Expansion of CD4⁺CD25⁺ regulatory T cells by intravenous immunoglobulin: a critical factor in controlling experimental autoimmune encephalomyelitis

Natural CD4⁺CD25⁺ regulatory T cells (nTreg) expressing the lineage marker Foxp3 are the key players in controlling immune responses and in maintenance of T-cell homeostasis.^1,2  Therapeutic induction of the Treg represents a novel approach in the treatment of autoimmune pathology.^3,4  CD4⁺CD25⁺Foxp3⁺ nTreg develop in thymus, in contrast to “adaptive” or “induced” Treg that develop in peripheral lymphoid tissues from CD4⁺ conventional T cells (Tconv) and are frequently Foxp3⁻. Although studies have highlighted the role of cytokines interleukin-2 (IL-2), transforming growth factor-β (TGF-β), and IL-10 in Treg development, other factors or mechanisms crucial to Treg homeostasis are not elucidated.⁵  Intravenous immunoglobulin (IVIg) is an established therapy for several immune disorders.^6,,–9  Several mutually nonexclusive mechanisms have been proposed to explain the beneficial effect of IVIg^7,8 ; however, the issue remains debated and an ongoing challenge. For instance, the FcγR-mediated effects of IVIg^10,–12  cannot entirely account for its proven benefit in several peripheral and central demyelinating diseases such as Guillain-Barré syndrome, chronic inflammatory demyelinating polyneuropathy (CIDP), and relapsing-remitting multiple sclerosis (MS), which are primarily mediated by autoreactive T cells.^6,7,13,14  Because the T cells do not express FcγR,¹⁵  the observed effects raise certain speculations, that is, if these effects could be attributed to a direct interaction of the variable region of the immunoglobulin (Ig) G molecules with the T cell or an indirect influence via other cell types such as dendritic cells (DC)

During the induction phase of experimental autoimmune encephalomyelitis (EAE), myelin reactive proinflammatory CD4⁺ T cells proliferate in the periphery, infiltrate the central nervous system (CNS) during the effector phase and, in concert with other inflammatory mediators, lead to demyelination characterized by a progressive paralysis.¹⁶  Natural remission and recovery from relapse in EAE is associated with the recruitment or generation of Treg in the CNS.^17,18  We and others have shown that IVIg protects against EAE development only when administered prophylactically.^14,19  We reasoned that IVIg manifests its protective effect in EAE through an early modulation of autoreactive T cells, and therefore we investigated the regulatory mechanisms, particularly the effect of IVIg on regulatory T cells.

We purchased C57BL/6J mice (females, 6-8 weeks of age) from Charles River Laboratories (L'Arbresle, France), and all animal-handling procedures conformed to European Union guidelines. The MOG-_35-55 (MEVGWYRSPFSRVVHLYRNGK) peptide was purchased from NeoMPS (Strasbourg, France). Tissue culture medium was RPMI 1640 supplemented with 2 mM l-glutamine, 100 IU/mL penicillin, 100 μg/mL streptomycin from Gibco (Cergy Pontoise Cedex, France), and 10% fetal calf serum from Eurobio (Les Ulis, France). The T-cell receptor (TCR)–hemagglutinin (HA) transgenic mice, expressing HA epitope of influenza virus, were kindly provided by Dr Benoît Salomon (Unité Mixte de Recherche 7087, Paris, France), and 6-week-old BALB/cByJ (BALB/c) mice were from Charles River Laboratories. The mice were bred in our animal facility under specific pathogen-free conditions.

