The objective of this study is to determine the mechanism of action of anti-CD52 mAb treatment in patients with relapsing-remitting multiple sclerosis (RRMS). Experimental autoimmune encephalomyelitis (EAE), an animal model of the disease, was used to address the role of T regulatory cells (Tregs) in the anti-CD52 mAb–induced suppression of the disease. In vitro studies on PBMCs from RRMS patients and matched healthy controls determined the effect of IL-7 on the expansion of CD4+CD25+CD127− Tregs and induction of their suppressive phenotype. This study using EAE animal models of MS has shown that mouse anti-CD52 mAb suppression of clinical disease was augmented by coadministration of IL-7 and partially reversed by anti-IL-7 mAb. In vitro human studies showed that IL-7 induced expansion of CD4+CD25+CD127− Tregs and increased their FOXP3, GITIR, CD46, CTLA-4, granzyme B, and perforin expression. Anti-CD52 mAb treatment of mice with relapsing-remitting EAE induced expansion of Foxp3+CD4+ Tregs and the suppression of IL-17A+CD4+ and IFN-γ+CD4+ cells in peripheral immune organs and CNS infiltrates. The effect was detected immediately after the treatment and maintained over long-term follow-up. Foxp3+CD4+ Treg-mediated suppression of IL-17A+CD4+ and IFN-γ+CD4+ cells in the spinal cord infiltrates was reversed after inducible Foxp3 depletion. Our results demonstrated that the therapeutic effect of U.S. Food and Drug Administration–approved anti-CD52 mAb is dependent on the presence of Tregs.
Studies reporting deficient T regulatory cell (Treg) suppression in relapsing-remitting multiple sclerosis (RRMS) (1) have demonstrated that Tregs from RRMS patients have a decreased ability to suppress T effector (Teff) cell proliferation, which correlates with their decreased FOXP3 and CTLA-4 expression (2). Furthermore, RRMS patients have a lower frequency of CD39+ Tregs, which selectively inhibit Th17 cell cytokine secretion (3). Thus, impaired Treg function may contribute to the Th17 autoimmune response in RRMS.
Alemtuzumab (Lemtrada) is a humanized mAb against the surface CD52 molecule, expressed on all lymphocytes (4). As an effective lymphocyte-depleting therapy, it has been U.S. Food and Drug Administration approved as a treatment for aggressive RRMS for patients with unsuccessful first-line therapy (5). The annual relapse rate and sustained accumulation of disability were significantly reduced in phase III clinical trials (6). Anti-CD52 mAb i.v. treatment has been established as one of the longest lasting suppressors of RRMS progression (7). It efficiently depletes T and B lymphocytes and, to a lesser extent, monocytes, macrophages, dendritic cells, and NK cells via Ab-dependent cellular cytotoxicity and complement-induced cell lysis (8, 9). Although effectively depleting all circulating CD52-bearing cells, this treatment does not affect hematopoietic stem cells, thus preserving the potential for an immune reconstitution (10). The long-lasting clinical disease suppression and an improvement in disability scores have led to the hypothesis that in addition to immune cell depletion, the subsequent differential reconstitution of CD4+ cell subsets may contribute to a decreased new brain lesion formation, demonstrated by brain magnetic resonance imaging scans in phase III clinical trials (5). Although its efficacy is well established and disease activity suppression lasts for years (11), its mechanisms of action has not been elucidated. Anti-CD52 mAb induces lymphocyte lysis, which confers an immunosuppressive effect. However, in the context of this pharmacologically induced lymphopenia, our previous studies have identified a differential reconstitution of T cell subsets, characterized by a 4-fold increase in the percentage of CD25+CD127− Tregs within the CD4+ lymphocytes at month 1, followed by a progressive increase in the numbers of TGF-β+, IL-10+, and IL-4+ regulatory CD4+ cells, and the inhibition of IFN-γ+ and IL-17A+ Teff CD4+ cells at months 12 and 24 posttreatment (4). The in vivo expanded CD4+CD25+CD127− Tregs after anti-CD52 mAb treatment maintained an immunosuppressive FOXP+CD39+Granzyme B+(GZMB+) TGF-β-1+ phenotype, which had a regulatory function in human in vitro studies (12). Several studies have demonstrated that Tregs from alemtuzumab-treated patients suppress IL-17A and IFN-γ production in PBMCs (13), and that alemtuzumab in vitro treatment of CD4+ cells induces increased suppressive function of Tregs by cell-to-cell contact (14).
