HLA-DR expression on T_regs. Thus, our data support the generally accepted idea that a high-dose treatment regimen is required in order to gain anti-inflammatory activity 7.

Our ex-vivo functional assays showed that IVIg treatment may not only stimulate the activation status, but may also enhance the suppressive function of circulating T_regs. We observed an increased suppressive function of T_regs that was highest on day 7 after IVIg treatment, which is in line with the time-dependent increase of the activation status after treatment. Supporting our data, previous studies have shown that T_regs with a higher HLA-DR 11 and FoxP3 15,16 expression exert higher suppressive capacity. Further studies are warranted to confirm this finding, as the suppressive function was studied in a small number of patients. Moreover, it will be extremely interesting to monitor the activation status and suppressive function of T_regs after high-dose IVIg therapy over a longer time-period to study the durability of this effect.

In order to assess the suppressive capacity of T_regs, we did not use proliferation as the classical read-out, as interleukin (IL)-2 produced by T_resp in the co-culture system may influence proliferation of T_regs, hence affecting overall [³H]-thymidine incorporation 17. Therefore, we analysed IFN-γ secretion by T_resp, which is independent of the extent of T_regs hyporesponsiveness and may therefore be more unbiased.

While a few previous studies have shown an increase of circulating T_reg numbers in patients treated with IVIg 18–21, we show an increase in both the activation status and functional suppressive capacity of T_reg upon high-dose IVIg therapy in humans in vivo. Although the effects are moderate this finding may be of importance, as in several autoimmune diseases the numbers of T_regs are normal, yet these cells display a functional defect 22–24. A possible explanation for the fact that we did not find an increase in the circulating T_reg numbers after IVIg treatment is that our patients did not have a deficit of T_reg numbers compared to healthy subjects (Fig. 1c), while the previous studies included patients with acute inflammatory diseases which had decreased numbers of circulating T_regs before IVIg treatment 18–20.

Compellingly, high-dose IVIg treatment selectively activated CD4⁺CD25⁺FoxP3⁺ T_regs, but not CD4⁺FoxP3^– T_conv. This observation may be interesting for the clinic, as conventional immunosuppressive drugs used for treating autoimmune diseases and allograft rejection, in particular calcineurin inhibitors, do not only suppress T_conv, but also T_regs 25–32. By selective activation of CD4⁺CD25⁺FoxP3⁺ T_regs, but not T_conv, IVIg may be superior to classical immunosuppressive drugs. Similar observations were also found in vitro and in mice models 11,12,14. In addition, IVIg selectivity for T_regs has also been suggested by Maddur and colleagues 33, who showed that IVIg inhibited T helper type 17 (Th17) cell differentiation and amplification, while it increased the numbers of T_regs among memory T cells.

The mechanism by which IVIg selectively activates T_regs is still unclear. Our previous data 11, together with the study by Ephrem et al. 12, suggest that the activation may be mediated by direct binding of IVIg to an unknown ‘IVIg receptor’ present on T_regs but not on T_conv. Because we observed that human T_regs do not express classical Fcγ receptors 11, this receptor must be a non-classical IgG-binding molecule. Alternatively, IVIg may bind to specific surface receptors by its F(ab′)2 fragment, as suggested by Maddur et al. 33, or IVIg may contain T_reg epitopes that can activate T_regs by presentation on major histocompatibility complex (MHC) II⁺ antigen-presenting cells towards T_regs, as proposed by De Groot and colleagues 34,35.

The current study focused upon patients treated with IVIg for different indications. Although the underlying diseases are variable, the positive effect of high-dose IVIg on T_reg activation is seen consistently in patients treated with high-dose IVIg. These results suggest that the effects of IVIg on T_reg activation are not restricted to a certain disease, but to high-dose treatment. It would, however, be interesting to confirm the dose-dependent effect of IVIg in patients with the same treatment indication in the future.

In summary, we have demonstrated that high-dose IVIg therapy in humans in vivo can activate T_regs selectively and enhance their suppressive capacity. As the effects in our patient cohort were moderate, we propose that T_reg stimulation may be one of the mechanisms by which IVIg controls inflammation in patients with autoimmune and systemic inflammatory disorders, but other mechanisms are probably also required to restore imbalanced immune homeostasis. The data presented in this study underline the multi-faceted character of the immunomodulatory mechanisms of IVIg.

Disclosure

The authors declare no financial or commercial conflicts of interest.

Supporting information

Additional Supporting Information may be found in the online version of this article at the publisher's website:

Fig. S1. Regulatory T cell (T_reg) activation status after high-dose intravenous immunoglobulin (IVIg) treatment. Mean fluorescence intensity (MFI) of forkhead box protein 3 (FoxP3) expression in T_regs from (a) autoimmune disease patients and (b) immunodeficient patients and percentages of human leucocyte antigen D-related (HLA-DR) expression on T_regs in (c) autoimmune disease patients and (d) immunodeficient patients that are obtained before, immediately after and at day 7 after IVIg infusion. Horizontal lines represent median; n.s.: non-significant.