C57BL/6J mice (weighing approximately 20 gm) were immunized with 200 μg MOG_35-55 peptide emulsified in Complete Freund's Adjuvant (CFA; Sigma-Aldrich, St. Quentin Fallavier, France) 1:1 by volume containing 800 μg of nonviable desiccated Mycobacterium tuberculosis H37RA (Difco Laboratories, L'Arbresk, France). A final volume of 200 μL was injected subcutaneously at 4 sites over the flanks. In addition, 300 ng of Pertussis toxin (List Biologic Laboratories, Meudon, France) was given intravenously on the same day and 2 days later. Clinical signs of EAE were assessed daily by the following scoring system: 0, no signs; 1, hindlimb weakness; 2, hindlimb weakness and tail paralysis; 3, hindlimb and tail paralysis; 4, hindlimb and tail paralysis and forelimb weakness; 5, moribund; and 6, death. From the day of the immunization until the peak of the disease (day 21-25), mice received daily intraperitoneal injections of 16 mg (0.8g/kg) IVIg (Sandoglobulin or a next-generation IVIg in development, IgPro10, both from CSL Behring) reconstituted at 50 mg/mL in phosphate-buffered saline (PBS). The control groups received either PBS alone or human serum albumin (HSA; Vialebex, LFB Biotechnologies, Les Ulis, France) intraperitoneally at equivalent molar amounts.

Blood of anesthetized mice was collected from the right ventricle in a syringe containing heparin. Mononuclear cells were isolated by floatation over Ficoll-Histopaque d = 1.077 (Sigma-Aldrich). Mice were perfused with cold PBS, followed by removal of spleen and draining lymph nodes (LN) and mechanical disaggregation. The brain and the spinal cord were mechanically disaggregated, and mononuclear cells of the CNS were isolated using 37.5% Percoll (Sigma-Aldrich).

For the analysis of cell surface marker expression, Abs: anti–CD4-fluorescein isothiocyanate (FITC) or anti–CD4-phosphatidylethanolamine (PE)-Cy5, anti-CD25 APC (clone PC61) or relevant isotype controls (all from BD Pharmingen, Le Pont de Claix, France) and PE-conjugated antihuman IgG (Beckman Coulter) were used. AntiFoxP3-PE and isotype control (rat IgG2a-PE; eBiosciences, Montrouge, France) were used per manufacturer's instructions. Cells were acquired and analyzed using fluorescence activated cell sorter (FACS)Calibur cytometer and the CellQuest Pro Software (BD Biosciences, Le Pont de Claix, France).

Spleen cells were cultured in triplicates alone, with 5 μg/mL MOG_35-55 or 1 μg/mL concanavalin A. Supernatants were collected after 24 hours, and cytokine concentration in the supernatant of cell culture was measured in sandwich enzyme-linked immunosorbent assay (ELISA; DuoSet; R&D Systems, Lille, France) according to the manufacturer's instructions.

Donor C57BL/6J mice were subjected to an EAE induction and treated or not with IVIg. At the peak of the disease (day 21), mice were killed and single-cell suspensions were prepared from the spleen, and the draining LN (axillary and inguinal). CD25⁺ T cells were enriched using anti-CD25 mAb (7D4; BD Biosciences) and antibiotin-microbeads (Miltenyi Biotec, Paris, France). The enriched fraction was labeled with streptavidin-PE and anti-CD4-FITC (3M4–5; BD Biosciences) and sorted for CD4⁺CD25⁺ high cells using FACSAria Cell Sorting System (BD Biosciences). The purity was more than 98%. Recipient mice (n = 4) received 0.25 or 0.5 million cells intravenously from IVIg-treated or untreated mice. After 24 hours, mice were subjected to an EAE induction as described in “EAE induction, assessment, and IVIg treatment.”

Single-cell suspension was prepared from the spleen. CD4⁺CD25⁺ T cells were isolated using CD4⁺CD25⁺ Treg-isolation-kit (Miltenyi Biotec) per manufacturer's instructions. For suppression assay, triplicate cultures of 5 × 10⁴ CD4⁺CD25⁻ cells, 1 μg/mL anti-CD3-plate bound (clone 145.2C11), 1 μg/mL anti-CD28, and the indicated number of CD4⁺CD25⁺ cells were incubated in complete medium. After 4 days, the cells were pulsed for 16 hours with 1 μCi (37 × 10³ becquerel) of [³H] thymidine. Radioactive incorporation was measured by standard liquid scintillation counting, and results were expressed as counts per minute (cpm). For proliferation, Treg (5 × 10⁴) were stimulated with 10 ng/mL rmIL-2, plate-bound anti-CD3, and soluble 1 μg/mL anti-CD28 Abs, with or without IVIg as indicated. Cultures were pulsed and harvested as explained above in this section.