Our human studies have identified a significant incremental increase in serum IL-7 concentrations from day 7 to 6 mo after anti-CD52 mAb–induced lymphocytopenia. In vitro studies have demonstrated the IL-7 induction of STAT5 phosphorylation and FOXP3 expression and the IL-7–induced proliferation of FOXP3+ Tregs (4).
This study has confirmed that Tregs mediate the therapeutic effect of anti-CD52 mAb in animal models of the disease. The therapeutic effect was enhanced by coadministration of IL-7 and partially reversed by coadministration of anti–IL-7 mAb. An in vitro human study on PBMCs from RRMS patients revealed that IL-7 induced expansion of CD4+C25+CD127− Tregs and increased their FOXP3, glucocorticoid-induced TNFR (GITR), CD46, CTLA-4, perforin, and GZMB expression, indicating suppressive function of expanded Tregs. Studies of experimental autoimmune encephalomyelitis (EAE) animal models confirmed expansion of Foxp3+ and IL-10+ Tregs and suppression of IL-17+ and IFN-γ+ cells in peripheral immune organs and in the CNS of treated mice. We propose that repopulating Tregs in the setting of anti-CD52 mAb–induced lymphopenia led to the suppression of inflammatory responses in RRMS and in animal models of disease.
Materials and Methods
Eight- to twelve-week-old female SJL/J and C57BL/6 mice were purchased from Charles River Laboratories. Depletion of Regulatory T cell (DEREG) mice were purchased from The Jackson Laboratory. All animal procedures were approved by the Institutional Animal Care and Use Committee of the University of North Carolina at Chapel Hill and Thomas Jefferson University.
Mouse anti-CD52 IgG2a mAb was provided by Sanofi Genzyme (Cambridge MA); the Ab production in mice was described by Turner et al. (15).
Actively induced EAE
We used two EAE models: proteolipid protein (PLP)139–151–induced relapsing-remitting EAE (RREAE) in SJL/J mice and myelin oligodendrocyte glycoprotein (MOG)35–55–induced chronic EAE in C57BL/6 and DEREG mice.
Eight- to twelve-week-old female SJL/J mice were immunized with 50 mouseg PLP139–151 peptide per mouse in CFA containing Mycobacterium tuberculosis (4 mg/ml). Pertussis toxin (200 ng) was administered i.p. on days 0 and 2 postimmunization (p.i.). Mice were monitored daily for the development of clinical signs, and starting from day 12 p.i., 10 mice per group received daily s.c. mouse anti-CD52 mAb (200 μg/d) or IgG2a isotype control (200 μg/d), mouse anti-CD52 mAb + IL-7 (1 μg/d), or mouse anti-CD52 mAb + anti–IL-7 mAb (150 μg/d) for 5 consecutive days. Clinical scores were assigned daily as follows: 1, limp tail; 2, hind limb weakness; 3, hind limb paralysis; 4, hind limb paralysis and forelimb weakness; 5, moribund mice; and 6, death.
Six- to eighteen-week-old female C57BL/6 or DEREG Foxp3-DTR (diphtheria toxin [DT] receptor) mice were immunized with 50 μg MOG35–55 (AnaSpec) in CFA consisting of IFA (Difco Laboratories) and 5 mg/ml M. tuberculosis H37RA (Difco Laboratories). A total of 200 ng pertussis toxin (List Biological) was injected i.p. on days 0 and 2 p.i. A total of 200 μg of mouse anti-CD52 mAb was injected s.c. each day per mouse for 5 d (days 11–15 p.i.). For Foxp3+ cell depletion, 200 ng (10 μg/kg) DT was injected i.p. for 4 consecutive days (days 12–15 p.i.). Clinical scores were recorded daily as described earlier.
Serum samples were collected from mouse anti-CD52 mAb–treated mice with RREAE on days 0, 17, and 23 p.i., and IL-7 levels were measured using ELISA (DY407; R&D Systems) in duplicate.