In vivo depletion of Treg was performed as previously described.²⁰  Briefly, mice were injected with 100 μg of anti-CD25 antibody, PC61 intraperitoneally at day 10 before EAE induction. Depletion was confirmed by flow cytometry. Mice were injected with IVIg or PBS as explained in “EAE induction, assessment, and IVIg treatment.”

Brachial, axillary, cervical, and inguinal LN and spleen of TCR-HA-transgenic mice were harvested. CD25⁺ cells were enriched as explained in “Purification of CD4⁺CD25⁺ T cells, in vitro suppression, and proliferation assay of Treg.” Purity was 65% of CD25⁺, more than 95% of them being Foxp3⁺. CD4⁺ cells were purified from the CD25⁻ fraction using the mouse CD4⁺ T-cell enrichment set (BD Biosciences) per manufacturer's instructions. To follow cell division, CD25⁻ and CD25⁺ cells were labeled with carboxyfluorescein diacetate succinimidyl ester (CFSE), and BALB/c mice were injected intravenously with either 2 × 10⁶ CD25⁺ cells or 10⁶ CD25⁻ cells or coinjected with 0.9 × 10⁶ CD25⁻ cells and 0.1 × 10⁶ CD25⁺ cells. The next day, mice were immunized with a subcutaneous injection of 2 μg HA_126-138 peptide in CFA in footpad. Mice received similar dose of IVIg intraperitoneally as in EAE for 10 days. Control group received PBS. At day 10, mice were killed, and the draining LN (popliteal), nondraining LN (axillary), and the spleen were mechanically dissociated, and cells were labeled with anti-CD4-PerCp, anti-CD25⁻ APC (PC61) and anti-Foxp3-PE (FJK-16s) and analyzed as described previously.

Two mice from each group were killed by intracardiac perfusion with 4% paraformaldehyde. Brain and spinal cord were rapidly dissected, fixed with 4% paraformaldehyde for overnight at 4°C, incubated for 2 days in 30% sucrose, and then frozen in optimal cutting temperature (OCT). Six-micrometer cryostat serial sections were prepared, fixed in acetone for 10 minutes, rinsed in PBS, and stained with haematoxylin/eosin. The tissue sections were examined by light microscopy in a blinded manner and were evaluated for the extent of inflammation.

Slides were viewed with a Nikon Eclipse E600 microscope (Nikon, Champigny-sur-marne, France) using a Plan Fluor lens at 10×/0.25 and 40×/0.75 and Mount Medium (Merck, Darmstadt, Germany). Images were acquired using a Nikon DXM 1200 digital camera and were processed with a LUCIA G (version 4.71; Nikon) image analysis software and Adobe Photoshop version 7.0 software (Adobe Systems, San Jose, CA).

Analysis of variance (ANOVA) or one-way between-groups ANOVA was used to determine the statistical significance of the data. A value of P less than .05 was considered to be statistically significant.

We induced EAE in C57BL/6J mice and treated them with IVIg, PBS, or HSA as explained in “EAE induction, assessment, and IVIg treatment.” The symptoms of EAE were observed in the PBS group from day 14 onward (Figure 1A); the disease typically peaked between day 21 and day 25 postimmunization, with an average clinical score of 3.2. Mice receiving IVIg from day 0 were almost completely protected from EAE (Figure 1A). The average clinical score was only 0.3 (P < .05), indicating a 10-fold decrease in the severity of the disease. Similar results were also observed using 2 different preparations of IVIg (Sandoglobulin and a next-generation IVIg in development, IgPro10, both from CSL Behring, Zurich, Switzerland). Although disease development was slightly slower in the HSA-treated group, compared with the PBS group, the maximal mean clinical score did not change significantly (3.4).