Blood samples were obtained from five untreated RRMS patients and five age-, sex-, and race-matched healthy controls (HCs) who had signed the Thomas Jefferson University–approved Institutional Review Board consent form. PBMCs were separated using Ficoll gradient. Fresh PBMCs were stained for the surface markers CD4, CD25, and CD127 (for gating purposes) and Treg markers CTLA-4, CD39 (eBioscience), GITR, and CD46 (BD Biosciences). Intracellular staining was performed after stimulation with PMA (50 ng/ml) and ionomycin (500 ng/ml) (Sigma-Aldrich) for 2 h, and brefeldin A (1:1000 dilution; eBioscience) was added for an additional 2 h. Cells were fixed, permeabilized, and stained with fluorescein-conjugated Abs against FOXP3, Ahr, GZMB, perforin (eBioscience), and IL-10 (BD Biosciences), as reported previously (4). The percentages of cells expressing each molecule in gated CD4+CD25+CD127− Tregs were determined using a BD FACSAria Fusion flow cytometer and FlowJo software following 48-h stimulation with IL-7 (100 ng/ml).
For the flow cytometry EAE experiments, the SJL mice with RREAE were sacrificed on days 23 and 60 p.i., and C57BL/6 and DEREG mice with chronic EAE were sacrificed on day 25 p.i. On the earlier-mentioned days, blood was collected via cardiac puncture, and mice were perfused with 50 ml of cold PBS containing heparin (10 U/ml; Sigma). PBMCs, lymph nodes (LNs), including superficial cervical, deep cervical, axillary, brachial, and inguinal LNs, spleen, spinal cord, and brain tissue were harvested. Spinal cord and brain tissues were cut into small pieces and digested in PBS-containing Collagenase D (5 mg/ml; Roche) for 45 min at 37°C with a short vortex every 15 min. After digestion, the cells were passed through a strainer and washed with PBS, followed by 38% Percoll (Sigma) gradient separation of the CNS mononuclear cell infiltrates.
The cells were stimulated with PMA and ionomycin for 2 h and with brefeldin A for an additional 3 h before intracellular staining. The cells were stained with anti-CD4 (RM4-5; BioLegend), anti-FOXP3 (FJK16S; eBioscience), anti–IFN-γ (XMG1.2), and anti–IL-17A mAb (TC11-18H10.1) from BioLegend. FOXP3 staining was done using FOXP3 Fixation/Permeabilization reagent (eBioscience). Isotype controls were used to determine the background. The percentage of cells expressing each molecule was determined in gated CD4+ cells using a BD FACSCalibur and Canto (BD Biosciences) Flow Cytometer, with FCS Express software (De Novo Software). We determined the percentages of the indicated cell subsets in PBMCs, LN, spleen, brain, and spinal cord, and present statistically significant changes.
The results of clinical scores and FACS studies were analyzed using two-way ANOVA and Tukey multiple comparisons test (GraphPad Software). The p values <0.05 were considered significant. The comparison of FACS data with two groups was performed using two-tailed Student unpaired t test.
Data availability statement
The data generated in this study are available on request from the corresponding author.
Anti-CD52 mAb treatment suppresses RREAE
Because the anti-CD52 mAb is an approved treatment for RRMS, the RREAE animal model, induced by PLP139–151 peptide immunization of SJL mice (16, 17), is the most suitable model for determining its mechanisms of action. Administration of mouse anti-CD52 mAb at 10 mg/kg s.c. for 5 d, starting at the peak of disease (day 12 p.i.), provides an optimal model for studies of the cellular mechanisms underlying its effects demonstrated in humans (15). In addition, we tested to what extent addition of IL-7 to mouse anti-CD52 mAb enhances the therapeutic effect by expanding Tregs, and whether addition of anti–IL-7 mAb ameliorates the treatment effect by suppressing the expansion of Foxp3+ Tregs.