To determine whether the mechanism by which IVIg protection involves Treg, we studied the changes in CD4⁺CD25⁺Foxp3⁺ T cells in treated and untreated mice by flow cytometry at different time points of the disease progression. We observed an increase in percentage of the CD4⁺CD25⁺Foxp3⁺ T cells as early as day 10 in the IVIg-treated group (Figure 1B,C). The increase was observed in LN and blood, but interestingly, it was most significant in the spleen, where a 2-fold increase (7.78 ± 0.26% in untreated vs 14.87 ± 1.48% in IVIg-treated group; P < .001) was observed. This enhancement was maintained in periphery during all the phases of the disease. Thus, IVIg clearly induced or expanded Treg population throughout the clinical phases of EAE. Interestingly, no Treg expansion was observed in a control group of mice that were infused with IVIg in the absence of MOG immunization (data not shown), indicating that this expansion is specific to T cells that are activated through TCR stimuli.

To study whether IVIg-expanded Treg had an enhanced suppressor function in vivo, we compared the capacity of Treg from treated and untreated group to protect recipient mice against EAE. In general, mice that were reconstituted with Treg from IVIg-treated or untreated mice developed milder EAE compared with nonreconstituted mice (Figure 2A). Further, mice that received 0.25 million Treg from IVIg-treated donor showed a mean clinical score of 0.5 plus or minus 1 compared with 1.25 plus or minus 0.8 in mice that received equal number of Treg from untreated mice (Figure 2A). The incidence of EAE development was 25% in the former group versus 75% in the latter group. Protective effect of the 2 Treg populations after a transfer of 0.5 million cells could not be discriminated because EAE development was totally inhibited in both cases (data not shown).

We further show that the expanded Treg population from IVIg-treated group was more efficient in suppressing the in vitro response of TCR-stimulated CD4⁺CD25⁻ T cells (Figure 2B) compared with Treg from control group. At a Treg versus responder ratio of 1:4, the Treg suppression capacity increased from 16% in the control group to 57% in IVIg-treated group. However, at a ratio of 1:2, we did not observe a difference in suppressive capacity of Treg among 2 groups because of a saturation in percentage of suppression (approximately 100%). Together, these in vivo and in vitro results give a clear indication that in addition to an increase in numbers, IVIg-mediated protection also correlates with a functional enhancement of Treg.

Suppressor cytokines like TGF-β and IL-10 are implicated in some models in the mechanism of action of Treg. To study the effect of IVIg treatment on secreted cytokine profiles, splenocytes of IVIg-treated and -untreated mice were prepared and stimulated with MOG_35-55, and culture supernatants were analyzed for IL-10 and TGF-β. We observed a 4-fold increase in IL-10 level (200 pg/mL and 50 pg/mL in treated and untreated group, respectively) and no significant difference in TGF-β levels (Figure 3A,B). To identify the source of IL-10 in the culture supernatants, we studied intracellular IL-10 expression in splenocytes and found that IL-10 was not produced from CD4⁺ T cells (data not shown).

Our results showing early expansion of CD4⁺CD25⁺Foxp3 cells suggested that the regulation of autoreactive T cells took place in peripheral lymphoid organs rather than in the CNS (ie, IVIg-induced Treg prevent CNS damage by restricting or energizing encephalitogenic T cells at the site of initiation of the immune response). To investigate this hypothesis, we compared the amount of secreted interferon-γ (IFN-γ) by splenocytes purified from IVIg-treated or untreated mice after MOG_35-55 stimulation in vitro. Figure 3C shows that the culture supernatants of MOG_35-55-stimulated lymphocytes from treated mice showed significantly reduced IFN-γ compared with those of untreated mice (700 pg/mL vs 3000 pg/mL, respectively).