Forty female SJL mice were immunized with PLP139–151 and divided into four groups (10 mice per group): the first group received mouse anti-CD52 mAb (200 μg/d), the second group received isotype control IgG2a (200 μg/d), the third group anti-CD52 mAb and IL-7 (1 μg/d), and the fourth group anti-CD52 mAb and anti–IL-7 mAb (150 μg/d) on day 12 p.i. for 5 consecutive days. Monitoring of clinical scores over 60 d, which corresponds to five human years posttreatment (18), demonstrated that mouse anti-CD52 mAb suppressed disease activity and prevented the second flare-up in comparison with isotype control (Fig. 1). IL-7 coadministered with anti-CD52 mAb further decreased the disease activity, while anti–IL-7 mAb partially reversed the mouse anti-CD52 mAb treatment effect (Fig. 1). IL-7 measurements in serum of anti-CD52 mAb-treated mice at days 17 and 23 p.i. revealed an increased IL-7 concentration, but the results did not reach statistical significance (Supplemental Fig. 1).
IL-7 induces expansion of human CD4+CD25+CD127− Tregs and increases their expression of FOXP3 and other suppressive markers
Because our human studies reported a longitudinally progressive increase in serum IL-7 levels after anti-CD52 mAb treatment from day 7 to 6 mo posttherapy, and in vitro studies documented IL-7–induced FOXP3 expression and FOXP3+CD4+ cell proliferation (4), we extended the study of the in vitro IL-7 effect in untreated RRMS patients and matched HCs. Fresh PBMCs from five RRMS patients and HCs were stained for CD4, CD25, CD127 (gating), and surface Treg markers CTLA-4, CD39, GITR, and CD46 and intracellular FOXP3, IL-10, GZMB, and perforin expression. Following IL-7 stimulation (100 ng/ml) over 2 d, we found a significantly increased percentage of CD4+CD25+CD127− Tregs within CD4+ cells in both RRMS patients and HCs (Fig. 2, top). Phenotyping of those cells revealed increased expression of FOXP3+, GITR+, CD46+ and coexpressing GITR+CTLA-4+, and Perforin+GZMB+ Tregs in RRMS patients (Fig. 2, bottom). The results suggest contact-dependent (GITR, CD46) and secreted molecule (Perforin, GZMB)-mediated suppression, induced by IL-7, in the setting of anti-CD52 mAb–induced lymphopenia.
mouse anti-CD52 mAb treatment depletes CD4+ T cells in the CNS and in peripheral immune organs in RREAE
We first tested to what extent mouse anti-CD52 mAb treatment of RREAE depletes the CD4+ lymphocytes, as demonstrated in patients with RRMS. On day 25 p.i., mice were sacrificed, and cells were isolated from LNs, spleen, brain, and spinal cord. Mouse anti-CD52 mAb caused a significant depletion of CD4+ T cells in all tested organs (Supplemental Fig. 2A). Percentages of CD4+ cells within peripheral immune organs and the CNS were significantly decreased, with relative sparing of CD4+ cells within LN organs (Supplemental Fig. 2B), suggestive of lymphocyte survival within LNs, as reported previously (19). We propose that homeostatic proliferation of CD4+ cells from LNs may contribute to their reconstitution after anti-CD52 mAb treatment-induced lymphocytopenia.
mouse anti-CD52 mAb treatment of RREAE increases the frequency of Foxp3+ and IL-10+ CD4+ cells and decreases IL-17A+ and IFN-γ+CD4+ cells in peripheral immune organs and the CNS
Studies of the mechanisms of action of anti-CD52 mAb in an animal model of the disease examined T cell subsets reconstitution and determined to what extent the results replicate findings on T cell reconstitution in RRMS patients (4).
Twenty SJL mice were immunized with PLP139–151 and treated with anti-CD52 mAb or IgG2a isotype control at the peak of the disease for 5 d (days 12–16 p.i.). Mice were sacrificed at day 23 p.i., 7 d after completion of the anti-CD52 mAb treatment, and cells were isolated from PBMCs, LNs, spleen, brain, and spinal cord. The percentages of CD4+ T cells expressing Foxp3, IL-10, IL-17A, and IFN-γ were determined by flow cytometry (Supplemental Fig. 3). Mouse anti-CD52 mAb treatment induced a significant increase in the percentage of regulatory Foxp+CD4+ cells in LNs and spinal cord (p < 0.0001 and p = 0.03, respectively; (Fig. 3A), as well as an increase in IL-10+CD4+ cells in PBMCs and LNs (p = 0.04 and p = 0.02; (Fig. 3B) and Foxp3+IL-10+CD4+ cells in LNs (p = 0.0006; (Fig. 3C). In contrast, mouse anti-CD52 mAb-treated mice had decreased percentages of IL-17A+CD4+ cells in PBMCs, LNs, brain, and spinal cord (p < 0.0001, p < 0001, p = 0.007, and p = 0.01; (Fig. 3D), IFN-γ+CD4+ cells in LNs (p < 0.0001; (Fig. 3E), and IL-17A+IFN-γ+CD4+ cells in LNs and brain in comparison with the control group at day 23 p.i. (p < 0.0001 and p = 0.004; (Fig. 3F). Given the immunodepleting effect of this treatment, absolute numbers of Tregs were decreased in LN, while the absolute number of Foxp3+CD4+ Tregs was significantly increased in spinal cord infiltrate of anti-CD52 mAb–treated mice. The treatment significantly decreased absolute cell numbers of IL-17A+ and IFN-γ+ CD4+ cells in most of the tested organs (Supplemental Fig. 4).