The data strongly indicate that Treg, if induced early during EAE, could prevent the activation/generation of effector T cells in lymphoid organs. We investigated the impact of this early suppression on subsequent lymphocytic infiltration and inflammation of the CNS, using histologic studies and flow cytometry. Although CNS tissue sections from untreated mice showed multiple foci of inflammation including lymphocytes and macrophages in perivascular and subpial areas, no histologic evidence of mononuclear cell infiltration was found in the treated group (Figure 4A). Further, we analyzed the T-cell infiltration in the CNS during peak and remission phase of EAE by flow cytometry. We observed a 6-fold increase of CD4⁺ T cells in the CNS of untreated mice compared with naive mice (Figure 4B); of these, 23% were Treg (Figure 4B,C). The CD4⁺ T cell numbers remained unchanged in IVIg-treated mice, and no Treg were detected. Because neither CD4⁺ T-cell infiltration nor tissue inflammation was detected in IVIg-treated mice, it was not surprising that Treg were not recruited in the CNS. Our results indicate that when given prophylactically, IVIg regulated the inflammatory process in periphery, resulting in an absence of infiltration of effector cells in the CNS. However, we cannot rule out its local effect in the CNS in conditions other than those presented in this study.

To further evaluate the implication of Treg in IVIg-mediated protection in EAE, we depleted mice of natural Treg population using a monoclonal antibody, PC61,²⁰  10 days before EAE induction and treatment with IVIg. Antibody PC61 depleted CD4⁺CD25⁺Foxp3⁺ T cells for a period of approximately 30 days. Depletion of Treg abolished the protective effect of IVIg, and the mice developed EAE of same severity as the control group (Figure 5A). The rate of reappearance of Treg was faster in immunized mice; however, their percentages were comparable in IVIg-treated and -untreated groups at the onset of EAE (Figure 5B). These results further indicate that IVIg-mediated protection in EAE was mediated by Treg.

To further explore the mechanisms that lead to IVIg-induced Treg expansion, we used an adoptive transfer model of TCR-transgenic T cells specific for a peptide of influenza virus HA (HA_126-138).²¹  We studied 3 different conditions of adoptive transfer in BALB/c mice in the presence or absence of IVIg infusion: CFSE-labeled HA-specific CD4⁺CD25⁻ Tconv alone, CD4⁺CD25⁺ T cells alone (Treg), and a mixture of CD4⁺CD25⁻ T cells (90%) with CD4⁺CD25⁺ T cells (10%; Tconv + Treg). Cell transfer was followed by immunization with HA_126-138 peptide in CFA. Mice were killed 10 days after immunization, and the CFSE⁺ T cells from the draining LN were gated and analyzed as shown in Figure 6A. Neither an induction of CD4⁺CD25⁺Foxp3⁺ T cells in the Tconv group nor an increase of their proliferation in the Treg group was observed by IVIg treatment (Figure 6B). We reason that because of the strong activation context, IVIg effect on Treg could not be distinguished. However, an increase in CD4⁺Foxp3⁺ proliferation in the presence of IVIg was observed when Treg were coinjected with Tconv (Figure 6B). Next, we assessed the proliferation of Tconv cells. As shown in Figure 6C, IVIg did not suppress the proliferation of Tconv in the absence of Treg. The coinjection of Treg had a mild suppressive effect on the proliferation of Tconv (13%) in the absence of IVIg. Interestingly, the coinjection of Treg and IVIg together yielded a significant reduction (30%) in the proliferation of Tconv (P < .001). Thus, the presence of IVIg enhanced the suppressive capacity of the Treg.

In parallel, we studied the effect of IVIg on the proliferation of Treg in vitro. We purified CD4⁺CD25⁻ T cells (Tconv; > 95% pure) and CD4⁺CD25⁺ T cells (Treg; > 80% pure) from naive mice, stimulated them with a mixture of IL-2, anti-CD3, and anti-CD28 Abs, and observed their proliferation in the presence or absence of IVIg. Although IVIg did not have any effect on the proliferation of Tconv, a significant increase (P < .001) was observed in Treg cultures (Figure 6D).