mouse anti-CD52 mAb treatment induces long-term increased percentages of Foxp3+ and IL-10+ CD4+ cells and decreased percentages of IL-17A+ and IFN-γ+ CD4+ cells in multiple organs
To address the long-term effect of the mouse anti-CD52 mAb therapy, we studied the response to treatment at days 12–16 p.i. at a late time point (day 60 p.i.). Consistent with the long-lasting suppression of clinical scores (Fig. 1), mouse anti-CD52 mAb treatment caused a significant increase in the percentages of Foxp3+CD4+ Tregs in LNs, spleen, brain, and spinal cord (p = 0.0006, p = 0.002, p = 0.02, and p = 0.0008; (Fig. 4A), as well as increased percentages of IL-10+CD4+ cells in PBMCs, LNs, and spleen (p = 0.002, p = 0.004, and p = 0.003; (Fig. 4B) and an increased percentage of Foxp3+IL-10+ double-positive CD4+ cells in LNs and spleen in comparison with control mice at day 60 p.i. (p = 0.0006 and p = 0.03; (Fig. 4C). In contrast, the percentages of IL-17A+CD4+ cells were decreased in PBMCs, brain, and spinal cord (p = 0.0004, p = 0.001, and p = 0.02; (Fig. 4D), IFN-γ+CD4+ T cells were decreased in the brain (p = 0.002; (Fig. 4E), and the percentage of IL-17A+IFN-γ+CD4+ cells was decreased in the brains of mouse anti-CD52 mAb–treated mice at day 60 p.i. (p = 0.01; (Fig. 4F).
Absolute numbers of Foxp3+CD4+ cells were increased in LN, spleen, brain, and spinal cord; however, the changes were statistically significant only in LN and spinal cord (Supplemental Fig. 5A). The absolute numbers of IL-10+CD4+ cells were significantly increased in PBMCs, LN, and spleen, and Foxp3+IL-10+ double-positive cells in the LN (Supplemental Fig. 5B, 5C). The numbers of IL-17+CD4+ cells were significantly decreased in PBMCs and brain, while the IFN-γ+ and IL-17+IFN- γ+ CD4+ cells were decreased in the brain infiltrates (Supplemental Fig. 5D–F), indicating a long-term effect on Treg expansion and IL-17+ and IFN-γ+ CD4+ cell depletion in treated mice.
A comparison between the fold changes between percentages of treated and untreated mice at two time points (days 23 and 60 p.i.) related to acute and long-term clinical effects indicated that the increase in Foxp3+CD4+ cells was maintained in PBMCs, spleen, brain, and spinal cord at day 60 p.i. in comparison with day 23, suggesting that long-term disease suppression is dependent on the presence of Foxp3+CD4+ cells. The decrease (fold change between percentages of treated and untreated mice) of IL-17A+CD4+ cells was maintained in spleen, brain, and spinal cord at day 60 p.i., suggesting treatment-induced long-lasting suppression of Th17 inflammatory responses (Figs. 3, 4).
mouse anti-CD52 mAb suppresses chronic EAE, which is abrogated by transient inducible depletion of Foxp3+ T cells
We next tested the therapeutic effect of mouse anti-CD52 mAb in chronic EAE induced in C57BL/6 mice by MOG35–55 peptide immunization (six mice per group). Mouse anti-CD52 mAb administered at the onset of disease (day 11 p.i.) for 5 d (200 μg per mouse) significantly decreased clinical scores in comparison with untreated mice (PBS) (Fig. 5A).
To directly confirm the critical role of Foxp3+ Tregs in the therapeutic effect of mouse anti-CD52 mAb, we used inducible transient depletion of Foxp3+ Tregs after DT administration to Foxp3-DTR-GFP knockin mice (20). These DEREG mice express simian DTR-enhanced GFP transgene under control of the Foxp3 promoter, allowing for Foxp3+ Treg depletion after DT administration (21). Rodents are resistant to DT due to their 105 times lower DTR affinity, which was confirmed by DT treatment of mice with EAE, which did not show any toxicity (data not shown). In contrast, transgenic expression of high-affinity primate DTR in a specific cell type allows for their specific ablation on DT injection. DT blocks protein synthesis and induces rapid apoptotic cell death (22). This system has been widely used to study the effect of transient depletion of specific cell types (CD4, CD19, dendritic cells) in the inflammatory response in mice with DTR transgene under the control of a cell-type–specific transcription factor (23). Because this mouse is generated on a C57BL/6 background, we used the chronic EAE model induced by MOG35–55 peptide in DEREG mice and treated them at the onset of disease (days 11–15 p.i.) with mouse anti-CD52 mAb (200 μg per mouse) as in (Fig. 5A. To deplete Foxp3+ Tregs, 200 ng DT was administered s.c. daily for 4 d with anti-CD52 mAb treatment for 5 d on days 11–15 p.i. Clinical scores were monitored in 10 mice per group daily. We confirmed a significant clinical suppression of chronic EAE after mouse anti-CD52 mAb treatment as demonstrated in C57BL/6 WT mice. Coadministration of DT, which depleted Foxp3+ Tregs, completely ameliorated the therapeutic effect (Fig. 5B).
Mouse anti-CD52 mAb treatment depletes CD4+ T cells in peripheral immune organs and in spinal cord inflammatory infiltrates in chronic EAE
Studies of the CD4+ cell numbers in the WT C57BL/6 mice LN, spleen, and spinal cord (six mice per group) revealed significant depletion of absolute CD4+ T cell numbers after anti-CD52 mAb treatment (Supplemental Fig. 5A). Similar results were obtained when determining the percentages of CD4+ cells in the LN, spleen, and spinal cord inflammatory infiltrates (Supplemental Fig. 5B). In DEREG mice treated with mouse anti-CD52 mAb (four mice per group), we also detected decreased absolute numbers of CD4+ cells in LN, spleen, and spinal cord, which were partially reversed in mice that received DT (Supplemental Fig. 5C). Similar results were obtained when determining the percentages of CD4+ cells in the spleen, LN, and spinal cord inflammatory infiltrates (Supplemental Fig. 5D).
mouse anti-CD52 mAb treatment of chronic EAE induces Foxp3+ Treg expansion and suppression of IL-17A+ and IFN-γ+CD4+ cells, which are abrogated after inducible depletion of Foxp3+ Tregs
Flow cytometry studies of the LN, spleen, and spinal cord infiltrates in control (PBS) and mouse anti-CD52 mAb-treated DEREG mice revealed a significant increase in absolute numbers of Foxp3+CD4+ cells in spleen and LNs of anti-CD52 mAb–treated mice, which were completely depleted in mice that received mouse anti-CD52 mAb + DT (p = 0.01 and p = 0.02; (Fig. 6A). Absolute numbers of proinflammatory IL-17A+CD4+ cells were significantly decreased in spleen (p = 0.03) and spinal cord (p = 0.01) of mouse anti-CD52 mAb–treated mice, which were partially reversed with DT administration (Fig. 6B). IFN-γ+CD4+ cell numbers were decreased in spleen, LNs, and spinal cord (p = 0.02, p = 0.07, and p = 0.001) and were partially reversed in mice that received DT (Fig. 6C). The percentages of Foxp3+CD4+ cells were increased in spleen and LNs of anti-CD52 mAb–treated mice (p = 0.0006 and p = 0.001) and were completely depleted after DT administration (Fig. 6D). The percentages of IL-17A+CD4+ (p = 0.02; (Fig. 6E) and IFN-γ+CD4+ cells in spinal cord were decreased by mouse anti-CD52 mAb treatment and reversed by DT administration (both p = 0003; (Fig. 6F).