We further investigated whether IVIg bind directly to Treg. As shown in Figure 7, IVIg directly interacted with CD4⁺ and CD4⁻ cells, as revealed by flow cytometry. However, within the CD4⁺ cells, IVIg binding to Treg was higher (mean fluorescence intensity [MFI] = 272) than that to Tconv (MFI = 132). To get an insight into the molecular interaction of the IgG molecule with cellular targets, we studied the protective effect of equimolar infusions of F(ab)′₂ and Fc preparations of IVIg. We did not observe a difference in protection against EAE and the Treg induction compared with intact IgG preparation (data not shown).

Initially used in primary and secondary immune deficiencies, IVIg is increasingly being used for the treatment of autoimmune and systemic inflammatory diseases. Although considerable progress has been made in understanding the mechanisms by which IVIg exerts immunomodulatory functions in autoimmune diseases, these remain not fully elucidated. In the current study, we demonstrate a novel mechanism of action of IVIg in controlling the disease progression of EAE through upmodulation of Treg. We show that IVIg, when given prophylactically, prevents T-cell infiltration in the CNS through regulation of autoreactive T cells in the periphery, thus preventing the onset of irreversible neurologic damage. We believe that our results are extremely relevant for shaping future clinical strategies for the treatment of autoimmune diseases.

The etiology of autoimmune disease is often associated with dysregulation of the Treg compartment.²²  Indeed, regulatory T cells in patients suffering from autoimmune disease are either functionally defective or reduced in numbers in periphery.²³  Investigations to develop immune-mediated therapies have pursued 2 main lines: targeting autoimmune effector cells and reestablishing tolerance to autoantigens. In the EAE model, we show that IVIg belongs to the second category by modulating the population of Treg. By extrapolation, we can speculate that the beneficial effect of IVIg in treating relapsing-remitting MS (RRMS) may be related to the reestablishment of the Treg compartment, which is reported to be defective in this pathology.²⁴  Furthermore, similar mechanisms of action of IVIg may be operational in pregnancy- and postpartum-related relapses,²⁵  which are also believed to be associated with a variation in pregnancy-related Treg.²⁶  In a recent study on the effects of IVIg in experimental autoimmune myasthenia gravis (EAMG), Treg expansion was not observed.²⁷  Because IVIg was given during chronic disease and not prophylactically, as in our EAE model, we believe that the kinetics of disease progression and regulation would take a different course.

Spontaneous recovery in MS is rare when neurologic deficits have persisted for longer than 6 months, and there are no established therapies that promote neural regeneration and reverse neurologic deficits. Therefore, disease-modifying therapies at an early stage are crucial before neurologic deficits have set in. In our study, we observed that after prophylactic IVIg treatment, the proliferation and function of myelin-specific T cells were suppressed in the peripheral lymphoid organs. There was a decrease in the secretion of pro-inflammatory cytokine IFN-γ. Thus, the IVIg-expanded Treg prevented the proliferation and trafficking of pathogenic T cells toward the CNS, resulting in an absence of encephalomyelitis. This is an important observation that further validates IVIg testing in suitable subsets of early onset MS.²⁸  Indeed, the cumulative probability of developing clinically definite MS was significantly lower in IVIg-treated persons presenting with first neurologic event in a randomized, double-blind, placebo-controlled trial.²⁹ 

Previous studies have shown that IVIg can modulate diverse lymphoid and myeloid cell populations.^30,31  Thus, at therapeutic concentrations used in autoimmune diseases (0.15 mM), IVIg induces a downmodulation of monocyte-derived DC, which in turn fail to support T-cell proliferation.³⁰  We asked whether the mechanism whereby IVIg leads to an expansion of Treg is central to the protection against EAE or if it is a mere bystander effect. Our results show that IVIg failed to protect mice that were depleted of Treg before EAE induction, thereby establishing unequivocally that the presence of natural Treg is crucial to the IVIg-mediated effect.

To get an insight into the mechanisms that led to Treg expansion by IVIg, we used adoptive transfer of HA-TCR transgenic T cells in wild-type (WT) recipients. The results clearly rule out a de novo generation of Treg. This conclusion is also supported by the observation that IVIg did not enhance TGF-β secretion that could have differentiated CD4⁺ T cells into Treg in WT-EAE experiments. Therefore, the proliferation of nTreg is the most probable explanation for the peripheral expansion of Treg in EAE. This interpretation is supported by in vitro results. Furthermore, IVIg interacted directly with the Treg and other cell types, including Tconv and CD4⁻ splenocytes. This is not surprising because IVIg is a source of polyreactive natural antibodies that interact with surface molecules of different T-cell types, contributing to its diverse mechanism of action. Previous reports have shown that IVIg bind to DC³⁰  and to activated T-cell lines specific for myelin binding protein.³²  It would be interesting to identify the cell surface molecules on Treg that interact with IVIg.

As a first step to understand the molecular interaction of the IgG molecule with cellular targets, we studied the protective effects of F(ab)′₂ and Fc preparations of IVIg. We did not observe a difference in protection against EAE and the Treg induction compared with intact IgG preparation. These questions may be better answered by studying protective effects of F(ab)′₂ and Fc preparations of IVIg against EAE in Fc receptor knockout mice. Mice lacking expression of FcγRIIB (FcγRIIB^−/−) have been successfully used in understanding the mechanisms of action of IVIg in ameliorating idiopathic thrombocytopenic purpura.^10,–12 

In conclusion, we have demonstrated that Treg are implicated in the IVIg-mediated protection against EAE. An increase in the population of CD4⁺CD25⁺Foxp3⁺ Treg is associated with the protection against EAE. Our observations explain the beneficial effects of IVIg in diverse autoimmune pathologies because Treg are pivotal in controlling auto reactivity. From the therapeutic perspective, our observations open up possibilities for testing IVIg in suitable subsets of early-onset MS. It has been observed that an absence of circulating IgG in XLA patients renders them susceptible to autoimmune pathologies.³³  Similarly, B cell–deficient mice, when induced with EAE, fail to recover, unlike their WT counterparts.³⁴  These observations indicate a dysregulation of the Treg compartment in the absence of IgG. Thus, the effect of therapeutic IgG on Treg modulation could reflect an important physiologic phenomenon.

We thank Drs D. Pham-Dinh and B. Bellon for assistance with EAE model; J.P. Duang Van Huyen, A. Nicoletti, P. Salameh, and H. Bouharoun-Tayoun for constructive suggestions at various stages of the study; and S. Graff-Dubois, A. Varthaman, J. Khallou-Laschet, A.T. Gaston, B. Wootla, and O. Cena for technical assistance.

This work was supported by Inserm; CNRS; CSL Behring, Zurich, Switzerland, Lebanese National Council for Scientific Research (LNCSR); and Coopération pour l'évaluation et le développement de la Recherche (CEDRE) Beirut, Lebanon. A.E. received a fellowship from CEDRE.

Contribution: A.E. performed laboratory experiments, analyzed data, and wrote the report. S.C. helped in data analysis, writing the report, and acquiring funding. C.M. performed the histology experiments. S.F., J.L.C., and B.L.S. furnished PC61 and TCR-HA transgenic mice and contributed to data analysis. G.C., S.D., S.E., J.B., and S.L.-D. performed laboratory experiments. L.M. contributed to writing the report. M.D.K. and S.V.K. contributed to conception of the study, acquisition of funding, and writing the report. N.M. designed and supervised the study, analyzed the data, and wrote the report.

Conflict-of-interest disclosure: The authors declare no competing financial interests.

Correspondence: Srini V. Kaveri, Inserm UMR S 872, Centre de Recherche des Cordeliers, Equipe 16 Immunopathology and therapeutic immunointervention, 15, Rue de l'Ecole de Médecine,75006 Paris, France; e-mail: srini.kaveri@crc.jussieu.fr; or Namita Misra, Inserm U764, 32, Rue des Carnets, 92140 Clamart, France; e-mail: namita.misra@u-psud.fr.