Our published study on alemtuzumab-treated RRMS patients suggested that the therapeutic effect is mediated by nearly complete depletion of lymphocytes from the peripheral circulation, and that differential reconstitution of T cell subsets may contribute to the delayed repopulation of CD4+ cells. The delayed reconstitution of CD4+ cells, resulting from deficient CD4+ cell homeostatic proliferation, susceptibility to apoptosis, a relative expansion of immunoregulatory CD4+CD25+CD127low Tregs, and induced peripheral IL-10+CD4+ Tregs, is proposed to mediate the alemtuzumab-induced long-lasting clinical disease suppression.
In this animal study, we show that mouse anti-CD52 mAb causes disease suppression in two animal models, PLP139–151–induced RREAE in SJL mice and MOG35–55–induced chronic EAE in C57BL/6 mice, in comparison with controls, which had a second flare-up in RREAE and chronic disease progression in the chronic EAE model. These animal models recapitulate some aspects of human disease and provide an opportunity to study CNS infiltrates, which are not accessible in human studies (24).
IL-7, a key cytokine promoting homeostatic proliferation in the setting of lymphocytopenia (25), is significantly increased in the serum in the longitudinal study of alemtuzumab-treated RRMS patients from day 7 to 6 mo posttreatment in comparison with the baseline levels (4). The results suggested its role in the expansion of Tregs after alemtuzumab-induced immunodepletion. Clinical data in this study from SJL/PLP139–151 RREAE mice confirmed that addition of IL-7 to the mouse anti-CD52 mAb enhanced its therapeutic effect, while addition of anti–IL-7 mAb decreased the anti-CD52 mAb treatment effect. Although the effect of IL-7 is complex and also induces expansion of Teff cells (26), our results provide clinical evidence that administration of IL-7 in the setting of anti-CD52 mAb–induced lymphocytopenia promotes Treg expansion and ameliorates clinical disease. The increased IL-7 serum levels of mice treated with anti-CD52 mAb did not reach statistical significance but are in line with the earlier human studies from anti-CD52 mAb–treated RRMS patients and with the earlier mouse clinical data. The differences in the increase in serum IL-7 in treated mice and patients may be related to species differences in production of this cytokine.
To further test the effect of IL-7 on the expansion and phenotype of Tregs in RRMS patients, we performed an in vitro study where CD4+CD25+CD127low Treg expansion was demonstrated after IL-7 stimulation in both RRMS patients and matched HCs. IL-7 increased expression of Treg transcription factor FOXP3 and FOXP3-regulated Treg marker GITR, which induces Treg proliferation and expansion (27). IL-7 also increased the expression of CD46, whose activation stimulates IL-10 secretion from inducible Tregs, and coexpression of GITR and CTLA-4 and secreted enzymes perforin and GZMB, which mediate Treg suppression via cell lysis. We propose that the phenotype of IL-7–expanded Tregs reflects the mechanism of their suppressive activity on homeostatic proliferation. Because the phenotype of Tregs detected in alemtuzumab-treated patients after 1 mo had also increased GITR, GZMB, and perforin expression, we propose that the Treg reconstitution is mediated via increased IL-7 serum levels in the setting of treatment-induced lymphocytopenia.
Ex vivo studies of the CD4+ cell subsets in RREAE have demonstrated that anti-CD52 mAb induced increased percentages of Foxp3+CD4+ cells in the peripheral immune organs and in the CNS. In addition, we have demonstrated suppression of IL-17A+CD4+ cells in PBMCs, LNs, brain, and spinal cord in mice with RREAE 5 d posttreatment. Consistent with the long-lasting suppression of the disease, we have also demonstrated an increased percentage of Foxp3+CD4+ cells in the peripheral immune organs and the CNS, as well as the suppression of IL-17A+CD4+ cells in the PBMCs and CNS infiltrates at day 60 p.i. These results are consistent with a study by Turner et al. (15), who reported a similar decrease in RREAE and chronic EAE scores after mouse anti-CD52 mAb treatment and depletion of CD4+ cells in spleen and CNS infiltrates. They reported a significant decrease in the number of spleen-derived IL-17A+ and IFN-γ+ CD4+ cells when stimulated in vitro with MOG35–55. However, they did not study repopulation of the CD4+ cell subsets, particularly Tregs. Consistent with our data, Pant et al. (28) have reported that mouse anti-CD52 mAb treatments of chronic EAE significantly decreased inflammatory infiltrate and demyelination in the spinal cord and increased IL-10, while decreasing IL-17, IFN-γ, and RORγτ gene expression in treated mice.
Finally, the reversal of anti-CD52 mAb treatment effect in DEREG mice with inducible transient depletion of Foxp3+ Tregs directly demonstrated a role of Tregs in the suppression of IL-17A+ and IFN-γ+CD4+ cells in the spinal cord infiltrates and in the therapeutic mechanism of anti-CD52 mAb. Because Moltedo et al. (29) reported transient generalized lymphopenia restricted to peripheral, but not thymic, T cells at day 2 after DT administration, which subsequently had increased T cell activation by day 4, we would like to emphasize that our study determined T cell responses at day 9 after DT administration when lymphocyte counts have normalized. Although because of limited Treg numbers we did not demonstrate recovery of Treg suppressive function in the in vitro suppressive assay, clinical suppression of EAE after anti-CD52 mAb–induced Treg persistence indicates that Tregs functionally suppress IL-17A+ and IFN-γ+ CD4+ responses in vivo. We did not treat DEREG mice with DT, because that had already been reported by Koutrolos et al. (30), who reported that DT-induced depletion of Tregs severely exacerbated EAE. The study confirmed Treg control of Teff cells in the CNS of mice with EAE. We chose to use transient selective Treg depletion in DEREG mice with EAE, similar to a study by Buenafe et al. (31), to provide a proof of principle that anti-CD52 mAb treatment effect is dependent on the presence of Tregs. Because the anti-CD52mAb effect in DEREG + DT–treated mice is similar to, but not more severe than, in WT mice (30), it is possible that anti-CD52 mAb therapeutic effect is partially mediated independently of Tregs.
In summary, the current results indicate that the mechanism of the long-lasting anti-CD52 mAb clinical disease suppression is mediated via Treg suppression of inflammatory IL-17A+ and IFN-γ+ CD4+ cells, as demonstrated in anti-CD52 mAb–treated RRMS patients (4).
Alemtuzumab is a very effective, long-lasting therapy approved for selected patients with aggressive RRMS with unsuccessful first-line therapies who are at risk for progressive multifocal leukoencephalopathy associated with anti–VLA-4 mAb and anti-CD20 mAb. However, its use is limited because of prevalent side effects, particularly secondary autoimmunity observed in 35% of treated patients. This study did not address B cell repopulation, which may lead to the expansion of Ab-producing memory B cells and Ab-mediated thyroid disease, thrombocytopenia, and glomerulonephritis (32). Our previously reported human study (4) demonstrated that increased frequencies of IL-4–, TGF-β–, and IL-10–producing CD4+ cells may contribute to B cell maturation (33), regulation of Ab isotypes (34), and the induction of Ab production (35). Future studies of B cell repopulation are needed to understand the induction of secondary autoimmunity and to advise about add-on therapies. In the current therapeutic landscape, anti-CD52 mAb is a treatment for selected patients.
We thank Dr. Tingting Zhan for assistance with statistical analysis, Mary Sweeney for help with patient recruitment, and Katherine Regan for editorial assistance.
This work was supported by grants from Sanofi Genzyme Inc. (to S.M.-P.), the National Institutes of Health (AI111592 to S.M.-P.; AI123193 to Y. Wan), and the National Multiple Sclerosis Society (RG-1802-30483 to Y. Wan).
N.K. and B.W. completed the experiments, analyzed the results, and wrote the paper. Y. Wan, M.S., S.K., X.Z., M.E., and E.K. performed the experiments. Y. Wang designed and supervised the study. S.M.-P. designed and supervised the study and wrote the paper.
The online version of this article contains supplemental material.
Abbreviations used in this article:
Depletion of Regulatory T cell
diphtheria toxin receptor
experimental autoimmune encephalomyelitis
myelin oligodendrocyte glycoprotein
relapsing-remitting experimental autoimmune encephalomyelitis
relapsing-remitting multiple sclerosis
T regulatory cell
The authors have no financial conflicts of interest.